Four-State Dairy Nutrition and Management Conference · Four-State Dairy Nutrition and Management...

Four-State Dairy Nutritionand Management Conference

Cooperative Extension for:Iowa State UniversityUniversity of IllinoisUniversity of MinnesotaUniversity of Wisconsin

June 10 & 11, 2009 Dubuque, Iowa

Table of Contents

Sponsors and Speakers

Sources of Yeast and Factors Influencing Yeast Viability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

HOT Feeding Strategies to Maximize Milk Yield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

Dealing with Issues of Risk: Is There More to It Than the Average? . . . . . . . . . . . . . . . . . . . . . . 14

A Meta-Analysis of the Responses to Feeding Diamond VYeast Culture to Lactating Dairy Cows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Research Update from the UMN College of Vet Medicine . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

What is the Most Critical Feeding Period: Far-off Dry,Pre-fresh Transition, or Post-fresh Transition? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Moderate Energy Diets and Forage Options for Dry Cows . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Managing Nitrogen for Profit and Stewardship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Should Protected Choline or Niacin be Fed toPeriparturien Dairy Cows? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

Physically Effective Fiber and Regulation of Ruminal pH:More Than Just Chewing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Inflammation and Transition Cow Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

Picking Up The Pieces After a Down Milk Market:Strategies for Deciding What Goes Back in the Ration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

Freshening the First Calf Heifer: Opportunities and Challenges. . . . . . . . . . . . . . . . . . . . . . . . . 87

DCAD Balancing for Lactating Cows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

Cow Comfort and Health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Are Cross Ventilated Dairy Barns Comfortable? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

New Concepts on Heat Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Optimizing Income Over Feed Supplement Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

Milking Management Systems: Your Computer Can Tell YouAbout More Than Just Reproduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

Sand Bedding and Sand Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

Maximizing Comfort for Uncomfortable Cows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

Five Steps to Creating the Ideal Transition Cow Barn . . . . . . . . . . . . . . . . . . . . . . . . . 135

Robotic Milking: What Do We Know . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

Automatic Calf Feeders. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Sponsors and SpeakersThe program committee deeply appreciates the following for their support and commitment tostrengthening the Midwest dairy industry.

Platinum Co-Sponsors

Arm & Hammer Animal Nutrition

Diamond V

Gold Co-Sponsors

Dairyland Laboratories, Inc.

Elanco Animal Health

Mycogen Seeds

Pioneer Hi-Bred International

Varied Industries Corp

Silver Co-Sponsors

Adisseo

AgSource

AgSource Cooperative Services

Ajinomoto USA, Inc.

Alpharma Animal Health

Alltech, Inc.

Amino Plus/Ag Processing, Inc.

Ana-Tech FA LLC

Balchem Corporation

BASF Plant Science

BioZyme, Inc.

Byron Seeds, LLC

Central Life Sciences

Cumberland Valley Analytical

Services

Dairy One Forage Lab

Digi-Star, LLC

Gladwin A. Read Co.

IVESCO

KS Dairy Consulting/Feed Supervisor

Software

Kemin Industries

MIN-AD, Inc

MSC Specialty Nutrition

Nutrition Horizons

Novus International, Inc

Papillon Agricultural Co.

Prince Agri Products, Inc.

Pro*Cal, Inc.

Quali Tech Inc.

Rock River Laboratory

SoyBest

SoyPLUS / SoyChlor

Zeeland Farm Service

Zinpro Performance Minerals

Bronze Co-Sponsors

Agri-Nutrition Consulting, Inc.

CHR Hansen, Inc.

Lallemand Animal Nutrition

Virtus Nutrition

Speakers:Barry Bradford

Kansas State UniversityMike Allen

Michigan State UniversityJohn Fetrow

University of MinnesotaBill Sanchez

Diamond VRic Grummer

University of Wisconsin-MadisonNoah Litherland

University of MinnesotaJoe Harrison

Washington State UniversityNigel Cook

University of Wisconsin-MadisonJoe Harner

Kansas State UniversityLance Baumgard

Iowa State UniversityVictor Cabera

University of Wisconsin Extension

Dick WallaceUniversity of Illinois

Jan ShearerIowa State University

Jim SalferUniversity of Minnesota

Jim PaulsonUniversity of Minnesota

Post Conference Workshops:Workshop #1Dave Kammel – Transition and Special Needs CowBarn Design Process

Workshop #2Mike Hutjens – Building Rations in Today’sEconomic Environment

Workshop #3Jeff Reneau – New Technologies for On-FarmTroubleshooting

Workshop #4Larry Tranel – Analyzing Dairy Farm Profits UsingDairy TRANS Software

Sources of Yeast and Factors Influencing Yeast Viability

Barry Bradford

Kansas State University

Yeast in dairy rations

• Yeast products are used in a significant portion of U.S. dairy herds

• Reasons:Reasons:– Digestibility

– Nutritional supportduring periods of stress

– Milk production

Mode(s) of action?

• Do NOT colonize the rumen

• Not likely to ever represent a significant portion of the microbes in the rumen

• Proposed to stimulate growth of certain bacteria by:– Scavenging oxygen

– Releasing growth factors

Two approaches to yeast

• Yeast culture: – Grow yeast and dry with media

– Feed to provide the growth factors in the media already released by the yeasty y y

• Active dry yeast– Grow yeast and dry, maintaining viability

– Feed an adequate number of viable yeast cells to allow for oxygen scavenging and/or growth factor release during growth in rumen

Yeast dose for ADY products

• Commercial recommendations range from 20 – 60 billion colony forming units (CFU)per cow/day

Sources of Yeast and Factors InfluencingYeast Viability

Barry BradfordKansas State University

1

Expressing viable yeast data

6

8

10

12

ow

per

day

Log 10 scale

3 E+10

4 E+10

5 E+10

6 E+10

ow

per

day

Linear scale

0

2

4

Diet A Diet B Diet C Diet D

CF

U/c

o

0 E+00

1 E+10

2 E+10

Diet A Diet B Diet C Diet D

CF

U/c

o

• 50 billion = 10.7 log10

Yeast sources

• Silage

• Fermentation byproducts

• Commercial• Commercialsupplements

Yeast in forages

5678

/gra

m

Yeasts and Molds in Standing Crops

012345

Standing corn Standing alfalfa

Log

10C

FU

C. Lin et al., J. Dairy Sci. 75, 2484 (1992)

Yeast in “normal” corn silage

3

4

5

6

7

8

10C

FU

/gra

m

0

1

2

Kim Kung Nishino

Log

1

S. C. Kim, A. T. Adesogan, J. Dairy Sci. 89, 3122 (2006)L. Kung, Jr. et al., J. Dairy Sci. 83, 1479 (2000)N. Nishino et al., J. Dairy Sci. 87, 2563 (2004)

Corn silage as a yeast source?

Ingredient Pounds as-fed Pounds dry matter

Corn silage 41 14

Ground corn 22 20

• Example diet:

Ground corn 22 20

Alfalfa hay 11 10

Whole cottonseed 5.7 5.1

Expeller soybean meal 5.1 4.6

Soybean meal 4.3 3.8

Micronutrient premix 1.0 1.0

6

8

10

12

cow

per

day

Corn silage as a yeast source?

Targetdose

0

2

4

Kim Kung Nishino

Log

10C

FU

/c

2

Yeast is yeast?

Yeast supplements

• Saccharomyces cerevisiae

Corn silage

• Candida species

• Saccharomyces dairensis

Products:

• Bread

• Beer

• Wine

Products:

• Silage heating and spoilage

• Infections during immunosuppresion

Yeast in exposed corn silage

56789

10

Yeast, log 10 CFU / gram

01234

0 36 68 144

Hours exposed to air

L. Kung, Jr. et al., J. Dairy Sci. 83, 1479 (2000)

Silage face heating

Courtesy of Dr. Tom Oelberg, Diamond V

• The effects of “wild” yeast on rumen function have not been evaluated

• Spoiled silage was tested in growing steers

Spoiled silage

steers

• Diet was 90%corn silage

• Silage was 0%, 25%,50%, or 75% spoiled

Spoiled silage

50%

55%

60%

65%

NDF Digestibility, %

16

17

18

Dry matter intake, lb/day

a

b

a

abb b

30%

35%

40%

45%

0% 25% 50% 75%

13

14

15

16

0% 25% 50% 75%

bcc

L. A. Whitlock et al. Cattlemen's Day 2000. Kansas State University.

Lactobacillus buchneri

3

4

5

Yeast in corn silage, log 10 CFU / gram

0

1

2

Untreated Low rate LB High rate LB

D. H. Kleinschmit, L. Kung, Jr., J. Dairy Sci. 89, 4005 (2006).v

3

Yeast in silages

• Large populations of yeast in silages are a sign of spoilage

• Should be avoided by packing and sealing silage as well as possible at harvestsilage as well as possible at harvest

• Keep silage faces as fresh as possible

• L. buchneri can suppress yeast growth

Yeast sources

• Silage



The wet milling process

Steep

SteepLi

Grind

Germ

Screen Spin

Bran Gluten Starch

Acid

LiquorGerm Bran Gluten Starch

ETHANOLCORN GLUTEN

FEEDGLUTEN

MEAL

CornOil

Yeast

The dry milling process

Grind

Distillers

Heat FermentYeast

Grains

Solubles

Distill

ETHANOL

DISTILLERS GRAINS WITH

SOLUBLES

Yeast in distillers grains

• The solids in corn distillers solubles are primarily yeast cells– Saccharomyces cerevisiae

SolublesC. Liu et al., ApplBiochem Biotechnol137-140, 875 (2007)

Yeast & mold in wet CDGS

R. M. Lehman, K. A. Rosentrater, Can J Microbiol 53, 1046 (2007)

4

Yeast species in wet CDGS

• Species identified:- 3 Candida species

- 1 Cryptococcus species

- Trichosporon asahiiTrichosporon asahii

• No Saccharomyces cerevisiae!- Unlikely to serve as a source of viable S. cerevisiae,

but may provide putative growth factors

R. M. Lehman, K. A. Rosentrater, Can J Microbiol 53, 1046 (2007)

Yeast in corn byproducts

• Yeast is not used in the production of corn gluten feed

• Few yeast cells appear to survive the process of CDGS production evenprocess of CDGS production, evenwithout drying

• As with silage, large populations of yeast in distillers grains are a sign of spoilage

Yeast sources

• Silage



Yeast dose for ADY products

• Commercial recommendations range from 20 – 60 billion CFU per cow/day.

• To meet these targets, recommended feeding rates of commercial ADY productsfeeding rates of commercial ADY productsrange from 1 gram to 10 grams / day.

• These feeding rates are based on guaranteed CFU / gram for the products.

Viable yeast in ADY products

• Samples of 6 commercial ADY products were purchased through normal distribution channels in spring and summer of 2007

• Three lots of each product were sampled

• Samples were analyzed within 2 weeks of receipt to determine CFU / gram

Viable yeast assessment

• Colonies formed on growth media are counted after dilution

5

CFU/gram relative to claim

100%

120%

140%

160%

180%

200%

Mean: 123% 8053% 12%* 84%* 79% 21%

0%

20%

40%

60%

80%

100%

Product A Product B Product C Product D Product E Product F

*Product claim in the U.S. is given in cells/gram, not CFU/gram

ADY viability - results

• Only 1 product met the guaranteed CFU/gram in each of the 3 samples

• Numerous samples contained less than 10% of the product claim10% of the product claim

• Conclusion: cows may not be getting a consistent dose of viable yeast

Sources of ADY variability?

• By design, we were blind to:– Length of storage prior to receipt

– Storage conditions at distributor

– Conditions in transitConditions in transit

• Does high-temperaturestorage affect yeastviability?

High-temperature storage

• A single sample of each of 6 products was stored in a sealed container at 104°F with ambient humidity

• Analyzed for CFU/gram after 1, 2, and 3 months of storage

High-temperature storage

800

1000

1200

1400

1600

min

gun

itsa

rant

ee)

Product AProduct BProduct CProduct DProduct EProduct F

0

200

400

600

800

0 1 2 3

Col

ony

for

m(%

ofgu

a

Storage time (months)

Product F

• Viable cell yield was significantly decreased by high-temperature storage

• Mean CFU/gram decreased by more than 85% with each month of storage

High-temperature storage results

85% with each month of storage

• All products were affected by high-temperature storage

6

Getting live yeast into cows

• Only Saccharomyces cerevisiae has been evaluated for effects on ruminal fermentation and productivity

• Silages and byproduct feeds do notSilages and byproduct feeds do notprovide viable S. cerevisiae

Getting live yeast into cows

• Active dry yeast products can be highly variable in the amount of viable yeast cells they provide.

• Minimize variability by:• Minimize variability by:

• Minimizing storage time at the distributor and on-farm

• Storing in a cool, dry place

• Ask questions of your supplier

Alternative approach

• Yeast culture products include yeast growth media

• Do not rely on viable yeast cells for putative benefitsputative benefits

• All yeast products: verifying consistency is difficult when the mode of action is unclear

Thank you!

7

n Take Home Messages• Feed intake is affected by the interaction of diet

characteristics, physiological state of animals, andenvironmental stressors.

• The dominant signals controlling feed intakelikely change throughout lactation. Control offeed intake is likely dominated by hepaticoxidation of non-esterified fatty acids duringtransition and propionate in late lactation, whileruminal distension likely controls feed intake ofpeak lactation cows. Thus, optimizing feed intakerequires different diets through lactation (i.e.grouping cows).

• Controlling mobilization of body fat stores duringtransition and limiting diet fermentability arekeys to maximize feed intake during transition.

• Peak milk yield is maximized by feeding low-filldiets that are highly fermentable. The fillingeffect of diets is affected most by concentration,digestibility, and fragility of forage NDF.

• Diets should be formulated to limit dietfermentability to provide consistent supply offuels as milk production declines post-peak andplasma insulin concentration and insulinsensitivity of tissues increase.

n IntroductionFeed intake is determined by many interactingfactors and prediction of feed intake is the “Achillesheel” of diet formulation. Many different dietcharacteristics interact with environment andphysiological state of cows, making it difficult topredict feed intake accurately. However,understanding the factors controlling feeding allowsus to manipulate diets to optimize feed intake.Eating is controlled by the integration of peripheralsignals in brain feeding centers. Dairy cow dietsmust contain a minimal concentration of relativelylow-energy roughages for proper rumen function andsignals from ruminal distension can control feedintake when the drive to eat is high and metaboliccontrol of feed intake is diminished (e.g. cows at peaklactation). Signals derived from metabolism of fuelsdominate the control of feed intake when signalsfrom distension diminish (e.g. cows in late lactation).Therefore, effects of diet on feed intake vary with thephysiological state of the animal. Furthermore, theyinteract with environmental stressors such as social(e.g. overcrowding) and thermal stress. The objective

of this article is to discuss factors controlling feedintake in lactating cows and how they can bemanipulated to optimize feed intake. “Optimal” feedintake might mean the maximum to attain highermilk yields for high-producing cows or less thanmaximum to increase efficiency of feed conversionfor lower producing cows.

n Hepatic Oxidation Theory (HOT)There is a large body of evidence (mostly in non-ruminant species) that food intake is controlled byoxidation of fuels in the liver. This has beenreviewed previously (Allen et al., 2005; Allen andBradford, 2006) and will be only briefly discussedhere. The liver is “hardwired” to feeding centers inthe brain via the hepatic vagus nerve. Feedingbehavior is controlled by the firing rate of the nerve,which is determined by oxidation of fuels in the liver;increased firing rate is associated with hunger, anddecreased firing rate is associated with satiety.Feeding behavior has been linked to ATP (a form ofenergy currency within cells) concentration in theliver with satiety occurring as fuels are oxidized andATP is produced, and hunger occurring as oxidationdecreases and ATP is depleted. The mechanism bywhich ATP concentration affects the firing rate of thehepatic vagus nerve has not yet been determined.Fuels oxidized in the liver vary across species but forruminants they include fatty acids (from the diet ormobilized from body reserves), propionate (producedby microbial fermentation in the gut), lactate(produced by muscle and gut tissues from glucose),and amino acids (from protein degradation). It isimportant to realize that the pattern of oxidation offuels (minute to minute) is what affects feedingbehavior because the amount of oxidation overlonger periods of time (hours or days) is relativelyconstant.

n Physiological Changes ThroughLactation

Because fatty acids are readily oxidized in the liver,the supply of non-esterified fatty acids (NEFA) frommobilization of body fat reserves likely suppressesfeed intake in the transition period. The degree of fatmobilization is affected by changes in plasma insulinconcentration and sensitivity of tissues to insulin.Plasma insulin concentration signals tissues tosynthesize fat if elevated, or mobilize fat if lowered.

8

HOT Feeding Strategies toMaximize Milk Yield

M. S. Allen and B. J. BradfordDepartment of Animal Science, Michigan State University,

East Lansing, MI 48824 Email: [email protected]

Changes in sensitivity of tissues to insulin throughthe lactation cycle modify this signal; decreasedsensitivity (increased resistance) results in greater fatmobilization and increased sensitivity results ingreater fat deposition at the same insulinconcentration. Plasma insulin concentrationdecreases 50% or more by calving, beginning severalweeks prepartum. Plasma NEFA concentrationincreases because fat is mobilized in response todecreased plasma insulin concentration. In addition,tissue sensitivity to insulin decreases in latepregnancy contributing to increased fat mobilization.Decreased plasma insulin concentration andsensitivity help the cow maintain constant plasmaglucose concentration despite declining feed intake inthe last week or so before calving. This is becauseutilization of glucose by tissues decreases, andutilization of NEFA by muscle increases, sparingglucose.

Plasma glucose concentration drops precipitously atcalving and partially recovers over the course of thenext several weeks. Plasma insulin concentration andsensitivity of tissues to insulin remain low in earlylactation so plasma NEFA concentration remainselevated for several weeks or more. The length oftime that NEFA remains elevated varies greatlyamong cows and depends upon the rate ofmobilization and removal from the blood by the liverand mammary gland. Transfer of NEFA to milk fatby the mammary gland is highly desirable becausestorage of NEFA as triglycerides in the liver results infatty liver, compromising glucose production, andoxidation of NEFA in the liver likely decreases feedintake according to HOT. This, in turn, delays theincrease in plasma glucose concentration followingcalving, extending intake suppression. This isbecause glucose stimulates insulin secretion by thepancreas, and plasma insulin concentration remainslow, extending the period of fat mobilization, andtherefore extending the period that feed intake issuppressed by oxidation in the liver. In addition, lowplasma glucose likely limits milk yield becauseglucose is required by the mammary gland for theproduction of milk lactose, the primary determinantof milk volume.

Hepatic oxidation of NEFA is a two-stage process;long carbon chains of fatty acids are partiallyoxidized to acetyl CoA, a two-carbon molecule,which is either completely oxidized or exported asketones. The ability of the liver to completely oxidizeNEFA is limited, so ketones are exported and theirconcentration in plasma is elevated when fatmobilization is high. Ketones can be beneficialbecause they can be used by some tissues for energy,sparing glucose, but can cause keto-acidosis ifconcentrations are very high.

Further increases in lipolysis following parturition,combined with higher starch diets, likely suppressfeed intake because rapid production and absorptionof propionate stimulates oxidation of acetyl CoA (seebelow). Because feed intake of fresh cows is likelycontrolled primarily by hepatic oxidation, diets withmoderately high forage fiber concentrations mightbenefit cows. Forage fiber increases rumen fill,decreasing the risk of abomasal displacement, andincreases acetate production, sparing glucoseutilization by extrahepatic tissues. While research isneeded to evaluate effects of concentration andfermentability of starch on feed intake response,starch sources with moderate ruminal fermentabilityand high digestibility in the small intestine such asdry ground corn will likely provide more glucoseprecursors by increasing feed intake.

Milk yield increases rapidly following parturition and,over the next several weeks, increasing plasma glucosestimulates insulin secretion, thereby decreasinglipolysis and plasma NEFA concentration. Becausefewer NEFA are available for oxidation, the acetyl CoAconcentration in the liver decreases, decreasing ketoneoutput by the liver. Lack of acetyl CoA and highglucose demand limit ATP accumulation in the liver,and satiety signals to the brain decrease. As milk yieldincreases further and feed intake control by hepaticoxidation diminishes, control is dominated bydistension from gut fill and cows should be offered adiet that is less filling and more fermentable. Thischange in the dominant mechanism of intakeregulation might occur only 7 to 10 days after calvingfor some cows in the herd or more than 3 weeks forothers; the best signs that hepatic oxidation is lesslimiting are lower plasma NEFA and ketoneconcentrations and steadily increasing feed intake.

As energy requirements decrease following peak milkyield, control of feed intake by gut distensiongradually diminishes and control by hepaticoxidation increases. Plasma insulin concentrationand sensitivity of tissues to insulin increase aslactation progresses and affect the feed intakeresponse to highly fermentable diets. Higher plasmainsulin concentrations that are indicative of adequatenutritional status likely provide negative feedback onhepatic gluconeogenesis. This relationship isconsistent with HOT because decreased use ofpropionate for glucose production leads to greaterpropionate oxidation and decreased feed intake.Individual cows with an adequate supply ofglucogenic precursors may respond to a furtherincrease in supply by decreasing DMI. Greatersensitivity of tissues to insulin likely increasesclearance of fuels from the blood sooner, partitioningmore energy to body reserves and decreasing theinterval between meals.

9

n Optimizing Fat MobilizationPlasma NEFA are used as an energy source bymaternal and fetal tissues, thereby sparing glucose,and also enrich the fat content of milk. However,plasma NEFA concentrations should be limitedbecause elevated NEFA can depress feed intake andsuppress immune function. To limit plasma NEFAconcentrations, rate of fat mobilization must becontrolled. Rate of fat mobilization is dependentupon the amount of fat reserves available formobilization as well as insulin concentration, tissuesensitivity to insulin, and stress. The importance ofcontrolling body condition at calving is wellrecognized. Cows with excessive body conditiongenerally mobilize fat very rapidly through transitionbecause their tissues are more insulin resistant andthey have greater fat stores to mobilize. Therefore itis very important to manage body condition to limitover-conditioned cows by reproductive management,grouping lactating cows, diet formulation, etc.Recent research indicates that allowing cows toconsume more energy than required during the dryperiod results in increased NEFA concentrations inearly lactation (Holtenius et al., 2003). Controllingenergy intake by feeding high-fill diets during thisrelatively short period might reduce depots of readilymobilized fat reducing the rate of fat mobilizationafter calving. Fat mobilization will be reduced byincreasing sensitivity of fat tissues to insulin(decreasing insulin resistance). Niacin decreases fatmobilization but likely needs to be supplemented athigher concentrations than currently recommendedunless provided in a protected form. Chromiumincreases insulin sensitivity and supplementalchromium has been demonstrated to decrease plasmaNEFA concentrations in lactating cows. Whilechromium supplements are restricted for use inlactating dairy cattle diets by the CFIA, the chromiumconcentration of feeds varies. A more rapid increasein plasma glucose following calving will likelyincrease insulin and decrease NEFA concentrationssooner. However, increasing insulin sensitivity of fattissue is preferable to increasing insulin concentrationbecause insulin can reduce glucose production by theliver. Hormones released during stress increase fatmobilization, elevating plasma NEFA concentrationfurther. Therefore, great attention should be paid toreducing all potential stressors of cows includingstressful interactions with farm workers,management procedures, and facilities (e.g. bedding,ventilation, bunk space).

n Propionate Control Of Feed IntakePropionate, produced by microbial fermentation inthe gut, is a primary fuel controlling feed intake inruminants fed diets containing high grainconcentrations. It is a primary endproduct of starchfermentation, and production rates vary greatly

among diets. Propionate can be produced andabsorbed at very high rates and very rapidly takenup by the liver, where it is a major fuel used toproduce glucose. However, when propionate isabsorbed faster than it can be utilized to produceglucose in the liver, it will likely be oxidized,generating ATP and a satiety signal to the brain. Thecapacity of the liver to produce glucose is affected byglucose demand (the difference between glucoserequired and glucose produced) because limitingenzymes in the liver are up-regulated to meetdemand. Because of this, propionate is less likely tobe oxidized (and decrease feed intake) at peaklactation when glucose demand is high, than in latelactation when glucose demand is lower. Althoughpropionate might be expected to have little effect onfeed intake of fresh cows because they have highglucose demand, decreasing oxidation of propionateper se, propionate also stimulates oxidation of acetylCoA. Fresh cows have a large supply of acetyl CoAin the liver from partial oxidation of NEFA. Someacetyl CoA is exported as ketones, but it is alsoreadily oxidized when propionate is taken up by theliver, quickly generating ATP and a satiety signal (seeAllen and Bradford, 2006 for more details). This is anapparent conundrum: propionate is a primary fuelused to produce glucose, which is needed to increaseinsulin and decrease NEFA, thereby alleviating thedepression in feed intake by NEFA oxidation in freshcows, but propionate suppresses feed intake bystimulating oxidation of acetyl CoA in fresh cows.However, there are diet formulation options that helpprevent the depression in feed intake, includingmanipulating the rate of propionate production toextend meal length, supplying other glucoseprecursors that stimulate oxidation of acetyl CoA to alesser extent, and providing alternate energy sourcesfor tissues to spare glucose. The goal is to maximizethe amount of glucose produced or spared per unit ofATP generated in the liver over time. Manipulatingthe pattern of oxidation of fuels in the liver canincrease plasma glucose and insulin concentrations,decreasing fat mobilization and the period of timefeed intake is suppressed by oxidation of NEFA in theliver.

n Altering Propionate Flux To The LiverRate of propionate production can be decreased byreducing starch concentration and fermentability andby increasing efficiency of microbial proteinproduction from organic matter, while absorptionrate is likely to be reduced by inhibiting ruminalmotility.

Dietary Starch Concentration Starch concentration of diets is often reduced bysubstituting forage or non-forage fiber sources(NFFS) such as beet pulp or soyhulls for cereal

10

grains. Dilution of starch in the diet has the addedbenefit of reducing the fermentation rate of the starchremaining when starch concentration is decreased byadding forage or NFFS, reducing the rate ofpropionate production. The optimal strategydepends upon the relative cost of ingredients,efficiency of feed utilization, and animal productionresponse. For instance, longer fiber particles fromforage compared to NFFS might increase fiberdigestibility by increasing ruminal pH throughstimulation of rumination and by increasing ruminalretention of fiber; however, forage fiber is very fillingand forages might limit feed intake compared toNFFS. Therefore, when ruminal distensioncontributes to the control of feed intake, substitutionof NFFS for grain might be a better choice thansubstitution of forage.

Site of Starch DigestionSubstitution of a less fermentable starch source is anoption when feed intake is depressed by a rapidlyfermented starch source. Altering dietary starchfermentability will likely be more desirable thanreplacing starch with fiber for ruminants with highglucose demand, such as early lactation cows,because postruminal starch digestion yields moreglucose precursors than ruminal fermentation offiber. It is important to note that the fraction ofglucose precursors provided by starch fermentationin the large intestine is much lower than in the rumenor small intestine because microbial cells containingstarch are lost in the feces. Therefore, carefulconsideration of site of starch digestion is veryimportant to maximize the yield of glucoseprecursors over time. Starch sources with lowerruminal digestibility should be highly digestible inthe small intestine to provide the greatest yield ofglucose precursors. For instance, dry ground andcracked corn both slow the rate of propionateproduction in the rumen compared to high moisturecorn, but the ground corn will provide more glucoseprecursors because of greater digestibility in thesmall intestine and total tract.

Rate of Propionate Absorption Ruminal motility affects the rate of propionateabsorption because mixing of ruminal contents isrequired to replenish its supply at the ruminalepithelium where it is absorbed. Therefore, ruminalmotility likely affects the rate at which propionatestimulates oxidation within meals. Ruminal motilityis affected by diet and is likely increased byphysically effective fiber and decreased by long-chainfatty acids and butyrate. Butyrate productionincreases when feed ingredients containing sugars areconsumed. Therefore, other diet components canalter feed intake by affecting flux of propionate to theliver.

Nitrogen Metabolism Consumption of ruminally degraded protein or totalprotein in excess of that required can decrease feedintake. Hepatic oxidation of ketogenic amino acidscan contribute to satiety according to HOT and ureaproduction from excess ammonia produces a carbonskeleton that can be oxidized. However, greaterdietary protein concentration can also increase feedintake by reducing propionate production.Increasing protein concentration could dilute dietstarch concentration and decrease energy spilling byruminal microbes, thus converting a greater fractionof fermented organic matter into microbial cells andless into VFA.

n Gut FillAs feed intake increases in early lactation, control offeed intake is dominated by ruminal distention andthe extent to which ruminal distention limits feedintake is linearly related to milk yield (Voelker Lintonand Allen, 2007). High producing dairy cows shouldbe fed diets with lower filling effect to maximize feedintake. The filling effect of a diet is determinedprimarily by the initial bulk density of feeds as wellas their filling effect over time in the rumen. Theoverall filling effect is determined by forage NDFcontent, forage particle size, fragility of forage NDFdetermined by forage type (legumes, perennialgrasses, annual grasses), and NDF digestibility withina forage family (Allen, 2000). Forage NDF is lessdense initially, digests more slowly, and is retained inthe rumen longer than other diet components. Feedintake of high producing cows is often dramaticallyreduced by increasing the forage NDF concentrationof the diet. Several studies in the literature reporteda decrease in DMI of up to 4 kg/d when dietNDF content was increased from 25 to 35% bysubstituting forages for concentrates. Althoughmost studies reported a significant decrease inDMI as forage NDF increased, the DMI responsewas variable, depending upon the degree towhich intake was limited by ruminal fill. Higherproducing cows are limited by fill to the greatestextent and the filling effect of forage fiber variesdepending upon particle size and fermentationcharacteristics.

Experiments that have evaluated effects of forageparticle size have generally shown small effects onDMI (Allen, 2000). However, one experimentshowed little effect of particle size of alfalfa silagewhen fed in high grain diets but a large reduction inDMI for the diet containing longer alfalfa silage whenfed in a high forage diet (Beauchemin et al., 1994).Feed intake might have only been limited by ruminalfill in the high forage diet, which could explain theinteraction observed.

11

Increasing diet NDF content by substituting non-forage fiber sources for concentrate feeds has shownlittle effect on DMI in studies reported in theliterature (Allen, 2000). Non-forage fiber sourcesinclude byproduct feeds with significantconcentrations of NDF such as soyhulls, beet pulp,cottonseeds, corn gluten feed, and distiller’s grains.Fiber in non-forage fiber sources is probably muchless filling than forage NDF because it is less fillingboth initially (smaller particle size) and over time inthe rumen because it digests and passes from therumen more quickly.

Forage NDF has a much longer ruminal retentiontime than other major dietary components. Retentiontime in the rumen is longer because of longer initialparticle size and greater buoyancy in the rumen overtime, which differs greatly across forages. As foragesmature, the NDF fraction generally becomes morelignified. Lignin is a component of plant cell wallsthat helps stiffen the plant and prevent lodging. It isalso essentially indigestible by ruminal microbes andlimits fermentation of cellulose and hemicellulose.Within a forage type, the degree to which NDF islignified is related to the filling effects of the NDF.Fiber that is less lignified clears from the rumenfaster, allowing more space for the next meal.However, ruminal retention time of NDF fromperennial grasses is generally longer than for legumeNDF in spite of being less lignified. Because of this, itis more filling and should not be included in highconcentrations in diets of cows for which feed intakeis limited by ruminal fill, unless it is of exceptionallyhigh quality. Corn is an annual grass, and cornsilage NDF digests and passes from the rumenquickly and can be an excellent source of forage NDFfor high producing cows.

While ruminal distention becomes a primarylimitation to feed intake as milk yield increases, itlikely has little effect on feed intake during thetransition period if feed intake is controlled primarilyby oxidation in the liver. Diets can be formulated tomeet requirements for energy and nutrients withlarge differences in the amount and turnover rate ofruminal digesta. Formulating diets to maintain gutfill with ingredients that are retained in the rumenlonger, and have moderate rates of fermentation andhigh ruminal digestibility will likely benefit transitioncows several ways. The ruminal digesta will providemore energy over time when feed intake decreases atcalving or from metabolic disorders or infectiousdisease. This will help maintain plasma glucose andprevent even more rapid mobilization of bodyreserves compared to when diets are formulated withingredients that disappear from the rumen quickly.Ruminal digesta is very important to bufferfermentation acids and buffering capacity is directly

related to the amount of digesta in the rumen.Therefore, diets formulated with ingredients thatincrease the amount of digesta in the rumen will havegreater buffering capacity and will maintain buffercapacity longer if feed intake decreases. Inadequatebuffering can result in low ruminal pH, decreasingfiber digestibility and acetate production, andincreasing propionate production, possiblystimulating oxidation in the liver and decreasing feedintake. Low ruminal pH also increases risk of healthproblems such as ruminal ulcers, liver abscess, andlaminitis, and causes stress, likely increasingmobilization of body reserves even further. Dietsformulated with ingredients that maintain digesta inthe rumen longer when feed intake decreases willlikely decrease risk of abomasal displacement.

n Unsaturated Fatty AcidsFeed and energy intake can be depressed bysupplementation of fat and the extent of depression isdependent upon fat type (Allen, 2000). Fat sourceswith more unsaturated fatty acids reduce intake tothe greatest extent and fatty acids that are highlysaturated have less effect.

n RecommendationsLimit mobilization of body fat by controlling bodycondition during mid to late lactation and limitingfeed intake of dry cows by feeding diets with highforage NDF concentration. A low concentration ofhighly fermentable starch might reduce NEFAconcentration prepartum.

Maintain rumen fill through transition. Diets withhigh concentrations of grain, non-forage fiber, andfinely chopped forages fed through the transitionperiod should be avoided. Feeding high-fill dietsprior to calving to control feed intake might reducedepots of readily mobilized fat and provide energy tohelp sustain plasma glucose through calving.Increased amounts of ruminal digesta also decreaserisk of displaced abomasum and increase bufferingcapacity, decreasing risk of acidosis. Forage fiber ismuch more filling than non-forage fiber or other dietcomponents but the filling effect of forage fiber variesgreatly. Some long fiber particles are necessary toform a mat and increase digesta retention in therumen, but excessive length of cut can increasesorting and can decrease feed intake. Digestioncharacteristics of forage fiber vary greatly by foragetype and maturity and have a large effect onretention time in the rumen. Wheat straw digestsand likely passes from the rumen slowly and it hasbeen used to dilute energy density of corn silage inTMRs for dry cows. Grass silage or hay is likelymore beneficial because the fiber is more digestibleand it provides energy for a longer time when feedintake decreases at calving. However, grass with

12

high potassium concentrations might require anionicsalts in prepartum diets to reduce milk feverfollowing calving.

Supplemental fat should not be fed through thetransition period because it can depress feed intakeby stimulating gut peptide release and increasing thesupply of fatty acids to be oxidized. An exceptionmight be the use of supplemental CLA to suppress fatproduction in the mammary gland, benefiting freshcows by sparing glucose.

Avoid feeding highly fermentable starch sources tofresh cows because rapid production and absorptionof propionate will stimulate oxidation of acetyl CoAand suppress feed intake. Starch sources withmoderate ruminal fermentability and highdigestibility in the small intestine, such as dry groundcorn, will provide glucose precursors and lesspropionate to stimulate oxidation and suppress feedintake.

Feed a less filling and more fermentable diet as gutfill begins to dominate the control of feed intake.This might be only 7 to 10 days after calving for somecows in the herd or more than 3 weeks for others andis likely indicated by lower plasma NEFA and ketoneconcentrations, visual observation of cow gutdistension, and steadily increasing feed intake. Whilegroup housing prevents measurement of feed intakefor individual cows, kits are available to measureNEFA and ketones concentrations on the farm.Because feed intake is limited by ruminal fill, feedingredients that can depress ruminal motility such asfat and sugar sources should be limited.

Feed a more filling, less fermentable diet as milkyield declines. As lactation progresses past mid-lactation, the highly fermentable diet that is optimalfor high-producing cows can depress feed intake asmilk yield and glucose demand decreases.Propionate is likely oxidized when it is producedfaster than it can be utilized, generating ATP and asatiety signal. Therefore, cows should be switched toa less fermentable and more filling diet as milk yielddeclines. This will increase feed intake and provide amore consistent supply of fuels, reducing insulin andpartitioning more energy to milk rather than bodycondition. Furthermore, the less fermentable, morefilling diet will decrease risk of milk fat depressionand late lactation abomasal displacement.Unsaturated fats likely decrease feed intake andshould be limited. Limit highly fermentable starchsources (e.g. high moisture corn, ground barley) bysubstituting less fermentable feeds such as dryground corn or non-forage fiber sources.

n Conclusions Consideration of physiological changes occurringthrough lactation and the physical and digestioncharacteristics of feeds beyond their nutrientcomposition is required to optimize feed intake forlactating cows. Understanding the control of feedintake is critical to diet formulation and the HepaticOxidation Theory is exciting for its potentialcontribution to our ability to formulate diets. Whilemore research is needed to better understand animalresponse to diets, the theory and concepts presentedin this paper will help to formulate diets to improveanimal health and farm profitability.

ReferencesAllen, M.S. 2000. Effects of diet on short-term regulation of

feed intake by lactating dairy cattle. J. Dairy Sci. 83:83:1598-1624.

Allen, M.S., B.J. Bradford, and K.J. Harvatine. 2005. Thecow as a model to study food intake regulation. Ann.Rev. Nutr. 25:523-547.

Allen, M.S. and B. J. Bradford. 2006. From the liver to thebrain: increasing feed intake in transition cows. Pp.115-124. Proc. 68th Meeting of the Cornell NutritionConference for Feed Manufacturers, Department ofAnimal Science, Cornell University, Ithaca, NY 14850

Beauchemin, K.A., B.I. Farr, L.M. Rode, G.B. Schaalje. 1994.Effects of alfalfa silage chop length and supplementarylong hay on chewing and milk production of dairycows. J. Dairy Sci. 77:1326-1339.

Holtenius, K., S. Agenas, C. Delavaud, and Y. Chilliard.2003. Effects of feeding intensity during the dry period.2. Metabolic and Hormonal Responses. J. Dairy Sci.86:883–891.

Voelker Linton, J.A. and M.S. Allen. 2007. Nutrient demandaffects ruminal digestion responses to a change indietary forage concentration. J. Dairy Sci. 90

13

14

Dealing with Issues of Risk: Is There More toIt Than the Average?

John FetrowUniversity of Minnesota

Dealing with risk in decision making on dairies

John Fetrow VMD, MBAProfessor of Dairy Production Medicine

College of Veterinary MedicineUniversity of Minnesota

Making decisions• Decisions (choices) on a dairy are complex;

they are almost never made for only one reason– profit– cash flow– available loans– management effort– labor / implementation effort– physical conditions / constraints– emotion, tradition, habit, hunch…

Making decisions

• How should risk be factored intodecision making?

• What is “risk”?• Can risk be quantified?• Can risk be managed?• Can risk be avoided?

What is risk?• Two kinds of risk (the two blur a bit)

– Risk of unpredictable calamity• Foot and mouth disease• Melamine in milk• Tornado

– For these there is insurance

– Predictable “risk” due to variation• Milk production• Milk price• Cow culling rates

– These need to be factored into management decision making

• If the pattern of outcome is known, then the risks and economics of decision options can be calculated.

Making decisions• For routine management decision making,

there are two possible errors:– Type I error: making a change when it does

not work (doesn’t pay)

– Type II error: NOT making a change when it would have worked (would have paid)

• Scientists worry about Type I; managers have to worry about both types.

Dealing with Uncertainty

• Lots of decisions have an “unknown”outcome– LDA surgery– Adding fans to a barn– More bedding– More feed in front of the cows– AI breeding with a better bull

15

Making a decision in the face of risk:LDA surgery in a heifer

• Left Displaced Abomasum surgery on a recently calved heifer– expense to do the surgery is known: $150

• three possible outcomes1. full recovery

• 85% of cases • the heifer is worth $1,800

2. poor outcome, the heifer is sold to slaughter• 10% of cases• Cull income of $400, buy a new heifer for $1,800

3. the heifer dies or is unfit for slaughter• 5% of cases• No income, buy a new heifer for $1,800

Heifer LDA example

• Full recovery: 85% of cases– $150 surgery cost

• Cull to slaughter: 10% of cases– $150 surgery + $1,800 buy new heifer -

$400 cull income = $1,550 total cost• Dead or unfit for slaughter: 5% of cases

– $150 surgery + $1,800 buy new heifer = $1,950 total cost

partial budget outcome (cost) versus probability

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

$150 $1,550 $1,950

value of outcome

prob

abili

tyof

outc

ome

In any particular case, one of the three outcomes will occur.

Full recovery DeadSlaughter

Heifer LDA example• If there are many such cases, what is the long term

(average) “expected value” of the decision to do surgery?– Full recovery: 0.85 * $150 = $128– Cull to slaughter: 0.10 * $1,550 = $155– Dead or unfit for slaughter: 0.05 * $1,950 = $98– -----------------------------------------------------------– Total expects cost = $128 + $155 + $98 = $380

• In the long run, doing surgery costs $380 per case.• Culling costs $1,800 to buy a new heifer - $400 cull income = $1,400

• Surgery costs less than culling on average – Surgery has a better expected value

• Remember: 15% of the time, surgery costs more thanculling would have cost.

Uncertainty • By using the probability of an outcome to

“weight” the value of an outcome and summingthe weighted value across all possible outcomes, one can calculate the “expected value” of a decision.– The probabilities of all outcomes must sum to 1.0

• Something must happen and two things cannot both happen.

• This approach assumes you will get to make a similar decision many times or that you makemany decisions on this basis so that you can “win” over the long run

Cow versus Herd• The LDA example operates at a decision made

at the level of a single cow.• In many cases, the herd is the unit of

interest for measurement of response and economic value

• What about decisions made that affect the herd?– Change in feed program– Change in facilities– Change in stocking density

• Dairy managers don’t realistically expect to make these kinds of decisions repeatedly

16

Cow versus Herd• The problem for decision making at the herd

level is that herds are much more variable than cows and so responses to herd interventions / changes are harder to predict.– If you increased access to water on 100 dairies,

would you expect a consistent response in milkproduction across all 100?

– What if you added more bedding?– What if you moved from 2x to 3x milking?– What if you fed rumen protected amino acids?

“During the mid 1990s, a better understanding of cow comfortmade us change our management so no barns were overcrowded andfeed space was not limited. Additionally, more water troughs, morefrequent bedding and scraping, and careful timing of feeding wereput into place. Production per cow improved to our current level of86 pounds.”

“Over two summers in Alabama, milk production was greater whensprinklers were used over feed alleys.”

“The new cross ventilation systems were able to maintain indoor airbelow ambient outdoor temperatures in the two dairy barns and provided convective cooling of the cows during warm summer days.Milk yield and feed efficiency improved in these two barns duringthe 4 summer (June – September) months of 2007 compared tothe previous three summers (2004-2006) even though 2007 had greater number of days above 30 °C.”

•The most common approach that dairymen use to makeherd level decisions is to look at other dairies that made the same decision and ask “did it work for them”?

•This is fundamentally an approach that tries to treat a “dairy” the same way we treated a “cow with an LDA”.

•If, on average, dairies that make the change did well, then we guess that we can also “expect” to do well.

•What is the risk that the change won’t work on yourdairy?

•What if you are the LDA cow that gets culled?•How likely and how costly would this be?

Decisions at a herd levelDairy Decision Making: the value of healthy skepticism

Most proposed changes are meant to address a problem:

something is not theway it should be

Most problems on a dairy should be ignored.

(at least for the moment)

Water through a plumbing system: problems

IN OUT

Problem: any dent in the pipe.Pipes aren’t supposed to be dented.

problems

17

Bottlenecks: a special class of problems

• rate limiting problem• interferes with achieving the dairyman’s goal

• any improvement pays off in better output

• before it is completely “fixed”, something else will become thebottleneck

Water through a plumbing system: bottleneck

IN OUT

Bottleneck: the rate limiting dent in the pipe.The bottleneck controls the rate of flow.

-Any reduction in the bottleneck (improvement) will increase theflow of water through the pipe.- The bottleneck does not have to be fully “fixed” to have a positive impact; any improvement helps.- Before the bottleneck is completely “fixed”, somethingelse will become the bottleneck.

bottleneck

Making changes on a dairy:the role of bottlenecks

• Changes only help if they open a bottleneck• We often turn to research studies to look for

possible positive ideas• Skepticism sets in when we experience the

results of applying “proven” findings on real dairies– suppose that three studies of different

interventions each show a 4 pound increase inproduction

– Great! I’ll use all three and get 12 pounds!• Bottlenecks don’t work that way

Clinical trials and bottlenecks

IN OUT

Studies of “vitamin Q” in early lactation.

Feeding vitamin Q is shown to increase milk production by 4 pounds.Good quality research, good sample size, well controlled,

published in a reputable journal, peer reviewed.Improved production is shown to be profitable.

bottleneck caused by relativedeficiency in vitamin Q

Feeding vitamin Q removes a bottleneck.

Clinical trials and bottlenecks

IN OUT

Vitamin Q in practical application in a specific herd.

- Feeding vitamin Q in this herd will not produce 4 pounds of milk.- Feeding vitamin Q is effective only when a relative deficiency invitamin Q is the herd’s current bottleneck.- Note that the herd can actually have a Q deficiency and if limitedaccess to water is the bottleneck, vitamin Q will still not work!

bottleneck caused by limitingaccess to water in fresh pen

Vitamin Q not deficientWhen they fix the waterers in thefresh pen, this will become the bottleneck.

At present, there is no vitamin Q bottleneck to remove!

Improving the dairy:opening bottlenecks

• know the dairy: identify likely bottlenecks– excellent records, careful observations, measurements,

others’ input• seek ideas that might open the bottleneck

– personal experience, research results, expert advice,successful dairies, reading material

• Ask if there is research that shows a consistent positive response for dairies with the bottleneck

• make prudent investments in a new way of doing things to open the bottleneck and measure/monitor the results– these are on-farm experiments– dairies that have stable performance can do more

experiments and improve more quickly

18

Considering research for interventions applied at the herd level

• There are products that have been extensively researched over time in many different herds– Research approach may have varied, but the general

responses (milk, DMI, gain, etc.) has been measured.– Study herds commonly were chosen either because

the herd had the bottleneck of interest or because access to the herd(s) was convenient.

• This kind of research background allows forconsideration of the variability of effect across herds. META-ANALYSIS– This makes it possible to deal with herds like we did

for cows with LDA

Objective of a meta-analysis• Assessment of and strength of evidence of

response across many studies– To determine whether an effect exists and in what

direction– Statistical approach to pool results from many studies

to estimate the distribution of response across herds• Similar to measuring response in many cows and pooling the

results for an average response PER COW• Here we are trying to get response PER HERD

• Investigation of heterogeneity– To examine if responses differ in different situations

• Meta-analysis depends on having the product / intervention studied in many herds / many studies

Sodium bicarbonate: buffer rumen acidosis

Sodium bicarbonate: buffer rumen acidosis

Sodium bicarbonate

This and similar slides are derived from economic evaluations by Dr. David Galligan at the University of Pennsylvania’s School of Veterinary Medicine. Models available at http://www.productiontools.org/type12.php

Sodium bicarbonate

19

Economic Evaluations: Per Herd

General Per cow/d Per month Per yearIncreased milk lbs 3.28 100 1,195Value of extra milk $0.43 $13 $155Cost extra feed $0.14 $4 $52Investment in product $0.07 $2 $24Net income from product: $0.22 $7 $79Return on TOTAL Investment 104% 104% 104%

Type 1 and 2 Error AnalysisBreakEven Response 0.30 lbs Per HerdCost of Decision Errors Response Freq. $/cow/d Per/month Per/yearType 1: Using product < 0.30 13% ($0.01) ($0) ($5)Type 2: Failing to use product > 0.30 87% $0.26 $8 $95

Sodium bicarbonate What about products that are not meant to open a bottleneck?

• There is probably a spectrum of products / interventions that range from those that obviously “fix” abottleneck to those that act to “enhance”performance in a system that is not “broken”.

• Should enhancement products be more widely adopted by the dairy industry?

• Fix the broken waterer• Push up feed• Bed cows better• Improve air quality• Dip teats• Add bicarbonate to

ration• Use chelated minerals• Feed yeast product• Use BST

rBST use in dairy herds rBST use in dairy herds

rBST use in dairy herds rBST use in dairy herds

$13.00 milk; 75 pounds; 5.67 feed cost / cwt

20

rBST use in dairy herds rBST use in dairy herdsEconomic Evaluations: Per Herd

General Per cow/d Per month Per yearIncreased milk lbs 8.60 261 3,137Value of extra milk $1.12 $34 $408Cost extra feed $0.12 $4 $45Investment in product $0.38 $11 $137Net income from product: $0.62 $19 $226Return on TOTAL Investment 124% 124% 124%

Type 1 and 2 Error AnalysisBreakEven Response 3.25 lbs Per HerdCost of Decision Errors Response Freq. $/cow/d Per/month Per/yea rType 1: Using product < 3.25 13% ($0.03) ($1) ($12)Type 2: Failing to use product > 3.25 87% $0.63 $19 $230

Zinpro Availa-4 Zinpro Availa-4 in milking rations

Response in herds to Diamond V Yeast Product

Ian Lean, Ahmad Rabiee and Mark Stevenson: Massey University

45 out of 50 studies showed a positive response

Diamond V Yeast Culture

Type I and II error analysis and expected valuesDIAMOND V YEAST CULTURE

2.10 mean milk response (lbs of milk)1.72 s.d. of milk production response

0.050$ product cost per cow per day0.130$ milk price per pound0.035$ cost of marginal feed per pound of milk

21

Diamond V Yeast CultureProfit perCow per Day by Milk Production Response

$(0.300)$(0.200)$(0.100)

$-$0.100$0.200$0.300$0.400$0.500$0.600

(1.3

4)

(0.8

2)

(0.3

1)

0.21

0.72

1.24

1.76

2.27

2.79

3.30

3.82

4.34

4.85

5.37

pounds milk

pro

fit profit

Diamond V Yeast CultureDistribution of Milk Production Responses

0.0%10.0%20.0%30.0%40.0%50.0%60.0%70.0%80.0%90.0%

100.0%

(1.1

7)

(0.6

5)

(0.1

4)

0.38

0.90

1.41

1.93

2.44

2.96

3.48

3.99

4.51

5.02

5.54

0.0%0.5%1.0%1.5%2.0%2.5%3.0%3.5%4.0%4.5%

milk pounds

cummulative proportion

incremental proportion

Diamond V Yeast CultureExpected Value by Milk Production Response

(0.002)

-

0.002

0.004

0.006

0.008

0.010

(1.1

7

(0.6

5

(0.1

4

0.3

8

0.9

0

1.4

1

1.9

3

2.4

4

2.9

6

3.4

8

3.9

9

4.5

1

5.0

2

5.5

4

milk pounds

expected value

Diamond V Yeast Culture• 84% probability that the outcome in the herd

will be at least breakeven• 16 cents per cow per day profit

• 16% chance that the result will not at least break even

• 1 cent per cow per day loss if used and not at least breakeven

• Average profit per cow is 15 cents per day

Summary• For a growing number of products in dairy nutrition,

there are sufficient separate field studies of the effects of the product to make it possible to perform a meta-analysis and to describe the average herd response to use and variation between herds in their responses to the product

• Those biological assessments make it possible to develop economic projections about the value of using the product– Type I: the risk and cost incurred if the product is used and

does not provide an adequate response– The risk and cost of NOT using the product when it would

have been valuable• These sorts of analyses are made possible by

repeated careful research into the biology of theproduct and responses

Issues of risk when cutting costs on a dairy.

22

Costs when margins are tight

• generally, costs are considered as categories (purchased feed, labor, supplies, etc.)– most often per cow or per cwt

• in tighter times, dairymen look harder at areas where costs can be cut

• often in big categories first– feed– labor– youngstock– equipment– etc.

A slightly different look at costs

• Making money on a dairy depends on1. making sure the barn is full of cows2. making each cow in the barn as financially

productive as possible

What does “financially productive” mean?

• Think of each “slot” for a cow as the limiting constraint for cows

• It may be defined by parlor capacity,freestall beds or bunk space, or umber of stanchions, etc.

• The goal is to extract the maximum profit from each slot (occupied by a milking cow)

The owner of this business wants to optimize the financial performance of each machine(output - costs).

Why is the owner of this business any different?”

Costs on a dairy• What if the costs on a dairy were split

between:– cow costs– dairy operating costs?

• Think of cow costs as those things that directly impact the quality and performance of the cow in the slot; things the cow knows about including the quality of the cow herself.

• in the end most things impact on the cow, but not directly• Dairy operating costs can be seen as “fixed”

in relation to what the cow perceives– Mortgage, equipment, utilities, etc. (even labor!)

23

“Cow costs”

• feed• replacements• bedding• veterinary and medicine• some supplies• BST

What is the “profit” per slot (cow)?Revenue • Milk the cow and sell the milk

– Volume, components, cell count, other price parametersExpense: “Cow costs”• buy the cow

– cost of replacements, cull and dead rate, cost of loans, cull beef price

• feed the cow – milking and dry

• care for the cow– supplies, drugs, vaccine, vet bills

“IOFRCC”: income over feed, replacement, and cow care cost:

Oh goodie! a new parameter…• income over

– feed• largest expense, driver of production, health, longevity

– replacement• cost to replace culled cow; lactation performance and longevity

in the herd– cow care

• health, longevity, production, reproduction• the money left over pays the rest of the bills (dairy

operating costs) and is the source of profit• if a dairyman maximizes IOFRCC and is reasonable

about decisions on the rest of the dairy, the dairyman is doing the best he can

One trick for tight margins• guard IOFRCC carefully

– be very careful of cost cutting here– if cost cutting reduces production or adds other

cow costs (replacements?) then you’ve probablyshot yourself in the foot

• dairy operating costs are more flexible– Postpone, reduce, do without

• simple question:• If I cut this cost, will the cow know about it?

– If the answer is yes, be careful!

Let’s talk milk and feed!

24

Where do profits come from?

Profit = revenue – expenses

(Milk price – cost per cwt) * cwts soldmargin * volume

The financial challenge

• increase the profit margin per cwt

and

• increase the number of cwts

Increasing the margin per cwt

• The first obvious reflex is to cut expenses.– If not done carefully, this commonly reduces milk

production and leads to reduced income.– Very often, the lost revenue is much greater than

the expense saved.• An alternative:

– Increase expenses and increase production– What is the value of additional milk production

from an existing cow?

Bob’s ration: 1994• Great dairyman, great cows, great forages• 100 cow Minnesota dairy milking 2x, pre-BST

– 70 tie stalls, 30 cows outside, switch milked• Cramped barn, poor ventilation, outside cows

exposed to the elements in 3-sided shed• Stuck at 19,000 pounds of milk for at least 5

years– Facilities blamed

• Ration balanced for production to save on feed costs– TMR mixer installed 2 years previously, no

response

Bob’s dilemma:should he hire this nutritionist?

• “I can only balance a milking ration for one cow, and she’s not in your herd”– Not interested in the herd’s current DHIA

production records• “I plan to increase the cost of your ration by

5% for every mouthful”

• How should Bob evaluate the economics of any new ration change?– feed cost per cow or feed cost per cwt?

tworationeconomics.xls

Is the new ration an improvement?(feed cost is more than $100 per cow per year higher,income over feed cost is worse by 25 cents /cwt!)

25

All cost measure are worse, profit hasn’t changed. What a new feeding philosophy really changed

currentexpense

currentprofit

newexpense

newprofit

+ =

currentexpense

currentprofit

The Profit Power of Marginal Milk

currentprofit

margin

newprofit

margin

newprofit

newexpense

currentfarm

financesmarginal

milk

note: total expenses increased!

How do we evaluate cost control?

• generally use comparative benchmarks• be careful:

– production drives the system– think about denominators

• What does supplies per cwt tell you?• What does purchased feed per cow mean?

Average feed cost per cwt• Meant to monitors the efficiency of the

feeding program– Feed purchasing: prices, discounts, inventory

control– Least cost ration balancing– Feed storage, waste, shrink, discard– Ration balance to minimize excess nutrients– Careful feed delivery and bunk management

Production dry 25 50 75 100

Maintenance feed costs $2.00

feed cost for the milk

total feed cost per day

feed costs per cwt milk

milk revenue produced

income over feed costs

Income Over Feed Costs atDifferent Production Levels

One herd: 5 cows

Iofc.xls

26


production dry 25 50 75 100

maintenance feed costs $2.00 $2.00 $2.00 $2.00 $2.00


total feed cost per day








total feed cost per day $5.50






feed cost for the milk $0.88 $1.75 $2.63 $3.50

total feed cost per day $5.50








total feed cost per day $2.88 $3.75 $4.63 $5.50









feed costs per cwt milk $11.50 $7.50 $6.17 $5.50




Average feed cost per cwt has almost nothing to do with the efficiency of feeding program

management.It has everything to dowith level of production.

27






milk revenue produced $5.00 $10.00 $15.00 $20.00

incomeover feed costs




feed cost for themilk $0.88 $1.75 $2.63 $3.50




income over feed costs $2.13 $6.25 $10.38 $14.50




feed cost for themilk $0.88 $1.75 $2.63 $3.50




income over feed costs $2.13 $6.25 $10.38 $14.50


To reduce feed $/cwt: increase production

In fact, any measure of performance per cwt is mostly a reflection of production level, not cost control.

Expense benchmarks need to be “pegged”against herds of similar production levels, not

just type of operation.

The best benchmark is the comparison of thefarm’s performance to its planning budget.

Opinion

• The single best measure of the financial efficiency of a feeding program is:

Income over feed cost per cow per day

Feed efficiency• if the current ration produces 80 pounds of milk with

50 pounds of dry matter intake:– Feed efficiency = 80/50 = 1.60

• if the cow would eat another pound of dry matter (total 51 lbs) and produce 1 more pound of milk (total 81 lbs), then feed efficiency = 81/51 = 1.58 would be worse

• profit added = 17 cents income – 9 cents for feed = 8 cents IOFC; which is better

• I’d like the more inefficient ration better.– real case is likely 82/51 = 1.61 efficiency and 34 – 9 = 25

cents profit• To increase feed efficiency, increase production per

cow and dilute maintenance feed per pound of milk

28

• Profit = revenue – expenses– Rule #1: increase milk production

What to do when margins are tight?


– Rule #2: decrease expenses

What to do when margins are tight?


– Rule #2: decrease expenses•Unless it breaks Rule #1!!

What to do when margins are tight? Protect or increase marginal milk

• Do not aim to reduce feed costs as a goal• Instead, aim to increase margin of income

over feed costs per cow• Do this by increasing production per cow

– Added milk dilutes maintenance and increases average margin per cwt

• Be wary of seeking cheap substitute feeds for “optimal” ration– Be prepared to measure the impact of the new

feed on total production when it is introduced

Saving money on the feed bill• In general, select the feeds that make

the most milk and pay as little aspossible for them.

• “First the best feed, then the bestprice.”– bids– cost comparisons– discounts or volume premiums– forward contracting

Milkorfeed.xls

Risks and cost control

• At times of tight margins, be skepticalof the temptation to cut costs in any category that the cow will know about.– Cutting “cow care” costs risks losing milk

production – In many cases the reduction in expenses is

less than the reduction in revenue.• Think income over feed cost, not just

feed cost.

29

Summary: factoring risk into decision making

• We have good science and analytic tools to evaluate the likelihood of profitability of some products at the herd level.

• The average response alone is not the best criteria for making herd level decisions.– If the product fixes bottlenecks, the herd must have the

same bottleneck to see a response

• Any change in herd management needs to consider revenue and risk of failure, not just expense.

Thanks!Questions before I go?

IntroductionA meta-analysis is becoming a popular statistical toolto summarize scientific data. In contrast to a singlestudy, a meta-analysis evaluates the entire data baseof studies together in one analysis. In the pastsummarizing data was a tedious and oftenimpossible task leading to lots of suggestions thatmore research is needed. Today many researchers areturning to a meta-analysis to make this process muchsimpler, more efficient and ultimately more precise.

To conduct a meta-analysis, a researcher gathers allpublished and non-published data that can becollected, follows a clearly defined set of steps forevaluating the quality of the studies, and applies arigorous statistical process to the data. Unlikenarrative reviews where the reviewer applies a levelof importance to each study, the meta-analysisassigns weights based on the size and variance of thestudy. The meta-analysis can thus provide atransparent, unbiased, and repeatable process toevaluate data. Additional benefits of the meta-analysis procedure allows us to determine if the lackof significance due to a treatment is real or due to lowpower, whether or not treatment responses differ dueto some type of outside influencing factor (orcovariate), and whether or not there is a publicationbias (i.e., missing studies that might have reported alack of effect if they had been published).

BackgroundThere are several yeast products in the market thatare used on commercial dairy farms. There havebeen several reviews of the effectiveness of yeastproducts on dairy cow performance but often theresults are mixed. One reason for the mixed resultsmay be due to the fact that data from different yeastproduct manufacturers are often pooled together inthe same analysis. The American Feed ControlOfficials define Yeast Culture as ‘dried productcomposed of yeast and the media on which it wasgrown, dried in such a manner as to preserve thefermenting activity of the yeast”. This is differentthan the AAFCO’s classification of Active Dry Yeast(often called live yeast) which is defined as yeast with> 15 billion live yeast cells/gm. Other categories ofyeast products include Dried Yeast, Irradiated DriedYeast, Brewers Dried Yeast, Torula yeast, and yeast

Extract. Although these products all have ‘yeast’ intheir name they are not equivalent to yeast cultureand yeast culture has no reliance on live yeast for itseffectiveness.

ObjectivesThe first objective of this paper was to limit ourreview to yeast culture products manufactured byDiamond V Mills only. This meta-analysis wasconducted only on studies that evaluated Diamond VMills YC, XP and XPC. YC was the original productdeveloped in 1943, XP product was launched in 1988and is 2.5 more concentrated than YC; XPC waslaunched in 2005 and is 4 times more concentratedthan XP (Table 1).

The second objective was to limit our review tolactating dairy cow studies only. Although there isoften a numeric increase in milk production fromfeeding Diamond V Yeast Culture products in manystudies, many of these individual studies result innon significant differences. A third objective of thismeta-analysis, therefore, was to determine if the lackof significance in individual studies was real or wasdue to low statistical power. A fourth objective wasto explore whether or not treatment responsesdiffered due to some type of outside influencingfactor (i.e., heterogeneity analysis). A fifth and finalobjective was to determine if there was publicationbias (i.e., missing studies that might have reported alack of effect if they had been published).

MethodsDiamond V turned to Bovine Research Australia toconduct this meta-analysis. This group has recentlypublished a couple of meta-analysis papers onmonensin, a meta-analysis review of vitamin E andhas been invited to publish their paper of an invitedtalk on Meta-Analysis presented at the 2008 ADSAmeetings.

This meta-analysis consisted of a critical review ofrandomized controlled trials that evaluated theeffectiveness of Diamond V’s three yeast products,YC, XP and XPC, on milk production, dry matterintake (DMI), and milk composition (milk fat andmilk protein, %). A total of 60 research papers andreports (Journal, published Abstracts, Reports and

A Meta-Analysis of the Responses toFeeding Diamond V Yeast Culture to

Lactating Dairy CowsBill Sanchez, Ph.D.

Diamond V Mills, Inc., [email protected]

30

Technical Bulletins) were provided by Diamond V(USA) to be considered in the meta-analysis. BovineResearch Australia also conducted an independentliterature search (internet: Pubmed, scholar Google,Agricola, Sciencedirect, Scirus and CAB) to exploreand identify other research papers and reports thatmay not have been provided for this study. Theirliterature search did not find any other papers.Randomized clinical trials that examined the effectsof YC, XP and XPC on milk production and drymatter intake (DMI) in lactating dairy cows wereconsidered in this study. A list of publications thatwere reviewed for this study is provided in Table 2:Trials were included in the analysis if they provided:

• information on the form and dose of Diamond Vyeast products (YC, XP and XPC)

• adequate description of the randomization process• sufficient data on milk production, dry matter

intake and milk composition• a measure of variance (SE or SD), and/or P value • studies including lactating dairy cows only

Outcomes measured. The initial analysis wasconducted on milk production and DMI by Rabiee etal. (2008). An identical analysis on milk compositionwas also conducted subsequent to that originalreport. A total of 32 studies met the eligibility criteriafor meta-analysis. Data recorded from suitablestudies in lactating dairy cows were;

• Milk production (kg/d/cow) (n = 49 trials)• Dry matter intake (kg/d/cow) (n = 28 trials)• Milk fat (%) (n = 26 trials)• Milk protein (%) (n = 25 trials)

Heterogeneity. A meta-regression was alsoconducted to explore if the outcomes of these studieshave been influenced by other factors thatcontributed to heterogeneity. The followingpublication quality factors were first evaluated(Table 3).

• Source of papers (peer-reviewed: [Journal] vs. notpeer-reviewed: [Abstracts, Technical Bulletins andReports])

• If there was an evidence of randomization (Yes orNo)

• If the outcomes of study have been adjusted forconfounders (Yes or No)

Then the following potential covariates wereevaluated. In most cases there were insufficient datato provide adequate power to evaluate all factors.

• Start time (before or after calving)- category• Duration of treatment before calving (days)-

continuous

• Duration of treatment after calving- continuous• Days in milk (DIM)- continuous• Stage of lactation (early vs. mid)- category• Types of diet (TMR & Component vs. PMR &

Pasture)- category• Number of milkings per day (2 vs. 3)- category• Parity of cows in the study (primiparous vs.

multiparous herds vs. mixed herds [multiparous& primiparous])- category

• BST (Yes or No)• Type of Diamond V products (YC vs. XP vs.

XPC)- category• Dose of Diamond V products (YC vs. XP vs.

XPC)- continuous • Delivery methods (mixed with diet vs.

topdressed)- category• Other supplements (No supplement vs. Monensin

& others)- category • Measure of variance (data with SD or SE vs. data

with estimated SD and SE, using P value)-category

These meta-regressions were conducted by firstscreening individual variables with a P-value of ≤0.20. for milk production and dry matter intake, allvariables meeting the first screening criteria wereentered into a backward stepwise weightedregression method (meta-regression) until allremaining variables were significant at P<0.05. Whenthe numbers of studies in subgroups were small (e.g.stage of lactation, delivery methods), these subgroupswere categorized into two main groups and thenmeta-regression analysis was conducted on the twomain subgroups. For milk fat and milk protein, onlystage of lactation was evaluated under heterogeneity.

Publication bias. The presence of publication bias wasinvestigated using funnel plots (Light and Pillemer,1984). These graphical displays are used to examinewhether the results of a meta-analysis may have beenaffected by publication or other types of bias. If thereis bias, for example because smaller studies withoutstatistically significant effects are unpublished, thiswill lead to an asymmetrical appearance of the funnelplot with a gap in a bottom corner of the graph. Inthese situations, the effect in the meta-analysis mayhave been overestimated. Funnel plots that wereconstructed to identify publication bias in the datawith less than 5 studies were not able to detect anybias, due to the small number of papers eventuallyincluded in this study (results not shown).Publication bias was also investigated statisticallyusing both Begg’s (Begg and Manumdar, 1994) andEgger’s test (Egger et al., 1997). However, becausethese tests have low power and there is some debateregarding the statistical properties, potentials andlimitations of these tests, these were not reported.

31

ResultsA total of 32 studies (milk production trials = 49; DMItrials = 28; milk fat trials = 26; and milk protein trials= 25) met the selection criteria and were eligible formeta-analysis. The meta-regression analysis on thequality of studies used for milk production effect sizeis presented in Table 4. These data showed that theeffect size of Diamond V yeast products (YC, XP andXPC) on milk production was not influenced by thesource of paper (peer-reviewed vs. not peer-reviewed) and adjustment for confounders. However,the subgroup analysis showed that the effect sizes ofthe peer-reviewed studies were slightly greater thanthose not peer-reviewed. Subgroup analysis ofstudies for confounder adjustment also showed thatthe effect sizes of studies that adjusted the outcomesfor confounders were smaller than those studies thatdidn’t adjust for confounders (Table 4. Rabiee et al.(2008) also tested whether or not the effect ofDiamond V products on milk production wereinfluenced by the stage of lactation, type and dose ofDiamond V products (YC, XP and XPC), duration ofsupplementation before and after calving, type of dietand parity and no heterogeneity was noted due tothese potential covariates.

Overall, Diamond V products (YC, XP and XPC)significantly increased the milk production (Table 5and Figure 1a; Weighted Mean Difference (WMD) =2.05 lbs per day per cow with a 95% Confidenceinterval (CI) = 0.70 to 1.15, P< 0.0001). The forest plotof the standardized effect size differences responsesfor milk production is shown in Figure 1a.Standardized differences are equal to the weightedmean difference divided by the standard deviationand is useful when scales differ (i.e., different milkproduction ranges in early vs. mid lactation cows). Avisual examination of the forest plot for milkproduction shows clearly that ~ 90% of the trialresponses are to the right of 0. There was potentialevidence of publication bias in milk productionoutcomes (Figure 1b). Small size studies that reportedan increase in milk production appeared to be morelikely to be published than small studies with areduction in milk production, however, it was theopinion of Rabiee et al. (2008) that this should notdramatically alter the point estimates made.

The effect of Diamond V products (YC, XP and XPC)on dry matter intake (DMI) was not significant (Table5 and Figure 2a; WMD = 0.55 lbs/day; P = 0.13; 95%CI= -2.91 to 0.66 lbs/day). This lack of an overalleffect may have been due to an interaction. Meta-analysis of subgroups (early vs. mid) showed thatDMI was significantly greater during the early stageof lactation (0.68 lbs/day for early lactation cows vs.– 1.12 lbs/day for mid lactation cows). Meta-regression analysis showed that DMI was

significantly influenced by the delivery methods ofDiamond V products (mixed vs. topdressed; data notshown), however, the relatively few observationsinvolved in this difference have led us to conductmore research before concluding that these effects arereal. A visual inspection of the funnel plot (Figure2b) did not reveal major publication bias.

There also was a significant increase in milk fat %from feeding Diamond V Yeast Culture (Table 5 andFigure 3a; WMD = 0.045 % unit increase; P = 0.027;95% CI = 0.005 to 0.085 % units). No evidence ofpublication bias (Figure 3b) were present andheterogeneity was not significant due to stage oflactation. However, this could be due to too fewstudies as a subgroup analysis indicated a trend for asignificant increase in early lactation. The effect of Diamond V Yeast culture on milkprotein % was not significant (P = 0.63) overall) andthis response was not affected by stage of lactation orpublication bias.

ConclusionsThe results of this meta-analysis showed that YC, XPand XPC increased milk and milk fat percentagesignificantly. The overall effect on DMI was notsignificant which may have been offset by the largeincrease in intake in early lactation and largereduction in mid lactation. Duration of treatmentbefore calving and supplementation at differentstages of lactation, early vs. mid did not appear toinfluence milk production. A sensitivity analysis(Table 6) provides the economic implications.

Limitations of the studyThere were several studies that failed to report ameasure of variance (SE or SD) and significance ofthe outcomes. In an attempt to use these data, Rabieeet al. (2008) estimated some of these variances. Trialdesign was limited to only randomized control trialswith continuous designs. This excluded all Latinsquare and cross-over designs leaving only 16 peer-reviewed papers (out of the 32 total) in the study,which may have limited the analysis, particularly forDMI and milk composition. It was not possible tocompare the ration and nutrient composition of thediets of the trials included in this study, due toinsufficient data in the Reports, Abstracts andTechnical Bulletins. The majority of Abstracts,Reports and Technical Bulletins provided limitedinformation on randomization, number of cows atdifferent stages of study (e.g. numbers of cowsexcluded), measures of variance and level ofsignificance of the data.

32

33

References available upon requestFigure 1a. Forest plot of the effect size of Diamond V yeast products (YC, XP & XPC) on milk production (kg/d) inlactating dairy cows. Units are in standardized mean differences calculated from a random effects model.

Figure 1b. Funnel plot for the detection of publication bias in milk production (kg/day) data in lactating dairycows. Units are in standardized mean differences calculated from a random effects model.

Figure 2a. Forest plot of the effect size of Diamond V yeast products (YC, XP & XPC) on dry matter intake (kg/d) inlactating dairy cows. Units are in standardized mean differences calculated from a random effects model.

Figure 2b. Funnel plot for the detection of publication bias in dry matter intake (kg/day) data in lactating dairycows supplemented with Diamond V products (YC, XP & XPC). Units are in standardized mean differencescalculated from a random effects model.

34

Standardised Mean diff.(95% CI) %

Figure 3a. Forest plot of the effect size of Diamond Vyeast products (YC, XP & XPC) on milk fat (%) inlactating dairy cows. Units are in standardizedmean differences calculated from a random effectsmodel.

Figure 3b. Funnel plot for the detection ofpublication bias in milk fat (%) data in lactating dairycows supplemented with Diamond V products (YC,XP & XPC). Units are in standardized meandifferences calculated from a random effects model.

Figure 4a. Forest plot of the effect size of Diamond Vyeast products (YC, XP & XPC) on milk protein (%) inlactating dairy cows. Units are in standardizedmean differences calculated from a random effectsmodel.

Figure 4b. Funnel plot for the detection ofpublication bias in milk protein (%) data in lactatingdairy cows supplemented with Diamond V products(YC, XP & XPC). Units are in standardized meandifferences calculated from a random effects model.

35

36

Table 1. Diamond v products and recommended doses for calves and dairy cows

YC XP XPCPre-Weaning Calves 1.25 oz/hd/d (35g/hd/d) 0.5 oz/hd/d (14g/hd/d) 0.125 oz/hd/d (3.5g/hd/d)Post-Weaning calves 2.5 oz/hd/d (70g/hd/d) 1 oz/hd/d (28g/hd/d) 0.25 oz/hd/d (7g/hd/d)Calf starter grain (Preand post weaning) - 2% 0.5%Non-lactating cows 5 oz/hd/d (140g/hd/d) 2 oz/hd/d (56g/hd/d) 0.5 oz/hd/d (14g/hd/d)Lactating cows 5 oz/hd/d (140g/hd/d) 2 oz/hd/d (56g/hd/d) 0.5 oz/hd/d (14g/hd/d)

Table 2. List of reviewed and eligible studies for meta-analysis

Studies /trials/ Eligible studies/trialsTypes of report Studies/trials reviewed (milk production & DMI)

Studies Trials Studies TrialsTotal number of studies 60 86 32 50Technical Bulletins (TB) 11 (18%) 11 (13%) 0 (0%) 0 (0%)Reports 24 (40%) 34 (40%) 16 (50%) 25 (50%)Published abstracts 13 (22%) 20 (23%) 7 (22%) 9 (18%)Journal papers (published) 12 (20%) 21 (24%) 9 (28%) 16 (32%)Studies with no productivity dataStudies with no Milk production data 6 9 1 1Studies with no DMI data 25 35 10 21Some of these studies were initially published as reports or TB and then published as abstracts or in peer-reviewed journals.

Table 3. Study qualities (factors) that were used to evaluate the outcome of study

Information onStudy quality randomization Adjusted for confounder

Yes No Yes NoSource of paper 23/34 11/34 7/34 27/34

Journal (peer-reviewed) (68%) (32%) (20.6%) (79.4%)TB, Abstracts, Reports 0/16 16/16 0/16 16/16

(Not peer-reviewed) (0%) (100%) (0%) (100%)Information on randomization 7/23 16/23Yes (30%) (70)No 0/27 27/27

(0%) (100%)

Table 4. Summary of meta-regression (Coefficient, 95% CI) of study factors that may have influenced the milkproduction results and meta-analysis results of the sub-populations

Outcome Group 1 Group 2Milk production data Coefficient P value N Chi-square N Chi-square

(95% CI) Effect size statistics=Q Effect statistics=Q(85% CI) (P value) size (P value)

(95% CI)Study qualitySource of paper -0.66 0.860 N = 16 Q = 9.6 N = 33 Q = 41.12Group 1=Journal (peer- (-0.80 – 0.67) 0.27 P = 0.844 0.22 P = 0.130reviewed) vs. (0.097 – (0.13 –Group 2=TB, abstract, 0.45) 0.30)report (not peer-reviewed P = 0.02 P <

0.0001Information on 0.294 0.319 N = 27 Q - 23.68 N = 22 Q = 27.59Randomization (-0.283 – 0.87) 0.21 P = 0.595 0.23 P = 0.152Group 1=Yes vs. (0.11 – (0.12 –Group 2= No 0.30) 0.34)

P < 0.0001 P < 0.0001Confounders -0.68 0.120 N = 43 Q = 38.95 N = 6 Q = 11.38Group 1=Adjusted for (-1.53 – 0.18) 0.20 P = 606 0.32 P = 0.044confounders vs. (0.13 – (0.17 –Group 2= Not adjusted 0.27) 0.66)for confounders P< 0.0001 P = 0.063

Table 5. Summary of the effect of Diamond V Yeast Culture on weighted mean differences, N, 95% confidencelimits, significance, and test for heterogeneity for milk, dry matter intake and milk fat and protein overall andseparately during early and mid lactation. Meta analyses and tests for heterogeneity were evaluated using a mixedmodel assuming a random effects model overall with common among-study variance component across subgroups.Subgroups were combined using a fixed effects model.

37

38

39

Research Update from theUMN College of Vet Medicine

John FetrowUniversity of Minnesota

University of MinnesotaCollege of Veterinary Medicine

Dairy Research Update

S. Godden, M. Donahue, P. Pithua, A. Lago,R. Bey, S. Wells, J. Fetrow, P. Rapnicki

Department of Veterinary Population Medicine,University of Minnesota

Outline

• Colostrum management:– Which is better: bottle or tube?– Colostrum cleanliness:

• Reducing bacterial levels in colostrum• Colostrum replacers• Pasteurizing colostrum

• Johne’s control programs:– Pasteurized milk

• Mastitis:– On-farm culture systems

Outline





Principles of Colostrum Management

Quality > 50 g/L IgGQuantity 4 quarts (10% BWt)Quickness < 6 hrsCleanliness < 100,000 cfu/ml TPC

Bottle vs Tube – Does the method of feeding matter?

(U of MN. Summer, 2007)

• Research Question: – Would volume fed affect

IgG absorption whenusing bottle or tube?

Methods – Calf Enrollment Procedure

Newborn calf removed from dam (< 1 hour old)

BOTTLE1 dose CR(100 g IgG)

in 1.5L

Random Assignment


in 1.5L

TUBE1 dose CR(100 g IgG)

in 1.5L

BOTTLE2 doses CR(200 g IgG)

in 3L *

TUBE2 doses CR(200 g IgG)

in 3L

40

Methods – Calf Enrollment Procedure

Newborn calf removed from dam (< 1 hour old)


in 1.5L

Random Assignment


in 1.5L

TUBE1 dose CR(100 g IgG)

in 1.5L

BOTTLE2 doses CR(200 g IgG)

in 3L *

TUBE2 doses CR(200 g IgG)

in 3L

Post-colostral blood samplecollected at 24 hr for IgG measures* 9 of 24 calves

did not consume3L by bottle

(tube fed rest)

A small volume fed using a nipple bottle had the highest efficiency of absorption of IgG

39.0%41.1%40.2%

51.7%

0%

10%

20%

30%

40%

50%

60%

Bottle Tube Bottle Tube

Colostrum Treatment Group

Appa

rent

Effic

ienc

yof

Abso

rptio

nof

IgG

(%)

100 g IgG in 1.5 L 200 g IgG in 3 L

a

bbb

Effect of Feeding Method and Large vs Small Volume on Serum IgG Concentrations at 24 hrs|

18.719.7

9.712.4

0

5

10

15

20

25

Bottle Tube Bottle Tube

Colostrum Treatment Group

Seru

mIg

Gco

ncen

tratio

n(m

g/m

l)

100 g IgG in 1.5 L 200 g IgG in 3.0 L

a

cc

b

Bottle vs Tube: Does Volume Fed Matter?

• Conclusions:

– Producers should feed large volumes of colostrum or colostrum replacer (these calves had the highestserum IgG levels)

– For calves fed a small volume, feeding with a bottleresulted in improved efficiency of absorption andhigher serum IgG levels

– For calves fed a large volume, method of feeding did not affect efficiency of absorption or IgG levels.

Outline





COLOSTRUM CLEANLINESS

3 Major Sources of Bacterial Contamination of Colostrum

1. Infected gland or fecal contamination

2. Contaminatedcollection, storage or feeding equipment

3. Bacterial proliferation in stored colostrum

Total Bacteria Counts in Minnesota Colostrum(Swan et al. 2007. JDSci. 90)

01020304050607080

10 100

1,000

10,00

0

100,0

00

1,000

,000

10,00

0,000

100,0

00,00

0

1,000

,000,0

00

10,00

0,000,0

00

Total Plate Count (cfu/ml)

Freq

uenc

yof

Sam

ples

Median TPC = 615 million cfu/ml (73 to 104 billion)

93% of samples > 100,000 cfu/ml TPC“We are feeding ‘fat-laden’ manure” Rob Trembley, 2006

We should try to reduce bacterial contamination in colostrum because…

• Pathogenic bacteria cause disease (E. coli scours, Johne’s disease, etc.)

• Bacteria can block IgG absorption across the intestine

High Bacteria Levels in Colostrum Associated with Lower Efficiency of Absorption of IgG

(TMF, Summer, 2007)

0%

10%

20%

30%

40%

50%

60%

0 1 2 3 4 5 6 7 8

Total Bacteria Count in Colostrum (Log10(TPC, cfu/ml))

App

aren

tEffi

cien

cyof

Abs

orpt

ion

ofIg

G(%

)Cutpoint of 100,000 cfu/ml

Conclusions: lower colostrum bacteria levels are associated with better passive transfer of IgG

Outline





Colostrum Replacement Products

• Must provide:– Minimum of 100 gm IgG / dose (plasma- or lacteal-derived Ig)

– Nutrients (fat, vitamins, minerals, etc.)– Cost: $25 to $30 (U.S.D.)– Convenient, consistent supply of IgG if sufficient clean, high

quality maternal colostrum is not available

41

Dose Response of serum IgG to IgG Mass Fed(U of MN Compiled data from Summer, 2006 LO’L CR Feeding Trials)

9.6

15.2

19.0

0

5

10

15

20

25

100 g 150 g 200 g

IgG mass fed (g)

Ser

umIg

G(m

g/m

l

Study 1FPT = 54%

Study 1FPT = 0%

Study 2FPT = 5%

Conclusion: We can reduce the risk of FPT by feeding higher doses (150 to 200g IgG) in Colostrum Replacers

Is colostrum a risk factor for transmission of Mycobact erium avium subsp.

paratube rculosis (MAP)?

• Though fecal-oral transmission is most common, MAP can be shed in colostrum and milk of subclinicallyinfected cows

(Sweeney et al. J.Clin.Micro. 1992. 56;Streeter et al., J. Clin. Micro. 1995. 30)

• Can one feeding of colostrum cause infection with MAP?

Risk of MAP Infection in Calves Fed Raw Colostrum or a Colostrum Replacer

(Pithua et al.,J.A.V.M.A. 2008. Accepted)

Newborn heifer calves from 12 herds

(N = 497)

colostrum replacer (n = 236)

maternal colostrum(n = 261)

Adult Period: 1st calving to 54 mos:- Fecal culture and serum ELISA for MAP at 30, 42 and 54 mos.

Acquire / SecureAPC, Inc.

Raw colostrum is an importantsource for transmission of MAP

Conclusion: We can reduc e the risk of Johne’stransmission by feeding Colostrum Replacers

Outline


• Reducing bacterial levels in colostrum• Colostrum replacers• Pasteuri zing colostrum



Can we reduce bacterial exposure through colostrum?Developing a method to pasteurize colostrum

• Continuous flow (72 ºC x 15 sec)or Batch (63 ºC x 30 min)– Unacceptable thickening – > 1/3rd loss of IgG (mg/ml)– Lower serum IgG in calves

(Green et al. JDSci. 2003. 86:246;Godden et al. JDSci. 2003. 86:1503)

• Batch pasteurize: 60 ºC x 60 min– No viscosity changes– No change in colostrum IgG (mg/ml)– Significantly reduce or eliminate

M. para tuberculosis, Sal monella, Mycoplasm a bovis, E. coli, Lis teria

(McMartin et al. JDSci. 2006. 89:2110Godden et al., JDSci. 2006. 89:3476)

42

Pilot study Feeding PasteurizedColostrum to Calves

(Johnson et al., JDSci. 2007)

• Feeding batch pasteurizedcolostrum (140 ºF x 60 min) resulted in…– No viscosity changes– No change in colostrum IgG (mg/ml)– Significantly reduction in bacteria

counts in colostrum– Improved passive transfer of IgG in

calves (mg/ml)

ObjectiveDescribe the effect of on- farm heat-treatment ofcolostrum in multiple commercial herds on:– Colostrum characteristics (bacteria levels, IgG)– Passive transfer of immunoglobulins in calves (serum IgG)

Materials and Methods

• Summer 2007

• Six commercial freestall dairyfarms in WI and MN

• Herd size: 1,200 – 2,500 cows

• All herds confirmed Johne’s positive either byfecal culture or ELISA within the past three years

Effect of feeding pasteurized colostrum on colostrum characteristics and passive transfer in calves

Fresh colostrum- split into 2 x 1gallon aliquots

Heat-treat (60 oC x 60 min)

Refrigerate < 48 hr

Fresh

Feed (n=25 calves)

Feed (n=25 calves)Refrigerate < 48 hr

All calve s fed 3.8 L colos trum us ing esophageal tube feeder at < 2 hrs old

< 0.00011929Failure PassiveTransfer (%)

< 0.000114.4816.94Serum IgG (mg/ml)572518Calves (n)

Calf

0.4259.1660.67<IgG (mg/ml)<0.00012.304.36TCC (log10 cfu/ml)< 0.00013.615.40TPC (log10 cfu/ml)

266266Batches (n)

Colostrum

Heat-Treated

Raw P- ValueColostrum

Parameter

0

10

20

30

40

50

60

70

80

90

1 2 3 4 5 6 AllFarm Number

Col

ostru

mIg

G(m

g/m

l)

Raw Heat-Treated

a aa a a

aa

aa

aa

a a

Means that differ significantly have a different letter

a

Effect of Treatment on Colostrum IgG

43

0.0

1.0

2.0

3.0

4.0

5.0

6.0


Log

10

Colo

stru

mTPC

(cfu

/ml)

Raw Heat-Treated

a

b

a

b

a

b

a

b

a

b

a

b

a

b

Means that differ significantly have a different letter

Effect of Treatment on Total Bacteria Counts in Colostrum

0

5

10

15

20

25

30


Ser

um

IgG

(mg

/ml)

Raw Heat-Treated

a

b

a

b

a a

a

b

a a

a a

ab

Means that dif fer signif icantly have a dif ferent letter

Effect of Treatment on Calf Serum IgG (mg/ml)

Conclusions

• Heat-treatment of colostrum at 140 °F (60 °C) for 60 minutes can be successfully adopted on commercial dairy farms.

• Heat-treatment of colostrum resulted in:– No harmful effect on colostrum IgG levels– Significant reduction in colostrum bacteria counts– Significant improvement in calf serum IgG

• Still to do:– Describe preweaning health– Describe adult: production, longevity, Johne’s disease

Summary of approaches to reduce pathogen exposure through colostrum

• Avoid pathogens from infected glands,fecal contamination:– Identify infected cows? (MAP)– Don’t let calf suckle dam– Udder prep– Don’t pool raw colostrum

• Reduce other sources of contamination:– Sanitation of milking, storage & feeding equipment

• Prevent bacterial proliferation in stored clostrum:– Feed (< 1-2 hrs), refrigerate (< 48 hrs) or freeze ASAP– Use of preservatives?

• Additional tools:– Colostrum replacers (feed 150 - 200 g IgG, efficacy tested)– Pasteurize colostrum

Outline





Objective• Describe the effect of feeding pasteurized non-saleable

milk (vs. conventional milk replacer) on short- and long-term performance in calves:– Calves:

• Health, growth

– Adults:• Milk production• Longevity• Risk for infection with Johne’s disease

44

Methods:

Calf Enrollment

• 439 calves enrolled from 2 MAP-infected dairies:– Heifers and bulls– Dec., 2001 to Aug., 2002

• Assigned to treatment at arrival atheifer grower (1-2 days):– Batch pasteurized non-saleable

milk (DairyTech, Inc.)– 20:20 milk replacer

• Milk from Johne’s infected dairy(10 - 12% seroprevalence)

Results from Preweaning Study

(Godden et al. 2005. JAVMA. 226:1547-1554)

Parameter 20:20 Milk Pasteurized PReplacer Milk Value

------------------------------------------------------------------------------Calves enrolled (n) 217 222ADG (kg/d) 0.35 0.47 < 0.001% Treated 32% 12% < 0.001% Died 11.6% 2.3% < 0.001Economics ($) . + $34------------------------------------------------------------------------------

Methods:

Follow-up of Adult CowsMilk Pasteurized

Replacer Milk Total----------------------------------------------------------------------------------Heifers enrolled 116 118 234

Heifers weaned 104 116 220

Heifers w 1st calving event 54 65 119----------------------------------------------------------------------------------

Large study herd disperses in fall, 2003.- Approx. half of heifers lost to follow-up.- What heifers we could trace were followed to herds in WI, IN and CA.

- Lessons learned?

Methods –

Data Collection for Adults

• 5 herds: MN, WI, IN, CA

• 1st calving to avg. 57 months (study concluded Jan. 2007)

• Collected:– Calving dates– Records of milk production– Records of culling or death– Tested for infection with MAP at avg. 25, 42 and 57 mos.

• serum ELISA • fecal culture

Adult Cow Performance- From 1st calving to avg. 57 months

Milk Pasteurized P Replacer Milk Value

------------------------------------------------------------------------------------------Cows w first calving event 54 65

Milk yield (lact > 150 DIM)Lact 1 (kg) 11,200 11,370 (+ 170) 0.81Lact 2 (kg) 11,246 13,257 (+2,011) 0.004Sum Lact 1 & 2 (kg) 22,028 23,964 (+1,935) 0.084

% culled or died 53.7% 41.5% 0.036

% MAP test positive 27.8% 21.5% 0.36-------------------------------------------------------------------------------------------

Risk for Culling or Death- From birth to avg. 57 months

Milk replacer

Pasteurized milk

Hazard Ratio Milk Replacer = 1.81 (P = 0.0037)

Age (Months)

45

Conclusions

• Calves fed pasteurized milk (vs conventional milk replacer)had…

– Preweaning period:• Lower risk of illness and death• Higher ADG (+ 0.25 lb/day)• Cost-benefit = $34/calf at weaning

– Adult period (calving to 57 mos):• Increased milk production • Lower risk for culling or death• Were NOT at increased risk for infection with MAP• Cost-benefit = TBD

Significance

• If raw milk is an important vehicle for transmission of MAP, then on-farm pasteurization was successful in eliminating the organism or else reduced it to levels below an infectious dose

• Short- and long-term benefits from feedingpasteurized milk thought to be due toimproved plane of nutrition:– Producers should adopt milk feeding

programs to increase plane of nutrition topreweaned calves

Outline





Effect of Using an On-Farm Culture Based

Treatment System on Antibiotic Use, Milk

Withholding Time, and Clinical and Bacteriological

Cure for Clinical Mastitis

Preliminary Results

Alfonso Lago, LV, Sandra Godden, DVM, DVSc, Russ Bey, PhD,Ken Leslie, DVM, MSc, Pamela Ruegg, DVM, MPVM, Randy Dingwell, DVM, DVSc

University of MinnesotaUniversity of Guelph University of Wisconsin

Project herds:8 Holstein herds in MN, WI and Ontario

(150 to 1800 cows)Objectives

• Reasoning:

– More than half of cultures from Clinical mastitis yield nogrowth or Gram-negative bacteria

– Some of these cases may not benefit from antibiotic therapy

– On-farm culture may allow us to make a rapid cow-sidetreatment decision

46

Objectives

• Describe effect of using OFC for strategic treatment of clinical mastitis on…– Antibiotic use– Risk for extended therapy / secondary treatment– Days to return to visibly normal milk– Days out of tank– Bacteriological cure rate– Risk for recurrence of clinical mastitis– Lactation SCC and milk production– Longevity in herd

Positive-Control Group

Culture-Based Group

No Growth Gram-neg

Gram-pos

Enrollment Day Next Day

Methods: Enrollment Process for mild and moderate clinical

mastitis cases

Mix Infection

Cefa-Lak® IMM

CultureBasedGroup

Treat quarteronly if on-farmculture showsGram + IMI

Cultureon Farm

Culture in lab

Follow-upSample 3Sample 2Sample 1

Day 0 to 365Day 21Day 14Enrollment (day 0)Grade 1 o r 2 Clinical Mastitis

- Farm records:Clinical Mastitis

- DHIA records:SCCMilk productionCulling/death events

Culturein lab

Culture in lab

Treat quarterimmediately after

milk samplecollected

Culturein lab

PositiveControlGroup

Methods: Follow-Up

Results:Etiology

Coliforms24%

tococcus15%

us

Othe r10%Coag Ne g S taph

10%No Growth

34%

No Growth

Streptoco ccus

Staph aure us7%

n = 452 q uarter clinical ca ses

100%

43%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Positive-Control Culture-Based

Results - % Quarter Cases Treated with Antibiotics:Less than Half for the Culture-Based Program

P < 0.001 36%

19%

0%

5%

10%

15%

20%

25%

30%

35%

40%


P = 0.0017

Results – Extend / Secondary Treatment:Less for the Culture-Based Program

7%16%No Growth

26%35%Gram-pos

36%

57%

Positive-Control

21%Gram-neg

19%All

Culture-Based

47

Results – Days to Clinical Cure:No Effect of Treatment Program

P = 0.2879

Days to Clinical Cure

Positiv e-Control

Culture-Bas ed ------

2.9 d2.7 dNo Growth

3.5 d3.2 dGram-pos

2.8 d

2.5 d

Positive-Control

3.3 dGram-neg

3.2 dAll

Culture-Based

Results – Days Out of the Tank: Tend for lower for the Culture-Based Program

P = 0.0763

Days Out of the Tank

Positiv e-Control

Culture-Bas ed ------

3.9 d5.6 dNo Growth

6.4 d6.2 dGram-pos

5.9 d

6.3 d

Positive-Control

4.8 dGram-neg

5.1 dAll

Culture-Based

71%

60%

0%

10%

20%

30%

40%

50%

60%

70%

80%


P = 0.2978

Results – Bacteriological Cure:No Effect of Program

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Cure

Risk

Coliforms Streps CNS S. aureus Other


Results – Bacteriological Cure:Not Different Program Effect Among Bacteria

P = 0.1636

P = 0.2436

P = 0.2610

P = 0.4172

P = 0.6651

• Significant reduction in antibiotic use:– Positive - Control = 100%– Culture - Based = 43%

• Less extended / secondary treatment:– Positive - Control = 36%– Culture - Based = 19%

• Tendency for a reduction in days out of the tank:– Positive - Control = 5.9 days– Culture - Based = 5.1 days

• No statistically significant differences of treatment program on:– Days to Clinical Cure and Bacteriological Cure

• Still to do:– Clinical Mastitis Recurrence, SCC, milk yield, longevity

Summary of Preliminary Results

Questions??

48

49

IntroductionAlthough the incidence rate of postpartum metabolicdisorders is actually quite low (McGuire et al., 2008),management of transition cows continues to be apriority among dairy producers. Early lactation iswhen most dairy cows are culled (Weigel, 2008) andeven though the overall rate of disorders is low, therate may be much higher on individual farms. Whenproblems occur on individual farms, it is oftendifficult to identify the cause. It may be nutritional, itmay be environmental (e.g. housing), or it may be aninteraction between nutrition and environment. Thisreview will examine energy feeding and how itaffects postpartum health and production of dairycows. I will briefly review data on the pre-freshtransition period, and then discuss the potential forthe far-off dry period and the early post-fresh periodto influence animal health and productivity. Aconclusion regarding the relative importance of thefar-off dry period, pre-fresh, and post-fresh transitionperiod will be made.

Feeding Pre-fresh Transition CowsDecreasing the forage-concentrate ratio of pre-freshdiets does increase dry matter intake (DMI, Grummeret al., 2007). Surprisingly, the increase in DMI occursfor sustained periods of time (i.e. 3 wk) even if cowsare in positive energy balance at the time theadditional grain is introduced. In other words, theredoes not seem to be a functional feedback mechanismto maintain energy balance when increasing energydensity in the diet during the pre-fresh transitionperiod. The obvious question is: does this increase ingrain and DMI provide some benefit to the cow suchthat her postpartum health and productivity isincreased. Potential benefits include: suppression ofadipose lipid mobilization as feed intake decreases atcalving, stimulation of acid production and rumenpapillae growth, and acclimation of rumen microbialpopulation to high starch diets. Such benefits,although widely cited, are not supported by theliterature. A summary of 10 studies conductedbetween 1995 and 2005 indicated that postpartumDMI, milk production, and liver fat are notinfluenced by prepartum forage: concentrate ratio(Grummer et al., 2007).

Figure 1 shows the energy balance of cows as theytransition from the dry period to the lactation period.It appears that despite the 30-35% decline in DMI ascalving approaches (Hayirli et al., 2002), cows do nottypically experience significant negative energybalance prior to calving. Therefore, improvingenergy balance prior to calving by feeding additionalconcentrate is not justified and it may be a negativefactor if the increase in DMI results in a greaterdecrease in feed intake as calving approaches.Grummer et al. (2004) suggested the magnitude ofdecrease in feed intake prior to calving may be morehighly associated with plasma NEFA and liver TG.Management factors that trigger decreases in feedintake (excessive diet or pen changes, overcrowding,or heat stress) may be more detrimental if cows havebeen fed higher concentrate diets and have thepotential for greater decreases in intake.

Feeding Far-off Dry CowsAlthough very little research has examined feedingstrategies of far-off dry cows, it is commonly believedthat excessive overfeeding of grain during the entiredry period should be avoided to minimize thelikelihood of over-conditioned cows, reduced feedintake, or metabolic disorders. Consequently,suggesting that far-off dry cows should be fed a dietthat meets energy requirements is neither novel norsomething that should be discouraged. However, itmay be premature to imply that far-off dry cow dietsmust be formulated “just right”, contain straw, or bea bulky diet that will lead the cow to only meetrequirements when fed ad libitum (Drackley andJanovick Guretzky, 2007). Consider results fromDann et al. (2006; a study that was instrumental incrystallizing the concept of the “Goldilocks diet” forfar-off dry cows [Drackley and Janovick Guretzky,2007]) versus those of a recent study conducted inour laboratory (Silva-del-Río et al., 2007; Table 1).Dann et al. (2006) conducted an experiment with a 3x 2 factorial arrangement of treatments: three energyfeeding strategies during the far-off dry period andtwo different energy feeding strategies during thepre-fresh transition period. There were no effects ofpre-fresh diet and no interactions of far-off and pre-fresh dry periods, so only the main effects of far-off

What is the Most Critical Feeding Period:Far-off Dry, Pre-fresh Transition,

or Post-fresh Transition?Ric Grummer

Department of Dairy ScienceUniversity of Wisconsin-Madison

[email protected]

dry period will be reported. Far-off dry periodtreatments were feeding a diet containing 1.59 McalNEl/kg DM at ad libitum (150% of NRCrequirements) or restricted (80% of NRCrequirements) feed intake or a low-energy dietcontaining straw fed ad libitum (1.30 Mcal NEl/kgDM, 100% of NRC requirements). Silva-del-Río et al.(2007) conducted an experiment with a 2 x 2 factorialarrangement of treatments: cows pregnant withsingletons or twins and a “close-up” diet withmoderate energy for 3 or 8 wk prepartum. The“close-up” diet contained 1.54 Mcal NEl/kg DM andthe far-off dry cow diet contained 1.32 Mcal NEl/kgDM. Therefore, treatments were 1.32 or 1.54 McalNEl/kg DM during the first 5 wk of the dry periodand were very similar to two of the far-off treatmentsemployed by Dann et al. (2006). There were fewinteractions between pregnancy status and diet, soonly the main effects of diet are shown (Table 1).

The results of the two trials are strikingly differentand there are no apparent reasons for the differencesin results. What conclusions can be drawn? Dann etal. (2006) concluded that “overfeeding during the far-off period had a greater negative impact onperipartum metabolism than did differences in close-up nutrition”. This conclusion is a bit puzzling sincethere were no negative effects of overfeeding duringthe close-up period and the differences due to far-offtreatments were small and of questionable biologicalsignificance (Table 1). Conversely, our data indicatedadditional energy feeding during the far-off periodwould cause a dramatic increase in postpartum milkyield. Based on our results, should we berecommending that producers feed transition dietsduring the entire dry period? The answer is probablynot. Based on both studies, there are two logicalconclusions. One might conclude that diets for far-offdry cows do not have to be formulated to be “justright” and there is substantial variability in diets thatcan be fed without adverse affects. This is probably areasonable conclusion in that it is quite conservative.It also takes into account lessons learned fromexperiments that examined feeding pre-freshtransition cows; it is problematic to base decisions onlimited data.

Another conclusion might be that these trials wereunder-replicated (n = 20-25 cows/treatment for thesetwo studies), making it difficult to draw firmconclusions. Typically, statistical analysis is reportedfor lactation data and blood or liver analysis.However, incidence of health disorders are oftenreported without statistical analysis with the authorsstating something to the effect: there was insufficientreplication to detect treatment effects. Healthdisorder data is binomially distributed, i.e., themeasurement is recorded as yes or no rather than as a

continuous variable (infinite number of outcomes)like milk production. It is well recognized thatsubstantially more replication is needed for detectingtreatment differences for parameters that arebinomially distributed. The critical question is: sincehealth affects milk production, can we have sufficientanimals on study to detect treatment differences onmilk production when we do not have sufficientanimals on study to detect differences on animalhealth? I would argue that we cannot and that it isvery dangerous to draw conclusions from individualstudies, particularly when replication was limited(e.g. 10-30 cows per treatment).

Feeding Fresh CowsAmazingly, an area of research that has received littleattention is feeding of the immediate postpartumcow. Why? Researchers avoid doing studies on freshcows because tremendous variability amongst cowsmakes it difficult to design experiments withsufficient replication. Most fresh cow studies areinitiated at 3 wk postpartum or later when cowvariability is reduced and there is less likelihood ofloosing a cow from the study! This is unfortunatebecause it easy to make an argument that nutrition ofthe cow during the first 3 wk postpartum may be themost important.

The most rapid decrease in energy balance andnegative energy balance nadir usually occurs duringthe first 3 wk postpartum (Figure 1). Aftersummarizing 26 studies, Brixy (2005) indicated thatpositive energy balance was reached byapproximately 50 days in milk and the minimumenergy balance occurred at about 11 days in milk. Wecollected data from twenty studies published in peerreviewed journal articles since 1988 (Grummer andRastani, 2003). The mean number of days in milkuntil energy balance was reached was 45 (standarddeviation = 21 d). The correlation between peak milkyield (r = 0.24, P = 0.16) or days to peak milk yield (r= 0.23, P = 0.17) and time to reach positive energybalance was extremely low indicating that some otherfactor besides energy output was responsible forvariability in the length of time it takes to reachpositive energy balance. The data did not allow us toexamine the relationship between energy intake orDMI and time to reach positive energy balance.However, we were able to examine the relationshipbetween energy density of the diet and days topositive energy balance. The data indicated thatthere was a stronger relationship between days topositive energy balance and energy density of thediet (r = 0.57, P < 0.0001) than peak milk yield. Thisdata provides evidence that energy intake may be amore important factor affecting return to positiveenergy balance than milk yield because energy intakeis a function of DMI and energy density of the diet.

50

We also examined individual cow data from aspecific research trial that included 24 primi- and 49multiparous cows from 2 through 21 wk postpartum(Grummer and Rastani, 2003). Average 4% fat-corrected milk yield was 29.2 kg/day for primiparouscows and 38.4 kg/day for multiparous cows for thefirst 21 wk postpartum. It is clear from data in Table 2that energy balance is more closely related to energyintake than energy output as measured by fat-corrected milk. Average time to reach positive EBwas the same for multiparous and primiparous cows,5 ± 2 wk. McGuire et al. (2008) performed a similaranalysis from a trial including 29 multiparousaveraging 46 kg milk/d for the first 12 wk oflactation. In agreement with our data, the correlationbetween energy balance and DMI (r = 0.751,P < 0.001) was much higher than with milk yield(r = 0.051; P < 0.037).

We can conclude several things from this research: 1.Energy status is most compromised during the first 3wk of lactation. 2. Return to positive energy balanceoccurs relatively quickly for most cows if they are feddiets that are nutritionally adequate (as is the case inthese research studies). 3. Energy balance is morelikely to be related to energy intake than milk yield.4. Minimizing negative energy balance is most likelyto be accomplished through successful feeding ratherthan through decreasing milk yield. 5. In alllikelihood, the most important time for feeding thecow correctly is during the first 3 wk postpartum.Unfortunately, little research has been conducted todefine “successful feeding” during this period.

Data from Rabelo et al. (2003, 2005) indicated thatenergy density of diets immediately postpartum aremore critical than energy density of dietsimmediately prepartum. They utilized a 2 x 2factorial arrangement of treatments. Cows were feddiets containing 1.55 (Dry Low- DL) or 1.65 McalNEl/kg DM (Dry High- DH) for the last 4 wk prior tocalving. Following calving, one half the cows fromeach group were fed diets containing 1.67 (High - H)or 1.74 Mcal NEl/kg DM (Low - L) for the first 3 wkafter calving. After that, all cows were fed H. Theexperiment was designed to determine how best totransition cows from far-off dry cow diets to a highenergy lactation diet.

Figure 2 shows the milk production results. Therewas no effect of prepartum treatment and there wasno interaction between prepartum treatment andpostpartum treatment. There was no main effect ofpostpartum treatment, but Figure 2 clearly shows thepostpartum treatment by time interaction (P < 0.001).There was a divergence of curves until 3 wkpostpartum. At that time, treatments wereterminated and the milk production difference

between treatments was maintained or wasnarrowed. For the first 35 d postpartum, cows on Hwere in a more favorable energy status as indicatedby higher plasma glucose concentrations (49.2 vs 45.9mg/dl; P < 0.001) and lower beta-hydroxybutyrateconcentrations (4.1 vs 6.3 mg/dl; P < 0.001). Therewas no effect of prepartum diet on triglycerideaccumulation in the liver at calving, however, cowsfed H postpartum had lower liver triglyceride in theliver at the end of the 3 wk treatment period (11.1 vs15.6 ug triglycerid/ug DNA; P = 0.07). By 35 dpostpartum, liver triglyceride was lower and therewas no difference between treatments (4.2 vs 4.7 ugtriglyceride/ug DNA; P = 0.84). However, it must bekept in mind that cows were on the same dietbetween 21 and 35 d postpartum. More importantly,energy balance should have been improving duringthis time; clearly triglyceride was being cleared fromthe liver from 21 to 35 d postpartum.

Finally, consider a study we did many years ago thatexamined the effects of dietary carbohydrate duringthe transition period on feed intake, metabolic health,and lactation performance of dairy cows (Minor etal., 1998). All cows (25 primiparous, 50 multiparous)were placed on the standard NFC diet (S) at 26 daysprepartum and then assigned to S or high NFC (H)diets at 19 days prepartum. After calving, all cowscontinued on either a S or H lactation diet.NDF/NFC content of pre- and post-fresh diets were49/24 and 26/42 for S and 30/44 and 22/47 for H.Milk and milk energy yield was higher for cows fedH (31.7 vs 33.0 kg/d, P < 0.10 and 22.2 vs 23.2Mcal/d, P < 0.05). Plasma β-hydroxybutyrateconcentrations were almost identical until afterparturition at which time they were dramaticallylower for cows fed H (Figure 3). Because of theexperimental design of this study, it is impossible totell if this postpartum difference was due toprepartum diets, postpartum diets, or both.However, it is not unreasonable to speculate thatpostpartum NFC may have had an important affecton b-hydroxybutyrate concentrations in earlylactation.

Additional research is needed to determine the mostappropriate feeding strategies of cows immediatelypostpartum. Emphasizing feeding nonfibercarbohydrate (NFC) at the expense of neutraldetergent fiber (NDF) may reduce the likelihood offatty liver and ketosis, but increase the likelihood ofacidosis. Conversely, emphasizing feeding of NDF atthe expense of NFC may decrease the likelihood ofacidosis but increase the likelihood of fatty liver andketosis. Unfortunately, it is too simplistic torecommend NFC and NDF levels in post-fresh diets.NFC and NDF are heterogeneous fractions whosecharacteristics vary greatly among feeds. Rates of

51

fermentation of NFC/NDF are not consistent acrossfeeds. Likewise, factors such as particle length ofNDF-rich feeds are variable and influence the rumenenvironment and rates of digestion. Unfortunately,inadequate information is available at this time tomake precise feeding recommendations for cowsduring the first three weeks postpartum.

ConclusionsFor decades, we preached that feeding during thepre-fresh transition period was the most critical timefor the prevention of metabolic problems aftercalving and that it was important to feed additionalgrain during that period. Data clearly shows thatcows tolerate a wide variety of diets during that timeand forage to concentrate ratio has little impact onpostpartum performance. Now there is considerableinterest in the far-off dry period and some feelfeeding during that period may be more crucial thanthe pre-fresh transition period. To our knowledge,there are only two studies that specifically havecompared only the far-off dry period feedingschemes, and it is safe to their results werecontradictory and it is too early to draw anyconclusions. There is a fairly substantial body ofevidence to indicate that cow should not be over-conditioned at calving. However, it is difficult tosignificantly increase condition score of dry cowsduring the relatively short time frame of the dryperiod when feed intake is low. The most criticaltime for feeding transition cows is probablyimmediately postpartum, but research is lacking toconfidently develop optimal energy/carbohydratefeeding strategies for that period.

ReferencesBrixy, J. D. 2005. Validation of a prediction equation for energy

balance in Holstein cows and heifers. M. S. Thesis. Universityof Idaho, Moscow.

Dann H. M., N. B. Litherland, J. P. Underwood, M. Bionaz, A.D’Angelo, J. W. McFadden and J. K. Drackley. 2006. Diets duringfar-off and close-up dry periods affect periparturientmetabolism and lactation in multiparous cows. J. Dairy Sci.89:3563-3577.

Drackley, J. K., and N. A. Janovick Guretzky. 2007. Controlledenergy diets for dry cows. Pages 7-16 in Proc. 8th WesternDairy Mgt. Conf., Reno, NV. Oregon St. Univ., Corvallis.

Grummer, R. R., A. Brickner, and N. Silva-del-Rio. 2007. Highforage or high grain for dry cows: what is best for animalhealth and reproduction? Pages 187-195 in Production Diseasesin Farm Animals. M. Furll ec. Merker Druck und. Kopier-Zentrum. Leipzig, Germany.

Grummer, R. R., and R. R. Rastani. 2003. When should lactatingdairy cows reach positive balance? Prof. Anim. Scientist.19:197-203.

Grummer, R. R., D. G. Mashek, and A. Hayirli. 2004. Dry matterintake and energy balance in the transition period. Pages 447-470 in Managing the Transition Cow to Optimize Health andProductivity. Veterinary Clinics of North America. N. B. Cookand K. V. Nordlund, eds. W. B. Saunders Co., Philadelphia, PA.

Hayirli, A., R. R. Grummer, E. V. Nordheim, and P. M. Crump.2002. Animal and dietary factors affecting feed intake duringthe prefresh transition period. J. Dairy Sci. 85:3430-3443.

McGuire, M. A., M. Theurer, and P. Rezamand. 2008. Putting thetransition period into perspective. Pages 257-264 in theProceedings of the 23rd Annual Southwest Nutrition andMagnagement Conference. University of Arizona, Tuscon.

Minor, D. J., S. L. Trower, B. D. Strang, R. D. Shaver, and R. R.Grummer. 1998. Effects of nonfiber carbohydrate and niacin onperiparturient metabolic status and lactation of dairy cows. J.Dairy Sci. 80:189-200.

Rabelo, E., R. L. Rezende, S. J. Bertics, and R. R. Grummer. 2003.Effects of transition diets varying in dietary energy density onlactation performance and ruminal parameters of dairy cows. J.Dairy Sci. 86:916-925.

Rabelo, E., R. L. Rezende, S. J. Bertics, and R. R. Grummer. 2005.Effects of pre- and post-fresh transition diets varying in dietaryenergy density on metabolic status of periparturient dairy cows.J. Dairy Sci. 88:4375-4383.

Silva-del-Río, N., P. M. Fricke, and R. R. Grummer. 2007. Effects oftwin pregnancy and dry period feeding strategy on milkproduction, energy balance and metabolic profiles in Holsteincows. J. Dairy Sci. 90 (Suppl. 1): 615.

Weigel, K. 2008. Genetic improvement of dairy cow longevity.http://www.extension.org/pages/Genetic_Improvement_of_Dairy_Cow_Longevity

Table 1. A comparison of two trials that comparedfeeding strategies for far-off dry cows.

Dann et al., 20061, 2 Silva-del-Río et al., 20071, 3

Far-off drycow 1.30 Mcal 1.59 Mcal 1.59 Mcal 1.32 Mcal 1.54 Mcaltreatment/parameters NEl/kg ad NEl/kg ad NEl/kg NEl/kg ad NEl/kg ad

libitum libitum restricted libitum libitum

Prepartum bodycondition, scale 1-5 3.04 3.16 2.94 3.25 3.25Milk, kg/d 39.4 36.9 37.0 43.3 48.5Fat, % 3.59 3.77 3.58 3.65 3.62Liver TG, % or 2.5 2.6 1.4 3.6 3.2µg/µg DNANEFA, µEq/L 786 792 627 393 461BHBA, mg/dL 8.1 9.0 6.6 6.4 7.8Total health 29 51 37 57 52disorders

1Dann et al. (2006): wk 1-8 postpartum for milkparameters and health disorders and d 1-10 for bloodand liver measurements. Silva-del-Río et al. (2007):wk 1-15 for milk parameters and health disorders, wk1-10 for blood measurements, and d 1 and 35postpartum for liver TG.2Prepartum body condition, P = 0.003; Liver TG,P = 0.14; BHBA, P = 0.03; other parameters P ≥ 0.15or insufficient animals for statistical analysis (healthdisorders).3Milk, P = 0.04; NEFA, P = 0.06; BHBA, P = 0.07; otherparameters P ≥ 0.15 or insufficient animals forstatistical analysis (health disorders).

52

Table 2. Correlation coefficients between energybalance and fat-corrected milk or net energy intakefrom weekly means of 49 multi- and 24 primiparouscows from 2-21 wk postpartum (Grummer andRastani, 2003).Item Fat-corrected milk Net energy intake

r P r PAll Cows -0.26 < 0.0001 0.58 < 0.0001Primiparous -0.15 0.001 0.75 < 0.0001Multiparous -0.33 < 0.0001 0.69 < 0.0001

Figure 1. Energy status of dairy cows during the dryperiod and early lactation.

Figure 2. Milk yield of cows fed diet containing 1.67(High - H) or 1.74 Mcal NEl/kg DM (Low - L) for thefirst 3 wk after calving. After that, all cows were fedH (Rabelo et al., 2003).

Figure 3. β-hydroxybutyrate concentrations in cowsfed standard (n) or high NFC (n) diets beginningprepartum and continuing postpartum (Minor et al.,1998). NDF/NFC content of pre- and post-fresh dietswere 49/24 and 26/42 for standard and 30/44 and22/47 for high NFC.

53

Take Home Messages• Dairy producers must procure forages with dry

cow nutrient requirements in mind.• Optimal forage combinations should be;

moderate in energy density, avoid mineralimbalances, and be free of anti-nutritional factors.

• Moderate energy diets for dry cows may improvepostpartum cow health.

• Feeding management of dry cow diets is criticallyimportant to the success of the dry cow program.

IntroductionResearch focusing on dry cow nutrition andmanagement has intensified over the past 10 years.It is generally accepted that nutritional managementin the dry period affects metabolic status in thesubsequent lactation. Moderating changes in energybalance and body fat mobilization in early lactationimproves cow health. Feeding strategies that avoidexcess consumption of energy during the dry periodand minimize body fat mobilization after calving willlikely optimize success in transition cows.

A dry cow requires about 15.5 Mcal NEL/day to meetenergy demands for maintenance and pregnancy.Nutritionists are challenged with packaging balanceddiets that can be consumed at an ad libitum ratewithout greatly exceeding energy requirements. Thischallenge is further compounded by today’s dairycow’s intense drive for feed intake and continuedagronomic improvements and forage digestibility.Moderate energy diets fed during the dry periodshow promise in reducing postpartum metabolicdisorders such as ketosis, fatty liver and displacedabomsum (Rukkwamsuk et al., 1999; Beever, 2006;Douglas et al., 2006). These moderate energy dietscontain low energy forages, such as wheat straw, todilute the dietary energy density and preventoverconsumption of energy (Drackley and JanovickGuretzky, 2007). Wheat straw has been successfullyused in this regard, but cost and availability maymake alternative forages more attractive in someareas.

The default forage option for dry cows hashistorically been the least expensive forage on thefarm that often possessed a low nutritive value.Dairy producers strive to optimize quality in theforages they harvest or purchase for lactating cows.Dairy producers should, however, purchase orproduce forages with dry cow nutrient requirementsin mind as well. Optimal forages for dry cowsshould be modest in energy density, high in quality,and have an appropriate mineral profile.

Altering Forage Content of Dry Cow DietsLimited work has compared forage type and amountfed to dry cows on post-partum health andperformance. Johnson and Otterby, (1981), fed drycows one of three diets; 1) alfalfa-grass hay only, 2)41% corn silage, 47% alfalfa silage , and 12% highmoisture corn, and 3) 25% corn silage, 28.5% alfalfasilage, and 46.5 % high moisture corn). Theseconsiderably different diets fed during the dry periodhad little effect on milk production, postpartum drymatter intake (DMI), or the incidence of milk feverand ketosis. Cows assigned to the high-grain diet hadthe highest decline in DMI prior to calving while thehay fed cows had the most consistent prepartumintake. Coppock et al. (1972) fed diets varying inforage to concentrate ratios 1) 75:25, 2) 60:40, 3) 45:55,4) 30:70, 28 d preprepartum and observed that cowsfed treatments 2, 3, and 4 had a significant depressionin DMI compared to cows in treatment 1. Energyintake for all four diets exceeded NRC requirements.Dry cows fed high forage diets ate less dry matterprepartum, had a tendency to eat more dry matterpostpartum than dry cows fed a low forage, higherenergy totally mixed ration (TMR) (Hocomb et al.,2001).

Moderate energy, high forage dry cow diets havereceived considerable interest over the last 10 years.The Illinois group has fed far-off dry cows dietscontaining 26% wheat straw (Dann et al., 2006) or31.8% wheat straw (Janovick-Guretzky et al., 2006)resulting in improvements in post-partum energyand indicators of cow health.

54

Moderate Energy Diets and Forage Optionsfor Dry Cows

Noah B. LitherlandUniversity of Minnesota

Department of Animal ScienceCollege of Food, Agricultural, and Natural Resource Sciences

155B Haecker Hall1364 Eckles Avenue

St Paul, MN [email protected]

To the author’s knowledge, experiments comparingforage types for dry cows in diets of equivalentnutrient composition have not yet been published.

Why is forage selection for dry cowsimportant?Forage selection for dry cows is often overlooked.Dry cow forage selection criteria often include price,availability, and convenience. However, let us notforget that forage type and processing affect;palatability, intake, digestibility, rate of passage,fermentation, feeding behavior, and have positiveand negative associative effects with otheringredients. These factors will have clear effects onthe success of the dry cow feeding program.

Factors to consider when selecting forages for a drycow program include; selecting a blend of foragesthat closely meet nutrient requirements,consideration of on farm agronomic factors that allowfor consistent production of forages, balance betweenforage production with nutrient management plan,make use of existing facilities and equipmentwhenever practical, have versatility of use in otheraspect of the farm such as inclusion in lactating cowand heifer diets, or use as bedding.

Representative and accurate forage sampling andanalysis for nutrient profile is critically important toaccurately formulate dry cow diets. Wet chemistryanalysis is suggested to determine mineral profile foraccurate calculation of dietary-cation anion difference(DCAD). Additionally, frequent dry mattermeasurement and adjustment will improve accuracyof mixing and delivery of diets.

Alternative forage options for dry cows must beexplored to provide forages with appropriate nutrientprofile at reduced costs. Options may include; wheat,oat, or barley straw, corn silage, alfalfa, grass hay,corn stalks, soybean stubble, sorghum silage,sorghum sudan grass, tropical corn silage, and forgesfor use in bio-energy production. Agronomicpractices such as harvesting at a mature stage andvarying fertilizer application rates are importantaspects of production of dry cow forages. Adequateprocessing, inclusion into a TMR, and feedingstrategy may be as important as the type of foragebeing fed.

Potential dry cow forage options Small grain strawWheat straw appears to be the gold standard todilute the energy density of dry cow diets.

Advantages: Very low energy density, slow rate ofpassage, excellent bulk, palatable, low potassium andcalcium, may help reduce moisture in TMR’s

containing numerous wet ingredients, readilyavailable in most areas, consistent nutrient profile,has received the most interest from researchers. Fieldexperience suggests reduced sorting of dry cowTMR’s that contain less than 30% wheat straw andwheat straw should be processed to result in 50% ofstraw in the bottom box and minimize straw found inthe top box (3-box system). Results of excessiveprocessing of straw on rate of passage are unknown.

Disadvantages: Must be adequately processed toprevent sorting. Wheat straw is a dry ingredient, sowater usually needs to be added to the TMR toachieve 50% DM.

Oat straw and barley straw can work well to dilutedietary energy density as well, although both are lessreadily available and nutrient composition appears tobe more variable. Oat straw does not chop as readilyas wheat straw, is higher in potassium, and has aslightly greater rate of in vitro digestibility. Barleystraw appears to be quite similar to wheat straw, butis less available in most areas.

Corn silageCorn silage works well in combination with wheatstraw to form a palatable, moderate energy, moderatepotassium, combination that is palatable. Varieties ofcorn silage that are modest in starch content andidentified as less optimal varieties for lactating cowsmay serve well in dry cow diets (Table 1). Forexample, assuming equal growing conditions, cornsilage varieties A and B are considerably higher instarch and energy content than varieties C and D.The moderate energy density of varieties C and Dmay make them better suited for dry cows as lessenergy dilution would be required compared withvarieties A and B. Producers must continue tofocus on length of chop, dry matter, and properensiling to produce a quality product that is free ofanti-nutritional factors such as mold and high ashcontent.

Advantages: Low protein, potassium and calcium,highly palatable, adds moisture to the TMR, low costingredient.

Disadvantage: May contribute excessive starch andenergy, low bulk, sortable ingredient, poorfermentation and low quality corn silage is disruptiveto rumen fermentation, total tract digestion, and maycompromise immune function.

Cool season grass hayGrass hay is readily available in most areas and hasagronomic advantages for manure application. Fieldexperience suggests that cool season grassesincluding; timothy, orchardgrass, reed canarygrass,

55

smooth brome grass, and tall fescue might work wellin dry cow diets, but research should confirm this.

Advantages: Readily available in the Midwest, lowcost, moderate protein and energy content, lowcalcium, high fill factor, and palatable.

Disadvantages: Wide variability in nutrientcomposition depending upon variety, stage ofmaturity at harvest, and harvesting technique.Obtaining an accurate sample is often difficult due toheterogeneity of plant species and stages of maturity.Tends to be high in potassium, difficult to process toappropriate particle size, higher energy and proteincontent limits grasses role to dilute the energydensity of the diet compared with straw. Whencompared to wheat straw, grass hay tends to behigher in energy density, lower in NDF content, andmore digestible. All of these factors result in greaterrates of passage and fewer constraints to rumen fillon energy intake. Additionally, variability in mineralcontent, such as high potassium, may limit inclusionrates of some grasses in dry cow diets. High levels ofpotassium are associated with higher dietary cation-anion difference which places periparturient cows atrisk for milk fever.

Corn stalksCorn stalks are one of the most readily available andperhaps underutilized forage sources in the uppermid-west. Corn stalks, if processed adequately, seemto work modestly well for producers during thewinter. Moisture accumulation in bales with thespring thaw and risk of mold growth may precludetheir use in dry cow diets in spring and summermonths. Some new round balers are equipped withrapid pick-up feeding systems that allow forharvesting a greater percentage of corn stalk residues.Harvesting corn stalks closer to the ground mayallow for more dirt to be incorporated into bales andincrease ash values. Producers who store baledforages inside have an advantage over those storingforages outside as quality and shrink lossesassociated with moisture and weather are reduced.

Advantages: Similar to wheat straw nutrientcomposition. Beneficial attributes include; low starchand sugar content, high fiber, low potassium andcalcium, excellent bulk, and palatable if finelyprocessed.

Disadvantages: High ash content, difficult to processto appropriate particle size to prevent sorting, maycontain mold depending upon harvest and storageconditions, and low moisture content.

Soybean stubble hayAdvantages: low cost, low starch and sugar content,readily available.

Disadvantages: Difficult to process to preventsorting, low yield (ton/acre), tends to be higher incalcium and ash than wheat straw.

Sorghum silageAdvantages: Lower starch content than corn silage,adds moisture, palatable, low sortability, and has avariety of applications including lactating cow andheifer diets.

Disadvantages: Lower starch digestibility, higherpotassium than corn silage, less grown in the upperMidwest than in the Southwest.

Sorghum-sudan grassAdvantages: Moderate energy density, high yielding(slightly lower than corn when harvested as silage).Sorghum-sudan grass can be cut 2-3 times per season.Quality declines rapidly with maturity resulting inmoderate energy forage.

Disadvantages: Dry hay production may be achallenge as thick stems increase drying time.Quality decreases rapidly with maturation. Largestems may be difficult to process and be sortedagainst. Variability in nutrient composition is achallenge. Sorghum-sudan grass may require specialequipment for harvesting.

Tropical corn silageAdvantages: Low starch, excellent yield, palatable.

Disadvantages: Limited research available, seed maybe a challenge to obtain, and slow dry down timeresults in late harvest.

Switchgrass Advantages: Similar to wheat straw in nutrientcomposition. Switchgrass may become readilyavailable for use as a renewable source, with dry cowforage as a diverse application.

Disadvantages: Limited data on nutrientcomposition. Large stem may be difficult to processand palatability may be a problem.

Sorting ChallengesThe limited demand on the daily schedule of a drycow may lead to amplified sorting behavior. Dietsorting leads to intake of inconsistent nutrient intake(Stone, 2004; DeVries et al., 2007). Cows that sortagainst long particles are likely to consume higheramounts of grain and lower amounts of fiber,resulting in intake of a more nutrient dense diet than

56

planned. Additionally, sorting of the TMR can reducethe nutrient density of the TMR remaining in the feedbunk, particularly if feed management is less thanideal.

Bulky forages such as wheat straw, grass hay, andcrop residues have inherent properties thatpredispose diets containing these ingredients tosorting. Appropriate reduction of particle size toprevent sorting is a challenge. Vertical mixers andtub grinders vary in their ability to adequatelyprocess long forages. Adding water to the diet toreduce the dry matter to 50 percent appears to reducesorting. Additionally, liquid supplements also helpreduce sorting. Consistent processing of ingredients,appropriate mixing and delivery of the TMR, andoptimal bunk management are important factors toconsider ensuring consumption of the diet that isintended. In addition to feeding management, cowfactors such as overcrowding and mixingprimiparous and multiparous cows should beconsidered

Do bulky forage offer advantages inmodifying feeding behavior?We know that dairy cows undergo manyphysiological and behavioral changes around thetime of calving that affect feed and energy intake.Factors regulating feed intake are due to complexinteractions between the cow and her environment.Increasing feed intake after calving is critical to thesuccess of the fresh cow. Cows are habitual creatures,and taking advantage of habit forming behaviors,such as feeding behavior, may confer an advantage inperiparturient dairy cows.

Lactating cows spend about 5.0 h per day feeding(Grant and Albright, 1995). Dry cows fed at an adlibitum rate (feed was available for 24 h per d) spent3.7 h per d eating at the feedbunk (Albright andPenninton, 1984). Perhaps increasing the amount oftime dry cows spend eating to mirror that of lactatingcows may be beneficial. Greter, (2008) fed dietscontaining increasing amounts of wheat straw togrowing heifers and found that DMI linearlydecreased with increasing amounts of straw wherefeeding time and meal duration increased.

Bulky forages that require greater eating andrumination time may help cows become accustomedto dedicating greater amounts of time during the dayto feeding. This behavior, if maintained, mayimprove postpartum intake and be one lessadaptation that a fresh cow must make to herenvironment. Rumen stretch and fill may also beadvantageous in dry cows fed diets containing bulkyforages throughout the dry period. Although therumen can adapt quickly to consumption of larger

amounts of feed (Allen et al., 2000), rapid increases infeed intake after calving are critically important tominimize the severity of energy balance.

Dietary modification to increase postpartum energyintake has been evaluated, yet changes in energyintake are mediated through changes in feedingbehavior. Only limited research has evaluated theeffects of transition on feeding behavior. Huzzey etal. (2005) observed feeding, drinking and standingbehavior in transition cows and observed a tendencyfor the average number of meals per day to be higherafter calving, but time spent eating declined from 87to 62 min/d after calving. Cows in this study werefed a TMR containing 48% corn silage, 26% alfalfahay, and 26% protein mineral supplement (CP, 15.0%,ADF 20.0%, NDF, 33.5%). Increased time spentfeeding and ruminating preprartum might beexpected if cows were fed a diet higher in fiber. Itwould be interesting to evaluate the effects ofprepartum feeding behavior when cows are fed highfiber diets on postpartum feeding behavior andenergy intake.

Importance of moderate potassium forages.Forage potassium concentration is dependent uponforage species, maturity, and soil type andfertilization. Most dairy producers recognize theimportance of selecting forages that are low inpotassium for dry cows to reduce dietary cation-anion difference (DCAD) and prevent milk fever.Hypocalcemia is associated with reduced DMI andmilk production and increased risk for displacedabomasum, delayed closing of the teat sphincter,compromised immune function, retained placenta,and metritis (Curits et al., 1983; Duffield et al., 2005;Kimura et al., 2006).

Diets with a low DCAD should be fed to cows threeweeks prior to calving to prevent hypocalcemia andmilk fever. Researchers in Canada evaluated fivespecies of cool season grasses (orchardgrass, meadobromegrass, tall fescue, smooth bromegrass, andtimothy) harvest twice per year during two growingseasons to evaluate mineral concentration (Tremblayet al., 2006). The five species had respectively,DCAD’s of 656, 540, 510, 490, and 384 mmol/kg-1/DM in spring growth, and 633, 569, 496, 447, and332 mmol/kg-1/DM in summer regrowth. Speciesdifferences in DCAD were primarily related todifferences in potassium (K) concentration. Theauthors concluded that timothy is the best suitedforage for dry cows. Additionally, work in New Yorkalso confirmed timothy as a low accumulator of Kunder varying fertilizer regimes (Cheney andCherney, 2005).

57

Forage potassium management The goal of K management program for cool seasongrasses for dry cows is to provide optimum K forplant functions without accumulating excessive plantK. Perennial grasses are luxury consumers of K,resulting in high K forages grown on fields withexcess soil K due to repeated animal manureapplications (Cherney et al., 1998). Most foragesproduced with adequate K fertility will not containexcessive K for lactating cows, however, foragescontaining greater than 2.0% K may put dry cows atrisk for hypocalcemia and milk fever. A rule ofthumb for dry cow forages is to keep forage K below1.5% of dry matter, especially when feeding highforage dry cow diets. It is important for dairyproducers to know the level of soil K in each field.Fields testing lowest in soil K can be set aside forproducing forage for dry cows.

Forage capacity to remove K is high, so soil test levelwill decline with time resulting in deficient soil K,which will decrease yields and persistency of thestand. The C4 plant species generally have lower Klevels than C3 species. Grasses are typically lower inK than legumes. The concentration of K in planttissue declines with increasing maturity. Hay istypically lower in K than silage due to leaching of Kduring drying. When soil K is high, higher rates ofnitrogen fertilizer tend to increase forage K.

Canadian researchers recently grew timothy hay onlow potassium soils and fertilized with two levels ofcalcium chloride (CaCl2) to produce low DCAD hayat two different levels (Charbonnear et al., 2008).Calcium chloride solution was applied with a pivotirrigation system. The DCAD values were 1.6 vs. 14.5mEq/100g of DM for the low-and high-DCADtimothy based diets, respectively. Dry cows fed thelow DCAD timothy hay had greater blood ionizedcalcium concentration prepartum, and at 0 and 8 hafter calving, and similar prepartum DMI. Cowswith ionized calcium concentration in blood <1 mMwere considered to be hypocalcemic and were fewer(35.0%) in the low DCAD group compared with thehigh DCAD group (66.7%) (Charbonnear et al., 2008).Despite timothy’s apparent advantage as forage fordry cows it is better suited for cold climates. TheCaCl2 fertilization of additional forages should beexplored to include other cool and warm seasongrasses and legumes. Diverse avenues forapplication of CaCl2 such as mixing CaCl2 solutionwith liquid manure should also be explored.

Restricted vs. ad libitum feeding. Wheredo high forage dry cow diets fit?Optimal dry cow feeding management strategiesremain controversial. Inconsistencies in dry cowfeeding programs include; amount of dry matter per

cow fed, time of feeding, access time to feed, feedpush-up frequency and bunk space per cow.Restricted feeding during the dry period may offerseveral advantages such as; control of energy intakebeyond dietary dilution, aggressive feeding behavior,reduced sorting (cows will clean up all feed), andreduced rate of passage and improvement in nutrientdigestibility. Possible disadvantages of restrictedfeeding of dry cows include; increased competition atfeeding time, reduced rumen fill, and slug feedingbehavior. Energy intake is routinely restricted inother species prior to parturition to improvepostpartum success. Questions remain regardingoptimal feeding management of dry cows.

Restricted fed cows during the dry period oftenperform as well if not better than cows fed ad libitumduring the dry period. Restricted feeding of diets todry cows also shows promise in improvedpostpartum dry matter intake and lower livertriglyceride concentrations (Dann et al., 2006;Douglas et al., 2006). In a study conducted inFlorida, restricted feed intake compared with adlibitum intake during the dry period had minimaleffects on postpartum performance; however, cowswith prepartum feed restriction tended to eat moredry matter in early lactation (Holcomb et al., 2001).Effects of ad libitum or restricted feeding of dry cowdiets containing different forage types on postpartumhealth, performance, and feeding behavior has notyet been evaluated. Lodge et al. (1975) restricted hayintakes for 6 wk prepartum to supply onlymaintenance energy versus hay at approximately 1.8times maintenance. Restricted cows lost more weightprepartum, but over wk 1 to 16 postpartum hadhigher DMI and greater milk yield with lower milkfat content and restricted cows tended to eat moreDM early in lactation (Lodge et al., 1975).

Feed cost for dry cows fed in restricted amountswould be lower than ad libitum cows. Additionally,restricted cows would produce less manure with ahigher dry matter content, resulting in cleanerbedding in dry cow facilities. Modest improvementsin environmental impact through reduced nutrientexcretion may be observed in restricted feedingprograms.

Dry cows and cellulosic biofuel?Forages discussed above as options for dry cowshave also received some interest as substrate forrenewable energy production. Research is needed tocontinue to explore forage options for dry cows ascompetition for cellulosic products and price forthese commodities may increase in the future.

58

ReferencesAlbright, J. L. and J. A. Penninton. 1984. The influence of

diet upon time of calving and behaviour in dairy cattle.In: J. Unshelm, G. Van Putten, and K. Zeeb (Ed.) Proc.Int. Congr. Appl. Ethology in Farm Animals. P 184.Kiel, Germany.

Allen, M. S. 2000. Effects of diet on short-term regulationof feed intake by lactating dairy cattle. J. Dairy Sci.83:1598-1624.

Cherney, J. H., D. J. R. Cherney, and T. W. Bruulsema. 1998.Potassium management. p 147-148. In J. H. Cherneyand D. J. R. Cherney (ed.) Grasses for dairy cattle. CABInt. Wallingford, UK.

Cherney, D. J. R., J. H. Cherney, and L. E. Chase. 2004.Lactation performance of Holstein cows fed fescue,orchardgrass, or alfalfa silage. J. Dairy Sci. 87:2268-2276.

Cherney, J and D. Cherney. 2005. Agronomic response ofcool-season grasses to low intensity harvestmanagement and low potassium fertility, Agron. J.97:1216-1221.

Curtis, C. R., H. N. Erb, C. J. Sniffen, R. D. Smith, P. A.Powers, M. C. Smith, M. E. White, R. B. Hillman, and E.J. Pearson. 1983. Association of parturienthypocalcemia with eight periparturient disorders inHolstein cows. J. Am. Vet. Med. Assoc. 183:559-561.

Dann, H. M., N. B. Litherland, J. P. Underwood, M. Bionaz,A. D’Angelo, J. W. McFadden, and J. K. Drackley. 2006.Diets during the far-off and close-up periods affectperiparturient metabolism and lactation in multiparouscows. J. Dairy Sci. 89:3563-3577.

Dann, H. M., M. P. Carter, K. W. Cotanch, C. S. Ballard, T.Takano, and R. J. Grant. 2007. Effect of partialreplacement of forage neutral detergent fiber with by-product neutral detergent fiber in close-up diets onperiparturient performance of dairy cows. J. Dairy Sci.90:1789-1801.

DeVries, T. J., K. A. Beauchemin, and M. A. G. vonKeyserlingk. 2007. Dietary forage concentration affectsthe feed sorting behavior of lactating dairy cows. J.dairy Sci. 90:5572-5579.

Dien, B. S., H. J. G. Jung, K. P. Vogel, M. D. Casler, J. F. S.Lamb, L. Iten, R. B. Mitchell, G. Sarath. 2006. Chemicalcomposition and response to dilute-acid pretreatmentand enzymatic saccharification of alfalfa, reedcanarygrass, and switchgrass. Biomass and Bioenergy30:880-891.

Douglas, G. N., T. R. Overton, H. G. Bateman II, H. M.Dann, and J. K. Drackley. 2006. Prepartal plane ofnutrition, regardless of dietary energy source affectsperiparturient metabolism and dry matter intake inHolstein cows. J. Dairy Sci. 89:2141-2157.

Drackley, J. K. and N. A. Janovick Guretzky. 2007.Controlled energy diets for dry cows. Western dairymanagement conference proceedings. 7-16.

Duffield, T., S. L. E. Blanc, and K. Leslie. 2005. Impact ofsubclinical metabolic disease on risk of early lactationculling. J. Dairy Sci. 88(Suppl. 1): 199-200 (Abstr.).

Grant, R. J. and J. L. Albright. 1995. Feeding behavior andmanagement factors during the transition period indairy cattle. J. Anim. Sci. 73:2791-2803.

Greter, A. M., T. J. DeVries, and M. A. G., von Keyserlingk.2008. Nutrient intake and feeding behavior of growingdairy heifers’ effects of dietary dilution. J. Dairy Sci.91:2786-2795.

Holcomb, C. S., H. H. Van Horn, H. H. Head, M. B. Hall, anC. J. Wilcox. 2001. Effects of prepartum dry matterintake and forage percentage on postpartumperformance of lactating dairy cows. J. Dairy Sci.84:2051-2058.

Janovick Guretzky, N. A., N. B. Litherland, K. M. Moyes,and J. K. Drackley. 2006. Prepartum energy intakeeffects on health and lactational performances inprimiparous and multiparous Holstein cows. J DairySci. 89 (Suppl. 1). (Abstr.).

Kimura, K., T. A. Reinhardt, and J. P. Goff. 2006.Parturition and hypocalcemia blunts calcium signals inimmune cells of dairy cattle. J. Dairy Sci. 89:2588-2595.

Johnson, D. G. and D. E. Otterby. 1981. Influence of dryperiod diet on early postpartum health, feed intake,milk production, and reproductive efficiency ofHolstein cows. J. Dairy Sci. 64:290-295.

Lodge, G. A., L. J. Fisher, and J. R. Lessard. 1975. Influenceof prepartum feed intake on performance of cows fedad libitum during lactation. J Dairy Sci. 64:290-295.

Stone, W. C. 2004. Nutritional approaches to minimizesubacute ruminal acidosis and laminitis in dairy cattle.J. Dairy Sci. 87:E13-E26.

Tremblay, G. F, H. Brassard, G. Belanger, P. Seguin, R.drapeau, A. Bregard, R. Michaud, and G. Allard. 2006.Agronomy Journal 98:339-348.

59

Table 1. Selected varieties (A-D) from the Minnesota corn silage varietal trial results (2008).

Yield, Ton/Acre Quality (concentration), % Milk YieldRelative Moisture, % DM Silage CP NDF IVD NDFD Starch lb/Ton lb/AcreMaturity

A 104 64.8 8.6 24.5 9.0 35 82 48 34 3,610 31,200B 101 63.0 7.3 19.6 8.8 37 79 45 32 3,300 24,200C 100 65.1 7.1 20.2 9.5 43 78 49 21 3,050 21,700D 97 66.2 6.6 19.5 9.7 46 78 51 19 3,000 19,900

IntroductionThis paper will focus on dairy feed management withnitrogen as a focus while : 1) providing an overviewof the importance of feed management and it’srelationship with whole farm nutrient management,and 2) share a few examples of how feedmanagement can reduce the import of nitrogen to thefarm.

Whole Farm Import of Nutrients Figure 1 depicts the concept of whole farm nutrientmanagement. From an environmental standpoint,ideally the input would equal the output from thefarm. This is rarely the case because only ~ 13 to 27 %of feed input of N, P, and K are exported in milk andanimals (Figure 2). The remainder of the N, P, and Kare excreted. The import/export imbalance is furtherimpacted by the increase in cow density at thefarmstead. From 1954 to 1987 there was a continualincrease in cow density on dairy farms across the US(Lanyon, 1992). Coincident with this increase in cowsper acre was an increased importation of feedstuffs tothe farm to achieve higher levels of milk production.Data shown in Table 1 indicate that the amount ofconcentrate (assumed to be imported) fed to dairyfarms increased ~ 10 to 50 fold between 1954 and1987.

Recent Changes in Dairy ProductionThe trend for increased herd size and fewer herds hasprogressed to the point in the US where in 2004, 50percent of the dairy cows were owned by 7.5 percentof the dairies (Cady, 2005). The reason for thecontinued restructuring of the dairy industry is theeconomic pressure at the farm level, since milk priceat the farm has not kept up with inflation (Fetrow etal., 2004). The continued decrease in profit per cowhas resulted in a need for increased efficiency andone of the most effective ways to improve efficiencyhas been to increase the number of cows per unit ofland (Rotz et al., 1999). It is common to observe anincrease in imported feed nutrients to the farm due tothe increase in number of cows per management unit(farm), and the need for increased nutrients as cowsachieve higher levels of milk production Lanyon,1992). In addition, inexpensive by-product feeds areoften imported which are high in N and P. These

changes accumulatively result in an increase innutrient import to most farms, without an equivalentexport of nutrients in milk.

While meeting the challenges of managing nutrientson dairies of today is necessary, it is important torecognize the overall decrease in national nutrientloading that has occurred over the last 60 years.Kohn (2004) reported (see Table 2) that in 1944 the USdairy herd peaked at 25 million cows, since thenthere has been a decline to 9 million cows in 2001.Since 1944, N excretion per cow per year hasincreased by about 12%; however, the total Nexcreted by all dairy cows in the US has decreased by60%.

Feeding for Reduced Crude Protein The transition from feeding the dairy cow for hercrude protein requirement has clearly progressedtoday to a more sophisticated approach offormulating for the estimated requirement of aminoacids (NRC Recommendation for Dairy Cattle – 2001- http://bob.nap.edu/books/0309069971/html/).While this transition has been occurring there hasbeen a simultaneous progression of a greaterawareness of the interrelationship of diet formulationand feed management on whole farm nutrientmanagement. The focus of this example will be todevelop the concept of ration balancing for increasedprofit and reduced environmental impact as it relatesto nitrogen. In particular, the merits of formulatingfor estimated amino acid requirements with the useof ruminally undegraded protein (RUP) sources.

Amino Acid FormulationAmino acid formulation for dairy cattle has beencommon practice since the availability of the CNCPS(Fox et al., 1990) model and CPM model. We haveused both models successfully to strategicallyformulate diets to evaluate the merits of sources ofRUP, ruminally protected amino acids, and freelysine-HCL (Xu, et al.,1998; Harrison, et al., 2000).Others (VonKeyeserlingk et al., 1999; Dinn et al.,1998) have had positive experiences with use of themodel to formulate diets to reduce the CP level in thediet while maintaining milk productivity.

60

Managing Nitrogen for Profit andStewardship

Joe HarrisonWSU-Puyallup Research and Extension Center

2606 West PioneerPuyallup, WA [email protected]

Additional studies (Harrison et al., 2002, andHarrison et al., 2003) continue to provide evidencethat formulating diets for available amino acids canprovide the opportunity to reduce CP levels in thediet and reduce on-farm import of nitrogen. A fieldstudy (Harrison et al., 2002) was conducted with ahigh producing herd in WA state to compare theirgeneral herd diet formulated at ~ 18 % CP to a dietthat was reformulated at ~ 17 % CP (Tables 3 and 4).Results showed that milk production could bemaintained while decreasing nitrogen import to thefarm (Tables 5 and 6). In addition, the dietreformulation resulted in an increase in IOFC(Table 7).

One of the primary environmental benefits of moreclosely meeting the protein and amino acid needs ofheifers and lactating cows is a reduction in potentialammonia emissions. James et al. (1999) fed diets of9.6 % or 11% CP to dairy heifers and observed adecrease in ammonia emission of 28.1%, anddecreased urea-N, total N, and percentage N excretedin urine of 29.6, 19.8, and 7.4%, respectively. Krober etal. (2000) fed diets that ranged from 12.4 % to 17.5 %CP to early lactation cows, and observed an increasein use of feed N for milk N from 27% to 35 % as thedietary CP percentage was reduced from 17.5% to12.4%, respectively. Krober at al. (2000) also observedthat ammonia N emission decreased from 231 to 160and 55 micrograms/sec per square meter of surfacearea when feeding diets of 17.5 %, 14.7 %, and 12.4 %CP, respectively. This translated to a decrease in totalnitrogen losses during seven weeks of storage of 89 to57 to 25 g/d per cow. While expensive, the on-farmmanagement strategy that has the greatest potentialto minimize ammonia N volatilization is thesegregation of feces and urine (Rotz eta al., 2006).This strategy does result in an estimated range inreduced annual net return of $64 to $88/cow wasobserved.

While individual feeding studies are necessary todefine the value of any given feeding strategy,integration of these practices with other farmmanagement factors and their impact at the wholefarm level is the ultimate goal of integrated nutrientmanagement. Rotz et al. (1999) used a simulationmodel to demonstrate the economic andenvironmental impact at the whole farm scale ofmore efficient feeding and use of proteinsupplements for milk production. When soybeanmeal was used along with a less ruminallydegradable protein source, volatile N loss was reducefrom 30 to 11.6 lb/acre of cropland, while leachingloss was estimated to be reduced by ~ 0.89 lb/acre.Use of the more expensive rumen undegradableprotein source increased net return by $46 to$69/cow per year. In this same report (Rotz et al.,

1999) less N was volatilized when a majority of theforage was based on corn silage vs. alfalfa silage.

SummaryThe society of today is increasingly expecting thatanimal agriculture has a minimal impact on theenvironment. Tools and knowledge (such as precisionnitrogen feeding) are available to successfully reducethe import of nitrogen to the farm while resulting inimproved profitability.

ReferencesCady, R. 2005. Dairy – consolidation or reorganization?

Dairy Herd Management. P. 8. June.Dinn, N.E., J. A. Shelford, and L. J. Fisher. 1998. Use of the

Cornell Net Carbohydrate and Protein System andrumen-protected lysine and methionine to reducenitrogen excretion from lactating cows. J Dairy Sci.81:229-237.

Fetrow, J., R Cady, and G Jones. 2004. Dairy productionmedicine in the United States. The Bovine Practitioner.38:(2) 8.

Harrison, J H, D Davidson, L Johnson, M L swift, MvonKeyserlingk, M Vazquez-Anon, and W Chalupa.2000. Effect of source of bypass protein andsupplemental Alimet and lysine-HCL on lactationperformance. J Dairy Sci 83(suppl 1):268.

Harrison, J., L. Johnson, D. Davidson, J. Werkhoven, A.Werkhoven, S. Werkhoven, M. Vazquez-Anon, G.Winter, N. Barney, and W. Chalupa. 2002. Effectivenessof strategic ration balancing on efficiency of milkprotein production and environmental impact. J. DairySci. 85:205 (Suppl. 1).

Harrison, J H, R L Kincaid, W Schager, L Johnson, DDavidson, L D Bunting, and W Chalupa. 2003. Strategicration balancing by supplementing lysine, methionine,and Prolak on efficiency of milk protein production andpotential environmental impact. J Dairy Sci. 86:60(Suppl 1).

Hart, J., M Gangwer, M Graham, and E Marx. 1996. Dairymanure as a fertilizer source. EM 8586. OR State Univ.Ext Serv.

James, T., D Meyer, E Esparza. EJ DePeters and H Perez-Monti. 1999. Effects of dietary nitrogen manipulationon ammonia volatilization from manure from Holsteinheifers. J Dairy Sci. 82:2430-2439.

Klopfenstein, T., R Angel, G. Cromwell, G. Erickson, D, Fox,C. Parsons, L. Satter, and A. Sutton. 2002. Animal dietmodification to decrease the potential for nitrogen andphosphorus pollution. CAST Issue Paper # 21. July2002.

Kohn, R. 2004. Use of animal nutrition to manage nitrogenemissions from animal agriculture. Second AnnualMid-Atlantic Nutrition Conference. Mar. 24-25,Timonium, MD, pp.25-30. University of Maryland,College Park.

Krober, TF, DR Kullimng, H Menzi, F Sutter, and MKreuzer. 2000. Quantitative effects of feed proteinreduction and methionine on nitrogen use by dairycows and nitrogen emission from slurry. J Dairy Sci.83:2941-2951.

Lanyon, L. 1992. Implications of dairy herd size for farmmaterial transport, plant nutrient management, andwater quality. J Dairy Sci. 75:334-344.

Nelson, C J. 1999. Managing nutrients across regions of theUnited States. J Dairy Sci. (suppl 2):90-100.

61

62

NRC. 2001. National Research Council. Nutrientrequirements of dairy cattle. Seventh Revised Edition.National Academy Press, Washington, D.C.

Rotz, C A., L D Satter, D R Mertens, and R E Muck. 1999.Feeding strategy, nitrogen cycling, and profitability ondairy farms. J Dairy Sci. 82:2841-2855.

Rotz, CA, J Oenema, and H van Keulen. 2006. Whole farmmanagement to reduce nutrient losses from dairyfarms: A simulation study. Appl. Eng. Ag. 22 (5): 773-784.

VonKeyserlingk, M. A. G., M. L. Swift, and J. A. Shelford.1999. Use of the Cornell Net Carbohydrate and ProteinSystem and rumen-protected methionine to maintainmilk production in cows receiving reduced proteindiets. Can. J. Anim. Sci. 79:397-400.

Xu, S., J.H. Harrison, W. Chalupa, C. Sniffen, W. Julien, H.Sato, T. Fujieda, K. Watanabe, T. Ueda, H. Suzuki. 1998.The Effect of Rumen-Bypass Lysine and Methionine onMilk Yield and Composition in Lactating Cows, J. DairySci. 81:1062.

Table 1. Changes in dairy farm numbers, cownumbers, and the concentrate consumed for three USdairy states from 1954 to 1987. Source: Lanyon (1992).

California1954 1987

No. dairy farms 34,031 3,631Milk cows 790,730 1,070,366Concentrate uselb/yr per cow 1,899 7,542lb/100 lb milk 23.98 42.02lb/yr per farm 43,747 2,223,069

Florida1954 1987

No. dairy farms 16,738 1,073Milk cows 158,877 176,993Concentrate uselb/yr per cow 3,216 9,469lb/100 lb milk 62.92 75.9lb/yr per farm 30,523 1,562,323

Pennsylvania1954 1987

No. dairy farms 82,708 15,096Milk cows 875,631 673,054Concentrate uselb/yr per cow 2,248 5,643lb/100 lb milk 35.9 40.0lb/yr per farm 23,793 251,123

Table 2. Production and nitrogen excretion for the USdairy herd in 1944 and 2001.

1944 2001Milk per cow (kg/d) 7 27N intake per cow (g/d) 360 490N excreted per cow (g/d) 326 364N excreted / N in milk (g/g) 10 3N in milk / N intake (g/g) 0.09 0.26Number of cows (106) 25 9Milk per cow (lb/yr) 4560 17934Total milk (109 lb/yr) 114 161N Excretion per cow (lb/yr) 262 293Total N excretion (109 lb/yr) 6.6 2.6Source: Kohn, R. (2004).

Table 3. Chemical composition for control and treateddiets (Harrison et al., 2002)Item Control TreatedCP, % DM 17.8 16.95Available CP, % DM 16.4 15.35Unavailable CP, % DM 1.4 1.55Neutral Detergent CP,% DM 2.3 2.65Adjusted CP, % DM 17.8 16.95Soluble Protein, % DM 6.4 6Soluble Protein, % CP 35.7 36.95ADF, % DM 22.55 22.65NDF, % DM 32.45 32.7NFC, % DM 39.05 39.8

Table 4. Composition of diets (Harrison et al., 2002).Item Control - % DM Treated - % DMAlfalfa Hay 29.32 26.23Corn Silage 19.55 19.99Corn grain, flaked 16.15 18.01Whole cottonseed 8.26 8.49Corn Distiller Grains 4.35 ----Beet pulp pellets 2.10 6.22Molasses 1.74 1.94Ener GII 1.48 .63Soybean Meal ---- 3.45Bakery Mix* 14.28 ----Bakery Waste ---- 7.97Soy Pass ---- 3.95Std Mineral/Vit 2.77 ----Std Minerals + Novus Premix** ---- 3.12*Bakery mix = Canola – 28.8% (as fed), soybean meal– 32.9% (as fed), and bakery waste – 32.8% (as fed).**contained Alimet and lysine HCL at a5.7% and 24%,respectively.

Table 5. Treatment response to diet reformulation(Harrison et al., 2002).Item Control Treated SE P<DMI,lb 56.7 55.2 ---- -----CP Intake, lb 10.1 9.35 ---- -----Milk, lb 99.9 101.9 0.53 .0073.5% FCM, lb 96.0 96.6 0.46 .32Fat, % 3.28 3.21 0.014 .001Milk Fat, lb 3.26 3.23 0.018 .63Protein, % 2.90 2.93 0.006 .0009Milk Protein, lb 2.88 2.95 0.015 .0004MUN, mg/dl 17.5 14.5 ---- ----Ratio Milk TrueProtein: IntakeProtein Ratio .285 .316 ---- ----BW, lb 1396 1395 1.80 .88Change in BW, lb 34 36 4.3 .70

Table 6. Environmental Characterization (Harrison etal., 2002).Item Control Treated % ChangeNitrogen Intake, gms/d 734 680 - 7.4Milk total N, gms/d* 240 246 + 2.5Predicted Urinary N, gms/d** 289 239 - 17.3Calculated Fecal N, gms/d*** 205 195 - 5.0*(Milk True protein - gms/6.38) X 1.17**Estimated per J Dairy Sci.85:227-233. Urinarynitrogen (gm/d) = 0.026 X BW (kg) X MUN (mg/dl)***Intake N- Milk N- Urine N

Table 7. Economic Evaluation. (Harrison et al., 2002).Item Control TreatedFeed Costs, $/day/cow 4.82 4.88Milk Income, $/day/cow 11.92 12.10IOFC*, $/day/cow 7.10 7.22*IOFC = Income over feed cost.

Figure 1. Schematic depicting the concept of wholefarm nutrient management. Ideally, inputs = outputs.

Source: Nelson (1999).Source: Hart et al. (1996).

63

IntroductionRequirements for choline and B vitamins have notbeen established. In the case of B vitamins, thedogma is that microbial synthesis and dietary sourcesescaping ruminal degradation provide adequatequantities and supplementation is not required.However, as milk production has increased, thatnotion has received additional scrutiny. Erdman(1992) indicated choline as potentially limiting fordairy cows. In contrast, he predicted that intestinalsupply of niacin was far in excess of that needed formilk production. Supplementation of free niacin todairy rations, commonly at a rate of 6 to 12 g/d, hasinconsistent and small effects on lactationperformance, and questionable economic return(Schwab et al., 2005). Limited data is available fordetermining the effects of protected niacin onlactation performance. Consequently, for niacin, thefocus of this paper will be on postruminal deliveryand its pharmacological effects on modifying lipidmetabolism.

CholineBackground. Choline is often referred to as a vitaminor micronutrient, but it does not fit the classicaldefinition. It can be synthesized endogenously, it isnot an enzyme cofactor in metabolic pathways, and itis typically supplemented in much larger quantitiesthan vitamins. Choline is an essential nutrient formany animal species including rats, pregnant sows,pre-ruminating sheep, and calves. Choline serves asa methyl donor in biochemical reactions and as aconstituent of phosphatidylcholine (PC). Methionineserves as a methyl donor for choline synthesis;therefore, choline and methionine can spare therequirement of each other. Involvement of biotin,folic acid, and vitamin B12 in one-carbon methylmetabolism yields potential interactions betweenthese vitamins and choline. Phosphatidylcholine canbe synthesized from tri-methylation ofphophatidylethanolamine or directly from choline.As a component of phospholipids, choline is essentialfor maintaining cell membrane structure andpermeability, and for transport of lipid from the liveras a constituent of very low density lipoproteins(VLDL). Choline deficiency leads to fatty liver inlaboratory animals.

Estimates of ruminal choline degradation are 80-98%(Atkins et al., 1988, Sharma and Erdman, 1989b).Differential milk yield response between ruminal andabomasal infusion of choline support invitro data andsuggest ruminal degradation of choline (Sharma andErdman, 1989a). The main fate of dietary choline isruminal conversion to trimethylamine followed byconversion to methane. Ruminal production ofcholine is negligible (Erdman, 1992). Protectedcholine supplements have been developed todecrease microbial degradation in the rumen andincrease delivery of choline to the small intestine butdocumentation of extent of ruminal protection anddegree of release in the small intestine is scarce.

Milk Production. A summary of milk and milkcomponent responses to postruminal infusion ofcholine or feeding protected choline are shown inFigures 1-3. In each, the control treatment mean hasbeen plotted against the choline treatment mean.Amounts of choline chloride supplied for most of thestudies was 15 g/day, although amounts may havebeen as high as 50 g/day in some of the postruminalinfusion trials. We plotted the data in this fashion totry and examine if the response was related to levelof production or milk components.

In the 16 treatment comparisons summarized inFigure 1, there was a significant increase (P < 0.05) inmilk yield in 6 comparisons and a trend (P < 0.15)toward an increase in one comparison. Response tocholine was not dependent on high milk yieldbecause some of the biggest responses were obtainedfrom moderate (30-35 kg/d) producing cows.

Reasons for the increases in milk yield or variabilityin milk yield response are not known. For studies inwhich protected choline was fed, it may be related tothe level of protection and/or degree of release in theintestine. It may be related to status of other methyldonors (i.e. methionine) or cofactors associated withone-methyl carbon metabolism (folic acid, vitaminB12). Further studies are needed to establish therelationship between supplemental choline and othernutrients involved in one-carbon methyl metabolism.

Some of the early studies by Erdman and co-workers(Erdman et al., 1984) hypothesized that choline

64

Should Protected Choline or Niacin be Fedto Periparturient Dairy Cows?

J.A.A. Pires and R.R. GrummerDepartment of Dairy Science

University of Wisconsin - [email protected]

65

supplementation may increase milk fat percentage.They suggested “choline aided the transport ofmobilized free fatty acids from adipose tissue throughthe liver to the mammary gland”. Figure 2 shows asummary of the responses of milk fat percentage torumen-protected or postruminally infused choline. Ofthe 12 studies summarized, fat percentage wasincreased statistically in 3 of the studies (P < 0.05) andtended to increase in 1 of the studies (P < 0.15).Interestingly, in three studies employing fresh cows inwhich milk fat test was high (> 4.0%) and fatmobilization would be high, there was no treatmenteffect. This does not support Erdman’s hypothesis ofenhanced hepatic lipid transport as a mechanism ofaction for increased milk fat.

Milk protein response to supplemental choline issummarized in Figure 3. The response is neutralwith only one study indicating a trend for a change; adecrease (P < 0.15). Given the potential ability ofcholine to spare methionine, it is surprising thatpositive treatment effects were never observed. Inthe great majority of studies, diet methionine statuswas not described.

Animal Health. Choline deficiency in rats has beenshown to cause an increase in accumulation oftriglyceride (TAG) in liver. Triglyceride istransported out of the liver as a constituent of VLDL.As previously mentioned, choline may sparemethionine (also a methyl donor) which is an aminoacid that is required for synthesis of protein (aconstituent of VLDL). Choline also serves as asubstrate for synthesis of PC, another constituent ofVLDL.

Fatty liver is a metabolic disorder that can affect upto 50% of high producing cows during the transitionperiod, potentially compromising health, productionand reproduction. Fatty liver develops when plasmanonesterified fatty acid (NEFA) concentrations arehigh due to depressed feed intake and alteredendocrine status associated with initiation ofparturition and lactation. The NEFA concentration atwhich TAG begins to accumulate in liver is not wellestablished, but is known that the hepatic uptake ofNEFA is directly associated with its concentration inblood. Research involving frequent blood and liversampling in periparturient Holstein cows has shownliver TAG accumulation within 1 day after calving,which was preceded by an acute increase in plasmaNEFA concentration and depressed feed intakeimmediately prior to and at calving (Vazquez-Añonet al., 1994). Therefore, if the flow of choline to theintestine of dairy cattle is insufficient during theperiparturient period when feed intake is low and fatmobilization is high, synthesis of VLDL could belimited and fatty liver could result.

Only recently have the effects of choline on liver TAGbeen measured directly. Hartwell et al. (2000) fedruminally protected choline to transition dairy cowsbut did not see any beneficial effect on liver TAGconcentration. However, the degree of ruminalprotection of the choline fed in that trial has beenquestioned by the manufacturer of the product (D.Putnam, personal communication). More recently, animproved protected choline product (D. Putnam,personal communication) was fed to transition dairycows and a statistically non-significant reduction inliver TAG was observed as level of supplementationwas increased (Piepenbrink and Overton, 2004).Liver TAG is a highly variable measurement in dairycattle immediately after parturition and this studymay not have had adequate animal numbers to detectstatistically significant treatment differences.Therefore, we attempted to assess whether cholinehad a role in preventing or alleviating fatty liverusing an experimental model that might be moresensitive for detecting a treatment effect.

To conduct these experiments, we used far-off drycows. In the first study, cows were energy-restrictedto approximately 30% of requirements formaintenance and pregnancy for 10 days. This wasdone to mimic feed intake depression prior to calvingand allow for lipid mobilization and development offatty liver. During the energy restriction, cows werefed an unsupplemented diet or one dietsupplemented with rumen-protected choline. Thisprotocol allowed access to whether choline has a rolein the prevention of fatty liver. For the secondexperiment, cows were energy-restricted for 10 days,similar to that in the first experiment. During thistime, all cows were fed the same diet. Following the10-day energy restriction, cows were fed ad libitumfor 6 days. During that time cows were fed anunsupplemented diet or one supplemented withrumen-protected choline. Depletion of TAG from theliver was monitored during the as libitum feeding.This protocol allowed us to determine whethercholine has a role in the alleviation of fatty liver(Cooke et al., 2007). Data from these trials indicatedthat choline had a role in prevention and alleviationof fatty liver. In the first study plasma NEFA werealso reduced by choline supplementation, therefore, itcannot be distinguished whether the beneficial effectsof choline on liver TAG were due to direct effects onthe liver or indirect effects on lower plasma NEFA.

Demonstrating that choline can prevent fatty liverusing a model such as we employed is an importantfinding, but it does not establish if there is abeneficial effect on animal health. To evaluate effectson animal health, trials employing large animalnumbers are required and such studies arenecessarily conducted on large commercial farms.

Lima et al. (2007) conducted two experiments onseparate farms. On one farm, using 363 cows, 0 or 15g/d of choline in a protected form was fed between25 days prior to expected calving until 80 days postcalving. Postpartum, DMI tended to be greater (22.6vs. 23.9 kg/d; P = 0.10) and fat-corrected milk yieldwas greater (44.6 vs. 42.8; P < 0.05) for cows fedcholine. Feeding choline reduced (P < 0.05) theincidence of ketonuria (10.7 vs. 28.8%), clinical ketosis(4.0 vs. 11.3%), and the relapse of clinical ketosis (2.3vs. 6.85). On the second farm, the same treatmentswere fed, but the duration was only from 25 daysprior to expected calving until calving. DMI and fat-corrected milk was not affected by treatment andcholine supplementation tended to increase milkyield (27.9 vs. 28.7 kg/d; P = 0.07). Parametersrelated to ketosis were not affected by treatment. Theresearchers speculated that the absence of a responseon the second farm my have been due to the absenceof choline supplementation after calving.

NiacinBackground. The vitamin niacin is a precursor of thecoenzyme nicotinamide adenine dinucleotide (NAD)which participates in a large number of oxidation-reduction reactions, both in anabolic(NADPH/NADP) and catabolic (NADH/NAD)pathways. Niacin can be found in two commonforms: nicotinic acid (NA) and nicotinamide (NAM).Both compounds have similar nutritional properties,and both can be used in the synthesis of NAD buthave distinct biological properties (Dipalma andThayer, 1991; Carlson, 2005).

Early studies from the 1960’s showed that NA hasanti-lipolytic effects in humans. Both oral orintravenous administrations of NA boluses lead todramatic and acute reductions of plasma NEFAconcentrations, followed by a rebound above baselinelevels and a subsequent return to baseline (Carlson,2005). In dairy cows, large oral boluses of NA causetransient decreases in NEFA concentration followedby a rebound (160 g, Waterman and Schultz, 1972;Waterman et al., 1972; 12 or 120 g, Jaster et al., 1983).Pharmacological doses of NA inhibit lipolysis inadipose tissue (Dipalma and Thayer, 1991; Carlson,2005), but have minimal direct effects on subsequentfatty acid metabolism in the bovine (Waterman andSchultz, 1973). The rebound is thought to occurwhen NA action in adipose tissue ceases, possiblydue to clearance of NA from blood (Waterman andSchultz, 1972). In contrast to NA, NAM does nothave anti-lipolytic properties in humans (Dipalmaand Thayer, 1991; Carlson, 2005). Accordingly, oraladministration of 12 g/d of NAM to feed-restrictedcows failed to reduce plasma NEFA or BHBAconcentrations (Jaster and Ward, 1990), even thoughrumen microbes are able to convert NAM to NA

(Harmeyer and Kollenkirchen, 1989; Campbell et al.,1994).

Productive performance and health. A meta-analysis of27 feeding studies involving free NAsupplementation to dairy rations showed noimprovement in lactation performance when NA wasgiven at a rate of 6 g/d (Schwab et al., 2005).Supplementation of 12g NA/d did not change inmilk production (0.4 kg; P = 0.12) and resulted inmodest increases in fat (25.8 g/d; P = 0.01) andprotein (17.4 g/d; P = 0.08) yields compared tocontrols (Schwab et al., 2005). The authorsquestioned the economic return from supplementingfree NA in dairy rations due to high variability ofresults and small production responses.

Plasma NEFA concentrations were reduced in onlyone out of 11 studies in which periparturient cowswere supplemented with niacin (6 to 12 g/d of freeNA or NAM; NRC, 2001). Accordingly, meta-analysisof multiple studies involving NA feeding showed nostatistical effects of NA in plasma NEFA and BHBAconcentrations (Schwab et al., 2005). The absence ofpositive results of niacin supplementation in mostfeeding trials contrasts with the positive effects of NAin reducing NEFA levels in humans, and with theacute reduction in plasma NEFA whensupraphysiological doses of NA were given to cows.There are several factors that may explain the lack ofsuccess with using niacin to reduce NEFA. The activeform of niacin modulating adipose tissue metabolismin humans is NA, therefore, effects may not beexpected in the bovine if NAM is fed. Supplementalniacin is extensively degraded or transformed in therumen. Only 6.7% to 17% of supplemental NA wasestimated to reach the duodenum in the bovine (Zinnet al., 1987; Campbell et al., 1994). Absorption of NAthrough the rumen is probably insignificant(Harmeyer and Kollenkirchen, 1989; Erickson et al.,1991; Campbell et al., 1994). The dosage of niacin fedto periparturient dairy cows (either as NA or NAM)usually ranges from 6 to 12 g/d (NRC, 2001; Schwabet al., 2005). This supplementation level is probablyinsufficient to elicit a significant and sustaineddecrease in NEFA, especially when taking intoconsideration the limited flow of supplemental NA tothe lower gut (Zinn et al., 1987; Campbell et al., 1994).Recent experiments involving the supplementation ofperiparturient dairy cows with high doses of rumenavailable NA have produced inconsistent results.Jersey cows supplemented with 48 g/d of free NAfrom 30 d prepartum until calving had lower levels ofplasma NEFA at calving and less DMI decline duringthe last week of gestation (French, 2004). However,these results were not replicated when both Holsteinand Jersey cows were supplemented with up to 98mg NA/d per kg BW, from 30 d prepartum to 21 d

66

postpartum (Chamberlain and French, 2006). Wesuspect that only a small fraction of the NA wasabsorbed because a free form of NA was used.Furthermore, the antilipolytic effects could have beentransient, and the inconsistent results on plasmaNEFA may have reflected the time of blood samplingrelative to NA feeding.

Current research. We hypothesized that, if delivered insufficient quantities to the intestine, NA would limitlipolysis in adipose tissue and induce sustainedreductions in plasma NEFA concentration duringperiods of negative energy balance. In experiment 1,we studied the effect of single NA abomasal bolus onplasma NEFA of feed-restricted cows (Pires andGrummer, 2007). Treatments were a bolus of 0, 6, 30or 60 mg NA/kg BW, corresponding to 0, 5, 24 and 49g of NA, given as a single abomasal infusion eachperiod. Cows were feed-restricted for 48 h prior toabomasal infusion to stimulate mobilization of fattyacids and elevate plasma NEFA concentration. AllNA doses caused dramatic reductions of plasmaNEFA, followed by a rebound during which NEFAincreased transiently above baseline levels (Figure 4).The initial pattern of plasma NEFA decrease suggeststhat blood NA concentrations initially reached athreshold that induced maximum inhibition ofadipose lipolysis. The rebound of plasma NEFAfollowed a pattern observed in other animal models.A second experiment was conducted to assesswhether successive abomasal infusions of NA couldinduce sustained reductions of plasma NEFAconcentration (Pires and Grummer, 2007). Six non-pregnant, non-lactating Holstein cows were feed-restricted for 48 h to increase plasma NEFA. At 48 hof feed restriction, cows received 9 hourly abomasalinfusions of 0, 6 or 10 mg NA/kg BW per h, whichcorresponded to 0, 4.9 and 8.3 g NA/h. Both NAtreatments reduced plasma NEFA in a similar pattern,from 550 to approximately 100 uEq/L. Again, adramatic rebound was observed after NA infusionswere discontinued at 8 h (Figure 5). We haveconducted additional experiments to determine iflower rates of NA infusion may promote moremoderate reductions of plasma NEFA concentrationwithout a rebound effect (unpublished); that does notappear to be the case.

ConclusionsSupplementation of dairy diets with protectedcholine has improved milk yield in moderateproducing cows (30 - 35 kg/d). There is someevidence supporting beneficial effects in milk fatpercent, while protein percent was unchanged in alltrials reviewed. In addition to productiveperformance, potential benefits of supplementalcholine on modulation of lipid metabolism must beconsidered. Controlled experiments suggest that

supplemental choline prevents liver TAGaccumulation and enhances depletion TAG followinga protocol for induction of hepatic TAG infiltration.Field studies involving large animal numbers showedincreased DMI and FCM production, and decreasedincidence of ketosis when cows were fed rumen-protected choline during both close-up period andearly lactation.

Recent research shows that NA is a powerfulantilipolytic agent in the bovine under negativeenergy balance due to feed restriction- if it isdelivered postruminally. Unfortunately, it seemsunlikely that there is a dose of NA that can lowerplasma NEFA and avoid the rebound effect whendelivery is terminated. Because of the potentialnegative consequences on plasma NEFA that mightoccur if an animal goes off feed, supplementation ofNA in any form cannot be recommended at this timefor control of lipid related metabolic disorders.

ReferencesAbeni, F., M. Speroni, M. G. Terzano, L. Migliorati, P

Cavassini, and G. Pirlo. 2007. Effects of rumenprotected choline on production responses in ItalianFriesian dairy cows. J. Dairy Sci. 90(Suppl. 1):354.

Atkins, K.B., R. A. Erdman, and J. H. Vandersall. 1988.Dietary Choline effects on milk yield and duodenalcholine flow in dairy cattle. J. Dairy Sci. 71:109-116.

Brusemeister, F., and K. Sudekum. 2006. Rumen-protectedcholine for dairy cows: the in situ evaluation of acommercial source and literature evaluation of effectson performance and interactions between methionineand choline metabolism. Anim. Res. 55:93-104.

Campbell, J. M., M. R. Murphy, R. A. Christensen, and T. R.Overton. 1994. Kinetics of niacin supplements inlactating dairy cows. J. Dairy Sci. 77:566-575.

Carlson, L. A. 2005. Nicotinic acid: The broad-spectrumlipid drug. A 50th anniversary review. J. Intern. Med.258:94-114.

Chamberlain, J. L. and P. D. French. 2006. The effects ofnicotinic acid supplementation during late gestation onlipolysis and feed intake during the transition period. J.Dairy Sci. 89 (Suppl. 1):232.

Cooke, R. F., N. S. Del Rio, D. Z. Caraviello, S. J. Bertics, M.H. Ramos, and R. R. Grummer. 2007. Supplementalcholine for prevention and alleviation of fatty liver indairy cattle. J. Dairy Sci. 90:2413-2418.

Davidson, S., B. Hopkins, J. Odle, V. Fellner, and L.Whitlow. 2006. Supplementation of diets with limitedmethionine content with rumen-protected forms ofmethionine, choline, betaine in early lactation cows. S.Davidson, B. Hopkins, J. Odle, C. Brownie, V. Fellner,and L. Whitlow. J. Dairy Sci. 89(Suppl. 1):142.

Dipalma, J. R. and W. S. Thayer. 1991. Use of niacin as adrug. Annu. Rev. Nutr. 11:169-187.

Emanuele, S., T. Hickley, and R. Carvalho. 2007. Effect ofrumen protected choline (Reashure) and rumenprotected methionine on milk yield, and composition inlactating cows. J. Dairy Sci. 90(Suppl. 1):352.

Erdman, R. A. 1992. Vitamins. Pages 297-308 in LargeDairy Herd Management. H. H. Van Horn and C. J.Wilcox, eds. American Dairy Science Association,Champaign, Il.

67

Erdman, R. A., and B. K. Sharma. 1991. Effect of dietaryrumen-protected choline in lactating dairy cows. J.Dairy Sci. 74:1641-1647.

Erickson, P. S., M. R. Murphy, C. S. McSweeney, and A. M.Trusk. 1991. Niacin absorption from the rumen. J. DairySci. 74:3492-3495.

French, P. D. 2004. Nicotinic acid supplemented at atherapeutic level minimizes prepartum feed intakedepression in dairy cows. J. Dairy Sci. Vol. 87 (Suppl.1):345 (Abstr.).

Harmeyer, J. and U. Kollenkirchen. 1989. Thiamin andniacin in ruminant nutrition. Nutrition research reviews2:201-225.

Hartwell, J. R., M. J. Cecava, and S. S. Donkin. 2000.Impact of dietary rumen undegradable protein andcholine on intake, peripartum liver triglyceride, plasmametabolites and milk production in dairy cows. J.Dairy Sci. 83:2907-2917.

Janovick Guretzky, N. A., D. B. Carlson, J. E. Garret, and J.K Drackley. 2006. Lipid metabolite profiles and milkproduction for Holstein and Jersey cows fed rumen-protected choline during the periparturient period. J.Dairy Sci. 89:188-200.

Jaster, E. H. and N. E. Ward. 1990. Supplemental nicotinicacid or nicotinamide for lactating dairy cows. J. DairySci. 73:2880-2887.

Jaster, E. H., D. F. Bell, and T. A. McPherron. 1983. Nicotinicacid and serum metabolite concentrations of lactatingdairy cows fed supplemental niacin. J. Dairy Sci.66:1039-1045.

Lima, F. S., M. F. Sa Filho, L. F. Greco, F. Susca, V. J. A.Magalhaes, J. Garrett, and J. E. P. Santos. 2007. Effectsof feeding rumen-protected choline (RPC) on lactationand metabolism. J. Dairy Sci. 90(Suppl. 1):174.

NRC. 2001. Nutrient requirements of dairy cattle. 7thRevised ed. Natl. Acad. Press, Washington, DC.

Ondarza, M. B., S. Emanuele, and D. Putnam. 2007. Effectof rumen protected choline (reashure) supplemented tohigh producing cows on milk production, milkcomponents, and intake. J. Dairy Sci. 90(Suppl. 1):353.

Piepenbrink, M. S., and T. R. Overton. 2003. Livermetabolism and production of cows fed increasingamounts of rumen-protected choline during theperiparturient period. J. Dairy Sci. 86:1722-1733.

Pires, J. A. A. and R. R. Grummer. 2007. The use of nicotinicacid to induce sustained low plasma nonesterified fattyacids in feed-restricted Holstein cows. J. Dairy Sci.90:3725-3732.

Pires, J. A. A., J. B. Pescara, and R. R. Grummer. 2007.Reduction of plasma NEFA concentration by nicotinicacid enhances the response to insulin in feed-restrictedHolstein cows. J. Dairy Sci. 90:4635-4642.

Schwab, E. C., D. Z. Caraviello, and R. D. Shaver. 2005. Ameta-analysis of lactation responses to supplementaldietary niacin in dairy cows. Professional AnimalScientist:239-247.

Sharma, B. K., and R. A. Erdman. 1988. Abomasal infusionof choline and methionine with or without 2-amino-2-methyl-1-propanol for lactating dairy cows. J. DairySci. 71:2406-2411.

Sharma, B. K., and R. A. Erdman. 1989. Effects of dietaryand abomasally infused choline on milk productionresponses of lactating dairy cows. J. Nutr. 119:248-254.

Thering, B. J., J. M. Ramos-Nieves, J. L. Lukas, D. E.Putnam, and T. R. Overton. 2007. Interrelationships ofdietary supplies of choline and methionine onproductive performance of Holstein dry cows. J. DairySci. 90(Suppl. 1):355.

Toghdory, A., S. Emanuele, T. Choorchi, and A Naerian.2007. Effect of choline and rumen protected choline onmilk production, milk composition and bloodmetabolites of lactating dairy cows. J. Dairy Sci.90(Suppl. 1):353.

Vazquez-Añon, M., S. Bertics, M. Luck, R. R. Grummer, andJ. Pinheiro. 1994. Peripartum liver triglyceride andplasma metabolites in dairy cows. J. Dairy Sci. 77:1521-1528.

Waterman, R. and L. H. Schultz. 1972. Nicotinic acidloading of normal cows: Effects on blood metabolitesand excretory forms. J. Dairy Sci. 55:1511-1513.

Waterman, R. and L. H. Schultz. 1973. 1 carbon 14-labeledpalmitic acid metabolism in fasted, lactating goatsfollowing nicotinic acid administration. J. Dairy Sci.56:1569-1574.

Waterman, R., J. W. Schwalm, and L. H. Schultz. 1972.Nicotinic acid treatment of bovine ketosis. I. Effects oncirculatory metabolites and interrelationships. J. DairySci. 55:1447-1453.

Zahra, L. C., T. F. Duffiled, K. E. Leslie, T. R. Overton, D.Putnam, and S. J. LeBlanc. 2006. Effects of rumen-protected choline and monensin on milk productionand metabolism of periparturient cows. J. Dairy Sci.89:4808-4818.

Zinn, R. A., F. N. Owens, R. L. Stuart, J. R. Dunbar, and B.B. Norman. 1987. B-vitamin supplementation of dietsfor feedlot calves. J. Anim Sci. 65:267-277.

Figure 1. Milk yield response, plotted as controltreatment vs. choline treatment, for cows fed rumen-protected choline or postruminally infused withcholine. References: Abeni et al., 2007; Davidson etal., 2006; Emanuele et al., 2007; Erdman and Sharma,1991; Hartwell et al., 2000; Janovick Guretzky et al.,2006; Lima et al., 2007; Ondarza et al., 2007;Piepenbrink and Overton, 2003; Sharma and Erdman,1988; Sharma and Erdman, 1989; Thering et al. 2007;Toghdory et al., 2007; Zahra et al., 2006.

68

Figure 2. Milk fat response, plotted as controltreatment vs. choline treatment, for cows fed rumen-protected choline or postruminally infused withcholine. References: Abeni et al., 2007; Davidson etal., 2006; Emanuele et al., 2007; Erdman and Sharma,1991; Hartwell et al., 2000; Janovick Guretzky et al.,2006; Ondarza et al., 2007; Piepenbrink and Overton,2003; Sharma and Erdman, 1988; Sharma andErdman, 1989; Toghdory et al., 2007; Zahra et al.,2006.

Figure 3. Milk protein response, plotted as controltreatment vs. choline treatment, for cows fed rumen-protected choline or postruminally infused withcholine. References: Abeni et al., 2007; Davidson etal., 2006; Emanuele et al., 2007; Erdman and Sharma,1991; Hartwell et al., 2000; Janovick Guretzky et al.,2006; Ondarza et al., 2007; Piepenbrink and Overton,2003; Sharma and Erdman, 1988; Sharma andErdman, 1989; Thering et al. 2007; Toghdory et al.,2007; Zahra et al., 2006.

Figure 4. Effects of abomasal infusion of single dosesof nicotinic acid on plasma NEFA. Fixed effects inthe statistical model: treatment (P = 0.001), time andtreatment x time interaction (P < 0.001). Treatmentdifferences within a time point are indicated by * (P <0.001; Pires and Grummer, 2007).

Figure 5. Effects of abomasal infusions of nicotinicacid at a rate of 0, 6, or 10 mg/h per kg of BW onplasma NEFA. Infusion of treatments started at 48 hof feed restriction (time 0) and was repeated at 1, 2, 3,4, 5, 6, 7, and 8 h thereafter. Fixed effects in thestatistical model: treatment (P = 0.06), time (P < 0.001)and treatment x time (P < 0.001). Treatmentdifferences within a time point are indicated by *(P < 0.001) and ‡ (P = 0.05; Pires and Grummer, 2007).

69

AbstractRuminal acidosis is caused by an imbalance betweenthe production of fermentation acids by microbes inthe rumen and the absorption, passage,neutralization, and buffering of those acids. Theproduction rate of fermentation acids is highlyvariable across diets and increases greatly withhighly fermentable starch sources. Hydrogen ionsare removed primarily by absorption from therumen, and the concentration gradient across theruminal epithelium is likely the major factor affectingabsorption of acids from the rumen. Coarse foragefiber affects ruminal pH by retaining digesta in therumen which provides the buffering capacityinherent in feedstuffs, increases salivary buffer flowthrough stimulation of rumination, and increases theconcentration gradient through stimulation ofruminal motility. Selecting optimal dietfermentability and maintaining an adequate digestapool are likely key factors in preventing subacuteruminal acidosis.

Keywords: subacute rumen acidosis, rumination,motility, concentration gradient, buffering

IntroductionA slightly acidic ruminal pH is desirable to maximizemilk yield of dairy cattle because diet digestibilityand yield of microbial protein produced in the rumenare maximized when highly fermentable diets arefed. Ruminal pH near or above 7.0 can be thought ofas lost opportunity; energy intake and microbialprotein production will be sub-maximal, limitingmilk yield and (or) increasing diet cost. However, asruminal pH decreases, appetite (Shinozaki, 1959),ruminal motility (Ash, 1959; Shinozaki, 1959),microbial yield (Hoover, 1986), and fiber digestion(Hoover, 1986; Terry et al., 1969) are reduced. Theoptimal ruminal pH is dependent on within-dayvariation determined by diet composition, productionlevel, and feeding systems. Ruminal pH measured atone time point or averaged throughout the day haslittle biological significance without considering thepattern of ruminal pH within a day; even minimumruminal pH means little if the duration at this pH isvery short. Both the time spent at low pH and theextent to which pH is depressed should be

considered (Mackie and Gilcrist, 1979). Within-dayvariation in ruminal pH varies greatly, increasingwith increased dietary starch concentration (Oba andAllen, 2000, 2003), and is expected to differ amongfeeding systems. Minimizing fluctuations in ruminalpH is expected to allow greater energy intake,microbial protein production and improved animalhealth. The objective of this paper is to identify anddiscuss factors affecting ruminal pH as a dynamicsystem, so that feeding systems can be managed tolimit subacute ruminal acidosis.

Ruminal acid productionRuminal hydrogenion concentration(described by pH) isdetermined by thebalance betweenfermentation acidproduction andhydrogen ionremoval byabsorption,neutralization,buffering, andpassage. Ruminalacids are endproducts of microbialfermentation andinclude acetic,propionic, andbutyric acids. Theyare mild acids withpKa’s around pH 4.8and are the primaryacids involved insubacute ruminalacidosis. Lactic acidalso is produced byruminal microbes(e.g. Streptococcusbovis, Lactobacilli),and its acid dissociation constant is 10-fold higherthan the major VFA, so it has a much greaterinfluence on ruminal pH below pH 5. Althoughlactic acid is clearly involved in acute acidosis, it isnot likely a common contributor to subacute ruminal

70

Physically Effective Fiber and Regulation ofRuminal pH: More Than Just Chewing1

M. S. Allen, J. A. Voelker and M. ObaMichigan State University, Department of Animal Science, East Lansing, MI 48824 [email protected]

1Allen, M. S., J. A. Voelker, and M. Oba. 2006. Effective fiber and regulation of ruminal pH: its more than just chewing. In:Production Diseases in Farm Animals, N. P. Joshi and T. H. Herdt, eds, Wageningen Academic Publishers, Wageningen,Netherlands. P. 270-278

acidosis (see Figure 1B) because it is normally only aminor fermentation end product that is metabolizedrapidly in the rumen (Gill et al. 1986).

Both the amounts and the patterns of acid productionand removal determine the physiological response toruminal pH. Total daily production of fermentationacids is determined primarily by organic matter (OM)intake and proportion of ruminally degraded organicmatter (RDOM) in the diet, and both OM intake andRDOM vary widely. Allen (1997) reported thatRDOM content of diets for lactating dairy cattleaveraged 50% and ranged from 29 to 67% of the totalOM for experiments using duodenally canulatedcows (48 treatment means). The amount of RDOMaveraged 9.8 kg/d and ranged from 5.7 to 15.4 kg/d.Because approximately 7.4 equivalents of acid areproduced per kg ruminally degraded organic matter(RDOM) when microbial cell yield from hexosefermented is 0.33 (Allen, 1997), approximatefermentation acid production in these experimentsvaried from 42 to 114 Eq/d. In addition to this widerange of potential acid production, both the patternof acid production and the extent and pattern of acidneutralization and removal contribute tophysiological response to ruminal pH. Theproduction of hydrogen ions per second for a cowconsuming 20 kg of 50 rumen-degraded OM is 0.86meq/sec, which is more than 10 times the freeruminal hydrogen ion pool size (0.08 meq at pH 6.0in 80 L). Therefore, understanding effects of diet andmanagement on the moment-by-moment turnover ofhydrogen ions is very important for limiting subacuteruminal acidosis.

The amount of fermentable OM in the rumen at anymoment, and thus the acid production at anymoment, depend upon DMI, meal patterns, flow rateof OM from the rumen, and rate of fermentation.Feeds vary in the fraction of OM available forfermentation and in rates of digestion and passage ofthe available fraction from the rumen. The sameruminal digestibility can be attained for a feed with aslow rate of digestion as for a feed with a very fastrate of digestion if it is retained in the rumen longenough. Also, feeds with very fast rates offermentation such as ground high-moisture corn andsteam-rolled barley cause greater variation in ruminalpH than feeds with more moderate rates offermentation. Fiber generally ferments more slowlythan starch and usually has a longer retention time.Therefore, digestible fiber provides a consistentsupply of energy to microbes and to the animal overtime. In contrast, finely ground grain ferments andpasses from the rumen quickly, providing pulses ofenergy to rumen microbes and VFA to the animal.Rapid energy availability within a day or across dietscan result in variation in the efficiency of microbial

cell yield from hexose because of the uncoupling ofdigestion and cell growth (Voelker and Allen, 2003),further contributing to within-day variation inruminal pH.

Feed intake amount and feed characteristicscontribute to variation in both amount and pattern offermentation acid production. Simultaneously,variation in the amount and pattern of acid removalmay either stabilize or increase variation in ruminalacid concentration.

Hydrogen ion removalA model of hydrogen ion production and removalfrom the rumen indicated that most (> 50%) acid isremoved from the rumen by absorption across therumen wall (Allen, 1997). Other routes of removalwere identified as incorporation into water (28%) viacarbonic acid from salivary bicarbonate, and flowfrom the rumen as dihydrogen phosphate (10%), VFA(~3%), ammonium (~2%), and particulate matter(<2%).

Conversion to water A large fraction of hydrogen ionsis removed by the carbonate buffer system; hydrogenions combine with bicarbonate to carbonic acid(H2CO3) which is rapidly converted to H2O and CO2.Although the pKa for bicarbonate is 6.1, the effectivepKa is much higher (~7.0) because the rumen is anopen system; CO2 is constantly lost by eructationwith a loss of bicarbonate and hydrogen ions fromthe system.

Passage The phosphate buffer system is different fromthe carbonate system because phosphate is removedby passage only (except for minor incorporation intomicrobial cells) and the pKa of hydrogen phosphateis higher at 7.2. At pH 6.0, approximately 94% of thehydrogen phosphate secreted is complexed asdihydrogen phosphate, which is removed by passagewith the liquid fraction. Hydrogen ions also passfrom the rumen associated with VFAs (~3%),ammonia (~2%), and feed residues (~2%; Allen, 1997).Therefore, passage rates of both solid and liquiddigesta can affect ruminal pH.

Saliva secretion Ruminal pH often decreases followingmeals and often increases during bouts of rumination(Figure 2). The decrease following meals is becauseof the production of fermentation acids from the OMconsumed while the increase during rumination isusually attributed to the secretion of buffers in saliva.Ruminal pH was highly responsive in the individualcow represented in the figure; the responsiveness ofruminal pH to meals and ruminating bouts dependson the presence of existing buffer reserves andtherefore is not always as dramatic. Saliva containsbicarbonate and phosphate that neutralize and buffer

71

acids in the rumen. The bicarbonate and phosphateconcentrations of saliva have been reported as 126and 26 meq/L, respectively (Bailey and Balch, 1961)and saliva composition is relatively constant and notgreatly affected by diet or feed intake (Erdman, 1988).Therefore differences in the flow of salivary buffersinto the rumen are a function of saliva flow. Totalchewing time per day and the fraction of totalchewing time spent ruminating were highly relatedto forage NDF content of the diet and the particlelength of the forage (Allen, 1997), confirming the roleof forage NDF in stimulating chewing. A meta-analysis of treatment means from the literaturerevealed that concentration of forage NDF in dietswas positively related to ruminal pH (r2 = 0.63,P < 0.001) while concentration of NDF from allsources was not related, showing the importance ofeffectiveness of forage NDF to maintain ruminal pH(Allen, 1997). Although this correlation is partly theresult of increased saliva flow during rumination,there are likely additional effects of effective fiber.Using saliva flow rates reported in the literature foreating, ruminating, and idle activites, Allen (1997)calculated that while increasing dietary forage NDFconcentration can increase total chewing time by 200min/d, salivary buffer flow only increased 2.3 eq/d,or about 5%. This suggests that either the methodsused to measure saliva flow drasticallyunderestimated the true difference between idle andchewing saliva flow rates, or additional mechanismsdetermine the relationship between forage NDF andruminal pH. It is possible that the effect of forageNDF on ruminal pH is the result of increased ruminalmotility and maintenance of ruminal digesta pool inaddition to increased salivary buffer flow.

Buffering of hydrogen ionsBasal fermentation of digesta in the rumen provides afairly consistent flux of acids but the total flux ispulsatile with a frequency and amplitude determinedby meal patterns and fermentability of the diet.Portal appearance of VFA, and therefore removal ofhydrogen ions by absorption, is also pulsatile and islinked to meal patterns (Benson et al., 2002). Eachmeal might provide substrate for the production ofover 7,000 meq of acid over time for cows consuming10 meals/d. Removal of hydrogen ions byneutralization is also expected to be pulsatile becausesaliva secretion increases during rumination andcows ruminate in bouts between meals throughoutthe day (Figure 2).

Diurnal variation in ruminal hydrogen ionconcentration varies depending upon the linkbetween meal patterns and rumination patterns. Thebuffering capacity of ruminal contents is a factorwhose effect on the free hydrogen ion pool andtherefore on ruminal pH is poorly understood.

Buffers in the rumen include undigested feedresidues and microbial cells, salivary buffers, andVFA. Hydrogen phosphate is less of a buffer in therumen than it is an alkalizer because it is nearlycompletely complexed with hydrogen ions below pH6. Fermentation acids and salivary buffers might beexpected to have a greater effect on ruminal pH whenpH is relatively high (i.e., 6.5), because pH is a log-scale and hydrogen concentration is an order ofmagnitude lower at pH 6.5 than at pH 5.5. However,pH at 5.5 is determined by the concentration of VFA,buffering of digesta, and the removal of hydrogenions through the bicarbonate system. This introducesgreat variation in the response of ruminal pH to VFAconcentration (Figure 1A). The reserve bicarbonatepool in the rumen is expected to be depleted quicklyas pH declines because CO2 is lost from the systemby eructation. Therefore, the lack of relationshipbetween ruminal VFA concentration and ruminal pHat pH less than 6.0 shown in Figure 1A is probablybecause of buffering by digesta and previous removalof hydrogen ions by the bicarbonate system andhydrogen phosphate, and less the result of actualbuffering by saliva. As ruminal pH declines below5.0, it is very resistant to change in either directionbecause of the high hydrogen ion concentration andthe great buffering capacity of VFA and digesta; anyadditional acid produced is slight relative to theexisting pool and is also more likely to be buffered.

Thebufferingcapacity (BC)of feedstuffsvariesconsiderably.Cereal grainshave low BC,low proteinfeeds andgrass forageshaveintermediateBC, andlegumeforages andhigh proteinfeeds havehigh BC(Jasaitis etal., 1987).Amongforages, BCtends toincrease withmaturity (Jasaitis et al., 1987) and with ensiling(Erdman, 1988). Normally only a fraction of the BCof feeds is used as most BC is at pH lower than

72

RuminalpH

Figure 2. The relationship among ruminal pH,meals, and chewing activity for one cow fed a35% NDF diet twice daily. Ruminal pH isrepresented by the top line. The weight of thefeed remaining was measured by a mangersuspended from a load cell and is representedby the middle line. Meals are represented bythe vertical bars. Increases in feed remainingthat were recorded during eating bouts weredue to downward pressure applied by the cowon the manger. Chewing activity is representedby the bottom line. Because many points arerepresented, chewing activity appears as blocksof eating and ruminating bouts. Ruminal pHdecreased rapidly following meals andincreased rapidly during rumination. FromAllen (1997).

normal ruminal pH (Allen, 1997). However, as pHdecreases below 5.5, an increasingly greater fractionof the total BC of digesta is used. Therefore, the BCof feeds and ruminal digesta are very important forstabilizing ruminal pH. Maintaining ruminal digestapool will likely be of special importance during theperiparturient period and other occurrences of lowfeed intake so that the buffering capacity of digesta ismaintained.

Absorption The rumen is the major site of VFAabsorption and absorbed molecules arepredominately in the undissociated form (Ash andDobson, 1963). Therefore VFA absorption results inthe net removal of hydrogen ions from the rumen.Rate of absorption of VFA is dependent on theabsorptive surface area, pH, and the concentrationgradient across the rumen epithelium. Absorptivesurface area is a function of rumen size, degree of fill(Djikstra et al., 1995), and papillae surface area(Dirksen et al., 1985). In addition, parakeratosisreduces effective surface area for absorption; acidosiscan have cumulative effects, diminishing surface areafor absorption and increasing risk for future bouts ofacidosis. The adaptive responses of ruminal papillaesize to diets varying in ruminally fermented OM hasbeen proposed as an important factor affecting thesusceptibility of some animals to ruminal acidosis(Dirksen et al., 1985). Surface area of ruminalpapillae is affected by diet (Dirksen et al., 1985; Xuand Allen, 1999) and offering high roughage dietsduring the dry period might decrease papillae surfacearea enough to decrease rate of VFA absorption.However, in practice, papillae surface area might notlimit rate of VFA absorption of transition cows.Fermentability of diets commonly recommended fordry cows, particularly those prior to parturition, islikely adequate to maintain absorptive surface area.Also, absorption rate displays a diminishing responseto surface area as surface area increases (Xu andAllen, 1999). In addition, relationships observedbetween papillae surface area and rate of VFAabsorption through the transition period isconfounded by other factors such as concentrationgradient and ruminal pH, making it difficult to assessthe importance of adapting ruminal papillae prior toparturition.

Effects of pH on fractional absorption rates of acetic,propionic, and butyric acids were reported byDijkstra et al. (1993) using evacuated and washedrumens. Rates of absorption were similar among theVFA at pH 7.2, but as pH was reduced to 4.5, rate ofabsorption nearly tripled for butyrate, nearly doubledfor propionate, and was not affected for acetate.Because they have similar acid dissociation constantswith similar fractions in the unionized form, thedifferences in absorption rates are probably due to

different concentration gradients caused bymetabolism in the ruminal epithelium (Dijkstra et al.,1993). Butyrate is extensively metabolized by theruminal epithelium, followed by propionate, butacetate is metabolized to a much lesser extent (Sanderet al., 1959). Fermentation acids are absorbed bypassive diffusion from high concentrations inside therumen to lower concentrations in the epithelial cells;therefore factors that increase VFA concentration atthe epithelial border inside the rumen and decreasethe concentration inside the cells are expected toincrease the rate of absorption. Concentration gradientmight be a major factor affecting rate of VFA absorptionfrom the rumen and therefore ruminal pH.

Because VFA are absorbed primarily in theundissociated state, and because a larger proportionof VFA exist in that form at lower pH, it is expectedthat fractional rate of absorption increases as pHdecreases. While this has been shown usingevacuated and washed rumens (Dijkstra et al., 1993),it might not be the case in practice. The washedrumen technique maximizes the concentrationgradient, decreasing the intracellular concentration ofVFA because of removal of the rumen contents andincreasing the concentration at the epithelial borderinside the rumen because the solutions of VFA addedare easily mixed. We developed a method to measurerelative rate of VFA absorption that does not requirerumen evacuation, by pulse dosing valerate and aliquid marker (Co EDTA) and calculating rate ofabsorption as the rate of change of the ratio ofvalerate concentration to cobalt concentration (Allenet al., 2000). Using this technique, we found that rateof absorption of valerate ranged from ~20 to 60 %/h(Voelker and Allen, 2003) and was the primary factorrelated to VFA concentration in the rumen whichranged from ~125 to 155 mM. As expected, ruminalVFA concentration was negatively correlated withboth valerate absorption rate (r2 = 0.48; P < 0.001) andliquid passage rate (r2 = 0.24; P < 0.01); 72% ofvariation in ruminal VFA concentration wasaccounted for by the combined rates of valerateabsorption and liquid passage.

We also recently reported that rate of VFA absorptionwas slower, not higher, under lower ruminal pH(Voelker and Allen, 2003); this contradicts dataobtained using washed rumens. Although rumenmotility was not measured in our study, valerateabsorption rate increased with greater passage rate ofindigestible NDF, and tended to increase with greaterliquid passage rate, both of which might be indicatorsof rumen motility. Furthermore, rate of indigestibleNDF passage decreased as ruminal pH decreased.Slower valerate absorption at lower pH was verified inanother recent study from our laboratory (Taylor andAllen, unpublished). This is additional evidence that

73

concentration gradient might be a dominant factor affectingrate of VFA absorption and therefore ruminal pH. Ruminalmotility affects rate of absorption because constantmixing of ruminal contents is needed to maintain VFAconcentrations at the ruminal epithelium whereabsorption occurs. Ruminal contractions are inhibitedat low ruminal pH (Shinozaki, 1959), decreasingmixing and rate of VFA absorption and decreasing pHfurther. Stimulation of ruminal contractions byphysically effective fiber is expected to increase mixingand rate of VFA absorption. Diets that provide moreconsistent rumen fill over time in animals prone todrastically reduced feed intake (periparturient or earlylactation animals) might reduce the occurrence ofsubacute acidosis upon refeeding by continuing tostimulate ruminal motility in spite of low feed intakeand by maintaining buffering capacity of ruminaldigesta. Another factor that might be related toconcentration gradient is milk yield; Voelker and Allen(2003) reported a positive relationship betweenvalerate absorption rate and FCM yield of cows(R = 0.49, P < 0.01). Cows with high milk yield havegreater blood flow and a greater ability to metabolizeabsorbed VFA, which will decrease theirconcentrations in the blood, increasing theconcentration gradient from the rumen to the blood.Greater ruminal motility is also expected to increaseblood flow to the rumen and utilization of VFA withinruminal tissue, further increasing the concentrationgradient and absorption rate for VFA.

Because more than 50% of acid produced in therumen is removed by absorption across the rumenwall (Allen, 1997), maximizing and maintaining VFAabsorption rate should be a primary strategy forstabilizing ruminal pH. Promoting ruminal motilityand maintaining the concentration gradient across therumen wall should contribute to this process.Physically effective fiber, especially from forages,should prevent acidosis and reduce fluctuations inpH not only by stimulating chewing and salivarybuffer flow but also by maintaining the ruminaldigesta pool and promoting ruminal motility. Foragefiber promotes mat formation and feed particleretention, which maintains a more constant supply ofVFA and should reduce fluctuations in ruminal pH.However, the need for fiber to promote rumination,ruminal motility, and digesta retention must bebalanced against its potential to limit feed intake dueto ruminal filling effects (Allen, 2000). Using foragefiber to maintain ruminal digesta pool will beespecially important when depressed feed intake isexpected and when feed intake is unlikely to belimited by filling effects, such as immediately beforeand after calving.

ConclusionsRuminal pH is a function of the rates of VFAproduction and absorption, and of hydrogen ionpassage, neutralization, and buffering. Diets andfeeding systems should be designed to providehighly fermentable diets while limiting variation inruminal pH. Important diet characteristics affectingacid production, buffering, and absorption arefermentability and coarseness of fiber; highlyfermentable feeds should be limited while adequateeffective fiber should be provided. The concentrationgradient across the ruminal epithelium probably hasa great effect on rate of VFA absorption and isaffected by rumen motility, ruminal pH, blood flow,and possibly by milk yield. This emphasizes theimportance of physically effective fiber to maintaindigesta pool size in order to provide adequatebuffering by ruminal digesta, stimulate ruminationand salivary buffer flow, and promote motility toenhance VFA absorption.

ReferencesAllen, M. S. 1995. Model of hydrogen flow through the

reticulorumen. Proc. XXIII Biannual Rumen FunctionConference, Chicago, IL.

Allen, M. S. 1997. Relationship between ruminalfermentation and the requirement for physicallyeffective fiber. J. Dairy Sci. 80:1447-1462.

Allen, M. S. 2000. Effects of diet on short-term regulation offeed intake by lactating dairy cattle. J. Dairy Sci.83:1598-1624.

Allen, M. S., L. E. Armentano, M. N. Pereira, and Y. Ying.2000. Method to measure fractional rate of volatile fattyacid absorption from the rumen. Abstracts Conferenceon Rumen Function 25:24, Chicago, IL, Nov. 14-16, 2000,Department of Animal Science, M ichigan StateUniversity, East Lansing,http://www.msu.edu/user/rumen/index.htm

Ash, R. W. 1959. Inhibition and excitation of reticulo-rumen contractions following the introduction of acidsinto the rumen and abomasum. J. Physiol. 169:39.

Ash, R. W. and A. Dobson. 1963. The effect of absorptionon the acidity of rumen contents. J. Phsiol. 169:39.

Bailey, C. B. and C. C. Balch, 1961. Saliva secretion and itsrelation to feeding in cattle. 2. The composition and rateof secretion of mixed saliva in the cow during rest. Br.J. Nutr. 15:383.

Benson, J. A., Reynolds, C. K., Aikman, P. C., Lupoli, B. &Beever, D. E. (2002) Effects of abomasal vegetable oilinfusion on splanchnic nutrient metabolism in lactatingdairy cows. J. Dairy Sci. 85: 1804-1814.

Cassida, K. A. and M. R. Stokes. 1986. Eating and restingsalivation in early lactation dairy cows. J. Dairy Sci., 69:1282.

Counotte, G. M. H., and R. A. Prins. 1981. Regulation oflactate metabolism in the rumen. Vet. Res. Communic.5:101.

Dado, R. G., and M. S. Allen. 1995. Intake limitations,feeding behavior, and rumen function of cowschallenged with rumen fill from dietary fiber or inertbulk. J. Dairy Sci. 78:118.

74

Dijkstra, J., H. Boer, J. van Bruchem, M. Bruining, and S.Tamminga. 1993. Absorption of volatile fatty acidsfrom the rumen of lactating dairy cows as influenced byvolatile fatty acid concentration, pH, and rumen liquidvolume. Br. J. Nutr. 69:385.

Dirksen, G. U., H. G. Liebich, and E. Mayer. 1985.Adaptive changes of the ruminal mucosa and theirfunctional and clinical significance. Bovine Pract.20:116.

Erdman, R. A. 1988. Dietary buffering requirements of thelactating dairy cow: a review. J. Dairy Sci., 71:3246.

Gill, M., R. C. Siddons, and D. E. Beever. 1986. Metabolismof lactic acid isomers in the rumen of silage-fed sheep.55:399.

Hoover, W. H. 1986. Chemical factors involved in ruminalfiber digestion. J. Dairy Sci., 69:2755.

Jasaitis, D. K., J. E. Wohlt, and J. L. Evans. 1987. Influenceof feed ion content on buffering capacity of ruminantfeedstuffs in vitro. J. Dairy Sci. 70:1391.

Knowlton, K. F., M. S. Allen, and P. S. Erickson. 1996.Lasalocid and particle size of corn grain for dairy cowsin early lactation. 2. Effect on ruminal measurementsand feeding behavior. J. Dairy Sci. 79:565.

Mackie, R. I., and F. M. C. Gilchrist. 1979. Change inlactate producing and lactate utilizing bacteria inrelation to pH in the rumen of sheep during stepwiseadaptation to a high concentrate diet. Applied andEnviron. Micro. 38:422.

Nocek, J. E., and J. B. Russell. 1988. Protein and energy asan integrated system. Relationship of ruminal proteinand carbohydrate availability to microbial synthesis andmilk production. J. Dairy Sci. 71:2070.

Nocek, J. E., and S. Tamminga. 1991. Site of digestion ofstarch in the gastrointestinal tract of dairy cows and itseffect on milk yield and composition. J. Dairy Sci.74:3598.

Oba, M., and M. S. Allen. 2000. Effects of brown midrib 3mutation in corn silage on productivity of dairy cowsfed two concentrations of dietary neutral detergentfiber: 1. Feeding behavior and nutrient utilization. J.Dairy Sci 83:1333-1341.

Oba, M. and M. S. Allen. 2003. Effects of corn grainconservation method on feeding behavior andproductivity of lactating dairy cows at two dietarystarch concentrations. J. Dairy Sci. 86:174-183.

Russell, J. B. 2002. Rumen Microbiology and Its Role inRuminant Nutrition, Department of Microbiology, 157Wing Hall, Cornell University, Ithaca, NY 14853, USA.

Russell, J. B. and M. S. Allen. 1983. Physiological basis forinteractions among rumen bacteria: Streptococcus bovisand Megasphaera elsdenii as a model. Page 239 in:Current perspectives in microbial ecology. M. J. Klugand C. A. Reddy, ed. Am. Soc. for Microbiol.,Washington, DC.

Sander, E. G., R. G. Warner, H. N. Harrison, and J. K. Loosli.1959. The stimulatory effect of sodium butyrate andsodium propionate on the development of the rumenmucosa in the young calf. J. Dairy Sci. 42:1600-1605.

Shinozaki, K. 1959. Studies on experimental bloat inruminants. 5. Effects of various volatile fatty acidsintroduced into the rumen on the rumen motility.Tohoku J. Agric. Res. 9:237.

Terry, R. A., J. M. A. Tilley, and G. E. Outen. 1969. Effect ofpH on the cellulose digestion under in vitro conditions.J. Sci. Food Agric. 20:317.

Voelker, J.A. and M.S. Allen. 2003. Pelleted beet pulpsubstituted for high-moisture corn: 3. Effects on ruminalfermentation, pH, and microbial protein efficiency inlactating dairy cows. J. Dairy Sci. . 86:3562-3570.

Xu, J. and M. S. Allen. 1999. Effects of dietary lactosecompared with ground corn grain on the growth rate ofruminal papillae and rate of valerate absorption fromthe rumen. S. Afr. J. Anim. Sci. 29(ISRP2).

75

IntroductionThe multitude of disorders that dairy cows faceduring the transition to lactating is a perennial sourceof concern for dairy producers, nutritionists, andveterinarians. Total disease incidence in the severalweeks after parturition accounts for a substantialproportion of all morbidity on many dairies(Ingvartsen, 2006), with particularly high rates ofmastitis, metritis, milk fever, displaced abomasum,ketosis, and fatty liver, among other problems. Notsurprisingly, these issues have been the focus ofmuch research in recent decades. During that time,substantial progress has been made in some areas(e.g. milk fever); however, incidence of otherdisorders (e.g. displaced abomasum) may be on therise (Goff, 2006).

It is well-documented that cows suffering from onetransition disorder are at greater risk for contractingothers, including such seemingly unrelatedconditions as mastitis and ketosis (Goff, 2006). Thetransition from gestation to lactation dramaticallyincreases requirements for energy, glucose, aminoacids, and other nutrients in dairy cattle.Simultaneously, feed intake is often depressed. Theresulting negative energy balance suppressesimmune function and promotes metabolic disorders,potentially explaining relationships betweeninfectious and non-infectious transition disorders.

The most widely adopted practice to avoid metabolicdisorders is the nutritional management of prepartumcows to prevent excess body condition. By limitingthe pool of stored fat available for mobilization,restricting energy intake during the far-off dry periodlimits the increase in plasma non-esterified fatty acid(NEFA) concentrations during the transition period,resulting in lower fat storage and ketone production inthe liver (Murondoti et al., 2004; NRC, 2001).However, results of controlled trials have beeninconsistent with regard to nutritional management ofdry cows; some studies have demonstrated a benefitfrom increased prepartum energy intake when bodycondition was not affected (Doepel et al., 2002),whereas restricting intake, even without affecting bodycondition, led to more favorable outcomes in otherstudies (Holcomb et al., 2001). These inconsistenciessuggest that our understanding of metabolic disordersremains incomplete.

Recent research has highlighted the role ofinflammation in infectious diseases and hassuggested that inflammation is involved in metabolicdiseases as well. A key role for inflammation innumerous transition cow disorders may help toexplain links between these diverse conditions, andmay also improve our ability to predict and preventmetabolic problems in transition cows. Thispresentation will provide an overview of findingsrelating to the role of inflammation in transitiondisorders and provide recommendations to smooththe transition to lactation.

Inflammatory responses to infectionDuring infections such as mastitis or metritis,immune cells in the body recognize invadingpathogens and become activated. When the infectionis caused by Gram-negative bacteria, endotoxinreleased by the bacteria also activates immune cells.The activation of local and systemic host defensemechanisms requires cross-talk between numeroustypes of immune cells. One component of thisresponse is inflammation. The host of signalingmolecules released by activated immune cellsincludes inflammatory mediators such as nitric oxide,prostaglandins, and cytokines. While many of thesemolecules promote local inflammation and increasedblood flow, inflammatory cytokines play a key role instimulating systemic inflammatory responses,including increased body temperature, increasedheart rate, and decreased feed intake (Dantzer andKelley, 2007). Cytokines are able to alter manyphysiological systems because nearly all cell typesexpress cytokine receptors. Key inflammatorycytokines include tumor necrosis factor alpha (TNFα),interleukin 1β, and interleukin 6; these inflammatorycytokines act through many of the same signalingcascades and often produce similar responses in cells.

One effect of cytokines is to activate production ofacute phase proteins. Primarily produced by theliver, this class of proteins includes haptoglobin,serum amyloid A, and C-reactive protein. Proteinsthat participate in the acute phase response toinfection are generally found in very low abundancein the bloodstream, but are greatly elevated duringsystemic activation of the immune system. Theimportance of acute phase proteins in the response toinfection is somewhat unclear, but they have gained

Inflammation and TransitionCow Disorders

Barry BradfordDepartment of Animal Sciences & Industry, Kansas State University

127 Call Hall, Manhattan, KS 66506 [email protected]

76

widespread acceptance as markers of inflammation(Petersen et al., 2004). Other proteins are known asnegative acute phase proteins because theirconcentrations decline dramatically during the acutephase response.

It is clear that mammary and uterine infections resultin both local and systemic inflammation. Coliformmastitis results in release of endotoxin into thebloodstream and increased plasma concentrations ofcytokines and acute phase proteins (Hoeben et al.,2000). Likewise, metritis is associated with an acutephase response in transition cows (Huzzey et al.,2009); in fact, plasma haptoglobin is elevated prior toclinical signs of metritis.

Is there a role for inflammation inmetabolic disorders as well?Inflammation has been proposed as a missing link inthe pathology of metabolic disorders in transitioncows (Drackley, 1999). The metabolic effects of acutesystemic inflammation include adipose tissuemobilization, breakdown of liver glycogen, and livertriglyceride accumulation, all of which occur duringthe transition period. More specifically, cytokinespromote the breakdown of fat stores throughdecreased feed intake (Kushibiki et al., 2003),impaired insulin sensitivity, and direct stimulation oflipolysis (Kushibiki et al., 2001). All of theseconditions are associated with ketosis and fatty liverin dairy cattle (Ingvartsen, 2006). Even moreintriguing is the evidence that TNFα decreases liverglucose production (Kettelhut et al., 1987) andpromotes triglyceride accumulation once mobilizedNEFA reach the liver (García-Ruiz et al., 2006). Thedirect effects of cytokines on liver metabolism may play akey role in promoting metabolic disorders in transitioncows, especially those already combating infectiousdisorders or with excessive body condition.

With the physiological stress associated with calvingand the risk for infection that accompanies bothcalving and the initiation of lactation, immuneresponses are common during the transition period.Abrupt dietary shifts during the transition period canalso contribute to systemic inflammation. Cows aregenerally fed diets with greater energy density at theonset of lactation, and if this change is too dramatic,it can result in ruminal production of endotoxin andsubsequent transfer of endotoxin into thebloodstream (Khafipour et al., 2009). Furthermore,monocytes are known to become more responsive toinflammatory stimulants during the transition period,resulting in greater secretion of inflammatorycytokines when stimulated (Sordillo et al., 1995).Mastitis, metritis, and acute acidosis can thereforeresult in systemic inflammation, elevated cytokineconcentrations, and altered liver metabolism.

Recent findings have supported previous speculationregarding the relationships between inflammatorymediators and metabolic disorders. Ametaj andcoworkers reported that plasma concentrations of anumber of inflammatory markers were increased incows that developed fatty liver (Ametaj et al., 2005).Similar findings were reported by Ohtsuka andcolleagues, who observed increased serum TNFαactivity in cows with moderate to severe fatty liver(Ohtsuka et al., 2001). Most recently, endotoxin-induced mastitis was shown to alter expression ofmetabolic genes in the liver, including decreasedexpression of genes important for glucose production(Jiang et al., 2008). In lactating cows, impairedglucose production would likely lead to increasedadipose tissue mobilization, elevated plasma NEFA,and increased ketone production by the liver.

A retrospective study of cows on 3 commercial Italiandairies suggests that liver inflammation is associatedwith a problematic transition to lactation (Bertoni etal., 2008). Cows were classified in quartiles fordegree of liver inflammation based on plasmaconcentrations of acute phase proteins. Those cowswith the strongest inflammatory profiles were at 8-fold greater risk for experiencing 1 or more transitiondisorders, had lower plasma calcium concentrations,took longer to re-breed, and produced less milk in thefirst month of lactation (Bertoni et al., 2008).

Relationships between oxidative stress andinflammationAlthough the importance of inflammation intransition disorders is becoming clear, the pathwaysthat cause this inflammation are less clear. Infectionscertainly initiate the process in some cows, but this isnot likely the cause of metabolic disorders in allcows. In particular, the dramatically higher incidenceof transition disorders in cows with excessive bodycondition (Morrow, 1976) is difficult to attributeexclusively to infections.

In addition to acute inflammatory events, chroniclow-grade inflammation may play a role in transitiondisorders. In the early 1990’s, it was discovered thatadipose tissue is capable of producing inflammatorycytokines such as TNFα (Hotamisligil et al., 1993).With the extensive list of “adipokines” discovered inthe ensuing 15 years, human metabolic disorders areincreasingly being viewed as products of low-gradeadipose tissue inflammation induced by obesity.Adipose tissue is now recognized as an importantsource of circulating TNFα, and plasma TNFαconcentrations are increased in obese individuals in anumber of species, including sheep (Daniel et al.,2003). Based on these findings, infection is no longera required component of an inflammation-basedetiology for metabolic disorders in the transition

77

period. Low-grade inflammation associated withobesity may help to explain “fat cow syndrome”(Morrow, 1976).

Lipid peroxides are also emerging as likely mediatorslinking plasma lipids to inflammation (Pessayre et al.,2004). Lipid peroxides are produced whenintracellular lipids encounter reactive oxygen species(ROS) such as hydrogen peroxide. Some ROS arealways produced in the liver; however, eventsoccurring in early lactation likely contribute toenhanced ROS production. One adaptation toincreasing delivery of NEFA to the liver in earlylactation is an increase in the capacity of peroxisomaloxidation (Grum et al., 1996), an alternative pathwayfor fatty acid oxidation. Enhanced peroxisomaloxidation increases total oxidative capacity of the cell,but the first step in this pathway produces hydrogenperoxide rather than NADH (Drackley, 1999), andtherefore it contributes to ROS production to agreater extent than mitochondrial oxidation.

Increased ROS production in early lactation cows,coupled with increased NEFA concentration,increases lipid peroxide formation; both the transitionto lactation and high body condition are associatedwith increased plasma markers of lipid peroxidation(Bernabucci et al., 2005). Lipid peroxides activateinflammatory cascades, which in turn alter nutrientmetabolism (Pessayre et al., 2004). In addition, ROSare especially harmful to immune cells and candecrease the ability of the immune system to respondto infections (Spears and Weiss, 2008).

In summary, a new model is emerging to explain thedevelopment of numerous transition disorders. Acombination of insults, including infection, chronicinflammation in obese cows, and lipid peroxideformation, promotes systemic inflammation duringthe transition period. Inflammation impairs immunefunction, making cows more susceptible to infectiousdisorders, and causes maladaptive shifts inmetabolism, increasing the risk of metabolicdisorders.

Potential interventions An inflammation-based understanding of transitiondisorders opens the door for novel strategies toaddress these problems. The complex interactions ofoxidative stress, inflammatory cascades, andmetabolic pathways allow for a broad array ofpotential treatments to prevent transition disorders,including antioxidants, metabolic modifiers, and anti-inflammatory drugs.

Antioxidants. Dietary antioxidants, notably vitamin Eand selenium, are important for their ability tocontribute to ROS neutralization, thereby impeding

the progression toward inflammation. Interestingly,plasma concentrations of a-tocopherol (vitamin E)decrease through the transition period (Weiss et al.,1990), and low antioxidant status is associated withtransition cow disorders (LeBlanc et al., 2004;Mudron et al., 1997). Supplementing vitamin Eprepartum improves antioxidant status (Weiss et al.,1990). Given the importance of antioxidants inmodulating inflammation, it is not surprising thatmultiple studies have shown that supplementingvitamin E in excess of traditional recommendationsdecreases the incidence and severity of clinicalmastitis (Smith et al., 1984; Weiss et al., 1990).Recently, a meta-analysis showed that supplementalvitamin E is also effective at preventing retainedplacenta (Bourne et al., 2007).

Low plasma vitamin E concentrations are associatedwith increased incidence of fatty liver and displacedabomasum (Mudron et al., 1997). Surprisingly, nopublished studies have evaluated the effects ofsupplemental vitamin E on liver metabolism orincidence of metabolic disorders. Given thatsupplemental vitamin E can decrease inflammatorycytokine production (Poynter and Daynes, 1998) andimprove liver antioxidant status in mice with fattyliver (Soltys et al., 2001), supplemental vitamin E mayimprove liver function in transition cows. With itsdemonstrated effects on immune function and itspotential to benefit liver function, it is recommendedthat vitamin E be supplemented at a rate of at least1,500 IU/day for close-up dry cows.

Selenium is the other most important dietaryantioxidant in dairy rations. Although responses toselenium are most dramatic when vitamin E status ismarginal, selenium has unique roles in ROSneutralization and must be considered independentlyto achieve optimal health. The FDA restrictsselenium supplementation in dairy rations to 0.3ppm, and most farms supplement at that level,limiting the attention paid to selenium in transitionhealth strategies. Feeding selenium yeast rather thaninorganic selenium sources is a common and effectivemeans of increasing selenium status of animals thatalready receive the legal limit of selenium (Salman etal., 2009). However, most evidence suggests athreshold response to selenium; once a minimumplasma concentration is reached, there may be nobenefit of further increases, and in many cases, thatthreshold seems to be reached with inorganicselenium sources (Spears and Weiss, 2008).Nevertheless, the use of organic selenium may beworth considering in areas with selenium-deficientsoils, such as the Great Lakes region.

Beta carotene, a precursor of vitamin A, can alsofunction as an antioxidant (Spears and Weiss, 2008),

78

and concentrations of both vitamin A and β-carotenetypically decrease during the transition period(LeBlanc et al., 2004). Although supplementingvitamin A at concentrations above currentrecommendations has improved udder health insome studies (NRC, 2001), in a head-to-headcomparison, supplementation of β-carotene duringthe transition period significantly decreased incidenceof both metritis and retained placenta compared tovitamin A supplementation(Michal et al., 1994).Cows fed 600 mg/day of β-carotene had equivalentplasma retinol concentrations to those supplementedwith 120,000 IU/day of vitamin A (Michal et al.,1994). Replacing vitamin A supplements in transitionrations with relatively high concentrations of β-carotene may be beneficial for transition cow health.

Metabolic modifiers. Agonists for peroxisomeproliferator-activated receptors (PPAR) can improveliver metabolism through several mechanisms.PPARg agonists (primarily targeting peripheralorgans) can decrease plasma NEFA concentration,whereas those targeting PPARa (the primary isoformin liver) promote fatty acid oxidation in liver, limitingtriglyceride accumulation and production of lipidperoxides (Kota et al., 2005). One PPARg agonist(2,4-thiazolidinedione) has been evaluated intransition cows, with positive effects on metabolichealth (Smith et al., 2007). Unfortunately, PPARagonists are unlikely to be approved for use ondairies in the near future.

Choline is a nutrient that may limit oxidative stress,although like PPAR agonists, it does not directlyneutralize ROS. Rather, choline likely limits lipidperoxide formation by decreasing plasma NEFAconcentration and promoting clearance oftriglycerides from the liver (Cooke et al., 2007). As aresult, supplemental rumen-protected choline hasbeen shown to increase plasma a-tocopherolconcentration during the transition period (Pinotti etal., 2003), presumably contributing to immunefunction and modulation of inflammation.

Anti-inflammatory agents. Direct inhibition ofinflammation through the use of non-steroidal anti-inflammatory drugs (NSAIDs) has shown promisefor treatment of metabolic disorders in laboratoryanimals. Indomethacin prevented hypoglycemiaafter administration of inflammatory cytokines(Kettelhut et al., 1987) and mice with induced fattyliver had decreased liver triglyceride content whentreated with two NSAIDs (Yu et al., 2006), amongother findings. Low blood glucose and fatty liver arerelated problems that many early lactation cows face,and these findings in rodents suggest that NSAIDsmay be useful in transition cows, as well.

The use of NSAIDs to combat transition cowdisorders has produced mixed results to date.Numerous NSAIDs have been evaluated for use inthe treatment of mastitis, and in general they areeffective at reducing body temperatures, but do notappear to decrease the severity of the infection(Morkoc et al., 1993). However, carprofen was shownto partially alleviate the decrease in ruminalcontractions during mastitis (Vangroenweghe et al.,2005), which could help prevent a subsequentdisplaced abomasum. Likewise, one study indicatedthat uterine involution was accelerated by flunixinmeglumine treatment for metritis (Amiridis et al.,2001), but another showed no beneficial effects, eithersystemically or in the reproductive tract (Drillich etal., 2007).

It is perhaps not surprising that anti-inflammatorytreatments have not consistently improved recoveryfrom infection; after all, inflammation is a keycomponent of the immune system’s attempt to fightoff the invading pathogen. The use of NSAIDs mayhave more promise for combating inflammation inmetabolic disorders, where it provides no obviousbenefits.

Only 2 published studies were found in whichNSAIDs were administered to transition cows priorto diagnosis of any disease. In the first study, cowstreated with acetyl-salicylate (aspirin) for the first 5days of lactation had significantly lower plasmaconcentrations of acute phase proteins and tended tohave greater peak milk production than controls(Bertoni et al., 2004). Conversely, a recent publicationshowed no benefit to administration of flunixinmeglumine for the first 3 days of lactation (Shwartzet al., 2009). In fact, this treatment depressed feedintake and milk yield over the first week of lactation.However, both of these studies were far too small(< 15 cows per treatment) to assess effects onincidence of transition disorders. Also, it is importantto note that while NSAIDs are often lumped together,different compounds have different modes of action,side effects, and clearance rates that together make itimportant to choose NSAIDs carefully when they areevaluated for effects beyond analgesia. Futureresearch in this area may provide for opportunities toconsider short-term NSAID treatment as apreventative measure against transition disorders.

ConclusionsGrowing evidence suggests that inflammation maybe a key factor in the development of many transitiondisorders. Because it results in suppressed immunefunction and altered nutrient metabolism,inflammation may provide a novel link betweeninfectious and metabolic disorders that are commonduring the transition period. This model suggests

79

that dietary antioxidants in dry cow rations should bere-evaluated on farms struggling with transition cowdisorders. Additional steps, such as theincorporation of β-carotene or rumen-protectedcholine, may also help to prevent oxidative stress andsubsequent inflammation. Future research mayprovide additional tools to directly combatinflammation in transition cows. Hopefully,continued progress on the pathology of transitiondisorders will help dairy producers to decrease thenumber of early lactation cows leaving the dairyherd.

ReferencesAmetaj, B. N., B. J. Bradford, G. Bobe, R. A. Nafikov, Y. Lu,

J. W. Young, and D. C. Beitz. 2005. Strong relationshipsbetween mediators of the acute phase response andfatty liver in dairy cows. Can. J. Anim. Sci. 85(2):165-175.

Amiridis, G. S., L. Leontides, E. Tassos, P. Kostoulas, and G.C. Fthenakis. 2001. Flunixin meglumine acceleratesuterine involution and shortens the calving-to-first-oestrus interval in cows with puerperal metritis. J. Vet.Pharmacol. Ther. 24(5):365-367.

Bernabucci, U., B. Ronchi, N. Lacetera, and A. Nardone.2005. Influence of body condition score on relationshipsbetween metabolic status and oxidative stress inperiparturient dairy cows. J. Dairy Sci. 88(6):2017-2026.

Bertoni, G., E. Trevisi, X. Han, and M. Bionaz. 2008. Effectsof inflammatory conditions on liver activity inpuerperium period and consequences for performancein dairy cows. J. Dairy Sci. 91(9):3300-3310.

Bertoni, G., E. Trevisi, and F. Piccioli-Cappelli. 2004. Effectsof acetyl-salicylate used in post-calving of dairy cows.Vet. Res. Commun. 28(0):217-219.

Bourne, N., R. Laven, D. C. Wathes, T. Martinez, and M.McGowan. 2007. A meta-analysis of the effects ofvitamin E supplementation on the incidence of retainedfoetal membranes in dairy cows. Theriogenology67(3):494-501.

Cooke, R. F., N. S. Del Rio, D. Z. Caraviello, S. J. Bertics, M.H. Ramos, and R. R. Grummer. 2007. Supplementalcholine for prevention and alleviation of fatty liver indairy cattle. J. Dairy Sci. 90(5):2413-2418.

Daniel, J. A., T. H. Elsasser, C. D. Morrison, D. H. Keisler, B.K. Whitlock, B. Steele, D. Pugh, and J. L. Sartin. 2003.Leptin, tumor necrosis factor-_ (TNF), and CD14 inovine adipose tissue and changes in circulating TNF inlean and fat sheep. J. Anim Sci. 81(10):2590-2599.

Dantzer, R. and K. W. Kelley. 2007. Twenty years of researchon cytokine-induced sickness behavior. Brain. Behav.Immun. 21(2):153-160.

Doepel, L., H. Lapierre, and J. J. Kennelly. 2002. Peripartumperformance and metabolism of dairy cows in responseto prepartum energy and protein intake. J. Dairy Sci.85(9):2315-2334.

Drackley, J. K. 1999. ADSA foundation scholar award.Biology of dairy cows during the transition period: thefinal frontier? J. Dairy Sci. 82(11):2259-2273.

Drillich, M., D. Voigt, D. Forderung, and W. Heuwieser.2007. Treatment of acute puerperal metritis withflunixin meglumine in addition to antibiotic treatment.J. Dairy Sci. 90(8):3758-3763.

García-Ruiz, I., C. Rodríguez-Juan, T. Díaz-Sanjuan, P. delHoyo, F. Colina, T. Muñoz-Yagüe, and J. A. Solís-Herruzo. 2006. Uric acid and anti-TNF antibodyimprove mitochondrial dysfunction in ob/ob mice.Hepatology 44(3):581-591.

Goff, J. P. 2006. Major advances in our understanding ofnutritional influences on bovine health. J. Dairy Sci.89(4):1292-1301.

Grum, D. E., J. K. Drackley, R. S. Younker, D. W. LaCount,and J. J. Veenhuizen. 1996. Nutrition during the dryperiod and hepatic lipid metabolism of periparturientdairy cows. J. Dairy Sci. 79(10):1850-1864.

Hoeben, D., C. Burvenich, E. Trevisi, G. Bertoni, J. Hamann,R. Buckmaier, and J. W. Blum. 2000. Role of endotoxinand TNF- alpha in the pathogenesis of experimentallyinduced coliform mastitis in periparturient cows. J.Dairy Res. 67(4):503-514.

Holcomb, C. S., H. H. Van Horn, H. H. Head, M. B. Hall,and C. J. Wilcox. 2001. Effects of prepartum dry matterintake and forage percentage on postpartumperformance of lactating dairy cows. J. Dairy Sci.84(9):2051-2058.

Hotamisligil, G. S., N. S. Shargill, and B. M. Spiegelman.1993. Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance.Science 259(5091):87-91.

Huzzey, J. M., T. F. Duffield, S. J. LeBlanc, D. M. Veira, D.M. Weary, and M. A. G. von Keyserlingk. 2009. Shortcommunication: Haptoglobin as an early indicator ofmetritis. J. Dairy Sci. 92(2):621-625.

Ingvartsen, K. L. 2006. Feeding- and management-relateddiseases in the transition cow: Physiologicaladaptations around calving and strategies to reducefeeding-related diseases. Animal Feed Science andTechnology 126(3-4):175-213.

Jiang, L., P. Sorensen, C. Rontved, L. Vels, and K.Ingvartsen. 2008. Gene expression profiling of liverfrom dairy cows treated intra-mammary withlipopolysaccharide. BMC Genomics 9(1):443.

Kettelhut, I. C., W. Fiers, and A. L. Goldberg. 1987. Thetoxic effects of tumor necrosis factor in vivo and theirprevention by cyclooxygenase inhibitors. PNAS84(12):4273-4277.

Khafipour, E., D. O. Krause, and J. C. Plaizier. 2009. A grain-based subacute ruminal acidosis challenge causestranslocation of lipopolysaccharide and triggersinflammation. J. Dairy Sci. 92(3):1060-1070.

Kota, B. P., T. H.-W. Huang, and B. D. Roufogalis. 2005. Anoverview on biological mechanisms of PPARs.Pharmacol. Res. 51(2):85-94.

Kushibiki, S., K. Hodate, H. Shingu, Y. Obara, E. Touno, M.Shinoda, and Y. Yokomizo. 2003. Metabolic andlactational responses during recombinant bovine tumornecrosis factor-_ treatment in lactating cows. J. DairySci. 86(3):819-827.

Kushibiki, S., K. Hodate, H. Shingu, Y. Ueda, Y. Mori, T.Itoh, and Y. Yokomizo. 2001. Effects of long-termadministration of recombinant bovine tumor necrosisfactor alpha on glucose metabolism and growthhormone secretion in steers. Am. J. Vet. Res. 62(5):794-798.

LeBlanc, S. J., T. H. Herdt, W. M. Seymour, T. F. Duffield,and K. E. Leslie. 2004. Peripartum serum vitamin E,retinol, and beta-carotene in dairy cattle and theirassociations with disease. J. Dairy Sci. 87(3):609-619.

80

Michal, J. J., L. R. Heirman, T. S. Wong, B. P. Chew, M.Frigg, and L. Volker. 1994. Modulatory effects of dietary{beta}-carotene on blood and mammary leukocytefunction in periparturient dairy cows. J. Dairy Sci.77(5):1408-1421.

Morkoc, A. C., W. L. Hurley, H. L. Whitmore, and B. K.Gustafsson. 1993. Bovine acute mastitis: Effects ofintravenous sodium salicylate on endotoxin-inducedintramammary inflammation. J. Dairy Sci. 76(9):2579-2588.

Morrow, D. A. 1976. Fat cow syndrome. J. Dairy Sci.59(9):1625-1629.

Mudron, P., J. Rehage, H. P. Sallmann, M. Mertens, H.Scholz, and G. Kovac. 1997. Plasma and liver alpha-tocopherol in dairy cows with left abomasaldisplacement and fatty liver. Zentralbl. Veterinarmed. A44(2):91-97.

Murondoti, A., R. Jorritsma, A. C. Beynen, T. Wensing, andM. J. H. Geelen. 2004. Unrestricted feed intake duringthe dry period impairs the postpartum oxidation andsynthesis of fatty acids in the liver of dairy cows. J.Dairy Sci. 87(3):672-679.

NRC. 2001. Nutrient Requirements of Dairy Cattle. 7th rev.ed. National Research Council. Natl. Acad. Sci.,Washington, DC.

Ohtsuka, H., M. Koiwa, A. Hatsugaya, K. Kudo, F. Hoshi,N. Itoh, H. Yokota, H. Okada, and S. Kawamura. 2001.Relationship between serum TNF activity and insulinresistance in dairy cows affected with naturallyoccurring fatty liver. J. Vet. Med. Sci. 63(9):1021-1025.

Pessayre, D., B. Fromenty, and A. Mansouri. 2004.Mitochondrial injury in steatohepatitis. Eur. J.Gastroenterol. Hepatol. 16(11):1095-1105.

Petersen, H. H., J. P. Nielsen, and P. M. Heegaard. 2004.Application of acute phase protein measurements inveterinary clinical chemistry. Vet. Res. 35(2):163-187.

Pinotti, L., A. Baldi, I. Politis, R. Rebucci, L. Sangalli, and V.Dell'Orto. 2003. Rumen-protected cholineadministration to transition cows: effects on milkproduction and vitamin E status. J. Vet. Med. A50(1):18-21.

Poynter, M. E. and R. A. Daynes. 1998. Peroxisomeproliferator-activated receptor alpha activationmodulates cellular redox status, represses nuclearfactor-kappa B signaling, and reduces inflammatorycytokine production in aging. J. Biol. Chem.273(49):32833-32841.

Salman, S., A. Khol-Parisini, H. Schafft, M. Lahrssen-Wiederholt, H. W. Hulan, D. Dinse, and J. Zentek. 2009.The role of dietary selenium in bovine mammary glandhealth and immune function. Animal Health ResearchReviews (Epub ahead of print):1-14.

Shwartz, G., K. L. Hill, M. J. VanBaale, and L. H. Baumgard.2009. Effects of flunixin meglumine on pyrexia andbioenergetic variables in postparturient dairy cows. J.Dairy Sci. 92(5):1963-1970.

Smith, K. L., J. H. Harrison, D. D. Hancock, D. A.Todhunter, and H. R. Conrad. 1984. Effect of vitamin Eand selenium supplementation on incidence of clinicalmastitis and duration of clinical symptoms. J. Dairy Sci.67(6):1293-1300.

Smith, K. L., S. E. Stebulis, M. R. Waldron, and T. R.Overton. 2007. Prepartum 2,4-thiazolidinedione altersmetabolic dynamics and dry matter intake of dairycows. J. Dairy Sci. 90(8):3660-3670.

Soltys, K., G. Dikdan, and B. Koneru. 2001. Oxidative stressin fatty livers of obese Zucker rats: Rapid ameliorationand improved tolerance to warm ischemia withtocopherol. Hepatology 34(1):13-18.

Sordillo, L. M., G. M. Pighetti, and M. R. Davis. 1995.Enhanced production of bovine tumor necrosis factor-alpha during the periparturient period. Vet. Immunol.Immunopathol. 49(3):263-270.

Spears, J. W. and W. P. Weiss. 2008. Role of antioxidants andtrace elements in health and immunity of transitiondairy cows. The Veterinary Journal 176(1):70-76.

Vangroenweghe, F., L. Duchateau, P. Boutet, P. Lekeux, P.Rainard, M. J. Paape, and C. Burvenich. 2005. Effect ofcarprofen treatment following experimentally inducedEscherichia coli mastitis in primiparous cows. J. DairySci. 88(7):2361-2376.

Weiss, W. P., J. S. Hogan, K. L. Smith, and K. H. Hoblet.1990. Relationships among selenium, vitamin E, andmammary gland health in commercial dairy herds. J.Dairy Sci. 73(2):381-390.

Yu, J., E. Ip, A. dela Peña, J. Y. Hou, J. Sesha, N. Pera, P.Hall, R. Kirsch, I. Leclercq, and G. C. Farrell. 2006.COX-2 induction in mice with experimental nutritionalsteatohepatitis: Role as pro-inflammatory mediator.Hepatology 43(4):826-836.

81

IntroductionBefore we start discussing ‘putting the pieces backtogether’ it is my true hope that there is nothing toput back together. Many herds have used sounddecision making practices and continued to feedingredients that provide a solid return on investment,even in low milk prices. Hopefully you were able tostand pat with your rations during this historic pricedecline and were able to demonstrate that yourrations were already optimally balanced. But if youwere like many of us that occasionally suffer from arare bone disorder, softbackboneitis, please keepreading.

We all know that cutting back on properly balancedrations will hurt future health and milk production.However, you may have observed several cases ofthis disorder at the same time dairy producers havebegun to inform their consultants that they are losinghundreds of thousands of dollars per month. I knowthat most attendees of this conference did not sufferfrom this disease themselves but if you saw theneighboring nutritionist succumb you may be askedto help their clients in the interim. In either case thismight be the rare opportunity to start with a ‘cleanslate’ and put rations back together from scratch.

The objective of this article and talk will be to discussstrategies of how to approach this opportunity. Thispaper should help you prioritize how to determinewhat goes back in the ration and when. I will discussthe strategy in general and not get into specificrecommendations. For specific suggestions pleasesee some of the excellent reviews on this topic(Hutjens, 2003; Hutjens, 2008; Kung, 2005).

First the BasicsWith a potential ‘clean slate’ to work with, the beststrategy is to keep things simple and focused. Thismeans finding areas that return the most profit toyour farms the quickest. In this sense I like to reflecton Covey Principles as a guide. For this situationthese include:

1. Focusing on the non-urgent but important. Thisincludes items like planning, reading andtraining (i.e., things that are important but wenever get to because they are not urgent).

2. On the dairy these are big rocks like:a. Prevention vs. problem solvingb. Planning vs. trouble shootingc. Feed Management vs. paper rationsd. Finding bottlenecks vs. stopping the bleeding

3. While there are many inputs that drive decisionmaking, the one main thing we all need toremember is that our golden goose is the cow.We need to continually work on keeping cowscomfortable, healthy and able to respond tobetter rations. Stating the obvious, this means:a. Free access to a healthy, well balanced

consistent and nutritious ration at all timesb. Free access to abundant, fresh, clean waterc. Free access to an unencumbered, well

groomed, clean, dry, bed in a pen that is notover crowded.

d. Adequate time during the day to lie down,chew her cud and make milk.

General StrategiesMy first recommendation regardless of the specificitem you choose to evaluate first is to always baseyour decisions on science. This means usingingredients that are proven with objective, ‘peerreviewed’ research. It also means using ingredientswith low risk relative to their return on investment.Risk comes from uncertainty which can be due tovariability in the ingredient itself, variability in theanimal’s response to the ingredient and uncertaintyof the true response to the ingredient due to limitedscientific research.

Feed Management StrategiesIn many cases it’s not what’s in the ration thatmatters most but how the ration is prepared anddelivered that will impact herd profitability thefastest. Below are a few items to consider even beforechanging the ration.

If the herd is still feeding a ‘one group ration’ it maybe time to move back to feeding multiple lactationrations.

By mixing more rations you will be able to add itemsback to the ration that work better in one phase oflactation or one group of animals and hold back onsome of the more expensive feed ingredients that do

Picking Up the Pieces After a Down MilkMarket: Strategies for Deciding What Goes

Back in the RationBill Sanchez, Ph.D.

Diamond V Mills, Inc., [email protected]

82

not work across the entire herd. Michigan Stateresearchers are finding ways to reduce ration costswith carbohydrate feeding strategies in herds thatfeed multiple lactation TMR’s.

Manage expensive push outs. If you can push outfeed more often you can blend some of it back intothe low pen rations rather than feeding it to theheifers. Test how much push out can be blendedback in to the later lactation pen diets withoutdropping milk from those pens.

Clean out the feed bunk daily especially in thesummer, as feed heats quickly in the feed bunk andcan decrease feed intake and milk production.

Audit the TMR. Find and eliminate the sources ofvariation. This includes the consistency of the mixwhich can often be improved by changing knives andreplacing shoe pads in the TMR mixer wagon. Use aPenn State Shaker Box to know for sure howconsistent the ration is; and be sure to check thebeginning, middle and end of the load as well as howit changes throughout the day.

Checking inventories will be even more critical in thesecond half of this year. It also may be time to lookvery closely at inventory shrink. If there are ways tocut shrink this means instant cash flow.

All of the above will help you monitor dry matterintake more closely – this is the key to measuring,monitoring and improving the efficiency of the drymatter fed.

Monitoring feed efficiency. A realistic goal for mostherds is 1.4 to 1.6 (calculated as milk or fat-correctedmilk divided by DMI).

But don’t stop with feed efficiency, use income overfeed costs as your final measure. This equates to netprofit from the feeding program.

Finally, over the longer term, consider a riskmanagement strategy for feed and milk. It may bethe best strategy to do nothing when it comes tolocking up feed costs and milk prices, however, makesure this decision is based on a strategy rather thanfear or lack of knowledge. Many producers that wework with locked in at least a break-even this year.Imagine the advantage they have today. If you arenew to feed and milk contracting, considerimplementing a ‘mock’ strategy in the next fewmonths to better understand the process beforeimplementing a system with real money. That waymuch of the fear will be taken out of the strategywhen you are ready to start.

Ration StrategiesNow that we have the basics and general feedmanagement strategies covered we can focus on themajor ration strategies. These include:

Essential nutrients. In this environment there will belots of herds that chose to focus on short-term profitsfor survival. In many of these herds it is important toimprove overall herd health before they will be ableto increase milk production.

1. Nutrients essential to health are first to get backin the ration.

2. These include any nutrient that plays a directrole in immunity.

3. I won’t insult you and list them all here but weall know that an unhealthy cow will not be ableto produce milk to her potential, even with aperfect ration.

Energy is an obvious ration addition that is likely tobe first limiting.

1. This is because many herds may have used thestrategy of letting the cow milk it off her backduring these tough times.

2. Now we have to get that condition back on thecows to keep them profitable through theupcoming summer heat stress and fall calvingseason.

3. Look for thinner late lactation and fresh cows tosupport this first step.

4. Herds with several different transition disordersmay simply be lacking in energy whichcompromises the immune system.

5. When looking at feed ingredients for energyremember that much of the energy comes from awell functioning rumen. Rumen fermentation ofstarch, sugars and fiber supply up to 85 percentof the cow’s energy.

6. Useful tool to compare the value of nutrientsand energy include Feedval, from the Universityof WI, and Sesame and Ping Pong from TheOhio State University. Often times there arebetter buys on ingredients if you can calculatethe relative nutrient values. Feedval and Sesamesoftware can help with this calculation. Pingpong helps to determine how many ingredientsamples are needed to control variationultimately making a more consistent TMR.

Forages. Excellent forage quality is the foundation ofgood rations.

1. Forage quality is one of the highest priorities. 2. If there is no way to purchase higher quality

forages, there are lots of ways to improve thequality of the forage now being fed.

83

3. This is the time to make sure the silage face isclean and unspoiled, moldy silage is not fed tolactating cows and hay is mixed and blendedinto the ration correctly.

4. It is also a time to know the DM content of everyingredient being fed. This requires checking drymatters on all silages and other high moistureingredients on a daily or at least weekly basisand adjusting rations accordingly.

Additives. This is where things get interesting.Hutjens (2003) describes feed additives as a group offeed ingredients that can cause a desired animalresponse in a non-nutrient role such as pH shift,growth or metabolic modifier. In general, many ofthese non-nutrient additives are in the ration becausewith the types of diets fed today it is impossible tofeed cows adequate nutrients to match theirpotential. Despite significant advances in dairynutrition, high producing cows in nearly every herdtoday have an assortment of digestive, metabolic andimmune challenges. Supplementing “feed additives”in dairy rations have been found to be effective inmanaging various challenges. We are still in ourinfancy of understating how nutrition interacts withthe immune system but there is certainly arelationship. The interesting and unfortunate part ofthis is that there are several products marketed thatsupposedly cure every known ailment and evenbring cows back to life. Because of this manyproducers are resistant to the whole class of ‘feedadditives’ and fail to use products that truly areeffective and provide a consistent profitable response.

Mechanism of action. To determine if a feedadditives offers true benefits it is important tounderstand how it might work before determining ifit might work. There are some feed additives thatwork even without a challenge. These are typicallyassociated with increased digestion. There is a largemultiplier effect when a small addition of aningredient can affect the digestibility of the entire dietor even just the forage component of the diet. I liketo think of these additives as higher quality hay. Fora dairy producer it is always an easy decision to buythe higher quality hay because of the improveddigestibility, improved intake and associated milkproduction.

These ‘digestibility’ additives should improve feedefficiency and they may or may not affect intake.This appears to depend on the stage of lactation.Sorting out whether or not these types of additiveswork by affecting rumen microbial populations ormore directly on the feed is important to evaluatingthem. This is because there is often limitingperformance data on these products and they willhave to be evaluated on in-vitro responses. Without

lactation studies it makes it more difficult to evaluatethe economic impact and real value of some of these.

Other types of additives have a metabolic or post-absorptive role. These work more like nutrients.They may not be essential in all situations, but theymay provide a real response during certain metabolicchallenges (i.e., excessive fat mobilization due to overconditioned cows).

Evaluating feed additives. Once we know how afeed additive might work we also need data showingthat it really does work. Too often product data isbased on marketing information rather than peerreviewed objective scientific data.

Dr. Mike Hutjens from the University of Illinois hasput together an excellent outline of ways to evaluatefeed additives. A review of current research on theseproducts, as well as a table that outlines additives insix categories, can be found online at:www.livestocktrail.uiuc.edu/dairynet/paperDisplay.cfm?ContentID=9999. In addition to some of thecriteria mentioned in this article Dr. Hutjens discusseswhat he refers to as the “me too syndrome” a termreferring to a product that has limited research, butmarketed on the concept that their product is similaror identical to the industry base standard additive.He recommends asking for product-specific researchdata to determine if the ‘me too’ feed additive hasbeen evaluated in controlled studies.

In addition to product specific research I recommendevaluating the actual product manufacturing andquality control of the manufacturing procedures.Companies with good manufacturing practices willtypically allow visits to their plants and take greatpride in showing off their manufacturing excellenceand quality control procedures. These companiestypically produce a much more consistent product.In contrast, some companies have their productsmanufactured in foreign countries that probably leadto less consistent products.

Another problem associated with evaluating feedadditives is what I call overlapping technologies. Weall have listened to the producer state that ‘if I fedevery additive out there my herd would be at 200 lbsof milk’. This statement is that it assumes allproducts are independent. The reality is they are not.Many work in the same way. Adding the firstadditive may indeed provide a response, yet thesecond one probably will not given that the first oneis already in the diet. As nutritionists, our job is tofirst feed the additive that works the best and has thegreatest return on investment. Again, it comes backto adequate scientific data to be able to make thisdecision.

84

Some of the answers can be gotten from in-vitrostudies, however, there needs to be a complete set ofwhole animal studies to evaluate the productcompletely. There should also be evidence thatresponse is repeatable. A single observation from aspecific research trial may not be applicable todifferent situations as product response could bedependent on nutritional, physiological, and/orenvironmental conditions. And for lactating cowsthere should be actual lactating cow data on theproducts. The data should be peer reviewed andpublished in scientific journals, hopefully conductedon US Universities with high producing dairy cows.

Once you have decided that there is enough scientificresearch supporting a plausible mechanism of actionand an animal response we next need to know if theproduct provides a return on investment. In thisevaluation, focus should not be just on price. Ifimproved milk production is the measurableresponse, comparisons should be made between thecost of an additive and the value of increased milkproduction. The target for return on investmentshould be at least 2:1. Health and reproductivebenefits are more difficult to evaluate, but could bevery profitable. These types of data are limited sobuyer be ware.

Because all of the above criteria are often limiting formany of the products on the market many are soldbased on farm trial data only. In my opinion this isunacceptable. It’s very challenging to evaluate anadditive on farm. The normal variation on a herd isoften greater than the expected response from theadditive. Because of the dynamics of farmoperations, keeping the test conditions under controland unbiased is very difficult.

Despite the difficulty of researching a product onfarm, producers have to make decisions on theeffectiveness of products to ensure they are getting areturn. Scientific data from peer reviewedpublications combined with on farm economicsshould always be the primary method of evaluation.Here are a few additional thoughts on this subject.

Understand the lag in response to a productassociated with both its introduction and removalfrom the diet. Most feed additives take time to causea biological response. In some cases this lag can be afew days to weeks in other cases it can be severalmonths. I remember during the last milk decline thatwe had a dairy producer saved 60 cents per cow byswitching to a competing nutritionist with less‘goodies’ in the diet. The herd actually went up acouple of lbs of milk after the first few months. Theywere extremely happy with the decision. However,eight months after the change they were down 3

points of milk fat and ten lbs of milk (and not sohappy).

Often we do not have time to sort through dataourselves so we turn to reviews of products andtechnologies. One problem with relying completelyon reviews is that broad categories of products arepooled together in the same review. In many casesthe actual products being pooled together are verydifferent with one type being effective and anothertype not so much. Due to pooling them together theclass of products may not look as good. Anotherproblem with evaluating categories of productstogether is that there may not be enough data toseparate out the differences that may truly existbetween them. For categories of products that thereviewer recommends we would naturally choose theless expensive product in the class which may in factbe different leaving us with no response in the field.

Regarding the review of studies another majorproblem is that the statistical power in individualstudies may be woefully short for the productionresponse we are evaluating. The reviewer mayevaluate the whole class by tallying up the number ofstudies with a significant P value to make arecommendation. This seems like a logical approach,however, with a pool of studies that have lowstatistical power there will be a lot of wrongconclusions.

One solution to this problem is a review called ameta-analysis. A meta-analysis is becoming apopular statistical tool to summarize scientific data.In contrast to a single study, a meta-analysisevaluates the entire data base of studies together inone analysis. In the past researchers used a narrativereview to summarize data. This was a tedious andoften impossible task leading to lots of suggestionsthat more research is needed. More recentlyreviewers have performed a more advancedsystematic review that includes a quantitativeanalysis. A meta-analysis is a specific type ofstatistical systematic review that can make theprocess much simpler, more efficient and ultimatelymore precise.

To conduct a meta-analysis, a researcher gathers allpublished and non-published data that can becollected, follows a clearly defined set of steps forevaluating the quality of the studies, and applies arigorous statistical process to the data. Unlikenarrative reviews where the reviewer applies a levelof importance to each study, the meta-analysisassigns weights based on the size and variance of thestudy. The meta-analysis can thus provide atransparent, unbiased, and repeatable process toevaluate data. Additional benefits of the meta-

85

analysis procedure allows us to determine if the lackof significance due to a treatment is real or due to lowstatistical power, whether or not treatment responsesdiffer due to some type of outside influencing factor(or covariate), and whether or not there is apublication bias (i.e., missing studies that might havereported a lack of effect if they had been published).One related tip when looking at summaries of data –look for the ‘negative’ study in the data. In the realworld of animal research the probability of error istoo great to not see any negative trials. This may be away to detect publication bias.

Other advanced techniques for evaluating productsinclude a portfolio analysis, similar to how a portfolioof stocks is evaluated. In this case similar productscan be evaluated based on their mean response andvariation (referred to as beta in stock marketlanguage). Once again we need sufficient data forthis type of analysis.

Finally, remember that there are two types ofstatistical errors that we can make. The Type I erroris known as a “false positive”. This is the type oferror that is most often avoided. This is the errorassociated with using a product that does not work.The Type II error is a "false negative". This is theerror of failing to use a product that truly doesprovide a profitable response. The problem with thistype of error is that if you make too many of themyou will fall behind your peers and be out ofbusiness due to a lack of competitiveness. Onceagain, sufficient data on a product is needed to beable to perform a Type I/II error analysis.

ConclusionWe have been observing a historic milk price declinein the dairy industry in 2009. During this time,many herds have made adjustments to their rationsleaving us with an opportunity to rebuild the ration,sometimes from scratch. Strategies for addingproducts to the ration to return the most profit toyour farms the quickest should be based on objective,‘peer reviewed’ research and careful planning.Nutritionists should have an understanding of riskrelative to return, as well as an understanding of thevariability of the product and the variation in theanimal’s response to the product. Essential nutrients,energy and better forages should be added first. Wellresearched feed additives often add additional valueto the ration. To evaluate their effectiveness for dairycows make sure there are thorough data on themechanism of action and a biological and economicresponse in lactating cows. Make an attempt to visitthe manufacturer of the product to evaluate theirmanufacturing expertise and quality controlprocedures. Be aware of overlapping technologies,the lag associated with a product response and

reviews that lump together differing products. Lookfor meta-analysis reviews that evaluate product-specific research with weighted means, heterogeneityanalysis (to determine whether or not treatmentresponses differ due to some type of outside factor)and an evaluation of publication bias. Finally, whendeciding on what to put back in the ration first don’tforget about the error that is associated with failing touse a product that is profitable.

Literature CitedDann, H. M., J. K. Drackley, G. C. McCoy, M. F. Hutjens,

and J. E. Garrett. 2000. Effects of Yeast culture(Saccharomyces cerevisiae) on prepartum intake andpostpartum intake and milk production of Jersey cows.J. Dairy Sci. 83:123-127.

DeFrain, J. M., A. R. Hippen, K. F. Kalscheur, and R. S.Patton. 2005. Effects of feeding propionate and calciumsalts of long-chain fatty acids on transition dairy cowperformance. J. Dairy Sci. 88:983-993.

Hutjens, M. F. 2003. Economics of feed additives.Proceedings of the 2003 Pennsylvania Dairy CattleNutrition Workshop. pp. 1-4.

Hutjens, M. F. 2008. Feed Additives: Which, When, andWhy. Illini Dairy Net. :www.livestocktrail.uiuc.edu/dairynet/paperDisplay.cfm?ContentID=9999.

Kung, L. Jr. 2005. Feed additives: modes of action andassessing their efficacy. Proceedings of the 2005Intermountain Nutrition Conference. pp. 125-129.

86

Take Home Messages• Avoid over-conditioning heifers in late gestation

to reduce calving problems and to help avoid lowpostpartum DMI.

• Adapting behavior and social structure prior tocalving should reduce stress at calving.

• Avoid feeding anionic salts to primiparous cows.• It may be advantageous to transport primiparous

cows from the grower to the owner 6-8 weeksprior to expected calving date to adapt behaviorand minimize heifers calving early.

• Providing exercise for pre-fresh primiparouscows might benefit physiological and behavioraladaptations.

IntroductionThe last sixty days prior to calving has the potentialto be a challenging time for first calf heifers. Theprimiparous cow must successfully adapt tonumerous physiological and behavioral changesrelated to calving and the initiation of lactation.Primiparous cows must adapt to changes in housing,feeding, management, and to milking. Strategies thatreduce stress around the time of calving shouldoptimize the chance for success. Additionalchallenges include appropriately co-minglingmultiparous and primiparous cows to minimizenegative social interactions. Our goal for a successfulprimiparous heifer program should be to reduce thenumber of metabolic, psychological, and pathogenichurdles during the transition period. Calving andinitiation of lactation are two unavoidable hurdles.Our focus should be to tear down additional hurdlesthrough adaptive nutrition and behavioralmanagement prior to calving. According to theUSDA national animal health monitoring survey(2007), 36.2 of herds surveyed consist of first-calfheifers. Primiparous heifers represent the presentand the future of dairy farms.

Prepartum feeding strategies that control prepartumenergy intake, moderate body fat mobilization at

calving, and optimize postpartum feed intake arecomponents of a successful primiparous cowtransition program.

A large (n = 1905) heifer field study was conducted inCalifornia to evaluate the effects of age at first calving(AFC) on postpartum performance and health(Ettema and Santos, 2004). Culling after calving andmortality in this study was similar among AFC andaveraged 17.6% and 3.9% respectively. Incidence ofmastitis averaged 19.4% among the groups of heifers,cases of lameness averaged 15%, and left displacedabomasums averaged 2.9%. Even on well managedherds, opportunities for improvement exist.

Deviations from the normal population must beidentified and addressed. Identifying the cause anddetermining the solutions of challenges that causehealth problems, increase premature culling andhinder the ability to perform should be a highpriority in all dairy herds.

Transitioning primiparous cows often receive lessattention than second and greater lactation cows.Perhaps primiparous cows have a lowered status onthe priority list due to seemingly lower healthproblems such as milk fever and displacedabomasum during the transition period whencompared to their older counterparts. Freshprimiparous cows have a greater prevalence ofmastitis than older cows, despite lower reportedincidence of mastitis later in lactation (Miltenburg etal., 1996; Barkema et al., 1998; Nyman et al., 2007).This increased incidence in mastitis in primiparouscows may indicate they are less able to cope withstressors during the periparturient period.

Improvements in transition programs forprimiparous cows will have an effect on herdperformance today and in subsequent lactations.Investigating strategies to promote favorable energybalance, reduce negative social and behavioral

Freshening the First Calf Heifer:Opportunities and Challenges

Noah B. LitherlandUniversity of Minnesota

Department of Animal ScienceCollege of Food, Agricultural, and Natural Resource Sciences

155B Haecker Hall1364 Eckles Avenue

St Paul, MN [email protected]

3400 1st Street North, St. Cloud MN [email protected]

87

interactions, reduce health challenges, and increasemilk production may improve the profitability ofprimiparous cows.

Energy IntakeOften primiparous cows are moved from the heifergrower and placed into the far-off dry cow group 60days prior to calving. The objective of this penmovement is to adapt first calf heifers to newsurroundings, expose them to social interactions witholder cows, and to feed an appropriate diet.Moderate energy or controlled energy diets are dietsbased on wheat straw and silage that are offered atan ad libitum rate, but do not allow cows to greatlyover-consume energy (Beever et al., 2006; Dann et al.,2006; Janovick-Guretzky et al., 2006). Moderateenergy diets for mature dry cows have beenimplemented with success and are being adopted byproducers. Are moderate energy diets appropriatefor primiparous cows? Primiparous cows exhibitnegative energy balance in early lactation similar tothat of multiparous cows (Lin et al., 1984).Primiparous cow’s produces less milk compared tomultiparous cows, but may be at risk for similarseverity of negative energy balance after calving(Cavestany et al., 2005; Wathes et al., 2007).

Work examining the role of energy intake during thedry period has been limited in primiparous cows.Researchers in Wisconsin found that primiparouscows fed a more moderate energy diet (59.7% TDN)prepartum had higher DMI postpartum than heifersfed higher energy (69.3% TDN) (Grummer et al.,1995). Higher energy feeding prepartum did notimprove milk yield or milk composition (Grummer etal., 1995). Heifers fed the higher energy dietprepartum had higher concentrations of blood non-esterified fatty acids (NEFA), _-hydroxy butyrate(BHBA), and tended to have higher concentrations ofliver triglycerides (Grummer et al., 1995). Heifers fedexcessive energy prior to calving tended to havelower DMI postpartum than those fed a high foragemoderate energy diet or those fed in restrictedamounts (Janovick Guretzky, 2006).

Field experience suggests that heifers consumingexcessive energy prior to calving have more difficultycalving. Additionally, excessive body fatmobilization may reduce energy intake postpartumand predispose heifers to metabolic disorders.Hoffman et al. (1996) determined that feedingpregnant first calf heifers low-energy, high fiberforages may help in controlling energy intake andassist in minimizing overconditioning at calving.

Metabolic differences between primiparous andmultiparous cows prepartum are numerous.Primiparous cows have not yet reached mature body

weight so changes associated with growth placedemands on anabolic pathways. Primiparous cow’sinvestment in colostrum quantity and quality andmilk yield is typically less than that of a mature cow.

Differences in metabolic traits, milk yield, and bodycondition score between primiparous andmultiparous cows during the periparturient periodwere investigated by workers in Great Britton(Wathes et al., 2007). Data from Wathes et al. (2007)indicated that primiparous cows had higherconcentrations of insulin-like growth factor-1 andlower _-hydroxybutyrate (BHBA) concentrations oneweek before and seven weeks after calving, higherleptin pre-partum, and both the peak in non-esterified fatty acids (NEFA) and the nadir in ureaconcentration occurred earlier after calving comparedwith multiparous cows. These authors concludedthat there are differences in the regulation ofmetabolism between primiparous and multiparouscows that promote nutrient partitioning into growthas well as milk during the first lactation (Wathes etal., 2007). Twenty Swedish herds participated in astudy to evaluate changes in metabolites andimmune variables associated with somatic cell countsin primiparous cows (Nyman et al., 2008). Resultsfrom this large study showed that greaterconcentration of BHBA and glucose before calvingwere associated with reduced Box-Cox transformedsomatic cell counts (bcSCC) at first test milksampling. However, greater concentrations of NEFAbefore calving and greater change in NEFA at calvingwere associated with greater bcSCC at first test-milking.

Protein intakeAdjustments are often made to increase the dietaryprotein content of the diet when multiparous cowsand first calf heifers are mixed to meet the higherprotein requirement for growth. Santos et al., 2001suggested that primigravid cows might benefit fromdiets with crude protein content greater than 12.7%.Primiparous cows were fed one of two dietscontaining a moderate amount of crude protein(12.7%) or a higher crude protein amount (14.7%).Primiparous cows fed the prepartum diet higher inprotein produced 2.0 kg/d more milk and 3.1 kg/dmore 3.5% fat corrected milk during the first 120 daysin milk. In the first month after calving, Nyman, etal. (2009) associated low milk urea nitrogen (MUN) (<4 mmol/L) with elevated somatic cell count at firsttest day in primiparous cows. Low MUN indicatesinsufficient dietary protein (Ipharraguerre et al., 2005;Schei et al., 2005).

Age at first calvingAge at first calving (AFC) is an important factor inthe cost or rearing replacements in dairy herds.

88

Optimal AFC in Holsteins was recommended to be ≤24 mo with body weight > 560 kg after calving at 24mo (Heinrichs, 1993; Tozer and Heinrichs, 2001). Ageat first calving has been shown to have an impact onhealth in primiparous cows. Waage et al. 1998)observed an increased risk of mastitis with increasedage at first calving. A similar association between ageat calving and ketosis was also observed (van Dam etal, 1988). Swedish and Danish researchershypothesized that metabolites and immune variablesmay differ between primiparous cows at different ageat first calving (Nyman et al., 2008). Primiparouscows calving at >27 mo of age had greater BHBA andNEFA values, and lower glucose, insulin, and ureanitrogen compared with heifers calving at <27months (Nyman et al., 2008). Heifers calving at anage <25 mo had greater conglutinin (collectininvolved in pathogen recognition) and urea nitrogenvalues and lower NEFA compared with heiferscalving at >25 mo (Nyman et al., 2008).

Should primiparous cows be fed anionicsalts?The default diet for primiparous cows during the dryperiod is often whatever the multiparous dry cowsare being fed. Housing primiparous and multiparousdry cows together creates a challenge for nutritionistsattempting to modify the dietary cation-aniondifference (DCAD) with anionic salts. Anionic saltsare used to decrease the DCAD to aid in preventionof hypocalcemia through increasing calciummobilization from bone and increased uptake ofcalcium from the intestine. Primiparous cows atcalving are typically 80-85 percent of their maturebody weight, so bone growth and remodeling is stilloccurring, making bone calcium more readilyavailable.

Researchers at Michigan State fed multiparous andprimiparous cow’s prepartum diets containinganionic salts to achieve DCAD’s of +15, 0, and -15meq/100g. Prepartum dry matter intake, energybalance, and body weight gains were lower andconcentration of liver triglyceride was higher forheifers but not cows fed the -15 DCAD diet (Moore etal., 2000). The authors concluded that heifers shouldnot be fed anionic salts before calving as theymaintained calcium homeostasis through thetransition period regardless of dietary DCAD, butconsumed less DM when fed the -15 DCAD (Mooreet al., 2000).

Behavioral adaptations In addition to the physiological and metabolicchanges associated with calving and the initiation oflactation, primiparous cows must adapt tocomingling with older, socially dominant cows, learnto use head locks and freestalls, become accustomed

to increased handling by humans, and adjust to themilking routine. Separate feeding and managementof primiparous and multiparous cows is warranted(Grant 2007; Daniels et al., 2008), but is often notpractical especially on smaller herds.

According to the NRC (2001) primiparous cowsconsume less feed and in a different pattern (peakinglater) than multiparous cows. Additionally, it isbelieved that primiparous cows are usually moretimid and occupy a lower rank in the social herdhierarchy (Wierenga, 1990). Interesting data fromSpain showed total eating time was longer whenprimiparous cows were housed with multiparouscows, however, primiparous cows housed alone hadalmost 1 more meal per day than did those housedwith multiparous cows (Bach et al., 2006). Feedingarea was limited to one feeder per 1.8 cows; however,more than 50% of the feeders at time of feeding wereoccupied by primiparous cows, suggesting they werenot intimidated by multiparous cows (Bach et al.,2006). Housing multiparous and primiparous cowstogether may offer advantages in exposing heifers tointensified competition prior to calving as well asassumption of learned behaviors.

Two producers in the upper Midwest that I recentlyvisited with have interesting philosophies on feedingbehavior in close-up heifers and cows. One producerpurposefully overcrowds the close-up group toincrease competition at the feed bunk andpurportedly stimulate intake. Data published byBach et al. (2006) seems to support this theoryalthough results might not be the same in allsituations. Another producer on a large dairyconstantly mixes pens of dry cows to reduce thegroups desire to develop a strong social hierarchy.The mixing of dry cow pens seemingly reducesnegative social interactions and helps fresh cowsacclimate to herd mates in the fresh cow pen.

Smooth return from the heifer growerWhether heifers are raised by the owner or farmedout to a heifer grower, a smooth transition into themilking herd is important to the success of the firstand subsequent lactations. Variables such asavoiding overconditioning, calving at an appropriateage, and experience with the use of headlocks andfree-stalls are important factors that will impacttransition success of first calf heifers. The amount oftime before expected calving date heifers are returnedto the milking herd may have some impact ontransition success.

In the most recent NAHMS survey (2007), heifersreturning to the milking herd from the heifer groweraveraged 21.6 months of age (USDA, 2007). Heifersare typically brought back to the milking herd as

89

early as 3 months or as late as 3 weeks prior toexpected calving date. Synchronization programshave helped in producing larger groups of heiferswith similar calving dates. Increased variability incalving date occurs when a heifer with a calving datethat is an extreme outlier from the mean or one withan incorrect expected calving date steps foot on thetruck.

Scheduling of heifer pickup may be a factorcontributing variability in the primiparous heifertransition program. Clearly the truck does not runevery day to pick up heifers so the arrival time at themilking herd is a potential source of variability.Adequate time should be allowed for all heifers toadapt to their new surroundings and to minimize thenegative impact on those that calve earlier thanexpected. Clearly constraints in facility capacity,labor, and management must be evaluated todetermine the optimal timing of heifer return fromthe grower.

USDA survey data indicates that small and mediumoperations sending heifers off-site to be raised weretransported fewer than 20 miles while large herdstransported heifers between 5 and 50 miles (USDA,NAHMS 2007). Of all operations surveyed, 10.6%transported heifers 50 miles or more. Stress fromtransportation and handling in transit from the heifergrower to the milking herd should be minimized.The impact of shipping stress on late termprimiparous cows has not been investigated.

Italian workers investigated the effects of shippingstress on immune function in 6 month old calves(Riondato et al., 2008). Highest leucocyte andneutrophil counts associated with increasedconcentrations of cortisol and catecholamines,indicated that stress was maximal upon arrival whencompared with that before departure, 24 h, and 1week after arrival (Riondato et al., 2007).Transportation is considered to be one of the maincauses of stress in calves (Momede et al., 1982).Biological systems activated in response to stress arethe hypothalamo-pituitary-adrenal (HPA) axis,responsible for active responses associated withneurogenic stress, such as transportation, and thesympatho-adrenal medullary system, responsible forpassive responses associated with perceivedenvironmental stress such as noise (Mitchell et al.,1988; Griffin, 1989; Minton, 1994).

Hartman et al. (1976) showed that shipping alteredthe capability of calves to synthesize antibodies.Effects of prepartum shipping on antibodyproduction in colostrum have not yet beeninvestigated. Some heifers are calved and thenshipped to the milking facility. Research has not been

conducted to evaluate the effect of transportationstress on cattle after calving compared with thosewith a dry period and that calve on-site.

Norwegian researchers evaluated feeding and properfree stall use by heifers transferred from heifergrowers back to the milking herds. On day 2 aftertransfer, 34% of the heifers refused to use free stallsand by day 15 there were still 23% of the heifersexhibiting this behavior (Kjoestad and Myren, 2001).Refusal to use stalls was significantly associated withrearing accommodations and free-stall layoutindicating the importance of previous experience andproper facility design.

Reducing udder edema in primiparouscows Udder edema is a periparturient disordercharacterized by excessive accumulation of fluids inthe intercellular tissue spaces of the mammary gland.Udder edema in primiparous cows is a chronicproblem in some herds. Udder edema begins severalweeks prior to calving and seems to be moreprevalent in primiparous than multiparous cows.Challenges with udder edema have been attributed togenetic predisposition, feeding excessive grain, andmineral imbalances (excessive sodium and potassiumintake) (NRC, 2001). Prepartum nutrition strategiessuch as selecting forages low in potassium andreducing sodium intake may help reduce theincidence of udder edema.

Research at the University of Florida conducted aretrospective observational study to evaluate riskfactors for udder edema in primaprous cows calvingwith and without udder edema (Melendez et al.,2006). First test day DHIA milk yield was lower incows that developed udder edema (3.6 kg/day) thanin normal cows and cows with udder edema were1.62 times more likely to develop udder edema in thesecond lactation (Melendez et al., 2006). These datasuggest that dry period feeding errors that put cowsat risk for udder edema must be taken into account.Additionally, first calf heifers with udder edemashould receive added attention as they may be atincreased risk for other disorders.

Prepartum exercise for primiparous cows?Exercising close-up heifers has become a standardprocedure on some dairy farms in the Midwest.Exercise lots for dry cows have been incorporatedinto dairy facilities. Additionally, versatile droverslanes in some facilities allows for ease of animalmovement for exercise around the barn. Manyproducers move first calf heifers through the parlor atmilking or when the parlor is not in use to familiarizeheifers with the sights and sounds of the parlor andto give them experience with being handled by

90

91

workers. This experience will become valuablepostpartum as these trained heifers will have one lesshurdle to contend with.

Researchers at Michigan State exercise trained pairsof late-pregnant and nonpregnant multiparous cowson a treadmill to determine if exercise improvedphysical fitness (Davidson and Beede, 2009). Theauthors hypothesized that exercise would improvecirculation or alter blood volume potentiallyincreasing blood flow to key organs during theperiparturient period, improve responses to stress,and improve muscle fitness to reduce response tofatigue. Cow’s exercised prepartum had lower heartrates and plasma lactate concentrations and moreeffectively maintained acid-base homeostasis duringtread mill tests when compared with non-exercisedcows (Davidson and Beede, 2009). Responses inexercised primiparous cows might be similar.

SummaryWe must continue to be vigilant in the quest forimproved performance, health and wellbeing of ourdairy cattle. Areas where opportunities forimprovement exist, such as optimal management ofprimiparous cows, should not be overlooked. A highdegree of variability is inherent in transition cowsoften shrouding solutions to transition cow failures.Feeding strategies that work for both primiparousand multiparous cows should be considered.

ReferencesBach, A. C. Iglesias, M. Devant, and N. Rafols. 2006.

Performance and feeding behavior of primiparous cowsloose housed alone or together with multiparous cows.J. Dairy Sci. 89:337-342.

Barkema, H. W., Y. H. Schukken, T. J. Lam, M. L. Beiboer,H. Wilmink, G. Benedcitus, and A. Brand. 1998.Incidence of clinical mastitis in dairy herds grouped inthree categories by bulk milk somatic cell counts. J.Dairy Sci. 81:411-419.

Beever, D. E. 2006. The impact of controlled nutritionduring the dry period on dairy cow health, fertility, andperformance. Anim. Repro. Sci. 96:212-226.

Cavestany, D., J. E. Blanc, M. Kulcsar, G. Uriarte, P.Chilibroste, A. Meikle, H. Febel, A. Ferraris, and E.Krall. 2005. Studies of the transition cow under apasture-based milk production system: Metabolicprofiles. J. Vet. Med. A Physiol. Pahtol. Clin. Med. 52:1-7.

Dann, H. M., N. B. Litherland, J. P. Underwood, M. Bionaz,A. D’Angelo, J. W. McFadden, and J. K. Drackley. 2006.Diets during the far-off and close-up periods affectperiparturient metabolism and lactation in multiparouscows. J. Dairy Sci. 89:3563-3577.

Daniels, K. J., J. R. Townsen, S. S. Donkin, E. A. Pajor, A. g.Fahey, and M. M. Schutz. 2008. Behaviors of transitiondairy cows and heifers. J. Dairy Sci. Vol. 86, Suppl. 1pp 352.

Davidson, J. A. and D. K. Beede. 2009. Exercise training oflate-pregnant and nonpregnant dairy cows affectsphysical fitness and acid-base homeostasis. J. Dairy Sci.92:548-562.

Etrema, J. F. and J. E. P. Santos. 2004. Impact of age atcalving on lactation, reproduction, health and incomein first-parity Holsteins on commercial farms. J. DairySci. 87:2730-2742.

Grant, R. J. 2007. Taking advantage of natural behaviorimproves dairy cow performance. Western DairyManagement Conference Proceedings pp 225-236.

Grant, R. J., and J. L. Albright. 1995. Feeding behavior andmanagement factors during the transition period indairy cattle. J. Anim. Sci. 69:2791-2803.

Griffin, J. F.T. 1989. Stress and immunity a unifyingconcept. Veterinary immunology andimmunopathology. 20:263-312.

Grummer, R. R., P. C. Hoffman, M. L. Luck, and S. J.Bertics. 1995. Effect of prepartum and postpartumdietary energy on growth and lactation of primiparouscows. J. Dairy Sci. 78:172-180.

Hartman, H., A. Bruer, H. Herzog, H. Meyer, H. Rhode, F.Schulze, and G. Steinbach. 1979. General adaptationsyndrome (Selye) in calf. 6. Stress conditions-theirimpact upon antibody levels, following active andpassive immunization, and upon the topographicdistribution of certain groups of germ in thegastrointestinal tract. Arch. Exp. Vet Med. 30:553-566.

Heinrichs, A. J. 1993. Raising dairy replacements to meetthe needs of the 21st century. J. Dairy Sci. 76:3179-3187.

Hoffman, P. C., N. M. Brehm, S. G. Price, and A. Prill-Adams. 1996. Effect of accelerated postpubertalgrowth and early calving on lactation performance ofprimiparous Holstein heifers. J Dairy Sci. 79:2024-2031.

Ipharraguerre, I. R., J. H. Clark, and D. E. Freeman. 2005.Varying protein and starch in the diet of dairy cows. 1.Effects on ruminal fermentation an intestinal supply ofnutrients. J. Dairy Sci. 88:2537-2555.

Janockick Guretzky, N. A., N. B. Litherland, K. M. Moyes,and J. K. Drackley. 2006. Prepartum energy intakeaffects health and lactational performance inprimiparous and multiparous Holstein cows. J. DairySci 89:267 (Abstract).

Kjoestad, H. P. and H. J. Myren. 2001. Failure to usecubicles and concentrate dispenser by heifers aftertransfer from rearing accommodation to milking herd.Acta. Vet. Scand. 42:171-180.

Lin, C. Y. A. J. McAllister, T. R. Batra, and A. J. Lee. 1984.Multitrait estimation of relationships of first lactationyields to body weight changes in Holstein heifers. JDairy Sci 68:2954-2963.

Melendez, P., C. C. Hofer, and G. A. Donovan. 2006. Riskfactors for udder edema and its association withlactation performance on primiparous Holstein cows ina large Florida herd, U.S.A. 2006. PreventiveVeterinary Medicine 76:211-221.

Miltenburg, J. D., D, deLange, A.P. P. Crauwels, J. H.Bongers, M. J. M. Tielen, Y. H. Schukken, and A. R. W.Elbers. 1996. Incidence of clinical mastitis in a randomsample of dairy herds in the southern Netherlands. Vet.Rec. 139:204-207.

Minton, J. E., 1994. Function of the hypothalamic-pituitary–adrenal axis and the sympathetic nervous sysemt inmodels of acute stress in domestic farm animals. J.Anim. Sci. 72:1891-1898.

Mitchell, G., J. Hatting, M. Ganhao. 1988. Stress in cattleassessed after handling, after transport and afterslaughter. The Veterinary Record. 123:201-2050.

Moore, S. J., M. J. VandeHaar, B. K. Sharma, T. E Pilbeam,D. K. Beede, H. F. Bucholtz, J. S. Liesman, R. L. Horst,and J. P. Goff. 2000. Effects of altering dietary cation-anion difference on calcium and energy metabolism inperipartum cows. J. Dairy Sci. 83:2095-2104.

92

NRC. 2001. Nutrient Requirements for Dairy Cattle. 7thRev ed. Natl. Acad. Sci., Washington DC.

Nyman, A. K., T. Ekman, U. Emanuelson, A. H. Gustafsson,K. Holtenius, K. P. Waller, and C. H. Sandgren. 2007.Risk factors associated with the incidence of veterinary-treated clinical mastitis in Swedish dairy herds with ahigh milk yield and a low prevalence of subclinicalmastitis. Prev. Vet. Med. 78:142-160.

Nyman, A. K., U. Emanuelson, K. Holtenius, K. L.Ingvartsen, T. Larsen, and K. Persson Waller. 2008.Metabolites and immune variables associated withsomatic cell counts of primiparous dairy cows. J. DairySci. 91:2996-3009.

Nyman, A. K., U. Emanuelson, A. H. Gustafsson, K.Persson Waller. 2009. Management practicesassociated with udder health of first-parity dairy cowsin early lactation. Preventive Veterinary Medicine88:138-149.

Riondato, R. A. D’Angelo, B. Miniscalco, C. bellino, R.Guglielmino. 2008. Effects of road transportation onlymphocyte subsets in calves. The Veterinary Journal.175:364-368

Santos, J. E. P., E. J. DePeters, P. W. Jardon, and J. T. Huber.2001. Effect of prepartum dietary protein level onperformance of primigravid and multiparous Holsteindairy cows. J. Dairy Sci. 84:213-224.

Schei, I., J. Volden, L. Baevre. 2005. Effects of energybalance and metabolizable protein level on tissuemobilization and milk performance of dairy cows inearly lactation. Livest. Prod. Sci. 95:35-47.

Tozer, P. R., and A. J. Heinrichs. 2001. What affects thecosts of raising replacement dairy heifers: A multiple-component analysis? J. Dairy Sci. 84:1836-1844.

USDA. 2007. Dairy 2007, Part II: Changes in the U.S.dairy cattle Industry, 1991-2007. USDA-APHIS-VS,CEAH. Fort Collins, CO. #N481.0308.

Van Knegsel, A. T. M., H. van den Brand, J. Dijkstra, W. M.van Straalen, T. Jorritsma, S. Tamminga, and B. Kemp.2007. Effect of glucogenic vs. lipogenic diets on energybalance, blood metabolites, and reproduction inprimiparous and multiparous dairy cows in earlylactation. J. Dairy Sci. 90:3397-3409.

Wathes, D. C., Z. Cheng, N. Bourne, V. J. Taylor, M. P.Coffey, and S. Brotherstone. 2007. Differences betweenprimiparous and multiparous dairy cows in the inter-relationships between metabolic traits, milk yield, andbody condition score in the periparturient period.Domest. Anim. Endocronol. 33:203-225.

Wierenga, J. K. 1990. Social dominance in dairy cattle andthe influences of housing and management. Appl.Anim. Behav. Sci. 27:201-229.

93

DCAD Balancing for Lactating CowsJoe Harrison

Washington State University

94

95

96

97

98

99

Cow Comfort and HealthNigel B. Cook

University of Wisconsin-Madison, School of Veterinary Medicine,Madison, Wisconsin, USA.

[email protected]

IntroductionThe dairy expansion era in North America, whichgathered pace throughout the nineties and continuesto this day, has resulted in the migration of dairycattle from traditional tiestall and stanchion barns tothe freestall facility, which has emerged as thedominant form of dairy cattle housing worldwide.

The basic premise of milking more efficiently througha parlor, while being able to keep larger groups ofcows together in management groups in larger herdsis sound from a management and economicperspective. Cows are ‘free’ to move between afeedbunk where a Total Mixed Ration is availableevery hour of every day, and a stall, designed toprovide her ample rest on a clean dry comfortablesurface. This is the ideal – the question is whether itis the reality.

In this article, I will attempt to summarize the aspectsof freestall design that are failing the cow, and justifythe solutions we have found and the current trendsemerging in the dairy industry of the Upper Mid-West. The challenge presented is to find the balancebetween excellent cow comfort and manageable cowcleanliness.

Freestalls and Time BudgetsSo, what do cows do in the freestalls that we havebuilt over the last 10 years? From an analysis of 250total 24-hour time budgets, we have collected from208 cows housed in 17 freestall barns in Wisconsin,the average time spent performing each of five keybehaviors is shown in Table 1. On average, cowsspend 2.6 h/d milking – reflecting the three times aday milking schedule most large freestall dairiesoperate at. Other components of the cow’s day arealso fixed and non-negotiable. The cow has to spenda large proportion of the day eating. The TMR fed,free stall housed dairy cow eats for an average of 4.4h/d (range 1.4-8.1). Note that this is about half thetime that a grazing cow spends eating per day –pasture cows average around 8-9 h/d eating. She alsoneeds to drink around 20-25 gallons of water per day(more in hot climates) and she will spend an averageof 0.4 h/d at or around a waterer. With these fixednon-negotiable time slots, we have already taken 4.4+ 0.4 + 2.6 = 7.4 hours out of the time budget, leavingunder 17 hours remaining in the pen.

Time left in the pen will be spent performing threeactivities – lying down, standing in an alley andstanding in a stall. The average freestall cow spends2.4 h/d standing in an alley socializing, movingbetween the feed bunk and stalls and returning fromthe parlor. Once in the stall, the average cow spends2.9 h/d standing in the stall (range 0.3-13.0) and 11.3h/d lying in the stall (range 2.8-17.6) on average –but note the wide ranges in these behaviors.

Table 1. The mean (range) 24-h time budgets for 208cows filmed over 250 filming periods on 17 freestallbarns in Wisconsin

Activity Mean RangeN=250 (h/d) (h/d)Time lying down in the stall 11.3 2.8-17.6Time standing in the stall 2.9 0.3-13.0Time standing in the alley 2.4 0.2-9.4Time drinking 0.4 0-2.0Time feeding 4.4 1.4-8.1Time milking 2.6 0.9-5.7

Lying behavior is typically divided into an average of7.2 visits to a stall each day (called a lying session),and each session is categorized by periods standingand lying – called bouts. The average cow has 13.6lying bouts per day and the average duration of eachbout is 1.2 h (range 0.3-2.9). Most cows will standafter a lying bout, defecate or urinate, and lie backdown again on the contra-lateral side.

From studies designed to make cattle work for accessto a place to rest, it would appear that cows targetaround 12 h/d target lying time (Jensen et al., 2005;Munksgaard et al., 2005), and this is in agreementwith the lying times found in well designed freestallfacilities (Cook et al., 2004). If this is the case, thenour industry is failing to provide the average cowsufficient rest, and our freestall ideal is not beingrealized.

What is the Cost of Inadequate Rest?It is commonly suggested that cow’s make more milkwhen they are lying down as blood flow through theexternal pudic artery increases by around 24-28%when lying compared to standing up (Metcalfe et al.,1992; Rulquin and Caudal, 1992), and failure toachieve adequate rest has negative impacts onlameness (Cook and Nordlund, In press), ACTH

100

concentrations (Munksgaard and Simonsen, 1996),cortisol response to ACTH challenge (Munksgaard etal., 1999) and growth hormone concentrations(Munksgaard and Løvendahl, 1993; Ingvartsen et al.,1999) – suggesting that there is a significant stressresponse.

Some workers have suggested that there is a linearrelationship between time lying and milk productionof the order of 2-3.5 lbs of milk increase for eachadditional hour of rest (Grant, 2004). While thismay be true, we have not seen such a relationshipand milk yield has not been significant in anyof the lying time models in our time budgetstudies.

It seems more likely that the requirement for rest is athreshold event and that all cows, regardless of yield,require a minimum period. A strong case can bemade that the true cost of failing to achieve this restis an increase in lameness, and lameness hassignificant impacts on production.

Let us take a look at mature equivalent (ME305) milkproduction (which standardizes milk output to 3rd

parity) and see how well older cows performcompared to younger cows – as a proxy measure ofhow successful our lameness management is.

ME305 averages for two groups of herds usingWisconsin DHIA testing – less than 100 cows (whichwe will assume are predominantly tiestall housed)and greater than 500 cows (which we will assume arefreestall housed) are shown in Table 2. There aresome interesting trends in the difference betweensecond and later lactation groups and first lactationgroups. While the freestall housed larger herds makemore milk, and the difference in ME between firstand second parity is similar (~500-600lbs) betweenboth herd size groups, the difference between 1stlactation and 3rd and greater lactation cows is muchwider in the large herd category – nearly 1,300lbsgreater. If this were due to a failure of the MEcalculation to properly account for parity effects, wewould expect the differences to be consistent acrossherd size. They are not and I believe that the greaterdifferences we are seeing in larger herds are due tothe environment in which we are keeping theanimals. Significant differences in lamenessprevalence have been recorded between tiestall andfreestall barns (Wells et al., 1993; Bergsten and Herlin,1996; Cook, 2003; Sogstad et al., 2005), and there isevidence to suggest that the freestall environment isfailing the larger older cows in our herds – lamenessbeing the primary reason for the disparity in ME milkproduction. In barn remodels, where we providemore comfortable stalls for older mature cows, we seethe ME gap close and sometimes invert. This occurs

coincident with a decrease in lameness prevalence,particularly in older cows.

Table 2. ME305 averages by parity group for DHIAherds by herd size (<100 cows or >500 cows) inWisconsin. Benchmarks April 1, 2008 (AgSourceCooperative Services, Verona, WI).

Parity Group Mature Equivalent 305 Milk Production (lbs)Herds<100 cows (n=3218) Herds>500 cows (n-155)

Average Difference Average Differencefrom 1st Lact from 1st Lact

1st Lactation 22,903 29,084 -2nd Lactation 22,374 529 28,462 6223rd+ Lactation 21,859 1,044 26,783 2,301

So, what can we do to improve the situation? We cancertainly make sure that there is adequate time forrest by limiting time out of the pen for milking,providing enough stalls for cows to achieve theirtarget rest by limiting overstocking and finally, bymaking sure that the stall is comfortable and easy touse. Indeed, we have used the ME gap theorydescribed above to help justify many stall renovationprojects.

The Importance of Stall SurfaceAnalysis of our time budget database highlights theimportance of stall surface. Cows bedded on sandexceeded our target of 12 h/d of rest, while cows onrubber crumb filled mattresses averaged only 10.7h/d (Figure 1).

Figure 1. Time budgets for cows bedded on sand(n=89) compared with cows on a rubber crumb filledmattress (n=119).

101

The reason for this is two-fold. Firstly, there are onaverage 42% fewer lame cows in sand beddedfreestall herds (Cook, 2003, Cook et al., 2004, Espejoet al., 2006), and secondly, lame cows stand longer inmattress stalls compared to sand stalls. We believethat the main reason for this is due to the difficultieslame cows have rising and lying down on a firmsurface (Cook and Nordlund, In Press). While sandprovides cushion, traction and support – whichfacilitates rising and lying movements for lame cows,enabling them to maintain normal patterns of rest,firm mattress surfaces make it difficult for cows torise and lie down because of the pain associated withthe contact point between a painful foot and a firmunforgiving surface. As a result, we see an extensionin standing time per day, a reduction in the numberof visits to a stall per day and as a consequence ofthree times a day milking and other stresses to thecow’s time budget, a reduction in lying time. Failureto provide adequate rest and recuperation for lamecows, results in chronic disease and an increase in theprevalence of lameness.

The difference in lameness prevalence is the mainreason for the large difference in milk productionobserved between sand and mattress freestalls (Table3), but there are also benefits in terms of milk quality.The numbers presented in Table 3 are for herdsvisited because of an udder health problem.However, the differences observed are very typical ofthe mattress to sand conversions we have beeninvolved in over the last 5 years and we use thesefigures for the construction of partial budgets tofinance the barn changes.

Table 3. Sand bedding benefits compared withmattress herds for 62 freestall herds investigatedby our Food Animal Production Medicine groupsince 2001.

Mattress Sand Sand

Factor Herds Herds BenefitRHA milkproduction percow (lbs) 24,260 25,926 +1,666Somatic CellCount (‘000/ml) 373 298 -75Cow Case MastitisRate (%) 62 45 -17

Sand must be managed to prevent a build up oforganic material over time (Cook and Reinemann,2007). Provided fresh sand is added once or twice aweek, gross contamination is removed each milking,the bed is leveled daily and sand is removed from therear of the beds every ~6 months or so, it remains thegold standard for the cow, not only in terms ofcomfort, but also in terms of milk quality. Whileorganic bedding materials may be ‘managed’, I find

them in every way inferior to sand, particularly whenmanaged in a deep loose bed.

Because sand is so forgiving it has often been saidthat the cow may compensate for other failures install design – such as inadequate space. In fact I usedto think the same – but I do not anymore. We haveseen too many improvements in production andhealth in sand bedded facilities when other stalldesign improvements have been made.

Providing Adequate WidthThe freestall barns built throughout the expansion erahave typically had resting spaces defined laterally bydivider loops located at 43 to 46 inches (109-117 cm)on center, and by a brisket board typically 66 inches(168 cm) from the rear curb. While we believe thatthese dimensions are appropriate for a 1200 lb (545kg) first lactation heifer, we believe that they areinadequate for larger mature cows. The evidence forsuch an opinion comes from three sources:

Firstly; direct observation. Anderson (2003) examinedthe ergonomics of stall design and showed howlimited resting space increased the disturbancesbetween neighbors and led to more restless lyingbehavior. Secondly; behavioral studies. The stallbehavior studies described by Tucker et al. (2004)used 15 Holstein cows averaging ~ 1,600 lbs (727 kg)body weight and showed a significant increase inresting time between 44 inch (112 cm) and 48 inch(122 cm) wide stalls, but no difference between 48(122 cm) and 52 inches (132 cm), proving that widthdoes have a significant effect on lying behavior, andsupporting wider dimensions than the industrystandard of ~45 inches (114 cm). Thirdly; herdperformance changes after stall remodeling. We haveremodeled a large number of freestall facilities in theUpper Mid-West over the last 5 years and myexperience has been that after stall surface changes,increased stall width for large mature Holstein cowshas been the second most important change made inboth sand and mattress facilities.

There is a commonly held belief that if we make thestall wider and longer, it will lead to increasedmanure contamination of the stall, inappropriate stalluse behavior (eg. backwards lying) and worseningudder health. If the stall is not sized appropriate forthe size of the animals using them and if the restingarea is poorly defined, these concerns may well berealized. It is therefore important to determine thesize of the animals using the pen, and to design theresting space correctly. Problems do occur whenmixed age groups are penned together. While smallheifers in larger stalls may well soil the platformmore, it makes no sense to punish two thirds of a penof mature cows to make sure the stalls are kept clean

102

for the heifers. A compromise must be reached –either with stall dimensions or cow groupingstrategy.

Diagonal lying is a complex behavioral issueresulting from a variety of stall design faults, but stallwidth is often blamed. I believe that the mostsignificant issues leading to diagonal lying areunrelated to stall width – they include adjacent cowsin head to head stalls (Anderson, 2003), too short aresting space length, brisket locators that are toohigh, inadequate lunge space, head bob restrictionsand neck rails that are too close to the rear curb(Cook and Nordlund, 2005). Failure to understandthese causative factors results in poorly designedstalls where cows become contaminated withmanure. Therefore, we need to make sure that theresting area is correctly defined, so that manurecontamination of the bed can be minimized.

Defining the Resting AreaThe stall resting space is defined laterally by thedivider loop and in front by the brisket locator. Weneed to provide just enough direction to align thecow correctly in the stall, but not inhibit theboundaries of the area so much that lying and risingbehavior has to be modified to the point that itreduces the ability of the cow to use the stall. Forthese reasons, we prefer a divider loop that has thefollowing characteristics (Figure 2):

1. Locates the upper edge of the lower divider railat 12 inches (30 cm) above the stall surface. Thisserves to align the cow, enables the cow to lungeeasily to the side over the top of the rail if shechooses to do so, and is high enough to reducetrauma to the front leg below the rail and limitcows putting their legs through the loop.

2. Has an angle in the lower rail that is located 20inches (51 cm) behind a correctly located brisketlocator. This location serves to align the cow, butlimits trauma to the hip area.

3. Has an interior loop diameter of 35 inches (89cm) (or exterior diameter of 39 inches (99 cm)).This will make sure that with the lower railcorrectly located, the upper rail will place theneckrail at around 50 inches (127 cm) above thestall surface.

The resting space is defined in front by the brisketlocator, which serves to position the cow when she isresting, so that her rear end is adjacent to thealleyway to reduce soiling of the bedding. Behavioralstudies have shown that cows prefer to lie in stallswithout a brisket locator, compared to stalls with an 8inch (20 cm) high piece of wood (Tucker et al., 2006).Many consultants have taken this to mean that weshould build stalls without brisket locators. This is a

mistake. While I will concede that in a short stall (lessthan 8 feet (2.44 m)), a poorly designed brisket locatorcan be removed resulting in an observableimprovement in stall usage, larger stalls require alocator to help position the cow.

Figure 2. An ideal divider loop positioned relative tothe rear curb and brisket locator.

The problem with brisket locator design is themovement of the cow’s front leg when she is rising inthe stall. To facilitate rising, she needs to take a fullforward stride and to do this, it is difficult for mostcows to lift their leg much higher than about 4 inches(10 cm) above the stall surface. Generally, we havemoved away from the traditional brisket board (awooden form used to pour the concrete curb), tomore rounded plastic, fiber glass or pvc pipes ormouldings. These have performed reasonably well,but because they are lower, smoother and lessrestrictive, some cows choose to ignore them. Becauseof these issues, we have returned to concrete for theanswer and I have developed a design that we callthe ‘brisket slope’. This serves to locate the cow,while being low enough to allow the cow to lie downwith her front legs outstretched, and sloped enoughto allow the front leg to land on the slope when rising(Figure 3).

Figure 3. The concrete brisket slope – designed tolocate the cow relative to the rear curb, allow her tostretch her legs forward and plant her front legforward when she rises.

103

Providing Adequate Lunge and Bob SpaceWhen the cow rises, she moves her head forward ina lunging motion to take the weight off her rear legs,to facilitate rising. The head is lowered, almost to thesurface she is resting on and then ‘bob’s upward. Werefer to the horizontal area in front of the restingspace as lunge space and the vertical area at the endof the lunge as the bob zone.

Failure to provide adequate lunge and bob space maynot result in a dramatic reduction in stall occupancy,but it does alter the way cows use the stalls in subtleways. Tucker et al. (2004) found no effect on lyingtimes for ~1600 lb (727 kg) cows housed in stalls 90inches (2.29 m) long or 108 inches (2.74 m) long.There are three possible reasons for this: a). It doesn’tmatter, b). Neither of the choices were long enoughfor front lunge (our recommendation is for cowsweighing ~1600 lbs (727 kg) to be housed in stalls 120inches (3.05 m) long), or c). The cows could sidelunge because the lower divider loop rail wascorrectly located to allow this option. I believe that b.and/or c. are the most likely explanations.

Having seen the results of numerous barns that haveextended side walls to allow 9.5-10 feet (~3.0 m) longside wall stalls, I am convinced that it does make adifference, especially for the largest oldest cows onthe farm. I am also convinced that some cows willwant to side lunge and we should allow that as anoption. This is particularly true of head to head stalls.The presence of one cow in front of another adds anelement of uncertainty in stall design in that somecows will not lie straight or lunge into a cow facingher. This leads to diagonal lying and side lunging.This also has an effect on how clean the stalls are ifwe change stall dimensions. In order to maximizewidth, without running into diagonal lying issues,we must provide adequate length for front lunge. Formature Holstein cows that means stalls that are 10feet (3.05 m) long facing a wall and at least 17 feet(5.18 m) head to head (Figure 4).

Figure 4. Lying position in head to head and sidewall facing stalls and how it is influenced by stallwidth and length and the presence of a socialobstruction in front of the stall.

Locating the Neck RailThe neck rail serves to provide lateral stability to thedivider loops while helping to position the cow whilestanding in the stall relative to the rear curb. It isimportant to realize that the cow on pasture rises andends up standing 2-3 feet (60-90 cm) in front ofwhere she was lying. Therefore, wherever we placethe neck rail, it will be in the way of the cow – even ifit is ‘floating’ or is made of some other material otherthan metal. The trick is to locate it so that the cow cancope and adjust and take a step backward, ratherthan a step forward when she rises. Neck rails do notinfluence lying time much, but they do influencestanding behavior when located between 55 inches(140 cm) and 92 inches 234 cm) from the rear curband between 40 inches (102 cm) and 50 inches (127cm) above the stall surface (Tucker et al., 2005), withlower rails closer to the rear curb increasing theamount of perching (standing half in and half out ofthe stall) observed. We associate these neck raillocations with a greater risk for injury also.

Neck rails that are located too far forward increasesoiling of the stall bed and frequently farmersrespond by moving the rail back closer to the curb.However, if there is insufficient space to risecomfortably below and behind the neck rail, cowshave difficulty standing without hitting the rail –which is just unacceptable. While a contaminatedstall may be a risk for udder infection, an unusedstall is most definitely a risk for inadequate rest,lameness problems and early herd removal. Wetherefore have to find the right balance betweencomfort and cleanliness.

Neck rail location recommendations are different inmattress and sand stalls because the raised rear curbmodifies the way cows stand in the stall – they arereluctant to stand on a raised concrete curb. Neckrails are located in mattress stalls directly above thecorrectly located brisket locator – so that the cow isable to stand squarely in the stall, but in deep beddedstalls with a raised rear curb, where the neck rail is atleast 48 inches (122 cm) above the surface, we movethe rail back a distance equivalent to the width of therear curb, so that the cow takes a step back andperches half in and half out of the stall. While we willnot tolerate this behavior in a flat, mattress stall, weare prepared to tolerate it in a deep loose beddedstall, because the front foot elevation is much less andthe problems of managing a deep bed soiled withurine and feces are too great. While there is someevidence to suggest extended time perching increasesthe risk for lameness (Weary, personalcommunication), this does not seem to be a factor insand stalls as 90% of the stall standing behavior isperching (Cook et al., 2005) and lameness prevalenceis almost half of that on mattresses. This probably

104

relates to the fact that stall standing time is about halfin sand stalls compared with mattress stalls.

Table 4. Target stall dimensions (inches) for cows ofdifferent body weight estimates.

Stall Dimension (inches) Body Weight Estimate (lbs) 1000 1200 1400 1600 1800

Total stall length facing a wall 96 96 108 120 120Distance from rear curb tobrisket locator 64 66 68 70 72Center-to-center stall dividerplacement (Stall width) 44 46 48 50 54Height of brisket locator abovestall surface 3 3 4 4 4Height of upper edge of bottomdivider rail above stall surface 11 11 12 12 12Height below neck rail 44 46 48 50 52Horizontal distance betweenrear edge of neck rail and rearcurb for mattress stalls 64 66 68 70 72Rear curb height 8 8 8 8 8

ConclusionI believe that stall designs which were commonplacein freestall barns up until a few years ago are failingthe cow and our industry in terms of performance,health and well-being. We can and should do better,and it is economical to do so. Numerous barnrenovations have proven this to be the case.However, doing what is right for the cow is not solelyan economic question. Welfare has never been andnever will be an argument based on economics. It is aduty that transcends such discussion. Fortunately, thebalance of welfare and economics are in alignmentwhen we consider improvements to cow comfort.

It is true to say that when we make stalls bigger, thereis greater risk for design error, leading to diagonallying and manure contamination of the stall bed.However, it is also true that a ‘one size fits all policyof restraint’ has also failed. In this discussion, I havedetailed the common errors made when making stallsbigger – using the wrong divider loop, not using abrisket locator or using a poorly designed one, failingto understand the importance of forward lunge andbob space to lying position and locating the neck railincorrectly. Such problems are common becausebuilders and farmers have not built such large stallsbefore and there is much confusing misinformationabout new design philosophies. Hopefully this articlehas put some of these issues to rest.

ReferencesAnderson, N. 2003. Dairy cattle behavior: Cows interacting

with their workplace. Pages 10-22 in Proc. 36th AnnualConvention American Association of BovinePractitioners, Columbus, Ohio.

Bergsten, C., and A.H. Herlin. 1996. Sole hemorrhages andheel horn erosion in dairy cows: The influence ofhousing system on their prevalence and severity. ActaVeterinaria Scandinavia 37, 395-408.

Cook, N.B. 2003. Prevalence of lameness among dairy cattlein Wisconsin as a function of housing type and stallsurface. Journal of American Veterinary MedicalAssociation 223, 1324-1328.

Cook, N.B., T.B. Bennett, and K.V. Nordlund. 2004. Effect offree stall surface on daily activity patterns in dairycows, with relevance to lameness prevalence. J. DairySci. 87:2912-2922.

Cook, N. B., and K. V. Nordlund. 2005. An update on dairycow freestall design. Bovine Practitioner 39:29-36.

Cook, N.B., T.B. Bennett, and K.V. Nordlund. 2005.Monitoring indices of cow comfort in free-stall houseddairy herds. . J. Dairy Sci. 88, 3876-3885.

Cook, N.B. and D. Reinemann. 2007. A tool box forassessing cow, udder and teat hygiene. Pages 31-43 in46th Natl. Mastitis Counc. Mtg. Proc. San Antonio,Texas. Natl. Mastitis Counc., Inc., Madison, WI.

Cook, N.B., and K.V. Nordlund. In press. The influence ofthe environment on dairy cow behavior, claw healthand herd lameness dynamics. Vet. J.doi:10.1016/j.tvjl.2007.09.016.

Espejo, L.A., and M.I. Endres. 2007. Herd-level risk factorsfor lameness in high-producing Holstein cows housedin freestall barns. J. Dairy Sci. 90:306-314.

Grant, R. 2004. Taking advantage of natural behaviorimproves dairy cow performance. Accessed 08/22/08 athttp://www.extension.org .

Ingvartsen, K.L., L. Munksgaard, V.K.M. Nielsen and L.Pedersen. 1999. Responses of repeated deprivation oflying down on feed intake, performance and bloodhormone concentration in growing bulls. Acta. Agric.Scans. A Anim. Sci. 49:260-265.

Jensen, M.B., L.J. Pedersen, and L. Munksgaard. 2005. Theeffect of reward duration on demand functions for restin dairy heifers and lying requirements as measured bydemand functions. Appl. Anim. Behav. Sci. 90:207-217.

Metcalfe, J.A., S.J. Roberts, and J.D. Sutton. 1992. Variationsin blood flow to and from the bovine mammary glandmeasured using transit time ultrasound and dyedilution. Res. Vet. Sci. 53:59-63.

Munksgaard, L., L. Ingvartsen, L.J. Pedersen, and V.K.M.Nielsen. 1999. Deprivation of lying down affectsbehavior and pituitary-adrenal axis responses in youngbulls. Acta. Agric. Scand. A Anim. Sci. 49:172-178.

Munksgaard, L., and P. Løvendahl. 1993. Effect of socialand physical stressors on growth hormone levels indairy cows. Can. J. Anim. Sci. 73:847-853.

Munksgaard, L., and H.B. Simonsen. 1996. Behavioral andpituitary adrenal-axis responses of dairy cows to socialisolation and deprivation of lying down. J. Anim. Sci.74:769-778.

Munskgaard, L., M.B. Jensen, L.J. Pedersen, S.W. Hansen,and L. Matthews. 2005. Quantifying behaviouralpriorities-effects of time constraints on behavior ofdairy cows. Appl. Anim. Behav. Sci. 92:3-14.

Phillips, C.J.C., and M.I. Rind. 2001. The effects onproduction and behaviour of mixing uniparous andmultiparous cows. J. Dairy Sci. 84:2424-2429.

Rulquin, H., and J.P. Caudal. 1992. Effects of lying orstanding on mammary blood flow and heart rate ofdairy cows. Ann. Zootech. (Paris) 41:101.

Sogstad, A.M., T. Fjeldaas, O. Osteras, and K. PlymForshell. 2005. Prevalence of claw lesions in Norwegiandairy cattle housed in tie stalls and free stalls.Preventive Veterinary Medicine 70, 191-209.

105

Tucker, C.B., D.M. Weary, and D. Fraser. 2004. Free-stalldimensions: effects on preference and stall usage. J.Dairy Sci. 87:1208-1216.

Tucker, C.B., D.M. Weary, and D. Fraser. 2005. Influence ofneck rail placement on freestall preference, use andcleanliness. J. Dairy Sci. 88:2730-2737.

Tucker, C.B., G. Zdanowicz, and D.M. Weary. 2006. Brisketboards reduce freestall use. J. Dairy Sci. 89:2603-2607.

Wells S.J., A.M. Trent, W.E. Marsh, and R.A. Robinson. 1993.Prevalence and severity of lameness in lactating dairycows in a sample of Minnesota and Wisconsin dairyherds. Journal of American Veterinary MedicalAssociation 202, 78-82.

IntroductionLow profile cross ventilated (LPCV) freestallbuildings provide a temperate environment thatranges within a dairy cow’s thermoneutral zone evenduring summer and winter months. LPCV buildingstypically maintain an air temperature 8-15∞F coolerthan ambient during the summer in the midwest, butthe relative humidity is often 75% or greater due toevaporative cooling and moisture generated by cows.In the winter the interior of an LPCV building is 10-30∞F warmer than outside air temperatures.

The ability to control a cow’s environment increasesmilk production, improves feed efficiency, raisesincome over feed cost, strengthens reproductiveperformance, allows for controlled lighting, reduceslameness, and lessens fly-control costs. The benefitsof LPCV buildings may be examined by reviewingscientific literature and understanding improvementsthat are possible when an environment complementsa cow’s thermoneutral zone.

Environmental Impact on NutrientRequirements and EfficiencyDairy cows that are housed in an environmentoutside their thermoneutral zone alter their behaviorand physiology in order to adapt. Adaptations arenecessary to maintain a stable core body temperature,but nutrient utilization and profitability arenegatively affected.

The upper critical temperature, or upper limit of thethermoneutral zone, for lactating dairy cattle isapproximately 70-80∞F for maximum nutritionalbenefits (NRC, 1981). When temperatures exceed therecommended range, cows combat heat stress bydecreasing feed intake (Holter at el., 1997), sweating,and panting. These mechanisms increase the cows’energy costs, resulting in up to 35% more feednecessary for maintenance (NRC, 1981). When drymatter intake decreases during heat stress, milkproduction also decreases. A dairy cow in a 100∞Fenvironment decreases milk production by 50% ormore as compared to thermoneutral conditions(Collier, 1985).

Relatively little research has been done on the effectof cold stress on lactating dairy cattle. The highmetabolic rate of dairy cows makes them susceptibleto heat stress in U.S. climates, so, as a result, thelower critical temperature of lactating dairy cattle isnot well established. Estimates range from as high as50∞F (NRC, 1981) to as low at -100∞F (NRC, 2001).Regardless, evidence shows that the performance oflactating cows decreases at temperatures below 20∞F(NRC, 1981).

One clear effect of cold stress is increased feed intake.While greater feed intake often results in greater milkproduction, cold-induced feed intake is caused by anincrease in the rate of digesta passage through thegastrointestinal tract. An increased passage rate limitsdigestion time and results in less digestion astemperatures drop (NRC, 2001). Cows also maintainbody temperature in cold environments by usingnutrients for shivering or metabolic uncoupling, bothof which increase maintenance energy costs. Thesetwo mechanisms decrease milk production by morethan 20% in extreme cold stress. However, even whencold stress does not negatively impact productivity,decreased feed efficiency hurts dairy profitability.

Smith et al (2008) assessed the effects ofenvironmental stress on feed efficiency andprofitability. They used a model which incorporatedthe temperature effects on dry matter intake, dietdigestibility, maintenance requirements, and milkproduction. Figure 1 shows the expected responsesof a cow producing 80 pounds of milk per day in athermoneutral environment. The model was alteredto assess responses to cold stress if milk production isnot decreased. In this situation, the decrease in dietdigestibility results in an 8% decrease in income overfeed cost as temperatures drop to -10∞F. With theseresearch results, cost benefits could be estimated forenvironmental control of LPCV facilities.

106

Are Cross Ventilated Dairy BarnsComfortable?

J.P. Harner1, J.F. Smith1Kansas State University: Seaton Hall 147

Biological and Agricultural Engineering DepartmentManhattan, KS 66506

[email protected]

Figure 1: Responses to Environmental Stress(thermoneutral production of 80 lbs/day, MR Cost of$0.12/lb dry matter, and milk value of $18/cwt)

Environmental Impact on ReproductionEven though cold stress has little effect onreproduction, heat stress reduces libido, fertility, andembryonic survival in dairy cattle. Environmentalconditions above a dairy cow’s thermoneutral zonedecrease the ability to dissipate heat and result in anincreased core body temperature. The elevated bodytemperatures negatively impact reproduction for boththe female and the male.

The impact of heat stress can be categorized by theeffects of acute heat stress (short-term increases inbody temperature above 103oF) or chronic heat stress(the cumulative effects of prolonged exposure to heatthroughout the summer). In acute heat stress, evenshort-term rises in body temperature result in a 25-40% drop in conception rate. An increase of 0.9∞F inbody temperature causes a decline in conception rateof 13% (Gwazdauskas et al.1972). As milk productionand feed intake increase, a greater internal heat loadis produced and the impact of heat stress onreproduction is dramatic (al-Katanani et al., 1999).

Whether the decline in pregnancy rates is voluntaryor not, a fewer number of pregnant cows createsholes in the calving patterns. In the fall an increasednumber of cows often become pregnant and,consequently, place additional pressures on thetransition facilities nine months later when an above-average group of cows must move through the close-up and fresh cow pens. Overcrowding these facilitiesleads to increases in post-calving health issues,decreased milk production, and impaired futurereproduction.

Creating a Thermoneutral Zone HousingEnvironment Changing the environment to reflect the

thermoneutral zone of a dairy cow minimizes theimpact of seasonal changes on milk production,reproduction, feed efficiency and income over feedcost. Evaporative cooling is often used to cool LPCVbuildings, and Harner and Smith (2008) discussspecific design details of the buildings when thiscooling method is utilized. The ability to lower airtemperature through evaporative cooling isdependent upon ambient temperature and relativehumidity. As relative humidity increases, the coolingpotential decreases, as shown in Figure 2. Coolingpotential is the maximum temperature drop possible,assuming the evaporative cooling system is 100%efficient. The cooling potential is greater as airtemperature increases and relative humiditydecreases. Evaporative cooling systems performbetter as the humidity decreases below 50 percent.

The cooling potential is a function of the air’s abilityto absorb moisture. Additional moisture in the airdecreases the air temperature and increases humidity.Theoretically, the lowest possible air temperatureoccurs when the air is at 100% humidity, orsaturation. Most designers assume the airtemperature exiting an evaporative cooling system isreached when the air has absorbed 75% of themoisture possible between inlet conditions andsaturation. Since the outdoor air temperatureconstantly changes, the exit temperature from anevaporative cooling system also changes.

LPCV buildings range in width from 200-600 feet,and the number of rows of freestalls vary from 8-24depending on the building width. The targeted airexchange rate through the buildings is 120 seconds orless, but buildings wider than 300 feet have exchangerates of 180-240 seconds.

Figure 2: Impact of Relative Humidity and Temperatureon Cooling Potential with an Evaporative CoolingSystem

107

0

5

10

15

20

25

30

35

10 20 30 40 50 60 70 80

Relative Humidity (%)

CoolingPo

tential(F

0

5

10

15

20

25

CoolingPo

tential(C

90 F / 32.2 C80 F / 26.7 C100 F / 37.8 C

The body heat generated by the cows warms theinterior of the building. The temperature rise is afunction of the airflow rate through the building.Different management strategies for environmentalcontrol are used during cold weather. The first modedecreases the air exchange rate by turning off fans inorder to prevent frozen manure on the alleys. Thisstrategy prevents potential lameness and injuryproblems but leads to a potential increase inammonia and moisture levels inside the building.The second management strategy utilizes a controllerto operate fans along the inlet side of the building.The disadvantage of this mode is that as the outdoorair temperature declines, the number of operatingfans remains constant. As a result, cold temperaturesare maintained inside the building, manure freezes inalleys near the inlet, and employees are exposed tocolder temperatures.

Though the interior of a LPCV building closelyresembles a naturally ventilated freestall (Harner andSmith, 2008), LPCV buildings incorporate baffles todivert air flow into the stall area. Depending on thenumber of baffles, air speed in the stall area isincreased from 2-3 miles per hour (mph) to6-8 mph during the summer months. Dairies thatutilize baffles observe better lay-down rates of cowsand report a corresponding increase in milkproduction.

Results of Environmental Studies in LPCVBuildingsTable 1 summarizes the temperature rise across LPCVbuildings in the upper Midwest from July 17 toAugust 16, 2007. A temperature increase of 0.85 oFper 100 feet of building width was observed. Sincethe humidity in the building was high due to theevaporative cooling system, approximately a 1 unitincrease in the temperature humidity index (THI)existed per 100 feet of building width.

Table 2 compares the average, maximum andminimum ambient temperatures with the interiorconditions of a 400-foot wide LPCV building in Iowa.The average ambient temperature and relativehumidity from July 17 to August 16, 2007, was 77∞Fand 77%, respectively. The average temperatureinside the LPCV building was approximately 3∞Fcooler than ambient, but the maximum temperaturewas 85∞F as compared to the outside temperature of96∞F. The ambient temperatures were 77∞F or greaterfor over 50% of the study. However, when measurednear the exhaust fans of the LPCV building, theambient temperature was greater than 77∞F only 28%of the time. Also, the ambient temperatures were lessthan 68∞F only 7% of the time, as compared to 12%inside the LPCV building. However, during the night,the indoor temperatures increased because the

evaporative cooling pad was turned off to allow thepad to dry and prolong pad usage.

Table 1: Average Temperature Rise Between Bafflesand Per Foot of Building Width

Nominal Average Temperature (°F)Building Rise/Foot of

Dairy ID Width ft Building Width*# 1 400 0.0085 °F/ft# 2 400 0.0077 °F/ft# 3 520 0.0110 °F/ft# 4 300 0.0095 °F/ft# 5 250 0.0057 °F/ft

Average 0.0085 °F/ft*Average values per dairy are based on 2,880 hourlyaverage measurements including nighttime data.

Table 2: Comparison of Ambient and InteriorTemperatures

Inlet Middle ExhaustAmbient Baffle Baffle Baffle

Average Temperature (°F) 77 73 74 74 Maximum Temperature (°F) 96 85 83 85 Minimum Temperature (°F) 58 58 59 58 Percent of Hours at 77 °F

or above 52 21 24 28Percent of Hours Between

68 to 77 °F 41 67 66 60Percent of Hours Below 68 °F 7 12 10 12

Figure 3 illustrates the average ambient, inlet andexhaust temperatures in a 400-foot wide LPCVbuilding in Iowa from July 17 to August 16, 2008.During the night hours the interior temperatures ofthe building is warmer than ambient. This occursbecause the evaporative cooling system is turned offand the body heat from the cows warm the air as itmoves from the inlet side to the exhaust side of thebuilding.

Figure 3: Comparison of Ambient, Inlet and ExhaustTemperatures

108

68

72

76

80

84

0 3 6 9 12 15 18 21

Time of Day

Tem

perature(F)

20

22

24

26

28

30Tem

perature(C

)AmbientInletExhuast

Temperature data was also logged during the winterof 2008. The data was averaged by hour and bafflelocation from January 18 to February 17, 2008, asshown in Figure 4. The ambient temperature duringthe winter period averaged 20 ∞F colder than barnconditions. Figure 4 shows a rapid warming of the airbetween the inlet and first baffle in two LPCVfacilities, and the air continued to warm untilexhausted from the building. Figure 4 also shows theexhaust air temperature as a function of inlet(outdoor) air temperature. As the outdoor airtemperature decreased, the variability in exhausttemperature increased. The exhaust air temperaturewas 25-45∞F when the inlet air temperature was -5oF.The variability is attributed to a difference in airexchange rates because air temperature is lower atthe exhaust as the air exchange rate increases.

Figure 4: Relationship Between Outdoor AirTemperatures and Outlet (Exhaust) Air Temperaturesin a 400-foot wide LPCV building

Figure 5 shows a correlation between the outdoor airtemperature and the temperature rise across an LPCVbuilding in Minnesota during the winter of 2008.Temperature rise is defined as the difference betweenthe exhaust and outdoor air temperature. Lessvariability exists in the temperature rises above 20∞Fsince there are more consistent strategies in fanoperation and less concern about freezing alleys.

Figure 5: Outdoor Air Temperatures andTemperature Rise in a 500-foot wide building

Figure 6 illustrates the average hourly temperaturesfrom January 18 to February 17, 2008, inside two 400-foot wide LPCV buildings in the upper Midwest. Thedifference in temperature rise from the inlet to theexhaust is explained by different stocking densitiesand air exchange rates.

Figure 6: Summaries of Temperatures in LPCVBuildings

Impact of Geographical LocationFigure 7 shows that ambient temperatures are withinthe thermoneutral zone 65-78% of the time for amajority of major dairy locations in the United States.The data was obtained from a military base near theselected locations. Dairy cows experience more hoursof ambient temperatures below their lowerthermoneutral zone limit (20°F) when housed ondairies in northern states and more hours of ambienttemperatures above their upper thermoneutral zonelimit (70°F) when housed on dairies in southernstates.

The exceptions are Gainesville, FL and Phoenix, AZwhere the ambient temperatures are within thethermoneutral zone only 50% of the year. Yearlyambient conditions result in cows being exposed toheat stress, cold stress or both 25% of the year. Figure 8 shows the potential benefits of LPCVbuildings. The housing environment may beefficiently maintained within the cow’s thermoneutralzone 85-93% of the time. The evaluation is based ontemperature only so, with evaporative cooling andlow relative humidity in Phoenix, AZ, the percentageof hours within the thermoneutral zone could begreater than 65 percent. Research shows that,regardless of location, LPCV buildings increased theannual number of hours the housing environmentmeasured within a cow’s thermoneutral zone by 17percent.

109

Figure 7: Percentage of Annual Ambient Hours inthe Thermoneutral Zone of Dairy Cows

Figure 8: Impact of LPCV Freestall Housing onPercentage of Annual Hours in the ThermoneutralZone

SummaryLPCV facilities are able to minimize fluctuations incore body temperature by providing an environmentwhich closely resembles a cow’s thermoneutral zone.

• Heat stress and cold stress significantly decreaseincome over feed cost. Limiting environmentalstress throughout the year increases feedefficiency.

• Temperatures inside a LPCV building withevaporative cooling are 8-15∞F cooler thanambient temperatures during afternoon hours.

• Temperatures inside a LPCV building during thewinter months are 15-30∞F warmer than ambienttemperatures, depending on the air exchangerate.

• Improving a cow’s environment greatly reducesthe impact of heat stress on present and futuremilk production.

ReferencesAl-Katanani, Yaser M., D.W. Webb, and P.J. Hansen. 1999.

“Factors Affecting Seasonal Variation in 90-DayNonreturn Rate to First Service in Lactating HolsteinCows in a Hot Climate.” Journal of Dairy Science82:2611-2616.

Armstrong, D.V. and P.E. Hillman. 1998. Effect of ColdStress on Dairy Cow Performance. Colorado NutritionConference. Jan 1998.

Berry, I.L., M.D. Shanklin, and H.D. Johnson. 1964. DairyShelter Design Based on Milk Production Decline asAffected by Temperature and Humidity. Transaction ofASAE; 7(3) pp. 329-331.

Bouraqui, R., M. Lahmar, A. Majdoub, M. Djemali, and R.Belyea. 2002. The Relationship of Temperature-Humidity Index with Milk Production of Dairy Cows ina Mediterranean Climate. Anim. Res. 51(2002) 479-491.

Collier, R. J. 1985. “Nutritional, metabolic, andenvironmental aspects of lactation.”

B. L. Larson, ed. Iowa State University Press, Ames, IA.Gwazdauskas, F.C., W.W. Thatcher and C.J. Wilcox 1972.

“Physiological, Environmental, and Hormonal Factorsat Insemination Which May Affect Conception.” Journalof Dairy Science 56:873-877.

Harner, J.P. and J.F. Smith. (editors). 2008. Proceedings ofthe Dairy Housing of the Future: Opportunites withLow Profile Cross Ventilated Housing. Sept 2008. SiouxFalls, ID.

Hillman, P.E., C.N. Lee and S.T. Willard. (2005).Themoregulatory Responses Associated with Lying andStanding in Heat-Stressed Dairy Cows. Transaction ofASAE. 48(2):795-801

Holter, J. B., J. W. West, and M. L. McGilliard. 1997.“Predicting ad libitum dry matter intake and yield ofHolstein cows.” J. Dairy Sci. 80(9):2188-2199.

NRC. 1981. “Effect of Environment on NutrientRequirements of Domestic Animals.” Natl. Acad. Sci.,Washington, DC.

NRC. 2001. “Nutrient Requirements of Dairy Cattle.” 7threv. ed. National Research Council. Natl. Acad. Sci.,Washington, DC.

Overton,M.W. 2006.“Cash Flows of InstitutingReproductive Programs: Cost vs. Reward.” 39thAnnual Convention of the American Association ofBovine Practitioners, 2006.

Smith, J.F. and J.P. Harner. 2008. Low-Profile Cross-Ventilated Freestall Facilities, Another Option for CowHousing, Proceedings of the 2008 Tri-State NorthwestDairy Shortcourse. Boise, ID. Jan, 2008. pp 81-98.

Smith, J.F., J.P. Harner, B.J. Bradford and M. Overton. 2008.Opportunities with Low-Profile Cross-VentilatedFreestall Facilities. Proceedings of the Dairy Housing ofthe Future: Opportunites with Low Profile CrossVentilated Housing. Sept 2008. Sioux Falls, ID.

110

IntroductionHeat stress negatively impacts a variety of dairy andbeef parameters including milk yield, growth andreproduction and therefore is a significant financialburden (~$900 million/year for dairy and > $300million/year in beef in the U.S.; St. Pierre et al., 2003).Advances in management (i.e. cooling systems;Armstrong, 1994) and nutritional strategies (West,2003) have alleviated some of the negative impact ofthermal stress on cattle, but production continues todecrease during the summer. Accurately identifyingheat-stressed cattle and understanding the biologicalmechanism(s) by which thermal stress reduces milksynthesis, growth and reproductive indices is criticalfor developing novel approaches (i.e. genetic,managerial and nutritional) to maintain productionor minimize losses during stressful summer months.

Biological Consequence of Heat StressThe biological mechanism by which heat stressimpacts production and reproduction is partlyexplained by reduced feed intake, but also includesaltered endocrine status, reduction in rumination andnutrient absorption, and increased maintenancerequirements (Collier and Beede, 1985; Collier et al.,2005) resulting in a net decrease in nutrient/energyavailable for production. This decrease in energyresults in a reduction in energy balance (EBAL), andpartially explains (reduced gut fill also contributes)why dairy cattle lose significant amounts of bodyweight when subjected to unabated heat stress.

DairyReductions in energy intake during heat stress resultin a majority of dairy cows entering into negativeenergy balance (NEBAL), regardless of the stage oflactation. Essentially, because of reduced feed andenergy intake the heat-stressed cow enters abioenergetic state, similar (but not to the same extent)to the NEBAL observed in early lactation. TheNEBAL associated with the early postpartum periodis coupled with increased risk of metabolic disordersand health problems (Goff and Horst, 1997; Drackley,1999), decreased milk yield and reduced reproductiveperformance (Lucy et al., 1992; Beam and Butler,1999; Baumgard et al., 2002, 2006). It is likely thatmany of the negative effects of heat stress onproduction, animal health and reproduction indicesare mediated by the reduction in EBAL (similar to thetransition period). However, it is not clear how muchof the reduction in performance (yield, daily gain and

reproduction) can be attributed or accounted for bythe biological parameters affected by heat stress (i.e.reduced feed intake vs. increased maintenance costs).

BeefIn general, heat stress-induced production losses forbeef cattle are not as severe as those for the dairyindustry. It is not entirely clear why growing cattletolerate higher THI conditions and exhibit a greaterheat strain threshold than lactating dairy cows, butmay involve: 1) reduced surface area to mass ratio, 2)reduced rumen heat production (because of themostly grain diet), and 3) reduced overall metabolicheat production (on a body weight basis). Inaddition, beef cattle will often experiencecompensatory gain after mild or short periods of heatstress (Mitlöhner et al., 2001). The combination ofthese factors translate into heat-related reduced gainthat is typically less than 10 kg, which amounts to ~7extra days in the feed lot (St-Pierre et al., 2003).Furthermore, the impact of heat stress onreproductive indices is typically not as severe in beefcattle due to the seasonal nature of breedingprograms (often occurring during the spring in theU.S.).

Metabolic Adaptations to Reduced FeedIntakeA prerequisite to understanding the metabolicadaptations which occur with heat stress, is anappreciation of the physiological and metabolicadjustments to thermal-neutral NEBAL (i.e.underfeeding or during the transition period). Thereis much less known about the metabolic andphysiological effects of hyperthermia in beef cattle ascompared to dairy cows, probably because theeconomic impact on the industry is less severe.Consequently, the changes in heat-related metabolismwill be compared and contrasted primarily to thebetter-known changes in lactating dairy cows.

Early lactation dairy cattle enter a uniquephysiological state during which they are unable toconsume enough nutrients to meet maintenance andmilk production costs and animals typically enterNEBAL (Moore et al., 2005a). Negative energybalance is associated with a variety of metabolicchanges that are implemented to support thedominant physiological condition of lactation(Bauman and Currie, 1980). Marked alterations inboth carbohydrate and lipid metabolism ensure

111

New Concepts on Heat StressL.H. Baumgard1, M.V. Skrzypek2, R.J. Collier2 and R.P. Rhoads2

1Iowa State University & 2University of Arizona, [email protected]

partitioning of dietary and tissue derived nutrientstowards the mammary gland, and not surprisinglymany of these changes are mediated by endogenoussomatotropin which naturally increases duringperiods of NEBAL (Bauman and Currie, 1980). Oneclassic response is a reduction in circulating insulincoupled with a reduction in systemic insulinsensitivity. The reduction in insulin action allows foradipose lipolysis and mobilization of non-esterifiedfatty acids (NEFA; Bauman and Currie, 1980).Increased circulating NEFA are typical in“transitioning” cows and represent (along with NEFAderived ketones) a significant source of energy (andare precursors for milk fat synthesis) for cows inNEBAL. Post-absorptive carbohydrate metabolism isalso altered by the reduced insulin action duringNEBAL with the net effect being reduced glucoseuptake by systemic tissues (i.e. muscle and adipose).The reduced nutrient uptake coupled with the netrelease of nutrients (i.e. amino acids and NEFA) bysystemic tissues are key homeorhetic (an acclimatedresponse vs. an acute/homeostatic response)mechanisms implemented by cows in NEBAL tosupport lactation (Bauman and Currie, 1980). Thethermal-neutral cow in NEBAL is metabolicallyflexible, in that she can depend upon alternative fuels(NEFA and ketones) to spare glucose, which can beutilized by the mammary gland to copiously producemilk.

Heat Stress and Production VariablesDairyHeat stress reduces feed intake and both daily gain inbeef cattle and milk yield in dairy cattle. The declinein nutrient intake has been identified as a major causeof reduced production (Fuquay, 1981; West, 2002,2003). However, the exact contribution of decliningfeed intake to the overall reduced milk yield oraverage daily gain remains unknown. To evaluatethis question in both dairy and beef cattle wedesigned experiments involving a group of thermalneutral pair-fed animals to eliminate the confoundingeffects of dissimilar nutrient intake. First we usedlactating Holstein cows in mid-lactation that wereeither cyclically heat-stressed (THI = ~80 for 16hrs/d) for 9 days or remained in constant thermal-neutral conditions (THI = ~ 64 for 24 hrs/d), but pair-fed with heat stressed cows to maintain similarnutrient intake (Rhoads et al., 2009). Cows werehoused at the University of Arizona’s ARC facilityand individually fed ad libitum a TMR consistingprimarily of alfalfa hay and steam flaked corn tomeet or exceed nutrient requirements (NRC, 2001).Heat-stressed cows had an average rectaltemperature of 40.6°C (~105.1°F) during theafternoons (maximum THI) of the treatment period.Heat-stressed cows had an immediate reduction (~5kg/d) in dry matter intake (DMI) with the decrease

reaching nadir at ~ day 4 and remaining stablethereafter (Figure 1). As expected and by design,thermal-neutral pair-fed cows had a feed intakepattern similar to heat-stressed cows (Figure 1). Heatstress reduced milk yield by ~14 kg/d withproduction steadily declining for the first 7 days andthen reaching a plateau (Figure 2). Thermal neutralpair-fed cows also had a reduction in milk yield ofapproximately 6 kg/d, but milk production reachedits nadir at day 2 and remained relatively stablethereafter (Figure 2). This indicates the reduction inDMI can only account for ~40-50% of the decrease inproduction when cows are heat-stressed and that~50-60% can be explained by other hyperthermia-induced changes. We have repeated this experimentmultiple times and the effects on DMI and milk yieldare remarkably consistent (Wheelock et al., 2006;Baumgard & Rhoads, unpublished)

BeefTo evaluate the differential effects of heat stress vs.reduced nutrient intake in beef cattle we studiedgrowing Holstein beef bulls (n=12, 4-5 months of age,136-182 kg BW; O’Brien et al., 2008). Bulls wereeither cyclically heat-stressed (29.4 to 40ºC, 25-40%humidity, and 12 hours of light [conditions slightlywarmer than during our dairy experiments]) or weremaintained in thermal-neutral conditions (18 to 20ºC,12 hours of light), but pair-fed (86% concentrate, 14%protein, 2x/d) with heat-stressed bulls to maintainsimilar nutrient intake. Heat-stressed bulls had anaverage rectal temperature of ~ 40.6°C (105.1°F)during the afternoons (peak ambient THI). Heatstress reduced DMI by ~12% (data not presented) andas expected (and by design) thermal-neutral pair-fedbulls had a feed intake pattern similar to heat-stressed cows (O’Brien et al., 2008). Heat stresseliminated body weight gain and thermal neutralpair-fed animals had a similar reduction inperformance (Figure 3; O’Brien et al., 2008).

Figure 1. Effects of heat stress and underfeeding(pair-feeding) thermal-neutral lactating Holstein cowson dry matter intake (Rhoads et al., 2009).

112

113

Figure 2. Effects of heat stress and underfeeding(pair-feeding) thermal neutral conditions on milkyield in lactating Holstein cows (Rhoads et al., 2009)

Figure 3. Effect of heat stress (HS) and underfeeding(UF; pair-feeding) during thermal neutral conditionson body weight gain (slopes of raw data) in growingHolstein beef bulls (O’Brien et al., 2008).

Dairy vs. BeefDespite being exposed to a slightly greater heat load,heat stress does not reduce DMI to the same extent ingrowing beef cattle as it does in lactating dairy cows(12 vs. 30%). In addition, the reduction in feed intakeaccounts for only ~50% of the decrease in milk yield(Figure 2), but appears to explain most (if not all) ofthe reduction in growth (Figure 3). Gaining a betterappreciation for the biological reasons underlying theaforementioned discrepancy between beef and dairymay theoretically provide insight on how to preventor ameliorate the exaggerated decrease in milksynthesis during hyperthermia.

Heat and Maintenance CostsEstimating EBAL during heat stress (for both dairyand beef cattle) introduces problems independent ofthose that are inherent to normal EBAL estimations(Vicini et al., 2002). Considerable evidence suggestincreased maintenance costs are associated with heatstress (7 to 25%; NRC, 2001), however due tocomplexities involved in predicting upper criticaltemperatures, no universal equation is available to

adjust for this increase in maintenance (Fox andTylutki, 1998). Maintenance requirements arethought to increase, as there is presumably a largeenergetic cost of dissipating stored heat. Notincorporating a heat stress correction factor results inoverestimating EBAL and thus inaccuratelypredicting energy status.

In the beef study, the pair-fed thermal-neutralcontrols did not gain or lose body weight (Figure 3),suggesting nutrient and energy intake satisfiedmaintenance requirements. The heat-stressed bullsconsumed similar quantities of the exact ration fed tothe pair-fed thermal-neutral control animals and alsohad static body weight. This latter observation mayindicate that, at least in growing bulls, heat stressdoes not increase overall maintenance requirements.If heat stress were to increase maintenance costs asreported (Fox and Tylutki, 1998; NRC, 2001) then theenergy requirements of heat-stressed bulls shouldhave exceed their pair-fed thermal-neutralcounterparts. In turn, the heat-stressed bulls wouldhave been consuming inadequate energy/nutrientsand should have (by definition) lost body weight.However, this was not the case and heat-stressedbulls did not lose body weight (Figure 3), indicatingthat maintenance costs may not have been increased.Further research is necessary to evaluate the effects ofheat on maintenance requirements and to determineif physiological state (growth vs. lactation) influencesenergy partitioning during thermal challenges.

Metabolic Adaptations to Heat StressDairyDue to the reductions in feed intake and presumedincreased maintenance costs, and despite the decreasein milk yield heat stressed cows enter into a state ofNEBAL (Moore et al., 2005b). In a similar dairy trialto the one described above, heat-stressed cowsentered into and remained in NEBAL (~4-5 Mcal/d)for the entire duration of heat stress (Figure 4;Wheelock et al., 2006). However, unlike NEBAL inthermal-neutral conditions, heat-stressed inducedNEBAL doesn’t result in elevated plasma NEFA(Figure 5). This was surprising as circulating NEFAare thought to closely reflect calculated EBAL(Bauman et al., 1988). In addition, using an IVglucose tolerance test, we demonstrated that glucosedisposal (rate of cellular glucose entry) is greater inheat-stressed compared to thermal neutral pair-fedcows. Furthermore, heat-stressed cows have a muchgreater insulin response to a glucose challenge whencompared to underfed cows (data not presented).Both the aforementioned changes in plasma NEFAand metabolic/hormonal adjustments in response toa glucose challenge can be explained by increasedinsulin effectiveness. Insulin is a potent anti-lipolyticsignal (blocks fat break down) and the primary driver

114

of cellular glucose entry. The apparent increasedinsulin action causes the heat-stressed cow to bemetabolically inflexible, in that she does not have theoption to oxidize fatty acids and ketones. As aconsequence, the heat-stressed cow becomesincreasingly dependant on glucose for her energeticneeds and therefore less glucose is directed towardsthe mammary gland.

BeefAlthough both the heat-stressed and pair-fed controlsquit growing, neither mobilized adipose tissue(plasma NEFA remained <100 µEq/L), which isagreement with a lack of body weight loss (O’Brien etal., 2008). However, despite similar changes inproduction and post-absorptive lipid variables, therewere heat stress-induced changes in post-absorptivecarbohydrate metabolism. Similar to lactating dairycows, heat-stressed growing bulls appear to have anincrease in glucose disposal rates and have a muchgreater insulin response to a glucose challenge(Figure 6, O’Brien et al., 2008).

Figure 4. Effects of heat stress and underfeeding(pair-feeding) thermal-neutral conditions oncalculated net energy balance in lactating Holsteincows (adapted from Wheelock et al., 2006.)

Figure 5. Effects of heat stress and underfeeding(pair-feeding) thermal-neutral conditions oncirculating non-esterified fatty acids (NEFA) inlactating Holstein cows (adapted from Wheelock etal., 2006).

Figure 6. Effects of heat stress (HS) and underfeeding(UF; pair-feeding) in thermal-neutral conditions onplasma insulin response to a glucose challenge ingrowing beef cattle (O’Brien et al., unpublished).

Theoretical Reasons for AlteredMetabolismWell-fed ruminants primarily oxidize (burn) acetate(a rumen produced VFA) as their principal energysource. However, during NEBAL cattle also largelydepend on NEFA for energy. Therefore, it appearsthe post-absorptive metabolism of heat-stressed cattlemarkedly differs from that of thermal-neutral cattle,even though they are in a similar negative energeticstate. The apparent switch in metabolism and theincrease in insulin sensitivity is probably amechanism by which cattle decrease metabolic heatproduction, as oxidizing glucose is more efficient(Baldwin et al., 1980). In vivo glucose oxidationyields 38 ATP (assuming the DG of ATP hydrolysis is-12.3 kcal/mole under cellular conditions; Berg et al.,2007) or 472.3 kcal of energy (compared to 637.1 kcalin a bomb calorimeter) and in vivo fatty acidoxidation (i.e. stearic acid) generates 146 ATP or 1814kcal of energy (compared to 2697 kcal in a bombcalorimeter). Despite having a much greater energycontent, due to differences in the efficiencies ofcapturing ATP, oxidizing fatty acids generates moremetabolic heat (~2 kcal/g or 13% on an energeticbasis) compared to glucose. Therefore, during heatstress, preventing or blocking adiposemobilization/breakdown and increasing glucose“burning” is presumably a strategy to minimizemetabolic heat production (Baumgard and Rhoads,2007).

For dairy cattle, the mammary gland requires glucoseto synthesize milk lactose and lactose is the primaryosmoregulator and thus determinant of milk volume.However, in an attempt to generate less metabolicheat, the body (primarily skeletal muscle) appears toutilize glucose at an increased rate. As aconsequence, the mammary gland may not receiveadequate amounts of glucose and thus mammarylactose production and subsequent milk yield is

115

reduced. This may be the primary mechanism whichaccounts for the additional reductions in milk yieldbeyond the portion explained by decreased feedintake (Figures 1 and 2).

Heat-stressed cattle require special attention withregards to heat abatement and other dietaryconsiderations (i.e. concentrate:forage ratio, HCO3-etc; Baumgard et al., 2007). In addition they may alsohave an extra requirement for dietary or rumen-derived glucose precursors. Of the three mainrumen-produced VFA’s, propionate is the VFAprimarily converted into glucose by the liver. Oneoption to increase rumen propionate production is byfeeding highly fermentable starches. However, thisstrategy may be risky as heat-stressed cattle arealready susceptible to rumen acidosis. Furtherresearch is needed to identify safe methods ofincreasing dietary or rumen derived glucoseprecursors during heat stress conditions.

SummaryClearly heat-stressed cattle implement a variety ofpost-absorptive changes in both carbohydrate andlipid metabolism (i.e. increased insulin action) thatwould not be predicted based upon their energeticstate. The primary end result of this alteredmetabolic condition is that heat-stressed cattle havean extra need for glucose (theoretically due to itspreferential oxidization in order to reduce metabolicheat). Therefore, any dietary component thatincreases propionate production (the primaryprecursor to hepatic glucose production), withoutreducing rumen pH, will probably increaseproduction.

Note: This article has been partially adapted from apaper first published by the authors in theProceedings in the 2007 University of ArizonaSouthwest Nutrition Conference and 2008 FloridaNutrition Conference.

ReferencesArmstrong, D.V. 1994. Heat stress interaction with shade and

cooling. J. Dairy Sci. 77:2044-2050.Baldwin, R.L., N.E. Smith, J. Taylor, and M. Sharp. 1980.

Manipulating metabolic parameters to improve growth rateand milk secretion. J. Anim. Sci. 51:1416-1428.

Bauman, D.E. and W.B. Currie.1980. Partitioning of nutrientsduring pregnancy and lactation: a review of mechanismsinvolving homeostasis and homeorhesis. J. Dairy Sci. 63:1514-1529.

Bauman, D.E., C.J. Peel, W.D. Steinhour, P.J. Reynolds, H.F. Tyrrell,C. Brown, and G.L. Harland. 1988. Effect of bovinesomatotropin on metabolism of lactating dairy cows: influenceon rates of irreversible loss and oxidation of glucose andnonesterified fatty acids. J. Nutr. 118:1031-1040.

Baumgard, L.H., L.J. Odens, J.K. Kay, R.P. Rhoads, M.J. VanBaaleand R.J Collier. 2006. Does negative energy balance (NEBAL)limit milk synthesis in early lactation? Proc. Southwest Nutr.Conf. 181-187.

Baumgard, L.H. and R.P. Rhoads. 2007. The effects of hyperthermiaon nutrient partitioning. Proc. Cornell Nutr. Conf. pp 93-104.

Beam, S.W., and W.R Butler. 1999. Effects of energy balance onfollicular development and first ovulation in postpartum dairycows. J. Reprod. Fertility 54:411-424.

Berg, J.M., J.L. Tymoczko and L. Stryer. Biochemistry. 6th Edition.W.H. Freeman.

Collier, R.J., L.H. Baumgard, A.L. Lock and D.E. Bauman. 2005.Physiological Limitations: nutrient partitioning. Chapter 16. In:Yields of farmed Species: constraints and opportunities in the21st Century. Proceedings: 61st Easter School. Nottingham,England. J. Wiseman and R. Bradley, eds. NottinghamUniversity Press, Nottingham, U.K. 351-377.

Collier, R.J., and D.K. Beede. 1985. Thermal stress as a factorassociated with nutrient requirements and interrelationships.In Nutrition of Grazing Ruminants. (ed) by L. McDowell.Academic Press, New York, NY. pp 59-71.

Drackley, J.K. 1999. Biology of dairy cows during the transitionperiod: the final frontier? J. Dairy Sci. 82:2259-2273.

Fox, D.G. and T.P. Tylutki. 1998. Accounting for the effects ofenvironment on the nutrient requirements of dairy cattle. J.Dairy Sci. 81:3085-3089.

Fuquay, J.W. 1981. Heat stress as it affects production. J. Anim. Sci.52:167-174.

Goff, J.P. and R.L. Horst. 1997. Physiological changes at parturitionand their relationship to metabolic disorders. J. Dairy Sci.80:1260-1268.

Lucy, M.C., C.R. Staples, W.W. Thatcher, et al.,.1992. Influence ofdiet composition, dry matter intake, milk production andenergy balance on time of postpartum ovulation and fertilityin dairy cows. Anim. Prod. 54:323-331.

McDowell, R.E., E.G. Moody, P.J. Van Soest, R.P. Lehmann and G.L.Ford. 1969. Effect of heat stress on energy and water utilizationof lactating cows. J. Dairy Sci.. 52: 188-194.

Mitlöhner, F.M., J.L. Morrow, J.W. Dailey, S.C. Wilson, M.L.Galyean, M.F. Miller, and J.J. McGlone. 2001. Shade and watermisting effects on behavior, physiology, performance, andcarcass traits of heat-stressed feedlot cattle. J. Anim. Sci.79:2327-2335.

Moore, C.E., J.K. Kay, M.J. VanBaale and L.H. Baumgard. 2005a.Calculating and improving energy balance during times ofnutrient limitation. Proc. Southwest Nutr. Conf: 173-185.

Moore, C.E., J.K. Kay, M.J. VanBaale, R.J. Collier and L.H.Baumgard. 2005b. Effect of conjugated linoleic acid on heatstressed Brown Swiss and Holstein cattle. J. Dairy Sci. 88:1732-1740.

National Research Council. 2001. Nutrient Requirements of DairyCattle, 7th rev. ed. Nat. Acad. Press, Washington, DC.

O’Brien, M.D., J.B. Wheelock, S.R. Sanders, G.C. Duff, R.P. Rhoadsand L.H. Baumgard. 2008. Differential effects of heat stressand reduced nutrient intake on production and metabolism inyoung growing beef cattle. J. Anim Sci. 86:E-Suppl 2: 349

Rhoads, M.L., R.P. Rhoads, M.J. VanBaale, R.J. Collier, S.R. Sanders,W.J. Weber, B.A. Crooker, and L.H. Baumgard. 2009. Effects ofheat stress and plane of nutrition on lactating Holstein cows: I.production, metabolism and aspects of circulatingsomatotropin. J. Dairy Sci. 92:1986-1997.

St. Pierre, N.R., B. Cobanov, and G. Schnitkey. 2003. Economiclosses from heat stress by US livestock industries. J. Dairy Sci.86:E52-E77.

West, J.W. 2002. Physiological effects of heat stress on productionand reproduction. Proc. Tri-State Nutr. Conf. 1-9.

West, J.W. 2003. Effects of heat-stress on production in dairy cattle.J. Dairy Sci. 86:2131-2144.

Wheelock, J.B., S.R. Sanders, G. Shwartz, et al.. 2006. Effects of heatstress and rbST on production parameters and glucosehomeostasis. J. Dairy Sci. 89. Suppl. (1):290-291 (abst.).

IntroductionLarge fluctuations in milk and supplemental feedprices create anxiety and uncertainties. Profit marginmay shrink rapidly when milk price goes down andsupplemental feed prices go up. Thus, it is importantthat correct decisions are made to maximize return onsupplemental feed expenses. The change in pricesand milk production response to feed supplementsare especially important because they impact directlythe profitability of dairy farms. Supplements mightrepresent the largest portion of expenses and milk isby far the most important revenue generator in adairy enterprise. Usually, more than 90% of dairyfarm revenue comes from the milk check and morethan 40% of the expenses are used on purchasedfeeds. Analyses from surface responses to incomeover feed cost for different crude protein (CP) levelshave been studied in the past (Roffler et al., 1986), butthe distinction between rumen undegradable protein(RUP) protein and rumen degradable protein (RDP)creates a need to further fine-tune the formulation ofsupplements for maximum income over feed cost.Volatile market conditions that are greatly impactingfeeds and milk prices require the ability to makestrategic ration formulation adjustments promptlyand efficiently to maintain dairy margins and remaineconomically competitive. An interactive simulationmodel may be beneficial in exploring those strategicdecisions in response to fluctuating milk and feedprices.

JustificationTraditional diet formulation is based on finding theleast cost ration that provides the minimum level ofrequired nutrients for a desired level of milkproduction (Tozer, 2000; Howard et al., 1968).Typically, diet formulation does not consider changesin milk production due to changes in CP, RUP andRDP that could be fine-tuned to maximize incomeover feed supplement costs. Rotz et al. (1999) foundthat profitability of dairy farms could be improved bydecreasing CP intake and adjusting RUP and RDPthrough a better selection of fed ingredients, whichvary according to market prices of feed stuffs. Thisprevious study also found that lower CP dietsdecrease N excretion and consequentlyenvironmental impacts, which has been confirmed bya number of recent publications (Broderick, 2003;Wattiaux and Karg, 2004). Rotz et al. (1999)

developed the dairy farm model (DAFOSYM)capable to estimate the income over supplementcosts, which nowadays has evolved to the integratedfarm system model (IFSM) (Rotz et al., 2007). TheIFSM, although very complete, is complex and servesthe scientific community more than field-based end-users. In addition, the IFSM model does not performoptimization, but rather simulation of scenarios. Thispaper main purpose is therefore to present a simpleformulation to optimize income over feedsupplement costs (IOFSC), implement theformulation in a user-friendly spreadsheet, andperform some case studies.

Maximum IOFSCAs it is usual in many dairy farms, forages producedeither on-farm or that are locally available can beconsidered a fixed proportion in the diet, at least inthe short-run. Consequently, the optimizationproblem can be simplified by discounting the drymatter (DM) and CP provided by fixed amounts offorages from the total needs. The problem to solvethen becomes that of optimizing the income over feedsupplement costs (IOFSC) given feed supplementcosts, milk price and cow's milk production responseto dietary CP.

Thus, the objective function is to maximize the IOFSCof a diet formulation:

max(MV – ∑ SVi) [1]

where MV is the milk value calculated as the milkprice (Mp) times the milk production (MP), SVi is thevalue of the i supplement defined as the price of thesupplement (Spi) times the quantity of suchsupplement (SQi) when there are N availablesupplements in the diet, which need to provide theexpected DM or dry matter intake (DMI) under theconstraint of an upper limit for RUP, RDP, and CP;Therefore,

∑=DMI [2]

RUP≤max RUP, RDP≤max RDP, CP≤max CP [3]

and the DMI is calculated following NRC (2001, Eq.1-2),

116

Optimizing Income Over FeedSupplement Costs

Victor E. Cabrera, Randy D. Shaver, and Michel A. WattiauxDepartment of Dairy Science, University of Wisconsin

1675 Observatory Dr., Madison, WI 53705, [email protected]

N

i = 1

N

i = 1

DMI = (0.372*FCM+0.0968*BW0.75)*(1-e(-0.192*((WOL+3.67))) [4]

where FCM is fat corrected milk, BW is body weightand WOL is week on lactation. An initial target MP isdefined to calculate FCM and from it DMI, however,depending on the selection of supplements, theexpected MP will change. Therefore final MP iscalculated using NRC (2001, Pg. 50),

MP=–55.61+1.15*DMI+8.79*RDP–0.36*RDP2+1.85*RUP[5]

The formulation must include a maximum limit offeed supplements, which are defined by uppermaximum levels. The concept of upper levels allowthe exclusion of some non-available ingredients(upper limit set at zero) or to put nutritional andbiological limits to some ingredients. Therefore,

SQi≤max SQi for i-1 to N for i=1 to N [6]

Supplements can be grouped as those that providemore energy than protein (e.g., corn grain) and thosethat provide more protein than energy (e.g., soy beanmeal). Both type of supplements need to be definedaccording to their composition of RUP, RDP and CP.Whereas CP is the sum of RUP and RDP, RUP andRDP are influenced by degradability and ruminaldisappearance of protein through two simultaneousprocesses, degradation and passage (NRC, 2001). Afraction "A" is assumed to be instantly degraded, afraction "C" is assumed to be completelyundegradable, and a fraction "B" is assumed to bedegradable overtime at a rate of degradation (Kd).Therefore, knowing a constant rate of degradation(Kd) and a rate of passage (Kp), it is possible tocalculate the RUP and RDP content of feed stuffs.Research has shown that the rate of passage Kp isdependent on the DMI, BW, feed characteristics, andproportions of forage and supplements in diet. Table1 provides an example of calculated proteincomposition for forages and supplements along withtheir calculated RUP and RDP values.

Table 1. Calculated protein availability of feed stuffsas rumen undegradable protein (RUP), rumendegradable protein (RDP) and crude protein (CP)based on standard values of some forages, energysupplements, and protein supplements.

CalculatedFeed Stuff A B C Kd Kp RUP RDP CP

(%) (%) (%) (%) (%) (%)Forages35-Corn

silage 51.00 5.62 8.80 4.40 5.93 3.15 5.62 8.8074-Mixed

silage 58.10 34.20 7.70 10.40 5.93 3.82 15.18 19.0083-Alfalfa

silage 57.30 35.30 7.40 12.20 5.93 4.15 17.75 21.90Energy Supplements27-Corn

grain 23.90 72.5 3.60 4.90 8.34 4.63 4.77 9.408-Barley

grain 30.20 61.20 8.60 22.70 8.34 3.11 9.29 12.40Protein Supplements106-Soybean

meal 22.50 76.80 0.70 9.40 8.34 18.37 31.53 49.9025-Corn gluten

meal 3.90 90.90 5.20 2.30 8.34 49.69 15.31 65.0023-Corn distiller

grains 28.50 63.30 8.20 3.60 8.34 15.57 14.13 29.70104-Soybean meal

expellers 8.70 91.30 0.00 2.40 8.34 32.83 13.47 46.30

Source: Adapted from NRC (2001, Table 15-1). Codenumbers preceding the feed name correspond toNRC codes. Calculated values are based on DMI of29.26 kg/d for a 625-kg BW cow with an estimatedmilk production of 50 kg/d.

A Case StudyConsider this situation for a group of Holstein cowswith an average of 100 days in lactation producing36.4 kg (80 lb) milk per day. According to theNational Research Council equations, these cowsshould be consuming on average 24.4 kg (53.66 lb) ofDM per day. We can safely assume that 50% of theDMI is provided through forage, which is composedof equal parts of corn silage and alfalfa silage.Therefore, these cows are already receiving 7.7% ofCP (1.8% RUP and 5.9% RDP) from the forages. If thetarget protein in diet is 18% (6.5% RUP and 11.5%RDP), then, the supplements could provide thedifference: up to 9.2% CP (4.8% RUP and 5.6% RDP).As previously discussed, supplement sources havedifferent compositions of CP, RUP, and RDP, whichnot only complete the protein requirements, but alsoimpact milk production. After setting the proportionof forage in the diet to a fixed value, the goal is thento find the combination of supplements that willmaximize the IOFSC.

Let's suppose that the group of cows described abovecurrently receives 9.5 kg (20.9 lb) of corn grain and2.7 (6.0 lb) of soybean meal as supplemental feeds.Under this diet, each cow in this group would haveon average $5.20 a day IOFSC using Wisconsin

117

February 2009 feed and milk prices: $0.207/kg($9.4/cwt) milk, $0.14/kg ($3.54/bu) corn and$0.33/kg ($300/ton) soybean meal. However, underthe same price structure, the IOFSC could be as muchas $5.54 a day if the sources and proportions ofsupplemental feeds are fine-tuned. A $0.34 per cowper day additional IOFSC could be realized if thesupplements are 8 kg (17.6 lb) of corn grain, 1.91 kg(4.2 lb) of soybean meal, and 2.27 kg (5 lb) of corndistiller grains at $0.154/kg ($140/ton).

User-Defined AnalysesThe actual combination of supplement will varysubstantially depending on the availability and costsof feed supplements, and price received for milk.Every situation is different and because of that wehave created a user-friendly computerized (PC-based) tool that could help with this task. The IOFSCalong with a fact sheet and instructions are freelyavailable at the University of Wisconsin ExtensionDairy Management website:http://www.uwex.edu/ces/dairymgt/ under"Management Tools" with title "Income Over FeedSupplement Costs." This Excel spreadsheet tool(Figure 1) uses linear programming to maximize theIOFSC for user-defined parameters (Equations 1through 6) of cow productive characteristics,availability of supplements, economic parameters,NRC (2001, Table 15-1) nutrient composition of feeds,and calculated RUP, RDP and CP. After finding theoptimal IOFSC, the tool could still further assist theuser to determine whether a decrease in CP studiedby substituting supplements is justified underdefined conditions. For convenience, the spreadsheetis divided in 10 successive sections that follow alogical order to understand, analyze, optimize theIOFSC, and substitute feed supplements. Userdefined data is entered in yellow marked cells andresults are seen in blue marked cells, figures andtables. The user selects whether to work on metric orEnglish units, which will be consistent across thespreadsheet.

In section 1, the user enters milk production (MP),body weight (BW) and days in milk (DIM) as inputsto calculate dry matter intake (DMI). Example: TheDMI of a milking cow producing 41 kg milk/day (90lb milk/day), weighting 636 kg (1,400 lb), and 180days in lactation is calculated at 26.3 kg/day (57.9lb/day). Further, if the user finds that this calculatedamount does reflect known conditions, a "spinbutton" is provided to adjust the calculated amountto desired or known level.

Section 2 lets the user select the sources andproportion of forages in the diet. The application thenuses the NRC feed composition table values tocalculate CP, RDP, and RUP provided by the forage

sources. Example: A diet containing 50% of forage, ofwhich 50% is corn silage and 50% is alfalfa silage willprovide 2.0 kg of CP (4.4 lb) or 7.7% CP of the diet(1.8% RUP and 5.9% RDP).

In sections 3 & 4, the user defines up to 3supplemental energy feed stuffs: corn, barley, andwheat; and up to 8 supplemental protein feed stuffs:soybean meal, corn gluten meal, corn gluten feed,corn distiller, whole soybeans, soybean mealexpellers, blood meal, and urea. Furthermore, theuser needs to define the price and the upper limit(maximum per day per cow) for each one of the feedsupplements. Supplements can be excluded from theoptimization by either entering zero or leaving ablank in the corresponding upper limit cell. The usercan enter the amounts of supplements in the currentdiet as an option. If this is done, the spreadsheet willprovide a comparison of the IOFSC for the currentdiet to the optimized one. Example: As shown theuser can set the price ($/bu) and the upper limit (lb)of the energy supplements to: corn (4 & 15) andwheat (7.4 & 10); the price ($/ton) and the upperlimit (lb) of the protein supplements to: soybeanmeal (250 & 15), corn gluten meal (550 & 2), corngluten feed (160 & 10), corn distiller grains (200 & 10),soybean meal expellers (402 & 15), blood meal ringdried (900 & 1), and urea (635 & 1). Optionally, theuser can set up the amount of these ingredientscurrently in the diet such as (lb): corn grain (10),wheat (1.5), soybean meal (5), corn gluten feed (5)and corn distiller grains (5).

Table 2. Exemplified prices, upper limits and currentamounts used to perform analyses with the IncomeOver Feed Supplement Costs application.

Upper CurrentFeed Stuff Price Limit in DietEnergy Supplements $/kg $/bu kg lb kg lb27-Corn grain 0.16 4.0 6.81 15 4.54 10Wheat grain 0.27 7.4 4.54 10 0.68 1.5

Protein Supplements $/kg $/ton kg lb kg lb106-Soybean meal 0.28 250 6.81 15 2.27 525-Corn gluten meal 0.61 550 0.91 224-Corn gluten feed 0.18 160 4.54 10 2.27 523-Corn distiller grains 0.22 200 4.54 10 2.27 5104-Soybean meal expellers 0.20 402 6.81 1514-Blood meal ring dried 0.99 900 0.45 1Urea 0.70 635 0.45 1

In section 5, the user sets an upper limit for dietaryRUP and RDP and enters an appropriate milk price.Example: The upper limit for RUP and RDP are set to6.5 and 11.5%, respectively, and the price of milk to$16/cwt.

Section 6 is where the user performs the optimizationanalysis by just clicking the "Maximize IOFSC"button (Figure 1) to perform a maximization that will

118

calculate the amount of each selected supplementalfeed to maximize IOFSC. Example: The maximumIOFSC for the working case is $11.64/cow/dayproducing 40.0 kg (88.32 lb) of milk/day using 5.22kg (11.5 lb) of corn, 0.17 kg (0.38 lb) of corn glutenmeal, 3.2 kg (7.0 lb) of corn gluten feed, and 4.54 kg(10 lb) of corn distiller grains. Using the current diet,the IOFSC is calculated to be $11.06/cow/day ($0.58lower than optimal), producing 38.25 kg (84.26 lb) ofmilk/day.

Once the user has the optimized quantities ofsupplemented feed, the spreadsheet in sections 7 to 10allows analyzing the impacts of substituting feedsupplements and performing sensitivity analyses ofprice change. Before performing a substitutionanalysis though, additional information is needed.

Figure 1. Income Over Feed Supplement Cost(IOFSC) application. Sections pertaining tomaximization of the IOFSC.

The user needs to define in section 7 the two feedsupplements to be dynamically substituted and apotential price range of change of the energysupplement, the protein supplement, or the milk.

Figure 2. Figure 1. Income Over Feed SupplementCost (IOFSC) application. Sections pertaining toperform substitution analyses.

Once these are defined, the substitution analysis canbe performed by just clicking the "PerformSubstitution" button (Figure 2). Results are displayedas figures and a table in sections 8 to 10. Example:Graph in section 8 (Figure 2) displays the sensitivityof milk production and IOFSC in response to achange in dietary CP (DM basis) as a result ofsubstituting corn grain ($4/bu) for corn gluten feed($160/ton) when milk price is $16/cwt. This graphshows that milk production continues to increase asdietary CP increases above 17.5%, while the IOFSCreaches a plateau and even decreases.

The example discussed shows that feeding more than17.5% CP may result in higher milk yield, butwithout additional net income. Graph in section 9(Figure 2) displays the sensitivity of IOFSC inresponse to a change in dietary CP (DM basis) as aresult of substituting corn grain for corn gluten feedwhen corn gluten feed price is $240/ton (50% aboveregular price, upper price) and when corn gluten feedprice is $80/ton (50% below regular price, lowerprice). Compared with graph in section 8, the IOFSCin response to the substitution of corn grain for corngluten feed reaches a plateau earlier when corngluten feed price is high ($240/ton) and later when

119

corn gluten feed price is low ($120/ton).Consequently, lower CP levels may be justified whenthe price of corn gluten feed is high. Section 10 (notshown in Figure 2) displays the tabular data ofgraphs presented in sections 8 and 9.

ReferencesBroderick, G.A. 2003. "Effects of Varying Dietary Protein

and Energy Levels on the Production of Lactating DairyCows." J. Dairy Sci 86: 1370-1381.

Howard, T. J.L. Albright, M.D. Cunningham, R.B.Harrington, C.H. Noller, and R.W. Taylor. 1968. Least-cost complete rations for dairy cows. J Dairy Sci 51: 595-600.

National Research Council (NRC). 2001. Nutrientrequirements for dairy cattle. 7th Revised Ed. TheNational Academies Press, Washington DC.

Roffler, R.E., J.E. Wray, and L.D. Satter. 1986. Productionresponses in early lactation to additions of soybean mealto diets containing predominantly corn silage. J. DairySci. 69:1055-1062.

Rotz, C.A., L.D. Satter, D.R. Mertens, and R.E. Muck. 1999.Feeding strategy, nitrogen cycling, and profitability ofdairy farms. J. Dairy Sci. 82:2841-2855.

Rotz, C.A. 2004. The integrated farm system model: a toolfor developing more economically and environmentallysustainable farming systems for the Northeast. ASABESection Meeting Paper NABEC04-0022.

Tozer, P.R. 2000. Least cost ration formulations for Holsteindairy heifers by using linear programming andstochastic programming. J. Dairy Sci. 83: 443-451

Wattiaux, M. A. and K. L. Karg (2004). "Protein Level forAlfalfa and Corn Silage-Based Diets: II. NitrogenBalance and Manure Characteristics." J. Dairy Sci.87(10): 3492-3502.

Appendix: Fact Sheet and Instructions of Use for IncomeOver Feed Supplement Costs (IOFSC)

The Income Over Feed Supplement Cost (IOFSC) is anExcel application (PC-based) that needs to be installed in auser local machine:

1. Install the IOFSC application

1.1. Prepare Excel Open a clean copy of Excel on your computerExcel 2003: a) Tools/Macro/Security... Select Low

Security levelb) Tools/Add-Ins... if not checked, checkSolver Add-in, then OK c) Tools/Solver (make sure solver opens inyour spreadsheet)d) close Solver. Your Excel 2003 is ready.

Excel 2007: a) Ribbon (upper left corner)/Excel options b) Trust Center/Trust Center Settings/MacroSettings/ c) Select Enable all macros, then OKd) Add-Ins/ Manage: Excel Add-Ins Go...e) If not checked, check Solver Add-in, thenOKf) Add-Ins/Solver (make sure solver opens inyour spreadsheet)g) Close Solver. Your Excel 2007 is ready.

1.2. Download and extract the application from the DairyManagement Websitea) Open a Web browser and visit the dairy management

website at: http://www.uwex.edu/ces/dairymgt/.b) Select Management Tools on the left menuc) Under the title Income Over Feed Supplement Cost,click on IOFSC.exe Filed) Save the executable file in a known location in the localmachinee) Double click in the file IOFSC.exe saved, select Runf) In the WinZip Self Extractor window, the driver "C:" willappear as default, but the user could click on the Browsebutton to save the files in any other directory. Click Unzipand a message will appear: "4 files were extractedsuccessfully." Click OK. g) Locate the folder where the files were extracted. Openfolder and open the file with name: IOFSC.xls. h) For Excel 2003 users: In the event that a SecurityWarning window appears asking to Disable or EnableMacros, select to Enable Macros. i) For Excel 2007 users: In the event that a SecurityWarning appears in the top of the spreadsheet, click on theOptions... button provided by this warning, select Enablethis content and click OK.j) Congratulations: the application is ready to be used,continue below.

2. Work with the IOFSC

2.1. General recommendationsa) Notice that the spreadsheet is divided in 10 sectionsb) In each section, cells marked in yellow are input datathat can be overwritten as desired. Drop box menus andoption button selections are also for personal choice. Cellsin blue are output data where results are displayed, butthe user is not allowed to change because they mayinclude formulae.c) The large blue button in Section 6 labeled "MaximizeIOFSC" and the large red button in Section 7 labeled“Perform a Substitution” are “action∏ buttons that willperform the analysis after data entry is completed.e) At the end of each printable page there is a gray buttonlabeled "Print this Page." By clicking on them the user willprompt the printing of the specified page. Each page isformatted to print in a letter size (8" x 11") paper of sheet.

120

IntroductionAs with any parameter to be monitored on dairyfarms, the analysis of values must be kept in contextwith the farm’s management goals and objectives.This is true for reproduction, milk production andcomponents, as well as parlor performance andmilking management. When considering milkingmanagement, first, we will assume that the basicprinciples of sound milking procedures and qualitymilk production are followed. So, the primary goalfor all dairy operations should be to have theirmilking teams apply the claws to clean, dry teats onwell-stimulated cows. Additionally, we will assumethat once the claw has been attached, it should bebalanced for efficient milk out and removed at arelatively consistent threshold for low milk flow(whether done manually or through automation).Unfortunately, there will always be some producersfor which these assumptions do not hold true. So, forthe sake of this discussion, we will disregard thisminority.

On many large dairies milking three times per day,parlor size limits the maximum number of cows thatcan be milked in a 24-hour period. Separate hospitalparlors have helped reduce the pressure on the mainparlor, but all parlors need to have sufficientdowntime for cleaning and disinfection. In thesemilking systems, monitoring parlor throughput iscritical. Stewart, et al., 1999, discussed manyessential parameters to monitor including milk/cow,milk/hour, cows/hour, milk/stall/hour, averageflow rates and average duration. They discussedways to improve parlor efficiency by decreasing uniton-time and minimizing unit idle-time. On manysmaller dairy operations, parlor size is not thelimiting bottleneck. On these farms, the milkingparlor may sit idle a significant portion of the day.Parlor throughput, while important for efficient useof labor, is not as critical. On these operations,milking teams facing a 20-30 minute delay orinterruption can still complete the milking andcleaning process in time for the next milking.Without the pressure of continuous cow flow,milk/hour, cows/hour and milk/stall/hour are notas useful. Both types of farms can utilize informationcurrently being presented in Flow Rate Reports.

This paper will not provide recommendations forproper milking procedures, the ultimate take-offsettings, or the ideal flow rates. Additionally, thispaper will not evaluate the different types of milkmeters and other measurement tools. We willassume that the data obtained from the milkingsystems outputs are relatively accurate, or at least theinformation is better than no data at all. Thequestions this paper will pursue are how to utilizedata from computerized milking managementsystems and how can statistical process controlanalysis of the data help monitor procedural drift inthe milking process. While the equipment functionand settings can have an impact on overallperformance the human factor can be the mostsignificant. The ultimate goal is improved parlorperformance, udder health and milk quality.

Definitions of TermsThe state-of-the art in milking parlor automationrelies on the data collected from automatic detachersthat may, or may not be equipped with milk metersand individual cow identification. TheBouMatic/Valley Ag Software effort initiated in 1998pushed the envelope of parlor performancemonitoring. This effort made use of existing datacollection capabilities but provided more detailedanalysis and reporting than other companies weredoing at the time. Most other companies haveincorporated many of these features in theirhardware and software. Stewart, Eicker andRapnicki, 2001, proposed standardized definitions forautomated collection of parlor performance data.Their definitions provided logical and practical termsfor use in parlor performance monitoring. Theywarned that not all manufacturers provide everymeasurement described and that some manufacturersmay be reporting the information using differentdefinitions or different precisions than noted in theirproposal. Therefore, they caution consultants whencomparing across manufacturers or even within agiven manufacturer with different software andhardware versions. Below is a summary of theirrecommendations (for their complete definitions withexplanations and caveats, the reader is encouraged toreview the entire paper).

121

Milking Management Systems: YourComputer Can Tell You About More

Than Just ReproductionRichard L. Wallace

University of IllinoisUrbana, Illinois

General Individual Cow Measurements/Calculations

Total Milk Weight: Total weight of milk produced bycow each time she entered the parlor (kg or lb).However, the smallest resolution that most meterscan measure is the size of one dump chamber(usually between 150 and 300 ml or .3 to .6 lbs), sothis apparent precision can be misleading. If re-attachments have occurred, this measure shouldreflect the total milk from all attaches during thesame parlor turn.

Duration: Length of time from claw-vacuum on toclaw-vacuum off (0.1 minute). The goal is toaccurately measure the length of time that the teatends are exposed to the milking unit. The length oftime from claw vacuum on to claw vacuum off isboth acceptable and practical for measuring milkingduration.

Parlor Stall: Unique numeric identifier for the parlorstall in which the cow was milked.

Attach Time-of-Day: Time-of-day that vacuum wasapplied to claw before attach (to the nearest 1-3seconds).

ID Time-of-Day: Time of day cow was identified (tothe nearest 1-3 seconds).

Individual Cow Milk Flow Pattern Measurements

Capture of milk weights during certain discreet timeintervals can be converted to flow rates when timestamps are applied. The time intervals listed belowhave served well for on-farm management.

Average Flow Rate: Total milk weight/total durationfor individual cow (0.1 kg/min or 01. lb/min) Thecalculation of overall average flow rate is useful inmonitoring both udder preparation and machinesettings.

Peak Flow Rate: Total milk produced in secondminute post-attach (60-120 seconds). Peak flow rateis useful in monitoring udder preparation andmachine settings.

Early Flow Rates: Flow rates in discrete intervalspost-attachment (0.1 kg/min or 01. lb/min) Flowrates early in the milking are potentially useful asmonitors of milk letdown and udder preparation.Care must be taken to avoid over interpretation ofindividual cow values due to milk meter function.The following intervals have been found to be useful:

Flow in first 15 seconds: Delay from attach to firstflow

Flow 15-30, or flow 30-60 seconds: First flow to peakflow

Flow 60-120 seconds: Peak flow

Low Flow Time: Calculated time from a lower flowthreshold (1.0 kg/min or 2.2 lb/min) to detach.Reporting time spent in low flow is potentially usefulto monitor presence of over-milking (manualsettings), poor letdown, overly “dry” take-offsettings, or other abnormalities of flow.

Unit Removals and Reattachment Data

For best parlor performance, units should remain onthe cows until milk-out is completed and thenpromptly removed. The number of cows requiring re-attachment should be minimal. Data can be collectedto monitor the following: premature unit removal,prolonged over-milking, appropriate re-attachment,and inappropriate re-attachment. Somemanufacturers provide times and productionamounts for all attaches and re-attaches. Othermanufacturers provide more limited data.

Manual Override of Automatic Take-Offs

Most manufacturers allow the user to manuallyoverride the automatic take-off settings and removethe unit prior to the level set by the sensor. In certaincases, this can be a source of abuse while in othercases it may be an appropriate human intervention. Itcan arise when workers are rushing the milking, canbe a sign of over-milking, or be a symptom ofimproperly functioning equipment. Somemanufacturers have a flag available to indicatewhether the manual override option has beeninvoked for an individual animal. The required datafor examination would include the flag itself,duration, and time of day when option was used.

Disabling Automatic Take-Offs to Allow Longer Unit On-Times

In many parlors, the most conscientious workers willdisable the automatic take-offs (set take-offs tomanual), either because of past mechanicalmalfunctions or the desire to be absolutely certainevery cow is completely milked out. To monitor ifthis is occurring, a flag can be set to indicate that theautomatic take-off was disabled.

Method of Evaluation and AnalysisWe need a consistent system to monitor changes inthe milking process over time. Most consultants canlook at daily milking system values and comparethem to “ideal” specifications, but daily variation canhave significant impact on many of these values.

122

How much deviation from the “ideal” specificationscan be tolerated? How long can these deviations betolerated?

Rasmussen, 1993, concluded that the milking unitcould be detached at a milk flow rate of 400 insteadof 200 g/min without having a negative influence onmilk yield. Machine-on time was shortened and teatcondition improved and udder health did not seemto be affected. In this study, the human use of thetechnology was not evaluated. This paper stimulatedinterest in the manipulation of ACR settings.Subsequently, Reid and Stewart, 1997, looked at theeffects on parlor performance by varying detachersettings. Parlor data was imported into Dairy Comp305 to evaluate the changes in detacher settings ontwo case farms. On both farms, flow threshold wassignificantly increased, the detach delay wasconsiderably decreased. The results of themanipulations were a reduction in average unit ontime by one minute in both parlors. Additionally, themanagers reported less stepping and kicking.

Stewart, et al., 1999, showed the effects of graduallyincreasing take-off settings on a 900-cow dairymilking 3X by monitoring milk produced per cow permilking, flow rates and unit on-time. Subsequentweekly values were compared to baseline valuesfrom an initial date. Flow rates increased (0.6lbs/min), duration decreased slightly (19.2 sec), andmilk production increased (1.6 lb/cow/milking). Theauthors assume that the additional 20 seconds percow translated to an additional 20 minutes permilking and the extra time would be used wisely toimprove milk quality. Without additional data andstatistical modeling, it is difficult to attribute theincreased milk volume to the altered take-off settings.There are too many other variables that can impactmilk production.

Stewart and Godden, etal, 2002, evaluated the effectsof switch point setting changes for automatic clusterremover units on average milking duration, milkflow and milk yield in a crossover study on fivecommercial dairy farms. Milk flow was significantlyincreased at higher switch point settings for all fiveherds. Higher automatic cluster remover switch pointsettings did not have a negative effect on milk yieldin any of the herds studied and were associated withincreased milk yield in two of the five herds. Theyconcluded that decreasing milking duration whileeither maintaining or increasing the volume of milkharvested should ultimately lead to improvedmilking efficiency and parlor performance.

Eicker and Stewart, 1998, theorized thatcomputerized parlor data could be used to monitorhow milkers used the equipment in the parlor. They

bemoaned the need for viable, efficient evaluationtechniques. They indicated that most methods ofmonitoring parlor efficiency either have been veryshort term (timing during a portion of a singlemilking) or very time consuming (viewing videotapes). Using data captured from milk meteringdevices offers methods of more routine, more rapidmonitoring of parlor efficiency.

Statistical process control (SPC) has been promoted asa way to monitor milk quality data from on-farmmeasurements (Reneau, 2000, Fuhrman, 2002). Datagenerally suitable for SPC applications are those thatare easy and practical to collect, those that arecollected on a frequent basis (daily), those that haveeconomic significance and those that, as directly as ispossible, reflects process behavior. The principles ofSPC are proven and have been used inmanufacturing businesses and the food processingindustry for over 70 years. Although the milkingprocess is a unique biological system and has morevariability SPC can be applied. If SPC is to be usefulas a part of a production system, the idea ofcontinuous improvement must be embraced.Experience has shown that application of SPCwithout commitment to the continuous improvementconcept will not be a very productive or satisfyingexperience. Once data is collected from automatedmilking systems, SPC is ideally suited to evaluate theperformance of the equipment and the milkingprocess.

Statistical Process Control is a set of several analyticaltools of which the control charts are an importantone. Control charts are helpful in signaling that a truechange has occurred in a process such as milkingparlor performance. The fundamental concept ofcontrol charts is to distinguish between inherentrandom variation and real changes in output, quality,or measured performance. Properly applied controlcharts can prevent the misinterpretation of inherentrandom variation due to “common causes” ofvariation. More importantly they provide a timelysignaling of real change due to “special cause”variation. Common causes affecting all data, arechronic, stable, and predictable within limits. Specialcauses affecting some data, are sporadic, unstable,and unpredictable. SPC methods can be used tosignal emerging problems, evaluate the positive ornegative impact of a change in a managementpractice or the implementation of a new product.

Accumulating the appropriate data for SPC chartinghas proven problematic. Most computerized milkingsystems are very effective at providing “snap shot”data of the most recent milking. Alternately, somesystems will allow the user to scroll back in time andview previous snap shots of milking performance.

123

With quite a bit of computer manipulation ofindividual reports some control charts are presentedto demonstrate the power and limitations of SPCcharting to monitor milking management systems.

Parlor performance data from the Dairy CattleResearch Unit at the University of Illinois will bepresented. There were 156 cows milked three timesper day in two different lots. Lot 1 holds 60 matureHolstein cows and Lot 3 holds 96 cows and is mostlyfirst calf Holstein heifers and Jerseys (10-12 of thetotal 96). The cows are milked in a parallel parlorwith Metatron P21 WestfaliaSurge equipment andDairyPlan 5.2 software. Data was saved fromOctober 31 through Dec 10. The cows are fed a TMRat an outdoor feed bunk with headlocks. There was asignificant snowstorm on December 1 PM andDecember 2 AM.

The data was extracted from DairyPlan pulled intoPCDART for reporting. Each day, the PCDART 819Milking Report – Flow Rate by Group (Figure 1) wassaved in a text file then merged into a spreadsheet tocreate a time series for SPC analysis. Figures 2-4demonstrate the SPC charts that can help identifyprocedural drift in the milking process. Just lookingat averages can be misleading. Averaging can blunttrue changes in the process. One AM milker haddecided that nearly all cows in Lot 1 needed to be setto manual take-off to get them completely milkedout. This practice was stopped on December 4.

ReferencesEicker, SW and SC Stewart. 1998. Computerized Parlor

Data Collection and Use: Monitoring the Cows, thePeople, and the Parlor. Proc 39th Annual Meeting of theNational Mastitis Council, pp. 98-107.

Fuhrman, TJ. 2002. Quality Milk Starts with QualityManagement. Proc 43rd Annual Meeting of the NationalMastitis Council. pp 131-139.

Rasmussen, MD. 1993. Influence of switch level ofautomatic cluster removers on milking performanceand udder health. Journal of Dairy Research 60(3):287-297.

Rasmussen, MD. 2004. Overmilking and Teat Condition.Proc 43rd Annual Meeting of the National Mastitis Council.pp 169-175.

Reid, DA and SC Stewart. 1997, The Effects on ParlorPerformance by variation in Detacher Settings. Proc38th Annual Meeting of the National Mastitis Council. pp.101-104.

Reneau, JK. 2000. Process Control: Timely Feedback forQuality Milk Production at the Farm. Proc 41st AnnualMeeting of the National Mastitis Council. pp 140-148.

Stewart, SC, SW Eicker, DA, Reid, and G Mein. 1999.Using Computerized Data to Find Time for MilkQuality, Proc 40th Annual Meeting of the NationalMastitis Council. pp 116-122.

Stewart, SC, SW Eicker and P Rapnicki. 2001. AutomatedCollection of Parlor Performance Data: InformationNeeded And Proposed Standardized Definitions. Proc ofThe 2nd International Symposium On Mastitis And MilkQuality. National Mastitis Council and AmericanAssociation Of Bovine Practitioners.

Stewart, S, S Godden, P Rapnicki, DA Reid, A Johnson, andS Eicker. 2002. Effects of automatic cluster removersettings on average milking duration, milk flow, andmilk yield. Journal of Dairy Science 85(4):818-823.

124

Figure 1. Example Flow Rate Report from PCDART with data generated by WestfaliaSurge Metatron P21 parlor.

Figure 2. Average Duration for Lot 1 cows at firstmilking. Notice long unit on-times, wide variationfrom day to day (process out of control) andmoderation after corrective action implemented onDecember 4.

Figure 3. Average Flow Rate for Lot 1 cows at firstmilking. Wide variation and low flow rate for maturehigh producing Holstein cows. Process showsimprovement after December 4.

Figure 4. Average Milk Production for Lot 1 cows atfirst milking. Manipulating take-offs by forcing cowson manual did not increase milk production and mayimpact teat health.

125

IntroductionSand is being used more frequently for bedding freestalls as producers continue to recognize theadvantages of sand. Cow health and comfort are twocommon reasons sited for bedding free stalls withsand. The common drawback with sand is handlingof sand laden manure. Smaller Kansas’ dairies with ascrape system opted to install a concrete basin thatseparated the sand and manure from the liquid(Harner, et al. 2003, 1997). These basins preventedsand barges in the holding pond or lagoon andenable the solid portion of the waste stream to behandled with conventional box spreaders. Dairiestended to switch to flush systems as herd sizeexpanded. Additionally, research and field studies atMichigan State University lead to the development ofa unit to reclaim sand from the flush stream. Moreby accident than sponsored research, gravity sandseparation systems were utilized on Californiadairies. However, the basic fundamentals of thegravity systems in California are similar to thosereported by Wedel and Bickert (1998, 1996, and 1994).Three common gravity sand separation systemscurrently include sand lanes, sand traps and sandbeaches. The cleanliness of the sand is dependent onthe quality of water being recycled from the lagoon.

Research of Reclaimed Sand QualityBernard and Bray (2002) found fresh and recycledsand had similar dry matter (fresh sand 0.6% andrecycled sand 1.2%) and organic matter contents(fresh sand 96.2 % and recycled sand 95.4 %). Theyfound minor differences in bacterial populations ofsand collected from freestalls bedded with fresh andrecycled sand. The bacteria populations in the sandsamples were different than those cultured from milksamples. It has been suggested recycled sand shouldcontain no more than 3% organic matter to minimizethe potential exposure to pathogens that causemastitis. Sand samples collected average 1.2 %organic matter (Benard and Bray 2002). Bernard et al(2003) found difference in bacteria concentrationsbetween recycled and fresh sand was dependent ontime of year. There was more variability with bacteriaconcentrations in recycled sand than with fresh sand.LeJune (Anon, 2005) reported the prevalence of E.coli:O157:H7 was 1.4 % on dairies bedding with sandand 3.1 % with sawdust. Pruna and Wenz (2003)reported bacteria counts in recycled sand bedding

remained constant the first 24 hrs after bedding(5.23E6) and peaked at 72 hrs (9.05E6). The peakbacteria counts did not correlate to increases inorganic matter or moisture content. Their conclusionwas bedding with recycled sand at 48 hr intervalsmight reduce the teat end exposure to bacteria.Bacteria concentrations above 107 cfu/g of beddingare thought to potentially cause mastitis (Bramley,1985). Hogan et al (1989) reported rates of clinicalmastitis caused by environmental organisms arerelated to bacteria concentrations in freestall bedding.Fulhage (2003) reported new sand generally exhibitsbacteria concentrations less than 105 cfu/ml andresidual levels in recycled sand of 106 cfu/ml. Thebacteria level in samples taken from stalls beddedwith new and recycled sand had similarconcentrations of 107 cfu/ml after 7 to 10 days.Bacteria concentrations were not influenced by sandtype after 7 to 10 days.

Harner and Brouk (2007) randomly sampledstockpiles of sand located on six dairies duringwinter months when there was less drying of thereclaimed sand. Samples were taken prior to beddingfreestall so the samples may not have represented thesand quality entering the freestalls. There wasnumerical and statistical difference for theconcentrations between fresh and reclaimed sand(Table 1). Bacteria concentrations on Farms C and Dexceeded 2.5 E6 cfu/ml. The remainder of the farmshad bacteria concentrations less than the 1E6 cfu/gthreshold. There were statistical differences forcoliforms and Staph species between Farms C and Dand the remainder of the farms. The fresh sandsample on Farm C had numerically higher coliformcounts than fresh sand samples taken at the otherfarms. Numerical difference in bacteriaconcentrations of other species was not observed inthe fresh sand. The data does suggest that bacteriapopulation counts may reach a level in reclaimedsand such that bedding with new sand should be apriority. It is important to monitor new and reclaimedsand quality to avoid the potential of mastitis due toelevated concentrations of bacteria. During thewinter months, longer conditioning or drying periodsmay be required when recycling sand.

126

Sand Bedding and Sand RecoveryJ.P. Harner1, M. J. Brouk and J.F. Smith

1Kansas State University: Seaton Hall 147Biological and Agricultural Engineering Department

Manhattan, KS [email protected]

Table 1 Average bacterial contamination of fresh andreclaimed sand samples from dairies

Sample Coliforms* Strep Staph Bacillus Total*Farm Type Species

------------------------------- cfu/ml ------------------------------- A Fresh 6,122a 256 628a 558 7,567a

Reclaim 3,694a 308,333 37,017a 151,417 500,461a

B Fresh 4,114a 2,597 1,414a 3,853 11,978a

Reclaim 186,917a 433,306 3,661a 87,139 711,022a

C Fresh 208a 228 8a 72 517a

Reclaim 1,650,336c 372,678 461,694a 72,778 2,557,486b

D Fresh 3,467a 792 122a 208 4,589a

Reclaim 838,889b 302,778 1,559,167b 59,778 2,760,611b

E Fresh 1,500a 25,500 24,667a 0 51,667a

Reclaim 118,722a 473,222 5,200a 4,083 601,228a

F Fresh 44a 131 89a 22 286a

Reclaim 159,342a 663,944 86,925a 25,778 935,989a

StdError 132,547 234,057 363,384 52,584 335,314*Means within the same column with unlikesuperscripts differ (P<0.01).

Kristula et al (2005) sample freestalls on commercialdairies using recycled sand and clean sand forbedding. Clean sand (CS) and recycled sand (RS) hadsimilar bacterial counts at given sampling periods.There was a significant increase in bacterial countsfrom day 0 to 1 for gram-negative bacteria, coliforms,and Streptococcus spp. in both winter and summer.However, no additional increases in counts weresignificant from day 1 to 7 during the summer orwinter. Coliform counts were lower on day 1 duringthe summer than from day 5 to 7. Klebsiella spp.counts were higher from day 3 to 7 when comparedto day 1. The number of Streptococcus spp. was highin both CS and RS during the sampling periods. Thenumber of coliform and Klebsiella spp. in both CSand RS was below the threshold thought to causemastitis during the sampling times. This study sawno difference in bacterial counts between recycledand clean sand the week following freestall bedding.

Handling Sand Laden ManureSeveral dairies are using gravity with a scrape - flushflume handling system. Therefore, guidelinesdiscussed are applicable to a flush flume and severalhave been installed in the upper Midwest. If a dairyis scraping and desiring to recover sand for reuse asbedding, then the mechanical system and the flushflume are options. There are many dairies using sandbedding and settling the sand and solids from thewaste stream in a settling basin to avoid this materialentering a storage but are currently not recyclingsand. Recycling sand requires additional water thatmust be pumped on to cropland, additional labor anddepending on size and terrain potentially asignificant investment. In some cases, purchasingsand is actually more profitable than recycling sandtherefore, the economics must be carefullyconsidered.

This paper presents thoughts on the three methodsthat are actually installed and operating on dairies.However, producers must remember the thoughtspresented are based strictly on field experience andinstallations and thus models, published papers, etcare not available. Sand quality diminishes as thepercent solids increases in the flush water. Withgravity systems it is fairly simple – DIRTER WATER –DIRTER SAND. Cleaner water is obtained by eitherremoval of solids, clean water dilution or acombination.

The main goal with gravity separation is to dissipatethe energy of the flush wave and reduce the wastestream velocity of the flush wave after it exits thetransfer system to 1 to 1.5 feet per second (fps).Reducing the wave velocity to less than 1 fps resultsin settling of the manure solids based on experienceswith municipalities. Following is a brief discussion ofsand lanes, sand traps and sand beaches with somegeneral recommendations.

Sand LanesThe early system observed in California is commonlyreferred to as a sand lane. The sand lane is a longnarrow flume designed to slow flush water velocityso sand settles out of the moving stream whileorganic solids remain suspended and exit the lane.Clean sand is removed from sand lane every 1 to 2days and stacked on a drying pad. Normally, at least30 days are allowed for drying. Any dirty sandobserved in the sand lane, is redistributed near theheadwater of the lane at a shallower depth where itrewashes when the next flush occurs. Field samplingof recovered sand indicates the organic matter is lessthan 2 percent. Organic matter may be reduced toless than 0.5 percent if the equipment operatorredistributes “dirty” sand in the lane rather thanstock piling. Water quality will also influence theorganic matter in the recycled sand.

The design criteria developed is based upon stormsewer design procedures developed formunicipalities. Engineers design the system tomaintain velocities above 2 feet per second (fps) toprevent the grit (inorganic material such as sand)from settling in the pipes (Wedel and Bickert 1998,1996, 1994). Other research has shown if the streamvelocity drops below 1 fps then organic materialsettles as well. Currently, sand lanes are design basedon trying to maintain the water velocity through thelane above 1 fps (to prevent organic matter fromsettling) but less than 2 fps to settle sand. Thetargeted flow velocity is 1.25 to 1.5 fps. Other factorsthat influence sand settling include particle densityand shape.

127

The flow velocity is controlled by lane width andslope and the sand settling is controlled by lanelength. Smaller particles require more time to settlethan larger ones. The main challenge of the sand laneis selecting length and slope. Based on the author’sexperience, lanes having a 0.25 to 0.5 percent slopeneed to be wider, than lanes with a slope of 0.2percent or less assuming the same flow volume.Current thought is to use 12 ft wide lanes at 0.25 to0.2 percent slope if the flush stream volume is lessthan 2,500 gpm or a pump system. A 16 to 20 ft laneat 1/4 percent slope is commonly used when theflush stream entering the basin is flowing at 5,000gpm or more. Recent installations have changed theslope to 0.15 % the last 1/3 of the lane length. Thegoal is slow the wave further and settle finer sandparticles. If the flush wave is entering the sand laneby a transfer pipe or narrow channel, the energy ofthe flush wave should be dissipated near the entranceto the sand lane. A concrete block or stem wall maybe used as energy adapter. The objective is todissipate the energy quickly and spread the flow overthe lane width allowing the flush wave to flow at 1 to1.5 fps for the total length of the sand lane.

The open channel of the sand lane avoids using pipeswhich can plug with sand. After the sand settles, themanure stream may enter a pipe and be transferredto a solid separator or the holding pond. Anotheradvantage of the sand lane is simplicity ofconstruction resulting in an economical recoverysystem. A broom finish is used on most sand lanes. Ifa flush flume is used, then the cost of the sand laneincreases due to the side wall heights of the lane.Normally, the top of the wall is at ground level andthe bottom is 6 inches below the bottom elevation ofthe flume.

Lack of manure storage when alleys must be scrapeddue to malfunctions of the flush system is the maindisadvantage to the sand lane. Finer particles aremore difficult to settle unless the length is increasedor the slope changed. Most use Manning’s open flowchannel equation to estimate the velocity of the watermoving through the channel. Since the flush wavemoves through the sand lane rather rapidly, dairies incolder climates have not experienced many problems.The sand will be wetter and thus the drying time willincrease. Generally, a dairy stops flushing due to iceaccumulation in cow traffic areas not iceaccumulation in the sand lane. The settled sand mayfreeze in clumps but this may be avoided byincreasing the frequency of cleaning or placing thesand on stack pad with surface drainage that allowsthe moisture in the interior of the pile to drain.

The general guidelines are:Lane width – 12 ft Lane curbs – minimum of 12 inchesLane slope – 0.25 to 0.2 percent with 0.25 the mostcommon.Lane length – generally at least 150 ft but varies –longer is better

Sand TrapsThe second generation of gravity separation systemsis referred to as sand traps. Sand traps are located 30to 40 feet past the lower end of the free stall building.This distance between the building and sand trap isnecessary to allow bedding equipment easy accessinto the alleys. At the entrance to the sand trap, waterflows over the edge and drops 2 to 4 feet. The basin isthe width of the free stall housing area (outside wallto feed line) and has a 24 foot long flat bottom. Aramp slopes up to ground level. As the flush waveexits the alley at the lower end of the building, thewave spreads and then overflows into the sand traplike a water fall. The turbulence of the water as itflows over the wall maintains the organic matter insuspension but allows the sand to settle into the trap.The water and organic matter drain away througheither one 18 inch pipe or two 12 inch pipes. Theadvantage of a sand trap is 1 to 2 weeks of sand mayaccumulate before removal. Also, the trap may beused to hold 1 to 3 days of scraped manure storagewhen the flush system doesn’t function. Thedisadvantage to the sand trap is the sand appearsdirty and wetter than sand recovered from a sandlane. Probably 60 days of drying time are neededbefore reusing as bedding. Sampling of the sandremoved from a trap indicates the organic matterranges from 1 to 3 percent. The critical designcomponent is the detention time in the basin. Thiscontrols the rate the water and organic matter drainsfrom the trap. If the detention time or discharge fromthe basin is too slow, organic matter settles out alongthe edges of the residing water. Rapid discharge or adecrease in the detention time may result ininadequate time for the sand to settle out. It appearsbased on several that have been installed, theoptimum discharge time is 2 to 3 minutes assumingan average head pressure of 1.5 ft. Normal pipe flowequations may be used to estimate the dischargetime. Sand traps handling water from a flush systemreleasing 7,500 or more gpm need a minimum of 250square inches of pipe cross-sectional area. This isequivalent to one 18 inch diameter pipe or two 12inch diameter pipes. A 12 inch pipe on each side (2pipes) is probably preferred since water maydischarge at two different locations. If small sandbarges form in the sand trap some times water istrapped when using one pipe.

128

The general guidelines for sand traps are:

Width 40 to 48 ftDepth 2 to 4 feetFlat Bottom Length 24 ft (min)Total Volume 2 times the volume of flush

water for cow and feed alleysDetention Time 2 to 3 minutes drain down based

on 12 inches of head pressure

Sand BeachThe latest type of gravity separation system is knownas the sand beach. As of the publication date of thisarticle, the author is aware of only two sand beachsystems in the United States. The best way tovisualize a sand beach is to think of a beach along theocean where the waves wash up the beach and thenrecede leaving a deposit on the beach. The sandbeach is similar to a sand lane with a 12 ft wide levelbottom that has a gradual slope to the solidseparation. The difference is the 12 inch curb oppositeof the flush wave entrance is replaced with a 50 footinclined on a 3 percent slope. It is the author’sopinion that a 4 to 6 % might be better. This is basedon the fact that the flush wave is exiting an 18 inchdiameter pipe at a high velocity actually flows up theincline for more than 50 ft (note: this was resolved byusing a sand dike at the top to control water flow).The flush water must enter the sand beachperpendicular to the basin such that water is forcedup the incline. The flush wave goes up the inclineleaving a sand deposit and then back washes organicmatter deposited on the sand as the water recedes.The receding wave exits the basin through the 12 ftlane. The existing basin uses the area between thepipe entrances for stock piling and drying sand.

The flush system utilized on the dairy with the sandbeach has a release rate of more than 8,000 gpm usingtower tanks with manual valves. The water exits thelanes and collects in a 4 foot deep sloping drop box.The water exits the box through an 18 inch drainpipe. This high release rate results in a rapid changeof velocities as the water exits the pipe. The inclinedissipates the remaining wave energy with the waveapproaching 0 fps before receding. Because the waverecedes, the organic matter that settles is re-suspended in the wave before discharging. Once thewater flows back into the 12 ft lane, the designedvelocity of the water should be maintained at 1 to 1 ?fps. Based on purchased sand volume at one dairy,the manager feels they are recovering more than 90percent of the sand with organic content less than 1percent.

Based on one known field installation; critical designcomponents appear to be:

Inclined slope 3 to 6 percent, maybe moreFlat bottom width 12 feetFlat bottom slope 0.15 to 0.25 percentIncline length 50 feet, probably 75 ft if using 3

% slope.

Other ConsiderationsSeparation of SLM from a waste stream requiresadditional water added to the system even if recycledwater is utilized. The amount of water added to thesystem is a function of solids removed and desiredsolid content in the recycled water. Based on currentmanufacturer recommendations, the recommendedsolid content in the recycled water varies from 1 and3 % for adequate sand and solid separation. Table 2shows the additional water on a per cow per daybasis that must be added to the system in order tomaintain the desired solids level in the recycledwater. This water may come from the parlor washwater, extraneous drainage such as roof or drivewayrunoff, surface rainwater, etc. The volume rangesfrom 204 gallons per day per cow (g/d/c) if there isno solid separator and the desired solids content is 1% to 8 g/d/c if 60 % of the solids are removed andthe recycled water contents 5 %. In general, a dairyshould plan an additional 50 g/d/c if a sandrecovery system is installed. Table 3 shows thevolume of manure and water entering a containmentstructure that annually must be pumped for a 500cow dairy depending on the solids content in therecycled water and separator efficiency. Sand qualitydeclines as the total solids in the flush waterincreases. Many mechanical systems use somerecycled water as well as fresh/clean sand at the finalstage of the separation process.

Table 1 The influence of separator efficiency on thegallons of clean water per cow that must be addeddaily in order to maintain a desired lagoon solidsconcentration

Desired Solids Separator EfficiencyContent in (Percent Total Solids Removed/Recycled Moisture Content of Solids)

Water 0 30/60 60/801 % 204 139 782 % 95 63 343 % 58 37 204 % 40 25 125 % 29 17 8

129

Table 2 Comparison of the quantity of materialhandled from a lagoon on a 500 cow dairy assuming140 lbs of manure per day per cow and variousseparator efficiencies and desired solids content inthe lagoon.

Desired Solids Separator EfficiencyContent in (Percent Total Solids Removed/Recycled Moisture Content of Solids)

Water 0 30/60 60/80Manure in Lagoon 3,100,000 gal 2,800,000 gal 1,900,000 gal

Material fromSeparator 0 310,000 gal 1,200,000 gal

1 % 40,500,000 gal 28,100,000 gal 16,100,00 gal2 % 20,600,000 gal 14,200,000 gal 8,000,000 gal3 % 13,900,000 gal 9,500,000 gal 5,500,000 gal4 % 10,600,000 gal 7,300,000 gal 4,000,000 gal5 % 8,600,000 gal 5,800,000 gal 3,300,000 gal

There are many gravity separation systems beingused across the United States. Random samplingindicates the organic matter ranges from ? to 3 %depending on the system and operational protocol.Most dairymen with these systems feel they arerecovery at least 80 % of the sand and many feel 90 %or more. It does appear that dairies reclaiming sandfor bedding purposes do a better job of stallgrooming and bed the stalls more frequently.

Employees understand the importance of followingprotocol and standard operating procedures. Mostoften these become routine in the milk parlor or feedcenter. There are economic risks associated whenprotocols are not followed. Many dairies agreed inprinciple to follow the protocol and operationalprocedures with the SLM system during the design,however, once operational, the commitment begins todecline. Most agree additional water is necessary toreclaim sand and are committed to this principle inthe design phase. However, an easy way to reducemanure handling cost, particularly with customapplicators, is to reduce the waste volume. Assume a500 cow dairy spends $0.01 per gallon of materialpumped, if the dairy decides to reduce daily wateradded to the system by 10 g/d/c, then the annualsavings is over $18,000 or $36 savings per cow. Thisreduction in additional water will affect the successof the SLM system.

Handling SLM requires commitment once thedecision is made to bed freestalls with sand. Longrange objectives need to be clearly defined. If a longrange objective is to flush, then the building must beconstructed on a slope which allows flushing to beadded. Another example is source of sand. One ofthe short and long range objectives should be toclearly define the source and type of sand being usedfor bedding the freestalls. All suppliers of equipmentfor the SLM system should be made aware of thesource of sand and characteristics. The system must

be designed around the source of sand. A commonmistake is, when the system is not successful, tochange the source of sand. When this happens, thedaily variable cost per cwt of milk is increasedbecause normally the sand is more expensive andhauling cost greater. Other objectives may includefrequency of bedding or grooming or desired sandrecovery level. It is not realistic to assume 100 % sandrecovery from any system. The abrasiveness of theconcrete and cow traffic on the sand particles resultsin particle size reduction. Table 3 shows the relationbetween percent sand recovery and the amountremaining after the sand has been recycled multipletimes. If 100 lbs of purchased sand is placed infreestalls and the sand separation unit has a recoveryefficiency of 95 %, then only 60 lbs remains after 10passes through the separation process for bedding.Clear objectives need to be outlined if expansion oradding of additional cows is planned. Often a fewcows are added to the herd without any expansion ofthe SLM system or increase in water usage. Whenthis occurs, the system may become out of balanceresulting in a decline in performance. Commitmentto long range planning means when a certain numberof cows are added, then additional capacity to theSLM system is operational when these cows comeinto the herd.

SummaryMany dairies are successfully handling sand ladenmanure using a variety of systems. A review of anunpublished survey from Kansas found some dairieshad been using sand for more than 25 years. Theseproducers feel there is an economic advantage withthe utilization of sand in freestall beds. Successfulhandling SLM requires more than a financialcommitment, success requires an operationalcommitment as well as system design approach.

ReferencesBernard, J.K. 2004. Bedding strategies for free stall bedding.

Proceedings 2004 Florida Dairy Production Conference.Bernard, J.K., D.R. Bray and J.W. West. 2003. Bacteria

concentrations and sand usage in free stall bedded withfresh or recycled sand. 5th International Dairy HousingConference. St. Joseph, MI: ASAE.

Bramely, A.J. 1985. The control of coliform mastitis. Pg 4-17in Proc. National Mastitis Council Annual Meeting.National Mastitis Council. Madison. WI.

Burcham, T.N., S.K. Gill and R.B. Moore. 1997. Comparisonof dairy manure separation technologies. Paper No. 97-4050. American Society for Agricultural Engineers, St.Joseph, MI.

Fulhage, C. 2003. Sand use and management as a free stallbedding material. ASAE Paper No. MC03-401. StJoseph MI:ASAE.

Harner, J.P., Mike J. Brouk, John F. Smith (2005) SandQuality in Free Stalls Paper number 054109, 2005ASAE Annual Meeting . @2005

130

Hogan, J.S., K.L. Smith, K.H. Hoblet, D.A. Todhunter, P.S.Schoenberger, W.D. Hueston, D.E. Pritchard, G.L.Bowman, L.E. Heider, B.L. Brockett and H.R. Conrad.1989. Bacteria counts in bedding materials used on ninecommercial dairies. J. Dairy Sci. 72:250-58.

Kristula, M. A., W. Rogers, J. S. Hogan and M. Sabo(2005)Comparison of Bacteria Populations in Clean andRecycled Sand used for Bedding in Dairy Facilities J.Dairy Sci. 88:4317-4325

Pruna, D.J. and J.R. Wenz. 2003. Characteristics of recycledsand bedding over time. 2003 Summer Abstract:Integrated Livestock Management. College ofVeterinary Medicine and Biomedical Sciences. ColoradoState University.

Wedel, A.W. and W.G. Bickert. 1998. Performancecharacteristics of a sand-manure separator. Proceedingsof 4th International Dairy Housing Conference. ASAE.St Joseph, MI.

Wedel, A.W. and W.G. Bickert. 1996. Separating sand fromsand-laden manure: factors affecting the process. PaperNo. 96-4016. American Society for AgriculturalEngineers, St. Joseph, MI.

Wedel, A.W. and W.G. Bickert. 1994. Handling and storagesystems for sand-laden manure from free stall barns.Proceedings of Third International Dairy HousingConference: Dairy Systems for the 21st Century.American Society for Agricultural Engineers, St. Joseph,MI.

131

How strange it is that sometimes we fail to maximizecomfort for the animals that need it most. It is not bychoice of course, but usually occurs as a failure torecognize what cows need when their health iscompromised by disease or injury. Just like it is forus humans, everything requires more effort when thebody isn’t functioning at 100%. Therefore, it is onlylogical that we make some extra effort to provideadditional creature comforts on dairies in the hospitaland foot care areas. This is the right thing to do, notonly from a performance standpoint, but also from awelfare perspective.

In the following the focus is on design of facilities forfoot care areas. However, some of these ideas wouldapply to hospital areas as well. Since it’s not only thefacilities that matter, we begin with a discussion ofcattle handling.

Basics in Cattle Handling It is important that personnel working with cattle inany capacity have a basic understanding of theirbehavior. There’s generally a very good reason whycattle don’t do as we would like in certaincircumstances. Taking a moment to look at thesesituations from the cow’s perspective often providesthe explanation and a solution. Cattle respond bestto gentle persuasion and worst to aggressive force.Patience is essential to success in cattle handling.

Proper selection of personnel. First of all, owners and/orsupervisors of personnel on dairies shouldunderstand that not all persons are “cow people”. Inother words, some people are better suited forpositions that require close or frequent contact withanimals. Furthermore, it is important to understandthat cows like people, have distinct personalities andeach is shaped by their genetic makeup and life

experiences. They also have good days and baddays. For reasons unknown to their handlers,animals may have feelings of fear or anxiousness thatmake them more difficult, if not dangerous to workwith at times. It’s during these moments thathandlers must be particularly sensitive to behavioralresponses in order to avoid possible injury tothemselves or the animal. Finally, all animals areunpredictable. One should expect that conditionswhich induce fear or anxiety are also likely topredispose to erratic or unanticipated behavioralresponses.

The expression “the fastest way to work cattle isslow” says a lot about how we should approachcattle handling. Cattle are basically very gentlecreatures. When we use what we know about theirnatural behavior and the way in which they perceivetheir environment we make cattle handling safer,more efficient and enjoyable.

Application of Proper HandlingTechniques with Lame CowsTrimmers normally charge for their services on percow basis. Therefore, the more cows they are able totrim or treat, the more money they will be able tomake. It doesn’t take a rocket scientist to see that theway to maximize profit in the trimming business is totrim and/or treat more cows. Couple the need towork quickly with an animal that is uncomfortabledue to a painful lameness disorder and the result is afrustrated trimmer and an anxious lame cow.

Cows have good memories of bad experiences. Thus,they’re not likely to race into the chute as soon as it’svacated. The method for encouraging theirmovement in the desired direction usually involvesthe use of electric prods or other goading devices.

132

Maximizing Comfort for Uncomfortable Cows1Jan Shearer, DVM, MS., 2Joe Harner, Ph.D. and 3John Smith, Ph.D.

1College of Veterinary MedicineIowa State University

Ames, Iowa [email protected]

2Department of Biological and Agricultural EngineeringKansas State University


3Department of Animal Sciences and IndustryKansas State University


Undisciplined use of these devices furthercomplicates things by re-enforcing in the cow’s mindthat the trim chute is a place where bad thingshappen to her. Indeed, the misery of getting into thechute is often worse than anything that occurs whilethey are there.

Patience is a virtue that is hard to come by in today’smodern dairy systems. It is severely complicated bypoor facility design that requires people to “break allthe rules” with respect to appropriate cattle handling.Facilities designed to utilize natural cow behaviorease cattle handling, improve efficiency and preventmistreatment of cows.

Cow Friendly Facilities for MaximizingSafety and Efficiency

Foot care working areas (See Figure 1.) have a fewbasic requirements: 1) provide sufficient space foranimals that have been isolated for foot care work, 2)provide for both safe and efficient movement of cattleto and from the foot care chute, 3) holding areasshould have drinking water and possibly hay or feedavailable, 4) soft non-slippery flooring in holdingpens, crowd pens and alleyways, 5) holding pens,crowd pens and alleyways equipped with shade, fansand sprinklers, misters, or foggers for cooling, 6)provisions for manure management in the chute areaand holding pens, etc. via flush or other system, 7)shade and fans for the trimmer and cow in the trimchute area, 8) water at the trim chute for cleaningfeet, 9) electricity with appropriate outlets for thesafe and convenient operation of power tools, 10) atrimmer’s table for equipment and supplies whileworking, 11) a storage cabinet for maintenance of footcare supplies near the trim chute, and a 12) holdingarea for cows that exit the trim chute after trimmingor treatment has been completed.

Holding Pens. Size and/or capacity of holding areasis one of the primary concerns in proper design of thefoot care pen holding area. If one assumes thatsomewhere in the range of 30 to 60 cows may betrimmed or treated per day, holding area capacityshould be large enough to accommodate 20 to 30cows (assumes 30 cows worked in the morning and30 in the afternoon). A 30-cow holding area, howeveris a large pen and may be difficult for 1 person to sortcows from alone. Therefore, large pens may besubdivided into 2 smaller pens for easier sorting ofcows. When 2 holding pens are available, at least oneshould lead to a crowd pen where cows may bedirected to the alleyway and eventually to the trimchute area. Each holding pen should have shade(with fans, sprinklers or misters as required tomanage heat stress), access to water and feed and asoft non-slippery flooring surface. Efforts to make

this area as comfortable as possible are advised sinceit is assumed that often times these animals are lameand may be required to be there for a period of timebefore being examined and treated.

Cattle leaving the trim chute may enter a holding penwhere they may be redirected back to their pen oforigin or to the hospital area for additional treatment.Provisions for this pen are the same as for thosesuggested above.

Crowd Pens. The crowd pen is designed to funnelcows from the holding pen to the alleyway whichleads to the trim chute. Crowd pens should bedesigned to hold a maximum of 3 cows. Whendesigned with straight panels or fences, one side ofthe crowd pen should remain straight, while theother approaches the alleyway at a 30 degree angle.A solid-sided sweep gate is useful and prevents cowsfrom escaping past the handler. When crowd pensare properly designed, one person can safely moveanimals to the alleyway without the need forprodding.

The Alleyway leading to the Trim Chute. Cattlegenerally move from a crowd pen to the trim chutethrough an alleyway. The alleyway to the trim chuteshould be approximately 20 feet in length which willcomfortably accommodate 2 to 3 cows. Solid-sidedalleyways have advantages but are rarely neededunless animals are unusually excitable. On the otherhand, a solid-curved alleyway prevents cattle fromseeing the chute until they are within a few feet ofentering. Since cattle tend to move from dark to lightareas, light coming through the head catch into a trimchute with solid sides is sufficient alone to encouragemost cows to enter. However, it is important to pointout that when alleyways are designed with solidsides, cows will balk if there are shadows in the alleyway due improper positioning of lights. Uniformityof lighting in the alley way is critical. Finally, properorientation of the head catch and trim chute areimportant considerations, since cattle will tend to shyaway from direct sunlight pouring through a head-catch.

Trim Chute and Trimming Station. In large herdswhere trimmers may spend as much as 6 to 8 hoursor more at the trim chute, a few “trimmer comforts”are in order. In summer conditions, the trimmingstation needs access to shade and a fan for the benefitof the trimmer as well as the cow restrained in thetrim chute. Fans should be located so that fresh air ismoved from the backside to the front-side of thetrimmer. When air is moved in the opposite directionthe trimmer is forced to breathe dust and cow hairthroughout the trimming process. Another option isto move air from side to side (i.e. from left to right).

133

Properly positioned this fan will move air across boththe cow and trimmer for improved comfort.

In winter, there should be a wind block andsupplemental heat as conditions require. Also, sincetrimmers may spend several hours standing at thetrim chute each day, a soft flooring surface(rubberized) is advised. Trimming stations also needa source of water for cleaning feet and cleansinglesions for proper examination and treatmentprocedures. Thus, the trim area should have a waterhose and nozzle as well as a floor design that willpermit drainage.

The trim area also needs a source of electricity for useof power tools and supplementary lighting in areaswhere natural light may be limited. For trimmer andcow safety’s sake, electrical connections should beground fault protected and located so that they donot readily come in contact with water (as from thewater hose). Proper lighting is essential for goodcorrective trimming work. Ineffective lighting oftenleads to corrective trimming errors and the failure todetect early lesions. Visualization of lesions at trimchutes is often times obscured by the orientation ofthe chute with the sun or light source. For example,with sunlight behind the operator, the trimmer’sshadow often obscures the view of lesions, whereaswhen the trimmer is forced to look toward sunlight(or into the direction of a light source), light is oftento dim to permit good observation of lesions.

On-farm trimmers need a secure place for theirequipment both while they are working and duringoff-hours. A 3 X 6 table is ideal for this purpose. Itprovides space for grinders, knives, blocks andadhesives, cloths for cleaning, topical medications orother treatments, sharpening devices, etc. Likewise, alockable storage cabinet provides for secure storageof equipment during off-hours.

General Location of Trimming Areas on Farms. Withrespect to location of trimming facilities, the ideallocation is one that is relatively close to the hospitalarea. In this way cows that may need treatmentbeyond trimming alone can be relocated to thehospital area and segregated according to treatmentand residue avoidance risks. For lame cows, housingwithin close proximity to milking facilities has majoradvantages. If milk is saleable a lot or pen close tothe milking parlor is best. If cows have been treatedand milk is not saleable, cows should be maintainedin lots or pens within close proximity to the hospitalmilking parlor. All of the creature comforts asdescribed above need to be included in these areas.

SummaryEvery dairy has, or will have; sick or lame cows andit should be every operation’s objective to maximizetheir comfort. This is necessary for reasons ofperformance and profit, but also because we owe tothem from a welfare perspective. Caring andcompassionate personnel who like cows andunderstand their behavior make the best caregivers.This is especially true for animals that are ill or lame.Designing facilities with the cow and hercompromised state in mind helps us better utilize thecow’s natural behavior making cattle handling safer,easier and more efficient.

References1. van Amstel, SR and JK Shearer: Manual for the

Treatment and Control of Lameness in Cattle. 2006,Blackwell Publishing, Ames, IA.

Figure 1. Above is a suggested layout for foot care area on adairy. Note that the hoof trimming table is withinclose proximity of the veterinary supply area forconvenient access to equipment or other materials asneeded (Joe Harner, KSU).

134

135

Five Steps to Creating The IdealTransition Cow Barn

Nigel B. CookUniversity of Wisconsin-Madison, School of Veterinary Medicine,

Madison, Wisconsin, [email protected]

IntroductionOver the last few years, my colleagues in the Food Animal Production Medicine group at the University ofWisconsin-Madison have used our clinical experiences troubleshooting fresh cow health problems on farms,research conducted by other groups and our own research findings, to formulate a plan for designing transitioncow barns which results in optimal health and performance. In this article, I will summarize the planning processwe have devised and used successfully to create these new facilities.

Where to start?

The planning process starts with one simple question:

‘How am I going to manage my cows at the point of calving?’

In order to limit the risk for dystocia and stillbirth, and avoid movement and social upheaval within the criticalperiod of 2-7 days before calving, there are only two possible strategies:

1. Move cows from a prefresh pen with freestalls to an individual or group calving pen at the point of calving –refereed to as ‘Just-in-time-calving’.

2. Manage socially stable group pens throughout the prefresh period and calve in the prefresh pen, which in thisscenario is a bedded pack.

Each strategy has some advantages and disadvantages laid out in the table below that should be discussed beforecontinuing with the plan.

Table 1. Comparison of two strategies for managing the cow at the point of calving

Just-in-time-calving Parameter Socially StablePrefresh/Calving Group Pen

Freestalls and individual or Type of Housing A series of group bedded packsgroup bedded pack15% less roof space required, but Space Requirement 15% more roof space required,more concrete and stall construction and Cost of Construction but less concrete and stallcosts construction costsLimited to stall bedding and bedding Bedding Costs High due to the use of multiple for individual or group calving penbedded packsNeed to check prefresh pen hourly 24/7 Need for supervision Less need for constant supervisionElevated if workers move cows too early Risk for dystocia Decreased, as cows do not have

to be moved when they are in laborGood, provided excellent bedding Disease control More difficult and may requiremanagement in the calving pen(s) separate pens for Johne’s cowsExcellent control Passive Immunity Transfer Depending on level of supervision,

opportunity for calves to suckthe wrong dam first

Each strategy has different keys for success.

136

For the Just-in-time-calving approach the keys to success are:

1. Control social structure in the prefresh pen by limiting new additions to a once per week cycle.2. Locate the calving pens close to the prefresh pen and away from heavy traffic areas3. Train maternity pen workers to identify the stages of labor and monitor what stage of labor the cow is in

when she is moved, the time that she is moved to the calving pen, the time when she calves, whether thecalving was assisted or not and when the calf received colostrum. Risk for stillbirth will increase if cows aremoved to the pen too early (before the water bag shows), the cow is continually disturbed once moved andwhen workers are too eager to assist. Target stillbirth rate for the herd is 4%.

4. Fresh dry bedding arrives with the cow and leaves with the cow to maintain hygiene in the calving pen5. Utilize a correctly designed pen, such as the one shown below in figure 1.

Figure 1. An ideal maternity pen layout, with a concrete apron against the feed bunk in the fore-ground, a beddedarea with sand and straw on top in the rear half of the pen, and a head gate in the far corner. The water trough islocated in the near right corner, away from the bedded area.

For the socially stable prefresh/calving group pen, the following is required:

1. A series of bedded pack pens with sufficient numbers to accommodate a weekly group of cows movingfrom far-dry to prefresh. Therefore, for a 21 day prefresh period, a minimum of three separate pens arerequired (figure 2).

2. No further animals are added until all of the animals in the pen calve (this is ideal, but in practice a fewstraggler cows may have to be moved between pens so that the flow can continue from week to week).

3. Each pen is sized to provide a minimum of 100 square feet of bedded area per cow at maximum fill.4. Each pen is bedded fresh daily, and the whole bed removed once the last cow calves.5. Sufficient supervision is still required to make sure that there is adequate control of colostrum feeding.

Figure 2. A series of bedded pack pre-fresh pens designed for a 1000 cow dairy. Each pen provides 2940 square feetof bedded area, with a maximum stocking rate of 30 cows per pen. A total of three pens provide capacity to copewith 140% of the weekly average calving rate, with a 3 week pen stay duration. Pens are filled in series – filling pen1 first to a maximum of 30 cows, then pen 2 and so on. Once the maximum stocking density is reached, no newcows are added. Cows may calve in the pen or in an adjacent calving pen and proceed to the post-fresh group.Once the pen is empty, it is cleaned out and re-bedded and the filling cycle repeats.

137

Once a decision has been made on a strategy for managing the cow at calving time, the planning process canproceed in 5 easy steps:

1. Size groups to accommodate the 90th percentile of the weekly calving rate2. Provide 30 inches of bunk space 21 days before and after calving3. Create a socially stable grouping structure – minimizing pen moves within the period 2-10 days before

calving4. Provide sand bedded stalls sized to accommodate the size of the cows using them5. Provide at least one stall per cow (or at least 100 square feet of bedded pack per cow)

Step 1. Sizing the pens correctly

The actual duration of stay within any given transition cow pen (which includes the far-dry, prefresh, maternity,calving, colostrum and postfresh management groups) is determined by two factors; the rate of calving and thetarget duration of stay in the pen.

Recommendations for pen sizes are typically based on the average flow of cows through the transition facility anddo not take into account farm management decisions which vary time spent in the pen. For that reason, manytransition cow pens are built that fail to accommodate the normal ebb and flow of calving rate over time. Werecommend that a facility would be best constructed to accommodate the surges in calving rate, withoutcompromise to stocking density within the pen. In essence we will be over building to some degree.

We have constructed a plan for sizing transition cow pens that allows us to accommodate cows in pens sized tocope with the normal increase in stocking density for 90% of the time. For 5 weeks a year (10% of the time), thefarm will need to modify days spent in the pen to maintain the targets for stocking rate or disease screening willneed to compensate for a lapse in prevention. The procedure is as follows:

1. Calculate the weekly rate of freshenings for the herd.

For herds that are remodeling we can graph this in programs like DC305 and file out the data into Excel. For newherds, we can estimate the number of calvings to be 104% of the rolling average number of cows in the herd, andthe weekly rate will be this number divided by 52.

For example, a 1000 cow dairy will freshen 20 cows and heifers per week on average.

2. Calculate the 90th percentile of the weekly calving rate.

138

Below (Figure 3) is an example of the weekly calving rate for a 1200 cow dairy. The dotted line is the 90th percentilethreshold (32), below which 90% of the cows calve and above which only 10% of the cows calve. The solid line isthe average (24). Note that by definition, if we build to accommodate the average, the facility will be overstockedhalf the time.

Figure 3. Calving rate by week for a 1200 cow dairy with average (solid line) and 90th percentile (dotted line)calculated.

For new facilities and for expansion herds, we need to use an estimate of the 90th percentile. Using data from 73large herds we have estimated that 140% of the average weekly calving rate is a reasonable estimate of the 90thpercentile.

For example, a 1000 cow dairy would freshen 28 cows and heifers per week for 140% of average (1.4 x 20).

3. Determine the target duration of stay in each transition cow pen.

Factors such as target dry days, time of return of heifers to the close-up or far-dry pens, days in prefresh, time inthe calving or maternity pen, and days in postfresh need to be decided. These are management decisions that willbe farm dependent.

4. Calculate the number of cows in each group.

For example, a 1000 cow dairy wishing to accommodate 28 cows and heifers per week in a postfresh pen sized toaccommodate these cows to 21 DIM would need 28/7 x 21 = 84 stalls.

We have brought these ideas together in a pen size calculator for use on farm available athttp://www.vetmed.wisc.edu/dms/fapm/fapmtools/5house/TransitionCowpenSizeCalculator.xls

The calculator is shown below for a 1000 cow dairy with a 60 day dry period and 21 days spent in the pre- andpost-fresh pens.

139

Once we know the inventory in each group that we need to build for, we can proceed with the rest of the buildingdesign.

Step 2. Provide adequate bunk space pre- and post-fresh.

Using the above requirements we know from the predicted inventory how many cows are in each pen and we cancalculate the feed bunk length of each pen knowing that we need to provide 30 inches of bunk space per cow in thepre- and post-fresh pens, and 24 inches in the far dry pen.

For example, a 1000 cow dairy would need a 21 day prefresh pen feed bunk that was 84 x 2.5 feet = 210 feet long.

To accommodate our requirement for bunk space, transition cow pens should be built with only 2 rows of stalls –either tail to tail or head to tail or head to head.

For ease of cow identification, head to tail is preferred for farms where pen workers need to check cows for signs oflabor every hour. Pens of around 30 stalls split with a 26 foot crossover with a water trough in the middle provideflexibility to cope with changing numbers of cows in each group over time.

Step 3. Minimize pen moves 2-10 days before calving

This will depend on the answer to the first question regarding whether Just-in-time-calving will be used, or aprefresh/calving pen. Different solutions have been found for different farms.

For example, the farm in Figure 4 opted for Just-in-time-calving in individual calving pens. In this strategy theprefresh pen is loaded with cows once a week in order to reduce social turmoil in the last week before calving.

Figure 4. Once weekly loading of 2-row head to tail prefresh pen in an 800 cow dairy with individual calving pensand a 3-row far dry pen.

In the farm shown in Figure 5, while they retained individual calving pens, they decided to create socially stablemature cow pens throughout the dry period with the provision of 5 x 25 cow pens. The pens are loaded at dry off,locked at 25 cows maximum and allowed to empty before refilling. Heifers are managed in a separate group pen.

140

Figure 5. Socially stable mature cow dry pens (5 x 2-row pens) with individual calving pens and a separate heiferpen.

Finally, the farm in Figure 6 created socially stable bedded packs in which prefresh cattle calve where heifers maybe grouped separate from cows.

Figure 6. Socially stable prefresh group bedded packs (6) with calving in the same pen, a hospital are and 2-rowfreestall pen for fresh cows.

141

Step 4. Sand bedded stalls sized to accommodate the cow

Sand is the optimal bedded surface for the dairy cow and provides cushion, traction and support at the time whenthe cow most needs it. Prefresh cows are the heaviest widest cows on the farm and require the largest stalls. MatureHolstein prefresh cows should have stalls at least 50 inches wide, preferably 52-54 inches. This creates problems inmixed age groups with heifers grouped in the same prefresh pen. However, I believe it is foolish to punish 2/3 ofherd in order to make sure that the stalls are small enough for heifers to stay clean. Most farms choose tocompromise with a 50 inch wide stall and remove manure from the rear of the stalls more frequently. Ideally,strategies that provide different size stalls for separate mature cow (52-54 inches) and heifer groups (48 inches) canbe utilized.

Post-fresh stalls for mature Holsteins are usually sized at 50 inches wide.

Step 5. Provide sufficient stalls and bedded pack space

For prefresh cows and maternity cows, bedded packs should provide 120 square feet per cow minimum. Becausethe bedded area should not be deeper than 35 feet, this usually results in ample bunk space and typically the areaoccupied by the pack is 15% larger than the area occupied by an equivalent freestall layout. Waterers should belocated in the bedded area, but shielded so that the cow must drink from the concrete feed alley side (Figure 7).

Figure 7. Bedded pack layout with ideal waterer location.

Prefresh cows decrease lying times by around 3h/d in the last week before calving compared to far dry cows. Atthis time, they do not need to be competing for a stall, so we recommend at least one stall per cow throughout thetransition period.

Conclusions and Economics

A 1000 cow facility built to accommodate the 90th percentile of average weekly calving rate, would require 61 morestalls than a facility built to accommodate the average. At $3,500 per stall, this equates to $213,500 or $214 per cow.

For us to convince the farm (or the banker) to build this barn, you would need to believe that a facility built toaccommodate 90% of the ebb and flow of calving rates would provide 1337 lb more milk per cow than a facility thatis overstocked 50% of the time, to pay back the extra cost in one year. Of course – the deal is even better than that,because we can pay off the barn over 5 years, making the required increase in milk only 280lb per cow.

We have the opinion that when it comes to transition cows you either pay at the beginning – to build the facilitythat encourages health and productivity, or you pay at the end, with broken cows and elevated culling rates. Thechoice is easy from our perspective and that of many Wisconsin dairy farmers that have already acted on ourrecommendations.

IntroductionThere is a tremendous and growing interest inautomatic milking systems (AMS) throughout theMidwest. Dairy producers using AMS in the UnitedStates and Canada indicated that the major reasonsthey installed robots were to allow a more flexiblework schedule and a reduction in the cost of hiredlabor (de jong et. al, 2003). Although relativelyuncommon in the United States, world wide therewere over 8000 milking units on over 2400 farms in2008 (Reinmann, 2008). We are only beginning tounderstand the key factors to making AMSsuccessful. This paper will summarize publishedresearch along with the author’s interviews of Lelyand Delaval AMS field representatives, a MinnesotaFarm Business Management instructor, several dairyproducers in Wisconsin and Minnesota with AMSand nutritionists that have worked with AMS.

OverviewIn the upper Midwest we are transforming frommanaging cows individually in tie stall barns togroup housing in free stall barns. With the use ofAMS it allows us again to manage cows individuallywith the assistance of technology. AMS have theability to collect milk production, milk conductivity,milk clarity, cow activity and even individual cowrumination data. This can be used by the herdmanager to make management decisions.

There are generally two types of AMS barn designs.These are free flow and guided flow. In a free flowsystem cows are allowed to travel anywhere in thebarn unimpeded and have free access to feeding,resting and/or milking at all times. In a guidedtraffic system, a series of one way gates “guide” cowsthrough the robot when they want access to thefeeding or resting areas. Another option with guidedtraffic is when there are pre-selection gates that sortout cows that are ready for milking and allowing theremaining cows to proceed to the feeding areawithout passing through the robot.

Just like a milking parlor, AMS have an idealthroughput per day. Below is a range of some of theparameters to expect when an AMS is operating atpeak efficiency (per robot):

• 140-190 attaches (milkings) per 24 hour period. • Average of 2.4-3.0 milkings/cow/day. However,

dairy producers are able to dictate milkings/cowfor each cow every day.

• 4000-5500 lbs of milk/AMS/day. This numbercan be widely variable depending on milkproduction per cow and other factors. Several ofthese will be discussed below.

Like any successful dairy, the entire managementsystem must be considered if maximum performanceis desired. Unlike a parlor system where cows areherded to the milking center and milked whetherthey like it or not, AMS must facilitate cows having agood milking experience every time so shevoluntarily returns to be milked again. Barn design,cow handling and manure systems must all focus onmaking it easy and a good experience for cows to bemilked by the robot.

Nutrition and Feeding ManagementOne of the most important factors in making AMSsuccessful is ration balancing/nutrition management.A commonly misunderstood concept is that cowscome to get milked when they feel pressure in theirudders. In reality, cows are enticed to visit the robotbecause of feed. Therefore, it is very important thatfeed presented in the AMS be very palatable so thatcows want to visit the robot. This presents a mind setchange for many nutritionists. In North Americatraditionally we try and feed all the nutrients througha total mixed ration (TMR). With AMS the mainbunk contains a partially mixed ration (PMR) withthe remaining nutrients being fed through the robot.A survey of 25 AMS herds in North Americaindicated that they fed an average of 65% forage inthe diet. Eleven of the 25 fed a forage percentagebetween 48-60% in the TMR (de jong et. al, 2003).

My interviews confirmed the research that indicatedthe importance of high quality feed through the AMS.All except one of the dairy producers I interviewedfed a pelleted concentrate mix through the robot.The other dairy producer used a combination of 1⁄2roasted soybeans and 1⁄2 whole shelled corn. Farmsaveraged feeding 6-7.5 lbs per cow through the AMS.All farms fed a minimum of 4 lbs/cow and two farmsfed as high as 12 lbs/cow through the robots withother farms being slightly lower. A survey showedsimilar results with 88% feeding a pelletedconcentrate (de jong et. al, 2003).

Do not feed any unpalatable ingredients such astallow, feather meal, blood meal or meat and bonemeal through the AMS. Feeds that nutritionists and

142

Robotic Milking: What Do We KnowJim Salfer

University of Minnesota Extension3400 1st Street North, St. Cloud MN 56303

[email protected]

AMS company representatives felt worked wellincluded oats, roasted beans, molasses, corn gluten,distillers grains, soybean meal, soyhulls and othersources of digestible fiber.

Table one shows the results of a case study in Ontariothat demonstrated the importance of high qualityfeed on visits to the AMS (Rodenburg, 2002). Thedairy switched from a low quality, low palatabilitypellet to a high quality, high palatability pellet. Thelow quality pellet had considerable fines andcontained ingredients such as fat. The high qualitypellet had minimal fines and contained morepalatable ingredients such as molasses. During myinterviews when I asked for keys to success, all AMSusers mentioned feed management as one major keyto success. Producers were very sensitive to qualityof feed fed through the robot. One producercommented that one batch of feed he received had asimilar formulation as previous batches, but hadincreased fines, and he immediately noticed adecrease in visits to the AMS and an increase in thenumber of cows needing to be fetched.

Table 1. Selected observations when switching from alow quality pellet to a high quality pellet1.

Item Low Quality Pellet High Quality PelletVoluntary visits/cow/day 3.40 4.04Voluntary milkings/cow/day 1.72 2.06Percent fetched 16.0 7.1Pounds milk/cow/day 54.5 55.61Rodenberg, 2002

A challenge is balancing the energy fed through theTMR with the concentrate through the AMS. Therewas a steep learning curve for both nutritionists anddairy producers I interviewed. Feeding too muchenergy and grain in the TMR caused a decrease in thevisits/cow through the robot. However, not feedinghigh enough energy in the TMR may have a limitingeffect on milk production. High energy TMR’s (>0.75Mcal/lb) may increase the number of cows that mustbe fetched for milking (Rodenburg, 2008).Nutritionists I visited with were very cognizant andfocused in maintaining diets that maintained cowrumen health. Their observation is that anything thatcaused sub-acute rumen acidosis or other digestiveupsets affected cows behavior and AMS visits.Producers and nutritionists monitored AMSvisits/cow, cows fetched and milk production as theymade changes to the feeding program.

Milk ProductionWith the increase milking frequency over 2times/day milking, there will be a slight increase inmilk yield per cow. However, because the milkinginterval is not consistent, the full benefit may not be

realized. A summary of research estimates that withaverage milking frequencies of 2.4 to 2.7 times perday, the expected milk response will be 3-5% higherthan 2x milking and 6 to 9 % lower than 3x milkingon 8 hour intervals (Rodenburg, 2008).

Cow Health and ComfortProducers emphasized the importance of maintainingcow health and comfort. Regular hoof trimming andavoiding acidosis or other digestive disorders areimportant to consistent AMS visits. Lame and evenslightly lame cows had fewer AMS visits than cowswith normal gaits (Borderas et. al, 2008). Allproducers I interviewed stressed that providingmaximum cow comfort to minimize lameness andinjuries improved visits to the robot.

Fetch RateOne of the biggest concerns producers had beforeinstalling AMS was fetching cows that do not visitthe robot. They were pleasantly surprised at theminimal time it takes to fetch non cooperative cows.All producers I interviewed indicated it took 5-15minutes per robot to fetch cows. This was usuallyaccomplished while they were in the pens scrapingstalls. Fetch rates are significantly lower with guidedtraffic compared to free flow systems on a similar diet(Bach et. al., 2008). The survey showed the fetch rateof 35 herds with a free flow system averaged 16.2% ±10.8% and in guided traffic systems averaged 8.52% ±5.9% (Rodenburg, 2008). Fetch rates are highlyvariable and can range from virtually 0% up to over40% for some herds (Rodenburg, 2008). Asmentioned earlier, feed management is the key toachieving consistently low fetch rates. Producersinterviewed with a free flow system had fetch rates of3-6 cows per robot. Late lactation, older cows orcows with health problems, especially lameness, havehigher fetch rates. A cow that typically attends theAMS voluntarily and now needs to be fetched canassist with early diagnosis of health problems(Rodenburg, 2008). One producer I interviewed hada guided flow system with a separate pen forfresh/lame cows and indicated that on most daysthere were no fetch cows. Robot owners in Canada(Rodenburg, 2008) and producers I interviewedindicated that fetch rates are lower and trafficthrough the AMS increases when humans are in thepens less. The goal is to not disturb the cows’ normalday. Fetching should be a last resort because it canteach cows bad habits (Rodenburg, 2008).

Training New Cows Immediately after moving into the barn, all cowsmust be pushed through the robots 3 times per dayfor 1-2 weeks. After the initial training, producers arepleasantly surprised how fast new cows learned touse the robot. However, it is quite variable between

143

cows. Most cows are trained by 7 days in milk.However some cows took as long as 45 days to adjustto the robot. After going through a dry period mostcows needed no training the following lactation.

Culling RateA few cows in most herds will need to be culledbecause of udder conformation or they do not adaptto the robot. However, this rate is very low with thenew generation of robots. This will typically be 0-3%of cows (Rodenburg, 2002). This is in agreement withinterviews with dairy producers in Wisconsin andMinnesota. Udder conformations that createchallenges are udders with a deep crease between thehalves along with rear teats that cross or touch. Theother challenge is cows with extreme reverse tiltwhere the rear teats are much higher than the frontteats. Very few cows must be culled because they justrefuse to adapt to the robot. Producers commentedthat they believe heifers adapted faster and were lesskicky with AMS than when milked in a parlor.

Cow Management and HandlingCows tend to become very calm and easy to handlewhen milking with AMS. All farms I interviewed useheadlocks for routine management intervention.During activities such as herd check, they fed thecows their PMR and 70-80% of the cows self locked.None of the producers indicated that catching cowsfor routine management interventions was achallenge. One producer had a sort pen with feedand water available and sorted cows to the pen forherd check. He set the computer to sort cows for 12hours before herd check and indicated that 80-90% ofthe needed cows were sorted within that time.

The key monitors that producers use when managingtheir herd were (Helgren and Reinemann, 2006):

• Deviation in milk yield• Milking interval since last successful milking• Milk conductivity

Milk QualityHelgren and Reinemann (2006) studied milk qualityof 12 AMS farms. The somatic cell count averaged268,000 cells/mL. This was not different than acohort of farms with conventional milking systems.Bacteria counts were lower than conventional farms.Somatic cell counts and bacterial counts decreased asthe amount of time a farm used AMS increased(Helgren and Reinemann, 2006).

Barn DesignMany factors must be considered in barn design.Since cows need to be coerced into milking, anythingthat makes visiting the AMS easier will improveperformance. Here are some considerations in barndesign:

• Consider a manure system that minimizes time inthe pens. Most producers install automaticscrapers or slats to eliminate having to go in thepen to scrape. However, one dairy producer Iinterviewed installed AMS in an existing scrapefree stall barn. He indicated that it took very littleextra time to scrape alleys compared to when hemilked in a parlor.

• AMS have been successful with all types of barnsincluding: sand bedded free stalls, mattressbedded free stall and loose housing.

• Minimize walking distance to the AMS. • Highly visible well lit areas around the robot are

preferred. • Providing comfortable amenities in the holding

pen near the AMS are very important. Oneproducer has extra fans to provide cooling in theholding pen for the AMS.

• Provide protection for submissive cows frommore dominant cows. This is especially truearound the AMS. Boss cows may try and controlthe AMS because of the palatable feed available.Provide protection like holding pens at theentrance of the AMS. Provide an exit lane orconsider a Y to allow cows two routes of exitfrom the AMS. Adding extra crossovers, wideralleys and more watering space will allowsubmissive cows easier access to the AMS.

• Do not move cows between pens. This requiressocial adjustment and cows will decrease visitsafter moving.

• Consider designing a barn where all robots arepositioned so the cows enter them on their left orright side. Another alternative is to have bothright and left entrance robots in the same pen.One study showed that 10% of cows had adifficult time adjusting to entering on theopposite side entry (Rodenburg, 2007).

• Consider a smaller freestall or bedded pack pennear the AMS for fresh and lame cows.

Labor vs. Capital Investment vs. LifestyleLike many decisions on dairy farms, the decision toinstall AMS needs to be based on family goals withtrade off’s between labor, capital investment andlifestyle desired. The “typical” producer adoptingAMS is a smaller dairy that does not desire to expandto several hundred cows. Many producers arelooking to expand to between 100 and 200 cows andneed or desire a new milking system. When makingthe decision on a milking system, the choice hastypically been either to build a modern labor efficientparlor and use it only a few hours each day or tobuild a smaller, lower cost, less labor efficientparlor. Some of these producers are opting to selectAMS as a method to minimize hired labor, increaselabor flexibility while expanding their herd.Robotic milking permits many farms to milk

144

120-180 cows without hiring non-family labor(Rodenburg, 2008).

As with any system, the capital investment with AMSis widely variable depending on automation selectedand facilities required. When asked about the costcomparison of AMS compared to a parlor, the realquestion is, “What system are you comparing theAMS to?” Compared to a very low cost swing parlorinstalled in an existing barn, AMS will be veryexpensive, but the cost could be similar to a newparlor in a new milking complex.

Reinemann (2009) compared the cost of milking 120cows in a double 8 parlor (D8) with 2 AMS. Heestimated that total labor and facilities costs to be$5.26 vs. $6.03 for the D8 and AMS respectively.When comparing the cost of milking 240 cows with aD8 with 4 AMS, the costs were $4.04 and $4.86respectively.

An alternative is to use a partial budget to comparelabor saved with AMS payments. Using a D8, mostproducers have one person in the pit and onescraping. With 3.5 turns per hour and milking 120cows, milking time will be 2.1 hours. With cleanuptime, total milking time per day will be 2.1 + 2.1 + 0.5x 2 times per day. This is about 9-10 hours total laborper day. Based on my interviews with producers,they estimated they spent about 0.5-1 hour per roboton milking type activities. This would equate tosaving approximately 7-9 hours of labor per day. Iflabor is valued at $12-$15/hour (including taxes,insurance and benefits) labor saved by the robots willbe between $30,000 and $49,000. The payments for 2AMS will be approximately $60,000 – 70,000depending on terms and automation. This shouldinclude the maintenance contract cost. However thisdoes not include any cost for the D8 milking systemor facilities to house the robot. This leaves $11,000 to$40,000 for annual payments on the D8 parlor andassociated facilities to have similar costs.

Using 7.5% interest amortized over 10 years the$11,000 to $40,000 will make payments onapproximately $77,000 to $280,000 worth of debt thatmight be associated with setting up the D8 facilities.

This is a wide range but conceptually this is one wayto compare the two milking systems. AMS willrequire much higher capital outlays than almostevery other milking system available. Therefore,careful financial planning with realistic expectationsof labor savings and milk production is required.

Other ideas to think about when making decisionsinclude:

• Future wage increases.

• Available family and non-family labor on thefarm.

• Ease of finding replacement help when planningextended time away such as vacations.

• Flexibility of labor.• Skills and abilities of owner/operators.• Potential future expansion – AMS barns must be

expanded in increments of 50-60 cows. Inaddition, every 60 cows will require a majorinvestment in a robot.

• Human health benefits such as carpal tunnel,shoulder and knee wear.

• Animal welfare concerns.• Immigrant labor reform.• Societal views of farm size and technology.

ConclusionAMS clearly have demonstrated they have the abilityto harvest high quality milk successfully. It has theopportunity, especially for smaller herds, to reducelabor, milk more frequently and provide flexibility ofhours of labor. Surveys of producers show AMSusers tend to be satisfied to very satisfied with theirdecision and flexibility of their time (de Jong et. al,2003). As with any system, it takes excellentmanagement for success. With AMS particularattention must be paid to nutrition management andcow health. It is important on all farms to figurewhat numbers, assumptions and concepts are realisticand helpful to use in analyzing the financial aspectsof this decision in the context of personal andbusiness needs, priorities and goals.

Acknowledgements I would like to thank the AMS specialists from Lelyand Delaval for their valuable input and materials,and Dr. Doug Reinemann, University of Wisconsinfor his ideas and assistance. I would also like tothank the many AMS users that allowed me to visittheir farms and for their valuable insight into theirsuccesses and challenges. And also the nutritioniststhat were so willing to share their ideas and keys tofeeding success with robots.

References Bach, A., M. Devand, C. Igleasias and A. Ferrer. 2008.

Forced traffic in automatic milking systems effectivelyreduces the need to get cows, but alters eating behaviorand does not improve milk yield of dairy cattle. J.Dairy Sci. 92:1272-1280.

Borderas, T.F., A. Fourmier, J. Rushen, A.M. Passille. 2008.Effect of lameness on dairy cows’ visits to automaticmilking systems. Can J of An Sci. 88:1-8.

de Jong, W., A. Finnema, D.J. Reinemann. 2003. Survey ofManagement Practices of Farms Using AutomaticMilking Systems in North America. ASAE An Int MtgTech Paper No. 033017.

Helgren, J.M., D.J. Reinemann. 2006. Survey of milk qualityon U.S. dairy farms using automatic milking systems.Transactions of the ASABE. 49(2):551-556.

145

Reinemann, D.J., 2008. Robotic Milking: Current Situation.Pages 75-80 in Proc NMC Ann mtg.

Reinemann, D.J., 2008. Personal communication.Rodenburg, J. 2002 Robotic Milkers: What, where and how

much!! Pages 1-18 in Proc Ohio Dairy ManagementConference. Dec 16-17 2002.

Rodenburg, J. 2008. Robotic milking systems: Are they theway of the future? WCDS Advances in DairyTechnology 20:35-54.

Rodenburg, J., and House, H. K. 2007. Field observations onbarn layout and design for robotic milking. Proc ofSixth International Dairy Housing Conf. ASABE PubNo 701P0507e.

146

Dairy heifer calves have traditionally been raised insmall group pens or individual stalls or hutches.They have been individually fed milk or milkreplacer until post weaning. These methods, whilevery successful, are also very labor intensive.

Computer controlled automatic calf feeding systemsare attracting increasing interest here in the Midwestbecause of the labor saving benefits. Developed inEurope due to animal welfare issues as well as laborefficiency, they are now available in the US.

Automatic calf feeders consist of a self-contained unitwhich heats the water, dispenses a programmedamount of milk replacer, mixes the milk replacer andwater in a container from which the calf can suck itout via a nipple feeding station. A singleprogrammable feeding unit with two nipple feedingstations will cost approximately $18 – 20,000. Anoptional computer and program for more feedingoptions and capabilities will cost an additional $4,000.Individual calf starter feeding stations whichdispense and monitor calf starter consumption willrun approximately $5,000 per station.

What is the payback for such a system? For this casescenario, let us assume we have a feeder withcomputer program and two nipple feeding stationsfor a total of $22,000 and that the building, office,water, electrical and site work is independent of theway we are feeding the calves, whether it is withbottles and buckets or the computer feeder. Verylikely the investment would be higher withindividual pens but less with outside hutches over afive year period.

A system of this type would provide the capability offeeding two pens of 25 calves per pen for up to 10

weeks. With automatic calf feeders, it is necessary tofeed calves by bottle for the first 7-10 days inindividual pens to get them started. Adaptation tothe nipple feeding station after ten days of age goesfairly well in most cases. It has been observed thatcalves will spend 30 – 50 minutes/day at the feedingstation.(Table 1,2,3 ) Typically calves will notconsume much if any for about six hours during thenight which also affects the number of animals perfeeding station. If we expect to move calves after 70days of age and they come in at 7-10 days of age,they will spend an average of 60 days in the feedingpen. We would start weaning at 49 days and wean byday 56 with calf starter consumption of two to threepounds per day. This would allow six groups of 25calves /pen/ year with two pens for a total of 300calves on the feeder per year.

$22,000 / 5 year depreciation = $4,400/ year

$4,400/ 300 calves per year = $14.66/ head feeder cost

Building cost is a separate item and would berequired for the feeding system. In a loose housingsystem, calves require 30 – 35 sq.ft. per head. Abuilding with a ten year life will currently run about$40-$45 per calf/year. Hutches, on the other handwith an initial cost of $350 and a five year life wouldcost $17.50 per calf per year. While labor time andcost is reduced with the automatic feeding system, itis not eliminated. Feeders need to be checked daily,be kept from freezing, pens bedded, calves checkedand the feeding system refilled with milk replacer oradditional milk supplied.

Management decisions for feeding calves bycomputer feeder are similar to feeding by bottle andbuckets. This would include the amount fed per day,the choice of milk or milk replacer as well as age atweaning. Feeding by computer feeder more easilyallows increased number of feedings per day andamount fed per day compared to feedingindividually. The number of feedings per day (4-8) aswell as the amount fed per day was investigated byDanish researcher M.B. Jensen. (Table 3) He found itwas better to reduce the number of feedings per dayrather than the amount per feeding. This feedingstrategy resulted in a greater amount of starter

147

Computer Controlled Automatic CalfFeeding Systems for Feeding Pre-weaned

Dairy CalvesJim Paulson, Dairy Extension Educator

University of Minnesota

consumed before weaning and less total time spent inthe feeding station.

Jensen also compared the amount fed per feedingand the number of feedings per day for the wholefeeding period calves were on milk. When acomputer controlled automatic feeding system isused, usually calves are fed .5 – 2.0 L per feedingover four to eight feedings per day. There is usually alag time of 1 -2 hours between feedings. Unrewardedvisits are visits by calves to the feeding nipple wherethey cannot receive any milk. A higher number ofunrewarded visits have been found in several studieswhere calves were offered many small portions,indicating a lack of satiety achieved by calves. Anincreasing number of visits also increased the totalamount of time calves occupied the milk feedingstations and increased the amount of observed cross-sucking done by calves. This problem is furtherexasperated by an increasing number of calves perfeeding station. Given a higher total daily intake ofsolids and a volume of eight liters versus four/day;calves were satisfied with five to six feedings/ day.This strategy reduced the number of unrewardedvisits per day by one half. Jensen concluded that the

appropriate number of portions fed per day maydepend on the total volume fed per calf per day.

Individual feeding of calves in a single pen or hutchallows for easier observation of calf health and milkconsumption. There is a concern of disease detectionin group raised calves. Automatic computercontrolled feeders can easily monitor milk intake ofindividual calves and provide alarm lists for calvesthat fall outside set parameters. Additionally, theautomatic feeder can monitor the number of visits,number of unrewarded and rewarded visits as wellas the rate of milk consumption. Svensson and Jensen(Table 4) found the most reliable indicator of calfhealth to be the number of unrewarded visits to thefeeding station and was more sensitive than theamount consumed per day and consumption rate.More visits tend to indicate a more active calf.

Automated calf feeders offer the opportunity to raisepre-weaned dairy calves with less manual labor thantraditional systems while still providing full growthpotential. Payback on calf feeding systems dependson number of animals to be fed per year as well aslabor costs and availability.

148

Table 1. Main effect of number of calves per feeder on variables collected via the computer-controlled milk feeder.

Calves per milk feeder12 24 F P——— Mean ± SE ———

Duration (min/24 h)All visits 36.50 ± 1.33 30.43 ± 1.34 F1, 23.8 = 10.08 =0.004Rewarded visits 22.84 ± 0.73 18.51 ± 0.71 F1, 20.9 = 17.50 <0.001Milk ingestion 18.40 ± 0.75 13.03 ± 0.79 F1, 12.7 = 23.08 <0.001After milk ingestion 3.58 ± 0.46 5.05 ± 0.69 NSUnrewarded visits 6.67 ± 1.12 7.96 ± 1.53 NSWith access, no milk 3.46 ± 0.55 2.35 ± 0.50 NSRate of ingestion (L/min) 0.301 ± 0.017 0.397 ± 0.022 F1, 13.6 = 11.70 =0.004Number (frequency per 24 h)Rewarded visits 6.45 ± 0.36 5.86 ± 0.48 NSUnrewarded visits 6.67 ± 1.12 7.96 ± 1.53 NSWith access, no milk 5.58 ± 0.95 4.31 ± 0.87 NS

Table 2. Main effect of number of calves per feeder on variables collected via video.

Calves per feeder 12 24 F P

——— Mean ±SE ———Duration (min/24 h)

Total occupation of the feeder 41.25 ± 3.38 26.70 ± 3.46 F1, 14.2 = 8.70 =0.01Occupying the feeder alone 35.61 ± 3.61 12.62 ± 4.85 F1, 10.8 = 14.42 =0.003Waiting for access to the feeder 5.70 ± 1.92 21.36 ± 4.97 F1, 12.1 = 10.10 =0.006

Number (frequency per 24 h)Attempts to access occupied feeder 1.82 ± 0.42 4.05 ± 0.78 F1, 16.5 = 7.05 =0.02Displacing calf from occupied feeder 1.40 ± 0.28 2.45 ± 0.36 F1, 18.1 = 5.39 =0.03Entering the feeder as soon as free 2.13 ± 0.41 4.05 ± 0.73 F1, 11.2 = 5.80 =0.03

149

Table 3. Main effect of number of milk portions on variables collected via the computer-controlled milk feeder.

Milk portions per 24 h 4 8 F P

——— Mean ± SE ———Duration (min/24 h)

All visits 30.27 ± 1.22 36.69 ± 1.35 F1, 167 = 12.95 <.001Rewarded visits 18.28 ± 0.66 23.10 ± 0.73 F1, 168 = 24.82 <.001Milk ingestion 15.03 ± 0.64 16.18 ± 0.66 NSAfter milk ingestion 2.86 ± 0.39 6.00 ± 0.57 F1, 161 = 35.71 <.001Unrewarded visits 7.00 ± 1.09 7.62 ± 1.12 NSWith access, no milk 2.72 ± 0.38 2.98 ± 0.42 NSRate of ingestion 0.372 ± 0.016 0.326 ± 0.016 F1, 163 = 7.80 =.06

Number (frequency per 24 h)Rewarded visits 5.02 ± 0.31 7.45 ± 0.38 F1, 164 = 85.61 <.001Unrewarded visits 7.00 ± 1.09 7.62 ± 1.12 NSWith access, no milk 4.81 ± 0.77 5.00 ± 0.81 NS

Table 4. Main effect of clinical status (healthy or diseased) on feeding behavior data retrieved from the milk feederunit in 68 preweaned dairy calves

Healthy DiseasedItem Mean SEM Mean SEM F PUnrewarded visits, frequency, 24-h 19.80 2.14 15.84 1.99 F1. 787 = 8.81 < 0.01Drinking rate, L/min 0.662 0.024 0.653 0.023 NSRewarded visits, frequency, 24-h 7.12 0.41 6.88 0.42 NSRewarded visits,1 frequency, 24-h F1. 696 = 3.52 < 0.10

Danish farm 4.25 0.67 4.40 0.71Swedish farm 9.98 0.46 9.37 0.45

1Interaction between clinical status and farm.

References:Jensen, M. B., and L. Holm. 2003. The effect of milk flow rate and milk allowance on feeding related behavior in

dairy calves fed by computer-controlled milk feeders. Appl. Anim. Behav. Sci. 82:87–100.Jensen, M. B. 2004. Computer-controlled milk feeding of dairy calves: The effects of number of calves per feeder

and number of milk portions on use of feeder and social behavior. J. Dairy Sci. 87:3428–3438.Jensen, M. B. 2006. Computer-controlled milk feeding of group-housed calves: The effect of milk allowance and

weaning type. J. Dairy Sci. 89:201–206.Svensson, C. and M.B. Jensen. 2007. Short Communication: Identification of diseased calves by use of data from

automatic milk feeders. J. Dairy Sci. 90:994-997.Ward, M. 2009. Personal communication.

Date post:	18-Jun-2020
Category:	Documents
Upload:	others
View:	14 times
Download:	11 times

Four-State Dairy Nutrition and Management Conference · Four-State Dairy Nutrition and Management...

Documents