1

2

3

4

5

6

7

8

9

10

11

12

13

Eighth Annual National Kidney Foundation Clinical Nephrology Meetings Abstracts

April 29-May 2,1999

Dietitians and New and Non-Renal Dietitians

Dietary Energy and Protein Intake and Their Relationship to Nutritional Status in Older vs. Younger Patients at Baseline in the Hemo Study J.D. Burr-owes, C. Bergen, D.B. Cockram, 5.7: Dwyel; L. Paranandi, D. Poole, Hemodialysis Study Group NIDDK

Development and Implementation of a Multidisciplinary Education Program for New Patients and Families Learning to Manage ESRD Beverly Ford, Marilyn Hilton, Amanda Thomas

TEAMS WORK: An Initial CQI Project for an Out-Patient Dialysis Center Leigh Gosnell, Charles Cooperberg, Mary Overby, Angela Carver Robin Malish, Cindy Roberts

Prognostic Importance of Enrollment Nutritional Markers in Hemodialysis (HD) Patients (PTS) at Eleven Years of Follow Up Arlene Henry, Cesar Martinez, Rajanna Sreedhara, Morrell M. Avram

A Patient Centred Approach to Motivate Fluid Compliance in Haemodialysis Patients-An Effective Use of Resources Susan Hyde

Serum Zinc in Newly Admitted HD Patients Related to Co-Morbid Illness and Selected Measures of Biochemical and Somatic Stores Mary Pat Kelly, Mary Ann Kight, Rudy Rodriquez, Vince Migliore

The Effect of Zemplar of Secondary Hyperparathyroidism in Chronic Hemodialysis Patients: A Case Study Anne S. McGhee, Stephen 0. Pastan

Anaerobic Threshold in Hemodialysis Renal Patients Paulo R. Moreira, Adriana E. Macagnan, Rodrigo D.M. Plentz, Fabrfcio E. Macagnan, Vilmar Baldissera

Nutritional Management of Low-Protein Diet Using Starchy Foods in Chronic Renal Failure Patients Sachiko Nagahama, Shozo Koshikawa, Yoshio Suzuki

Nutritional Rehabilitation in Hemodialysis M. Nicholson, G. Stephens

Clinical Indicators Associated with Poor Oral Intake of Patients with Chronic Renal Failure Alison L. Steiber

Post-Dialysis Albumin-A Better Nutrition Indicator? Tanya Wapensky Jane Tien, Susan Chung, Bryan Wong

Clinical, Nutritional, and QOL Outcomes After One Year in a Prospective Study of Dialysis Patients Reporting PICA Patricia Ward, Nancy G. Kutner

106 Jooumal ofRenal h’utrition, Vol 9, No 2 (April), 1999: pp 106-l 10

MEETING ABSTRACTS 107

1 DDX4RYENERGYANDFRVlBININB4KEANDTHEJRREIATKNSHIPTo NUIRmONALSl-ATUSlNOUXRLCERVS. youNcwPATlENISATBAsEuNE lNlT!EHEMOSILDY. Lp Burrowq, C Bergen, DB Cockram, JT Dwyer, L Paranandi, D Poole, and the Hemodialysis (HEMO) Study Group, NIDDK, NIH, Bethesda, MD

Dietary energy (DEI) and protein (DPI) intakes of the first 1398 patients randomized in the NIH-sponsored HEM0 Study were analyzed for differences between older (>=65 yrs) and younger (~65 yrs) patients. Nutritional status was assessed using serum albumin (SA) (by nephelometry), serum creatiniie (SC), total cholesterol (TC), equilibrated normalized protein catabolic rate (enPCR), appetite assessment (AA), body mass index (BMI), interdialytic weight gain (IDW), triceps, biceps, and subscapular skinfolds, and calf and mid- arm muscle circumferences. Adjustments were made using the Index of Co-existing Disease (ICED) Score (for comorbidities) and the Karnofsky Index (KI) (for functional status) to control for their possible effects on dietary intake. Two-day diet recalls estimated DE1

(kcal/kg/day) and DPI (g/kg/day), standardized using adjusted body weight.

Univariate analysis showed DPI, DEI, SA, SC, IDW, and KI were lower in older patients, and ICED score and TC were higher. BMI, eKt/V. enPCR, AA, and skinfolds were similar. After adjusting for sex, race, ethnicity, KI, ICED score, diabetic status, duration of dialysis (yrs), renal diagnosis, flux, and eKt/V, multivariate anaiysis showed an association between age (per decade) and the following nutrition variables:

Estimate SE P value Adj R-sq -.026 .OtB ,003 .04 -.918 ,198 .OOOl .09 -a5 .009 .OOtll s.5 408 .co6 NS .I2

BMI +413 .127 .ool .08 BMI increased significantly with increasing age, while DPI. DEI,

and SA declined. Energy recommendations for healthy older people are less than younger people; protein is the same whereas for dialysis patients, dietary recommendations are higher regardless of age.

Further research is needed to determine if age-specific protein and energy recommendations are needed for dialysis patients.

2 DEVELOPMENT AND IMPLEMENTATION OF A MULTIDISCIPLlNARY EDUCATION PROGRAM FOR NEW PATIENTS AND FAMILIES LEARNING TO MANAGE ESRD. Beverly Ford,-Marilyn Hilton, Amanda Thomas. Nephrology Center of Augusta, TRC, Augusta. Georgia.

The healthcare team collaboratively presents from the beginning of ESRD, an educational program designed for new patients and their families. By establishing from the beginning the knowledge and support needed for compliance, it is expected that morbidity and mortality may be reduced. Research indicates that the more a patient learns about kidney disease and its treatment, the more likely the patient will comply with the requirements, and more easily integrate these requirements into their lives. This objective was accomplished by compiling a comprehensive, easy to understand booklet entitled “Managing Your Kidney Disease”, to be provided to new patients entering the out patient dialysis setting Topics included are: Treatment Options, Who’s Who on Your Healthcare Team Coping with Kidney Disease, What Happens During Hemodialysis, Caring for Your Hemodialysis Access. What Happens During Peritoneal Dialysis, Caring For Your Peritoneal Catheter, Diet and Fluids; Potassium Sodium Protein and Phosphorus, and Helpfol Hints.

Ather initiation of dialysis, new patients and their families are invited to a New Patient Meeting. Here the nurse coordinator, social worker, dietitian and physician assistant are available to further explain the new information. Experienced dialysis patients participate in this meeting to relate their experiences and offer advice and support.

Capitalizing on the initial concerns, interest and need for information; early education collectively by the team is e.xpected to promole compliance from initiation of dialysis and throughout duration. The program is expected to motivate patients as they develop positive attitudes necessary for assmning responsibility for their care.

An open-ended survey was administered to collect data reflecting their opinions regarding the effectiveness of the education program. All pahents participating reported that the healthcare team provided information beneficial in helping them understand their disease, questions important to them were answered and that fears and concerns were addressed.

3 TEAMS WORK: An initial CQI project for an out-patient

dialysis center. Leigh Gosnell, Charles Cooperberg, Mary

Overby, Angela Carver, Robin Malish, Cindy Roberts. West

P&grew Dialysis Center, FMC-NA, Durham, NC.

A multi-disciplinary CQI team had the opportunity to improve

the labs resulting system and decrease the number of repeat labs

identified at our out-patient dialysis center. By using the Clarify,

Plan, Do the improvement, Act on the results (C-PDCA)

process, the lab drawing error rate and the physician review of

labs was improved. The CQI team members included a Medical

Director, Charge Nurse, Dietitian, Patient Care Technician, and an Office Coordinator. Using a ftshbone technique, root causes were identified and goals established. Staff education, new lab

rounding summaries, consultation with lab couriers, and

assignment of specific tasks contributed to improved resulting

and review of lab data for patients. With the success from this

project, we will continue to use the C-PDCA process for our CQI teams.

4 PROGNOSTIC IMPORTANCE OF ENROLLMENT NUTRlTlONAL

MARKERS IN HEMODIALYSIS (HD) PATIENTS (PTS) AT ELEVEN YEARS OF FOLLOW UP. Arlene Henrv, Cesar Martinez, Rajanna Sreedhara, Morrell M. Avram. The Division of Nephrology, The Long Island College Hospital, Brooklyn. New York

Nutrition has become one of the most important markers in predicting outcome in dialysis pts. We and others have shown that lower levels of

serum nutritional markers are linked to the higher risk of death in dialysis patients. We now report on the survival experience of 534 HD prs

followed up to I I years. Mean age was 59.9 i: 15.5 (SD) yrs. Forty seven per cent were male. Fifty seven per cent were African American, 16%

were Hispanic and 27% were white. Enrollment levels of nutritional markers are shown in the Table. Observed survival of HD

pts as measured by Kaplan Meier method for prealbumin and parathyroid hormone (PTH) are shown in the Figures. Pts with lower levels of enrollment prealbumin and PTH had significantly decreased survival compared to

those with higher levels of prealbumin and

00, 0 Ls .t dwp 6 8 02 r.2 dFollLp B ‘O ‘z

PTH. Survival by albumin (~3.5 vs. >3.5 g/dL) and creatinine (s IO vs.

IO-13 vs.>13 mg/dL) were also significantly higher in pts with higher enrollment values. Baseline serum nutritional markers continue to predict patient survival on HD for I I years. Improving nutritional status in pre- ESRD state may lead to improved patient survival. The routme

management of maintenance dialysis pts should be designed to prevent or correct malnutrition.

108 MEETING ABSTRACTS

5 A PATIENT CENTRED APPROACH TO MOTIVATE FLUID COMPLIANCE IN HAEMODIALYSIS PATIENTS -AN EFFECTIVE USE OF RESOURCES. Susan Hvde

Excessive interdialytic weight gain in haemodialysis patients is potentially life threatening and a waste of resources. In 1997, we audited patient knowledge of fluid allowance, fluid content in foods, medical implications and problems with fluid compliance by asking them to complete a questionnaire and found significant deficiencies.

A Unit standard was introduced of a maximum interdialytic weight gain of 4% of patient d@ body weight, and the average recorded for each patient. Patients and staff were then actively involved in discussion before individual patient strategies were planned to aid compliance. As a result, each patient received a Fluid Advice Sheet, a personalised Progress Chart and counselling about complications of fluid overload. The Unit presented all patients with a colourful 100ml. cup to aid fluid measurement at home. Large posters were displayed in the Unit, accompanied by teaching aids (designed by a patients daughter).

27 patients out of an original 3 1 are still dialysing on the Unit, and we re-audited in 1998 using the original protocol, and again recorded average interdialytic weight gains. Results demonstrate the strategies have led to an improvement in patient compliance, and a greater understanding of daily fluid allowances. There was a 55% reduction in those with excessive gains and 50% of those already within the Unit standards had further reduced their interdialytic weight gains.

We conclude that these strategies have encouraged patients to comply with their interdialytic weight gain targets and have proved an efficient use of Unit resources.

6 SERUM ZINC IN NEWLY ADMITTED HD PATIENTS RELATED TO CO-MORBID ILLNESS AND SELECTED MEASURES OF BIOCHEMICAL AND SOMATIC STORES Marv Pat Kelly, Mary Ann Kight, Rudy Rodriquez, Vince

Migliore. UC Renal Center, San Francisco General Hospital, San Francisco, CA

Serum zinc 550 mcg/dL has been demonstrated in 8/60 (13%) HD patients screened for antioxidant status (JASN, 8:66A, 1997), and 15/43 (35%) of patients screened for B-Vitamins (unpublished data) in our center. To further understand this phenomenon, serum zinc [atomic spectrophotometry] was drawn on admission to determine zinc status at the onset of HD, whether co-morbid illnesses increased risk for imbalance [student’s T test], and which biochemical, somatic status indicators correlated with zinc [Pearson]. Nutrition Physical Exam (CRN Q,i 1:9, 1987) was also performed.

Of 32 new patients accepted in 1998, 18 had a diabetic history, 9 were (+) for HCAB with 6 HIV (+). Serum zinc was low at 53.1 SD 13.3 mcg/dL. Using lab norms (R 50-150 mcg/dL), 13 patients were deficient, 17 patients were depleted (lowest normal quadrant 51- 75), leaving 2 with zinc 276. Diabetics had lower serum zinc (48.9 +

2.31 vs 58.3 + 4.14; p<O.O44), but no difference was shown in

HCAB, HIV. Zinc was higher in transfer vs new patients (65.4 + 9.49 vs 50.8 + 1.98; p<O.O21), but 2 of 5 transfers remained deficient.

Zinc correlated with ALB (r = 0.495, p<O.O04); Ferritin (r = 0.400, p<O.O43), and TransSat (r = 0.387. pcO.029). No correlations existed between zinc and somatic indicators weight, BMI. or % Usual Weight. Presence of seborrheic-like dermatitis was associated with zinc 555 in 8 of IO (80%) patients with eyebrow

involvement, 6 of 8 (75%) with scalp lesions, and all patients (5 of 5) with nasolabial fold change. Five of 21 (24%) patients with zinc 5 55 were without observable skin lesions.

Incidence of zinc deficiency on admission was 41%. Higher serum zinc in transfer patients suggested a modest, but not universal nutritional resiliency with dialysis therapy. Biochemical correlates of zinc depletion included ALB, Ferritin, and TransSat. Presence of seborrheic-like dermatitis lesions of the eyebrow, scalp, and nasolabial fold identified 76% of patients with zinc 555.

7 TED?, EFFECT OF ZEMPLAR ON SECONDARY HYPERPARATHY-

ROIDISM IN CHRONIC HEMODIALYSIS PATIENTS: A CASE

STUDY. Anne S. McGhee* and Stephen 0. Pastan+.

Gambro HealthCare Peachtree, Atlanta, GA.*

Division of Nephrology, Emory University School

of Medicine, Atlanta, GA.+

A significant number of ESRD patients have

persistent hyperphosphatemia and hypercalcemia as

well as hyperparathyroidism despite intravenous

(IV) Vit D3 and phosphate binder therapy. For

this reason, other analogues of Vit D3 were

formulated such as 19-nor-D2 Zemplar (paracalci-

trol injection or 19-nor-D2). Zemplar can

suppress PTH secretion from the parathyroid

gland as effectively as calcitriol but without

the hypercalcemic effect because it does not

cause as much absorption of calcium from the in-

testines. A Zemplar protocol and tracking device

were developed. Dosing was based on 0.4 mcg per

kg of body weight. SR, a 37-yr. old black single

male with ESRD secondary to hypertension had a

history of hypercalcemia and hyperparathyroidism

despite treatment with IV Calcitriol. He started

Zemplar therapy in Aug. 1998. Eis PTH decreased

from 1120(June 1998) to 396(Aug 1998) on 5 mcg of

Zemplar and continued to fall to 226 3 weeks la-

ter. Zemplar was held but the next month PTB re-

bounded to 848. SR was placed back on 5 mcg of

Zemplar with a resulting fall in his PTH in Ott

1998 to 448. The Nov PTH was 272 and he cootin-

ues with a maintenance dose of Zemplar of 2.5mcg

IV with each treatment. Because of possible re-

bound effect, consider maintenance dosing.

8

ANAEROBIC THRESHOLD IN HEMODIALYSIS RENAL PATIENTS Moreira, Paul0 R.; Macagnan, Adriana E.; Plentz,

Rodrigo D.M.; Macagnan, Fabrfcio E.; Bald&era, Vilmar. Department of Physical Therapy - University of Cruz Alta Y’ UNICRUZ J Brazil; Department of Physiology - University of Spo Carlos J Spo Paul0 J Brazil.

The purpose of the present study was to determine the anaerobic threshold in chronic renal patients submitted to hemodialysis, related to

ventilatory threshold, during a maximal cycloergometric test. Eleven patients of the hemodialysis program were subjects of the study (7 male and 4 female). averaging 41 a 12,4 and 26.7 a 6.3 years of age; 64 ’ 15.2 and 45.7 a 5.2 kg ofweight; 168 a 9.5 and 154 a 6.9 cm ofheight, respectively. Time of dialysis was 64.23 a 55.68 months for both genders. Some of the lab aspect factors that were also considered in the study-included blood Pa02, PaC02 and pH, obtained by artherial

gasometry during the resting period. To determine lactate threshold a gradual progressive ergometric test was performed in a bicycle

ergometer (electromagnetic) with increments of 25 Watts at each 2 minutes. Blood sample was obtained from the ear during rest; at the end of each increment and at 5 and 10 minutes of recovery. V02

measurements were performed using a Metabolic Analysis system TEEM 100. Lactate threshold was reached at average of 68% of the mean load of 100 ’ 25 watts and at 74% of rhe maximal V02 (18.52 s

5.9 mlkglmin) related to a mean of 2.6 a I .29 Mmol/L. Mean RQ (VCO2/VO2) at lactate threshold was I .07 ’ 0.11. VE at lactate threshold was 55% of the maximal value that was reached (34.3 a 8.54 limin) and FC was 74% of the maximal one (147 a 28.9 bpm). Oxygen pulse (VOZIFC) at lactate threshold was 5.91 a 1.68 mlibpm.

Determination of the lactate threshold was performed by the analysis of the inflexion point in the lactate curve, related to OBLA (onset of blood lactate accumulation), which is suggested to be fixed at 4

MmoliL, showing that patients in the study were classified at a low lactate level. By the correlation analysis between the values that were found by the ventilatory threshold method and the lactate threshold method one could see a significant correlation for the following factors: VE, VO2, FC, RQ and VOZ/FC!

MEETING ABSTRACTS 109

9 NUTRITIONAL MANAGEMENT OF LOW-PROTEIN DIET USING STARCHY FOODS IN CHRONIC RENAL FAILURE PATIENTS Sachiko Naaahama, Shozo Koshikawa, Yoshio Suzuki. Dept. of Dietetics, Sagami Women’s Univ., Kanagawa, Japan.

We examined the background of nutritional management of chronic renal failure patients on low-protein diet in which the use of starchy foods is indispensable. Two cases of chronic renal failure treated by a low-protein diet (0.5 &g/day) using starchy foods were reviewed. From the dietitian’s clinical records, the nutrition guidance records, diet records, and clinical data of the patients were analyzed, and the approaches for effective nutritional management were examined. Case I was an 80 year-old male (Cr 6.4, BUN 76.1 and Ccr 3.2) who was in an energy-deficient state at the initial nutrition guidance (energy intake <lo00 kcal, protein intake >35 g). After using starchy products effectively for 1 year, his daily intake approached that prescribed by the physician. Thereafter repeated guidance were conducted considering the patient’s eating preferences and habits, until the patient and family fully understood the low-protein diet. Four years later, the physician reported a good clinical course with Cr 6.6, BUN 33.8 and Ccr 5.3. Case 2 was a 21 year-old male college student (Cr 6.6, BUN 73.6 and Ccr 11.1). At admission, he was greatly dissatisfied with the low-protein diet. Starchy foods were gradually increased to raise the energy level, while the patient’s requests were considered. After 1 month, renal function was improved (Cr 7.7, BUN 45.1 and Ccr 12.2), and the patient was discharged. When conducting nutritional management, the dietitian should understand the patient’s situations and give detailed guidance of the usage of starchy foods.

10 NUTRITIONAL REHABILI’IADON IN HEMODIALYSIS. &

Heal& Alliance of Greater Nicholson and G Stephens. Cincinnati, Cincinnati, Ohio.

Hypoalbuminemia, which is associated with increased morbidity and mortality in hemodialysis, is fresuently present in patients who initiate dialysis iu our hospital based unit. Using performance improvement techniques, we developed a multidisciplinary treatmem program for prctein calorie mahmtrition and measured it’s effectivmmss over two years.

A chart review was couducted to identify potential causes of msltmtrition. Jnadequate protein iutake @cr ~1.2) was found iu 94%, inadequate dialysis (KVWl.2) in 29%, and a contributing medical condition in 2%. A survey of those with hypoalbuminemia indicated that 54% lacked insurauce coverage or fmancial resources to obtain nutritional supplements. Iuterventions to improve nutritiooal status included a program to increase Kfl, mttritional counseling aud provision of donated dietary supplements to those with inadequate financial resources, Effectiveness of this program was measured by determining

prevalence of hypoalbumiuemia and by serial determinatiou of albumin for a cohort of patients after initiation of dialysis. The proportion of patients with albumin exceeding 3.5 gmMl increased from 50% to 19% over 18 months (x2= 23.1, p I 0.001); the median albumin coucentratiou increased from 3.0 gru/dl at start of dialysis to 3.4 gm/dl after 3 mouths of treatment (p< 0.02).

Hypoalbuminemia can be corrected in hemodialysis patients by a multidisciplinary approach that includes provision of adequate dialysis, nutritional counseling and dietary supplements. Lack of insurance coverage for enteral supplements is an obstacle to frill program implememation.

ii Clinical Indicators Associated with Poor Oral Intake of Patients with Chronic Renal Failure. Alison L. Steiber. Dialysis Center of Lincoln, Lincoln, Nebraska

The purpose of this study was two fold: 1) to determine the incidence of patients with CRF who consume < 75% of their estimated nutritional needs and 2) to identify factors which were associated with an oral intake of < 75% of the estimated nutritional needs of patients with CRF. This was a prospective, descriptive, correlational study of data obtained from patients with a diagnosis of CRF at BryanLGH Medical Center East Campus, Lincoln, Nebraska. The sixty-six patients included in this study were patients who met the following criteria: 1) a primary or secondary underlying diagnosis of CRF and 2) were not receiving parenteral or enteral (tube feeding) nutritional support upon admission. Admission data (Age, Sex, %IBW, Wt. Loss, Type of Dialysis, GI History, BUN, Creat and Diet) were collected horn the patients medical record and two day kcaliprotein counts were conducted on consecutive patients admitted to the hospital. The kcal/protein counts were initiated within 24 hours of admission and consisted of 6 meals and all snacks the patient received. Only 15% of the patients met greater 5 75% of their estimated kcal needs and 12% met ~75% of their estimated protein needs. The mean kcabkg intake was 11 kc&kg with a range of O-27 kcaVkg and the mean protein/kg was 0.42 gm protein’kg with a range of O-10 gm protein/kg. Of the variables studied for associations with decreased nutrient intake, only emesis mildly correlated with kcal intake. Less than one-quarter of the patients in this study met 2 75% of their kcal and protein needs. The average kcal and protein intakes found in this study were far beneath current recommendations for HD, CAPD and pre-dialysis patients. However, the resultsof this study could lead to earlier and more aggressive intervention in patients with CRF who are at risk for poor oral intake.

12 POST-DIALYSIS ALBUMIN - A BETTER NUTRITION

INDICATOR? Tanva Waoensky, Jane Tien, Susan Chung, Bryan Wong Vivra Renal Care (now Gambro) Oakland, CA.

Serum albumin levels drawn post-dialysis (post-HD) may be a better indicator of nutritional status in hemodialysis (HD) patients (pts) since the confounding effect of excess fluid is removed before levels are drawn. Post-HD albumin levels are generally higher than those drawn pre-HD. However, it is not known whe- ther levels remain stable after treatment.

This prospective study was designed to determine whether post-HD albumin levels remain stable for up to 2 hours post-HD. A total of 96 patients completed the study in one of 2 groups: Grp A (n = 25) or Grp B (n = 71). All pts had albumin levels (bromcresol green method - gm/dl) and weights checked pre- and post-HD. Grp A pts stayed an extra 2 hours after treatment and had albumin and weight checked at that time. Our resub showed:

Albumin levels were significantly higher post-HD for all

groups &related strongly with weight loss: ( Grp A: r = 0.76, Grp 6: r = 0.64, Grps A+B: r = 0.67) Albumin levels remained fairly stable in Grp A pts from

ost-HD to 2 hrs post-HD. 1 Post-HD 1 2 hrs Post-HD 1 p value *

Group A 1 4.31 +/- 0.49 1 4.22 +/- 0.32 1 p = 0.09 -. .-. ..- 2 tarlea T-test, NS

Since albumin levels do not change significantly for up to 2 hrs after dialysis, we recommend drawing albumin levels post-dialysis to better identify nutritional or medical hypoalbuminemia.

110 MEETING ABSTRACTS

13 CLINICAL, NUTRITIONAL, AND QOL OUTCOMES AFTER ONE YEAR IN A PROSPECTIVE STUDY OF DIALYSIS PATIENTS REPORTING PICA. Patricia Ward. Nancy G. Kutner. Emory Univ. Sch. Of Med. Atlanta, Ga., USA.

This stidy examines tbe prevalence and outcomes of pica behaviors in an incident patient cohort measured at baseline (1996-97, n=226) and at follow-up (1997-98, n=180). Pica refers to compulsive ingestion of food substances such as ice, flour, or starch or of nonnutritive substances such as dirt or white clay. Data sources in the study include structured interviews, medical recdrd review, and 3-day diet diaries.

Current pica and/or history of pica were reported by 16.0% of the sample at baseline. At the l-year follow-up, three-fourths of these patients reported that they continued to practice pica. In addition, 10 patients who had not acknowledged the practice at baseline reported practicing pica at follow-up.

Similar to the baseline assessment, ice craving and consumption was most cmumn, reported by 76% of the patients who were practicing pica at follow-up. The remaining subjects reported otber pica (e.g. white dirt) or ice craving and consumption in combination with other pica substances. At the l-year follow-up, as at baseline, African-American women were significantly more likely to report pica than were other race/gender groups.

At follow-up, examination of clinical outcome variables indicated low average hematocrits and a drop in dietary iron based on diet diary averages of patients reporting ice pica. Most of the HD patients practicing ice pica had a usual lDWG that exceeded 2 kg or was 3.5% or more of their dry weight, increasing their risk for volume overload between dialysis treatments. The average fluid intake for hemdialysis patients with ice pica was not remarkable at baseline nor follow-up. However, for patients not reporting pica behavior at baseline but reporting ice pica at follow-up the average intake increased 16 ounces per day. Most patients practicing ice pica said that they were bothered by their fluid reshictions and the majority of patients reporting either ice or other pica said they had been bothered by cramps during tbe past 4 weeks. Helping patients manage pica behaviors is an important challenge for the renal dietitian.