7 B.S. Narasinga Rao Dietary Fibre in Indian Diets and Its Nutritional Significance TABLE 1 Components of Dietary Fibre Major class Property Chemical steriods ellulose Water insoluble Linear 1-4 3-glycan Non-celluose polysaccharides. : Pectin Water soluble Galacturonic aCid. neutral sugars Hemicellulose Water Insoluble Xylose, arabinose. galactose. mannose Gums Water soluble Xylose. arabinose, rhamnose Mucilages Water soluble Galactose. galalumeric acid, rhamnose Lignin Water Insoluble Polymer of hydroxy phenyl propone derivatives. Dietary fibre is defined as plant cell components, present as part of diet, and resistant to digestive secretions of the gastrointestinal tract. Thus they are considered as "unavailable carbo- hydrates". Dietary fibre, however, is not a single entity, but consists of a wide range of complex polysaccha- rides. Dietary fibre in any food is a mixture of cellulose, lignin and non cellulosic polysaccharides namely hemicellulose, plant gums, pectins, and mucin (Table 1). Dietary fibre is usually determined by fractionation procedures and the one proposed by Southgate is now widely used. The estimate of amount of dietary fibre in a food will depend upon the analytical method used for the estimation and the values so obtained need not necessarily be considered always pre- cise. Dietary fibre estimation is still considered to be only semi- quantitative. "Crude fibre", which is reported in food composition tables now in gen- eral use, represents material left after treating food with hot acid and hot alkali and perhaps corresponds to a part of cellulose and lignin compo- nent of dietary fibre and does not include all other components of die- tary fibre as defined earlier. The total dietary fibre content in foods as deter- mined by modern methods is five to 20 times higher than the earlier reported values for the "crude fibre" content. Chemically dietary fibre is polysaccharide whose basic units are neutral sugars such as glucose, man- nose, xylose, arabinose and their derivatives or galacturonic acid. Lignin is a complex material composed of phenolic derivatives (Table 1). Plant foods contain different proportions of these dietary fibre components which are now determined by the standard fractionation procedure proposed by Southgate. Limited data derived from analysis of Indian foods for dietary fibre employing Southgate's method have been reported (Table 2). Dietary fibre component of com- mon foods consumed in India vary from 2 to 25 g per 100 g. Cereals and pulses are rich sources of dietary fibre, and contain 8-20g fibre per 100 g. Dietary fibre in composite diets is contributed by cereals, pulses, vege- tables, fruits, nuts and oilseeds. In Indian diets, however, cereals and mil- lets are more important sources of dietary fibre than vegetables and fruits, etc. Dietary fibre content of some diets, consumed in India based on different cereals is given in Table 3. Nearly 90 percent of dietary fibre in Indian diets is contributed by cereals, not only due to fairly high fibre con- tent of unrefined cereals/millets, but also due to a high consumption of the unrefined cereals/millets in Indian diets. An Indian adult may thus con- sume 50-120 g dietary fibre/day through his habitual diet depending upon the type and the quantity of cereal/millets consumed (Tables 3 and 4). Diets of pre-school children may contain 20-25 g of dietary fibre/ day (Table 5). The desirable level of daily dietary fibre intake by an adult is generally believed to be around 40 g. It would appear that on this basis. die- tary fibre content of habitual Indian diets which ranges 55-120 g is on the
Transcript
7
B.S. Narasinga Rao
Dietary Fibre in Indian Diets and ItsNutritional Significance
TABLE 1ComponentsofDietaryFibre
Major
classProperty Chemicalsteriods
Cellulose
Water insolubleLinear 1-4 3-glycan
Non-celluosepolysaccharides. :Pectin
Water solubleGalacturonic aCid. neutral sugars
Hemicellulose
Water InsolubleXylose, arabinose. galactose. mannose
Gums
Water solubleXylose. arabinose, rhamnose
Mucilages
Water solubleGalactose. galalumeric acid, rhamnose
Lignin
Water InsolublePolymer of hydroxy phenylpropone derivatives.
Dietary fibre is defined as plant cellcomponents, present as part of diet,and resistant to digestive secretionsof the gastrointestinal tract. Thus theyare considered as "unavailable carbo
hydrates". Dietary fibre, however, isnot a single entity, but consists of awide range of complex polysaccharides. Dietary fibre in any food is amixture of cellulose, lignin and noncellulosic polysaccharides namelyhemicellulose, plant gums, pectins,and mucin (Table 1). Dietary fibre isusually determined by fractionationprocedures and the one proposed bySouthgate is now widely used. Theestimate of amount of dietary fibre in afood will depend upon the analyticalmethod used for the estimation andthe values so obtained need not
necessarily be considered always precise. Dietary fibre estimation is stillconsidered to be only semiquantitative.
"Crude fibre", which is reported infood composition tables now in general use, represents material left aftertreating food with hot acid and hotalkali and perhaps corresponds to apart of cellulose and lignin component of dietary fibre and does notinclude all other components of dietary fibre as defined earlier. The totaldietary fibre content in foods as determined by modern methods is five to20 times higher than the earlierreported values for the "crude fibre"content. Chemically dietary fibre ispolysaccharide whose basic units are
neutral sugars such as glucose, mannose, xylose, arabinose and theirderivatives or galacturonic acid. Ligninis a complex material composed ofphenolic derivatives (Table 1). Plantfoods contain different proportions ofthese dietary fibre components whichare now determined by the standardfractionation procedure proposed bySouthgate. Limited data derived fromanalysis of Indian foods for dietaryfibre employing Southgate's methodhave been reported (Table 2).
Dietary fibre component of common foods consumed in India varyfrom 2 to 25 g per 100 g. Cereals andpulses are rich sources of dietary fibre,and contain 8-20g fibre per 100 g.Dietary fibre in composite diets is
contributed by cereals, pulses, vegetables, fruits, nuts and oilseeds. InIndian diets, however, cereals and millets are more important sources ofdietary fibre than vegetables andfruits, etc. Dietary fibre content ofsome diets, consumed in India based
on different cereals is given in Table3. Nearly 90 percent of dietary fibre inIndian diets is contributed by cereals,not only due to fairly high fibre content of unrefined cereals/millets, butalso due to a high consumption of theunrefined cereals/millets in Indian
diets. An Indian adult may thus consume 50-120 g dietary fibre/daythrough his habitual diet dependingupon the type and the quantity ofcereal/millets consumed (Tables 3and 4). Diets of pre-school childrenmay contain 20-25 g of dietary fibre/day (Table 5). The desirable level ofdaily dietary fibre intake by an adult isgenerally believed to be around 40 g.It would appear that on this basis. dietary fibre content of habitual Indiandiets which ranges 55-120 g is on the
TABLE 2DietaryFibre Contentof Some CommonIndianFoods
a:taken from Nutritive Value of Indian Foods: b: from Kamath and Belavadi, J.Sc.
Food Agr. 31, 191 (1980).
TABLE 3Dietary Fibre Contentof Average Rural IndianDiet Based on
DifferentCereals
Cereal/Millet
EnergyCrudefibreDietaryOverestimation
(Kcalld)(g/d)fibreof" energy
(g/d)
intake(%)
Rice
24273.352.0 8.0Wheat
2433881070 16.2
Sorghum
244911.089.212.8
Bajra
25168.8122.3 18.0
Ragi
233322.0113.515.7
Daily intake of cereal: 552 g.
a: By considering dietary fibre as a part of carbohydrate and computed as follows:Dietary fibre - crude fibre
x 4 x 100
Energy intakehigher side. Hence a reappraisal ofthe beneficial and undesirable effects
of such high levels of dietary fibre inIndian diets becomes necessary.
Beneficial Effects ofDietary Fibre
Health benefits of dietary fibre arebeing increasingly recognised. Someof the diseases like colon cancer, cardiovascular diseases prevalent indeveloped countries are attributed tolow fibre content of their diets.
Dietary fibre has the tendency toabsorb water and to act as "bulkingagent". It facilitates faster transit offoods in the gastrointestinal tract andreduces the retention time of faecesin the colon. Some of the well established functions of fibre are
indicated in Table 6. Dietary fibrecould prevent colon cancer and otherbowel disorders by decreasing retention time of faeces in the colon. It
could bind bile salts and help inincreasing the loss of cholesterol andact as a hypocholesterolemic agentand therefore useful in dietary management of cardiovascular diseases.
Similarly, some of the dietary components, particularly gums, tend to slowdown glucose absorption and areuseful in management of certaintypes of diabetes. Dietary fibre maybind xenobiotics and toxins and
reduce toxicity of food borne toxins.Because of these favourable func
tions of fibre in the diet, high fibrediet is considered to be beneficial for
maintaining good health. Some of
the cereal based Indian diets contain
80-120 g dietary fibre, mostly (90percent) derived from cereals. Therelative effectiveness of such highfibre diets (wherein fibre is mostlyderived from cereals) in conferringthe above health benefits on Indians
in comparison with diets with highfibre content wherein fibre is derived
from fruits and vegetables, needsassessment.
TABLE 4DietaryFibreContent ofRegionalDiets
Regional
dietsEnergyCrudeDietaryOverestimation
(Kcal/d)
fibrefibrebofa energyintake
(g/d)
(g/d)(%)
Andhra Pradesh:Very low income
25584.457.7 8.3Low income
25924.753.0 7.5
West Bengal:Very low income
26214.556.4 7.9
Low income
26494.356.9 7.9
Uttar Pradesh:Very low income
26507.248.0 62
Low income26196.667.4 9.3
Maharashtra:Very low income
25959.181.6 11.2
Low income
25547.5686 9.5
a:
By considering dietary fibre as a part of carbohydrate by difference and computed as
given in Table 3.b:Actually determined in cooked diet.(Nageswara Rao and Narasinga Rao,Nutr. Met.
24,244, 1980)Possible Undesirable Effects
Nutrient bioavailability: Notwithstanding the health benefits ofdietary fibre discussed above, fearshave been expressed that high fibrediet may reduce nutrient bioavailability. This may become critical in dietshigh in fibre, but poor in nutrients.Diets of poor Indians in rural areas arerather high in fibre (80-120 g/day) butmarginal or even deficient in severalnutrients. Although the effects of dietary fibre on various nutrients like"available" carbohydrates, fats, proteins and minerals have been
studied, evidence ot significant deleterious effect has been seen only incase of divalent metals like Ca, Mg, Znand iron.
The reported effects of dietaryfibre on the bioavailability of proteinsand fats are too small to be nutrition
ally significant. However, in case of
TABLE 5Dietary
FibreContent of DietsofPre-schoolChildren ofDifferentIncomeGroups
Rice
WheatSorghumRagiSocio-
Energy CrudeDietary Energy CrudeDietary Energy CrudeDietaryEnergyCrudeDietaryeconomic
10110.912.010121.921.610152.318.79944.323.0Middle income
7921.013.07932.123.67763.220.47745.425.1Low income
7010.912.77032.326.07072.822.067£f5.628.0Rural income
6100.812.26122.124.56152.620.85895.126.3
Health consequence
TABLE 6
Some Properties of Dietary Fibre and TheirHealth Consequences
Function
Water absorbing and
bulking property
Increased transit time of
food in the gut
Bile acid and steroid
binding
Retardation of carbohydrateabsorption and impairedglucose tolerance
Binding of toxins
Binding of divalent cations
Energy diluent to formulatelow calorie diets
Reduced risk of inflammatorybowel disease
Hypocholesterolemic agent and reducingthe risk of cardiovascular diseases
Management of certain typeof diabetes
As a detoxifying agent
Reduced bioavailability of Ca,Mg, Zn, Fe
carbohydrates, although the net availability is not affected, dietary fibre,particularly the soluble gel forms (pectins and gums) have been shown to'modulate absorption of available dietary carbohydrates. Although themechanism of this effect is unclear at
present, various mechanisms likeincreased viscosity, delayed stomachemptying time and osmotic effect ofdiets containing soluble fibre havebeen proposed. The effect of dietaryfibre in delaying carbohydrate absorption is currently exploited in thedietary management of diabetes forpreventing excessive rise in bloodglucose.
The effect of fibre on divalent metal
absorption is considered nutritionallysignificant. These polysaccharideswith reactive groups like hydroxyl andcarboxyl can bind divalent cations.Besides, phytates, oxalates, tannins
which are closely associated with thefibre can also bind minerals and
reduce their bioavailability. This effecton mineral bioavailability assumesimportance in high fibre diets consumed in India. The effect on mineral
bioavailability depends on the fibretype and the minerals; all fibre typesmay not affect all minerals to the sameextent. Minerals, so affected are Ca,Zn, Fe, Mg. Reduced bioavailability ofminerals particularly iron observed incereal-based Indian diets may bepartly due to high fibre and partly dueto high phytate and tannins present insuch diets. The relative contribution
of fibre, phytate and tannins in reducing bioavailability of minerals fromcereals based Indian diets needs tobe assessed. As far as vitamins are
concerned, nothing can be predictedabout the behaviour of fibre on vita
min bioavailability from the currentlyavailable data. However, as with othermacro-nutrients, available informationindicates that the effect of fibre, onvitamin availability, if any, need notcause any concern. On the otherhand, fibre due to its undergoingcolonic microbial fermentation maycontribute some vitamins to the host.
There is, however, a need to studysystematically to what extent highfibre present in Indian diets contributes to low absorption of iron,calcium, etc., with well-designed studies in humans. It is also to be seen
whether dietary fibre at the levelpresent does modify vitamin availability and significantly affect alreadyexisting deficiencies of vitaminsamong our population.
Energy availability from highfibre diets: There has been some
concern that high levels of dietaryfibre may compromise bioavailability ofenergy from such diets. This concernstems from the consideration that die
tary fibre, by definition, is"indigestible" and hence its energycontent may not be utilisable by thebody. Current estimates of energycontent of foods assumes that onlythe "crude fibre" is indigestible and, inthe present conventional methods offood analysis, the rest of the fibre isincluded as a part of dietary carbohydrate which is computed bydifference, i.e. by deducting the protein, fat, moisture, crude fibre and ash
content per 100 g from 100. By thisprocedure, dietary fibre excluding thecrude fibre is assumed to yield 4 kcal/g. We know now that dietary fibre content of food is several-fold higher than"crude fibre" and the energy from therest of the dietary fibre apart fromcrude fibre may not be fully available.Thus current estimates of energycontent of foods and diets could beoverestimates and there is, therefore,
a need to reevaluate the energy content of foods after taking into accounttheir total (and not just "crude fibre")dietary fibre content.
High fibre diet can compromiseavailability of energy from diets in twoways: dietary fibre has the property ofswelling on water absorption andincreasing the bulk of the diet anddecreasing its energy density.
With a given capacity of stomach,the amount of food that can be eaten,
and the energy intake therefrom willbe correspondingly low on a bulkyhigh fibre low-calorie-density diet.This is particularly so with young children who need relatively morecalories per kg body wt (100 kcal/kg)than adults (40 kcallkg). Young children will have to eat two or three times
as much of the bulky diet as an adultper kg body weight to meet their calorie needs. The principle of dilutingenergy density of diet with inclusionof fibre is in fact used to formulate low
calorie diets for the control of obesity.Another way calorie intake on a
high fibre diet may be low is due tounavailability of energy from dietaryfibre. This may be significant if dietaryfibre content is in the range of 80-120g/day as in habitual Indian diets,based on different cereals and millets.As discussed above, most of the dietary fibre is taken as availablecarbohydrate in computing energycontent of foods given in our foodcomposition table. On this basis, current computation of energy intake onhigh fibre diet would be an overestimate. Dietary fibre, however, bydefinition is unavailable carbohydratenot subject to enzyme hydrolysis andabsorption in the small intestine.Then we would have overestimated
energy intake on cereal based dietsby eight and 12 percent (i.e. 200-300kcal/day/adult). If it were really so, itwould hav& far reaching consequences on present estimates of
energy adequacy in our populationand also other derived parametersbased on energy intakes, namely proportion of population below povertyline, etc. Let us examine this issue inthe light of available data.
Microbial Breakdown of
Dietary Fibre
Although dietary fibre is conceptually indigestible and unavailable, itspossible breakdown by gastrointestinal microflora has been investigatedfor more than half a century. It iswidely known that plant cell-wall materials which constitute dietary fibre, arebroken down by symbiotic microorganisms in rumen from which theyderive their energy. There is also considerable evidence to indicate that
such complex carbohydrates, as cellulose, hemicellulose and noncellulosic polysaccharides are brokendown by microorganisms of the intestinal flora, particularly in the colon ofnon-ruminants like rat, rabbit and man.This is based on several studies
including in vitro studies, balancestudies, where faecal loss of caloriesand fibre components are measuredin faeces of rat, man, rabbit, etc.Balance studies with rice and sor
ghum-based diets in rats by Kamathand Belavadi has indicated that only20 and 37 percent of fibre from thesetwo diets are excreted in faeces
respectively. It is seen that 70-80 percent of the fibre is broken down. It is
known that a substantial part of dietaryfibre is broken down by the microfloraof the lower gut releasing lower fattyacids like acetic propionic, butyricacids, C02 hydrogen and methane.These fatty acids can be absorbedand utilised as energy sources. Thelatter has also been demonstrated byfeeding fibre to rats on energyrestricted diets. Bacteria recoveredfrom colon, faeces of both men andanimals has been shown to possesshydrolysing activity against a variety ofpolysaccharides. The dietary fibrecomponents thus broken downinclude pectins, cellulose, hemicellulose, some gums, non-cellulosicglucans, mucopolysaccharides andmucin glycoproteins.
The extent of digestibility of dietaryfibre would vary from one fibre component to the other and would also
M. Viswanathan, V. Mohan, A. Ramachandran
Diabetes In Indians
TABLE 1DiabetesPrevalenceinMigrantIndians
Country
Author & YearAge GroupPrevalence(%)
Singapore
CheahJS& >156.1Tan BY, 1980 South Africa
Jackson WPU 1978>106.0Trinidad
Poon-KingT, 1968All ages2.4F~i
Cassidy, 1967~205.7
Source: Taylor R, Zimmet P. In: Diabetes in Epidemiological Perspective.
Mann JI, Pyorala K, Teuscher A, eds.,Churchill Livingstone, Edinburgh, 1983, p.63.
depend on the profile of the gutmicroflora in the individual. The extent
of the digestibility of dietary fibre alsowould vary from one food source tothe other, apparently because of thechemical composition of the fibretype. Fibre from vegetables and fruitsare reported to be digestible to theextent of 50 to 80 percent. From allavailable information it would appearthat at least 50 percent of the dietaryfibre present in diets may be brokendown by colonic microorganismsreleasing acetic, propionic and butyricacids which can be absorbed and
used as sources of energy. In thelight of available data, it would berather inadvisable to conclude that
the whole of dietary fibre in Indiandiets is unavailable. There is an
urgent need to study systematicallythe colonic breakdown of dietary fibrepresent in Indian diets which arederived mostly (90 percent) fromcereals and millets.' Such studies
must be carried out not only in animals, but in subjects habituated tothese diets since it is reported thatindividuals accustomed to eating highfibre diets are more efficient in break
ing down and utilising these fibrecomponents than those who are notused to such diets. For the time beingthe question may be kept open andfor practical purposes, about 50 percent of these fibres may be assumedto be available for providing energy. Ifthis were so, the actual overestimation in present computations ofenergy intake by our population willbe in the range of only four to six i.e.120-150 kcal/day/C.U.
There is a clear need for a reappraisal of energy intakes of ourpopulation subsisting on high fibrecereal/millet diets based on accurate
information on the availability ofenergy from the dietary fibre presentin these diets.
The author is former director of the National
Institute of Nutrition, Hyderabad.
We are grateful to UNICEF for amatching grant towards the cost
of this publication.
In an earlier issue of this Bulletin
wherein dietary guidelines for affluentIndians were considered (NFl Bull., 3,July 1988), reference had been madeto the growing dimensions of theproblem of diabetes in India withincreasing affluence. In this paper wepresent some of our observations ondiabetes in Indians, based on ourstudies at the Diabetes ResearchCentre at Madras.
During the last two decades, it hasbecome increasingly apparent thatdiabetes as seen among Indianspresents some significant featureswhich are somewhat different fromthose associated with the disease in
western countries (Mohan,Ramachandran and Viswanathan,Tropical Diabetes, Diabetes Annuals,Elsevier Science Publishers,Amsterdam, 1985,1986 and 1988).These differences include (1) a lowerfrequency of juvenile onset of insulindependent diabetes mellitus (100M),(2) a high prevalence of non-insulindependent diabetes mellitus (NIDDM)which also presents at a younger agein Indians and (3) the occurrence ofspecial forms of malnutrition relateddiabetes mellitus (MRDM).
The majority of diabetics in Indiabelong to the non-insulin dependentdiabetes mellitus (NIDDM) variety,which comprises over 95 percent ofall diabetics in our country. This articleis concerned only with the NIDDMform of diabetes.
We shall first present data regarding prevalence of diabetes in Indiansand then highlight the possible factors involved in its causation.
Prevalence of Diabetes
The Indian Council of Medical
Research (ICMR) multi-centric studyon the epidemiology of diabetesbased on six centres in India
(Epidemiology of Diabetes inDeveloping Countries, Ahuja, MMS.ed., New Delhi, Interprint, 1979)showed that the prevalence of diabetes in urban areas was higher (2.5percent) as compared to the ruralareas (1.5 percent). This pointed to,but did not conclusively prove, therole of environmental factors in thecausation of diabetes. Urbanisation is
generally associated with decreasedphysical activity and a change to morerefined foodstuffs, not to mentionovernutrition and obesity. However,all urban societies cannot necessarilybe assumed to be affluent. The general impression gained from theabove ICMR study was that the overall prevalence of diabetes amongIndians in India was not very high.
In contrast, the prevalence of diabetes among migrant Indians hasbeen reported to be high (Table 1).
It may be argued that the migrantIndians investigated in the abovestudies were generally more affluentthan the subjects of the ICMR study.
TABLE 2Recent SurveysonDiabetes inIndians
Place of
survey Author & YearMethod ofPrevalence
screening
of diabetes
Southall, London, UK
Mather andQuestionnaire2.2%
Keen, 1985 Daryaganj, New Delhi, India
Verma et ai, 1986Questionnaire3.1%
Kudremukh, S. India
RamachandranOGTT5.0%
etal,1988Recent Studies
Interest in diabetes in Asian Indians
received impetus after the recentSouthall Diabetes Survey (Mather andKeen, B.M.J, 1985: 291 :1081) whichshowed that the crude prevalence of'known' diabetes was two-fold higheramong Indians (2.2 percent) compared to Whites (1.2 percent) andAfro-Carribeans (1.2 percent). In the40-64 age group, the prevalence wasat least five-fold higher in Indians compared to Europeans. These datacome from two large ethnic groupswho live together in a well-integratedcommunity and use the same primaryhealth care system. It is, therefore,likely that these differences in prevalence rates are genuine. The Southallsurvey was soon followed by theDaryaganj Diabetes Survey in India(Verma, N.P.S., et ai, B.M.J,1986;293; 429). In this study, it was seenthat the prevalence of diabetes wasequal to, if not higher, in Daryaganj ascompared to Southall. The meanincome group of the Daryaganj studypopulation was Rs. 2,500/month. Thissuggests that if similar enviromentalfactors such as affluence, obesity,etc. are present, the prevalence ofdiabetes in India is not different from
that seen in migrant Indians.
The Kudremukh study: Inboth the Southall and Daryaganj studies, estimates of diabetes were basedon the prevalence of 'known' diabetes which was assessed by aquestionnaire. No blood tests werecarried out in either study. Nor werethe body weights recorded. The prevalence rates of diabetes in both the
Southall and Daryaganj surveyscould, therefore, be underestimates.
There was hence the need to take upan epidemiological study where thepopulation is actually tested for diabetes. A diabetes screening surveyof the population of Kudremukh, atownship which mainly consisted ofemployees of an iron ore factory inSouth India was carried out by theDiabetes Research Centre, Madras,with the active cooperation of the factory authorities. Every fifth member ofthe population was tested. Overt diabetes was seen in five percent of thepopulation and another 2.2 percenthad impaired glucose tolerance
(Ramachandran, et ai, B.M.J, 1988, inpress). The subjects investigated inthis study were generally more affluent than the general run of the ruraland urban slum population in thecountry. The figures were evenhigher in a subsample of this population with a higher income highlightingthe possible aggravating role ofaffluence. Table 2 summarises the
results of the Southall, Daryaganj andKudremukh surveys.
We may now briefly consider factors that could be possibly involved insuch relatively high diabetes prevalence among some Indian populationgroups.
• Familial aggregation of diabetes in Indians: At the Diabetes
Research Centre, we found thatabout 45-50 percent of diabetics hada first degree relative with diabetes(Viswanathan, et ai, Diab. Assn. Ind.,1977; 17; 9). What was more strikingin that study was that the prevalenceof diabetes among parents of diabetics seemed to be very high. Thisdegree of familial aggregation is rarelyseen among Europeans. To testwhether Indians have greater familialaggregation of diabetes, a study offamily histories of Asian Indian andEuropean diabetics living in Southallwas carried out (Mohan, et ai, PracticalDiabetes, 1986; 3: 254). It was seenthat Indians had a somewhat higherfrequency of diabetes in parents thanEuropeans (28 percent vs 20 percent). In 10 percent of Indian patients,both parents had diabetes comparedto one percent among Europeans.Vertical transmission of diabetes
through three or more generationswas seen in 14 percent of Indianscompared to four percent ofEuropeans. These studies suggestthat Indians have a higher familialaggregation of diabetes thanEuropeans.
• High prevalence of"MODY" and dominantly inherited diabetes: A relatively recentinteresting observation has beenmade that the form of diabetes
known as Maturity Onset Diabetes ofYouth (or MODY) is common amongIndians in South Africa (Asmal, et ai,S. Afr. Med. J, 1981; 60: 93) and inSouth India (Mohan, et ai, DiabetesCare, 1985; 3: 371). MODY is characterised by autosomal dominantinheritance including threegeneration transmission of diabetesand is the only form of diabeteswhere a genetic factor has beenestablished with some degree of certainty. About 50 percent of theoffspring of MODY ultimately developdiabetes. It is to be expected that inpopulations with a high prevalence ofMODY and dominantly inherited diabetes, the prevalence of diabeteswould be high.
Is defective insulin secretion
responsible for the diabetes? UsingMODY as a model of NIDDM, wefound defective insulin secretion in
MODY as assessed by insulin and Cpeptide responses to a glucose load(Mohan, et ai, Diabetes Care, 1985; 8:69). This defective insulin secretionseems to be genetically inherited. Asmentioned earlier, 50 percent of theoffspring of MODY could beexpected to ultimately developdiabetes.
Studies of insulin secretion were
therefore out in these offspring ofMODY patients at a stage whenthey had normal glucose tolerance.It was seen that decreased pancreatic beta cell function could be
demonstrated even in normoglycemic offspring of MODY (Mohan, etai, Diabetes Care, 1986; 9: 53). Thissuggested that defective insulinsecretion precedes glucose intolerance and is probably geneticallydetermined.
• High prevalence of diabetes among offspring ofconjugal diabetic parents(OCDP): The offspring of conjugaldiabetic parents (OCDP) form an interesting study group to study theprevalence of diabetes because bothparents have diabetes. AmongEuropeans, several studies haveshown that even when both parentsare diabetic, the prevalence of diabetes among OCDP is usually in therange of three to 12 percent(Tattersall and Fajans, Diabetes,1975; 24: 452). We studied the prevalence of diabetes among IndianOCDP and found that 50 percenthave overt diabetes and 12 percenthave impaired glucose tolerance.Thus 62 percent have abnormal glucose tolerance (Viswanathan, et ai,Diabetologia, 1985; 28: 907). Thesestudies on OCDP seem to indicate a
greater susceptibility to diabetesamongst Indians.
• Increased insulin resistance in Indians: It is well known
that one of the most important factorsin the aetiopathogenesis of NIDDMtype of diabetes is insulin resistance.Serum immunoreactive insulin
responses to a glucose load werestudied in well matched groups ofIndian and European subjects withand without diabetes. It was seen that
despite similar body weight, Indianshad higher plasma insulin responsesto a glucose load (Mohan, et ai,Diabetologia, 1986; 29: 235). Thissuggested that Indians may be moreinsulin resistant than Europeans. Totest this hypothesis, euglycemicclamp studies were carried out inmatched groups of Indian andEuropean NIDDM patients. It wasseen that Indians had lower glucosedisposal constants (M value) thanEuropeans (Sharp, et ai, Harm.Metab. Res., 1987; 19: 84). This confirmed that Indians had evidence ofincreased insulin resistance. It is,however, not clear at present whetherthe greater insulin resistance amongIndians has a genetic basis or is theresult of environmental and dietaryfactors.
• Role of environmental/nutritional factors: It is tempting totry to explain the high prevalence of
diabetes among Indians on geneticgrounds. But several observationssuggest that this cannot be the soleexplanation. Why have the prevalence rates for diabetes been higherin the recent Kudremukh and
Daryaganj surveys which covered therelatively affluent groups of the population as compared to the originalICMR survey which may be presumedto have largely captured the generalrun of the population composed of amajority of the poor? Increased prevalence of diabetes among migrantIndians in foreign countries as compared to the prevalence of diabetesamong the host populations need notalso necessarily imply a genetic factor;the diets, the occupational patternand the living style of the differentethnic groups could be different.
Of the various environmental fac
tors, perhaps, the most important oneto study is the role of diet. Several dietary factors could promotediabetogenesis and unmask geneticpredisposition to the disease. Theseinclude an increase in total calories,the free sugar intake, and a changeover to more refined foods (and therefore diets with lower dietary fibre).
We do not have precise dataregarding the total calorie intake in thegroups showing high prevalence ofdiabetes. Both the Daryaganj andKudremukh surveys reviewed herewere carried out in relatively affluentareas and a higher calorific intake inthe diet of these populations cannotbe excluded.
The role of free sugar (sucrose) inthe etiology of diabetes is far fromclear. A survey of the literature showsthat there are about 50 studies which
support the sucrose hypothesis andan equal number that do not.Carefully planned dietary surveys covering the whole population wouldhave to be done to answer the question whether increased sucroseintake is one of the factors contribut
ing to diabetogenesis.The role of dietary fibre in the man
agement of diabetes is wellestablished. Dietary fibre helps to flatten post-prandial surges of plasmaglucose (Viswanathan, et ai, J. Diab.Assn., India, 1978; 18: 119). Theinclusion of vegetable proteins in theform of bengal gram, green gram,etc., has also been shown to have
several beneficial effects in the man
agement of diabetes (Viswanathan, etai, J. Diab. Assn., India, 1975; 15: 45).However the role of dietary fibre inprotecting against diabetes, or conversely that of a low fibre diet incausing or precipitating diabetes, isless well established.
Special forms of diabetes associated with undernutrition have beenidentified - classified as malnutritionrelated diabetes mellitus, MRDM(WHO study group report on,diabetesmellitus, Technical Report Series727) and these merit further study.
The most interesting feature of theMRDM forms of diabetes is that in
spite of requiring insulin, sometimesin large doses, these patients do notdevelop ketoacidosis when insulininjections are withdrawn. It was earlierbelieved that it may be either due tothe small adipose tissue mass or todelayed mobilisation of free fattyacids. Recently, we have shown(Mohan, et aI, Metabolism, 1983: 32:1091) that these patients have someamount of residual pancreatic betacell function which may be the mechanism which protects them fromketosis. An alternative explanationeffered is that the glucagon levels inthese patients are paradoxically lowafter a glucose load in contrast topatients with ketosis prone 100M typeof diabetes where the glucagon levels rise after a glucose load (Rao, et ai,Diabetes, 1983: 32: 1163).
Clearly, there is considerablescope and need for more intensivestudies on diabetes in India. An
important area for such further studywould be the question of dietary management of Indian vegetariandiabetics. Recent developmentswould suggest that vegetarianism, farfrom being a handicap in dietary management of diabetes, may well turnout to be a favourable factor.
Studies at our centre during thelast decades have led to the evolution
of the High Carbohydrate High Fibre(HCHF) diet in the management ofdiabetes. The HCHF diet basicallycomprises a calories restricted dietwhich provides about 68.7 percent ofcarbohydrate, 18 percent of proteinand 13.3 percent fat. The diet has ahigh dietary fibre content (52 gms)which is approximately double that ofthe Standard American Diabetes
Reviews and Comments
Control of Vitamin A Deficiency - Prioritiesfor Future Research in India
c. Gopalan
Association Diet. The diet is com
prised of whole cereals (chiefly rice)along with pulses and legumes likebengal gram and green gram andgreen leafy vegetables. Extensivestudies at our centre have shown that
the HCHF diet is effective in achievingquick and effective control of diabetes (Viswanathan, J. Diab. Assn.India, 1968: 8: 353: 360). The dietalso helps to bring down serum cholesterol and triglyceride levels, whichare important risk factors for atherosclerosis (Viswanathan et aI, J. Diab.Assn. India, 1981: 30: 90). The dietimproves the peripheral insulin sensitivity to insulin and does notexcessively strain the pancreatic betacells (Viswanathan, et aI, J. Diab.Assn. India, 1983: 23: 45). Thepatients' acceptance and adherenceto the diet is excellent. In summary,the HCHF diet is ideal for countrieslike India where the diet is cerealbased.
The authors are senior scientists at the Diabetes
Research Centre, Madras
NUTRITIONNEWS
We congratulate:Dr. Rajammal Devadas on her
appointment as Vice-Chancellor,Home Science (Deemed) University,Coimbatore.
Dr. Vinodini Reddy on her appointment as Director, National Institute ofNutrition, Hyderabad.
FOUNDATIONNEWS
The following publications of theFoundation have been released:
• Scientific Report-8 - Profiles ofUndernutrition and Under development- A Study of Poor Communitiesin Seven Regions of the Country.
Scientific Report-9 - MaternalNutrition, Lactation and Infant Growthin Urban Slums.
A limited number of free copies ofthese publications will be available onspecific written request.
The current global approach for theprevention of keratomalacia arisingprimarily from vitamin A deficiency,through the distribution of two massive annual oral doses of syntheticvitamin A - one each at six-monthlyintervals - to children under three
years of age, was developed and pioneered in India.
However, looking back on theseefforts which were initiated a quarterof a century ago, and now looking atthe results, we may legitimately askwhether all the expectations whichprompted these efforts on the part ofIndian nutrition scientists have in factbeen fulfilled.
The control of nutritional blindness
through the 'short-cut' of administration of synthetic vitamin A had beenenvisaged as a short-term approach- not as the permanent solution ofthe problem. It was always recognisedthat the ultimate solution lay in thepromotion of the optimal use of Bcarotene rich foods - green leafyvegetables - in the dietaries of poorchildren. It must be confessed that
the euphoria and complacencecreated by the introduction of the prophylaxis through massive dosage ofsynthetic vitamin A has to a considerable extent retarded research
designed to develop and promotethe better use of inexpensive Bcarotene rich foods in the country. Ifsuch research has not altogethercome to a standstill, it is proceeding,at best, at snail pace as an effort of lowpriority.
Secondly, the implementation ofthe prophylaxis programme is obviously tardy especially in states likeBihar. What is most disconcerting isthat we do not have any authenticindication as to what real impact theprophylaxis programm~ has had onthe nutritional blindness problem.The official figures of annual incidence of cases of nutritionalblindness will not stand scientific scru
tiny. We do not even seem to havereliable data on changes in the annual
incidence of keratomalacia in our lead
ing ophthalmic and paediatrichospitals since the introduction of theprogramme. In the absence of suchdata, we are in no position to counteror confirm the claims that are fre
quently made.It would seem that while on the
one hand we are relying heavily onsynthetic vitamin A administration asthe answer to vitamin A deficiencywhich, as pointed out above, was notwhat was originally intended, on theother hand, we continue to beentirely dependent 'on a foreign commercial source for our supply ofvitamin A concentrate. The claims
that a major part of the vitamin A weneed for our programmes is being'indigenously manufactured' will againnot stand scrutiny - unless weaccept that 'bottling' and participationin the subsidiary and final stages ofmanufacture amounts to 'indigenousmanufacture'; the truth appears to bethat the technical know-how for the
essential step in the manufacture ofsynthetic vitamin A rests with a foreign commercial source whichvirtually holds monopoly in thisregard. It is not clear as to why Indianscientists have shown no enthusiasm
in achieving self-reliance in thisregard and why no "TechnologyMission" has been set up for thisvalid purpose.
Recent research indicates that the
implication of vitamin A deficiency maybe more far-reaching than what wehad earlier imagined. This is all themore reason that this subject shouldnow receive renewed attention. At
the same time we must also guardagainst possible indiscriminate,excessive use of synthetic vitamin A,which is not without its harmful
effects. This caution is speciallynecessary in view of current attemptsto promote synthetic vitamin A notjust for the prevention of keratomalacia in poor children but as the mosteffective single step for the reductionof child mortality.
