|
|
||||||||
Biometry & Nutrition Unit, Agharkar Research Institute (A.N.K., S.R., R.S.K.)
Gupte Clinic (S.G.), Pune, INDIA
Address correspondence to: Dr. (Mrs) Shobha Rao, Head, Animal Science Division, Agharkar Research Institute, G.G. Agarkar Road, Pune 411 004. E-mail: raoari{at}yahoo.com
| ABSTRACT |
|---|
|
|
|---|
Objective: To examine this relationship among urban affluent mothers and to compare the findings with those on rural Indian mothers.
Subjects: Data collected on urban affluent mothers (n = 236) was compared with rural mothers (n = 633).
Design: Mothers were contacted at 18 ± 2 and 28 ± 2 wk of gestation for anthropometry, dietary intakes [24-hr recall, Food Frequency Questionnaire (FFQ)] and after delivery for neonatal anthropometry.
Results: Despite large differences in nutritional status of urban and rural mothers ( pre-pregnant weight 55.9 ± 9.2 Vs 41.5 ± 5.2 kg, respectively) maternal fat intakes at 18 wk were associated with birth weight (p < 0.05), length (p < 0.01) and triceps skin fold thickness (p < 0.05) of the newborn in urban and rural mothers. Consumption of fruits was associated with birth length (p < 0.05) in urban (18wk) and with birth weight (p < 0.01) and length (p < 0.01) in rural (28wk) mothers, when their energy intakes were low. Maternal consumption of milk too, was associated with newborn's triceps (p < 0.01) in urban (28wk) while with birth weight (p < 0.05) and length (p < 0.05) in rural (18wk) mothers. The findings mainly underscore the importance of consumption of micronutrient rich foods, when energy intakes are limiting during pregnancy, for improving birth size.
Conclusions: Creating nutritional awareness and motivating rural mothers for consuming micronutrient rich foods like green leafy vegetables and seasonal fruits that are easily available in rural areas, will be a much affordable solution for combating the problem of low birth weight rather than waiting for improvement in the existing nationwide programs for pregnant women.
Key words: birth size, India, maternal nutrition, micronutrient rich foods
| INTRODUCTION |
|---|
|
|
|---|
Relationship of maternal nutrition with birth size is influenced by many biological as well as socio-economic factors. In India, poor fetal growth has been attributed to widespread maternal under-nutrition [10]. Although energy and protein are believed to be the major macronutrients that are associated with birth size, worldwide studies of supplementation of these nutrients during pregnancy have produced variable and sometimes conflicting results [11]. Associations of maternal fat intakes with birth size are rarely examined [12] and there are hardly any data for fat reported as an entity separate from energy.
Reported studies on pregnant women in India [13,14] have examined micronutrients supplementation with folate, iron and zinc in pregnancy and have shown beneficial impact on fetal growth. However, there are limited data from controlled trials examining effects of other micronutrients on fetal growth. In fact, in our recent study [15] on the rural undernourished Indian women, strong associations of birth size were not observed with maternal consumption of macronutrients but only with consumption of micronutrient rich foods such as green leafy vegetables (GLV) and fruits. In view of the considerable rural-urban differences that exist with regard to the food consumption patterns and more importantly with respect to nutritional status of women it would be important to investigate this relationship among women from urban affluent populations in India. Maternal consumption of micronutrients may perhaps be equally important as that of macronutrients in determining fetal growth regardless nutritional status, social class or place of residence of mothers. Therefore the aim of this study was to examine this relationship prospectively among urban affluent mothers and compare the findings with those obtained earlier from a rural prospective study [15].
| MATERIALS AND METHODS |
|---|
|
|
|---|
Anthropometry (weight, height, skin folds at four sites and head and mid-arm circumference) for women, before and during pregnancy at 18 ± 2 and 28 ± 2 wk gestation, as well as for newborns was obtained. During pregnancy women were assessed for dietary intake (24 hr recall, FFQ) and activity as well. Methods for these measurements were kept similar in the urban study and are described below.
Urban Study
Mothers.
Pregnant women coming for antenatal care in a private hospital situated in an elite area of Pune city which was catering mainly to high income class, were studied during March 1998 to March 2001. Women were listed (n = 265) before 12 wk of gestation after seeking their consent to participate in the study. During the first visit (18 ± 2 wk of gestation) detailed information on socio-economic variables was obtained. The information on occupation, education of self and spouse and possession of economic assets such as owning a house, car, telephone etc. was obtained on 3-point scale each so as to discriminate between income classes. Women then identified from affluent class were enrolled and visited at their house second time during their pregnancy (28 ± 2 wk of gestation). Thus out of 265 women who gave consent 236 were actually enrolled after ensuring their income class. The drop out of the enrolled women in subsequent visits was mainly due to the fact that they were not available in a defined window of visits. The study was approved by Institutional Research Advisory Committee.
Maternal Nutritional Status
At both the visits women were measured for weight (Atco India, 200 g least count), skinfold thickness (triceps and biceps, Langes Caliper, least count 0.1 mm), mid-arm and head circumference (non-stretchable tapes, 0.1 cm least count) and height (Stadiometer UNA, India 0.1 cm least count). Pre-pregnancy weight recorded before 10th wk of gestation was obtained from hospital records.
Dietary Intake
Estimates of dietary intakes using a conventional 24-hr recall method depend on correct recall of the foods consumed, their portion sizes and nutrient contents. The precision of the estimate needs not only skilled interviewing, but also the objective assessment for the portion sizes of the food consumed and their nutritive values. We therefore generated a database on weights of portion sizes using food models (for a staple food like roti) and standard spoons used in urban community for serving various food items. The nutrient contents of the foods consumed were obtained using the food composition tables [16]. Thus using modified 24-hr recall method, daily intakes of macronutrient i.e. energy, protein, fat & carbohydrate were estimated for each women twice during their pregnancy (18 ± 2 and 28 ± 2 wk of gestation). In addition, to have information on habitual pattern of diet, food frequency questionnaire (FFQ) was also administered. Frequencies of consumption of various (111) foods in 10 food categories, ranging from "never to "twice/thrice a day were obtained for the reference period of one month prior to each visit. The 10 food categories were: chapati / roti, rice, pulses / legumes, other / green leafy vegetables, fruits, meat/fish, dairy products, snacks (sweet & spicy), fats, foods eaten outside home. These food categories were mutually exclusive. Frequency of consumption of individual food item was converted into score and then score for each food category was obtained by adding the scores of various foods in it. Thus, in our study while 24 hr recall offered objective estimates of macronutrients, FFQ allowed us to compare the groups of women consuming relatively low and high frequency of specific food groups.
Similarly, physical activity was assessed with the use of activity questionnaire consisting of four major categories such as resting, domestic, occupational and recreational activities with several activities listed in each major category. Using published data for the energy expenditure of activities [17] a weighted score was derived, which reflected as a base unit an activity level of 1 kcal per minute for a 30 minute slot of time. For example, walking for 60 minutes would have a score of 6.8 (2 x 3.4; 2 slots 30 min. each & 3.4 Kcal/min is energy cost for walking). The scores were computed for each individual activity in the questionnaire and adding over all the activities performed in a day, total daily activity score was obtained for each women, at both time points (i.e. 18 ± 2 and 28 ± 2 wk of gestation).
Neonatal Anthropometry
Of the enrolled 236 mothers, 4 reported abortions, 4 delivered twins and 2 could not be contacted after the delivery for neonatal measurements. Of the 226 live deliveries, complete neonatal anthropometry (weight, length, triceps skin fold thickness, mid-arm & head circumference obtained on all singleton births within 72 hr. after delivery) was available on 165 babies (omitting 34 deliveries that took place outside Pune, 8 in other hospitals, 14 could not be contacted within 72 hours and 5 were refusals). However, for these 61 omitted cases, parents reported birth weight of their baby.
Statistical Methods
Differences between group means were tested using t-test and ANOVA. Multiple regression analysis was used to examine trends in birth-size according to maternal dietary intakes, and to assess the relative contributions of other factors to the variability in birth measurements. In regressions, birth measurements, maternal measurements and macronutrient intakes were analyzed as continuous variables while intake of specific food categories based on the food frequency questionnaire was analyzed as grouped variable. For example, the frequency of consumption of non-vegetarian food with categories as never eating, less than once a week & more than once a week was considered as a grouped variable with values 0,1 & 2 respectively. Neonatal measurements were significantly higher among boys (p < 0.05) and of the mothers with multiparous (p < 0.05) and having higher gestation (p < 0.05). All analyses, therefore, were adjusted for the sex (of the baby), gestational age at delivery, and maternal parity. Neonatal skin folds measurements were skewed and required log-transformation to satisfy assumptions of normality. Analysis was carried out using SPSS/PC+, version 11.0. Values unless otherwise stated are Mean ± SD.
| RESULTS |
|---|
|
|
|---|
Pre-pregnancy nutritional status of urban mothers (Table 1) was good and the average BMI was 22.6 ± 3.6 kg/m2. Only 1% women were below 38 kg and below 145 cm height, i.e. the cut-offs below which risk for LBW is considered to be high [18]. Mean weight gain during pregnancy up to 18th wk gestation was however low, 2.07 ± 2.6 kg, but it increased to 7.53 ± 3.4 kg by 28th wk of gestation.
|
Maternal Intakes
Macronutrients (24 h Recall).
Mean energy and protein intakes at 18th wk gestation were inadequate and were low (8.09 ± 2.47 MJ, 51.2 ± 17.3 g respectively) as compared to RDA for Indian pregnant women [20]. The intakes were low mainly due to high prevalence of nausea (50.8%) reported within first 3–4 months. However, intakes improved considerably by 28th wk of gestation (9.03 ± 2.37 MJ of energy, 57.9 ± 18.7 g of protein) meeting the nutrient requirements. Their diets were rich in fats (31% of total energy) but were low in carbohydrates (58% of total energy). Milk, fried snacks, sweets and bakery products were major contributors (20 %) to dietary fat intakes. Apart from cereals and pulses, milk and non-vegetarian items contributed about 13 % to protein intakes. The activity pattern remained similar throughout the pregnancy (Table 2).
|
|
|
|
Fruits.
Fruit consumption in urban affluent women was relatively high. Commonly consumed fruits were apple (55%), banana (49%), sapota (38%), mango (24%) and orange/sweet lemon (32%) etc. It was observed that the frequency of consumption of fruits at 18th wk was related to birth length (p < 0.05) after adjusting for gestation, sex and parity (Table 6). The association remained significant even after adjusting for prepregnant weight (height or BMI). The relative risk of delivering short (lower tertile of length) babies was 2.8 (95% CI, 1.32–5.79) times higher in mothers who ate fruits less than once a day than in mothers who had higher consumption of fruits (more than once a day). Fruits are rich sources of micronutrients especially vitamins and antioxidants and its association at 18th wk when energy intakes are low assumes significance.
|
Comparison with Rural Women
Comparison with the rural data [15] shows that rural mothers were significantly thin (41.5 ± 5.2 kg), and short (1.52 ± 0.05 m); p < 0.001 for both, as compared to urban affluent women (Fig. 1) and many were chronically energy deficient (34% below < 17 BMI) before conception. The rural neonatal size (2.67 ± 0.35 kg birth weight, 47.9 ± 2.3 cm length and 4.1 ± 0.9 mm triceps skinfold thickness) was also significantly (p < 0.001) lower in almost all anthropometric measurements except head circumference (33.3 ± 1.3 cm). Prevalence of LBW (26.9%) and prematurity (12.3%) among rural mothers was significantly (p < 0.05) higher as compared to urban mothers. Maternal energy and protein intakes were low (below 70% of RDA, ICMR [20]) at 18th wk and did not improve by 28th wk of gestation unlike that seen in case of urban mothers. Similarly, the consumption of foods which are important sources of micronutrients was lower in rural women as compared to urban women. For example, the frequency of having milk at least once a day (29.7% Vs 100%), frequency of having fresh fruits once a day (40% Vs 76%) and frequency of having green leafy vegetables every alternate day (30% Vs 64%) was significantly (p < 0.001 for all) lower.
|
Commonly consumed GLVs by rural mothers were fenugreek (57%), spinach (33%), colocasia (15%), coriander (16%) and onion stalk which are rich source of iron, folate and other micronutrients. While fruits consumed (eaten more than once a wk) were those which were also rich in micronutrients and especially antioxidants like Vit C. These fruits were rarely purchased as they were freely available in the fields e.g. tamrind (48% of women ate), country guava (40%), custard apple, zizapus (58%) etc. Association of consumption of these food groups with birth size is therefore, indicative of the importance of combinations of micronutrients when maternal diets are inadequate in energy.
| DISCUSSION |
|---|
Urban affluent mothers had a good pre-pregnant nutritional status and their weight, height were comparable to those of urban affluent women from other parts of India [21–23]. Weight gain (18th wk or 28th wk of gestation) was not associated with birth weight or length but pre-pregnant nutritional status was strongly related with birth weight in rural as well as urban mothers indicating importance of good pre-pregnant nutritional status.
The neonatal size observed for urban babies was also comparable to that reported in other well-to-do Indian populations [24] and was closer to the 30th percentile of NCHS standards. Rural babies had significantly low birth weight, skin fold thickness at triceps and ponderal index in comparison to urban babies, but had comparable mean values for length, head circumference and mid-arm circumference. Preservation of head circumference in rural babies in spite of chronic undernutrition of their mothers renders some support to the observations of brain sparing on pups born to undernourished dams, reported from animal studies [25,26]. In addition, our observation also indicates preservation of length in rural babies highlighting similar importance for skeletal growth in the event of maternal undernutrition.
Maternal intakes of energy and protein of urban affluent mothers were low at 18th wk of gestation, and were not associated with birth size. However, fat intakes at 18th wk of gestation were associated with birth size among urban mothers. Although fat intake has been shown previously to correlate with birth weight [12], there are few data for fat, reported as an entity separate from energy. The role of specific fatty acids is shown to be important for fetal growth [27,28]. Therefore, it cannot be judged whether it was due to the role of fat as macronutrient or fatty acids as micronutrient.
Fruits are rich in micronutrients especially, vitamin C and other antioxidants. We observed strong association of birth length with consumption of fruits at 18th wk of gestation in urban mothers and at 28 wk among rural mothers. The relative risk of delivering short (lower tertile of length) babies was 2.8 (95% CI: 1.32–5.79) times higher in urban and 2.1 (95% CI: 1.44–3.07) times higher in rural mothers who ate fruits less than once a day than in mothers who had higher consumption of fruits (more than once a day). In addition to fruits, increased consumption of green leafy vegetables and milk products in rural women was also associated with better neonatal size. The observations therefore, suggest that consumption of foods rich in micronutrients may be beneficial for improving birth size when maternal intakes are limited in macronutrients.
Susser [29] and Stein [30] have reported that the relationship between maternal energy intake and birth weight was not seen among non-famine Dutch women but was seen among those who experienced famine conditions. We observed that although same nutrients or foods were associated with birth size, their associations differed at different times in gestation and with different neonatal measures among urban well nourished and rural undernourished mothers. For example, fat intakes at 28 wk were associated with birth weight in urban mothers but with birth length in rural mothers. Similarly, consumption of fruits at 18 wk was observed to be associated with length in urban mothers, but at 28 wk with birth weight as well as length in rural mothers. Energy intakes of the mothers in both places were drastically low at these times, and may indicate important role of antioxidants present in fruits. Milk consumption too shows significant associations with triceps in urban group (at 28 wk) but with birth weight and length in rural mothers (at 18 wk). Additionally, strong associations of GLV consumption were seen with birth size in rural mothers but not in urban mothers. The observations thus indicate that relationships of maternal intakes of micronutrients with birth size may be conditioned by maternal intakes of macronutrients in addition to their pre-pregnant nutritional status.
In view of the strong relationship between pre-pregnancy nutritional status and birth size, it may be worthwhile to have programs to improve nutritional status of mothers before conception than supplementing during pregnancy. Alternatively, our observations suggest that the existing nationwide programs providing energy-protein supplementation to pregnant women may be modified to improve it's qualitative aspect (micronutrients). Simple food based interventions may be feasible solution for combating the problem of low birth weight in rural India. It is however, beyond doubt that further research is needed for understanding the relationship of different micronutrients at different time points in gestation with different neonatal measures in undernourished mothers for planning effective interventions.
| ACKNOWLEDGMENTS |
|---|
|
|
|---|
Received October 20, 2005. Accepted July 12, 2006.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |