Hostname: page-component-848d4c4894-ndmmz Total loading time: 0 Render date: 2024-05-18T05:23:54.168Z Has data issue: false hasContentIssue false

Fibre intake and evolution of BMI: from pre-pregnancy to postpartum

Published online by Cambridge University Press:  24 August 2012

Michele Drehmer*
Affiliation:
Graduate Studies Program in Epidemiology, Department of Social Medicine, Federal University of Rio Grande do Sul, Rua Ramiro Barcelos no. 2600 – 4° andar – sala 419, CEP 90035-003, Porto Alegre, RS, Brazil
Suzi Alves Camey
Affiliation:
Graduate Studies Program in Epidemiology, Department of Social Medicine, Federal University of Rio Grande do Sul, Rua Ramiro Barcelos no. 2600 – 4° andar – sala 419, CEP 90035-003, Porto Alegre, RS, Brazil Department of Statistics, Institute of Mathematics, Federal University of Rio Grande do Sul, Porto Alegre, RS, Brazil
Maria Angélica Nunes
Affiliation:
Graduate Studies Program in Epidemiology, Department of Social Medicine, Federal University of Rio Grande do Sul, Rua Ramiro Barcelos no. 2600 – 4° andar – sala 419, CEP 90035-003, Porto Alegre, RS, Brazil
Bruce B Duncan
Affiliation:
Graduate Studies Program in Epidemiology, Department of Social Medicine, Federal University of Rio Grande do Sul, Rua Ramiro Barcelos no. 2600 – 4° andar – sala 419, CEP 90035-003, Porto Alegre, RS, Brazil
Mauro Lacerda
Affiliation:
Department of Statistics, Institute of Mathematics, Federal University of Rio Grande do Sul, Porto Alegre, RS, Brazil
Andréia Poyastro Pinheiro
Affiliation:
Graduate Studies Program in Epidemiology, Department of Social Medicine, Federal University of Rio Grande do Sul, Rua Ramiro Barcelos no. 2600 – 4° andar – sala 419, CEP 90035-003, Porto Alegre, RS, Brazil
Maria Inês Schmidt
Affiliation:
Graduate Studies Program in Epidemiology, Department of Social Medicine, Federal University of Rio Grande do Sul, Rua Ramiro Barcelos no. 2600 – 4° andar – sala 419, CEP 90035-003, Porto Alegre, RS, Brazil
*
*Corresponding author: Email michele.drehmer@gmail.com
Rights & Permissions [Opens in a new window]

Abstract

Objective

To evaluate the effect of fibre intake on the evolution of maternal BMI from pregnancy to postpartum and to identify dietary patterns associated with fibre intake.

Design

Cohort study. Food intake was obtained using an FFQ. Focused principal component analysis was used focusing on the variables: postpartum weight retention and total dietary fibre intake. Poisson regression models with robust variance were built in order to measure the effect of fibre intake during the postpartum period on obesity risk.

Setting

Primary care clinics in southern Brazil.

Subjects

Pregnant women (n 370) were followed until the 5th month postpartum.

Results

The highest contribution to fibre intake came from the consumption of beans. Consumption of bread and rice indicated a common Brazilian food pattern along with beans. Participants retained a median of 4·4 (interquartile range 0·6, 7·9) kg of weight gained during pregnancy. Obesity risk, defined as an unfavourable evolution of BMI during pregnancy and postpartum, was present in 189 (55·1 %) women. Individual food items did not have an important effect on weight retention. In Poisson regression adjusting for maternal age, pre-pregnancy BMI and total gestational weight gain, inadequate postpartum fibre intake increased obesity risk by 24 % (relative risk = 1·24; 95 % CI 1·05, 1·47).

Conclusions

Important maternal weight retention occurred in these women. Adequate fibre intake may reduce obesity risk in the period following childbirth.

Type
HOT TOPIC – Nutrition in pregnancy
Copyright
Copyright © The Authors 2012 

The postpartum period appears to be a critical period for obesity development in women of reproductive age due to maternal weight retention determined by a series of interrelated factors(Reference Oken, Taveras and Popoola1, Reference Lacerda and Leal2). Gestational weight gain above the recommendations of the Institute of Medicine (IOM) is considered the most important predictor for postpartum weight retention, increasing the short-, medium- and long-term risk obesity(Reference Kac, Benicio and Velasquez-Melendez3Reference Siega-Riz, Viswanathan and Moos8).

Among other factors associated with postpartum weight variation, maternal age, parity, social and demographic factors as well as pre-pregnancy nutritional status represent non-modifiable factors or ones that occur before pregnancy(Reference Oken, Taveras and Popoola1, Reference Lacerda and Leal2). Modifiable factors related to lifestyle, such as diet, physical activity, smoking and lactation, and their effects on the postpartum weight evolution, are still relatively underexplored in the literature(Reference Rasmussen and Yaktine9Reference Kinnunen, Pasanen and Aittasalo12).

Maternal energy intake needs to be increased in both pregnancy and postpartum in order to meet physiological demands. The American Dietetic Association recommends that food intake in pregnancy and the postpartum period should be based on whole grains, fruit and vegetables, aiming at meeting recommended vitamin and mineral contents(13). Nutritional restriction is discouraged even for obese women in the postpartum period, and methods that emphasize a fast weight loss are not recommended(14). Approximately 13–20 % of women do not return to their pre-pregnancy weight, showing significant postpartum weight retention(Reference Kac, Benicio and Velasquez-Melendez15Reference Gunderson17). More than two-thirds of women face some degree of weight retention at 6 weeks postpartum(Reference Walker, Sterling and Timmerman18). Postpartum maternal food recommendations are based on the needs to promote breast-feeding, but the best way to achieve the ideal maternal weight over the postpartum period is not clear(Reference Castro, Kac and de Leon19). The only international postpartum weight-loss recommendation is about 2 kg/month after the first month postpartum for women with low or normal pre-pregnancy BMI(14). It is well known that a fibre-rich diet is related to weight control and obesity prevention(Reference Lindstrom, Peltonen and Eriksson20Reference Bhargava22). Pregnant and postpartum women should have a higher fibre intake compared with the general adult population. The positive effects of fibre consumption for pregnant and postpartum women include promoting heart health and decreasing diabetes risk and constipation effects as well as reducing pre-eclampsia risks, together with providing nutrient-rich food with low energy density(13, 23).

For the population in general, convincing evidence exists demonstrating the protective effect of a fibre-rich diet against weight gain and a minimal daily fibre intake is recommended by the ‘Global Strategy for Healthy Eating, Physical Activity and Health’ of the WHO(24, Reference Swinburn, Caterson and Seidell25). It is still not clear in the current literature whether the consumption of a fibre-rich diet could help prevent postpartum maternal weight retention. A recent finding points to a possible protective effect. Oken et al. describe that daily intake of fibre was associated with a protective effect on postpartum weight retention of ≥5 kg (OR = 0·72, 95 % CI 0·56, 0·94)(Reference Oken, Taveras and Popoola1).

Based on the aforementioned, the present study aimed to evaluate the effect of postpartum fibre intake on the change of maternal BMI from pre-pregnancy to postpartum, as well as to identify dietary patterns associated with fibre intake among women assessed in the 4th–6th month after childbirth.

Methods

Study design and sample

The ECCAGE Project (Study of Food Intake and Eating Behaviour of Pregnant Women; the acronym is derived from the study's Portuguese name of Estudo do Consumo e do Comportamento Alimentar em Gestantes) is a cohort study of pregnant women enrolled in primary health-care units of the Brazilian Unified Health System, and was conducted in the cities of Porto Alegre and Bento Gonçalves between 2006 and 2007. The cohort consists of three phases: (i) baseline (phase I), in which pregnant women between the 16th and 36th pregnancy week were consecutively enrolled in the study; (ii) the follow-up period immediately postpartum (phase II) during which information was obtained from medical records; and (iii) further follow-up between the 4th and 6th postpartum months (phase III). The present study refers to phase III, which included a convenience sample: sixty-one out of 311 women from Bento Gonçalves who were followed up in the 4th–6th month postpartum and 392 out of 401 women from Porto Alegre who were followed up in the 4th–6th month postpartum, for a total sample of 453 women.

In these phases, women had an evaluation of their current weight, as well as a retrospective evaluation of weight variation before, during and after pregnancy, in order to verify maternal weight retention and BMI evolution in the 4th–6th month postpartum. Additionally, a measurement of food consumption during the postpartum period was accessed by means of an FFQ.

In a previous paper, information regarding the methodology of all study phases was comprehensively described, including interview logistics, measurement of sociodemographic variables and maternal eating habits, weight and height measurements, and quality control(Reference Nunes, Ferri and Manzoli26). The study flowchart is shown in Fig. 1.

Fig. 1 Flowchart showing participants in the study. *Two excluded women were simultaneously without weight retention data and consumed >25 104 kJ/d or >16 portions of fruits and vegetables/d; four excluded women simultaneously consumed >25 104 kJ/d and >16 portions of fruits and vegetables/d

Exclusion criteria were: twin pregnancy; lack of postpartum weight retention data; and implausible data for weight retention (below −10 kg or above 20 kg), daily energy consumption (>25 105 kJ/d)(Reference Castro, Kac and de Leon19) or daily consumption of fruits and vegetables, as standardized according to the Food Guide for the Brazilian Population (>16 portions/d)(Reference Buss, Nunes and Camey27, 28) (Fig. 1).

The study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects/patients were approved by the Committee for Ethical Research of the Universidade Federal do Rio Grande do Sul, of the Health Center of Escola Murialdo and by the Municipal Health Departments of Porto Alegre and Bento Gonçalves. Informed consent was signed by all pregnant women who agreed to participate in the study.

Main predicting variables – fibre intake and gestational weight gain

An FFQ previously validated for pregnant women(Reference Giacomello, Schmidt and Nunes29) was used to evaluate food consumption. This instrument comprises eight consumption frequency options, varying from ‘more than three times a day’ to ‘never or nearly never’. The list contains eighty-eight food items for which standardized portions of household measures or food units evaluating the consumed quantity are offered.

Estimates of fibre (g/d) and energy consumption (kJ/d), based on the FFQ, were obtained by summing estimates of fibre and energy consumption from each food, these being calculated as: consumption estimate = consumption frequency/d × portion size × number of portions × (portion nutrient/portion size)(Reference Zulkifli and Yu30, Reference Willett31).

Determination of the quantity in grams of a standardized portion was done using the Brazilian Table for Evaluation of Food Consumption in Household Measurements (Reference Pinheiro32). Food not contemplated in this table was weighed or measured in order to obtain a default value in grams. The Brazilian Food Composition Table (33) was used as a reference to obtain the food's nutritional composition (energy value and value in grams, milligrams or micrograms of macro- and micronutrients). A second food composition table, the Food Composition Table: A Support for Nutritional Decision (Reference Philippi34), was consulted when any food appearing in the FFQ or any nutrient under investigation was not considered in the Brazilian Food Composition Table. Food labels were consulted only as a third option to obtain nutritional information.

Food items were separated by food groups according to their description in the Food Guide for the Brazilian Population (28), permitting estimation of the number of portions of fruits and vegetables consumed daily.

In order to evaluate fibre intake adequacy at the 4th–6th month postpartum, the Adequate Intake recommendation of the Dietary Reference Intakes was used, which recommends consumption of at least 29 g fibre/d in the postpartum period. Women who were breast-feeding were classified as having ‘reached’ or ‘not having reached’ the recommendation of 29 g fibre/d. Mothers who were no longer breast-feeding were classified according to the Adequate Intake value for fibre recommended for adult women (25 g/d). The same logic was used for the classification of maternal energy consumption adequacy. Since there are no Brazilian guidelines on intake of fibre during lactation, the American ones were employed(13).

Total gestational weight gain was calculated as the difference between the weight at the end of pregnancy obtained from medical records (phase II) and the reported pre-pregnancy weight (phase I). Final pregnancy weight was considered as the last weight recorded at most two weeks before delivery. Pregnant women whose weight was not registered during the last two weeks before delivery had their values attributed using the simple imputation method, which is, multiplying the weekly weight gain average (adopting the first weight recorded from the second trimester as the basal component) by the number of weeks between the last weight registered during the prenatal period and childbirth. Pregnant women who had no weight registry beyond the 28th week were excluded. Further details regarding gestational weight gain were previously published(Reference Drehmer, Camey and Schmidt35).

Total weight gain was classified according to the 2009 IOM recommendations. Total weight gain between 12·5 and 18·0 kg was considered adequate for women with pre-pregnancy BMI below 18·5 kg/m2, between 11·5 and 16·0 kg for women with pre-pregnancy BMI between 18·5 and 24·9 kg/m2, and between 7·0 and 11·5 kg/m2 for women with pre-pregnancy BMI between 25·0 and 29·9 kg/m2. Total weight gain between 5·0 and 9·0 kg was considered appropriate when pre-pregnancy BMI was higher than or equal to 30·0 kg/m2(Reference Rasmussen and Yaktine9).

Outcome variable – weight retention and BMI evolution from pre-pregnancy to postpartum

Participants reported their pre-pregnancy weight in phase I of the study. In phase III, weight was measured by a trained interviewer according to a standardized protocol(Reference Nunes, Ferri and Manzoli26). Weight retention (kg) was defined as the difference between measured phase III maternal weight and reported pre-pregnancy weight.

The outcome – obesity risk – was defined to indicate an unfavourable evolution of BMI from pre-pregnancy to 4th–6th month postpartum. This risk was present with the following:

  1. 1. Change from a normal weight pre-pregnancy to overweight or obesity in postpartum;

  2. 2. Change from pre-pregnancy overweight to obesity in postpartum;

  3. 3. Change from class I obesity pre-pregnancy to class II obesity in postpartum;

  4. 4. Change from class II obesity pre-pregnancy to class III obesity in postpartum;

  5. 5. Weight gain among women who reported pre-pregnancy overweight or obesity but not enough to alter nutritional status in postpartum;

  6. 6. Weight loss among women who reported pre-pregnancy overweight or obesity but not enough to alter nutritional status in postpartum.

Covariates

The sociodemographic covariates were maternal age (≤19 years, 20–29 years and ≥30 years), years of schooling (≤4 years, 5–8 years and ≥9 years), family income (expressed in minimum monthly wages, one minimum wage being equivalent to $US 175/month; categorized as ≤1 minimum wage, >1 and <3 minimum wages and ≥3 minimum wages), marital status (married, living with a partner, single/separated/divorced) and parity (primiparous and multiparous). Behavioural and nutritional covariates were also included in the following categories: smoking (smoker or non-smoker), breast-feeding (exclusive breast-feeding, non-exclusive and no breast-feeding), pre-pregnancy BMI (classification according to IOM 2009) and energy consumption (has reached or has not reached the recommendations according to the Dietary Reference Intakes). The postpartum follow-up was categorized into ‘four months’ and ‘five to six months’.

Statistical analysis

Participants’ characteristics were described by absolute and relative frequencies and Pearson's χ 2 test was used to test the association with obesity risk. Continuous variables were described as mean and standard deviation, or median and interquartile range (IQR; 25th percentile, 75th percentile), unless otherwise stated.

Focused principal component analysis (FPCA) was used to investigate the relationship between maternal weight retention and food consumption as well as the relationships among food items. This method was also used to investigate which foods had a stronger relationship with dietary fibre intake, as well as to characterize food patterns formed by the consumption of such foods(Reference Falissard, Corruble and Mallet36, Reference Canuto, Camey and Gigante37).

For the FPCA a list of nineteen selected foods items, among the eighty-eight listed in the FFQ, were correlated with fibre intake and with postpartum weight retention. Low-glycaemic-index foods, fibre-source foods belonging to the fruit and vegetable groups and the food items rice and bread were included because they were consumed at least twice daily by more than 50 % of the sample. Foods not consumed by at least 75 % of the sample were excluded.

In the FPCA, correlations are represented as concentric circles, in which circles with a smaller radius indicate stronger correlations. The outcome variable (fibre intake or postpartum weight retention) is listed at the centre of the circle. Positive correlations were plotted using $$$$ symbols and negative ones with ○ symbols. This analysis was carried out in the R program (R Development Core Team, 2008) through the psy library(Reference Falissard, Corruble and Mallet36).

Poisson regression models with robust variance were built in order to measure the effect of fibre intake during the postpartum period on obesity risk. Adjusted models included covariates with P ≤ 0·20 in the univariate analysis. Covariates (age, schooling, monthly family income, parity, pre-pregnancy BMI, gestational weight gain, energy intake and postpartum follow-up time) were included as continuous variables in the multivariate model in order to increase power, whereby only variables with P < 0·05 were kept in the final model. Results were described as relative risks (RR) and 95 % confidence intervals. The SPSS statistical software package version 18 was used.

Results

Mean age of participants was 24·8 (sd 6·0) years, average schooling was 7·7 (sd 2·8) years and average monthly family income was $US 430·7 (sd $US 338·3). Mean postpartum follow-up time was 4·7 (0·6) months and 87 % of the interviews took place until the fifth postpartum month. Average pre-pregnancy BMI was 24·4 (sd 4·6) kg/m2 and the total gestational weight gain was 14·2 (sd 6·2) kg. Other social, demographic, behavioural and maternal nutrition features of the total sample, stratified by obesity risk, are described in Table 1. Variables significantly associated with obesity risk were maternal age, parity, marital status, pre-pregnancy BMI, total gestational weight gain, fibre and bean intakes in phase III (all P < 0·001 except for marital status, fibre and bean intakes, where P = 0·021, 0·018 and 0·034, respectively; Table 1).

Table 1 Sociodemographic, behavioural and maternal nutrition factors in the postpartum period associated with obesity risk among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007

*Obesity risk – evolution of BMI from pre-pregnancy to 4th–6th month postpartum: change from a normal weight pre-pregnancy to overweight or obesity in postpartum, from overweight pre-pregnancy to obesity in postpartum, from class I obesity pre-pregnancy to class II obesity in postpartum, from class II obesity pre-pregnancy to class III obesity in postpartum, weight gain among women who reported pre-pregnancy overweight or obese but not enough to alter nutritional status in postpartum, weight loss among women who reported pre-pregnancy overweight or obesity but not enough to alter nutritional status in postpartum.

†Pearson's χ 2 test.

‡MW = minimum wages (equivalent to $US 175/month).

§Total gestational weight gain classification according to the US Institute of Medicine (2009)(Reference Rasmussen and Yaktine9).

∥Fibre or energy recommendations according to Dietary Reference Intakes, considering breast-feeding or non-breast-feeding women (Institute of Medicine (2005)(23) and American Dietetic Association (2008)(13)).

¶Thirteen per cent of the sample were interviewed at the beginning of the sixth postpartum month.

Median postpartum maternal weight retention was 4·4 (IQR 0·6, 7·9) kg. Obesity risk incidence, according to BMI evolution in the postpartum period (n 189), was 51·1 % (95 % CI 46·0, 56·2 %). Figure 2 shows cohort participants stratified by obesity risk category.

Fig. 2 Change from pre-pregnancy to postpartum in obesity risk category (in clockwise order: $$$$ , normal weight to overweight or obesity; $$$$ , overweight to obesity; $$$$ , class I obesity to class II obesity; $$$$ , overweight or obesity with BMI increase without changing category; $$$$ , overweight or obesity with BMI decrease without changing category; $$$$ , without obesity risk) among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007

The median daily energy intake postpartum was 11 950 (IQR 9037, 15 347) kJ/d. The percentage of total energy intake arising from dietary carbohydrates was 63·6 % (sd 10·1 %); from proteins, 15·4 % (sd 2·9 %); and from fats, 23·6 % (sd 5·7 %). Median fibre intake was 42 (IQR 28·9, 55·4) g/d, with 77 % (95 % CI 72·7, 81·3 %) ingesting a quantity greater than or equal to the recommended minimum.

Figure 3 represents the FPCA centred on the fibre intake variable, showing the correlations between selected food items in the FFQ and total dietary fibre quantity. Beans were the food that showed the strongest correlation with total fibre in the diet. On another correlation level, bread and rice were found to significantly correlate with fibre in the diet, indicating a common Brazilian food pattern along with beans. Consumption of spices like onion, garlic, tomato and red pepper, food items typically included in the preparation of common Brazilian meals, were also significantly correlated with dietary fibre.

Fig. 3 Correlations of foods ( $$$$ , positive correlation; ○, negative correlation) with total fibre content of the diet: focused principal component analysis among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007. Legend: a1 = rice; a3 = beans; a7 = bread; c1 = lentils, c2 = lettuce, c3 = kale; c4 = cabbage, c5 = orange, c6 = banana, c7 = papaya, c8 = apple, d3 = mango, e1 = tomato, e2 = chayote, e7 = onion; e8 = garlic, e9 = red pepper; f1 = carrot, f2 = sugarbeet

Figure 4 represents the FPCA centred on the postpartum weight retention variable, showing the correlations between total energy value percentages of foods and maternal weight retention. Individual food items did not impact on postpartum weight retention. The only variable with a significant positive correlation was white bread. This figure shows food patterns similar to those of Fig. 3.

Fig. 4 Correlations of the percentage of total energy intake of given foods ( $$$$ , positive correlation; ○, negative correlation) with postpartum weight retention: focused principal component analysis among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007. Legend: a1 = rice; a3 = beans; a7 = bread; c1 = lentils, c2 = lettuce, c3 = kale; c4 = cabbage, c5 = orange, c6 = banana, c7 = papaya, c8 = apple, d3 = mango, e1 = tomato, e2 = chayote, e7 = onion; e8 = garlic, e9 = red pepper; f1 = carrot, f2 = sugarbeet

Table 2 shows the results for the association between inadequate fibre intake and obesity risk. In the unadjusted model, women who did not reach the daily recommended fibre intake had an obesity risk 30 % (RR = 1·30; 95 % CI 1·06, 1·60) higher than those who did reach the recommended fibre intake. When adjusted for maternal age, pre-pregnancy BMI and weight gain during pregnancy, the association between fibre intake and obesity risk remained significant (RR = 1·24; 95 % CI 1·05, 1·47).

Table 2 AssociationFootnote * between fibre intake and obesity risk in postpartum among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007

Ref., referent category.

Model 1: adjusted for to maternal age, pre-pregnancy BMI. Model 2: Model 1 + total gestational weight gain.

* Adjusted through Poisson regression models with robust variance.

According to Dietary Reference Intakes (Institute of Medicine (2005)(23) and American Dietetic Association (2008)(13)).

Discussion

The present study points to a high incidence of obesity risk, as defined by the evolution of maternal pre-pregnancy BMI through the 4th–6th month postpartum. Less than recommended consumption of fibre was associated with this risk. The only specific food item shown to be correlated with postpartum weight retention was white bread, consumption frequency of which was at least twice daily in more than 50 % of the sample. Beans were the food with the greatest daily fibre contribution to the diet, forming a typical Brazilian food pattern together with rice and white bread. Fruits and vegetables formed another consumption pattern. Although representing an important potential source of fibre, they were generally consumed in small quantities and thus showed only a weak correlation with the daily fibre ingestion in the present sample.

The high risk for future obesity occasioned by pregnancy is a cause for concern. Previous evidence shows that excessive weight gain, above the recommendations of the IOM, together with the failure to lose weight during the postpartum period, are important obesity predictors for women of reproductive age(Reference Siega-Riz, Viswanathan and Moos8, Reference Rooney and Schauberger38). Recent anthropometric data from the 2008–2009 Family Budget Study are alarming in the sense of demonstrating that weight excess is present in 48 % of the Brazilian female adult population, exceeding the frequency of weight deficit thirteen times. Obesity, present in 16·9 % of Brazilian women, corresponds to about one-third of these weight excess cases(39).

The literature indicates excessive gestational weight gain as the main weight retention predictor postpartum(Reference Olson, Strawderman and Hinton6, Reference Althuizen, van Poppel and Seidell7, Reference Rooney and Schauberger38, Reference Scholl, Hediger and Schall40Reference Phelan42). Nevertheless, some authors stress that the relationship between weight gain during pregnancy and posterior postpartum weight retention may be subject to the ‘part–whole’ bias, meaning that weight gain is a component of weight retention and pointing to a high correlation between these variables(Reference Lacerda and Leal2).

Weight retention after pregnancy may be physiologically more dangerous than weight changes in other life stages because weight excess retained postpartum seems to deposit preferentially in a central manner, increasing the risk of insulin resistance and of cardiovascular illness(Reference Gunderson, Murtaugh and Lewis43). Only one previous study reported a protective effect of fibre-rich foods in reducing the risk of postpartum weight retention(Reference Oken, Taveras and Popoola1). The present study showed similar results by demonstrating that inadequate fibre consumption increases obesity risk corresponding to an unfavourable BMI evolution postpartum.

An individual analysis of the foods consumed in the postpartum period showed that white bread was the only food significantly correlated with maternal weight retention. To our knowledge, no other study has indicated this association. It is important to underscore the effects of white bread on glycaemic changes, acting as a possible obesity cofactor(Reference van Bakel, Kaaks and Feskens44). Foods with a high glycaemic index, like white bread, appear to cause less satiety, more hunger after meals and increase daily energy intake, while foods with a high fibre content have been proposed to be more satiating, reduce hunger and decrease energy intake(Reference Rolls, Ello-Martin and Tohill45).

When assessing specific food items in relation to their fibre contribution in the sample's diet, beans showed the strongest association. This finding confirms previous evidence that indicates beans as a primary fibre source in the typical Brazilian diet among adults. Of note, beans are also an important source of Fe and, when associated with rice, combine to represent a source of high-quality plant protein, commonly referred to as the ‘perfect protein’(28).

This typical Brazilian eating pattern associated with fibre intake, identified in the present is study, gave the greatest fibre contribution to the postpartum diet, highlighting the minimal consumption of fruits and vegetables in this low-income population. Both national and international studies underscore the association between low income and the high consumption of beans, along with low consumption of fruits and vegetables(28, Reference Monteiro, Mondini and Costa46Reference Mitchell, Lawrence and Hartman49). Nevertheless, our findings do not allow comparisons between the reported patterns and different income ranges, due to the homogeneity of our sample regarding income.

Among the limitations of the present study we can mention the inclusion of adolescent pregnant women, which may reduce the estimates’ validity since these women are still going through a developmental stage. The lack of data on body fat percentage and physical activity is another limitation since these are relevant factors associated with postpartum weight retention(Reference Oken, Taveras and Popoola1, Reference Amorim, Lacerda and Kac50). Regarding the assessment of food consumption, the limited ability of the FFQ to estimate real intake, as characterized in a relative validity study comparing FFQ and 24 h dietary recalls(Reference Giacomello, Schmidt and Nunes29), is another limitation. Furthermore, lactating women may overestimate food consumption(Reference Forsythe and Gage51Reference Baer, Blum and Rockett53). Another limitation is the use of reported, rather than measured, pre-pregnancy weight. However, studies indicate a high correlation between referred and measured weight(Reference Gunderson, Abrams and Selvin4, Reference Abrams, Carmichael and Selvin54Reference Schmidt, Duncan and Tavares57).

Keeping in mind the above mentioned limitations, the present study, the first to our knowledge in the Brazilian context, suggests that protection against maternal weight excess in the 4th–6th month after childbirth may be viable by reaching the daily values of recommended fibre intake, choosing the typical Brazilian dietary pattern that is characterized by a high fibre content, especially due to beans consumption. It is important to note that the current belief (built up mainly by the media) that a healthy diet is expensive and necessarily quite different from the usual one, is mistaken(58).

The risk of obesity in women in reproductive age is concerning. If dietary fibre can be a modifiable factor related to weight control postpartum, then public policies focusing on interventions for this specific population should be put into practice. Commitment of the government and of the public sector is essential to encourage, support and protect healthy eating. The publication of an addendum in the Food Guide for the Brazilian Population, including a specific topic with recommendations regarding pregnancy and postpartum, is of utmost importance, since this stage of life represents a good opportunity for nutritional education(Reference Phelan42). Maternal nutrition not only influences birth outcomes but ultimately impacts on the long-term health status of the offspring(Reference Oken and Gillman59Reference Salsberry and Reagan61).

Conclusions

Obesity risk related to weight gain or retention in the postpartum period was shown to be high in this sample of southern Brazilian women and represents a major public health concern given the ongoing obesity epidemic. Fibre intake appears to reduce the risk of an even worse situation. In this population, the main fibre source is associated with the typical Brazilian meal, especially with bean consumption. In this sense, the revival of traditional food practices as well as the consumption of healthy regional food, considering the cultural identity, behavioural and affective aspects, should be encouraged.

Acknowledgments

Sources of funding: The research was financed by the Programa de Apoio a Núcleos de Excelência (PRONEX) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES). Conflicts of interest: There is no financial conflict of interest for all authors. The authors have no undisclosed financial or personal interest in any company or organization connected in any way with the research represented herein. Authors’ contributions: The data reported in this paper were collected as part of the first author's PhD thesis that was supervised by M.I.S. and S.A.C. All authors were responsible for the design and performance of the ECCAGE Study. M.A.N. designed the research (project conception, development of overall research plan, study oversight), interpretation of data and critical review of the manuscript. M.L. contributed to analysis and interpretation of the data and preparation of the manuscript. B.B.D. and A.P.P. critically reviewed the manuscript. All authors have read and approved the final manuscript.

References

1.Oken, E, Taveras, EM, Popoola, FAet al. (2007) Television, walking, and diet: associations with postpartum weight retention. Am J Prev Med 32, 305311.CrossRefGoogle ScholarPubMed
2.Lacerda, E & Leal, MC (2004) Risk factors associated with postpartum weight gain and retention: a systematic review. Rev Bras Epidemiol 7, 187200.CrossRefGoogle Scholar
3.Kac, G, Benicio, MH, Velasquez-Melendez, Get al. (2004) Gestational weight gain and prepregnancy weight influence postpartum weight retention in a cohort of Brazilian women. J Nutr 134, 661666.CrossRefGoogle Scholar
4.Gunderson, EP, Abrams, B & Selvin, S (2000) The relative importance of gestational gain and maternal characteristics associated with the risk of becoming overweight after pregnancy. Int J Obes Relat Metab Disord 24, 16601668.CrossRefGoogle Scholar
5.Ohlin, A & Rossner, S (1994) Trends in eating patterns, physical activity and socio-demographic factors in relation to postpartum body weight development. Br J Nutr 71, 457470.CrossRefGoogle ScholarPubMed
6.Olson, CM, Strawderman, MS, Hinton, PSet al. (2003) Gestational weight gain and postpartum behaviors associated with weight change from early pregnancy to 1 y postpartum. Int J Obes Relat Metab Disord 27, 117127.CrossRefGoogle ScholarPubMed
7.Althuizen, E, van Poppel, MN, Seidell, JCet al. (2006) Design of the New Life(style) study: a randomised controlled trial to optimise maternal weight development during pregnancy. BMC Public Health 6, 168.CrossRefGoogle ScholarPubMed
8.Siega-Riz, AM, Viswanathan, M, Moos, MKet al. (2009) A systematic review of outcomes of maternal weight gain according to the Institute of Medicine recommendations: birthweight, fetal growth, and postpartum weight retention. Am J Obstet Gynecol 201, 339.e1339.e14.CrossRefGoogle Scholar
9.Rasmussen, KM & Yaktine, AL (editors) (2009) Weight Gain During Pregnancy: Reexamining the Guidelines. Washington, DC: National Academies Press.Google Scholar
10.Castro, MBT, Kac, G & Sichieri, R (2009) Determinantes nutricionais e sóciodemográficos da variação do peso no pós-parto: uma revisão da literatura. Rev Bras Saude Matern Infant 9, 125137.CrossRefGoogle Scholar
11.Linne, Y & Rossner, S (2003) Interrelationships between weight development and weight retention in subsequent pregnancies: the SPAWN study. Acta Obstet Gynecol Scand 82, 318325.CrossRefGoogle ScholarPubMed
12.Kinnunen, TI, Pasanen, M, Aittasalo, Met al. (2007) Reducing postpartum weight retention – a pilot trial in primary health care. Nutr J 6, 21.CrossRefGoogle ScholarPubMed
13.American Dietetic Association (2008) Position of the American Dietetic Association: nutrition and lifestyle for a healthy pregnancy outcome. J Am Diet Assoc 108, 553561.CrossRefGoogle Scholar
14.Subcommittee for a Clinical Applications Guide, Institute of Medicine, National Academy of Sciences (1992) Nutrition During Pregnancy and Lactation: An Implementation Guide. Washington, DC: National Academies Press.Google Scholar
15.Kac, G, Benicio, MH, Velasquez-Melendez, Get al. (2004) Nine months postpartum weight retention predictors for Brazilian women. Public Health Nutr 7, 621628.CrossRefGoogle ScholarPubMed
16.Gunderson, EP & Abrams, B (1999) Epidemiology of gestational weight gain and body weight changes after pregnancy. Epidemiol Rev 21, 261275.CrossRefGoogle Scholar
17.Gunderson, EP (2009) Childbearing and obesity in women: weight before, during, and after pregnancy. Obstet Gynecol Clin North Am 36, 317332.CrossRefGoogle Scholar
18.Walker, LO, Sterling, BS & Timmerman, GM (2005) Retention of pregnancy-related weight in the early postpartum period: implications for women's health services. J Obstet Gynecol Neonatal Nurs 34, 418427.CrossRefGoogle ScholarPubMed
19.Castro, MBT, Kac, G, de Leon, APet al. (2009) High-protein diet promotes a moderate postpartum weight loss in a prospective cohort of Brazilian women. Nutrition 25, 11201128.CrossRefGoogle Scholar
20.Lindstrom, J, Peltonen, M, Eriksson, JGet al. (2006) High-fibre, low-fat diet predicts long-term weight loss and decreased type 2 diabetes risk: the Finnish Diabetes Prevention Study. Diabetologia 49, 912920.CrossRefGoogle ScholarPubMed
21.American Dietetic Association (1997) Position of the American Dietetic Association: health implications of dietary fiber. J Am Diet Assoc 97, 11571159.CrossRefGoogle Scholar
22.Bhargava, A (2006) Fiber intakes and anthropometric measures are predictors of circulating hormone, triglyceride, and cholesterol concentrations in the women's health trial. J Nutr 136, 22492254.CrossRefGoogle ScholarPubMed
23.Institute of Medicine, Food and Nutrition Board (2005) Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington, DC: The National Academies Press.Google Scholar
24.World Health Organization (2003) Diet, Nutrition and the Prevention of Chronic Disease. Joint FAO/WHO Expert Consultation. WHO Technical Report Series no. 916. Geneva: WHO.Google Scholar
25.Swinburn, BA, Caterson, I, Seidell, JCet al. (2004) Diet, nutrition and the prevention of excess weight gain and obesity. Public Health Nutr 7, 123146.Google ScholarPubMed
26.Nunes, MA, Ferri, C, Manzoli, Pet al. (2010) Nutrition, mental health and violence: from pregnancy to postpartum cohort of women attending primary care units in southern Brazil – SCDBP. BMC Psychol 10, 66.Google Scholar
27.Buss, C, Nunes, MA, Camey, Set al. (2009) Dietary fibre intake of pregnant women attending general practices in southern Brazil – the SCDBP Study. Public Health Nutr 12, 13921398.CrossRefGoogle Scholar
28.Ministério da Saúde (2006) Guia Alimentar para a População Brasileira. Brasília: Ministério da Saúde.Google Scholar
29.Giacomello, A, Schmidt, MI, Nunes, MAAet al. (2008) Validação relativa de Questionário de frequência alimentar em gestantes usuárias de serviços do Sistema Único de Saúde em dois municípios no Rio Grande do Sul, Brasil. Rev Bras Saude Matern Infant 8, 445454.CrossRefGoogle Scholar
30.Zulkifli, SN & Yu, SM (1992) The food frequency method for dietary assessment. J Am Diet Assoc 92, 681685.CrossRefGoogle ScholarPubMed
31.Willett, W (1998) Nutritional Epidemiology, 2nd ed. New York: Oxford University Press.CrossRefGoogle Scholar
32.Pinheiro, A (2004) Tabela para avaliação de consumo alimentar em medidas caseiras. São Paulo: Atheneu.Google Scholar
33.Department de Foods e Nutrition Experimental da Faculdade de Ciências Farmacêuticas – USP (2007) Tabela Brasileira de Composição de Alimentos. http://www.fcf.usp.br/tabela (accessed February 2012).Google Scholar
34.Philippi, ST (2002) Tabela de Composição de Alimentos: Suporte para Decisão Nutricional, 2 ed. São Paulo: Coronário.Google Scholar
35.Drehmer, M, Camey, S, Schmidt, MIet al. (2010) Socioeconomic, demographic and nutritional factors associated with maternal weight gain in general practices in Southern Brazil. Cad Saude Publica 26, 10241034.CrossRefGoogle ScholarPubMed
36.Falissard, B, Corruble, E, Mallet, Let al. (2001) Focused principal component analysis: a promising approach for confirming findings of exploratory analysis? Int J Methods Psychiatr Res 10, 191195.CrossRefGoogle Scholar
37.Canuto, R, Camey, S, Gigante, Det al. (2010) Focused principal component analysis: a graphical method for exploring dietary patterns. Cad Saude Publica 26, 21492156.CrossRefGoogle ScholarPubMed
38.Rooney, BL & Schauberger, CW (2002) Excess pregnancy weight gain and long-term obesity: one decade later. Obstet Gynecol 100, 245252.Google ScholarPubMed
39.Ministério da Saúde & Ministério do Planejamento, Orçamento e Gestão (2010) Pesquisa de Orçamentos Familiares 2008–2009. Antropometria e estado nutricional de crianças, adolescentes e adultos do Brasil. Rio de Janeiro: Instituto Brasileiro de Geografia e Estatística.Google Scholar
40.Scholl, TO, Hediger, ML, Schall, JIet al. (1995) Gestational weight gain, pregnancy outcome, and postpartum weight retention. Obstet Gynecol 86, 423427.CrossRefGoogle ScholarPubMed
41.Gunderson, EP & Abrams, B (1999) Epidemiology of gestational weight gain and body weight changes after pregnancy. Epidemiol Rev 21, 261275.CrossRefGoogle Scholar
42.Phelan, S (2010) Pregnancy: a teachable moment for weight control and obesity prevention. Am J Obstet Gynecol 202, e1e8.CrossRefGoogle ScholarPubMed
43.Gunderson, EP, Murtaugh, MA, Lewis, CEet al. (2004) Excess gains in weight and waist circumference associated with childbearing: The Coronary Artery Risk Development in Young Adults Study (CARDIA). Int J Obes Relat Metab Disord 28, 525535.CrossRefGoogle ScholarPubMed
44.van Bakel, MME, Kaaks, R, Feskens, EJMet al. (2009) Dietary glycaemic index and glycaemic load in the European Prospective Investigation into Cancer and Nutrition. Eur J Clin Nutr 63, Suppl. 4, S188S205.CrossRefGoogle ScholarPubMed
45.Rolls, JB, Ello-Martin, JA & Tohill, BC (2004) What can intervention studies tell us about the relationship between fruit and vegetable consumption and weight management? Nutr Rev 62, 117.CrossRefGoogle ScholarPubMed
46.Monteiro, CA, Mondini, L & Costa, RB (2000) Changes in composition and appropriate nutrition of family diet in the metropolitan areas of Brazil (1988–1996). Rev Saude Publica 34, 251258.CrossRefGoogle ScholarPubMed
47.Ministério da Saúde (2009) Indicadores de Vigilância Alimentar e Nutricional Brasil 2006. Série B. Textos Básicos de Saúde. Brasília: Ministério da Saúde.Google Scholar
48.Caire-Juvera, G, Ortega, MI, Casanueva, Eet al. (2007) Food components and dietary patterns of two different groups of Mexican lactating women. J Am Coll Nutr 26, 156162.CrossRefGoogle ScholarPubMed
49.Mitchell, DC, Lawrence, FR, Hartman, TJet al. (2009) Consumption of dry beans, peas, and lentils could improve diet quality in the US population. J Am Diet Assoc 109, 909913.CrossRefGoogle ScholarPubMed
50.Amorim, AR, Lacerda, E & Kac, G (2007) Uso e interpretação dos indicadores antropométricos na avaliação do Estado Nutricional de Gestantes. In Epidemiologia Nutricional, pp. 3147 [G Kac, R Sichieri and DP Gigante editors]. Rio de Janeiro: Atheneu.Google Scholar
51.Forsythe, HE & Gage, B (1994) Use of a multicultural food-frequency questionnaire with pregnant and lactating women. Am J Clin Nutr 59, 1 Suppl., 203S206S.CrossRefGoogle ScholarPubMed
52.Erkkola, M, Karppinen, M, Javanainen, Jet al. (2001) Validity and reproducibility of a food frequency questionnaire for pregnant Finnish women. Am J Epidemiol 154, 466476.CrossRefGoogle ScholarPubMed
53.Baer, HJ, Blum, RE, Rockett, HRet al. (2005) Use of a food frequency questionnaire in American Indian and Caucasian pregnant women: a validation study. BMC Public Health 5, 135.CrossRefGoogle ScholarPubMed
54.Abrams, B, Carmichael, S & Selvin, S (1995) Factors associated with the pattern of maternal weight gain during pregnancy. Obstet Gynecol 86, 170176.CrossRefGoogle ScholarPubMed
55.Kleinman, KP, Oken, E, Radesky, JSet al. (2007) How should gestational weight gain be assessed? A comparison of existing methods and a novel method, area under the weight gain curve. Int J Epidemiol 36, 12751282.CrossRefGoogle Scholar
56.Nucci, L, Duncan, B, Mengue, Set al. (2001) Assessment of weight gain during pregnancy in general prenatal care services in Brazil. Cad Saude Publica 17, 13671374.CrossRefGoogle Scholar
57.Schmidt, MI, Duncan, BB, Tavares, Met al. (1993) Validity of self-reported weight – a study of urban Brazilian adults. Rev Saude Publica 27, 271276.CrossRefGoogle Scholar
58.Ministério da Saúde, Coordenação-Geral da Política de Alimentação e Nutrição (2005) O que é uma alimentação saudável? Considerações sobre o conceito, princípios e características: uma abordagem ampliada. Brasília: Ministério da Saúde.Google Scholar
59.Oken, E & Gillman, MW (2003) Fetal origins of obesity. Obes Res 11, 496506.CrossRefGoogle ScholarPubMed
60.Oken, E, Taveras, EM, Kleinman, KPet al. (2007) Gestational weight gain and child adiposity at age 3 years. Am J Obstet Gynecol 196, e1e8.CrossRefGoogle ScholarPubMed
61.Salsberry, PJ & Reagan, PB (2005) Dynamics of early childhood overweight. Pediatrics 116, 13291338.CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1 Flowchart showing participants in the study. *Two excluded women were simultaneously without weight retention data and consumed >25 104 kJ/d or >16 portions of fruits and vegetables/d; four excluded women simultaneously consumed >25 104 kJ/d and >16 portions of fruits and vegetables/d

Figure 1

Table 1 Sociodemographic, behavioural and maternal nutrition factors in the postpartum period associated with obesity risk among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007

Figure 2

Fig. 2 Change from pre-pregnancy to postpartum in obesity risk category (in clockwise order: $$$$, normal weight to overweight or obesity; $$$$, overweight to obesity; $$$$, class I obesity to class II obesity; $$$$, overweight or obesity with BMI increase without changing category; $$$$, overweight or obesity with BMI decrease without changing category; $$$$, without obesity risk) among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007

Figure 3

Fig. 3 Correlations of foods ($$$$, positive correlation; ○, negative correlation) with total fibre content of the diet: focused principal component analysis among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007. Legend: a1 = rice; a3 = beans; a7 = bread; c1 = lentils, c2 = lettuce, c3 = kale; c4 = cabbage, c5 = orange, c6 = banana, c7 = papaya, c8 = apple, d3 = mango, e1 = tomato, e2 = chayote, e7 = onion; e8 = garlic, e9 = red pepper; f1 = carrot, f2 = sugarbeet

Figure 4

Fig. 4 Correlations of the percentage of total energy intake of given foods ($$$$, positive correlation; ○, negative correlation) with postpartum weight retention: focused principal component analysis among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007. Legend: a1 = rice; a3 = beans; a7 = bread; c1 = lentils, c2 = lettuce, c3 = kale; c4 = cabbage, c5 = orange, c6 = banana, c7 = papaya, c8 = apple, d3 = mango, e1 = tomato, e2 = chayote, e7 = onion; e8 = garlic, e9 = red pepper; f1 = carrot, f2 = sugarbeet

Figure 5

Table 2 Association* between fibre intake and obesity risk in postpartum among women of two cities in southern Brazil (n 370), ECCAGE cohort study, 2006–2007