Abstracts

Introduction

In the last decades, noncommunicable diseases (NCDs) have emerged as the leading cause of the global burden of disease ¹1. Institute for Health Metrics and Evaluation. Global Burden of Disease 2021: findings from the GBD 2021 Study. Seattle: Institute for Health Metrics and Evaluation; 2024.. Annually, NCDs result in approximately 17 million premature deaths worldwide, with cardiovascular diseases being the primary cause of mortality globally ²2. World Health Organization. Invisible numbers: the true extent of noncommunicable diseases and what to do about them. Geneva: World Health Organization; 2022.. Factors such as high blood pressure, plasma glucose levels, and LDL cholesterol were among the top 10 leading risk factors for disability-adjusted life years in 2021, with high blood pressure standing as the main contributor ¹1. Institute for Health Metrics and Evaluation. Global Burden of Disease 2021: findings from the GBD 2021 Study. Seattle: Institute for Health Metrics and Evaluation; 2024..

Breastfeeding has clear short-term (reducing morbidity and mortality from infectious diseases ³3. Horta BL, Victora CG. Short-term effects of breastfeeding: a systematic review on the benefits of breastfeeding on diarrhoea and pneumonia mortality. Geneva: World Health Organization; 2013.) and long-term benefits. In the long-term, breastfeeding would be positively associated with human capital ⁴4. Horta BL, Loret de Mola C, Victora CG. Breastfeeding and intelligence: a systematic review and meta-analysis. Acta Paediatr 2015; 104:14-9.^,55. Victora CG, Horta BL, Loret de Mola C, Quevedo L, Pinheiro RT, Gigante DP, et al. Association between breastfeeding and intelligence, educational attainment, and income at 30 years of age: a prospective birth cohort study from Brazil. Lancet Glob Health 2015; 3:e199-205.^,66. Brion MJ, Lawlor DA, Matijasevich A, Horta B, Anselmi L, Araújo CL, et al. What are the causal effects of breastfeeding on IQ, obesity and blood pressure? Evidence from comparing high-income with middle-income cohorts. Int J Epidemiol 2011; 40:670-80. and would decrease the risk of NCDs. A recently published meta-analysis reported that breastfeeding would protect against overweight/obesity (pooled odds ratio - pooled OR = 0.85 and 95% confidence interval - 95%CI: 0.77; 0.93) ⁷7. Horta BL, Rollins N, Dias MS, Garcez V, Pérez-Escamilla R. Systematic review and meta-analysis of breastfeeding and later overweight or obesity expands on previous study for World Health Organization. Acta Paediatr 2023; 112:34-41.. Moreover, breastfeeding would also decrease the odds of type 2 diabetes (pooled OR = 0.67, 95%CI: 0.56; 0.80) ⁸8. Horta BL, Lima NP. Breastfeeding and type 2 diabetes: systematic review and meta-analysis. Curr Diab Rep 2019; 19:1.. Furthermore, Lewandowski et al. ⁹9. Lewandowski AJ, Lamata P, Francis JM, Piechnik SK, Ferreira VM, Boardman H, et al. Breast milk consumption in preterm neonates and cardiac shape in adulthood. Pediatrics 2016; 138:e20160050. observed that preterm subjects who had been exclusively breastfed had increased left and right ventricular end-diastolic volume index, in relation to those who had not been breastfed. The plausibility of the association between breastfeeding and overweight/obesity is emphasized by the finding that breastfeeding moderates the association of the fat mass and obesity-associated (FTO) genotype with obesity ¹⁰10. Horta BL, Victora CG, França GVA, Hartwig FP, Ong KK, Rolfe EL, et al. Breastfeeding moderates FTO related adiposity: a birth cohort study with 30 years of follow-up. Sci Rep 2018; 8:2530..

On the other hand, most studies on the long-term consequences of breastfeeding are observational and have been conducted in high-income countries ¹¹11. Horta BL, Loret de Mola C, Victora CG. Long-term consequences of breastfeeding on cholesterol, obesity, systolic blood pressure and type 2 diabetes: a systematic review and meta-analysis. Acta Paediatr 2015; 104:30-7., in which socioeconomic status is positively associated with duration of breastfeeding ⁶6. Brion MJ, Lawlor DA, Matijasevich A, Horta B, Anselmi L, Araújo CL, et al. What are the causal effects of breastfeeding on IQ, obesity and blood pressure? Evidence from comparing high-income with middle-income cohorts. Int J Epidemiol 2011; 40:670-80.. Therefore, the observed associations could be due to residual confounding by socioeconomic status. In the 1982 Pelotas (Brazil) birth cohort, breastfeeding duration was not related to socioeconomic status ¹²12. Victora CG, Matijasevich A, Santos IS, Barros AJ, Horta BL, Barros FC. Breastfeeding and feeding patterns in three birth cohorts in Southern Brazil: trends and differentials. Cad Saúde Pública 2008; 24 Suppl 3:S409-16.. Consequently, studies using the cohort data provide evidence that are not susceptible to residual confounding.

This study aimed to assess the long-term association of duration of breastfeeding and metabolic cardiovascular risk factors at 30 years old (mean arterial pressure, carotid intima-media thickness, pulse wave velocity, glycemia, and total, HDL, and non-HDL cholesterol), using data from a birth cohort study conducted in Pelotas, a southern Brazilian city. The study hypothesis was: duration of breastfeeding was positively associated with HDL and inversely associated with the other cardiovascular risk factors evaluated.

Methods

This study used data from the 1982 Pelotas (Brazil) birth cohort. In 1982, the maternity hospitals in Pelotas were visited daily and all births recorded. Those live births (n = 5,914) whose family lived in the urban area of the city were examined and the mothers were interviewed. Initially, the cohort aimed to investigate perinatal, infant, and early childhood morbidity and mortality ¹³13. Victora CG, Barros FC. Cohort profile: the 1982 Pelotas (Brazil) birth cohort study. Int J Epidemiol 2006; 35:237-42.. Over time, the study expanded to include follow-ups into later stages of life, with cohort participants undergoing multiple assessments ¹³13. Victora CG, Barros FC. Cohort profile: the 1982 Pelotas (Brazil) birth cohort study. Int J Epidemiol 2006; 35:237-42.^,1414. Barros FC, Victora CG, Horta BL, Gigante DP. Metodologia do estudo da coorte de nascimentos de 1982 a 2004-5, Pelotas, RS. Rev Saúde Pública 2008; 42 Suppl 2:7-15.. In 2012-2013, we tried to locate all cohort members and invited the subjects to attend the research clinic to be interviewed and examined ¹⁵15. Horta BL, Gigante DP, Gonçalves H, Motta JS, Loret de Mola C, Oliveira IO, et al. Cohort profile update: the 1982 Pelotas (Brazil) Birth Cohort Study. Int J Epidemiol 2015; 44:441e..

The study protocol of the 1982 Pelotas cohort was approved by the Research Ethics Committee of the Faculty of Medicine of the Federal University of Pelotas (protocol number: Of. 16/12). Written informed consent was obtained from all participants.

Assessment of breastfeeding

Information on duration of breastfeeding and age at introduction of complementary feeding was gathered in the 12, 24 and 48 months of the cohort visits, and the earliest information on the age at which breastfeeding completely stopped was used. Duration of predominant breastfeeding was defined as the age when foods other than breast milk, water and tea were introduced. Considering that exclusive breastfeeding was an unusual practice at that time and had a short duration in the sample, it was included in the predominant breastfeeding group.

Outcomes assessment

At 30 years, metabolic cardiovascular risk factors were evaluated. Blood pressure was assessed twice on the left arm, with a specific cuff for individuals with obesity, using a digital sphygmomanometer (Omron HEM 705CPINT; https://www.omron.com/). The measurement was made at the beginning and the end of the anthropometric evaluation, and the mean of these measures was used in the analyses. For participants who were taking antihypertensive drugs, assuming a linear blood pressure reduction due to antihypertensive treatment, 8 and 5mmHg were added to the values of systolic and diastolic blood pressure, respectively, for each additional drug ¹⁶16. Law MR, Wald NJ, Morris JK, Jordan RE. Value of low dose combination treatment with blood pressure lowering drugs: analysis of 354 randomised trials. BMJ 2003; 326:1427.^,1717. Wald DS, Law M, Morris JK, Bestwick JP, Wald NJ. Combination therapy versus monotherapy in reducing blood pressure: meta-analysis on 11,000 participants from 42 trials. Am J Med 2009; 122:290-300.. Afterwards, mean arterial pressure was computed by adding double of the diastolic pressure to the systolic number and then dividing that value by three. Carotid intima-media thickness was assessed at the posterior wall of right and left common carotid in longitudinal planes using a Toshiba Xario ultrasound (https://www.toshiba.com/tic/industries-served/medical). A 10mm-long section of the common carotid artery was imaged proximal to carotid bulb ¹⁸18. Touboul PJ, Hennerici MG, Meairs S, Adams H, Amarenco P, Bornstein N, et al. Mannheim carotid intima-media thickness and plaque consensus (2004-2006-2011). An update on behalf of the advisory board of the 3rd, 4th and 5th watching the risk symposia, at the 13th, 15th and 20th European Stroke Conferences, Mannheim, Germany, 2004, Brussels, Belgium, 2006, and Hamburg, Germany, 2011. Cerebrovasc Dis 2012; 34:290-6.. The Carotid Analyzer for Research (Medical Imaging Applications, MIA-LLC; https://mia-llc.com/) automatically estimated the mean value of 90 measurements (frames) taken in the 10mm long section studied. Carotid-femoral pulse wave velocity was measured twice using a portable ultrasound, Sphygmocor (Atcor Medical; https://atcormedical.com/) in the supine position. The distance of pulse wave transit was measured with a flexible tape as the distance from suprasternal notch to femoral and carotid application point of the tonometer. Pulse wave velocity was estimated as the distance between the measurement sites divided by transit time delay between femoral and carotid pulse wave. The mean of the measurements was used in the analysis. Random blood glucose and total and HDL cholesterol were obtained using an automatic enzymatic colorimetric method in a chemistry analyzer (BS-380, Shenzhen Mindray Bio-Medical Electronics Co.; https://www.mindray.com/en). Non-HDL cholesterol was computed by subtracting HDL from total cholesterol.

Statistical analysis

Statistical analyses were performed using Stata, version 15 (https://www.stata.com/). Means were compared using analysis of variance (ANOVA), and the chi-square test was used to compare proportions. Simple and multiple linear regressions were used to evaluate the association between breastfeeding and cardiometabolic outcomes. In the linear regression, the normality of residuals and homoscedasticity were graphically tested. Multivariable models were adjusted for the following confounding factors collected at the perinatal study, informed by previous findings in the literature and theoretical knowledge: sex; family income (≤ 1; 1.1-3; 3.1-6; 6.1-10; > 10 minimum wages); maternal schooling (0-4; 5-8; 9-11; ≥ 12 years); maternal smoking in the pregnancy (yes; no); maternal hypertension or diabetes during pregnancy; maternal age; mode of delivery (vaginal; cesarean), maternal body mass index (BMI) before pregnancy; gestational age, and birthweight -measured soon after childbirth by hospital staff. Furthermore, the models were controlled for household asset index at 24 months, calculated via factor analysis of a list of household durable goods, and European genomic ancestry proportion, based on the analysis of approximately 370,000 single nucleotide polymorphisms (SNPs) accessible for samples of the Pelotas cohort, the HapMap Project, the Human Genome Diversity Project ¹⁹19. Kehdy FS, Gouveia MH, Machado M, Magalhães WC, Horimoto AR, Horta BL, et al. Origin and dynamics of admixture in Brazilians and its effect on the pattern of deleterious mutations. Proc Natl Acad Sci U S A 2015; 112:8696-701.. For mean arterial pressure, analyses were also adjusted for use of oral contraceptive and height at 30 years, whereas for glycemia, time since last meal was also included in the multivariable model. Significance level was set at 0.05. For ordinal data, both heterogeneity and linear trend were tested, showing the one with the lowest p-value.

Results

In 2012-2013, we managed to interview 3,701 participants, averaging 30.2 years old, which added to the 325 deaths identified among the cohort members, represented a follow-up rate of 68.1%. Participants were slightly more likely to be female and from intermediate socioeconomic groups compared to the original cohort; although these differences were statistically significant, they were small between the groups, being at most 6.5 percentage points. Regarding breastfeeding duration, there was no significant difference in the follow-up (Table 1).

Table 2 shows that about seven out of 10 participants had a family income at birth ≤ 3 minimum wages. Maternal schooling was heterogeneous, whereas about one of every three mothers had ≤ 4 years of schooling, 13.8% had completed middle school (≥ 12 years of schooling). Regarding breastfeeding, 21.1% of the participants were breastfed for less than one month and 30% breastfed for six or more months. Duration of predominant breastfeeding was short, and only about four out of 10 participants were predominantly breastfed for three months or more. At 30 years, 57.7% of the participants were overweight, mean arterial pressure was 90.8mmHg, and mean total cholesterol was 191.3mg/dL.

Table 3 shows that family income at birth was not associated with the prevalence of breastfeeding in the first six months of life. Maternal age, birthweight, and maternal BMI before pregnancy were positively associated with the prevalence of breastfeeding in the first six months. Nonetheless, offspring of mothers who smoked in the pregnancy and those born by cesarean section were less likely of being breastfed in the first six months of life. Mean arterial pressure was positively associated with maternal BMI before pregnancy and hypertension or diabetes during pregnancy. Moreover, carotid intima-media thickness was lower among females, inversely associated with socioeconomic status and positively related to maternal BMI and birthweight, whereas pulse wave velocity was positively associated with maternal BMI before pregnancy. Table 4 shows that glycemia, total and non-HDL cholesterol were lower among females. Blood glucose was lower among participants whose mother did not smoke during pregnancy. Total and HDL cholesterol were directly associated with socioeconomic status at birth and were lower in those with low birthweight.

Table 5 shows that mean arterial pressure, carotid intima-media thickness, and pulse wave velocity did not show any clear association with infant feeding in the crude and adjusted analyses. Table 6 shows that HDL cholesterol was positively related to predominant breastfeeding in the crude analysis, but the association disappeared after controlling for confounding variables. On the other hand, total and non-HDL cholesterol and random blood glucose were not related to infant feeding.

Discussion

In this population-based birth cohort, duration of total and predominant breastfeeding was not associated to metabolic cardiovascular risk factors (mean arterial pressure, carotid intima-media thickness, pulse wave velocity, glycemia, and blood lipids) at 30 years.

Meta-analyses of observational and randomized studies have shown that breastfed participants had lower systolic blood pressure ¹¹11. Horta BL, Loret de Mola C, Victora CG. Long-term consequences of breastfeeding on cholesterol, obesity, systolic blood pressure and type 2 diabetes: a systematic review and meta-analysis. Acta Paediatr 2015; 104:30-7. and were less likely to have type 2 diabetes ⁸8. Horta BL, Lima NP. Breastfeeding and type 2 diabetes: systematic review and meta-analysis. Curr Diab Rep 2019; 19:1.^,1111. Horta BL, Loret de Mola C, Victora CG. Long-term consequences of breastfeeding on cholesterol, obesity, systolic blood pressure and type 2 diabetes: a systematic review and meta-analysis. Acta Paediatr 2015; 104:30-7., whereas total cholesterol was not associated with breastfeeding ¹¹11. Horta BL, Loret de Mola C, Victora CG. Long-term consequences of breastfeeding on cholesterol, obesity, systolic blood pressure and type 2 diabetes: a systematic review and meta-analysis. Acta Paediatr 2015; 104:30-7.. Most studies on the long-term consequence of breastfeeding were observational and conducted in high-income countries, a scenario in which high socioeconomic status is associated with longer duration of breastfeeding ⁶6. Brion MJ, Lawlor DA, Matijasevich A, Horta B, Anselmi L, Araújo CL, et al. What are the causal effects of breastfeeding on IQ, obesity and blood pressure? Evidence from comparing high-income with middle-income cohorts. Int J Epidemiol 2011; 40:670-80., and residual confounding could have overestimated the magnitude of the associations.

Indeed, findings from the Promotion of Breastfeeding Intervention Trial (PROBIT) ²⁰20. Kramer MS, Matush L, Vanilovich I, Platt RW, Bogdanovich N, Sevkovskaya Z, et al. Effects of prolonged and exclusive breastfeeding on child height, weight, adiposity, and blood pressure at age 6.5 y: evidence from a large randomized trial. Am J Clin Nutr 2007; 86:1717-21. - in which exchangeability between the comparison groups is expected - indicated that allocation to breastfeeding promotion group was not associated to blood pressure and BMI at a mean age of 6.5 years. The null result of the mentioned study could be considered as an indication that the association of breastfeeding with BMI and blood pressure is not causal. Still, compliance to the intervention group was low. Among the intervention group, 49.8% of the children were breastfed in the first six months of life, whereas this proportion was 36.1% among the control. Because noncompliance to allocated group tends to decrease the statistical power of analyses, the null results of the mentioned study should not be considered as proof of absence of association.

A potential mechanism linking breastfeeding to cardiovascular risk factors involves its association with obesity, which is mediated by breastfeeding effect on the infant gut microbiota ²¹21. Petraroli M, Castellone E, Patianna V, Esposito S. Gut microbiota and obesity in adults and children: the state of the art. Front Pediatr 2021; 9:657020., promotion of healthier dietary preferences ²²22. Mennella JA. Ontogeny of taste preferences: basic biology and implications for health. Am J Clin Nutr 2014; 99:704S-711S., and influence on satiety regulation ²³23. Brown A, Lee M. Breastfeeding during the first year promotes satiety responsiveness in children aged 18-24 months. Pediatr Obes 2012; 7:382-90.. A previous study conducted in the same population showed that breastfeeding moderated the association between the fat mass and obesity-associated genotype variant rs9939609 and adiposity measures, being inversely associated with visceral fat thickness ¹⁰10. Horta BL, Victora CG, França GVA, Hartwig FP, Ong KK, Rolfe EL, et al. Breastfeeding moderates FTO related adiposity: a birth cohort study with 30 years of follow-up. Sci Rep 2018; 8:2530.. Therefore, regarding the relation between visceral fat and metabolic measures, we expected breastfeeding could be a protective factor for the cardiometabolic outcomes assessed, but the great increase of overweight and obesity prevalence may have mitigated the protective effect ²⁴24. Lima NP, Horta BL, Motta JV, Valença MS, Oliveira V, Santos TV, et al. Evolução do excesso de peso e obesidade até a idade adulta, Pelotas, Rio Grande do Sul, Brasil, 1982-2012. Cad Saúde Pública 2015; 31:2017-25..

This study has two important strengths: we were able to evaluate the long-term association of breastfeeding with several cardiometabolic factors in a large population-based cohort with no clear social pattern of breastfeeding, and adjusted the analysis for different socioeconomic variables, reducing the possibility of residual confounding. Moreover, selection bias is unlikely due to the low attrition rate at 30 years, the nondifferential follow-up rates regarding breastfeeding duration, and the reasonably similar follow-up rates for baseline characteristics (Table 1). Information was prospectively collected by trained fieldworkers and the metabolic measures were assessed with high standard instruments, following standardized procedures, which minimize measurement bias. As a limitation, we were not able to separately evaluate individuals who were never breastfed. Because of the considerably low frequency of this category, it was grouped with < 1 month of breastfeeding. Furthermore, glycemia was evaluated using non-fasting blood glucose and it may have introduced a nondifferential misclassification even after adjusting for time since last meal.

Concisely, our study suggests no association between breastfeeding duration and cardiometabolic risk factors in adulthood. While no significant links were found with cardiovascular indicators at 30 years old, it is central to understand that breastfeeding plays an important role in reducing infant and child mortality. Additionally, its positive association with human capital development is noteworthy, as individuals who experience breastfeeding tend to show higher intelligence quotient scores, better school performance, and increased income in adulthood. Therefore, the diverse benefits derived from breastfeeding underscore the critical importance of promoting and supporting this practice. Our findings do not diminish the overall significance of breastfeeding; instead, they emphasize the need for a comprehensive understanding of the multifaceted advantages.

Acknowledgments

This article is based on data from the 1982 Pelotas Birth Cohort Study, conducted by the Post-Graduation Program in Epidemiology at Federal University of Pelotas with the collaboration of the Brazilian Public Health Association (ABRASCO), Wellcome Trust, International Development Research Center, World Health Organization, Overseas Development Administration, Brazilian National Program for Centers of Excellence (PRONEX), the Brazilian National Research Council (CNPq), and Brazilian Ministry of Health.

References

Publication Dates

History