Oral Session
Maternal, Perinatal and Pediatric Nutrition
Allison Daniel, MSc
PhD Candidate
The Hospital for Sick Children
Sara Shama, BSc
MSc Student
Hospital for Sick Children
Celine Bourdon, MSc
Project Manager
Hospital for Sick Children
Alex Kiss, PhD
Scientist
Sunnybrook Research Institute
Martha Mwangome, PhD
Research Scientist
KEMRI-Wellcome Trust Research Programme
Robert Bandsma, MD, PhD
Hospital for Sick Children, Canada & University of Toronto, Canada
Deborah O'Connor, PhD, RD
Senior Associate Scientist
Hospital for Sick Children
Objectives : Given increased incidence of maternal postpartum obesity and limited understanding of the degree to which maternal body composition is related to energy and macronutrient content of breastmilk, we conducted a systematic review to assess the associations between maternal body mass index (BMI) and breastmilk energy, fat, and/or total protein content.
Methods : A protocol was registered prior to the completion of this systematic review (PROSPERO 2018 CRD42018098808). The electronic databases MEDLINE, Embase, CINAHL, and Web of Science were searched. The three outcomes of interest were breastmilk energy (kcal/dL), fat (g/dL), and total protein (g/dL) from mothers of term-born infants. Articles that reported maternal BMI or weight and height, collected breastmilk, and measured energy, fat, or protein concentration in breastmilk one to six months postpartum were eligible. If breastmilk was collected at multiple timepoints, the measures closest to three months postpartum were used. A simulated dataset was generated in Stata 14 based on respective sample sizes and means and variance of maternal BMI and breastmilk energy, fat, and protein content of the included studies, assuming normality of each of these variables based on the literature. Linear regressions were performed to examine associations between maternal BMI (kg/m2) and breastmilk energy and individual macronutrients.
Results : A total of 2,012 articles were identified from the search and reference lists of included studies. After full-text screening by two reviewers, 39 articles were included with breastmilk samples from 2,837 mothers. Based on the simulated data, maternal BMI was not significantly associated with breastmilk energy concentration (β=-0.0044, 95% CI [-0.14, 0.14], p >0.9). However, BMI was positively associated with breastmilk fat concentration (β=0.015, 95% CI [0.0022, 0.027] p=0.02) and was negatively associated with total protein concentration (β=-0.0068, 95% CI [-0.011, -0.0027], p=0.001). Thus, every 1kg/m2 increase in BMI is associated with a 0.0022g/dL increase in fat and a 0.011g/dL decrease in total protein concentration.
Conclusions : Statistically significant associations were found between maternal BMI and breastmilk fat and protein content. However, the effect size was small suggesting that these differences may not be clinically meaningful.
Funding Sources : N/A