• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

Q.31How does breastfeeding affect glycaemic control?

Bibliographic InformationStudy Type & Evidence LevelNumber of PatientsPatient CharacteristicsIntervention & ComparisonFollow-up & Outcome MeasuresEffect SizeStudy SummaryReviewer Comments
Saez de Ibarra, l., Gaspar, R., Obesso, A., Herranz, L.


Study Type: Cohort

Evidence level: 2+
36 women with type 1 diabetesType 1 diabetes, mean age 32 ±3 years, mean duration of diabetes 15±7 years. All women were on multiple dialy injections (3 to 4 per day) with short and intermediate acting insulins. Patients given instructions on maintaining good glycaemic control, including advice to eat before baby feeding times (or to have a glass of juice or milk if this is not possible).

Country: Spain
Intervention: Breastfeeding (Patients were asked to keep a daily record of breastfeeding hours and if juice or milk was taken before breastfeeding).

Comparison: Not breastfeeding
Follow-up period:

Outcome Measures:
Glycaemic control in four periods:
Preconception period, first seven days postpartum, first month postpartum, second month postpartum. Measured using glucose meters (4 times daily capillary blood glucose self monitoring).

Hypoglycaemic episodes (these were considered to be related to a breastfeeding session when it occurred during breastfeeding or up to 30 minutes after).
28 women breastfed in the first month. 24 women breastfed in the second month.

Mean glucose values were significantly lower during the first week postpartum (6.7±1.1 mmol/l) than at preconception (7.7±0.9 mmol/l) or during the second month postpartum (7.6±1.3 mmol/l). The percentage of glucose reading below 3 mmol/l did not differ during the four periods.

Insulin requirements were significantly lower during the first postpartum week (0.56±0.15U/kg/day) than the preconception period (0.68±0.16U/kg/day) and remained significantly lower over the first (0.56±0.15 U/kg/day) and second (0.56±0.11 U/kg/day) postpartum months.

Breastfeeding mothers had lower mean glucose levels during the first postpartum week (borderline significance, 6.6±0.6 vs 7.0±0.9 mmol/l; P = 0.050). There was no significant difference between breastfeeding and nonbreastfeeding mothers in mean glucose levels in the first or second months postpartum, in the mean standard deviation of blood glucose readings or in insulin dose.

The 13 women who kept records of breastfeeding hours had fewer hypoglycaemic episodes related to breastfeeding sessions (4.0±3.5) than unrelated to breastfeeding sessions (12.2±7.1, P = 0.002).
Insulin requirements after delivery and throughout the first two months postpartum are lower than before pregnancy. Glucose levels are lower during the first week postpartum. These changes apply both to breastfeeding and non- breastfeeding mothers.

Hypoglycaemic episodes do not occur more frequently during or immediately after nursing sessions.
The comparison between breastfeeding and bottlefeeding mothers may be limited by small numbers of bottlefeeding mothers.
Ferris, A., Dalidowitz, M., Ingardia, C. et al


Study Type: Cohort

Evidence level: 2+
30 (16 breastfeeding and 14 bottle feeding)Type 1 diabetes

Country: USA

Comparison: Bottle feeding
Follow-up period: 6 weeks

Outcome Measures:
Insulin dose Fasting blood glucose levels (mean)
Insulin doses did not differ between the two groups.

6-week postpartum mean fasting plasma glucose levels: Mothers who exclusively breastfed (n = 6) 4.6 ± 2.2 mmol/L
Stopped before 6 weeks (n = 9) 8.1±2.1 mmol/l
Bottle fed (n = 14) 6.7 ± 1.7 mmol/l
Mothers with type 1 diabetes who exclusively breastfed had lower fasting blood glucose levels by 6 weeks postpartum than mothers who stopped breastfeeding or mothers who bottle fed.Small numbers in subgroups. P values not given.
Bibliographic InformationStudy Type & Evidence LevelAim of StudyNumber of Patients & Patient CharacteristicsPopulation CharacteristicsOutcome measuresResults & CommentsStudy SummaryReviewer Comment
Davies, A., Clark, J., Dalton, K., Edwards, O.


Study Type:
OtherCase series

Evidence Level: 3
Intervention: Breastfeeding

Comparison: Not breastfeeding
24 insulin dependent women

Country: UK
Insulin requirementsBreastfeeding was established after 18 pregnancies and continued in 16 cases until the visit to the postnatal clinic six weeks later.

Mean (SD) insulin dose
In early pregnancy: 44(15)
At end of pregnancy: 96 (36)
Postpartum: 33 (13)
At 6 week postnatal clinic: 37(13)

In early pregnancy: 46(10)
At end of pregnancy: 72(8)
Postpartum 41(10)
At postnatal clinic 46(12)

The mean reduction in insulin dose in breastfeeding mothers from prepregnancy dose was 11.6 units (95% CI 8.9 to 14.3 units) and 5.2 units (1.1 to 9.3 units) in women who bottlefed.
The study is too small to show a difference in the mean reduction in insulin between breastfeeding and bottlefeeding mothers.
Bibliographic InformationStudy Type & Evidence LevelNumber of PatientsPatient CharacteristicsIntervention & ComparisonFollow-up & Outcome MeasuresEffect SizeStudy SummaryReviewer Comments
Briggs,G.G.; Ambrose,P.J.; Nageotte,M.P.; Padilla,G.; Wan,S.


Study Type:

Evidence level: 2−
7 women:

5 women with type 2 diabetes

2 controls (women without diabetes)
Five women with type 2 diabetes and two controls (women without diabetes) who were started on metformin on the first day after caesarean section

Country: USA
Intervention: Metformin use after caesarean section

Comparison: Women without diabetes who were also started on metformin after caesarean section
Follow-up period:
Not reported

Outcome Measures:
Metformin concentration in serum and milk

Blood glucose concentrations in the neonate
The mean peak and trough serum metformin concentrations were 1.06 μg/mL (range 0.68–1.90 μg/mL) and 0.42 μg/mL (range 0.26–0.51 μg/mL), respectively, whereas the mean peak and trough metformin concentrations in breast milk were 0.42 μg/mL (range 0.38–0.46 μg/mL) and 0.39 μg/mL (range 0.31–0.52 μg/mL), respectively.

The mean milk:serum ratio was 0.63 (range 0.36–1.00) and the mean estimated infant dose as a percentage of the mother's weight- adjusted dose was 0.65% (range 0.43–1.08%).

In 3 infants, the blood glucose concentrations 4 hours after feeding were within the normal limit, ranging from 47–77 mg/dL.
Metformin is excreted into breast milk, but the amounts seem to be clinically insignificant.
Feig,D.S.; Briggs,G.G.; Kraemer,J.M.; Ambrose,P.J.; Moskovitz,D.N.; Nageotte,M.; Donat,D.J.; Padilla,G.; Wan,S.; Klein,J.; Koren,G.


Study Type:

Evidence level: 2+
8 mothers with type 2 diabetes from Toronto.

13 mothers with type 2 diabetes from Long Beach
The study cohorts included women with type 2 diabetes who had recently delivered at Mount Sinai Hospital in Toronto, Canada or at the Women’s Pavilion, Miller Children’s Hospital in Long Beach, Canada and who were currently breast- feeding.

In Toronto, the women were put on insulin therapy during pregnancy and continued using insulin after delivery.

In Long Beach, the women were treated with insulin during pregnancy and were changed to nonmicronised glyburide (5 mg/day) or immediate-release glipizide (5 mg/day) immediately after delivery by their personal physicians independently of the study.

Country: Canada
Intervention: Glyburide and glipizide.

Comparison: Standard breast milk or serum concentrations of glyburide ng/ml) and a fixed concentration of tolbutamide for breast milk and serum).
Follow-up period:
Two years.

Outcome Measures:
Detection of glyburide and glipizide in breast milk.
Neither glyburide nor glipizide were detected in the breast milk and blood glucose was normal in the three infants (one glyburide and two glipizide) who were wholly breast-fed when the drug concentrations were at steady state.

In the single-dose glyburide study, the mean maximum theoretical infant dose as a percent of the weight- adjusted maternal dose was < 1.5% and < 0.7% for the 5 mg and 10 mg doses, respectively.

The mean maximum theoretical infant dose as the percent of weight-adjusted maternal dose was < 28% and < 27% for the glyburide and glipizide subjects, respectively.
Neither glyburide nor glipizide were detected in breast milk, and hypoglycemia was not observed in the three nursing infants. Both glyburide and glipizide, at the doses tested, appear to be compatible with breast-feeding.
Plagemann A, Harder T, Franke K, Kohlhoff R.


Study Type: Cohort

Evidence level: 2+
112 children of women with diabetes (type 1 diabetes, n =83 and gestational diabetes, n = 29).Children born to women with diabetes who were delivered during 1980–1989 at the Clinic of Obstetrics and Gynecology, Berlin-Kaulsdorf.

They were part of the prospective cohort study on consequences of maternal diabetes during pregnancy for the children’s development.

Country: Germany
Intervention: Feeding on diabetic breast milk only.

  1. Feeding on some diabetic breast milk.
  2. Banked donor breast milk
Follow-up period: 2.5 years

Outcome Measures:
  1. Nutritional:
  2. Anthropometrical:
  3. Glucose levels
There was a positive correlation between the volume of diabetic breast milk ingested and risk of overweight at 2 years of age (OR 2.47, 95% CI 1.25 to 4.87).

In contrast, the volume of banked donor breast milk ingested was inversely correlated to body weight at follow-up (P = 0.001).

Risk of childhood impaired glucose tolerance decreased by increasing amounts of banked donor breast milk ingested neonatally (OR 0.19, 95% CI 0.05 to 0.70).

Stepwise regression analysis showed volume of diabetic breast milk to be the only significant predictor of relative body weight at 2 years of age (P = 0.001).
Early neonatal ingestion of breast milk from diabetic mothers may increase risk of becoming overweight and, consequently, developing impaired glucose tolerance during childhood.

Additional studies are needed to assess long-term consequences that might result from the type of neonatal nutrition in children of diabetic mothers.
This is a well designed and executed study taking into account potential confounding factors.

The sample size is reasonable and the likelihood of the role of chance is expected to be significantly small.
Rodekamp E;Harder T;Kohlhoff R;Franke K;Dudenhausen JW;Plagemann A;


Study Type: Cohort

Evidence level: 2+
317 children of women with type 1 diabetes.The participants of this study came from a population of 741 children of women with diabetes during pregnancy (type 1 diabetes and gestational diabetes) who delivered during 1980–1989 at the Clinic of Obstetrics and Gynecology, Berlin-Kaulsdorf.

They were part of the prospective cohort study on consequences of maternal diabetes during pregnancy for the children’s development.

A total of 112 infants (83 born to mothers with pre-existing diabetes and 29 with gestational diabetes) were followed-up with complete data on nutrition throughout the neonatal period, i.e., both the early neonatal period (1st neonatal week) and the late neonatal period (2nd–4th neonatal week), as well as data on duration of breast- feeding.

Gestational diabetes was diagnosed between the 26th and 28th gestational weeks.

Country: Germany
Intervention: Diabetic breast milk only.

  1. Some diabetic breast milk.
  2. No diabetic breast milk.
Follow-up period: 2.5 years.

Outcome Measures:
  1. Nutritional:
  2. Anthropometrical:
  3. Glucose levels
Regression analysis showed no significant influence of maternal third trimester glucose on the child’s body weight (OR 0.86, 95% CI 0.58 to 1.27) or impaired glucose tolerance (OR 0.97, 95% CI 0.56 to 1.67).

Relative body weight was lowest in children who had not ingested diabetic breast milk during the neonatal period and highest in those fed with diabetic breast milk only.

Exclusive breast-feeding was associated with increased childhood relative body weight (P = 0.011).

Breast-fed children of women with diabetes had increased risk of overweight (OR 1.98, 95% CI 1.12 to3.50).

Breast-feeding duration was positively associated with childhood relative body weight (P = 0.004) and 120-min blood glucose during an oral glucose tolerance test (P = 0.022).

Adjustment for the breast milk volume ingested during the early neonatal period (1st week of life) eliminated all these relationships with late neonatal breast-feeding and its duration.

No relationship was observed between maternal blood glucose in the middle of the third trimester and the subsequent outcome.
Neither late neonatal diabetic breast milk intake nor the duration of breast-feeding has an independent influence on childhood risk of overweight or impaired glucose tolerance in children of women with diabetes. The 1st week of life appears to be the critical window for nutritional programming in children of women with diabetes by ingestion of maternal diabetic breast milk.The study was generally: well designed with detailed background information and statistical analysis and reporting of results.

The only issue not clear is “some diabetic breast milk”, the authors did not explain.

From: Evidence tables

Cover of Diabetes in Pregnancy
Diabetes in Pregnancy: Management of Diabetes and Its Complications from Preconception to the Postnatal Period.
NICE Clinical Guidelines, No. 63.
National Collaborating Centre for Women's and Children's Health (UK).
London: RCOG Press; 2008 Mar.
Copyright © 2008, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.