Background
Concerns have been expressed in relation to a possible negative effect of fentanyl epidural analgesia on breastfeeding rates. A 2016 systematic review was inconclusive and suggested cohort studies be performed with control groups to allow for other potential confounding factors.

Aims
To assess the effect of fentanyl epidural analgesia in labour on breastfeeding practices.

Material and Methods
A prospective cohort study, conducted at a metropolitan hospital, Ipswich, Queensland. Eligible women were first‐time mothers, wishing to breastfeed after a vaginal birth at term. Women who had fentanyl epidural only (107) were compared with women who had subcutaneous morphine only (90), and women who had neither subcutaneous morphine nor epidural (107).

Results
Fewer women in the fentanyl epidural analgesia group were fully breastfeeding when leaving hospital (54.2% compared to 69.2% in the no analgesia group and 77.8% in the subcutaneous morphine group; P < 0.05). At 6 weeks postpartum, there was no evidence of difference between the groups (43% for fentanyl epidural analgesia, 52.3% for no analgesia group and 48.9% for subcutaneous morphine group; P = 0.63). Rates of induction of labour, instrumental birth and admission to special care nursery differed significantly between groups and when controlled for these, there was no association between fentanyl epidural analgesia and breastfeeding at discharge (adjusted odds ratio (OR) 0.74; 95% CI 0.39–1.41) or at 6 weeks postpartum (OR 0.75; 95% CI 0.41–1.38). Conclusion Our study did not find an association between fentanyl epidural analgesia in labour and rates of fully breastfeeding at discharge and at 6 weeks postpartum.