The association between intrapartum interventions and immediate and ongoing breastfeeding outcomes: an Australian retrospective population-based cohort study

Abstract

Abstract Background The use of intrapartum interventions is becoming increasingly common globally. Interventions during birth, including caesarean section (CS), epidural analgesia and synthetic oxytocin infusion, can be important in optimizing obstetric care, but have the potential to impact breastfeeding. This study aimed to identify whether women who have certain intrapartum interventions have greater odds of unfavourable breastfeeding outcomes, both the immediate post-partum period and in the months after birth. Methods This was a population-based cohort study of singleton livebirths at ≥37 weeks’ gestation between 2010 and 2018 in Victoria, Australia using routinely-collected state-wide data from the Victorian Perinatal Data Collection (VPDC) and the Child Development Information System (CDIS). The interventions included were pre-labour CS, in-labour CS, epidural analgesia, and synthetic oxytocin infusion (augmentation and/or induction of labour). Outcomes were formula supplementation in hospital, method of last feed before hospital discharge and breastfeeding status at 3-months and 6-months. Descriptive statistics and multivariable logistic regression models adjusting for potential confounders were employed. Results In total, 599,191 women initiated breastfeeding. In-labour CS (aOR 1.96, 95%CI 1.93,1.99), pre-labour CS (aOR 1.75, 95%CI 1.72,1.77), epidural analgesia (aOR 1.45, 95%CI 1.43,1.47) and synthetic oxytocin infusion (aOR 1.24, 95%CI 1.22,1.26) increased the odds of formula supplementation in hospital. Long-term breastfeeding data was available for 105,599 infants. In-labour CS (aOR 0.79, 95%CI 0.76,0.83), pre-labour CS (aOR 0.73, 95%CI 0.71,0.76), epidural analgesia (aOR 0.77, 95%CI 0.75,0.80) and synthetic oxytocin infusion (aOR 0.89, 95%CI 0.86–0.92) decreased the odds of exclusive breastfeeding at 3-months post-partum, which was similar at 6-months. There was a dose-response effect between number of interventions received and odds of each unfavourable breastfeeding outcome. Conclusion Common intrapartum interventions are associated with less favourable breastfeeding outcomes, both in hospital and in the months after birth. This confirms the importance of only undertaking interventions when necessary. When interventions are used intrapartum, an assessment and identification of women at increased risk of early discontinuation of breastfeeding has to be performed. Targeted breastfeeding support for women who have intrapartum interventions, when they wish to breastfeed, is important.