Presented at the Neonatal Society 2010 Summer Meeting.
Prior E, Hyde MJ, Santhakumaran S, Philipps L, Gale CRK, Modi N
Neonatal Medicine, Imperial College London, Chelsea and Westminster Hospital, 369 Fulham Road, London, SW10 9NH, UK
Background & Aims: Despite the well documented benefits of breastfeeding for mother and baby, the rate of breastfeeding remains suboptimal in many countries, including the UK. In an attempt to address this, the World Health Organisation established the Baby Friendly Hospital Initiative (BFHI) to introduce practices supportive of breastfeeding into hospitals. Disappointingly breastfeeding rates remain stubbornly low. Several factors are likely to contribute to this, one of which may be delivery by Caesarean section (CS) (1). Given the rising rate of CS in the UK and worldwide, we aimed to conduct a systematic review to synthesise currently available evidence about effect of mode of delivery on breastfeeding.
Methods: A literature search was conducted in Medline using pre-defined search terms. Studies were included if they measured any breastfeeding outcomes by mode of delivery. Authors were contacted to obtain further information if required. Where studies were comparable in exposure definition and outcome measure, a meta-analysis was carried out in RevMan5 using the inverse variance method. This was carried out separately for adjusted and unadjusted results. Where heterogeneity was present a random effects meta-analysis was carried out, otherwise a fixed effects analysis was performed. Forest plots were used to illustrate results and funnel plots were to investigate publication bias.
Results: 186 papers were located of which 29 were eligible for inclusion in the review. The pooled unadjusted odds ratio (OR) for breastfeeding initiation comparing all CS with all vaginal delivery (VD) was 0.66 (95% CI 0.58, 0.75; p < 0.001; 20 studies; CS n = 67177 VD n = 222812). Among studies which controlled for confounding factors, the adjusted OR for breastfeeding initiation was 0.55 (95% CI 0.42, 0.71; p < 0.001; 11 studies). Comparing emergency CS with elective CS, the unadjusted OR for breastfeeding initiation was 1.22 (95% CI 1.16, 1.28; p < 0.001; 4 studies; emergency n = 17835 elective n = 11763). There was no difference in breastfeeding at 6 months in CS and VD women who had successfully initiated breastfeeding (unadjusted OR 1.04; 95% CI 0.80, 1.35; p = 0.79; 7 studies; CS n = 7605 VD n = 18734).
Figure 1: Caesarean section delivery is associated with lower rates of initiation of breastfeeding. Forest plot of the unadjusted odds ratio of breastfeeding initiation by mode of delivery.
Conclusion: Our meta-analysis has shown that breastfeeding initiation rate is influenced by mode of delivery. In particular, elective CS is associated with a significant reduction. This may be due to hospital practice, maternal characteristics or physiological differences in mother and baby consequent upon mode of delivery. If a mother initiates breastfeeding there appears to be no effect of mode of delivery on breastfeeding at 6 months post-partum. We suggest that mothers delivering by CS require extra support focussed on increasing breast feeding initiation. Staff providing support must be made aware of the particular needs of this increasingly prevalent group of women and their babies.
1. Rowe-Murray HJ, Fisher JR. Birth (Berkeley, Calif) 2002; 29:124