Previous caesarean delivery and the risk of unexplained stillbirth: retrospective cohort study and meta-analysis.

OBJECTIVE: To determine whether caesarean delivery in the first pregnancy is a risk factor for unexplained antepartum stillbirth in a second pregnancy. DESIGN: A population-based retrospective cohort study and meta-analysis. SETTING: All maternity units in Scotland. PARTICIPANTS: A cohort of 128 585 second births, 1999-2008. METHODS: Time-to-event analysis and random-effects meta-analysis. MAIN OUTCOME MEASURE: Risk of unexplained antepartum stillbirth in a second pregnancy. RESULTS: There were 88 stillbirths among 23 688 women with a previous caesarean delivery (2.34 per 10 000 women per week) and 288 stillbirths in 104 897 women who had previously delivered vaginally (1.67 per 10 000 women per week, P = 0.002). When analysed by cause, women with a previous caesarean delivery had an increased risk of unexplained stillbirth (hazard ratio, HR 1.47; 95% confidence interval, 95% CI 1.12-1.94; P = 0.006) and, as previously observed, the excess risk was apparent from 34 weeks of gestation onwards. The risk did not differ in relation to the indication of the caesarean delivery, and was independent of maternal characteristics and previous obstetric complications. We identified three other comparable studies (two in North America and one in Europe), and meta-analysis of these studies showed a statistically significant association between previous caesarean delivery and the risk of antepartum stillbirth in the second pregnancy (pooled HR 1.40; 95% CI 1.10-1.77; P = 0.006). CONCLUSIONS: Women who have had a previous caesarean delivery are at increased risk of unexplained stillbirth in the second pregnancy. TWEETABLE ABSTRACT: Caesarean first delivery is associated with an increased risk of unexplained stillbirth in the next pregnancy.The work was supported by the NIHR Cambridge Comprehensive Biomedical Research Centre. The funding bodies had no role in any aspect of the conduct, analysis or presentation of this study.This is the accepted manuscript. The final version is available at http://dx.doi.org/10.1111/1471-0528.1346