Implications (in light of other evidence)
Our finding regarding the increased risk for subsequent stillbirth in pregnancies with a previous CD is consistent with several other studies18, 24-28. However, most of the previous studies had inconsistencies and weaknesses in the methods and statistical analysis. For example, most studies did not exclude stillbirth from the first birth which is strongly associated with subsequent stillbirth21.
Our findings agree with those from a recent Norwegian cohort study of 294 598 singleton second births, which reported a 33% increased odds of antepartum stillbirth (aOR:1.33 [95% CI, 1.08–1.63)] and a 84% increased odds of intrapartum stillbirth (aOR:1.84 [95% CI, 1.00–3.38])18, although the timing of CD was not considered in this study.
When analysed by cause of fetal death, excess risk was apparent of unexplained stillbirth from 34 weeks’ gestation as previously observed10, 26. It should be noted however that both studies failed to adjust for maternal comorbidities and our results attenuated to almost no effect in the adjusted model.
We found no increased odds of intrapartum stillbirth in women who had VBAC compared to those with repeat CD. It must be acknowledged however that there were few cases of intrapartum stillbirth (n=13) in women who had a VBAC. To our knowledge, none of the previous studies assessed the risk of stillbirth while adjusting for potential confounders in women who had a VBAC, possibly due to the rarity of this event. Findings from a systematic review and meta-analysis suggested a higher rate of perinatal mortality in women who had VBAC compared with a planned repeat CD (0.13% versus 0.05%), but no data were reported on stillbirth19.
Two studies reported inconsistent results for the association between perinatal death and neonatal death in women with a first CD undergoing VBAC29, 30. Smith et al. assessed the risk of perinatal mortality, including intrapartum stillbirth and neonatal death using a large Scottish registry data of 313,238 singleton births and reported 11 times greater odds (aOR:11.7 [95% CI, 1.4–101.6]) compared to women with a repeat CD29. However, the reported CI was very wide, and the authors did not adjust for maternal comorbidities. The second study by O’Neill et al .30 evaluated the risk of neonatal and infant death in women with a VBAC compared to women with a repeat CD using data from the Danish registry, including 61,626 births. The authors reported no increased odds of late neonatal death (aOR:0.97 [95% CI, 0.22–4.32) or infant death (aOR:1.12 [95% CI, 0.79 –1.59])30.
Our analysis demonstrates that maternal conditions (including hypertensive disorders of pregnancy) and intrapartum asphyxia have a significant impact on the association between CD and subsequent stillbirth. This provides information that could formulate a causal hypothesis for the observed association. A case control study found that the uterine artery Doppler waveform between 18-22 weeks’ gestation is more likely to be notched (14.5% vs. 6%, p<0.001) and have a higher average pulsatility index (1.25 vs 1.16, p<0.02) in women who had an elective CD compared to a VB31. Notably, women who had a prior CD also have a higher incidence of preeclampsia32, 33. Further research is required to further understanding of a causal relationship between CD and stillbirth; these studies suggest that investigation of impaired placental blood flow merits further exploration.
In case of antepartum stillbirth, it is common practice and a safe option for women with a previous CD to have VBAC as most of the risk of VBAC relates to the fetus34. Both the Royal College of Obstetricians and Gynaecologists and the American College of Obstetricians and Gynecologists encourage that VB should be offered over CD in case of antepartum stillbirth.34 This could explain the observed association between VBAC and antepartum stillbirth.