Jiamin Zhang, Weichao He, Haixuan Zhao, Shaorong Li, Tianhong Chen, Xiaoxia Chen, Minyi Yuan, Jing-Yun Yu, Lifen Pan
After adequate confounding control, NA was not found to independently increase the risk of PPH following vaginal delivery. The observed positive association was largely explained by instrumental delivery and episiotomy. Concern about PPH should not restrict appropriate use of NA in clinical practice, although attention to its indirect effects is warranted.
OBJECTIVE: To investigate the association between neuraxial analgesia (NA) and postpartum hemorrhage (PPH) after vaginal delivery, and to assess the mediating role of instrumental delivery and episiotomy.
METHODS: This multicenter retrospective cohort study included 148 302 women who underwent vaginal delivery between January 2023 and December 2025 among 75 institutions in Dongguan, China. The exposure was NA. The primary outcome was PPH (blood loss ≥500 mL within 2 h after vaginal delivery). Multimodel approaches including generalized estimating equations, inverse probability weighting, and propensity score matching were used to control for confounding. Mediation analysis was performed to evaluate indirect effects via instrumental delivery and episiotomy. A meta-analysis of published studies was conducted for external validation.
RESULTS: The crude PPH rate was 2.1% in the NA group and 1.7% in the nonanalgesia group. In unadjusted models, NA was positively associated with PPH (odds ratio [OR], 1.21 [95% confidence interval (CI), 1.13-1.31]). After full adjustment for covariates, the association attenuated and became nonsignificant (OR, 1.07 [95% CI, 0.99-1.16]). Propensity score matching (OR, 1.05 [95% CI, 0.96-1.15]) and inverse probability weighting (OR, 1.04 [95% CI, 0.96-1.13]) yielded similar results. Continuous blood loss analysis showed a mean increase of ≈3.5 mL. Mediation analysis indicated that the indirect effects via instrumental delivery and episiotomy accounted for 73.6% and 78.8% of the total effect, respectively, and the direct effect was not significant (P = 0.16). In the external meta-analysis, the pooled adjusted OR was 1.00 (95% CI, 0.77-1.28).
CONCLUSION: After adequate confounding control, NA was not found to independently increase the risk of PPH following vaginal delivery. The observed positive association was largely explained by instrumental delivery and episiotomy. Concern about PPH should not restrict appropriate use of NA in clinical practice, although attention to its indirect effects is warranted.