RESUMO
The aim of our study was to identify characteristics associated with postpartum hemorrhage (PPH defined as blood loss >1000 mL) in twin pregnancies in order to select patients at higher risk to be treated. This retrospective study includes multiple pregnancies between 2015 and 2020. The possible association between pregnancy characteristics and the primary endpoint (occurrence of PPH) was conducted using chi-square or Fisher exact test and Wilcoxon test. Then, univariate logistic models were performed considering as outcome the PPH, and the odds ratios with 95% CI were estimated. Finally, a multivariate logistic model was implemented, including all significant covariates. Seven hundred seven twin pregnancies giving birth beyond 32 weeks were included and of those, 120 (16.97%) had a PPH. The univariate analysis showed that factors significantly associated with PPH were: Preterm delivery, episiotomy, neonatal weight, and mode of delivery. The multivariate analysis showed that the most important factors were episiotomy and neonatal weight. The results show that the performance of episiotomy and the neonatal weight are the factors that most impact the risk of PPH in twin pregnancies. The correct identification of factors associated with PPH in twins could ideally allow to modify the clinical management and positively affect the rate of complications.