Shock index as a predictor of postpartum haemorrhage after vaginal delivery: Secondary analysis of a multicentre randomised controlled trial

Objective To describe the shock index (SI) distribution during the first 2 hours after delivery and to evaluate its performance when measured 15 and 30 minutes after delivery for predicting postpartum haemorrhage (PPH) occurrence in the general population of parturients after vaginal delivery. Design Secondary analysis of a multicentre randomised controlled trial testing prophylactic administration of tranexamic acid versus placebo in addition to prophylactic oxytocin to prevent PPH. Setting 15 French maternity units in 2015–2016. Sample 3891 women with a singleton live fetus ≥35 weeks, born vaginally. Methods For each PPH‐related predicted outcome, we calculated the area under the receiver operating characteristic curve (AUROC) values of the SI at 15 and 30 minutes after delivery and its predictive performance for SI cut‐off values of 0.7, 0.9 and 1.1. Main outcome measures Quantitative blood loss ≥1000 ml (QBL ≥1000 ml) measured in a graduated collector bag and provider‐assessed clinically significant PPH (cPPH). Results Prevalence of QBL ≥1000 ml and cPPH was respectively 2.7% (104/3839) and 9.1% (354/3891). The distributions of the SI at 15 and 30 minutes after delivery were similar with a median value of 0.73 and 97th percentile of 1.11 for both. The AUROC values of the 15‐minute SI for discriminating QBL ≥1000 ml and cPPH were respectively 0.66 (lower limit of the 95% confidence interval [LCI] 0.60) and 0.56 (LCI 0.52); and for the 30‐minute SI 0.68 (LCI 0.61) and 0.49 (LCI 0.43). Conclusions The shock index at 15 and 30 minutes after delivery did not satisfactorily predict either QBL ≥1000 ml or clinical PPH.