The ability of shock index to predict refractory postpartum haemorrhage: A secondary analysis of the prospective and multicentre HELP MOM study

Abstract

Background: The predictability of severe and refractory postpartum haemorrhage (PPH) remains a challenge for clinicians and researchers.

Objective: We aimed to evaluate the ability of the shock index to predict refractory PPH.

Design: This study was a secondary analysis of a multicentre, prospective, observational study investigating the association between severe postpartum haemorrhage (PPH) and the subsequent development of mental disorders.

Setting: Participants were patients who experienced severe PPH, characterised by blood loss of at least 1500 ml requiring the administration of sulprostone. The shock index, defined as the ratio of heart rate to SBP, was recorded at two time points: at the start of sulprostone infusion (T0) and at the time of the most severe symptoms after enrolment but before the occurrence of refractory PPH (T1). Refractory PPH was defined by the need for four or more blood products, interventional radiology, or laparotomy (excluding caesarean section).

The main outcome: The predictive ability of the shock index was assessed using univariate and multivariate logistic regression and area under the receiver operating characteristic curve (AUROC).

Results: From November 2014 to November 2016, 332 patients experienced severe PPH, of which 316 were included in the final analysis. The prevalence of refractory PPH was 35.4% (112/316). We found that high shock index levels at T0 were independently associated with the occurrence of refractory PPH: odds ratio (OR) 3.07 [95% confidence interval (CI), 1.22 to 7.89, P = 0.017]. In addition, high shock index levels at T1 were also independently associated with the occurrence of refractory PPH: OR 5.28 (95% CI, 2.25 to 12.8), P < 0.001. The AUROC of shock index levels measured at T0 and T1 were 0.614 (95% CI, 0.549 to 0.678) and 0.681 (95% CI, 0.616 to 0.746), respectively.

Conclusion: The shock index measured at the start of sulprostone infusion and at the worst time after enrolment, has poor discriminative power to predict this event.