Abstract
BACKGROUND: Postpartum hemorrhage (PPH) is a common yet potentially serious complication of delivery. Recently, the HEMSTOP bleeding score has emerged as a tool for detecting preoperative inherited bleeding disorders. Our objective is to investigate its relevance within the obstetrical context for predicting PPH occurrences among unselected pregnant women. METHODS: Prospective cohort study conducted in a tertiary maternity hospital between 2014 and 2016 including women without any known bleeding disorder nor antithrombotic therapy who completed the HEMSTOP questionnaire before delivery. Primary outcome was primary PPH ≥ 500 mL following delivery. We studied the bleeding score ranging from zero to seven both as a continuous and an ordinal variable. We used multivariable analysis with adjustment on PPH risk factors. We also estimated the measures of the bleeding score prognostic accuracy. RESULTS: PPH occurred in 116 of 2536 women (4.6%). Elevated bleeding scores were associated with increased PPH risk (adjusted RR = 1.58; 95% CI, 1.23 to 2.02; P < 0.001 for a continuous score) demonstrating a progressive rise in risk alongside score escalation (adjusted RR = 1.58; 95% CI, 1.01 to 2.46 for a score of one, adjusted RR = 2.11, 95% CI 0.86 to 5.20 for a score of two and adjusted RR = 7.20, 95% CI 2.54 to 20.41 for a score of three compared with a score of zero; P < 0.001). The area under the curve for the bleeding score accuracy in predicting PPH was 0.56 [95% CI 0.52 to 0.61]. CONCLUSIONS: The HEMSTOP bleeding score is weakly predictive of PPH. Women scoring ≥ three are at high risk of PPH.