350: Validation of risk assessment tools for postpartum hemorrhage in women undergoing cesarean deliveries

Abstract

While published risk assessment tools for postpartum hemorrhage (PPH) have been widely used, these risk assessment tools were created based on expert opinion. Data regarding the validity of these risk assessment tools are limited, particularly in women undergoing cesarean delivery. We sought to examine the validity of three known risk assessment tools (California Maternal Quality Care Collaborative [CMQCC], Association of Women’s Health, Obstetric and Neonatal Nurses [AWHONN], and New York Safety Bundle for Obstetric Hemorrhage [NYSBOH] in women undergoing cesarean delivery. We conducted a retrospective cohort study evaluating all women undergoing cesarean delivery at ≥23 weeks’ gestation between 2012 and 2017 at an urban hospital with an average of 1,200 cesarean deliveries per year. Women with missing information were excluded. Severe PPH was defined as estimated blood loss ≥1000 mL requiring transfusion of packed red blood cells. Women were stratified into low-risk, medium-risk, and high-risk groups using risk assessment tools. Odds ratios (OR) with 95% confidence interval (95%CI) were calculated according to risk stratification. Risk factors were examined using multivariable logistic regression according to risk assessment tools. Receiver operating characteristic curves with the area under a curve (AUC) and 95%CI were created according to each risk assessment tool. Of 6,925 women who underwent cesarean delivery, 293 (4.2%) had severe PPH. All risk assessment tools correctly classified high-risk women as having higher odds of severe PPH compared to women who were classified as low risk (see Table 1). The analyzed risk factors included in risk assessment tools were significantly associated with increased odds of severe PPH (p<.05) except for macrosomia (all tools), prolonged second stage (AWHONN and NYSBOH), induction or augmentation of labor (AWHONN), and body mass index (all tools) (see Table 2). However, all risk assessment tools had low AUCs (CMQCC 0.64 [0.61-0.67]; AWHONN 0.60 [0.58-0.63]; NYSBOH 0.64 [0.61-0.67]), demonstrating poor ability to predict severe PPH. Our findings question the utility of risk assessment tools in accurately predicting women at high risk for severe PPH. Further studies are needed to create more reliable risk assessment tools.View Large Image Figure ViewerDownload Hi-res image Download (PPT)