Accuracy of visual estimation of blood loss in obstetrics using clinical reconstructions: an observational simulation cohort study.

INTRODUCTION

Postpartum hemorrhage is the leading cause of maternal mortality worldwide, and optimal management requires accurate blood loss estimations. The aim of this study was to assess whether differences exist between visually estimated blood loss vs. actual blood loss based on delivery mode, blood volume or distribution/location and knowledge of patient's current cardiovascular status.

METHODS

For this observational cohort study, photographs were taken of 18 blood loss scenarios for vaginal delivery and cesarean delivery, and six photographs were duplicated and annotated with maternal vital signs. Scenarios were categorized into 50% (500 mL), 100% (1000 mL) and 200% (2000 mL) of the defined blood loss volume for postpartum hemorrhage and the photographs were shown to participants to visually estimate blood loss volumes.

RESULTS

The mean ± standard deviation estimates of actual 500 mL, 1000 mL and 2000 mL blood loss volumes were 1208 ± 438 mL, 1676 ± 630 mL and 2637 ± 1123 mL, respectively (P <0.001 among groups). The difference was significantly greater in vaginal delivery than cesarean delivery scenarios (1064 ± 849 mL vs. 284 ± 456 mL; P <0.001). Estimated blood loss volume was not influenced by blood loss distribution/location, or by provider group or experience. The cardiovascular status of the patient impacted estimations only if tachycardia and hypotension were present.

CONCLUSIONS

Most providers significantly overestimated blood loss volumes (by nearly 700 mL). Provider and scenario factors that impact inaccuracies in visual estimated blood loss identified in this study can be used to guide education and training.