Predicting Blood Loss for Vaginal Deliveries: When Are Quantitative Measures More Accurate? [29M

Abstract

INTRODUCTION: Hospitals have been adapting quantitative blood loss (QBL) techniques over traditionally estimated blood loss (EBL), which has been shown to underestimate large volumes of blood loss. Our purpose was to determine the milliliters of blood loss where QBL better approximates EBL in vaginal deliveries. METHODS: Retrospective cohort study was conducted at a tertiary center examining blood loss from singleton term vaginal births. EBL and QBL groups each contained 6 separate months of deliveries and compared to a standard: calculated blood loss (CBL), determined by maternal size and hematocrit decrease. Absolute differences between calculated and recorded blood loss were computed and modeled against CBL. Multiple linear regression was used to control for confounders. RESULTS: CBL between EBL (n=1085) and QBL (n=1076) groups were not significantly different (P=.051), with a median of 685 (interquartile range 426-948) mL vs 697 (457-967) mL. Absolute difference between CBL and recorded blood loss was smaller in the EBL group [402 (178-644) mL] than QBL group [497 (268-716) mL]. Absolute differences were plotted: EBL (r^2 .84, P<.001) and QBL (r^2 .63, P<.001) groups exhibited linear relationships. Intersection of the linear models corresponded to 370 mL; beyond this value, QBL group demonstrated a smaller difference in comparison to EBL group. CONCLUSION: QBL measurements are not more accurate than EBL for uncomplicated vaginal deliveries with little blood loss. However, when blood loss measures greater than 370 mL, QBL is a better approximate. Further studies need to be done to validate role of CBL in determining blood loss for all delivery types.