Abstract The objective of this study is to evaluate whether full quantitative blood loss (FBL), which incorporates the weight of blood soaked materials, improves the prediction of hemorrhage-related clinical decision-making compared with volumetric blood loss (VBL) using calibrated drapes alone. We performed a secondary analysis of an IRB-approved retrospective cohort including all deliveries over 2 years at a large, urban academic center. During the study period, a quantitative blood loss protocol was introduced without other changes to hemorrhage management. VBL consisted of calibrated drape measurement with visual estimation of laps and clots, while FBL additionally included weighing of soaked materials. The primary outcome was hemorrhage intervention, defined as transfusion, surgical management of atony, balloon tamponade, ICU admission, or hysterectomy. Predictive performance of VBL and FBL was compared using receiver operating characteristic curves and area under the curve (AUC). Secondary outcomes included uterotonic use, correlation with peripartum hemoglobin change, and a subgroup analysis of vaginal deliveries with blood loss of 500 to 1,000 mL. Among 8,359 deliveries, 1,524 (18.4%) had both VBL and FBL recorded, allowing direct comparison. Overall, hemorrhage intervention occurred in 5.4% of deliveries. There was no significant difference in predicting intervention in either the direct-comparison group or the full cohort. In the paired sample, VBL better predicted uterotonic use than FBL (AUC: 0.713 vs. 0.680, p¼0.02). FBL identified more deliveries as having abnormal blood loss and more cases meeting PPH criteria; however, this did not translate into higher intervention rates. Both methods correlated modestly with hemoglobin change, explaining <13% of variance, with no meaningful difference between approaches. When calibrated drapes are used, adding weighing of blood-soaked materials provides minimal additional clinical value in predicting hemorrhage interventions or physiologic impact. These findings support streamlined blood loss quantification strategies, particularly in low-resource settings where simplicity and efficiency are critical.
Wiley et al. (Fri,) studied this question.