Interpreting comparative effectiveness of vacuum‐assisted hemorrhage control and uterine balloon tamponade: The impact of temporal and clinical confounding

We read with interest the retrospective cohort study by Lee and colleagues comparing a vacuum-assisted hemorrhage-control device (VHD) with uterine balloon tamponade (UBT) for refractory atonic postpartum hemorrhage, in which VHD was associated with lower composite morbidity (45.0% versus 75.8%; unadjusted odds ratio 0.26).1 The authors are appropriately cautious in noting that this difference lost significance after adjustment. We would go further: the design does not permit a fair comparison of the two devices for two related reasons. First, device and calendar time are almost completely confounded. VHD became available at the institution only in October 2023, whereas UBT was used across the entire 2020 to 2025 window, so the VHD group is essentially a later-era cohort. During that same interval, the unit implemented ROTEM-guided transfusion and E-MOTIVE-aligned bundle care, both of which independently reduce hemorrhage morbidity. Because nearly every VHD patient was treated under the more mature pathway, the device effect and the secular improvement in hemorrhage care cannot be separated within this single-center series. The authors note temporal bias, but its structure here is close to collinearity, which no covariate adjustment can resolve; a fair comparison would require restriction to the overlapping period or a design that breaks this link. Second, residual confounding by indication remains, and the adjusted model is underpowered. UBT patients had far more placenta previa (45.5% versus 5.0%) and cesarean delivery (78.8% versus 50.0%), markers of more complex bleeding; the crude association became non-significant after adjusting for only these two variables (adjusted odds ratio 0.38, 95% CI 0.10 to 1.41).2 With 53 women and few events, the model can support only two or three covariates,3 leaving other determinants of severity, such as initial blood loss, coagulopathy, and bleeding etiology, unaddressed. The composite endpoint compounds this, mixing components of very different severity and including device failure, a device-specific outcome whose definition differs between modalities and risks circularity; reporting the components separately would be more informative.4 We therefore read these findings as hypothesis-generating rather than as evidence of device superiority. A concurrent-period analysis, component-level reporting, and ultimately a prospective multicenter comparison would be needed to determine whether VHD offers a genuine advantage over UBT in refractory atonic postpartum hemorrhage. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.