The value of angiogenic biomarkers in determining preeclampsia-related perinatal deaths: A cross-sectional study in a high burden setting of southern Mozambique

Estimating the contribution of preeclampsia (PE) to perinatal mortality in low-resource settings remains difficult due to limited diagnostic capacity. Lack of specific histopathologic changes in stillbirth, foetuses and/or placenta often hampers assignment of PE as cause of death. While PE angiogenic biomarkers (sFlt-1, PlGF) predict adverse pregnancy outcomes, their post-mortem diagnostic utility is unclear. We evaluated the potential role of these biomarkers in identifying PE-related perinatal deaths. This cross-sectional study was conducted among pregnant women who delivered a stillborn infant or experienced an early neonatal death at Maputo Central Hospital in Mozambique. Concentrations of sFlt-1, PlGF and their ratio (sFlt-1/PlGF) were assessed in maternal blood and post-mortem foetal/neonatal and placental blood. Causes of death were determined via minimally invasive tissue sampling (MITS) and diagnostic accuracy was assessed using ROC curves for standardized and optimal biomarker cut-offs. A total of 100 women with perinatal deaths (98 stillbirths and 2 early neonatal deaths) were included in the study between March 2021 and April 2022. Maternal sFlt-1/PlGF ratios showed the highest diagnostic accuracy for PE-related deaths (area under curve [AUC]=82%), followed by placental (AUC = 75%) and foetal blood (AUC = 61%). PlGF levels were significantly reduced in PE-associated deaths. sFlt-1/PlGF cutoffs of ≥85 and ≥110 in maternal and placental blood were reliable cut-offs for identifying PE-related foetal/neonatal deaths. Optimal ratio cut-offs were ≥50 (maternal blood), ≥ 230 (placental blood), and ≥140 (foetal blood), with maternal and placental sFlt-1/PlGF ratios showing significant associations with PE-related death (OR=10.58 and 5.98 respectively). Sensitivity and specificity in maternal and placental blood were 84% and 73%, and 67% and 78%, respectively, with a positive predictive value (PPV) of 84% in both. In conclusion, maternal and placental angiogenic biomarkers enable reliable identification of preeclampsia-related perinatal deaths and could significantly enhance cause-of-death determination in low-resource, high-burden settings. These findings underscore the potential value of biomarker measurement for both risk stratification and mortality surveillance.