sFlt-1/PlGF ratio for prediction of preeclampsia in clinical routine: A pragmatic real-world analysis of healthcare resource utilisation

Abstract

Background We investigated the impact of the soluble fms-like tyrosine kinase 1 (sFlt-1)/placental growth factor (PlGF) ratio to predict short-term risk of preeclampsia on clinical utility and healthcare resource utilisation using real-world data (RWD), and compared findings with health economic modelling from previous studies. Methods and findings This retrospective analysis compared data from the German population of a multicentre clinical study (PROGNOSIS, n = 203; sFlt-1/PlGF ratio blinded and unavailable for decision-making) with RWD from University Hospital Leipzig, Germany (n = 281; sFlt-1/PlGF ratio used to guide clinical decision-making). A subgroup of the RWD cohort with the same inclusion criteria as the PROGNOSIS trial (RWD prediction only, n = 99) was also included. sFlt-1/PlGF ratio was measured using fully automated Elecsys® sFlt-1 and PlGF immunoassays (cobas e analyser; Roche Diagnostics). A similar proportion of women in the RWD and PROGNOSIS cohorts experienced preeclampsia (14.95% vs. 13.79%; p = 0.7938); a smaller proportion of women in the RWD prediction only cohort experienced preeclampsia versus PROGNOSIS (6.06%; p = 0.0526). In women with preeclampsia, median gestational age at delivery (weeks) was comparable in the RWD and PROGNOSIS cohorts (34.0 vs. 34.3, p = 0.5895), but significantly reduced in the RWD prediction only cohort versus PROGNOSIS (27.1, p = 0.0038). sFlt-1/PlGF ratio at baseline visit was not statistically significantly different for the RWD and PROGNOSIS cohorts, irrespective of preeclampsia outcome. Hospitalisations for confirmed preeclampsia were significantly shorter in the RWD cohort versus PROGNOSIS (median 1 vs. 4 days, p = 0.0093); there was no significant difference between RWD prediction only and PROGNOSIS (3 days, p = 0.9638). All-cause hospitalisations were significantly shorter in the RWD (median 1 day; p<0.0001) and RWD prediction only (1 day; p<0.0001) cohorts versus PROGNOSIS (3 days). Conclusions This study supports the findings of previous studies, showing that routine clinical use of the sFlt-1/PlGF ratio may result in shorter duration of hospitalisations, with potential economic benefits.