Abstract
Purpose: To investigate the predictive performance of placental growth factor (PlGF) and soluble FMS-like kinase 1 (sFlt-1) on birth weight and small for gestational age (SGA), in a large, population-based cohort.
Methods: Women enrolled in the population-based, prospective Odense Child Cohort Study with early (GA < 20 weeks) and/or late (≥20 weeks) pregnancy blood samples (n = 1937) were included. The association between log-transformed values of the biomarkers and birth weight Z-score was studied using multivariate regression models. The prediction of SGA overall, and in women developing preeclampsia, by biomarkers was evaluated using receiver operating characteristic analyses.
Results: No substantial associations between early pregnancy biomarkers and SGA were seen. PlGF measured in late pregnancy demonstrated the strongest association with birth weight Z-score (adjusted β-coefficient = 0.43 [95%CI = 0.35; 0.50]). The area under curve (AUC) for predicting SGA was higher for sFlt-1/PlGF compared to sFlt-1 (0.74 versus 0.63, p = .006) and reached excellent prediction for SGA after preeclampsia (AUC 0.94). Optimal sFlt-1/PlGF ratio cut-offs had higher negative predictive value (NPV) and positive predictive value (PPV) for SGA (cut-off > 5.0; NPV = 99.1%, PPV = 5.4%) compared to each marker individually.
Conclusion: The sFlt-1/PlGF ratio is a potential predictor of SGA in population-based screening, particularly when preeclampsia is also present.
Disclosure statement
Henrik T. Christesen has received an open research grant from BRAHMS GmbH. Louise B. Andersen and Henrik T. Christesen have received speaker fees from BRAHMS GmbH. The other authors report no conflicts of interest. This is a secondary analysis of a dataset, the first paper published in hypertension in pregnancy [Citation42].