OP13.04: Predictive performance of combined screening plus PlGF for fetal growth restriction

To evaluate the predictive performance of combined screening for fetal growth restriction versus combined screening plus PIGF. A retrospective cohort was conducted from 2016 to 2022. Single pregnancies with combined screening for aneuploidies and determination of PIGF and pregnancy resolution were included. Data from patients evaluated in the maternal-fetal medicine department and for whom complete information was found in the Fetal Medicine Foundation risk calculator were included. The risk calculation for fetal growth restriction was carried out considering the maternal demographic variables, individual risks, and three biochemical markers (PAPPA, free B -human chorionic gonadotropin, and PIGF) subsequently the risk was recalculated including only PAPPA and free B -human chorionic gonadotropin; two groups were made: 1. Combined screening and 2. Combined screening plus PIGF, each of these was subdivided into low and high risk using the cutoff point > 1:100. Descriptive statistics were performed for the study variables, frequencies, percentages, measures of central tendency, and dispersion were calculated. Sensitivity, specificity, predictive values, and likelihood ratios of both tests were evaluated, odds ratios were calculated for the main outcome and ROC curves were performed. Four hundred patients were included. The positive predictive value (PPV) of the combined screening was 15.79% (9/57,) the negative predictive value (NPV) 90.67%, sensitivity 21.95% and specificity 86.63%, the area under the ROC curve was 0.577, for the combined screening plus PIGF the PPV was 18.52% (10/54), NPV 91.04%, sensitivity 24.39% and specificity 87.74%, the area under the ROC curve 0.578. The predictive performance of combined screening + PIGF versus combined screening shows a difference of less than 10%, both tests have a positive predictive value and low discriminatory capacity, through the ROC curves we identified cut-off points with a higher performance which could improve the prediction of FGR.