[External validation of FullPIERS model for prediction of adverse outcomes among women with pre-eclampsia in French maternity of 2014 to 2018].

Objectives: FullPIERS score (Pre-eclampsia Integrated Estimate of RiSk) is built in order to predict maternal outcomes among women with preeclampsia. Our objective is to assess external validity of this score among a population of French women.

Methods: A retrospective study was conducted for all patients with pre-eclampsia (severe or not) admitted to a tertiary hospital center. A receiver operating characteristic (ROC) curve was constructed to determine accuracy of the model at 48 hours and 7 days after admittance and area under the curve (AUC) was calculated.

Results: In total, 276 patients meet inclusions criterion. Among them, 217 were included and 61 (28 %) had one or more complication. Only 8 patients had two complications at two different moments (one at 48 hours and one at 7 days). The FullPiers score predicted maternels events at 48 hours with an area under the curve at 0.80 (IC95 % [0.74-0.85]). At 4.2 cut off, the sensitivity was 71 % and specifity 88 %. At 7 days, the area under the curve was 0.74 (IC95 % [0.67-0.79]), admiting a cutoff point for FullPIERS probability of 3.4, sensitivity was 59 % and specificity 85 %.

Conclusions: Our study show utility of FullPIERS score in french maternity. This model might be a useful tool for predicting complications in women with pre-eclampsia. Using of this score in clinical practice might help improving optimization of patient's care.