Construction of the risk prediction model for gestational hypertension patients with HELLP syndrome

CHEN Chunrong; CHEN Ningjing; YANG Ning; CHEN Weifeng; GAO Yuling; WU Biyu

doi:10.12102/j.issn.1009-6493.2025.04.003

Abstract：Objective:To analyze the influencing factors of HELLP syndrome in patients with gestational hypertension,and to construct the risk prediction model of gestational hypertension patients with HELLP syndrome.Methods:A total of 470 patients with gestational hypertension patients with HELLP syndrome treated in the Quanzhou First Hospital,the Second Affiliated Hospital of Fujian Medical University and the Children's Hospital,Quanzhou Maternal and Child Health Hospital from February 2020 to August 2023 were selected as the study objects.The clinical information of the subjects were collected.Multivariate Logistic regression was used to screen the risk factors of gestational hypertension patients with HELLP syndrome,and R software was used to establish a random forest model for predicting gestational hypertension patients with HELLP syndrome.Results:The incidence of HELLP syndrome was 9.57%in 470 gestational hypertension patients.Logistic regression analysis showed that age,onset gestation week,education level,whether regular childbirth examination,PLT,PLGF were all independent risk factors of HELLP syndrome in patients with gestational hypertension(P<0.05).The area under the receiver operating characteristic curve(AUC)of patients with HELLP syndrome predicted by random forest model had no significant difference from that of Logistic regression model.The prediction accuracy of Logistic regression model was 81.3%after the cross-validation of 5 fold.Conclusions:Age,gestational week,education level,regular childbirth examination,PLT and PLGF are the influencing factors of HELLP syndrome in gestational hypertension patients.The random forest model based on the above factors has a good predictive effect on the risk of HELLP syndrome in gestational hypertension patients.