Preeclampsia prediction model using demographic, clinical, and sonographic data in the second trimester of Japanese nulliparous women

Le résumé fourni par la source

Abstract Aim This study aimed to clarify the factors influencing preeclampsia (PE) development in nulliparous Japanese women and to develop a PE prediction model using second trimester sonographic and clinical data readily available to obstetricians. Methods This historical cohort study examined the obstetric records of nulliparous women who delivered at Yamanashi Prefectural Central Hospital from January 2019 to May 2023. A model was constructed to predict the PE development rate, with a focus on 796 nulliparous women. The assessed outcome was PE, excluding superimposed PE. Data on maternal age, assisted reproductive technology, mean arterial pressure, uterine artery notching, and umbilical artery resistance index were extracted. Multivariable logistic regression analysis was conducted on these five factors. Results The incidence of PE was 4.3% (34/796). Multivariable analysis indicated significant odds ratios for the association of PE with mean arterial pressure (adjusted odds ratio: 1.06, 95% confidence interval: 1.03–1.10) and uterine artery notching (adjusted odds ratio: 6.28, 95% confidence interval: 2.82–14.0) in nulliparous women. The PE prediction formula was established as follows: Probability of PE development (%) = (odds/1 + odds) × 100, odds = ex and x = −11.3 + 0.039 × maternal age (years) + 0.91 × assisted reproductive technology + 0.061 × mean arterial pressure (mmHg) + 1.84 × uterine artery notching + 1.84 × umbilical artery resistance index. The sensitivity and specificity of this model were 58.8% and 84.5%, respectively (area under the curve: 0.79). Conclusions This study is the first to provide a prediction formula targeting the Japanese population. Our specialized model for nulliparous women could guide obstetricians to educate women regarding the precise prospect of PE development.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.