Early prediction of preeclampsia among the advanced-age pregnant women: A retrospective multicenter study.

BACKGROUND

Preeclampsia (PE) remains a leading cause of maternal and perinatal morbidity. Women of advanced maternal age (AMA, ≥ 35 years) constitute a rapidly expanding high-risk subpopulation in China. Existing first-trimester prediction models were derived in general obstetric populations and often rely on specialized biophysical or proprietary biomarker assays, with limited validation in AMA women.

METHODS

This multicenter retrospective study enrolled 2,582 AMA pregnant women from three tertiary centers in southern China, partitioned into a training cohort (n = 1,327), an internal validation cohort (n = 569), and two external validation cohorts (n = 399 and 287). Predictors were selected using LASSO and multivariable logistic regression and assembled into a nomogram. Performance was evaluated by area under the receiver operating characteristic curve (AUC), calibration, Brier scores, decision curve analysis, and clinical impact curves.

RESULTS

Seven independent predictors were retained: pre-pregnancy BMI, parity, mode of conception, uric acid, white blood cell count, red blood cell count, and hemoglobin. The nomogram achieved AUCs of 0.819, 0.795, 0.787, and 0.756 across the four cohorts, with close calibration and favorable net benefit. Performance was preserved when women with chronic hypertension were included and across early-onset and late-onset PE subgroups. Notably, maternal age itself was not discriminative within the AMA stratum (P = 0.626).

CONCLUSIONS

This nomogram relying exclusively on routine clinical and laboratory parameters provides an accessible tool for individualized early-pregnancy PE risk assessment and risk-stratified antenatal management in AMA women.