Risk prediction for preeclampsia in CKD patients: development of a model in a retrospective cohort.

Abstract

Background: Chronic kidney disease (CKD) may affect women of childbearing age and may lead to substantial maternal and foetal morbidity and mortality in pregnancy. There is a lack of prediction models for  preeclampsia and adverse pregnancy outcomes in pregnant women with CKD. This study aimed to create a prediction nomogram for these issues.

Methods: This retrospective cohort study included clinical data from 627 women with CKD and their 627 pregnancies at Peking University First Hospital between January 1, 2009, and December 31, 2022. Multivariate logistic regression analysis was conducted to identify independent prognostic factors and develop a nomogram for predicting the occurrence of preeclampsia. The identified risk factors were utilised to construct the nomogram, which was subsequently internally validated using receiver operating characteristic (ROC) analysis and calibration curve assessment.

Results: According to our multivariate analysis, age, blood urea nitrogen (BUN), serum creatinine (Scr), mean arterial pressure (MAP), 24-h proteinuria, and CKD stage were identified as predictors of preeclampsia. Additionally, Scr, MAP, BUN, and 24-h proteinuria were found to be predictors of adverse pregnancy outcomes. The nomogram for predicting preeclampsia had an area under the ROC curve of 0.910, while the nomogram for predicting adverse pregnancy outcomes had an area under the ROC curve of 0.906. Both models demonstrated excellent discriminatory ability.

Conclusions: A nomogram based on 24-h proteinuria, serum creatinine, serum urea and age, and MAP allows predicting the occurrence of preeclampsia and other adverse pregnancy-related outcomes in CKD patients.