PREDICT Tool for Pregnancy-Associated CKD Progression.

Authors

Elizabeth Ralston, Mairéad Hamill, Shalini Santhakumaran, Michelle Hladunewich, Graham Smith, Lavanya Bathini, Nivethika Jeyakumar, Amit X Garg, Kate Bramham, PREDICT Investigation Group9, RaDaR Consortium, PREDICT Investigation Group

Published in

Kidney international reports. Volume 11. Issue 9. Pages 106636. Epub Jun 06, 2026.

Abstract

Prepregnancy counselling is recommended for women with chronic kidney disease (CKD) to discuss potential adverse outcomes; however, no tools exist to estimate individual risk. We aimed to develop and externally validate 2 prediction models for outcomes prioritized by people with CKD and health care professionals: The primary outcome was the probability of ≥25% reduction in estimated glomerular filtration rate (eGFR) or kidney replacement therapy (KRT) within 12 months postpartum. The secondary outcome was the probability of small-for-gestational-age (SGA) (< 3rd percentile) infant and/or preterm delivery (< 34 weeks).
The development cohort used linked data from the National Registry of Rare Kidney Disease (RaDar), UK Renal Registry (UKRR) and NHS Hospital Episode Statistics (HES). Individuals with eGFR < 90 ml/min per 1.73 m² within 24 months preconception and deliveries between 1997 and 2021 were included. Validation cohorts were as follows: (i) Ontario Pregnancy Cohort (2007-2022) and (ii) 3 UK pregnancy-CKD studies (2010-2018). Candidate predictors were selected from known risk factors. Clinically relevant cut-points were determined with people with CKD.
The development cohort included 746 women (median prepregnancy eGFR: 58 ml/min per 1.73 m²); validation cohorts included 6974 and 380 women. Optimal cut-points of 0.15 (sensitivity: 90%, negative predictive value (NPV): 85%) and 0.10 (sensitivity: 90%, NPV: 80%) were selected for the primary and secondary outcomes. External validations demonstrated high sensitivity and NPV for the primary outcome. Although comparability is limited by differing end points, Kidney Failure Risk Equation (KFRE) predictions (2-year) were lower than our PREDICT model (1-year) (median risk 1.4% vs. 48%).
We developed high performing models for individuals with CKD to predict coselected adverse kidney and neonatal outcomes from contemporaneous cohorts. Individualized pregnancy risk assessment tools could support future parents and health care professionals to make informed choices.

PMID:
42438723
Bibliographic data and abstract were imported from PubMed on 13 Jul 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.