Preterm Birth and Other Adverse Pregnancy Outcomes in Women With Chronic Kidney Disease: A Systematic Review with Exploratory Meta-Analyses in Lupus and Diabetic Nephropathy

Abstract

Background: Chronic kidney disease complicates interpretation of pregnancy risk because exposure definitions, denominators, underlying diseases, and clinical practices vary. This review evaluates associations with preterm birth and other adverse pregnancy outcomes while preserving these distinctions. Objective: To evaluate associations between pre-existing chronic kidney disease and preterm live birth and other adverse maternal and neonatal outcomes, and to determine which disease-specific estimates could be combined without conflating populations, outcome definitions, or crude and adjusted effects. Methods: PubMed and Europe PMC were searched with supplementary renal disease and microalbuminuria searches. Comparative observational studies were assessed using a prespecified protocol registered as CRD420261503668. Extraction distinguished established CKD, reduced kidney function, and disease-matched comparisons. Exploratory sensitivity analyses used crude risk ratios, restricted maximum likelihood, and modified Hartung–Knapp intervals. Results: Thirty-four reports were mapped; eight contributed to three disease-matched crude analyses. The prespecified adjusted primary CKD estimate could not be pooled. Pooled risk ratios were 1.12 (95% confidence interval 0.13–9.35) for lupus-nephritis preterm birth, 2.79 (0.84–9.25) for lupus-nephritis preeclampsia, and 1.35 (1.21–1.51) for diabetic-nephropathy cesarean delivery. Adding a quiescent-nephritis cohort yielded a preterm-birth ratio of 1.24 (0.39–3.97). One cohort contributed 92.1% of the cesarean-analysis weight. Certainty was very low. Conclusion: Diabetic nephropathy was associated with higher crude cesarean-delivery risk, whereas lupus-nephritis estimates remained highly imprecise. Confounding, repeated pregnancies, and restricted evidence availability limit clinical inference.