Early Versus Late Initiation of Dialysis in End-Stage Kidney Disease Patients with Diabetes Mellitus: A Systematic Review, Meta-Analysis, and Meta-Regression on Mortality Risk

Introduction: Chronic kidney disease represents a significant global health burden, with dialysis as the most prevalent modality for end-stage kidney disease (ESKD) treatment. One of the major causes of ESKD is diabetes mellitus. Diabetic patients undergoing dialysis have higher mortality risk so optimal timing for its initiation is critical in maximizing survival and quality of life. This study aimed to explore the mortality risk of early versus late dialysis initiation in ESKD patients with diabetes. Methods: Systematic searches were conducted according to the PRISMA 2020 guidelines on PubMed, Scopus, ProQuest, and several databases through Web of Science up to 20 May 2025 (PROSPERO CRD420251074686). Effect sizes were presented as hazard ratios (HRs) with 95% confidence intervals (CIs) and 95% prediction intervals (PIs), pooled using a restricted maximum likelihood random-effects model. Subgroup and meta-regression analyses were also performed to search for potential confounding variables. Results: Eight studies involving 303,116 patients were included. Two studies defined early and late dialysis initiation using an estimated glomerular filtration rate (eGFR) cut-off of 5.0 mL/min/1.73 m2, while the remaining studies used cut-offs ranging from 7.0 to 7.7 mL/min/1.73 m2. The pooled hazard ratio (HR) showed no significant difference in the mortality risk between early and late initiation of dialysis in ESKD patients with diabetes mellitus (HR 1.02, 95% CI 0.79–1.31, p = 0.90, I2 = 97.87%, 95% PI 0.45–2.32). Sensitivity analysis showed that the pooled HR was robust. Subgroup analysis demonstrated no significant difference in the pooled HR according to different study designs. Meta-regressions also showed that the year of population sampling, mean age, and follow-up duration of mortality risk did not have significant associations with the pooled HR. Conclusions: Early dialysis initiation does not appear to confer a survival benefit in ESKD patients with diabetes mellitus. However, given the limited and heterogeneous evidence, further high-quality studies are needed. We suggest that dialysis initiation in this specific population should be guided by clinical indications.