Accelerated biological aging and homeostatic dysregulation in diabetic kidney disease: Evidence from NHANES and Mendelian randomization

Accelerated biological aging may drive diabetic kidney disease (DKD) pathogenesis, but comprehensive evidence integrating observational and causal approaches remains limited. This study integrated cross-sectional analyses of 2418 diabetic adults (National Health and Nutrition Examination Survey 1999–2010) with bidirectional Mendelian randomization (MR). Associations between homeostatic dysregulation (HD), biological age acceleration (Klemera–Doubal method residual) (BioAgeAccel), and Phenotypic age acceleration (PhenoAgeAccel) with DKD were assessed using weighted logistic regression, restricted cubic splines, receiver operating characteristic curve analysis and mediation. Two-sample MR utilized genetic instruments for BioAgeAccel and PhenoAgeAccel with FinnGen DKD data (3283 cases/181,704 controls). Significant nonlinear threshold effects were observed: DKD risk increased markedly at BioAgeAccel > 2.13 years and HD > 3.16 units, while PhenoAgeAccel exhibited a linear dose-response relationship. Accelerated biological aging status (BioAgeAccel > 0) substantially increased DKD odds (OR = 3.25, 95% CI: 2.52–4.18), as did PhenoAgeAccel > 0 (OR = 1.44, 95% CI: 1.13–1.84). Similarly, highest-quartile HD conferred elevated risk (Q4 OR = 2.85, 95% CI: 1.87–4.36). Associations remained consistent across subgroups stratified by age, sex, body mass index, income, comorbidities, and lifestyle factors, though effect modification occurred with stroke history (BioAgeAccel/PhenoAgeAccel) and ethnicity/age (HD). Urinary albumin-to-creatinine ratio mediated 89% to 93% of associations between all aging metrics and DKD. Discriminatory accuracy was highest for HD (area under the curve (AUC) = 0.678, 95% CI: 0.654–0.703) versus BioAgeAccel (AUC = 0.660, 95% CI: 0.637–0.682) and PhenoAgeAccel (AUC = 0.562, 95% CI: 0.539–0.586). MR confirmed a causal effect of PhenoAgeAccel on DKD (IVW OR = 1.078, 95% CI: 1.029–1.129, P = .0014) but not BioAgeAccel. Accelerated biological aging independently predicts DKD risk, with PhenoAgeAccel demonstrating causal effects. HD emerges as a potent clinical predictor, while albuminuria mediates most aging-related renal damage. These findings advocate integrating biological aging assessment into DKD risk stratification frameworks.