Diabetic kidney disease (DKD) substantially contributes to premature mortality in individuals with Type 2 diabetes (T2D). Although the 2024 American Diabetes Association (ADA) guideline incorporates the risk stratification of DKD progression into routine diabetes care, its predictive validity for mortality remains underexplored in real-world populations. Using data from 6936 adults with T2D enrolled in the US National Health and Nutrition Examination Survey (NHANES) from 1999 to 2018, this study examined whether baseline DKD progression risk categories-based on estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (UACR)-can predict all-cause and cardiovascular mortality. Risk of DKD progression was classified as low, moderately increased, high, or very high. Mortality outcomes were identified via linkage to the National Death Index until December 31, 2019. Cox proportional hazards models were used to estimate hazard ratios (HRs), adjusting for demographics, comorbidities, lifestyle behaviors, laboratory values, and medication use. Over a median follow-up of 98 months, 1781 all-cause and 623 cardiovascular deaths occurred. Compared with the low-risk group, the very high-risk group had significantly higher risks of all-cause mortality (adjusted HR: 2.53; 95% CI: 2.15-2.98) and cardiovascular mortality (adjusted HR: 2.47; 95% CI: 1.87-3.27), with a clear dose-response gradient across risk categories (p for trend < 0.001). Notably, this graded relationship remained stable across most examined subgroups, and significant interactions were observed specifically among individuals aged < 65 years and those using renin-angiotensin system (RAS) inhibitors. These findings confirm the robust prognostic utility of the ADA-endorsed risk stratification matrix of DKD progression for mortality in T2D and highlight its potential to inform risk-based individualized management strategies in clinical practice. This study also provides pragmatic evidence supporting the implementation of guideline recommendations.
Qiu et al. (Thu,) studied this question.