Abstract
BACKGROUND: The global prevalence of diabetes is on an upward trajectory. The management of complications related to the condition has seen limited progress in recent years. Klotho, characterized as an anti-aging protein that mitigates oxidative stress and inflammation, has previously been correlated with all-cause mortality in the broader United States population. The objective of this research was to investigate the persistence of this relationship among diabetic patients. METHODS: This study meticulously analyzed data (2007-2016) sourced from the National Health and Nutrition Examination Survey, encompassing a cohort of 3,560 individuals. To elucidate the links of Klotho with all-cause mortality in diabetic patients, a multivariate Cox proportional hazards regression model was employed. The relationship was further explored using the restricted cubic spline model, threshold analysis, and subgroup analysis. Additionally, a mediation analysis was conducted to unravel the influence of age on the observed correlations. RESULTS: Throughout the observation period, which had a median duration of 84 months, the incidence of all-cause mortality reached 18.511%. The Cox model analysis revealed a statistically significant association between Klotho levels and all-cause mortality. Further, the application of restricted cubic splines revealed a nuanced, nonlinear relationship between exposure factors and outcome across the entire study population (nonlinear P < 0.001), pinpointing a critical threshold at 829.138 pg/mL. Subgroup analyses showed consistent correlation between Klotho levels and mortality across various groups. Intriguingly, mediation analysis indicated that age was a significant mediator, accounting for 76.1% of the observed correlation of Klotho levels with all-cause mortality among diabetic patients. CONCLUSIONS: Low levels of Klotho were found to be strongly associated with an increased risk of all-cause mortality in individuals with diabetes (Klotho levels < 829.138 pg/ml), and a nonlinear relationship was observed between these two variables. These associations were largely mediated by age.