Abstract
BACKGROUND: Cardiovascular-kidney-metabolic (CKM) syndrome is a recently proposed condition encompassing metabolic dysfunction, chronic kidney disease, and cardiovascular diseases including coronary artery disease (CAD). Although concomitant metabolic dysfunction-associated steatotic liver disease (MASLD) exacerbates CKM syndrome, MASLD is not included in the original stratification (stages 0-3) for predicting cardiovascular disease (stage 4). METHODS: We investigated whether the addition of MASLD in the original health stage of CKM syndrome improves the risk stratification of CAD in Japanese participants who underwent annual health checkups. A total of 18 358 participants were categorized by the original health stages (stages 0-3) or modified health stages incorporating MASLD without chronic kidney disease in stage 2 and MASLD with chronic kidney disease in stage 3 to analyze the predictive ability for CAD during a 10-year period. RESULTS: Kaplan-Meier survival curve analysis showed that the modified classification more effectively stratified the risk of CAD than did the original CKM health stage. In Cox hazard proportional analyses after adjustment of confounders, hazard ratios in the original and modified stages 2/3 were 1.63 (95% CI, 1.27-2.11; P<0.001), 2.41 (95% CI, 0.34-17.4; P=0.381), 1.58 (95% CI, 1.23-2.05; P<0.001), and 2.56 (95% CI, 1.66-3.95; P<0.001), respectively. Risk discrimination significantly improved in the modified classification evaluated by integrated discrimination improvement. Furthermore, machine learning-mediated feature importance analyses using random forest and extreme gradient boosting identified MASLD as a key predictor of CAD with a high value of the Shapley Additive Explanations, indicating strong contribution in models. CONCLUSIONS: The modified health stage classification of CKM syndrome incorporating MASLD improves the accuracy of predicting new onset of CAD in a Japanese general population.