Abstract
OBJECTIVE: Malnutrition is one of the most common complications in acute decompensated heart failure (ADHF). This study investigated the predictive value of a modified prognostic nutritional index (PNI)-the C-reactive protein-to-albumin-to-lymphocyte (CALLY) index-for short-term mortality in ADHF patients, while accounting for the potential interactive effects of participants' glycemic status. METHOD: The data were derived from the Jiangxi-ADHF II study cohort, which included 1,225 ADHF patients. The Boruta algorithm was employed to identify key prognostic features associated with mortality in ADHF and rank their predictive importance. Subsequently, multivariate Cox regression analysis and receiver operating characteristic curve analysis were conducted to evaluate and compare the prognostic significance of the PNI and CALLY index in predicting short-term mortality in ADHF patients. Exploratory subgroup analyses, including diabetes subgroups, were performed to assess the generalizability of these findings across populations. RESULTS: During the 30-day observation period, 109 (8.9%) participants experienced mortality. Using the Boruta algorithm, the CALLY index was identified as a key factor associated with ADHF-related mortality. In mortality risk assessment, the CALLY index demonstrated a stronger inverse association with mortality risk in ADHF patients compared to PNI. Quartile-based analysis revealed significantly higher mortality risks associated with low CALLY index relative to low PNI (HR: Q1 4.21 vs. 3.32). For mortality outcome prediction, the CALLY index (AUC = 0.80) was significantly superior to the PNI. Exploratory subgroup analyses further revealed that glycemic metabolic status may act as a significant interaction term in the association between the CALLY index and short-term prognosis in ADHF: compared to non-diabetic ADHF patients, those with comorbid diabetes exhibited a stronger inverse association between the CALLY index and 30-day mortality risk. This finding implies that diabetes significantly amplifies the mortality risk associated with low CALLY index. CONCLUSION: In conclusion, the CALLY index, modified based on the PNI, serves as a valuable prognostic tool for short-term outcomes in ADHF patients, with special attention required regarding the potential inhibitory effect of diabetes status on the CALLY index. The promotion of early risk stratification awareness and implementation of CALLY index screening in ADHF patients should be encouraged, particularly in those with comorbid diabetes.