Abstract
OBJECTIVE: The study evaluated the effectiveness of the sarcopenia indices neutrophils/lymphocytes, platelets/lymphocytes, AST/ALT, and creatinine (Cr)/ cystatin C (CysC)*100 in predicting mortality in hospitalized patients with Alzheimer's disease (AD) aged 60 years or older. MEASUREMENTS: This retrospective observational survey was undertaken in a teaching hospital in western China from January 1, 2017, to December 30, 2022. The neutrophil/lymphocyte, platelet/lymphocyte, AST/ALT, and Cr/CysC*100 ratios were used to assess the presence of sarcopenia, with the upper quartiles used as the cutoff value. Information on all-cause mortality was obtained through telephone interviews or electronic medical records between June 1, 2024, and June 20, 2024. Overall survival (OS) represented the time from hospital admission to death/final follow-up. Cox proportional hazards models were applied to determine the relationships between the above parameters and mortality from all causes. RESULTS: The information on 523 patients with AD was retrieved from the electronic medical record system. Of these, 329 were finally enrolled, all of whom were hospitalized and over the age of 60 years. The use of Cr/Cys C*100 as a sarcopenia indicator was found to be effective in predicting mortality (24.39% vs. 13.77% for patients with sarcopenia vs. those without, P = 0.024). However, the application of neutrophils/lymphocytes, platelets/lymphocytes, and AST/ALT as indicators showed no marked differences between the sarcopenia and non-sarcopenia participants. After further logistic regression analysis and correction of possible variables, participants with sarcopenia had an increased risk of death relative to those without (HR = 2.179, 95%CI: 1.175-4.044). CONCLUSIONS AND IMPLICATIONS: This study showed that only Cr/CysC*100 was effective in the prediction of mortality in older individuals with AD and sarcopenia and that neutrophils/lymphocytes, platelets/lymphocytes, and AST/ALT were not effective as predictors.