Abstract
PURPOSE: Community-acquired pneumonia (CAP) is common among the elderly; it typically has a poor prognosis and high mortality. This study evaluated the factors predicting CAP-related in-hospital mortality in the elderly to identify a simpler and more accurate predictor. PATIENTS AND METHODS: This was a single-center, retrospective study. The data used in this study was collected from all older patients (≥65) with CAP admitted to our hospital between January 2012 and April 2020. RESULTS: A total of 2028 older patients with CAP were included; 121 (5.97%) died in hospital. Of the patients in the study, 1267 (62.5%) were men and 261 (12.9%) had a history of malignant tumors. After performing univariate and multivariate Cox regression analyses, sex, history of malignant tumor, CURB-65 score, neutrophil-to-lymphocyte ratio (NLR), hemoglobin level, and NLR*CURB-65 levels were associated with CAP mortality. By comparing the area under the receiver operating characteristic (ROC) curves of the predicted factors, the NLR*CURB-65 level used to predict CAP mortality in the elderly was 0.755, and was superior to other measurements. All included patients were then dichotomized into two groups based on NLR*CURB-65 level (≤9.06 and >9.06) according to the ROC analysis. Patients with a high NLR*CURB-65 level had higher in-hospital mortality than those with a low NLR*CURB-65 level. The two divided groups showed significant differences in age, sex, smoking history, comorbidity, and laboratory findings. This indicates that NLR*CURB-65 is a predictive index that could reflect the comprehensive condition of older patients with CAP. CONCLUSION: NLR*CURB-65 is a simpler and more accurate predictor of CAP-related in-hospital mortality in the elderly.