ABSTRACT
Objective:To explore the prognostic factors predicting the recovery of elderly patients with hospital-acquired pneumonia(HAP).Methods:Data of HAP inpatients aged over 60 years in the First Affiliated Hospital of Nanjing Medical University between October 2015 and October 2020 were collected.Body mass index(BMI), neutrophil/lymphocyte rate(NLR), Charlson Comorbidity Index(CCI), Combined Comorbidity Score(CCS)and other data were retrospectively analyzed.The predictive value of the related factors was evaluated by using the Student's t test, the Logistic regression model and the receiver operating characteristic curve(ROC). Results:A total of 200 cases were enrolled in this study, and grouped into survival group(n=158)and death group(n=42). There were significant differences between the death group and the survival group in length of age, BMI hospital stay, state of consciousness, swallowing function, indwelling gastric tube, the use of proton pump inhibitors, leukocyte count, lymphocyte count, neutrophil count, hemoglobin, NLR, albumin, alanine aminotransferase, aspartate aminotransferase, D-dimer, C-reaction protein, procalcitonin(all P<0.05). Scores of CCI and CCS were higher in the death group than in the survival group[(6.79±2.86) vs.(3.42±1.98), (6.21±3.08) vs.(1.66±1.94), t=-7.193, -9.116, both P<0.001]. Multivariate Logistic analysis showed that age>86 years( OR=1.155, 95% CI: 1.014-1.316), BMI<21.77 kg/m 2( OR=0.651, 95% CI: 0.480-0.883), neutrophil count>10.10×10 9/L( OR=1.208, 95% CI: 1.025-1.422), C-reaction protein>59.32 mg/L( OR=1.055, 95% CI: 1.018-1.093), CCS>6.21 scores( OR=2.859, 95% CI: 1.559-5.244)were risk factors for death inpatients aged 60 years and older with HAP.Area under the ROC curve showed that CCS could better predict the mortality of elderly patients than CCI, and the areas under the ROC curve were 0.831(95% CI: 0.753-0.909)and 0.898(95% CI: 0.850-0.946)respectively(both P< 0.001). Conclusions:The elderly patients with HAP are inclined to multiplecomplications and high mortality rates.Combined application of multiple assessment systems and clinical indicators can improve the ability to predict the outcome of HAP.