INTRODUCTION: Respiratory infections account for significant mortality and morbidity throughout the world. In spite of advances in the development of newer antibiotics, respiratory infections continue to be an important cause for mortality. It is important to study the predictors of mortality in patients with respiratory infection in Indian patients as there is scarce data METHODS: A prospective observational study was done in a tertiary care center from August 2017 to February 2018. Patients admitted to ICU with acute respiratory infection were included in the study. Data on demographic variables such as age, gender, diagnosis, and reason for ICU admission were recorded. We also noted the presence of co-morbidities. Modified CPI score was calculated on day 1 and day 2. Cox regression univariate and multivariate analysis along with Kaplan Meier analysis was done for assessing factors associated with mortality. RESULTS: 303 patients satisfied the inclusion criteria and were included in the study. The mean age of the cohort was 56.05 ± 16.37 years and 62% were men. The most common diagnosis was Pneumonia (66%) followed by COPD (43.5%). Mean duration of hospital stay was 7.29 ± 3.76 days. The mortality rate was 17.8%. On multivariate Cox regression analysis CPI score >4 on day 1, CPI score >6 on day 2, number of antibiotics >2 were independently associated with increased hazard for mortality. CONCLUSION: We found on multinominal Cox regression analysis, modified CPI score >4 on day 1, CPI score >6 on day 2, Usage of >2 antibiotics were independently associated with increased hazard for mortality in patients admitted to ICU with a respiratory infection.
