Assessment of pneumonia severity index and curb-65 to predict inpatient mortality from bacterial and viral etiologies of community-acquired pneumonia in the 21st century

SESSION TITLE: Pneumonia: New Drugs, Old Bugs, and Other Things for Your Consideration SESSION TYPE: Rapid Fire Original Inv PRESENTED ON: 10/10/2023 12:55 pm - 01:40 pm PURPOSE: Community-acquired pneumonia (CAP) is one of the leading causes of hospitalization and ICU admissions worldwide. In the US, the annual incidence of CAP is high and increases with age. Two of the most common pneumonia severity scoring systems that are used to identify high-risk patients are the Pneumonia Severity Index (PSI) and CURB-65. The aim of this study is to assess the discrimination of PSI and CURB-65 to predict inpatient mortality in patients hospitalized with CAP due to bacterial and non-COVID viral etiologies. METHODS: This was a retrospective analysis of a cohort of patients aged 18 years or older who were hospitalized with CAP from June 1st 2014 to May 31st 2016 from eight adult hospitals in Louisville, Kentucky. Patients were excluded if they had multiple etiologies detected. Vitals, lab values, and demographics were gathered from admission to calculate PSI and CURB-65. In-patient mortality data was then gathered. Receiver operator curve analyses and area under the curve (AUC) were performed to assess discrimination between PSI and CURB-65 by CAP etiology. 95% confidence intervals were determined by bootstrap resampling. RESULTS: A total of 7,208 patients met inclusion into the study. Viral CAP was present in 341 patients and bacterial CAP was present in 1,116 patients. In-hospital mortality was 3.5% for viral CAP, and 6.7% for bacterial CAP (p=0.040). By AUC, PSI demonstrated similar predictability for inpatient mortality based on etiology: 0.848 (95% CI 0.718-0.945) for viral CAP, and 0.820 (95% CI 0.765-0.866) for bacterial CAP. By AUC, CURB-65 also demonstrated similar predictability for inpatient mortality based on etiology: 0.831 (95% CI 0.698-0.929) for viral CAP, and 0.726 (95% CI 0.667-0.778) for bacterial CAP. Additionally, PSI and CURB-65 demonstrated similar predictability when compared within each etiologic group. CONCLUSIONS: The predictive ability for in-hospital mortality of these scoring systems does not substantially change with the etiology of CAP, despite the higher rate of in-hospital mortality among patients hospitalized with bacterial CAP compared to viral. Additionally, PSI and CURB-65 predict in-hospital mortality similarly when compared to each other within patient cohorts with viral or bacterial pneumonia. CLINICAL IMPLICATIONS: PSI and CURB-65 can continue to be utilized to assess in-hospital mortality risk and compare patient populations, regardless of etiology. DISCLOSURES: No relevant relationships by James Bradley No relevant relationships by Rodrigo Cavallazzi No disclosure on file for Thomas Chandler No disclosure on file for Stephen Furmanek No relevant relationships by Pooja Gandhi No relevant relationships by Julio Ramirez No relevant relationships by Matthew Wallace