CURB-65 as an initial prognostic score in Internal Medicine patients

Abstract

Background: There are few validated tools to assess severity in patients admitted to an Internal Medicine service. The aim of this study was to evaluate if CURB-65 score, previously validated as mortality index in community acquired pneumonia, could also be used in those patients.

Methods: We analyzed prospectively all patients admitted to Hospital Sierrallana (Spain) from 1 March to 31 October 2010. Variables of the CURB-65 score (confusion, serum urea>7mmol/L (42mg/dl), respiratory rate≥30/min, systolic blood pressure<90mmHg and/or diastolic blood pressure≤60mmHg, and age≥65) and other clinical and epidemiological data and laboratory findings were recorded. Charlson comorbidity index was also estimated. Prognostic variables were identified using multiple logistic regression with 30days mortality as the outcome measure.

Results: 539 patients were studied (51% males; mean age: 78±14years; mortality 12%). A high CURB-65 score was a significant predictor of 30day mortality (p<0.001). Eighty-six percent of the patients who died had high CURB-65 score at admission, and none of them had low score. Sensitivity and specificity for high CURB-65 score were 86% and 70%, respectively, and negative predictive value was 97%. Receiver-operator characteristic curve showed an area under curve of 0.79 for CURB 65-score. Charlson index also correlated with mortality, but its performance was worse than that of CURB-65.

Conclusion: Our findings suggest that CURB-65 score may be a simple and useful tool to help clinicians in establishing the prognosis of patients admitted to general Internal Medicine wards.