CURB-65 criteria were proposed to help physicians determine the need for hospitalization by the British Thoracic Society. This study examines the CURB-65 criteria in a Utah Emergency Department population of community-acquired pneumonia patients.
We abstracted the computerized patient records and charts of all adult, immuno-competent, emergency department CAP patients 6/95-6/99 with a radiologically confirmed infiltrate, excluding nursing homes patients and those with cavitary disease or active malignancy. We evaluated the CURB 65 (confusion, urea, respiratory rate, blood pressure, age >65) criteria in relation to hospitalization, length of stay, and 30 day mortality obtained from Utah vital statistics data. We also examined SpO2 < 90% added to the CURB criteria.
Of the 883 patients (mean age: 58.6±22.5; 53.3% male; 4.7% died) 539 were admitted to the hospital (64.7%). There were no deaths among outpatients.
We confirm CURB 65 as a 30 day mortality predictor. Adding SpO2 <90% to CURB-65 improved its performance in predicting admission and length of stay.
Age >65 and SpO2 <90% may improve the value of the CURB criteria in predicting need for hospitalization. CURB-650 n = 2751 n = 2182 n = 2153 n = 1004 n = 225 n = 3Admitted/LOS (days)82 / 4.5149 /5.1188 / 5.296 / 5.621 / 7.13 / 4.5Mortality0%3.2%4.7%14%32%33%SpO2<90: admissions/LOS47 / 5.072 / 4.9116 / 5.161 / 5.610 / 4.53 / 4.5
N.C. Dean, None.