Performance of CURB-65 and CURB-age in community-acquired pneumonia

Background: Community-acquired pneumonia (CAP) is common and associated with significant mortality. In this study, we validated a newly proposed severity assessment rule for CAP, CURB-age, and also compared with to the currently recommended criteria in UK, CURB-65. Methods: We conducted a prospective study in three hospitals in Norfolk and Suffolk, UK. One hundred and ninety patients were included and followed up for 6 weeks. Results: Of 190 patients, 100 were men (53%). The age range was 18-101 years (median 76 years). Sixty-five (34%) had severe pneumonia by CURB-65 and 54 (28%) had severe pneumonia by CURB-age. There were 54 deaths during follow-up. There were 32 deaths (50%) in severe and 22 deaths (18%) in non-severe group by CURB-65. There were 27 deaths each in both the groups by CURB-age (50% of severe cases and 20% of non-severe cases). For CURB-65, sensitivity, specificity, and positive and negative predictive values were 59.3% (45.0-72.4), 75.7% (67.6-82.7), 49.2% (36.6-61.9) and 82.4% (74.6-88.6), respectively. For CURB-age, the respective values were 50.0% (31.1-63.9), 80.1% (72.4-86.5), 50.0% (36.1-63.9) and 80.1% (72.4-86.5). Exclusion of patients aged < 65 years did not alter the results. Conclusions: Despite better specificity in correctly identifying 6-week mortality for CAP, CURB-age appears to be less sensitive than CURB-65. Our findings further assure the usefulness of CURB-65 for predicting mortality in CAP.