Fauzijah Sri Rahmawati, Yani Jane Sugiri, Sanarto Santoso, Asri Maharani

Abstract
Background : Community-acquired pneumonia (CAP) remains as a significant problem in healthcare. Decision to admit the patient
become a serious attention. This has led to the development of prognostic scoring system aimed at assisting in risk stratification of
patients presenting with CAP. The aim of this study was to test validity of PSI and CURB-65 severity scoring systems in CAPpatients
who admitted at Dr. Saiful Anwar Hospital Malang.

Methods : In this prospective descriptive study, 158 admitted CAP patients were enrolled. The severity of CAP were evaluated by
PSI and CURB-65 scoring system at the same time. In-hospital death or death within 30 days of discharge was the main outcome
study.

Results : PSI could predict death from CAP with 31.37% sensitivity and 85.98% specificity; while CURB has sensitivity of 13.73%
and specificity of 98.13%. PSI has higher area under the ROC curve,0,682 (95% CI 59,6%-76,8%, dengan p=0,000) compare with
CURB-65,0,625 (95% CI 52,8%-72,3%, dengan p=0,011). PPV (77,78%), LR+ (7,343), and OR of CURB-65 are superior to PSI
(PPV 51,61%; LR+ 2,238; OR 2,804).

Conclusion : Sensitivity to predict death from CAP of PSI was better than that of CURB-65, but specificity of CURB-65 was higher
than that of PSI. CURB-65 was better than PSI in predicting the mortality of CAPin this population. (J Respir Indo. 2013; 33:26-33)

Keywords : Community-acquired pneumonia, PSI, CURB-65, mortality.

Click for details