Clinical risk scores and blood biomarkers as predictors of long-term outcome in patients with community-acquired pneumonia: a 6-year prospective follow-up study.

Abstract

OBJECTIVE: Prediction of long-term outcomes in patients with community-acquired pneumonia (CAP) is incompletely understood. We investigated the value of clinical risk scores [pneumonia severity index (PSI) and CURB-65] (Confusion, Urea, Respiratory rate, Blood Pressure, Age >65 years) and blood biomarkers of different physiopathological pathways in predicting long-term survival in a well-characterized cohort of patients with CAP enrolled in an antibiotic stewardship trial. DESIGN, SETTING AND SUBJECTS: Patients admitted with CAP to six medical centres in Switzerland were prospectively followed for 6 years. Cox regression models and area under the receiver operating characteristics curve (AUC) were used to investigate associations between initial risk assessment and all-cause mortality. MAIN OUTCOME MEASURE: All-cause mortality during a 6-year follow-up period. RESULTS: Six-year mortality in the present cohort (median age 73 years) was 45.1% [95% confidence interval (CI) 41.8-48.3%]. Initial PSI and CURB-65 scores both had excellent long-term prognostic accuracy, with a stepwise increase in mortality per risk class. The hazard ratios (95% CI) of the highest PSI and CURB-65 classes (reference: lowest class) were 38.0 (14.0-103.0) and 7.8 (2.2-14.5), respectively, after 6 years. The addition of inflammatory (pro-adrenomedullin) and cardiac (pro-atrial natriuretic peptide) blood biomarkers measured upon hospital admission further improved the prognostic capabilities of the PSI (AUC increase from 0.79 to 0.83; P < 0.0001) and the CURB-65 score (AUC increase from 0.73 to 0.80; P < 0.001). CONCLUSION: Risk assessment using clinical scores allowed accurate long-term prognostication, which was further improved by the addition of two inflammatory (pro-adrenomedullin) and cardiac (pro-atrial natriuretic peptide) blood biomarkers. These data provide a rationale for a more risk-adapted, 'personalized' strategy for long-term management of patients with CAP.