Validation of a novel risk score for severity of illness in acute exacerbations of COPD

BACKGROUND: Clinicians lack a validated tool for risk stratification in acute exacerbations of COPD (AECOPD). We sought to validate the BAP-65 (elevated BUN, altered mental status, pulse > 109 beats/min, age > 65 years) score for this purpose.

METHODS: We analyzed 34,699 admissions to 177 US hospitals (2007) with either a principal diagnosis of AECOPD or acute respiratory failure with a secondary diagnosis of AECOPD. Hospital mortality and need for mechanical ventilation (MV) served as co-primary end points. Length of stay (LOS) and costs represented secondary end points. We assessed the accuracy of BAP-65 via the area under the receiver operating characteristic curve (AUROC).

RESULTS: Nearly 4% of subjects died while hospitalized and approximately 9% required MV. Mortality increased with increasing BAP-65 class, ranging from < 1% in subjects in class I (score of 0) to > 25% in those meeting all BAP-65 criteria (Cochran-Armitage trend test z = -38.48, P < .001). The need for MV also increased with escalating score (2% in the lowest risk cohort vs 55% in the highest risk group, Cochran-Armitage trend test z = -58.89, P < .001). The AUROC for BAP-65 for hospital mortality and/or need for MV measured 0.79 (95% CI, 0.78-0.80). The median LOS was 4 days, and mean hospital costs equaled $5,357. These also varied linearly with increasing BAP-65 score.

CONCLUSIONS: The BAP-65 system captures severity of illness and represents a simple tool to categorize patients with AECOPD as to their risk for adverse outcomes. BAP-65 also correlates with measures of resource use. BAP-65 may represent a useful adjunct in the initial assessment of AECOPDs.