Hospital Noninvasive Ventilation Case Volume and Outcomes of Acute Exacerbations of Chronic Obstructive Pulmonary Disease.

RATIONALE: Higher hospital case volume may produce local expertise ("practice makes perfect"), resulting in better patient outcomes. Associations between hospital noninvasive ventilation (NIV) case volume and outcomes for patients with acute exacerbations of chronic obstructive pulmonary disease (COPD) are unclear.

OBJECTIVES: To determine associations between total hospital NIV case volume for all indications and NIV failure and hospital mortality among patients with acute exacerbations of COPD.

METHODS: Using the 2011 California State Inpatient Database and multivariable hierarchical logistic regression, we calculated hospital-level risk-adjusted rates for NIV failure (progression from NIV to invasive mechanical ventilation) and hospital mortality among patients with acute exacerbations of COPD.

MEASUREMENTS AND MAIN RESULTS: We identified 37,516 hospitalizations for acute exacerbations of COPD in 252 California hospitals in 2011. Total hospital NIV use for all indications ranged from 2 to 565 cases (median, 64; interquartile range, 96). Hospital NIV failure rates for acute exacerbations of COPD ranged from 3.7 to 31.3% (median, 8.5%; interquartile range, 4.2). At the hospital level, higher total hospital NIV case volume was weakly associated with higher hospital NIV failure rates for acute exacerbations of COPD (r = 0.13; P = 0.03). Higher hospital NIV failure rates were weakly associated with higher hospital mortality rates for acute exacerbations of COPD (r = 0.15; P = 0.02), but higher total hospital NIV case volume was not associated with hospital mortality for exacerbations of COPD (r = -0.11; P = 0.08). At the patient level, patients admitted to high-NIV versus low-NIV case-volume hospitals had greater odds of NIV failure (quartile 4 vs. quartile 1 adjusted odds ratio [aOR], 1.95; 95% confidence interval [CI], 1.12-3.40). Compared with initial treatment with invasive mechanical ventilation, NIV failure was associated with higher odds of death (aOR, 1.81; 95% CI, 1.35-2.44). However, admission to high-NIV versus low-NIV case-volume hospitals was not significantly associated with patient in-hospital mortality (quartile 4 vs. quartile 1 aOR, 0.76; 95% CI, 0.57-1.02).

CONCLUSIONS: Despite strong evidence for use of NIV in the management of acute exacerbations of COPD, we observed no significant mortality benefit and higher rates of NIV failure in high-NIV case-volume hospitals. Further investigation of patient selection and hospital factors associated with NIV failure is needed to maximize favorable patient outcomes associated with use of NIV for acute exacerbations of COPD.