Relationship between inhaled β2-agonists and ventilator-associated pneumonia: A cohort study*

Abstract

To determine the impact of aerosolized bronchodilators on ventilator-associated pneumonia. Prospective cohort study. A 30-bed medical and surgical intensive care unit. All intubated patients requiring mechanical ventilation for >48 hrs were eligible during a 13-month period. Nebulized β2-agonists were administered at the intensive care unit physician's discretion. Ventilator-associated pneumonia definition included clinical and quantitative microbiological criteria. Only first ventilator-associated pneumonia episodes were analyzed. Risk factors for ventilator-associated pneumonia were determined using univariate and multivariate analyses. The influence of inhaled β2-agonists on ventilator-associated pneumonia occurrence was also adjusted for confounding factors using Cox's proportional-hazards model. Ventilator-associated pneumonia was diagnosed in 137 (31%) of the 439 enrolled patients. Ventilator-associated pneumonia was early-onset in 14 (10%) patients. The incidence rate of ventilator-associated pneumonia was 20 per 1,000 ventilator days. Ventilator-associated pneumonia was polymicrobial in 16 (11%) patients, and related to multidrug-resistant bacteria in 42 (28%) patients. Most cases of ventilator-associated pneumonia were caused by Gram-negative bacteria. Inhaled β2-agonists were significantly more frequently used in patients with ventilator-associated pneumonia compared with those without ventilator-associated pneumonia (49% vs. 34%, odds ratio [95% confidence interval] = 1.9 [1.2-2.8], p = .003). Multivariate analysis identified aerosolized β2-agonists (odds ratio [95% confidence interval] = 1.7 [1.1-2.6], p = .012), Simplified Acute Physiology Score II at intensive care unit admission (odds ratio [95% confidence interval] = 1.01 [1.001-1.02] per point, p = .031), and red blood cell transfusion (odds ratio [95% confidence interval] = 2 [1.3-3.1], p = .001) as independent risk factors for ventilator-associated pneumonia. Cox's proportional-hazards model also identified inhaled β2-agonists as a risk factor for ventilator-associated pneumonia (odds ratio [95% confidence interval] = 1.52 [1.06-2.19], p = .021). Use of aerosolized bronchodilators in intensive care unit mechanically ventilated patients is an independent risk factor for ventilator-associated pneumonia.