Bacteremia in patients with ventilator-associated pneumonia is associated with increased mortality: A study comparing bacteremic vs. nonbacteremic ventilator-associated pneumonia

Objective: To assess whether bacteremic ventilator-associated pneumonia (B-VAP) differs in terms of risk factors, organisms, and outcomes from nonbacteremic VAP (NB-VAP). Design: A retrospective, single-center, observational, cohort study. Setting: Multidisciplinary teaching intensive care unit. Patients: Adult patients requiring mechanical ventilation, identified as having VAP in a 44-month prospective surveillance database. Interventions: Each B-VAP patient was matched with two controls with VAP and negative blood cultures based on the microbial etiology responsible for VAP, Acute Physiology and Chronic Health Evaluation 11 score on admission (+/- 3 points), diagnostic category, and length of stay before pneumonia onset. Measurements and Main Results: B-VAP was documented in 35 (17.6%) of 199 microbiologically confirmed VAP episodes. B-VAP developed later (median 8 vs. 5 days, p =.03) and was more frequent in previously hospitalized patients (34.3% vs. 11.0%, p <.01) and in older patients (57.4 +/- 15.2 vs. 49.5 +/- 19.3 yrs, p =.02). B-VAP was more often caused by methicillin-resistant Staphylococcus aureus (12 [20.7%] vs. 13 [5.1%] episodes, p <.01), whereas Haemophilus influenzae, was associated with NB-VAP (52 [20.4%] vs. 0, p <.01). Multivariate analysis confirmed an association between B-VAP and both methicillin-resistant S. aureus (odds ratio 3.18; 95% confidence interval 1.15- 8.76, p <.01) and prior hospitalization (odds ratio 2.56; 95% confidence interval 1.01-6.54, p =.05). After adjustment for potential confounders, B-VAP (hazard ratio for death 2.55; 95% confidence interval 1.25-5.23, p =.01) and vasopressor use (hazard ratio 2.43; 95% confidence interval 1.23-4.82, p =.01) remained associated with mortality. The estimated relative risk of death for bacteremic cases was 2.86 (95% confidence interval 1.09-7.51), since mortality for cases and matched NB-VAP controls was 40.6% (13 of 32) and 19.3% (11 of 57), respectively. Conclusions: B-VAP occurs later during intensive care unit stay, is more frequent in previously hospitalized patients, is more often caused by methicillin-resistant S. aureus, and is independently associated with increased intensive care unit mortality.