Acute lung injury in patients with subarachnoid hemorrhage: incidence, risk factors, and outcome.

OBJECTIVE: Pulmonary complications account for significant morbidity and mortality following aneurysmal subarachnoid hemorrhage; however, the effect of acute lung injury is largely unknown. The goal of this study was to determine the incidence of acute lung injury in a large cohort of patients with subarachnoid hemorrhage as well as determine the risk factors for acute lung injury and its effect on mortality and length of stay. Ventilator management was analyzed to determine the proportion of patients with subarachnoid hemorrhage and acute lung injury who a received a low-tidal volume ventilation strategy.

DESIGN: Retrospective cohort study.

SETTING: University-affiliated county hospital in Seattle, WA.

PATIENTS: Six-hundred and twenty patients with aneurysmal subarachnoid hemorrhage documented on computed tomography or angiography.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN RESULTS: One-hundred and seventy patients met criteria for acute lung injury (incidence, 27%; 95% confidence interval, 24-31%). On multivariate analysis, severity of illness, clinical grade of hemorrhage, packed red blood cell transfusions, and severe sepsis in the intensive care unit were independently associated with development of acute lung injury. After adjustment for important confounders, development of acute lung injury was associated with a statistically significant increase in hospital mortality (odds ratio, 1.63; 95% confidence interval, 1.03-2.57). Acute lung injury was also independently associated with an increased intensive care unit length of stay (15%, 95% confidence interval, 5-27%). Thirty percent of patients with acute lung injury received low tidal volume ventilation. Patients receiving low tidal volume ventilation had worse oxygenation and higher positive end-expiratory pressure requirements compared with those who did not, but there were no significant differences in arterial pH or Pco2.

CONCLUSIONS: Acute lung injury is common in patients with subarachnoid hemorrhage and is independently associated with a worse clinical outcome. Research is needed to determine the causes of acute lung injury in this population and whether these patients are candidates for evidence-based ventilator strategies to reduce mortality.