Incidence of acute lung injury in the United States*

Abstract

OBJECTIVE: Recent estimates of acute respiratory distress syndrome (ARDS) incidence have varied from 1.3 to 22 per 100,000 person years (105 person.years); the incidence of acute lung injury (ALI) has varied from 17.9 to 34 cases per 105 person.years. Potential reasons for this wide range include differences in the definition of the syndrome, in the populations sampled, and in the assumptions made to estimate incidence. We hypothesized that careful, prospective, protocol-driven case identification that included the milder hypoxemia criterion for ALI would yield incidence numbers greater than previously reported. DESIGN: Prospective cohort study with extrapolation to the U.S. population. SETTING: National Heart, Lung, and Blood Institute-sponsored ARDS Network composed of 20 hospitals. PATIENTS: As part of the National Institutes of Health-sponsored ARDS network, 20 hospitals prospectively identified patients with ALI from 1996 to 1999. Screening logs from this study were used to estimate ALI rates per intensive care unit (ICU) bed per year. We used the registry and data from the American Hospital Association to estimate the incidence of ALI in the United States. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The ALI per ICU bed incidence in the ARDS network registry varied from 0.7 to 5.8 cases of ALI per ICU bed per year with an average of 2.2 cases of ALI per ICU bed per year. We tested the robustness of the incidence estimate by performing a variety of sensitivity analyses. When we used the conservative assumptions that the ARDS network screening logs were complete at each of the participating hospitals and that ALI cases are limited to academic hospitals with > or =20 adult ICU beds, the incidence of ALI in adults in the United States is 22.4 cases per 105 person.years. Under the less conservative assumption that ALI cases occurred only at hospitals with > or =20 ICU beds, regardless of their academic status, the incidence of ALI in the United States is estimated at 64.2 cases per 105 person.years. CONCLUSIONS: Based on this analysis, which used prospective clinical trial screening data and conservative assumptions about where patients with ALI are cared for, the incidence of ALI in the United States appears to be higher than previously reported.