Objective: To assess the possibility that acute respiratory distress syndrome (ARDS) of pulmonary and nonpulmonary origins represent two distinct syndromes.
Design: Analysis of data collected prospectively from an inception cohort of 117 patients with ARDS.
Setting: Adult intensive care unit (ICU), university/postgraduate hospital.
Measurements and main results: Differences were sought in mortality and 6-month functional outcome between patients developing ARDS due to pulmonary (group 1) and nonpulmonary (group 2) pathology. Group 1 patients displayed a trend toward increased ICU and in-hospital mortality (42.1% vs. 23.2%, p = .10, and 47.4% vs. 27.9%, p = .11, respectively). No difference was found in ICU length of stay (46.3 +/- 4.9 vs. 39.0 +/- 4.8 days for groups 1 and 2, respectively) nor in duration of positive-pressure ventilation (26.2 +/- 4.3 vs. 20.6 +/- 3.2 days). However, the need for pressure-controlled inverse ratio ventilation was significantly greater in group 1 (16.9 +/- 3.2 vs. 9.1 +/- 1.3 days; p = .033). In survivors, reductions in total lung capacity at 6 months (68.1 +/- 4.6 vs. 61.8 +/- 4.6% predicted for groups 1 and 2, respectively; p = .4), accessible lung volume (74.4 +/- 4.4 vs. 68.9 +/- 4.9% predicted; p = .56), and forced expiratory volume (77.8 +/- 2.9 vs. 80.3 +/- 2.4% predicted; p = .57) did not differ between groups. Gas transfer coefficient was well preserved (84.5 +/- 4.6 vs. 86.6 +/- 4.7% predicted; p = .80).
Conclusions: These data suggest a trend toward higher mortality and ventilatory requirements in ARDS of direct etiology, generating a hypothesis worthy of further exploration.