Objective: Outcomes in patients with ARDS/acute lung injury (ALI) may be dependent on the underlying cause. We describe the case mix, clinical behavior, and outcomes of patients with ALI/ARDS resulting from pulmonary causes (ALI/ARDSp) and extrapulmonary causes (ALI/ARDSexp).
Design: Retrospective study conducted between January 2001 and June 2005.
Setting: Respiratory ICU (RICU) of a tertiary care hospital in northern India.
Patients: All patients fulfilling the criteria for ALI/ARDS and requiring mechanical ventilation for > 24 h.
Measurements and results: Of the 180 patients (ARDS, 140 patients; ALI, 40 patients), 123 patients had ALI/ARDSp, whereas 57 patients had ALI/ARDSexp. The most common cause of ALI/ARDSp was infective pneumonia, whereas the most common cause of ALI/ARDSexp was sepsis. At ICU admission, although patients with ALI/ARDSexp were sicker than those with ALI/ARDSp, there was no difference between the two groups of patients in the development of new organ dysfunction/failure (Δ sequential organ failure assessment [SOFA] scores) or the time to develop the first organ dysfunction/failure (assessed by SOFA scores). The median length of RICU stay was similar in the two groups (5 days [interquartile range (IQR), 6 days] vs 5 days [IQR, 9.5 days], respectively, in patients with ALI/ARDSp and ALI/ARDSexp; p = 0.4). The hospital mortality rate was 47.8% and was not significantly different between the two groups (ALI/ARDSp group, 43.1%; ALI/ARDSexp group, 57.9%; p = 0.06). Multivariate analysis showed the following risk factors for death in the ICU: female gender (odds ratio [OR], 0.49; 95% confidence interval [CI], 0.25 to 0.94); SOFA scores (OR, 1.18; 95% CI, 1.07 to 1.3); and ΔSOFA scores (OR, 1.24; 95% CI, 1.09 to 1.41). There was no significant effect of the category of ARDS on outcome (OR, 1.6; 95% CI, 0.8 to 3.2).
Conclusions: Although patients with ALI/ARDSexp are sicker on ICU admission, the underlying cause of ARDS does not affect the length of ICU stay or hospital survival time.