[Cohort study on incidence of ARDS in patients admitted to the ICU and prognostic factors of mortality].
OBJECTIVE: Analyze acute respiratory distress syndrome (ARDS) in patients admitted to an Intensive Care Medicine Service (ICMS) and prognostic factors of mortality in these patients. DESIGN: Prospective study of all the patients admitted consecutively in the ICMS from January 1998 to February 2003. SCOPE: ICMS of a third level university site with 32 beds in its General Area and 10 beds in the Traumatology Area. PATIENTS: Patients who met the ARDS criteria of the European-North American Consensus Conference at any time during admission in ICMS. ENDPOINTS OF INTEREST: Mortality at 28 days. RESULTS: One hundred and ninety-one patients (3.4 of all the admissions in ICMS) had ARDS criteria. The origin of ARDS was intrapulmonary in 63%. A total of 77% of the patients had multiorgan dysfunction and 26% respiratory superinfection. Median stay in the ICMS was 20 days. Mortality at 28 days was 48% and hospital mortality 58%. Multivariant analysis showed that the variables associated independently with an increase in mortality were the following: APACHE II > 22 (odds ratio [OR] 2.7; 95% CI: 1.3-5.8; p = 0.007), minimum PaO2/FIO2 during evolution of ARDS < 81 mmHg (odds ratio 5.5; 95% CI: 2.6-11.9; p < 0.0001), dysfunction > or = 3 organs (odds ratio 11.8; 95% CI: 2.5-55.4; p = 0.002). CONCLUSIONS: ARDS is an entity with elevated mortality whose prognosis is associated not only with the seriousness of pulmonary function deterioration but also of systemic function, on which some treatment could modulate its evolution.