Delirium in a Latin American intensive care unit. A prospective cohort study of mechanically ventilated patients.

OBJECTIVE: To establish the prevalence of delirium in a general intensive care unit and to identify associated factors, clinical expression and the influence on outcomes.

METHODS: This was a prospective cohort study in a medical surgical intensive care unit. The Richmond Agitation-Sedation Scale and Confusion Assessment Method for the Intensive Care Unit were used daily to identify delirium in mechanically ventilated patients.

RESULTS: In this series, delirium prevalence was 80% (N = 184 delirious patients out of 230 patients). The number of patients according to delirium psychomotor subtypes was as follows: 11 hyperactive patients (6%), 9 hypoactive patients (5%) and 160 mixed patients (89%). Multiple logistic regression modeling using delirium as the dependent outcome variable (to study the risk factors for delirium) revealed that age > 65 years, history of alcohol consumption, and number of mechanical ventilation days were independent variables associated with the development of delirium. The multiple logistic regression model using hospital mortality as the dependent outcome variable (to study the risk factors for death) showed that severity of illness, according to the Acute Physiology and Chronic Health Evaluation II, mechanical ventilation for more than 7 days, and sedation days were all independent predictors for excess hospital mortality.

CONCLUSION: This Latin American prospective cohort investigation confirmed specific factors important for the development of delirium and the outcome of death among general intensive care unit patients. In both analyses, we found that the duration of mechanical ventilation was a predictor of untoward outcomes.