Survival and long-term functional outcome in 1,155 consecutive neurocritical care patients.

Objective: To analyze survival, mortality, and long-term functional disability outcome and to determine predictors of unfavorable outcome in critically ill patients admitted to a neurologic intensive care unit (neuro-ICU).

Design: Retrospective cohort study with post-neuro-ICU health-related evaluation of functional long-term outcome.

Setting: Ten-bed neuro-ICU in a tertiary care university hospital.

Patients: A consecutive cohort of 1,155 patients admitted to a neuro-ICU during a 36-month period.

Interventions: None.

Measurements And Main Results: A total of 1,155 consecutive patients, of whom 41% were women, were enrolled in the study. The predominant reasons for neuro-ICU care were cerebrovascular diseases, such as intracerebral hemorrhage (20%), subarachnoid hemorrhage (16%), and complicated, malignant ischemic stroke (15%). A total of 213 patients (18%) died in the neuro-ICU. The Glasgow Outcome Scale and modified Rankin scale were dichotomized into two groups determining unfavorable vs. favorable outcome (Glasgow Outcome Scale scores 1-3 vs. 4-5 and modified Rankin scale scores 2-6 vs. 0-1). Factors associated with unfavorable outcome in the unselected cohort according to logistic regression analysis were admission diagnosis, age (p < .01), and a higher score in the simplified Therapeutic Intervention Scoring System (TISS-28) at time of admission (p < .01). Functional long-term outcome was evaluated by telephone interview for 662 patients after a median follow-up of approximately 2.5 yrs by evaluating modified Rankin scale and Glasgow Outcome Scale scores. Factors associated with unfavorable functional long-term outcome were admission diagnosis, sex, age of >70 yrs (odds ratio, 8.45; 95% confidence interval, 4.52-15.83; p < .01), TISS-28 of >40 points at admission (odds ratio, 4.05; 95% confidence interval, 2.54-6.44; p < .01), TISS-28 of >40 points at discharge from the neuro-ICU (odds ratio, 3.50; 95% confidence interval, 1.51-8.09; p < .01), and length of stay (odds ratio, 1.01; 95% confidence interval, 1.00-1.03; p = .02).

Conclusion: We found admission diagnosis, age, length of stay, and TISS-28 scores at admission and discharge to be independent predictors of unfavorable long-term outcome in an unselected neurocritical care population.