Background and Goal of Study: Glasgow Coma Scale (GCS) is a simple, objective and fast
method of evaluating patients' neurologic state. There are several severity of disease score systems that include GCS as a neurologic indicator. Although it was demonstrated that it has prognostic value alone, there are few studies about GCS in evaluation of outcome of surgical patients in Intensive Care Unit. The aim of this study was to evaluate the prognostic value of GCS in a cohort of patients admitted to a surgical intensive care unit (SICU) and identify risk factors for lower GCS in postoperative period.
Materials and methods: After study approval by the institutional ethics committee, an observational retrospective study was conducted. Patients submitted to noncardiac,
nonneurologic elective and emergency surgery admitted at SICU (from Jan 2006 to July 2013) were included. Exclusion criteria: age < 18 years old; length of stay < 12h; patients readmitted in the context of initial admission in the study period. Patients were divided in two groups according to their GCS at SICU admission, GCS≥9 and GCS< 9. Patient´s demographics, intraoperative and postoperative data were collected. Descriptive analysis was performed and the MannWhitney U test, Fischer´s exact test or Chisquare were used. Univariate and multivariate analyses were done with logistic binary regression with calculation of an Odds Ratio and its 95% Confidence Interval. 
Results and discussion: From a total of 4565 patients, 4398 were included. There were 54
(1.23%) patients with GCS< 9. These patients were more often admitted for nonelective
surgery (p< 0.001). The group with GCS< 9 had more often a revised cardiac risk index >2 (p< 0.001),  but the only score's item that differ between groups was history of congestive heart disease (p=0.001). They were more likely to be admitted on mechanical ventilation, with higher FiO2 and higher PaO2 (p< 0.001). Patients who had GCS< 9 at admission to SICU had a higher incidence of major cardiac events (9.3% vs 2.3%, p< 0.001), had a longer stay in SICU (43 hours vs 20 hours, p< 0.001) and 14.8% of these patients died at SICU (p< 0.001). Characteristics independently associated with a GCS< 9 were: PaO2 (OR 1.02, p< 0.001) and APACHE II (OR 1.38, p< 0.001). 
Conclusion(s): PaO2 and APACHE II were considered independently associated with GCS< 9. Patients with GCS< 9 had higher incidence of major cardiac events, stayed onger in the SICU and had a higher mortality rate.

Evaluation of Glasgow Coma Scale after non-cardiac and non-neurologic surgery among intensive care patients