Persistent perioperative hyperglycemia as an independent predictor of poor outcome after aneurysmal subarachnoid hemorrhage.

OBJECT The authors of previous studies have shown that admission hyperglycemia or perioperative hyperglycemic events may predispose a patient to poor outcome after aneurysmal subarachnoid hemorrhage (SAH). The results of experimental evidence have suggested that hyperglycemia may exacerbate ischemic central nervous system injury. It remains to be clarified whether a single hyperglycemic event or persistent hyperglycemia is predictive of poor outcome after aneurysmal SAH. METHODS Ninety-seven patients undergoing treatment for aneurysmal SAH were observed, and all perioperative variables were entered into a database of prospectively recorded data. Daily serum glucose values were retrospectively added. Patients were examined at hospital discharge (14-21 days after SAH onset), and Glasgow Outcome Scale (GOS) scores were prospectively documented. The GOS score at last follow-up was retrospectively determined. Serum glucose greater than 200 mg/dl for 2 or more consecutive days was defined as persistent hyperglycemia. Outcome was categorized as "poor" (dependent function [GOS Score 1-3]) or "good" (independent function [GOS Score 4 or 5]) at discharge. The independent association of 2-week and final follow-up outcome (GOS score) with the daily serum glucose levels was assessed using a multivariate analysis. RESULTS In the univariate analysis, increasing age, increasing Hunt and Hess grade, hypertension, ventriculomegaly on admission computed tomography scan, Caucasian race, and higher mean daily glucose levels were associated with poor (dependent) 2-week outcome after aneurysmal SAH. In the multivariate analysis, older age, the occurrence of symptomatic cerebral vasospasm, increasing admission Hunt and Hess grade, and persistent hyperglycemia were independent predictors of poor (dependent) outcome 2 weeks after aneurysmal SAH. Admission Hunt and Hess grade and persistent hyperglycemia were independent predictors of poor outcome at last follow-up examination a mean 10 +/- 3 months after aneurysmal SAH. Isolated hyperglycemic events did not predict poor outcome. Patients with persistent hyperglycemia were 10-fold more likely to have a poor (dependent) 2-week outcome and sevenfold more likely to have a poor outcome a mean 10 months after aneurysmal SAH independent of admission Hunt and Hess grade, occurrence of cerebral vasospasm, or all comorbidities. CONCLUSIONS Patients with persistent hyperglycemia were seven times more likely to have a poor outcome at a mean of 10 months after aneurysmal SAH. Isolated hyperglycemic events were not predictive of poor outcome. Serum glucose levels in the acute setting of aneurysmal SAH may help predict outcomes months after surgery.