The deleterious effects of methamphetamine use on initial presentation and clinical outcomes in aneurysmal subarachnoid hemorrhage

Abstract

Object The objective of this study was to retrospectively look at methamphetamine (MA) use in patients with aneurysmal subarachnoid hemorrhage (SAH) to determine if MA use affects clinical presentation and outcomes after aneurysmal SAH. Methods A retrospective review of patients admitted to the Oregon Health & Science University neurosurgical service with aneurysmal SAH during the past 6 years was undertaken. Variables analyzed included MA use, age, sex, cigarette use, Hunt and Hess grade, Fisher grade, admission blood pressure, aneurysm characteristics, occurrence of vasospasm, hospital length of stay (LOS), cerebral infarction, aneurysm treatment, and Glasgow Outcome Scale (GOS) score. Data differences between MA users and nonusers were statistically analyzed using multivariate logistic regression analysis. A separate comparison with randomly selected age-matched nonuser controls was also performed. Results Twenty-eight (7%) of 374 patients with aneurysmal SAH were identified as MA users. Methamphetamine users were younger than nonusers (45.2 vs 55.9 years, respectively; p <0.001). Despite a younger age, MA users had significantly higher Hunt and Hess grades than nonusers (3.0 vs 2.5, respectively; p <0.020) and age-matched controls (3.0 vs 2.0, respectively; p <0.001). Earliest available mean arterial pressure was significantly higher in MA users (122.1 vs 109.7, respectively; p = 0.005) than all nonusers but not age-matched controls. Methamphetamine users had significantly higher vasospasm rates than nonusers (92.9% vs 71.1%, respectively; p = 0.008) but similar rates as age-matched controls (92.9% vs 89.3%, respectively; p = 0.500). Glasgow Outcome Scale score did not differ significantly between users and nonusers (3 vs 4, respectively; p = 0.170), but users had significantly lower GOS scores than age-matched controls (3 vs 5, respectively; p <0.001). There was no statistically significant difference in the LOS between users and nonusers (18 days vs 16 days, respectively; p = 0.431) or users and age-matched controls (18 days vs 14 days, respectively; p = 0.250). In the multivariate analysis, MA use (OR 3.777, p = 0.018), age (p <0.001), Fisher grade (p = 0.011), Hunt and Hess grade (p <0.001), and cerebral infarction (p <0.001) were predictors of poor GOS score. The only predictor of vasospasm was age (p <0.001), although a strong predictive trend in MA use (p = 0.149) was found. Predictors of a hospital LOS >15 days included age (p = 0.002), Fisher grade (p = 0.002), Hunt and Hess grade (p <0.001), and cerebral infarction (p <0.001). Predictors of cerebral infarction include male sex (p = 0.022) and Hunt and Hess grade (p = 0.006), with vasospasm demonstrating a strong trend (p = 0.056). Conclusions A history of MA use may predict poorer outcomes in patients who present with aneurysmal SAH. Methamphetamine users have significantly worse presentations and outcomes when compared with age-matched controls.