Predictors of major neurologic improvement after thrombolysis in acute stroke

Background: Major neurologic improvement at 24 hours after administration of recombinant tissue plasminogen activator (rt-PA) in acute stroke may predict good outcome at 3 months. Objective: To identify predictors of major neurologic improvement at 24 hours after IV rt-PA administration and its relationship with outcome at 3 months. Methods: The authors analyzed patients with acute stroke treated with IV rt-PA from two academic centers in London, Ontario, and 33 affiliated hospitals between 1999 and 2003. Major neurologic improvement was defined by a >= 8-point improvement in NIH Stroke Scale (NIHSS) score or an NIHSS score of 0 or 1 at 24 hours. Good outcome was defined as a 3-month modified Rankin Scale of 0 to 1. Results: Of 219 patients with acute stroke treated with rt-PA, 61 (28%) had major neurologic improvement at 24 hours. Glucose levels < 8 mmol/L (OR 4.98, 95% CI 1.6 to 15.2), lack of cortical involvement on 24 hour CT scan (OR 3.97, 95% CI 1.87 to 8.43), and female sex (OR 2.4, 95% CI 1.12 to 5.13) were associated with major neurologic improvement after adjusting for covariates. Patients with major neurologic improvement had a shorter hospital stay (6.7 vs 14.3 days; p = 0.001). Major neurologic improvement was an independent predictor of good outcome at 3 months (OR 12.8, 95% CI 4.72 to 34.6). Conclusions: Major neurologic improvement after rt-PA was observed in 28% of patients and independently predicted good outcome at 3 months. Female sex, glucose levels < 8 mmol/L, and absence of cortical involvement at 24 hours CT scan were associated with major neurologic improvement.