<i>Background:</i> Treating hyperglycemia in acute ischemic stroke may be beneficial, but knowledge on its prognostic value and optimal target glucose levels is scarce. We investigated the dynamics of glucose levels and the association of hyperglycemia with outcomes on admission and within 48 h after thrombolysis. <i>Methods:</i> We included 851 consecutive patients with acute ischemic stroke treated with intravenous thrombolysis in the Helsinki University Central Hospital during 1998–2008. Outcome measures were unfavorable 3- month outcome (3–6 on the modified Rankin Scale), death, and symptomatic intracerebral hemorrhage (sICH) according to NINDS criteria. Hyperglycemia was defined as a blood glucose level of ≧8.0 mmol/l. Four groups were identified based on (a) admission and (b) peak glucose levels 48 h after thrombolysis: (1) persistent normoglycemia (baseline plus 48-hour normoglycemia), (2) baseline hyperglycemia (48-hour normoglycemia), (3) 48-hour hyperglycemia (baseline normoglycemia), and (4) persistent hyperglycemia (baseline plus 48-hour hyperglycemia). <i>Results:</i> 480 (56.4%) of our patients (median age 70 years; onset-to-needle time 199 min; National Institutes of Health Stroke Scale score 9), had persistent normoglycemia, 59 (6.9%) had baseline hyperglycemia, 175 (20.6%) had 48-hour hyperglycemia, while persistent hyperglycemia appeared in 137 (16.1%) patients. Persistent and 48-hour hyperglycemia independently predicted unfavorable outcome [odds ratio (OR) = 2.33, 95% confidence interval (CI) = 1.41–3.86, and OR = 2.17, 95% CI = 1.30–3.38, respectively], death (OR = 6.63, 95% CI = 3.25–13.54, and OR = 3.13, 95% CI = 1.56–6.27, respectively), and sICH (OR = 3.02, 95% CI = 1.68–5.43, and OR = 1.89, 95% CI = 1.04–3.43, respectively), whereas baseline hyperglycemia did not. <i>Conclusions:</i> Hyperglycemia (≧8.0 mmol/l) during 48 h after intravenous thrombolysis of ischemic stroke is strongly associated with unfavorable outcome, sICH, and death.