A double-blind study was carried out in 30 patients suffering from spasticity due to cerebrovascular lesions to compare the long-term efficacy and tolerability of tizanidine hydrochloride with that of baclofen. A 2-week titration phase identified the optimum dose of tizanidine (max. 20 mg/day) or baclofen (max. 50 mg/day) in each patient. Patients were then treated with this dose for a 50-week maintenance phase. Efficacy and tolerability parameters were evaluated first on a monthly and then on a bimonthly basis. Both tizanidine and baclofen caused an improvement in the symptoms associated with spasticity. In end-point analysis, 87% of patients showed an improvement (p less than 0.01) in excessive muscle tone - the major efficacy parameter in this study - in the tizanidine group, while 79% improved (p less than 0.01) in the baclofen group. Side-effects in the tizanidine group were mild and transient and no patients discontinued the study; in the baclofen group, 3 patients discontinued the study due to severe side-effects. However, both drugs were assessed as effective and fairly well tolerated in the long-term. Although there were no statistically significant differences between the two drugs, the global assessment of antispastic efficacy revealed a nearly significant difference (p = 0.057) in favour of tizanidine and the global assessment of tolerability was also in favour of tizanidine.