The objective of this study was to compare the long-term tolerability and efficacy of tolcapone and entacapone in patients with fluctuating Parkinson's disease (PD). Tolcapone and entacapone are two currently available catechol- O -methyltransferase inhibitors that have demonstrated efficacy in the treatment of advanced PD. There are little published data on long-term experience and no direct comparisons. We compared the results of two separate, simultaneous, long-term open label extensions, one for tolcapone and the other for entacapone. The inclusion/exclusion criteria were similar. Data were collected prospectively at 6, 12, 24, and 36 months. Efficacy measures included the Unified Parkinson's Disease Rating Scale (UPDRS) total score, subscores, items 32 (duration of dyskinesia) and 39 (duration of "off" time), and levodopa dose. The two groups were compared using a Mann-Whitney U test for change from baseline and analysis of variance. Tolerability was defined as the ability of patients to maintain therapy and was compared using a Kaplan-Meier analysis. Eleven patients enrolled in the entacapone study and 14 in the tolcapone study. The tolcapone group had more severe disease with significantly higher UPDRS motor score, duration of "off," and levodopa dose requirement. Tolcapone was more effective in lowering UPDRS motor and complication subscores, duration of "off" time, and levodopa doses. UPDRS motor scores and change in levodopa dose in the tolcapone group remained below baseline level for 36 months; however, they were above baseline in the entacapone group from 6 months on. Tolerability was the same for both treatments. Tolcapone appears to have greater and longer efficacy with regard to motor symptoms, "off" time, and change in levodopa requirements than entacapone. These findings indicate that tolcapone continues to have a place in the treatment of advanced PD. However, the risks associated with this drug, particularly hepatic injury, and the requirement for rigorous blood monitoring, need to be considered when choosing an appropriate treatment for patients with advanced PD.