Randomized trial of pallidotomy versus medical therapy for Parkinson's disease

Jerrold L. Vitek, Roy A E Bakay, Alan Freeman, Marian Evatt, Joanne Green, William McDonald, Michael Haber, Huiman Barnhart, Natalie Wahlay, Shirley Triche, Klaus Mewes, Vijay Chockkan, Jian Yu Zhang, Mahlon R. DeLong

Research output: Contribution to journalArticlepeer-review

163 Scopus citations


Thirty-six patients with Parkinson's disease (PD) were randomized to either medical therapy (N = 18) or unilateral GPi pallidotomy (N = 18). The primary outcome variable was the change in total Unified Parkinson's Disease Rating Scale (UPDRS) score at 6 months. Secondary outcome variables included subscores and individual parkinsonian symptoms as determined from the UPDRS. At the six month follow-up, patients receiving pallidotomy had a statistically significant reduction (32% decrease) in the total UPDRS score compared to those randomized to medical therapy (5% increase). Following surgery, patients' showed improvement in all the cardinal motor signs of PD including tremor, rigidity, bradykinesia, gait and balance. Drug-induced dyskinesias were also markedly improved. Although the greatest improvement occurred on the side contralateral to the lesion, significant ipsilateral improvement was also observed for brady-kinesia, rigidity and drug-induced dyskinesias. A total of twenty patients have been followed for 2 years to assess the effect of time on clinical outcome. These patients have shown sustained improvement in the total UPDRS (p < 0.0001), "off" motor (p < 0.0001) and complications of therapy subscores (p < 0.0001). Sustained improvement was also seen for tremor, rigidity, bradykinesia, percent on time and drug-induced dyskinesias.

Original languageEnglish (US)
Pages (from-to)558-569
Number of pages12
JournalAnnals of Neurology
Issue number5
StatePublished - May 1 2003


Dive into the research topics of 'Randomized trial of pallidotomy versus medical therapy for Parkinson's disease'. Together they form a unique fingerprint.

Cite this