10 Year Clinical Outcomes of Subthalamic Nucleus versus Pallidal Deep Brain Stimulation for Parkinson’s Disease: VA/NINDS CSP #468F
Jill Ostrem1, Ping Luo2, Frances Weaver2, Kenneth Follett3, Nicholas Galifianakis1,4, Eugene Lai5, Jeff Bronstein6,7, John Duda8, Kathryn Holloway9,10, Aliya Sarwar11, Matthew Brodsky12, Kathryn Chung12,13, Meredith Spindler14, Domenic Reda2, Johannes Rothland4, Amanda Snodgrass15, William Marks Jr16
1University of California San Francisco, 2Edward Hines Jr, VA Hospital, 3University of Nebraska Medical Center, 4San Francisco VA Medical Center, 5Houston Methodist Neurological Institute, 6University of California Los Angeles, 7West Los Angeles VA Medical Center, 8Philadelphia VA Medical Center, 9McGuire VA Medical Center, 10Virginia Commonwealth University, 11Baylor College of Medicine/ MEDVAMC, 12Oregon Health Science University, 13Portland VA Medical Center, 14University of Pennsylvania, 15VA CSP Clinical Research Pharmacy Coordinating Center, 16Stanford University School of Medicine

To report extended long-term outcomes of DBS of the GPi and STN for patients with Parkinson disease (PD) in the landmark CSP468 VA/NINDS multicenter randomized controlled trial.

There are no published studies from prospective randomized trials examining outcomes beyond 3 years comparing STN to GPi DBS.  Here we present outcomes after DBS (out to 10 years in some patients).

A subset of patients originally randomized to GPi or STN DBS were followed up to 10 years with visits completed at 2 years (GPi n=85; STN n=70), 7 years (GPi n=68; STN n=49), and 10 years (GPi n=49; STN n=28). The primary outcome was change in UPDRS motor subscale in the off medication/on stimulation state between targets and included multiple secondary outcomes.


Baseline characteristics were comparable. Improvement in motor function compared to baseline was maintained at 2, 7, and 10 years for GPi(43.2 to 25.8, p < 0.001; 35.4, p<0.001; and 34.0, p=0.10) and STN(43.2 to 27.7, p< 0.001; 34.4 p<0.001, and 28.3 p<0.001), respectively. Improvements were generally similar between targets over time, but with a trend favoring STN(p = 0.09). The tremor subscales showed the greatest reduction overtime followed by rigidity subscores.  Bradykinesia subscores showed greater improvement at 7 and 10 years with STN DBS(p=0.03). UPDRS I,  II, and IV scores and quality of time based on motor diaries also showed significant long-term improvement regardless of target. PDQ39 total score no longer showed improvement at 7 or 10 years (either target). Lastly, both targets had significant medication reduction with no target difference over time (p=0.70).

DBS therapy had a significant and stable effect on motor function regardless of target over 10 years. This is remarkable, given that PD is a progressive neurodegenerative disease. This is the longest follow up describing DBS outcomes comparing the two targets in a randomized cohort.