(AS1) Symptom Specific Sweetspot, DTI, and Resting State Fmri Analysis of the STN and Gpi Regions in the CSP 468 Randomized Controlled Trial of DBS Stimulation in Parkinson’s Patients

Introduction: The Veteran’s Affairs Cooperative Studies Program #468 was an NIH and VA funded multicenter, randomized, controlled, blinded clinical trial comparing outcomes of STN and GPi deep brain stimulation in 299 Parkinson's disease patients. This study was unique in the broad sampling of the target areas, making it an ideal dataset to explore the effect of lead location on outcome.

Methods: The patient cohort with appropriate imaging (n=160) was utilized to assess the effect of lead location on the change from baseline in the blinded off-medication UPDRS-III and select subscores including tremor, rigidity, and bradykinesia at 6 months. Cranial Suite and LeadDBS v2.6 were used to normalize postoperative scans to MNI space and generate volumes of tissue activation (VTA). Outcome-weighted VTAs were incorporated into statistically significant sweetspots (SS) per the Wilcoxon T-test. Outcome-weighted fiber tracts and rsfMRI connectivity maps correlated with significant clinical improvement were identified with t-tests. All models were cross validated using a “leave-one-out” methodology.

Results: The STN-SS correlated with UPDRS-III improvement (R=0.46, p=0.001) was located at the junction of the posterior-ventral STN, Zona Incerta, and Substantia Nigra Reticularis. A distinct region anticorrelated with UPDRS-III improvement for the STN (R=-0.27, p=0.034) was located anterodorsally to the STN-SS in the associative STN. A GPi-SS correlated with UPDRS-III improvement (R=0.26, p=0.046) was located at the border of the posterior-lateral GPi and anteromedial GPe. Statistically significant and distinct symptom-specific sweetspots were also identified within the STN (R-Tremor=0.63, p< 0.001, R-Bradykinesia=0.28, p=0.05) and GPi (R-Tremor=0.32, p=0.029, R-Rigidity=0.54, p< 0.001, R-Bradykinesia=0.34, p=0.021). Fiber tracking and rsfMRI connectivity fingerprints correlating with these sweetspots were identified.

Conclusion : Outcome-weighted VTAs from this large patient cohort identified subregions of the STN and GPi that were associated with UPDRS-III and symptom-specific improvement. Interestingly, these sweet spots were not in the anatomic locations traditionally targeted for stimulation.