MRI TRACT VOLUMETRY AS A PREDICTOR OF MOTOR OUTCOME IN PARKINSON’S DISEASE

Predicting motor improvement after subthalamic nucleus deep brain stimulation (STN-DBS) in Parkinson’s disease (PD) remains challenging. MRI-based tractography allows quantitative assessment of white matter tracts, which may serve as imaging biomarkers. Our goal was to determine whether volumetric and fiber length measures of specific white matter tracts predict clinically meaningful motor improvement (≥50% MDS-UPDRS reduction) following STN-DBS. In 33 PD patients treated with unilateral STN-DBS, diffusion MRI tractography reconstructed five major associative tracts bilaterally: ILF, IFOF, UF, SFOF, and SLF. Tract volume and mean fiber length were extracted. Logistic regression with backward elimination identified independent predictors of motor improvement. Model performance was assessed with ROC analysis, 5-fold stratified cross-validation, and calibration metrics. Greater volume of the left SFOF (OR = 4.86, p = 0.005) and longer fiber length of the right SFOF (OR = 1.90, p = 0.045) were independently associated with achieving ≥50% MDS-UPDRS improvement. Patients with both favorable measures had a substantially higher probability of meaningful motor benefit. The final model demonstrated good discrimination (AUC = 0.78, 95% CI: 0.57–0.95) and acceptable calibration (Brier score 0.12), with stable performance on cross-validation. Preoperative SFOF tract volume and fiber length predict motor improvement after STNDBS in PD. These findings support the potential of tractography-based structural metrics as imaging biomarkers to enhance patient selection and optimize surgical planning.



Figure 1. Visual comparison of SFOF in PD patient (a, c) and healthy control (b, d).