Habenula as a Marker of Thalamic Asymmetrical Skew and a Landmark for the Vim-Vc Boundary in MRgFUS Thalamotomy.

BACKGROUND: Traditional targeting based on the anterior commissure-posterior commissure (AC-PC) line in MR-guided focused ultrasound (MRgFUS) thalamotomy does not account for individual thalamic asymmetry, potentially leading to suboptimal outcomes or adverse effects. OBJECTIVE: To investigate whether habenular torsion, an anatomical marker of thalamic asymmetry, predicts clinical outcomes and sensory side effects following ventral intermediate nucleus (Vim) thalamotomy. METHODS: We retrospectively analyzed 157 patients (essential tremor: 121; Parkinson's disease: 36) who underwent unilateral MRgFUS thalamotomy. Angular metrics, including the Torsion Index (inter-hemispheric habenular angular difference) and the Lesion-Habenula difference (based on a lesion axis originating from the midline-habenula intersection), were measured on axial MRI 2.0 mm dorsal to the AC-PC line. Tremor improvement (CRST) and ventralis caudalis (Vc)-related sensory symptoms were evaluated immediately and at the 1-year follow-up. RESULTS: Tremor improved by over 70% immediately after treatment, with no significant difference between torsion groups (p = 0.490), indicating that torsion did not compromise efficacy. However, Anterior Torsion (Torsion Index < 0) was significantly associated with a higher incidence of both immediate (43.2% vs. 14.5%, p < 0.001) and persistent (25.0% vs. 5.1%, p = 0.001) Vc symptoms. ROC analysis identified the Torsion Index (AUC 0.72-0.75) and Lesion-Habenula difference (AUC 0.76-0.84) as robust predictors for sensory side effects, significantly outperforming the standard AP ratio (AUC 0.47-0.50). CONCLUSION: Habenular torsion is a critical anatomical biomarker for predicting post-operative sensory complications. Incorporating the Torsion Index into preoperative planning allows for a personalized, anatomy-driven targeting approach, potentially enhancing the safety and efficacy of stereotactic thalamotomy.