Focused Ultrasound in Parkinson's Disease

MRI-guided focused ultrasound (MRgFUS) has emerged as an incisionless alternative to DBS and radiofrequency lesioning for movement disorders. By converging over 1,000 ultrasound beams through the intact skull to a precise intracranial target, MRgFUS generates a thermoablative lesion under real-time MRI guidance and thermometry — without incisions, burr holes, or implanted hardware. Originally developed and proven for essential tremor, FUS is now rapidly expanding into Parkinson's disease across multiple brain targets. This article reviews the technology, available targets (VIM, STN, GPi, pallidothalamic tract), evidence from landmark trials, FDA regulatory milestones, and the practical considerations that guide patient selection.

Technology & Procedure

How MRgFUS Works

• Transducer helmet: A hemispherical phased array (typically 1,024 elements) generates convergent ultrasound beams that pass through the intact skull and focus on a target ~3 mm in diameter
• MRI guidance: The patient lies in a 1.5T or 3T MRI scanner throughout the procedure. Real-time MR thermometry measures tissue temperature at the focal point during each sonication
• Staged temperature escalation: Low-energy sonications confirm targeting (patient reports symptom changes, examiner tests motor function). Once the optimal target is confirmed, temperature is raised to 54–60°C to produce a permanent thermocoagulative lesion
• Awake procedure: Patient is fully conscious — enabling real-time clinical assessment of tremor suppression, motor improvement, and detection of side effects (speech, sensation, strength) before committing to the permanent lesion
• Duration: Approximately 2–4 hours. Most patients are discharged same day or next day

Technical Requirements & Limitations

• Skull density ratio (SDR): The ratio of cortical bone to trabecular bone must be ≥0.40 (some centers use ≥0.45) for adequate ultrasound transmission. Low SDR causes beam aberration and insufficient heating. Pre-procedure CT scan is required to assess SDR
• Skull characteristics: Some patients (estimated 5–15%) cannot be treated due to unfavorable skull density, thickness, or shape. This may disproportionately affect certain racial groups
• Unilateral vs bilateral: Historically, bilateral lesioning was avoided due to concerns about cumulative speech, swallowing, and cognitive deficits (lessons from radiofrequency era). The 2024–2025 bilateral FUS data suggest staged bilateral procedures may be feasible with careful patient selection
• Irreversibility: Unlike DBS (where stimulation can be adjusted or turned off), FUS creates a permanent lesion. If the lesion is suboptimal or causes side effects, it cannot be reversed. This is the most fundamental disadvantage vs DBS
• MRI compatibility: Patient must be able to lie still in MRI for 2–4 hours with a stereotactic frame; claustrophobia and inability to cooperate are contraindications

FUS Targets in Parkinson's Disease

Evidence: Key Trials

VIM Thalamotomy for PD Tremor

VIM thalamotomy targets the ventral intermediate nucleus of the thalamus — the same target used in DBS for tremor. It was the first FUS indication studied in PD.

• FUS ET (Elias et al., NEJM 2016): The pivotal sham-controlled RCT that established MRgFUS thalamotomy — conducted in essential tremor, not PD, but provided the safety and feasibility foundation. Hand tremor improved by 47% vs sham at 3 months (P<0.001), sustained at 12 months (40%). Common AEs: paresthesias (38%), gait disturbance (36%); most resolved by 12 months (persistent: 14% paresthesias, 9% gait)
• FUST PD (2017): Phase 2 pilot RCT of 27 patients with medication-refractory tremor-dominant PD, randomized 2:1 to FUS VIM thalamotomy vs sham. On-medication hand tremor improved 62% with FUS vs 22% sham (P=0.04). UPDRS motor improved 8 points vs 1 point. 65% responder rate sustained at 1 year. Persistent AEs: orofacial paresthesia (20%), hemiparesis (10% — from thermal spread to internal capsule)
• Important limitation: VIM thalamotomy treats only tremor. Patients with progressive bradykinesia, rigidity, and motor fluctuations will not benefit — this target is appropriate only for the tremor-dominant PD subtype where tremor is the dominant source of disability

STN Subthalamotomy

The subthalamic nucleus is the most common DBS target for PD. FUS subthalamotomy ablates the STN on one side, aiming to replicate the broader motor benefits of STN-DBS without implanted hardware.

• FUS PD (Martínez-Fernández et al., NEJM 2020): Sham-controlled RCT of 40 patients with markedly asymmetric PD, randomized 2:1 to unilateral FUS subthalamotomy vs sham
  • MDS-UPDRS III (more affected side, off-medication): −9.8 points FUS vs −1.7 sham (between-group difference 8.1; 95% CI 6.0–10.3; P<0.001)
  • Benefit sustained at 12 months (mean reduction 11.6 points)
  • Improvements in rigidity, bradykinesia, and tremor on the treated side
• Adverse events were concerning:
  • Off-medication dyskinesia: 22% (persisting in 3/6 at 4 months) — from disruption of STN inhibitory output, analogous to hemiballismus/STN lesion syndromes
  • Speech disturbance: 56% (persisting in 3 patients)
  • Weakness on treated side: 19% (persisting in 2)
  • Gait disturbance: 50%
• Key limitations: Small (N=40), nearly single-center (36/40 from one site), blinding was ineffective (patients and assessors correctly guessed assignment), short follow-up (4 months for primary), limited to highly asymmetric disease. The high rate of neurological AEs has raised safety concerns and limited enthusiasm compared to DBS
• STN subthalamotomy remains investigational and is not FDA-approved

GPi Pallidotomy

Pallidotomy — ablation of the posteroventral GPi — has a long history in PD surgery (Leksell, 1950s; Laitinen revival 1990s). MRgFUS enables incisionless pallidotomy with MRI precision.

• Phase I (Jung et al., J Neurosurg 2018): 8 patients; demonstrated feasibility. UPDRS-III improved significantly; UDysRS improved. One patient developed visual field cut (optic tract proximity)
• Systematic review and meta-analysis (Abbas et al., 2024): 5 studies, 112 patients. UPDRS-III overall: −10.2 points (P<0.001). UDysRS: −18.9 points (P<0.001). UPDRS-IV: −5.1 (P<0.001). AEs were mostly transient — headache, gait difficulty, sonication-related pain; no statistically significant serious AE rate
• FDA-approved 2021 for unilateral FUS pallidotomy in PD (dyskinesia and motor symptoms)
• Clinical niche: Patients where dyskinesia is the dominant problem (analogous to GPi DBS), who are not candidates for or decline DBS, and have predominantly unilateral symptoms

Pallidothalamic Tractotomy (PTT)

The pallidothalamic tract carries inhibitory output from GPi to the thalamus — it is the "output highway" of the basal ganglia. Ablating this white matter tract effectively disconnects the hyperactive GPi output without lesioning the nucleus itself.

• Phase 3 trial (NCT04728295, MDS 2025 late-breaker): 54 patients received unilateral MRgFUS PTT; 40 proceeded to staged bilateral treatment
  • Unilateral (3 months): MDS-UPDRS III OFF improved ~50% (20.9 → 10 points)
  • Bilateral: Sustained improvement at 12 months
  • Improvements in rigidity, bradykinesia, and dyskinesia
• FDA-approved July 2025 — staged bilateral pallidothalamic tractotomy for advanced PD motor complications (rigidity, bradykinesia, dyskinesia). This is the first bilateral FUS indication for PD
• Advantages over nuclear targets: White matter tract lesioning may produce therapeutic effect with smaller lesion volumes, potentially reducing AEs. The PTT is anatomically distinct from adjacent eloquent structures
• Limitations: Very new — limited long-term follow-up, no head-to-head comparisons with DBS or other FUS targets. Open-label design (no sham arm in bilateral phase). Expert commentary has urged caution pending more data

FUS vs DBS: A Clinical Comparison

Adverse Events Across FUS Targets

• Common transient AEs (across all targets): Headache during sonication, pin-site pain (from stereotactic frame), nausea, dizziness — usually resolve within hours to days
• Paresthesia/numbness: 20–38% (thalamotomy); usually resolves. Persistent in ~14% at 12 months. Reflects thermal spread to medial lemniscus or adjacent thalamic nuclei
• Gait disturbance/ataxia: 9–50% depending on target. Higher with STN subthalamotomy. May reflect cerebellar tract involvement
• Dysarthria: Particularly concerning with STN subthalamotomy (56%) and bilateral procedures. Persistent dysarthria reported in 7–14% of bilateral FUS ET patients
• Dyskinesia (off-medication): Specific to STN subthalamotomy (~22%) — analogous to hemiballismus from STN lesions. A unique and potentially persistent complication not seen with DBS (where stimulation can be reduced)
• Hemiparesis: Thermal spread to internal capsule. Reported in ~10% of FUST PD; usually transient but can persist
• Visual field deficits: Risk with pallidotomy (optic tract proximity). Careful targeting and real-time monitoring mitigate this risk
• Lesion fading: Some patients experience symptom recurrence over months as the perilesional edema resolves and the effective lesion size decreases. Long-term durability data beyond 3 years are limited

Trial Comparison Table

Future Directions

• Head-to-head FUS vs DBS RCTs: Critically needed. No randomized comparison exists between FUS and DBS for any PD indication. Until such data are available, evidence-based target selection between the two modalities relies on indirect comparisons and expert consensus
• Long-term durability: Most FUS data extend only 1–3 years. Whether lesion effect persists or fades over 5–10 years (as disease progresses and compensatory mechanisms shift) remains unknown
• Bilateral FUS safety: The July 2025 bilateral PTT approval is a landmark, but real-world safety data on speech, cognition, and swallowing outcomes with bilateral treatment will be essential
• Optimizing targets: Comparative studies between STN subthalamotomy, pallidotomy, and PTT for non-tremor PD features are underway (Paschen et al., Mov Disord 2025)
• Low-intensity FUS (LIFU): Investigational non-ablative approach using pulsed ultrasound to modulate neural activity or open the blood-brain barrier for targeted drug delivery — preclinical and early clinical studies ongoing
• Expanded access: Skull density ratio limitations, cost ($20,000–30,000+ per procedure), and limited MRgFUS-equipped centers remain barriers to broader adoption