For the roughly one-third of epilepsy patients whose seizures refuse to respond to medication, surgical options carry real risk and many are simply not candidates. A noninvasive brain stimulation technique capable of reaching deep neural targets without a scalpel could fundamentally shift the treatment calculus — and a new controlled trial offers the most rigorous evidence yet that low-intensity focused ultrasound (LIFU) may be that tool.
This pilot randomized sham-controlled crossover trial enrolled 12 adults with drug-resistant epilepsy (DRE), each receiving both active LIFU and sham stimulation targeting their individually mapped seizure onset zone, in randomized order, with a 4-week washout period between conditions. The primary crossover analysis produced a non-significant result — a proportional seizure frequency estimate of 0.49 (95% CI −0.04 to 1.01, p = .068) — meaning the controlled phase narrowly missed statistical significance. However, an open-label longitudinal extension phase revealed a sustained reduction of approximately 14 percentage points per seizure-month following LIFU (β = −14.0, 95% CI −22.2 to −5.8, p = .001). MRI confirmed no structural tissue damage at any follow-up point, and neuropsychological assessments showed no deterioration in mood, cognition, or quality of life.
Contextualizing these results requires acknowledging several layered limitations. Twelve participants is underpowered for definitive efficacy conclusions — the crossover p-value of .068 is tantalizingly close but statistically inconclusive, and the open-label phase lacks a concurrent control arm, leaving responder bias and natural seizure fluctuation as confounders. That said, LIFU occupies a genuinely distinct mechanistic niche among neuromodulation options: unlike transcranial magnetic stimulation or transcranial direct current stimulation, focused ultrasound can penetrate to centimeter depths with millimeter-scale precision, enabling direct modulation of cortical and subcortical seizure onset zones without thermal injury at low intensities. Compared to implanted devices like responsive neurostimulation or vagus nerve stimulators — the current standard for DRE candidates who fail surgery — LIFU's noninvasive profile and clean safety record here are notable. This trial is best characterized as a well-designed proof-of-concept with promising longitudinal signal; adequately powered phase II trials with longer follow-up are the essential next step before any clinical translation.