Spinal cord ultrasound stimulation modulates corticospinal excitability in humans
Lin Hou, Yuming Lei
Brain Stimulation 2025 · 10.1016/j.brs.2025.05.134
Abstract
Background Low-intensity focused ultrasound offers deep tissue penetration and holds promise as a noninvasive tool for neuromodulating brain circuits. However, its effects on neural activity within the human spinal cord remain largely unexplored. Objective To investigate the effects of spinal cord ultrasound stimulation (SCUS) on corticospinal excitability by varying acoustic parameters: spatial-peak pulse-average intensity (I SPPA ), duty cycle (DC), and pulse repetition frequency (PRF). Methods Two experiments were conducted involving 62 healthy adult participants. In Experiment 1 (N = 36), participants were randomly assigned to either a low-intensity group (2.5 W/cm 2 , N = 18) or a relatively high-intensity group (10 W/cm 2 , N = 18) to examine parameter-dependent SCUS effects. SCUS was delivered as a 500 ms pulse train to the C8 spinal cord segment using two DCs (10 % and 30 %) and two PRFs (500 and 1000 Hz). Stimulation was applied concurrently with posterior-anterior (PA) oriented transcranial magnetic stimulation (TMS) over the left primary motor cortex (M1), and motor-evoked potentials (MEPs) were recorded from the right first dorsal interosseous (FDI) muscle. Active and sham SCUS trials were interleaved and compared. In Experiment 2 (N = 26), the specificity of SCUS effects was assessed in relation to TMS coil orientation and target muscle using the sonication parameter set of 10 W/cm 2 I SPPA , 1000 Hz PRF, and 30 % DC. MEPs from the FDI were recorded using anterior-posterior (AP) and latero-medial (LM) coil orientations, while MEPs from the abductor digiti minimi (ADM) muscle were assessed using PA, AP, and LM orientations. MEPs were also recorded from the biceps brachii with PA orientation. Active and sham SCUS conditions were interleaved and compared. Results SCUS significantly suppressed MEP amplitudes compared to sham stimulation under relatively high-intensity (10 W/cm 2 ), high-PRF (1000 Hz) conditions, regardless of DC. No significant effects were observed at lower intensity (2.5 W/cm 2 ) or lower PRF (500 Hz). Follow-up experiments confirmed consistent inhibitory effects across multiple TMS coil orientations (PA, AP, LM) and in C8-innervated hand muscles (FDI and ADM), with no significant modulation observed in the more rostrally innervated biceps brachii muscle, supporting the specificity of C8-targeted SCUS. Conclusions These findings demonstrate that SCUS can modulate corticospinal excitability in a parameter-specific manner, with suppression observed primarily under relatively high-intensity and high-PRF conditions. SCUS represents a promising noninvasive technique for targeted neuromodulation of spinal neural circuits in humans.
Abstract via europepmc.
Exposures
Exposure 1: SCUS to C8 spinal cord, 500 kHz, 2.5 and 10 W/cm2 ISPPA
Target: cervical spinal cord — “C8 spinal cord”
Device: Sonic Concepts · Sonic Concepts Inc., US · DPX-500-4CH four-element annular array, driven by a four-channel TPO ✓
| Waveform | pulsed | |
|---|---|---|
| Fundamental frequency (kHz) | 500 | ✓✓✓ |
| Pulse duration (ms) | 0.1, 0.2, 0.3, 0.6swept | ✓✓✓⚑ |
| Pulse repetition frequency (Hz) | 500, 1,000swept | ✓✓✓ |
| Duty cycle (%) | 10, 30swept | ✓✓✓ |
| Sonication duration (s) | 0.5 | ✓✓✓ |
| Free-field pressure (kPa) | 280, 550swept | ✓✓✓⚑ |
|---|---|---|
| Free-field Isppa (W/cm²) | 2.5, 10swept | ✓✓✓ |
| Free-field Ispta (W/cm²) | 0.01, 0.04, 0.05, 0.14swept | ✓✓✓⚑ |
| In-situ estimate | simulationmean or range across subjects | |
| In-situ pressure (kPa) | 130, 250swept | ✓✓✓⚑ |
| In-situ Isppa (W/cm²) | 0.53, 2.12swept | ✓✓✓ |
| In-situ Ispta (W/cm²) | not reported |
A 500 ms SCUS pulse train was applied to the C8 spinal segment, with a single TMS pulse delivered 100 ms before the end of the train; 15 active and 15 sham trials per session in randomised order, 10 s interstimulus interval. Four DC/PRF combinations (10 %/500 Hz, 30 %/500 Hz, 10 %/1000 Hz, 30 %/1000 Hz), one per session.
Flags from extraction
exposures[0].free_field.pressure_kpa— 0.55 and 0.28 MPa are simulated focal pressures in a free-water field for the 10 and 2.5 W/cm2 conditions, recorded as a list; two separate sentences carry the two values.exposures[0].free_field.ispta_w_cm2— ISPTA values 0.01, 0.04, 0.05, 0.14 W/cm2 read from the parameter table (header 'in water'); the quote is the whole table.exposures[0].in_situ.pressure_kpa— Simulated mean pressure at C8 in three CT datasets for 2.5 and 10 W/cm2 conditions (0.13 and 0.25 MPa), recorded as a two-element list of the two conditions rather than a range; ISPPA likewise 0.53 and 2.12 W/cm2.exposures[0].timing.pulse_duration_ms— Pulse durations 0.1-0.6 ms listed as four values tied to DC/PRF combinations; stated in the text and table, not computed.n_subjects— 62 participants over two experiments; Experiment 2 is given as N=26 in the abstract but as 16 per arm (N=10 for biceps) in Methods.n_sessions_per_subject— Four sessions per participant in Experiment 1 only; Experiment 2 session count not stated.blinding— Gaussian noise was said to maintain blinding but the paper does not state a blinding level.exposures[0].target.label— Spinal cord target; the paper is on spinal cord ultrasound stimulation with TMS over M1 as the readout.