Non-invasive ultrasonic neuromodulation of neuronal excitability for treatment of epilepsy
Zhengrong Lin, Long Meng, Junjie Zou, Wei Zhou, Xiaowei Huang, Shan Xue, Tianyuan Bian, Tifei Yuan, Lili Niu, Yanwu Guo, Hairong Zheng
Theranostics 2020, 10, 5514-5526 · 10.7150/thno.40520
Abstract
Non-invasive low-intensity pulsed ultrasound has been employed for direct neuro-modulation. However, its range and effectiveness for different neurological disorders have not been fully elucidated. Methods: We used multiple approaches of electrophysiology, immunohistochemistry, and behavioral tests as potential epilepsy treatments in non-human primate model of epilepsy and human epileptic tissues. Low-intensity pulsed ultrasound with a frequency of 750 kHz and acoustic pressure of 0.35 MPa (the spatial peak pulse average intensity, I SPPA = 2.02 W/cm 2 ) were delivered to the epileptogenic foci in five penicillin-induced epileptic monkey models. An ultrasound neuro-modulation system with a frequency of 28 MHz and acoustic pressure of 0.13 MPa (I SPPA = 465 mW/cm 2 ) compatible with patch-clamp systems was used to stimulate the brain slices prepared from fifteen patients with epilepsy. Results: After 30 min of low-intensity pulsed ultrasound treatment, total seizure count for 16 hours (sham group: 107.7 ± 1.2, ultrasound group: 66.0 ± 7.9, P in vitro . Ultrasound stimulation could inhibit epileptiform activities with an efficiency exceeding 65%, potentially due to adjusting the balance of excitatory-inhibitory (E/I) synaptic inputs by the increased activity of local inhibitory neurons. Conclusion: Herein, we demonstrated for the first time that low-intensity pulsed ultrasound improves electrophysiological activities and behavioral outcomes in a non-human primate model of epilepsy and suppresses epileptiform activities of neurons from human epileptic slices. The study provides evidence for the potential clinical use of non-invasive low-intensity pulsed ultrasound stimulation for epilepsy treatment.
Abstract via europepmc.
Exposures
Exposure 1: In vivo epileptic-focus stimulation (penicillin-induced epilepsy monkey model)
Target: frontal lobe — “right frontal lobe (epileptogenic focus, penicillin injection site)”
Device: Sonic Concepts · Sonic Concept · H116
| Waveform | pulsed | |
|---|---|---|
| Fundamental frequency (kHz) | 750 | ✓✓✓ |
| Pulse duration (ms) | 0.3 | ✓✓✓ |
| Pulse repetition frequency (Hz) | 1,000 | ✓✓✓ |
| Duty cycle (%) | not reportedpulse duration × PRF gives 30% | |
| Sonication duration (s) | 0.2 | ✓✓✓ |
| Free-field pressure (kPa) | not reported | |
|---|---|---|
| Free-field Isppa (W/cm²) | not reported | |
| Free-field Ispta (W/cm²) | not reported | |
| In-situ estimate | not reported | |
| In-situ pressure (kPa) | not reported | |
| In-situ Isppa (W/cm²) | not reported | |
| In-situ Ispta (W/cm²) | not reported | |
| Pressure, domain unspecified (kPa) | 350 | ✓✓✓⚑ |
| Isppa, domain unspecified (W/cm²) | 2.02 | ✓✓✓ |
A single-element ultrasound transducer with TBD of 300 µs, PRF of 1000 Hz, sonication duration (SD) of 200 ms and inter-stimulation interval (ISI) of 5 s was applied to the epileptogenic focus for 30 minutes total per session. Sham consisted of the transducer fixed in place for 30 minutes without ultrasound output; a separate control experiment placed the transducer away from the brain (audible-sound-only) for 8 hours.
Consistency checks: intensity pressure inconsistent unspecified domain.
Exposure 2: Ex vivo brain slice stimulation (mouse cortical slices for parameter selection; human TLE patient temporal-lobe slices)
Target: brain slice — “cortical/temporal-lobe brain slices (mouse, for parameter selection; human TLE patient biopsy, main experiment)”
Device: custom-built · ultrasound neuro-modulation chip (interdigital transducers on LiNbO3 substrate)
| Waveform | pulsed | |
|---|---|---|
| Fundamental frequency (kHz) | 28,000 | ✓✓✓ |
| Pulse duration (ms) | 5 | ✓✓✓ |
| Pulse repetition frequency (Hz) | 100 | ✓✓✓ |
| Duty cycle (%) | not reportedpulse duration × PRF gives 50% | ⚑ |
| Sonication duration (s) | 60 | ✓✓✓ |
| Free-field pressure (kPa) | not reported | |
|---|---|---|
| Free-field Isppa (W/cm²) | not reported | |
| Free-field Ispta (W/cm²) | not reported | |
| In-situ estimate | not applicable | |
| In-situ pressure (kPa) | not applicable | |
| In-situ Isppa (W/cm²) | not applicable | |
| In-situ Ispta (W/cm²) | not applicable | |
| Pressure, domain unspecified (kPa) | 130 | ✓✓✓ |
| Isppa, domain unspecified (W/cm²) | 0.465 | ✓✓✓ |
| Ispta, domain unspecified (W/cm²) | 0.233 | ✓✓✓ |
Sixty-second low-intensity pulsed ultrasound waveforms (28 MHz, TBD 5 ms, PRF 100 Hz) that produced an inhibitory effect in mice cortical slices were therefore used to stimulate the human epileptic slices; spontaneous activity was recorded for 60 s before and 60 s during ultrasound. A supplementary chip at 6.57 MHz was also tested but without quantified pressure/timing parameters reported.
Consistency checks: f0 out of range.
Flags from extraction
n_subjects— This paper spans three very different subject populations that cannot be summed to one total: 5 monkeys (in vivo), 15 TLE patients plus 4 non-epileptic glioma patients providing biopsy tissue for ex vivo brain slices, and 4 mice used only for parameter selection; subject_unit is set to 'animal' as an approximation.anaesthesia— Ketamine/atropine anesthesia is explicitly described for the surgical/injection procedure (1.5 h), but the paper does not clearly state whether the monkeys remained anesthetized during the subsequent video-EEG monitoring and 30-minute ultrasound treatment.exposures[0].unspecified_domain.pressure_kpa— The paper does not state whether the 0.35 MPa / 2.02 W/cm2 values were measured in free field or in situ; the skull was surgically removed and the transducer coupled directly to the dura, but no domain language ('free field' or 'in situ') is used.exposures[1].timing.duty_cycle_pct— ISPTA is stated to be 'calculated by multiplying duty cycle to the ISPPA' but the duty cycle value itself is not given in the text (233/465 = 50% is implied, but not stated), so it is left not_reported rather than computed.exposures[1].in_situ— Ultrasound was delivered directly to an isolated brain slice bathed in ACSF via a chip beneath it, with no intervening skull or tissue path; set to not_reported (rather than null) because the record as a whole is not tissue-only (it also includes the in vivo monkey exposure).