Enhancement of sleep slow wave activity using transcranial electrical stimulation with temporal interference: an interim analysis of the STRENGTHEN study.
Level 3 - non-randomized controlled study
Non-randomized controlled interventional trial (single-blind, non-random allocation)
PubMed 42386897 · doi:10.1038/s43856-026-01744-x
What was done
Interim analysis of the single-blind, non-randomized STRENGTHEN trial examining whether non-invasive Transcranial Electrical Stimulation with Temporal Interference (TES-TI) enhances slow wave activity (SWA, 0.5–4 Hz) during NREM sleep in healthy adults aged 18–50 (N = 28). Stimulation targeted the left ventromedial prefrontal cortex using two high-frequency carriers with a 1 Hz difference frequency (TES 15kHz-TI 1Hz) during ~10 three-minute periods in the first half of the night over a 4-week protocol (1–2 nights/week). Participants were allocated across four parallel groups: Group 1 received control stimulation without temporal interference (TES 15kHz, n = 7) plus meditation; Group 2 received TES-TI 2 nights/week plus sham meditation (n = 5); Group 3 received TES-TI 1 night/week plus meditation (n = 10); and Group 4 received TES-TI 2 nights/week plus meditation (n = 6). SWA and higher frequencies were recorded using simultaneous high-density EEG and polysomnography.
What was found
TES-TI (pooled N = 21 across Groups 2–4) increased SWA, with the enhancement outlasting the active stimulation period, while higher frequencies (sigma and beta) decreased. During stimulation, SWA was greater with TES-TI than with pure TES 15kHz control stimulation (Group 1, N = 7). The increase in SWA between the first and last intervention night was positively associated with subjective ratings of restorative sleep. The abstract reported no exact numerical values, effect sizes, confidence intervals, or p-values.
Why it matters
This provides proof-of-concept human data that non-invasive temporal interference electrical stimulation can selectively boost deep slow-wave oscillations during NREM sleep, a key electrophysiological marker of restorative sleep.
Limits
Allocation was non-randomized and single-blind, introducing risk of selection bias. The sample size was small (total N = 28, with individual intervention groups as small as 5 to 6 participants). Co-interventions (meditation vs. sham meditation and differing weekly frequencies) complicate group comparisons. The abstract provides no quantitative statistical metrics, and functional or cognitive outcomes beyond subjective restorative ratings were not evaluated.