Search for a command to run...
To the Editor: Transcranial alternating current stimulation (tACS), a noninvasive brain stimulation (NIBS) technique, alters neural activity and behavior in humans by delivering low-intensity sinusoidal alternating current at a specific frequency (Hertz) to targeted brain areas through electrodes placed on the scalp.[1] tACS regulates endogenous cortical oscillations through entrainment mechanisms to improve brain network communication.[2] Over the past decade, it has been applied in various fields and has demonstrated promising outcomes in reducing psychiatric and neurological symptoms.[3] In recent years, the application of tACS in regulating arousal, sleep, and sleep-related plasticity has gained increasing attention. Regulating the cortical components of the sleep control system using tACS can facilitate or inhibit sleep onset processes. Additionally, tACS noninvasively affects sleep macrostructure, brain oscillations, and arousal.[4] As a treatment, tACS offers significant operational advantages and potential to improve physiological sleep. Despite its use in studies on sleep regulation, stimulation parameters, including electrode positioning, current intensity, frequency, and intervention protocols, remain highly variable, which can substantially impact outcomes related to sleep improvement. A scoping review and meta-analysis was conducted to examine clinical trials of tACS in sleep regulation, both qualitatively and quantitatively. The purpose of this review is to highlight the overall efficacy of tACS and assess the influence of individual parameters on sleep. The study adheres to the PRISMA 2020 checklist [Supplementary Table 1, https://links.lww.com/CM9/C785]. We conducted a comprehensive search of the MEDLINE, Embase, PsycINFO, CENTRAL, and ClinicalTrials.gov databases from their inception to May 9, 2024. Detailed information regarding the search terms, strategy, inclusion and exclusion criteria, and data extraction methods is provided in Supplementary Material 1 and Supplementary Table 2, https://links.lww.com/CM9/C785. Meta-analysis was performed using Review Manager software (RevMan 5.3, Cochrane Collaboration, Oxford, England). For continuous outcomes, the mean difference (MD) with a 95% confidence interval (CI) was calculated. When outcomes were assessed across studies but measured using different scales, the standardized mean difference (SMD) was applied. Heterogeneity was evaluated using the Q statistic and I2. A nonsignificant Q statistic (P >0.1) combined with an I2 ≤50% was interpreted as low heterogeneity among studies, favoring the use of a fixed-effects model. Conversely, a random-effects model was used when significant heterogeneity was present. Statistical significance was defined as a P value less than 0.05. Publication bias was analyzed through funnel plots and Egger’s test. Egger’s test, which quantitatively assesses funnel plot asymmetry, considered a P value less than 0.1 indicative of significant publication bias. Sensitivity analysis using the leave-one-out method was performed to assess the robustness of the results. Subgroup analyses explored potential sources of heterogeneity, including variations in the study methods, age, health conditions, stimulation onset, and assessment tools. Quality evaluation and risk of bias assessment were conducted using the Cochrane Risk of Bias 2.0 (RoB 2) tool for randomized trials and the Risk Of Bias In Non-randomized Studies–of Interventions (ROBINS-I) tool for non-randomized studies, respectively. A total of 4950 references were identified through the electronic database search, of which 16 studies met the inclusion criteria [Supplementary Figure 1 and Supplementary Table 3, https://links.lww.com/CM9/C785]. These studies, published between 2015 and 2024, included sample sizes ranging from 8 to 120 participants, with five studies involving more than 30 participants [Supplementary Figure 2, https://links.lww.com/CM9/C785]. Key characteristics of the included studies are summarized in categories [Supplementary Table 4, https://links.lww.com/CM9/C785]. The parameters of tACS varied across the studies in sleep treatment. Supplementary Figure 3, https://links.lww.com/CM9/C785 provides a schematic representation of electrode positions and sizes used in earlier studies. Beyond electrode placement, the current intensity and frequency of stimulation also exhibited significant variation [Supplementary Figure 4, https://links.lww.com/CM9/C785]. Studies were classified into two categories: (1) usual tACS and (2) closed-loop tACS, based on the stimulation type [Supplementary Table 4, https://links.lww.com/CM9/C785]. Closed-loop tACS is more advanced and precise than usual tACS. However, this approach lacks essential technical support and requires further development and refinement, making usual tACS the more commonly used method in current research. Despite its prevalence, usual tACS protocols exhibit high variability in duration, intervals, and repetitions due to the open nature of the stimulation. The studies were conducted under diverse brain states, with seven of them applying stimulation primarily during nonrapid eye movement (NREM) sleep. The primary outcome indicators in tACS sleep regulation research can be categorized into four domains: (1) macroscopic sleep structure, (2) subjective sleep quality, (3) neural oscillatory rhythm, and (4) cognitive function. Ten studies reported on sleep macrostructure. This was assessed using polysomnography (PSG) or questionnaires, capturing various metrics such as total sleep time (TST), sleep onset latency (SOL), sleep efficiency (SE), and wakefulness after sleep onset (WASO). Improvements in sleep macrostructure following tACS showed great heterogeneity. Some studies reported deepening and prolongation of sleep, indicated by an increased proportion of time spent in N2 and N3 sleep stages. Conversely, other studies found no effect on sleep stages. Subjective sleep quality data were collected in 12 studies using questionnaires, with the Pittsburgh sleep quality index and insomnia severity index being the most frequently utilized. Most studies demonstrated a positive effect of tACS on subjective sleep quality, with participants experiencing better sleep and greater sleepiness after the intervention. Brain oscillations, measured using electroencephalography (EEG) or PSG (including slow waves, delta, theta, alpha, beta, and spindle waves), were reported in nine studies. The definitions of the amplitude for each EEG band varied across studies and did not consistently align with the standards set by the American Academy of Sleep Medicine. The findings on brain oscillations, which involved multiple EEG band activities, exhibited high variability [Supplementary Table 5, https://links.lww.com/CM9/C785]. Sleep-related cognitive changes were investigated in eight studies, which indicated that tACS may support cognitive function improvements while regulating sleep. Regarding safety, most studies reported no serious adverse effects. The adverse effects observed were generally mild, such as mild scalp pain, localized pressure, or burning sensations, resolved shortly after stimulation, and did not require specific treatment, indicating that tACS is well tolerated. Following the screening of studies included in the scoping review, 11 studies were selected for the meta-analysis [Supplementary Figure 1, https://links.lww.com/CM9/C785]. The estimated effect size compared the sleep performance of individuals receiving tACS interventions to that of placebo controls [Figure 1 and Supplementary Figure 5, https://links.lww.com/CM9/C785]. Funnel plot analysis [Supplementary Figure 6, https://links.lww.com/CM9/C785] and Egger’s test (P = 0.62) provided no evidence of publication bias. Risk of bias was assessed for the 11 studies using the ROB2 [Supplementary Figure 7, https://links.lww.com/CM9/C785].Figure 1: Summary the forest plot and effect estimate of the tACS objective and subjective sleep after-effect in meta-analysis. REM: Rapid eye movement; tACS: Transcranial alternating current stimulation.* P <0.05.For objective sleep outcomes, there was no statistically significant improvement in TST (MD 2.23; 95% CI = −6.29, 10.75; P = 0.61), SE (MD 0.66; 95% CI = −1.67, 3.00; P = 0.58), or NREM 2 latency (MD −8.39; 95% CI = −38.37, 21.59; P = 0.58) following tACS intervention. Similarly, no statistical significant difference was observed in other metrics such as SOL, the percentage of N1, N2, N3, and REM, latency of REM, and WASO. Objective sleep outcome indicators exhibited no significant heterogeneity among studies (P >0.01 and/or I2 <50%). For subjective sleep outcomes, a statistically significant increase in subjective TST was noted after tACS intervention (MD 0.38; 95% CI = 0.04, 0.72; P = 0.03; heterogeneity, I2 = 44%; P = 0.17). However, SOL showed no statistical significance (MD −9.39; 95% CI = −22.47, 3.69; P = 0.16; heterogeneity, I2 = 2%; P = 0.13). The subjective sleep quality, measured in nine studies using the SMD approach, was also not statistically significant (SMD −0.34; 95% CI = −0.70, 0.02; P = 0.07; heterogeneity, I2 = 61%; P = 0.009). Sensitivity analyses showed no change in heterogeneity, suggesting robust results. Subgroup analyses were conducted to investigate potential sources of heterogeneity. The results indicated that factors such as age, health conditions, and the timing of stimulation contributed to the observed variability to some extent [Supplementary Figure 8, https://links.lww.com/CM9/C785]. This review synthesizes previous tACS intervention programs and their primary findings related to sleep. Unlike earlier studies, this review includes a meta-analysis with a quantitative assessment of sleep outcome indicators. However, we found that most indicators showed no significant changes. This outcome may be attributed to the low level of available evidence and small sample sizes. Smaller samples are susceptible to inflated effect size estimates and publication bias, reducing the statistical power and generalizability of the findings.[5] Addressing these limitations, future large-scale, multicenter studies are necessary to guide the optimization and application of this technology. Our analyses revealed substantial heterogeneity observed in tACS modulation of sleep. Given the high heterogeneity in existing findings and the complexity of the sleep–wake network, further well-designed clinical trials are needed. Studies involving closed-loop tACS and personalized parameter adjustments—such as those based on computational modeling of electric field intensities—are particularly promising. Furthermore, the precise mechanisms by which tACS affects sleep remain undefined; current conclusions rely primarily on theoretical models from previous studies and empirical observations. From the reviewed literature, we propose two potential mechanistic pathways [Supplementary Material 2 and Supplementary Figure 9, https://links.lww.com/CM9/C785]. Therefore, monitoring multimodal data, including multi-omics, imaging techniques, and sleep electrophysiology over extended periods, will enable detailed observation of macroscopic and microscopic changes in brain morphology and function before and after tACS intervention. These approaches will help reveal the mechanisms by which tACS improves sleep and provide a theoretical foundation for its clinical application. Funding This work was supported by the Ministry of Science and Technology of the People' s Republic of China (No. 2021ZD0201900), the National Natural Science Foundation of China (General Program, No. 82270106), and the National Clinical Research Center for Geriatrics (West China Hospital of Sichuan University) (No. Z2024LC002). Conflicts of interest None.