5G Radio-Frequency-Electromagnetic-Field Effects on the Human Sleep Electroencephalogram: A Randomized Controlled Study in CACNA1C Genotyped Healthy Volunteers.

The introduction of 5G technology as the latest standard in mobile telecommunications has raised concerns about potential health effects. In particular, individuals who self-identify as electrohypersensitive are concerned about sleep disturbances, headaches, and related brain impairments due to exposure to radiofrequency electromagnetic fields (RF-EMF). While the long-term health effects of EMF exposure remain unclear, several independent studies have shown that 2G to 4G EMFs acutely alter electroencephalographic (EEG) oscillations in the ca. 9-16 Hz range during wakefulness and sleep.

Notably, repeated spectral increases in the EEG spindle frequency range (11-16 Hz) have been observed during non-rapid eye movement (NREM) sleep. Sleep spindles modulate interactions between the brain and the external environment, essentially dampening responsiveness to sensory stimuli—effectively isolating the brain from external disturbances during sleep. The pulse modulation of 2G to 4G EMF signals appears to be critical for inducing EEG changes during sleep. The high inter-individual variability and strong intra-individual stability of EMF-induced sleep EEG changes suggest a possible genetic predisposition.

Based on other research, the likely mechanism of EMF exposure effects on the brain involves the depolarization of neuronal membrane potentials, which activates voltage-gated calcium ion channels (Ca²⁺), and leads to increased intracellular Ca²⁺ concentration. The influx of Ca²⁺ regulates processes such as hormone secretion, neurotransmitter release, gene transcription, and neuronal activity. RF-EMFs can activate L-type voltage-gated calcium channels (LTCC), which are associated with sleep quality and oscillatory EEG activity.

Different allelic variants of the CACNA1C gene, which encodes the α1C subunit of the L-type calcium channel, have been linked to prolonged sleep latency and reduced sleep quality. This subunit determines the voltage sensitivity and conductivity of LTCCs. LTCCs are expressed in nearly all types of neurons in the brain and regulate neuronal firing, learning, memory, addictive behaviors, and neural development.

The aim of this study was to investigate whether pre-sleep exposure to realistic, standardized 5G EMF signals affects the spectral properties of spindles in the NREM sleep EEG. In addition, the study aimed to determine whether EMF-induced changes are modulated by the rs7304986 variant of the CACNA1C gene (T/C or T/T alleles). A novel methodological approach, the Fitting Oscillations & One Over f (FOOOF) analysis, was used. This analysis provides a validated, intuitive method for reliable and meaningful extraction of individual spectral EEG features.

Thirty-four healthy, right-handed volunteers, predominantly women, were enrolled in this study and genotyped for rs7304986 (15 T/C carriers and 19 T/T carriers). Participants completed questionnaires regarding mobile phone use, medication use, sleep behavior, and general and neurological health status.

All participants underwent three experimental nights with different standardized exposure conditions in a randomized, double-blind crossover design:

1. 30-minute exposure before bedtime to active 5G EMF at a 700 MHz carrier frequency, 20 MHz bandwidth, and 12.5 Hz power control;
2. 30-minute exposure before bedtime to active 5G EMF at a 3.6 GHz carrier frequency, 100 MHz bandwidth, and 12.5 Hz power control;
3. 30-minute sham exposure with no active field.
    

The exposure system (sXh5G, IT'IS Foundation, Zurich, Switzerland) was calibrated to ensure that the specific absorption rate (SAR) for the head did not exceed 2 W/kg. The output power was 4.28 W at 700 MHz and 1.63 W for the 3.6 GHz signal. Identical power control was applied to both signals, introducing low frequency amplitude modulation at 12.5 Hz. Peak exposures were concentrated in the cortical tissue closest to the antenna, with a much steeper SAR decay at the higher frequency.

For EEG recordings, the researchers used 128-channel Electrical Geodesics sensor nets for overnight high-density EEG (hd-EEG) monitoring (Electrical Geodesics Inc., EGI, Eugene, OR).

T/C carriers reported longer sleep latency compared to T/T carriers. Statistical analysis of nighttime sleep variables revealed an interaction between “exposure” and “genotype” for later sleep stages, although post hoc t-tests showed no significant differences.

Further analysis identified distinct negative and positive peaks in specific power ratios, suggesting a shift in spindle peak frequency rather than a general increase in spectral power density.

When analyzing the periodic components of oscillatory spindle activity in NREM sleep EEG, a significant interaction between “exposure” and “genotype” was found for the center frequency of sleep spindle activity.

Finally, a topographical comparison (i.e., high-density EEG analysis) showed a widespread shift toward higher spindle frequencies in T/C allele carriers after exposure to the 3.6 GHz field. This effect covered a large cluster of central, parietal, and occipital cortical areas in 50 out of 109 EEG channels.

The percentage increase in spindle center frequency was 1.43 ± 6.5 × 10⁻⁴ %, corresponding to an average shift from 13.62 ± 0.1 Hz in the sham condition to 13.82 ± 0.1 Hz after 3.6 GHz exposure. The acceleration of spindle center frequency in T/C genotype participants after 3.6 GHz exposure was consistent, with a large effect size (Cohen's d mean ± SD = 0.78 ± 0.18; Cohen's d [min, max] = [0.28, 1.28]; Cohen's d > 0.57 in 48 out of 50 channels).

Using the recently developed FOOOF algorithm, this study identified 5G-induced changes in spindle peak components in the NREM sleep EEG. A significant interaction was found between exposure and the genetic variant in the center frequency of sleep spindles.

The researchers observed a widespread shift in the center frequency of sleep spindles toward faster oscillatory activity in T/C allele carriers after exposure to a 5G RF-EMF with a carrier frequency of 3.6 GHz, implicating L-type voltage-gated calcium channels in the physiological response to RF-EMF. A smaller shift was also observed after exposure to the 700 MHz signal.

If present, the effect of the lower frequency 700 MHz field is only marginally detectable with the current methodology. The discrepancy between the deeper penetration of the 700 MHz signal (as indicated by the simulated SAR distribution in the brain) and the more pronounced effects on sleep spindles observed after 3.6 GHz exposure remains unclear.

These results highlight the need for a comprehensive investigation into the complex properties of the new 5G signals. Knoblauch et al. (2005) demonstrated the circadian regulation of spindle center frequency, showing a decrease (from about 13.85 to 13.7 Hz) coinciding with melatonin secretion. Since an increase in spindle center frequency was observed after 5G exposure in this study, a circadian effect of RF-EMFs, such as reduced melatonin production, cannot be ruled out (although melatonin levels were not measured in this study).

These findings provide preliminary evidence that the LTCC type Cav1.2 may play a mechanistic role in the interaction between EMF exposure and the human brain. This hypothesis could be further tested by examining the effects of RF-EMF on sleep EEG after selective pharmacological modulation of these channels. (Furthermore, only individuals with the higher-sensitivity LTCC variant were susceptible to EMF exposure, suggesting that genetic factors play a role in electrosensitivity, editor's note.) (AT)