127 Electrophysiological effects of deep brain stimulation in anorexia nervosa epochs across time (i.e. with multiple observations per person). Repeated-measures comparisons were used to test effects over time, which was performed in two ways. First, we used repeated measures model to compare change of power (alpha and beta bands) over time within the patient group. When power was first averaged across channels in four regions (anterior L and R, posterior L and R), the results were Bejamini-Hochberg False Discovery Rate (FDR) corrected for the eight band x region combinations. The second method for repeated effects compared difference scores of T2-T1 and T4-T1 (which were FDR corrected with across channels for multichannel comparisons). These were tested against an alpha level of 0.05/2 to correct for the two comparisons across time. Between-subject analysis was performed to test for significant effects of AN against HC. For graph parameters (single observation per person) simple t-tests were used to compare cases versus controls. Comparing band power (alpha, beta) between AN (at T1) and HC was performed on the power averaged across channels within four regions, using simple t-tests and corrected for the eight band x region combinations. Finally, power spectra were also compared between cases and controls for alpha peak frequency differences by averaging the spectra over the regions (see above) and tested with simple t-tests using FDR corrected for the eight band x region combinations. RESULTS Sample characteristics From 2016 to 2020, four female AN patients were enrolled. Mean age was 39 (SD=10) and illness duration 21 years (SD=3). Average baseline BMI was 12.5 (SD=1.0) kg/m², indicating extremely severe AN. Eight control subjects were matched for age. The controls had a BMI between 18.5 and 25 kg/m², were right-handed, and had no history of mental illness. As reported earlier, in the patient group, at T1, monopolar DBS using the middle two contacts was switched on (pulse width 90μs, frequency 130ms) at a mean voltage of 3.0V (2.5–3.5V). The mean voltages at T2, T3 and T4 were respectively 3.8V (3.0–5.0V), 3.8V (3.0–4.5V) and 3.8V (2.7–4.8V). Adjustment of the stimulation intensity occurred in steps of 0.5V, later fine-tuning in steps of 0.1V. Stimulated contacts, pulse width and frequency remained unchanged during the study. Significant clinical improvements were seen during the study, including increases in BMI and decreases in psychiatric symptom questionnaire scores, as described previously (1). This was mainly attributable to two responders, although all patients reported subjective improvements. No intraoperative adverse events were observed. However, 28 severe adverse events (SAE’s) occurred, with two being probably ((hypo)manic symptoms) and nine possibly (self-destructive behaviour) related to the intervention. Most SAE’s were related to (somatic) AN severity rather than DBS (n=11) (1).
RkJQdWJsaXNoZXIy MjY0ODMw