间歇性Theta爆发刺激联合脑电生物标志物监测治疗成人注意缺陷多动障碍：一项随机双盲假对照试验

Intermittent Theta-Burst Stimulation Combined with EEG Biomarker Monitoring for the Treatment of Adult Attention-Deficit/Hyperactivity Disorder: A Randomized Double-Blind Sham-Controlled Trial

摘要: 成人注意缺陷多动障碍(ADHD)的执行功能损害和核心症状对神经调控治疗的反应异质性较大，缺乏基于脑电生物标志物的个体化疗效预测。间歇性Theta爆发刺激(iTBS)可调节背外侧前额叶皮质(DLPFC)活性，但联合多模态评估(行为、电生理、临床)的随机对照试验证据有限。本文旨在评估左侧DLPFC iTBS对成人ADHD执行功能、核心症状及脑电功率谱的疗效，并探索静息态额叶theta/beta比值(TBR)对治疗反应的预测价值。采用随机、双盲、假刺激平行对照、优效性设计，纳入2023年1月至2024年12月两家三甲医院精神科收治的100例成人ADHD患者(18~45岁)，按1:1随机分配至iTBS组(n = 50)或假刺激组(n = 50)。iTBS组接受左侧DLPFC iTBS (80%静息运动阈值，每序列600脉冲，每日2序列，间隔15分钟，连续10个工作日)，假刺激组接受假线圈刺激。主要结局：ADHD自评量表(ASRS)总分从基线至干预结束(第2周末)的变化。次要结局：(1) 临床总体印象–严重度(CGI-S)；(2) 执行功能行为评定量表成人版(BRIEF-A)；(3) 剑桥神经心理测试自动化成套(CANTAB)中停止信号反应时(SSRT)和空间工作记忆错误；(4) 静息态脑电图(64导联)的额叶TBR (Fz、Cz、Pz平均)及额叶theta功率变化；(5) 持续注意力测验(CPT)的遗漏错误和虚报错误。评估时间点：基线(T0)、治疗第5天(T1，中期)、治疗结束(T2)、治疗后4周(T3)及12周(T4)。采用线性混合模型(LMM)分析主要及次要结局，以年龄、性别、基线TBR作为协变量。同时进行亚组分析(高TBR ≥ 4.0 vs.低TBR)。安全性记录不良事件。92例患者完成全程随访(iTBS组47例，假刺激组45例)。基线两组人口学、临床及电生理指标无显著差异。LMM显示，ASRS总分的组别 × 时间交互效应显著(F = 27.85, P < 0.001)。T2时，iTBS组ASRS较基线下降15.2 ± 4.8分，假刺激组下降4.5 ± 4.1分(组间差值10.7分，95% CI: 8.6~12.8，Cohen’s d = 1.72)。BRIEF-A总分的改善也显著优于假刺激组(P < 0.001)。CANTAB SSRT在iTBS组缩短51.2 ± 24.5 ms，假刺激组缩短9.8 ± 22.1 ms (P < 0.001)。脑电图方面，iTBS组治疗后额叶TBR显著降低(从4.85 ± 1.02降至4.02 ± 0.95)，假刺激组无变化(4.79 ± 0.98至4.70 ± 1.01，组间P = 0.002)。高TBR亚组(≥4.0)较之低TBR亚组显示出更大的ASRS改善(下降幅度差异6.3分，P = 0.01)。不良事件轻微，无组间差异。疗效维持至T4。左侧DLPFC iTBS可显著改善成人ADHD的核心症状和执行功能，同时降低额叶theta/beta比值。基线高TBR可预测更优的治疗反应。iTBS有望成为ADHD个体化神经调控治疗策略。

Abstract: Adult attention-deficit/hyperactivity disorder (ADHD) exhibits marked heterogeneity in response to neuromodulation, both in terms of executive function impairment and core symptoms, yet there remains a lack of electroencephalography (EEG)-based biomarkers for individualized efficacy prediction. The efficacy of intermittent Theta-burst stimulation (iTBS) has been shown to modulate dorsolateral prefrontal cortex (DLPFC) activity; however, evidence from randomized controlled trials (RCTs) combined with multimodal biomarkers (behavioral, electrophysiological, clinical) remains limited. This study aims to evaluate the effects of left dorsolateral prefrontal cortex (DLPFC) iTBS on executive function, core symptoms, and electroencephalographic (EEG) power spectra in adult ADHD, and to explore the predictive value of resting-state frontal theta/beta ratio (TBR) for treatment response. This randomized, double-blind, sham-controlled, parallel-group superiority trial enrolled 100 adults with ADHD (aged 18~45 years) from the Department of Psychiatry at two tertiary hospitals between January 2023 and December 2024. Participants were randomly allocated 1:1 to active iTBS group (n = 50) or sham group (n = 50). Active iTBS over left DLPFC (80% resting motor threshold, 600 pulses/session, two daily sessions with 15-min inter-session interval, 10 weekdays) was compared with sham coil. Primary outcome was change in ADHD Self-Report Scale (ASRS) total score from baseline to post-intervention (end of week 2). Secondary outcomes included: (1) Clinical Global Impression-Severity (CGI-S); (2) Behavior Rating Inventory of Executive Function-Adult Version (BRIEF-A) global score; (3) Cambridge Neuropsychological Test Automated Battery (CANTAB) stop-signal reaction time (SSRT) and spatial working memory errors; (4) resting-state EEG frontal TBR (average of Fz, Cz, Pz) and frontal theta power change; (5) Continuous Performance Test (CPT) omission and commission errors. Assessments were performed at baseline (T0), mid-treatment (day 5, T1), post-intervention (T2), and at 4-week (T3) and 12-week follow-ups (T4). Linear mixed models (LMM) were used to analyze primary and secondary outcomes, covarying for age, sex, and baseline TBR. Subgroup analyses (high TBR ≥ 4.0 vs. low TBR) were pre-specified. Adverse events were monitored. Ninety-two participants completed the full protocol (47 active, 45 sham). No significant differences were found between the two groups at baseline in terms of demographic, clinical, or electrophysiological characteristics. LMM revealed a significant group × time interaction for ASRS (F = 27.85, P < 0.001). At T2, ASRS decreased by 15.2 ± 4.8 points in the active iTBS group vs. 4.5 ± 4.1 in sham group (between-group difference 10.7 points, 95% CI: 8.6~12.8, Cohen’s d = 1.72). BRIEF-A improvement was also superior in the active iTBS group (P < 0.001). CANTAB SSRT shortened by 51.2 ± 24.5 ms in active iTBS group vs. 9.8 ± 22.1 ms in sham group (P < 0.001). EEG analyses showed a significant reduction in frontal TBR after treatment in active iTBS group (from 4.85 ± 1.02 to 4.02 ± 0.95) but not in sham group (4.79 ± 0.98 to 4.70 ± 1.01, between-group P = 0.002). The high-TBR subgroup (≥4.0) exhibited greater ASRS improvement (difference 6.3 points, P = 0.01) than low-TBR subgroup. Adverse events were mild, with no between-group differences. Therapeutic effects were sustained through T4. Left DLPFC iTBS significantly improves core symptoms and executive function in adult ADHD, accompanied by a reduction in frontal theta/beta ratio. Baseline high TBR predicts better treatment response. iTBS may serve as a personalized neuromodulation strategy for ADHD.