Chronic Insomnia Disorder (CID) is a highly prevalent sleep disorder, affecting approximately 24% of the adult population. It is characterized by persistent difficulties in sleep initiation, sleep maintenance, or early morning awakening, accompanied by significant daytime functional impairments such as fatigue, inattention, mood disturbances, and cognitive decline. CID is highly comorbid with anxiety and depression and has been identified as a potential risk factor for cardiovascular disease, imposing a substantial burden on patients' quality of life, social functioning, and the public health system. Currently, although Cognitive Behavioral Therapy for Insomnia (CBT-I) is recommended as the first-line treatment, it has limitations including limited accessibility and high demands on patient adherence. Pharmacological treatments, while rapid in onset, face issues such as dependency risks, side effects, and insufficient improvement in sleep continuity. In recent years, repetitive Transcranial Magnetic Stimulation (rTMS), as a non-invasive neuromodulation technique, has shown potential in treating CID. However, its precise neuromodulatory mechanisms remain unclear, and there is a lack of predictive tools for treatment response based on objective brain function indicators. functional Near-Infrared Spectroscopy (fNIRS), with its portability, resistance to interference, and good spatiotemporal resolution, offers a new technological approach for investigating cortical functional changes and identifying biomarkers of treatment response.

Based on the "hyperarousal" theory of CID, this study aims to systematically evaluate the clinical efficacy and safety of low-frequency rTMS targeting the right dorsolateral prefrontal cortex (DLPFC) combined with zolpidem for CID through a randomized controlled trial design; to elucidate the neural mechanisms of rTMS in modulating brain activity using fNIRS from both task-state and resting-state perspectives; and to explore potential biomarkers based on baseline fNIRS features for predicting individualized treatment response.

A total of 84 patients with CID were enrolled and randomly assigned to either the rTMS group (n=40) or the control group (n=44). All patients received 10 mg of zolpidem nightly. The rTMS group additionally underwent two weeks (5 sessions/week) of 1 Hz rTMS applied to the right DLPFC (stimulation intensity at 80% resting motor threshold, 1200 pulses per session). Clinical scale data, including the Pittsburgh Sleep Quality Index (PSQI) and Insomnia Severity Index (ISI), and fNIRS brain function data (including resting-state and Verbal Fluency Task, VFT) were collected before and after the treatment period. Repeated measures ANOVA was used to evaluate inter-group efficacy differences. Neural mechanisms were analyzed through changes in fNIRS metrics. Responder analysis (PSQI reduction rate ≥ 50%) was employed to explore the predictive value of baseline brain function characteristics.

1. Clinical Efficacy: The rTMS group showed significantly superior time-by-group interaction effects compared to the control group in PSQI total score, sleep duration, sleep efficiency, sleep disturbance, and ISI total score (p < 0.05). Unique within-group improvements were observed in the rTMS group for the daytime dysfunction and sleep disturbance sub-items. rTMS was well-tolerated, with only mild adverse reactions reported.

2. Neural Mechanisms: Following rTMS treatment, patients demonstrated significantly increased activation in the left DLPFC during the VFT. In the resting state, significant decreases in the amplitude of low-frequency fluctuations (ALFF) were observed in the right DLPFC and the right frontopolar area (FPA). Furthermore, functional connectivity was significantly decreased between the right DLPFC-left Broca's area and the left FPA-left Broca's area (all p < 0.05, FDR corrected).

3. Efficacy Prediction: Treatment responders (n=15) were characterized at baseline by higher task-state activation levels in the left DLPFC region and lower resting-state ALFF values in the right DLPFC region. Both of these baseline metrics were significantly correlated with the degree of PSQI improvement (p < 0.05).

Low-frequency rTMS targeting the right DLPFC can safely and effectively improve sleep quality and daytime function in patients with CID, serving as a valuable adjunct to existing pharmacotherapy. Its neural mechanism may involve a dual-pathway "inhibition-compensation" regulation: inhibiting right prefrontal hyperarousal and associated aberrant functional connectivity on one hand, while enhancing compensatory activation of the left prefrontal cortex during executive function tasks on the other. Baseline fNIRS features (high left DLPFC task-state activation, low right DLPFC resting-state activity) show promise as objective biomarkers for predicting rTMS efficacy, providing a theoretical basis and practical reference for precision neuromodulation therapy in CID.