Efficacy and exploratory analysis of potential mechanisms of stellate ganglion block in alleviating sleep disturbance in generalized anxiety disorder: a randomized controlled trial excluding comorbid depression

Objective To investigate the efficacy and mechanisms of stellate ganglion block (SGB) in treating generalized anxiety disorder (GAD) with sleep disturbance, excluding patients with comorbid depression. Methods This double-blind randomized controlled trial (RCT) enrolled 128 patients with GAD (Hamilton Anxiety Scale [HAMA] > 14, Generalized Anxiety Disorder 7-item Scale [GAD-7] ≥ 5) and sleep disturbance (Pittsburgh Sleep Quality Index [PSQI] ≥ 15), randomized to receive SGB ( n = 64, 4 ultrasound-guided 1% lidocaine injections) or conventional treatment ( n = 64, cognitive behavioral therapy [CBT] + estazolam 1–2 mg/day). Outcomes included anxiety (HAMA), depression (Hamilton Depression Scale [HAMD]), sleep quality (PSQI), polysomnography (PSG), and neurotransmitter levels (norepinephrine [NE], serotonin [5-HT], neuropeptide Y [NPY]). Results After 4 weeks, SGB demonstrated higher efficacy (98.4% vs. 89.1%, p = 0.028) and greater reductions in HAMA (9.36 ± 2.34 vs. 11.87 ± 2.71, p < 0.001) and HAMD scores (6.87 ± 2.01 vs. 8.09 ± 2.04, p < 0.001). PSQI improved significantly in the SGB group (5.74 ± 1.64 vs. 8.03 ± 1.86, p < 0.001), with increased total sleep time (TST) (429.76 ± 33.22 vs. 391.13 ± 30.76 min, p < 0.001) and efficiency (90.23 ± 13.29% vs. 86.34 ± 12.84%, p < 0.001). Neurotransmitter analysis showed reduced NE (289.43 ± 51.68 vs. 253.78 ± 57.12 pg./mL, p < 0.05) and increased 5-HT (138.56 ± 19.73 vs. 124.93 ± 18.44 ng/mL, p < 0.05) and NPY (453.21 ± 73.41 vs. 402.34 ± 68.12 pg./mL, p < 0.05). Adverse events were comparable (6.25% vs. 3.13%, p = 0.403). Conclusion SGB effectively improves GAD symptoms and sleep quality in patients without comorbid depression, potentially via modulation of NE, 5-HT, and NPY pathways. The exclusion of psychiatric comorbidities enhances the specificity of these findings.