Effect of high-frequency stellate ganglion block on sleep and affective symptoms in insomnia disorder: a prospective pilot study

Abstract

Background

Insomnia disorder imposes substantial health burdens, while current pharmacotherapies risk adverse effects and cognitive behavioral therapy face implementation barriers. Stellate ganglion block (SGB) modulates autonomic function and sleep architecture but lacks evidence for optimized protocols. This pilot study evaluates the feasibility and preliminary clinical outcomes of a high-frequency SGB protocol for improving sleep, anxiety, and depression in insomnia disorder.

Methods

A single-center, prospective single-arm trial enrolled 72 patients diagnosed with insomnia disorder according to ICSD-3 criteria. Participants received daily ultrasound-guided SGB (0.8% lidocaine, 3 ml) unilaterally for 10 consecutive days. Outcomes included Pittsburgh Sleep Quality Index (PSQI), Generalized Anxiety Disorder-7 (GAD-7), and Patient Health Questionnaire-9 (PHQ-9) scores assessed at baseline (T1), immediately post-intervention (T2), and at 4-week (T3) and 12-week (T4) follow-up. Statistical analysis used repeated-measures ANOVA.

Results

The mean PSQI score improved from 15.35 ± 2.39 at baseline to 10.59 ± 3.23 post-intervention (mean reduction 4.76, 95% CI [4.04, 5.49], P < 0.001), exceeding the minimal clinically important difference (MCID) of 3 points. This improvement was sustained at 12 weeks (mean reduction 4.08, 95% CI [3.32, 4.85]). Anxiety (GAD-7: baseline 7.47 ± 4.57) and depression (PHQ-9: baseline 9.07 ± 5.72) scores showed statistically significant reductions post-treatment (GAD-7: T2-T4 P < 0.05; PHQ-9: T2-T4 P < 0.05). However, the magnitude of improvement did not reach established MCID (4 points for GAD-7, 5 points for PHQ-9), with mean reductions ranging from 0.99 to 2.18 points for GAD-7 and 1.72 to 2.04 points for PHQ-9. This suggests that while statistically detectable, the clinical relevance of affective symptom improvements may be limited, particularly given the relatively mild baseline symptom severity.

Conclusion

This pilot study demonstrates that a high-frequency SGB protocol is feasible to implement in patients with insomnia disorder. Within-group improvements in sleep quality that exceeded the MCID were observed and sustained through 12-week follow-up. However, due to the uncontrolled single-arm design, these results should be interpreted as hypothesis-generating rather than confirmatory. The observed improvements may be attributable to placebo effects, natural history, or regression to the mean. These findings provide effect size estimates to inform the design of future randomized sham-controlled trials, which are necessary to establish causal efficacy.

Trial registration

This study was registered on the Chinese Clinical Trial Registry on March 19th, 2025 (Registration No.: ChiCTR2500099219).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12883-026-04781-0.

Introduction

Insomnia disorder represents one of the most prevalent sleep disorders, affecting approximately one-third of the global adult population [1]. The clinical definition of insomnia disorder is a persistent sleep disorder characterized by difficulty initiating sleep, maintaining sleep, or returning to sleep after waking, despite having adequate opportunities and a suitable environment for sleep, and this sleep difficulty leads to impairment of daytime functioning [2]. This condition may present as a primary complaint or coexist with other physical or psychiatric disorders such as chronic pain and depression. Insomnia disorder is associated with multiple adverse health outcomes, including diminished quality of life and elevated risks of psychophysiological disorders [3]. Insomnia disorder significantly impairs daytime functioning, leading to compromised cognitive performance and occupational impairment, frequently accompanied by symptoms of irritability, anxiety, and fatigue [4, 5]. Beyond substantially affecting work performance and quality of life, insomnia disorder markedly increases susceptibility to various psychosomatic disorders and imposes substantial socioeconomic burdens, underscoring its critical importance for public health management [6].

The management of insomnia disorder primarily encompasses pharmacological and non-pharmacological interventions. The U.S. Food and Drug Administration has approved several pharmacological agents for insomnia disorder, including benzodiazepine receptor agonists, classified as traditional benzodiazepines (BZDs) and non-BZDs [7]. They demonstrate efficacy in short-term improvement of subjective total sleep time and reduction of daytime sleepiness and fatigue in most patients. However, pharmacotherapy is associated with significant adverse effects, including the risks of dependence, ataxia, excessive sedation, increased fall incidence, fractures, cognitive decline, and elevated dementia risk [8]. Cognitive Behavioral Therapy for Insomnia (CBT-I) is endorsed as a first-line treatment for insomnia disorder by multiple international clinical guidelines [9]. This evidence-based approach modifies maladaptive sleep-related cognitions and behaviors to reconstruct sustainable sleep patterns, while eliminating risks associated with pharmacological side effects [10, 11]. Nevertheless, the clinical implementation of CBT-I encounters several challenges: extended treatment duration, substantial resource requirements, and demanding patient adherence and engagement thresholds [12]. These logistical constraints currently limit its widespread dissemination in real-world healthcare settings.

The stellate ganglion, formed by the fusion of the inferior cervical ganglion and the first thoracic ganglion (C7-T1), constitutes a critical component of the sympathetic chain. Stellate ganglion block (SGB) involves the injection of local anesthetics near the SG to achieve reversible neural blockade. This intervention modulates autonomic function within SG-innervated territories, enhances cerebral perfusion, reduces local reactive oxygen species levels, and regulates circadian melatonin secretion—collectively facilitating the normalization of pathological sleep architecture and improvement of overall sleep continuity in patients with insomnia disorder [13]. Current research on SGB for insomnia disorder primarily focuses on sleep quality and mood state improvements. Insomnia frequently co-occurs with psychiatric conditions such as anxiety and depression, a comorbidity potentially rooted in shared neurobiological substrates. Recent large-scale neuroimaging studies have identified common structural brain alterations, such as cortical thickness abnormalities and patterns of brain asymmetry, associated with internalizing psychopathology [14, 15]. These findings suggest that the hyperarousal state characteristic of insomnia may be linked to broader neural circuitry dysfunction extending beyond sleep-regulating centers. This conceptualization provides a compelling rationale for interventions that target autonomic regulation, as modulation of these shared pathways could potentially yield transdiagnostic therapeutic benefits for both sleep disturbances and co-occurring affective symptoms. Given this shared neurobiological substrate, interventions targeting autonomic regulation, such as SGB, may hold promise for addressing both sleep and affective disturbances. The core rationale for this high-frequency protocol is to achieve sustained inhibition of sympathetic hyperactivity—a key pathophysiological feature of insomnia—thereby preventing sympathetic rebound during treatment intervals and enhancing cumulative therapeutic effects. Although rigorous controlled studies directly comparing the efficacy of high-versus low-frequency protocols are currently lacking, clinical observations and several small-scale reports suggest that short-term intensive SGB courses may yield positive effects in certain patients with refractory insomnia and anxiety symptoms (based on our preliminary clinical experience and some published case series observations) [16, 17].

Current research on SGB for insomnia disorder has largely focused on improvements in global sleep quality, with insufficient exploration of its effects on specific sleep dimensions (e.g., sleep onset, maintenance, efficiency, and daytime function). Moreover, limited attention has been given to the role of high-frequency SGB in alleviating common comorbid affective symptoms such as anxiety and depression, and there is a lack of systematic tracking regarding the durability of therapeutic benefits. In view of these gaps, the present study adopts a high-frequency SGB protocol (daily administration for 10 consecutive days) to explore whether a high-frequency SGB protocol is associated with improvements in sleep and affective symptoms. We further establish a multidimensional assessment framework encompassing sleep quality, anxiety, and depression, and incorporate the minimal clinically important difference as a criterion for clinical significance. This study systematically investigates the efficacy profile and durability of high-frequency SGB in patients with insomnia disorder, aiming to provide effect size estimates and feasibility evidence to inform future randomized controlled trials.

Methods

Study design

This study constitutes a single-center, single-arm prospective pilot trial designed to investigate changes in depressive symptoms, anxiety manifestations, and sleep parameters following high-frequency SGB therapy in patients diagnosed with insomnia disorder. The trial was conducted at the Third Affiliated Hospital of Guangzhou University of Chinese Medicine, with the research protocol receiving formal approval from the institutional ethics committee (Approval No.: PJ-XS-20250227-003) and being prospectively registered with the Chinese Clinical Trial Registry on March 19th, 2025 (Registration No.: ChiCTR2500099219). Participants underwent comprehensive assessment at four critical timepoints: baseline (pre-treatment) (T1), immediately post-intervention (T2), 4-weeks post-intervention (T3), and 12-weeks post-intervention (T4), evaluating both subjective and objective sleep quality alongside anxiety and depression status. A dedicated data manager oversaw all data entry and management procedures to guarantee timeliness, accuracy, completeness, and standardization of records. Data processing was executed through specialized software, with rigorous confidentiality protocols maintained throughout the entire research continuum.

As this was a pilot study aimed at assessing feasibility and generating preliminary data for future trial planning, a formal hypothesis-testing sample size calculation was not performed [18]. The target sample size of 72 patients was determined based on: (1) expected recruitment capacity at our sleep medicine clinic during the study period (March-September 2025); and (2) recommendations for pilot studies suggesting that approximately 50–70 participants can provide stable estimates of effect sizes and variability to inform future sample size calculations [18]. Based on effect size estimates from previous small-sample studies of SGB in insomnia (ranging from Cohen’s d = 0.3 to 0.7) [16, 17], the present study with 72 participants has 80% power to detect a moderate effect size (Cohen’s d = 0.5) for within-group pre-post comparisons at α = 0.05, but is not powered for confirmatory subgroup analyses or between-group comparisons.

Patients

The study cohort comprised patients diagnosed with insomnia disorder according to the International Classification of Sleep Disorders, Third Edition (ICSD-3) criteria, all recruited from the Sleep Medicine Clinic of the Third Affiliated Hospital of Guangzhou University of Chinese Medicine between March 2025 and September 2025. Final diagnosis of insomnia disorder according to ICSD-3 criteria was confirmed by a board-certified neurologist and sleep medicine specialist. Inclusion criteria were: (1) aged 18 to 65 years regardless of gender; (2) meeting ICSD-3 diagnostic criteria for insomnia disorder, manifesting at least one of the following symptoms: sleep initiation difficulty (sleep latency > 30 min), sleep maintenance disturbance (≥ 2 nocturnal awakenings), early morning awakening, impaired sleep quality, or reduced total sleep time (< 6.5 h), with symptom frequency ≥ 3 times weekly. Exclusion criteria encompassed: (1) comorbid severe cardiovascular or cerebrovascular diseases, hepatic or renal insufficiency, endocrine disorders, or other significant systemic conditions; (2) known allergy to lidocaine or other local anesthetics; (3) coagulation abnormalities or infection at puncture sites; (4) consciousness or cognitive impairments; (5) diagnosis of other primary sleep disorders; (6) patients who fulfill the diagnostic criteria for alcohol use disorder or substance use disorders (encompassing opioids, stimulants, cannabis, and prescription medication abuse); (7) use of any medication known to significantly influence sleep architecture within 4 weeks prior to enrollment, with the exception of prescribed hypnotics that have been used at a stable dose for at least 4 weeks and without dosage adjustment in the 4 weeks prior to enrollment; (8) poor compliance preventing completion of protocol-required data collection and follow-up procedures.

Primary outcome measure

Pittsburgh Sleep Quality Index (PSQI) [19]: PSQI serves as a valuable instrument in clinical practice, effectively screening for the severity of sleep disturbances in patients while providing clinicians with diagnostic clues regarding abnormalities in specific sleep functional domains. Within the research domain, the PSQI is widely employed in epidemiological surveys and clinical studies to differentiate sleep quality characteristics among distinct population groups. Its longitudinal research value is particularly significant, primarily manifested in its capacity to systematically track the clinical outcomes and natural disease progression of sleep-wake disorders. This instrument served as the primary outcome measure for this study. It encompasses seven dimensions of sleep quality: subjective sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbances, use of sleep medications, and daytime dysfunction. Each dimension is scored from 0 to 3, yielding a global score range of 0–21. A higher score corresponds to worse sleep quality. The minimal clinically important difference (MCID) is defined as the smallest difference in treatment outcome that is perceived as clinically meaningful by either patients or physicians under specific treatment conditions [20]. The MCID for the PSQI is 3 points [21]. Assessments were conducted at baseline (pre-treatment), immediately post-intervention, 4-weeks post-intervention, and 12-weeks post-intervention.

Secondary outcome measure

Generalized Anxiety Disorder-7 (GAD-7) [22]: This instrument serves as a screening tool for generalized anxiety disorder and provides quantitative assessment of anxiety symptom severity, representing an internationally validated anxiety screening measure. The scale comprises seven items, each scored 0–3, with total scores ranging from 0 to 21 (sum of all item scores). The Chinese version of GAD-7 has demonstrated satisfactory reliability and validity, with its psychometric properties and clinical utility empirically validated. The MCID for the GAD-7 is 4 points [23]. This instrument was used as a secondary outcome measure to evaluate changes in anxiety symptoms. Assessments were administered at baseline (pre-treatment), immediately post-intervention, 4-weeks post-intervention, and 12-weeks post-intervention. Scores of 5, 10, and 15 are taken as the cut-off points for mild, moderate, and severe anxiety, respectively [24].

Patient Health Questionnaire-9 (PHQ-9) [25]: As an abbreviated version of the Patient Health Questionnaire, this instrument was developed in accordance with DSM-IV diagnostic criteria for depression to facilitate rapid screening and severity assessment of depressive symptoms. The scale comprises nine items corresponding to the core diagnostic symptoms of major depressive disorder, quantifying symptom frequency and functional impact to support clinical evaluation and treatment decisions. Each item is scored 0–3, yielding a total score range of 0–27. The MCID for the PHQ-9 is 5 points [26]. The PHQ-9 served as a secondary outcome measure to assess depressive symptoms. Assessments were conducted at baseline (pre-treatment), immediately post-intervention, 4-weeks post-intervention, and 12-weeks post-intervention. Scores of 5, 10, 15, and 20 represent cut-off points for mild, moderate, moderately severe, and severe depression, respectively [27].

Procedures

Written informed consent was obtained from all participants or their legally authorized representatives. Patients were positioned supine with slight neck extension and rotation toward the non-blocked side. The operator first palpated the cricoid cartilage (approximately at the level of the sixth cervical vertebra), then moved laterally along the posterior border of the sternocleidomastoid muscle to locate the anterior tubercle of the sixth cervical transverse process (palpable as a bony prominence). Under real-time ultrasound guidance, the needle was advanced with continuous visualization of its entire trajectory. Upon reaching the peristellar ganglion region, 3 ml of 0.8% lidocaine solution (prepared by mixing 2 ml of 2% lidocaine with 3 ml of normal saline) was injected. Successful blockade was confirmed by the development of Horner’s syndrome on the blocked side, characterized by miosis, ptosis, facial anhidrosis, conjunctival injection, and nasal congestion. To ensure patient comfort and minimize infection risk, a unilateral alternating blockade protocol was implemented with one procedure performed daily at fixed intervals for ten consecutive days. Patients who were using hypnotic medications at the time of enrollment had been on a stable dose for at least 4 weeks prior to study entry and were instructed to maintain the same dosage throughout the intervention and follow-up period. Following each blockade, patients maintained supine positioning for ten minutes to monitor for and rule out immediate complications. All SGB procedures were performed by anesthesiologists with over twenty years of clinical experience. Following each blockade, patients were monitored for at least 30 min for any adverse events, including hoarseness, chest tightness, dysphagia, hematoma, or signs of local anesthetic toxicity. Ultrasound operators assessed the injection site daily for signs of hematoma, swelling, or tissue irritation before each subsequent procedure. Each patient independently completed scale assessments via a secure online electronic data capture platform using personal electronic devices before treatment initiation, immediately after the treatment course, and during follow-up at 4 and 12 weeks.

Statistical analysis

All statistical analyses were performed using GraphPad Prism software (Version 10.20). Continuous variables were first assessed for normality through the Shapiro-Wilk test. Normally distributed data are presented as mean ± standard deviation (Mean ± SD), while non-normally distributed data are expressed as median with interquartile range [M, IQR]. For longitudinal comparisons of continuous variables (PSQI, GAD-7, PHQ-9) across the four time points (T1 ~ T4), repeated measures analysis of variance (RM-ANOVA) was employed. Mauchly’s test of sphericity was used to verify the assumption of sphericity; if violated, the Greenhouse–Geisser correction was applied to adjust the degrees of freedom. Following a significant main effect of time, post-hoc pairwise comparisons were conducted using the Bonferroni correction to control for type I error. Categorical variables are reported as frequencies (n) and percentages (%). The significance level for all statistical tests was set at α = 0.05, with P < 0.05 considered statistically significant.

To explore whether concomitant hypnotic medication influenced treatment response, we performed exploratory subgroup analyses comparing medication users (n = 42) and non-users (n = 30). Due to the limited sample size, we did not further differentiate between specific hypnotic subtypes (e.g., zolpidem, eszopiclone, estazolam) in this analysis. For each outcome (PSQI, GAD-7, PHQ-9), two-way repeated measures ANOVA was conducted with time (T1, T2, T3, T4) as the within-subjects factor and medication use (yes/no) as the between-subjects factor. The time-by-medication interaction was used to assess whether medication use modified the pattern of improvement observed following SGB. When Mauchly’s test of sphericity was violated (P < 0.05), Greenhouse-Geisser correction was applied. Additionally, independent samples t-tests were used to compare between-group differences at each time point and for change scores (ΔT1-T2 and ΔT1-T4), with results presented as mean differences and 95% confidence intervals. Given the limited sample size, these analyses were underpowered and are considered hypothesis-generating.

Results

The demographic characteristics of the study population are presented in Table 1. Between March 2025 and September 2025, a total of 80 patients were screened. Among these, 3 participants withdrew from the study due to personal reasons, and 5 were lost to follow-up or dropped out after treatment, resulting in 72 patients ultimately included in the analysis. The mean age of the included patients was 48.78 years, with 32 males (44.44%) and 40 females (55.56%). The mean height and weight were 163.58 cm and 60.78 kg, respectively. The mean body mass index (BMI) was 22.62 ± 3.40 kg/m², with weight categories distributed as follows: underweight (10 patients, 13.89%), normal range (35 patients, 48.61%), overweight (22 patients, 30.56%), and obese (5 patients, 6.94%). Chronic insomnia disorder (ICSD-3 defined as symptom duration ≥ 3 months) was the predominant type, affecting 60 patients (83.33%). Occupational status analysis indicated that 34 participants (47.22%) were employed and 38 (52.78%) were unemployed. Additionally, 58.33% (42/72) of the patients had received hypnotic medications during the study period. The most commonly used hypnotic medications among the 42 users were zolpidem (47.62%, 20/42), eszopiclone (23.81%, 10/42), and estazolam (28.57%, 12/42). According to the PHQ-9 grading, the number (and proportion) of patients with mild, moderate, moderately severe, and severe symptoms were 29 (40.28%), 10 (13.89%), 9 (12.50%), and 6 (8.33%), respectively. According to the GAD-7 grading, the number (and proportion) of patients with mild, moderate, and severe symptoms were 30 (41.67%), 12 (16.67%), 6 (8.33%), respectively.

Table 1.

Baseline characteristics of patients

Characteristic	Total (n = 72)
Age (years)	48.78 ± 13.32
Sex
Male	32 (44.44)
Female	40 (55.56)
Height (cm)	163.58 ± 8.60
Weight (kg)	60.78 ± 11.37
BMI (kg/m2)	22.62 ± 3.40
Underweight	10 (13.89)
Normal weight	35 (48.61)
Overweight	22 (30.56)
Obese	5 (6.94)
Type of insomnia disorder
Acute insomnia disorder (< 3 months)	12 (16.67)
Chronic Insomnia disorder (≥ 3 months)	60 (83.33)
Employment
Employed	34 (47.22)
Unemployed	38 (52.78)
Use of sleeping Medication	42 (58.33)
Zolpidem	20 (47.62)
Eszopiclone	10 (23.81)
Estazolam	12 (28.57)
Patient health questionnaire-9 (PHQ-9)
None	18 (25.00)
Mild	29 (40.28)
Moderate	10 (13.89)
Moderately severe	9 (12.50)
Severe	6 (8.33)
Generalized anxiety disorder-7 (GAD-7)
None	24 (33.33)
Mild	30 (41.67)
Moderate	12 (16.67)
Severe	6 (8.33)

Data presented as n (%), mean ± SD. Abbreviations: BMI, body mass index

Exploratory subgroup analysis compared medication users (n = 42) and non-users (n = 30). No significant time-by-medication interactions were detected for any outcome (PSQI: P = 0.250; GAD-7: P = 0.415; PHQ-9: P = 0.489). However, given the limited sample size in each subgroup and the non-randomized nature of medication use, these analyses are underpowered to detect modest interaction effects. The absence of statistical significance should not be interpreted as evidence of no medication effect; rather, these findings suggest that within this pilot sample, the pattern of improvement was not markedly different between groups. Adequately powered studies are needed to definitively assess medication-by-treatment interactions (Table 2).

Table 2.

Subgroup analysis by hypnotic medication use

Outcome	Time point	Medication user (n = 42)	Non-user (n = 30)	Between-group difference (95% CI)	P for interaction
PSQI	T1	15.60 ± 2.58	15.00 ± 2.08	0.60 (-0.54, 1.73)	-
	T2	11.07 ± 3.51	9.90 ± 2.73	1.17 (-0.36, 2.70)	-
	T3	10.86 ± 3.72	10.57 ± 2.76	0.29 (-1.31, 1.89)	-
	T4	11.38 ± 3.66	11.10 ± 2.86	0.28 (-1.32, 1.88)	-
	△T1-T2	4.52 ± 2.54	5.10 ± 2.45	-0.58 (-1.77, 0.62)	0.250
	△T1-T4	4.21 ± 2.84	3.90 ± 2.37	0.31 (-0.95, 1.58)	0.250
GAD-7	T1	6.93 ± 4.26	8.23 ± 4.94	-1.30 (-3.47, 0.87)	-
	T2	5.07 ± 3.60	5.60 ± 3.38	-0.53 (-2.20, 1.15)	-
	T3	5.33 ± 3.98	5.97 ± 3.51	-0.63 (-2.44, 1.17)	-
	T4	6.05 ± 4.53	7.10 ± 4.12	-1.05 (-3.13, 1.03)	-
	△T1-T2	1.86 ± 2.01	2.63 ± 3.45	-0.78 (-2.06, 0.51)	0.415
	△T1-T4	0.88 ± 2.17	1.13 ± 3.21	-0.25 (-1.52, 1.01)	0.415
PHQ-9	T1	8.05 ± 4.79	10.50 ± 6.62	-2.45 (-5.30, 0.39)	-
	T2	6.07 ± 3.89	8.37 ± 5.51	-2.30 (-4.65, 0.06)	-
	T3	6.40 ± 4.11	8.17 ± 5.21	-1.76 (-4.05, 0.53)	-
	T4	6.60 ± 4.06	8.40 ± 5.04	-1.80 (-3.95, 0.34)	-
	△T1-T2	1.98 ± 2.44	2.13 ± 3.31	-0.16 (-1.51, 1.19)	0.489
	△T1-T4	1.45 ± 2.70	2.10 ± 4.05	-0.65 (-2.23, 0.94)	0.489

PSQI, Pittsburgh Sleep Quality Index; GAD-7, Generalized Anxiety Disorder-7; PHQ-9, Patient Health Questionnaire-9. T1, baseline; T2, immediately post-intervention; T3, 4-weeks post-intervention; T4, 12-weeks post-intervention. Data are presented as mean ± SD. The P for interaction was derived from the time-by-medication use interaction term in the two-way repeated measures ANOVA (Greenhouse-Geisser correction)

At the pretreatment baseline (T1), the mean total PSQI score for the 72 patients was 15.35 ± 2.39. Following SGB treatment, PSQI scores measured at T2, T3, and T4 timepoints were all significantly lower than baseline levels (Fig. 1A). Post-hoc pairwise comparisons with Bonferroni correction confirmed that PSQI scores were significantly lower at all follow-up time points compared to baseline (T1 vs. T2: mean difference = 4.764, 95% CI [4.040, 5.487], P < 0.001; T1 vs. T3: mean difference = 4.611, 95% CI [3.870, 5.352], P < 0.001; T1 vs. T4: mean difference = 4.083, 95% CI [3.320, 4.847], P < 0.001). The reduction in total PSQI score from baseline to each follow-up time point exceeded 3 points, indicating clinical improvement in sleep quality (Table S1). Analysis of PSQI component scores revealed the following baseline values: subjective sleep quality (2.53 ± 0.56), sleep latency (2.81 ± 0.57), sleep duration (2.47 ± 0.75), sleep efficiency (2.69 ± 0.66), and daytime dysfunction (1.82 ± 0.89). Analysis of PSQI component scores revealed significant improvements across all seven domains (Table S2). Post-treatment evaluation (Fig. 1B) showed significant improvement from baseline in subjective sleep quality, sleep duration, and daytime dysfunction scores, with reductions of 40.32%, 36.03%, and 41.21% respectively (Table S2). The marked improvement in daytime dysfunction suggests that SGB may reduce overall sympathetic hyperarousal, thereby alleviating daytime fatigue and cognitive impairment commonly associated with insomnia. This aligns with the conceptualization of insomnia as a 24-hour hyperarousal disorder and supports the potential of SGB to address both nocturnal and diurnal symptoms.

Fig. 1.

PSQI scores before and after high-frequency SGB. A Comparison of PSQI global scores at different time points. Higher PSQI scores indicate worse sleep quality. Data presented as mean ± SD. B Component analysis of PSQI dimensions. SGB, stellate ganglion block; PSQI, Pittsburgh Sleep Quality Index. T1, baseline (pre-treatment); T2, immediately post-intervention; T3, 4-weeks post-intervention; T4, 12-weeks post-intervention

Additionally, the mean GAD-7 score at baseline (T1) was 7.47 ± 4.57 for the 72 patients. Following SGB treatment, GAD-7 scores at T2, T3, and T4 were all significantly reduced compared to baseline (Fig. 2A). Post-hoc comparisons with Bonferroni correction showed that GAD-7 scores were significantly lower at all follow-up time points compared to baseline (T1 vs. T2: mean difference = 2.181, 95% CI [1.398, 2.963], P < 0.001; T1 vs. T3: mean difference = 1.875, 95% CI [1.136, 2.614], P < 0.001; T1 vs. T4: mean difference = 0.986, 95% CI [0.225, 1.748], P = 0.007). The reduction in GAD-7 scores from baseline to each follow-up time point was less than 4 points, indicating a statistically significant decrease, although the clinical significance, according to the MCID, was not achieved (Table S1). The mean PHQ-9 score at baseline measured 9.07 ± 5.72. Post-treatment PHQ-9 scores at T2, T3, and T4 likewise demonstrated significant reductions from baseline (Fig. 2B). Post-hoc comparisons with Bonferroni correction showed that PHQ-9 scores were significantly lower at all follow-up time points compared to baseline (T1 vs. T2: mean difference = 2.042, 95% CI [1.228, 2.855], P < 0.001; T1 vs. T3: mean difference = 1.931, 95% CI [1.116, 2.745], P < 0.001; T1 vs. T4: mean difference = 1.722, 95% CI [0.764, 2.680], P < 0.001). The reduction in PHQ-9 scores from baseline to each follow-up time point was less than 5 points, indicating only a statistically significant decrease (Table S1).

Fig. 2.

GAD-7 and PHQ-9 scores before and after high-frequency SGB. A Comparison of GAD-7 scores at different time points. B Comparison of PHQ-9 scores at different time points. SGB, stellate ganglion block; GAD-7, Generalized Anxiety Disorder-7; PHQ-9, Patient Health Questionnaire-9. T1, baseline (pre-treatment); T2, immediately post-intervention; T3, 4-weeks post-intervention; T4, 12-weeks post-intervention

Adverse events were systematically monitored during the 10-day treatment course and are summarized in Table 3. All adverse events were mild and transient, resolving spontaneously within 30–60 min without intervention. No procedure was discontinued due to adverse events. Daily ultrasound evaluation prior to each subsequent block revealed no evidence of hematoma formation, tissue irritation, or infection at injection sites throughout the study period. No serious adverse events were observed. Horner’s syndrome, observed in all patients (100%), was considered a sign of successful blockade rather than an adverse event, confirming accurate needle placement and effective sympathetic inhibition.

Table 3.

Adverse Events During the 10-Day Treatment Course

Adverse Event	n (%)	Onset	Duration	Management	Outcome
Hoarseness	4 (5.6%)	Immediate post-block	30–60 min	Observation	Spontaneous resolution
Chest tightness	2 (2.8%)	5–15 min post-block	20–40 min	Observation	Spontaneous resolution
Hematoma	0 (0%)	-	-	-	-
Injection site pain	0 (0%)	-		-	-
Dysphagia	0 (0%)	-	-	-	-
Local anesthetic toxicity	0 (0%)	-	-	-	-
Horner’s syndrome (expected)	72 (100%)	Immediate	4–8 h	-	-

Horner’s syndrome was observed in all patients as a sign of successful blockade and is not considered an adverse event

Discussion

As a safe conventional therapeutic approach, SGB has gained widespread clinical adoption [28]. Notably, the clinical indications for SGB have expanded significantly from initial applications in pain management to encompass diverse domains including immunomodulation, psychiatric disorders, and endocrine dysregulation [29, 30]. The principal finding of this prospective pilot study is that participants with insomnia disorder showed a clinically meaningful improvement in sleep quality, as measured by the PSQI, following a 10-day high-frequency SGB protocol. This improvement, which exceeded the established MCID, was sustained at 12-week follow-up. From a clinical perspective, high-frequency SGB may offer a viable short-term intensive option for patients who are intolerant to pharmacotherapy or have limited access to CBT-I resources. Notably, the improvement was similar between hypnotic users and non-users, suggesting that SGB can be effectively combined with ongoing medication without apparent interference. Furthermore, statistically significant reductions in symptoms of anxiety (GAD-7) and depression (PHQ-9) were observed post-intervention, though the improvements did not reach the MCID, suggesting limited clinical meaningfulness in this cohort with relatively mild baseline affective symptoms. Exploratory subgroup analyses showed no significant time-by-medication interactions for any outcome, suggesting that the improvements following SGB were not substantially influenced by concomitant hypnotic use in this sample. While this finding reduces concern about medication confounding, it is important to note that the analysis was exploratory and underpowered, and residual confounding cannot be ruled out. Regarding safety, all adverse events were mild and transient, with no serious adverse events or procedure discontinuations. The absence of cumulative local tissue injury over 10 consecutive daily injections—confirmed by daily ultrasound evaluation—suggests that this high-frequency protocol can be administered safely with appropriate ultrasound guidance and interval monitoring. The observed hoarseness (5.6%) and chest tightness (2.8%) rates are comparable to or lower than those reported in single-injection SGB studies [28], possibly reflecting the use of low-volume (3 mL) injections and real-time ultrasound guidance to avoid intravascular or paratracheal spread. However, given the limited sample size, rare but serious complications (e.g., vertebral artery dissection, seizure, pneumothorax) cannot be ruled out and warrant continued vigilance in future studies.

Insomnia disorder is conceptualized as a disorder of 24-hour hyperarousal, encompassing cognitive, cortical, and autonomic nervous system hyperactivity. This heightened state of arousal directly opposes the physiological quiescence required for sleep initiation and maintenance. The therapeutic mechanism of SGB in this context is hypothesized to involve inhibition of sympathetic nervous system outflow. By reversibly blocking the stellate ganglion, it may disrupt the cervical sympathetic chain, thereby potentially attenuating systemic sympathetic hyperactivity and facilitating a shift toward parasympathetic dominance [31]. This restoration of autonomic balance is considered critical for rectifying disruptions in the sleep-wake cycle [32]. It is also theorized that this process could be accompanied by a downstream reduction in hypothalamic-pituitary-adrenal (HPA) axis activity and stress hormone levels such as cortisol [33], which may contribute to ameliorating the hyperarousal-driven sleep dysfunction characteristic of insomnia disorder. Furthermore, melatonin, an indoleamine neuroendocrine hormone synthesized and secreted by the pineal gland, plays a pivotal role in circadian phase regulation [34]. Individuals diagnosed with insomnia disorder commonly exhibit disrupted melatonin secretion rhythms, and it has been suggested that SGB may be associated with restoration of physiological rhythmicity, potentially improving sleep-wake disturbances caused by abnormal melatonin fluctuations. Previous studies have reported associations between SGB and improved sleep outcomes: for example, a study in patients undergoing radical gastrointestinal cancer surgery demonstrated that SGB improved postoperative sleep disturbances by attenuating inflammatory responses, elevating melatonin levels, and stabilizing perioperative hemodynamics [35]. Similarly, Yang et al. revealed that SGB effectively enhanced sleep quality in post-mastectomy patients by modulating melatonin rhythm disruptions induced by elevated sympathetic tone [36]. Collectively, current findings suggest SGB demonstrates considerable promise in sleep disorder management, thus warranting expanded clinical investigation.

Insomnia and anxiety often form a self-perpetuating vicious cycle that exacerbates sleep disturbances and elevates risks for other mental health complications [37]. Within depression etiology, stress constitutes a pivotal etiological factor [38]. The physiological alterations induced by stress responses encompass activation of the autonomic nervous system and multiple neuroendocrine axes [39], among which the HPA axis is particularly significant [40]. Previous research indicates that suppressing HPA axis overactivity or sympathetic hyperactivity may confer therapeutic benefits for depression [41]. Furthermore, Shi et al. demonstrated that SGB increases cerebral blood flow and suppresses the HIF-1α/NLRP3 inflammatory signaling pathway, thereby improving anxiety and depressive states in rodent models [42]. Clinically, multiple studies substantiate SGB’s efficacy in alleviating anxiety symptoms among patients with post-traumatic stress disorder (PTSD) [29, 43]. Based on these mechanistic insights and empirical evidence, SGB emerges as a safe and mechanistically grounded potential intervention strategy for anxiety disorders.

Compared to previous studies, the present research explores the application of a high-frequency SGB protocol in patients with insomnia disorder and examines associated changes in sleep and emotional states. The rationale for this approach includes the potential for cumulative effects through sustained neuromodulation, which may have contributed to the observed improvements. The daily administration of SGB for 10 consecutive days may provide continuous sympathetic inhibition, preventing the sympathetic rebound that can occur with single or intermittent blocks. This sustained blockade could facilitate neural circuit reorganization within key sleep-wake regulatory regions, such as the amygdala and prefrontal cortex, potentially explaining the durable improvements observed at 12 weeks [44]. Animal studies have suggested that SGB may regulate brain-derived neurotrophic factor expression, offering a potential mechanistic hypothesis for the rapid alleviation of mood disorders observed in some clinical contexts [45]. Furthermore, the daily treatment protocol may reduce serum inflammatory cytokine levels and maintain HPA axis homeostasis, demonstrating particular advantages for autoimmune-related mood disturbances [33]. However, vigilance against potential risks is warranted, including progressive increases in complication rates and compensatory sympathetic hyperactivity. Clinical implementation should therefore strictly adhere individualized protocols, and maintain careful risk-benefit assessment throughout treatment. While our findings show a statistically significant reduction in anxiety symptoms, this interpretation should be considered within the context of anxiety’s heterogeneous nature. Anxiety manifests in diverse phenotypes, each potentially linked to distinct structural and functional brain alterations [46, 47]. It is plausible that SGB, by modulating common neural circuits underlying hyperarousal—such as amygdala-prefrontal pathways—may preferentially impact specific symptom dimensions. This theoretical framework offers a more nuanced perspective on the mechanism, suggesting that SGB’s therapeutic effects might be mediated through modulation of shared neurobiological pathways rather than uniform reduction across all anxiety subtypes. Future studies incorporating refined anxiety assessments and neuroimaging modalities could test this hypothesis empirically.

This study has several important limitations that should be considered when interpreting the findings. First and foremost, the single-arm, unblinded design without a sham-control group represents the primary limitation; the notable improvements observed may be influenced by placebo effects, natural symptom fluctuation, regression to the mean, or nonspecific therapeutic effects of repeated interventions, precluding causal efficacy attribution. Second, the absence of objective sleep measurements (polysomnography or actigraphy) prevents definitive exclusion of comorbid sleep disorders such as obstructive sleep apnea, periodic limb movement disorder, or parasomnias, which affect up to 30–40% of patients presenting with insomnia symptoms and may respond differently to autonomic modulation. Third, the lack of structured diagnostic psychiatric interviews (e.g., MINI, SCID-5) means that undiagnosed psychiatric comorbidities—particularly major depressive disorder, generalized anxiety disorder, or post-traumatic stress disorder—may have been present and could have influenced treatment response, limiting generalizability to strictly defined primary insomnia populations. Fourth, the sample size, while adequate for a pilot study, was insufficient for robust subgroup analyses; the exploratory comparisons of hypnotic users versus non-users were underpowered (as evidenced by wide confidence intervals crossing zero), and due to sample size constraints, we could not differentiate between specific hypnotic subtypes or conduct stratified analyses by insomnia subtypes, age groups, or sex, which may exhibit differential treatment responses. Fifth, the 12-week follow-up period is insufficient to assess long-term durability of effects beyond three months. Sixth, the study lacked mechanistic biomarkers (e.g., heart rate variability, salivary cortisol, melatonin, inflammatory cytokines) to elucidate the pathways underlying the observed clinical changes. Seventh, the single-center design and predominance of chronic insomnia (83.33%) may limit generalizability to other geographic regions, healthcare settings, or acute insomnia populations. Finally, the optimal treatment parameters—including frequency, duration, number of sessions, and need for maintenance treatments—remain undefined, and the risk-benefit profile of repeated high-frequency SGB courses over longer periods has not been established; rare but serious complications cannot be ruled out given the limited sample size. Despite these limitations, this pilot study provides critical feasibility data, effect size estimates, and variability parameters necessary for designing adequately powered randomized controlled trials, and the observed signal of potential benefit, particularly for sleep quality, warrants further investigation through rigorous sham-controlled designs incorporating objective outcome measures, comprehensive psychiatric screening, and mechanistic biomarkers.

Conclusion

In summary, this pilot study establishes the feasibility of a high-frequency SGB protocol in patients with insomnia disorder. Within-group improvements in sleep and affective outcomes were observed, but the uncontrolled design precludes causal attribution. These findings provide preliminary signal of potential benefit and critical data (effect sizes, variability estimates) to power future definitive trials. Until confirmed in sham-controlled studies, these results should not be interpreted as evidence of efficacy. Future research should prioritize large-scale, randomized, sham-controlled trials to establish causal efficacy and confirm the durability of treatment effects. These investigations should incorporate objective sleep measures such as polysomnography or actigraphy, rigorous screening for comorbid psychiatric and sleep disorders, and validated biomarkers (e.g., heart rate variability, salivary cortisol) to elucidate underlying mechanisms. Systematic evaluation of optimal treatment parameters—including frequency, duration, and long-term maintenance—will also be essential for developing evidence-based, personalized therapeutic protocols. Collectively, such research will help determine whether high-frequency SGB can be integrated into clinical practice as a safe and effective intervention for insomnia disorder and its commonly associated affective symptoms.

Abbreviations

SGB

Stellate ganglion block

PSQI

Pittsburgh sleep quality index

MCID

Minimally clinically important difference

GAD-7

Generalized anxiety disorder-7

PHQ-9

Patient health questionnaire-9

BZDs

Benzodiazepines

CBT-I

Cognitive behavioral therapy for insomnia

BMI

Body mass index

HPA

Hypothalamic-pituitary-adrenal

Authors’ contributions

Xinjian Zhang contributed to the study concept and design. Shuangrui Wang and Yongjia Wang enrolled patients. Shuangrui Wang, Xiaoyu Zhang, and Xia Li analyzed the data and drafted the manuscript. Xiaomin Fan, Shunqing Hu and Xiangyu Liu critically revised the manuscript for important intellectual content. All authors have read and agreed to the published version of the manuscript.

Funding

No funding was received to perform this study.

Data availability

The data used to support the findings of this study are available from the corresponding author upon request.

Declarations

Ethics approval and consent to participate

The study was conducted according to the guidelines of the Declaration of Helsinki, approved by the Ethics Committee of the Third Affiliated Hospital of Guangzhou University of Chinese Medicine (Approval No.: PJ-XS-20250227-003), and registered with the Chinese Clinical Trial Registry on March 19th, 2025 (Registration No.: ChiCTR2500099219). Informed consent was obtained from all subjects involved in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.