Congqi Shoushen Tuina Method for generalized anxiety disorder: Clinical outcomes and serum metabolomic profiles

Abstract

BACKGROUND

Generalized anxiety disorder (GAD) is a common emotional disorder in clinical practice, with a high incidence rate, and it can seriously affect people’s work and life. The Congqi Shoushen Tuina Method can be used to treat the head and abdomen simultaneously, and can also regulate the body’s energy and spirit; meanwhile, the clinical treatment of GAD is significant.

AIM

To assess clinical outcomes of the Congqi Shoushen Tuina Method in GAD and to examine related serum metabolomic profiles.

METHODS

This experiment enrolled 20 GAD patients as the GAD group, who received Congqi Shoushen Tuina Method (40 minutes per session, once daily, 5 times weekly, for 2 consecutive weeks). The primary outcome was the Hamilton Anxiety Scale; secondary outcomes included the Self-Rating Anxiety Scale and Pittsburgh Sleep Quality Index. Meanwhile, 20 healthy volunteers served as the control group to explore serum metabolomics profile alterations in the GAD group before and after intervention, compared with the control group.

RESULTS

After a 2-week intervention, the GAD group’s Hamilton Anxiety Scale scores dropped significantly, along with notable decreases in Self-Rating Anxiety Scale and Pittsburgh Sleep Quality Index scores. Serum untargeted metabolomics revealed distinct metabolites in GAD patients vs healthy controls, dominated by lipids and lipid-like molecules (19.86%), followed by organic heterocyclic compounds (15.87%) and others; lipid molecules (e.g., phosphatidylcholine, lysophosphatidylcholine, Lysophosphatidylethanolamine) showed the most prominent differences. Congqi Shoushen Tuina Method reversed levels of some metabolites, especially lipids. This suggests that therapy alleviates GAD symptoms by regulating metabolite expression, particularly lipids metabolism. Exploratory biomarker identification also pinpointed 5 key differential metabolites: Levulinic acid, Corypalline, sphingomyelin (d18: 1/20:0), lysine-valine, and cucurbitin.

CONCLUSION

Congqi Shoushen Tuina Method modulates serum metabolites in GAD patients, especially lipid metabolism, alleviating anxiety symptoms and improving sleep quality. Identified metabolites may support future prediction and diagnosis of GAD.

Key Words: Generalized anxiety disorder; Metabolic dysregulation; Non-targeted metabolomics; Tuina; Congqi Shoushen Tuina Method

Core Tip: The Congqi Shoushen Tuina Method, a distinguished traditional Chinese medicine-specific manipulation therapy, is characterized by its core principles of treating the head and abdomen in tandem and coordinating qi dynamics and mental state synergistically. Mechanistically, it exerts systemic regulatory functions by modulating patients’ serum metabolomic signatures, resulting in a marked amelioration of anxious emotions and associated complications, including sleep disturbances. Given its superior safety, good tolerability, and high patient compliance, this therapeutic approach holds substantial promise for broad clinical translation and application.

INTRODUCTION

Generalized anxiety disorder (GAD) is characterized by continuous and excessive worries and inattention, which is manifested as a feeling of threat, anxiety, irritability, sleep disorders, a feeling of tension, palpitations, dry mouth, sweating, etc.[1]. In recent years, due to the increasing pressure on life and work, more people are troubled by anxiety, and the incidence of GAD is getting higher, with an increasingly younger age of onset[2]. Its incidence is closely related to genetics, social psychology, etc.[3,4]. However, a growing body of evidence indicates that metabolic dysregulation is involved in the pathogenesis and progression of mental disorders[5,6]. When the body is exposed to external stimuli, it elicits a corresponding stress response, which dysregulates the nervous and immune systems, disrupts metabolic homeostasis, and consequently induces a spectrum of clinical symptoms[7]. Studies have demonstrated that dysregulation of glucose, lipid, and amino acid metabolism can trigger neuroinflammatory responses in the body, thereby elevating the risk of developing mental disorders such as anxiety and depression[6,8]. This condition not only severely impairs patients’ mental health but also exerts profoundly negative impacts on their daily life, work, and social interactions, ultimately compromising their overall quality of life. Currently, psychotherapy and pharmacotherapy are the first-line clinical interventions for anxiety disorders; yet, these approaches are associated with notable limitations, including the potential for dependence and addiction[9,10].

GAD belongs to the category of “depression syndrome” in traditional Chinese medicine (TCM), and its core pathogenesis is characterized by stagnation of qi movement, emotional imbalance, and failure of the spirit to remain inward. As a characteristic TCM therapy, Tuina exerts effects on corresponding acupoints via manipulations such as pressing and kneading, which can unblock meridians, harmonize qi and blood, regulate emotional state, improve patients’ overall condition, and alleviate clinical symptoms[11,12]. Tuina is comfortable and safe, has high patient acceptance and reliable efficacy, and can effectively relieve patients’ illnesses without toxic side effects[13,14].

The research team interprets the depression syndrome from the perspectives of qi and spirit and has developed the Congqi Shoushen Tuina Method. By adopting the strategy of simultaneous treatment of the head and abdomen, this therapy regulates qi movement throughout the body, harmonizes qi and blood, and restores the inward retention of the spirit, thereby achieving the effects of balancing Yin-Yang, smoothing qi movement, and calming the spirit to stabilize the will. Yin and Yang are the fundamental basis of life; their dynamic balance is the guarantee of physical health, and the foundation for the normal operation of qi and blood, as well as mental activities. Qi is the root of human life activities. As recorded in Synopsis of the Golden Chamber, “when the primordial qi of the five zang-organs flows unobstructed, the body will be in a state of peace and harmony”. The spirit is the master of life activities, encompassing consciousness, mental cognition, emotional state, and spirit itself, and regulates all physiological and psychological activities of the human body. Lingshu·Ben Zang states, “when will and intention are in harmony, the spirit will be focused and undistracted; the soul and hun will not disperse, regrets and anger will not arise, and the five zang-organs will be free from pathogenic factors”. Only when the spirit is clear and refreshed, mental activities are inwardly focused, and qi and blood flow unobstructed, can the body maintain normal emotional functions.

Previous clinical studies have found that the Congqi Shoushen Tuina Method has significant efficacy in treating GAD, regulating patients’ anxiety and significantly improving sleep quality, gastrointestinal symptoms, etc., with no toxic side effects[15]. However, it remains unknown whether the Congqi Shoushen Tuina Method can play a role in treating GAD by altering the body’s metabolic level. This study aims to further explore the impact of the Congqi Shoushen Tuina Method on serum metabolomics in patients with GAD and provide a basis for clinical promotion and application.

MATERIALS AND METHODS

Experimental design

The present study aimed to explore serum metabolic differences between patients with GAD and healthy individuals, and to identify the regulatory effects of tuina intervention on the metabolomics of GAD patients. This study recruited participants in the outpatient clinic of the Massage Department of the Third Affiliated Hospital of Henan University of TCM from February 2023 to December 2023, including 20 GAD patients and 20 healthy volunteers. Patients with GAD were assigned to the GAD group, while healthy volunteers were allocated to the control group. This study was designed and implemented in accordance with the Declaration of Helsinki of the World Medical Association and was approved by the Ethics Committee of the Third Affiliated Hospital of Henan University of TCM. All participants signed informed consent.

Inclusion criteria: (1) Comply with the diagnostic criteria for GAD in the United States Diagnostic and Statistical Manual of Mental Disorders; (2) 18 years old ≤ age ≤ 60 years old, no limit for men and women; (3) Hamilton Anxiety Scale (HAMA) score: 14 points ≤ HAMA score ≤ 29 points; (4) Self-Rating Anxiety Scale (SAS) score: 50 points ≤ SAS < 70 points; (5) No anti-anxiety drugs or other psychiatric drugs were taken 4 weeks before treatment, and no other related treatment was received; and (6) Voluntarily cooperate to complete the treatment and research, and sign an informed consent form.

Exclusion criteria: (1) Exclude secondary anxiety in physical diseases such as hyperthyroidism, hypertension, coronary heart disease, etc.; (2) Exclude stimulant overdose, withdrawal reactions of hypnotic sedation drugs or anti-anxiety drugs, obsessive-compulsive disorder, phobia, hypochondriasis, neurasthenia, mania, depression, or anxiety associated with schizophrenia; (3) Pregnant or breastfeeding women; (4) Patients with severe skin diseases or severe damage from the surgery; and (5) Patients who are intolerant to treatment.

Shedding criteria: (1) The participant who has incomplete data affects the therapeutic efficacy or safety assessment; (2) The participant who has adverse reaction events or other serious diseases or adverse events during treatment but cannot continue treatment; (3) The participant who has poor compliance and cannot complete the treatment smoothly; (4) The participant who has dropped out and lost follow-up during observation; and (5) The participant who has requested to withdraw or lost follow-up during treatment.

Intervention

The GAD group gave Congqi Shoushen Tuina Method treatment. The specific operations are as follows: (1) The patient is lying in a supine position and loosens the trouser belt to expose the Qihai and Guanyuan acupoints and lays the treatment towel flat on the treatment regions. The performer sits on the patient’s right side and rotates and point-presses clockwise using the index, middle, and ring fingers. The tip of the middle finger is closely adsorbed and fixed at the Qihai acupoint, and the other two fingers are placed naturally. The strength should be based on the patient’s comfort using force to penetrate deeper at about 100 times/minute; it takes about 5 minutes. The same method was used to apply pressure to the Guanyuan acupoint for 20 minutes. Point-press the Tianshu acupoint for 2 minutes; rub the abdomen for about 1 minute; rub the flanks from top to bottom 3 times; (2) Press and knead the Baihui acupoint, knead the Taiyang acupoint, and press the five meridians for about 4 minutes. Grasp the Jianjing acupoint, and point-press the Fengchi acupoint for about 3 minutes; and (3) The patient is guided to breathe in an abdominal manner during the treatment process, nasal inhalation, and mouth exhalation. The abdomen is slightly bulging when inhaling, and the perineum area and the entire abdomen are slightly relaxed when exhaling. Repeated and uninterrupted operations as shown above. Each treatment time is about 40 minutes, once a day, 5 times a week, for 2 consecutive weeks.

Observation indicators

Primary outcome: HAMA. There are 14 items in this scale, and all items use a 5-point scoring scale (0-4). Evaluation criteria: If the total score is ≥ 29 points, it may be severe anxiety; if the total score is ≥ 21 points, it will definitely be obvious anxiety; if the score is ≥ 14 points, it will definitely be anxious; if the score is less than 7 points, there will be no anxiety symptoms.

Secondary outcome: SAS. This scale contains 20 items that reflect subjective feelings of anxiety, and all items are scored on a 1-4 scale. Evaluation criteria: The cutoff value of the SAS standard deviation is 50 points, of which 50 to 59 points are mild anxiety, 60 to 69 points are moderate anxiety, and above 69 points are severe anxiety. Pittsburgh Sleep Quality Index (PSQI). This scale consists of 18 self-evaluation items, each scored on a 0-3 scale. Evaluation criteria: 0-5 points, excellent sleep quality; 6-10 points, good sleep quality; 11-15 points, common sleep quality; 16-21 points, poor sleep quality.

Serum test

Liquid chromatography-mass spectroscopy technology is used for testing, and the specific detection process is shown in Figure 1. Operation procedure: Sample collection, metabolites extraction, on-machine detection, data processing. Sample collection: Blood was collected in a centrifuge tube at 37 °C (or room temperature) for 1 hour for coagulation and stratification. Then centrifuge at room temperature for 10 minutes at 3000 rpm, and transfer the supernatant to a clean centrifuge tube. Centrifuge at 12000 rpm for 10 minutes at 4 °C, aliquot the supernatant into 1.5 mL centrifuge tubes, store 0.2 mL per tube, and freeze it at -80 °C. Prepare sufficient dry ice to send samples. Sample volume requirements: 200 μL/sample.

Figure 1  Metabolomics test flow chart.

Metabolites extraction: 100 μL of each sample was transferred into an Eppendorf tube, followed by the addition of 400 μL of extraction solution containing isotope-labeled internal standards (methanol/acetonitrile = 1:1, v/v). The mixture was vortexed thoroughly for 30 seconds and then ultrasonicated for 10 minutes under an ice-water bath condition. After incubation at -40 °C for 1 hour, the samples were centrifuged at 12000 rpm and 4 °C for 15 minutes, and the resulting supernatants were collected for subsequent analysis. Equal volumes of supernatants from all samples were pooled to prepare quality control (QC) samples, which were then subjected to instrumental analysis.

On-machine detection: Target compounds were subjected to chromatographic separation using a Vanquish ultra-performance liquid chromatography system (Thermo Fisher Scientific, CA, United States) equipped with a Waters ACQUITY ultra-performance liquid chromatography BEH Amide column (2.1 mm × 50 mm, 1.7 μm). The mobile phase consisted of phase A (aqueous phase) containing 25 mmol/L ammonium acetate and 25 mmol/L ammonia solution, and phase B (acetonitrile). The sample tray temperature was maintained at 4 °C, and the injection volume was set at 2 μL. Data acquisition of both full-scan (MS¹) and tandem mass spectrometry (MS²) spectra was performed on an Orbitrap Exploris 120 mass spectrometer, which was controlled by Xcalibur software (Version 4.4, Thermo Fisher Scientific, CA, United States). Data processing: Raw data were converted to mzXML format using ProteoWizard software, followed by metabolite identification and visualization using the BiotreeDB database (version 3.0).

Adverse events

This study uses recording adverse events as safety observations, including the patient’s name, time of occurrence, and adverse events and their severity. The test’s safety was evaluated after completion.

QC and data processing

QC: QC of the experiment was implemented by using QC samples to evaluate methodological stability. In the two-dimensional principal component analysis score plot (Figure 2), the QC samples exhibited excellent clustering, indicating good methodological stability and high-quality data obtained from the present experiment.

Figure 2 Metabolic spectrum characterization detection process. QC: Quality control; PC: Principal component.

Raw data processing: To minimize detection errors, a series of preprocessing steps was performed on the raw data as follows: (1) Outlier filtering. Individual peaks were filtered to eliminate background noise. Outlier filtering was conducted based on the relative standard deviation (also referred to as the coefficient of variation); (2) Missing value filtering. Individual peaks were subjected to filtering. Only peak area data with no more than 50% missing values in a single group or across all groups were retained; (3) Missing value imputation. Missing values in the raw data were subjected to numerical simulation (missing value recoding). Specifically, the missing values were imputed with half the minimum value of the dataset; and (4) Data normalization. Normalization was performed using an internal standard.

Statistical analysis

All data were entered into Excel tables and analyzed statistically using SPSS 25.0. All measurement data were subjected to the Shapiro-Wilk test prior to analysis. Data that followed a normal distribution were statistically described as mean ± SD, and differences were compared using a two-sample t-test; if the data did not follow a normal distribution, then the median (P25, P75) was used for statistical description. The difference between groups was assessed using Mann-Whitney U test. The χ² test was used for categorical data, including the gender composition and total effective efficiency of patients in this trial. After the original data obtained from non-targeted metabolomics detection were converted into mzXML format using ProteoWizard, metabolites were identified using the co-written R package. The database used was Bio treeDB (V3.0), and visual analysis was performed by using the self-written R package. The differential metabolite analysis process is shown in Figure 3.

Figure 3 Principal component analysis score plot. Green dots are designated as quality control samples, while blue dots represent analytical samples. The clustering degree of quality control samples is positively correlated with methodological stability and reflects the quality of the experimental data.

RESULTS

Baseline characteristics of participants

A total of 20 GAD patients and 20 healthy volunteers were included. There was no significant difference in age. The age range of the GAD group was 22-55 years old, with an average of (35.00 ± 11.02) years old, 4 males and 16 females. The course of the disease was 6-90 months, with an average of (31.55 ± 20.62) months. The control group ranged in age from 21 to 60 years, with an average of 33.65 ± 12.35 years, and included 9 males and 11 females. Comprehensive details regarding the baseline characteristics of the study population are illustrated in Table 1.

Table 1 Baseline characteristics, mean ± SD.

	GAD group	Control group	P value	t/F
Age (years)	35.00 ± 11.02	33.65 ± 12.35	0.717	-0.365
Gender (male/female)	4/16	9/11	0.088	1.000
Disease course (months)	31.55 ± 20.62
HAMA	23.95 ± 4.52
SAS	55.69 ± 4.22
PSQI	12.65 ± 3.588

HAMA: Hamilton Anxiety Rating Scale; SAS: Self-Rating Anxiety Scale; PSQI: Pittsburgh Sleep Quality Index; GAD: Generalized anxiety disorder.

Analysis of outcome indicators

Primary outcome indicators: HAMA Scale scores were compared before and after treatment, and the patient’s scores were significantly reduced after treatment (P < 0.01). The results are presented in Table 2.

Table 2 Comparison of primary and secondary outcome indicators, mean ± SD.

		Before treatment	After treatment	P value	t
Primary	HAMA	23.95 ± 4.52	9.75 ± 5.94	0.000	13.33
Secondary	SAS	55.69 ± 4.22	36.88 ± 6.70	0.000	12.15
	PSQI	12.65 ± 3.59	6.05 ± 2.80	0.000	10.08

HAMA: Hamilton Anxiety Rating Scale; SAS: Self-Rating Anxiety Scale; PSQI: Pittsburgh Sleep Quality Index.

Secondary outcome indicators: From the comparison results of the patient’s SAS and PSQI before and after treatment, it was found that the patient’s scores on each scale were significantly lower after treatment (P < 0.01). The results are presented in Table 2.

Analysis of non-targeted metabolomics test results

Analysis approach of orthogonal partial least squares-discriminant: Orthogonal partial least squares-discriminant (OPLS-DA) was performed for two comparison cohorts. The control group vs the pre-treatment GAD group, and the pre-treatment vs post-treatment GAD group. The scatter plot of the OPLS-DA model is presented in Figure 4A and B, from which it can be observed that the sample clusters of different groups are completely separated, with distinct inter-group discrimination, indicating significant differences in serum metabolic profiles between the compared cohorts. A total of 200 permutation tests were conducted to prevent model overfitting and verify the reliability and rationality of the constructed models. For the OPLS-DA model comparing the control group with the pre-treatment GAD group, the permutation test dot plot yielded R²Y and Q² values of 0.81 and -0.62, respectively (Figure 4C). For the model comparing the pre-treatment and post-treatment GAD groups, the corresponding R²Y and Q² values were 0.94 and -0.15 (Figure 4D). Furthermore, all sample points in the permutation plots were located below the original points on the fitting line. These results demonstrated that the established data analysis models were reasonably applicable without overfitting issues, and thus could be used for subsequent screening of differential metabolites.

Figure 4 Orthogonal partial least squares-discriminant analysis results. A and B: The orthogonal partial least squares-discriminant model scatter plot. In Figure 4A, orange circles represent the samples from the generalized anxiety disorder (GAD) group, while purple diamonds represent the samples from the control group. In Figure 4B, orange circles denote the samples from the pre-intervention GAD group, and blue diamonds denote the samples from the post-intervention GAD group. The samples in each group are roughly distributed within the same interval, and the sample sets between the two groups are completely separated, indicating that the serum metabolic spectrograms of the two groups are significantly different; C and D: The orthogonal partial least squares-discriminant model permutation result dot plot. GAD: Generalized anxiety disorder; ZC: Control group; GYH: Congqi Shoushen Tuina Method group.

Analysis of serum metabolic spectrum: The serum metabolomic profile was comprehensively analyzed before and after intervention in GAD patients and compared with that of the control group. Following data normalization, all samples were subjected to subsequent analysis, and clustered heatmaps and volcano plots were generated and visualized as shown in Figure 4. Comparing the GAD group and control group before treatment (Figure 5A and B), it was obvious that there were significant differences in the expression of serum metabolites in the two groups. Comparisons of results before and after intervention revealed significant differences in metabolite expression levels (Figure 5C and D), indicating that the Congqi Shoushen Tuina Method may exert therapeutic effects on GAD patients by regulating their metabolite expression profiles.

Figure 5 Heatmaps and volcano plots. A and B: Heatmaps, where the horizontal coordinates represent different groupings of samples, the vertical coordinates represent all metabolites, and the color blocks at different positions represent the relative expression of metabolites at the corresponding positions; red indicates high expression of the substance content, and blue indicates low expression of the substance content. Figure A shows the comparison between the control group and the generalized anxiety disorder group before intervention, and Figure B shows the comparison between the generalized anxiety disorder group before intervention and after intervention; C and D: Volcano plots. Each point in the plots represents a detected metabolite: The horizontal axis denotes the fold change of each metabolite between groups, while the vertical axis denotes the P value derived from Student’s t-test. The color of each scatter point indicates the expression change trend of metabolites: Red represents upregulated metabolites, blue represents downregulated ones, and gray represents metabolites with no significant differences in expression. VIP: Variable importance in the projection.

Analysis of differential metabolite concentration changes: Differential metabolites were screened using the criteria of variable importance in the projection > 1 and P value < 0.05. Compared with the control group, the GAD group showed 174 differential metabolites whose expression levels changed after intervention. Variation trends in the relative content of different metabolites across groups are shown in Figure 6A. A total of 9 different clusters were identified, including 18 metabolites, 34 metabolites, 13 metabolites, 15 metabolites, 32 metabolites, 15 metabolites, 17 metabolites, 21 metabolites, and 9 metabolites, respectively. Clusters 3, 5, 7, 8, and 9 exhibited reversal of dysregulation after treatment, encompassing a total of 92 metabolites, including 30 upregulated metabolites after treatment, 62 downregulated metabolites, including 26 lipids and lipid molecules (accounting for 28.2%), 13 organic heterocyclic compounds (accounting for 14.13%), 10 organic acids and their derivatives (accounting for 10.87%), 7 fatty acids (accounting for 7.6%), and 6 amino acids and polypeptides (accounting for 6.52%) (Figure 6B).

Figure 6 K-means plot and differential metabolites proportion. A: K-means plot, the horizontal coordinate in the figure represents the sample group name, the vertical coordinate indicates the relative content of metabolites after normalization, and the cluster represents the same change trend as a cluster; B: The classification and proportion of differential metabolites.

Exploratory identification of candidate Biomarkers for GAD following Congqi Shoushen Tuina Method intervention: The authors used the above metabolites for subject-operating-characteristics curve analysis. The results are presented in Figure 7. The area under the Levulinic acid curve is 0.785 (0.639-0.931); the Corypalline is 0.7675 (0.611-0.924); sphingomyelin (SM, d18:1/20:0) is 0.7525 (0.589-0.916), lysine-valine (Lys-Val) is 0.73 (0.569-0.891); cucurbitin is 0.72 (0.557-0.883). We propose that these metabolites hold promise as biomarkers for diagnosing GAD and evaluating the therapeutic efficacy of the Congqi Shoushen Tuina Method intervention. However, their clinical applicability requires further validation in larger-scale independent cohorts.

Figure 7 Receiver operating characteristic curve. The results of different differential metabolites: Levulinic acid, Corypalline, sphingomyeli (d18:1/20:0), lysine-valine, and cucurbitin. SM: Sphingomyelin; Lys-Val: Lysine-valine.

Metabolic pathway analysis: Through a comprehensive analysis of the pathways in which differentially expressed metabolites are located, it was found that the pathways affected by this therapy are mainly alanine, aspartate and glutamate metabolism, propanoate metabolism, glycine, serine and threonine metabolism, and glyoxylate and dicarboxylate metabolism (Figure 8).

Figure 8 Bubble diagram for metabolic pathway analysis and rectangular treemap for metabolic pathway analysis. A: Bubble diagram for metabolic pathway analysis. Each bubble in the bubble chart represents a metabolic pathway. The horizontal coordinates of the bubble and bubble size represent the influence factors of the pathway in topological analysis, the larger the bubble size, the larger the influence factor; the vertical coordinate of the bubble and bubble color represent the P value of the enrichment analysis [take the negative natural logarithm, i.e. -ln(P)]. The darker the bubble color, the smaller the P value and the more significant the enrichment degree; B: Rectangular treemap for metabolic pathway analysis. Each block in the figure represents a metabolic pathway. The size of the block indicates the influence factor of the pathway in topological analysis, the larger the block size, the larger the influence factor; the color of the block represents the P value of the enrichment analysis [take the negative natural logarithm, i.e. -ln(P)], the darker the block color, the smaller the P value, and the more significant the degree of enrichment.

Adverse events

During the treatment period, no obvious abnormalities in vital signs were observed in either group, and no symptoms or other adverse reactions occurred. It shows that this therapy is safe and reliable.

DISCUSSION

GAD is a common clinical mental disorder triggered by stress responses to external stimuli. These responses induce aberrant activity in the body’s nervous system, elicit alterations in the endocrine system, and ultimately lead to psychological and behavioral changes[16,17]. Susceptibility to GAD is influenced by genetic, environmental, and other factors, and its incidence has been steadily rising in recent years. Some studies have indicated that females have a higher prevalence of GAD due to elevated neuroticism levels[18]; however, most relevant trials fail to stratify their analyses by gender[19,20]. As another potential contributing factor, gender’s specific role in influencing the onset and severity of GAD has not been fully elucidated in existing research. Future studies should design more rigorous experiments to clarify the impact of gender and other covariates on GAD pathogenesis. Prolonged anxiety can disrupt cellular functions, impair normal neuronal development, interfere with intercellular signal transmission and neurotransmitter activity, and induce neuroendocrine and metabolic dysregulation, thereby giving rise to a wide array of clinical symptoms. Clinically, we have observed that apart from emotional disturbances, the most distressing symptoms in GAD patients are insomnia and gastrointestinal discomfort, which severely impair their daily work and quality of life[15].

TCM has unique advantages in treating GAD. Moreover, a growing body of research has demonstrated that Tuina can alleviate anxiety symptoms and effectively regulate patients’ mental health status[21,22]. The research team believes that GAD is mostly caused by stagnation of qi activity, abnormal circulation of qi and blood, and excessive outward dispersion of spirit and energy. The disease is related to abnormal “qi” and “spirit”. The theory of “Congqi Shoushen” was proposed, and tuina therapy was innovated, using fingers instead of needles and selecting Guanyuan acupoint, Qihai acupoint, Baihui acupoint, and Fengchi acupoint as the main acupoints, and coordinating with pressing the five meridians, grasping the Jianjing acupoint, kneading the Taiyang acupoint, etc.[23]. Furthermore, clinical observations have indicated that this therapy yields favorable clinical efficacy in the treatment of GAD. It exerts corresponding therapeutic effects by regulating the serum levels of 5-hydroxytryptamine and substance P, with no obvious adverse reactions observed, thus confirming its safety and reliability[24]. In this study, the Congqi Shoushen Tuina Method was used to treat GAD, and the HAMA score of patients was significantly reduced, as were the SAS score and PSQI index. These findings further verify the efficacy of Congqi Shoushen Tuina Method in the intervention of GAD.

Modern research has found that metabolic disorders may lead to the occurrence and development of mental illnesses[5,25]. Metabolomics can reveal altered metabolic pathways by detecting changes in metabolites and provide a basis for a deeper understanding of disease physiology and pathology[26]. This study uses metabolomics to examine the effect of the Congqi Shoushen Tuina Method on the expression of serum metabolites in GAD patients, aiming to reveal the mechanism underlying its therapeutic effect. It was found that there were a variety of differential metabolites in the body of GAD patients after metabolomic testing, among which mainly were mainly lipids and lipid-like molecules (19.86%), organic heterocyclic compounds (15.87%), organic acids and their derivatives (11.61%), benzene (9.31%), and fatty acids (6.12%). After intervention with the Congqi Shoushen Tuina Method, the differentially expressed metabolites showed partial reversal; among these, lipid molecules showed the greatest restoration toward baseline levels.

We explored the biomarkers of GAD intervened by Congqi Shoushen Tuina Method, and found that five metabolites, including Levulinic acid, Corypalline, SM (d18:1/20:0), Lys-Val, and cucurbitin, were significantly different between GAD and healthy human serum. Furthermore, the levels of five metabolites in the serum of GAD patients exhibited significant alterations before and after treatment. Levulinic acid and SM (d18:1/20:0) belong to fatty acids, lipids, and lipid-like molecules. Changes in biological metabolites, including lipids, apolipoproteins, etc., are closely related to the development of common mental diseases such as anxiety, depression, etc.[27,28]. Lipid molecules may activate innate immune cells, promote the release of proinflammatory factors and bioamine catabolism, and lead to microglia-induced hypothalamic inflammation; they can also increase blood-brain barrier permeability, leading to chronic inflammation in the brain and ultimately increasing the risk of mental illness ultimately[29]. Fatty acids can alter the fluidity of the nerve membrane and the binding of neurotransmitters to receptors, thereby obstructing neurotransmitter synthesis and release, regulating neurogenesis and myelination, and thereby affecting neural function[30]. Corypalline belongs to alkaloids and is a tyrosine-based alkaloid. It can regulate the physiological pathological process of the body by affecting the synthesis of neurotransmitters such as catecholamines, etc., and thus participate in the onset of anxiety[31,32]. Lys-Val belongs to amino acids and peptides, and amino acids are key precursors for the synthesis of neurotransmitters in the central nervous system (CNS). Its abnormal levels can lead to neurotransmitter imbalances and induce or exacerbate anxiety[33,34]. Cucurbitin belongs to the organic acids and derivatives. Organic acids can participate in the body’s energy metabolism, maintain neuronal function, and regulate neuronal transmission[35], while organic acids produced by intestinal flora can affect CNS functions through the brain and intestinal axis[36]. Further studies are warranted to validate the scientific validity of these findings, which are expected to provide objective biomarkers for the prediction and diagnosis of GAD, and lay a solid foundation for its clinical application.

The differentially expressed metabolites affected by this therapy are mainly enriched in four metabolic pathways: Alanine, aspartate, and glutamate metabolism; propionate metabolism; glycine, serine, and threonine metabolism; and glyoxylate and dicarboxylate metabolism. These four pathways mediate neurotransmitter balance, gut-brain crosstalk, and energy metabolism, thereby forming a synergistic regulatory network. Neurotransmitter metabolic disorder is one of the most direct pathogenic mechanisms of GAD. Glutamate serves as a major excitatory neurotransmitter[37], while glycine and γ-aminobutyric acid are inhibitory neurotransmitters[38]. Contemporary research has demonstrated that tuina acts on the body surface, stimulates cutaneous tactile and pressure receptors, activates the body’s neuro-immune-endocrine pathways to alleviate neuroinflammation[39,40], and regulates the levels of neurotransmitters such as glutamate and γ-aminobutyric acid, thus restoring the excitation-inhibition balance of the CNS[41]. D-serine acts as an agonist of ionotropic glutamate receptors, and participates in the regulation of learning, memory, and synaptic plasticity. Its metabolic disorder can induce anxiety-related cognitive impairment[42]. Tuina manipulations applied to relevant acupoints can alleviate oxidative stress damage induced by chronic stress, improve hippocampal neuronal synaptic plasticity, and consequently ameliorate patients’ emotional and sleep quality symptoms[43]. The gut-brain interaction effect represents a core peripheral regulatory mechanism underlying GAD pathogenesis. Its dysfunction can induce or exacerbate anxiety-like behaviors through signal transmission between the gut and the CNS[44,45]. Propionate, a member of short-chain fatty acids, is a metabolic product generated during the functional activity of gut microbiota[46]. It can affect neuroinflammation via the gut-brain axis and regulate CNS excitability[47]. The propionate metabolic pathway may thus serve as a key target for Tuina-mediated regulation of the gut-brain axis. Existing studies have shown that abdominal Tuina manipulations can enhance gastrointestinal motility, improve the intestinal microenvironment, and regulate the composition of gut microbiota, thereby repairing the intestinal barrier. Furthermore, these manipulations transmit signals to the CNS via the gut-brain axis, triggering complex transformations within the nervous system and the release of gut-brain peptides, thereby reducing neuroinflammatory responses and improving related symptoms[24]. Meanwhile, by regulating abdominal adipose tissue and promoting adiponectin secretion, Tuina enhances its regulatory effects on the CNS[48]. Glyoxylate and dicarboxylate metabolism participate in the tricarboxylic acid cycle. Dicarboxylates, as intermediates of the tricarboxylic acid cycle, affect neuronal energy supply[49]. Relevant studies have also shown that tuina can promote local energy metabolism, modulate systemic metabolic levels, and exert corresponding therapeutic effects[50,51]. The multi-target effects of tuina in GAD intervention are consistent with the conclusions of previous studies that tuina can comprehensively regulate the body’s neuroendocrine system[52-54]. However, the exact underlying mechanisms still require further in-depth investigation.

This study has several limitations. First of all, it relied solely on healthy controls and did not establish a sham tuina control group, which hinders rigorous differentiation between tuina’s specific therapeutic effects and its non-specific effects. Future studies may adopt a double-control design: (1) A sham tuina control group, in which GAD patients receive superficially similar gentle kneading or touching on non-acupoint areas to eliminate the placebo effect; and (2) A healthy control group to identify disease-specific metabolic characteristics. Secondly, the relatively small sample size compromises the generalizability of the results. Future research should increase sample size to further verify the reliability of this study’s findings and provide more accurate data for clinical practice. Finally, this study included GAD patients of all age groups without performing sex-stratified analysis. Meanwhile, only two treatment courses were observed and no follow-up was conducted. Subsequent studies should refine the experimental design, conduct stratified analyses across subgroups, and explore the long-term effects of tuina on GAD through follow-up investigations to obtain more representative data.

CONCLUSION

The Congqi Shoushen Tuina Method used in this study has demonstrated good efficacy and can offer new ideas and therapies for the clinical treatment of GAD. The use of metabolomics technology to reveal the impact of this therapy on patients’ metabolism, may provide objective biomarkers for the clinical diagnosis of GAD. Meanwhile, it provides a basis for further exploring the impact of tuina on the metabolomics of GAD patients.