Effects of Gua Sha therapy on thoracolumbar fascia thickness and clinical outcomes of patients with chronic nonspecific low back pain: A randomized controlled trial

Abstract

Background:

Chronic nonspecific low back pain (CNSLBP) is associated with thoracolumbar fascia (TLF) dysfunction. However, the structural effects of Gua Sha, a Traditional Chinese Medicine technique, remain unclear. This study aimed to explore the acute and short-term effects of Gua Sha therapy on TLF thickness, pain intensity, and related physiological parameters in patients with CNSLBP.

Methods:

Thirty-two participants with CNSLBP were randomized to receive Gua Sha or hot pack therapy, a commonly used conservative treatment for low back pain, once a week for 4 weeks. The effects of the 2 treatments were compared. TLF thickness, pain, and related parameters were measured at baseline and immediately after the first and fourth interventions. A 2 (group) × 3 (time) repeated measures ANOVA was used for data analysis.

Results:

With increasing intervention, both groups showed significant improvements in pain intensity and dysfunction (P < .001), significant reductions in tissue hardness and pressure pain threshold (P < .05), and significant increases in skin temperature and lumbar flexibility (P < .001). However, only the Gua Sha group significantly reduced TLF thickness immediately after the first intervention (MD = 0.388, 95% CI: 0.101–0.675; P = .01) and immediately after the fourth session (MD = 0.607, 95% CI: 0.199–1.015, P = .005). The heart rate variability-related indicators did not reach statistical significance (P > .05), but their trends were favorable.

Conclusion:

Gua Sha can effectively relieve pain, improve function, and regulate tissue mechanical properties in CNSLBP patients and its effects may be achieved through multiple pathways. Although the single and 4-session interventions were not significantly better than heat in improving fascial thickness, it performs better in pain and flexibility clinical outcomes, supporting its potential value as a complementary therapy. Future studies with larger samples and longer periods are needed to clarify its mechanism of action and optimize treatment options.

1. Introduction

Chronic nonspecific low back pain (CNSLBP) is the leading cause of disability worldwide. It affects the daily lives of patients and places a heavy economic burden on society.^[1,2] Several cases of lower back pain improve within weeks, but the overall global burden, measured by prevalence and disability, continues to increase. In 2017, low back pain affected an estimated 577 million people globally,^[3] increasing to 619 million by 2020.^[4] Recent projections suggesting that the prevalence will reach 843 million cases by 2050,^[4] underscoring the urgent need for effective, accessible, and evidence-based interventions.

Despite its complex etiology and diverse causes, increasing evidence emphasizes the key role of the thoracolumbar fascia (TLF) in the pathophysiology of CNSLBP.^[5] Structural abnormalities of the TLF, including thickening,^[6] reduced shear strain,^[7] and morphological disorders,^[8] are closely associated with pain and functional limitations. Due to the complexity of NSLBP and the lack of identifiable organic lesions, clinical treatment increasingly emphasizes nondrug, noninvasive treatments, such as physical therapy^[9,10] and exercise therapy.^[11–13]

Among the various manual therapies, Gua Sha, a traditional East Asian intervention involving repeated unidirectional stroking of the skin using a smooth-edged instrument, has shown promise in managing musculoskeletal pain.^[14–16] Mechanistically, it has been proposed to enhance local microcirculation,^[17] modulate inflammatory mediators,^[18] and improve soft tissue extensibility.^[19] However, evidence on the effect of Gua Sha on TLF thickness in patients with CNSLBP is limited.

Previous studies have demonstrated that a single session of Gua Sha therapy can provide immediate analgesia for patients with low back pain.^[18,20] Like Gua Sha, myofascial release has therapeutic effects on low back pain, reducing muscle stiffness and TLF thickness.^[21] Tamartash et al^[22] and Arguisuelas et al^[23] reported that 4 sessions of myofascial release therapy improved pain and physical function in patients with chronic CNSLBP. Given the similarities between the pain relief effects of Gua Sha and myofascial release, the important role of TLF in CNSLBP, and the potential therapeutic value of Gua Sha, we hypothesized that Gua Sha may have similar improvement effects on TLF thickness and patient pain.

The present study aimed to investigate the acute and short-term effects of Gua Sha therapy on TLF thickness and related clinical outcomes in patients with CNSLBP. We hypothesized that a single session of Gua Sha would produce immediate structural and symptomatic improvements, and the effect of the intervention would be stronger after 4 consecutive weeks.

2. Materials and methods

2.1. Research design

This study was a 4-week, randomized, controlled clinical trial to explore the immediate and short-term effects of Gua Sha therapy on TLF thickness, pain severity, functional impairment, and related physiological measures in patients with CNSLBP. The study was conducted at the Nursing Clinic of Ganzhou Traditional Chinese Medicine Hospital from December 16 to January 20, 2024. Ethical approval was obtained from the Medical Ethics Committee of Ganzhou Traditional Chinese Medicine Hospital (approval number: GZSZYYKYLL20240083) and the Ethics Committee of Thailand (approval number: HE67217). This trial was prospectively registered in the Thai Clinical Trials Registry (TCTR20241124004). All procedures adhered to the principles of the Declaration of Helsinki and the Good Clinical Practice guidelines.

2.2. Randomization and blinding

Stratified randomization was used to ensure that the baseline characteristics of the Gua Sha and control groups were comparable. Patients meeting the inclusion criteria were divided into groups based on sex, age, body mass index, history of low back pain, and pain severity. Independent researchers randomly assigned each patient to a group at a 1:1 ratio using a randomization website (www.random.org).

This study employed a single-masked design. An independent researcher who was blinded to the group assignments of the patients measured all outcomes. Data analysts were also unaware of the group allocations until all the analyses were completed. Blinding the participants was not feasible due to the nature of the Gua Sha intervention; however, this design minimized potential participant bias and enhanced the objectivity of the research findings.

2.3. Sample size calculation

The sample size was recalculated based on a 2 (group) × 3 (time) repeated measures ANOVA using G*Power 3.1.9.7. F tests; ANOVA: Repeated measures, within-between interaction. Assuming a medium effect size (f = 0.25), alpha = 0.05, power = 0.8, number of groups = 2, number of measurements = 3, correlation among repeated measures = 0.6, and nonsphericity correction ε = 1, a total sample size of 24 participants (12 per group) was required. Allowing for a 25% attrition rate, 32 participants (16 per group) were recruited.

2.4. Participants

The study participants were recruited from a pool of volunteers within the Ganzhou Traditional Chinese Medicine Hospital community, including patients, employees, and residents. The recruitment methods included advertisements, WeChat groups, and posters. An orthopedic doctor assessed all registered participants. The assessment included medical history review and physical examination. All eligible participants provided informed consent before the initiation of the study.

To be eligible for inclusion, participants had to be 18 to 65 years old, have chronic lower back pain for at least 3 months with a visual analog scale (VAS) pain score of 4 or higher, and be free of severe medical conditions. Participants signed an informed consent form and agreed to participate in weekly sessions for 4 weeks.

Participants were excluded if they were pregnant, unable or unwilling to participate, had uncontrolled hypertension or other cardiovascular diseases, had specific types of low back pain, such as radicular pain or spinal stenosis, a history of lumbar surgery, or insufficiently severe low back pain. Participants were withdrawn from the study if they requested it, experienced severe adverse events, such as skin damage, large bruises, blisters, infection, or severe pain, or if their pain persisted despite standard pain management.

As shown in the CONSORT flowchart (Fig. 1), 62 volunteers were screened and 32 were enrolled. Five participants withdrew from the study, 2 from the Gua Sha group and 3 from the control group, leaving 27 participants who completed the trial.

Figure 1.

CONSORT flowchart.

2.5. Experimental protocol

This study adopted a randomized controlled trial design and selected hot packs as the active control group. The 2 reasons for choosing a hot pack are ethical and scientific. Ethically, a no-treatment group should not be established when effective treatments (such as hot packs) exist. Regarding scientific considerations, the hot pack can be used more accurately to evaluate the efficacy of Gua Sha and reduce the placebo effect. Hot packs have been proven to relieve pain and inflammation^[18,24] and are widely used in clinical practice.^[25] The efficacy may not be as good as Gua Sha, but its safety and analgesic effect make it a suitable control choice.

2.5.1. Gua Sha group

Participants in the Gua Sha group received treatment once a week for 4 consecutive weeks. The participants were placed in the prone position during the treatment, and the Gua Sha therapist applied olive oil to their waist and back as a lubricant. A smooth ox horn Gua Sha board was scraped along the waist and back fascia. The Gua Sha site was selected from the waist and back skin from the seventh thoracic vertebra to the coccyx, and the Gua Sha site length was approximately 25 to 30 cm. The specific order of Gua Sha was as follows: scrape along the bladder meridian on both sides of the spine from the seventh thoracic vertebra to the fifth lumbar vertebra, scrape from the first lumbar vertebra to the fifth lumbar vertebra at a position approximately 3 inches away from the midline of the waist and back spine, and scrape along the bladder meridian on both sides of the spine from the fifth lumbar vertebra to the coccyx.

The Gua Sha pressure was adjusted according to the level of comfort of the participants. Pressure was measured using the Algo Med System, and it ranged from light (487 ± 21 g) to moderate (626 ± 11 g).^[26] Each site was scraped for 2 to 3 minutes, and the total duration of treatment was 15 minutes. The residual oil was removed, and the participants were offered warm water to maintain their body temperature after the Gua Sha was completed.

2.5.2. Control group (hot pack)

Participants in the control group also received weekly 15-minute sessions for 4 weeks. The intervention involved placing a heated sea salt pack on the lower back.

The procedure was as follows: 800 g of sea salt were heated in a ceramic container using a microwave, medium-high setting for 3 minutes until the temperature reached approximately 45°C, as previous research has demonstrated the pain-relieving effects of heat at this temperature.^[27] Participants lay prone, and the therapist applied the heated pack to the painful area while massaging gently for 5 minutes to distribute heat evenly. This was followed by static application for 10 minutes. A towel was placed between the skin and pack if the participant reported excessive warmth. Participants were advised to drink warm water to promote systemic relaxation after treatment.

2.6. Outcome measurements

All outcome measurements were conducted at 3 time points: baseline (T0), immediately after the first intervention (T1), and immediately after the fourth session (T2). Assessments were performed in a quiet, temperature-controlled environment (24–26°C) by a trained physiotherapist blinded to group allocation. The primary outcome was the thickness of TLF. Secondary outcomes included pain intensity, functional disability, tissue hardness, skin temperature, pressure pain threshold (PPT), lumbar flexibility, and heart rate variability (HRV).

2.6.1. TLF thickness

The TLF thickness measurements were conducted using a medical ultrasound imaging system (Konted C10, China). B-mode imaging was performed using a 7.5 to 14 MHz linear probe and a 3.0-cm display depth. The reliability and validity of ultrasound imaging for measuring TLF thickness have been well established, with a high inter-rater reliability (intraclass correlation coefficient [ICC] > 0.80)^[28,29] and a strong correlation with magnetic resonance imaging.^[30]

In the sagittal plane, the measurement sites were bilaterally at the L2–L3 level, 2 cm lateral to the interspinous ligament (Fig. 2). This method has been previously used in similar studies.^[6,31] It has been demonstrated with moderate intra-observer (ICC: 0.67–0.77) and high interobserver (ICC: 0.82–0.92) reliabilities.^[28,32] To improve the accuracy, we measured the fascia on the left and right sides of the waist separately and saved the image data. For each image, we selected 4 equally spaced measurement points, used the built-in measurement tool of the ultrasound system, accurately evaluated fascial thickness at each measurement point, and calculated the average value. The final value was taken as the average value of fascia thickness on the left and right sides.

Figure 2.

Ultrasound measurement of TLF thickness in patients with NSLBP. The black arrow indicates the direction of ultrasound transducer placement. ES = erector spinae, L2 = lumbar 2, L3 = lumbar 3, NSLBP = non-specific low back pain, Sub = subcutaneous tissue, TLF = thoracolumbar fascia, US = ultrasound.

2.6.2. Tissue hardness

Tissue hardness of the erector spinae muscles of the lower back was measured using an Algometer Combo (OE-220, Japan). This method is highly reliable (ICC = 0.97).^[33] Trained researchers identified and marked the points on both sides of the lower back, where the pain threshold was relatively low. A 10-cm diameter plastic disk on the Algometer Combo was placed vertically on these points. Researchers have gradually applied vertical pressure through the long axis of the algometer. The device automatically recorded tissue hardness after a beeping sound signal. This process was repeated thrice for each side, and the average left and right mean values were calculated as the final result.

2.6.3. Skin temperature

The skin temperature of the lower back was measured bilaterally at a distance of 2 to 5 cm from the L3 spinous process using a noncontact infrared thermometer (Hikvision DS-2TP31B-3AUF). This method has been shown to have high reliability (ICC = 0.949–0.998).^[34] Three measurements were performed on each side, and the average left and right mean values were calculated as the final result.

2.6.4. PPT

The bladder meridian acupoints PPT beside the L3 vertebra was measured using an algometer (Algometer Combo, OE-220, Japan). Studies have shown that this method has high reliability (ICC = 0.77–0.94).^[35] The specific measurement steps were as follows: the participant held the stop switch while he/she was lying prone on a treatment bed, and the lower back was uncovered. The researcher gradually applied pressure with the algometer at the measurement point until the participant felt little discomfort and pushed the switch button. At this time, the pressure value (kg/cm²) was recorded, which was the PPT at this point. To improve the measurement accuracy of the lower back muscles, the PPT was measured 3 times on each side, and the average values of the left and right sides were calculated as the final result.

2.6.5. Lumbar flexibility

Lumbar flexibility was evaluated using the modified Schober test, a reliable clinical method (ICC = 0.83–0.94).^[36] The participants stood with their feet approximately 30 cm apart. A horizontal mark was made on the L5 spinous process, and 2 more marks were made 5 cm below and 10 cm above it. The participant bent forward as far as possible with knees straight, and the distance between the top and bottom marks was measured. The measurement was repeated 3 times at 30-second intervals, and the average was calculated. The final value was obtained by subtracting the average measurement from the initial 15 cm distance.

2.6.6. Pain intensity

Pain intensity was assessed using a 100-mm VAS, with one end of the scale marked as “no pain” and the other end marked as “the most severe pain.” The participants marked a point on a scale based on their current pain experience. The researchers measured the distance between the marked point and the “no pain” end and used this distance as the VAS score. To ensure the accuracy of the assessment, participants performed the VAS assessment in a quiet environment. They were provided detailed instructions on using the VAS, a well-established and reliable tool for measuring pain severity.^[37] VAS has demonstrated moderate to high reliability (ρ = 0.60–0.77) and validity (ρ = 0.76–0.84)^[38] in various clinical studies.

2.6.7. Functional disability

The Oswestry disability index (ODI) is a widely used self-report questionnaire used to assess functional disability associated with low back pain. The ODI is a 10-item self-assessment scale used to assess the degree of functional impairment within the past week in patients with low back pain. Each item was scored from 0 to 5, with higher scores indicating more severe functional limitations. All participants completed the ODI independently, in a quiet environment. The ODI has good internal consistency (ICC = 0.93) and test-retest reliability.^[39]

2.6.8. Autonomic nervous system

The uBio Macpa device (Firstbeat Technologies), validated for HRV measurement, was used according to the European Society of Cardiology and the North American Society of Pacing and Electrophysiology guidelines. Five-minute HRV data were acquired using a finger clip sensor (measurement range: 40–200 bpm, error ≤ 2%). The reliability of this method has been extensively demonstrated in the literature.^[40–42] Time-domain (RMSSD, SDNN), frequency domain (LF, HF, LF/HF ratio), and HRV indices, reflecting autonomic activity, were assessed. The participants rested in a quiet chair for 10 minutes. The sensor is placed on the index finger. The device automatically recorded and analyzed the aforementioned data after 2.5 minutes.

2.7. Data analysis

IBM SPSS Statistics 28.0 (IBM Corporation, Armonk) was used for statistical analysis. First, all parameters were subjected to Shapiro–Wilk normality and variance homogeneity tests. Mixed-model repeated measures ANOVA (2 × 3 design) was used to evaluate the time and interaction effects (time × group) of the Gua Sha and control groups at different time points (T0, T1, T2).

The Greenhouse-Geisser or Huynh-Feldt correction was applied when sphericity assumptions were violated. Post hoc comparisons were conducted using Bonferroni correction to control for Type I error. The effect size was reported as partial eta squared (η²_p), with values of 0.01, 0.06, and 0.14 considered small, medium, and large effects, respectively.^[43,44]

Given the exploratory nature of the study and the small sample size, to enhance statistical power and more accurately assess between-group differences at posttreatment time points, an additional exploratory ANCOVA analysis was performed. This model adjusted for baseline (T0) values and used post-intervention values (T1 and T2) as dependent variables. All statistical tests were 2-sided, with a significance set at α = 0.05.

3. Results

None of the participants who completed the study reported adverse reactions. The independent t-test revealed no statistically significant differences in the baseline characteristics, including age, height, weight, body mass index, blood pressure, heart rate, previous history of low back pain, and pain values between the Gua Sha (n = 14) and control (n = 13) groups (P > .05), indicating that the 2 groups were comparable (Table 1).

Table 1.

Baseline characteristics of participants.

Outcomes	Gua Sha group (mean ± SD)	Control group (mean ± SD)	P-value
Number (people)	14	13
Gender (female/male)	13/1	11/2	.515
Age (yr)	41.29 ± 7.57	42.78 ± 13.31	.722
Height (cm)	161.00 ± 7.08	160.54 ± 7.81	.873
Weight (kg)	60.29 ± 8.86	58.15 ± 9.01	.243
Body mass index	23.21 ± 2.82	22.48 ± 2.31	.217
Systolic blood pressure (mm Hg)	112.64 ± 13.55	118.69 ± 13.36	.254
Diastolic blood pressure (mm Hg)	76.29 ± 9.99	80.00 ± 14.24	.437
Heart rate (beats/min)	77.93 ± 10.46	76.54 ± 13.08	.762
History of low back pain (yr)	6.64 ± 2.73	6.62 ± 3.01	.980
Visual analog scale pain (scores)	5.36 ± 1.01	5.31 ± 0.95	.897

SD = standard deviation.

P < .05 denotes statistical significance.

As shown in Table 2, the time effects on TLF thickness, skin temperature, tissue hardness, PPT, lumbar flexibility, VAS score, and ODI score were significant (P < .05). These indicators changed significantly over time during the intervention. However, the time effects for the stress index and HRV-related indicators (LF, HF, LF/HF, SDNN, and RMSSD) were not significant (P > .05), indicating no significant changes throughout the intervention period. Regarding the time × group interaction effects, skin temperature, tissue hardness, lumbar flexibility, VAS score, and ODI score showed significant differences (P < .05), indicating divergent patterns of change in the Gua Sha and control groups over time. In contrast, the interaction effects for TLF thickness, PPT, stress index, and HRV-related indicators were not significant (P > .05), suggesting similar trajectories of change in both groups.

Table 2.

Mixed design ANOVA results of various indicators for the Gua Sha and control groups at baseline, after the first intervention, and after the fourth intervention.

Outcomes	Group	T0	T1	T2	Time effects			Time × group effects
		Mean ± SD	Mean ± SD	Mean ± SD	F	P-value	PES	F	P-value	PES
TLF thickness H (mm)	GG	4.09 ± 0.96	3.70 ± 0.87	3.48 ± 0.85	(1.74, 43.55) = 6.808	.004	0.214	(1.74, 43.55) = 0.287	.722	0.011
	CG	4.18 ± 1.00	3.93 ± 1.10	3.78 ± 0.75
Skin temperature (℃)	GG	35.98 ± 0.18	37.06 ± 0.21	36.99 ± 0.12	(2, 50) = 109.987	<.001	0.815	(2, 50) = 19.228	<.001	0.435
	CG	35.98 ± 0.27	38.22 ± 0.83	38.63 ± 0.84
Tissue hardness (N/mm²)	GG	60.26 ± 8.44	57.54 ± 5.03	53.19 ± 4.76	(2, 50) = 42.610	<.001	0.630	(2, 50) = 4.780	.013	0.160
	CG	63.03 ± 5.68	55.64 ± 5.27	49.07 ± 4.39
PPT (kg/cm²)	GG	6.50 ± 1.22	6.28 ± 1.42	5.11 ± 1.59	(2, 50) = 10.723	<.001	0.300	(2, 50) = 0.711	.496	0.028
	CG	5.67 ± 1.50	5.28 ± 1.44	4.68 ± 1.32
Lumbar flexibility (cm)	GG	5.16 ± 0.98	5.25 ± 0.94	6.31 ± 0.75	(2, 50) = 36.995	<.001	0.597	(2, 50) = 3.659	.033	0.128
	CG	5.00 ± 0.91	5.35 ± 0.81	5.78 ± 0.83
VAS (score)	GG	5.36 ± 1.01	4.36 ± 0.63	3.21 ± 0.89	(2, 50) = 97.354	<.001	0.796	(2, 50) = 3.599	.035	0.126
	CG	5.31 ± 0.95	4.54 ± 0.88	3.85 ± 0.90
ODI G (score)	GG	6.86 ± 1.61	5.71 ± 1.07	3.86 ± 1.03	(1.45, 36.31) = 90.715	<.001	0.784	(1.45, 36.31) = 90.715	<.001	0.342
	CG	6.08 ± 1.19	5.38 ± 0.87	4.69 ± 1.11
Stress index (score)	GG	39.57 ± 9.46	36.71 ± 10.82	35.29 ± 13.99	(2, 50) = 0.122	.886	0.005	(2, 50) = 0.962	.389	0.037
	CG	39.38 ± 8.98	40.92 ± 10.19	41.38 ± 12.13
LF (ms)	GG	7.19 ± 0.68	7.31 ± 0.84	7.14 ± 0.90	(2, 50) = 1.076	.349	0.041	(2, 50) = 0.176	.839	0.007
	CG	6.93 ± 0.73	7.22 ± 0.91	7.04 ± 0.95
HF (ms)	GG	6.10 ± 0.78	6.24 ± 0.82	6.27 ± 0.75	(2, 50) = 0.473	.626	0.02	(2, 50) = 0.247	.782	0.01
	CG	5.83 ± 0.81	6.04 ± 0.68	5.82 ± 0.71
LF/HF	GG	1.18 ± 0.10	1.18 ± 0.13	1.15 ± 0.12	(2, 50) = 0.132	.877	0.01	(2, 50) = 0.220	.803	0.01
	CG	1.21 ± 0.10	1.20 ± 0.14	1.20 ± 0.09
SDNN (mHz)	GG	40.31 ± 11.90	47.38 ± 10.23	41.25 ± 9.72	(2, 50) = 2.765	.073	0.10	(2, 50) = 0.022	.979	0.00
	CG	40.42 ± 16.07	47.48 ± 19.13	40.14 ± 13.96
RMSSD (mHz)	GG	30.82 ± 13.78	30.68 ± 3.19	37.72 ± 14.03	(2, 50) = 0.496	.612	0.02	(2, 50) = 1.399	.256	0.05
	CG	29.63 ± 17.03	32.58 ± 17.31	29.13 ± 11.72

CG = control group, G = Huynh-Feldt, GG = Gua Sha group, H = Greenhouse-Geisser, HF = high frequency, LF = low frequency, PES = partial eta squared, RMSSD = root mean square of successive differences, SD = standard deviation, SDNN = standard deviation regular heartbeat interval, SE = standard error, T0 = baseline data, T1 = data after the first intervention, T2 = data after the fourth intervention, VAS = visual analog scale.

P < .05, statistical significance.

To further explore between-group differences in post-intervention, an ANCOVA was conducted with baseline (T0) as a covariate. As shown in Table 3, at T1, there were no statistically significant differences between groups in TLF thickness, tissue hardness, or other clinical parameters, except for skin temperature (P < .05). At T2, significant between-group differences favoring were observed in tissue hardness (P = .011), lumbar flexibility (P = .033), pain intensity (VAS, P = .008), and functional disability (ODI, P < .001), with better results in the Gua Sha group. This result is consistent with the results of the time × group effect. However, differences in fascial thickness remained nonsignificant at both time points (T1: P = .466; T2: P = .297), indicating no superior effect of Gua Sha over hot pack therapy in modifying TLF thickness.

Table 3.

Between-group differences at post-intervention time points (T1 and T2) after adjusting for baseline values using ANCOVA.

Index	Group	Baseline (T0; mean ± SE)	After the first intervention (T1)			After the fourth intervention (T2)
			Post-intervention adjustment (mean ± SE)	Mean difference (95% CI)	P-vaule	Post-intervention adjustment (mean ± SE)	Mean difference (95% CI)	P-vaule
TLF thickness (mm)	GG	4.09 ± 0.26	3.74 ± 0.14	−0.15 (−0.56 to 0.26)	.466	3.51 ± 0.16	−0.25 (−0.73 to 0.23)	.297
	CG	4.18 ± 0.28	3.89 ± 0.14			3.76 ± 0.17
Skin temperature (℃)	GG	35.98 ± 0.05	37.06 ± 0.16	−1.16 (−1.63 to −0.69)	<.001	36.99 ± 0.16	−1.64 (−2.12 to −1.16)	<.001
	CG	35.98 ± 0.07	38.22 ± 0.16			38.63 ± 0.17
Tissue hardness (N/mm²)	GG	60.26 ± 2.26	58.14 ± 1.10	3.15 (−0.17 to 6.45)	.062	53.50 ± 1.18	4.75 (1.21–8.29)	.011
	CG	63.03 ± 1.57	54.99 ± 1.14			48.75 ± 1.22
Pressure pain threshold (kg/cm²)	GG	6.50 ± 0.32	6.02 ± 0.31	0.45 (−0.50 to 1.40)	.336	4.94 ± 0.38	0.07 (−1.07 to 1.22)	.895
	CG	5.67 ± 0.42	5.57 ± 0.32			4.86 ± 0.39
Lumbar flexibility (cm)	GG	5.16 ± 0.26	5.20 ± 0.17	−0.2 (−0.70 to 0.30)	.41	6.27 ± 0.13	0.43 (0.04–0.83)	.033
	CG	5.00 ± 0.25	5.40 ± 0.17			5.83 ± 0.14
VAS (score)	GG	5.36 ± 0.27	4.34 ± 0.14	−0.21 (−0.62 to 0.19)	.294	3.20 ± 0.16	−0.67 (−1.14 to −0.19)	.008
	CG	5.31 ± 0.26	4.55 ± 0.14			3.86 ± 0.17
ODI (score)	GG	6.86 ± 0.43	5.48 ± 0.12	−0.15 (−0.52 to 0.23)	.42	3.67 ± 0.22	−1.22 (−1.90 to −0.54)	.001
	CG	6.08 ± 0.33	5.63 ± 0.13			4.89 ± 0.23

P < .05, statistical significance, indicated in bold.

CI = confidence interval, ODI = Oswestry disability index, PPT = pressure pain threshold, T1 = after the first intervention, T2 = after the fourth intervention, TLF = thoracolumbar fascia, VAS = visual analog scale.

The within-group comparisons are presented in Table 4. TLF thickness significantly decreased after both the first and fourth interventions in the Gua Sha group (both P < .05), while skin temperature significantly increased in both groups (P < .001). Pain intensity (VAS) and functional disability (ODI) scores were significantly decreased in both groups after the first session (both P < .001). A significant reduction in tissue hardness was observed only in the control group after the first session (P < .05). No significant changes in the PPT or lumbar flexibility were found at T1 in either group. At T2, significant improvements in skin temperature, tissue hardness, PPT, lumbar flexibility, VAS, and ODI scores were observed in both groups after the intervention 4 times (all P < .05). TLF thickness decreased slightly in the control group, but a significant reduction was achieved only in the Gua Sha group (P < .05).

Table 4.

Within-group changes from baseline (T0) to post-intervention time points (T1 and T2) in the Gua Sha and control groups.

Outcomes	Group	T0 vs T1		T0 vs T2
		Mean difference (95% CI)	P-value	Mean difference (95% CI)	P-value
TLF thickness (mm)	GG	−0.388 (0.101–0.675)	0.010	0.607 (0.199–1.015)	.005
	CG	0.254 (−0.044 to 0.552)	0.092	0.401 (−0.023 to 0.824)	.062
Skin temperature (℃)	GG	−1.075 (−1.453 to −0.697)	< 0.001	−1.004 (−1.353 to −0.655)	<.001
	CG	2.231 (−2.623 to −1.839)	< 0.001	−2.642 (−3.005 to −2.28)	<.001
Tissue hardness (N/mm²)	GG	2.715 (−0.401 to 5.831)	0.085	7.062 (3.183–10.941)	.001
	CG	7.393 (4.16–10.626)	< 0.001	13.958 (9.933–17.984)	<.001
PPT (kg/cm²)	GG	−0.031 (−0.672 to 0.61)	0.92	1.141 (0.341–1.94)	.007
	CG	0.387 (−0.278 to 1.052)	0.242	0.992 (0.162–1.821)	.021
Lumbar flexibility (cm)	GG	−0.093 (−0.47 to 0.284)	0.616	−1.157 (−1.479 to −0.835)	<.001
	CG	−0.346 (−0.737 to 0.045)	0.08	−0.777 (−1.111 to −0.443)	<.001
VAS (score)	GG	1.000 (0.647–1.353)	< 0.001	2.143 (1.779–2.507)	<.001
	CG	0.769 (0.402–1.136)	< 0.001	1.462 (1.084–1.839)	<.001
ODI (score)	GG	1.143 (0.754–1.532)	< 0.001	3.000 (2.406–3.594)	<.001
	CG	0.692 (0.289–1.096)	0.002	1.385 (0.769–2.001)	<.001

CG = control group, CI = confidence interval, GG = Gua Sha group, ODI = Oswestry disability index, PPT = pressure pain threshold, T0 = measurement before the start of the experiment, T1 = data after the first intervention, T2 = data after the fourth intervention, TLF = thoracolumbar fascia, VAS = visual analog scale.

P < .05, statistical significance.

While the stress index and HRV-related indicators didn’t show statistically significant time or time × group interaction effects (P > .05), a closer look at the trend graphs (Fig. 3) reveals compelling patterns. The Gua Sha group exhibited a decreasing trend in their stress index, contrasting with an increasing trend observed in the control group. Furthermore, the Gua Sha group displayed a more pronounced downward trend in LF/HF, whereas the control group remained relatively stable. We also noted an increasing trend in HF and RMSSD for the Gua Sha group, while these indicators decreased in the control group. Interestingly, LF and SDNN showed similar trends across both groups. These preliminary findings suggest that Gua Sha intervention may contribute to symptom alleviation in patients with CNSLBP through its influence on the autonomic nervous system.

Figure 3.

Trend analysis of the regulatory effect of Gua Sha and control group on the autonomic nervous system. (A) Stress index; (B) low frequency; (C) high frequency; (D) LF/HF ratio; (E) SDNN; (F) RMSSD. RMSSD = root mean square of successive differences, SDNN = standard deviation regular heartbeat interval, T0 = measurement before the experiment started; T1 = data after the first intervention; T2 = data after the fourth intervention.

4. Discussion

This study employed ultrasound imaging to quantitatively assess the effects of Gua Sha on TLF thickness in patients with CNSLBP. While significant within-group reductions in TLF thickness were observed in the Gua Sha group, no statistically significant group-by-time interaction was found. This indicates that the effect was not statistically superior to the control group, but with further reductions observed as the number of treatment sessions increased. A similar but not significant effect was observed for the hot pack group. This finding not only supports the clinical efficacy of Gua Sha but also provides biomechanical evidence for its mechanism of action.

Relative to that of asymptomatic patients, the TLF of patients with CNSLBP is significantly thicker.^[5] It is often accompanied by pathological changes such as densification of the connective tissue around the muscles^[6,45] and reduction in shear strain capacity.^[7] This affects the elasticity of the tissue, known as sliding properties. This study found that Gua Sha therapy can effectively reduce TLF thickness after the first and fourth interventions, which is consistent with the reports of previous intervention studies. Previous studies have shown that the 4-session myofascial release technique^[22] and single-session myofascial manipulation^[21,46] can significantly reduce TLF thickness and improve the low back pain symptoms of patients. These studies indicate that myofascial restriction is an essential pathological factor for CNSLBP.

We hypothesized that Gua Sha therapy may reduce TLF thickness through multiple mechanisms. First, the Gua Sha action may indirectly cause viscoelastic mechanical stretching of the fascia. This stretching is analogous to the repeated stretching of a rubber band, which may cause microdeformation and reduce TLF thickness of the TLF. This is consistent with the fact that Gua Sha may improve posterior chain flexibility in patients with Parkinson’s disease.^[47] Secondly, Gua Sha may also modulate sympathetic nervous system activity, increase parasympathetic activity, reduce muscle tension, and improve tissue extensibility.^[48,49] Previous studies support these findings, but there were no significant differences in individual responses to Gua Sha treatment in this study. In addition, previous studies have confirmed that Gua Sha can effectively alleviate the symptoms of low back pain,^[16] and its mechanistic principles may be closely related to pain gating theory.

In contrast, Yang et al^[9] found that 6 weeks of percussion massage therapy did not significantly improve TLF thickness in firefighters with CNSLBP. This difference can be attributed to several factors. First, the study participants included physically demanding workers with a high training intensity, and their overall muscle-fascial health may have been better than that of the general population. Second, our study participants were older and less active and may have been more sensitive to the effects of Gua Sha. In addition, differences in treatment methods may have played a crucial role. Percussion massage involves primarily localized vibrations, while Gua Sha involves scraping and stretching of the skin and subcutaneous tissues, which can produce a more significant mechanical stretch on the fascia than vibration alone.

Skin temperature increased significantly in the Gua Sha and hot pack groups, suggesting that both interventions could promote local blood circulation. However, the increased skin temperature in the Gua Sha group may be related to the mechanical stimulation of scraping, which accelerates local blood circulation and vasodilatation of small blood vessels in the subcutaneous tissues. This finding is consistent with previous studies showing that Gua Sha reflexively induces vasodilatation by stimulating skin receptors, thereby increasing the local blood flow.^[50]

Both Gua Sha and hot packs reduced tissue hardness in patients with CNSLBP. However, the reduction in tissue hardness was significantly greater in the Gua Sha group than in the hot pack group after 4 weeks of intervention (P < .05). A reduction in tissue hardness was observed after a single Gua Sha intervention, but the difference was not statistically significant (P > .05). This may be attributed to the relatively short-lived and inadequate effect of a single Gua Sha session in inducing significant changes in the structural and mechanical properties of soft tissues, eliciting only a mild initial physiological response. After 4 consecutive weeks of Gua Sha intervention, the reduction in tissue hardness reached statistical significance (P < .05). This suggests that Gua Sha may exert a cumulative effect on improving tissue stiffness in patients with CNSLBP. Similarly, Chao et al^[31] reported a significant reduction in perceived tissue hardness after a single session of percussive massage.

With an increase in the number of interventions, the PPT of the participants in both the Gua Sha and control groups gradually decreased, suggesting increased pain sensitivity. However, the degree of PPT reduction in the Gua Sha group was more evident than that in the Control group. This phenomenon may be related to the local tissue inflammatory response and central nervous system sensitization induced by Gua Sha. Previous studies have shown that repeated mechanical stimulation can sensitize the peripheral and central nervous systems and lower the PPT.^[51] This suggests that Gua Sha can relieve pain in the short term, but its long-term application may increase the risk of potential harm from chronic pain. Therefore, pain responses should be closely monitored in clinical scenarios and treatment protocols should be adjusted according to individual differences to avoid overstimulation.

Both groups showed improvement in lumbar flexibility after the intervention. However, the improvement in the Gua Sha group was more significant after the 4 interventions, with an average improvement of 1.16 cm. This shows that Gua Sha can relieve pain, soften tissues, and effectively improve lumbar spine mobility. This finding is consistent with those of previous studies. For example, 1 study found that the Gua Sha can rapidly and significantly improve posterior chain muscle mobility and flexibility.^[47]

Additionally, a randomized controlled trial involving 30 patients showed that connective tissue massage improved chronic lumbar spine mobility.^[52] Another crossover study involving 24 college athletes confirmed that self-myofascial relaxation can improve myofascial sliding and increase lumbar spine flexibility.^[53] Gua Sha may enhance the flexibility of the lumbar spine by releasing the muscle fascia and improving local blood circulation.

Although no statistically significant time or interaction effects were found in HRV-related indicators, a downward trend in the stress index and LF/HF ratio, and upward trends in HF and RMSSD were observed in the Gua Sha group. These trends suggest a potential modulation of the autonomic nervous system; however, due to the lack of statistical significance, these findings should be interpreted as exploratory. Notably, these observations are consistent with previous studies showing that Gua Sha can improve the stress index in healthy individuals and athletes and enhance parasympathetic activity.^[48] Further studies with larger samples are warranted to confirm the autonomic regulatory effects of Gua Sha.

HRV changes are affected by a combination of complex factors. On the 1 hand, the physiological regulation ability of the individual plays a key role. Different patients have different functional bases of the autonomic nervous system. This inherent individual specificity makes their responses to Gua Sha stimulation different.^[54] On the psychological state also plays an important role. Psychological factors such as mood swings and daily stress levels may interfere with the stability of HRV and affect the evaluation of the effect of Gua Sha.^[55] At the same time, lifestyle factors such as sleep quality,^[56] nutrition,^[57] and exercise^[58] are also closely related to HRV. The diversity of these factors across different individuals further increases the complexity of HRV changes, making it challenging to present the significant effect of Gua Sha on HRV and stress index-related variables in this study.

Furthermore, for clinical outcomes, significant group × time interactions were observed for both VAS and ODI scores, indicating a differential effect between the groups over time. Specifically, after 4 interventions, the Gua Sha group demonstrated significantly greater reductions in pain and disability. The average decrease in VAS score in the Gua Sha group was 0.67 points greater than in the control group, and the average decrease in ODI score was 1.22 points greater than in the control group. These findings suggest that Gua Sha therapy offers significant advantages over hot pack therapy in relieving chronic pain and improving patient function. This finding is consistent with those of previous studies. Gua Sha can significantly relieve pain intensity (P < .05).^[20] A randomized study by Saha et al^[16] involving 50 patients with low back pain also showed that Gua Sha can effectively reduce the pain intensity of patients with low back pain. Additionally, a crossover trial^[18] found that Gua Sha had a longer-lasting anti-inflammatory effect than hot packs in 12 older patients with CNSLBP, relieving pain, improving mobility, and reducing disability. These findings are also consistent with the conclusion that 4 sessions of lumbar myofascial release can effectively relieve CNSLBP.^[59]

The analgesic and tissue-repair effects of Gua Sha may involve multiple mechanisms. On the 1 hand, the skin stimulation caused by Gua Sha can activate the endogenous analgesic system of the body, release endorphins and other substances, and produce analgesic effects similar to morphine.^[60] On the other hand, Gua Sha can improve local blood circulation,^[17,50] promote tissue repair, and affect pain transmission by regulating neurotransmitters such as serotonin and norepinephrine. In addition, studies have shown that Gua Sha may reduce the concentrations of pro-inflammatory cytokines such as TNF-α and IL-6 in the serum, inhibit the inflammatory response, and relieve pain.^[61]

However, this study had several limitations that should be considered when interpreting the findings. First, the short intervention duration (4 weeks) may not fully capture the long-term effects of Gua Sha therapy. Nevertheless, this study demonstrated both acute and short-term benefits, providing a foundation for future research with extended follow-up periods. Second, the established efficacy of hot packs as a control intervention may have minimized the observed differences between groups, but Gua Sha did not show unique structural improvements in TLF thickness. Third, individual patient responses to Gua Sha could vary owing to factors such as baseline pain severity or fascial health status, but the randomized design and stratified sampling helped mitigate this variability. Fourth, the potential limitations of measurement tools (e.g., ultrasound imaging variability) may have introduced noise, but the use of standardized protocols and blinded assessor’s enhanced reliability.

Despite these limitations, the study’s rigorous methodology, including randomization, blinded assessments, and objective outcome measures, supported the validity of the findings. The results highlight the potential of Gua Sha as complementary therapy for CNSLBP, warranting further investigation with larger samples and longer durations to confirm its clinical utility.

5. Conclusion

Gua Sha therapy provides both acute and short-term benefits for patients with CNSLBP. Significant within-group reductions in TLF thickness were observed, indicating potential structural modulation through mechanical stretching and tissue remodeling. Although the differences in TLF thickness were not statistically significant, Gua Sha resulted in greater improvements in clinical outcomes, including pain intensity, functional disability, tissue hardness, skin temperature, and lumbar flexibility.

These effects may be attributed to a combination of biomechanical, circulatory, and neuromodulatory mechanisms. Despite the short intervention period and the use of active control, the findings support Gua Sha as an effective complementary therapy for CNSLBP, offering both immediate relief and functional recovery. These results highlight the promise of Gua Sha and underscore the need for further validation in larger, long-term clinical trials.

Acknowledgments

The authors would like to thank the Thai Medical Trading Company for loaning the ultrasound imager used in this study, as well as the staff of the Nursing Clinic at Ganzhou Traditional Chinese Medicine Hospital and all volunteer participants.

Author contributions

Conceptualization: Beibei Wang, Wichai Eungpinichpong.

Data curation: Beibei Wang.

Formal analysis: Yang Hu.

Investigation: Beibei Wang, Dehui Lai.

Methodology: Yang Hu.

Project administration: Dehui Lai, Wichai Eungpinichpong.

Resources: Yang Hu.

Software: Dehui Lai.

Supervision: Yang Hu, Wichai Eungpinichpong.

Validation: Dehui Lai.

Writing – original draft: Beibei Wang.

Writing – review & editing: Wichai Eungpinichpong.