Immediate Effects of Dynamic Cupping on Median Nerve Mechanosensitivity in Healthy Participants: A Randomized Controlled Trial

Abstract

Objective: To assess the immediate effects of dynamic cupping on median nerve mechanosensitivity, measured by the upper limb neurodynamic test 1 (ULNT1), in healthy participants. Methods: After completing the questionnaire, 60 healthy participants were randomly assigned to two groups: the intervention group (IG; n = 30), which received dynamic cupping therapy, and the control group (CG; n = 30), which received no intervention. In the first assessment (M0), the ULNT1 was conducted on the dominant upper limb. The elbow extension range of motion was measured at the onset of symptoms and at the maximum tolerated point using a smartphone (iPhone 6, iOS 12.4.5, Apple Inc., Cupertino, CA, USA) as a goniometer substitute. Immediately following the intervention or control, both groups were assessed again (M1). Results: There were no significant differences between groups in terms of the range of motion for elbow extension at the onset of symptoms (IG: 23.8 ± 21.4° vs. CG: 24.8 ± 22.5°; p = 0.946) and at the maximum tolerated point of the ULNT1 (IG: 57.0 ± 19.9° vs. CG: 67.0 ± 19.4°; p = 0.236). Conclusions: These findings indicate that dynamic cupping does not appear to affect the mechanosensitivity of the median nerve in healthy participants. These results suggest that dynamic cupping may not be effective for immediate changes in nerve mechanosensitivity in asymptomatic individuals, but further research is needed to explore its effects in symptomatic populations, such as patients with carpal tunnel syndrome or cervical radiculopathy.

1. Introduction

Peripheral nerve plasticity and its response to physical stress play a crucial role in clinical care for individuals with nerve dysfunctions. Mechanosensitivity of neural tissues is a key protective mechanism, allowing nerves to adapt to stress, tension variations, or compression, which can lead to symptoms such as pain, paresthesia, or restricted movement [1,2]. Increased nerve sensitivity requires less force to provoke discomfort, resulting in an amplified response. Understanding and managing mechanosensitivity is essential in both therapeutic and rehabilitation contexts [2,3,4].

The mechanosensitivity of nerves can be assessed clinically by palpating neural structures and/or applying tension to the nervous system through neurodynamic tests, which demonstrate good reliability, precision, and diagnostic accuracy [2,5,6]. These tests are particularly useful for identifying nerve-related dysfunctions and guiding treatment strategies. The upper limb neurodynamic test 1 (ULNT1) is a provocative test designed to evaluate the mechanosensitivity of the brachial plexus and median nerve. This test involves a series of sequential movements that promote sliding and apply progressive tension to the brachial plexus and median nerve [1,4,7]. Key factors to consider during the assessment include the location of symptoms, resistance to movement, range of motion, and sensory and structural differentiation maneuvers [8].

Assessing the mechanosensitivity of the median nerve is particularly relevant in the context of cupping therapy because this nerve is often implicated in common musculoskeletal conditions such as carpal tunnel syndrome and cervical radiculopathy. These conditions are characterized by increased nerve mechanosensitivity, which can lead to pain, tingling, and functional limitations [2,3]. Understanding how cupping therapy affects nerve mechanosensitivity could provide valuable insights into its potential therapeutic applications for these and other nerve-related disorders.

Cupping therapy is an ancient technique with documented applications spanning thousands of years [9]. This non-invasive mechanical massage technique utilizes negative pressure to create a localized vacuum, promoting blood circulation and mobilizing underlying tissues [10,11]. Research suggests that cupping therapy can reduce fascial and muscular tension [12], increase pain thresholds [12,13], enhance local circulation [14], and modulate inflammation [15]. These effects suggest potential benefits for neuromuscular conditions, particularly those involving mechanosensitivity alterations.

Despite its widespread use, the impact of cupping therapy on peripheral nerve mechanosensitivity remains unclear. While previous studies have demonstrated its efficacy in increasing range of motion [16] and alleviating musculoskeletal pain [12,13], no research has specifically investigated its immediate effects on the median nerve’s mechanosensitivity. Given that musculoskeletal conditions often involve nerve-related symptoms, exploring this relationship is crucial. This represents a significant research gap, as musculoskeletal conditions often involve nerve-related symptoms, and understanding the relationship between cupping therapy and nerve mechanosensitivity could enhance its clinical application.

Recent studies have further emphasized the importance of neurodynamics in physiotherapy, particularly in the management of nerve-related pain and dysfunction. For example, Nee et al. (2012) demonstrated that neural tissue management can provide immediate clinically relevant benefits for patients with nerve-related neck and arm pain [8]. Additionally, Schmid et al. (2009) highlighted the reliability of neurodynamic tests in assessing upper limb nerve function [5]. These findings underscore the need for further research into interventions, such as cupping therapy, that may influence nerve mechanosensitivity.

This study aims to assess the immediate effects of dynamic cupping on the mechanosensitivity of the median nerve, as measured by ULNT1, in healthy participants. By addressing this research question, we seek to contribute to the growing body of evidence on cupping therapy’s role in neuromuscular health and its potential applications in clinical practice.

2. Materials and Methods

This report adhered to the CONSORT guidelines for reporting clinical trials.

2.1. Ethical Considerations

The protocol was approved by Fernando Pessoa University’s Ethics Committee (FISIO-1-28102019) and registered in ClinicalTrials.gov (NCT04688892). All participants provided written informed consent following the ethical principles outlined in the Declaration of Helsinki. Participants were informed about the study’s purpose, procedures, and their right to withdraw at any time. Confidentiality and anonymity were strictly maintained.

2.2. Participants

Sixty healthy volunteers, aged 18 to 35 years, were recruited for this study. Participants of both sexes were included if they demonstrated normal joint mobility in the upper body quadrant, as assessed through active movements of the spine, shoulders, elbows, and hands. Each movement was visually assessed by the examiner to ensure it fell within normal joint range of motion and was free from compensatory patterns. Exclusion criteria comprised the following: structural deformities in the upper body region; recent (within the past 6 months) complaints or injuries in this area; a history of surgical procedures or pathologies (musculoskeletal, cardiac, renal, metabolic, neurological, or oncological) affecting the upper quadrant that could impair nerve function; a lack of a mechanosensitive response to ULNT1 [17]; pregnancy; recent use of non-steroidal anti-inflammatory drugs or narcotics; and alcohol consumption within the preceding 12 h.

2.3. Randomization

Participants were randomly divided into two groups using the GraphPad online tool (https://www.graphpad.com/quickcalcs/, accessed on 10 December 2020). The intervention group (IG, n = 30) underwent dynamic cupping therapy applied along the median nerve pathway, while the control group (CG, n = 30) rested for 5 min without intervention. The randomization process ensured unbiased allocation by generating a random sequence. The inclusion of a control group was essential to account for potential confounding factors, such as natural variability in nerve mechanosensitivity over time or placebo effects.

2.4. Protocol

All participants completed a Characterization Sociodemographic and Clinical Questionnaire to identify possible exclusion criteria, which included personal data (2 items), training habits (1 item), and clinical background (9 items). Anthropometric measurements, including height and weight, were recorded to calculate body mass index (BMI). The study was carried out on the dominant upper limb. To identify the dominant upper limb, a test was carried out in which a ball (Kempa^® Leo) was thrown at the participant, who had to return it using the upper limb [18].

Measurements were collected at baseline (M0) and immediately after the intervention (M1). The ULNT1 was administered to the dominant upper limb of each participant. Participants were positioned supine on a massage table (Posturarte^® Olympic, Braga, Portugal), with their bodies aligned and the test forearm exposed for smartphone placement via an armband.

A smartphone (iPhone 6, iOS 12.4.5, Apple Inc.) was used as a goniometer substitute and was calibrated in the neutral position (0° elbow extension) before starting the test. The device was secured to the forearm with a strap to ensure stability during measurements, and the Compass application was used to measure elbow extension. The smartphone’s built-in motion sensors (tri-axial accelerometer, gyroscope, and magnetometer) allowed for the precise detection and quantification of angular movement. The lightweight design of the smartphone (129 g) minimized any potential influence on elbow range of motion. The use of smartphones as goniometers has been validated in previous studies, demonstrating high reliability and accuracy in measuring joint range of motion [19,20,21].

To ensure stability, the head and cervical spine were supported at the maximum comfortable lateral inclination using a foam pillow (LojaPro^®) [22]. Before the application of the test, the quality of the coupling (e.g., the possibility of sliding the armband during elbow extension or the comfort of the armband) was assessed, ensuring its consistency and feasibility for carrying out the test. All participants were submitted to a previous application of the ULNT1 on the contralateral side to familiarize them with the mechanosensitive changes (stretching, resistance, pain, or paresthesia) that can occur during ULNT1. Participants were instructed to report the onset of symptoms, defined as the first sensation of stretching, resistance, pain, or paresthesia during the test, and the maximum tolerated range of motion during the test. The ULNT1 was executed in the following sequence: (1) maximal contralateral cervical inclination; (2) shoulder abduction to 90° while preventing scapular elevation; (3) external rotation of the shoulder to 90° combined with elbow flexion at 90° (establishing the initial position or 0° range of motion); (4) forearm supination; (5) extension of the wrist and fingers; and (6) elbow extension [4,17,23]. As an evaluation parameter, the elbow extension range of motion at the onset of symptoms and the maximum tolerated range of motion were selected (Figure 1).

According to Cruz and Morais (2016) [19], the assessment of the elbow extension at the maximum tolerated point of the ULTN1 through a smartphone app showed great intra-rater agreement. The ULNT1 demonstrated moderate (kappa = 0.54) to substantial (kappa = 0.76) inter-rater concordance to collect information on mechanosensitivity in the studies by Schmid et al. (2009) [5] and Wainner et al. (2003) [24], respectively.

Regarding the instrument used to assess elbow extension, some studies have compared the measurement of range of motion with smartphones and universal goniometers, reporting a high correlation between them (r = 0.79–0.99) [16,20] and similar results [20].

Neurodynamic tests are considered to have moderate to substantial reliability [5] and ULNTs have acceptable clinimetric priorities [5].

As a data collection instrument, a smartphone was chosen because it has built-in motion sensors (such as a tri-axial accelerometer, electromechanical gyroscope, and magnetometer) allowing the detection and quantification of the linear and angular movement of the device in the 3 planes of space. The measurement of the elbow extension range of motion was performed using a smartphone (iPhone 6, iOS 12.4.5, Apple Inc.) with built-in sensors that allowed for measuring the position and orientation of the device in space. The Compass application, native to the iPhone 6 operating system, was used. This hardware–software set had already demonstrated good criterion validity and intra-examiner reliability in measuring movements in the 3 planes of space in the cervical [21].

2.5. Intervention

Participants in the IG received dynamic cupping therapy (dry-cupping with moving-cupping) applied along the entire pathway of the median nerve, including the arm, forearm, and palmar aspect of the wrist. A plastic suction cup (5.08 cm diameter) with a hand pump from K.S. Choi Corp^® (Los Angeles, CA, USA) was used. The participants remained in the supine position on a Posturarte^® Olympic massage table (Posturarte, Braga, Portugal) without inclination. A small amount of massage cream (ATL^®, Carnaxide, Portugal) was applied to facilitate smooth gliding. The cupping procedure involved slow, rhythmic sliding movements with two pumps of negative pressure applied for 5 min (Figure 2).

In the CG, the participants remained at rest in a supine position on a massage table (Posturarte^® Olympic) without inclination for 5 min. To avoid bias, the ULNT1 assessments before (M0) and after (M1) the intervention were always performed by the same researcher, who was blinded to the group allocation. The interventions, however, were carried out by a separate researcher who was trained and experienced in the technique. This separation of roles helped minimize potential bias and maintained the integrity of the study. Immediately after the intervention or control, the M1 assessment was performed in both groups.

2.6. Statistical Procedures

The data were analyzed using the Statistical Package for the Social Sciences (SPSS v. 26.0) software for Windows. The Kolmogorov–Smirnov test was used to assess the distribution of the collected data and the Levene test was used to assess the homogeneity of the variances in the studied groups. Although the variables did not follow a normal distribution according to the Kolmogorov–Smirnov test, the sample size, the homogeneity of the variances in the two groups under test, the absence of marked violation of the degree of asymmetry and kurtosis, and parametric tests were selected. The mean and standard deviations were used to describe the characteristics of the participants (age, body mass index, sex, and hand dominance) and measure the outcomes.

Student’s t-test for independent samples was used to compare groups, and Student’s t-test for paired samples was used to assess the differences between the assessment moments. To analyze the effect size, Cohen’s d was used. Values were considered small (d = 0.2), medium (d = 0.5), or large (d = 0.8) based on benchmarks suggested by Cohen (1988) [25]. A p-value equal to or lower than 0.05 was considered significant.

3. Results

Sixty participants participated in this study with a mean age of (22 ± 2.3) years and a mean body mass index (BMI) of (22.3 ± 3.4) kg/m². A total of 30 healthy participants of both sexes took part in both groups. The intervention group consisted of 12 males (40%) and 18 females (60%), while the control group had 15 males (50%) and 15 females (50%). Regarding hand dominance, the intervention group included 2 left-handed participants (6.7%) and 28 right-handed participants (93.3%), while the control group had 1 left-handed participant (3.3%) and 29 right-handed participants (96.7%). The descriptive characterization of the participants (age, BMI, sex, and hand dominance) is shown in

Table 1. Characteristics of the participants of both groups regarding age and body mass index.

	Control Group (CG)	Intervention Group (IG)	p
Variables	Mean ± Standard Deviation	Mean ± Standard Deviation
N	30	30
Age (years)	22.0 ± 2.9	21.5 ± 1.7	0.446
Body mass index (kg/m2)	22.2 ± 3.0	22.6 ± 3.8	0.338
Sex (male/female)	15/15	12/18	0.425
Dominant hand (right/left)	29/1	28/2	0.555

.

There were no significant differences between groups under analysis regarding age, BMI, sex, or hand dominance.

A flow diagram of this parallel randomized controlled trial is presented in Figure 3.

Comparing the two groups at M0, no significant differences were found regarding elbow extension at the onset of symptoms and the maximum tolerated point from the ULNT1.

No significant differences were found in the elbow extension range of motion at the beginning of the symptoms and maximum tolerated point of the ULNT1 in the intra-group comparisons (

Table 2. Intra-group differences in ULNT1 before (M0) and after intervention (M1).

								95% CI for Mean Difference
M0		M1	T	df	P	Mean Difference	SE Difference	Lower	Upper	Cohen’s d	SE Cohen’s d
M0 OOS	-	M1 OOS	0.714	59	0.478	2.225	3.118	−4.015	8.465	0.092	0.146
M0 MTP	-	M1 MTP	0.365	59	0.717	1.133	3.107	−5.083	7.350	0.047	0.153

Note. Student’s t-test for paired samples. Abbreviations: CI: confidence interval; OOS: onset of symptoms; MTP: maximum tolerated point.

) and inter-group comparisons (

Table 3. Intra-group differences in ULNT1 before (M0) and after intervention (M1).

						95% CI for Mean Difference
	T	df	P	Mean Difference	SE Difference	Lower	Upper	Cohen’s d	SE Cohen’s d
M0 OOS	−0.121	58	0.904	−0.667	5.505	−11.686	10.352	−0.031	0.258
M1 OOS	−0.068	58	0.946	−0.383	5.671	−11.735	10.968	−0.017	0.258
M0 MTP	0.018	58	0.985	0.100	5.448	−10.806	11.006	0.005	0.258
M1 MTP	−1.197	58	0.236	−6.067	5.067	−16.208	4.075	−0.309	0.261

Note. Student’s t-test for independent samples. Abbreviations: CI: confidence interval; OOS: onset of symptoms; MTP: maximum tolerated point.

).

The comparison of elbow extension range of motion before and after the intervention for each group is presented in Figure 4, assessing the beginning of symptoms and the maximum tolerated point of the ULNT1.

4. Discussion

The objective of this study was to assess the immediate effects of dynamic cupping on the mechanosensitivity of the median nerve in healthy participants, as measured by the ULNT1. The results showed no significant differences in elbow extension range of motion at the onset of symptoms or at the maximum tolerated point between the intervention and control groups. These findings suggest that dynamic cupping does not appear to affect the mechanosensitivity of the median nerve in healthy participants.

One of the factors responsible for the worsening of painful symptoms (hypersensitivity, hyperalgesia, and in more severe cases, allodynia) to chemical, thermal, and mechanical stimuli, due to greater disability and an aggravation of several musculoskeletal syndromes, is the increase in the mechanosensitivity of the peripheral nerves [8,26]. One of the most used strategies to trigger or increase responses in mechanosensitized nerves is to carry out a neurodynamic test where compression and/or tension is applied [17]. The ULNT1, as used in this study, is a well-established method for assessing mechanosensitivity, particularly in the median nerve and brachial plexus [1,4,7]. However, we acknowledge that the application of dynamic cupping along the median nerve pathway may have also affected surrounding tissues, including tendons, blood vessels, and other nerves (e.g., ulnar and radial nerves). While we focused on the median nerve, the non-specific nature of cupping therapy makes it difficult to isolate its effects solely on the nerve. Future studies should consider using more targeted interventions or additional evaluation tools, such as nerve conduction studies or quantitative sensory testing, to better understand the specific effects of cupping on nerve mechanosensitivity.

Despite the known benefits of cupping therapy on musculoskeletal conditions, our study did not find significant changes in nerve mechanosensitivity. Markowski et al. (2014) [27] reported an increase in the active range of motion of lumbar spine flexion and Straight Leg Raise after static cupping, suggesting that reduced muscle tension and improved circulation contributed to the effects observed. Similar findings were reported by Sadek (2016) [28] in lumbar spine mobility. A systematic review by Bridgett et al. (2018) [29] highlighted that the relationship between cupping therapy and range of motion remains unclear. However, it does suggest that an increase in range of motion may occur due to the muscle relaxation effects induced by cupping therapy. This increased range of motion observed in studies might be attributed to several factors that cupping therapy promotes, including enhanced cell nutrition, increased blood flow, and biomechanical changes in the skin that facilitate joint mobility [14,16]. Cupping therapy has also been shown to modulate inflammation and reduce pain thresholds in both healthy individuals and patients with musculoskeletal conditions [12,13]. For example, Emerich et al. (2014) [12] demonstrated that cupping therapy significantly increased pain thresholds in neck pain patients, suggesting a potential analgesic effect. Similarly, Rozenfeld and Kalichman (2016) [13] reported that dry cupping could reduce myofascial pain and improve tissue mobility, likely due to its effects on local blood flow and fascial release. These findings suggest that cupping therapy may have a role in managing conditions involving altered mechanosensitivity, although our study did not observe immediate effects on the median nerve in healthy participants. Cupping therapy may influence the peripheral nervous system through mechanisms such as improved blood flow, reduced muscle tension, and modulation of inflammation. These effects could potentially alter nerve mechanosensitivity. Future research should explore the effects of dynamic cupping in symptomatic populations, such as patients with carpal tunnel syndrome or cervical radiculopathy, where mechanosensitivity may be more pronounced due to underlying pathologies.

Michalsen et al. (2009) [30] applied cupping therapy with scarification and reported a reduction in pain and other symptoms associated with carpal tunnel syndrome in the short term. However, the long-term effects of cupping therapy on this condition remain unclear. It may be relevant for the investigation by Michalsen et al. (2009) [30] to assess whether mechanosensitivity is affected in this pathology, especially since cupping therapy appears to reduce symptoms. Additionally, Lowe (2017) [31] suggested that cupping therapy could influence neurophysiological processes, including nociceptor activity, which may explain its analgesic effects in some conditions.

These studies highlight the potential benefits of cupping for mobility and pain relief, but our study suggests that such effects may not extend to immediate mechanosensitivity changes. This discrepancy could be due to differences in study populations, as our participants were healthy individuals without underlying nerve pathologies. In symptomatic populations, such as those with carpal tunnel syndrome or cervical radiculopathy, cupping therapy may have a more pronounced effect on mechanosensitivity due to the presence of inflammation or nerve compression [8,30].

The large degree of variability in the onset of symptoms values may reflect individual differences in nerve mechanosensitivity and the subjective nature of symptom reporting. While the smartphone-based measurement method is reliable, variations in arm positioning and individual pain thresholds could also contribute to this variability. Future studies should consider using additional objective measures, such as quantitative sensory testing, to complement the ULNT1 and reduce reliance on subjective reporting.

Factors such as the therapist’s experience, the technique used, and the pressure applied during cupping therapy could influence the results. In this study, the therapist who performed the dynamic cupping intervention was trained and experienced in the technique, ensuring consistency in the application of the intervention. The same therapist performed all moving cupping procedures with two pumps of negative pressure, which helped standardize the technique and minimize variability in the intervention.

Limitations/Suggestions

This study has several limitations. First, the exact level of negative pressure applied during dynamic cupping was not measured. According to a systematic review by Kim et al. (2020) [32], the pressure levels used in cupping therapy vary widely, ranging from 60 mmHg to 750 mmHg, depending on the device, technique, and study design. However, without direct measurement of the pressure in this study, it is difficult to standardize the intervention or compare our results with those of other studies. Future research should include the objective measurement of negative pressure to ensure consistency and reproducibility.

Second, the study only assessed the immediate effects of the intervention. Longer intervention periods and follow-up assessments are needed to evaluate the sustained effects of dynamic cupping. Additionally, the sample consisted only of healthy participants with a normal range of motion. While asymptomatic individuals do exhibit a certain degree of mechanosensitivity, this sensitivity tends to increase with injuries affecting the somatosensory component of the peripheral nervous system [8]. Therefore, the lack of any underlying pathology may have influenced the results of the investigation.

Third, the use of the ULNT1 as the sole evaluation tool may not fully capture the effects of dynamic cupping on nerve mechanosensitivity. Although ULNT1 is a well-established method for assessing median nerve mechanosensitivity, incorporating additional evaluation tools, such as pain threshold testing, neurophysiological studies, or blood flow measurements, could provide a more comprehensive assessment of the intervention’s effects. Future studies should consider using a combination of evaluation methods to better understand the mechanisms underlying the effects of dynamic cupping.

Fourth, the study design included a control group to account for potential confounding factors, such as natural variability in nerve mechanosensitivity over time or placebo effects. However, the inclusion of a placebo group or sham cupping intervention could have further strengthened the study design by controlling for the non-specific effects of cupping therapy. Future randomized controlled trials should consider incorporating placebo or sham interventions to better isolate the specific effects of dynamic cupping.

Further research is needed to clarify the effectiveness of dynamic cupping in improving mechanosensitivity. Randomized controlled trials should include more groups, such as a placebo group and additional intervention techniques, whether combined with cupping or not. These studies should also involve longer session durations and intervention periods, such as 10 or 15 min over the course of a week, with different intensities (one or three pumps), as well as follow-up assessments. Additionally, it would be beneficial to enroll participants with symptoms related to nerve pain and disability to evaluate the effects of dynamic cupping on peripheral nerves that exhibit increased mechanosensitivity.

For example, future studies could explore the effects of dynamic cupping in patients with carpal tunnel syndrome, cervical radiculopathy, or other nerve-related conditions. Combining cupping therapy with other interventions, such as neural mobilization or acupuncture, may also provide synergistic effects on mechanosensitivity and pain relief [17,31].

Finally, the application of dynamic cupping along the median nerve pathway may have also affected surrounding tissues, including tendons, blood vessels, and other nerves (e.g., ulnar and radial nerves). While we focused on the median nerve, the non-specific nature of cupping therapy makes it difficult to isolate its effects solely on the nerve. Future studies should consider using more targeted interventions or additional evaluation tools, such as nerve conduction studies or quantitative sensory testing, to better understand the specific effects of cupping on nerve function and mechanosensitivity.

5. Conclusions

The findings from this study indicate that dynamic cupping appears to have no effect on the mechanosensitivity of the median nerve, as measured through the ULNT1, in healthy participants. These results suggest that dynamic cupping may not be effective for immediate changes in nerve mechanosensitivity in asymptomatic individuals, but further research is needed to explore its effects in symptomatic populations.