Understanding tightened muscle in knee osteoarthritis and the impacts of Fu’s subcutaneous needling: A pilot trial with shear-wave elastography and near-infrared spectroscopy

Abstract

Background:

Given the scarce reports on the interplay between Fu’s subcutaneous needling (FSN), tightened muscle, and therapeutic effects, we developed a clinical research protocol to synchronously collect data on clinical efficacy and muscle characteristics in patients with knee osteoarthritis, exploring the mechanism of FSN action. The primary aim was to assess the feasibility and safety of this protocol, guiding future trials and their sample size calculations.

Methods:

In this prospective, single-blind, self-controlled study, 19 patients with early to mid-stage unilateral knee osteoarthritis underwent FSN therapy on both knees over 1 week (4 sessions, every other day). We measured local elastic modulus, muscle thickness, blood flow volume, and oxygen consumption rate of bilateral vastus lateralis muscles using shear-wave elastography and near-infrared spectroscopy (NIRS) before and after the first and fourth treatments. Additionally, real-time NIRS indicators (oxygenated hemoglobin [O₂Hb], deoxyhemoglobin [HHb], total hemoglobin [THb], and tissue saturation index [TSI]) were recorded during these treatments. Pain intensity (visual analogue scale [VAS]), functional status (Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC]), and active range of motion were evaluated before these treatments.

Results:

All 19 participants completed the trial without serious adverse events. After 3 FSN treatments, significant changes were observed in VAS and WOMAC scores (VAS: P < .001; WOMAC: P < .001), and knee flexion (P < .001) and external rotation (P = .02), except for internal rotation. No meaningful significant differences were observed in muscle characteristics at baseline or between pre- and post-treatment periods. NIRS results during treatments indicated significant increases in local O₂Hb and THb post-FSN therapy (First treatment: O₂Hb: P = .005; THb: P = .006. Fourth treatment: O₂Hb: P = .002; THb: P = .004); however, no significant increases were observed for HHb (First treatment: P = .06; Fourth treatment: P = .28). No linear correlation was found between therapeutic effects and changes in tightened muscle indices.

Conclusion:

FSN reduces pain and improves joint function in knee osteoarthritis, while also enhancing blood flow and oxygenation in the vastus lateralis muscle of the affected side. Further revisions of this protocol are warranted based on our insights.

1. Introduction

Fu’s subcutaneous needling (FSN), a unique acupuncture method, is widely used in China for chronic musculoskeletal disorders.^[1–3] It specifically targets tightened muscles with myofascial trigger points (MTrPs) in a pathological tension state.^[4] FSN clinical technique involves swaying movement of FSN needle over the subcutaneous connective tissue layer above these muscles, along with reperfusion approaches, which are tailored to muscle function and require the targeted muscle to briefly exert force against resistance and then quickly relax (Fig. 1). It can be seen that the operation of the FSN therapy markedly differs from traditional acupuncture. What, then, is the effective mechanism underlying its unique operation? This question highlights a critical knowledge gap in the field.

Figure 1.

Examples of swaying movement and reperfusion approach in the vastus lateralis muscle. (A) Horizontal reciprocating fan-shaped swaying movement is performed in the subcutaneous loose connective tissue with the needle. (B) While the vastus lateralis muscle is in a relaxed state, the physician continuously performs the swaying movement. (C) The patient cooperates with the physician to perform knee extension resistance movements, maintaining for 10 seconds before relaxing, during which the swaying movement continues. This approach of contracting and then relaxing the target muscle is known as the reperfusion approach.

Clinically, FSN commonly leads to muscle softening, patient-reported relaxation, and signs of enhanced blood circulation like local redness and decreased chills. Given the prevalence of these observations, a correlation with clinical efficacy seems highly plausible. Through literature review, we have identified several studies that may shed light on the mechanisms of FSN therapy: Subcutaneous swaying movement of FSN needle notably stretch the loose connective tissue, a key element in both direct (e.g., collagen connections between epimysia) and indirect (e.g., neurovascular bundles) muscle connections, as well as in compartmental boundaries like the fascia.^[5] Mechanical traction on this tissue can remodel fibroblast cytoskeletons, reducing tissue tension.^[6] Given the distribution of loose connective tissue, changes in its tension not only affect the tissue itself but also the tension state of skeletal muscles and the surrounding environment of blood vessels within its network. Fu et al^[7] found that both distal and proximal subcutaneous swaying movement of FSN needle significantly reduced the endplate noise associated with skeletal muscle tension and pain in the taut band of biceps femoris MTrP region in New Zealand rabbits. He et al^[8] found that subcutaneous swaying movement of FSN needle at the lateral margin of the clavicle significantly enlarges the diameter of the adjacent vertebral artery and increases its blood flow. In summary, the above studies indicate that subcutaneous swaying movement of FSN needle can relax taut skeletal muscles and reduce their compression on surrounding blood flow by intervening at MTrPs. This superficial to deep effect is likely mediated by changes in the tension of subcutaneous loose connective tissue. Regarding reperfusion approaches, both human and animal studies have long established that a brief contraction followed by rapid relaxation of skeletal muscle increases local blood flow.^[9] Therefore, we considered whether the mechanism of FSN action could involve reducing muscle hardness (transitioning from tense to relaxed) and enhancing blood circulation. Based on these considerations, we formulated the research hypothesis: “FSN action potentially operates by reducing the hardness of tightened muscles and improving their blood circulation” and attempted to initiate research to explore it.

In line with our hypothesis, Chiu et al^[10] validated FSN effectiveness in reducing skeletal muscle hardness around the knee and alleviating knee pain, Nakajima^[11,12] demonstrated that increasing blood flow and reducing stiffness in lower limb muscles could relieve knee pain. Building on this, considering the physiological heterogeneity in individual muscles, a self-controlled before-and-after study design with unilateral knee osteoarthritis (KOA) patients was deemed most suitable for our small-sample exploratory pilot trial. Therefore, we designed this trial protocol to preliminarily explore 3 main aspects: firstly, whether there are differences in hardness and blood circulation metrics between the tightened muscle on the affected side and its contralateral, relatively healthy counterpart; secondly, if FSN demonstrates specific therapeutic effects on the tightened muscle of the affected side compared to this counterpart; and thirdly, the potential correlation between the impact of FSN on muscle hardness and blood circulation and its therapeutic effectiveness.

2. Methods

This trial utilized a prospective, self-controlled before-and-after study design with blinding for data collector and data analyst. The lower limbs of participants with unilateral knee osteoarthritis were categorized into healthy and affected side groups based on the side of affliction, thereby excluding the need for random allocation and concealment of group assignments. The protocol was approved by the Ethics Committee of Guangdong Provincial Hospital of Traditional Chinese Medicine (Approval No.: BF2022-011-01) and registered with the Chinese Clinical Trial Registry (Registration No.: ChiCTR2300076488). All participants signed an informed consent form prior to joining. The study process is illustrated in Figure 2.

Figure 2.

Study process flowchart. HHb = deoxyhemoglobin, O₂Hb = oxyhemoglobin, ROM = range of motion, THb = total hemoglobin, TSI = tissue Saturation Index, VAS = Visual Analogue Scale, WOMAC = Western Ontario and McMaster Universities osteoarthritis index.

2.1. Sample size calculation

As a pilot study, a rigorous sample size calculation was not mandatory.^[13] Initially, we aimed for 30 participants, divided into healthy and affected side groups (n = 30 each). However, the study was prematurely terminated after completing 19 participants, primarily due to the following reasons: First, the research was temporarily suspended due to the continuous impact of COVID-19 in Guangzhou in mid to late 2022. At that time, there was a sharp increase in COVID-19 cases in the city. The primary trial personnel were consistently unable to access hospitals due to administrative policies, many communities were under lockdown following outbreaks among residents, and other citizens chose to avoid hospital visits to minimize the risk of infection. Consequently, the recruitment and progression of the trial faced significant challenges, and it was impossible to predict how the situation would evolve. Second, based on data from 19 participants who underwent 3 sessions of FSN therapy, including efficacy indicators (VAS and WOMAC function scores) and exploratory muscle indices (elastic modulus, muscle thickness, blood flow, and blood oxygen consumption rate of the lateral vastus muscle on the affected side), we set a 2-tailed α = 0.05 and n = 19 as parameters. Using the standard procedures^[14] in G*power 3.1.9.7 software (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany) for post hoc power analysis under conditions of uncertain effect size in paired t-tests, it was determined that in terms of efficacy indicators, the existing sample size could achieve over 99% power. This suggests reliable efficacy results and a reduced risk of type II errors, even without meeting the anticipated sample size. However, calculations for the exploratory muscle indices indicated that the desired power was far from being achieved, as the power almost always fell below 30%. Even expanding to the originally planned sample size of 30 participants would not yield an ideal power. Additionally, areas needing improvement were identified in our current protocol. Considering these factors and to avoid unnecessary research costs, we decided to terminate the study early.

2.2. Research objectives

Acupuncture treatment, as a complex clinical intervention, necessitates a preliminary trial to verify the feasibility and internal process coordination of the trial plan before conducting extensive formal research.^[15,16] FSN, more complex than traditional acupuncture, and the inclusion of 2 additional types of equipment examinations in this trial, underscore the complexity of this clinical study. Therefore, a pretrial assessment is crucial before embarking on large-scale trials. The specific objectives of this trial are:

1. Assessing the feasibility of FSN therapy, in conjunction with shear-wave elastography (SWE) and near-infrared spectroscopy (NIRS) technologies, to monitor changes in target muscle characteristics. This encompasses evaluating patient screening and consent processes, recruitment challenges, timeline estimations, research process coordination, participant compliance, and equipment suitability for research goals.

2. Evaluating the feasibility and safety of the FSN protocol for unilateral KOA in early to mid-stages, and providing sample size parameters for future clinical studies.

3. Conducting preliminary exploration of 3 established research questions based on study findings.

2.3. Research setting

The study was conducted at the Traditional Therapy Center and Ultrasound Department of Fangcun Hospital, Guangdong Provincial Hospital of Traditional Chinese Medicine. Room temperature and lighting conditions can affect the stability of the equipment data acquisition, thus we maintained the temperature between 22°C and 24°C, tailored to the individual needs of the participants, and kept the curtains in the experimental room fully closed. Besides necessary communication with participants, such as explaining procedures and issuing treatment instructions, researchers aimed to maintain a completely quiet environment, while participants were encouraged to remain physically and mentally calm.

2.4. Participant recruitment

Patients were recruited from May to September 2022 through outpatient and WeChat online advertising. Inclusion criteria were: Early to mid-stage unilateral knee osteoarthritis (KOA),^[17,18] Age 40 to 75 years, irrespective of gender, Body mass index (BMI) under 28 kg/m², Visual Analog Scale (VAS) score over 30 mm, and Spontaneous pain in the affected knee, evidenced by stiffness or palpable pain in the vastus lateralis compared to the contralateral side. Exclusion criteria included: Non-primary KOA, Acute inflammation in the affected knee or severe deformities/poor lower limb alignment, Recent alternative treatments for KOA, Comorbid conditions causing lower limb pain, Uncontrolled or undiagnosed internal diseases, Engagement in professional sports or manual labor, Bleeding disorders, Severe systemic or local lower limb skin diseases, and Pregnancy or lactation. Additionally, ultrasound was used to measure the subcutaneous layer thickness in the bilateral vastus lateralis, excluding those with measurements over 1cm. Eligible participants who understood the research procedures and volunteered were asked to sign the informed consent form.

2.5. Intervention

Qualified participants underwent consistent FSN therapy on both lower limbs, performed by a licensed physician with over 3 years of experience. The primary tightened muscles in patients with KOA are the quadriceps femoris, gastrocnemius, and tibialis anterior,^[19] with the quadriceps femoris being particularly associated with KOA.^[20–22] We selected these muscles for intervention, specifically choosing the vastus lateralis of the quadriceps femoris due to its thinner subcutaneous layer, facilitating observation. The FSN protocol involved 4 sessions (bi-daily over 1 week). The procedure was as follows: Participants assumed either a supine or prone position, depending on the treatment requirements. After skin disinfection at the needle entry point (usually the most painful area of each tightened muscle), a disposable FSN needle (Nanjing Paifu Medical Technology Co., Ltd., China, medium) was then prepared. The needle was inserted using a device (Nanjing Paifu Medical Technology Co., Ltd.) that penetrated the skin surface (Fig. 3). After device removal, the needle was advanced parallel to the skin in the subcutaneous loose connective tissue until fully inserted. Once inserted, the needle tip was retracted by 3 mm and concealed within the external cannula. Each participant comfort was checked before proceeding with fan-shaped swaying of the needle in the tissue. The reperfusion approach was synchronized with swaying: 20 seconds of swaying followed by 10 seconds of reperfusion, forming one treatment set, with each tightened muscle receiving 3 sets. The reperfusion approach varied based on muscle: gastrocnemius (prone position, ankle flexion resistance), tibialis anterior (supine position, ankle dorsiflexion resistance), and vastus lateralis (supine, knee extension against resistance with a foam shaft under the popliteal fossa).

Figure 3.

Insertion device usage examples. (A) Place the needle in the groove of the insertion device. (B) Slide the groove downwards to set the needle in the ready-to-eject position. (C) The physician uses one hand to gather the subcutaneous layer at the insertion point, while holding the insertion device in the other hand, aligning the needle tip with the insertion point. (D) Press the blue trigger, causing the needle to be ejected, with its tip penetrating the skin to the subcutaneous layer.

In a real clinical setting, the plastic cannula outside the stainless steel core of the FSN needle is often left under the skin for 3 to 8 hours post-treatment to enhance efficacy. However, in this trial, to avoid confounding factors that could affect treatment efficacy evaluation, we did not retain the cannula in participants.

Regarding treatment duration, prior study indicates that FSN treatment significantly improves KOA patients’ pain and functionality after at least 3 sessions.^[19] Although our study focused on the impact of 3 FSN sessions on evaluation indices, 4 treatments were conducted. This was due to observed fluctuations in elastic modulus values post-treatment, likely caused by muscle contractions during reperfusion approaches. In contrast, data collected before each treatment session was more stable. Thus, a fourth session was added not only to ensure accurate data representation of the 3 FSN treatments’ effects but also to maintain participant compliance for the complete course of the study.

2.6. Outcomes

Ultrasound and NIRS examinations were conducted by a medically trained practitioner, while outcome data were collected by a proficient evaluator who was unaware of the participants’ treatment process. The physicians responsible for handling the examination equipment, collecting outcomes, and administering acupuncture treatment were independent of each other.

2.7. Efficacy outcomes

2.7.1. Visual analogue scale (VAS)

In terms of pain, The focus was on the change in the reliable VAS score^[23] from baseline to before the fourth treatment, specifically whether the mean difference reached the minimal clinically important improvement (MCII) of 19.9 mm.^[24] The VAS comprised a 100 mm line, ranging from 0 (no pain) to 100 (extreme pain). Participants marked their pain level using “×” or “|,” and researchers measured and recorded these marks with a dedicated ruler.

2.7.2. Western Ontario and McMaster Universities Osteoarthritis index (WOMAC)

The primary focus in terms of physical function was on the differences in the valid and reliable WOMAC^[25] physical function subscale scores from baseline to before the fourth treatment. Particular attention was paid to whether the mean difference achieved the MCII of 6 points, derived from the MCII of 9.1 points for the 100-point maximum WOMAC physical function dimension.^[24] The WOMAC scale encompassed 3 dimensions: pain, stiffness, and physical function, with a total of 24 items, using the Likert 5-point scoring system ranging from “none” to “very severe.”

2.7.3. Active range of motion

This included measuring the active flexion, internal rotation, and external rotation angles of the participants’ affected knee joints at baseline and before the fourth treatment.

2.8. Safety outcome

Continuous recording of treatment-related adverse reactions was conducted from the first to the fourth treatment. Participants were instructed to report any perceived acupuncture-related adverse reactions, including pain, ecchymosis, dizziness, or nausea. Physicians evaluated and provided appropriate responses and suggestions for these reactions.

2.9. Exploratory outcomes

Before and after the first and fourth treatments, the Supersonic Aixplorer and the PortaLite system were used to measure local muscle thickness, mean elastic modulus, blood flow volume, and blood oxygen consumption rate of the bilateral vastus lateralis. Additionally, the PortaLite system captured real-time changes in oxygenated hemoglobin, deoxyhemoglobin, total hemoglobin, and tissue saturation index in the vastus lateralis of affected side during the first and fourth treatments. The detailed methodology is described in Supplemental Content 1, http://links.lww.com/MD/M584 (see Exploratory Indices Acquisition Process, Supplemental Content 1, http://links.lww.com/MD/M584, which details the acquisition process for all exploratory outcomes).

2.10. Statistical analysis

All 19 participants in the final analysis completed the protocols without missing data. Data analysis was performed using IBM SPSS Statistics 27.0 (IBM, Armonk, NY) following a per-protocol analysis approach. Statistical tests were bilateral, with P < .05 indicating significance. Normally distributed measurement data were described using mean and standard deviation, while non-normal data were described using median and interquartile ranges. Count data were presented as counts and frequencies. For continuous variables in inter-group and intra-group comparisons, paired t-tests or Wilcoxon tests were used depending on data normality. For count data, Fisher exact test was applied. Correlation analysis was employed to examine the relationship between the change in efficacy indices and exploratory indices.

3. Research results

3.1. Baseline characteristics

Baseline characteristics are shown in Table 1 (see Table 1, Supplemental Content 2, http://links.lww.com/MD/M585, which details the baseline characteristics of all participants). Of 19 participants, 84.20% were female and 15.80% were male. The incidence of the disease was 47.40% on the left side and 52.60% on the right side. The average age was 61.16 years, with an average BMI of 24.03. The average duration of the disease was 4.83 months. Grouping was based on the affected and contralateral sides of the participants, ensuring balanced demographic baseline characteristics.

3.2. Feasibility and safety

From May 12 to September 30, 2022, 55 volunteers were screened, with 19 qualifying and all consenting to participate. All 19 participants completed the study without protocol violation or deviation. The longest duration for visits was observed during the first and fourth sessions, involving both examinations and FSN therapy, taking about 4 to 6 hours. This varied according to the participants’ comprehension and the physiological state of the examination site. Regarding safety, 3 out of 19 participants experienced minor intervention-related adverse reactions, mainly ecchymosis at the needle site, which significantly resolved within a week.

3.3. Efficacy outcomes

Table 2 (see Table 2, Supplemental Content 3, http://links.lww.com/MD/M586, which illustrates the changes in efficacy outcomes after 3 sessions of FSN therapy) summarizes the efficacy outcomes. After 3 FSN therapy treatments, the VAS pain assessment decreased by 22.63 mm (95% CI, −28.39 to −16.87, P < .001), meeting the MCII. The WOMAC pain subscale decreased by 4.68 points (95% CI, −6.16 to −3.21, P < .001), stiffness subscale by 1.05 points (95% CI, −1.66 to −0.44, P = .005), and physical function subscale by 17.42 points (95% CI, −23.22 to −11.62, P < .001), also meeting the MCII. In an active knee joint ROM, flexion and external rotation angles increased by 7.37 degrees (95% CI, 2.83 to 11.91, P = .003) and 3.05 degrees (95% CI, 0.54 to 5.57, P = .02), respectively, showing statistical significance. The internal rotation angle increase of 3.63 degrees (95% CI, −0.94 to 8.21, P = .11) was not statistically significant.

3.4. Exploratory outcomes

3.4.1. Static acquisition indices

Initially, to understand muscle characteristic differences between the affected and contralateral sides, baseline bilateral indexes were compared. The vastus lateralis on the affected side showed higher local elastic modulus and lower muscle thickness, blood flow volume, and oxygen consumption rate compared to the contralateral side, but with non-significant differences (see Table 3, Supplemental Content 4, http://links.lww.com/MD/M587, which illustrates the baseline differences in bilateral static acquisition indices). Secondly, to determine if FSN had differential impacts on bilateral muscles, we assessed the differences in terms of impact size and trend. The impact size was assessed by comparing changes observed before-and-after single treatments (specifically, comparing conditions before the first and fourth treatments with those immediately afterwards, to reflect the instantaneous effects; same below) and following 3 treatments (evaluating changes from baseline to pre-fourth treatment, to indicate the cumulative effects; same below). No significant statistical differences were found in these comparisons (see Table 4, Supplemental Content 5, http://links.lww.com/MD/M588, which compares the differences between bilateral static acquisition indices at different time periods). Impact trends, categorized as increase, no change, or decrease, were assessed for changes post-single and post-three treatments. Except for a statistically significant difference in the trend of bilateral local muscle thickness after 3 treatments (P = .005), no other indices showed significant trends at different time points (see Table 5, Supplemental Content 6, http://links.lww.com/MD/M589, which compares the trends between the value changes of bilateral static acquisition indices at different time periods). Finally, to understand any correlation between the impact of FSN on muscle indices and efficacy changes, we conducted a correlation matrix analysis on the differences in VAS and WOMAC scores post 3 treatments, as well as changes in static muscle indices on the affected side. The results revealed a significant positive correlation between the differences in VAS and WOMAC scores (R = 0.54, P = .02), as well as between the differences in local muscle thickness and the blood oxygen consumption rate (R = 0.52, P = .02). No linear correlations were found among other variables (see Table 6, Supplemental Content 7, http://links.lww.com/MD/M590, which illustrates the correlations between the changes in efficacy indices and changes in static acquisition indices on the affected side after the first 3 treatments).

Furthermore, as shown in Figure 4 and table 7 and 8 (see Tables 7 and 8 in Supplemental Contents 8, http://links.lww.com/MD/M591 and 9, http://links.lww.com/MD/M592, respectively, which illustrate the differences in the bilateral static acquisition indices at different time periods), the local mean elastic modulus of bilateral vastus lateralis increased post-single treatments but decreased following 3 treatments, Notably, the immediate increase after the fourth treatment was statistically significant on both the healthy side (P < .001) and the affected side (P = .002). Local muscle thickness increased after single treatments and further increased following 3 treatments, with the differences not being statistically significant. Surprisingly, blood flow volume decreased after a single treatment, while the blood oxygen consumption rate increased as expected. However, both of them showed an increase following 3 treatments, with none of these differences being statistically significant. (Special note: In this trial, near-infrared spectroscopy was used to measure local muscle blood flow and oxygen consumption rate. The numerical value of blood flow was derived from the rising slope of total hemoglobin during venous occlusion, while the numerical value of blood oxygen consumption rate was derived from the falling slope of oxygenated hemoglobin during arterial occlusion. Therefore, it could be inferred that the numerical value of blood flow was positive, whereas the numerical value of oxygen consumption rate was negative. However, when considering the magnitude of oxygen consumption rate, it should have been based on the absolute value, unaffected by the direction of the slope. In all images and tables involving blood oxygen consumption rates presented in this study, the values were displayed based on their original negative values. Thus, during data analysis, this negative value required a conceptual shift. For example, although the blood oxygen consumption rates showed downward trends in Figure 4, they actually signified an increase in oxygen consumption).

Figure 4.

Trend chart showing changes over time in static acquisition indices of bilateral vastus lateralis muscles.

3.4.2. Dynamic acquisition indices

During the first and fourth treatments, real-time dynamic acquisition of oxygenated hemoglobin, deoxyhemoglobin, total hemoglobin, and tissue saturation index in the vastus lateralis of the affected sides was conducted using the PortaLite. Post-treatment, the values of local oxygenated hemoglobin and total hemoglobin values increased significantly (First treatment: O₂Hb: P = .005; THb: P = .006. Fourth treatment: O₂Hb: P = .002; THb: P = .004). Deoxyhemoglobin and tissue saturation index also increased during both periods, but these increases were not statistically significant (First treatment: HHb: P = .06; TSI: P = .42. Fourth treatment: HHb: P = .28; TSI: P = .56). Detailed results can be found in Table 9 (see Table 9, Supplemental Content 10, http://links.lww.com/MD/M593, which illustrates the dynamic acquisition indices during the first and fourth treatments of vastus lateralis on the affected side).

4. Discussion

In clinical practice, FSN therapy primarily targets skeletal muscles experiencing pathological states (tension or pain) to achieve therapeutic outcomes. This approach attributes the etiology of many diseases to skeletal muscle issues, particularly chronic musculoskeletal pain. The treatment focus in FSN therapy, termed “tightened muscle,” is crucial for understanding its primary clinical mechanisms. The hardness and blood flow of the tightened muscle are aspects of particular interest to clinicians in this field. To our knowledge, this study represents the first attempt to explore the clinical mechanisms of FSN therapy in these areas.

The primary aim of this trial design was to investigate potential characteristics in muscle hardness and blood circulation indicators of the vastus lateralis on the affected side (specifically, those conforming to the palpation characteristics of tightened muscle) compared to the contralateral side in KOA patients. Additionally, the study sought to explore the specific impact of FSN on the affected vastus lateralis and its potential correlation with clinical efficacy. This pilot study primarily aimed to explore the feasibility and safety of the trial protocol and to provide a reference for similar future research. Our main findings are as follows: The study protocol, while feasible, requires further modification; The FSN treatment protocol utilized in the trial not only demonstrated good clinical efficacy in treating KOA but also proved to be safe with minimal adverse reactions.; FSN can provide instantaneous enhancement in blood flow and oxygenation in tightened muscle. Below, we will systematically discuss and analyze the findings of this trial and briefly summarize some important research insights.

Our trial demonstrated the feasibility of the protocol in several aspects. The baseline gender ratio, average age, and average BMI of the 19 participants corresponded with the epidemiological characteristics of knee osteoarthritis patients both internationally and domestically.^[26–28] Despite challenges in recruiting participants due to specific requirements for BMI, subcutaneous fat thickness, and unilateral disease onset, a qualification rate of about 35% and a 100% consent rate indicate that recruiting eligible participants from community and outpatient settings is feasible. Although some patients initially had a fear of FSN needle upon seeing it, and both the first and fourth visits involved lengthy research processes with the physician, a 100% completion rate reflects good participant compliance and high acceptance of FSN therapy. Among the 19 participants, 3 experienced common acupuncture-related adverse reactions of local ecchymosis, all of which gradually resolved within a week. No other needle-related adverse reactions were observed, indicating the sufficient safety of FSN therapy.

Regarding clinical efficacy, pain and functional impairment are recommended core dimensions for assessing symptoms of KOA.^[29] We primarily assessed pain improvement using the 19.9mm MCII on the VAS scale and function improvement using the 6-point MCII on the WOMAC function subscale. Our findings indicate that the FSN therapy significantly improved both pain and function in participants, achieving the MCII. Compared to minimal clinical important difference (MCID), which may indicate symptom improvement or deterioration, MCII, emphasizing clinical improvement, is more recommended in musculoskeletal research.^[30] MCII is defined as the smallest patient-reported outcome change score of clinical significance from the patient perspective.^[24] The achievement of MCII indicates that the statistical differences brought about by the intervention are meaningful and perceptible to patients. However, as this trial involved simultaneous treatment of both lower limbs, we cannot rule out the possibility of additive effects on efficacy. This limitation may restrict the use of our data for sample size calculations in future clinical studies focused on unilateral treatment.

In our study, no significant findings emerged regarding the 3 research questions. We attribute this to several factors: Sample Size: The small-sample size in our trial inevitably led to type II errors in muscle index analysis, as indicated by post hoc power analysis using G*power software. Equipment Limitations: For SWE technology in ultrasound, typical applications like fibrosis, tumors, nodules,^[31] tendon diseases in musculoskeletal areas,^[32] and neuromuscular conditions like Parkinson disease and stroke^[33] display different colors on quantitative shear modulus maps based on tissue hardness. However, the vastus lateralis muscle belly in our KOA patients, displayed an overall blue color representing soft consistency, even though it might feel relatively tense or painful upon palpation. From this, it appears that the pathological tension state of our research target is underrepresented in SWE, suggesting a lower-than-expected tension level. Indeed, the average shear-wave elastic modulus of skeletal muscles ranges from 0.6 to 240.0 kPa,^[34] further indicating that our participants had low baseline elasticity values. This is understandable as our participants were mostly elderly, and multiple studies show a negative correlation between muscle elasticity and age.^[35–37] This characteristic of low elastic modulus may introduce a floor effect in our use of SWE to detect differences in stiffness between the participant bilateral muscles or in the stiffness changes brought about by FSN. Additionally, the NIRS device used had only 3 channels with low spatial resolution, which might miss significant findings in large skeletal muscle like vastus lateralis. Meanwhile, in the field of disease monitoring, NIRS devices are commonly used for peripheral vascular disease, muscle myopathies, among others,^[38] with few comparative studies on healthy and affected side muscles in diseased individuals like ours. A study by Han Cui^[39] using NIRS to detect differences in forearm muscle blood flow and oxygen consumption in patients in recovery phase of cerebral infarction found no significant differences in blood flow but significant differences in oxygen consumption (P < .01). The aforementioned applications of NIRS in detecting significant findings in skeletal muscle research suggest a potential ceiling effect when observing changes in blood flow and oxygen consumption rates in our trial participants using NIRS device. This is likely because the skeletal muscle condition of our participants is better than that of the patients in the aforementioned categories. Examination Point Localization: In our trial, the examination sites for ultrasound and NIRS had to be consistent. NIRS examination required arterial and venous occlusion using a thigh blood pressure cuff at the thigh root for performing arteriovenous occlusion to calculate local blood flow volume and blood oxygen consumption rate. Therefore, limited by the cuff position, we had to locate the most painful point on the lower third of the vastus lateralis as the examination point for both devices. However, extensive palpation revealed that the most painful point of the entire vastus lateralis muscle on the affected side was mostly on the upper third. This suggests that not obtaining indices from the most typical muscle locations might have also affected our results. Encouragingly, the significant increase in oxygenated hemoglobin detected in real-time by the NIRS device following treatments effectively demonstrates the impact of FSN therapy on instantaneously enhancing local blood flow and oxygenation in skeletal muscles. This provides valuable insights for future studies.

Undoubtedly, certain limitations are evident in this study. Firstly, the small-sample size and trial design constraints limit the representativeness of our findings. Furthermore, cross-education indicates that improved motor function in the limb receiving intervention can positively affect the contralateral limb.^[40] Similarly, knee pain in one limb may influence muscle strength in the opposite limb.^[41] Considering this inter-limb interaction, although they are minor, our approach of treating both lower limbs simultaneously could introduce confounding factors in evaluating the treatment effectiveness and muscle response. Therefore, further investigation with a more robust trial design is necessary to validate the clinical effectiveness and muscle response observed in this study.

Acknowledgments

We extend our sincere gratitude to all participants in this trial. Special thanks to Professor Zhonghua Fu’s for substantial support of the trial conditions and to Researcher Han Cui for invaluable insights regarding the use of NIRS devices. Our appreciation also goes to Rui Bao for expert assistance in data analysis and to physician Siyi Feng for guidance in ultrasound application. We are also thankful to Engineers Haitao Liu and Shengrong Wang for their professional guidance on utilizing NIRS devices.

Author contributions

Conceptualization: Xiaolin Yang, Jian Sun.

Methodology: Xiaolin Yang, Jian Sun.

Project administration: Xiaolin Yang, Hanlin Wang.

Supervision: Jian Sun.

Visualization: Hanlin Wang.

Writing – original draft: Xiaolin Yang.

Writing – review & editing: Xiaolin Yang, Hanlin Wang.