Knee osteoarthritis (KOA) is a degenerative arthritis, primarily defined by the breakdown of joint cartilage and underlying bone. The clinical syndromes include knee pain, swelling, joint deformity, and disability.[1,2] According to previous studies, the incidence of KOA is more than 50% in people over 60 years old in China. Female presented higher incidence of KOA than male. Approximately, 14 million people are suffering from KOA in the United States, and causing up to $420 million per year in the treatment of KOA. With the increasing prevalence of obesity and the aging of the population, KOA constitutes a global burden and has become a serious public health problem. The most common causes of KOA include aging, gender, obesity, skeletal muscle mass, menopausal status, genetics, and mechanical factors. Currently, there are both surgical and conservative treatment options for KOA, with treatment aimed at pain relief, joint function protection, and improve life quality improvement.
Common conservative treatment options include weight loss, knee strength training, nonsteroidal anti-inflammatory drugs administration, hyaluronic acid administration, acupuncture, moxibustion, extracorporeal shockwave, and acupotomy (AT).[7–10] Among these conservative therapies, AT, a new category of acupuncture combine the traditional acupuncture science with modern anatomy, modern surgical principles, and pathology theories. The AT consists of a handle, cylindrical body, and flat head (Fig. 1). It can effectively eliminate adhesion, contracture, and relieve soft tissue tension to recover normal tissue function by cutting and detaching local lesions.[12,13] In clinical practice, AT is a common treatment for KOA in China, due to the reasons that AT is easy to operate, low cost, and highly effective. At present, traditional AT locates the site and depth of the puncture site based on the experience, body surface bony landmarks, and the tactile feedbacks estimated by the physician from the AT. Therefore, such operations are carried out without direct vision, leading to the possible impact on the clinical efficacy due to the inaccuracy of the therapeutic target locations, and may even damage some blood vessels and nerves.
AT visualization is the unavoidable future. Some previous studies found that ultrasound-guided acupotomy (UGAT) improves the efficacy and cause fewer complicating syndromes, comparing to traditional AT. However, these clinical researches are still at the primary stage, and no meta-analysis regarding this topic has been conducted previously, while the usage of UGAT for KOA treatment still has no convincing supportive study.
In this study, systematic reviews and meta-analyses of randomized controlled trials (RCTs) were conducted in order to compare the effectiveness and safety of UGAT with other therapies (traditional AT, acupuncture, and electroacupuncture [EA]) in the treatment of KOA.
2. Materials and methods
2.1. Research strategies and selection criteria
Independently researches were carried out by 2 researchers (JXW and YTL) in 6 academic databases: PubMed, the Cochrane Library, EMBASE, the China national knowledge infrastructure, the Wan Fang databases, and the Chinese biomedical literature database to identify relevant articles, from their inception to January 1, 2022. The following search terms were used: (“acupotomy,” “needle scalpel,” “needle knife,” “acupotome”), (“ultrasound,” “ultrasonography,” “ultrasonic,” “ultrasonics”), and (“Knee Osteoarthritides,” “Osteoarthritides, Knee,” “Knees, Osteoarthritis Of”). Detailed search criteria for PubMed were listed in Appendix S1, Supplemental Digital Content, https://links.lww.com/MD/I331. In addition, potentially relevant studies were checked individually to ensure that there were no missing papers. There was no language limitation.
2.2. Studies selection
2.2.1. Inclusion criteria.
- Types of participants: participants were patients with clinically confirmed KOA regardless of their age, race, sex, education, and employment.
- Types of interventions: UGAT (material of needles, acupoints, and ultrasonic equipment were not constrained).
- Types of control groups: conventional therapy generally used for KOA such as traditional AT, acupuncture, EA, and ultrasound-guided drug injection therapy were included.
- Types of outcome measures: primary outcome measures were score changes on the visual analog scale (VAS) and Lysholm knee score (LKSS); secondary outcome measures were clinical effective rate (ER), and adverse events.
- Types of studies: RCTs.
2.2.2. Exclusion criteria.
- Quasi-RCTs and randomized crossover studies.
- Duplicate reports, animal studies, case reports, and letters.
- Data in abstract forms, or data that were not completed.
- Involvement of additional treatments, or control group with additional treatments.
- Small sample size (less than 15 participants).
2.3. Data extraction and management
Two reviewers (JXW and YTL) independently undertook the extracted relevant data of each report that met the inclusion criteria. Any disagreements/inconsistency were resolved by thorough discussions, until consensus was achieved. This standardized form included 3 major domains: general information (such as name of first author, publication time, and location of study), experimental design (such as participant numbers, sex, age, trial methods, detailed procedures of the treatment and control groups, type of ultrasound device, total period, and outcome), and conclusion.
2.4. Assessing of the bias
Two reviewers (JXW and YTL) independently assessed the bias risks in each literature studied using the Cochrane “Risk of bias” tool. It included 6 domains of bias: selection bias (method of randomization and allocation concealment), performance bias (blinding of personnel and participants), detection bias (blinding of the result assessor), attrition bias (incomplete result data), reporting bias (selective reporting), and other biases (baseline balance and fund). The risk of bias for every domain was categorized as entailing a low, unclear, and high.
2.5. Statistical analyses
All statistical analyses were conducted by Review Manager version 5.3 software (https://training.cochrane.org/online-learning/core-software/revman). For continuous outcomes (VAS and LKSS), mean differences (MD) and 95% confidence interval (CI) were subjected to quantitative analysis data of each group. Relative risk (RR) or odds ratio combined with 95% CI were evaluated for dichotomous data (ER and adverse events). For each study, P value lower than .05 was considered as statistically significant. Statistical heterogeneities among studies were evaluated using the chi-squared test and the Higgins I2 measure. When test values indicated moderate heterogeneity (P < .1 and I2 > 50%) random rather than fixed-effects models were applied. To compare with different control groups, further analyses were conducted. To estimate the robustness of the findings, we conducted a sensitivity analysis. Moreover, visual observation of funnel plot was subjected to the evaluation of potential publication bias if sufficient studies were included (n ≥ 10).
2.6. Quality of evidence
We used the grading of recommendations assessment, development, and evaluation methodology to evaluate the qualities of evidences for all meta-analyses, and categorized them into high, moderate, low, or very low.
3.1. Literature selection
In accordance with the retrieval strategies, 115 potentially relevant studies were retrieved. Further, we excluded 27 duplicates studies by using EndNote X7 software (https://endnote.com/downloads) and excluded 40 studies due to inappropriate diagnosis and wrong intervention measures. After reading the full contents of all remaining articles, 40 studies were excluded for reasons such as no RCTs were included or insufficient data. Finally, 8 studies17–24 were included in this systematic review. Figure 2 shows the flow diagram describing the process of studies selection.
3.2. Characteristics of included literature
The main features of the included studies were listed as in Table 1. Trials included in this study were all carried out in China. All results were published from 2013 to 2019, with a total of 543 participants. Among them, the sample size ranged from 48 to 123. The baseline data of the included studies demonstrated no significant difference. Based on the comparisons, 6 studies[17–22] compared UGAT with traditional AT, 1 study compared UGAT with EA, and 1 study compared UGAT with manual acupuncture (MA). Moreover, the average treatment time ranged from 2 to 5 weeks.
Table 1 -
Features of the included studies.
|Ma and Ding (2013)
|UGAT 1 time per week (3 wk)
||AT 1 time per week (3 wk)
|Ruan and Zeng (2017)
|UGAT 1 time per week (5 wk)
||AT 1 time per week (5 wk)
|E: 56.3 ± 4.6
C: 56.1 ± 4.7
|UGAT 1 time per wk (3 wk)
||AT 1 time per week (3 wk)
||VAS, ER, LKSS
|E: 54.7 ± 5.4
C: 53.4 ± 4.3
|UGAT 1 time per wk (3 wk)
||AT 1 time per wk (3 wk)
|UGAT 1 time per wk (3 wk)
||AT 1 time per wk (3 wk)
|E: 57.3 ± 6.8
C: 58.7 ± 6.9
|UGAT 1 time per wk (2 wk)
||AT 1 time per wk (2 wk)
|Ding et al (2016)
|UGAT 1 time per wk (3 wk)
||EA 5 times per wk (30 min, 3 wk)
|Li et al (2018)
|E: 65.7 ± 4.5
C: 63.5 ± 4.1
|UGAT 1 time per wk (2 wk)
||MA 5 times per wk (30 min, 2 wk)
AT = traditional acupotomy, C = control group, E = experimental group, EA = electroacupuncture, ER = effective rate, LKSS = Lysholm knee score, MA = manual acupuncture, NR = not reported, UGAT = ultrasound-guided acupotomy, VAS = visual analogue scale.
3.3. Methodological quality of included studies
As shown in Figures 3 and 4, the results of risk of bias assessment of all included studies were summarized. Among the 8 included studies, 2 studies[18,22] employed computer random number generator, 3 studies[21,23,24] used applied random number table, and the remaining 3 studies[17,19,20] only mentioned “random” without provide details of the randomization procedure. Two studies[21,22] used sealed opaque envelopes, and other studies did not describe allocation concealment. Given this type of trials were quite difficult to blind participants and personnel completely, we therefore rated them as high risk, with the exception of only 2 studies[18,22] clearly reported the blinding of outcome assessment. Regarding incomplete data, 3 studies reported dropout numbers and reasons. Additionally, the presentation of “reporting bias and other biases” were lacked in all included studies.
3.4. Results of meta-analysis
The 8 studies[17–24] (including 543 patients) included to evaluate the pooled estimation for assessing the VAS score revealed that UGAT could decrease the VAS score more significantly in comparing with the control group (MD = −0.81, 95% CI = [−1.15, −0.47], P < .00001). However, there was moderate statistical among these RCTs (χ2 = 14.39, P = .04, I2 = 51%), while a random-effects model was applied. We performed a sensitivity analysis and found that 1 study (Ma and Ding) was removed, and the remaining studies were considered homogeneous and the results were stable (MD = −0.76, 95% CI = [−0.97, −0.55], P < .00001; χ2 = 4.07, P = .80, I2 = 0%). For the study of heterogeneity, subgroup analysis was subjected according to the different types of comparison groups: traditional AT, EA, and MA. Subgroup analysis demonstrated positive effects of UGAT on pain relief in comparing with AT (MD = −0.82, 95% CI = [−1.29, −0.36], P = .0005; P = .02, I2 = 64%; 6 study, 423 participants), EA (MD = −0.70, 95% CI = [−1.37, −0.03], P = .0005; 1 study, 60 participants), and AT (MD = −1.00, 95% CI = [−1.74, −0.26], P = .0005; 1 study, 60 participants) (Fig. 5).
Two studies[19,20] reported that the LKSS was used to evaluate knee joint function. The meta-analysis indicated that UGAT could further improve knee function compared with AT (MD = 8.26, 95% CI = [1.56, 14.97], P = .02). The result was obvious heterogeneity (χ2 = 4.06, P = .04, I2 = 75%) and a random effects model was applied (Fig. 6).
3.4.3. Clinical ER.
Six studies[18–21,23,24] reported the clinical ERs of these 2 groups of patients. Data extracted showed no significant heterogeneity (χ2 = 5.58, P = .35, I2 = 10%) in these RCTs, therefore a fixed-effect model was applied to evaluate the pooled RR and 95% CI. The data demonstrated that the clinical efficacy rate improved significantly in the UGAT group versus the control group (RR = 1.14, 95% CI = [1.06, 1.23], P = .005). According to the different types of comparison groups, subgroup analyses were conducted. Subgroup analysis also showed that UGAT was statistically significantly better than AT (RR = 1.13, 95% CI = [1.04, 1.23], P = .004) with no heterogeneity (χ2 = 2.42, P = .49, I2 = 0%) and MA (RR = 1.35, 95% CI = [1.02, 1.79], P = .04). However, no difference was detected between UGAT and EA based on 1 study (RR = 1.04, 95% CI = [0.89, 1.21], P = .64) (Fig. 7).
3.4.4. Adverse events.
Four studies[17,20–22] reported adverse events occurred in the 2 groups during the treatments, including hematomas, local painful sensation, bleeding, swelling of the joint, and slight nerve damage. All adverse events were shown in Table 2. Analysis of data showed no heterogeneity (χ2 = 1.70, P = .64, I2 = 0%), and the fixed-effects model demonstrated that adverse event rates were significantly lower for UGAT (odds ratio = 0.27, 95% CI = [0.12, 0.63], P = .002) compared to AT (Fig. 8).
Table 2 -
The adverse events about all included RCTs.
|Slight nerve damage
|Local painful sensation[17,22]
|Swelling of the joint[21,22]
RCTs = randomized controlled trials.
3.4.5. Level of evidence.
Based on the results of the grading of recommendations assessment, development, and evaluation criteria, the qualities of evidences for most comparisons were rated within the region of very low to low (Table 3). Most of the studies did not report blinding of personnel and participants, allocation concealment, and the outcomes and reason of the adverse event was not sufficiently described, leading to the result that the qualities of evidences were initially downgraded. Also, the significant heterogeneity and small sample sizes led to imprecision also downgraded the evidence level.
Table 3 -
The quality of evidence.
||No. of study (subjects)
||Quality of the evidence
||+/−/−/−/; very low
||+/−/−/−/; very low
A: study design defects may affect the results; B: significant heterogeneity; C: very few participants may affect the results; D: publication bias may affect the results.
DD = design defects, ER = effective rate, HE = heterogeneity, IN = indirectness, LKSS = Lysholm knee score, PB = publication bias, RI = data sparse or incomplete, VAS = visual analog scale.
4.1. Principal findings
This study was the first meta-analysis aimed to evaluate the effectiveness and safety of the UGAT therapy for treatments of KOA. After a comprehensive research of the major academic databases, 8 studies with 543 participants were included in this meta-analysis. From the available data, the present meta-analysis demonstrated that UGAT were effective for KOA as follows: With respect to LKSS and VAS score, UGAT was superior in improved knee function and pain relief compared with the control group based on the very low-quality evidence. In the subgroup analysis, the results were not significantly changed in different types of comparison groups. Sensitivity analysis demonstrated that the data in this meta-analysis were relatively stable. The low-quality evidences showed that UGAT was more effective than AT and MA. However, no statistical significance was demonstrated between UGAT and EA based on 1 study. UGAT significantly reduced incidence of adverse events compared with AT. These seemingly positive results should be interpreted with caution due to the low methodology quality, a limited sample size, and the language limitation of the included trials.
4.2. Mechanism of AT on KOA
The underlying mechanisms of AT for KOA was not clear; however, it demonstrated a positive therapeutic effect on KOA. The theory of traditional Chinese medicine suggests that trauma to the sinews, and cumulative fatigue entangled in the fascia’ muscle may form “strips” and “nodules,” which would obstruct the channels and eventually lead to knee diseases. AT therapy was beneficial to release “strips,” “nodules,” and relieving spasms in the surrounding ligaments and tendons to recover joint functions.[26,27] In addition, recent studies indicated that the mechanism of AT therapy potentially associated with decreased inflammatory cytokine levels, improved skeletal muscle fibrosis, regulating biomechanical properties of ligament, inhibition of chondrocyte apoptosis, and repairing the damage of cartilage to some extent.
4.3. Advantages of UGAT in KOA
Traditional AT treatment was a closed procedure that its efficacy and surgical safety depend on the operator’s familiarity with local anatomy and technical level. Hence, visualization of AT therapy was the best way to ensure the safety and effectiveness. With the development of ultrasound-guided techniques, they have been commonly applied as an effective tool for interventional diagnosis and treatments of musculoskeletal and articular diseases. AT guided by ultrasound for KOA had several advantages as follows. First, it has the characteristics of convenient, noninvasive, non-radiation, and flexible; second, the real-time ultrasonic images were able to display the position of AT and nearby structures precisely, which had good accuracy and safety; third, ultrasound images were easily observed from layers of the tissue structures, therefore the location and type of lesions were accurately identified; finally, it was also conducive to learning and dissemination. Therefore, UGAT had broad application prospects in the future.
4.4. Limitations of these studies
In this study, potential limitations of meta-analysis revealed possible impacts in the interpretation of our findings. First, sample sizes of most included studies were relatively small, causing the decrease of the overall precision of the estimations. Second, the methodological quality was low for most studies, especially lack of the details of blinding and allocation concealment. Meanwhile, no protocol or clinical trial registration were recorded for studies included in this, therefore the possibility of selective reporting were not clear. Third, there was significant heterogeneity across included studies. Although subgroup analysis was conducted, heterogeneity has not been resolved. The methodological bias, differences in treatment points selection, and duration of treatment may have contributed to the heterogeneity. Finally, ultrasound sonographic relevant indices, such as joint effusion and synovial thickness, were not assessed, and adverse events were underestimated.
In conclusion, despite the limitations of the methodology, our results of meta-analysis indicated the potential use of UGAT therapy as an effective and safe method in the clinical treatments of KOA, and could be applied in the management of KOA. To strengthen supportive evidence, more large-scale, long-term follow-up and high-quality RCTs were recommend in future.
Conceptualization: Sixiong Lin, Renpan Zhang.
Formal analysis: Sixiong Lin, Chuanshi Lai.
Methodology: Sixiong Lin, Chuanshi Lai, Jinxin Wang, Yating Lin, Yinyong Tu, Yuanfang Yang.
Software: Sixiong Lin.
Writing – original draft: Sixiong Lin, Chuanshi Lai, Renpan Zhang.
Writing – review & editing: Sixiong Lin, Renpan Zhang.
. Bhatia D, Bejarano T, Novo M. Current interventions in the management of knee osteoarthritis. J Pharm Bioallied Sci. 2013;5:30–8.
. Bijlsma JW, Berenbaum F, Lafeber FP. Osteoarthritis: an update with relevance for clinical practice. Lancet. 2011;377:2115–26.
. Tang X, Wang S, Zhan S, et al. The prevalence of symptomatic knee osteoarthritis in China: results from the China Health and Retirement longitudinal study. Arthritis Rheumatol. 2016; 68:648–53.
. Vina ER, Kwoh CK. Epidemiology of osteoarthritis: literature update. Curr Opin Rheumatol. 2018;30:160–7.
. Chen A, Gupte C, Akhtar K, et al. The global economic cost of osteoarthritis: how the UK compares. Arthritis. 2012;2012:1–6.
. Silverwood V, Blagojevic-Bucknall M, Jinks CC, et al. Current evidence on risk factors for knee osteoarthritis in older adults: a systematic review and meta-analysis. Osteoarthritis Cartilage. 2015;23:507–15.
. Liu SC, Qiao XF, Tang QX, et al. Therapeutic efficacy of extracorporeal shock wave combined with hyaluronic acid on knee osteoarthritis. Medicine (Baltimore). 2019;98:e14589.
. Tío L, Orellana C, Pérez-García S, et al. Efecto del con-droitin sulfato en la sinovitis de pacientes ton artrosis derodilla. Med Clin. 2017;149:9–16.
. Kolasinski SL, Neogi T, Hochberg MC, et al. 2019 American college of rheumatology/arthritis foundation guideline for the management of osteoarthritis of the hand, hip, and knee. Arthritis Care Res. 2020;72:149–62.
. Bennell KL, Hunter DJ, Hinman RS. Management of osteoarthritis of the knee. BMJ. 2012;345:e4934.
. Fang T, Li Q, Zhou F, et al. Effect and safety of acupotomy in treatment of knee osteoarthritis: a systematic review and meta-analysis. J Tradit Chin Med. 2020;40:355–64.
. Liu J, Zeng W, Lin Q, et al. Proteomic analyses reveals the mechanism of acupotomy intervention on the treatment of knee osteoarthritis in rabbits. Evid Based Compl Alternat Med. 2022;2022:5698387.
. You Y, Yang FY, Liu NG. Acupotomy therapy for shoulder adhesive capsulitis: a systematic review and meta-analysis of randomized controlled trials. Evid Based Compl Alternat Med. 2019;2019:2010816.
. Zhu XY, Shen YF, Liu ZX, et al. Ultrasound-guided percutaneous release procedures in the lumbar ligamentum flavum by acupotomy: a cadaveric study. Evid Based Compl Alternat Med. 2019;2019:2807901.
. Qiu Z, Jia Y, Shen Y, et al. Acupotomy by an ultrasound-guided technique: a protocol for a systematic review. Med (Baltim). 2019;98:e17398.
. Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations. BMJ. 2004;328:1490.
. Ma YH, Ding K. Application of ultrasound-guided technique in the treatment of knee osteoarthritis with needle knife. Beijing Med. 2013;35:545–7.
. Ruan YJ, Zeng ZM. Observation of therapeutic effect of ultrasound-guided needle knife in the treatment of refractory knee osteoarthritis and comparative study of soft tissue ultrasonography before and after. J Ext Ther TCM. 2017;26:6–8.
. Deng ZY. Clinical efficacy and improvement of clinical symptoms of ultrasound-guided needle knife in the treatment of knee osteoarthritis. Clin Med Eng. 2019;26:1025–6.
. Li MT. Clinical Observation of Ultrasound-Guided Needle Knife in the Treatment of Knee Osteoarthritis. M.S. thesis, Hubei University of Traditional Chinese Medicine2019:17–25.
. Zhong TH. Observation and Study on the Therapeutic Effect of Ultrasound-Guided Needle Knife in the Treatment of Knee Osteoarthritis. M.S. thesis, Guangzhou, China: Guangzhou University of Traditional Chinese Medicine, 2016.
. Zhang D. Study on the Mechanism of Needle Knife in the Treatment of Knee Osteoarthritis Based on the Theory of Tendon Injury. M.S. thesis, Beijing, China: Beijing University of Chinese Medicine, 2019.
. Ding Y, Wang YX, Shi X, et al. Effect of ultrasound-guided acupotomy
vs electro-acupuncture onknee osteoarthritis: a randomized controlled study. J Tradit Chin Med. 2016;36:450–5.
. Li LN, Hu GQ, Gao CC, et al. Clinical effect of ultrasound-guided needle knife in the treatment of knee osteoarthritis. China Med Herald. 2018;15:38–141.
. Wei S, Chen ZH, Sun WF, et al. Evaluating meridian-sinew release therapy for the treatment of knee osteoarthritis. Evid Based Complement Alternat Med. 2013;2013:182528.
. Zhang R, Li L, Chen B, et al. Acupotomy versus nonsteroidal anti-inflammatory drugs for knee osteoarthritis: protocol for a systematic review and meta-analysis. Medicine (Baltim). 2019;98:e17051.
. Jun S, Lee JH, Gong HM, et al. Efficacy and safety of miniscalpel acupuncture on knee osteoarthritis: a randomized controlled pilot trial. J Pharmacopuncture. 2018;21:151–8.
. Lin M, Li X, Liang W, et al. Needle-knife therapy improves the clinical symptoms of knee osteoarthritis by inhibiting the expression of inflammatory cytokines. Exp Ther Med. 2014;7:835–42.
. Huang YR, Jin YL, Li N, et al. Effects of acupotomy, electroacupuncture and Yuanli on protein expression of Bcl-2, bax and caspase-3 in rectus femoris muscle of rabbits with knee osteoarthritis. Acupunct Res. 2018;39:105–23.
. Wang LJ, Shi XW, Zhang W, et al. Effect of needle knife intervention on tensile mechanics of femoral quadriceps tendon in rabbits with knee osteoarthritis. Zhongguo Gu Shang. 2019;32:462–8.
. Huang YR, Jin YL, Li N, et al. Effects of acupotomy, electroacupuncture or round-sharp acupuncture needle interventions on expression of Bcl-2, bax, caspase-3 proteins of rectus femoris in rabbits with knee ostarthritis. Zhen Ci Yan Jiu. 2014;39:100–23.
. Zhang W, Gao Y, Guo CQ, et al. Effect of acupotomy versus electroacupuncture on ethology and morphology in a rabbit model of knee osteoarthritis. J Tradit Chin Med. 2019;39:229–36.
. Yoon SH, Kwon CY, Jo HG, et al. Safety of acupotomy in a real-world setting: a prospective pilot and feasibility study. J Integr Med. 2022;20:514–23.
. Boesen M, Ellegaard K, Henriksen M, et al. Osteoarthritis year in review 2016: imaging. Osteoarthritis Cartilage. 2017;25:216–26.