Acupotomy therapy for knee osteoarthritis: An overview of systematic reviews

Abstract

Background:

This study aimed to evaluate the methodological quality, report quality, and evidence quality of a meta-analysis (MA) and systematic review (SR) of the efficacy of acupotomy in the treatment of knee osteoarthritis (KOA), and provided a reference for clinical decision-making.

Methods:

We searched 8 databases to collect systematic reviews and meta-analyses on the efficacy of acupotomy in the treatment of KOA from January 30, 2018, to January 31, 2023. The methodological quality of the studies was assessed using the assessment of multiple systematic reviews (AMSTAR) 2 scale, the quality of the literature reports was scored using the Preferred Reporting Items for Systems Reviews and Meta-Analyses 2020 Version (PRISMA 2020),and the quality of the evidence was graded using the grading of recommendations assessment, development, and evaluation (GRADE) scale.

Results:

Nine systematic reviews including 35 outcome indicators were included. AMSTAR 2 evaluated the methodological quality of the included studies, and 1 was of low quality, 8 were of very low quality, and the entries with poor scores were 2, 3, 4, 8, 10, 12, and 13. By PRISMA 2020, there were some reporting deficiencies, and quality problems were mainly reflected in the abstract, information sources, search strategy, synthesis methods, reporting bias assessment, certainty assessment, reporting biases, certainly of evidence, registration and protocol. The GRADE classification results showed that there were 2 medium-quality evidences, 7 low-quality evidences, and 26 very low-quality evidences. The main factors of degradation were limitations, imprecision, and publication bias.

Conclusion:

Acupotomy had been a promising complementary treatment for KOA. However, due to the low quality of the SRs/MAs supporting these results, high-quality studies with rigorous study designs and larger samples were needed before widespread recommendations could be made.

1. Introduction

Osteoarthritis (OA) was the most common form of arthritis and chronic joint disease,^[1] affecting an estimated 250 million people worldwide.^[2] Among them, knee osteoarthritis (KOA) was the most common OA disease in middle-aged and elderly people,^[3] accounting for 85% of the global burden of osteoarthritis, with a global prevalence rate of about 4%, ranking 11th among disabling diseases.^[4] The main clinical manifestations were knee pain, stiffness, knee dysfunction, and proprioceptive changes, which seriously affected patients’ quality of life.^[5,6] Studies around the world had shown that with the rise in risk factors such as age, obesity, lack of exercise, and joint injury, the incidence of the disease is increasing, which increased the economic burden on families and society.^[7,8]

The management of KOA focused on pain relief and control. The Osteoarthritis Society International (OARSI) recommended conservative treatment as first-line treatment for KOA.^[9] Conservative treatment of KOA pain mainly included drug and non-drug therapies. Non-drug treatments included acupuncture, massage, physical therapy, exercise therapy, and functional exercise. The main drug treatments included analgesics, nonsteroidal anti-inflammatory drugs, and corticosteroid injections.^[10] Although the above drug treatment was effective to a certain extent, the adverse reactions caused by the drugs also limited the choice of treatment methods for patients with KOA.^[11] Acupotomy, as a non-drug therapy, had the advantages of minimal trauma, light pain, short course of treatment, and fast curative effect, and was a common method for the clinical treatment of KOA.^[12]

Currently, several randomized controlled trials (RCTs) had been conducted to assess the efficacy of acupotomy therapy for knee osteoarthritis (KOA), and multiple systematic reviews (SRs)/meta-analyses (MAs) had been published on the related findings. However, these SRs/MAs exhibited discrepancies in publication years, interventions, selected outcome measures, and resulting conclusions, making it difficult for clinicians to obtain reliable information intuitively. The quality of evidence in SRs/MAs was influenced by both the original studies and the methodological quality of the SRs/MAs themselves, thereby lacking objective evaluation of the reliability of their conclusions. Hence, this overview the SRs/MAs on acupotomy therapy for KOA to assess their reporting quality, methodological quality, and evidence quality, aiming to provide more reliable evidence-based guidance for clinical decision-making regarding acupotomy therapy for KOA. Our study aimed to guide clinical practice in formulating optimal treatment plans and providing treatment options for more KOA patients.

2. Materials and methods

2.1. Registration

A predetermined, written protocol of this overview was registered in the International Prospective Register of Systematic Overview (PROSPERO) database (https://www.crd.york.ac.uk/PROSPERO/), registration number: CRD42023445938. This overview of SRs/MAs was performed in accordance with the guidelines introduced by the Cochrane Collaboration Search Strategy and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.

2.2. Inclusion criteria

• Type of studies: This study included SRs/MAs of RCTs of patients who were diagnosed with KOA using definitive diagnostic criteria. Repeated publications and SRs/MAs that were not rigorous were also excluded.

• Interventions: The needle knife was the main intervention measure, and the observation group used a simple needle knife or needle knife combined with other therapies without limiting the equipment or treatment methods of the needle knife.

• The control group treatments were traditional Chinese and Western therapies other than needle knife therapy.

• Outcome indicators: SRs/MAs should have at least 1 clear outcome, such as the total efficacy rate, pain score, Western Ontario and McMaster University Osteoarthritis Index score (WOMAC), WOMAC pain score, Lysholm score, Japanese Orthopaedic Association (JOA) score, and Oxford Knee Score (OKS).

2.3. Exclusion criteria

We excluded studies that met the following criteria: a SR/MA that did not use needle-knife as an intervention, comparison of different types of needle-knife therapy, and other types of research, such as animal experiments, protocols, conference papers, case reports, and guidelines; the literature was repeated, and the full text could not be obtained.

2.4. Data sources and search strategy

We comprehensively searched the following 8 databases: Embase, PubMed, Web of Science, Cochrane Library, SinoMed database, China National Knowledge Infrastructure, Wanfang, and China Science and Technology Journal Database, and screened qualified SRs that had been released from January 30, 2018, to January 31, 2023. The key words mainly included knee osteoarthritis, knee articular bone, osteoarthritis, knee osteoarthritis, knee osteoarthritides, knee arthritis, osteoarthritis, knee degeneration, senile osteoarthritis, knee arthralgia, bone arthralgia, KOA, OA, needle knife, small needle knife, needle knife therapy, needle knife release, needle knife release technique, small needle knife closed release technique, system evaluation, and meta analysis. The search format was adjusted to adapt to different databases. In addition, we manually searched for relevant references in the review articles, and the search strategies for each database were presented in Appendix 1, Supplemental Digital Content, http://links.lww.com/MD/N571.

2.5. Data management and extraction

Two reviewers (Jiangchun Zhang and Tingting Pang) independently conducted literature screening (Appendix 2, Supplemental Digital Content, http://links.lww.com/MD/N571). All search results were imported into Endnote 20 to remove duplicates, and inconsistent articles were removed based on the title and abstract. Finally, the full text was read out and eligible SRS were included. The unresolved differences were resolved by a third evaluator (Yufeng Wang). Two reviewers (Li Dong and Yueting Wang) independently extracted the basic characteristics of the literature, including the author, publication year, diagnostic criteria, sample size, intervention, comparison, outcome, adverse effects, and methodological evaluation tools. Two reviewers (Li Dong and Yueting Wang) crosschecked what was extracted and consulted a third reviewer (Junjie Yao) for any discrepancies.

2.6. Quality assessment

Two reviewers (Ai Lin Li and Tingting Pang) independently evaluated the quality of the included systematic reviews and meta-analyses, and checked each other. In the case of disputes, 2 reviewers would discuss and reach a consensus, or a third reviewer (Yufeng Wang) would assist in the decision.

Two investigators (Jiangchun Zhang and Tingting Pang) used the AMSTAR 2 scale,^[13] Risk of Bias in Systematic Reviews (ROBIS) tool,^[14] PRISMA 2020 inventory (Appendix 3, Supplemental Digital Content, http://links.lww.com/MD/N571),^[15] and grading of recommendations assessment, development, and evaluation (GRADE) rating system to independently evaluate the methodological quality,^[16,17] risk level of bias, reporting quality, and quality of evidence included in the systematic reviews and meta-analyses. The results were descriptively analyzed. The final manuscript would be strictly drafted according to the requirements of the PRISMA checklist (Appendix 3, Supplemental Digital Content, http://links.lww.com/MD/N571).

3. Results

3.1. Results on literature search and selection

A total of 151 studies were retrieved from the 8 databases, and 88 studies were obtained after excluding duplicate studies. After reading the title and abstract, 21 literatures were obtained, 12 articles were screened after reading the full text, and finally, 9 articles were included.^[18–26] The literature screening process is illustrated in Figure 1.

Figure 1.

The PRISMA flow diagram of study selection.

3.2. Description of characteristics

A summary of the data extracted from the 9 SRs/MAs is presented in Table 1. These study included SRs/MAs published from 2019 to 2022. Five of them were written in Chinese,^[22–26] and the remaining 4 were written in English.^[18–21]

Table 1.

Characteristics of the included.

Intervention (D/F)	Comparison (D/F)	Adverse effects (number of RCTs, E/C)	Methodological evaluation tool	Primary outcomes	Main conclusion
Acupotomy (not reported)	Intra-articular HA (not reported)	The pain, bleeding, hematoma, swelling of the joint and infection of the incision after treatment with acupotomy (4/1)	Cochrane risk of bias tool	①②③④⑤⑥⑩	Acupotomy is a safe and effective treatment for KOA. Acupotomy’s potential for KOA treatment in alleviating knee pain or related symptoms for short-term and long-term.
Acupotomy (1–5 wk/once a week)	Acupuncture, electroacupuncture, sodium hyaluronate (not reported) radiofrequency electrotherapy and NSAIDs (1–5 wk/not reported)	Not reported	Cochrane risk of bias tool	①③⑥⑦⑧	Acupotomy therapy may be effective in reducing pain and improving the physical function of individuals with KOA.
Acupotomy (3–8 wk/not reported)	Conventional Western medicine (not reported)	Pain, palpitations, and increased blood pressure (3/1)	Cochrane risk of bias tool	①③⑥⑩	Acupotomy may be an effective and safe treatment for KOA.
Acupotomy combined with acupuncture and moxibustion	Other treatment methods other than acupotomy	Not reported	Cochrane risk of bias tool	①③⑥	Acupotomy combined with acupuncture and moxibustion for KOA can increase treatment efficiency and relieve postoperative pain.
Acupotomy (not reported)	Acupuncture (not reported)	Not reported	Cochrane risk of bias tool	①②⑥	Acupotomy efficient than acupuncture for the treatment of KOA.
Acupotomy combined with maneuver (not reported)	Oral NSAID (not reported)	Gastrointestinal reaction (3/1)	Cochrane risk of bias tool	①②③⑥⑦⑩	Acupotomy combined with maneuver in the treatment of KOA has significant curative effect, which can more effectively reduce the degree of knee pain, improve the function of knee joint, and is relatively safe under the standard operation.
Acupotomy combined with massage (not reported)	Other treatments other than acupotomy combined with massage (not reported)	Not reported	Cochrane risk of bias tool	①⑥	Acupotomy combined with massage has a significant clinical effect in the treatment of KOA.
Acupotomy (not reported)	Acupuncture, electroacupuncture, sodium hyaluronate (not reported)	Not reported	Cochrane risk of bias tool	①②③⑥⑦⑨	Acupotomy has certain advantages in improving clinical symptoms of KOA, protecting the knee cartilage, recovering the knee function.
Ordinary acupotomy (not reported)	Ordinary acupuncture (not reported)	Not reported	Cochrane risk of bias tool	①③⑥	Acupotomy therapy is an effective method for the treatment of KOA, and its curative effect is better than that of acupuncture.

① = total effectiveness rate, ② = the cure rate, ③ = WOMAC Scale, ④ = pain score, ⑤ = NRS, ⑥ = VAS, ⑦ = Lysholm Score, ⑧ = JOA, ⑨ = OKS, ⑩ = adverse reaction.

ACR = American College of Rheumatology, C = Criteria for the Diagnosis and Therapeutic Effect of TCM Diseases, D = Diagnostic Criteria of the Osteoarthritis Treatment Guide, D = duration, F = frequencies, G = guiding principle of clinical research on new drugs in the treatment of knee osteoarthritis score, JOA = Japanese Orthopaedic Association, OKS = Oxford Knee Score, VAS = visual analogue scale, WOMAC = Western Ontario and McMaster University Osteoarthritis Index score, wk = week.

One of these SRs/MA papers was published by a Korean author^[21] and the remaining 8 papers were published by authors from China. The number of RCTs included in these SRs varied widely, ranging from 9 to 43, and the total number of participants ranged from 655 to 4268. Interventions in the therapy group were mainly acupotomy or acupotomy combined with other therapies, while traditional Chinese or Western therapies other than acupotomy therapy were used in the control group. In terms of the quality assessment scales, all 9 used the Cochrane risk of bias criteria, and all 9 SRs/MAs reached a positive conclusion; detailed basic characteristics of the included SRs are shown in Table 1.

3.3. Results of the methodological quality

AMASTAR-2 was used to assess the methodological quality of the SRs included in this study, 1SR was rated as having low quality, and 8 SRs were rated as having critically low quality due to more than 1 serious deficiency of critical items 2, 4, 7, 9, 11, 13, and 15 and multiple deficiencies in non-critical items. In critical items, 8 SRs did not specify registration in the advance research project (item 2), and none of the SRs provided a comprehensive literature search strategy (item 4). All SRs (100%) selected appropriate effect sizes and statistical methods during the meta-analysis, and investigated the sources of heterogeneity and reasonable explanations (Item 11). Eight SRs (88.89%) reported publication bias (Item 15). Of the non-critical items, all SRs (100%) did not explain the reasons for the type of design included in the study (Item 3), 7 SRs (77.78%) mentioned that two people independently performed literature screening (Item 5), and 1 SR (11.11%) reported the source of funding for the included RCTs (Item 10). None of the SRs assessed the potential effects of RoB in a single RCT (Item 12). All SRs (100%) had heterogeneity or explained the heterogeneity reasonably (Item 14), and all SRs (100%) claimed no conflict of interest and identified funding sources (Item 16). Detailed results were presented in Table 2.

Table 2.

AMSTAR-2 for methodological quality of the included SRs/MAs.

Included studies	AMASTAR2																Quality
	Item 1	Item 2	Item 3	Item 4	Item 5	Item 6	Item 7	Item 8	Item 9	Item 10	Item 11	Item 12	Item 13	Item 14	Item 15	Item 16
Fang et al (2020)[18]	Y	N	N	PY	N	Y	Y	PY	Y	N	Y	N	N	Y	Y	Y	Critycally low
Sun et al (2020)[20]	Y	N	N	PY	Y	Y	PY	PY	PY	N	Y	N	N	Y	Y	Y	Critycally low
Lee et al (2021)[21]	Y	N	N	PY	N	Y	Y	PY	Y	N	Y	N	N	Y	Y	Y	Critycally low
Qu et al (2021)[19]	Y	PY	N	PY	Y	Y	PY	N	Y	N	Y	N	N	Y	Y	Y	Low
Zhang et al (2019)[22]	Y	N	N	PY	Y	Y	PY	N	PY	N	Y	N	N	Y	N	Y	Critycally low
Li et al (2021)[26]	Y	N	N	PY	Y	Y	Y	PY	Y	Y	Y	N	N	Y	Y	Y	Critycally low
Tang et al (2022)[24]	Y	N	N	PY	Y	Y	Y	PY	PY	N	Y	N	N	Y	Y	Y	Critycally low
Zhou et al (2022)[23]	Y	N	N	PY	Y	Y	Y	PY	PY	N	Y	N	N	Y	Y	Y	Critycally low
Li et al (2022)[25]	Y	N	N	PY	Y	Y	Y	PY	PY	N	Y	N	N	Y	Y	Y	Critycally low
YES [n (%)]	9 (100)	0 (0)	0 (0)	0 (0)	7 (77.78)	9 (100)	6 (66.67)	0 (0)	4 (44.44)	1 (11.11)	9 (100)	0 (0)	0 (0)	9 (100)	8 (88.89)	9 (100)

Bold values are critical items of AMSTAR-2.

AMSTAR = assessment of multiple systematic reviews, MA = meta-analysis, N = no, PY = partial yes, SR = systematic review, Y = yes.

3.4. Results of the reporting quality

The reporting quality of the included studies was evaluated using the PRISMA2020 statement. The manuscript structures of the SRs included in this overview were relatively complete. Item 1 (the title), Item 4 (introduction) and discussion sections had good integrity (100%); however, other sections had some deficiencies, such as Item 2 (abstract), Items 6 (information sources), 7 (search strategy), Item 14 (reporting bias assessment), Item 15 (certainty assessment), Item 21 (reporting biases), Item 22 (certainty of evidence), and Item 24 (registration and protocol). The additional details were provided in Table 3.

Table 3.

The quality assessment results of the PRISMA checklist.

Items	Included studies									YES [n (%)]
	Fang et al (2020)[18]	Sun et al (2020)[20]	Lee et al (2021)[21]	Qu et al (2021)[19]	Zhang et al (2019)[22]	Li et al (2021)[26]	Tang et al (2022)[24]	Zhou et al (2022)[23]	Li et al (2022)[25]
Item 1	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 2	PY	PY	PY	PY	PY	PY	PY	PY	PY	0 (0)
Item 3	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 4	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 5	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 6	Y	PY	PY	PY	PY	PY	PY	PY	PY	1 (11.11)
Item 7	N	N	N	N	N	N	N	N	N	0 (0)
Item 8	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 9	Y	Y	Y	Y	N	N	Y	Y	Y	7 (77.78)
Item 10a	Y	Y	PY	PY	PY	PY	PY	Y	PY	3 (33.33)
Item 10b	Y	PY	PY	PY	PY	Y	PY	Y	PY	3 (33.33)
Item 11	Y	Y	Y	Y	PY	Y	Y	Y	Y	8 (88.89)
Item 12	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 13a	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 13b	N	N	N	N	N	N	N	N	N	0 (0)
Item 13c	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 13d	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 13e	N	N	Y	N	N	N	N	N	N	1 (11.11)
Item 13f	N	N	N	N	N	N	N	Y	N	1 (11.11)
Item 14	N	N	N	N	N	N	N	N	N	0 (0)
Item 15	N	N	N	N	N	N	N	N	N	0 (0)
Item 16a	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 16b	N	N	N	N	N	N	N	N	N	0 (0)
Item 17	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 18	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 19	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 20a	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 20b	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 20c	N	N	N	N	N	N	N	Y	N	1 (11.11)
Item 20d	N	N	N	N	N	N	N	Y	N	1 (11.11)
Item 21	N	N	N	N	N	N	N	N	N	0 (0)
Item 22	N	N	N	N	N	N	N	N	N	0 (0)
Item 23a	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 23b	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 23c	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 23d	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 24a	N	Y	N	N	N	N	N	N	N	1 (11.11)
Item 24b	N	N	N	N	N	N	N	N	N	0 (0)
Item 24c	N	N	N	N	N	N	N	N	N	0 (0)
Item 25	Y	Y	Y	Y	Y	Y	Y	Y	Y	9 (100)
Item 26	N	Y	Y	Y	N	N	N	N	N	3 (33.33)
Item 27	N	PY	N	N	N	N	N	N	N	0 (0)

N = no, PRISMA = Preferred Reporting Items for Systems Reviews and Meta-Analyses, PY = partial yes, Y = yes.

3.5. Results of ROBIS evaluation

For ROBIS, phase 1 assessed the relevance of the research topic, and all SRs/MAs were rated as having a low risk of bias. Phase 2 assessed the risk of bias in the 4 domains. Domain 1 assessed the study eligibility criteria, and 5 out of 9 SRs/MAs were rated as having a low risk of bias. Domain 2 assessed the identification and selection of studies, and 1 out of 9 SRs/MAs were rated as having a low risk of bias. Domain 3 assessed the collection and study appraisal, and 5 SRs/MAs had a low risk of bias. One out of 9 SRs/MAs was rated as having an uncertain risk of bias in domains 2 and 3. Domain 4 assessed the synthesis and findings, and 5 SRs/MAs had a low risk of bias. Phase 3 considered the overall risk of bias in the reviews, and 2 SRs/MAs had a low risk of bias. The additional details were provided in Table 4.

Table 4.

Results of the ROBIS assessments.

Review	Phase 1	Phase 2				Phase 3
	Assessing relevance	Study eligibility criteria	Identification and selection of studies	Data collection and study appraisal	Synthesis and findings	Risk of bias in the review
Fang et al (2020)[18]	Low	Low	Unclear	Low	Low	Low
Sun et al (2020)[20]	Low	Low	Low	Low	Low	Low
Lee et al (2021)[21]	Low	High	High	Unclear	Low	High
Qu et al (2021)[19]	Low	High	High	Low	Low	High
Zhang et al (2019)[22]	Low	High	High	High	High	High
Li et al (2021)[26]	Low	Low	High	Low	High	High
Tang et al (2022)[24]	Low	High	High	High	High	High
Zhou et al (2022)[23]	Low	Low	High	High	Low	High
Li et al (2022)[25]	Low	Low	High	Low	High	High

ROBIS = Risk of Bias in Systematic Reviews.

3.6. Evidence quality

The 9 SRs included 35 outcomes. The results showed that 2 (5.71%) were rated as moderate quality, 7 (20%) as low quality, and 26 (74.29%) as critically low quality. The risk of bias (35/35, 100%), inconsistency (19/35,54.29%), and publication bias (29/35, 82.86%) were the main factors for obtaining the results. The details were listed in Table 5.

Table 5.

Results of the GRADE for quality of evidence.

Included studies	Outcomes	Number of RCTs (participants)	Effect estimate (95% CI)	Certainty assessment					Quality of evidence
				Risk of bias	Inconsistency	Indirectness	Imprecision	Publication bias
Fang et al (2020)[18]	Cure rate	10 (766)	OR 2.04 (1.46, 2.85)	−1†	0	0	0	0	Moderate
	Total effective rate	11 (1082)	OR 2.25 (1.55, 3.28)	−1†	0	0	0	0	Moderate
	Pain score	6 (472)	SMD −1.02 (−1.72, −0.31)	−1†	−2‡	0	0	−1¶	Critically low
	WOMAC score	4 (238)	SMD −0.74 (−1.11, −0.37)	−1†	0	0	−1‖	−1¶	Critically low
Sun et al (2020)[20]	VAS pain score	22 (1969)	MD −1.11 (−1.51, −0.71)	−1†	−2‡	0	0	0	Critically low
	WOMAC pain score	9 (880)	MD −2.32 (−2.94, −1.69)	−1†	−1§	0	0	0	Low
	Total effective rate	23 (2276)	RR 1.15 (1.09, 1.21)	−1†	−1§	0	0	0	Low
	Lysholm’s score	3 (464)	MD 12.75 (2.61, 22.89)	−1†	−2‡	0	0	−1¶	Critically low
	JOA score	5 (436)	MD 6.93 (4.11, 9.76)	−1†	−2‡	0	0	−1¶	Critically low
Lee et al (2021)[21]	VAS pain score	4 (429)	MD −1.85 (−1.92, −1.79)	−2*	−1§	0	0	−1¶	Critically low
	WOMAC pain	5 (428)	SMD −0.91 (−1.11, −0.70)	−2*	−1§	0	0	−1¶	Critically low
	Effective rate	16 (1369)	RR 1.15 (1.10, 1.20)	−2*	−1§	0	0	−1¶	Critically low
Qu et al (2021)[19]	Effective rate	11 (970)	OR 3.09 (1.11, 8.56)	−1†	−1§	0	0	−1¶	Critically low
	VAS pain score	6 (472)	MD −1.44 (−2.47, −0.42)	−1†	−1§	0	0	−1¶	Critically low
	WOMAC score	6 (462)	MD −4.76 (−8.50, −1.03)	−1†	−1§	0	0	−1¶	Critically low
	IL-1β level	3 (219)	MD −5.80 (−7.16, −4.44)	−1†	−1§	0	−1‖	−1¶	Critically low
Zhang et al (2019)[22]	Effective rate	9 (726)	OR 3.07 (1.93, 4.88)	−1†	0	0	0	−1¶	Low
	Cure rate	5 (426)	OR 2.28 (1.33, 3.90)	−1†	0	0	0	−1¶	Low
	VAS pain score	3 (170)	SMD −1.31 (−2.12, −0.50)	−1†	−2‡	0	−1‖	−1¶	Critically low
Li et al (2021)[26]	Cure rate	7 (835)	RR 2.14 (1.45, 3.17)	−1†	−1§	0	0	−1¶	Critically low
	Total effective rate	9 (1017)	RR 1.20 (1.14,1.26)	−1†	0	0	0	−1¶	Low
	VAS pain score	5 (602)	MD −1.78 (−2.64, −0.93)	−1†	−2‡	0	0	−1¶	Critically low
	Lysholm’s score	3 (304)	MD 15.41 (9.04, 21.78)	−1†	0	0	−1‖	−1¶	Critically low
	WOMAC score	3 (385)	MD −9.46 (−11.82, −7.11)	−1†	0	0	−1‖	−1¶	Critically low
Tang et al (2022)[24]	Total effective rate	16 (1655)	RR 1.18 (1.14, 1.22)	−2*	0	0	0	0	Low
	VAS pain score	5 (408)	MD −1.16 (−1.42, −0.90)	−2*	0	0	0	−1¶	Critically low
Zhou et al (2022)[23]	Total effective rate	14 (1158)	OR 3.39 (2.31, 4.99)	−1†	0	0	0	−1¶	Critically low
	VAS pain score	7 (388)	SMD −1.11 (−1.33, −0.90)	−1†	0	0	−1‖	−1¶	Critically low
	WOMAC score	3 (182)	SMD −0.69 (−0.99, −0.39)	−1†	0	0	−1‖	−1¶	Critically low
	OKS	2 (143)	MD −5.74 (−8.20, −3.27)	−1†	−1§	0	−1‖	−1¶	Critically low
	Lysholm’s score	2 (211)	MD 4.34 (3.30, 5.39)	−1†	0	0	−1‖	−1¶	Critically low
	TNF-α	3 (163)	MD −15.80 (−19.17, −12.43)	−1†	0	0	−1‖	−1¶	Critically low
Li et al (2022)[25]	Total effective rate	15 (1264)	RR 1.16 (1.11, 1.22)	−1†	0	0	0	−1¶	Low
	VAS pain score	11 (1078)	MD −1.24 (−1.58, −0.90)	−1†	−2‡	0	0	−1¶	Critically low
	WOMAC score	6 (570)	MD −9.06 (−13.34, −4.78)	−1†	−2‡	0	0	−1¶	Critically low

0 = no downgrade, −1 = lower the level, −2 = down 2 levels.

GRADE = grading of recommendations assessment, development, and evaluation, MD = mean difference, OKS = Oxford Knee Score, OR = odds ratio, RR = relative risk, SMD = standard mean difference, TNF-α = tumor necrosis factor alpha, VAS = visual analogue scale, WOMAC = Western Ontario and McMaster University Osteoarthritis Index.

^* There are 2 or more high risk of bias in randomization, blindness, allocation concealment, completeness, or selective reporting of outcome data.

^† There is a high risk of bias in randomization, blindness, allocation concealment, completeness, or selective reporting of outcome data.

^‡ There was significant heterogeneity for the included studies (75%≤I² ≤ 100%).

^§ There was moderate heterogeneity for the included studies (50%≤I² ≤ 75%).

^‖ The sample size is small, and the CI is wide.

^¶ The funnel plot was asymmetric or <9 included studies or all were positive results.

4. Discussion

Acupotomy therapy was simple, convenient, effective, and inexpensive and has been widely used in the treatment of KOA in the past 20 years. Recommendation 7 of the “Chinese Medicine Diagnosis and Treatment Guide for Knee Osteoarthritis (2020 Edition)” states: For KOA patients with knee pain, contracture flexion deformity, functional restriction and remission, acupotomy therapy was recommended (2B). Acupotomy could be performed on the suprapatellar bursa, subpatellar fat pad, internal knee eye, external knee eye, tibial collateral ligament, iliotibial bundle, and the anseropodium sac. By cutting, separating, spatulating, adjusting, and releasing tendon ligaments and other soft tissues, biomechanical balance of the knee joint could be achieved. It was suitable for patients with knee pain, morning stiffness, muscle adhesion, functional limitation, contracture flexion deformity, obvious attack period, and remission period. KOA could relieve knee pain and improve joint function.^[27] In recent years, numerous SRs/MAs had also been performed to elucidate their efficacy and safety; however, the evaluation indicators, methodological quality, and results varied across studies. Based on the above reasons, this systematic evaluation reevaluation study was conducted to evaluate the methodology and evidence-level quality of the analysis. This review was the first systematic reevaluation study to comprehensively evaluate the clinical efficacy of needle knife in KOA.

4.1. Therapeutic effect of acupotomy on KOA

Nine SRs/MAs of KOA were included in this study, Published in 2019 to 2022 and included 10 outcome indicators, most of which were subjective evidence, such as total effective rate, cure rate, visual analogue scale (VAS) score, WOMAC Scale, Lysholm Score, JOA, and OKS; these indicators were all assessed according to the subjective feelings of the test subjects, which had great limitations in nature; only 1 adverse reaction indicator was evidence of objectivity, and the outcome indicators were mostly low and extremely low, and the results of the meta-analysis showed that the clinical efficacy of acupotomy treatment of KOA was better than that of the control group.

4.2. The quality of SRs/MAs methodology for acupotometric treatment of KOA needs to be improved

According to the AMASTAR-2 evaluation, 1 study was of low quality and the remaining 8 were of extremely low quality. None of the eight documents explained the research methods before the systematic evaluation and only 1 article was registered in the previous stage. In addition to following the methodological criteria of systematic evaluation, registering and writing a systematic evaluation of the program could reduce study bias. In addition, all studies partially reported key Item 4, mostly for not stating whether to include or consult appropriate experts in related fields, and not searching gray literature, the above problems might lead to incomplete literature search. Key Item 11 was completed at 100.0%, key Item 9 (44.44%), and key Item 13 (0). In the future, researchers should fully evaluate the risk of bias and publication deviation of the data and provide a more reliable interpretation of the results to enhance the credibility of the evaluation results.

According to the PRISMA 2020 evaluation, only 1 of the included studies was pre-registered, and because of the subjective content of the descriptive analysis in the systematic review/meta-analysis, failure to perform pre-registration in accordance with the specifications might result in selective reporting. Article 7 of the PRISMA statement requires presenting the full search strategies for all databases, but none of the systematic studies met the requirements for full reporting. Since comprehensive and extensive literature was a prerequisite for writing a good review, failure to report search strategies would reduce the transparency and reproducibility of the study.

4.3. The level of evidence for SRs/MAs of needle-knife therapy for KOA is low

According to the GRADE evidence evaluation, 2 moderate-quality studies suggested that needle knife treatment was superior to intra-articular HA treatment in terms of cure rate and total effective rate. There were 7 pieces of evidence of low quality, the rest of which were of very low quality (74.29%). The main reasons for this degradation focused on the limitations, imprecision, and publication deviation of the original studies. Considering the original clinical trial, the part only reported the randomized method, while the allocation concealment was not reported by the standard method; therefore, the corresponding downgrading factor could not be avoided when performing the SRs/MAs. The degradation in imprecision was mostly due to the small sample size of this clinical outcome index; therefore, there was the possibility of publication deviation.

4.4. The safety of acupotomy for KOA remains to be determine

Only 1 of the 9 studies included in this study quantified the safety indicators of acupotomy for KOA,^[18] and the evidence level was very low. Although this study reported no statistically significant difference in the incidence of adverse events for acupotomy treatment of KOA compared with injection of intra-articular HA, no definitive conclusions could be drawn based on the available evidence. The safety of acupotomy for KOA needed further study.

4.5. Recommendations for SRs/MAs of the efficacy of acupotomy in treating KOA

In the 9 included articles, all reported positive results, which indicated the advantages of the treatment of KOA, reflected in clinical effective rate, relief of knee pain, and improvement of knee function. However, due to the low methodological quality and evidence strength of the included SRs/MAs, the results of this study should be treated with caution. When preparing future SRs/MAs analyses of acupotomic efficacy for KOA, it was recommended that standardized production procedures should be carried out in strict accordance with the Cochrane Handbook and PRISMA statements to ensure optimal quality and results of systematic reviews and meta-analyses.

5. Strengths and limitations

This review identified some shortcomings in the preocular acupotomy treatment of KOA through the AMSTAR 2 and GRADE evaluation process, and proposes some solutions that would help guide future high-quality studies. Only SRs/MAs published in Chinese and English were included in this study, which may be incomplete, and the evaluation process was subjectively evaluated by the researchers according to the evaluation criteria, which might be biased due to differences in the level and understanding of the evaluators.

6. Conclusions

The existing systematic review/meta-analysis resulted indicate that acupotomy therapy held promise as a treatment option and might improve symptoms in patients with knee osteoarthritis (KOA). However, the current systematic reviews/meta-analyses were greatly limited by low methodological quality. Therefore, researchers needed to validate these findings through relevant studies with rigorous experimental designs, standardized protocols, and larger sample sizes to provide high-quality evidence for clinical decision-making and practice regarding acupotomy therapy for KOA.

Author contributions

Conceptualization: Jiangchun Zhang, Yufeng Wang.

Data curation: Junjie Yao, Li Dong, Yueting Wang.

Formal analysis: Jiangchun Zhang, Tingting Pang.

Methodology: Ailin Li, Li Dong, Yueting Wang.

Supervision: Yufeng Wang.

Writing – original draft: Jiangchun Zhang.

Writing – review & editing: Jiangchun Zhang, Yufeng Wang.