Acupotomy, a type of acupuncture that combines conventional acupuncture using a small scalpel, has recently attracted attention. This manipulation can be considered minimally invasive [1]. Owing to its shorter treatment period, faster recovery period, and minimal injury to the patient than during a surgical operation, the utilization of acupotomy in clinical practice is drawing increasing attention. For acupotomy to be used more systematically in clinical practice, the range of diseases in which acupotomy can be highly effective and the effects of treatment must be organized and analyzed by examining the grounds for various clinical uses in which acupuncture therapy has been used so far. Previously, Yuk et al. [2] selected clinical practices of acupotomy performed between 1999 and May 2014 and studied the disease type, acupoints (treatment sites), treatment intervention, evaluation index, and analysis according to the sample size of the study target by year. However, from May 2014 to the present, no high-quality trend analysis studies on acupotomy have been conducted. Therefore, to outline the trend of clinical studies, we selected randomized controlled trials (RCTs) among acupuncture studies published in English or Korean between May 2014 and April 2022, analyzed the published year, acupotomy terminology, study sample size, disease type, treatment sites, retention time, insertion depth of the acupotomy, size of acupotomy, treatment frequency, and duration, treatment intervention, adverse effects, acupotomy manipulation methods, and results, and assessed RCTs using the Cochrane risk-of-bias (RoB) tool to clarify the clinical research trend of acupotomy therapy.

1. Database and search method

The analysis examined clinical research studies published over approximately 8 years between May 2014 and April 2022. International research works were retrieved from PubMed, Wanfang, and China National Knowledge Infrastructure (CNKI), and Korean research works were retrieved from the National Digital Science Library (NDSL), Research Information Sharing Service (RISS), Korean Studies Information Service System (KISS), and Oriental Medicine Advanced Searching Integrated System (OASIS). In PubMed, the search words were the following English terms: “acupotomy,” “acupotomies,” “acupotomology,” “acupotome,” “stiletto needle,” “small needle-knife,” “needle knife,” “needle scalpel,” “miniscalpel,” “miniscalpel acupuncture,” “miniscalpel needle,” “stiletto needle,” “sword like needle,” “acupuncture knife,” “knife-shaped acupuncture,” “dochim,” “inchim,” and “ceramic acupuncture therapy.” In CNKI, the search words were the following Chinese words: “zhendao,” “daozhen,” “taozhen,” “taozhenshu,” and “xiaozhendao” (in Chinese). In Korean databases (NDSL, RISS, KISS, and OASIS), in addition to the English terms listed above, the search terms “Do-Chim” (in Korean) or “Chim-Do” (in Korean) were used.

2. Inclusion and exclusion

RCTs on acupotomy published between May 2014 and April 2022 were selected and reviewed. Among the selected clinical articles, clinical listings without patients, studies that were published in other languages, abridged or incomplete articles (if the original full text could not be obtained), and experimental studies on animals, such as rabbits, were excluded.

Two independent researchers checked the titles and abstracts of the selected studies and created the final sample by reviewing the articles’ full texts against the predefined inclusion and exclusion criteria. Disagreements that could not be resolved were referred to a third researcher to determine the suitability of the article.

As this was a retrospective study, informed consent was not necessary.

3. Data extraction

In the selected clinical studies, the published year, study sample size, disease type, and adverse effects were analyzed. In addition, based on Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) treatment sites, the retention time, insertion depth, treatment frequency, and duration, treatment intervention, and type of acupotomy manipulation method were analyzed.

4. Assessing the quality of research

The RoB in RCTs was assessed using the Cochrane RoB tool [3]. All items were evaluated when specified in the text of the selected study. In cases of disagreement between the evaluators, the opinions of third-party researchers were sought.

1. Search results

As a previous study analyzed the research trends in acupotomy between 1999 and May 2014, studies published between May 2014 and April 2022 were searched in international and Korean databases [2]. A total of 926 articles were retrieved from PubMed (n = 100), CNKI (n = 481), and Wanfang (n = 345). Further, a total of 233 articles were retrieved from Korean databases NDSL (n = 60), RISS (n = 66), KISS (n = 25), and OASIS (n = 82). Among a total of 1159 articles reviewed in this study, 534 articles were excluded for being duplicates, 274 were not an RCT, 416 studies were not published in English or Korean, and 22 were not related to human acupotomy. Finally, 37 studies were selected. After screening, 21 studies were further excluded because the full text of those studies could not be obtained, and another two studies were excluded because the full text was not published in English or Korean. Thus, 14 studies were selected for review (Fig. 1).

Figure 1. Flow diagram according to PRISMA. PRISMA, Preferred Reporting Items for Systematic reviews and Meta-Analyses; CNKI, China National Knowledge Infrastructure; NDSL, National Digital Science Library; RISS, Research Information Sharing Service; KISS, Korean Studies Information Service System; OASIS, Oriental Medicine Advanced Searching Integrated System; RCT, randomized controlled trials.

2. Analysis of selected articles

All 14 articles included in this study evaluated musculoskeletal diseases, which included four articles on knee diseases (28.56%), three articles on lumbar diseases (21.42%), one article on cervical diseases (7.14%), two articles on ankle joint diseases (14.28%), and four single articles on the shoulder joint, wrist joint, and whole spinal and muscular diseases (7.14%) (Tables 1, 2).

Table 1 . Analysis of acupotomy articles.

Disease	First author (y)	Terminology of the acupotomy	Samples (M:F)	Treatment intervention	Insertion depth and size (D × L mm)	Retention time	Treatment frequency and period
Knee
Knee osteoarthritis	Jun (2018) [5]	Miniscalpel acupuncture	I (1:11) C (4:8)	AC CA + EA	NR (0.5 × 50)	X	1/wk 3 wk
Knee osteoarthritis	Wang (2018) [6]	Acupotomy	I (8:24) C (7:24)	AC AC	NR (0.8 × 40)	X	1/wk 3 wk
Knee osteoarthritis	Hua (2021) [7]	Acupotomy	I (23:40) C (18:43)	AC AC	Reached the surface of the bone (0.8 × 40)	X	1/wk 4 wk
Knee osteoarthritis	Ding (2016) [8]	Acupotomy	I (30) C (30)	AC CA + EA	NR (NR)	X	3/3 wk 3 wk
Low back
Lumbar disk herniation	Jeong (2020) [10]	Acupotomy	I (33:40) C (34:39)	AC CA	50–60 mm (0.75 × 80)	X	4/2 wk 2 wk
Ankylosing spondylitis	You (2020) [11]	Needle-knife	I1 (28:2) I2 (29:1) C (29:1)	AC + etanercept injection AC Drug	NR (NR)	X	1 time NS
Sacral nerve dysfunction	Zhang (2018) [9]	Acupotomy	I (16:22) C (17:20)	AC + fire needle + PNB PNB	NR (NR)	X	1/wk 4 wk
Ankle & foot
Tarsal tunnel syndrome	Fu (2020) [12]	Acupotomy	I (22:18) C (18:22)	AC WM	5 mm (1.0 × 50)	X	1/wk 1–3 wk
Plantar fascilitis	Li (2014) [13]	Miniscalpel-needle	I (29) C (25)	AC WM	NR (0.8 × 50)	X	1 time NS
Neck
Chronic neck pain	Zheng (2014) [14]	Miniscalpel-needle	I (45:37) C (48:25)	AC CA	Not deeper than the posterior tubercle of the transverse processes (NR)	X	1/wk 3 wk
Shoulder
Frozen shoulder	Wang (2021) [15]	Acupotomy	I (0:30) C (0:30)	AC AC	Bone surface of the coracoid process (NR)	X	1/wk 5 wk
Wrist
Carpal tunnel syndrome	Zhang (2019) [16]	Miniscalpel-needle	I (5:18) C (6:17)	AC + WM WM	Transverse carpel ligament (NR)	X	1 time
Whole spine
Adolescent idiopathic scoliosis	Wei (2015) [17]	Acupotomy	I (18:40) C (10:39)	AC + Daoyin + Tuina Milwaukee brace	NR (NR)	X	1/wk 10 wk
Muscle
Myofascial pain syndrome	Ma (2010) [4]	Miniscalpel-needle	I1 (7:8) C2 (8:7) C (6:7)	AC + SNS CA + SNS SNS	NR (0.8 × 50)	1 min	1–2/wk 1 wk

M, male; F, femal; D, diameter; L, length; I, intervention group; C, control group; AC, acupotomy; CA, common acupuncture; EA, electro acupuncture; PNB, pudendal nerve block; WM, Western medicine; SNS, self-neck-stretching exercises; NR, not recorded..

Table 2 . Analysis of acupotomy according to disease classification.

Disease	First author (y)	Treatment sites	Depth of acupotomy	Size (D × L mm)
Knee
Knee osteoarthritis	Jun (2018) [5]	EX-LE2, EX-LE210, LR8, GB33, medial, and lateral regions of the patella, medial, and lateral sides of the quadriceps tendon	NR	0.5 × 50
Knee osteoarthritis	Wang (2018) [6]	Tenderness points of three yang meridians and three yin meridians of the foot and funicular nodules	NR	0.8 × 40
Knee osteoarthritis	Hua (2021) [7]	BL40, KI10, GB34, LR8, LR7, and SP9	Reached the surface of the bone	0.8 × 40
Knee osteoarthritis	Ding (2016) [8]	A-shi point, tendon ligament attached points, bone spurs, etc.	NR	NR
Low back
Lumbar disk herniation	Jeong (2020) [10]	Corresponding disk level based on the imaging findings/20–30 mm away from the spinous process	50–60 mm	0.75 × 80
Ankylosing spondylitis	You (2020) [11]	Positioned by C-arm X-ray machine	NR	NR
Sacral nerve dysfunction	Zhang (2018) [9]	BL31, BL32, BL33, and BL34	NR	NR
Ankle and foot
Tarsal tunnel syndrome	Fu (2020) [12]	1. Intersection point between the line and the posterior and inferior edge of the medial malleolus 2. Intersection point between the line and the inner edge of the calcaneus 3. Intersection of the line and the inferior edge of the medial malleolus 4. Intersection with the medial edge of the calcaneus	Not more than 0.5 cm	1.0 × 50
Plantar fascilitis	Li (2014) [13]	The most painful tender point over the medial tubercle of the calcaneum	NR	0.8 × 50
Neck
Chronic neck pain	Zheng (2014) [14]	Trigger point	Not deeper than the posterior tubercle of the transverse processes	NR
Shoulder
Frozen shoulder	Wang (2021) [15]	Bone surface of the coracoid process	Bone surface of the coracoid process	NR
Wrist
Carpal tunnel syndrome	Zhang (2019) [16]	Transverse carpel ligament	Transverse carpel ligament	NR
Whole spine
Adolescent idiopathic scoliosis	Wei (2015) [17]	Lumbodorsal fascia ligament, thoracolumbar junction ligament, and soft tissue around the neck, thoracic facet, and adhesions due to scarring	NR	NR
Muscle
Myofascial pain syndrome	Ma (2010) [4]	Trigger point	NR	0.8 × 50

D, diameter; L, length; EX-LE2, quadriceps tendon; EX-LE210, patella ligament; LR8, medial collateral ligament; GB33, lateral collateral ligament; BL40, Weizhong; KI10, Yingu; GB34, Yanglingquan; LR7, Xiguan; SP9, Yinlingquan; BL31, Shangliao; BL32, Ciliao; BL33, Zhongliao; BL34, Xialiao; NR, not recorded..

Of the total 14 articles, one (7.14%) was published in 2010, two (14.28%) in 2014, one (7.14%) in 2015, one (7.14%) in 2016, three (21.42%) in 2018, one (7.14%) in 2019, four (28.56%) in 2020, and one (7.14%) in 2021 (Table 1).

Terms describing acupotomy include acupotome, acupotomology, acupotomy, needle knife, needle scalpel, miniscalpel, acupuncture, miniscalpel needle, stiletto needle, sword-like needle, and Xiaozhendao [1]. Thus, the articles reviewed in this study involved various terms for acupotomy. Of the 14 articles, seven (49.98%) referred to acupotomy, four (28.56%) to miniscalpel-needle, and three single articles (7.14% each) each to miniscalpel acupuncture, acupotomology, and needle-knife (Table 1).

In this study, 3 (21.42%) involved <50 participants, 7 (49.98%) with 50–100, 3 (21.42%) with 100–150, and 1 (7.14%) with 150–200. The total number of participants involved in the studies was 1,127, and the average was 80.5 (Table 1).

Of the 14 studies, only one study included a retention time of approximately 1 minute [4], and in the other 13 studies, acupotomy was removed immediately after the procedure without retention time. Regarding treatment points, in all studies, the participants were treated in their anatomical structures, tender points, and acupoints. Several studies used anatomical structures, tender points, and acupoints simultaneously. Most studies used treatment sites near the disease area.

Four studies did not specify the insertion depth or size, three studies specified both the insertion depth and size, three studies specified only the insertion depth, and four studies specified only the size (Tables 1, 2).

In knee osteoarthritis, the acupotomy was inserted deep enough to touch the bone surface, using acupotomy with a diameter of 0.5–0.8 mm and a length of 40–50 mm; however, one study did not mention the insertion depth or its size [5-8]. In lower back diseases, the acupotomy (0.75 mm in diameter and 80 mm in length) was inserted 50–60 mm deep into the lumbar disk herniation; however, two studies mentioned neither the insertion depth nor its size [9-11]. In ankle and foot diseases, an acupotomy with a diameter of 0.8–1.0 mm and a length of 50 mm was used, and the study showed that it was inserted not deeper than 50 mm [12,13]. In chronic neck pain, the acupotomy was not inserted deeper than the posterior tubercle of the transverse processes; however, the size of the acupotomy was not mentioned [14]. In the frozen shoulder, the acupotomy was inserted until the bone surface of the coracoid process; however, the size of the acupotomy was not mentioned [15]. In carpal tunnel syndrome, the acupotomy was inserted until the transverse carpal ligament; however, its size was not mentioned [16]. In adolescent idiopathic scoliosis, neither the insertion depth nor the size of the acupotomy was mentioned [17]. Finally, in myofascial pain syndrome, acupotomy of 0.8 mm in diameter and 50 mm in length was used [4].

Regarding the frequency of acupotomy, seven studies reported that acupotomy was performed once a week, which is the most common. In three studies, acupotomy was administered three times a week for 3 weeks, four times a week for 2 weeks, or 1–2 times during a week (Table 1). In three studies, acupotomy was performed only once. Six single studies reported that the treatment lasted 1, 2, 3, 5, 10, or 1–3 weeks, depending on the severity of the symptoms. The average treatment period was 4.2 weeks, excluding the studies in which acupotomy was conducted only once or the treatment period varied depending on the symptoms.

Among the total of 14 studies, 11 compared acupotomy with other treatments, and three studies used acupotomy for both experimental and control groups. Among the 11 studies that used acupotomy only in an experimental group, 6 used acupotomy alone, and the remaining 5 reported the use of various complex treatment methods, including acupotomy. The combined treatment methods involved etanercept injection, pudendal nerve block therapy, steroid injections, Daoyin and Tuina, and self-neck stretching exercises [4,9,11,16,17]. In one study, the experimental groups were divided into two groups: in one group, etanercept injection was administered together with acupotomy, and in the other group, only acupotomy was used; the results were compared with a control group receiving Western medicine treatment [11]. Among the three studies featuring acupotomy in both experimental and control groups, two compared the differences between the application of meridian sinew and anatomical theories [6,7], and one compared the differences between pricking and dissection techniques (Table 3) [15].

Table 3 . Treatment intervention and adverse effects reported in acupotomy articles.

Disease	First author (y)	Treatment intervention	Type of acupotomy manipulation method	Adverse effects
Knee
Knee osteoarthritis	Jun (2018) [5]	I: AC C: CA + EA	Not mentioned	Not occurred
Knee osteoarthritis	Wang (2018) [6]	I: AC (meridian sinew theory) C: AC (anatomical theory)	I: longitudinal dissection and subcutaneous sweeping C: pressurizing to separate and puncturing	I: 2 patients (6.3% of 32) showed slight palpitation and discomfort (d/t fear) (not recur 2nd and 3rd treatments) No serious adverse event was reported C: 3 patients (9.7% of 31) complained of severe pain
Knee osteoarthritis	Hua (2021) [7]	I: AC (meridian sinew theory) C: AC (anatomical theory)	I: released the adhesions vertically and subcutaneous weeping C: pressing releasing and puncturing	Total of 6 cases (4.8% of 124) of mild adverse event: 5 (4%) of subcutaneous hemorrhage and 1 (0.8%) of tingling. All recovered completely without medical intervention
Knee osteoarthritis	Ding (2016) [8]	I: AC C: CA + EA	Releasing the subcutaneous tissue	Not mentioned
Low back
Lumbar disk herniation	Jeong (2020) [10]	I: AC C: CA	Not mentioned	No severe adverse events related to the intervention 3 (0.7% of 429) mild adverse events in the AC group, 4 (0.9% of 426) mild adverse events in the CA group related to the intervention (post intervention muscle pain)
Ankylosing spondylitis	You (2020) [11]	I 1: AC + etanercept injection I 2: AC C: WM (p.o med)	Acupotomy lysis	Not mentioned
Sacral nerve dysfunction	Zhang (2018) [9]	I: AC + fire needle + PNB C: PNB	Dredge and strip longitudinally	Not occurred
Ankle & foot
Tarsal tunnel syndrome	Fu (2020) [12]	I: AC C: WM(triamcinolone acetonide acetate 1 mL, lidocaine hydrochloride 2 mL, 0.9% sodium chloride 2 mL, total 5 mL injection)	I: release adhesion to relieve entrapment	Not occurred
Plantar fascilitis	Li (2014) [13]	I: AC C: WM (steroid injection)	Release of plantar fasciitis by moving up and down	5 patients (9.2%) reported mild pain, 1 (1.8%) reported subcutaneous bleeding (total 11.1%)
Neck
Chronic neck pain	Zheng (2014) [14]	I: AC C: CA	Moved up and down longitudinally	6 patients (7.3% of 30) in experimental group, 7 patients (9.6% of 30) in control group Slight pain and somatic reactions (sweating)
Shoulder
Frozen shoulder	Wang (2021) [15]	I: AC (pricking technique) C: AC (dissection technique)	I: pricking C: dissection	Not occurred
Wrist
Carpal tunnel syndrome	Zhang (2019) [16]	I: AC + WM (steroid injection) C: WM (steroid injection)	Release the nerve entrapment	1 patient (of 25) in the experimental group had mild pain, but disappeared within 24 hours (4% in the experimental group, 2% in total)
Whole spine
Adolescent idiopathic scoliosis	Wei (2015) [17]	I: AC + Daoyin + Tuina C: Milwaukee brace	Release and dredge and cross the contracture band	Not occurred
Muscle
Myofascial pain syndrome	Ma (2010) [4]	I: AC + SNS C1: CA + SNS C2: SNS	Moved up and down longitudinally	9 (60% of 15) in the AC group (coldness, burning sensation, and muscle pain) 11 (73.3% of 15) of the CA group (coldness, burning sensation, and muscle pain) Did not last more than a few days

I, intervention group; C, control group; AC, acupotomy; CA, common acupuncture; EA, electroacupuncture; WM, Western medicine; PNB, pudendal nerve block; SNS, self-neck-stretching exercises..

Of the total 14 studies, 2 did not describe in detail how acupotomy was performed. Among the remaining 12 articles, 11 described the method of acupotomy manipulation with various expressions, such as “release something,” “dissection,” “move up and down,” “relax,” “dredge,” “sweep,” “separate,” and “strip,” for synechotomy. However, Wang and Liu [15] stated that the experimental group took the “pricking” method not for synechotomy (Table 3).

All studies were evaluated with at least one index: up to seven indices were used. Ultimately, a total of 33 indices were used. The visual analog scale (VAS) was used as the evaluation index in 11 studies, and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) was used in three studies. Specifically, WOMAC was used in three of the four knee osteoarthritis studies, and the effective rate was used in another two studies involving tarsal tunnel syndrome and frozen shoulder (Table 4).

Table 4 . Evaluation index and results of acupotomy articles.

Disease	First author (y)	Evaluation index	Result
Knee
Knee osteoarthritis	Jun (2018) [5]	1. VAS 2. SF-MP 3. WOMAC	1. NS-D, S-E (p < 0.01), S-C (p < 0.01) 2. NS-D, S-E (p < 0.01), S-C (p < 0.01) 3. NS-D, S-E (p < 0.05), S-C (p < 0.05)
Knee osteoarthritis	Wang (2018) [6]	1. WOMAC 2. VAS	1. S-D (p < 0.05), S-E (p < 0.001), S-C (p-value not mentioned) 2. S-D (p < 0.001), S-E (p < 0.001), S-C (p < 0.001)
Knee osteoarthritis	Hua (2021) [7]	1. WOMAC 2. VAS	1. NS-D (p = 0.77), S-E (p < 0.01), S-C (p < 0.01) 2. NS-D (p = 0.78), S-E (p < 0.01), S-C (p < 0.01)
Knee osteoarthritis	Ding (2016) [8]	1. VAS 2. ADL 3. HSS	1. S-D (p < 0.01), S-E (p < 0.01), S-C (p < 0.01) 2. NS-D (p > 0.05) 3. S-D (p < 0.01), S-E (p < 0.01), S-C (p < 0.01)
Low back
Lumbar disk herniation	Jeong (2020) [10]	1. VAS 2. RMDQ 3. MMST 4. EQ-5D 5. CID 6. PGIC	1. S-D (p < 0.05) 2. S-D (p < 0.01) 3. S-D (p < 0.01) 4. NS-D 5. S-D (p < 0.05) 6. S-D (p < 0.05)
Ankylosing spondylitis	You (2020) [11]	1. VAS 2. ESR 3. CRP 4. mHHS 5. BASDAI 6. BASFI 7. ROM of hip joint	1. S-D (p < 0.05, compare experiment 1 group with experiment 2 and control groups), NS-D (p > 0.05, compare experiment 2 with the control group) 2. S-D (p < 0.05, compare experiment 1 group with experiment 2 and control groups), NS-D (p > 0.05, compare experiment 2 group with the control group) 3. S-D (p < 0.05, compare experiment 1 group with the control group), NS-D (p > 0.05, compare experiment 1 group with experiment 2 group), NS-D (p > 0.05, compare experiment 2 group with the control group) 4. S-D (p < 0.05, compare experiment 1 and 2 groups with the control group), NS-D (p > 0.05, compare experiment 1 with 2) 5. S-D (p < 0.05, compare experiment 1 group with experiment 2 and control groups), NS-D (p > 0.05, compare experiment 2 group with the control group) 6. S-D (p < 0.05, compare experiment 1 and groups with the control group), NS-D (p > 0.05, compare experiment 1 with 2) 7. S-D (p < 0.05, compare experiment 1 group with the control group about flexion and extension/abduction and adduction/rotation at extension/rotation at flexion of 90), S-D (p < 0.05, compare experiment 2 group with the control group about flexion and extension/abduction and adduction), NS-D (p > 0.05, compare experiment 1 group with 2 about all ROM, compare experiment 2 group with the control group about rotation at extension/rotation at flexion of 90)
Sacral nerve dysfunction	Zhang (2018) [9]	1. Anorectal pain VAS 2. Defecation disorder score 3. Anal incontinence score 4. Lumbar pain or soreness VAS 5. Abdominal pain and distension score VAS	1. S-D (p < 0.01), S-E (p < 0.05), S-C (p < 0.05) 2–5. NS-D (p > 0.05), S-E (p < 0.05), S-C (p < 0.05)
Ankle and foot
Tarsal tunnel syndrome	Fu (2020) [12]	1. CR 2. IVR 3. ER	1. S-D (p < 0.01) 2. S-D (p < 0.05) 3. S-D (p < 0.05)
Plantar fascilitis	Li (2014) [13]	1. VAS (at morning)	1. S-D (p < 0.01), S-E (p < 0.01), S-C (p < 0.01)
Neck
Chronic neck pain	Zheng (2014) [14]	1. VAS 2. NDI 3. PCS 4. MCS	1, 2. S-D (p < 0.001) 3. S-D (p < 0.05) 4. NS-D
Shoulder
Frozen shoulder	Wang (2021) [15]	1. VAS 2. ER	1. NS-D, S-E (p < 0.05), S-C (p < 0.05) 2. NS-D, S-E (100%, p-value not mentioned), S-C (100%, p-value not mentioned)
Wrist
Carpal tunnel syndrome	Zhang (2019) [16]	1. BCTQ 2. CSA of the median nerve 3. DML 4. CMAP 5. SNAP 6. SNCV	1–4, 6. S-D (p < 0.05), S-E (p < 0.05), S-C (p < 0.05) 5. NS-D, S-E (p < 0.05), S-C (p < 0.05)
Whole spine
Adolescent idiopathic scoliosis	Wei (2015) [17]	1. Cobb angle 2. AEMG ratio 3. Pulmonary function	1. S-D (p < 0.05), S-E (p < 0.05), S-C (p < 0.05) 2. Significantly increased (p < 0.05)/significantly decreased (p < 0.05) 3. Significantly increased (p < 0.05)/significantly decreased (p < 0.05)
Muscle
Myofascial pain syndrome	Ma (2010) [4]	1. VAS 2. PPT 3. ROM	1–3. S-D (compare experiment and control 1 with control 2, compare experiment with control 1) (p < 0.05), S-E (p < 0.05), S-C1 (p < 0.05), S-C2 (p < 0.05)

VAS, visual analog scale; SF-MP, Short-Form McGill Pain Questionnaire; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index; ADL, activities of daily living score; HSS, hospital for special surgery index; RMDQ, Roland Morris Disability Questionnaire; MMST, Modified-Modified Schober Test; EQ-5D, EuroQol Five Dimensions; CID, clinically important difference; PGIC, patient global impression of change; ESR, erythrocyte sedimentation rate; CRP, c-reactive protein; mHHS, modified Harris hip score; BASDAI, bath ankylosing spondylitis disease activity index; BASFI, bath ankylosing spondylitis functional index; ROM, range of motion; CR, cure rate; IVR, invalid rate; ER, effective rate; NDI, neck disability index; PCS, physical component score; MCS, mental component score; BCTQ, boston carpal tunnel questionnaire; CSA, cross-sectional area; DML, distal motor latency; CMAP, compound muscle action potential; SNAP, sensory nerve action potential; SNCV, sensory nerve conduction velocity; AEMG, average electromyogram; PPT, pressure pain threshold; NS-D, not statistically different between two groups after treatment; S-E, significant in the experiment group compared with before treatment; S-C, significant in the control group compared with before treatment; S-D, significant difference between two groups after treatment..

For knee diseases, all four studies were based on knee osteoarthritis, and VAS, and WOMAC were adopted as primary variables. Among the four studies, one reported the effectiveness of acupotomy [8], and one reported that acupotomy made no significant difference when compared with the control group receiving acupuncture treatment [5]. The remaining two single studies used acupotomy according to the meridian sinew and anatomical theories; in the experimental and control groups, the therapeutic effects reported by the two studies were contradictory [6,7].

In lower back disease, three single articles examined lumbar disk herniation [10], ankylosing spondylitis [11], and sacral nerve dysfunction [9]. Here, VAS was adopted as the primary outcome, and all three studies reported the effectiveness of acupotomy.

Regarding ankle and foot diseases, two studies analyzed tarsal tunnel syndrome [12] and plantar fascilitis [13]. One study did not clearly mention the primary outcome; in the other study, VAS was adopted as the primary outcome. Both studies reported the effectiveness of acupotomy.

In neck diseases, one study evaluated chronic neck pain: VAS was adopted as the primary outcome, and acupotomy was reported to be effective [14].

In shoulder diseases, one study examined frozen shoulder, and VAS was adopted as the primary outcome. Wang and Liu [15] compared the therapeutic effects of the pricking and dissection techniques of acupotomy in the treatment of coracoid pain in frozen shoulder and reported no significant difference between the experimental and control groups.

In wrist diseases, one study analyzed carpal tunnel syndrome. The boston carpal tunnel questionnaire result was adopted as the primary outcome, and acupotomy was reported to be effective [16].

In whole spine diseases, one study focused on adolescent idiopathic scoliosis. Wei et al. [17] did not clearly mention the primary outcome, but the results of the experimental group were more statistically significant in most evaluation indexes, including the Cobb angle, than those of the control group.

In muscle diseases, one study evaluated myofascial pain syndrome: VAS was adopted as the primary outcome, and acupotomy was reported to be effective [4].

Of the 14 studies, seven reported adverse effects, five reported the absence of adverse effects, and two did not mention adverse effects. Among the seven studies reporting adverse effects, six mentioned pain as an adverse effect. Although not mentioned in all studies, this side effect was also found in the control group. In addition, two studies reported subcutaneous bleeding and hemorrhage, and three single studies reported palpitation and discomfort, coldness and burning sensation, and somatic reactions, such as sweating. No significant difference in the type or incidence of adverse effects was found in the experimental and control groups. However, some studies did not report a specific group experiencing an adverse effect (Table 4).

The RoB was assessed for 14 RCTs using the Cochrane RoB tool. In “random sequence generation,” the studies using a table of random numbers and a randomized code showed a low RoB; however, two studies that did not have clear data showed an uncertain RoB. One study that was conducted according to the order of visits without using a random number table showed a high RoB [6]. In “allocation concealment,” studies that specified that the allocation procedure was performed by an independent researcher showed a low RoB, and two studies that did not have clear data showed an uncertain RoB [7,11]. In “blinding of participants and personnel,” owing to the characteristics of acupotomy, most studies were inevitably evaluated as having a high RoB. However, four studies had an uncertain RoB. Two studies used acupotomy in both the experimental and control groups, and as only the treatment sites differed, some blinding would be possible for the participants [6,7]. In the study conducted by Wang and Liu [15], both the experimental and control groups underwent acupotomy, and as only the manipulation methods were different, some blinding would be possible for the participants. In a study conducted by Zheng et al. [14], those who received acupuncture, or acupotomy were not included; thus, some blinding would be possible for the participants. In “blinding of outcome assessment,” six studies evaluated by independent researchers showed a low RoB; however, nine studies that did not have clear data showed an uncertain RoB. In “incomplete outcome data,” when analyzing the outcomes, an uncertain RoB appeared if the excluded participants or those who withdrew during the study were not considered. In “selective reporting,” all studies showed a low RoB. Finally, in “other bias,” due to a lack of data, it was not possible to accurately judge the bias risk (Figs. 2, 3).

Figure 2. Risk of randomized controlled trials bias summary.

Figure 3. Risk of randomized controlled trials bias.

By using various search terms for acupotomy, this study selected RCTs published between May 2014 and April 2022 and analyzed and summarized the research trends of acupotomy use by analyzing the published year, acupotomy terminology, sample size, disease type, retention time, and treatment sites, insertion depth and size, treatment frequency, and duration, type of acupotomy manipulation method, evaluation index, therapeutic effects, adverse effects, and RoB assessment.

The search for RCTs on acupotomy published in English or Korean between May 2014 and April 2022 identified a total of 14 studies: 1 (7.14%) in 2010, 2 (14.28%) in 2014, 1 (7.14%) in 2015, 1 (7.14%) in 2016, 3 (21.42%) in 2018, 1 (7.14%) in 2019, 4 (28.56%) in 2020, and 1 (7.14%) in 2021. Although RCTs are steadily conducted annually, a total of two RCTs were published in Korean journals in 2018 and 2020, and RCTs conducted in approximately 8 years are insufficient [5,12]. Considering the increasing interest in acupotomy and the frequency of its clinical use, further studies are urgently needed.

Because various terms are used to refer to acupotomy, many search terms were used in this study [1]. Regarding the terms of acupotomy used in the selected studies, 7 (49.98%) used “acupotomy,” 4 (28.56%) used “miniscalpel-needle,” 1 (7.14%) used “miniscalpel acupuncture,” 1 (7.14%) used “acupotomology,” and 1 (7.14%) used “needle-knife.” In addition, many acupotomy terms are used, and a study on clear definitions of acupotomy terms is expected, alongside further research on various forms and functions involved in acupotomy.

When classifying the selected studies based on the sample size, two studies had <50 participants, 7 had 50–100, and 4 had 100–200. As many aspects are involved in acupotomy, evaluating the effectiveness of clinical studies based on small samples is difficult; therefore, more reports involving a large sample size are expected.

All of the studies included in this review focused on musculoskeletal diseases. Among them, knee diseases were the most common (four studies), followed by lumbar diseases (three studies), and ankle diseases (two studies). Single studies concerning neck, shoulder, wrist, and muscular diseases were also included. Based on these results, acupotomy can be considered more effective and of greater interest in musculoskeletal diseases than other diseases. Therefore, research for acupotomy for diseases other than musculoskeletal diseases appears necessary.

Clinically, acupotomy does not have a retention time during operation; however, one of the included studies reported a retention time of 1 minute [4]. Thus, research on the therapeutic effects based on the difference in acupotomy retention times for specific diseases is also expected.

From the perspective of treatment sites, most studies consider anatomical structures to be treatment sites. As most studies used acupotomy for synechotomy, the anatomical structure was targeted. Some studies have employed acupotomy on acupoints; however, rather than entirely using the characteristics of acupoints, it is also necessary to consider that acupotomy is performed on acupoints near the disease area.

In the analysis of the insertion depth and size of the acupotomy, the size of the acupotomy was selected according to the anatomical location, particularly the disease type. The insertion depth was the deepest in lower back diseases. In all studies, the length, and insertion depth of the acupotomy were not mentioned precisely. If there is a clear anatomical structure target, it is explained as a specific bone touch rather than mentioning the insertion depth, and if it is difficult to accurately touch the target anatomical structure with the acupotomy, the insertion depth tends to be mentioned. The diameter of the acupotomy appears to be determined in various ways, depending on the treatment site and purpose of the treatment. Owing to the essential nature of the acupotomy, the relationship between a specific disease and the insertion depth and size of the acupotomy is crucial; therefore, future high-quality studies must accurately and clearly express the insertion depth and size of the acupotomy.

Regarding the frequency of acupotomy, in most studies, acupotomy was provided once a week. The treatment duration varied for each study and disease, which can be explained by the recovery period of synechotomy.

Various expressions were used to describe the manipulation method involved in acupotomy. In most of the reviewed studies, manipulations were made for synechotomy after inserting the acupotomy. However, Wang and Liu [15] used a manipulation method called “prick” not only for synechotomy but also for the comparison of the therapeutic effects of the pricking and dissection techniques of the acupotomy. In the future, after the acupotomy manipulation method is unified, a high-quality provincial study is expected to compare the therapeutic effects of the manipulation method of acupotomy for each disease. In addition, Wang and Liu [15] reported a comparative study on the effectiveness of the manipulation method in acupotomy; therefore, further studies on the therapeutic effects of the manipulation method in acupotomy are also expected.

As regards therapeutic effects, 11 of 14 studies reported the significance of acupuncture. Three studies reported the absence of a statistically significant difference. Jun et al. [5] did not find a statistically significant difference between the experimental group of patients who underwent acupotomy after knee osteoarthritis treatment and the control group receiving acupuncture treatment because the experimental group received treatment less frequently than the control group; thus, the effect of acupotomy can be more economical and less time consuming. Both Wang et al. [6] and Hua et al. [7] applied acupotomy in both the experimental and control groups, comparing the meridian sinew, and anatomical theories. The studies showed contradictory results, and Wang et al. [6] reported a statistically significant difference between the experimental group receiving acupotomy following the meridian sinew theory and the control group receiving acupotomy following the anatomical theory. Conversely, Hua et al. [7] did not find a statistically significant difference after treatment between the two groups. Nonetheless, according to Hua et al. [7], acupotomy based on the meridian sinew theory has comparable effectiveness to acupotomy based on the anatomical theory in the treatment of knee osteoarthritis. Considering this difference in the results, more research is needed. Wang and Liu [15] compared the therapeutic effects of pricking and dissection acupotomy techniques in the treatment of coracoid pain of frozen shoulder, and no statistically significant difference was found after treatment between the two groups. In other words, the pricking acupotomy technique had the same therapeutic effect on coracoid pain of the frozen shoulder as the dissection acupotomy technique.

Furthermore, the Cochrane RoB tool was used in the quality evaluation of the reviewed RCTs. RCTs that contained incomplete information or data were insufficiently mentioned had an uncertain RoB. In “blinding of participants and personnel,” most of the studies were evaluated as having a high RoB because of acupotomy-related characteristics. However, in the studies conducted by Hua et al. [7] and Wang et al. [6], acupotomy was performed in both the experimental and control groups; however, the RoB was considered low because only the treatment sites changed from the perspective of the meridian sinew and anatomy. In addition, in the study by Wang and Liu [15], both the experimental, and control groups received acupotomy; however, the RoB was considered low because only the acupotomy manipulation method was different. In the study by Zheng et al. [14], the sample group could not see the treatment process, so the RoB was also considered low because those who had previously received acupuncture or acupotomy were not included. In “blinding of outcome assessment,” outcome should be evaluated by independent researchers in many future studies to lower bias.

As limitations of this review, all relevant databases have not been searched, and the review did not cover all studies. In addition, the studies included are limited to those published in English or Korean; thus, the results of this study may not comprehensively cover the current research trends on acupotomy.

As the use of acupotomy in clinical practice is increasing, the need for high-quality academic research is also rising. However, as discussed in this study, many limitations are related to the literature search and the use of the research data due to the variety of terms used when describing acupotomy. Thus, standardizing acupotomy terms is necessary. More importantly, various terms are used, such as the synechotomy method for acupotomy; however, the exact procedure method has not been described. Thus, a unified reporting standard plan, such as the STRICTA of acupuncture, is needed. Therefore, high-quality acupotomy-related studies can be published only when these issues are addressed. This study aims to provide basic data to encourage researchers to publish improved studies considering recent trends in acupotomy research.

A total of 14 RCTs published between May 2014 and April 2022 were selected. The research trends of acupotomy were analyzed and summarized based on several items. However, because acupotomy-related terms varied, the standardization of terms, and unified reporting standard plan are necessary for future high-quality research.

Conceptualization: HJL. Data curation: JHH, HJL. Formal analysis: HJL. Investigation: JHH, HGL, JHK, ESH, GYC. Methodology: HJL, SHW, JHL, YKL, JSK. Project administration: HJL, SHW, JHL, YKL, JSK. Resources: SHC. Supervision: HJL, SHW, JHL, YKL, JSK. Validation: HJL. Writing – original draft: JHH, HJL. Writing – review & editing: JHH, HJL.

The authors have no conflicts of interest to declare.

None.

This research did not involve any human or animal experiment.