Scoliosis is a 3-dimensional deformity in which the spine is curved or deflected laterally from the central axis of the body [1]. It is also characterized by rotational deformation of the vertebrae and loss of the normal curvature on the sagittal plane [1]. The causes of scoliosis are mostly unknown, and scoliosis cases in a clinic will most commonly be idiopathic [2].

The treatment for this condition is determined by the age of the patient, remaining growth period, type of scoliosis, magnitude of the curve, and patient’s own thoughts about the form of the back [3,4]. In Europe and the USA, wearing a brace is generally considered to be an acceptable and effective treatment for mild to moderate cases [5,6]. But there are many limitations associated with wearing a brace, such as diminished quality of life, physiological problems, decreased pulmonary functions, and developmental problems [7,8]. Spinal fusion is recommended when the angle of scoliosis is over 40° [9–11]. However, complications from such a surgical procedure, including spinal nerve damage, postoperative pain, and decreased pulmonary functions, [12] as well as reductions in the normal active range of movement have been reported [13].

Acupotomy is an acupuncture treatment where a bladed needle is used to detach soft tissue adhesion and restores tissue to its dynamic state. It is considered to be a combination of microsurgery and conventional acupuncture [14], and can be used for treating musculoskeletal disorders that have accumulated damage over a relatively long time [15].

Recently, clinical trials have been conducted on the use of acupotomy treatment for scoliosis, but their availability has not yet been systematically reviewed. Therefore, this systematic review evaluated the effectiveness and safety of acupotomy treatment for scoliosis.

Search strategy

Two reviewers independently performed a comprehensive literature search using multiple online databases (including the China National Knowledge Infrastructure, the WanFang databases, the China Science and Technology Journal Database, MEDLINE via PubMed, EMBASE via Elsevier, the Korea Citation Index, and the Research Information Service System), from their inception to March 17, 2021. The search terms included “scoliosis,” “acupotomy,” “acupotome,” “miniscalpel,” “needle knife,” and “acupuncture knife.”

Study selection

Randomized controlled trials (RCTs) and observational studies involving patients with scoliosis were included, regardless of patient sex, age, and race. We also included studies with scoliosis patients who had other spinal diseases, such as lumbar disc herniation.

In this review, acupotomy was defined as the use of an acupuncture needle with a flat blade at the tip of the needle [14]. We included studies using acupotomy as an experimental intervention, as well as those that used acupotomy combined with other treatments. Physical therapy was considered to include manipulation therapy such as Chuna, chiropractic therapy, spine manipulation, and exercise therapy such as Daoyin, active training, and resistance training, based on the Medical Subject Headings terms.

Types of outcomes

The primary outcome was the degree of scoliosis as measured by Cobb’s angle. The secondary outcomes were pain measured by the visual analog scale (VAS) score, and the McGill Pain Questionnaire score, and the functional outcomes as measured by the Oswestry Disability Index (ODI), and the total effective rate (TER). The TER is a processed outcome measure of certain assessment criteria, such as pain relief, improvements in physical signs, and reduction in the degree of Cobb’s angle. If other outcomes were reported, they were analyzed.

Data extraction

The extracted data included basic study information, such as the first author, publication year, informed consent, sample size, details of the participants, outcome measures, and results. Details of acupotomy, such as the needle type and treatment site based on the Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) guidelines were also extracted [17].

Quality assessment

The quality of the methodology of the studies was evaluated using the Cochrane risk-of-bias (RoB) tool [18]. In case of other sources of bias, the risk was evaluated according to the statistical homogeneity of participant characteristics at baseline. The RoB tool results are presented as a RoB graph and RoB summary using the Cochrane Collaboration’s software program Review Manager (RevMan), Version 5.3, for Windows (Copenhagen, The Nordic Cochrane Centre, 2012).

Data analysis

The RevMan software was used to create a quantitative synthesis of the included studies using the same intervention, comparisons, and outcome measures. For the continuous outcomes (Cobb’s angle, VAS, and pulmonary function index), results were presented as mean differences (MDs) with 95% confidence intervals. For the dichotomous outcomes (TER), results were presented as risk ratios (RRs) with a 95% confidence interval. Heterogeneity between the studies was evaluated using the I² statistic and chi-square test. An I² value less than 40% indicates that finding “may not be important;” 30–60%, “may represent moderate heterogeneity;” 50–90%, “may represent substantial heterogeneity;” or 75–100%, “may represent considerable heterogeneity” [19]. In cases of substantial, and considerable heterogeneity (I² ≥ 50%), a random-effects model was used, and a fixed-effects model was used when the I² value was

< 50% (fewer than 5 included studies).

Study search results

There were 223 studies retrieved, of which 159 remained following the removal of duplicates. There were 145 studies removed after reviewing titles and abstracts, and full texts of 14 studies were reviewed. In total, 12 RCTs [20–31] were included in this review, and 4 RCTs [21,22,27,28] were included in the meta-analysis (Fig. 1).

Characteristics of the included studies

All 12 of the included studies were published in China from 2011 to 2020: 11 were published in Chinese and 1 in English. These included 9 articles published in journals, 2 Master’s theses, and 1 conference abstract. The sample sizes of the studies ranged from 34 to 120 participants, and the mean ages of the participants in the studies ranged from 8.9 ± 0.6 years to 63.5 ± 5.3 years. The diagnostic tools used were reported by 4 studies [23,24,29,30] and the inclusion criteria used were reported for 10 studies [20–25,27,29–31]. In most studies, acupotomy was performed once a week [20,21,23,26,29–31] and was combined with physical therapy [21–31]. A total of 8 studies [21–26,30,31] performed manipulation therapy, 3 studies [26–28] performed exercise therapy, and 2 studies [24,29] performed traction therapy with acupotomy. The control interventions included prescribing medications, [20,23] manipulation therapy, [24,26] having participants wear a brace, [21,22,25,26,31], and traction therapy [24,27–30].

Five studies [23,25,27–29] reported VAS scores, 9 studies [21–23,25–29,31] reported Cobb’s angles, and 8 studies [20,24–28,30,31] reported TERs as outcomes. Additional outcomes included changes in the average electromyogram (AEMG) ratio of a surface electromyogram [21,22], pulmonary function index values [21,22], ODI scores [23], average height increases [25,26], McGill Pain Questionnaire scores [27], T1 pelvic angles [29], Nash–Moe spinal rotation degrees [29], and Japanese Orthopaedic Association scores [30]. The characteristics of the included studies are summarized in Table 1.

Methodological quality of included studies

Random sequence generation in 2 studies [27,29] had a low risk of bias due to the use of a computerized random number generator, and 1 study [31] had a high risk of bias due to the assignment of treatment based on the order of admission. Allocation concealment in only 1 study [21] had a low risk of bias due to the use of encoded, sealed, and opaque envelopes. No studies reported blinding for the participants and personnel. Blinding in a study is difficult when performing an acupotomy procedure, all studies were evaluated as having a high risk of bias. Only 1 study [29] reported blinding the outcome assessment. In the incomplete outcome data, 2 studies [21,22] had a high risk of bias because they did not explain their reason(s) for drop out. No studies documented selective reporting. In the evaluation of other sources of bias, all of the studies were evaluated as having a low risk of bias (Figs. 2 and 3). A total of 10 studies [20–25,27,29–31] received consent from patients, and 3 studies [22,27,31] received approval from an institutional review board.

Effectiveness of acupotomy as monotherapy

Wu [20] reported that patients receiving acupotomy treatment as compared with pharmacological treatment 3 weeks after surgery, exhibited significant improvements in the levels of pain (p < 0.01), muscle spasms (p < 0.01), tenderness and swelling in the lesion (p < 0.01), and extent of functional limitations (p < 0.01). In addition, the acupotomy treatment group had significantly higher TERs, as estimated by pain and heaviness, compared with the group who received pharmacology (p < 0.01).

Effectiveness of acupotomy as a combined therapy

Acupotomy combined with physical therapy

There were 4 studies [21,22,25,31] included in this review where acupotomy treatment combined with physical therapy was compared with wearing a brace. In a meta-analysis of 2 studies, [21,22] the combination of acupotomy and physical therapy was associated with a significantly greater reduction in the value of Cobb’s angle compared with wearing of a brace (MD, 1.19; 95% CI, 0.59–1.79; I² = 0%), as well as significant increases in pulmonary function index measurements such as vital capacity, forced expiratory volume in the 1^st second, divided by forced vital capacity, and maximal voluntary ventilation [MD, 5.94 (95% CI, 3.30–8.57; I² = 0%); MD, 4.92 (95% CI, 2.96–6.88; I² = 0%); and MD, 6.79 (95% CI, 4.01–9.58; I² = 0%), respectively; Figs. 4 and 5]. Wei et al [21] reported that the AEMG ratio decreased in the treatment group and increased in the control group following 12 months of treatment (p < 0.01). In a subsequent study, Wei et al [22] reported a significant reduction in the AEMG ratio in the treatment group and a significant increase in the AEMG ratio in the group who wore a brace after treatment (p < 0.01). Wei et al [22] further reported that Cobb’s angle was significantly reduced after 12 months in both groups (p < 0.05), but there was no significant difference between the 2 groups (p > 0.05). Following 24 months of treatment, the change in value of Cobb’s angle had increased in the acupotomy treatment group and had decreased in the group who wore a brace compared with the value at 12 months (p < 0.05). Nie et al [25] reported that the acupotomy treatment group, compared with the group who wore a brace, was associated with a significantly higher TER, estimated by Cobb’s angle (p < 0.05), a significantly reduced VAS score (p < 0.05). Quan [31] reported that the acupotomy treatment group, compared with the group who wore a brace, was associated with a significantly higher TER, estimated by pain (p = 0.007), a significant increase in the average height (p < 0.001), and a significant reduction in Cobb’s angle (p < 0.001).

Zheng [26] reported a significant improvement in patients receiving acupotomy with physical therapy compared with those wearing a brace and receiving physical therapy in terms of Cobb’s angle (p < 0.05) and TERs, estimated by pain and heaviness (p < 0.05). In addition, a significant increase in height was observed following treatment (p < 0.05).

There were 2 studies [27,28] comparing acupotomy combined with physical therapy and treatment with 4-dimensional traction. A meta-analysis revealed that the acupotomy treatment group was associated with a significant decrease in Cobb’s angle and VAS scores compared with the group who received traction [MD, −5.59 (95% CI, −6.87 to −4.30, I² = 0%); MD, −2.72 (95% CI, −2.91 to −2.54; I² = 86%), respectively; Figs. 6 and 7], and with higher TERs, estimated by Cobb’s angle (RR, 1.32; 95% CI, 1.11–1.58; I² = 0%; Fig. 8). Wang et al [27] also reported a significantly reduced ODI score in the acupotomy treatment group compared with the group who received traction (p = 0.018).

Sun et al [23] reported that patients receiving acupotomy with physical therapy had a significantly reduced Cobb’s angle compared with patients receiving pharmacological treatment (p = 0.022). However, no significant difference was observed in the scores for VAS (p = 0.160) or ODI (p = 0.937) between the 2 groups.

Zhang et al [30] reported that patients receiving acupotomy with physical therapy had significantly higher TERs, estimated by pain, Cobb’s angle, ability to work, and the straight leg raise test, compared with patients receiving combination therapy with intravenous injection, electroacupuncture, traction treatment, and manual therapy (p < 0.05).

Acupotomy combined with traction therapy

Jiang [29] reported that patients receiving acupotomy and traction therapy compared with traction therapy alone had a significantly reduced Cobb’s angle (p = 0.044), VAS score (p = 0.005), and T1 pelvic angle (p = 0.003). However, no significant difference was observed in the Nash–Moe spinal rotation degree between the 2 groups (p = 0.843).

Acupotomy combined with physical therapy and traction therapy

Huang et al [24] reported that patients receiving acupotomy treatment combined with physical therapy and traction therapy versus physical therapy and traction therapy alone, had a significantly higher TER, estimated by Cobb’s angle (p < 0.01).

Safety data

Adverse events were reported in 5 studies [20–22,28,29]. Wei [21] and Wei [22] looked for changes in vital signs, such as blood pressure, pulse, heart rate, and respiration, and for adverse events, such as fracture, dislocation, syncope, and hemorrhage. However, none of these symptoms were reported. Wu [20] looked for changes in vital signs, and there were no adverse events. The adverse events observed by Jiang [29] included a lidocaine skin allergy, muscle strain, and needle knife overreaction. During the entire research period, there were no serious adverse reactions. Min [28] reported 1 adverse event in the treatment group and 5 in the control group, but there were no reports on the criteria used for adverse events.

Analysis of acupotomy procedure

Analysis of information for acupotomy procedures is presented in Table 2. Three studies [21,22,26] used No. 2 Zhu Hanzhang-type acupotomy needles, and only 1 study [29] reported the length of the needle. One study [23] used 2 types of needles: a blade-shaped needle knife and a round-head needle knife. Two studies [30,31] reported only the number of the needle knife without mentioning the brand. The most frequently reported treatment site was the concave or convex side of the area affected by scoliosis [22,23,28,29], followed by the low back fascia [21,22,25], and the thoracolumbar junction [21,22,25]. Five studies [20–23,27] reported the insertion number, with the most frequent reports performing 5 to 7 insertions [21,22]. Seven studies [20,23,26,28–31] described their methods of stimulation.

Six studies [23,27–31] reported using texture or sensation to monitor performance of the correct procedure, and each reported on the loose texture [23,27–31] and clicking sound [28] felt by the practitioner. Only 2 studies [27,28] mentioned the needle retention time, all of which used a time of 1 minute. In 8 studies, [20–23,26,29–31] the frequency of treatment was reported as once a week. In each study, the total number of treatment sessions varied, with a minimum of 2 sessions [20] and a maximum of 10 sessions were reported [21,22].

There were 5 studies [20,23,24,29,30] that mentioned the use of anesthesia. Three studies [20,24,30] used lidocaine as an anesthetic, and in 1 study [30], lidocaine was injected with sodium bicarbonate, vitamin B12, and triamcinonide. There were no studies that reported the use of ultrasound guidance during the acupotomy procedure or the practitioner’s qualifications or level of experience.

A total of 223 studies were identified in the database search, from which 12 RCTs were selected. All the included studies were published in China from 2011 to 2020. Among the selected RCTs, only 1 study [20] was conducted using acupotomy treatment alone. In the physical therapy, manipulation therapy [21–26,30,31] was frequently used in combination with acupotomy treatment than exercise therapy [26–28]. Cobb’s angle was the outcome most frequently used to evaluate the effectiveness of the treatment in 9 studies [21–23,25–29,31], followed by TER in 8 studies [20,24–28,30,31]. Of the 8 studies using TERs as outcomes, 4 studies [20,26,30,31] used the improvement of symptoms, such as pain and heaviness, as an index, whereas 4 studies [24,25,27,28] used the reduction of Cobb’s angle as an index. Even in the studies that used the same index to evaluate TER, there were cases where the evaluation criteria were different.

All RCTs reported that the experimental group had significant improvements in most outcomes (Cobb’s angles, VAS scores, TER) compared with the control group. In a meta-analysis, when acupotomy treatment combined with physical therapy was compared with wearing a brace, Cobb’s angle was significantly reduced, and pulmonary function was significantly improved in the acupotomy treatment group, and when acupotomy treatment combined with physical therapy was compared with traction therapy, Cobb’s angles and VAS scores were significantly reduced in the acupotomy treatment group, and acupotomy was associated with higher TERs.

Of the 5 studies that mentioned adverse events, only 1 study [28] reported adverse events in both the treatment and control groups. Four studies reported no adverse events, such as fracture, dislocation, syncope, and changes in vital signs. Therefore, acupotomy appears to be a safe treatment. However, considering that only 5 of the 12 studies stated in the methods that they observed adverse events, the safety of acupotomy treatment for scoliosis could not be determined.

Based on the STRICTA guidelines, the most frequently used acupotomy needle was the No. 2 Zhu Hanzhang-type [21,22,26], and the concave and convex sides of the area affected by scoliosis were the most frequently reported treatment sites [22,23,28,29] A total of 7 studies [20,23,26,28–31] described stimulation methods, and 6 studies [23,27–31] reported the sensations to indicate performance of the correct procedure. Two studies [27,28] reported the retention time as 1 minute, and in most studies, an acupotomy was performed once a week [20,21,23,26,29–31] Five studies [20,23,24,29,30] reported the presence of anesthesia, and lidocaine was the most frequently used anesthetic. No study reported the use of an ultrasound guidance in the acupotomy procedure, nor did they mention the qualifications or experience of the practitioner.

Acupotomy may have a higher risk of adverse events compared with conventional acupuncture because the needle is thicker and a blade is attached to the tip of the needle. In addition, because acupotomy is more invasive, its safety and effectiveness may be affected by the practitioner’s anatomical knowledge, experience and skill. To compensate for this risk, the use of ultrasound as a guide has recently been reported [32–35]. Using ultrasound to guide treatment may improve the safety and effectiveness of acupotomy by allowing the practitioner to check the treatment site and the position of the needle knife [32].

As noted in the RoB assessment, 9 studies [20–26,28,30] did not report random sequence generation, 15 studies [20,22–31] did not report allocation concealment, 15 studies [20–28,30,31] did not reported blinding of outcome assessments, and none of the included studies used selective reporting.

The results of this review indicate that acupotomy may be an effective treatment for scoliosis. However, since participants were not blinded in all studies and some of the outcomes used for evaluation were subjective outcomes, the results may have been overestimated. In addition, most studies did not report their randomization method, so the possibility of bias in patient selection cannot be excluded.

The arrangement of the spine is closely related to the muscles surrounding it [36–38]. Acupotomy directly relieves chronic soft tissue adhesion and the abnormal tension of muscles and ligaments around the spine; it also improves microcirculation [39]. Following acupotomy treatment, the muscles and ligaments are relaxed and are easier to reset [31] which can result in reduced pain [34]. The use of acupotomy treatment at the concave side where there is tenderness can reduce muscle pressure and fascia tension on the side with the overload. Furthermore, acupotomy can help restore the mechanical balance of the spine and can be employed in combination with manual therapy or other methods of reduction [40,41].

This study has several limitations. Firstly, all included studies were conducted in China, therefore, the results cannot be applied to acupotomy treatment for scoliosis worldwide. However, acupotomy is mainly used in China. Secondly, the quality of most of the included studies methods was poor. It is possible that the study results were affected by the placebo effect or the intention of the researcher. Thirdly, the heterogeneity of the included studies was high, therefore, a comprehensive quantitative evaluation of all included studies was not possible. Fourthly, in most included studies, acupotomy was used as a component of combined interventions which does not allow examination of the therapeutic effects of using acupotomy as a single intervention. However, acupotomy was used mostly as a combination treatment, clinically. The acupotomy can reduce abnormal tension of the soft tissue, but in order to maintain and stabilize the shape of the spine, it is thought that other treatments are needed to increase the endurance and strength of the spinal muscles [36,42]. Finally, there were only 4 studies that reported both the criteria used to determine adverse events and observations of adverse events. In addition, the criteria used for adverse events in these studies were heterogeneous. Standardized reporting of adverse events is needed.

Considering these limitations, further studies with better methodological quality are needed to confirm the effectiveness and safety of acupotomy treatment for scoliosis. Moreover, due to the invasive nature of acupotomy, the qualifications and experience of the practitioner are important, and the use of imaging devices, such as ultrasound, is needed. In addition, it is recommended that a standardized system should be established for reporting adverse events of acupotomy by referring to the studies of Jiang et al [43] and Yoon et al [44] who systematically reviewed adverse events in acupotomy procedures.

Acupotomy may be helpful in the treatment of scoliosis but the results of this review should be interpreted carefully due to the low methodological quality of the included studies. In addition, there were only a few reports that included safety data, so even though there are safety concerns, a conclusion about the safety of acupotomy treatment for scoliosis could not be determined in this review. Therefore, further studies of a higher quality should be conducted to confirm the safety and effectiveness of acupotomy treatment for scoliosis.

Flow chart of this study.

Risk of bias graph.

Risk of bias summary.

+ = low risk of bias, ? = unclear risk of bias, - = high risk of bias.

Forest plots of correction degree of Cobb’s angle: Acupotomy and physical therapy versus wearing of brace.

PT, physical therapy.

Forest plots of pulmonary function index: Acupotomy and physical therapy versus wearing of brace

PT, physical therapy.

Forest plots of Cobb’s angle: Acupotomy and physical therapy versus traction therapy.

PT, physical therapy.

Forest plots of VAS score: Acupotomy and physical therapy versus traction therapy.

PT, physical therapy; VAS, visual analog scale.

Forest plots of total effective rate: Acupotomy and physical therapy versus traction therapy.

PT, physical therapy.

Characteristics of Included Studies.

Acupotomy Treatment Methods for Scoliosis.