Warm-needle acupuncture for limb spasticity post-stroke:a systematic review

Qiong-Shuai Zhang, Yu Zhang, Guang-Cheng Ji, Jian-Nan Li, Xun Qi, Shan Xu, Jia-Zhen Cao, Xiao-Hong Xu＊, Bai-Lin Song＊

1Department of Acupuncture and Tuina, Changchun University of Chinese Medicine, Changchun 130117, China.2Department of Rehabilitation, The Second Affiliated Hospital of Heilongjiang University of Chinese Medicine, Harbin 150000, China.3Department of Rehabilitation, The Third Affiliated Hospital of Changchun University of Chinese Medicine, Changchun 130117, China.4Graduate School, Changchun University of Chinese Medicine, Changchun 130117, China.5Department of Traditional Chinese Medicine,Changchun University of Chinese Medicine, Changchun 130117, China.6Heart Rehabilitation Center of Department of Cardiology,Affiliated Hospital of Changchun University of Traditional Chinese Medicine,Changchun 130000,China.

Abstract

Keywords:Acupuncture, Warm-needle acupuncture, Limb spasticity, Stroke, Sensorimotor improvement,Activities of daily living

Background

Limb spasticity post-stroke (LSPS) is a common complication in patients with stroke [1] and is mainly manifested as increased muscle tension accompanied by pain and limited mobility [2-4].Studies have shown that spasticity occurs in approximately 24%-60% of post-stroke patients [5-8].LSPS seriously affects patients' quality of life and creates a heavy burden for patients and their families.At present,although physical therapy, oral or injection drugs, and other modern rehabilitation technologies are used to treat LSPS, the course of treatment is generally long and the curative effect is unsatisfactory.

Acupuncture and moxibustion, traditional medical treatments in China [9, 10], are now widely used for the rehabilitation of post-stroke patients.Warm-needle acupuncture, a type of acupuncture and moxibustion treatment that is used to treat many diseases, involves the fixing of moxa on the handle of the needles and igniting it after the needles are inserted into the human body.To date, an increasing number of studies have shown the efficient therapeutic effects of warm-needle acupuncture on LSPS [11, 12]; however, only one systematic review [13] has been completed and the literature in that review was published before 2016.

However, many new studies have emerged since 2016, which may affect the results and conclusions of the preview review, so an updated review is urgently needed.The results of the current systematic review provide evidence for clinicians for the use of warm-needle acupuncture in the positive treatment of LSPS.

Methods

Search strategy

We conducted a search of English and Chinese databases, including PubMed, Embase, the Cochrane Library, the Web of Science, the China National Knowledge Infrastructure, the Chinese Scientific and Journal Database, the Wanfang database, and the Chinese Biomedical Literature Database.We searched all the databases above from inception to May 21st,2020.The search strategy included aspects of stroke,spasticity, and warm-needle acupuncture and an example of search strategy for PubMed is shown in Table 1.

Eligibility criteria

We included studies according to the following criteria.

Study participants and control groups.We included study participants with a clear diagnosis of LSPS,according to the World Health Organization criteria[14]and those who had increased limb muscle tension and a Modified Ashworth Scale (MAS) score of 1-2.We placed no restrictions on patient age, gender, race,nationality, or other demographics and issues.Patients in the control group could use any type of treatment other than warm-needle acupuncture, including Chinese herbal medicine; western medicine such as baclofen, tizanidine, diazepam, dantrolene, or placebo;and modern rehabilitation therapies.

Intervention.The treatment of patients in the studies'experimental groups involved warm-needle acupuncture described as follows:during the retention of the needle in the human body,the needle is wrappedwith moxa or coated with one section of a moxa roll approximately 2 cm in length and then burned.The needle is withdrawn from the body after the moxa completely burns out and the ash is cleared.

Table 1 Search strategy for PubMed

Figure 1 Screening process.CNKI, China National Knowledge Infrastructure;VIP,Chinese Scientific and Journal Database;CBM,Chinese Biomedical Literature Database.

Outcomes.Primary outcomes:total clinical effective rate (total clinical effective rate = (number of cured cases + number of effective cases) / number of total cases), MAS [15-17], Clinic Spasticity Index (CSI),and clinical instruments mainly used for evaluating spasticity.Secondary outcomes:motor function and activities of daily living (ADLs) were measured using the Fugl-Meyer Assessment (FMA) [18, 19] and Barthel Index(BI)[20,21],respectively.

Study type.The studies included in the systematic review were randomized controlled trials(RCTs).

Exclusion criteria

We excluded the following from the systematic review:repetitive literature, case reports, comments, letters,systematic reviews, and experiments with animals;articles in which tuina was used in both the experimental and the control groups; and articles with data that cannot be used for analysis.

Data extraction

Two authors (JZC and XQ) used the pre-piloted forms to extract information, including first author, year of publication, treatments used in experimental and control groups, the total number in each group, the total number with positive results from each group,treatment times (TTs), outcomes, and other information(Figure 1).

Quality of the included studies

Two authors (GCJ and QSZ) independently assessed the methodological quality of the included studiesusing the Cochran Handbook 5.1.0 criteria [22] for judging the risk of bias.

Table 2 Characteristics of the studies included in the systematic review

Table 2 Characteristics of the studies included in the systematic review(continued)

Statistical analysis

We used pooled odds ratio (OR) with 95% confidence intervals (CIs) for dichotomous data and standard mean differences (MDs) with 95% CIs or weighted mean differences to analyze continuous variables to assess the effects of warm-needle acupuncture on LSPS.WhenI2＜50% andP＞0.1, indicating no or mild significant heterogeneity,we used the fixed-effect model; otherwise, we selected the random-effects model.Analysis was conducted using RevMan 5.3.5.

Results

From the databases we searched, we retrieved 393 studies published between 1993 and 2020.Two reviewers (YZ and JNL) who were independently working placed 393 articles in the Endnote software(Version 9.1,Clarivate Analytics,USA)and eliminated 151 duplicate entries.The reviewers then read the titles and abstracts of the remaining 242 studies to exclude the obvious ones that did not meet the criteria.Finally,they read the full texts to screen the qualified literature,finishing with 24 studies that met all the criteria; these studies were included in the systematic review.Disagreements in the screening process were arbitrated by a third reviewer (QSZ).Figure 1 shows the screening process.

Description of the included studies

The characteristics of the 24 studies included in the systematic review are shown in Table 2.These studies were published from 2008 to 2020.Sixteen studies[12,23-37] used a random-number table to perform randomization; the remainder did not describe the specific random method used.Nineteen studies [12,23-27, 30, 31, 33, 35-44] used warm-needle acupuncture alone in the experimental group and 5 studies [28, 29, 32, 34, 45] used warm-needle acupuncture and physical therapy.For the control groups, general acupuncture was used in 12 RCTs [12,24,26,29,31,33,36,37,40,42-44],physical therapy was used in 6 RCTs [28, 30, 32, 38, 41, 45],electroacupuncture was used in 3 RCTs [23, 35, 39]and the remaining 3 RCTs used general acupuncture and physical therapy [34], acupoint sticking [25], or electromagnetic radiation[27].

The sample sizes in the experimental and control groups ranged from 24 to 80 participants.Twenty-two RCTs [12, 23-29, 31-33, 35-45] reported outcomes using MAS, 20 RCTs [12, 23, 25, 27, 29, 30, 32-45]reported using FMA[17],RCTs[12,23,25-27,30-32,34-36, 39-43, 45] reported using BI, and 6 RCTs [24,25, 27, 28, 33, 34] reported using CSI.TTs ranged from 14 to 42 days.

Methodological quality of the included studies

The methodological quality of the studies was moderate.Sixteen studies [12, 23-37] used random-number tables to perform randomization, six studies used random-sequence generation but without describing the methods, and the remaining three studies used inpatient number or treatment type, which we considered to have a high risk of bias.Because warm-needle acupuncture is an obvious external treatment, it is hard to blind participants and personnel to its use; therefore, all included studies are at a high risk of bias(Figure 2).

Meta-analysis of outcomes

Total effective rate.Eleven studies[30-33,35,37,38,40, 41, 44, 45] with 880 enrolled participants reported the total effective rate of warm-needle acupuncture for LSPS.Heterogeneity was low (I2= 0%,P= 0.99).Therefore, we used the fixed-effects model for quantitative synthesis; the combined result was significant (OR = 3.61, 95% CI (2.51, 5.19),P＜0.001), indicating that warm-needle acupuncture has better effects than the treatments of the control groups,including general acupuncture, modern rehabilitation therapies,and electroacupuncture(Figure 3).

Figure 2 Risk-of-bias graph(A)and risk-of-bias summary(B)

Figure 3 Forest plot of the comparison between warm-needle acupuncture and control group treatments for a total effective rate

Figure 4 Forest plot of the comparison between experimental and control groups for Modified Ashworth Scale.Subgroup analysis was conducted according to treatment times ≤20 days and treatment times ＞20 days.

Figure 5 Forest plot of the comparison between experimental and control groups for Fugl-Meyer Assessment.Subgroup analysis was conducted according to treatment times ≤20 days and treatment times ＞20 days.

MAS subgroups with different treatment times.Fifteen studies used MAS to assess limb spasticity.Each of the 1,404 patients in these studies received 14-42 days of warm-needle acupuncture.Compared with the control groups, MAS scores increased in the warm-needle acupuncture group; the combined difference was significant (MD = -0.78, 95% CI(-1.00, -0.56),P＜0.001).Each patient in the 13 studies received more than 20 days of warm-needle acupuncture, whereas in only 2 studies, patients received fewer than 20 days of this treatment.Therefore, we conducted subgroup analysis.The results of a heterogeneity analysis of TTs ≤20 days showed thatI2= 54%, indicating moderate heterogeneity(P=0.14).For TTs ＞20 days,I2=94%,indicating high heterogeneity (P＜0.001).Therefore,warm-needle acupuncture appears to have better outcomes than the treatments in the control groups in decreasing limb spasticity.In addition, TTs ＞20 days has an even better effect(P＜0.001)(Figure 4).

Fugl-Meyer Assessment subgroups with different treatment times.Twenty-one studies [12, 23-27, 29,30, 32-40, 42-45] reported the FMA scores of 1,872 participants, 937 of whom received warm-needle acupuncture treatment.Scores increased in the treatment group as compared with those in the control group, and the combined difference was significant(MD = 9.51, 95% CI (6.99, 12.03),P＜ 0.001).Subgroup analysis was conducted based on TTs ≤20 days and TTs ＞20 days.Seven studies performed warm-needle acupuncture no more than 20 days,whereas 14 studies performed it more than 20 days.The meta-analysis of seven studies with TTs ≤20 days shows that heterogeneity of was moderate (I2= 61%,P= 0.02), while that of 14 studies with TTs ≤20 days is high (I2= 93%,P＜0.001) in another subgroup.The results of the two subgroups shows that TTs ＞20 days(MD = 11.71, 95% CI (9.38, 14.04),P＜0.001) can have a better effect than TTs ≤20 days (MD = 4.15,95% CI (2.06, 6.24),P= 0.001) in improving sensorimotor impairment(Figure 5)

Clinic Spasticity Index subgroups with different treatment times.Seven studies[24,25,27,28,31,33,34] used CSI to evaluate LSPS, with 275 patients in the experimental group receiving warm-needle acupuncture and 275 in the control group receiving other treatments.The combined difference was significant (MD = 1.06, 95% CI (0.51, 1.60),P=0.0002),indicating that warm-needle acupuncture is an effective treatment for LSPS (Figure 6).We further conducted subgroup analysis in terms of TTs ≤20 days and TTs ＞20 days.Five studies included TTs＜20 days and 2 included TTs ＞20 days.We used the random-effects model to perform subgroup analysis,and the results showed that the therapeutic effects of warm-needle acupuncture with TTs ≤20 days (MD =1.17, 95% CI (0.29, 2.04),P= 0.009) was better than TTs ＞20 days (MD = 0.85, 95% CI (0.13, 1.58),P=0.02).

Barthel Index subgroups with different treatment times.We conducted subgroup analysis for BI, which was reported in 18 studies [12, 23, 24, 26-28, 30-32,34-37, 39, 40, 42, 43, 45].The combined difference was significant (MD = 907, 95% CI (6.27, 13.12),P＜0.001), indicating that warm-needle acupuncture is an effective treatment for LSPS (Figure 7).Six studies included TTs＜20 days and 12 studies included TTs ＞20 days.We used the random-effects model to perform the analysis; the results showed that TTs ＞20 days(MD = 12.33, 95% CI (8.02, 16.65),P＜0.001) is better than TTs ≤20 days (MD = 4.88, 95% CI (3.27,6.49),P＜ 0.001), indicating that more TTs may increase the scores of ADLs of patients with LSPS.

Publication bias of total effective rate

The funnel plot comparing warm-needle acupuncture to other treatments is not symmetrical, indicating that publication bias may exist(Figure 8).

Discussion

Acupuncture and moxibustion have been widely used in China for the treatment and prevention of diseases for thousands of years, with each having good clinical effects on patients.Warm-needle acupuncture combines these two traditional Chinese treatments that are typically used for nervous, muscular, and joint diseases.

A previous systematic review [46] has shown that warm-needle acupuncture may be a promising intervention to reduce limb spasm; however, the study also pointed out that the conclusion was based on inconclusive evidence and that larger sample sizes and rigorously designed RCTs are needed.In addition, the effect of different TTs was not assessed.

In our new quantitative synthesis,the results showed that warm-needle acupuncture can effectively decrease limb spasticity and increase motor function as well as improve ADLs.To the best of our knowledge, we have performed the first subgroup analysis of TTs for warm-needle acupuncture for LSPS.TTs of more than 20 days may have a better curative effect,although the heterogeneity of this analysis is high.A growing number of recent studies have focused on warm-needle acupuncture.Studies [47-49] have shown that this treatment can accelerate the establishment of collateral circulation, relieve microvascular spasm caused by stroke, increase peripheral collateral blood flow into the ischemic area, and improve cerebral blood flow.The lighted moxibustion transfers heat along the needle body to the tip and through the acupuncture point directly to the muscle tissue, promoting blood vessel expansion and blood circulation, thereby improving local microcirculation and metabolism and relaxing local muscles to reduce spasticity.

This systematic review has some limitations.The quality of the studies included is this new review is moderate because eight studies did not describe or used inappropriate randomization methods.Significant heterogeneity may also be caused by researchers'inconsistent methods of measurement.In addition,because warm-needle acupuncture is an external treatment, it is hard to blind its use to participants and personnel.No studies have reported adverse events about the treatment,and none performed follow-ups.

Furthermore, all studies were conducted in China;no other countries were represented.The databases we searched were in English and Chinese; literature published in other languages was not included,possibly leading to publication bias.Therefore, the results of this meta-analysis should be treated with caution.

Figure 6 Forest plot of the comparison between the experimental and control groups using the Clinic Spasticity Index.Subgroup analysis was conducted according to treatment times ≤20 days and treatment times ＞20 days.

Figure 7 Forest plot of the comparison between the experimental and control groups using the Barthel Index.Subgroup analysis was conducted according to treatment times ≤20 days and treatment times ＞20 days.

Figure 8 Evaluation of publication bias of total effective rate

Conclusion

On the basis of the evidence of this systematic review and meta-analysis, we conclude that warm-needle acupuncture may be effective in treating LSPS compared with other interventions and that the effectiveness of TT ≥20 days may be better than that of＜20 days.However, owing to the poor methodological quality of the included literature, we strongly recommend that clinical trials should be ongoing and that their reports should comply with STRICTA guidelines[50].