Chinese herbal medicine for chronic neck pain due to cervical degenerative disc disease

Abstract

Background

Chronic neck pain with radicular signs or symptoms is a common condition. Many patients use complementary and alternative medicine, including traditional Chinese medicine, to address their symptoms.

Objectives

To assess the efficacy of Chinese herbal medicines in treating chronic neck pain with radicular signs or symptoms.

Search methods

We electronically searched CENTRAL (The Cochrane Library 2009, issue 3), MEDLINE, EMBASE, CINAHL and AMED (beginning to October 1, 2009), the Chinese Biomedical Database and related herbal medicine databases in Japan and South Korea (1979 to 2007). We also contacted content experts and handsearched a number of journals published in China.

Selection criteria

We included randomized controlled trials with adults with a clinical diagnosis of cervical degenerative disc disease, cervical radiculopathy or myelopathy supported by appropriate radiological findings. The interventions were Chinese herbal medicines, defined as products derived from raw or refined plants or parts of plants, minerals and animals that are used for medicinal purposes in any form. The primary outcome was pain relief, measured with a visual analogue scale, numeric scale or other validated tool.

Data collection and analysis

The data were independently extracted and recorded by two review authors on a pre‐developed form. Risk of bias and clinical relevance were assessed separately by two review authors using the twelve criteria and the five questions recommended by the Cochrane Back Review Group. Disagreements were resolved by consensus.

Main results

All four included studies were in Chinese; two of which were unpublished. Effect sizes were not clinically relevant and there was low quality evidence for all outcomes due to study limitations and sparse data (single studies). Two trials (680 participants) found that Compound Qishe Tablets relieved pain better in the short‐term than either placebo or Jingfukang; one trial (60 participants) found than an oral herbal formula of Huangqi ((Radix Astragali)18 g, Dangshen (Radix Codonopsis) 9 g, Sanqi (Radix Notoginseng) 9 g, Chuanxiong (Rhizoma Chuanxiong)12 g, Lujiao (Cornu Cervi Pantotrichum) 12 g, and Zhimu (Rhizoma Anemarrhenae)12 g) relieved pain better than Mobicox or Methycobal and one trial (360 participants) showed that a topical herbal medicine, Compound Extractum Nucis Vomicae, relieved pain better than Diclofenac Diethylamine Emulgel.

Authors' conclusions

There is low quality evidence that an oral herbal medication, Compound Qishe Tablet, reduced pain more than placebo or Jingfukang and a topical herbal medicine, Compound Extractum Nucis Vomicae, reduced pain more than Diclofenac Diethylamine Emulgel. Further research is very likely to change both the effect size and our confidence in the results.

Plain language summary

Herbal medicine for cervical degenerative disc disease

Significance of the review

Degenerative changes of the cervical spine are quite common and can cause severe neck pain, impairment and decreased quality of life. Degenerative disc disease of the cervical spine can result in severe pain, instability and radiculopathy (pain spreading down the arms and into the head), myelopathy (spasticity and weakness of arms or hands, which may include "numb and clumsy" hands) or both. Chinese oral and topical herbal medicines are being used to treat many neck disorders. Some have been tested in clinical trials. 

Description of the trials

Two Chinese oral herbal medications were tested in three randomized controlled trials that included 701 adults with chronic neck pain with radicular signs or symptoms or myelopathy. One oral herbal medication was compared with Mobicox (non‐steroidal anti‐inflammatory medication) and Methycobal (drug to reduce numbness, tingling in the arms), and the other (Compound Qishe Tablet) with placebo and Jingfukang. A topical herbal medicine (Compound Extractum Nucis Vomicae) was compared with Diclofenac Diethylamine Emulgel (non‐steroidal anti‐inflammatory medication).

Findings

Oral herbal medications may reduce neck pain more than placebo and Jingfukang. A topical herbal medicine (Compound Extractum Nucis Vomicae) also relieved neck pain in the short term (four weeks), but the trail had a high risk of bias.

Limitations

All four included studies were in Chinese and two of these studies were unpublished. Half of the trials had a low risk of bias, but they only tested the effects of short term use (up to eight weeks). The size of the studies was small. There is a need for trials with adequate numbers of participants that address the long‐term efficacy or effectiveness of Chinese herbal medicine compared to placebo. 

Conclusion

For chronic neck pain with or without radicular symptoms, there is low quality evidence that Compound Qishe Tablet is more effective than placebo for pain relief, measured at the end of the treatment. However, the size of the studies was small and the effect was measured in the short‐term. Further research is very likely to change both the effect size and our confidence in the results. There is a need for trials with adequate numbers of participants that address long‐term efficacy or effectiveness of herbal medicine compared to placebo.

Background

Description of the condition

Degenerative disc disease of the cervical spine can result in severe pain, instability, radiculopathy or myelopathy (Grob 1998), impairment and decreased quality of life. Cervical degenerative disc disease includes spondyloarthrosis, apophyseal joint osteoarthritis and disc degeneration. Degenerative changes increase from 5% to 10% in people between the ages of 20 and 30 years, to more than 50% by 45 years of age. Approximately 90% of men over the age of 50 and 90% of women over the age of 60 have radiographic evidence of cervical spondylosis (Murphy 2000). Population‐based data from Rochester, Minnesota, USA, indicate that cervical radiculopathy has an annual incidence rate of 107.3 per 100,000 for men and 63.5 per 100,000 for women, with a peak at 50 to 54 years of age (Radhakrisnan 1994). Data on the natural history of cervical radiculopathy are limited. In the same study from Rochester, USA, 26% of 561 patients with cervical radiculopathy underwent surgery within three months of the diagnosis (typically for the combination of radicular pain, sensory loss and muscle weakness), while the remainder were treated conservatively. Recurrence, defined as the re‐appearance of symptoms of radiculopathy after a symptom‐free interval of at least six months, was common and occurred in 32% of patients during a median follow‐up of 4.9 years (Heckmann 1999; Radhakrisnan 1994; Saal 1996; Sampath 1999). There are many conditions that can cause chronic neck pain. Degeneration of the cervical spine is the most common one. So, this review focuses on the chronic neck pain due to degenerative disc disease.

Description of the intervention

Conventional medical treatments for cervical degenerative disc disease are limited by their modest effectiveness (Wolff 2002). Conservative treatments include non‐steroidal anti‐inflammatory medication (NSAIDs), narcotics, muscle relaxants, physical therapy and transcutaneous electrical nerve stimulation (TENS). The main objectives of conservative treatments are to relieve pain, improve function and improve health‐related quality of life (Wolff 2002). Surgical treatment for cervical disc disease is indicated when symptoms are refractory to conservative treatments and neurological symptoms are progressive (Agrillo 1999). The goals of surgical treatment should be decompression, restoration of alignment and stability. However, many patients who undergo surgical treatment for cervical disc disease may continue to suffer from neck pain. In addition, many patients may not be good candidates for surgery because of their co‐morbidities, advancing age or preference for conservative treatment (Carette 2005). As a complementary and alternative medicine (CAM), herbal medicines have the potential to avoid the adverse effects of medications and surgery (Bensoussan 1996). Natural substances, including herbal medicines, have being used to promote healing and alleviate neck pain in western countries (Audette 2008). In China, many patients with cervical disc disease are increasingly turning to CAM, including herbal medicines, to alleviate their symptoms and reduce the side effects of medications. However, Chinese herbal medicines are not totally without side effects and some patients who take them might experience nausea, vomiting or dizziness.

Herbal medicines have been used for centuries in China. While in the strictest sense of the term, herbal medicine only includes plant materials (Gagnier 2006), the study of herbal medicine under a Chinese Herbal Medicine framework always includes the study of herbs, animal or insect products or minerals, since they are often combined in one formula. Herbal medicines are defined in this review as products derived from raw or refined plants or parts of plants (e.g. leaves, stems, buds, flowers, roots or tubers), minerals (e.g. borneol) and animals (e.g. prepared centipede or earthworm) and used for the treatment of disease. Synonyms for herbal medicines include: herbal remedies, herbal medications, herbal products, herbal preparations, herbal formulas, medicinal herbs and phytopharmaceuticals. There are four kinds of herbal therapies: i) single herb, ii) Chinese proprietary medicines, iii) mixtures of different herbs and iv) any one of the aforementioned three therapies plus Western pharmaceuticals, also known as integrative medicinal treatment (Liu 2006; Vickers 1999). Chinese proprietary medicines are usually based on well‐established and long‐standing recipes and formulated as tablets or capsules for commerce, convenience or palatability. The mixture of herbs prescribed by Chinese herbalists depends on the differentiation of symptoms according to the signs and symptoms and other disease information derived from the four examinations (observation, listening and smelling, inquiring, feeling the pulse and palpation) and analysed to detect the cause and location of the disease and the relationship between pathogenic factors and the vital energy. The disease is then diagnosed as a syndrome using an approach that is different from conventional Western medicine. Chinese herbal medicine is often combined with Western pharmaceuticals to produce an integrative medicinal treatment that may be more effective than either treatment alone (e.g. an injection of Chuanxiong (Rhizoma Chuanxiong), or rhubarb (Radix et Rhizoma Rhei), an injection of a mixture of Chuanxiong and Danshen (Radix Salviae Miltiorrhizae) or other herbal preparations plus Western medicine).

How the intervention might work

In the basic theory of traditional Chinese medicine (TCM), chronic neck pain with radicular signs or symptoms or myelopathy in cervical degenerative disc diseases is believed to be caused by the obstruction of Qi flowing and blood circulating in the neck area due to pathogenic wind, cold, and dampness invading the neck, inducing various symptoms of the neck, such as pain, numbness, cold feeling, etc. TCM doctors prescribe herbal formulae based on findings from the four examinations to deal with the pathogenic factors. Some active substances in herbal medicine can promote Qi flowing and blood circulating to relieve pain.

Why it is important to do this review

Given the heavy public health and economic burden that neck pain due to cervical degenerative disc disease causes and the fact that many sufferers consult CAM practitioners for their symptoms, a systematic review of these practices was justified.

Objectives

To assess the efficacy of Chinese herbal medicines/formulae in treating chronic neck pain with radicular signs or symptoms due to cervical degenerative disc disease.

Methods

Criteria for considering studies for this review

Types of studies

Randomized controlled trials (RCTs) were included, regardless of blinding, publication status or language. Randomized cross‐over trials were included only if the trial reported a wash‐out period.

Types of participants

Patients, aged 18 to 65 years old with a duration of symptoms of longer than six months, regardless of previous treatment, with a clinical diagnosis of cervical degenerative disc disease or cervical radiculopathy (pain along the cutaneous distribution of one or more cervical roots, which may include weakness and hyporeflexia), or myelopathy (spasticity and weakness in the upper limbs, which may include "numb and clumsy" hands), and supported by appropriate radiological findings.

Studies that included patients with symptoms that simulate radiculopathy but result from non‐spondylotic pathological changes, such as peripheral entrapment syndromes, brachial plexitis, herpes zoster, thoracic outlet syndrome, sympathetic mediated pain syndrome, intraspinal or extraspinal tumour, epidural abscess, cardiac ischemias, trauma, and infection were excluded. Studies that included patients with neurologic conditions that mimic cervical spondylotic myelopathy, such as peripheral polyneuropathy, motor neuron disease, multiple sclerosis, cerebrovascular disease, and syringomyelia were also excluded.

Types of interventions

All RCTs that examined Chinese herbal medicine as a treatment for neck pain due to cervical degenerative disc diseases were included. Comparisons included:

• Any Chinese herbal medicine/formula versus placebo or no treatment

• Any Chinese herbal medicine/formula versus other treatment

• Integrative medicinal treatment versus placebo or no treatment

• Integrative medicinal treatment versus Western medicine

Types of outcome measures

The outcome measures of interest were pain, functional and patient satisfaction, as recommended by the Cochrane Back Review Group (Deyo 1998; Furlan 2009) and were measured with validated quantitative tools. Short‐term outcomes were outcomes assessed at or before the three‐month follow‐up after randomization. Intermediate‐term outcomes were outcomes assessed between three and 12‐month follow‐up after randomization. Long‐term outcomes were outcomes assessed at 12 months or longer after randomization.

Primary outcomes

• Pain severity, measured with a visual analogue scale, numeric scale or other validated tool.

• Functional Status, measured by validated tools

• Patient Satisfaction

Secondary outcomes

• Neurologic Outcomes

• Adverse Events as reported in the trial.

Search methods for identification of studies

Electronic searches

We electronically searched CENTRAL (The Cochrane Library 2008, issue 2), MEDLINE , EMBASE CINAHL and AMED (from their beginning to April 15, 2008), the Chinese Biomedical Database and related herbal medicine databases in Japan and South Korea (1979 to 2007). See Appendix 1 for specific search strategies. These search strategies were adapted and used for the Japanese and South Korean databases.

Searching other resources

The following journals published in Chinese were handsearched:

• China Journal of Orthopaedics and Traumatology (1987‐2007)

• Chinese Journal of Traditional Medical Traumatology and Orthopedics (1993‐2007)

• Journal of Sichuan of Traditional Medicine (1982‐2007)

• New Journal of Traditional Chinese Medicine (1969‐2007)

• Shanghai Journal of Traditional Chinese Medicine (1955‐2007)

• The Journal of Traditional Chinese Orthopedics and Traumatology (1989‐2007).

We also handsearched Conference proceedings relevant to this topic, screened the reference lists of all included studies or relevant review articles, and contacted experts in the field to ensure we identified all potentially relevant published and unpublished material. We complimented our search by handsearching Chinese journals as many of these journals were not in electronic format.

Searches in CENTRAL (The Cochrane Library 2009, issue 3), MEDLINE , EMBASE CINAHL and AMED were updated on October 1, 2009.

Data collection and analysis

Selection of studies

The Cochrane Back Review Group Trials Search Co‐ordinator searched the English databases; one author (XC) conducted the literature searches in the Chinese databases and stored the search results (duplicates removed) in a reference bibliographic management system. Two review authors (XC, KT) independently screened the titles and abstracts to select potential references. Those that appeared to be relevant, or whose relevance could not be determined from the information available, were retrieved for further screening. The two review authors then independently read the selected papers and made a final selection decision. They made independent decisions at each step and discussed their decisions to reach consensus.They would have consulted a third author (YW) if disagreements persisted. However, there were no disagreements. The names of journals and authors were not masked in the identification phase as XC and KT were not familiar with articles in this area. The names of journals and authors were not masked in the data extraction phase, as the review authors were already familiar with these articles from the identification phase.

Colleagues agreed to search the Japanese database (OH) and the South Korean database (KDY). They screened the identified databases as above and had planned to complete the risk of bias assessment and data extraction had they found any studies that met the inclusion criteria, but they did not identify any.

Data extraction and management

The data described below were independently extracted and recorded by two review authors (XC, KT) on a data extraction form developed by a review team that has conducted a number of reviews on neck pain. This form was translated to Chinese and the information initially recorded in Chinese. After the data extraction process was completed, the extracted information was translated to English. We used consensus to make decisions. We would have consulted a third author (YW) if disagreements persisted. However, there were no disagreements. If the number randomized and the number analysed were inconsistent, we calculated and recorded the percentage lost‐to‐follow‐up. We planned to extract data to allow an intention‐to‐treat (ITT) analysis (the analysis should include all the participants in the groups to which they were originally randomly assigned). However, ITT analyses were not performed in any of the studies we found. For binary outcomes, we planned to record the number of participants experiencing the event in each group of the trial. For continuous outcomes, we planned to extract the arithmetic means and standard deviations for each group. If the data were reported using geometric means, we planned to extract the standard deviations on the log scale. Medians and ranges were extracted and reported if available. If necessary, we contacted the authors for additional information.

We collected these data:

(1) Characteristics of trials: date, location, period of data collection, year of publication, publication status, case definitions, sponsor of trial (commercial, academic, pharmaceutical, or unknown), definitions of cure and efficacy if available.

(2) Characteristics of participants : number of participants in each group, age (as described in the article: mean, median or range), numbers of participants with the main symptom if available.

(3) Characteristics of interventions: type, dose and duration of therapy.

(4) Characteristics of outcome measures : overall symptoms, adverse events resulting from the treatment (e.g. mortality, life threatening, toxic response, or anaphylaxis).

Two review authors (XC, KT) independently assessed the risk of bias of each relevant article, using the twelve criteria recommended by the Cochrane Back Review Group (Furlan 2009). See Appendix 2 for the definitions of the criteria. Criteria were scored yes (criteria met), no (criteria not met), or unsure (not enough information to make the decision). Articles that met six of the twelve criteria were considered to have a low risk of bias. We used consensus to make decisions and consulted a third author (YW) if disagreements persisted.

Clinical relevance was assessed using the five questions recommended by the Cochrane Back Review Group (Furlan 2009) and are outlined in Appendix 3.

As with other decisions, the clinical relevance of the studies was independently assessed by two review authors who used consensus to make final decisions and consulted with a third author if disagreements persisted.

Assessment of heterogeneity

Descriptive statistics were used to provide a summary description of the groups, interventions, outcomes, adverse effect of treatments and cost of care if available. All results reported were based on a small sample size. 

As the number of clinical trials included in the review is small, we did not test for potential publication bias with a funnel plot or other corrective analytical methods (Egger 1997).

There were insufficient data available to use quantitative analyses to summarize the data. However, we assessed the overall quality of the evidence for each outcome using an adapted GRADE approach, as recommended by the Cochrane Back Review Group (Furlan 2009). The quality of the evidence for a specific outcome was based on the study design, risk of bias, consistency of results, directness (generalizability), precision (sufficient data) and potential bias for the reporting of results across all studies that measured that particular outcome.

Qualitative Analysis of Trial Results

High quality: Further research is very unlikely to change our confidence in the estimate of effect, based upon consistent findings from at least two RCTs with low risk of bias and generalizable to the population in question. There are sufficient data with narrow confidence intervals. There are no known or suspected reporting biases.

Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate; one of the domains is not met.

Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate; two of the domains are not met

Very low quality: Great uncertainty about the estimate; three of the domains are not met.

No evidence: No evidence from RCTs.

Subgroup analysis and investigation of heterogeneity

As the review only included four trails, subgroup analyses could not be performed. Heterogeneity appeared in the sample group, the type of Chinese herbal formulae used and the outcomes measured. It was not possible to pool the results of these studies.

Results

Description of studies

Results of the search

We identified and screened 1061 citations and retrieved two RCTs (Li 2007; Mo 2005) for more detailed evaluation. Two other RCTs (Wang 2004; Wang 2005) were identified through consultation with experts. One of the review authors (YW) knew of experts conducting clinical trials in China. All four included studies were in Chinese and two of these studies were unpublished.

Included studies

We did not find any studies that included participants with acute or subacute neck pain. We included four trials (1,100 participants) on chronic degenerative cervical disc disease (confirmed by X‐ray) with radicular signs or symptoms or myelopathy. See Characteristics of included studies table for further details on study characteristics, interventions and outcomes.

Excluded studies

After reviewing the titles and the abstracts, 1057 reports were excluded as they were either not RCTs, or were not related to cervical degenerative disc disease, cervical radiculopathy, or cervical myelopathy. There were no studies excluded after reviewing the full report.

Two hundred and thirty unique citations were identified when the literature search was updated in October 2009. Two appeared to be potentially relevant but were excluded on closer examination because the intervention was incorrect.

Risk of bias in included studies

The twelve criteria recommended by the Cochrane Back Review Group Furlan 2009 were used to assess the risk of bias. The number of criteria met varied from 3/12 to 6/12 with a mean score of 4.5/12. However, only 50% of the studies (2/4) were considered to have a low risk of bias (Wang 2004; Wang 2005). None of the included studies had an adequate description of withdrawals and drop outs. None of the studies reported or appeared to use an intention‐to‐treat analysis (See Figure 1 for overview).

1.

Review authors' judgements about each risk of bias item for each included study.

Allocation

All the studies were described as randomized. However, the descriptions of allocation were ambiguous. Therefore, we contacted the first authors of these four articles and all but one (Li 2007) responded. We found that all of the studies fell short on the appropriateness of allocation concealment (Li 2007; Mo 2005; Wang 2004; Wang 2005).

Blinding

Double blinding was described in two trials (Wang 2004; Wang 2005). The placebo had a similar shape, smell, colour, and package as the active drugs. After the placebo and the active drugs were produced, a number was given according to the random number table by the statistics department, the Second Military Medical University (Wang 2004; Wang 2005). None of the participants (including patients, doctors and statisticians) knew which treatment the patients received.

Incomplete outcome data

There was no additional follow‐up except for the follow‐up at the end of treatment in two trials (Li 2007; Mo 2005). Four weeks after the end of treatment, 217/240 (Wang 2004) and 416/440 (Wang 2005) were still in the study. There was no report of an intention‐to‐treat analysis.

Selective reporting

Selective reporting was found in all of the four trials. Three trials reported gender breakdown of patients, average age and duration of symptoms with no range (Wang 2004; Wang 2005; Li 2007). One reported overall range, average age and gender breakdown of patient without the detailed information in each group (Mo 2005).

Other potential sources of bias

See Figure 1 for overview.

Effects of interventions

All four included trials assessed pain severity. Functional status and patient satisfaction were not assessed in all four trials. There were no continuous outcomes reported. We could not extract the arithmetic means and standard deviations for each group. The primary outcomes were expressed with an ordinal scale (NRS).

(Qishe versus placebo)

Pain relief (NRS) at short term follow‐up

There is one study (meeting 6/12 criteria) in this category. Wang 2004 studied chronic neck pain with radicular signs or symptoms. However, there were serious methodologic issues with this study, such as the lack of allocation concealment and ITT analysis.

There was low quality evidence (1 trial, 240 participants) that Compound Qishe Tablets are more effective for pain relief than placebo for patients with chronic neck pain, measured eight weeks after the end of treatment (Wang 2004). The quality of evidence was reduced because of study limitations and sparse data (single study).

(Huangqi versus Mobicox and Methycobal)

Pain relief (NRS) at short term follow‐up

There is one study with a high risk of bias in this category. Mo, et al (Mo 2005) compared Huangqi (prepared Radix Astragali 18 g, Dangshen, Radix Codonopsis 9 g, Sanqi, Radix Notoginseng 9 g, Chuanxiong, Rhizoma Chuanxiong 12 g, Lujiao, Cornu Cervi Pantotrichum 12 g, Zhimu, Rhizoma Anemarrhenae 12 g) to Mobicox and Methycobal and showed that Huangqi was significantly better than Mobicox and Methycobal at four weeks post randomization.

There is low quality evidence (1 trial, 60 participants) that Huangqi treatment (prepared Radix Astragali 18 g, Dangshen, Radix Codonopsis 9 g, Sanqi, Radix Notoginseng 9 g, Chuanxiong, Rhizoma Chuanxiong 12 g, Lujiao, Cornu Cervi Pantotrichum 12 g, Zhimu, Rhizoma Anemarrhenae 12 g) is more effective for pain relief than Mobicox and Methycobal for patients with chronic neck pain, measured at the end of the treatment (Mo 2005). The quality of evidence was reduced because of study limitations and sparse data (single study).

Qishe and placebo Jingfukang versus Jingfukang and placebo Qishe

Pain relief (NRS) at short term follow‐up

There is one study (meeting 6/12 criteria) in this category. Wang et al (Wang 2005) compared Compound Qishe Tablet plus placebo Jingfukang to Jingfukang (a Chinese proprietary medicine for neck pain, composed of Qianghuo (Rhizoma Notopterygii), Chuangxiong (Rhizoma Chuanxiong), Gegen (Radix Puerariae), Qinjiao (Radix Gentianae Macrophyllae), Weilingxian (Radix Clematidis), Cangzhu (Rhizoma Atractylodis), Danshen (Radix Salviae Miltiorrhizae), Baishaoyao (Radix Paeoniae Alba), Dilong (Lumbricus), Honghua (Flos Carthami), Ruxiang (prepared Olibanum), and Huangqi (Radix Astragali) and placebo Qishe and showed that Compound Qishe Tablet treatment with placebo Jingfukang was significantly better than Jingfukang with placebo Qishe, measured at eight weeks after the end of treatments. However, there were serious methodologic issues with this study such as the lack of allocation concealment and ITT analysis.

There was low quality evidence (1 trial, 440 participants) that Compound Qishe Tablet was more effective for pain relief than Jingfukang for patients with chronic neck pain, measured eight weeks after the end of the treatment (Wang 2005). The quality of evidence was reduced because of study limitations and sparse data (single study).

Integrative medicinal treatment versus placebo or no treatment

There were no studies identified for this comparison.

(Extractum Nucis Vomicae versus Diclofenac Diethylamine Emulgel)

Pain relief (NRS) at short term follow‐up

There is one study with a high risk of bias in this category. Li et al (Li 2007) compared Extractum Nucis Vomicae to Diclofenac Diethylamine Emulgel and showed that Extractum Nucis Vomicae was significantly better at relieving pain than Diclofenac Diethylamine Emulgel at the end of four weeks treatment.

There is low quality evidence (1 trial, 360 participants) that Extractum Nucis Vomicae is more effective for pain relief than Diclofenac Diethylamine Emulgel for patients with chronic neck pain, measured at the end of the treatment (Li 2007). The quality of evidence was reduced because of study limitations and sparse data (single study).

Adverse effects/Cost of care

Adverse effects were reported in all four studies (Li 2007; Mo 2005; Wang 2004; Wang 2005) and included watery stool, abdominal pain, stomach ache, pruritus, reddish skin, small blisters on the skin. (See Characteristics of included studies table for full details). No life threatening adverse effects were noted in these studies.

Cost of care was only discussed in one of the studies (Mo 2005). The cost of herbal treatment was RMB (Chinese currency) 232.9 to 625.4, with an average of RMB 376.07±115.32, and Mobicox and Methycobal RMB 242.4 to 475.8, with an average of RMB 375.25±66.33.

Clinical relevance

Clinical relevance was assessed using the five questions recommended by the Cochrane Back Review Group (Furlan 2009). The results are displayed in Table 1. Many of the criteria were not met. The only category met by all four studies was the descriptions of the treatments. Only two studies described the patients in sufficient details. However, the details were not specific enough in terms of the nerve roots involved.

1. Assessment of Clinical Relevance of the RCTs.

Questions	Li 2007	Mo 2005	Wang 2004	Wang 2005
1. Are the patients described in detail so that you can decide whether they are comparable to those that you see in your practice?	No	No	Yes	Yes
2. Are the interventions and treatment settings described well enough so that you can provide the same for your patients?	Yes	Yes	Yes	Yes
3. Were all clinically relevant outcomes measured and reported?	No	No	No	No
4. Is the size of the effect clinically important?	No	No	No	No
5. Are the likely treatment benefits worth the potential harms?	No	No	No	No

Discussion

Summary of main results

This review was only able to report on short‐term effects of Chinese herbal formulae since pain relief was measured either immediately after the end of the treatment or four to eight weeks later. However, all the studies reported that the herbal medicine tested were better than the comparison treatment. There was only one study comparing herbal medicine to placebo (Wang 2004); the others compared it to other medications or combinations of other medications. We planned to assess the evidence of Chinese Herbal formulae on function, patient satisfaction or neurological outcomes but this information was not available from the four studies.

Overall completeness and applicability of evidence

All four included studies were in Chinese and two of these studies were unpublished. Thus, the results may not be generalizable to western countries. One of the challenges faced in the review was searching for articles. Our first round of searching the electronic databases did not identify any studies. A subsequent update yielded only one study through the conventional database after reviewing the titles and abstracts of over 900 articles. Only one study was found through the Chinese Biomedical database after spending a considerable amount of effort. After consulting with many experts who conducted clinical trials in this area (by YW), only two further papers were obtained. Our experience in this review indicates that better organization of published articles (such as including Chinese journals in electronic format, using Roman characters for Chinese words) is required for the field of complementary medicine in non‐English databases. Many hours were spent searching by hand with no yield. It also indicated to us that this is a wide open area of potential research as many articles were found, but only a few RCTs. We complimented our search in the Chinese databases as many journals were not in electronic format. Items for clinical relevance were poorly reported by these studies, which raises serious doubts about the applicability of their results to clinical practice.

Quality of the evidence

On reviewing the four articles, we found that in general, the write‐up was poor. The authors did not follow the CONSORT statement in reporting the results of their clinical trials. Although the CONSORT statement has been translated, it is not well known in the Chinese research community and better promotion is required to enhance the quality and the write‐up of the research.

The most difficult challenge faced in this review was clinical heterogeneity. The heterogeneity appeared at many levels, such as in the sample group, the type of herbal medicine used and the outcomes measured. It was not possible to pool the results of these studies. Another challenge was the lack of data for calculating an effect size for even a single study.

Weaknesses of this review rest with limitations in the primary studies. We were unable to make any firm statements about the quality of the evidence, since the therapies were only reported in a single trial. We were not able to identify in any of the two studies in which there were drop‐outs reported whether intention‐to‐treat analysis was used. In reviewing many of the articles in this field in Chinese, this concept does not appear to be familiar to the researchers in this area. Furthermore, details of drop‐outs were not reported in any of the studies.

Potential biases in the review process

Our approach to summarizing the literature has several strengths. We used a comprehensive, librarian‐assisted search of multiple databases. We used a team of international healthcare professionals to decide on article relevance and assess risk of bias. We avoided any professional bias inherent in having a single profession evaluate its literature.

Agreements and disagreements with other studies or reviews

Due to limited RCTs in herbal medicine treating chronic neck disorders, there is no similar systematic review in the field.

Authors' conclusions

Implications for practice.

For chronic neck pain, with or without radicular symptoms, there is low quality evidence that Chinese herbal medicine is more effective than placebo for pain relief, measured at the end of the treatment. However, there was only one study (240 participants) and the effect was only measured in the short term. The effect size was small and it was not clinically relevant. Therefore, further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change it.

Implications for research.

There is a need for trials with adequate sample size that address the long‐term efficacy or effectiveness of Chinese herbal medicine compared to placebo.

History

Protocol first published: Issue 2, 2007
 Review first published: Issue 1, 2010

Date	Event	Description
10 June 2008	Amended	converted to new review format

Appendices

Appendix 1. Electronic database searches

MEDLINE 
 1. exp "Clinical Trial [Publication Type]"/ 
 2. randomized.ab,ti. 
 3. placebo.ab,ti. 
 4. dt.fs. 
 5. randomly.ab,ti. 
 6. trial.ab,ti. 
 7. groups.ab,ti. 
 8. or/1‐7 
 9. Animals/ 
 10. Humans/ 
 11. 9 not (9 and 10) 
 12. 8 not 11 
 13. exp Cervical Vertebrae/ 
 14. exp Intervertebral Disk/ 
 15. exp Neck/ 
 16. exp Neck Pain/ 
 17. exp Nerve Compression Syndromes/ 
 18. exp Spinal Cord Compression/ 
 19. exp Spinal Cord Diseases/ 
 20. exp Spinal Diseases/ 
 21. cervical$.tw. 
 22. neck$.tw. 
 23. spinal$.tw. 
 24. spine.tw. 
 25. spondyl$.tw. 
 26. (nerve adj compression adj syndrome$).tw. 
 27. or/13‐26 
 28. alternative medicine.mp. 
 29. alternative medicine.tw. 
 30. botanical.tw. 
 31. (Chinese adj herbal).mp. 
 32. Chinese herbal drugs.tw. 
 33. Chinese medicine.tw. 
 34. complementary medicine.mp. 
 35. complementary medicine.tw. 
 36. exp Complementary Therapies/ 
 37. exp Drugs, Chinese Herbal/ 
 38. exp Drugs, Non‐Prescription/ 
 39. medicinal herbs.mp. 
 40. exp Medicine, Chinese Traditional/ 
 41. exp Medicine, Herbal/ 
 42. exp Plants, Medicinal/ 
 43. exp Medicine, Oriental Traditional/ 
 44. exp Medicine, Traditional/ 
 45. herb$.tw. 
 46. herbal drugs.tw. 
 47. herbal medicine.tw. 
 48. herbal preparations.tw. 
 49. herbs medicinal.tw. 
 50. phytomedicine.tw. 
 51. plants extract.tw. 
 52. plants medicinal.tw. 
 53. traditional Chinese medicine.tw. 
 54. traditional medicine.tw. 
 55. or/28‐54 
 56. 12 and 27 and 55

EMBASE 
 1. Clinical Article/ 
 2. exp Clinical Study/ 
 3. Clinical Trial/ 
 4. Controlled Study/ 
 5. Randomized Controlled Trial/ 
 6. Major Clinical Study/ 
 7. Double Blind Procedure/ 
 8. Multicenter Study/ 
 9. Single Blind Procedure/ 
 10. Phase 3 Clinical Trial/ 
 11. Phase 4 Clinical Trial/ 
 12. crossover procedure/ 
 13. placebo/ 
 14. or/1‐13 
 15. allocat$.mp. 
 16. assign$.mp. 
 17. blind$.mp. 
 18. (clinic$ adj25 (study or trial)).mp. 
 19. compar$.mp. 
 20. control$.mp. 
 21. cross?over.mp. 
 22. factorial$.mp. 
 23. follow?up.mp. 
 24. placebo$.mp. 
 25. prospectiv$.mp. 
 26. random$.mp. 
 27. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).mp. 
 28. trial.mp. 
 29. (versus or vs).mp. 
 30. or/15‐29 
 31. 14 or 30 
 32. human/ 
 33. Nonhuman/ 
 34. exp ANIMAL/ 
 35. Animal Experiment/ 
 36. 33 or 34 or 35 
 37. 32 not 36 
 38. 31 not 36 
 39. 37 and 38 
 40. 38 not 39 
 41. exp Cervical Spine/ 
 42. exp Intervertebral Disk/ 
 43. exp Neck/ 
 44. exp Neck Pain/ 
 45. exp Nerve Compression/ 
 46. exp Spinal Cord Compression/ 
 47. exp Spinal Cord Disease/ 
 48. exp Spinal Disease/ 
 49. cervical$.tw. 
 50. neck$.tw. 
 51. spinal$.tw. 
 52. spine.tw. 
 53. spondyl$.tw. 
 54. (nerve adj compression adj syndrome$).tw. 
 55. or/41‐54 
 56. alternative medicine.mp. 
 57. alternative medicine.tw. 
 58. botanical.tw. 
 59. (Chinese adj herbal).mp. 
 60. Chinese herbal drugs.tw. 
 61. Chinese medicine.tw. 
 62. complementary medicine.mp. 
 63. complementary medicine.tw. 
 64. exp alternative medicine/ 
 65. exp Non Prescription drug/

CENTRAL 
 1. MeSH descriptor Neck, this term only in MeSH products 
 2. MeSH descriptor Neck Injuries explode all trees in MeSH products 
 3. MeSH descriptor Whiplash Injuries, this term only in MeSH products 
 4. MeSH descriptor Neck Muscles, this term only in MeSH products 
 5. MeSH descriptor Neck Pain, this term only in MeSH products 
 6. neck in All Fields, from 1800 to 2005 in all products 
 7. whiplash in All Fields in all products 
 8.. MeSH descriptor Spine explode all trees in MeSH products 
 10. MeSH descriptor Spinal Diseases explode all trees in MeSH products 
 11. (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10) 
 12. alternative medicine in All Fields from 1800 to 2006 in all products 
 13. Chinese herbal drugs in All Fields from 1800 to 2006 in all products 
 14. Chinese medicine in All Fields from 1800 to 2006 in all products 
 15. complementary medicine in All Fields from 1800 to 2006 in all products 
 16. non prescription drugs in All Fields from 1800 to 2006 in all products 
 17. herb in All Fields from 1800 to 2006 in all products 
 18. medicinal plants in All Fields from 1800 to 2006 in all products 
 19 .(#12 or #13 or #14 or #!5 or #16 or #17 or #18) 
 20. (#11and #19)

CBMdisc on Chinese Biomedical Literature 
 1. exp Randomized Controlled trials / all subheadings 
 2. exp Random Allocation 
 3. Random* 
 4. Control* 
 5. or/ 1‐4 
 6. exp traditional Chinese medicine/ all subheadings 
 7. exp integrated Chinese and western medicine/ all subheadings 
 8. or/ 6‐7 
 9. degenerative cervical disc diseases 
 10. Human 
 11. 5 and 8 and 9 and 10

Appendix 2. Criteria for a judgment of 'yes' for the sources of risk of bias  

1. Was the method of randomisation adequate? A random (unpredictable) assignment sequence. Examples of adequate methods are computer‐generated random numbers table and use of sealed opaque envelopes. Methods of allocation using date of birth, date of admission, hospital numbers, or alternation should not be regarded as appropriate.

2. Was the treatment allocation concealed? Assignment generated by an independent person not responsible for determining the eligibility of the patients. This person has no information about the persons included in the trial and has no influence on the assignment sequence or on the decision about eligibility of the patient.

3. Was the patient blinded to the intervention? The review author determines if the treatments were considered equally credible and acceptable to patients in order to score a "yes."

4. Was the care provider blinded to the intervention? The review author determines if enough information about the blinding is given in order to score a "yes."

5. Was the outcome assessor blinded to the intervention? The review author determines if enough information about the blinding is given in order to score a "yes."

6. Was the drop‐out rate described and acceptable? The number of participants who were included in the study but did not complete the observation period or were not included in the analysis must be described and reasons given. If the percentage of withdrawals and drop‐outs does not exceed 20% for immediate and short‐term follow‐ups, 30% for intermediate and long‐term follow‐ups and does not lead to substantial bias a "yes" is scored.

7. Did the analysis include an intention‐to‐treat analysis? All randomized patients are reported/analysed in the group to which they were allocated by randomisation for the most important moments of effect measurement (minus missing values), irrespective of noncompliance and co‐interventions.

8. Are reports of the study free of suggestion of selective outcome reporting? This item was scored "yes" if all the results from all pre‐specified outcomes were adequately reported. 

9. Were the groups similar at baseline regarding the most important prognostic indicators? In order to receive a "yes," groups have to be similar at baseline regarding demographic factors, duration and severity of complaints, percentage of patients with neurological symptoms, and value of main outcome measure(s).

10. Were co‐interventions avoided or similar? Co‐interventions should either be avoided in the trial design or be similar between the index and control groups.

11. Was the compliance acceptable in all groups? The review author determines if the compliance to the interventions is acceptable, based on the reported intensity, duration, number and frequency of sessions for both the index intervention and control intervention(s).

12. Was the timing of the outcome assessment in all groups similar? Timing of outcome assessment should be identical for all intervention groups and for all important outcome assessments.

Appendix 3. Questions to determine if results are clinically relevant

1. Are the patients described in detail so that you can decide whether they are comparable to those that you see in your practice? 
 2. Are the interventions and treatment settings described well enough so that you can provide the same for your patients? 
 3. Were all clinically relevant outcomes measured and reported? 
 4. Is the size of the effect clinically important? 
 5. Are the likely treatment benefits worth the potential harms?

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Li 2007.

Methods	RCT Number Analyzed/Randomized: 360/360 Index treatment: N =185 Comparison treatment: N =175 Power Analysis: not reported Funding Source: not reported
Participants	Chronic neck pain with radicular signs or symptoms Patients recruitment: out‐patients from the Traditional Chinese Medicine Hospital of Shenzhen Index treatment: 50.8±9.3 year old; 96 male, 89 female; Duration of pain: 18.4±5.1 months. Comparison treatment: 48.6±10.2 year old; 90 male, 85 female; Duration of pain: 17.8±5.9 months. DDD confirmed by X‐ray
Interventions	INDEX TREATMENT: Compound Extractum Nucis Vomicae made in the Shenzhen Hospital of Traditional Chinese Medicine, composed of Maqianzi(Semen Strychni), Chanshu(Venenum Bufonis), Honghua(Flos Carthami), Chuanwu(Radix Aconiti), and Tiannanxing(Rhizoma Pinelliae seu Arisaematis). Extractum Nucis Vomicae ointment was applied to the painful area of the neck, 3 times daily, for 4 weeks. (Dose not reported) COMPARISON TREATMENT: Diclofenac Diethylamine Emulgel 3 times a day for 4 weeks (Dose and composition not reported) Treatment Schedule: for 4 weeks Duration of Follow‐up: 4 weeks after randomization
Outcomes	PAIN RELIEF: (Numerical Rating Scale, 0 to 3) Baseline mean: Compound Extractum Nucis Vomicae 2.73±0.52 for neck pain, and 2.42±0.60 for radicular pain; control 2.58±0.62 for neck pain, and 2.50±0.57 for radicular pain End of study mean: Compound Extractum Nucis Vomicae 0.59±0.14 for neck pain, and 0.61±0.17 for radicular pain; control 0.86±0.21 for neck pain, and 0.88±0.23 for radicular pain Absolute Benefit: Compound Extractum Nucis Vomicae 2.14 for neck pain, and 1.81 for radicular pain; control 1.72 for neck pain, and 1.62 for radicular pain Reported Results: Compound Extractum Nucis Vomicae treatment was significantly better than Diclofenac Diethylamine Emulgel at 4 weeks. Reason for Dropouts: No drop‐outs Adverse Effects: Herbal treatment, 14 cases had pruritus, 9 cases had reddish skin, and among them 2 cases had small blister on the skin. Diclofenac Diethylamine Emulgel, 10 cases had pruritus, and 6 cases had reddish skin. Cost of Care: not reported
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	Sequence was generated from a random table. This was confirmed in the author's reply to our survey about their allocation concealment
Allocation concealment?	High risk	This was not described and we verified with the author through our survey that this was not done.
Blinding? patients blinded ‐ All outcomes	High risk	Not reported
Blinding? Careproviders blinded ‐ All outcomes	High risk	Not reported
Blinding? Outcome assessor blinded ‐ All outcomes	High risk	Not reported
Incomplete outcome data addressed? All outcomes ‐ ITT analysis?	High risk	Not reported
Incomplete outcome data addressed? All outcomes ‐ drop‐outs?	Low risk	No drop‐outs
Free of selective reporting?	Unclear risk	Information such as age range and duration of symptoms missing
Similarity of baseline characteristics?	Low risk	reported to be similar
Co‐interventions avoided or similar?	Unclear risk	Unclear whether co‐interventions were avoided from the description in the article
Compliance acceptable?	Unclear risk	Not assessed
Timing outcome assessments similar?	Low risk	all at 4 weeks after randomization

Mo 2005.

Methods	RCT Number Analyzed/Randomized: 60/60 Index treatment: N = 30 Comparison treatment: N = 30 Power Analysis: not reported Funding Source: Supported by Shanghai Health Board
Participants	Chronic neck disorder with radicular signs or symptoms, or with spasticity and weakness in the upper limbs Patients recruitment: From the special department of neck pain of the hospital Age range: 18 to 65 year old with a mean of 51.97 year old. There were 40 males and 20 females. The age range, mean age, gender and duration of pain were not reported in each group. DDD was verified by x‐ray
Interventions	INDEX TREATMENT: Huangqi (prepared Radix Astragali)18 g, Dangshen (Radix Codonopsis) 9 g, Sanqi (Radix Notoginseng) 9 g, Chuanxiong (Rhizoma Chuanxiong)12 g, Lujiao (Cornu Cervi Pantotrichum) 12 g, and Zhimu (Rhizoma Anemarrhenae) 12 g. These herbs were boiled in water and decocted to 300 ml. This solution was then applied topically twice a day. COMPARISON TREATMENT: Mobicox 7.5 mg once a day; Methycobal 0.5 mg 3 times a day for 4 weeks Treatment Schedule: Over 4 weeks Duration of Follow‐up: 4 weeks after randomization
Outcomes	PAIN RELIEF: (Numerical Rating Scale) Baseline mean: Unclear End of study mean: Unclear Reported Results: Herbal treatment was significantly better than Mobicox and Methycobal at 4 weeks. Reason for Dropouts: There were no drop‐outs Adverse Effects: One case had watery stool after herbal treatment, which was improved by further herbal treatment. Two cases had discomfort in the upper abdomen after Mobicox and Methycobal treatment. The discomfort disappeared when the patient took the drugs 30 min after eating. Cost of Care: Herbal treatment, RMB 232.9 to 625.4, with an average of RMB 376.07±115.32. Mobicox and Methycobal, RMB 242.4 to 475.8, with an average of RMB 375.25±66.33.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	Sequence was generated from a random table. This was confirmed in the author's reply to our survey about their allocation concealment
Allocation concealment?	High risk	This was not described in the article and we confirmed with the author through our survey that allocation concealment was not done
Blinding? patients blinded ‐ All outcomes	High risk	Not reported
Blinding? Careproviders blinded ‐ All outcomes	High risk	Not reported
Blinding? Outcome assessor blinded ‐ All outcomes	High risk	Not reported
Incomplete outcome data addressed? All outcomes ‐ ITT analysis?	High risk	Not reported
Incomplete outcome data addressed? All outcomes ‐ drop‐outs?	Low risk	There were no drop‐outs
Free of selective reporting?	High risk	Details of participants were not provided
Similarity of baseline characteristics?	High risk	Unclear from descriptions in the article
Co‐interventions avoided or similar?	Unclear risk	Unclear whether co‐interventions were avoided from the description in the article
Compliance acceptable?	Unclear risk	Not assessed.
Timing outcome assessments similar?	Low risk	All at 4 weeks after randomization

Wang 2004.

Methods	RCT Number Analyzed/Randomized: 217/240 Index treatment: N = 109/120 Comparison treatment: N = 108/120 Power Analysis: not reported Funding Source: Shanghai Developing Company of Modern TCM Technique
Participants	Chronic neck disorder with radicular signs or symptoms Patients recruitment: out‐patients from the Fujian Academy of Traditional Chinese Medicine, the First Hospital Affiliated to Guangxi College of Traditional Chinese Medicine, Zhejiang Hospital of Traditional Chinese Medicine, and the First Hospital Affiliated to Hunan College of Traditional Chinese Medicine Index treatment: the age range was 25 to 65 year old, mean age was 54.86±7.47 years old; 52 male, 68 female; duration of pain was 5.06±4.62 years. Comparison treatment: the age range was 27 to 65 year old, mean age was 54.65±8.90 years old; 45 male, 75 female; duration of pain was 4.94±4.16 years. DDD was verified by x‐ray
Interventions	INDEX TREATMENT: Compound Qishe Tablet 25 tablets twice a day for 4 weeks. COMPARISON TREATMENT: Placebo 25 tablets twice a day for 4 weeks. Treatment Schedule: for 4 weeks Duration of Follow‐up: 8 weeks after randomization
Outcomes	PAIN RELIEF: (Numerical Rating Scale, 0 to 3) with 0 indicated no pain, 1 mild, 2 moderate, and 3 severe. Definitions of recovery was defined as having no neck or radicular pain. Baseline: Compound Qishe Tablet group: 1 case without neck pain, 27 cases with mild neck pain, 65 cases with moderate neck pain, and 16 cases with severe neck pain, and 2 cases without radicular pain, 40 cases with mild radicular pain, 63 cases with moderate radicular pain, and 4 cases with severe radicular pain. Control group: 27 cases with mild neck pain, 67 cases with moderate neck pain, and 14 cases with severe neck pain, 1 case without radicular pain, 46 cases with mild radicular pain, 52 cases with moderate radicular pain, and 9 cases with severe radicular pain. End of study: Compound Qishe Tablet group: 42 cases without neck pain, 57 cases with mild neck pain, 10 cases with moderate neck pain, and 0 case with severe neck pain, and 60 cases without radicular pain, 46 cases with mild radicular pain, 3 cases with moderate radicular pain, and 0 case with severe radicular pain. Control group: 10 cases without neck pain, 50 cases with mild neck pain, 45 cases with moderate neck pain, and 3 cases with severe neck pain, and 17 case without radicular pain, 54 cases with mild radicular pain, 35 cases with moderate radicular pain, and 2 cases with severe radicular pain. Reported Results: Compound Qishe Tablet is more effective for pain relief than placebo for patients with chronic neck pain, measured at the end of the study (eight weeks of follow‐up after four weeks of treatment). Reason for Dropouts: withdrew, refused treatment, accident, lost to follow‐up Index treatment: 4 withdrew, 3 refused treatment, 1 accident, 3 lost to follow‐up Comparison treatment: 3 withdrew, 6 refused treatment, 1 accident, 2 lost to follow‐up Adverse Effects: Index treatment: 2 cases with diarrhea, 1 case with abdominal pain and diarrhea, and 1 case with stomachache. Comparison treatment: 1 case with diarrhea and abdominal pain. Cost of Care: not reported
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	Sequence was generated from a random table. This was confirmed in the author's reply to our survey about their allocation concealment
Allocation concealment?	High risk	This was not described in the article and we confirmed with the author through our survey that allocation concealment was not done
Blinding? patients blinded ‐ All outcomes	Low risk	Described in the report
Blinding? Careproviders blinded ‐ All outcomes	Low risk	Described in the report
Blinding? Outcome assessor blinded ‐ All outcomes	Low risk	Described in the report
Incomplete outcome data addressed? All outcomes ‐ ITT analysis?	High risk	Not reported
Incomplete outcome data addressed? All outcomes ‐ drop‐outs?	Unclear risk	The number of drop‐outs could be calculated but the reasons were unclear. Index treatment: 11 drop‐outs Comparison treatment: 12 drop‐outs
Free of selective reporting?	High risk	Information such as age range and duration of symptoms were missing
Similarity of baseline characteristics?	Low risk	reported to be similar
Co‐interventions avoided or similar?	Unclear risk	Unclear whether co‐interventions were avoided from the description in the article
Compliance acceptable?	Unclear risk	Not assessed.
Timing outcome assessments similar?	Low risk	All at the 8 week follow‐up after 4 weeks of treatment

Wang 2005.

Methods	RCT Number Analyzed/Randomized: 416/440 Index treatment: 311/330 Comparison treatment: 105/110 Power Analysis: not reported Funding Source: Shanghai Developing Company of Modern TCM Technique
Participants	Chronic neck disorder with radicular signs or symptoms Patients recruitment: out‐patients from the Fujian Academy of Traditional Chinese Medicine, the First Hospital Affiliated to Guangxi College of Traditional Chinese Medicine, Zhejiang Hospital of Traditional Chinese Medicine, and the First Hospital Affiliated to Hunan College of Traditional Chinese Medicine, Affiliated Hospital of Jiangxi College of Traditional Chinese Medicine, Hubei Academy of Traditional Chinese Medicine Index treatment: the age range was 30 to 62 year old; mean age was 49.71±11.07 years old; 121 male, 219 female; duration of pain was 4.77±5.42 years. Comparison treatment: the age range was 29 to 63 year old; mean age was 50.45±10.78 years old; 32 male, 78 female; duration of pain was 4.89±5.26 years.
Interventions	INDEX TREATMENT: Compound Qishe Tablet 1 bag (25 tablets) twice a day for 4 weeks with placebo Jingfukang 1 bag (granules) twice a day for 4 weeks COMPARISON TREATMENT: Placebo Qishe Tablet 1 bag (25 tablets) twice a day for 4 weeks with Jingfukang 1 bag (granules) twice a day for 4 weeks CO‐INTERVENTION: unclear Treatment Schedule: for 4 weeks Duration of Follow‐up: 8 weeks after randomization
Outcomes	PAIN RELIEF: (Numerical Rating Scale, 0 to 3) with 0 indicated no pain, 1 mild, 2 moderate, and 3 severe. Baseline: Compound Qishe Tablet group: 0 cases without neck pain, 38 cases with mild neck pain, 229 cases with moderate neck pain, and 44 cases with severe neck pain; and 18 cases without radicular pain, 88 cases with mild radicular pain, 189 cases with moderate radicular pain, and 16 cases with severe radicular pain. Control group:, 0 cases without neck pain, 14 cases with mild neck pain, 83 cases with moderate neck pain, and 8 cases with severe neck pain; and 1 case without radicular pain, 28 cases with mild radicular pain, 74 cases with moderate radicular pain, and 2 cases with severe radicular pain. End of study: Compound Qishe Tablet group: 84 cases without neck pain, 202 cases with mild neck pain, 24 cases with moderate neck pain, and 1 case with severe neck pain; and 166 cases without radicular pain, 128 cases with mild radicular pain, 17 cases with moderate radicular pain, and 0 cases with severe radicular pain. Control group: 28 cases without neck pain, 61 cases with mild neck pain, 16 cases with moderate neck pain, and 0 cases with severe neck pain; and 51 case without radicular pain, 42 cases with mild radicular pain, 12 cases with moderate radicular pain, and 0 cases with severe radicular pain. Reported Results: Compound Qishe Tablet treatment was significantly better than Jingfukang at eight weeks after completing the four weeks of treatment. Reason for Dropouts: withdrew, refused treatment, accident, lost to follow‐up Index treatment: 19 cases Comparison treatment: 5 cases Adverse Effects: Index treatment, 11 cases with diarrhea, 9 cases with abdominal pain, and 2 cases with stomachache. Comparison treatment, 1 case with diarrhea, 1 case with stomachache. Cost of Care: not reported
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	Sequence was generated from a random table. This was confirmed in the author's reply to our survey about their allocation concealment
Allocation concealment?	High risk	This was not described in the article and we confirmed with the author through our survey that allocation concealment was not done
Blinding? patients blinded ‐ All outcomes	Low risk	Described in the report
Blinding? Careproviders blinded ‐ All outcomes	Low risk	Described in the report
Blinding? Outcome assessor blinded ‐ All outcomes	Low risk	Described in the report
Incomplete outcome data addressed? All outcomes ‐ ITT analysis?	High risk	Not reported
Incomplete outcome data addressed? All outcomes ‐ drop‐outs?	Unclear risk	Reasons for drop‐outs were mentioned but the number of drop‐outs associated with each reason was not given Index treatment: 19 cases Comparison treatment: 5 cases
Free of selective reporting?	High risk	Information such as age range and duration of symptoms were missing
Similarity of baseline characteristics?	Low risk	Reported to be similar
Co‐interventions avoided or similar?	Unclear risk	Unclear whether co‐interventions were avoided from the description in the article
Compliance acceptable?	Unclear risk	Not described.
Timing outcome assessments similar?	Low risk	All assessed at the end of the 8 weeks after end of treatment

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Teng 2009	combined treatment of Xiaoyusan plaster (compound herbal medicine) and Daoyin (an exercise modality)
Ying 2008	compared two different acupuncture methods for supraclavicular nerve‐entrapment syndrome

Contributions of authors

Xuejun Cui: conceived the idea, designed the review, developed the literature search strategy, searched the literature, wrote the body of the review, contacted experts for trials and clarification of methods, screened trials for inclusion, assessed risk of bias and prepared the submission for the Cochrane Back Review Group.

Kien Vinh Trinh: Searched the literature, screened trials for inclusion, assessed risk of bias, edited and contributed to the body of the review and prepared the submission and re‐submission to the Cochrane Back Review Group.

Yongjun Wang: Screened trials for inclusion, edited final version of protocol and review. 

An‐Fu Hsiao revised the protocol.

Sources of support

Internal sources

• No sources of support supplied

External sources

• The State Administrationof Traditional Chinese Medicine of the People's Republic of China(06‐07JQ05), China.

Declarations of interest

This subject is supported by the State Administration of Traditional Chinese Medicine of the People's Republic of China (06‐07JQ05)

New