THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE Volume 16, Number 8, 2010, pp.

889898 Mary Ann Liebert, Inc. DOI: 10.1089/acm.2009.0470

Review Article

Is Oral Administration of Chinese Herbal Medicine Effective and Safe as an Adjunctive Therapy for Managing Diabetic Foot Ulcers? A Systematic Review and Meta-Analysis
1,2 1 Min Chen, MD, Hui Zheng, MD,3 Li-Ping Yin, MD, PhD, and Chun-Guang Xie, MD, PhD2

Abstract

Objective: This meta-analysis aimed to assess the effectiveness and safety of Chinese herbal medicine (CHM) as an adjunctive method to standard therapy for patients with diabetic foot ulcers (DFU). Methods: Randomized controlled trials (RCTs) of CHM to treat DFU were searched in the following electronic databases: MEDLINE; EMBASE; Chinese Biomedical Database (CBM); Cochrane Central Register of Controlled Trials (CENTRAL); Allied & Complementary Medicine Resources (AMED); and Cumulative Index to Nursing and Allied Health Literature (CINAHL). Two (2) researchers independently assessed the quality and validity of included trials and extracted outcome data for synthesis. Results: Six (6) trials were included for analysis. Compared to using standard therapy alone, CHM combined with standard therapy signicantly increased the number of patients whose ulcers healed (risk ratio [RR], 0.62, [95% condence interval (CI), 0.390.97]) and number of patients with at least a 30% reduction in the ulcer area (RR, 0.81 [95%CI, 0.710.92]). In addition, the two therapies combined signicantly decreased the number of patients without any improvement (RR, 0.34 [95%CI, 0.210.53]). However, with respect to blood ow volume in the dorsal artery of the foot, no signicant difference between the two therapies was observed (standardized mean difference, 1.71 [95% CI 1.254.67]), but the result favored the CHM combined with standard therapy group. Only 2 of 6 trials reported adverse events, which included nausea, epigastric pain, and dry mouth. Conclusions: CHM may be effective and safe as an adjunctive therapy for treating DFU. However, a rm conclusion could not be reached because of the poor quality of the included trials. Further trials with higher quality are justied.

Introduction

iabetic foot ulcers (DFU) are one of the serious complications of diabetes; these ulcers develop in about 1525% of patients with diabetes.1,2 DFU cause not only emotional burdens and pain problems, but also reduce quality of life (QoL).3,4 In addition, 28%89% of amputations performed for nontraumatic lesions are associated with diabetes, with considerable variations globally. Differences in the prevalence of diabetes and availability of health care may contribute to these variations.5 Furthermore, the economic burden of DFU is heavy, both to society and to patients and their families. The annual cost of managing DFU and amputations was estimated to be $10.9 billion in the United States, whereas the cost was estimated to be $385 million in the United Kingdom, based on the same methodology.6 The
1 2 3

global burden of DFU is expected to be raised, because of the increasing incidence of diabetes.7 Therefore, it is necessary and urgent to nd some cost-effective treatments for DFU. Multidisciplinary treatment of DFU has been suggested, because DFU is such a complicated condition. Recommendations for treating DFU included patient education; holistic evaluation and foot-ulcer assessment; identifying and optimizing systematic, local, and extrinsic factors (i.e., debridement, infection control, a moist wound environment, or pressure redistribution); and adjunctive therapies.8 In China, herbal medicine for wound healing has been applied for thousands of years. In modern times, Chinese herbal medicine (CHM) has been used as an adjunctive method for managing DFU. Several experimental studies justied its merits, such as enhancing broblast viability9 and antidiabetic properties of

TCM Internal Medicine Department, Chengdu University of Traditional Chinese Medicine, Chengdu, Sichuan, China. First Afliated Hospital of Chengdu University of Traditional Chinese Medicine, Chengdu, Sichuan, China. Acupuncture and Moxibustion Department, Chengdu University of Traditional Chinese Medicine, Chengdu, Sichuan, China.

889

890 CHM.10,11 In addition, some clinical trials evaluated the role of traditional CHM for treating DFU.1214 Their results showed that CHM may improve the conditions patients with DFU. However, these trials were case reports, case-series studies, or quasi-randomized controlled trials (quasi-RCTs), of which the results were not convincing enough to rationalize the use of CHM to treat DFU. It has been reported that there were *25% of patients, hospitalized in internal medicine wards, who were consuming herbal or dietary supplements.15 So, it is urgent and important to nd answers to two questions: (2) Is CHM effective as a kind of therapy for patients with DFU? (2) Is currently used CHM safe for managing DFU? A systematic review of randomized controlled trials (RCTs) may be a reasonable solution for addressing these questions, because RCTs are recognized as the gold standard for testing the effectiveness of therapies. However, few RCTs are found in English-based databases; for example, at the time of this review, there were only 3 RCTs found on PubMed using diabetic foot and herbal medicine as keywords. Many RCTs evaluating herbal medicine for DFU were published in China, and most of them were reported without abstracts in English. Therefore, a systematic review of RCTs using CHM for patients with DFU was necessary, and it was also essential to include a Chinese-literature database in the review process. Consequently, the aim of this research was to conduct a systematic review of RCTs for patients with DFU, and to address the questions regarding whether CHM is effective and safe as an adjunctive therapy for managing DFU. Methods Study selection Types of studies. Only randomized controlled trials (RCTs) were included. Quasi-RCTs, nonrandomized controlled trials, observational studies, and cohort studies were ruled out. Quasi-RCTs use false randomization methods, which do not provide each participant with an equal probability of enrollment into different arms of the studies. In this review, such methods were dened as generating a random allocation sequence based on date of birth, hospital record number, date of enrollment, and other factors. Types of participants. Trials were included in which participants met the following criteria: rst, patients were diagnosed as having diabetes according to the diagnostic criteria recommended by World Health Organization (WHO) and the American Diabetes Association (ADA); second, patients were also diagnosed as having DFU, which included neuropathic (warm with intact pulses, diminished sensation, callus, etc.), ischemic, and neuroischemic (pulseless, necrosis or gangrene, etc.) ulcers.16 Studies including other types of foot ulcers (ulcers caused by venous disease, vasculitis, or any conditions other than diabetes) were excluded. Types of interventions. The use of CHM in diabetes control is promising but still far from proven, because there is still insufcient evidence to draw denitive conclusions.17,18 Thus, for participants with DFU, it is unethical and unsafe to use CHM alone in an experimental group. So trials

CHEN ET AL. were included that evaluated use of CHM combined with standard therapy for DFU. The standard therapy was dened as standard antidiabetic treatment with or without use of antibiotics and debridement. In addition, management of osteomyelitis (X-ray examination, debridement, etc.) was also considered as standard therapy, because approximately 15% of DFU are complicated by osteomyelitis.19 Trials using CHM orally combined with treatment options other than standard therapy were excluded. These treatment options referred to pressure-relieving interventions, wound dressings, decompressive surgery, etc. Trials evaluating the effectiveness of topical use of CHM (e.g., using CHM as wound dressings) or both topical and oral administration of CHMs were also excluded. Types of outcome measures. Considering that the outcome measures of most studies of CHM were focusing on number of patients whose ulcers healed, reduced or not reduced, the following outcomes were considered: Primary outcomes were number of patients whose ulcers healed and number of patients without any improvement in the condition. Secondary outcomes were number of patients with at least a 30% reduction in the ulcer area and improved blood ow volume in the dorsal artery of the foot. Data sources and search strategy A systematic review was conducted to collect clinical trials that used Chinese herbal medicine to treat DFU. The following electronic databases were searched from inception to March 2009: MEDLINE; EMBASE; Chinese Biomedical database (CBM); Cochrane Central Register of Controlled Trials (CENTRAL); Allied & Complementary Medicine Resources (AMED); Cumulative Index to Nursing; and Allied Health Literature (CINAHL). Given that CHM is widely used in China, and many studies were published in Chinese, the CBM was searched to enlarge the scope of this review. CBM is available online at http:/ /cbmwww.imicams.ac.cn/ and allows rapid searching of journal abstracts to quickly locate clinical trials published in China after 1978. The following search terms were used as as keywords and medical subject headings: diabetic foot, diabetic feet, and foot ulcer; and drugs, Chinese herbal, medicine, Chinese traditional, medicine, East Asian traditional, Chinese herbs, tang, and decoction. The following search strategy was used: (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) Search diabetic foot/all subheadings. Search diabetic feet. Search foot ulcer/all subheadings. Search #1 OR #2 OR #3. Search drugs, Chinese herbal/all subheadings. Search medicine, Chinese traditional/all subheadings. Search medicine, East Asian traditional/all subheadings. Search Chinese herbs. Search tang. Search decoction. search #6 OR #7 OR #8 OR #9 OR #10; #12 search #4 AND #11.

This search strategy was used with a method lter for clinical controlled trials and including articles published in all languages was considered.

CHINESE HERBAL MEDICINE FOR DIABETIC FOOT ULCERS Data extraction and quality assessment Two independent reviewers (H.Z. and L-P.Y.) reviewed the selected trials and extracted the details of data on participants, generation of random allocation sequence, allocation concealment, blinding, interventions, comparisons, outcome measures, intention-to-treat analysis (ITT), and follow-up. Two (2) reviewers (H.Z. and L.-P.Y.) independently assessed the methodological quality of the included studies using the 5-point Jadad scale.20 The scale focused on whether a study is described as randomized and double-blinded, the methods for generation of the allocation sequence and double blinding, and whether there was a description of dropouts during the study. Studies with 3, 4, or 5 points were considered high quality. In addition, these 2 reviewers also evaluated the internal validity of the studies with an 11-item scale developed by Cochrane Back Review Group.21 The 11 items focused on the following information of the trials: adequacy of randomization; allocation concealment; baseline equivalence; blinding of patients; care providers and outcome assessors; cointerventions; patients compliance; dropout rate; timing of outcome assessment; and whether the analysis was based on the ITT principle. Discrepancies between the 2 reviewers assessments were resolved by discussion among all of the researchers for this meta-analysis. Data synthesis and analysis The data were analyzed with Review Manager, version 5.0.18 (Nordic Cochrane Centre, Copenhagen, Denmark). Comparisons of treatment effects were made between standard therapy combined with CHM and standard therapy used alone. Heterogeneity among trials was tested using a chi-squared test with a p-value < 0.10 to dene a signicant degree of heterogeneity. I2 statistics were used with a cutoff point of 25%. If heterogeneity among trials existed, the data were pooled from the included trials with a random-effect model; otherwise, the xed model would be selected. The dichotomous data were reported as relative risk (RR) with a corresponding 95% condence interval (95% CI). For continuous data, a standardized mean difference was calculated. A sensitivity analysis was planned a priori, according to study quality ratings on the Jadad scale. In addition, subgroup analyses were planned, according to type of CHM decoction and wound grade of the DFU. Results Study identication and characteristics In the initial screening, 67 articles were identied in which the title, the abstract, or both mentioned the use of Traditional Chinese Medicine (TCM) for managing DFU. Most of the articles were from the CBM. Of the 67 articles, 8 were excluded for reporting the same trials or multiple citations in the databases. In addition, 52 articles were excluded because they were quasi-RCTs, reported trials of nondiabetic foot ulcers, and other unacceptable factors. Finally, 6 RCTs were included for analysis. Among the 52 excluded articles, 13 were articles describing RCTs evaluating the effectiveness of topical use of CHM (10 trials) or both topical and oral ad-

891

ministration of them (3 trials). In these trials, a total of 999 participants were excluded; 500 participants were allocated in experimental groups (standard therapy combined with topical use of CHM or both topical and oral administration of them), while 499 participants were allocated to control groups (standard therapy used alone or standard therapy combined with placebo). Results of these excluded trials seemed to favor the experimental group. Figure 1 shows more details of the screening process, including how many articles were found in the databases. The 6 included trials were small-sample-size trials; the largest one included a total of 96 patients.22 All of them used TCM plus standard therapy as an experimental intervention, whereas standard therapy alone was used as a comparison. The only difference between groups was the administration of CHM in intervention groups. Standard therapies used in the studies included: diabetes education, diet, exercise; standard antidiabetic treatment; antibiotics; cleaning and desloughing the ulcers daily; and prescribing blood-vessel dilation medicine. All of the studies originated in China; ve of them were from the mainland and one of them was from Hong Kong. In total, 439 patients were included; 221 of them were traditional CHM plus standard therapy groups, the other 218 patients were in standard therapy groups. The studies used different remedies of TCM; however, some herbs were used in at least 2 trials. For example, Radix Angelicae sinensis, Flos Carthami Tinctorii, and Semen Persicae were used in 4 trials;2225 Radix astragali were used in 3 trials.2426 Important data on the characteristics of the 6 studies are summarized in Table 1. Quality of the included trials Methodological quality of the included trials varied from high to low. Five (5) of them were of low quality, with the scores of the 11-item scale ranging from 3 to 4 and the scores of the Jadad scale as 1.2226 All of the low-quality trials were conducted in mainland China from 2006 to 2008, and the results were consistent with an article reporting the quality of RCTs of TCM in mainland China.27 Only 1 trial was designed as a double-blinded RCT, and it was also the only trial labeled as high quality.28 More details on the scores for each trial were presented in Table 2. Results of the meta-analysis All the included trials compared CHM combined with standard therapy (CHMST) with standard therapy alone (ST). However, outcome measurements of some trials were different from the others, so the data of studies that had used the same outcome measurements were combined. The results suggested that there was a signicant difference between CHMST and ST in number of patients whose ulcers healed (RR, 0.62 [95% CI, 0.390.97]). More patients were healed in CHMST groups than in ST groups. Signicant differences were also noted between CHMST and ST in number of patients without any improvement (RR, 0.34 [95% CI, 0.210.53]), number of patients with at least a 30% reduction in the ulcer area (RR, 0.81 [95% CI, 0.710.92]). In addition, these results favored the CHMST groups. With respect to blood ow volume in the dorsal artery of the foot

892 Randomization

CHEN ET AL.

During literature retrieval, 19 articles were reported as RCTs, which were actually quasi-RCTs. In these studies, random sequences were generated according to either hospital number or the chronologic order of patients medical records. In the included studies, only 2 described the randomization method used and applied correctly; whereas the other 4 articles did not provide any information about generating random allocation sequences. According to a previous survey, articles reporting generation of random allocation sequences comprised 7.9% of all analyzed articles.27 In the present meta-analysis, the percentage increased; however, most of the researchers applied the wrong approach to generate random allocation sequences. Given the availability of free software to generate random numbers, the current authors suggest that Chinese investigators use them in future trials. Allocation concealment is still a big problem in trials examining CHM. The previous survey reported 0.3% of the articles providing description of how concealment was implemented.27 In this review, none of the 6 included articles described allocation concealment. Therefore, in future studies, an independent data monitoring board is recommended for taking charge of the concealment process. For Chinese investigators, a national center of good clinical practice in China is an option for independent data monitoring. FIG. 1. Study screening process. AMED, Allied & Complementary Medicine Resources; CINAHL, Cumulative Index to Nursing; and Allied Health Literature. Blinding Blinding is an essential method for preventing research outcomes from being inuenced by either the placebo effect or the observer bias. Ideal blinding should mask participants, care providers, outcome assessors, and statistical analysts. Placebo plays an important role in blinding participants and care providers; however, only 1 trial used placebo to achieve satisfactory blinding.28 The 5 trials provided only conventional treatment for patients in the control groups, while providing CHM combined with conventional treatment for patients in experimental groupsthis method made these studies open trials with a high risk of bias. One of the main reasons for not using placebos in these trials may be the difculties involved in preparing them to have the same color and avor as an experimental CHM. Therefore, adopting advanced technologies and facilities of pharmaceutical preparations may solve the problem, making placebo CHMs indistinguishable from real ones. A more practical method is to prepare a CHM in the form of capsule or pill. For example, Yunnan Baiyao capsules and Danqi Piantang Jiaonang were successful cases of preparing CHM in the form of a capsule.29,30 In a word, in trials evaluating CHM, researchers should pay attention to the quality of the placebo. In addition, quality of blinding ought to be assessed, for instance, by asking patients to guess which group they were in. Follow-up process and ITT analysis This meta-analysis results showed that none of the included trials described the follow-up process clearly and did not state whether they analyzed the data based on the ITT principle. In addition, information on dropout rates was was not provided. Therefore, in future reports of RCTs, losses to follow-up and how they were dealt with should be clearly stated. For instance, researchers should report the dropout

(which was measured using color Doppler ultrasound), there was no signicant difference between the above two groups (standardized mean difference, 1.71 [95% CI 1.254.67]), but the results favored CHMST groups, in which a larger mean volume of blood ow was reported. Figure 2 shows more details. Adverse events Only 2 of 6 trials reported adverse events,24,28 which included nausea, epigastric pain, dry mouth, and diarrhea. None of these trials reported statistically signicant difference between treatment and control groups in terms of adverse events. One (1) trial reported that nausea was reduced after the patients were told to take the herbal medicine after meals.24 The other trial reported that, although epigastric pain, dry mouth, and diarrhea were experienced, the patients liver and renal functions were not affected.28 Discussion These meta-analysis data suggest that CHM may work effectively as an adjunctive therapy for patients with DFU. However, because the included studies were of generally poor quality and had small sample sizes, it was not possible to draw a rm conclusion. This meta-analysis also suggests that CHM may be safe for managing DFU. Nonetheless, as herbal remedies were different across studies, and some studies did not report adverse events, it is difcult to draw a denite conclusion.

Table 1. Characteristics of Included Trials Interventionsb STCHM Bu-yang-huan-wu decoction (Radix astragali 60 g, Radix Angelicae sinensis 15 g, Radix Paeoniae Lactiorae 10 g, Lumbricus 10 g, Radix Ligustici Chuanxiong 10 g, Semen Persicae 10 g, and Flos Carthami Tinctorii 10 g); CHM taken orally twice a day for 21 days Comparisonc Outcome measuresd B

Study ID

Participants (experimental/control)a

Deng et al. (2007)

Sample size (40/35) Diabetes duration (11.7 4.8) Ulcer size (not mentioned) Diabetes control (HbA1c) (10.6% 2.1%/10.5% 1.9%) Duration of DFU (5.1 1.8/4.8 1.9 years) Baseline equivalence ( p > 0.05)

He (2008)

Sample size (40/38) Diabetes duration (11.2 9.8/11.8 10.1 years) Ulcer size (not mentioned) Diabetes control (not mentioned) Duration of DFU not mentioned Baseline equivalence ( p > 0.05)

ST 1. Diabetes education, diet, exercise 2. Standard antidiabetic treatment 3. Antibiotics 4. Prostaglandin E1, 60 mg per day by IV drip Treatments given once a day for 21 days ST 1. Standard antidiabetic treatment 2. Antibiotics 3. Cleaning and desloughing ulcers daily 4. Anisodamine, 20 mg per day, by IV drip. Treatments given once a day for 30 days ST 1. Standard antidiabetic treatment 2. Antibiotics 3. Cleaning and desloughing ulcers daily Treatments given once a day ST 1. Standard antidiabetic treatment 2. Antibiotic; 3. Blood vessel dilation medicine Treatments given once a day for 30 days

A, B, C, D

893

Leung et al. (2008)

Sample size (40/40) Diabetes duration (8.4 7.6/12.4 8.8 years) Ulcer size (28.7 31.3/26.7 27.3 cm2) Duration of DFU (7.8 8.2/12.9 24.6 weeks) Diabetes control (19,14,4/17,17,1) Baseline equivalence ( p > 0.05)

B, E, F, G

Song et al. (2006)

Sample size (25/25) Diabetes duration (14.0 9.0 years) Ulcer size (70.9 69.9 cm2) Duration of DFU (32.5 30.5 days) Diabetes control (14.4 7.4 mmol/L) Baseline equivalence ( p > 0.05)

STCHM Tao-hong-si-wu decoction (Semen Persicae 12 g, Radix Ligustici Chuanxiong 12 g, Radix Angelicae sinensis 12 g, Radix Rehmanniae 12 g, Radix astragali 12 g, Flos Carthami Tinctorii 10 g, Radix Paeoniae Lactiorae 10 g, Hirudo seu Whitmania 10 g, Lumbricus 10 g, Rhizoma Atractylodis Macrocephalae 10 g, Radix Salvia miltiorrhizae 15 g, Gummi Olibanum 6 g, Myrrha 6 g, and Radix Glycyrrhizae Uralensis 6 g); CHM orally taken twice a day for 30 days STCHM Radix astragali, Rhizoma atractylodis, marcocephalae, Radix stephaniae tetrandrae, Radix polygoni multifori, Radix rehmanniae, Radix smilax china, Fructus corni, Rhizoma dioscoreae, Cortex moutan, Rhizoma alismatis, Rhizoma smilacis glabrae, and Fructus schisandrae; CHM given orally twice a day for 24 weeks STCHM Si-miao-yong-an decoction (Radix Scrophulariae Ningpoensis 15 g, Radix Angelicae sinensis 15 g, Flos Lonicerae japonicae 18 g, Radix Glycyrrhizae Uralensis 6 g, Semen Persicae 10 g, Flos Carthami Tinctorii 10 g, Rhizoma Dioscoreae Septemlobae 12 g, Cortex Phellodendri 10 g, and Alismatis Rhizoma 10 g) CHM given orally twice a day for 30 days

A, B, C

(continued)

Table 1. (Continued) Interventionsb STCHM Radix astragali 30 g, Radix Rehmanniae 20 g, Herba Siegesbeckiae 20 g, Draconis Sanguis 5 g, Lumbricus 15 g, Radix Ligustici Chuanxiongm 15 g, and Ramulus Cinnamomi Cassiae 10 g; CHM given orally twice a day for 8 weeks Comparisonc Outcome measuresd A, B, D

Study ID

Participants (experimental/control)a

Wang et al. (2006)

Sample size (30/30) Diabetes duration (5.5 3.7/5.9 3.2 years) Ulcer size (not mentioned); Duration of DFU (not mentioned) Diabetes control (11.2 4.27/10.9 3.8 mmol/L) Baseline equivalence ( p > 0.05)

Xing (2008)

Sample size (46/50) Diabetes duration (8.61 1.97/7.68 2.21years) Ulcer size (134.12 132.75/111.04 110.4 cm2) Duration of DFU (15,16,15/16,16,18) Diabetes control (not mentioned) Baseline equivalence ( p > 0.05)

A, B, C, D

894

STCHM She-xiang-huo-xue capsule (Radix Notoginseng, Radix Angelicae sinensis, Draconis Sanguis, Flos Carthami Tinctorii, Semen Persicae, Secretio Moschus, Eupolyphaga seu Opisthoplatia, Rhizoma Drynariae, Dipsaci Radix and Radix Achyranthis Bidentatae); CHMs given orally twice a day for 28 days

ST 1. Diabetes education, diet, exercise 2. Standard antidiabetic treatment; 3. Antibiotics 4. Cleaning and desloughing ulcers daily 5. Blood-vessel dilation medicine Treatments given once a day for 8 weeks ST 1. Diabetes education, diet, exercise 2. Standard antidiabetic treatment 3. Antibiotics 4. Cleaning and desloughing ulcers daily 5. Blood vessel dilation medicine. Treatments given once a day for 28 days

a In trials in which baseline data were reported per group, we used slashes to separate them. As suggested in the table (experimental/control), on the left side of the data belong to the to experimental group, whereas data on the right side belong to the control group. In trials that did not report baseline data separately, we used the data of the whole sample as reported. Data were listed as mean standard deviation or mean only, the units were attached behind the numbers. b In the intervention groups, both standard therapy and Chinese herbal medicine were used. The unit g refers to gram. Some trials did not specify the dose used, so the doses were not listed in this table. c In the control groups, only standard therapy was used. The only difference between groups was the administration of CHM in the intervention group. d A refers to number of patients whose ulcers healed; B refers to number of patients without any improvement; C refers to number of patients with at least a 30% reduction in ulcer area; D refers to blood ow volume in the dorsal artery of the foot; E refers to limb salvage; F refers to rate of ulcer healing; G refers to changes in the surface. temperature. DFU, diabetic foot ulcers; ST, standard therapy; CHM, Chinese herbal medicine; IV, intravenous.

CHINESE HERBAL MEDICINE FOR DIABETIC FOOT ULCERS Table 2. Quality of the Included Trials 11-item scalea Study ID Song et al. Deng et al. He Xing Wang et al. Leung et al. A 1 0 0 0 1 1 B 0 0 0 0 0 0 C 1 1 1 1 1 1 D 0 0 0 0 0 1 E 0 0 0 0 0 1 F 0 0 0 0 0 0 G 1 1 1 1 1 1 H 0 0 0 0 0 1 I 0 0 0 0 0 0 J 1 1 1 1 1 1 K 0 0 0 0 0 0 Total 4 3 3 3 4 7 a 1 1 1 1 1 1 b 0 0 0 0 1 1 Jadad scaleb c 0 0 0 0 0 1 d 0 0 0 0 0 1 e 0 0 0 0 0 0

895

Total 1 1 1 1 2 4

a The 11-item scale was developed by the Cochrane Back Review Group. A was the method of randomization adequate?; B was the treatment allocation concealed?; C were the groups similar at baseline regarding the most important prognostic indicators?; D was the patient blinded to the intervention?; E was the care provider blinded to the intervention?; F was the outcome assessor blinded to the intervention?; G were co-interventions avoided or similar?; H was the compliance acceptable in all groups?; I was the dropout rate described and acceptable?; J was the timing of the outcome assessment in all groups similar?; K did the analysis include an intention-totreat analysis? The item is scored 1 point if the answer is yes, otherwise, it is assigned 0 points. A total score of 6 or more is recognized as high quality. b Points were awarded as follows: a the study was described as randomized, 1 point; b the randomization scheme was described and appropriate, 1 point, or inappropriate randomization method, reduce 1 point; c the study was described as double blinded, 1 point; d the method of double-blinding was appropriate, 1 point; e there was a description of withdrawals and dropouts, 1 point.

rate in each group and illustrate whether the dropout rate was equivalent among groups. In addition, researchers should report which methods for managing missing data were used, such as last observations carried forward (LOCF), multiple imputation (MI), etc. Standardization of herbal remedies During the analysis, clinical heterogeneity was noted among the trials, although the meta-analysis showed no statistical heterogeneity among them. One of the most important reasons for this was the diversity of the remedies. Every trial used a different remedy, which made the effect of the CHM hard to assess. So, standardization of herbal remedies is required in future studies. An remedy needs to go through following process before it is considered a xed herbal remedy: (1) National TCM specialists need to reach a consensus on the therapeutic principles and constituent sof the remedy, ensuring that the remedy conforms to the rules of TCM (e.g., the remedy improves the qi and Blood Stagnation and is suitable for patients DFU diagnosed as having a Stagnation Pattern. (2) Herbs to be used in the remedy should be recommended, logically, based on the best evidence available. (3) Third, a small-sample observational study should be performed to examine the remedy, before it is used in an RCT. The results of the current analysis showed that Radix Angelicae sinensis, Flos Carthami Tinctorii, Semen Persicae, and Radix astragali may be ingredients of a xed remedy for patients with DFU because these herbs were used in several RCTs and achieved positive effects. Apart from the abovementioned points, the denition of standard therapy in the included trials was not clear enough. For example, some major complications, such as osteomyelitis, and how they were managed were not mentioned. Therefore, it was not clear how frequently the patients feet

were X-rayed and whether both groups had equal access to X-ray examination if osteomyelitis was present, which tends to occur in 15% of patients with DFU.19 Moreover, only 1 trial used limb savage as an endpoint, others did not mention how many patients had amputations in each group. As mentioned above, amputation placed a heavy burden on both the patients families and society, and should be considered as an important endpoint for evaluating an intervention. Innovations and limitations of this meta-analysis To the best knowledge of the authors, this is the rst metaanalysis of CHM used as an adjunct for patients with DFU. It was noted that a planned Cochrane systematic review of Chinese medicine is ongoing; however, no results were obtainable as of this writing. In addition to the databases this Cochrane review protocol aimed to search, two databases CBM and AMEDwere searched in the current metaanalysis. This meta-analysis had some limitations: (1) Only 6 RCTs were included and it was a small-sample study. (2) Only trials using CHM orally were included, whereas many studies use CHM as wound dressings or wound dressings combined with oral administration. (3) Only reports of RCTs were assessed, not the trials themselves. (4) No sensitivity and subgroup analyses were applied. The rst limitation will only be solved in future analyses, when there are many of RCTs, ideally, high-quality RCTs. With respect to the second limitation, rst, including only orally used CHM trials helps avoid confounding issues. For example, if the current meta-analysis included trials using CHM both orally and topically, and they worked well, it might have been difcult to determine CHM worksorally or topically? Second, according to the literature retrieved, CHM for topical use was prepared as plasters, usually, with their ingredients unknown. On the contrary, ingredients of

896

CHEN ET AL.

FIG. 2. Outcome measurements after treatment. Chinese herbal medicine combined with standard therapy versus standard therapy alone. *Patients without any improvement refers to those who had no or little reduction in the ulcer area (e.g., less than 30% of the ulcer area), still needed amputation, etc. CI, condence interval.

CHM for oral use were listed in detail. In addition, oral administration is a popular form of CHM, so only studies using CHM orally were selected for inclusion in the current metaanalysis. In future studies, when the ingredients of CHM for topical use are disclosed and clearly presented, it will be necessary and meaningful to include trials using CHM to-

pically for analysis. With regard to the third limitation, failure to report the methods of a trial does not necessarily mean that researchers did not use these methods. Some methodological insufciencies may lie in the reports of trials rather than in their performance.31 However, the current authors were not able contact most of the researchers, because there

CHINESE HERBAL MEDICINE FOR DIABETIC FOOT ULCERS were no exact e-mails, fax numbers, or addresses to get in touch with them although phone numbers were provided within the reports. However, according to the reports in the articles, it was still possible to analyze the data and discern the limitations of the trials. Considering the fourth limitation, the current authors agree that it is important to conduct subgroup analyses to different grades of patients with DFU divided by ulcer classication systems (e.g., the Wagner or the university of Texas wound classication systems).32 However, lack of enough statistical power because of the small sample size, normally, will be not be enough to ensure the primary hypothesis being adequately tested.33 So subgroup analyses were not planned for this review. Conclusions This meta-analysis showed that CHM may be benecial as an adjunctive for patients with DFU. However, it is difcult to draw a rm conclusion, because of insufcient highquality evidence. This meta-analysis does suggest that largesample-size and well-designed RCTs are needed to justify use of CHM in clinical practice further. Acknowledgments This work was nancially supported by grants from the National Natural Science Foundation of China (Grant No. 30772806). Disclosure Statement There are no conicts of interest for any of the authors. References
1. Boulton AJ, Kirsner RS, Vileikyte L. Clinical practice: Neuropathic diabetic foot ulcers. NEMJ 2004;351:4855. 2. Singh N, Armstrong DG, Lipsky BA. Preventing foot ulcers in patients with diabetes. JAMA 2005;293:217228. 3. Ribu L, Rustoen T, Birkeland K, et al. The prevalence and occurrence of diabetic foot ulcer pain and its impact on health-related quality of life. J Pain 2006;7:290299. 4. Nabuurs-Franssen MH, Huijberts MS, Nieuwenhuijzen Kruseman AC, et al. Health-related quality of life of diabetic foot ulcer patients and their caregivers. Diabetologia 2005; 48:19061910. 5. Global Lower Extremity Amputation Study Group. Epidemiology of lower extremity amputation in centres in Europe, North America and East Asia: The Global Lower Extremity Amputation Study Group. Br J Surg 2000;87:328337. 6. Boulton AJ, Vileikyte L, Ragnarson-Tennvall G, Apelqvist J. The global burden of diabetic foot disease. Lancet 2005;366: 17191724. 7. Wild S, Roglic G, Green A, et al. Global prevalence of diabetes: Estimates for the year 2000 and projections for 2030. Diabetes Care 2004;27:10471053. 8. Registered Nurses Association of Ontario. (RNAoO). RNAoO: Assessment and Management of Foot Ulcers for People with Diabetes. Toronto: RNAoO), 2005. 9. Lau TW, Chan YW, Lau CP, et al. Radix Astragali and Radix Rehmanniae, the principal components of two antidiabetic foot ulcer herbal formulae, elicit viability-promoting effects on primary broblasts cultured from diabetic foot ulcer tissues. Phytother Res 2009;23:809815.

897

10. Lau CH, Chan CM, Chan YW, et al. In vitro antidiabetic activities of ve medicinal herbs used in Chinese medicinal formulae. Phytother Res 2008;22:13841388. 11. Chan CM, Chan YW, Lau CH, et al. Inuence of an antidiabetic foot ulcer formula and its component herbs on tissue and systemic glucose homeostasis. J Ethnopharmacol 2007;109:1020. 12. Wong MW, Leung PC, Wong WC. Limb salvage in extensive diabetic foot ulcerationa preliminary clinical study using simple debridement and herbal drinks. Hong Kong Med J 2001;7:403407. 13. Lu W, Zhong GL. Clinical observation on treatment of diabetic foot by integrative Chinese and Western Medicine [in Chinese]. Zhongguo Zhong Xi Yi Jie He Za Zhi 2003;23:911913. 14. Xie XS, Wang YJ, Zuo C, et al. A case report of an effective treatment for diabetic foot ulcers with integration of Traditional Chinese Medicine and Western medicine. J Diabetes Complications 2009;23:360364. 15. Goldstein LH, Elias M, Ron-Avraham G, et al. Consumption of herbal remedies and dietary supplements amongst patients hospitalized in medical wards. Br J Clin Pharmacol 2007;64:373380. 16. Watkins PJ. ABC of diabetes: The diabetic foot. BMJ. 2003; 326:977979. 17. Yeh GY, Eisenberg DM, Kaptchuk TJ, Phillips RS. Systematic review of herbs and dietary supplements for glycemic control in diabetes. Diabetes Care 2003;26:12771294. 18. Wang E, Wylie-Rosett J. Review of selected Chinese herbal medicines in the treatment of type 2 diabetes. Diabetes Educ 2008;34:645654. 19. Ramsey SD, Newton K, Blough D, et al. Incidence, outcomes, and cost of foot ulcers in patients with diabetes. Diabetes Care 1999;22:382387. 20. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: Is blinding necessary? Control Clin Trials 1996;17:112. 21. van Tulder M, Furlan A, Bombardier C, Bouter L. Updated method guidelines for systematic reviews in the Cochrane Collaboration back review group. Spine 2003;28:12901299. 22. Xing YD. Clinical observation of She-xiang-huo-xue capsule for patients with diabetic foot. J Liaoning Univ Trad Chin Med 2008;10:142143. 23. Song YH, Liu MJ, Li YQ, Chen L. Clinical observation of 25 cases of diabetes with diabetic foot treated by Simiao Yongan Tang. J Sichuan Trad Chinese Med 2006;24:64. 24. Deng JH, Zhang JD, Wang GS, et al. Clinical observation of using Bu-yang-huan-wu decoction to treat diabetic foot. J Sichuan Trad Chinese Med 2007;25:8182. 25. He ZD. Clinical observation on 40 cases of diabetic gangrene treated with jiawei Taohong Siwu Tang. Zhejiang J Trad Chinese Med 2008;43:202203. 26. Wang XG, Jiang J. Clinical analysis of using intergrative medicine for patients with diabetic foot. Jilin J Trad Chin Med 2006;26:4041. 27. Wang G, Mao B, Xiong ZY, et al. The quality of reporting of randomized controlled trials of Traditional Chinese Medicine: A survey of 13 randomly selected journals from mainland China. Clin Ther 2007;29:14561467. 28. Leung PC, Wong MW, Wong WC. Limb salvage in extensive diabetic foot ulceration: An extended study using a [sic] herbal supplement. Hong Kong Med J 2008;14:2933. 29. Chen C, Venketasubramanian N, Gan RN, et al. Danqi Piantang Jiaonang (DJ), a Traditional Chinese Medicine, in poststroke recovery. Stroke 2009;40:859863.

898
30. Tang ZL, Wang X, Yi B, et al. Effects of the preoperative administration of Yunnan Baiyao capsules on intraoperative blood loss in bimaxillary orthognathic surgery: A prospective, randomized, double-blind, placebo-controlled study. Int J Oral Maxillofac Surg 2009;38:261266. 31. Hill CL, LaValley MP, Felson DT. Discrepancy between published report and actual conduct of randomized clinical trials. J Clin Epidemiol 2002;55:783786. 32. Oyibo SO, Jude EB, Tarawneh I, et al. A comparison of two diabetic foot ulcer classication systems: The Wagner and the University of Texas wound classication systems. Diabetes Care 2001;24:8488.

CHEN ET AL.
33. Cook DI, Gebski VJ, Keech AC. Subgroup analysis in clinical trials. Med J Aust 2004;180:289291.

Address correspondence to: Chun-Guang Xie, MD, PhD First Afliated Hospital of Chengdu University of Traditional Chinese Medicine 39# Shi-er-qiao Road Chengdu 610075, Sichuan China E-mail: chunguang.xie@yahoo.com

Copyright of Journal of Alternative & Complementary Medicine is the property of Mary Ann Liebert, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.