Printer Friendly

A systematic review of randomised clinical trials of Tripterygium wilfordii for rheumatoid arthritis.


Tripterygium wilfordii is a Chinese herb with immunosuppressive effects and an established history of use in the treatment of rheumatoid arthritis (RA). We have carried out a systematic review of randomised clinical trials (RCTs) which assess the effectiveness of T. wilfordii in this indication. We included only randomised and controlled studies which tested the effectiveness of T. wilfordii monopreparations in the treatment of RA. Studies in any language were included. A search of five electronic databases from inception to February 2005 identified 18 articles which could potentially meet our inclusion criteria. Only 16 of these could be retrieved from the scientific literature and after reading these in full, only two unique RCTs meeting our inclusion criteria were identified. Both indicated that T. wilfordii has beneficial effects on the symptoms of RA. However, the literature indicates that T. wilfordii is associated with serious adverse events which make the risk-benefit analysis for this herb unfavourable. Therefore, we cannot recommend its use.

[c] 2006 Elsevier GmbH. All rights reserved.

Keywords: Rheumatoid arthritis; Tripterygium; systematic review; adverse events


Like many other chronic conditions rheumatoid arthritis (RA) is associated with a high level of complementary and alternative medicine use and in particular, herbal treatments (Resch et al. 1997). Among these, the extract from the Chinese herb Tripterygium wilfordii has a long history of use in RA. However, early hot water decoctions produced many adverse effects and two new preparations, an ethyl acetate extract and a chloroform-methanol extract named T2, were developed in the 1970s in an attempt to reduce toxicity (Tao and Lipsky 2000). Root extracts are therapeutically active and appear to induce fewer side effects than extracts from the leaves and stem of the herb and more than 70 compounds have been identified in the roots including diterpinoids, triterpinoids, sesquiterpinoids, alkaloids, [beta]-sitosterol, dulcitol and glycosides (Tao and Lipsky 2000). Numerous preclinical studies have demonstrated immunosuppressive and anti-inflammatory effects in vitro and in vivo (Asano et al. 1998; Chang et al. 1993, 1997; Tao et al. 1998; Ho et al. 1999) and the diterpinoids appear to account for most of this activity, in particular the triptolide and tripdiolide fractions (Tao and Lipsky 2000).

Several studies describing themselves as clinical trials of T. wilfordii have been published, often in Chinese, and including some multiple publications. A Cochrane review of herbal therapies for RA, conducted in 2000, included only one randomised clinical trial (RCT) of T. wilfordii, and based on the limited evidence available, concluded that the extract may be effective but that there were safety concerns (Little and Parsons 2000). Apparently reflecting similar concerns, T. wilfordii as a single herb was eliminated from the English Edition of the Chinese Phamacopoeia in the same year. Nevertheless, the herb remains available for sale on the Internet where it is recommended as a treatment for RA. It is used mainly as a monopreparation and is not available commercially as part of any herbal formulas. Some herbalists may, however, sometimes use T. wilfordii as part of complex herbal formulae and the herb is included in at least two TCM formulae. These formulae are to "improve the liver and kidney" (Herba Epimedii, Radix Achyranthis bidentatae, Radix Dipsaci, Rhizoma Polygonati, Cornu Cervi degelatinatum, Radix Gentianae macrophyllae, Agkistrodon acutus, at 15 g each; Caulis Spatholobi, Semen Coicis, at 30 g each; Radix Aconiti praeparata, T. wilfordii, at 10 g each; Herba Asari 6 g) or to "release humidity and heat" (Cortex Phellodendri, Rhizoma Atractylodis, Ramulus Mori, Radix Achyranthis bidentatae, Agkistrodon actus, Radix Stephaniae tetrandrae, Talcum, at 15 g each; Semen Coicis, Caulis Piperis futokadsurae, at 30 g each; T. wilfordii 20 g).

In the light of the continuing use of the herb, we have carried out an updated systematic review of RCTs testing the effectiveness of T. wilfordii for the treatment of RA which includes publications up to February 2005.


Literature searches were made of the MEDLINE, EMBASE, AMED, CINAHL and CCTR (Cochrane) databases from inception of database to February 2005. The search terms were RA, arthritis, arthrosis, ankylosing spondylitis, rheumatoid, rheumatism, felty's syndrome, caplan's syndrome and T. wilfordii, Tripterygium, thunder god vine, lei gong teng, seven-step vine. Studies in any language were screened against the following inclusion criteria: (a) human subjects, (b) use of a control procedure, (c) subjects randomised between treatment conditions, (d) testing a monopreparation of T. wilfordii, (e) allowing assessment of the efficacy of T. wilfordii and (f) subjects diagnosed with RA. Initial assessment against inclusion criteria was made by reading abstracts. Articles apparently meeting the inclusion criteria at this stage were then read in full by one of the two authors according to the language of publication. Research articles in English were read by PHC and research articles in Chinese by HL. These authors then discussed the articles and made decisions to include or exclude. Data were extracted from included articles by the same two authors using a pro forma data extraction sheet. Quality of reporting was assessed using the five-point Jadad scale (Jadad et al. 1996).


The searches described located 83 articles of which 18 appeared to be clinical trials. Two of these could not be located despite carrying out worldwide Internet searches. The first of these (Zhang et al. 1981) is described in the title as a preliminary clinical study so may not be randomised. The second (Huang and Shen 1989) is described in the title as an RCT but is cited only as a conference proceeding. Abstracts are not available for either reference. The other 16 articles were retrieved, read in full and 12 were excluded. Reasons for excluding trials were failure to randomise (Tao et al. 1990; Su et al. 1990; Xu et al. 1996) comparing different forms of T. wilfordii without any other control group (Li et al. 1996; Du et al. 1998), using a combination product in a way which prevented evaluation of T. wilfordii (Wu et al. 2001; Deng et al. 1998; Cibere et al. 2003), absence of a control group (Tao et al. 2001; RA Cooperative Group 1977), or using T. wilfordii as the control treatment in a trial of another intervention of unknown efficacy (Ji et al. 2002; Zhou et al. 2004). Of the remaining four articles, three described the same clinical trial (Tao et al. 1987a,b, 1989). The two trials included are summarised in Table 1 and are described here in narrative form.

Tao et al. (1987a,b, 1989) tested the efficacy of the standardised polyglycoside extract T2, a chloroform-methanol extract of T. wilfordii manufactured by Jiansu Taizhou Pharmaceuticals. Diterpinoid content ([micro]g/mg extract) of T2 has been characterised as 0.36 triptolide, 0.68 tripdiolide, 0.03 triptonide and 0.04 triptophenolide (Tao and Lipsky 2000). Patients enrolled into the trial had adult-onset RA of at least 6 months standing, had active symptoms and had been non-responsive to NSAID treatment of 2 months or longer. Active symptomology was defined as meeting at least four of the following criteria: tenderness score > 11; duration of morning stiffness > 30 min; four or more swollen joints; mean grip strength < 10.7kPa (female) or < 13.3kPa (male); 15 m walking time > 17s; erythrocyte sedimentation rate (ESR) > 30 mm/h. Seventy patients were randomised to two parallel groups. Baseline characteristics, including disease stage ascertained by X-ray, and duration of disease were similar at baseline. Group A was treated with T. wilfordii extract (60 mg/day) for 12 weeks and with placebo for the subsequent 4 weeks. Group B was treated with placebo for 12 weeks followed by T. wilfordii for 4 weeks. A stable dose of NSAID (ibuprofen 600mg/day or indomethacin suppository 100 mg/day) was continued throughout the trial in all patients except those whose symptoms were markedly relieved during the trial. A stable dose of prednisone (up to 7.5 mg/day) was allowed to continue in two patients. Clinical assessment and determination of ESR and rheumatoid factor (RF) were carried out every 4 weeks. The clinical assessment included physician-rated joint tenderness (0-3 scale summed for 66 diarthrodial joints), total number of swollen joints (out of 60 diarthrodial joints), grip strength (mean of both hands), 15 m walking time, duration of morning stiffness on the day of the assessment, and patient-rated and physician-rated overall assessment (classified as worsen, no change, improved or significantly improved). Immunoglobulin G (IgG), Immunoglobulin M (IgM), Immunoglobulin A (IgA) and serum C-reactive protein (CRP) were measured at baseline and after 12 weeks.

In Group A, 27 of 35 patients completed the 12-week T. wilfordii treatment and 24 completed the subsequent 4-week placebo treatment. In Group B, 31 of 35 completed the 12-week placebo treatment and 27 completed the subsequent 4-week T. wilfordii treatment. In a per protocol analysis, Group A compared to Group B showed statistically significant improvement (p<0.05 or better) after 12 weeks in tenderness score, swelling count, duration of morning stiffness, mean grip strength, 15 m walking time, ESR, CRP, IgG, IgM, IgA, and physician-rated and patient-rated overall assessments. There was no statistically significant change in RF titre.

The trial design does not allow comparison between groups or treatment conditions following the crossover at 12 weeks. Within-group comparisons before and after the second phase verum treatment in Group B showed improvements in tenderness score, swelling count, grip strength, ESR, RF, and physician-rated and patient-rated overall assessments, but not duration of morning stiffness or 15 m walking time. Within-group comparisons for Group A who received placebo in the second phase showed statistically significant improvement only in grip strength but patients' overall assessment significantly deteriorated during the second phase.

Adverse effects (Table 2) were more frequently associated with verum than placebo treatment and markedly so for skin rashes and cheilosis (n = 22), diarrhoea (n = 8) and amenorrhoea (n = 6). Four patients from Group A withdrew from T. wilfordii treatment because of adverse reactions. One of these withdrawals was a 27-year-old man who developed feverishness and aplastic anaemia following mistaken medication with 180 mg/day of the T. wilfordii extract for 2 weeks. He recovered 2 weeks after discontinuation of T. wilfordii treatment and daily treatment with 30 mg prednisone. Another patient, a 58-year-old woman with long-standing RA and diffuse fibrotic pulmonary tuberculosis, had developed heart failure during the previous 3 years. She developed anorexia after 1 month of T. wilfordii treatment, withdrew, and died of chronic heart failure 1 month later.

Tao et al. (2002) randomised 35 patients with RA of at least 1 year standing to receive either a high dose (360 mg/day) of a T. wilfordii ethanol/ethyl alcohol extract, a low dose (180 mg/day) or placebo for 20 weeks. The extract was prepared from peeled roots of T. wilfordii at the University of Texas Southwestern Medical Center. The T. wilfordii extract was standardised to a previously tested ethanol/ethyl alcohol extract (Tao et al. 2001) (30 [micro]g triptolide plus tripdiolide per 60 mg capsule). Patients met American College of Rheumatology (ACR) criteria and were in functional class II, III or IV, and had active symptoms despite treatment with disease-modifying antirheumatic drugs (DMARDs). Active symptoms were defined as "at least two swollen joints and two out of three of the following: at least six painful/tender joints, morning stiffness of 30 min or longer duration, ESR of 28 mm/h or higher". Stable doses of prednisone and NSAIDs were continued throughout the trial. Disease activity and treatment response according to ACR criteria were evaluated at baseline and every 4 weeks by an assessor blinded to treatment allocation. Duration of morning stiffness and RF were also determined.

At baseline the patients in the high-dose group had suffered RA for longer than those in either of the other two groups (p<0.01) and both T. wilfordii groups had more males than the placebo group (p = 0.0146). Eight of 11 patients in the high-dose group, seven of 12 in the low-dose group and six of 12 in the placebo group completed 20 weeks of treatment. The number of patients completing 4 weeks of treatment and qualifying for the modified intention to treat analysis was 10 in the high-dose group, 10 in the low-dose group and 12 in the placebo group. Of these eight (80%) in the high-dose group, four (40%) in the low-dose group and none in the placebo group achieved an improvement meeting ACR-20 criteria. The difference in response rate between high dose and placebo groups reached statistical significance (p = 0.0001). High dose was more effective than low dose (p = 0.027) and low dose more effective than placebo (p = 0.0287). Five patients in the high-dose group and one patient in the low-dose group met criteria for ACR-50 and one patient in the high-dose group met criteria for ACR-70. For those patients who did reach ACR-20, the mean time taken was 7 weeks in the high-dose group and 12 weeks in the low-dose group. The individual components of ACR-20, namely number of tender joints, number of swollen joints, patient assessed pain, patient and physician global assessments, and patient-rated physical function, and ESR and CRP all showed significant improvements in the high-dose group relative to the placebo group by 4 weeks of treatment. The secondary measure, duration of morning stiffness, declined from 145 min at baseline to 37 min at 4 weeks and then to 26 min by 20 weeks in the high-dose group. RF titre also decreased significantly within group for the high dose but inter-group statistical comparisons were not made for RF or morning stiffness. Data from the open-label extension to the trial are not presented here.

There were more adverse reactions (Table 2) in the high-dose group (15 among 11 patients) than either low dose (12 among 12 patients) or placebo group (6 among 12 patients) reported for the first 20 weeks of the trial. The authors analysed this in terms of number of patients experiencing more than one adverse event in each group, an analysis that rather misleadingly suggests little difference between groups. The most common adverse reactions exclusively associated with verum treatment were diarrhoea (seven cases), hair loss (three cases) and nausea (three cases). One 50-year-old woman in the high-dose group developed headache, diarrhoea and vaginal spotting and recovered 1 week after discontinuing treatment. One 29-year-old woman from the low-dose group who was subsequently treated with the high dose for 20 weeks in the open-label extension to the trial developed amenorrhoea. Menstruation resumed 1 year after discontinuing treatment.


The trial conducted by Tao et al. (1987a, b, 1989) is described as a crossover trial but does not conform to the usual design for such a trial. The unequal time periods in the two treatment phases means that data cannot be combined in such a way as to exploit the additional statistical power which a crossover design can offer. It also lacked a washout period which is required to allow for carry-over effects in patients receiving verum treatment first. Such carry-over effects seem to be evident in the continuing improvement in hand grip observed in such patients. The trial, therefore, largely reduces to a simple parallel group design with only data for the first 12 weeks being of scientific interest. The within-group changes observed in Group B who received T. wilfordii treatment in the second phase appear to be clinically significant but the trial design does not allow them to be compared statistically with changes in a comparable control group. Data for Phase 2 in Group B (verum 4 weeks) could perhaps, have been compared statistically to that for Phase 1 in Group A (verum 12 weeks) to ascertain the effect of treatment period and the results reported apparently suggest that a longer treatment period produces more improvement. The validity of such a comparison is, however, conditional upon adequate blinding being maintained throughout both phases of the trial and an assumption that neither disease progression nor any other factor differentially affected the outcome during the intervening 3 months.

Tao et al. (2002) carried out a power calculation which suggested that 90 patients were required but the trial was terminated prematurely because the investigators relocated after 35 patients were enrolled. These authors used an ethanol/ethyl alcohol extract on the basis of their previous experience which suggested that the therapeutic effect was similar to other extracts but adverse reactions were less severe. Although this appears to be the case when judged by withdrawals because of adverse reactions in the three arms of the trial, adverse reactions were more frequent in the verum group and the risk of amenorrhoea still appears to be present.

The limited evidence available from the two studies included in this review suggests that T. wilfordii is an effective treatment for RA. This view is supported by case studies and uncontrolled studies in the literature (e.g. Li et al. 1996; Tao et al. 2001). Unfortunately, the extract is also associated with serious adverse events. The association with dysmenorrhoea is well documented (e.g. Tao et al. 1987b; Jiang et al. 2002; Zhou et al. 1999; Gu 1989). There is an effect on male fertility in rats and humans which has been observed at around one-third the dose normally used in the treatment of RA (Qian 1987), an effect sufficiently important to stimulate research into the use of the extract as a male contraceptive (Zhen et al. 1995). An adverse effect on renal function in older patients is also documented (Zhou et al. 1999). There is at least one incident where induction of hypertension appears to have occurred (Tao et al. 2001). There is a case study of a previously healthy young man who developed profuse vomiting and diarrhoea, leukopenia, renal failure, profound hypotension and shock after ingestion of the extract (Chou et al. 1995). Therapeutic concentrations of the extract have a haematotoxic effect which inhibits the growth factor response in bone marrow cells (Pyatt et al. 2000). Abnormal development of mouse embryos has been observed in vitro (Chan and Ng 1995) and there is at least one report of human occipital meningoencephalocele associated with use of the extract by the mother (Takei et al. 1997). The immunosuppressive effects of T. wilfordii may also leave patients susceptible to infectious diseases (Guo et al. 1981).

The triptolide fraction appears to be important in producing the therapeutic effects of T. wilfordii extracts but is also implicated in the toxic effects. A clinical trial comparing isolated triptolide (0.5-0.75 mg/day) with the ethyl alcohol extract (120 mg/day) in the treatment of RA reports significant improvements in both groups but more frequent and more severe adverse effects in the group treated with triptolide (Su et al. 1990). Toxicological data are available on both the ethyl acetate and the methanol-alcohol (T2) extracts. For the ethyl alcohol extract the lethal dose for 50% of mice (LD50) is 608-858 mg/kg depending on the source of the plant material and the time of harvest (Zheng et al. 1983). Subacute toxicity testing of rats for 6 months at one-sixteenth to one-fourth of the LD50 dose revealed pathological changes mainly in the lymphatic and reproductive systems (Wang et al. 1991). For T2 the LD50 in mice is reported as 159.7 mg/kg and in mice, rats and dogs treated for 80 days with doses of 1/4 to 1/24 of LD50, pathological changes were mainly observed in the testes (Tao and Lipsky 2000).

On this basis of the evidence presented here, we conclude that the risk-benefit analysis for the use of T. wilfordii in RA is negative.


There is only very limited evidence from rigorous clinical trials regarding the effectiveness of T. wilfordii as a treatment for RA. Based on the two RCTs which do exist, it appears that the extract is effective in treating the symptoms of RA. However, the occurrence of serious adverse events is sufficiently frequent to make the risk-benefit analysis for this herbal negative. We cannot, therefore, recommend its use in the treatment of RA or any other disease.


Asano, K., Matsuishi, J., Yu, Y., Kasahara, T., Hisamitsu, T., 1998. Suppressive effects of Tripterygium wilfordii Hook F., a traditional Chinese medicine, on collagen arthritis in mice. Immunopharmacology 39 (2), 117-126.

Chan, W.Y., Ng, T.B., 1995. Adverse effect of Tripterygium wilfordii extract on mouse embryonic development. Contraception 51 (1), 65-71.

Chang, M.L., Yang, L.L., Chang, D.M., Kuo, S.Y., Chu, S.J., 1993. [The influence of Chinese traditional medicine on the production and activity of interleukin 1 (IL-1)]. Zhonghua Min Guo Wei Sheng Wu Ji Mian Yi Xue Za Zhi 26 (1), 15-24.

Chang, D.M., Chang, W.Y., Kuo, S.Y., Chang, M.L., 1997. The effects of traditional antirheumatic herbal medicines on immune response cells. J. Rheumatol. 24 (3), 436-441.

Chou, W.C., Wu, C.C., Yang, P.C., Lee, Y.T., 1995. Hypovolemic shock and mortality after ingestion of Tripterygium wilfordii Hook F.: a case report. Int. J. Cardiol. 49 (2), 173-177.

Cibere, J., Deng, Z., Lin, Y., Ou, R., He, Y., Wang, Z., et al., 2003. A randomized double blind, placebo controlled trial of topical Tripterygium wilfordii in rheumatoid arthritis: reanalysis using logistic regression analysis. J. Rheumatol. 30 (3), 465-467.

Deng, Z., Lin, Y., Ou, R., He, Y., Wang, Z., 1998. A further study on compound Tripterygium wilfordii Hook film in treating rheumatoid arthritis at active stage. J. Guangzhou Univ. Tradit. Chin Med. 15, 180-183.

Du, X., Zhang, H., Fu, X., 1998. [Clinical study on Tripterygium wilfordii complex ester tablet in treating rheumatoid arthritis]. Zhongguo Zhong Xi Yi Jie He Za Zhi 18 (2), 88-91.

Gu, C.X., 1989. [Cause of amenorrhea after treatment with Tripterygium wilfordii F.]. Zhongguo Yi Xue Ke Xue Yuan Xue Bao 11 (2), 151-153.

Guo, J.L., Yuan, S.X., Wang, X.C., Xu, S.X., Li, D.D., 1981. Tripterygium wilfordii Hook in rheumatoid arthritis and ankylosing spondylitis. Preliminary report. Chin. Med. J. 94, 405-412.

Ho, L.J., Chang, D.M., Chang, M.L., Kuo, S.Y., Lai, J.H., 1999. Mechanism of immunosuppression of the antirheumatic herb TWHf in human T cells. J. Rheumatol. 26 (1), 14-24.

Huang, Cf., Wu, C.B., Shen, Z.K., 1989. Study on the treatment of rheumatoid arthritis with polyglycosides of Tripterygium wilfordii Hook F.; a randomized, controlled, crossover clinical trial. In: Proceedings of the Third National Meeting of Rheumatology of China, Shijiazhuang, China, 1989. Chinese Association of Medicine, Beijing, China, 1989, pp. 31-33.

Jadad, A.R., Moore, R.A., Carrol, D., Jenkinson, C., Reynolds, D.J.M., Gavaghan, D.J., McQuay, H.J., 1996. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Contrib. Clin. Trials 17, 1-12.

Ji, H., Yao, S., Luo, Q., Cao, X., 2002. TCM treatment of rheumatoid arthritis by supplementing the kidney and invigorating the blood circulation. J. Tradit. Chin. Med. 22 (4), 252-255.

Jiang, L., Zhao, N., Ni, L., 2002. Retrospective study of adverse events in patients with rheumatoid arthritis treated with second-line drugs. Zhonghua Liu Xing Bing Xue Za Zhi 23 (3), 213-217.

Li, R.L., Liu, P.L., Wu, X.C., 1996. [Clinical and experimental study on sustained release tablet of Tripterygium wilfordii in treating rheumatoid arthritis]. Zhongguo Zhong Xi Yi Jie He Za Zhi 16 (1), 10-13.

Little, C.V., Parsons, T., 2000. Herbal therapy for treating rheumatoid arthritis. Cochrane Database Syst. Rev. (4).

Pyatt, D.W., Yang, Y., Mehos, B., Le, A., Stillman, W., Irons, R.D., 2000. Hematotoxicity of the Chinese herbal medicine Tripterygium wilfordii Hook F. in CD34-positive human bone marrow cells. Mol. Pharmacol. 57 (3), 512-518.

Qian, S.Z., 1987. Tripterygium wilfordii, a Chinese herb effective in male fertility regulation. Contraception 36, 335-345.

RA Cooperative Group, 1977. Clinical trial of Tripterygium wilfordii Hook in the treatment of 31 patients with rheumatoid arthritis. Wu Han Yi Xue Pao 6, 51-53.

Resch, K.L., Hill, S., Ernst, E., 1997. Use of complementary therapies by individuals with 'arthritis'. Clin. Rheumatol. 16, 391-395.

Su, D.F., Li, R.L., Sun, Y.J., 1990. Comparative study of triptolide and the ethyl acetate extract of Tripterygium wilfordii Hook F. in the treatment of rheumatoid arthritis. Zhong Cao Yao 10, 144-146, 131.

Takei, A., Nagashima, G., Suzuki, R., Hokaku, H., Takahashi, M., Miyo, T., et al., 1997. Meningoencephalocele associated with Tripterygium wilfordii treatment. Pediatr. Neurosurg. 27 (1), 45-48.

Tao, X., Lipsky, P.E., 2000. The Chinese anti-inflammatory and immunosuppressive herbal remedy Tripterygium wilfordii Hook F. Complementary and alternative therapies for rheumatic diseases II. Rheumatic Disease Clinics of North America 26, 29-50.

Tao, X.L., Dong, Y., Zhang, N.Z., 1987a. [A double-blind study of T2 (tablets of polyglycosides of Tripterygium wilfordii Hook) in the treatment of rheumatoid arthritis]. Zhonghua Nei Ke Za Zhi 26 (7), 399-402.

Tao, X.L., Sun, Y., Dong, Y., Xiao, Y.L., Hu, D.W., Shi, Y.P., 1987b. A double blind, controlled trial of Tripterygium wilfordii glycosides in the treatment of rheumatoid arthritis. Arthritis Rheum. 30 (Suppl. 5), S59.

Tao, X.L., Sun, Y., Dong, Y., Xiao, Y.L., Hu, D.W., Shi, Y.P., et al., 1989. A prospective, controlled, double-blind, cross-over study of Tripterygium wilfordii Hook F. in treatment of rheumatoid arthritis. Chin. Med. J. (Engl.) 102 (5), 327-332.

Tao, X., Sun, Y., Zhang, N., 1990. [Treatment of rheumatoid arthritis with low doses of multi-glycosides of Tripterygium wilfordii]. Zhong Xi Yi Jie He Za Zhi 10 (5), 289-292.

Tao, X., Schulze-Koops, H., Ma, L., Cai, J., Mao, Y., Lipsky, P.E., 1998. Effects of Tripterygium wilfordii Hook F. extracts on induction of cyclooxygenase 2 activity and prostaglandin E2 production. Arthritis Rheum. 41 (1), 130-138.

Tao, X., Cush, J.J., Garret, M., Lipsky, P.E., 2001. A phase I study of ethyl acetate extract of the Chinese antirheumatic herb Tripterygium wilfordii Hook F. in rheumatoid arthritis. J. Rheumatol. 28 (10), 2160-2167.

Tao, X., Younger, J., Fan, F.Z., Wang, B., Lipsky, P.E., 2002. Benefit of an extract of Tripterygium wilfordii Hook F. in patients with rheumatoid arthritis: a double-blind, placebo-controlled study. Arthritis Rheum. 46 (7), 1735-1743.

Wang, Z.F., Zhang, Y.H., Huang, G.Z., 1991. An experimental, pathological study of subchronic intoxication of ethyl acetate extract of Tripterygium wilfordii Hook. Tong Ji Yi Ke Da Xue Pao 20, 247-250.

Wu, Y.J., Lao, Z.Y., Zhang, Z.L., 2001. [Clinical observation on small doses Tripterygium wilfordii polyglycoside combined with methotrexate in treating rheumatoid arthritis]. Zhongguo Zhong Xi Yi Jie He Za Zhi 21 (12), 895-896.

Xu, D.S., Shen, Z.Y., Lu, W., 1996. [Clinical and experimental study of RA mixture in treatment of rheumatoid arthritis]. Zhongguo Zhong Xi Yi Jie He Za Zhi 16 (1), 14-17, 23.

Zhang, N.Z., et al., 1981. A preliminary clinical study of T2 in the treatment of rheumatoid arthritis. Arthritis Rheum.

Zhen, Q.S., Ye, X., Wei, Z.J., 1995. Recent progress in research on Tripterygium: a male antifertility plant. Contraception 51 (2), 121-129.

Zheng, J.R., Liu, J.H., Xu, L.F., et al., 1983. Studies on toxicity of total glycosides in Tripterygium wilfordii. Acta Acad. Med. Sin. 5, 73-78.

Zhou, J., Zhu, Q., Yang, X., 1999. [Clinical observation of side effects of Tripterygium preparation]. Zhongguo Zhong Xi Yi Jie He Za Zhi 19 (2), 77-79.

Zhou, X., Zhou, Z., Jin, M., Wang, H., Wu, M., Song, Y., et al., 2004. Clinical study of qingluo tongbi granules in treating 63 patients with rheumatoid arthritis of the type of yin-deficiency and heat in collaterals. J. Tradit. Chin. Med. 24 (2), 83-87.

P.H. Canter (a,*), Hyang Sook Lee (b), E. Ernst (a)

(a) Complementary Medicine, Peninsula Medical School, Universities of Exeter & Plymouth, 25 Victoria Park Rd, Exeter, Devon EX2 4NT, UK

(b) Department of Meridianology, College of Oriental Medicine, SangJi University, Wonju, Korea

*Corresponding author. Tel.: +44 1392 439 035; fax: +44 1392 424 989.

E-mail address: (P.H. Canter).
Table 1. Summary of randomised clinical trials included in the
systematic review

Reference Design score Subjects

Zhou et al. (2004) Randomised, placebo 3 70 patients with adult
and Tao et al. controlled, double onset RA for > 6 months,
(1987a, b) blind, quasi- active symptoms,
 crossover trial non-responsive to NSAID
 treatment. Mean age in
 treatment groups 46 and
 48 years
Tao et al. (2002) Randomised, placebo 5 35 patients with RA
 controlled, double meeting ACR criteria for
 blind, parallel at least 1 year,
 group trial functional class II, III
 or IV, and active
 symptoms despite DMARDs
 treatment, aged 18-75

Reference Treatment Main results

Zhou et al. (2004) 60 mg/day for 12 weeks Group A vs. group B after 12
and Tao et al. (T2 polyglycosides of weeks; improved tenderness
(1987a, b) Tw, chloroform- score, swelling count,
 methanol extract) then duration of morning
 4 weeks placebo (group stiffness, mean grip
 A), or 12 weeks strength, 15 m walking time,
 placebo then 4 weeks ESR, CRP, IgG, IgA,
 Tw (group B) physician-rated and patient-
 rated overall assessments
 (p < 0.05 or better). No
 statistically significant
 change in RF titre
Tao et al. (2002) 180 mg/day Tw or 360 Modified ITT analysis of
 mg/day Tw or placebo patients completing 4 weeks
 for 20 weeks (ethanol/ treatment: response rate for
 ethyl alcohol root high dose > placebo (p =
 extract of Tw) 0.0001), high dose > low
 dose (p = 0.027), low dose >
 placebo (p = 0.0287) (8 of
 10 high dose, 4 of 10 low
 dose, 0 of 12 placebo met

Tw, Tripterygium wilfordii; ESR, erythrocyte sedimentation rate; CRP,
serum C-reactive protein; IgG, immunoglobulin M; IgA, immunoglobulin A;
RF, rheumatoid factor; ACR, American College of Rheumatology; DMARDs,
disease-modifying antirheumatic drugs.

Table 2. Adverse events reported in randomised clinical trials included
in the systematic review

 Number of
 (number in each
Reference Extract treatment group)

Zhou et al. T2 N = 70 (1st 12 Adverse events in first 12
(2004) and chloroform- weeks: placebo week period (placebo, T2)
Tao et al. methanol n = 31, T2 n =
(1987a, b) extract of Tw 27)
 Skin rash and cheilosis
 (1, 15)
 Diarrhoea (0, 6)
 Anorexia (0, 2)
 Abdominal pain (1, 2)
 Amenorrhoea (0, 5)
 Postmenopausal vaginal
 bleeding (0, 1)
 Total (2, 31)
 (4 weeks after Adverse events in 4 weeks
 crossover: after crossover (placebo,
 placebo n = 24, T2)
 T2 n = 25)
 Skin rash and cheilosis
 (0, 7)
 Diarrhoea (0, 2)
 Anorexia (1, 0)
 Amenorrhoea (5, 1)
 Withdrawals due to AEs:
 four from T2 treatment in
 first 12 weeks including
 one case of feverishness
 and aplastic anaemia after
 mistaken medication with
 180 mg/day of T2 for 2
 weeks and one case of
 anorexia and death through
 heart failure. Reasons for
 the other two withdrawals
 are not reported
Tao et al. Ethanol/ethyl N = 35 (placebo Adverse events in first 20
(2002) alcohol root n = 12, low dose weeks (placebo, low dose,
 extract of Tw n = 12, high high dose)
 dose n = 11)
 Diarrhoea (0, 4, 3)
 Headache (2, 1, 1)
 Hair loss (0, 2, 1)
 Blisters (1, 0, 1)
 Nausea (0, 2, 1)
 Flatulence (1, 0, 1)
 Constipation (1, 1, 1)
 Heartburn (0, 2, 0)
 Facial rash (0, 0, 1)
 Tinnitus (0, 0, 1)
 Abdominal pain (0, 0, 1)
 Vaginal spotting (0, 0, 1)
 Indigestion (1, 0, 1)
 Total (6, 12, 15)
 Open label extension:
 similar AEs to the above
 plus one case of
 amenorrhoea after 20 weeks
 on high dose
COPYRIGHT 2006 Urban & Fischer Verlag
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Canter, P.H.; Lee, Hyang Sook; Ernst, E.
Publication:Phytomedicine: International Journal of Phytotherapy & Phytopharmacology
Article Type:Clinical report
Geographic Code:9JAPA
Date:May 1, 2006
Previous Article:Antitumor and immunostimulating effects of Anoectochilus formosanus Hayata.
Next Article:The first HMPC community herbal monographs.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters