Cochrane Database of Systematic Reviews

Tripterygium wilfordii Hook F (a traditional Chinese medicine) for primary nephrotic syndrome

  • Review
  • Intervention

Authors

  • Yizhi Chen,

    1. Chinese People's Liberation Army (PLA) General Hospital (301 Hospital), Chinese PLA Medical Academy, Division of Nephrology, State Key Discipline and State Key Laboratory of Kidney Diseases (2011DAV00088), Beijing, China
    Search for more papers by this author
  • Zhixiang Gong,

    1. Chinese PLA 532 Hospital, Division of Infectious Diseases, Huangshan, Anhui, China
    Search for more papers by this author
  • Xiangmei Chen,

    1. Chinese People's Liberation Army (PLA) General Hospital (301 Hospital), Chinese PLA Medical Academy, Division of Nephrology, State Key Discipline and State Key Laboratory of Kidney Diseases (2011DAV00088), Beijing, China
    Search for more papers by this author
  • Li Tang,

    1. Chinese People's Liberation Army (PLA) General Hospital (301 Hospital), Chinese PLA Medical Academy, Division of Nephrology, State Key Discipline and State Key Laboratory of Kidney Diseases (2011DAV00088), Beijing, China
    Search for more papers by this author
  • Xuezhi Zhao,

    1. Shanghai Chang Zheng Hospital, Division of Nephrology, Shanghai Kidney Disease Quality Control Center, Kidney Institute of Chinese PLA, Shanghai, Shanghai, China
    Search for more papers by this author
  • Qing Yuan,

    1. 309th Hospital of the People's Liberation Army, Organ Transplant Center, Beijing, China
    Search for more papers by this author
  • Guangyan Cai

    Corresponding author
    1. Chinese People's Liberation Army (PLA) General Hospital (301 Hospital), Chinese PLA Medical Academy, Division of Nephrology, State Key Discipline and State Key Laboratory of Kidney Diseases (2011DAV00088), Beijing, China
    • Guangyan Cai, Division of Nephrology, State Key Discipline and State Key Laboratory of Kidney Diseases (2011DAV00088), Chinese People's Liberation Army (PLA) General Hospital (301 Hospital), Chinese PLA Medical Academy, Fuxing Road 28, Haidian District, Beijing, 100853, China. caiguangyan@sina.com.

    Search for more papers by this author

Abstract

Background

Tripterygium wilfordii Hook F (TwHF), a traditional Chinese herbal medicine used as an immunosuppressive agent, has been prescribed in China for patients with primary nephrotic syndrome (NS) for more than two decades. Although patients with primary NS in China have benefited from TwHF treatment, its properties have not yet been fully understood.

Objectives

To assess the benefits and harms of TwHF for patients with primary NS.

Search methods

We searched the Cochrane Renal Group's specialised register (August 2012), Cochrane Register of Controlled Trials (CENTRAL, The Cochrane Library 2012, Issue 8), EMBASE (1966 to August 2012), and MEDLINE (1966 to August 2012). We also searched CBM (Chinese Biological Medical Database) (1978 to November 2010), CNKI (Chinese National Knowledge Infrastructure) (1979 to November 2010), VIP (ChongQing WeiPu Chinese Science and Technology Periodical Database) (1989 to November 2010), WanFang Database (1980 to November 2010), and reference lists of articles (6 November 2010).

Selection criteria

Only randomised controlled trials (RCTs) were included. Two standardised preparations of TwHF were investigated: ethanol-ethyl acetate extract and chloroform-methanol extract. All other TwHF preparations were excluded because of reported toxicities. Other traditional Chinese herbal medicines were also excluded. All included RCTs had a follow-up of at least three months.

Data collection and analysis

Data extraction and risk of bias assessment were undertaken independently by two authors. Where details of randomised sequence generation and allocation concealment were absent or inadequately reported, we contacted original study investigators for verification and details of the procedure. For dichotomous outcomes (remission and drug-related adverse events) we used risk ratio (RR) with 95% confidence intervals (95% CI) and mean difference (MD) for continuous outcomes (urinary protein excretion, serum albumin and serum creatinine).

Main results

Ten studies enrolling 630 participants were included. Overall, the quality of evidence was suboptimal due to the small number of included studies enrolling small numbers of participants; short follow-up in each study; only a few studies in each comparison category; and major concerns with methodological bias. Four studies (293 participants) contributed to the comparison of TwHF versus non-TwHF. TwHF significantly increased complete remission (RR 1.46, 95% CI 1.18 to 1.80) and complete or partial remission (RR 1.26, 95% CI 1.10 to 1.44) without escalating the adverse events profile at the last follow-up (12 to 16 months). Four studies (223 participants) compared TwHF with prednisone. There were no statistically significant differences between complete remission, partial remission, and complete or partial remission. Two studies (114 participants) contributed to the comparison of TwHF versus cyclophosphamide (CPA) at the last follow-up (3 to 12 months). There were no statistically significant differences between complete, partial, or complete or partial remission. One study (46 participants) reported TwHF was associated with a significantly lower serum creatinine compared with CPA (MD -14.00 μmol/L, 95% CI -26.43 to -1.57). No serious adverse events of TwHF were observed. One study (37 participants) reported TwHF was associated with a significantly lower risk of psychosis when compared to prednisone (RR 0.11, 95% CI 0.01 to 0.75), and two studies showed a significantly lower risk of hair loss with TwHF when compared to CPA ((2 studies, 114 participants): RR 0.11, 95% CI 0.02 to 0.59).

Authors' conclusions

TwHF may have an add-on effect on remission in patients with primary NS. There was insufficient evidence to assess if TwHF was as effective as prednisone or CPA. More methodologically sound and sufficiently powered studies, with adequate follow-up would help to better inform management options for the use of TwHF for primary NS. TwHF should be further directly compared with other widely used immunosuppressive agents after the superiority over placebo or no treatment has been clearly established.

Plain language summary

Tripterygium wilfordii Hook F (a traditional Chinese herbal medicine) for primary nephrotic syndrome

Primary nephrotic syndrome (NS) is a relatively rare kidney disease (diagnosed in up to 7/100,000 children and 3/100,000 adults annually). However the resulting kidney damage causes loss of proteins in urine (proteinuria) leading to low level albumin in the bloodstream, which can cause raised lipids and severe, generalised swelling. Primary NS can also lead to blood clotting (thromboembolism), infection, and acute kidney injury, and may become life threatening. Treatment for primary NS aims to relieve symptoms, avoid complications, and prolong life. Immunosuppressive treatments are of importance for primary NS. In China, a traditional Chinese herbal medicine, Tripterygium wilfordii Hook F (TwHF) has been used for over two decades as an immunosuppressive agent to decrease proteinuria and preserve kidney function. TwHF is prescribed alone or in combination with corticosteroids or other immunosuppressive agents such as cyclophosphamide (CPA).

In this review we investigated the use of TwHF as an immunosuppressive therapy for patients with primary NS. We only considered two standardised TwHF extracts which have lower toxicity profiles and fewer adverse effects than other non-standardised preparations. We reviewed the evidence from 10 randomised studies enrolling 630 Chinese patients with primary NS. Of these, four studies (293 patients) reported that when TwHF was used there was a significantly increased number of patients achieving complete, and complete or partial remission without increasing adverse events. There were no significant differences in complete, partial, or complete or partial remission when TwHF was compared with prednisone in four studies (223 patients). There were also no significant differences in complete, partial, or complete or partial remission when TwHF was compared with CPA in two studies (114 patients). One study reported TwHF significantly improved kidney function (decreased serum creatinine) when compared with CPA. TwHF was associated with a lower risk of psychosis than prednisone; and there was less likelihood of hair loss when compared to CPA. The quality of evidence was poor; there were only 10 studies, enrolling a total of 630 patients; follow-up was short; the maximum number of studies included in a comparison was four; and we had major concerns over the quality of the included studies. TwHF may have an add-on effect on remission in patients with primary NS; however there is insufficient evidence to assess if TwHF is as effective as prednisone or CPA.

Summary of findings(Explanation)

Summary of findings for the main comparison. Summary of findings: Tripterygium wilfordii Hook F (TwHF) versus no TwHF
  1. 1 Small number of studies and patients, short follow-up, and major concerns with methodological bias.
    2 Language bias should be considered because all included studies were conducted in Chinese patients and published in Chinese language.

TwHF versus non-TwHF for primary nephrotic syndrome
Patient or population: patients with primary nephrotic syndrome
Settings:
Intervention: Tripterygium wilfordii Hook F (TwHF)
Comparison: no TwHF
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
Number of participants
(studies)
Quality of the evidence
(GRADE)
Assumed riskCorresponding risk
Non-TwHF TwHF
Complete remission (follow-up: 12 to 16 months) Study population RR 1.46
(1.18 to 1.80)
293 (4)⊕⊕⊝⊝
Low 1,2
436 per 1000 637 per 1000
(514 to 785)
Medium risk population
406 per 1000 593 per 1000
(479 to 731)
Partial remission (follow-up: 12 to 16 months) Study population RR 0.77
(0.49 to 1.21)
293 (4)⊕⊕⊝⊝
Low 1,2
229 per 1000 176 per 1000
(112 to 277)
Medium risk population
257 per 1000 198 per 1000
(126 to 311)
Complete or partial remission (follow-up: 12 to 16 months) Study population RR 1.26
(1.1 to 1.44)
293 (4)⊕⊕⊝⊝
Low 1,2
664 per 1000 837 per 1000
(730 to 956)
Medium risk population
662 per 1000 834 per 1000
(728 to 953)
*The basis for the assumed risk (the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence
High quality: further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
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
Very low quality: we are very uncertain about the estimate

Summary of findings 2 Summary of findings: Tripterygium wilfordii Hook F versus prednisone

Summary of findings 2. Summary of findings: Tripterygium wilfordii Hook F versus prednisone
  1. 1 Small number of studies and patients, short follow-up, and major concerns with methodological bias.
    2 Language bias should be considered because all included studies were conducted in Chinese patients and published in Chinese language.

Tripterygium wilfordii Hook F (TwHF) versus prednisone for primary nephrotic syndrome
Patient or population: patients with primary nephrotic syndrome
Settings:
Intervention: TwHF
Comparison: prednisone
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
Number of participants
(studies)
Quality of the evidence
(GRADE)
Assumed riskCorresponding risk
Prednisone TwHF
Complete remission (follow-up: 6 to 18 months) Study population RR 1.09
(0.83 to 1.43)
223 (4)⊕⊕⊝⊝
Low 1,2
462 per 1000 504 per 1000
(383 to 661)
Medium risk population
500 per 1000 545 per 1000
(415 to 715)
Partial remission (follow-up: 6 to 18 months) Study population RR 1.07
(0.69 to 1.65)
223 (4)⊕⊕⊝⊝
Low 1,2
269 per 1000 288 per 1000
(186 to 444)
Medium risk population
290 per 1000 310 per 1000
(200 to 478)
Complete or partial remission (follow-up: 6 to 18 months) Study population RR 1.05
(0.92 to 1.2)
223 (4)⊕⊕⊝⊝
Low 1,2
731 per 1000 768 per 1000
(673 to 877)
Medium risk population
722 per 1000 758 per 1000
(664 to 866)
*The basis for the assumed risk (the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; RR: risk ratio;
GRADE Working Group grades of evidence
High quality: further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
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
Very low quality: we are very uncertain about the estimate

Summary of findings 3 Summary of findings: Comparison of TwHF and CPA

Summary of findings 3. Summary of findings: Comparison of TwHF and CPA
  1. 1 Small number of studies and patients, short follow-up, and major concerns with methodological bias.
    2 Language bias should be considered because all included studies were conducted in Chinese patients and published in Chinese language

Tripterygium wilfordii Hook F (TwHF)versus cyclophosphamide (CPA)for primary nephrotic syndrome
Patient or population: patients with primary nephrotic syndrome
Settings:
Intervention: TwHF
Comparison: CPA
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
Number of participants
(studies)
Quality of the evidence
(GRADE)
Assumed riskCorresponding risk
CPA TwHF
Complete remission (follow-up: 3 to 12 months) Study population RR 1.06
(0.82 to 1.37)
114 (2)⊕⊕⊝⊝
low 1,2
654 per 1000 693 per 1000
(536 to 896)
Medium risk population
653 per 1000 692 per 1000
(535 to 895)
Partial remission (follow-up: 3 to 12 months) Study population RR 1.02
(0.46 to 2.29)
114 (2)⊕⊕⊝⊝
low 1,2
173 per 1000 176 per 1000
(80 to 396)
Medium risk population
169 per 1000 172 per 1000
(78 to 387)
Complete or partial remission (follow-up: 3 to 12 months) Study population RR 1.05
(0.9 to 1.23)
114 (2)⊕⊕⊝⊝
low 1,2
827 per 1000 868 per 1000
(744 to 1000)
Medium risk population
822 per 1000 863 per 1000
(740 to 1000)
*The basis for the assumed risk (the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) was based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence
High quality: further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
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
Very low quality: we are very uncertain about the estimate

Background

Description of the condition

Primary nephrotic syndrome (NS) is a relatively rare, but important kidney disease in children and adults (Eddy 2003; Hull 2008). It has an annual incidence of 2 to 7/100,000 children (Eddy 2003) and 3/100,000 adults (Hull 2008). There are five main distinct histological variants of primary NS: minimal change nephropathy (MCN); focal segmental glomerulosclerosis (FSGS); membranous nephropathy (MN); mesangioproliferative glomerulonephritis (MsPGN) (immunoglobulin A (IgA) nephropathy and non-IgA nephropathy); and membranoproliferative glomerulonephritis (MPGN). Acute complications related to primary NS include thromboembolism, infection, and acute kidney injury (AKI). Deep and renal vein thromboses could lead to pulmonary embolism. Bacterial and viral infections such as pneumonia and cellulitis are an ongoing threat to the health of immunocompromised patients. Long-term sequelae of primary NS relate mainly to adverse events of immunosuppressive treatments on bones, growth and the cardiovascular system (Eddy 2003; Hull 2008).

Description of the intervention

Tripterygium wilfordii Hook F (TwHF) has been used widely in traditional Chinese medicine to treat rheumatoid arthritis and other autoimmune and inflammatory diseases. Results from studies in Chinese patients (Tao 1987; Tao 1989) stimulated further investigation of TwHF for rheumatoid arthritis in western countries (Goldbach-Mansky 2009; Tao 2001; Tao 2002). TwHF has been prescribed as an immunosuppressive agent for primary NS in China for more than 20 years (Chen 1985; Jiang 1994). Preliminary observations suggested that remission was achieved in 8/13 (62%) children with primary NS and maintained in four children (31%) for up to three years (Jiang 1994). A meta-analysis of three studies enrolling 150 adult patients with primary refractory NS indicated that TwHF were associated with a higher complete remission (Odds ratio (OR) 2.81, 95% CI 1.26 to 6.30) and complete or partial remission (OR 3.25, 95% CI 1.20 to 8.76) (Xu 2009a).

Non-standardised TwHF preparations have been reported as causing serious adverse events and death (Wen 1999). A search of the Chinese Biological Medicine (CBM) Database (from 1982 to 1996) identified 38 patients who experienced adverse events related to non-standardised TwHF preparations (Wen 1999). The adverse events included leukopenia (9/38, 24%), aplastic anaemia (5/38, 13%), gastrointestinal tract disturbances (10/38, 26%), liver dysfunction (4/38, 11%), skin lesions such as erythema and papules (5/38, 13%) and other rare complications. Serious adverse events were primarily related to haematological system, and 8/38 (21%) died (leukopenia (4/8, 50%); aplastic anaemia (2/8, 25%); severe liver dysfunction (2/8, 25%)) (Wen 1999). Toxicities caused by non-standardised TwHF preparations which were reported before 1996 were difficult to compare with adverse events reported for standardised TwHF preparations such as ethanol-ethyl acetate and chloroform-methanol extracts of TwHF (Goldbach-Mansky 2009). Peeling the roots of this plant and using a standardised extraction method can fundamentally reduce toxicity and increase tolerability (Tao 2002). More recently, the most common adverse events of standardised TwHF preparations were gastrointestinal tract disturbances, leukopenia, thrombocytopenia, rash, skin pigmentation, and malfunction of reproductive system (Bao 2011). Most of these could be resolved by dose adjustment or drug discontinuation (Bao 2011; Goldbach-Mansky 2009).

How the intervention might work

Kupchan 1972 first demonstrated that triptolide and tripdiolide, two diterpenoid triepoxides from TwHF, possess antileukaemic properties. More recently, triptolide and tripdiolide were considered to be the principal active components of TwHF (Tao 1995). Two preparations of TwHF have been used extensively in experimental and clinical studies because of their lower toxicity profiles: chloroform-methanol extract (T2 or TwHF multiglycoside) and ethanol-ethyl acetate extract (Goldbach-Mansky 2009; Sharma 1997; Tao 1998; Tao 2001; Tao 2002; Wan 2010).

Sharma 1997 demonstrated the in vitro inhibitory effect of chloroform-methanol extract on increased glomerular albumin permeability in isolated glomeruli. The protective effect of chloroform-methanol extract on glomerular filtration barriers may be independent from anti-inflammatory and immunosuppressive properties of TwHF (Sharma 1997). Hong 2002 showed that in vitro triptolide could effectively inhibit the pro-inflammatory and immunoregulatory potential of TNF-α activated human renal proximal tubular epithelial cells by inhibiting C3, B7h, and CD40 expression. The chloroform-methanol extract, presumably through reducing production of injurious cytokines, ameliorated proteinuria and acute mesangial injury in Thy1.1 glomerulonephritis rat model (Wan 2005). Triptolide displayed a prominent in vivo anti-proteinuric effect in puromycin aminonucleoside (PAN)-induced nephrosis rat model characterised by improvement of foot process effacement, decrease of podocyte injury marker (desmin) and restoration of nephrin and podocin expression and distribution (Zheng 2008). This in vitro study further confirmed the protective effect of triptolide on immortalised mouse podocytes from PAN-induced disruption of actin cytoskeleton and microfilament-associated synaptopodin and abnormality of nephrin and podocin expression (Zheng 2008). The effects of triptolide may be attributed to the suppression of reactive oxygen species generation and subsequent p38 mitogen-activated protein kinase pathway activation, and restoration of RhoA activity (Zheng 2008).

Why it is important to do this review

Although TwHF has been used widely as an immunosuppressive agent for primary NS in China for more than 20 years, the efficacy and safety of TwHF have not been systematically reviewed.

Objectives

This review aimed to assess the benefits and harms of TwHF for the treatment of primary NS in adults and children, either as a sole agent or combined with other drug therapies.

Methods

Criteria for considering studies for this review

Types of studies

All randomised controlled trials (RCTs) investigating the efficacy and safety of TwHF administered to patients with primary NS were considered for inclusion. Quasi-RCTs (studies in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable methods) were excluded. All the included RCTs had a follow-up of at least three months.

Types of participants

Inclusion criteria

Adult and children diagnosed with primary NS were included. The diagnosis of NS were defined by the authors in each single study. In absence of an explicit definition of primary NS, proteinuria above 3.5 g/24 h was used as the cut-off value for adults, and in children urine protein-creatinine ratio above 200 mg/mmol, proteinuria above 300mg/dL, or protein on urine dipstick above 3+ was used (KDIGO 2012).

Exclusion criteria

Patients with secondary NS were excluded.

Types of interventions

  • Two preparations of TwHF extracts (ethanol-ethyl acetate extract and chloroform-methanol extract) have been investigated in clinical studies and reported to carry lower toxicity profiles compared with other TwHF preparations. This review considered only ethanol-ethyl acetate extract and chloroform-methanol extract of TwHF. Other TwHF preparations were excluded because of their less favourable toxicity profiles (Tao 2002; Wen 1999).

  • TwHF versus placebo, no treatment, non-immunosuppressive agents (such as angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, anticoagulants, antiplatelet agents, or fish oil), or immunosuppressive drugs (glucocorticoids, alkylating agents, azathioprine, cyclosporine, tacrolimus, mycophenolate mofetil, or leflunomide).

  • TwHF in addition to the listed immunosuppressive and non-immunosuppressive agents versus the same drugs. Co-interventions were permitted where all study arm patients received the same co-interventions.

  • Various doses, frequencies of administration, and therapeutic durations of the same TwHF preparation were included.

  • Any preparations of other traditional Chinese herbal medicines, including single herb, compound herbs, extracts, raw materials, Chinese proprietary medicines, and practitioner-prescribed herbal formulae (individualised treatment) according to the study report, were excluded.

Types of outcome measures

Primary outcomes

Complete remission, partial remission, and complete or partial remission at the last follow-up. Complete and partial remission were assessed according to the definition provided in each single study. In the absence of an explicit definition of remission, we applied the following parameters (KDIGO 2012).

  • In children

    • Complete remission was defined as urine protein-creatinine ratio < 20 mg/mmol or < 1+ of protein on urine dipstick for three consecutive days.

    • Partial remission was defined as proteinuria reduction of ≥ 50% from the presenting value and absolute urine protein-creatinine ratio of 20 to 200 mg/mmol.

  • In adults

    • Complete remission was defined as proteinuria < 0.3 g/24 h, accompanied by a normal serum albumin concentration and normal serum creatinine.

    • Partial remission was defined as reduction of proteinuria to 0.3 to 3.5 g/24 h and a decrease > 50% from baseline, accompanied by an improvement or normalisation of the serum albumin concentration and stable serum creatinine (change in creatinine < 25%)

Secondary outcomes
  • Urinary protein excretion at last follow-up

  • Serum albumin at last follow-up

  • Serum creatinine at last follow-up

  • Adverse events of TwHF

Search methods for identification of studies

Electronic searches

  • Cochrane Renal Group's specialised register were searched by the Trial Search Coordinator of Cochrane Renal Group in August 2012

  • Cochrane Register of Controlled Trials (CENTRAL) in The Cochrane Library (2012 Issue 8)

  • PUBMED (1966 to August 2012)

  • EMBASE (1966 to August 2012)

  • CBM (Chinese Biological Medical Database) (1978 to November 2010)

  • CNKI (Chinese National Knowledge Infrastructure) (1979 to November 2010)

  • VIP (ChongQing WeiPu Chinese Science and Technology Periodical Database) (1989 to November 2010)

  • WanFang Database (1980 to November 2010)

See Appendix 1 for search strategies used.

Searching other resources

We searched the reference lists of relevant studies and related Chinese journals (6 November 2010).

Data collection and analysis

Selection of studies

The search strategy was applied to obtain titles and abstracts of relevant studies. The titles and abstracts were screened independently by two authors, who discarded studies that were not applicable; however studies and reviews that may have included relevant data or information were retained initially. Two authors independently assessed retrieved abstracts and, if necessary the full text, to determine which satisfied the inclusion criteria. Disagreements were resolved by consultation with a third author.

Data extraction and management

Data extraction was conducted independently by two authors using standard data extraction forms. Studies reported in non-English language journals were translated before assessment. Where more than one publication of one study existed, the publication with the most complete data was included. Where relevant outcomes were published only in earlier versions, these data were included in the analyses. Any discrepancies between published versions were to be highlighted. Any further information required from the original investigators was requested by phone or written correspondence and any relevant information obtained was included in this review. Disagreements were resolved by consultation with a third author.

Assessment of risk of bias in included studies

The following items were independently assessed by two authors using the risk of bias assessment tool (Higgins 2011). For each item, the judgement (low risk of bias, high risk of bias, or unclear risk of bias) was followed by a description of the design, conduct or observations underlying the judgement (see Appendix 2).

  • Was there adequate sequence generation (selection bias)?

  • Was allocation adequately concealed (selection bias)?

  • Was knowledge of the allocated interventions adequately prevented during the study (detection bias)?

    • Participants and personnel

    • Outcome assessors

  • Were incomplete outcome data adequately addressed (attrition bias)?

  • Are reports of the study free of suggestion of selective outcome reporting (reporting bias)?

  • Was the study apparently free of other problems that could put it at a risk of bias?

Measures of treatment effect

For dichotomous outcomes (e.g. remission) results were expressed as risk ratios (RR) with 95% confidence intervals (CI). Where continuous scales of measurement were used to assess the effects of treatment (e.g. urinary protein excretion, serum albumin, serum creatinine), the mean difference (MD) were used, or the standardised mean difference (SMD) if different scales have been used. If possible, we also aimed to determine the applicability of the results to individual patients, such as calculation of absolute risk reductions with therapy in relation to different baseline risk of the event with no therapy or a different therapy. Adverse events were tabulated and assessed using descriptive techniques. Where possible, risk difference with 95% CI was to be calculated for each adverse event, compared with either no treatment or another agent.

Unit of analysis issues

Studies with multiple intervention arms were analysed by entering each pair-wise comparison separately. For dichotomous outcomes, both the number of events and the total number of patients were halved. For continuous outcomes, only the total number of participants were halved, while the means and standard deviations were left unchanged (Higgins 2011).

Dealing with missing data

Data such as numbers of participants screened, randomised, intention-to-treat, and as-treated and per-protocol populations were evaluated and analysed. Attrition rates (drop-outs, losses to follow-up and withdrawals) were investigated. Issues of missing data and imputation methods (such as last-observation-carried-forward) were critically appraised (Higgins 2011). Participant results missing due to drop-out, intention-to-treat analysis was performed and compared with per-protocol analysis for the dichotomous data; while the continuous data remained unchanged due to the difficulty of application of ITT for this type of data (Moher 2010). For missing statistics such as standard deviations, these studies were not considered in the meta-analysis unless the missing data could be appropriately imputed using methods recommended by the Cochrane Collaboration (Higgins 2011). We also contacted principal investigators to request the missing data.

Assessment of heterogeneity

Heterogeneity was analysed using a Chi² test on N-1 degrees of freedom, with an alpha of 0.10 used for statistical significance (Higgins 2011). I² values of 25%, 50% and 75% corresponded to low, medium and high levels of heterogeneity.

Assessment of reporting biases

Assessment of publication bias was not conducted because of the small numbers of included studies in the three comparisons.

Data synthesis

Data were pooled using the random-effects model but the fixed-effect model was also used to ensure robustness of the model chosen and susceptibility to outliers.

Subgroup analysis and investigation of heterogeneity

The following subgroup analyses were planned.

  • Type of NS (steroid-resistant NS (SRNS) or frequently relapsing NS (FRNS)).

  • NS pathology.

  • Age of participants (e.g. children or elderly patients).

Sensitivity analysis

  • We conducted sensitivity analyses to explore the influence of diagnostic criteria of NS on the treatment effect.

Results

Description of studies

See Characteristics of included studies and Characteristics of excluded studies.

Results of the search

The search strategy identified 2938 citations (6 November 2010) (Figure 1). Of these, 50 studies were considered as potentially eligible. All were conducted in China and reported in Chinese language. A total of 30/50 studies were excluded after further evaluation. We attempted to contact the investigators from the remaining 20 studies to obtain missing data. The authors of two studies could not be contacted and subsequently excluded (10%). We were successful in eliciting responses from 18 studies (90%). Eight of these studies were further excluded after contact with the investigators.

Figure 1.

Flow diagram of study selection

Included studies

Our review included 10 RCTs published as full reports (A-Da 2008; Chen 2003; Han 1999; Li 2000; Song 1998; Wen 1993; Yuan 2003; Zhang 2007a; Zhao 1997; Zhao 2009).

Baseline characteristics

Age and gender appeared to be balanced in all included studies. Three studies did not report if proteinuria and serum creatinine were comparable at baseline (Li 2000; Wen 1993; Yuan 2003); and seven studies reported comparability at baseline (A-Da 2008; Chen 2003; Han 1999; Song 1998; Zhang 2007a; Zhao 1997; Zhao 2009).

Sample size

The 10 included studies randomised a total of 630 patients (345 in the experimental group and 285 in the control group). The median sample sizes was 64 patients (range from 36 to 92).

Diagnostic criteria

Eight studies only enrolled patients with primary NS. We contacted the investigators of the remaining two studies and confirmed that all enrolled patients had been diagnosed with primary NS (Li 2000; Zhao 1997). Sensitivity analyses for diagnostic criteria were conducted to detect the potential impact of unpublished information (Li 2000; Zhao 1997).

Age

Eight studies enrolled adult patients; one study included elderly patients (Yuan 2003); and one enrolled both adults and children (Li 2000).

Clinical conditions

Four studies included patients with primary refractory NS (Chen 2003; Han 1999; Song 1998; Zhao 2009). The remaining six studies included patients with primary NS.

Pathologic conditions

Common pathologic diagnoses reported in nine included studies were MCN, FSGS, MsPGN (including IgA nephropathy), MN, and MPGN. One study included only adults with primary IgA nephropathy-induced primary NS (A-Da 2008).

TwHF treatment

Initial doses of TwHF ranged from 0.5 mg/kg/d (Wen 1993) to 2.0 mg/kg/d (Song 1998; Zhao 2009), and were administered for four (Chen 2003; Zhao 2009) and eight weeks (A-Da 2008; Song 1998; Zhao 1997).

Follow-up

The median follow-up was 12 months (range: 3 to 18 months).

Comparisons

A presentation of comparisons in these 10 included studies was presented in additional Table 1. Nine studies featured two-arm design. Zhao 1997 conducted a three-armed study that compared 34 patients receiving TwHF plus prednisone (0.6 mg/kg/d), 26 patients receiving TwHF, and 26 patients receiving prednisone (1.0 mg/kg/d). The prednisone group (1.0 mg/kg/d) was split into two equal-sized subgroups. Two comparisons were conducted: (A) TwHF plus prednisone (0.6 mg/kg/d) (N = 34) versus prednisone (1.0 mg/kg/d) (N = 13) and (B) TwHF (N = 26) versus prednisone (1.0 mg/kg/d) (N = 13). We conducted primary analyses according to experimental and control interventions:

Table 1. Study treatment, control and comparison groups
  1. CPA: cyclophosphamide; CSA: cyclosporine; TwHF: Tripterygium wilfordii Hook F

StudyTreatmentControlfollow-up (months)Co-interventions
TwHF versus non-TwHF
Li 2000TwHF + prednisone + CPAPrednisone + CPA12Same co-interventions and doses in each arm
Song 1998TwHF + prednisone + CPAPrednisone + CPA12-15Same co-interventions and doses in each arm
Wen 1993TwHF + prednisone (1.0 mg/kg/d)Prednisone (1.0 mg/kg/d)12-16Same co-interventions and doses in each arm
Zhang 2007aTwHF + prednisone (1.0 mg/kg/d)Prednisone (1.0 mg/kg/d)12Same co-interventions and doses in each arm
TwHF versus prednisone
A-Da 2008TwHFPrednisone6Same co-interventions and doses in each arm
Han 1999TwHF + CSAPrednisone + CSA12Same co-interventions and doses in each arm
Yuan 2003TwHF + prednisone (0.4 to 0.5 mg/kg/d)Prednisone (0.8 to 1.0 mg/kg/d)12-18Same co-interventions and doses in each arm
Zhao 1997TwHF + prednisone (0.6 mg/kg/d)Prednisone (1.0 mg/kg/d)10-12Same co-interventions and doses in each arm
Zhao 1997TwHFPrednisone (1.0 mg/kg/d)10-12Same co-interventions and doses in each arm
TwHF versus CPA
Chen 2003TwHF + prednisoneCPA + prednisone3Same co-interventions and doses in each arm
Zhao 2009TwHF + prednisoneCPA + prednisone12Same co-interventions and doses in each arm

Excluded studies

We excluded 40 studies from this review.

Risk of bias in included studies

See Figure 2 and Figure 3.

Figure 2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies

Figure 3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study

Allocation

Sequence generation

Although all included studies were RCTs, explicit methods of randomised sequence generation was described in only one. A-Da 2008 used a random number table for sequencing. We determined that drawing of lots was used in nine studies by contacting study investigators.

Allocation concealment

None of the included studies described allocation sequence concealment methods after contact with study investigators.

Blinding

None of the included studies reported blinding methods.

Incomplete outcome data

None of the studies described losses to follow-up or treatment withdrawals.

Selective reporting

The primary outcome (remission) was reported in all included studies. Secondary outcome data on urinary protein excretion, serum albumin, serum creatinine, and drug-related adverse events were presented in two (20%), two (20%), three (30%), and five studies (50%), respectively.

Other potential sources of bias

There was insufficient information to determine other potential sources of bias.

Effects of interventions

See: Summary of findings for the main comparison Summary of findings: Tripterygium wilfordii Hook F (TwHF) versus no TwHF; Summary of findings 2 Summary of findings: Tripterygium wilfordii Hook F versus prednisone; Summary of findings 3 Summary of findings: Comparison of TwHF and CPA

Results are presented in Summary of findings for the main comparison, Summary of findings 2 and Summary of findings 3. Subgroup and sensitivity analyses are presented in Table 2 (TwHF versus non-TwHF) and Table 3 (TwHF versus prednisone).

Table 2. Subgroup and sensitivity analyses comparing Tripterygium wilfordii Hook F (TwHF) with no TwHF
  1. CPA: cyclophosphamide; FRNS: frequently relapsing nephrotic syndrome; NA - Not applicable; TwHF: Tripterygium wilfordii Hook F; CR: complete remission; PR: partial remission; CR+PR: complete or partial remission

Outcome or subgroupStudies (participants)Statistical methodEffect estimate (95% CI)HeterogeneitySubgroup difference
CR (TwHF versus no TwHF for non-FRNS)3 (235)Risk ratio (IV, Random, 95% CI)P = 0.004, RR = 1.41 (1.11 to 1.79)P = 0.33, I² = 10%P = 0.37, I² = 0%
CR (TwHF versus no TwHF for FRNS)1(58)Risk ratio (IV, Random, 95% CI)P = 0.03, RR = 1.87 (1.07 to 3.26)NA
PR (TwHF versus no TWHF for non-FRNS)3 (235)Risk ratio (IV, Random, 95% CI)P = 0.39, RR = 0.80 (0.48 to 1.33)P = 0.79, I² = 0%P = 0.81, I² = 0%
PR (TwHF versus no TwHF for FRNS)1 (58)Risk ratio (IV, Random, 95% CI)P = 0.45, RR = 0.70 (0.28 to 1.76)NA
CR+PR (TwHF versus no TwHF for non-FRNS)3 (235)Risk ratio (IV, Random, 95% CI)P = 0.005, RR = 1.24 (1.07 to 1.43)P = 0.64, I² = 0%P = 0.62, I² = 0%
CR+PR (TwHF versus no TwHF for FRNS)1 (58)Risk ratio (IV, Random, 95% CI)P = 0.06, RR = 1.35 (0.99 to 1.84)NA
CR (TwHF + prednisone versus prednisone)2 (143)Risk ratio (IV, Random, 95% CI)P = 0.14, RR = 1.43 (0.86 to 2.29)P = 0.19, I² = 42%P = 0.62, I² = 0%
CR (TwHF + prednisone + CPA versus prednisone + CPA)2 (150)Risk ratio (IV, Random, 95% CI)P = 0.002, RR = 1.65 (1.21 to 2.24)P = 0.60, I² = 0%
PR (TwHF + prednisone versus prednisone)2 (143)Risk ratio (IV, Random, 95% CI)P = 0.72, RR = 0.89 (0.47 to 1.68)P = 0.73, I² = 0%P = 0.53, I² = 0%
PR (TwHF + prednisone + CPA versus prednisone + CPA)2 (150)Risk ratio (IV, Random, 95% CI)P = 0.21, RR = 0.67 (0.35 to 1.26)P = 0.89, I² = 0%
CR+PR (TwHF + prednisone versus prednisone)2 (143)Risk ratio (IV, Random, 95% CI)P = 0.04, RR = 1.22 (1.01 to 1.49)P = 0.35, I² = 0%P = 0.72, I² = 0%
CR+PR (TwHF + prednisone + CPA versus prednisone + CPA)2 (150)Risk ratio (IV, Random, 95% CI)P = 0.008, RR = 1.29 (1.07 to 1.55)P = 0.71, I² = 0%
CR (sensitivity analysis for diagnostic criteria)3 (201)Risk ratio (IV, Random, 95% CI)P = 0.02, RR = 1.50 (1.07 to 2.00)P = 0.23, I² = 31% --
PR (sensitivity analysis for diagnostic criteria)3 (201)Risk ratio (IV, Random, 95% CI)P = 0.47, RR = 0.83 (0.49 to 1.39)P = 0.86, I² = 0% --
CR+PR (sensitivity analysis for diagnostic criteria)3 (201)Risk ratio (IV, Random, 95% CI)P = 0.006, RR = 1.26 (1.07 to 1.48)P = 0.56, I² = 0% --
Table 3. Subgroup and sensitivity analyses comparing Tripterygium wilfordii Hook F (TwHF) with prednisone
  1. CSA: cyclosporine; IgAN: immunoglobulin A nephropathy; FRNS: frequently relapsing nephrotic syndrome; NA - Not applicable; TwHF: Tripterygium wilfordii Hook F; CR: complete remission; PR: partial remission; CR+PR: complete or partial remission

Outcome or subgroupStudies (participants)Statistical methodEffect estimate (95% CI)HeterogeneitySubgroup difference
CR (TwHF versus prednisone for non-FRNS)3 (159)Risk ratio (IV, Random, 95% CI)P = 0.94, RR = 0.99 (0.73 to 1.34)P = 0.90, I² = 0%P = 0.17, I² = 47.7%
CR (TwHF versus prednisone for FRNS)1 (64)Risk ratio (IV, Random, 95% CI)P = 0.13, RR = 1.60 (0.87 to 2.93)NA
PR (TwHF versus prednisone for non-FRNS)3 (159)Risk ratio (IV, Random, 95% CI)P = 0.91, RR = 1.03 (0.60 to 1.77)P = 0.85, I² = 0%P = 0.82, I² = 0%
PR (TwHF versus prednisone for FRNS)1 (64)Risk ratio (IV, Random, 95% CI)P = 0.71, RR = 1.07 (0.69 to 1.65)NA
CR+PR (TwHF versus prednisone for non-FRNS)3 (159)Risk ratio (IV, Random, 95% CI)P = 0.93, RR = 0.99 (0.86 to 1.15)P = 0.78, I² = 0%P = 0.06, I² = 71.7%
CR+PR (TwHF versus prednisone for FRNS)1 (64)Risk ratio (IV, Random, 95% CI)P = 0.04, RR = 1.38 (1.01 to 1.90)NA
CR (TwHF versus prednisone for non-IgAN)3 (187)Risk ratio (IV, Random, 95% CI)P = 0.42, RR = 1.13 (0.84 to 1.51)P = 0.55, I² = 0%P = 0.53, I² = 0%
CR (TwHF versus prednisone for IgAN)1 (36)Risk ratio (IV, Random, 95% CI)P = 0.74, RR = 0.89 (0.44 to 1.78)NA
PR (TwHF versus prednisone for non-IgAN)3 (187)Risk ratio (IV, Random, 95% CI)P = 0.98, RR = 1.01 (0.63 to 1.61)P = 0.94, I² = 0%P = 0.51, I² = 0%
PR (TwHF versus prednisone for IgAN)1 (36)Risk ratio (IV, Random, 95% CI)P = 0.46, RR = 1.50 (0.51 to 4.43)NA
CR+PR (TwHF versus prednisone for non-IgAN)3 (187)Risk ratio (IV, Random, 95% CI)P = 0.53, RR = 1.06 (0.89 to 1.27)P = 0.20, I² = 35%P = 0.94, I² = 0%
CR+PR (TwHF versus prednisone for IgAN)1 (36)Risk ratio (IV, Random, 95% CI)P = 0.70, RR = 1.08 (0.74 to 1.57)NA
CR (TwHF versus prednisone for adults)

3 (186)

 

Risk ratio (IV, Random, 95% CI)P = 0.55, RR = 1.10 (0.81 to 1.48)P=0.48, I²= 0%P = 0.91, I² = 0%
CR (TwHF versus prednisone for elderly)1 (37)Risk ratio (IV, Random, 95% CI)P = 0.87, RR = 1.05 (0.56 to 1.97)NA
PR (TwHF versus prednisone for adults)

3 (186)

 

Risk ratio (IV, Random, 95% CI)P = 0.62, RR = 1.12 (0.71 to 1.77)P = 0.93, I² = 0%P = 0.53, I² = 0%
PR (TwHF versus prednisone for elderly)1 (37)Risk ratio (IV, Random, 95% CI)P = 0.62, RR = 0.71 (0.18 to 2.74)NA
CR+PR (TwHF versus prednisone for adults)

3 (186)

 

Risk ratio (IV, Random, 95% CI)P = 0.37, RR = 1.06 (0.93 to 1.22)P = 0.23, I² = 31%P = 0.60, I² = 0%
CR+PR (TwHF versus prednisone for elderly)1 (37)Risk ratio (IV, Random, 95% CI)P = 0.80, RR = 0.95 (0.62 to 1.44)NA
CR (TwHF versus prednisone)3 (159)Risk ratio (IV, Random, 95% CI)P = 0.94, RR = 0.99 (0.73 to 1.34)P = 0.90, I² = 0%P = 0.17, I² = 47.7%
CR (TwHF + CSA versus prednisone + CSA)1 (64)Risk ratio (IV, Random, 95% CI)P = 0.13, RR = 1.60 (0.87 to 2.93)NA
PR (TwHF versus prednisone)3 (159)Risk ratio (IV, Random, 95% CI)P = 0.91, RR = 1.03 (0.60 to 1.77)P = 0.85, I² = 0%P = 0.82, I² = 0%
PR (TwHF + CSA versus prednisone + CSA)1 (64)Risk ratio (IV, Random, 95% CI)P = 0.71, RR = 1.07 (0.69 to 1.65)NA
CR+PR (TwHF versus prednisone)3 (159)Risk ratio (IV, Random, 95% CI)P = 0.93, RR = 0.99 (0.86 to 1.15)P = 0.78, I² = 0%P = 0.06, I² = 71.7%
CR+PR (TwHF + CSA versus prednisone + CSA)1 (64)Risk ratio (IV, Random, 95% CI)P = 0.04, RR = 1.38 (1.01 to 1.90)NA
CR (sensitivity analysis for diagnostic criteria)3 (137)Risk ratio (IV, Random, 95% CI)P = 0.40, RR = 1.17 (0.81 to 1.69)P = 0.42, I² = 0% --
PR (sensitivity analysis for diagnostic criteria)3 (137)Risk ratio (IV, Random, 95% CI)P = 0.65, RR = 1.14 (0.65 to 1.98)P = 0.70, I² = 0% --
CR+PR (sensitivity analysis for diagnostic criteria)3 (137)Risk ratio (IV, Random, 95% CI)P = 0.19, RR = 1.16 (0.93 to 1.45)P = 0.32, I² = 11% --

TwHF versus non-TwHF

Four studies compared TwHF to non-TwHF (Li 2000; Song 1998; Wen 1993; Zhang 2007a).

Primary outcomes

TwHF significantly increased complete remission (Analysis 1.1 (4 studies, 293 participants): RR 1.46, 95% CI 1.18 to 1.80; I² = 3%) and complete or partial remission (Analysis 1.3 (4 studies, 293 participants): RR 1.26, 95% CI 1.10 to 1.44; I² = 0%). The effect of TwHF on partial remission did not reach statistical significance (Analysis 1.2 (4 studies, 293 participants): RR 0.77, 95% CI 0.49 to 1.21; I² = 0%).

Secondary outcomes

One study (Zhang 2007a) reported no significant difference in urinary protein excretion (Analysis 1.4 (1 study, 78 participants: MD -0.03 g/24 h, 95% CI -0.21 to 0.15) and serum albumin (Analysis 1.5 (1 study, 78 participants: MD 0.66 g/L, 95% CI -2.82 to 4.14). Two studies (Song 1998; Zhang 2007a) reported a significant decrease in serum creatinine; however heterogeneity was high (93%) and therefore we did not pool these studies (Analysis 1.6).

One study (Song 1998) reported three types of adverse events of TwHF: infection, liver transaminase elevation, and leukopenia/bone marrow suppression. There were no significant differences in these adverse events (Analysis 1.7).

Subgroup analyses

We were unable to perform subgroup analyses for pathological diagnosis of NS and age of participants due to lack of data.

SRNS or FRNS

Song 1998 included patients with SRNS or FRNS. There were no statistically significant subgroup differences in complete remission (P = 0.37), partial remission (P = 0.81) and complete or partial remission (P = 0.62) (Table 2).

Co-interventions

Wen 1993 and Zhang 2007a compared TwHF + prednisone versus prednisone; Li 2000 and Song 1998 compared TwHF + prednisone + CPA versus prednisone + CPA. There were no statistically significant subgroup differences in complete remission (P = 0.62), partial remission (P = 0.53), and complete or partial remission (P = 0.72).

Sensitivity analyses for diagnostic criteria

The diagnosis of primary NS was not confirmed until we contacted the principal investigator in one study (Li 2000). This study was excluded from the sensitivity analysis. There were no significant changes of complete, partial, and complete or partial remission (additional Table 2).

TwHF versus prednisone

Four studies compared TwHF to prednisone (A-Da 2008; Han 1999; Yuan 2003; Zhao 1997).

Primary outcomes

There were no significant differences in complete remission (Analysis 2.1 (4 studies, 223 participants): RR 1.09, 95% CI 0.83 to 1.43, I² = 0%), partial remission (Analysis 2.2 (4 studies, 223 participants): RR 1.07, 95% CI 0.69 to 1.65, I² = 0%), or complete or partial remission (Analysis 2.3 (4 studies, 223 participants): RR 1.06, 95% CI 0.92 to 1.22, I² = 14%).

Secondary outcomes

Urinary protein excretion, serum albumin, and serum creatinine were not reported.

Two studies (Han 1999; Yuan 2003) showed no significant difference in liver transaminase elevation (Analysis 2.4.1 (2 studies 101 participants): RR 1.51, 95% CI 0.16 to 14.22; I² = 47%). Yuan 2003 reported no significant difference for leukopenia or bone marrow suppression, elevated blood pressure, elevated blood glucose, femoral head necrosis, and infection. One study (Yuan 2003) reported the risk of psychosis was significantly lower in the TwHF group compared to the prednisone group (Analysis 2.4.7 (1 study, 37 participants): RR 0.11, 95% CI 0.01 to 0.75).

Subgroup analyses
SRNS or FRNS

Han 1999 only enrolled patients with SRNS or FRNS. There were no statistically significant subgroup differences in complete remission (P = 0.17), partial remission (P = 0.82) and complete or partial remission (P = 0.06) (additional Table 3).

Pathological diagnosis of NS

A-Da 2008 included only patients with IgA nephropathy-induced primary NS. There were no significant subgroup differences in complete remission (P = 0.53), partial remission (P = 0.51), and complete or partial remission (P = 0.94).

Aged participants

Yuan 2003 only enrolled elderly patients. There were no significant subgroup differences in complete remission (P = 0.91), partial remission (P = 0.53), and complete or partial remission (P = 0.60)

Co-interventions

A-Da 2008, Yuan 2003 and Zhao 1997 compared TwHF and prednisone; and Han 1999 compared TwHF + cyclosporine versus prednisone + cyclosporine. There were no statistically significant subgroup differences in complete remission (P = 0.17), partial remission (P = 0.82) and complete or partial remission (P = 0.06).

Sensitivity analyses for diagnostic criteria

The diagnosis of primary NS was not confirmed until we contacted the principal investigator in one study (Zhao 1997). This study was excluded from the sensitivity analysis. There were no significant changes of complete, partial, and complete or partial remission (additional Table 3).

TwHF versus CPA

Two studies compared TwHF with CPA (Chen 2003; Zhao 2009).

Primary outcomes

There were no significant differences in complete remission (Analysis 3.1 (2 studies, 114 participants): RR 1.06, 95% CI 0.82 to 1.37; I² = 0%), partial remission (Analysis 3.2 (2 studies, 114 participants): RR 1.02, 95% CI 0.46 to 2.29; I² = 0%), and complete or partial remission (Analysis 3.3 (2 studies, 114 participants); RR 1.05, 95% CI 0.90 to 1.23; I² = 0%).

Secondary outcomes

One study reported no significant differences in urinary protein excretion (Analysis 3.4 (1 study, 46 participants): MD -0.30 g/24 h, 95% CI -1.34 to 0.74) and serum albumin (Analysis 3.5 (1 study, 46 participants): MD 1.00 g/L, 95% CI -1.91 to 3.91). This study reported TwHF was associated with a significant decrease in serum creatinine (Analysis 3.6 (1 study, 46 participants): MD -14.00 μmol/L, 95% CI -26.43 to -1.57).

There were no significant differences in the following adverse events: gastrointestinal discomfort, liver transaminase elevation, menstrual disorders, leukopenia and haemorrhagic cystitis. The risk of hair loss was significantly lower for the TwHF group compared to the CPA group (Analysis 3.7.5 (2 studies, 114 participants): RR 0.11, 95% CI 0.02 to 0.59; I² = 0%). Subgroup and sensitivity analyses were not performed due to the lack of data.

Discussion

Summary of main results

This systematic review provides some support for the effectiveness of TwHF for the treatment of primary NS; however the quality of evidence was suboptimal because of the small number of included studies enrolling small numbers of participants; short follow-up in each study; only a few studies in each comparison category; and major concerns with methodological bias. TwHF could significantly increase complete remission by 46% (18% to 80%) and complete or partial remission by 26% (10% to 44%). In a population with an expected rate of complete remission of 436/1000, TwHF could increase the number of patients achieving complete remission to 637/1000 (514 to 785). Complete or partial remission could be increased from 664 to 837/1000 (730 to 956). Meta-analyses of head-to-head comparisons demonstrated that there were no significant differences in the primary outcomes when TwHF was compared with either prednisone or CPA. Few studies reported data on secondary outcomes - proteinuria, serum albumin and serum creatinine; one study reported TwHF could significantly decrease serum creatinine compared with CPA. Issues of safety varied considerably and still remain unclear due to the poor reporting of adverse events and the variety of toxicity profiles. Compared to prednisone, TwHF was associated with a lower risk of psychosis. Patients treated with TwHF had a lower risk of hair loss when compared with CPA. However there were no significant differences in other adverse events, including reversible amenorrhoea, leukopenia, liver transaminase elevation, and gastrointestinal discomfort. A double-blind, placebo-controlled study examined the safety and efficacy of TwHF in the treatment of rheumatoid arthritis (Tao 2002). Although the dose in Tao 2002 was 180 mg/d to 360 mg/d for 20 weeks - higher than in any of the 10 included studies (0.5 to 2.0 mg/kg/d for 4 to 8 weeks as initial therapy) - no serious adverse events were observed, and no patients withdrew because of adverse events.

Overall completeness and applicability of evidence

All included studies were conducted in Chinese patients and published in Chinese language journals. Evidence is still needed to examine the effects of TwHF for other ethnicities. There was little evidence related to the use of TwHF in children with NS. In Chinese clinical practice many physicians were reluctant to use TwHF for children to avoid reproductive system complications (Li 2000). Pathological variations may differ in response to TwHF treatment; however due to the paucity of evidence we unable to use subgroup analyses to determine whether those variations could convey different responses to TwHF treatment. Although four studies enrolled patients with SRNS or FRNS, heterogeneity in the interventions among these four studies did not allow us to examine the effect of TwHF on patients with SRNS or FRNS.

Quality of the evidence

This review included 10 studies (630 participants). Assessment according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group indicated that the results could be considered as low quality level. The suboptimal methodological quality of the included studies was the major contributing factor. No study demonstrated adequate allocation concealment. Effect estimates from studies with inadequate concealment of allocation have been shown to overestimate beneficial effects by 18% (95% CI 5% to 29%) than that from studies with adequate concealment of allocation (Pildal 2007). Blinding was not reported in any study. The outcome measurements were all objective and were not likely to be influenced by lack of blinding; however lack of blinding could introduce potential bias in clinical studies (Higgins 2011). Although primary outcome completeness rates were 100%, this was remarkably lower for secondary outcomes (20% to 50%). Consequently, selective outcome reporting bias should be considered when interpreting the results.

Both the numbers of included studies and participants were relatively small, and all tested comparisons involved no more than four studies. Follow-up duration was no more than 18 months. The paucity of evidence did not allow us to perform subgroup analyses to explore modification effects by age, pathological variations, clinical conditions, and co-interventions. Such modification effects may potentially influence the efficacy of TwHF and should be investigated in the future. None of the included studies were designed to evaluate the efficacy of within intervention differences, such as dose and duration variation. More and better evidence is needed to explore benefits that may be conferred by different doses and durations of TwHF treatment.

Potential biases in the review process

Assessment of publication bias was not conducted because of the small number of included studies in the three comparisons. All studies were reported in Chinese language journals. Therefore, publication bias and language bias should be carefully considered when interpreting these results.

Agreements and disagreements with other studies or reviews

Xu 2009a meta-analysed results from three studies enrolling 150 adult patients with primary refractory NS. TwHF significantly increased complete remission (OR 2.81, 95% CI 1.26 to 6.30) and complete or partial remission (OR 3.25, 95% CI 1.20 to 8.76). The efficacy of TwHF on primary IgA nephropathy was examined in another meta-analysis of four studies with 188 patients (Chen 2010a). TwHF significantly increased complete remission (RR 1.53, 95% CI 1.09 to 2.16) and complete or partial remission (RR 1.27, 95% CI 1.08 to 1.48).

Authors' conclusions

Implications for practice

Overall, the quality of evidence was suboptimal due to the small number of included studies enrolling small numbers of participants; short follow-up in each study; only a few studies in each comparison category; and major concerns with methodological bias. This review has highlighted the fact that no well-designed RCT has been performed to test the efficacy and safety of TwHF for primary NS. In the majority of the included studies TwHF was used in conjunction with other immunosuppressive agents such as prednisone and CPA. Therefore, the individual effect of TwHF is still unclear. TwHF may have an add-on effect on remission in patients with primary NS. There was insufficient evidence to assess if TwHF was as effective as prednisone or CPA.

Implications for research

There is need for more methodologically sound studies with an emphasis on adequate sample size and follow-up. Furthermore, priority should be given to the choice of definite endpoints (all-cause mortality, risk of end-stage kidney disease) rather than surrogate outcomes (decline in kidney function). Recording and reporting adverse events should also be improved in future clinical studies. TwHF should be further directly compared with other widely used immunosuppressive agents after the superiority over placebo or no treatment has been clearly established. Ideally, optimal doses and durations of TwHF treatment that are most beneficial and least harmful to patients of different races, ages, and clinical and pathological severity still remain to be clarified.

Acknowledgements

We would like to thank the referees for their comments and feedback on this review. We would also like to acknowledge the contribution of Ribao Wei and JIanhui Zhou who helped to prepare this systematic review.

Data and analyses

Download statistical data

Comparison 1. TwHF versus no TwHF
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Complete remission4293Risk Ratio (IV, Random, 95% CI)1.46 [1.18, 1.80]
2 Partial remission4293Risk Ratio (IV, Random, 95% CI)0.77 [0.49, 1.21]
3 Complete or partial remission4293Risk Ratio (IV, Random, 95% CI)1.26 [1.10, 1.44]
4 Urinary protein excretion1 Mean Difference (IV, Random, 95% CI)Totals not selected
5 Serum albumin1 Mean Difference (IV, Random, 95% CI)Totals not selected
6 Serum creatinine2 Mean Difference (IV, Random, 95% CI)Totals not selected
7 Adverse effects1 Risk Ratio (IV, Random, 95% CI)Totals not selected
7.1 Liver transaminase elevation1 Risk Ratio (IV, Random, 95% CI)0.0 [0.0, 0.0]
7.2 Leukopenia or bone marrow suppression1 Risk Ratio (IV, Random, 95% CI)0.0 [0.0, 0.0]
7.3 Infection1 Risk Ratio (IV, Random, 95% CI)0.0 [0.0, 0.0]
Analysis 1.1.

Comparison 1 TwHF versus no TwHF, Outcome 1 Complete remission.

Analysis 1.2.

Comparison 1 TwHF versus no TwHF, Outcome 2 Partial remission.

Analysis 1.3.

Comparison 1 TwHF versus no TwHF, Outcome 3 Complete or partial remission.

Analysis 1.4.

Comparison 1 TwHF versus no TwHF, Outcome 4 Urinary protein excretion.

Analysis 1.5.

Comparison 1 TwHF versus no TwHF, Outcome 5 Serum albumin.

Analysis 1.6.

Comparison 1 TwHF versus no TwHF, Outcome 6 Serum creatinine.

Analysis 1.7.

Comparison 1 TwHF versus no TwHF, Outcome 7 Adverse effects.

Comparison 2. TwHF versus prednisone
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Complete remission4223Risk Ratio (IV, Random, 95% CI)1.09 [0.83, 1.43]
2 Partial remission4223Risk Ratio (IV, Random, 95% CI)1.07 [0.69, 1.65]
3 Complete or partial remission4223Risk Ratio (IV, Random, 95% CI)1.06 [0.92, 1.22]
4 Adverse effects2 Risk Ratio (IV, Random, 95% CI)Subtotals only
4.1 Liver transaminase elevation2101Risk Ratio (IV, Random, 95% CI)1.51 [0.16, 14.22]
4.2 Leukopenia or bone marrow suppression137Risk Ratio (IV, Random, 95% CI)10.45 [0.62, 176.40]
4.3 Elevated blood pressure137Risk Ratio (IV, Random, 95% CI)0.16 [0.02, 1.19]
4.4 Elevated blood glucose137Risk Ratio (IV, Random, 95% CI)0.27 [0.06, 1.13]
4.5 Femoral head necrosis137Risk Ratio (IV, Random, 95% CI)0.19 [0.01, 3.71]
4.6 Infection137Risk Ratio (IV, Random, 95% CI)0.24 [0.03, 1.92]
4.7 Psychosis137Risk Ratio (IV, Random, 95% CI)0.11 [0.01, 0.75]
Analysis 2.1.

Comparison 2 TwHF versus prednisone, Outcome 1 Complete remission.

Analysis 2.2.

Comparison 2 TwHF versus prednisone, Outcome 2 Partial remission.

Analysis 2.3.

Comparison 2 TwHF versus prednisone, Outcome 3 Complete or partial remission.

Analysis 2.4.

Comparison 2 TwHF versus prednisone, Outcome 4 Adverse effects.

Comparison 3. TwHF versus CPA
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Complete remission2114Risk Ratio (IV, Random, 95% CI)1.06 [0.82, 1.37]
2 Partial remission2114Risk Ratio (IV, Random, 95% CI)1.02 [0.46, 2.29]
3 Complete or partial remission2114Risk Ratio (IV, Random, 95% CI)1.05 [0.90, 1.23]
4 Urinary protein excretion1 Mean Difference (IV, Random, 95% CI)Totals not selected
5 Serum albumin1 Mean Difference (IV, Random, 95% CI)Totals not selected
6 Serum creatinine1 Mean Difference (IV, Random, 95% CI)Totals not selected
7 Adverse effects2 Risk Ratio (IV, Random, 95% CI)Subtotals only
7.1 Gastrointestinal discomfort2114Risk Ratio (IV, Random, 95% CI)0.49 [0.24, 1.04]
7.2 Liver transaminase elevation2114Risk Ratio (IV, Random, 95% CI)0.69 [0.13, 3.69]
7.3 Menstrual disorders114Risk Ratio (IV, Random, 95% CI)1.33 [0.40, 4.43]
7.4 Leukopenia2114Risk Ratio (IV, Random, 95% CI)0.32 [0.08, 1.32]
7.5 Hair Loss2114Risk Ratio (IV, Random, 95% CI)0.11 [0.02, 0.59]
7.6 Haemorrhagic cystitis168Risk Ratio (IV, Random, 95% CI)0.13 [0.01, 2.38]
Analysis 3.1.

Comparison 3 TwHF versus CPA, Outcome 1 Complete remission.

Analysis 3.2.

Comparison 3 TwHF versus CPA, Outcome 2 Partial remission.

Analysis 3.3.

Comparison 3 TwHF versus CPA, Outcome 3 Complete or partial remission.

Analysis 3.4.

Comparison 3 TwHF versus CPA, Outcome 4 Urinary protein excretion.

Analysis 3.5.

Comparison 3 TwHF versus CPA, Outcome 5 Serum albumin.

Analysis 3.6.

Comparison 3 TwHF versus CPA, Outcome 6 Serum creatinine.

Analysis 3.7.

Comparison 3 TwHF versus CPA, Outcome 7 Adverse effects.

Appendices

Appendix 1. Electronic search strategies

DatabaseSearch terms
CENTRAL
  1. tripterygium:ti,ab,kw

  2. triptolide:ti,ab,kw

  3. "thundergod vine":ti,ab,kw

  4. "leigong teng":ti,ab,kw

  5. "lei gong teng":ti,ab,kw

  6. (#1 OR #2 OR #3 OR #4 OR #5)

  7. "nephrotic syndrome":ti,ab,kw

  8. MeSH descriptor Nephrosis, Lipoid explode all trees

  9. nephrosis next lipoid:ti,ab,kw

  10. lipoid next nephrosis:ti,ab,kw

  11. "minimal change disease":ti,ab,kw

  12. glomeruloneph*:ti,ab,kw

  13. nephritis:ti,ab,kw

  14. nephropathy:ti,ab,kw

  15. glomerulosclerosis:ti,ab,kw

  16. (FSGS or MsGN):ti,ab,kw

  17. (#7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16)

  18. (#6 AND #17)

MEDLINE
  1. Tripterygium/

  2. tripterygium.tw.

  3. thundergod vine.tw.

  4. (leigong teng or lei gong teng).tw.

  5. triptolide.tw.

  6. or/1-5

  7. Nephrotic Syndrome/

  8. Nephrosis Lipoid/

  9. nephrotic syndrome.tw.

  10. lipoid nephrosis.tw.

  11. minimal change disease.tw.

  12. minimal change glomerulon$.tw.

  13. minimal change nephr$.tw.

  14. Glomerulonephritis, Membranous/

  15. (membranous adj2 glomerulon$).tw.

  16. (membranous adj2 nephropathy).tw.

  17. extramembranous glomerulon$.tw.

  18. Glomerulosclerosis, Focal Segmental/

  19. focal segmental glomerulosclerosis.tw.

  20. FSGS.tw.

  21. Glomerulonephritis, IgA/

  22. IgA glomerulon$.tw.

  23. IgA nephropathy.tw.

  24. mesangioproliferative glomerulon$.tw.

  25. MsGN.tw.

  26. mesangi$ proliferative glomerulon$.tw.

  27. Glomerulonephritis, Membranoproliferative/

  28. membranoproliferative glomerulon$.tw.

  29. mesangiocapillary glomerulon$.tw.

  30. or/7-29

  31. and/6,30

EMBASE
  1. Tripterygium/

  2. Tripterygium wilfordii extract/

  3. tripterygium.tw.

  4. thundergod vine.tw.

  5. (leigong teng or lei gong teng).tw.

  6. Triptolide/

  7. triptolide.tw.

  8. or/1-7

  9. Nephrotic Syndrome/

  10. Lipoid Nephrosis/

  11. nephrotic syndrome.tw.

  12. lipoid nephrosis.tw.

  13. Focal Glomerulonephritis/

  14. focal segmental glomerulosclerosis.tw.

  15. FSGS.tw.

  16. Immunoglobulin A Nephropathy/

  17. IgA nephropathy.tw.

  18. IgA glomeruloneph$.tw.

  19. Proliferative Glomerulonephritis/

  20. mesangioproliferative glomerulon$.tw.

  21. MsGN.tw.

  22. mesangi$ proliferative glomerulon$.tw.

  23. Membranoproliferative Glomerulonephritis/

  24. membranoproliferative glomerulon$.tw.

  25. mesangiocapillary glomerulon$.tw.

  26. Membranous Glomerulonephritis/

  27. (membranous adj2 glomeruloneph$).tw.

  28. (membranous adj2 nephropathy).tw.

  29. extramembranous glomerulon$.tw.

  30. Minimal Change Glomerulonephritis/

  31. minimal change disease.tw.

  32. minimal change glomerulon$.tw.

  33. minimal change nephr$.tw.

  34. or/9-33

  35. 35. and/8,34

Appendix 2. Risk of bias assessment tool

Potential source of bias Assessment criteria

Random sequence generation

Selection bias (biased allocation to interventions) due to inadequate generation of a randomised sequence

Low risk of bias: Random number table; computer random number generator; coin tossing; shuffling cards or envelopes; throwing dice; drawing of lots; minimization (minimization may be implemented without a random element, and this is considered to be equivalent to being random).
High risk of bias: Sequence generated by odd or even date of birth; date (or day) of admission; sequence generated by hospital or clinic record number; allocation by judgement of the clinician; by preference of the participant; based on the results of a laboratory test or a series of tests; by availability of the intervention.
Unclear: Insufficient information about the sequence generation process to permit judgement.

Allocation concealment

Selection bias (biased allocation to interventions) due to inadequate concealment of allocations prior to assignment

Low risk of bias: Randomisation method described that would not allow investigator/participant to know or influence intervention group before eligible participant entered in the study (e.g. central allocation, including telephone, web-based, and pharmacy-controlled, randomisation; sequentially numbered drug containers of identical appearance; sequentially numbered, opaque, sealed envelopes).
High risk of bias: Using an open random allocation schedule (e.g. a list of random numbers); assignment envelopes were used without appropriate safeguards (e.g. if envelopes were unsealed or non-opaque or not sequentially numbered); alternation or rotation; date of birth; case record number; any other explicitly unconcealed procedure.
Unclear: Randomisation stated but no information on method used is available.

Blinding of participants and personnel

Performance bias due to knowledge of the allocated interventions by participants and personnel during the study

Low risk of bias: No blinding or incomplete blinding, but the review authors judge that the outcome is not likely to be influenced by lack of blinding; blinding of participants and key study personnel ensured, and unlikely that the blinding could have been broken.
High risk of bias: No blinding or incomplete blinding, and the outcome is likely to be influenced by lack of blinding; blinding of key study participants and personnel attempted, but likely that the blinding could have been broken, and the outcome is likely to be influenced by lack of blinding.
Unclear: Insufficient information to permit judgement

Blinding of outcome assessment

Detection bias due to knowledge of the allocated interventions by outcome assessors.

Low risk of bias: No blinding of outcome assessment, but the review authors judge that the outcome measurement is not likely to be influenced by lack of blinding; blinding of outcome assessment ensured, and unlikely that the blinding could have been broken.
High risk of bias: No blinding of outcome assessment, and the outcome measurement is likely to be influenced by lack of blinding; blinding of outcome assessment, but likely that the blinding could have been broken, and the outcome measurement is likely to be influenced by lack of blinding.
Unclear: Insufficient information to permit judgement

Incomplete outcome data

Attrition bias due to amount, nature or handling of incomplete outcome data.

Low risk of bias: No missing outcome data; reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes not enough to have a clinically relevant impact on observed effect size; missing data have been imputed using appropriate methods.
High risk of bias: Reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size; ‘as-treated’ analysis done with substantial departure of the intervention received from that assigned at randomisation; potentially inappropriate application of simple imputation.
Unclear: Insufficient information to permit judgement

Selective reporting

Reporting bias due to selective outcome reporting

Low risk of bias: The study protocol is available and all of the study’s pre-specified (primary and secondary) outcomes that are of interest in the review have been reported in the pre-specified way; the study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were pre-specified (convincing text of this nature may be uncommon).
High risk of bias: Not all of the study’s pre-specified primary outcomes have been reported; one or more primary outcomes is reported using measurements, analysis methods or subsets of the data (e.g. subscales) that were not pre-specified; one or more reported primary outcomes were not pre-specified (unless clear justification for their reporting is provided, such as an unexpected adverse effect); one or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta-analysis; the study report fails to include results for a key outcome that would be expected to have been reported for such a study.
Unclear: Insufficient information to permit judgement

Other bias

Bias due to problems not covered elsewhere in the table

Low risk of bias: The study appears to be free of other sources of bias.
High risk of bias: Had a potential source of bias related to the specific study design used; stopped early due to some data-dependent process (including a formal-stopping rule); had extreme baseline imbalance; has been claimed to have been fraudulent; had some other problem.
Unclear: Insufficient information to assess whether an important risk of bias exists; insufficient rationale or evidence that an identified problem will introduce bias.

What's new

Last assessed as up-to-date: 13 August 2012.

DateEventDescription
11 August 2013AmendedContact details updated.

Contributions of authors

  1. Draft the protocol: Yizhi Chen, Qing Yuan

  2. Study selection: Yizhi Chen, Zhixiang Gong

  3. Extract data from studies: Yizhi Chen, Zhixiang Gong

  4. Enter data into RevMan: Yizhi Chen, Zhixiang Gong

  5. Carry out the analysis: Yizhi Chen

  6. Interpret the analysis: Yizhi Chen, Guangyan Cai, Xiangmei Chen, Xuezhi Zhao, Li Tang, Qing Yuan

  7. Draft the final review: Yizhi Chen, Guangyan Cai, Xiangmei Chen, Xuezhi Zhao, Li Tang, Qing Yuan

  8. Disagreement resolution: Xiangmei Chen

  9. Update the review: Yizhi Chen

Declarations of interest

None known.

Sources of support

Internal sources

  • No sources of support provided, Not specified.

External sources

  • This work was partially supported by the National Science and Technology Department 863 Project (2012AA02A512), the National Key New Drug Innovation Project (2013ZX09104-003), the Beijing Science and Technology Project (D13110700470000), and the National Science and Technology Support Program (2011BAI10B00, 2011BAI10B03), China.

    The sponsors had no role in the study design, data collection and analysis, decision to publish, or manuscript preparation.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

A-Da 2008

Methods
  • Study design: parallel RCT

  • Follow-up period: 6 months

Participants
  • Country: China

  • Setting: single centre in Xinjiang province

  • Health status: primary NS with IgA nephropathy

  • Number: group 1 (18); group 2 (18)

  • Mean age ± SD (range): 22.3 ± 1.8 (18 to 35) years

  • Sex (M/F): group 1 (12/6); group 2 (14/4)

  • Kidney function: NA

Interventions

Group 1

  • TwHF

  • Dose: 1.0 to 1.5 mg/kg/d for 8 weeks gradually reducing to 20 mg/

Group 2

  • Prednisone

  • Dose: 0.8 to 1.0 mg/kg/d for 8 weeks reducing by 10% to 20% every 1 to 2 weeks to < 15 mg/d

Outcomes
  • Complete, partial and total remission rates at 6 months

Notes
  • Baseline comparison: adequate

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number table
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskSame number of patients randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
High riskNot reported
Other biasUnclear riskInsufficient information

Chen 2003

Methods
  • Study design: parallel RCT

  • Follow-up period: 3 months

Participants
  • Country: China

  • Setting: single centre in Henan province of China

  • Health status: refractory primary NS

  • Number: group 1 (36); group 2 (32)

  • Mean age ± SD: group (1:28.2 ± 12.3 years); group 2 (29.6 ± 11.5 years)

  • Sex (M/F): group 1 (21/15); group 2 (22/10)

  • Kidney function: Normal

Interventions

Group 1

  • TwHF

    • Dose: 2.0 mg/kg/d for 4 weeks; 1.5 mg/kg/d for 2 to 4 weeks; 1.0 mg/kg/d for 4 to 6 weeks

  • Prednisone

    • Dose: 40 mg/d for 12 weeks

Group 2

  • CPA

    • Dose: 2 mg/kg/d/oral for 12 weeks; total dosage < 6 to 8 g

  • Prednisone

    • Dose: 40 mg/d for 12 weeks

Outcomes
  • Complete, partial and total remission rates at 3 months

  • Adverse events at 3 months

Notes
  • Baseline comparison: adequate

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskLots drawn
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskThe same number of patients were randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
Low riskReported
Other biasUnclear riskInsufficient information

Han 1999

Methods
  • Study design: Parallel RCT

  • Follow-up period: 12 months

Participants
  • Country: China

  • Setting: single centre in Shandong province

  • Health status: refractory primary NS

  • Number: group 1 (33); group 2 (31)

  • Mean age; range: group 1 (31; 13 to 51 years); group 2 (30; 14 to 49 years)

  • Sex (M/F): group 1 (23/10); group 2 (20/11)

  • Kidney function: NA

Interventions

Group 1

  • TwHF

    • Dose: 60 mg/d then reducing over 6 to 12 months

  • CSA

    • Dose: 3.0 to 5.0 mg/kg/d reducing over 3 to 6 months

Group 2

  • Prednisone

    • Dose: 10 to 20 mg/d for 3 to 6 months

  • CSA

    • Dose: 3.0 to 5.0 mg/kg/d reducing over 3 to 6 months

Outcomes
  • Complete, partial and total remission rates at 12 months

  • Adverse events at 12 months

Notes
  • Baseline comparison: adequate

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskLots drawn
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskSame number of patients randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
Low riskReported
Other biasUnclear riskInsufficient information

Li 2000

Methods
  • Study design: parallel RCT

  • Follow-up period: 12 months

Participants
  • Country: China

  • Setting: single centre in Liaoning province

  • Health status: NS

  • Number: group 1 (48); group 2 (44)

  • Mean age; range: group 1 (27.2; 13 to 44 years); group 2 (26.9; 12 to 48 years)

  • Sex (M/F): group 1 (25/23); group 2 (23/21)

  • Kidney function: NA

Interventions

Group 1

  • TwHF

    • Dose: 1.5 mg/kg/d reducing over 12 months

  • Prednisone

    • Dose: 1.0 mg/kg/d for adults or 1.5 mg/m²/d for children, total dose < 80 mg/d; total duration 6 to 12 months

  • CPA

    • Dose: 200 mg IV alternate days, total dosage < 6 to 8 g

Group 2

  • Prednisone

    • Dose: 1.0 mg/kg/d for adults or 1.5 mg/m²/d for children, total dose < 80 mg/d; total duration 6 to 12 months

  • CPA

    • Dose: 200 mg IV alternate days, total dosage < 6 to 8 g

Outcomes
  • Complete, partial and total remission rates at 12 months

Notes
  • Baseline comparison: NA

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskLots drawn
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskSame number of patients randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
High riskNot reported
Other biasUnclear riskInsufficient information

Song 1998

Methods
  • Study design: parallel RCT

  • Follow-up period: 12 to 15 months

Participants
  • Country: China

  • Setting: single centre in Shandong province

  • Health status: refractory primary NS

  • Number: group 1 (30); group 2 (28)

  • Mean age ± SD: group 1 (25.3 ± 8.7 years); group 2 (24.1 ± 10.5 years)

  • Sex (M/F): group 1: (19/11); group 2 (18/10)

  • Kidney function: NA

Interventions

Group 1

  • TwHF

    • Dose: 2.0 mg/kg/d for 6 to 8 weeks, 1.5 mg/kg/d for 4 weeks, reducing to 1.0 mg/kg/d; total duration 9 to 12 months

  • CPA

    • Dose: 0.5 to 1.0 g/m² every month for 3 to 6 months then every 3 months; total duration 9 to 12 months

  • Prednisone

    • Dose: 0.5 mg/kg/d reducing over 6 months

Group 2

  • CPA

    • Dose: 0.5 to 1.0 g/m² every month for 3 to 6 months, then every 3 months; total duration 9 to 12 months

  • Prednisone

    • Dose: 0.5 mg/kg/d reducing over 6 months

Outcomes
  • Complete, partial and total remission rates at 12 to 15 months

  • SCr at 12 to 15 months

  • Adverse events at 12 to 15 months

Notes
  • Baseline comparison: Adequate

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskLots drawn
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskSame number of patients randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
Low riskReported
Other biasUnclear riskInsufficient information

Wen 1993

Methods
  • Study design: parallel RCT

  • Follow-up period: 12 to 16 months

Participants
  • Country: China

  • Setting: single centre in Henan province

  • Health status: Primary NS

  • Number: group 1 (35); group 2 (30)

  • Age range: 16 to 60 years

  • Sex (M/F): 40/25

  • Kidney function: NA

Interventions

Group 1

  • TwHF

    • Dose: 30 to 60 mg/d reducing over 3 to 6 months

  • Prednisone

    • Dose: 1.0 mg/kg/d for 8 weeks reducing by 5 mg/d every 2 weeks to 10 to 15 mg/d for 3 months, then 10 to 15 mg on alternate days for 2 months then twice weekly for 1 month

Group 2

  • Prednisone

    • Dose: 1.0 mg/kg/d for 8 weeks reducing by 5 mg/d every 2 weeks to 10 to 15 mg/d for 3 months, then 10 to 15 mg alternate days for 2 months then twice weekly for 1 month

Outcomes
  • Complete, partial and total remission rates at 12 to 16 months

Notes
  • Baseline comparison: NA

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskLots drawn
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskSame number of patients randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
High riskNot reported
Other biasUnclear riskInsufficient information

Yuan 2003

Methods
  • Study design: parallel RCT

  • Follow-up period: 12 to 18 months

Participants
  • Country: China

  • Setting: single centre in Hunan province

  • Health status: primary NS; elderly patients

  • Number: group 1(19); group 2 (18)

  • Mean age (range): 63.5 (60 to 71) years

  • Sex (M/F): 25/12

  • Kidney function: NA

Interventions

Group 1

  • TwHF

    • Dose: 1 mg/kg/d reducing over 12 to 18 months

  • Prednisone

    • Dose: 0.4 to 0.5 mg/kg/d reducing over 12 to 18 months

Group 2

  • Prednisone

    • Dose: 0.8 to 1.0 mg/kg/d reducing over 12 to 18 months

Outcomes
  • Complete, partial and total remission rates at 12 to 18 months

  • Adverse events at 12 to 18 months

Notes
  • Baseline comparison: NA

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskLots drawn
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskSame number of patients randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
Low riskReported
Other biasUnclear riskInsufficient information

Zhang 2007a

Methods
  • Study design: parallel RCT

  • Follow-up period: 12 months

Participants
  • Country: China

  • Setting: single centre in Hunan province

  • Health status: Primary NS

  • Number: group 1 (40); group 2 (38)

  • Mean age ±SD: group 1 (24.8 ± 11.8 years); group 2 (31.1 ± 12.0 years)

  • Sex (M/F): group 1 (24/16); group 2 (21/17)

  • Kidney function: SCr < 176 μmol/L

Interventions

Group 1

  • TwHF

    • Dose: 60 mg/d reducing over 12 months

  • Prednisone

    • Dose: 1.0 mg/kg/d reducing over 12 months

Group 2

  • Prednisone

    • Dose: 1.0 mg/kg/d reducing over 12 months

Outcomes
  • Complete, partial and total remission rates at 12 months

  • Urinary protein excretion at 12 months

  • Serum albumin at 12 months

  • SCr at 12 months

Notes
  • Baseline comparison: adequate

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskLots drawn
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskSame number of patients randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
High riskReported for the experimental group only
Other biasUnclear riskInsufficient information

Zhao 1997

Methods
  • Study design: parallel, three-arm RCT

  • Follow-up period: 10 to 12 months

Participants
  • Country: China

  • Setting: Single centre in Jiangsu province

  • Health status: NS

  • Number: group 1 (34); group 2 (26); group 3 (26)

  • Mean age ± SD; range: group 1 (24 ± 2.4; 16 to 45) years); group 2 (24 ± 2.3; 15 to 43 years); group 3 (24 ± 2.0; 16 to 44) years)

  • Sex (M/F): group 1 (18/16); group 2 (12/14); group 3 (15/11)

  • Kidney function: NA

Interventions

Group 1

  • TwHF

    • Dose: 0.6 mg/kg/d for 8 weeks reducing over 10 to 12 months

  • Prednisone

    • Dose: 0.6 mg/kg/d for 8 weeks reducing over 10 to 12 months

Group 2

  • TwHF

    • Dose: 1.0 mg/kg/d for 8 weeks reducing over 10 to 12 months

Group 3

  • Prednisone

    • Dose: 1.0 mg/kg/d for 8 weeks reducing over 10 to 12 months

Outcomes
  • Complete, partial and total remission rates at 10 to 12 months

Notes
  • Baseline comparison: Adequate

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskLots drawn
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskSame number of patients randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
High riskNot reported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
High riskNot reported
Other biasUnclear riskInsufficient information

Zhao 2009

  1. a

    CPA - cyclophosphamide; CSA - cyclosporin; eGFR - estimated glomerular filtration rate; IgA - immunoglobulin A; NA - not applicable; NS - nephrotic syndrome; RCT - randomised controlled trial; SCr - serum creatinine; TwHF - Tripterygium wilfordii Hook F

Methods
  • Study design: parallel RCT

  • Follow-up period: 12 months

Participants
  • Country: China

  • Setting: single centre in Hebei province

  • Health status: refractory primary NS

  • Number: group 1 (26); group 2 (20)

  • Age range: 12 to 55 years

  • Sex (M/F): group 1 (20/6); group 2 (12/8)

  • Kidney function: normal or declining slightly

Interventions

Group 1

  • TwHF

    • Dose: 1.5 to 2.0 mg/kg/d for 4 weeks then 1 mg/kg/d and reducing

  • Prednisone

    • Dose: standard treatment

Group 2

  • CPA

    • Dose: 2 mg/kg/d, total dose < 8 to 10 g

  • Prednisone

    • Dose: standard treatment

Outcomes
  • Complete, partial and total remission rates at 12 months

  • Urinary protein excretion at 12 months

  • Serum albumin at 12 months

  • SC at 12 months

  • Adverse events at 12 months

Notes
  • Baseline comparison: Adequate

  • Drop-out rate: 0

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskLots drawn
Allocation concealment (selection bias)High riskInvestigators were probably aware of allocation sequencing before or until assignment
Blinding (performance bias and detection bias)
All outcomes
High riskBlinding methods not described
Incomplete outcome data (attrition bias)
All outcomes
Low riskSame number of patients randomised and analysed
Free of selective reporting of remission (reporting bias)
Complete, partial and total remission rates at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of proteinuria (reporting bias)
Urinary protein excretion at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of serum albumin (reporting bias)
Serum albumin at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of serum creatinine (reporting bias)
Serum creatinine/eGFR at 3, 6, 12 and(or) 24 months
Low riskReported
Free of selective reporting of adverse events (reporting bias)
Adverse events of drugs at 3, 6, 12 and(or) 24 months
Low riskReported
Other biasUnclear riskInsufficient information

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    CPA: cyclophosphamide; NS: nephrotic syndrome; RCT: randomised controlled trial; TwHF: Tripterygium wilfordii Hook F

Chang 2005Wrong intervention: TwHF + Chinese proprietary medicine (Fu Fang Dan Shen injection) versus prednisone
Du 2008Wrong intervention: TwHF + Chinese proprietary medicine (breviscapine injection) + prednisone versus prednisone
Han 2001Not RCT: allocation according to medical condition
Hu 2008Not RCT: sequential participant allocation
Huang 2003Wrong intervention: non-standard TwHF decoction
Li 2003Wrong intervention: Radix astragali mongolici injection; TwHF part of treatment arm intervention
Lin 2005Wrong outcomes: NS remission rates (complete, partial, total) at 3, 6, 12 and 24 months not reported
Liu 2003aWrong outcomes: NS remission rates (complete, partial, total) at 3, 6, 12 and 24 months not reported
Liu 2003bNot RCT, wrong population: alternate participant allocation; included secondary NS patients
Liu 2010Not RCT: allocation according to medical condition
Lu 2003Not RCT: unequal participant allocation
Ma 1991Not RCT: random allocation to control arm only
Mo 2009Wrong intervention: TwHF + leflunomide versus prednisone + CPA
Pang 1999Not RCT: sequential participant allocation
Song 2008Not RCT: unable to contact authors
Sun 2008Wrong intervention: TwHF + Chinese proprietary medicine (ligustrazine injection) + prednisone versus prednisone
Tang 2004Wrong intervention: TwHF + ticlopidine + prednisone versus prednisone
Wang 1996Not RCT: random allocation to control arm only
Wang 2000Not RCT: participant allocation according to medical condition
Wang 2001Wrong intervention: TwHF + Chinese proprietary medicine (Shu Xue Tong injection) versus prednisone + cyclophosphamide + dipyridamole
Wang 2002Wrong population: patients with secondary NS
Wang 2005aWrong intervention: Chinese proprietary medicine (lumbrokinase tablet) part of the intervention
Wang 2005bNot RCT: alternate participant allocation
Wang 2006Wrong intervention: Liu Wei Di Huang pill, Bao Shen tablet, Jin Kui Shen Qi pill, Bai Ling capsule; TwHF was treatment group intervention
Wen 1996Wrong intervention: TwHF + heparin + prednisone versus prednisone
Xiao 2003Not RCT: outcomes summarised from medical records
Xiao 2007Wrong duration: Treatment duration was two months
Xu 1998Wrong intervention: Liu Wei Di Huang and Shen Qi pills; TwHF treatment group intervention
Xu 2009bNot RCT: sequential participant allocation
Yang 1995Wrong intervention: TwHF + dipyridamole + prednisone versus prednisone
Yang 2002Wrong intervention: TwHF + levamisole + prednisone versus prednisone
Yang 2010Not RCT: sequential participant allocation
Ye 2010Wrong intervention: TwHF + leflunomide versus prednisone + CPA
Zhai 2004Wrong population: paediatric patients with secondary NS
Zhang 2007bWrong intervention: TwHF + Chinese proprietary medicine (breviscapine injection) + prednisone versus prednisone
Zhang 2010Wrong population: elderly patients with secondary NS
Zheng 1987Wrong intervention, not RCT: Random allocation to control only; non-standard TwHF extract
Zhong 2002Not RCT: participant allocation according to medical condition
Zhou 1999Not RCT: unable to contact the authors
Zhou 2000Wrong intervention: six herbs investigated; TwHF treatment group intervention

Ancillary