Systematic Review Of Randomized Controlled Trials Of Traditional Chinese Medicine Treatment Of Type 2 Diabetes And Methodological Evaluation Studies

Chinese materia medica (CMM) has long been used for patients with type 2 diabetes mellitus (T2DM) alone in China under the theory of traditional Chinese medicine (TCM) or combined with the hypoglycemic agents (HA). In TCM, T2DM is also known as Xiao-Ke and its main pathogenesis is deficiency in origin and excess in superficiality (ben xu biao shi) manifesting different symptoms and signs which can be divided into separate syndrome differenciation patterns with different therapeutic principles and CMMs. Nowadays people are increasingly concerned about its efficacy and safety on treatment of diabetic patients.[Objective]The aims of this thesis were to discuss the methodological quality of the randomized controlled trials (RCTs) of CMMs for treatment of T2DM, and to evaluate its efficacy and safety by conducting systematic review.I Methods]1. Electronic databases of CNKI, VIP, SinoMed, Wanfang Data, Medline, EMbase and Cochrane Library were searched from inception through January 2012. RCTs with treatment duration≤3 months were selected.2. The data extraction form was created based on CONSORT statement checklist. Descriptive analysis was carried out to describe the reporting rate and the trend of each PICOS item changed before and after the year of 2006 in the included RCTs.3. The power of effect size was estimated based on sample size and outcomes parameters to assess its reliability and effectiveness.4. The methodological quality of the eligible studies was assessed according to the Cochrane risk bias checking list.5. In this systematic review, we included the trials of comparisons of CMM versus placebo, CMM versus HA, CMM plus HA versus HA, CMM versus no treatment and CMM versus non special treatment, and included> 1 of the following outcomes:mortality, incidence of diabetes-associated complications, glycemic control (HbAlc, FPG,2-h PG), adverse effects, and health-related quality of life (QOL), symptom improving score and syndrome improving score.[Results]1. A total of 232 studies meeting the inclusion criteria were identified, of which 170 RCTs were used for meta-analysis. 2. The clinical trials registration rate is lower than 1%, and only 5% of the included studies reported Ethics Committee approval.3. The TCM diagnostic criteria and the inclusion criteria of syndrome differentiation were reported in 58.2%(135/232) and 32.8%(76/232) of the included RCTs respectively.4. The proportion of placebo-control trials and the first-line drugs in the active-control drugs rose from 6.9%(7/101) and 25.5%(24/94) before the year of 2006 to 13.7%(18/131) and 51.3%(58/113) after 2006 (P<0.001).5. The reporting rate for outcomes of HbAlc, FPG,2-h PG and QOL score were 55.6% (129/232),95%(220/232),76%(176/232),0.4%(1/232) respectively. Among superiority trials, only 14%(12/81) of the trials had the power more than 80%.6. The reporting rates of random number generation, blinding and ITT analysis were 17.2% (40/232),9.5%(22/232) and 2.2%(5/232) respectively. The proportion of reporting allocation concealment and outcome attrition rose from 1%(1/101) and 2%(2/101) before the year of 2006 to 7.6%(10/131) and 15.3%(20/113) after 2006 (P< 0.05).7. Semen Trigonellae saponins Capsules and Extracts of Touchi had beneficial effects on HbAlc and 2-h PG level control, and the Korean red Radix ginseng and Semen Trigonella saponins Capsule had effect on reducing FPG level compared to placebo.8. Some CMMs combined with HA were more conducive than HA alone to reducing HbAlc, FPG,2-h PG, symptom score and QOL socre level.9. Any conclusions of CMMs instead of HA on glycemic control can not be reached, for 52 RCTs with high risk bias were included in the comparisons of CMM versus HA.10.27.65%(47/170) of the included studies reported adverse effects.[Conclusion]1. Implications for research:Clinical trial should be registered at or before the onset of participant enrollment. Important outcome measures such as end point outcomes, HbAlc, QOL score should be included in efficacy and effects evaluation studies. Data and safety should be carried out a mid-term monitoring by an independent data monitoring committee, and the outcomes results should be reported standardized based on CONSORT statement.2. Implication for practice:Some CMMs have efficacy and effects in respect to blood glucose control for diabetic patients. However, we cannot recommend any of the examined CMMs for clinical routine use for the limitation of low quality of clinical trials; and the conclusions on the safety of CMMs for treatment of T2DM needed additional information.