| Objective:By gathering information from Randomized Control Trials (RCT) on the treatment of Non-Erosive Reflux Disease (NERD) using Traditional Chinese Medicine (TCM), both published and unpublished up to October2013, and analyzing the data using evidence base medicine methodology, to evaluate the effectiveness of TCM treatments for NERD. Thus providing medical evidence for the effectiveness of treating NERD with TCM, as well as offer new insight to further TCM clinical research.Methodology and Data Analysis:A strict set of inclusive and exclusive criteria was established. Thorough information retrieval was conducted using the following databases:Pubmed, Chinese Biology Medical Literature database (CBM), CMCI, VIP Chinese Technique Magazine database, CNKI database, Wan Fang Resource database, Japana Cenra Revuo Medicina, Science Citation Index (SCI), various medical journals as well as selected grey literature. After secondary screening of the full articles, selected studies were subjected to the Jadad scale quality assessment, and relevant data extracted. The data was divided into3groups:"TCM vs conventional medicine","TCM vs Proton Pump Inhibitors (PPIs)","TCM vs a combination of gastroprokinetic agents and PPIs". All3groups were evaluated using the following set of criteria:"Effectiveness of treatment","Rate of relapse","Presence of heartburn","Presence of acid reflux","Presence of eructation","Presence of chest pain". All data was analyzed using the RevMan5.2software provided by Cochrane, which involved calculating the Odds Ratio (OR) or Risk Ratio (RR) and95%Confidence Interval (CI) of Technical data; the Weighted Mean Difference (WMD) or Standardized Mean Difference (SMD) and95%CI of Measured Data.Results:A total of40articles were retrieved after eliminating literature which were not related to the objective of this research,for example, animal testing and non-RCTs. The full articles were then screened for a second time with regards to the set of eliminating criteria, and11articles finally selected to enter the research. Out of the11RCTs selected, some were methodologically flawed, mainly due to vague randomization schemes. It was not stated, in some of the RCTs, whether a commitment scheme or blind experiment was used; reasons for subjects being removed from or leaving the trials were left unmentioned; and only4articles reported cases of adverse reactions. Below are the results of meta-analysis after extraction of relevant data:in the group "TCM vs Conventional Medicine", effectiveness of treatments for11articles showed:Test of Heterogeneity:I^2=37.0%, P=0.10, no significant heterogeneity present, using the fixed effect model, combined effect size RR=1.06,95%CI=[1.01,1.11], P=0.10, there is a significant difference statistically in terms of effectiveness between the treatment and control groups, the effectiveness of treatment being higher in the treatment group. When comparing the rate of relapse, the4articles used gave the following results:Test of Heterogeneity:I^2=0.0%, P=0.93, no significant heterogeneity present, using the fixed effect model, combined effect size RR=0.38, 95%CI=[0.28,0.53], P＜0.00001, there is a significant difference statistically in terms of rate of relapse between the treatment and control groups. The rate of relapse being lower in the treatment group. Comparing the presence of heartburn,5articles were used and gave the following results: Test of Heterogeneity:I^2=94.0%, P<0.0001, significant heterogeneity present, using the random effect model, combined effect size MD=0.02,95%CI=[-0.55,0.58], P=0.96, there is no significant difference statistically in terms of presence of heartburn between the treatment and control groups. The effectiveness of alleviating heartburn for both groups is similar. When comparing the presence of acid reflux,5articles were used and gave the following results:Test of Heterogeneity: I^2=94.0%, P＜0.0001, significant heterogeneity present, using the random effect model, combined effect size MD=-0.02,95%CI=[-0.66,0.61], P=0.95, there is no significant difference statistically for the presence of acid reflux between the treatment and control groups. The effectiveness of reducing acid reflux for both groups is similar. While investigating the presence of eructation,3articles were used:Test of Heterogeneity:I^2=96.0%, P＜0.00001, significant heterogeneity present, using the random effect model, combined effect size MD=-0.56,95%CI=[-1.24,0.12], P=0.10, there is no significant difference statistically for the presence of acid reflux between the treatment and control groups. The effectiveness of alleviating eructation for both groups is similar.Comparing the presence of chest pain,4articles were used:Test of Heterogeneity:I^2=73.0%, P=0.01, significant heterogeneity present, using the random effect model, combined effect size MD=-0.30,95%CI=[-0.57,-0.02], P=0.04, there is a significant difference statistically for the presence of chest pain between the treatment and control groups. The effectiveness of reducing chest pain is better for the treatment group.In the group "TCM vs PPIs",6of the articles used only PPIs as their control, effectiveness of treatments showed:Test of Heterogeneity:I^2=44.0%, P=0.11, no significant heterogeneity present, thus the fixed effect model was used, combined effect size RR=1.06,95%CI=[1.01,1.11], P=0.10, there is significant difference statistically in terms of effectiveness between the treatment and control groups, the effectiveness of treatment being higher in the treatment group. Comparing the presence of heartburn,3articles were involved and gave the following results:Test of Heterogeneity:I^2=96.0%, P<0.0001, significant heterogeneity present, using the random effect model, combined effect size MD=0.16,95%CI=[-0.61,0.94], P=0.68, there is no significant difference statistically in terms of presence of heartburn between the treatment and control groups. The effectiveness of alleviating heartburn for both groups is similar. When comparing the presence of acid reflux,33articles were used and gave the following results:Test of Heterogeneity:I^2=96.0%, P＜0.0001, significant heterogeneity present, using the random effect model, combined effect size MD=-0.13,95%CI=[-0.76,1.01], P=0.78, there is no significant difference statistically for the presence of acid reflux between the treatment and control groups. The effectiveness of reducing acid reflux for both groups is similar. While investigating the presence of eructation,2articles were used:Test of Heterogeneity:I^2=92.0%, P＜0.0001, significant heterogeneity present, using the random effect model, combined effect size MD=-0.39,95%CI=[-1.15,0.37], P=0.32, there is no significant difference statistically for the presence of eructation between the treatment and control groups. The effectiveness for alleviating eructation for both groups is similar. Comparing the presence of chest pain,2articles were used:Test of Heterogeneity:I^2=89.0%, P=0.003, significant heterogeneity present, using the random effect model, combined effect size MD=-0.3295%CI=[-0.85,0.21], P=0.23, there is no significant difference statistically for the presence of chest pain between the treatment and control groups. The effectiveness of reducing chest pain is similar for both groups. In the group "TCM vs a combination of gastroprokinetic agents and PPIs", effectiveness of treatments for4articles showed:Test of Heterogeneity:I^2=31.0%, P=0.32, no significant heterogeneity present, using the fixed effect model, combined effect size RR=1.0495%CI=[0.97,1.12], P=0.24, there is no significant difference statistically in terms of effectiveness between the treatment and control groups, the effectiveness of treatment is similar for both groups. When comparing the rate of relapse, the2articles entered gave the following results:Test of Heterogeneity:I^2=0.0%, P=0.56, no significant heterogeneity present, using the fixed effect model, combined effect size RR=0.35,95%CI=[0.21,0.59], P＜0.00001, there is significant difference statistically in terms of rate of relapse between the treatment and control groups. The rate of relapse being significantly lower in the treatment group.Conclusion:The quality of RCTs involving treatments of NERD with TCM is generally unsatisfactory at present, with many methodological flaws, especially when it comes to precisely stating the type of randomization method and blind experiment employed. While the data shows that in terms of effectiveness of treatment, TCM is more effective than the mere usage of conventional modern drugs, the difference is not tremendously significant. Furthermore, there is a lack results from large clinical trials and pathological observations. There is, therefore, a need to conduct high quality, large sampled clinical trials, to provide further evidence for the effectiveness and safety concerns involving the treatment of NERD with TCM. Yet at the same time, it is necessary to take into consideration the unique characteristics of TCM, combine it with the essence of evidence based medicine, and seek to develop a constructive research method for TCM. Thereby improving the standards of TCM clinical research, provide further reliable scientific evidence to aid clinical decision-making, and discover innovative methods for the development of TCM. |
PDF Full Text Request