Common Mechanism of Pathogenesis in Gastrointestinal Diseases Implied by Consistent Efficacy of Single Chinese Medicine Formula: A PRISMA-Compliant Systematic Review and Meta-Analysis.

Gastrointestinal (GI) disorders often manifest similar symptoms with overlapping clinical diagnosis and unmet medical needs. Traditional Chinese medicine (TCM) has history-proven benefits for GI diseases; albeit language barrier prevents Western readers from accessing the original reports in Chinese. The TCM formula Si-Ni-San (SNS) consists of 4 herbs targeting on homeostatic disturbances characterized by "reflux" and "irritable" problems. Here we used SNS as a therapeutic tool to explore the common mechanisms of pathogenesis in non-neoplastic GI diseases.Data sources from PUBMED, Chinese National Knowledge Infrastructure, and Wanfang databases were searched for clinical trials. Comparisons were SNS as intervention and Western conventional medicine as control, which treat patients with upper GI disorders (gastroesophageal reflux disease, peptic ulcer, chronic gastritis, duodenogastric reflux), lower GI diseases (irritable bowel syndrome, ulcerative colitis), and functional dyspepsia. Participants and studies in accordance with the Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement were eligible. We used the Jadad scale to assess methodological qualities, the fixed or random-effect model to evaluate therapeutic efficacy, and the funnel plots to explore publication bias. Outcome was clinical efficacy defined by symptom relief with normal GI endoscopy, radiology, and pathology.We included 83 studies involving 7762 participants: 1708 versus 1397 of the upper GI disorders in 34 studies, 901 versus 768 of the lower GI diseases in 19 studies, 1641 versus 1348 of functional dyspepsia in 30 studies, and 328 versus 287 of relapse rate in 8 studies. Six studies had a Jadad score >2 points and the rest were <2 points. Pooled data showed significant efficacy of SNS for the upper GI disorders (odds ratio [OR] = 3.9, 95% confidence interval [CI] = 3.09-4.92), lower GI diseases (OR = 4.91, 95% CI = 3.71-6.51), and functional dyspepsia (N = 2989; OR = 3.94, 95% CI = 3.17-4.90). The relapse rate was 12.9% for SNS, significantly <46.5% for conventional therapies (OR = 0.16, 95% CI = 0.11-0.25).The consistent efficacy of the single TCM formula implicates common mechanisms of pathogenesis in GI disorders.

INTRODUCTION

Individuals with digestive problems are often diagnosed on symptoms grounds alone. Dyspepsia, heartburn, non-cardiac chest pain, abdominal pain, chronic diarrhea, and constipation are common symptoms complained by individuals who have no histopathological explanation. Common digestive diseases such as gastroesophageal reflux disease (GERD), chronic gastritis, duodenogastric reflux, and irritable bowel syndrome (IBS) are clinically symptom-based diagnosis with considerable overlap and symptom fluctuation over time.[1] GERD symptoms in individuals with IBS are 4-fold that of individuals without IBS.[2,3] Similarly, IBS symptoms frequently co-exist with biopsy-proved celiac disease,[4] Crohn disease, and ulcerative colitis.[5] Furthermore, IBS individuals usually suffer from dyspepsia[6,7] and chronic idiopathic constipation.[8] It remains uncertain whether all these common digestive disorders share common mechanisms of pathogenesis.[2,9]

Traditional Chinese medicine (TCM) is typically symptoms-based approach with history-proven therapeutic efficacy. TCM physicians have used classic formula comprising of several ingredient herbs to achieve symptom relief,[10] sustain metabolic homeostasis,[11,12] and prolong patients’ survival.[13] One of the classic formulas designed about 220 B.C. for relieving digestive symptoms is Si-Ni-San (SNS), consisting of 4 herbs Radix Bupleuri, Radix paeoniae Alba, Fructus Aurantii Immaturus, and Radix Glycyrrhizae. Most of the clinical studies of SNS have been reported in Chinese, not readable by any non-Chinese. Hereby, we conducted a systematic review and meta-analysis to evaluate the clinical efficacy of SNS on common digestive disorders, and to use TCM as a tool to validate the hypothesis that all common GI problems have shared mechanisms of pathogenesis.

METHODS

The present study was approved by the Ethics Committee Board of Guilin Medical University (GLMC030811HL) and conducted in accordance with the Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement. We had also reviewed each article and found 5 articles mentioned in the Method section that ethical approval and written informed consent were obtained.

Search Strategy and Databases

We searched up to March 2014 the following electronic databases: the Chinese National Knowledge Infrastructure (CNKI), the Chinese Science and Technology Periodical Database, the Wanfang Database and the PubMed. All potentially relevant articles including reference lists of retrieved articles were investigated as full text in English or Chinese. For ambiguous or missing information, we contacted the authors where possible. For duplicate publications, the original publication was used. Medical terms used in literature search were as follows: gastroesophageal reflux diseases, peptic ulcer, duodenogastric reflux disease, chronic gastritis, irritable bowel syndrome, functional dyspepsia, and ulcerative colitis, and in combination with “Si-Ni-San (Sini San or sinisan)”. Figure 1 shows the ingredient herbs and their ratios of SNS.

FIGURE 1

Ingredient herbs and their ratio for Si-Ni-San formula.

Eligibility Criteria

Studies meeting the following criteria were included: randomized controlled trials stating the phrase “randomization” (no restriction was imposed on studies with respect to blinding and type of design such as parallel or cross-over); participants with GERD, peptic ulcer, duodenogastric reflux, functional dyspepsia, chronic gastritis, IBS, or ulcerative colitis, irrespective of age, sex, ethnic origin, and geography; the patients were diagnosed using the latest guideline by the year of the study conducted (pregnant, lactating women, and patients with serious medical conditions were excluded); and intervention was SNS, whereas control could be western conventional medicine, studies with co-intervention were excluded if they were given to both groups.

Studies meeting the following criteria were excluded: duplication (the same data of patients with the same authors published in different journals); information of diagnostic criteria, participants, interventions, or outcomes were not defined; observational studies, reviews, and case series reports; studies not meeting the inclusion criteria. Eligibility assessment was performed independently by 2 investigators (LW and JW), using pre-designed eligibility forms, with all questions resolved by consensus with other authors.

Data Extraction and Quality Assessment

Two reviewers (LY and LW) independently conducted the literature search, study selection, and data extraction. The extracted data included authors, title of the study, publication date, study design, characteristics of participants, details of intervention, outcome measures, intervention durations, adverse events, and any relapse of uncomfortable symptoms. Disagreement was resolved by discussion and consensus with the TCM experts (SY and JW). The quality assessment of the trials selected for inclusion was evaluated using the Jadad score.[16] The final Jadad score ranged from 0 to 5 points, with high scores indicating high quality. Studies with a Jadad score of 2 or less were considered to have low quality and those with a Jadad score of ≥3 were considered to have high quality.[17]

Outcome Measures

Outcome was clinical efficacy defined by symptom relief with normal endoscopies results: relief of the clinical symptoms according to the latest guidelines implemented by the year the study conducted; normalization of GI endoscopies, radiology, and pathology.

For functional dyspepsia diagnosed on symptoms grounds alone, we strictly followed the Rome III criteria. The formula to calculate the clinical effect index (EI) is as follows: EI = (pretreatment scores – post-treatment scores)/pretreatment scores × 100%, whereas the treatment scores were calculated by the degree of clinical symptoms.

Relapse rate and adverse events were extracted for the evaluation of sustained effectiveness and safety concern.

Data Synthesis

Revman 5.1 software provided by the Cochrane Collaboration was used to combine results from >2 separate trials to generate forest plots of pooled efficacy rates, pooled odds ratio (OR), and 95% confidence interval (CI). Before the results of the studies were combined, statistical heterogeneity among studies was estimated using the chi-square test and I2 test (P > 0.05 and I2 <50% indicate acceptable heterogeneity between the pooled studies). Fixed-effect model can be appropriate when there is statistical homogeneity (P > 0.1, I2 <50%) among the studies, and random-effect model has to be pursued when statistical heterogeneity (P < 0.1, I2 >50%) exists in the trials. Intervention effects were expressed OR and the associated 95% CI as calculated for dichotomous outcomes. The funnel plot was used for publication bias. A symmetric inverted funnel indicates that publication bias is unlikely, whereas an asymmetric funnel signifies the possibility of either publication bias or a systematic difference between smaller and larger study effects.

RESULTS

Study Description

A total of 859 articles were initially identified and eventually 83 randomized controlled studies, involving 7763 patients (4,250 in SNS groups and 3,513 in control groups) were in accordance with our inclusion criteria (Figure 2). Among the 83 studies, listed in the Appendix, http://links.lww.com/MD/A331, 6 studies were postgraduate candidate thesis, and 77 journal articles. All the 83 studies were conducted in China. Trials treating GERD were observed in 7 studies, peptic ulcer in 6 studies, functional dyspepsia in 30 studies, chronic gastritis in 6 studies, duodenogastric reflux in 15 studies, IBS in 15 studies, and ulcerative colitis in 3 studies. The duration of all studies ranged from 15 days to 90 days. Eight RCTs had reported the relapse rate after treatment discontinuation (615 patients, 328 in SNS groups and 287 in control groups). Clinical characteristics are summarized in Table 1 .

FIGURE 2

Flow chart of the study selection process.

Table 1

Summary of the Characteristics of the Included Trials

Outcome of Interventions

Effects of SNS Versus Conventional Therapy on Upper GI Diseases

Thirty-four independent trials (SNS: 1708; control, 1397 patients) reported SNS-treated GERD, peptic ulcer, chronic gastritis, and duodenogastric reflux with homogeneity in the consistency of the trial results (P = 1.00, I2 = 0%); therefore, fixed-effects model was used for statistical analysis. As shown in Figure 3A, higher efficacy rate was attributed to SNS than conventional therapy for duodenogastric reflux (OR = 3.83, 95% CI = 2.71–5.41), GERD (OR = 3.93, 95% CI = 2.42–6.38), chronic gastritis (OR = 5.09, 95% CI = 2.83–9.14), and peptic ulcer (OR = 2.99, 95% CI = 1.65–5.43). The combined OR was 3.90 (95% CI = 3.09–4.92) with significant overall effect (Z = 11.44, P < 0.001). The funnel plot was roughly symmetric, indicating little publication bias for the 4 diseases (Figure 3B).

FIGURE 3

Efficacy rates and publication bias of the included 34 studies on upper gastrointestinal diseases. (A) Meta-analysis of the efficacy rate of Si-Ni-San versus conventional therapy in the treatment of upper GI diseases. (B) Publication bias of the included studies.

Effects of SNS Versus Conventional Therapy on Lower GI Diseases

SNS-treated 901 patients and 768 control subjects were included in 19 studies of lower GI diseases (16 in IBS, 3 in ulcerative colitis). Fixed-effects model was used for statistical analysis (P = 0.81, I2 = 0%). Consistently, SNS showed higher efficacy rates than conventional treatment (IBS: OR = 4.81, 95% CI = 2.71–5.41; ulcerative colitis: OR = 2.40, 95% CI = 1.21–4.75). Pooled results showed an OR as 4.91 (95% CI = 3.71–6.51) with overall effect as 2.50 (P < 0.001) (Figure 4A). The funnel plot demonstrated no apparent asymmetry, suggesting publication bias unlikely (Figure 4B).

FIGURE 4

Efficacy rates and publication bias of the included 19 studies on lower GI diseases. (A) Meta-analysis of efficacy rate of Si-Ni-San versus conventional therapy in the treatment of lower GI diseases. (B) Publication bias of the included trials.

Effects of SNS Versus Conventional Therapy on Functional Dyspepsia

Thirty studies of functional dyspepsia involving 2989 participants (1641 in SNS group) were qualified for the comparison with significant heterogeneity in the 2 groups (P = 0.99; I2 = 0%); thus, fixed-effects model was used for statistical analysis. The outcomes favored SNS group by pooled data (OR = 3.94, 95% CI = 3.17–4.90) and test for overall effect (Z = 12.39, P < 0.001) (Figure 5A). The funnel plot was roughly symmetric, indicating little publication bias of the studies (Figure 5B).

FIGURE 5

Efficacy rates and publication bias of the included 30 studies on functional dyspepsia. (A) Meta-analysis of efficacy rate of Si-Ni-San versus conventional therapy in the treatment of functional dyspepsia. (B) Publication bias of the included trials.

Relapse Rate of SNS Versus Conventional Therapy on Treating GI Diseases

Among the 83 studies, 8 have addressed the relapse problems (3 studies in IBS, 2 in functional dyspepsia, 2 in peptic ulcer, 1 in GERD). The observation period ranged from 3 months to 6 months. As shown in Figure 6A, meta-analysis of the 8 studies strongly favored SNS than conventional therapy for clinical efficacy (OR = 3.54, 95% CI = 2.29–5.47). In contrast, relapse rate was more common in conventional group than the SNS-treated subjects (OR = 0.16, 95% CI = 0.11–0.25), with overall effect of 8.11 (P < 0.001); the relapse rate was 12.9% for SNS, significantly lower than 46.5% for conventional therapy (Figure 6A). Funnel plot provided evidence of publication bias (Figure 6B, C).

FIGURE 6

Relapse and efficacy rates of the included 8 studies on gastrointestinal diseases. (A) Relapse and efficacy rates of Si-Ni-San versus conventional therapy in the 8 studies. (B) Publication bias of the 8 studies on the relapse rate. (C) Publication bias of the 8 studies on the efficacy rate.

None of the included 83 studies reported mortality or acute incidents such as hemorrhage and perforation.

Methodological Quality and Adverse Effects

Based on randomization, blinding and description of withdrawal, the Jadad score varied greatly from 1 to 4 points, whereas 6 studies (7.2%) were classified as high quality. Seven mentioned randomization, 1 described blinding, and 5 (6.0%) provided information of dropout or withdrawal. Seven studies reported adverse events: 3 reported no adverse events and 4 reported more frequent adverse effects in the SNS-treated groups than control groups. The adverse effects were gastrointestinal symptoms, such as nausea, vomit, and abdomen uncomfort. All the adverse effects were mild and tolerable and did not result in treatment withdrawal. Table 2  summarizes the results of the methodological quality item for each included studies.

Table 1 (Continued)

Summary of the Characteristics of the Included Trials

DISCUSSION

Summary of Main Findings

This is the first attempt to synthesize clinical data of single formula for 7 different GI disorders. In this systematic review, SNS was used as a tool to validate the common pathogenesis of the 7 GI disease entities. The efficacy of the single TCM formula SNS is consistently validated for functional dyspepsia and the other 6 GI disorders, indicating that all the 7 GI disorders may have shared mechanisms of common pathogenesis.

Common Mechanisms of Pathogenesis

GI symptoms often manifest similar symptoms and diagnosed on symptoms ground alone. A vast number of treatment strategies were introduced to relieve the symptoms of GI diseases.[18,19] However, few of them could provide complete control of reflux symptoms,[20] indigestion, abdominal pain,[21] diarrhea, and constipation. Several studies have demonstrated the overlaps among different GI diseases;[1,2,6] multiple mechanisms such as abnormal GI motility,[22] visceral hypersensitivity,[23] impaired GI mucosa barrier,[24] and central nervous system factors[25] are likely involved to explain the phenomenon, yet few are holistic and reasons for overlaps remain speculative. Consequently, the definitions of non-neoplastic GI disorders remain confounding with unmet clinical needs.

The consistent efficacy of this single Chinese formula SNS on 7 GI diseases may provide a novel insight and alternative prospective. In this study of synthesized data of SNS, TCM serves as a tool to validate the common mechanisms of digestive disorders. The first potential reason for the apparently increased risk of overlaps in GI disorders may link to “reflux.” “Reflux,” on one hand, could be defined as the regurgitation of the lower digestive track contents into upper organs.[26] Researchers found disturbed motility in functional dyspepsia and IBS.[27,28] Many reports have also demonstrated specific association between GERD and functional dyspepsia,[29,30] IBS,[2,31] and ulcerative colitis.[32] However, the bacterial dysbiosis and relocation might be an important etiology factor for reflux disorders.[33,34] Walker review article highlights the upper gastrointestinal bacteria and associations with disease such as IBS and coeliac disease.[35] Yang and colleagues’ findings also raise the issue of a possible role for microbiome dysbiosis in the pathogenesis of reflux-related GI disorders.[36]

It is well-known that there are several annulus muscles functioning as “gates” or one-way moving “check-points” along the GI tract. These muscles include the orbicularis oris muscle, preventriculus, pylorus, oddi sphincters, ileocecal valve, orifice of vermiform appendix, and the anus. All the “gates” are fixed with sphincter or smooth muscle as barriers, which can resist effacement and opening when challenged by lower contents. Failure to do so results in episodes of lower gut juice refluxing into upper digestive tracts.[37,38] Therefore, one shared mechanism relies on the “reflux” because of the inability of such sphincters and smooth muscles. Sphincters are important to GI functions.[39-41] A manometric study has found that IBS patients exhibited significantly lower esophageal sphincter pressures compared with age and sex-matched controls.[39] Other researchers reveal a fluctuation among GI hormones,[42] glucose,[43,44] and oxidative free radicals[24] in patients with damages to the sphincters and smooth muscles. Interestingly, TCM formulae such as SNS exhibit consistent efficacy for maintaining the normal function of the sphincters, and thus may correct most, if not all, reflux-associated disorders.[45,46]

The second common mechanism of pathogenesis refers to irritable stimulation. Irritable comorbidity, including emotional irritation, anger, and depression, is prevalent in GI diseases. GI patients with persistent emotional irritation, especially anger and anxiety, are usually suffering from GI disorders.[47] Epidemiologic, psychophysiological, and functional neuroimaging studies have partially elucidated the mechanisms underlying the relation between cognitive-affective processes on the one hand and GI function and symptom reporting on the other. A nationwide cohort study in Taiwan suggests that psychiatric patients using antidepressant agents have increased risk of upper GI bleeding.[48] In IBS, 50% to 90% of those seeking treatment have comorbidity of lifetime psychiatric disorders, especially depressive and anxiety disorders.[49] In another systematic review and meta-analysis, patients with IBS had significantly higher levels of anxiety and depression than healthy controls.[50] Furthermore, irritable GI causes visceral hypersensitivity. GI patients demonstrated lower sensory thresholds for diarrhea, constipation, and abdominal pain[51,52] when taking ice foods[23] and experiencing climate change of weather. Intriguingly, SNS has history-proven beneficial effects on reliving GI irritability.[53-55]

The third common mechanism underlying the apparent overlaps of GI diseases is the stasis of GI microcirculation. Previous researches have demonstrated catecholamine and dopamine fluctuation in function dyspepsia,[56] IBS,[57,58] and peptic ulcer.[59] Elikowski et al[60] and Mitsuyama et al[61,62] found a disturbed blood viscosity in IBS and colitis, causing the imbalance of myogenic homeometric autoregulation and resulting in a stasis of abdominal circulation. Accordingly, radix paeoniae Alba, one component of SNS, has benefits on artery pressure,[63] inflammation, allergy,[64] and smooth muscle dilation,[65] and consequently improves GI microcirculation. The abnormal contractions of abdominal vascular smooth muscle and the inappropriate hormone secretion constitute rationale for the SNS-induced efficacy on abnormal motility, visceral hypersensitivity, bowel irritability, and mucosal barrier disruption.[66-68]

Overlap of functional gastrointestinal disorders, GERD, peptic ulcer, and IBD may exist more than by chance. But pathogenesis of these diseases is very complex and multifactorial. Such as Helicobacter pylori is one of the most important causes of GU, but no for GERD. Intriguingly, the effects of TCM, herbal formula such as SNS in particular, are also multiple and plural. Therefore, the multiple effects of SNS might target the multifactorial pathogenesis in common digestive disorders.

CONCLUSION AND LIMITASIONS

In this study, we have used synthesized clinical data of the single TCM formula as a tool indirectly to validate the common mechanisms of pathogenesis of GI disorders. The findings are positive for common GI disorders implicated by similar pathogenesis. The present study has limitations such as inherited risk bias and low quality of some included trials. Validation of our findings warrants high-quality clinical studies based on different geographic locations or using different therapeutic agents.

Table 2 (Continued)

The Methodological Quality of the Included Trials

Table 2 (Continued)

The Methodological Quality of the Included Trials