Potential effectiveness of traditional Chinese medicine for cardiac syndrome X (CSX): a systematic review and meta-analysis

Background Treatment of cardiac syndrome X with unknown pathological mechanism remains a big challenge for clinicians. Complementary and alternative medicine may bring a new choice for its management. The aim of this study is to evaluate the clinical effects of traditional Chinese medicine on cardiac syndrome X patients. Methods We systematically searched databases such as Cochrane CENTRAL, PubMed, EMBASE, CBM, Chinese National Knowledge Infrastructure (CNKI), WanFang and VIP, and handsearched relevant journals to identify randomized controlled trials. Following the steps of systematic review recommended by the Cochrane group, we assessed the quality of included studies, extracted valid data and undertook meta-analysis. Results Twenty one moderate-to low-quality randomized controlled trials involving 1143 patients were included. The results showed that traditional Chinese medicine could improve angina [OR=1.34, 95% CI: 1.2 to 1.50], electrocardiogram (ECG), endothelin-1 (ET-1) levels, prolong exercise duration in treadmill tests, and reduce angina frequency per week compared with routine treatment. No other side effect was reported except two cases of stomach pain. Conclusion Compared with conventional treatment, traditional Chinese medicine shows the potential of optimizing symptomatic outcomes and improving ECG and exercise duration. The efficacy of TCM may find explanation in its pharmacological activity of adjusting the endothelial function. TCM, as a kind of alternative and complementary medicine, may provide another choice for CSX patients.

Results: Twenty one moderate-to low-quality randomized controlled trials involving 1143 patients were included. The results showed that traditional Chinese medicine could improve angina [OR=1.34, 95% CI: 1.2 to 1.50], electrocardiogram (ECG), endothelin-1 (ET-1) levels, prolong exercise duration in treadmill tests, and reduce angina frequency per week compared with routine treatment. No other side effect was reported except two cases of stomach pain.
Conclusion: Compared with conventional treatment, traditional Chinese medicine shows the potential of optimizing symptomatic outcomes and improving ECG and exercise duration. The efficacy of TCM may find explanation in its pharmacological activity of adjusting the endothelial function. TCM, as a kind of alternative and complementary medicine, may provide another choice for CSX patients.

Background
Cardiac syndrome X (CSX, also called microvascular angina) refers to typical stable angina that is exclusively or predominantly induced by effort. Upon diagnostic investigation, findings are comparable to those of myocardial ischemia, showing normal or near-normal coronary arteries on angiography and an absence of any other specific cardiac diseases (e.g., variant angina, cardiomyopathy, or valvular diseases) [1]. Approximately 10% to 20% of affected patients suspected to have angina reportedly turned out to have negative coronary angiogram results [2]. Although the long-term prognosis of CSX does not necessarily include increased mortality, patients' quality of life is invariably affected and the incidence rates of cardiovascular and cerebrovascular events are increased [3]. Exacerbated and recurrent angina may also lead to physical discomfort, frequent hospital readmissions, or even repeat coronary angiography, imposing patients with a huge economic burden [4]. Moreover, patients with CSX tend to show high scores on psychological inventories that measure anxiety and depression [5].
The etiology and pathology of CSX remain unclear. Although hypotheses claiming that the disease is associated with endothelial dysfunction, inflammation, oxidative stress, or estrogen deficiency have been formulated, they lack evidence [6,7]. Internationally, conventional drugs against ischemia, such as beta-blockers, calcium antagonists, xanthine derivatives, angiotensin-converting enzyme inhibitors, estrogen, and statins, are recommended for clinical use in patients with CSX. However, their respective curative effects remain controversial [8][9][10]. We believe that alternative and complementary medicine may provide additional treatment options. As more papers on treatment with Traditional Chinese Medicine (TCM) were published in recent years, it was suggested that Chinese herbs might bring health benefits for patients with CSX [3]. TCM has a unique system of interpreting the etiology and pathology of CSX [11], allowing this syndrome to be treated accordingly.
Moreover, modern pharmacological studies have found that the active ingredients in many of these herbs have functions related to endothelial protection, antiinflammation, antioxidative stress, and improvement of estrogen function, all of which are believed to be effective in the pathology of CSX [12][13][14][15][16][17]. Therefore, the relatively inexpensive TCM treatment for CSX is worthy of attention. This study aimed to systematically and objectively evaluate the clinical curative effect and safety of TCM for     Herbal supplements in the decoction: 1 Nourishing Qi and Blood Decoction (huangqi 30 g, chaihu 12 g, gualoupi 15 g, xiebai 15 g, houpou 15 g, jiangbanxia 9 g, zhiqiao 9 g, huanglian 3 g, tanxiang 9 g, dansen 9 g, chuanxiong 9 g, danggui 6 g, baishao 12 g, baizhu 15 g, zhigancao 6 g, dangshen 12 g).
CSX based on a general understanding of previous research on CSX.

Methods
This systematic review is conducted according to the Preferred Reporting Items for Systematic Reviews and Metaanalyses: Additional file 1: The PRISMA Statement.

Ethics
Data for this study was acquired through previously published work, no patient or hospital data was accessed.
Therefore, written consent and institutional ethical review was not required for this research.

Search strategy
We systematically searched for studies on CSX published in either Chinese or English. The databases that were searched included the Cochrane Central Register of Controlled Trials (Issue 3,   We also checked the references of literature for possible identification of relevant studies. Electronically inaccessible journal articles were manually retrieved.

Inclusion criteria
The study is a randomized controlled trial (RCT). The study includes participants diagnosed with CSX by the criteria listed in ANGINA PECTORIS AND NORMAL CORONARY ARTERIES: CARDIAC SYNDROME X [18]. The study includes intervention and comparison of any of the following: A. TCM + routine treatment* vs. routine treatment B. TCM + routine treatment vs. routine treatment + placebo. C. TCM vs. routine treatment  *Routine treatment includes nitrate medications, beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and calcium channel blockers, as recommended in the essay Therapeutic Options for the Management of Patients with Cardiac Syndrome X [19]. Outcomes to be measured include symptom improvement and electrocardiograph (ECG) changes and/or treadmill test results. Any adverse event defined in "ICH-GCP 1997" [20] should be recorded.

Exclusion criteria
We excluded studies with unclear diagnostic criteria and without full texts or the use of other TCM therapeutic methods except herbs. Intervention for the treatment group was limited to Chinese herbs. Combinations of herbs and other forms of treatment (e.g., acupuncture or moxibustion) were excluded.

Study selection
Two researchers (W.J.Y. and X.L.) independently undertook the aforementioned search and managed the search results using Note Express 2 software. Ineligible studies were excluded after reading the title and abstract, and the full texts of the remaining studies were scanned for confirmation. When disagreement arose between the two researchers, they sought help from a third reviewer (S.H.C.).

Data collection
Both researchers (W.J.Y. and X.L.) independently extracted data from all included studies and input them into two Excel sheets. One sheet contained the general information of the studies, including the authors' names (the first author was listed if there was more than one), publication date, sample size, age of participants, gender, interventions and comparisons, treatment courses, and outcomes. The other sheet included items for quality assessment listed in the Cochrane risk of bias tool [21]. The above information was then carefully checked by two other researchers (C.J. and Z.S.). Any disagreement was resolved through discussion, and errors were corrected.

Statistical analysis
The two researchers (W.J.Y. and X.L.) used Review Manager 5.0.2, provided by the Cochrane corporative network, to analyze the data. Effect measures were presented using odds ratios (OR) for dichotomous data and weighted mean differences (WMD) or standardized  mean differences (SMD) for continuous data, both with 95% confidence intervals (CI). The chi-square test was used to test heterogeneity across studies [22] with a significance level of 0.05. Data were analyzed with a fixedeffect model if no statistical heterogeneity was observed (P > 0.05 or I 2 < 50%). In the presence of heterogeneity, the two researchers (W.J.Y. and X.L.) checked the data entered and explored the variation by conducting subgroup analysis. If the variation could not be explained, we performed a random-effects meta-analysis and interpreted the effect measure with care. A funnel plot was formed to detect publication bias [23].

Subgroup analysis, sensitivity, and meta-regression
Subgroup analysis was conducted to show the magnitudes of the effects in different subgroups and determine whether there was a different effect of an intervention in different situations. Meta-regression analysis was conducted to identify possible causes for differences across the groups.

Results
A total of 903 studies (900 in Chinese and 3 in English) were identified through electronic searches; 101 of them were included for further assessment after both researchers examined the title and abstract of all studies and excluded duplicates, animal experiments, reviews, and others. We read the full texts of all 101 articles and discovered and excluded other ineligible studies. The final result included 21 articles , all in Chinese (see Figure 1). Of the 21 RCTs, 12 compared TCM plus routine treatment with routine treatment alone, eight compared TCM with routine treatment, and the remaining RCT (Liang 2005) compared TCM combined with routine treatment versus placebo, which were combined with routine treatment. TCM interventions included Chinese patented drugs (12 studies), decoction (seven studies), and injection (two studies). With regard to outcome measures, angina improvement was reported in 15 studies, angina frequency in nine studies, and treadmill test results in 11 studies. Eight trials used ECG improvement as the primary outcome measure, two trials used TCM syndrome improvement, and seven used changes in cytokine levels such as endostatin ET-1 or C-reactive protein (CRP). Only one  study reported fractional flow reserve. Ten studies reported safety outcomes (see Table 1).

Quality assessment of the included studies
All 21 articles mentioned the word "randomization," but only two studies Bi 2003) elaborated on the randomization method (random number table). Only two studies (Peng 2011 andNiu 2008) mentioned the performance of single blinding. None mentioned allocation concealment or intention-to-treat analysis (see Table 2).

Efficacy and safety analysis
The 21 RCTs were divided into two subgroups for further analysis with consideration of clinical heterogeneity across the studies. Group A (13 studies) compared TCM plus routine treatment with routine treatment alone, while Group B (eight studies) evaluated the effects of TCM relative to routine treatment. One study that compared the effects of TCM plus routine treatment with that of placebo plus routine treatment was assigned to Group A. Descriptions and interpretations of angina and ECG improvement for patients in all studies followed the rules prescribed in the Efficacy Criteria for Angina of Coronary Heart Disease [45]. TCM syndromes were determined against the same criteria: Guideline for Clinical Research of Chinese Medicine (New Drug) [46].

Subgroup analysis
Group A: TCM + RT vs. RT Nine studies involving 239 participants in the treatment group reported angina improvement. No statistical heterogeneity was found among these studies (P = 0.13 > 0.05, I 2 = 36%) (see Figure 2). The results showed that TCM combined with routine treatment was more effective than routine treatment alone (OR = 1.34, 95% CI = 1.2-1.50).
Eight studies reported the patients' angina frequency per week. A random-effect model was applied because obvious heterogeneity was observed among these studies, as seen in Figure 3 (P < 0.00001, I 2 = 94%). The results indicated that patients in the treatment group experienced angina almost five times less frequently than those in the control group (WMD = −4.91, 95% CI = −6.56 to −3.25).
Nine studies reported outcomes of the Bruce Protocol Treadmill Test. Either exercise duration or the time to 1-mm ST segment depression was recorded. Exercise duration (measured in seconds) is known to be a health predictor for patients with coronary heart disease [47] and was reported in seven of the nine studies (excluding Wu 2010 and Li 2009). A random-effect model was applied because obvious heterogeneity was observed (P < 0.0001, I 2 = 79%) (see Figure 4). Patients in the TCM plus routine treatment group had a nearly 1-min improvement in exercise duration (WMD = 77.31, 95% CI = 39.70-114.93).
Five studies reported changes in ET-1 levels. As seen in Figure 5, statistical heterogeneity was observed and the units of outcomes varied; thus, a random-effect model was used (P = 0.002, P < 0.05, and I 2 = 77%). Pooled results indicated greater effects of TCM combined with routine medicine in decreasing ET-1 levels (SMD = −1.12, 95% CI = −1.73 to −0.50).
Spearman correlation analysis using SAS9.1 software showed no correlation between angina improvement and reduced angina frequency or between angina improvement and changes in ET-1 levels.

Sensitivity analysis and meta-regression
No statistical heterogeneity is observed in Figures 2, 6, or 7. However, clinical heterogeneity across trials still needs special attention. The final effect sizes were influenced by multiple factors such as sample size, medication course, publication year, and even different forms of Chinese medicine. As seen in Figures 3 and 4, metaregression was conducted to test the reliability of the pooled analysis and search for possible causes for this heterogeneity. We conducted meta-regression by residual maximum likelihood (REML) with Knapp-Hartung modification. Tau2 was equal to 0.02809 in Table 3, 6.82 in Table 4, and 1700 in Table 5 as REML estimates of between-study variance. I 2 was 0.02809%, 91.27%, and 61.44% in terms of the proportion of residual variation due to heterogeneity. The adjusted R 2 value was equal to −684.96%, 15.63%, and 0.65%, with the proportion of between-study variance explained. Data in Tables 3, 4, and 5 suggest that these four clinical aspects of heterogeneity had no statistically significant effect on the final results. However, only four studies used an identical form and dosage of TCM intervention (Tongxinluo capsules), which shows that clinical heterogeneity existed and that its influence cannot be ruled out. According to statistical tests, the results shown in Figure 2 are more likely to reflect the real effects of TCM. Interpretation of the results in Figures 3 and 4 is controversial.

Safety analysis and publication bias
Ten of the 21 RCTs reported routine blood or urine examination results as well as liver and kidney function test results. No side effects were reported with the  exception of one study (Li 2009) that reported two cases of stomachache in the treatment group. Funnel plot analysis (Figures 8 and 9) showed no significant publication bias.

Findings
The results of this study suggest that women are more likely to suffer from CSX based on the fact that patients were randomly recruited. Improving patient outcomes is the primary goal of CSX management. Compared with routine Western medicine, TCM has the potential of being more effective in relieving symptoms, improving ECG results, and prolonging exercise duration in treadmill tests; however, these advantages were not obvious based on the evidence collected so far. The results in Groups A and B show that Chinese medicine with or without routine Western medicine could improve the degree of angina and reduce the frequency of angina attacks by five times per week. Besides these presenting symptoms of the patients, objective measurements such as ECG and treadmill testing also provided evidence. Because CSX and coronary heart disease share the same symptom of chest pain, the effective solution for the treatment of CSX may provide useful information in terms of symptom improvement for coronary heart disease. Studies in this field will be meaningful.
One mechanism of action of the Chinese herbal medicines prescribed to patients with CSX is believed to involve regulation of endothelial function. A number of cell factors, including ET-1, ET-21, CRP, hs-CRP, and ET-1, were measured and recorded in most of the involved studies. In addition, a comprehensive study (by one researcher, C.J.) of the pharmacological actions of all 48 types of herbs involved in TCM showed that these herbs can improve or protect endothelial function. Analysis of the forest plot also provided hints regarding this hypothesis. However, the undetected association between ET-1 levels and angina improvement in this study may have been caused by the small size of the studies.
Furthermore, we identified a total of 259 case reports and case series from our search, almost 12 times the number of RCTs conducted on this topic. Observational studies have been published since the early 1990s, whereas the first RCT was carried out in 2002. This shows TCM researchers' growing interest in CSX studies and the difficulties they faced in conducting high-quality RCTs. However, observational studies might also be valuable resources for further research.
Finally, it should be pointed out that the outcomes reported in the included studies in this review were all short-term outcomes, such as symptom relief before and after treatment. The long-term effects of TCM on patients with CSX and its role in the prognosis were scarcely discussed. This may provide a starting point for future studies.

Limitations
This research only included articles published in English and Chinese. Studies published in other languages were not considered. The sample sizes of the present studies were small, which may lead to bias. The 21 trials included in this review were of moderate to low quality. As a result, the evidence generated needs to be interpreted with caution.

Implications for further study
These clinical trials show that the basic mechanism of Chinese herbs in relieving chest pain mainly involves endothelial function and the ET-1 pathway. This should draw our attention because Chinese herbs may be helpful for developing and optimal for the treatment of CSX or even coronary heart disease. The methodology of RCTs should be modified in terms of double blinding and allocation concealment. Female patients suffering from CSX should receive more attention.

Conclusion
This review is the first to systematically evaluate the effects of TCM in the treatment of CSX, addressing the lack of this type of research. In conclusion, TCM shows potential in treating CSX, but its efficacy seems to be minor thus far. Rigorous and multicenter, large-scale clinical trials must be carried out to reveal the exact effectiveness of TCM.