Evaluation of the Add-On Effect of Chinese Patent Medicine for Patients with Stable or Unstable Angina: A Systematic Review and Meta-Analysis

Chinese herbal medicine (CHM) has been widely used as an adjunct to western medicine in treating angina in China. We carried out this systematic review to evaluate the effectiveness of CHM on top of western medicine for angina. This meta-analysis included 46 randomized control trials with 4212 patients. For trials that included stable angina patients, the CHM group had significant lower incidence of total heart events (relative risk (RR) = 0.50, 95% confidence interval (CI) 0.33–0.78), myocardial infarction (RR = 0.32, 95% CI 0.14–0.72), heart failure (RR = 0.37, 95% CI 0.15–0.91), and angina (RR = 0.46, 95% CI 0.30–0.71) than that of control group. For trials that included unstable angina patients, CHM led to significantly lower occurrence of total heart events (RR = 0.46, 95% CI 0.32–0.66), myocardial infarction (RR = 0.37, 95% CI 0.26–0.54), and angina (RR = 0.36, 95%CI 0.26–0.51). Likewise, for trials that included stable or unstable angina patients, the rates of myocardial infarction (RR = 0.34, 95% CI 0.17–0.68) and angina (RR = 0.46, 95% CI 0.30–0.70) in CHM group were significantly lower than that in control group. In conclusion, CHM is very likely to be able to improve the survival of angina patients who are already receiving western medicine.


Introduction
Angina is pain or constricting discomfort that typically occurs in the front of the chest and is brought on by physical exertion or emotional stress [1]. It is the main symptomatic manifestation of myocardial ischemia, caused by an imbalance between myocardial blood supply and oxygen demand [1,2]. The prevalence of angina in the population appears to increase in the past decades [3]. A meta-analysis of data from 31 countries indicated the population weighted prevalence was 6.7% in woman and 5.7% in man [4]. Angina is a common initial presentation of coronary disease [5], and it may exert a major impact on quality of life, ability to work, and costs to society [6,7].
Angina is clinically classified into stable angina (SA) and unstable angina (UA), and treatment strategy is different between them. SA is a chronic medical condition and the aim of management for it is to abolish or minimize symptoms, improve quality of life, and decrease long-term morbidity and mortality [1] while UA is an acute coronary syndrome, which should be treated as an emergency [8]. The current antiangina medical management includes pharmacological strategies, revascularization strategies, and lifestyle interventions. Chinese herbal medicine (CHM) is also widely used for treating angina in China [9].
Because angina is a life-threatening event and very effective western medicine treatments are available, CHM is usually used in addition to baseline treatment with western 2 Evidence-Based Complementary and Alternative Medicine medicine. Common CHM prescribed for treating angina includes Tongxinluo capsules, Fufangdanshen dripping pills, Shengmai capsules, and Shexiangbaoxin tablets. Tongxinluo capsules mainly take effect by dilating coronary arteries, increasing blood vessel perfusion flow, and strengthening cardiac contractility [10,11]. Fufangdanshen dripping pills may alleviate angina via the antimyocardial ischemia and antiatherosclerotic effect [12,13]. Likewise, not only can Shengmai capsules raise coronary blood flow, it may also improve the tissue tolerance to oxygen privation of cardiac muscle [14]. Apart from expanding blood vessel and increasing crown arteries current capacity, Shexiangbaoxin tablets may also alleviate arteriosclerosis and steady plague [15]. However, there is still a knowledge gap to clearly establish evidence that CHM is effective in improving the outcomes of angina patients. We therefore performed this systematic review of randomized controlled trials (RCT) to evaluate the add-on effect of CHM in angina patients.

Materials and Methods
This study was undertaken according to the recommendation of Cochrane handbook for systematic reviews of interventions [16] and reported according to the PRISMA statement [17].

Inclusion Criteria.
Studies are eligible if (1) RCTs which compared CHM + western medicine versus western medicine, or compared CHM versus no treatment/placebo; (2) the participants were patients with stable or unstable angina (diagnosed by typical angina chest pain, and ischemic ST-segment depression by electrocardiography); (3) the intervention may be any preparations containing at least one herb that is included in the latest version of the Chinese Pharmacopeia; (4) the follow-up time should be at least 7 days. The primary outcome of this study is mortality (death from myocardial infarction (MI) and other causes). The secondary outcomes include recurrent MI, heart failure, quality of life, use of revascularization, deterioration or improvement in symptoms of angina, and adverse events. Trials were eligible if one of the hard outcomes (mortality, recurrent MI, or heart failure) was reported.  Figure 1). There was no limitation on language or publication status. The search strategy included the following key words: "Chinese herbal, " "traditional Chinese medicine, " "herb, " "angina, " "stenocardia, " "clinical trials, " and "randomized controlled trial. " The reference lists of relevant trials and review articles, abstracts from major relevant conferences, and relevant trials registers were checked for additional studies.

Selection of Studies.
Study eligibility was independently determined by two authors. All the citations were inputted into reference management software Endnote, and the duplicates were removed. The authors then evaluated the eligibility of remaining studies by examining the titles, abstracts, and full articles progressively. Discrepancies were resolved by discussion.

Data Extraction and Quality
Assessment. Data were extracted independently by two authors using a standard form. Data extracted include (1) general information (e.g., title, authors, reference, language, year of publication, and setting); (2) trial characteristics related to methodological quality (e.g., design, duration of followup, sequence generation, allocation sequence concealment, and blinding); (3) intervention and comparison (dose, route, and timing); (4) patients (e.g., baseline characteristics and diagnostic criteria); (5) outcomes (e.g., estimates, standard error, and P value). Discrepancies were resolved by discussion. The authors of original studies were consulted for missing information where necessary.
The methodological quality of included randomized trials was assessed and reported by the Cochrane Collaboration's tool to assess the risk of bias [16]. The methodological quality assessed included: (1) sequence generation, (2) allocation sequence concealment, (3) blinding, (4) incomplete outcome data, (5) selective outcome reporting, and (6) other potential sources of bias.

Data
Analysis. All analyses were conducted using the Review Manager software compiled by the Cochrane Collaboration [16]. Dichotomous outcomes were expressed as relative risk (RR). 95% confidence intervals (CIs) were calculated for all estimates. Tests for heterogeneity were performed with chi-squared test at a significance level of = 0.1. 2 statistic was calculated to estimate variation across studies. We regarded 2 < 25% as an indicator of low heterogeneity level, 25-50% as moderate level, and >50% as high level. The estimates were pooled with a fixed-effect model if there was no significant heterogeneity. Whenever a significant heterogeneity presented, a random-effect model was used to pool the results. We intended to explore the potential sources of heterogeneity by subgroup analysis and metaregression if the number of trials was sufficient. We assessed the publication bias by funnel plot, and adjustment was made to reduce the effect of publication bias in the estimate of effectiveness. Figure 1 showed the study selection in this study. Our search in bibliographic databases yielded 15866 citations, of which 2660 were classified as potentially relevant and were subjected to full text assessment. Finally, a total of 46 studies were included , of which 9 studies included patients with SA [18][19][20][21][22][23][24][25][26], 31 studies included patients with UA , and 6 studies included patients with SA or UA [58][59][60][61][62][63]. This systematic review totally included 4212 patient, with 2141 patients receiving the combination of CHM and western  medicine and 2071 patients receiving western medicine alone. The duration of treatment and followup ranged from 1 day to 48 days, and 7 months to 36 months, respectively. Table 1 demonstrated the characteristics of included studies.

Risk of Bias.
Among included studies, only 6 trials concealed allocation sequence, two studies did not conceal the allocation sequence generated, and the rest studies were unclear. As to randomization, two studies had high risk of bias, and the rest studies had either low or uncertain risk of bias. All the included studies had low risk of bias except one study that had high risk of bias for selective outcome reporting. The details are shown in Table 2 and

Add-On Effect of Chinese Herbal Medicine in Patients with
SA. For trials that only included SA patients, we analyzed the following outcomes: total heart events (3 trials), MI (7 trials), cardiac arrhythmia (2 trials), heart failure (3 trials), angina (6 trials), and death (2 trials). All pooled results showed homogeneity ( > 0.1). Though the SA patients in CHM group had a lower death rate (1.74%) than SA patients in control group (5.22%), the difference was not statistically significant ( = 0.190). Twenty-three out of 150 SA patients who were treated with the combination of CHM with western medicine got total heart events, with an incidence rate of 15.33%, which was significantly lower (RR = 0.50, 95% CI 0.33-0.78; = 0.002) than that of SA patients who were treated with western medicine alone (30.61%). Compared with western medicine alone, the combination of CHM with western medicine significantly reduced the occurrence of myocardial infarction, from 6.67% to 1.85, with the pooled RR equal to 0.32 (95% CI 0.14-0.72, = 0.006). The incidences of heart failure (RR = 0.37, 95% CI 0.15-0.91; = 0.031), cardiac arrhythmia (RR = 0.27, 95% CI 0.13-0.57; = 0.001), and angina (RR = 0.46, 95% CI 0.30-0.71; < 0.001) were also significantly lower in SA patients who were treated with the combination of CHM with western medicine than that of patients treated with western medicine alone. There were only one study that explored the difference of difference on the outcomes of fatal events, sudden cardiac death, and adverse 4 Evidence-Based Complementary and Alternative Medicine            events such as bleeding and stomach discomfort between patients in CHM group and patients in control group, none of them were different between two groups. Visual inspection of the funnel plots revealed no asymmetry except the metaanalysis of the occurrence of angina. Details of statistical results can be found in Table 3 and Figures 3, 4, 5, 6, and 7.

Add-On Effect of Chinese Herbal Medicine in Patients with UA.
For trials that only included UA patients, we analyzed the following outcomes: total heart events (7 trials), death (1 trial), MI (19 trials), cardiac arrhythmia (1 trial), heart failure (2 trials), angina (12 trials), need for cardiac surgery (2 trials), and need for PCI (3 trials). All pooled results showed homogeneity. There are 337 patients in CHM group, of whom 36 patients developed total heart events (10.68%), while 78 out of 333 patients (23.42%) in the control group occurred total heart events. Patients in CHM group had a significant lower incidence of total heart events than patients in the control group, with a pooled RR equal to 0.46 (95% CI 0.32-0.66, < 0.001). Compared with western medicine alone, the combination of CHM with western medicine significantly reduced the occurrence of myocardial infarction, from 12.42% in control group to 4.26% in CHM group, with a pooled RR of 0. 37      Two studies presented the data on the adverse events on the treatment, and there is no significant difference on bleeding and stomach discomfort between groups. Visual inspection of funnel plots suggesting there's no publication bias. Details of statistical results can be found in Table 3 and Figures 8, 9, 10, 11 and 12.   CHM with no CHM treatment in patients with either SA or UA showed that the myocardial infarction rate was 3.13% (10/320) and 8.91% (31/348) in CHM treatment group and control group, respectively, with a pooled RR of 0.34 (95% CI 0.17-0.68; = 0.002). There was enough evidence to illustrate that patients with CHM treatment had a significant lower myocardial infarction rate than the control group. No evidence of statistical heterogeneity across the studies was identified using the 2 test ( 2 = 0, = 0.984 > 0.1). Metaanalysis showed that patients with CHM treatment had a significant lower angina rate than no CHM treatment group, with a pooled RR of 0.46 (95% CI 0.30-0.70; = 0.000). There was no evidence of statistical heterogeneity ( 2 = 0, = 0.874 > 0.1). Visual inspection of funnel plots suggesting no publication bias. Details of statistical results can be found in Table 3 and Figures 13, 14, 15 and 16.

Discussion
This systematic review synthesized evidence from 4212 patients in 46 RCTs. The main findings of this systematic review are as follows. (1) The combination of CHM with western medicine may significantly reduce the occurrence of total heart events, MI, cardiac arrhythmia, heart failure, and angina compared with western medicine alone in patients with SA. (2) For patients with UA, the combination therapy is superior to western medicine alone on total heart events, myocardial infarction, angina, and sudden cardiac death. But there is no significant difference on the incidence of death, cardiac arrhythmia, hart failure, fatal events, cardiac surgery, and percutaneous coronary intervention. (3) For patients with SA or UA, the combination therapy may lead to significant lower occurrence of myocardial infarction, heart failure, and angina as compared with western medicine monotherapy, but no significant differences were indentified on the incidence of total heart events, death, cardiac arrhythmia, fatal events, sudden cardiac death, cardiac operation, bleeding, and stomach discomfort.
The overall risk of bias of included studies was moderate. Inadequacy in reporting methods for randomization and allocation concealment is a major problem among most included RCTs, which may lead to selection bias in our study [16]. The lack of blinding is another problem in most included trials; however, the impact on conclusion was less critical as we focused on objective outcomes. Visual inspection of funnel plots indicated no symmetry on most outcomes, except for the meta-analysis of angina in patients with SA.  We therefore believe publication bias is unlikely to be a serious threat to the estimates.
Apart from the limitations on the quality of included studies, the estimates of some outcomes, such as the incidence of fatal events, sudden cardiac death, and bleeding in patients with UA are also limited by relatively small sample size. For these outcomes, only one study with less than 100 patients was included in meta-analysis, and the precision of estimates was therefore influenced [75].
To the best of our knowledge, this is the most comprehensive assessment of add-on effect of CHM in patients with angina. We employed a contemporaneous search strategy in both international and Chinese databases to ensure most studies were included. This enabled us to locate a much higher number of studies compared to other existing reviews on this topic. Additionally, the study selection, data extraction, and quality assessment in this study were independently carried out by two authors to ensure high validity.  This study has important implication for clinical practice and CHM research. For practitioners, this systematic review demonstrated consistent, add-on benefits of using CHM on top in western medicine treatment for preventing all-cause and cardiac mortality amongst angina patients. However, this conclusion was drawn from moderate or low quality trials and the estimates on some outcomes were based on limited number of patients. Large scale, rigorously designed RCTs are still wanted to confirm these conclusions.