Evidence and Potential Mechanisms of Traditional Chinese Medicine for the Adjuvant Treatment of Coronary Heart Disease in Patients with Diabetes Mellitus: A Systematic Review and Meta-Analysis with Trial Sequential Analysis

Traditional Chinese medicine (TCM) has long been used to treat diabetes mellitus and angina. It has also gained widespread clinical applications in China as a common adjuvant treatment. Although there is high-quality evidence that TCM is effective in regulating glucose and lipid metabolism, the cardiovascular protective effect of TCM in the treatment of diabetes mellitus has not been fully elucidated, especially in patients with both diabetes mellitus and coronary heart disease (CHD). We systematically assessed the efficacy and safety of TCM for the adjuvant treatment of patients with CHD and diabetes mellitus and examined the pharmacological effects and potential mechanisms of TCM medication/herbs on diabetes mellitus with CHD. We found that TCM could improve the control effect of conventional treatment on cardiac function, hemorheology, blood glucose, blood lipid, and inflammation, thus reducing the frequency of angina and the incidence of cardiovascular events and all-cause mortality. These findings indicate that TCM may be used as a complementary approach for patients with diabetes mellitus and CHD. Nevertheless, more rigorously designed randomized controlled trials and long-term evaluations are needed to support these findings.


Introduction
The global population of patients with diabetes mellitus (DM) is approximately 463 million, which is expected to rise to 700 million by 2040 [1]. As an independent risk factor for coronary heart disease (CHD), DM is considered as "coronary heart disease equivalent" [2], which is associated with increased CHD morbidity and mortality [3]. In patients with DM, the risk of mortality due to CHD is twice that of those who do not have DM [4]. Clinical outcomes and complications following coronary interventions are also worse in such high-risk patients [5]. Several medical resources are spent on DM every year. The total global health expenditure due to DM was estimated to be $673 billion in 2015 [6], with cardiovascular diseases attributable to diabetes accounting for the highest proportion (47.9%) of overall costs [7]. However, the need to control cardiovascular diseases is still unmet. At present, the treatment for DM, especially in those with a high atherosclerotic cardiovascular disease risk, mainly focuses on comprehensively controlling multiple risk factors [8]. However, the complexity of multidrug regimens may induce the chance of nonadherence among patients, while long-term use of high-intensity antiplatelet drugs and lipid-lowering statins may increase the incidence of side effects, such as bleeding, muscle complaints, increased liver or muscle enzymes, or various neurological symptoms [9]. Therefore, clinicians have begun considering traditional Chinese medicine (TCM) as an adjuvant or alternative treatment for DM.
TCM has been widely used to treat diabetes and angina for 1000 years. The holistic and multitarget approaches of TCM have unique advantages in controlling complex diseases, such as DM [10]. In recent years, large-scale clinical trials have demonstrated that TCM effectively reduces hyperglycemia [11][12][13] and improves vascular inflammation, lowers lipids, and improves cardiac function [14,15]. A large-sample prospective cohort study also showed that TCM alone is as efficacious as Western medicines in preventing diabetes from being complicated with coronary artery disease [16]. Moreover, patients who used more types of TCM tended to use much less Western medicine recommended by current guidelines [17]. At the same time, some therapeutic mechanisms underlying TCM's potency have been uncovered with the application of modern science and technology.
A number of clinical trials in China have shown that TCM is effective in treating DM or CHD. However, because of the insufficient sample size, short experimental period, and lack of empirical validity of efficacy evaluation in many studies, the results have been questioned and rejected. To date, there are very few systemic analyses that have focused on the comprehensive regulation effect of TCM adjuvant treatment on both DM and CHD. Considering the aforementioned limitations, we selected more credible studies for a systematic review and meta-analysis to comprehen-sively evaluate the clinical efficacy and safety of TCM for patients with CHD and DM.

Materials and Methods
We conducted and reported this systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [18].
2.1. Search Strategy. The first three authors independently searched for papers published from January 2005 to May 2021 on the following databases: the China National Knowledge Internet, Wanfang, VIP databases, the China Biology Medicine, Web of Science, Pubmed, Medline; Embase, International Pharmaceutical Abstracts, and the Cochrane Library. Search terms were as follows: ("coronary heart disease" or "cardiovascular disease" or "angina" or "myocardial ischemia" or "myocardial infarction" or "acute coronary syndrome" or "coronary atherosclerosis") and ("diabetes" or "diabetes mellitus" or "Xiaoke syndrome") and ("randomized controlled trial" or "controlled clinical trial" or "random" or "randomly" or "randomized" or "control" or "RCT") and ("TCM" or "traditional Chinese medicine" or "Chinese medicinal herb" or "Chinese herbal medicine" or "decoction" or "formula" or "prescription" or "Chinese patent medicine" or "Chinese patent drug" or "Chinese herbal compound prescription"). The PRISMA flow chart for the study is presented in Figure 1.   [19]. The risk of bias was assigned as low, unclear, or high for the following items: random sequence generation,         Journal of Diabetes Research the third author (Yeung) was consulted. Relevant reports' information on the author, year, number of participants, intervention (weeks, frequency, and dosage), and passive control condition were extracted and recorded in an excel spreadsheet. The authors of the identified papers were contacted for additional information if necessary. The primary outcomes were all-cause mortality and cardiovascular (CV) events. The frequency of angina contributed to efficacy assessment and was included as a secondary outcome. Hemorheology indices and cardiac function were also included as secondary outcomes, as abnormalities of these laboratory measurements are associated with CV events. The secondary outcomes also included the levels of fasting blood glucose, postprandial blood glucose (PBG), glycosylated hemoglobin (HbA1c), total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), C-reaction protein (CRP), interleukin-6 (IL-6), and tumor necrosis factor α (TNF-α). RevMan 5.3 (Nordic Cochrane Center, Copenhagen, Denmark) and Stata version 15.0 (Stata Corp., College Station, TX, USA) were used to analyze the data. Continuous outcomes were pooled to find the weighted mean differences (WMDs) accompanied by 95% confidence intervals (CIs), when the outcomes were based on the same scale; otherwise, the standard mean differences (SMDs) were used. Categorical outcomes were pooled to find relative risks (RRs) and were accompanied by 95% CIs. I 2 statistics were used to measure heterogeneity. A fixed-effect model was used if I 2 < 50%; otherwise, the random-effect model was used. Publication bias was explored through a funnelplot analysis.   Journal of Diabetes Research 2.5. Trial Sequential Analysis. In this systematic review and meta-analysis, the curative effect was exaggerated because of random errors, especially when the number of included trials or the total sample size was too small. Thus, the Trial Sequential Analysis (TSA) software version 0.9 Beta (Copenhagen Trial Unit, Copenhagen, Denmark) was used to conduct TSA analysis on the primary outcomes that could be combined and to estimate the meta-analysis sample size and the value strength of the outcome effect. In this study, an overall 5% risk of a type I error was maintained with a power of 80% [20]; the predicted risk reduction was set as 20%, and the relative event rate of the control group was derived from the meta-analysis data.

Characteristics of the Included Trials.
A total of 21 RCTs  were included in this systematic review. The study by Zhang [34] showed that the HDL levels decreased after treatment, which was contrary to the description of elevated HDL levels. Therefore, this study was excluded from our analysis. Therefore, 20 RCTs were included. One was in English, and the others were in Chinese. The publication years of these studies ranged from 2007 to 2021. The main characteristics of the individual trials are summarized in Table 1. A total of 3565 patients were included; the number of participants in each study ranged from 100 to 591 patients. Only one trial [23] did not indicate the exact num-ber of female and male patients. The overall male to female ratio was approximately 4 : 3. The study period ranged from 8 weeks to 4 years. In all 20 studies, the diagnostic criteria for both DM and CHD were specified. Three studies focused on patients with stable angina pectoris, unstable angina pectoris, or acute coronary syndrome. Chinese medicines used in these studies included both herbal medicines and Chinese patent medicines. The intervention measures in the control group mainly included conventional Western medicines in line with clinical practice and lifestyle modifications, such as exercise and dietary control. Only one study reported that the placebo was delivered to participants in the control group. All 20 studies evaluated multiple outcomes at the end of treatment. Three, two, and the remaining 15 studies reported cardiovascular events, all-cause mortality, and at least two secondary outcome measures, respectively. The specific composition of TCM in the studies is listed in Table 2. Moreover, 12 trials reported adverse events.

Methodological
Quality. All 20 trials claimed that the groups were randomized; nine trials [22-24, 26-28, 30, 32, 35] clearly stated the aforementioned using a random number table, while the others did not specify the method used for random sequence generation. No trials described allocation concealment. Only one trial explicitly mentioned blinding the participants and personnel [36]. The blinding of outcome assessment was not specified in any of the trials. Five studies reported cases of loss to follow-up [23, 28, 33,

Primary Outcomes.
Three studies reported mortality and cardiovascular events. One study found no death events because of the short follow-up period. Compared to patients in the control group, the beneficial effect of TCM participation on mortality was observed (RR, 0.50; 95% CI, 0.32, 0.78; and P = 0:002) with low heterogeneity (I 2 = 47%, P = 0:17; Figure 3). However, the TSA result indicated a possibility of a false positive conclusion, and more trials should be included to confirm the efficacy (Figure 4). The results were more reliable when it came to reducing cardiovascular events, showing that TCM adjuvant therapy could effectively reduce the incidence of cardiovascular events (RR, 0.47; 95% CI, 0.34, 0.65; and P < 0:001) without heterogeneity (I 2 = 0%; P = 0:82; Figure 5). The TSA results also supported this conclusion, as shown in Figure 6. The pooled results (Z curve, blue lines) crossed both the conventional boundary of benefit (dotted line) and the TSA boundary value curve (full line), indicating that a positive conclusion was reached even though the required information size (n = 3102) had not yet been reached ( Figure 6).

Frequency of Angina Pectoris.
Three studies reported data on the change in angina frequency with large heterogeneity. Considering that heterogeneity may be caused by different prescriptions, the pooled effect size analysis was performed according to Ning et al. [29] and Shi et al. [40], with a significant benefit of TCM (MD, -0.88; 95% CI, -1.13, -0.63; and P < 0:001), both of which used the Yindan Xinnaotong soft capsules (Figure 7).

Cardiac Function.
Three studies showed an improvement of cardiac function with TCM. Compared to the control group, TCM decreased the left ventricular end-systolic

Adverse Events.
Among the 12 studies that mentioned adverse events, four declared that no significant adverse   [36]. Other adverse events included fever [31], mild headache [25], edema, and itchy skin [21,30,31]. There was no statistically significant difference in the incidence of adverse events between the experimental and control groups.
3.6. Publication Bias. Funnel plots were drawn in cases where there were >10 articles to examine publication bias regarding the secondary outcomes. Because of the large heterogeneity of the studies, which was difficult to judge by funnel plots alone, Egger's test was performed.
More than 10 studies reported information concerning the TC, TG, LDL, HDL, and 2hPG levels, of which all showed P values > 0.05 by Egger's test, indicating no publication bias in the included studies. The funnel plot and Edger's test results for TC, TG, LDL, HDL, and 2hPG are presented in Supplementary materials (available here).

Discussion
Diabetes combined with CHD increases the incidence of and mortality due to CV events. This review revealed that the use of TCM in addition to conventional treatment showed potential benefits in the treatment of DM with multiple types of CHD. Compared to the existing integrated management of glucose and cardiovascular risk factors, the addition of many TCM prescriptions formed by the delicate compatibility of herbal ingredients showed better efficacy in reducing cardiovascular events. The Standards of Medical Care for Type 2 Diabetes in China 2019 included a dedicated section focusing on TCM treatment [42].   64  60  65  50  50  54  81  75  75  60  51  60  67   64  60  50  50  54  81  75  60  75  60  51  60  67   64  60  54  81  75  75  60  51  60  67   64  60  54  81  75  75     12 Journal of Diabetes Research The use of TSA analysis in this paper overcomes the shortcomings of a traditional systematic review or meta-analysis by avoiding hasty conclusions and ignoring the accumulation of random errors. However, it also has certain limitations: it cannot compensate for the defects in the methodological quality of the trial or the errors caused by outcome reporting bias [43].  64  60  65  50  50  54  81  75  75  60  51  60  67   64  60  50  50  54  81  75  60  75  60  51  60  67   64  60  54  81  75  75  60  51  60  67   64  60  54  81  75  75  60  51  60 Figure 12: Forest plot of blood glucose levels.

Journal of Diabetes Research
Articles have compared the efficacy of TCM in the treatment of diabetes and CHD [14], but the outcomes were clinical effectiveness rates, which varied greatly in different studies with no unified standard.
However, combined with the results of previous studies, the use of TCM for the treatment of patients with DM and CHD should be explored further to promote its application.
TCM is characterized by multitargeted effects. The study herein showed beneficial ameliorative effects in the risk factors (blood glucose, lipid, cardiac function, hemorheology, and inflammatory factors) of DM combined with CHD. At present, many Chinese medicines are used to treat cases of DM combined with CHD, but the related clinical studies are insufficient, and the underlying mechanism requires further research. High-quality, randomized, double-blind clinical trials are needed to understand the associated mechanisms.

Limitations.
Most of the clinical trials included in this study were high-risk trials. Although all trials claimed to use randomization, only nine described specific randomization methods. Improper randomization often leads to selection bias. At the same time, only one out of 20 studies used a double-blind design, indicating the possibility of potential errors. As the allocation is not concealed, the effect of the intervention may be exaggerated by 30-40% [44]. The reason may be that it is difficult to make a convincing placebo of TCM. This is also a prominent defect of most clinical trials of TCM.
Sensitivity analysis revealed that three studies [25,29,36] had a large effect on the heterogeneity of different outcome measures. The work by Zhao et al. [36] was a randomized double-blind trial with intervention for 4 years, which was much longer than the corresponding period in other clinical trials. Interestingly, they reported a clear efficacy on regulating blood lipid indicators [45]. The study suggested that long-term TCM administration may modulate blood lipids better than conventional treatment. The patients included in the study by Li et al. [25] were aged 65-85 (mean age, 74.54) years, which was higher than the corresponding mean ages in other trials. The poor glycemic control, which was possibly attributed to islet cell function, was gradually lost in elderly patients with DM. The patients in the study by Ning et al. [29] received nitrate ester, which can improve hemodynamics and can explain the better control in FIB than other study.
The main sources of heterogeneity were the following: (1) Study protocol designs were not rigorous. (2) The interventions differed greatly in composition, dosage, dosage form, and intervention time. In the theory of TCM, individuals adapted in different formulas for treating DM and CHD are different. Meanwhile, the different dosage forms of TCMs also indicate problems in the quality control of TCMs. (3) Basal interventions differed.
Considering the complicated sources of heterogeneity, we found it difficult to reduce the high heterogeneity through subgroup analysis.
On the one hand, different TCM interventions are main sources of heterogeneity. On the other hand, there are differences in the type of heterogeneity. In addition, although patient baselines were comparable in independent studies, patient conditions, including the severity of CHD, varied between studies.

Conclusions
Through this systematic review and meta-analysis of adjuvant TCM treatment of CHD in patients with DM, we found that TCM could improve the control effect of conventional treatment on blood glucose, blood lipid, and inflammation, thereby reducing the frequency of angina and the incidence of cardiovascular events and all-cause mortality; this indicates that TCM may be used as a complementary approach to the tertiary prevention of DM. Nevertheless, more rigorously designed RCTs and long-term evaluations are needed to support these findings. The relationship between cardiovascular benefits and TCM control of inflammation, cardiac function, and hemodynamics should be further examined.

Data Availability
The data used to support the findings of this study are included within the article.