Xiongshao for restenosis after percutaneous coronary intervention in patients with coronary heart disease

Abstract

Background

Percutaneous coronary intervention (PCI) with stent placement is a standard treatment for coronary heart disease (CHD). In‐stent restenosis after PCI remains an important clinical problem. Xiongshao capsule has been reported to be beneficial in preventing restenosis after PCI in CHD patients. However, the strength of evidence to support its use is unclear.

Objectives

To systematically assess the efficacy and safety of Chinese herbal medicine Xiongshao capsule in preventing restenosis after PCI in patients with CHD.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library (Issue 3 of 2012), MEDLINE (OVID) (1948 to week 1 March 2012), EMBASE (OVID) (1980 to week 10 2012), ISI Web of Science with Conference Proceedings (1970 to 14 March 2012), LILACS (1982 to 15 March 2012), Chinese biomedical literature database (1980 to May 2012), China National Knowledge Infrastructure (1994 to May 2012), Chinese Medical Current Contents (1994 to May 2012), VIP Database for Chinese Technical Periodicals (1989 to May 2012), Chinese Master's Theses Full‐Text Database (1994 to May 2012), China Doctor Dissertation Full‐Text Database (1994 to May 2012), and China Proceedings of Conference Full‐Text Database (1994 to May 2012). We also searched ongoing trials and research registries.

Selection criteria

All randomised controlled trials comparing Xiongshao capsule alone/plus conventional western medicine with the same conventional western medicine alone/plus placebo in participants with CHD who met the recognised diagnostic criteria and had successfully undergone a PCI procedure were included.

Data collection and analysis

Two review authors independently selected trials and extracted data. Two review authors independently assessed the risk of bias of included trials using The Cochrane Collaboration tool, and any disagreements were resolved by discussion with a third review author. Data were pooled for meta‐analysis using the fixed‐effect model, and the results were expressed as risk ratios (RRs) with 95% confidence intervals (CIs).

Main results

Four trials involving 649 participants were included in this review. Two of these trials (459 participants) were designed as randomised, double‐blind, placebo‐controlled trials with an adequate methodological description; the other two trials (190 participants) described an inadequate methodological design. All four trials with 649 participants were included in the meta‐analysis. Significant differences were noted in rates of restenosis, recurrence angina pectoris, and serious cardiovascular adverse events between Xiongshao capsule plus conventional western medicine and the same conventional western medicine alone; RR values (95% CIs) were 0.41 (0.22 to 0.75), 0.47 (0.31 to 0.72), and 0.47 (0.25 to 0.90), respectively. Xiongshao capsule plus conventional western medicine showed more significant reductions in restenosis (RR 0.52, 95% CI 0.33 to 0.80), recurrence angina pectoris (RR 0.26, 95% CI 0.18 to 0.38), and serious cardiovascular adverse events (RR 0.45, 95% CI 0.28 to 0.70) than the same conventional western medicine plus placebo. Safety outcomes and adverse events of the Xiongshao capsule were reported in two trials, which reported no adverse events.

Authors' conclusions

The summary estimates indicate a protective effect of Xiongshao on restenosis and suggest that Xiongshao capsule may be used to prevent restenosis after a PCI procedure in CHD patients. However, this evidence is derived from small randomised trials, all conducted in China, and two of the included trials showed important methodological limitations that undermine the validity of the findings. Additional high‐quality research trials with sufficient sample size are required.

Plain language summary

Xiongshao capsule preventing restenosis after percutaneous coronary intervention in patients with coronary heart disease

Coronary heart disease (CHD) is the leading cause of death and disability in the world. Percutaneous coronary intervention (PCI) with stent placement is the standard nonsurgical treatment for CHD. However, its main limitation is the repeated blocking of blood vessels following a stent procedure. Data from experimental and human studies have suggested that Xiongshao capsule, a Chinese herbal medicine product, may be beneficial in preventing restenosis of blood vessels after a PCI procedure. We systematically reviewed currently available evidence with the goal of performing a reliable assessment of the effects and safety of Xiongshao in repeated blocking of blood vessels after PCI.

We identified four trials that tested treatments based on Xiongshao capsule. Two trials compared Xiongshao capsule plus conventional western medicine with the same conventional western medicine plus placebo; the other two trials compared Xiongshao capsule plus conventional western medicine with the same conventional western medicine alone for preventing restenosis after a PCI procedure in CHD patients. These trials reported that use of Xiongshao capsule caused a significant reduction in the incidence rates of restenosis, recurrence angina pectoris, and serious cardiovascular adverse events with no substantive adverse effects following a PCI procedure. Therefore Xiongshao capsule may represent a choice for the prevention of in‐stent restenosis following a PCI procedure. However, although summary estimates indicate a protective effect of Xiongshao on restenosis, evidence is derived in part from small randomised trials, all conducted in China, with some methodological limitations that undermine the validity of the findings. Additional high‐quality research trials with sufficient sample size are required.

Summary of findings

Summary of findings for the main comparison. Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) for restenosis after percutaneous coronary intervention in patients with coronary heart disease.

Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) for restenosis after percutaneous coronary intervention in patients with coronary heart disease
Patient or population: patients with restenosis after percutaneous coronary intervention in coronary heart disease Settings: hospitals Intervention: Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM)
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No of Participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM)
All‐cause death: not measured	See comment	See comment	Not estimable	‐	See comment	All‐cause death not measured
Cardiac mortality Follow‐up: 6‐24 months	See comment	See comment	Not estimable	190 (2 studies)	⊕⊕⊕⊝ moderate1	No cardiac death occurred
Serious adverse events Follow‐up: mean 6 months	Study population		RR 0.47 (0.25 to 0.9)	108 (1 study)	⊕⊕⊕⊕ high2,3
	400 per 1000	188 per 1000 (100 to 360)
	Moderate
	400 per 1000	188 per 1000 (100 to 360)
Restenosis Coronary angiography Follow‐up: mean 6 months	Study population		RR 0.41 (0.22 to 0.75)	190 (2 studies)	⊕⊕⊕⊕ high4,5
	312 per 1000	128 per 1000 (69 to 234)
	Moderate
	313 per 1000	128 per 1000 (69 to 235)
Recurrence angina pectoris Treadmill exercise test Follow‐up: 6‐24 months	Study population		RR 0.47 (0.3 to 0.72)	190 (2 studies)	⊕⊕⊕⊝ moderate6,7
	479 per 1000	225 per 1000 (144 to 345)
	Moderate
	477 per 1000	224 per 1000 (143 to 343)
Health‐related quality of life: not measured	See comment	See comment	Not estimable	‐	See comment	Health‐related quality of life not measured
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (with its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: risk ratio.
GRADE Working Group grades of evidence: High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

¹Both trials did not describe method of concealment of allocation. One trial did not report method of randomisation.
 ²Concealment of allocation not reported.
 ³RR 0.47, 95% CI 0.25 to 0.90.
 ⁴Both trials did not describe method of concealment of allocation. One trial did not report method of randomisation.
 ⁵RR 0.41, 95% CI 0.22 to 0.75.
 ⁶Both trials did not describe method of concealment of allocation. One trial did not report method of randomisation. Furthermore, blinding was not performed.
 ⁷RR 0.47, 95% CI 0.3 to 0.72.

Summary of findings 2. Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo for restenosis after percutaneous coronary intervention in patients with coronary heart disease.

Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo for restenosis after percutaneous coronary intervention in patients with coronary heart disease
Patient or population: patients with restenosis after percutaneous coronary intervention in coronary heart disease Settings: hospitals Intervention: Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No of Participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo
All‐cause death: not measured	See comment	See comment	Not estimable	‐	See comment	All‐cause death not measured
Cardiac mortality Follow‐up: 6‐12 months	See comment	See comment	Not estimable	432 (2 studies)	⊕⊕⊕⊕ high	Cardiac death not occurred
Serious adverse events Follow‐up: 6‐12 months	Study population		RR 0.45 (0.28 to 0.7)	432 (2 studies)	⊕⊕⊕⊕ high1
	237 per 1000	107 per 1000 (66 to 166)
	Moderate
	249 per 1000	112 per 1000 (70 to 174)
Restenosis Coronary angiography Follow‐up: mean 6 months	Study population		RR 0.52 (0.33 to 0.8)	432 (2 studies)	⊕⊕⊕⊕ high
	233 per 1000	121 per 1000 (77 to 186)
	Moderate
	246 per 1000	128 per 1000 (81 to 197)
Recurrence angina pectoris with 1 month follow‐up Treadmill exercise test Follow‐up: mean 30 days	Study population		RR 0.31 (0.1 to 0.92)	314 (1 study)	⊕⊕⊕⊕ high2
	83 per 1000	26 per 1000 (8 to 76)
	Moderate
	83 per 1000	26 per 1000 (8 to 76)
Recurrence angina pectoris with 3 months' follow‐up Treadmill exercise test Follow‐up: mean 3 months	Study population		RR 0.37 (0.19 to 0.71)	314 (1 study)	⊕⊕⊕⊕ high3
	191 per 1000	71 per 1000 (36 to 136)
	Moderate
	191 per 1000	71 per 1000 (36 to 136)
Recurrence angina pectoris with 6 months' follow‐up Treadmill exercise test Follow‐up: mean 6 months	Study population		RR 0.26 (0.18 to 0.38)	432 (2 studies)	⊕⊕⊕⊕ high4
	460 per 1000	120 per 1000 (83 to 175)
	Moderate
	495 per 1000	129 per 1000 (89 to 188)
Health‐related quality of life: not measured	See comment	See comment	Not estimable	‐	See comment	Health‐related quality of life not measured
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (with its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: risk ratio.
GRADE Working Group grades of evidence: High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

¹RR 0.45, 95% CI 0.28 to 0.70.
 ²RR 0.31, 95% CI 0.10 to 0.92.
 ³RR 0.37, 95% CI 0.19 to 0.71.
 ⁴RR 0.26, 95% CI 0.18 to 0.38.

Background

Description of the condition

In developed countries, cardiovascular disease is and will remain one of the primary causes of death and disability (Andrew 2009). Cardiovascular disease is also expected to become the main cause of death and disability in developing countries (Reddy 2004). Coronary heart disease (CHD), the main cardiovascular disease, is responsible for about half of cardiovascular deaths. For example, CHD was responsible for one of five deaths in the United States in 2003, and it is the single largest killer of Americans. In fact, about 40% of people who experience a coronary attack in a given year will die from it (Thom 2006). As one of the major treatments for CHD, percutaneous coronary intervention (PCI) is currently the foremost modality, and it plays an important role in relief of symptoms of coronary ischemia (Grech 2003). The main limitation of PCI is in‐stent restenosis (Ma 2010).

Restenosis is the healing response of the arterial wall to mechanical injury, which can be defined clinically or angiographically. Clinically it is defined as the presentation of recurrent angina or objective evidence of myocardial ischemia, whereas angiographic restenosis is the presence of > 50% diameter stenosis in the stented segment (Mitra 2006; Teirstenin 1997). The diagnosis of restenosis is usually based on angiographic evidence in a clinical setting. Patients who have undergone balloon angioplasty have been reported to have a 20% to 50% incidence rate of restenosis, and most require revascularization within six months (Grech 2003; Serruys 1994). The risk of restenosis has a negative impact on the long‐term survival of patients treated with this procedure (Serruys 1994). When coronary artery stenting was introduced in CHD, a significant improvement on the reduction of restenosis rates was noted compared with angioplasty alone (Mitra 2006). With the development and universal application of drug‐eluting stents (DESs), the recurrence rate of restenosis has been reduced to less than 10% (Perkins 2010; Stone 2004). However, the DES has been shown to lead to adverse side effects such as a higher rate of stent thrombosis and acute or chronic vascular occlusion (Aziz 2007; Lagerqvist 2007; Moses 2003). According to Liistro's report, late (more than six months) total occlusion after paclitaxel eluting stent implantation occurs in approximately 4% of patients (Liistro 2001). Furthermore, after completed elution of active agents from the DES, the risk of restenosis rapidly increases. In addition, many studies have shown that DES does not reduce late cardiac‐related death and myocardial infarction (MI) (Camenzind 2006; Kim 2008; Pfisterer 2006). Although dual antiplatelet therapy after PCI is effective in reducing cardiac‐related death, MI, and stroke (Ten 2001), premature discontinuation may result in insufficient antiplatelet coverage, which greatly increases risks of stent thrombosis, MI, and death (Cindy 2007). So treatment for restenosis after PCI remains a challenging clinical issue.

Description of the intervention

Chinese herbal medicines (CHMs) include extracts from mixtures of herbs, single herbs, compound Chinese herbs, Chinese patent medicine, and Chinese medicine formulae (Liu 2010). CHMs use a variety of herbs in different combinations to restore a balance of body, energy, and spirit to treat disease. Their pharmacological actions are different from those of pharmaceutical chemicals (western medicine). Many pharmacological and clinical studies have indicated that CHMs can be used for dilating coronary vessels, improving blood circulation, preventing inflammation and immune reactions and platelet aggregation, reducing peroxidative injury, and inhibiting proliferation and migration of vascular smooth muscle cells (Li 2008b; Men 2008; Tang 2001; Wang 2010). As a method of supplemental therapy, CHMs have been tested in clinical trials in China for prevention and treatment of in‐stent restenosis after PCI (Gong 2003; Liang 2010; Ren 2010).

Xiongshao is composed primarily of two constituents (Chuanxiong andChishao) and comes under the Xuefu zhuyu formulate. This traditional Chinese medicine formulate has been and remains the main treatment for cardiovascular disease in China. Many modern studies have showed that the Xuefu zhuyu formulate can be used for improving microcirculation, dilating cardiac vessels, preventing thrombosis, removing blood stasis, and suppressing the aggregation of platelets (Geng 2011; Lei 2002; Luo 2006; Tang 2010), and it is helpful in resisting the loss of lumens after PCI (Fang 2002; Li 2006; Shi 1997). Pharmacological and clinical studies have found that Xiongshao is effective in preventing in‐stent restenosis (Chen 2005; Lu 2006b; Xu 2001; Xu 2004a). Xiongshao is available in capsule form, and several biologically active ingredients have been identified to date, including ferulaic acid, protocatechuic acid, caffeic acid, chrysophanic acid, gallic acid, progallin A, and hendecanoic acid (Wang 2001). A pill of Xiongshao capsule contains about 0.25 g valid ingredients, which is equal to 30 g Chuanxiong and 15 g Chishao (Xu 2001).

How the intervention might work

Various clinical factors, including lesion location, residual stenosis, numbers of stents, stent length, occlusion, and bifurcating lesion, were the main factors associated with restenosis after PCI (Mitra 2006). The mechanism of restenosis involved mainly two processes － neointimal hyperplasia and vascular remodeling － which jointly participated in the pathological course of atherosclerotic lumen stricture (Costa 2005). Injury to the endothelium was one of the initiating elements of pathological vascular remodeling, and hyperlipaemia could intensify its formation (Bauters 1996; Casscells 1994). The mechanisms of restenosis also relate to platelet aggregation, inflammatory cell infiltration, release of growth factors, and medial smooth muscle cell (SMC) modulation and proliferation (Bayata 2010; Carter 1994). Lu's study showed that Xiongshao capsule could improve pathological vascular remodeling and inhibit intima hyperplasia by regulating the content of collagens within blood vessels (Lu 2006b). Another study found that Xiongshao could significantly inhibit pathological vascular remodeling after balloon injury, and it reduced late lumen loss in a porcine coronary balloon injury experiment (Xu 2001). Furthermore, Xiongshao capsule could prevent the formation of atherosclerosis by inhibiting intimal proliferation after balloon injury in rabbits with experimental atherosclerosis (Xu 2004a). In studies of its molecular mechanism, Xiongshao capsule was shown to reduce levels of total cholesterol (TC) and low‐density lipoprotein cholesterol (LDL‐C) in an atherosclerosis rabbit model, and to inhibit platelet aggregation after injury to the endothelium (Li 2008a). Xiongshao capsule could markedly inhibit the proliferation of SMC caused by endothelin‐1 (ET‐1), and it could induce apoptosis of SMC (Xu 2000). Moreover, Xu's study showed that Xiongshao capsule could regulate vascular activity and improve the function of endothelium‐dependent vascular dilation, which apparently led to decreased plasma ET‐1 and increased serum Nitric Oxide (NO) levels (Xu 2004a).

Why it is important to do this review

Over the past three decades, PCI technology has largely revolutionized the field of cardiology through the use of balloon angioplasty and different coronary stents. As a nonmedical treatment for coronary artery disease, coronary angioplasty with stenting is currently most popular. However, in‐stent restenosis has become the major limitation in its application (Hamid 2007). Although the DES has proved superior to the bare‐metal stent (BMS) in reducing the incidence of restenosis, DES‐related restenosis continues at a rate of 5% to 10%, and is especially more frequent in complex lesion subsets (Vaknin‐Assa 2010). On the other hand, prolonged dual antiplatelet therapy may lead to the risk of upper gastrointestinal bleeding (Lemos 2004), and results of randomised controlled trials (RCTs) show that treatment with DES was not associated with a reduction in death or in MI (Gupta 2010; Kirtane 2009). Today more than three million PCIs and consecutive implantation of one or more stents are performed worldwide every year. At least 0.15 to 0.3 million cases of in‐stent restenosis in these patients will be recurrent per year, even if DES implantation is used in all cases (Joost 2010). So the issue of restenosis will continue to be notable and remains a clinical challenge in the field of cardiology.

For many years, Xuefu zhuyu concentrated pill, Xiongshao capsule, Shuxinyin capsule, Fufang Danshen drop, and other treatments have been used widely in CHM to prevent restenosis after PCI. Most of these treatments have been reported to have positive effects on reducing the rate of restenosis (He 2010; Ren 2008; Shi 1997; Wang 2002; Yu 1998; Zeng 2004). Xiongshao capsule is one of the most frequently used treatments in CHM, and many significant results have shown a reduced incidence of restenosis and clinical end point events after PCI in participants with CHD when compared with control groups (Chen 2005; Lu 2006b; Xu 2000). A few studies have reported toxicity between Xiongshao capsule and other drugs, and it occasionally has caused upset stomach or dry stool when used clinically (Chen 2005; Lu 2006b). However, the effectiveness and adverse effects of Xiongshao capsule have not been assessed systematically. The objective of this review is to assess the efficacy and safety of the CHM Xiongshao capsule in preventing restenosis after PCI in study participants with CHD.

Objectives

To systematically assess the efficacy and safety of the Chinese herbal medicine Xiongshao capsule in preventing restenosis after PCI in study participants with CHD.

Methods

Criteria for considering studies for this review

Types of studies

Only RCTs were included.

Types of participants

Participants of either sex and any age or ethnic origin had been given a clinical diagnosis of coronary heart disease based on recognised diagnostic criteria. They were required to have successfully undergone a PCI operation and had been allocated to receive treatment with Xiongshao capsule/plus conventional western medicine or control compared with placebo +/‐ conventional western medicine, and to have received treatment for at least one month.

Studies that included participants with graft vessel disease, serious left main coronary artery disease, serious cardiac inadequacy (ejection fraction [EF] < 35%), serious cardiac valve disease, and serious hepatic failure or renal failure were excluded (CSCCMA 2009).

Types of interventions

Xiongshao capsule used alone or in combination with conventional western medicine compared with the same conventional western medicine alone or plus placebo. Conventional treatment included oral aspirin, clopidogrel, angiotensin‐converting enzyme (ACE) inhibitors, β‐adrenergic receptor blockers, statins, or nitrates, or any other agents recommended by percutaneous transluminal coronary intervention guidelines (CSCCMA 2009).

Types of outcome measures

Primary outcomes

• Cardiac mortality.

• All‐cause mortality.

• Serious adverse events (acute myocardial infarction (AMI), target lesion or vessel revascularization, repeat PCI, coronary artery bypass (CABG)).

Secondary outcomes

• Recurrence angina pectoris.

• Restenosis (defined angiographically).

• Health‐related quality of life (ideally determined using validated instruments).

All outcomes were assessed early (within one month) and later (after three+ months).

Search methods for identification of studies

A comprehensive and exhaustive search strategy was formulated in an attempt to identify all relevant studies regardless of published or unpublished and without language restrictions.

Electronic searches

The following main electronic databases were searched:

• Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library (Issue 3 of 2012).

• MEDLINE (OVID) (from 1948 to week 1 March 2012).

• EMBASE (OVID) (from 1980 to week 10 2012).

• ISI Web of Science with Conference Proceedings (from 1970 to 14 March 2012).

• LILACS (from 1982 to 15 March 2012).

• CBMdisc (Chinese Biomedical Literature Database (from 1980 to May 2012).

In addition, the following special Chinese full‐text databases were searched:

• China National Knowledge Infrastructure (CNKI, from 1994 to May 2012).

• Chinese Medical Current Contents (CMCC, from 1994 to May 2012).

• VIP Database for Chinese Technical Periodicals (VIP; from 1989 to May 2012).

The following unpublished literature databases were searched:

• Chinese Master's Theses Full‐Text Databases (CMFD; from 1994 to May 2012).

• China Doctor Dissertation Full‐Text Database (CDFD; from 1994 to May 2012).

• China Proceedings of Conference Full‐Text Database (CPCD; from 1994 to May 2012).

Search strategies for English language databases are listed in Appendix 1. Strategies for searching the Chinese databases are listed in Appendix 2.

Searching other resources

Ongoing and unpublished trials were searched (30 May 2012) in the following clinical trial registers:

• WHO International Clinical Trials Registry Platform (ICTRP) on http://apps.who.int/trialsearch/.

• International Standard Randomised Controlled Trial Number Register (ISRCTN) on www.controlled‐trials.com/isrctn/.

• Clinical Trials (www.clinicaltrials.gov).

• Chinese Clinical Trial Register (www.chictr.org).

In addition, the reference list for each identified study was reviewed. Bibliographies and reviews for relevant studies were also examined, and experts and pharmaceutical companies in the field were contacted to discuss relevant ongoing trials.

Data collection and analysis

Selection of studies

Two review authors (HYC and LJM) independently scanned abstracts, titles, or both sections of every record retrieved according to the inclusion criteria. For items of literature with unclear information in the title or in the abstract, the full text was retrieved for clarification. All potentially relevant articles were investigated as full text. The most complete literature was selected, or the data were pooled, when two or more publications pertained to a single trial. The divergent views of two review authors were arbitrated by a third review author (GHZ). When resolution of disagreement was not possible, articles were added to those 'awaiting assessment' and were clarified through contact with original authors.

Data extraction and management

Two review authors (HYC and LJM) independently extracted data using the data extraction form, and results were checked for accuracy by a third review author (JFC). Discrepancies were resolved by discussion if necessary. We maximized the yield of information by simultaneous evaluating all available data when duplicate publications and companion papers of a primary article were identified. We extracted the following information from each included trial, if possible.

• Basic information: titles, authors, publication status, etc.

• Baseline characteristics: diagnostic criteria of CHD, mean age, gender proportions, sample numbers, routine risk factors for CHD, types of coronary stents, complications, and characteristics of vascular lesions.

• Description of methodology: method of randomisation, method of concealment of allocation, exclusions post randomisation, blinding, losses to follow‐up.

• Intervention characteristics: doses, routes and names of administered drugs, period of treatment, follow‐up period.

• Outcomes: primary and secondary outcomes, adverse events.

In addition, we contacted the original authors for additional information if data from these trials were insufficient or missing.

Assessment of risk of bias in included studies

Two review authors (JFC and LJM) independently assessed the risk of bias in included trials using The Cochrane Collaboration tool, and the following criteria were applied (Higgins 2011). We contacted the original authors to confirm and validate methods of allocation concealment and randomisation procedures if descriptions of methodological quality were unclear in reports of included trials:

• Was the allocation sequence randomly generated?

• Was the treatment allocation adequately concealed?

• Was knowledge of the allocated interventions adequately prevented during the study?

• Were incomplete outcome data adequately addressed?

• Were reports of the study free of suggestion of selective outcome reporting?

• Was the study apparently free of other problems that could put it at a high risk of bias?

In addition, baseline characteristics, co‐interventions avoided, and level of compliance between comparison groups in all included studies were assessed.

Measures of treatment effect

For dichotomous outcomes

Dichotomous data (restenosis, mortality, recurrence of angina, adverse events, and major cardiac adverse effects) were expressed as risk ratios (RRs) with 95% confidence interval (CIs). Data were pooled by the fixed‐effect model, but the random‐effects model was also used when heterogeneity among studies was obvious.

For continuous data

Continuous outcomes (such as quality of life) were analysed using weighted mean differences (95% CIs) or standardized mean differences (95% CIs) when different measurement scales were applied. Skewed data and nonquantitative data were presented descriptively.

Unit of analysis issues

All four included trials were RCTs with a standard design in the present review, therefore no unit of analysis issues had to be managed. If trials with nonstandard designs, such as cluster‐randomised trials, become available in the future, we will assess the unit of analysis by using the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Dealing with missing data

We contacted the original authors and obtained relevant missing data as far as possible. We carefully evaluated the importance of numerical data such as numbers of randomly assigned patients and intention‐to‐treat (ITT) and per‐protocol populations. We carefully evaluated attrition rates, for example, exclusions post randomisation, drop‐outs, losses to follow‐up, and withdrawals, and critically appraised issues of missing data.

We planned to perform sensitivity analyses to assess the impact on overall treatment effects of included trials that did not report an intention‐to‐treat analysis, with high rates of participant attrition or with other missing data.

Assessment of heterogeneity

We assessed the heterogeneity between included trials by using the Chi‐square test on N‐1 degrees of freedom with a significance level of 0.1 (Lau 1997). We also specifically examined heterogeneity with the I² statistic, quantifying inconsistency across studies to assess the impact of heterogeneity on the meta‐analysis (Higgins 2003). When heterogeneity was not significant (P ≥ 0.1), the results were pooled using a fixed‐effect model and the Mantel‐Haenszel method (Petitti 1994). When high levels of heterogeneity were noted among trials (I² > 70% or P < 0.1), we planned to analyse study components, including participants, diseases, interventions, comparisons, outcomes, etc, in the included studies, and to try to explain the source of heterogeneity by subgroup analysis or sensitivity analysis. A random‐effects meta‐analysis with the DerSimonian and Laird method was planned for use as an overall summary if heterogeneity among trials was considered moderate.

Assessment of reporting biases

We did not assess the potential publication by using funnel plots because only four trials are included in the present review (Higgins 2011; Sterne 2001). Selective reporting was unclear because of lack of protocols for the RCTs.

Data synthesis

We statistically summarised pooled effects if exacted outcomes were available and were sufficiently similar and of sufficient quality. We performed statistical analyses according to the statistical guidelines referenced in the current version of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

When the test of heterogeneity was not significant, a meta‐analysis of data was performed using fixed‐effect models, otherwise random‐effects models. However, we provided a narrative, qualitative summary when heterogeneity was substantial.

All data were analysed by using the statistical software RevMan 5.1 (Oxford, England) of The Cochrane Collaboration, and all P values were two‐sided.

Subgroup analysis and investigation of heterogeneity

Subgroup analyses were performed to explore the source of heterogeneity based on the following:

• Types of PCI (such as percutaneous transluminal coronary angioplasty [PTCA]) and BMS and DES.

• Follow‐up period (such as 3, 6, 12 months).

• Types of control (such as using conventional western medicine alone, conventional western medicine plus placebo).

• Course of treatment using Xiongshao capsule (such as 1, 3, 6 months).

Sensitivity analysis

We did not conduct sensitivity analyses in the present review because few trials were included. We will perform sensitivity analyses in the future to explore the influence of the following factors on heterogeneity:

• Repeating the analysis excluding unpublished studies.

• Repeating the analysis taking account of study quality.

• Reanalysing the data using different statistical analysis models (fixed‐ and random‐effects models).

• Repeating the analysis excluding studies using the following filters: cluster‐randomisation, high rate of participant attrition, missing data on primary outcomes.

Results

Description of studies

For a detailed description of studies, see Characteristics of included studies; Characteristics of excluded studies.

Results of the search

In total, 1977 records were searched out by using the electronic search strategies listed in Appendix 1 and Appendix 2. Of these, 1588 records came from English databases, and 389 records from Chinese databases. After duplicates among different databases were removed, 1612 potentially relevant records (1284 English records, 328 Chinese records) were kept for further assessment. We excluded 1594 records after closer scrutiny of the title and available abstract. The full texts of 18 remaining records were downloaded for careful assessment. Finally we excluded five articles; the reasons for exclusion are listed in Characteristics of excluded studies. We prepared a PRISMA flow diagram to describe the articles identified through our searches (Figure 1).

1.

Study flow diagram.

Included studies

A total of four studies with 13 citations were included in the review (see Characteristics of included studies). All of the four included studies were of parallel, two‐arm, randomised, controlled design; one study (Chen 2006) used a multi‐centre, randomised, double‐blind, placebo‐controlled design, and was supported by a grant from the national Tenth Five‐Year Projects Plan (No. 2001BA701A20). All included studies were conducted in China, and were performed in a hospital setting.

A total of 649 participants were included in the four trials (Chen 2006; Liu 2012; Lu 2006; Xu 2002). Numbers of participants in each trial ranged from 82 (Liu 2012) to 335 (Chen 2006). Ages of participants ranged from 30 to 75 years. All 649 participants showed objective evidence of myocardial ischemia or AMI (at least one > 50% stenosis was documented by coronary angiogram) and had undergone a successful PTCA and/or coronary stenting procedure. Of the four included studies, two (Liu 2012; Xu 2002) compared Xiongshao capsule plus conventional western medicine with the same conventional western medicine alone, and the other two (Chen 2006; Lu 2006) compared Xiongshao capsule plus conventional western medicine with the same conventional western medicine plus placebo. Conventional western medicine agents, such as aspirin, ticlopidine, ditiazem (Herbesser), nitroglycerin, heparin, etc, were taken as routine administration before, during, and after the operation. Then Xiongshao capsule or placebo was initiated on the day of the PCI procedure for both groups (250 mg, three times per day). All included trials reported cardiac mortality, restenosis, recurrence angina pectoris, and serious cardiovascular adverse events such as AMI, target lesion or vessel revascularization, and repeat PCI or CABG. No trial reported outcomes of all‐cause mortality and health‐related quality of life. Two trials reported outcomes of safety and side effects such as blood pressure; heart rate; routine tests of blood, urine, and stool; test of kidney and liver function; etc.

Excluded studies

In total, five articles were excluded on the basis of their full‐text versions. Reasons for exclusion were as follows: Only abstracts were published in two articles (Chen 2002Xu 2002), and a bulletin of medical research in another article (Chen 2006). No outcome data were available in two articles (Xu 2004; Xu 2007). Further information on this can be found in Characteristics of excluded studies.

Risk of bias in included studies

Of the four included studies, we assessed two studies (Chen 2006; Lu 2006) as having a low risk of bias across five aspects of the assessment because of details of design and methodology provided. The other studies were judged as having a high risk of bias because one or more main aspects of the assessment had a high risk of bias. All included trials had prespecified inclusion and exclusion criteria, except for one (Lu 2006). Two trials stated that they performed an ITT analysis to evaluate the data (Chen 2006; Lu 2006). Two trials did not state that they performed an ITT analysis, and we tried to contact the original author with no response. It is difficult to judge whether all outcomes were published because no trial protocol was available for all included studies. Assessment of risk of bias for each included trial is described in the risk of bias tables in the Characteristics of included studies. Authors' judgements about the risk of bias for each included trial are provided in Figure 2 and Figure 3.

2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

3.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Allocation

Three studies reported an adequate sequence generation by using random digit tables or computerized random digit generators (Chen 2006; Lu 2006; Xu 2002). Two studies reported allocation concealment (Chen 2006; Lu 2006). In other studies the method of allocation concealment was unclear because of lack of useful information (Liu 2012; Xu 2002). Detailed information is shown in Figure 2.

Blinding

Two studies (Chen 2006; Lu 2006) were designed with a placebo control group; therefore blinding of patients and personnel (physicians) was substantial. As a result of this, the outcome assessment was not likely to be influenced by the allocation concealment used. So we assessed these studies (Chen 2006; Lu 2006) as having a low risk of bias for performance and detection bias. Detailed information is shown in Characteristics of included studies and Figure 2.

Incomplete outcome data

Although the flow of patients was not reported in any of the included studies, two studies (Chen 2006; Lu 2006) provided information on participants regarding withdrawal, drop‐outs, exclusions post randomisation, or losses to follow‐up, and attrition rates between comparison groups were similar. In two studies (Liu 2012; Xu 2002), the same numbers of participants were randomly assigned and analysed. So we assessed all included studies that had low risk for incomplete outcome data. Further information is given in Characteristics of included studies and Figure 2.

Selective reporting

No study protocols were available for any of the included studies. According to described outcomes in the methods section and reported outcomes in the results section of the publication, we made judgements on this domain. We believed all included studies to be free of selective reporting because the same outcomes were described in the methods and reported in the results. Further information is provided in Characteristics of included studies and Figure 2.

Other potential sources of bias

Baseline information such as general risk factors for CHD, diagnostic criteria, angiographic data, etc, between comparisons was considered to be similar in the four included studies; furthermore most of them listed a statistical table of their clinical and angiographic baseline characteristics. In addition, usage of Xiongshao capsule and conventional western medicines was similar, and co‐intervention was not reported in each included trial. So we considered all four included trials to be free of other potential sources of bias. Detailed information is shown in Characteristics of included studies and Figure 2..

Effects of interventions

See: Table 1; Table 2

No data were given on the primary outcome of all‐cause mortality and on secondary outcomes of health‐related quality of life in any of the included trials. No cardiac mortality was reported in any included trials (Analysis 1.1; Analysis 2.1). We were able to perform meta‐analyses on only three outcomes in this review.

1.1. Analysis.

Comparison 1 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM), Outcome 1 Cardiac mortality.

2.1. Analysis.

Comparison 2 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo, Outcome 1 Cardiac mortality.

1. Xiongshao capsule plus conventional western medicine versus conventional western medicine

Serious adverse events

One trial reported outcomes of serious cardiovascular adverse events including CABG, nonfatal target lesion MI, and repeat PCI after follow‐up for six months (Xu 2002). In comparing Xiongshao capsule plus conventional western medicine with conventional western medicine alone, the ITT analysis involving 108 patients showed an obvious difference between comparison groups (RR 0.47; 95% CI 0.25 to 0.90; Analysis 1.2) .

1.2. Analysis.

Comparison 1 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM), Outcome 2 Serious adverse events.

Restenosis

Restenosis rates at six months after the PCI operation were reported in two trials with 190 participants (Liu 2012; Xu 2002) and showed a statistically significant difference between the two treatments (RR 0.41, 95% CI 0.22 to 0.75, Analysis 1.3).

1.3. Analysis.

Comparison 1 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM), Outcome 3 Restenosis.

Recurrence angina pectoris

Two trials (Liu 2012; Xu 2002) of 190 participants, including 94 in the intervention groups and 96 in the control groups, investigated recurrence angina pectoris at six months' follow‐up. A meta‐analysis of two trials showed a significant effect in favour of Xiongshao capsules plus conventional western medicine when compared with conventional western medicine alone (RR 0.47, 95% CI 0.30 to 0.72, Analysis 1.4) .

1.4. Analysis.

Comparison 1 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM), Outcome 4 Recurrence angina pectoris.

2. Xiongshao capsule plus conventional western medicine versus conventional western medicine plus placebo

Two trials involving 459 participants compared Xiongshao capsule plus conventional western medicine with conventional western medicine plus placebo (Chen 2006; Lu 2006). The numbers of trial participants were 335 (Chen 2006) and 124 (Lu 2006), respectively. Interventional duration was six months, and outcomes including serious adverse events, restenosis, and recurrence angina pectoris were measured. The follow‐up periods of different outcomes were different, ranging from one month to two years in different trials.

Serious adverse events

Two trials reported outcomes of serious adverse events after the PCI operation at six months' follow‐up (Lu 2006) or two years' follow‐up (Chen 2006), respectively. Different types of serious adverse events were noted between the two trials. Serious adverse events including CABG, nonfatal target lesion MI, and target lesion revascularization were reported in Chen's trial (Chen 2006), and CABG, nonfatal target lesion MI, and repeat target vessel angioplasty were reported in Lu's trial (Lu 2006). Subgroup analysis revealed that participants administered Xiongshao capsule plus conventional western medicine had significantly fewer serious adverse events than participants given conventional western medicine plus placebo at six months' follow‐up (RR 0.42, 95% CI 0.19 to 0.95) or two years' follow‐up (RR 0.46, 95% CI 0.26 to 0.76) after the PCI operation (Analysis 2.2).

2.2. Analysis.

Comparison 2 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo, Outcome 2 Serious adverse events.

Restenosis

Two trials involving 432 participants reported outcomes of restenosis rates between comparison groups after six months' follow‐up (Chen 2006; Lu 2006). In the two trials analysed, Xiongshao capsule plus conventional western medicine were more effectual in preventing restenosis after PCI than was conventional western medicine plus placebo (RR 0.52, 95% CI 0.33 to 0.80, Analysis 2.3).

2.3. Analysis.

Comparison 2 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo, Outcome 3 Restenosis.

Recurrence angina pectoris

Recurrence angina pectoris in one trial involving 314 participants was investigated at one month, three months', and six months' follow‐up after PCI (Chen 2006). Another trial involving 118 participants investigated this outcome only at six months' follow‐up after PCI (Lu 2006). Significant differences in preventing the incidence of recurrence angina pectoris after PCI favoured Xiongshao capsule plus conventional western medicine at various follow‐up periods. The RR value (95% CI) was 0.31 (95% CI 0.1 to 0.92, Analysis 2.4) at one month, 0.37 (95% CI 0.19 to 0.71, Analysis 2.5) at three months', and 0.26 (95% CI 0.18 to 0.38,Analysis 2.6) at six months' follow‐up, respectively.

2.4. Analysis.

Comparison 2 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo, Outcome 4 Recurrence angina pectoris with 1 month follow‐up.

2.5. Analysis.

Comparison 2 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo, Outcome 5 Recurrence angina pectoris with 3 months follow‐up.

2.6. Analysis.

Comparison 2 Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo, Outcome 6 Recurrence angina pectoris with 6 months follow‐up.

3. Safety and adverse events

Outcomes of safety with Xiongshao capsule were reported in one trial (Chen 2006) involving 335 participants. Results showed that blood pressure, heart rate, routine tests of blood, urine, and stool, and tests of liver and kidney function revealed no statistically significant difference between comparison groups at six months after PCI. Two trials (Chen 2006; Lu 2006) involving 459 participants reported adverse events according to trial duration. Although two participants suffered from gastric discomfort in the Xiongshao capsule group, this had resolved without interruption of Xiongshao capsule treatment. Granulocytopaenia was reported in one participant in the placebo control group and was considered to be related to concomitant administration of the western medicine ticlopidine.

Discussion

Summary of main results

Four randomised trials were included in this review. Two trials compared Xiongshao capsule plus conventional western medicine with conventional western medicine. Others compared Xiongshao capsule plus conventional western medicine with conventional western medicine plus placebo. All included trials had the same intervention durations and the same doses of Xiongshao capsule were used. Slight clinical heterogeneity was noted in the interventions of included trials.

In this systematic review, we found evidence from two trials (Liu 2012; Xu 2002) indicating that Xiongshao capsule combined with conventional western medicine was significantly better in preventing restenosis and recurrence angina pectoris when compared with conventional western medicine alone (RR 0.41, 95% CI 0.22 to 0.75; RR 0.47, 95% CI 0.30 to 0.72). Furthermore, in the meta‐analysis of these two outcomes, no statistical heterogeneity was noted among the comparisons (all I² = 0%). Results from one trial (Xu 2002) suggest that Xiongshao capsule combined with conventional western medicine was significantly better in reducing serious cardiovascular adverse events when compared with conventional western medicine alone (RR 0.47; 95% CI 0.25 to 0.90) after PCI.

In the analysis of Xiongshao capsule plus conventional western medicine versus the same conventional western medicine plus placebo(Chen 2006; Lu 2006), participants given Xiongshao capsule plus conventional western medicine had a significant reduction in rate of restenosis, recurrence angina pectoris, and serious cardiovascular adverse events when compared with those given the same conventional western medicine plus placebo.

Overall completeness and applicability of evidence

This review includes a meta‐analysis of four trials involving 649 participants with coronary heart disease who had undergone successful PCI. This analysis reveals that Xiongshao capsule may be effective in preventing serious cardiovascular adverse events, restenosis, and recurrence angina pectoris. Although the age and gender of participants in the included trials were representative of CHD patients who undergo successful PCI operations, the sample numbers of included trials was not enough; furthermore, all participants were recruited from Chinese populations. This fact could have an impact on the applicability of the intervention.

Quality of the evidence

Most of the trials included in this review provided detail information on study design, randomisation, allocation concealment, and baseline data. Two of four included trials (Chen 2006; Lu 2006) were designed as randomised, double‐blind, placebo‐controlled trials. One was identified as multi‐centre, large‐scale RCT. Overall trial quality was reasonable, and trials were assessed as having a low risk of bias with The Cochrane Collaboration 'Risk of bias' tool. Although the other two included trials (Liu 2012; Xu 2002) did not mention the method of allocation concealment used, which was considered to be one of the most important causes of bias in a randomised controlled trial (Altman 2001), the effect of allocation concealment on outcomes was similar. These trials can therefore be considered to provide reasonably good evidence on six months' follow‐up of preventing restenosis and recurrent angina pectoris and reducing the incidence of serious cardiovascular adverse events after PCI in CHD patients.

Potential biases in the review process

We attempted to minimise bias in the review process. First, a comprehensive search without language restrictions was conducted to identify all relevant studies. Second, two review authors independently assessed the studies for inclusion and carried out data extraction to ensure that all relevant data had been obtained. Two review authors independently assessed the bias risk of included studies according to the rules of the Cochrane Handbook for Systematic Reviews of Interventions, and disagreements were resolved by consensus. In addition, we strictly followed the protocol outline in performing the review. Nevertheless, potential biases exist. For example, the process of assessing risk of bias might involve some degree of subjective judgement on the part of the assessors. Publication bias was not assessed by using a funnel figure in this review because fewer than ten studies were used for each comparison variable. Publication bias was possible because we found that all included studies had been published, and most reported outcomes were positive. Many of the outcomes predefined in the protocol, such as health‐related quality of life and all‐cause mortality, were not measured in the included studies; therefore we could not review the data for the purpose of weighing the effect estimated.

Agreements and disagreements with other studies or reviews

We did not identify any other systematic reviews with a focus on Xiongshao capsule for restenosis after PCI in patients with CHD. Two systematic reviews (Ren 2008; Zheng 2012) on Chinese herbal medicines for restenosis after PCI were published in 2008 and 2012. One review (Zheng 2012) included 52 trials involving 34 Chinese herbal medicines, and the other review (Ren 2008) included 17 trials involving 15 Chinese herbal medicines. Patients with coronary heart disease who had successfully undergone PCI were included in the reviews (Ren 2008; Zheng 2012). The conclusion of the reviewers was that current evidence from randomised trials shows that Chinese herbal medicines may be helpful in preventing restenosis (Ren 2008; Zheng 2012). Although other reviews (Ren 2008; Zheng 2012) did not focus on the same objectives defined for this review, their conclusions are similar.

Authors' conclusions

Implications for practice.

Based on this systematic review of four trials with 649 participants, summary estimates indicate that Xiongshao capsule may have a protective effect in preventing restenosis and recurrence angina pectoris and in reducing the incidence of serious cardiovascular adverse events, and Xiongshao capsule may be an option for preventing restenosis without adverse effects after PCI in CHD patients. However, data on the effects of Xiongshao capsule on all‐cause mortality and on quality of life after PCI in study participants with coronary heart disease are lacking. Because of the relatively small samples of the included studies, some binary outcomes were highly significant between comparison groups and showed large relative risk reductions; these results are likely implausible and should be considered with great caution for practice.Therefore no strong evidence has been found for preventing restenosis after PCI with Xiongshao capsule.

Implications for research.

Further research should consider the following points:

• Methods of random sequence generation and allocation concealment should be adequately described.

• Blinding of outcome assessors should be applied to reduce detection bias.

• Sample size calculation should be applied in the study.

• Clinical trial registries should be encouraged to provide the available protocol.

• Participants with withdrawal/drop‐out during the trial should be clearly described.

• The follow‐up period should be long enough to assess the long‐term efficacy and safety of the treatment.

• Adverse effects and economic evaluation should be appraised in the study.

What's new

Date	Event	Description
11 March 2014	Amended	Author's contact details have changed

Appendices

Appendix 1. Search strategies for the English language databases

CENTRAL

#1 (xiongshao)
 #2 MeSH descriptor Drugs, Chinese Herbal, this term only
 #3 (xiong shao)
 #4 (chinese near/3 (medic* or drug* or capsul*))
 #5 MeSH descriptor Medicine, Chinese Traditional, this term only
 #6 (chuanxiong)
 #7 (chishao)
 #8 "radix paeoniae rubra"
 #9 (chuanxingol*)
 #10 (paeoniflorin*)
 #11 Xuefu Zhuyu
 #12 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11)
 #13 MeSH descriptor Coronary Disease explode all trees
 #14 (coronary near/2 disease*)
 #15 restenosis
 #16 (#13 OR #14 OR #15)
 #17 (#12 AND #16)

MEDLINE OVID

1. xiongshao.tw.
 2. Drugs, Chinese Herbal/
 3. xiong shao.tw.
 4. (chinese adj3 (medic* or drug* or capsul*)).tw.
 5. Medicine, Chinese Traditional/
 6. chuanxiong.tw.
 7. chishao.tw.
 8. radix paeoniae rubra.tw.
 9. chuanxingol*.tw.
 10. paeoniflorin*.tw.
 11. Xuefu Zhuyu.tw.
 12. or/1‐11
 13. exp Coronary Disease/
 14. (coronary adj2 disease*).tw.
 15. restenosis.tw.
 16. or/13‐15
 17. 12 and 16

EMBASE OVID

1 xiongshao.tw.
 2 Drugs, Chinese Herbal/
 3 xiong shao.tw.
 4 (chinese adj3 (medic* or drug* or capsul*)).tw.
 5 Medicine, Chinese Traditional/
 6 chuanxiong.tw.
 7 chishao.tw.
 8 radix paeoniae rubra.tw.
 9 chuanxingol*.tw.
 10 paeoniflorin*.tw.
 11 Xuefu Zhuyu.tw.
 12 or/1‐11
 13 exp Coronary Disease/
 14 (coronary adj2 disease*).tw.
 15 restenosis.tw.
 16 or/13‐15
 17 12 and 16

Web of Science

#14 #13 AND #10
 #13 #12 OR #11
 #12 Topic=(restenosis) ;
 #11 Topic=(coronary near/2 disease*)
 #10 #9 OR #8 OR #7 OR #6 OR #5 OR #4 OR #3 OR #2 OR #1
 #9 Topic=("Xuefu Zhuyu")
 #8 Topic=(chuanxingol*)
 #7 Topic=("radix paeoniae rubra")
 #6 Topic=(chishao)
 #5 Topic=(paeoniflorin*)
 #4 Topic=(chuanxiong)
 #3 Topic=((chinese near/3 (medic* or drug* or capsul*)))
 #2 Topic=(xiong shao)
 #1 Topic=(xiongshao)

LILACS

(tw: xiongshao or tw:"xiong shao" or tw: chinese medic* or TW; chinese drug* or TW: chinese capsule* or tw: chuanxiong or tw:chishao or tw: "radix paeoniae rubra" or tw: chuanxingol* or tw: paeoniflorin* or tw: "Xuefu Zhuyu")  and (tw: "coronary disease*" or tw: restenosis)

Appendix 2. Search strategies ‐ Chinese databases

CBMdisc

1. [缺省]冠心病
 2. [缺省]介入治疗
 3. [缺省]介入手术
 4. [缺省]支架
 5. #2 or #3 or 4#
 6. [缺省]再狭窄
 7. #1 and #5 and #6
 8. [缺省]芎芍
 9. #7 and #8

CNKI

全文=冠心病 and (全文=支架 or 全文=介入治疗 or 全文=介入手术) and 全文=再狭窄 and 全文=芎芍

CMCC

冠心病/fld=摘要 and (支架/fld=摘要 or 介入治疗/fld=摘要 or 介入手术/fld=摘要) and 再狭窄/fld=摘要 and 芎芍/fld=摘要

VIP

U=冠心病*（U=支架+U=介入治疗+U=介入手术）*U=再狭窄*U=芎芍

CMFD , CDFD AND CPCD

全文=冠心病 and (全文=支架 or 全文=介入治疗 or 全文=介入手术) and 全文=再狭窄 and 全文=芎芍

Data and analyses

Comparison 1. Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM).

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Cardiac mortality	2	190	Risk Ratio (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
1.1 follow‐up 6 months	1	108	Risk Ratio (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
1.2 follow‐up 2 years	1	82	Risk Ratio (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
2 Serious adverse events	1	108	Risk Ratio (IV, Fixed, 95% CI)	0.47 [0.25, 0.90]
3 Restenosis	2	190	Risk Ratio (IV, Fixed, 95% CI)	0.41 [0.22, 0.75]
4 Recurrence angina pectoris	2	190	Risk Ratio (IV, Fixed, 95% CI)	0.47 [0.30, 0.72]
4.1 follow‐up 6 months	1	108	Risk Ratio (IV, Fixed, 95% CI)	0.50 [0.29, 0.86]
4.2 follow‐up 2 years	1	82	Risk Ratio (IV, Fixed, 95% CI)	0.42 [0.21, 0.85]

Comparison 2. Xiongshao capsule (XSC) plus conventional western medicine (CWM) versus conventional western medicine (CWM) plus placebo.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Cardiac mortality	2	432	Risk Ratio (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
1.1 follow‐up 6 months	1	118	Risk Ratio (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
1.2 follow‐up 1 year	1	314	Risk Ratio (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
2 Serious adverse events	2	432	Risk Ratio (IV, Fixed, 95% CI)	0.45 [0.28, 0.70]
2.1 follow‐up 6 months	1	118	Risk Ratio (IV, Fixed, 95% CI)	0.42 [0.19, 0.95]
2.2 follow‐up 1 year	1	314	Risk Ratio (IV, Fixed, 95% CI)	0.46 [0.26, 0.79]
3 Restenosis	2	432	Risk Ratio (IV, Fixed, 95% CI)	0.52 [0.33, 0.80]
4 Recurrence angina pectoris with 1 month follow‐up	1	314	Risk Ratio (IV, Fixed, 95% CI)	0.31 [0.10, 0.92]
5 Recurrence angina pectoris with 3 months follow‐up	1	314	Risk Ratio (IV, Fixed, 95% CI)	0.37 [0.19, 0.71]
6 Recurrence angina pectoris with 6 months follow‐up	2	432	Risk Ratio (IV, Fixed, 95% CI)	0.26 [0.18, 0.38]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Chen 2006.

Methods	Study design: randomised, double‐blind, parallel, placebo‐controlled trial. Method of randomisation: SAS software program. Concealment of allocation: pharmacy control of allocation. Exclusions post randomisation: 9 patients were excluded in treatment group because of insufficient follow‐up (3) and protocol violation (6). 12 patients in the placebo group with 3 incomplete follow‐up and 9 noncompliance. Drop‐out: 6 patients (2 patients in treatment group, 4 patients in control group). Losses to follow‐up: none. Intention‐to‐treat analysis: yes.
Participants	Country: China. Setting: hospital. Numbers: 335 participants (166 treatment, 169 control). Age: treatment (mean±standard deviation, 58.52±10.30 years), control (58.74±9.91 years). Sex: male/female 124/33 in treatment group, 123/34 in control group. Inclusion criteria: Patients were 35‐75 years old, had angina and/or objective evidence of myocardial ischemia or acute myocardial infarction, and had at least one significant (.50%) stenosis that was documented on a recent coronary angiogram and was treated successfully with PCI. Exclusion criteria: presence of restenosis lesion or graft vessel lesion; chronic completely obstructive lesion (>3 months); severe left main artery lesion; severe heart failure [ejection fraction (EF) < 35%]; uncontrolled level III hypertension; severe valvular heart disease; insulin‐dependent diabetes mellitus; diseases of hepatic, renal, hematologic, and neurologic systems, psychological abnormalities, and malignancies; pregnancy or breast‐feeding; unable to provide written consent or expected poor compliance; participants in other clinical trials.
Interventions	Conventional western medicine: Pre‐operation: aspirin 100 mg/day, ticlopidine 250 mg bid, Plavix 300 mg (first dose) or 75 mg (later doses) qd. Intra‐operation: heparin 5000 U. Post‐operation: Low molecular heparin 0.4 ml or 0.6 ml subcutaneous injection bid for 3 days. Ticlopidine 250 mg bid for 2 weeks, then qd for 10 weeks. Aspirin 100 mg/day. Treatment: conventional western medicine plus Xiongshao capsules [2 capsules (250 mg/capsule) tid]. Control: conventional western medicine plus placebo (2 capsules tid). Duration of intervention: 6 months. Co‐intervention: no. Type of PCI: percutaneous transluminal coronary angioplasty (PTCA) or stent implantation.
Outcomes	Primary: cardiac mortality, serious adverse events. Secondary: restenosis, recurrence angina pectoris.
Notes	Baseline characteristics (sex, age, course of disease, types of coronary pathological changes, angiographic data, numbers of coronary pathological changes, history of disease, levels of blood lipoid) were balanced between groups (P > 0.05).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random sequence was generated by using SAS software program.
Allocation concealment (selection bias)	Low risk	Pharmacy control of allocation.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind, and Xiongshao capsules and placebo were indistinguishable in all aspects of their outward appearance.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Comparison groups were marked as "A" or "B" group, and outcome assessors did not know the details of which intervention each participant received.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing outcome data, and ITT analysis was performed.
Selective reporting (reporting bias)	Unclear risk	The study protocol is not available; it is difficult to judge whether all outcomes were published.
Other bias	Low risk	Baseline information between groups shows no statistical differences in sex, age, course of disease, types of coronary pathological changes, angiographic data, numbers of coronary pathological changes, history of disease, and levels of blood lipoid. Co‐intervention avoided. The study appears to be free of other sources of bias.

Liu 2012.

Methods	Study design: randomised controlled trial. Method of randomisation: no description. Concealment of allocation: no description. Exclusions post randomisation: none reported. Losses to follow‐up: none. Intention‐to‐treat analysis: not reported.
Participants	Country: China. Setting: hospital. Numbers: 82 participants (41 treatment, 41 control). Age: mean 60.8 years, range 48‐76 years. Sex: male/female 45/37. Inclusion criteria: patients with coronary heart disease diagnosed by coronary arteriography who successfully underwent PCI operation Exclusion criteria: none described.
Interventions	Treatment: conventional western medicine plus Xiongshao capsules. Control: conventional western medicine (aspirin, dipyridamole, diltiazem, clopidogrel hydrogen sulfate, nitroglycerin, heparin, etc; selectivity applied according to illness state). Duration of intervention: 6 months. Co‐intervention: none. Type of PCI: PTCA or stent implantation.
Outcomes	Primary: cardiac mortality. Secondary: restenosis, recurrence angina pectoris.
Notes	Baseline characteristics (sex, age, course of disease, types of coronary pathological changes, numbers of coronary pathological changes, history of disease, levels of blood lipoid) were balanced between groups (P > 0.05).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomised, but detailed information on the randomisation method was not provided.
Allocation concealment (selection bias)	High risk	No description about the allocation procedure.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not reported.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Not reported.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Same numbers of participants randomly assigned and analysed.
Selective reporting (reporting bias)	Unclear risk	The study protocol is not available; it is difficult to judge whether all outcomes were published.
Other bias	Low risk	Baseline information between groups shows no statistical differences in sex, age, course of disease, types of coronary pathological changes, numbers of coronary pathological changes, history of disease, and levels of blood lipoid. The study appears to be free of other sources of bias.

Lu 2006.

Methods	Study design: randomised, double‐blind, parallel, placebo‐controlled trial. Method of randomisation: SAS software program. Concealment of allocation: pharmacy control of allocation. Exclusions post randomisation: none. Drop‐out: 6 patients (2 patients in treatment group, 4 patients in control group). Losses to follow‐up: none, Intention‐to‐treat analysis: yes.
Participants	Country: China. Setting: hospital. Numbers: 124 participants (62 treatment, 62 control). Age: treatment (mean±standard deviation, 58.94±10.79 years), control (57.10±9.81 years). Sex: male/female 48/14 in treatment group, 48/14 in control group. Inclusion criteria: age from 30 to 79 years, coronary heart disease diagnosed by coronary arteriography (stenosis > 50%), successfully underwent PCI operation. Exclusion criteria: transplanted vascular lesion, serious lesion with left main artery, serious cardiac dysfunction (EF < 35%), serious uncontrolled hypertension, serious valvular heart disease, complicating serious diseases such as liver, kidney, and hematopoietic system disorders, nervous system and mental diseases, malignancy (cancer), pregnancy, or lactation. Refusal to sign informed consent. Entry into other clinical trials.
Interventions	Conventional western medicine: Pre‐operation: aspirin 100 mg/day, ticlopidine 250 mg bid, Plavix 300 mg (first dose) or 75 mg (later doses) qd. Intra‐operation: heparin 5000 U. Post‐operation: Low molecular heparin 0.4 ml or 0.6 ml subcutaneous injection bid for 3 days. Ticlopidine 250 mg bid for 2 weeks, then aspirin 100 mg/day qd for 10 weeks. Treatment: conventional western medicine plus Xiongshao capsules (2 capsules/day, 250 mg/capsule, tid). Control: conventional western medicine plus placebo (2 capsules/day tid). Duration of intervention: 6 months. Co‐intervention: none. Type of PCI: PTCA or stent implantation.
Outcomes	Primary: cardiac mortality, serious adverse events. Secondary: restenosis, recurrence angina pectoris.
Notes	Baseline characteristics (sex, age, course of disease, types of coronary pathological changes, numbers of coronary pathological changes, history of disease, levels of blood lipoid) were balanced between groups (P > 0.05).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random sequence was generated by using SAS software program.
Allocation concealment (selection bias)	Low risk	Pharmacy control of allocation.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind, and Xiongshao capsules and placebo were indistinguishable in all aspects of their outward appearance.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Comparison groups were marked as "A" or "B" group, and outcome assessors did not know the details of which intervention a participant received.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing outcome data. Attrition numbers between comparison groups were reported in detail.
Selective reporting (reporting bias)	Unclear risk	The study protocol is not available, it is difficult to judge whether all outcomes were published.
Other bias	Low risk	Baseline information between groups shows no statistical differences in sex, age, course of disease, types of coronary pathological changes, numbers of coronary pathological changes, history of disease, and levels of blood lipoid. The study appears to be free of other sources of bias.

Xu 2002.

Methods	Study design: randomised controlled trial. Method of randomisation: random number table. Concealment of allocation: not reported. Exclusions post randomisation: none. Losses to follow‐up: none. Intention‐to‐treat analysis: yes.
Participants	Country: China. Setting: hospital. Numbers: 108 participants (53 treatment, 55 control). Age: treatment (mean±standard deviation, 55.3±9.0 years), control (58.7±9.12 years). Sex: male/female 41/12 in treatment group, 46/9 in control group. Inclusion criteria: age from 35 to 75 years, coronary heart disease diagnosed by coronary arteriography (stenosis > 50%), successfully underwent PCI operation. Exclusion criteria: insulin‐dependent diabetes mellitus, serious uncontrolled hypertension, complicating serious liver diseases, and entry into other clinical trials.
Interventions	Treatment: conventional western medicine (aspirin, ticlopidine, diltiazem [Herbesser], heparin, nitroglycerin) plus Xiongshao capsules (2 capsules/day, 250 mg/capsule, tid). Control: conventional western medicine. Duration of intervention: 6 months. Co‐intervention: none. Type of PCI: PTCA or stent implantation.
Outcomes	Primary: cardiac mortality, serious adverse events. Secondary: restenosis, recurrence angina pectoris.
Notes	Baseline characteristics (sex, age, diagnosis, location of coronary pathological changes, numbers of coronary pathological changes, and types of coronary pathological changes) were balanced between groups (P > 0.05).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random number table.
Allocation concealment (selection bias)	Unclear risk	No description.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Conventional western medicine plus Xiongshao capsules in treatment group and conventional western medicine in control group. It is impossible to blind participants and personnel.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Primary outcomes such as all‐cause mortality, serious adverse event, and restenosis are substantive or objective. Outcome assessment was not likely to be influenced by lack of blinding.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Same numbers of participants randomly assigned and analysed.
Selective reporting (reporting bias)	Unclear risk	The study protocol is not available. It is difficult to judge whether all outcomes were published.
Other bias	Low risk	Baseline information between groups shows no statistical differences in sex, age, course of disease, types of coronary pathological changes, numbers of coronary pathological changes, history of disease, and levels of blood lipoid. The study appears to be free of other sources of bias.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Chen 2006	A published abstract. No detailed data can be extracted.
Chen 2002	A bulletin of medical research on Xiongshao capsule for restenosis after PCI. No available data can be extracted.
Xu 2002	A published abstract. No detailed data can be extracted.
Xu 2004	A study of Xiongshao capsule in the function of the vascular endothelium of patients after PCI. No available clinical data can be extracted.
Xu 2007	A correlation study on blood stasis syndrome with degree of restenosis after PCI. No available clinical data can be extracted.

Differences between protocol and review

• In the protocol, we planned to perform subgroup analyses based on type of PCI, but in this review, subgroup analysis based on type of PCI was not performed because it was difficult to separate participants with different types of PCI from those in the original trials.

• In the protocol, we planned to perform sensitivity analyses to explore the source of heterogeneity, but in the review, we did not adhere to this plan because only four trials were included with low clinical heterogeneity among these trials.

• In the protocol, we planned to use funnel plots to assess publication bias, but in the review, we did not do this because only four trials were included.

Contributions of authors

Guo Hua Zheng: writing the protocol; designing the search strategies; undertaking searches; screening search results; analysing data; interpreting data; writing the review.

Jian Ping Liu: planning the protocol; providing guidance on the review.

Jian Feng Chu: appraising quality of papers; extracting data from papers; analysing data; providing a clinical perspective; writing the protocol and the review.

Li Juan Mei: undertaking searches; screening search results; appraising quality of papers; entering data into RevMan; analysing data; appraising quality of papers; writing the protocol and the review.

Hai Ying Chen: undertaking searches; screening search results; extracting data from papers; collecting data for the review; providing general advice on the review.

Sources of support

Internal sources

• National Natural Science Foundation of China (Grant No: 81072726), China.

  Supported by The National Natural Science Foundation of China

External sources

• No sources of support supplied

Declarations of interest

None known.

Edited (no change to conclusions)