Yang 2009a.

Methods	Study design: randomised controlled trial. Method of randomisation: random number table. Concealment of allocation: no description. Exclusions post randomisation: none. Length of follow‐up: 3 months. Losses to follow‐up: none. Intention‐to‐treat analysis: not mentioned.
Participants	Participants were diagnosed as having AMI based on the Acute Myocardial Infarction Diagnosis and Treatment Guidelines formulated by the Chinese Society of Cardiology of the Chinese Medical Doctors Associations in 2001, who successfully underwent PCI. Tong‐xin‐luo group: 30 participants (male/female: 22/8; age mean: 58 years). Control group: 29 participants (male/female: 25/4; age mean: 56 years ).
Interventions	Control group: conventional treatment: • Pre‐operation: aspirin 300 mg qd, clopidogrel 600 mg qd. • Intra‐operation: heparin 1000 IU/kg, then 1000 IU/hour during operation. • Post‐operation: low molecular weight heparin 5000 IU subcutaneous injection bid for 7 days; aspirin 300 mg qd. Tong‐xin‐luo group: conventional treatment plus Tong‐xin‐luo capsules (3 capsules, tid). Duration of intervention: more than 24 months. Co‐intervention: none.
Outcomes	Occurrence of myocardial infarction and heart failure; Mortality (death) due to any cardiovascular event; All cause mortality; Use of revascularisation.
Notes	Withdrawal of one participant before randomisation was reported.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomised, random number table was used as randomisation method.
Allocation concealment (selection bias)	Unclear risk	No description about the allocation procedure.
Blinding of participants and personnel (performance bias) All outcomes	High risk	No description about blinding. No placebo.
Blinding of outcome assessment (detection bias) All outcomes	High risk	No description about blinding. No placebo.
Incomplete outcome data (attrition bias) Angiographic restenosis	Unclear risk	The outcome was not reported.
Incomplete outcome data (attrition bias) Outcomes other than angiographic restenosis	Low risk	Same numbers of participants randomly assigned and analysed. No missing outcome data.
Selective reporting (reporting bias)	High risk	The study protocol is not available, but some outcomes of interest described in the methods were not reported.
Other bias	Low risk	Baseline information showed no statistical differences in sex, age, comorbidities, numbers of coronary pathological changes and classification of the New York Heart Function Assessment between the two groups (P > 0.05). The study appeared to be free of other sources of bias.