Table 4

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study weaknesses
McGuinness JB et al, 1976, Scotland	898 patients admitted to CCU.	Prospective diagnostic cohort	Sensitivity of initial	51%
	400 with AMI.		ECG
Starck M and Vacek JL, 1987, USA	221 ED chest pain patients. 39 with AMI.	Prospective diagnostic cohort	Sensitivity of initial	62%	Air Force hospital, possible selection bias.
			ECG		No raw cardiac enzyme data confirming how AMI diagnosed.
Sharkey SW et al, 1988, USA	34 patients admitted to CCU. 34 with AMI.	Prospective diagnostic cohort	Sensitivity of initial	61%	CCU population not ED.
			ECG		Small population size.
Fesmire F et al, 1989, USA	440 ED chest pain patients. 100 with AMI.	Prospective diagnostic cohort	Sensitivity of initial	47%	No evidence of timing of the initial ECG.
			ECG
Rouan G et al, 1989, USA	918 ED chest pain patients. 811 with AMI.	Prospective diagnostic cohort	Sensitivity of initial	13%	Interrator agreement of ECG interpretation not measured from separate participating ED.
			ECG
					Inclusion of AMI and ischaemic ECG changes.
Gibler B et al, 1992, USA	616 ED chest pain patients. 108 with AMI.	Prospective diagnostic cohort	Sensitivity of initial	36%	Recruitment criteria unclear.
			ECG		Unclear if series or selection of patients recruited
Young P and Green T, 1993, USA	222 ED chest pain patients. 43 with AMI.	Retrospective survey	Sensitivity of initial	28%	Retrospective study.
			ECG		Study population mostly elderly
Zalenski R et al, 1993, USA	149 ED chest pain patients. 34 with AMI.	Prospective diagnostic cohort	Sensitivity of initial	47.1%	Small population
			ECG
Fesmire F, 1998, USA	1000 ED chest pain patients. 204 with AMI.	Prospective diagnostic cohort	Sensitivity of initial	55.4%	? 34 patients had AMI following admission to ED.
			ECG
Kudenchuk PJ et al, 1998, USA	3027 ED chest pain patients. 1149 with AMI.	Prospective diagnostic cohort	Sensitivity of initial	69%	Initial ECG taken by paramedics, not in ED.
			ECG