Background: Accurate identification of low-risk emergency department (ED) chest pain patients who may be safe for discharge has not been well defined. Goldman criteria have reliably risk-stratified patients but have not identified any subset safe for ED release. Cardiac troponin I (cTnI) values have also been shown to risk-stratify patients but have not identified a subset safe for ED release.
Objective: To test the hypothesis that ED chest pain patients with a Goldman risk of < or =4% and a single negative cTnI (< or =0.3 ng/mL) at the time of ED presentation would be safe for discharge [<1% risk for death, acute myocardial infarction (AMI), revascularization].
Methods: A prospective cohort study was performed in which consecutive ED chest pain patients were enrolled from July 1999 to November 2000. Data collected included patient demographics, medical and cardiac history, electrocardiogram, and creatine kinase-MB and cTnI. Goldman risk stratification score was calculated while patients were still in the ED. Hospital course was followed daily. Telephone follow-up occurred at 30 days. The main outcome was death, AMI, or revascularization (percutaneous transluminal coronary angioplasty/stents/coronary artery bypass grafting) within 30 days.
Results: Of 2,322 patients evaluated, 998 had both a Goldman risk < or =4% and a cTnI < or =0.3 ng/mL. During the initial hospitalization, 37 patients met the composite endpoint (3.7%): 6 deaths (0.7%), 17 AMIs (1.7%), 18 revascularizations (1.8%). Between the time of hospital discharge and 30-day follow-up, 15 patients met the composite endpoint: 4 deaths (0.4%), 6 AMIs (0.6%), and 5 revascularizations (0.5%). Overall, 49 patients met the composite endpoint (4.9%; 95% CI = 3.6% to 6.2%): 10 deaths (1.0%; 95% CI = 0.4% to 1.6%); 23 AMIs (2.3%; 95% CI = 1.4% to 3.2%), and 23 revascularizations (2.3%; 95% CI = 1.4% to 3.2%) within 30 days of presentation.
Conclusions: The combination of two risk stratification modalities for ED chest pain patients (Goldman risk < or =4% and cTnI < or =0.3 ng/mL) did not identify a subgroup of chest pain patients at <1% risk for death, AMI, or revascularization within 30 days.