The Thrombolysis in Myocardial Infarction (TIMI) score, derived from unstable angina/non–ST-segment elevation acute myocardial infarction patient population, predicts 14-day cardiovascular events. It has been validated in emergency department (ED) patients with potential acute coronary syndrome with respect to 30-day outcomes. Our objective was to determine whether the initial TIMI score could risk stratify ED patients with potential acute coronary syndrome with respect to the 1-year outcomes. This was a prospective cohort study of patients presenting to the ED with chest pain who underwent electrocardiography. Patients with ST-segment elevation myocardial infarction (acute myocardial infarction) were excluded. Follow-up was conducted by telephone and record review >1 year after the index visit. The main outcome was the 1-year mortality, nonfatal acute myocardial infarction, or revascularization stratified by the TIMI score. Of 2,819 patients, 253 (9%) met the composite outcome. The overall incidence of the composite 1-year outcome of death (n = 119), acute myocardial infarction (n = 96), and revascularization (n = 90) according to TIMI score was TIMI 0 (n = 1,162), 4%; TIMI 1 (n = 901), 8%; TIMI 2 (n = 495), 13%; TIMI 3 (n = 193), 23%; TIMI 4 (n = 60), 28%; and TIMI 5 to 7 (n = 8), 88% (p <0.001). In conclusion, in addition to risk stratifying ED patients with chest pain at the initial ED evaluation, the TIMI score can also predict the 1-year cardiovascular events in this patient population.
The Thrombolysis in Myocardial Infarction (TIMI) score is a simple 7-item risk score. It was originally developed by Antman et al with data derived from the TIMI 11B and Efficacy and Safety of Subcutaneous Enoxaparin in Non–Q-wave Coronary Events (ESSENCE) trials, where it was found to predict the 14-day adverse cardiovascular outcomes in patients with unstable angina or non–ST-segment myocardial infarction. The TIMI risk score was further applied and validated in subsequent clinical trials, which found similar results in the short and long term. Since it was developed and studied using a patient population with acute coronary syndrome (ACS), it required additional study in the emergency department (ED) to determine its value in a largely undifferentiated patient population with symptoms of potential ACS. Studies have determined that it can simplify risk stratification in ED patients presenting with chest pain and is associated with the 30-day outcome. No study has examined the association of the TIMI risk score in the ED with the 1-year outcome. The goal of our investigation was to determine, in a prospective cohort study of ED patients, whether the TIMI risk score can accurately risk stratify patients with potential ACS with respect to the 1-year cardiovascular events. In particular, we focused on the outcomes in patients without previous cardiac events.
Methods
This was a prospective cohort study of ED patients with potential ACS, conducted at the Hospital of the University of Pennsylvania in Philadelphia. Subject recruitment occurred from July 13, 2003 to March 31, 2008. The Local Institutional Committee on Research Involving Human Subjects approved the study protocol. Our hospital is an urban tertiary care center receiving approximately 57,000 patient visits each year.
The inclusion criteria included all patients presenting to the ED with a primary complaint of nontraumatic chest pain or possible ischemic equivalents who underwent electrocardiography for evaluation of possible ACS. Patients with previous cardiac events (acute myocardial infarction [AMI] or revascularization), ST-segment elevation myocardial infarction, or cocaine use in the past week were excluded. All patients with symptoms suggestive of potential ACS were screened.
For all patients, the core criteria recommended in the standardized reporting guidelines were collected. These reporting guidelines were developed by the Emergency Medicine Cardiac Research and Education Group–International (EMGREG-I), with representation from the major cardiology and emergency medicine societies, including the Society for Academic Emergency Medicine, the American Heart Association, the American College of Emergency Physicians, and the American College of Cardiology. Trained research assistants screened and enrolled the patients for 16 to 17 hours/day, 7 days/wk. Treatment proceeded independent of the study.
The following information was collected during the ED stay: characteristics of the chest pain and associated symptoms, medical history, presenting TIMI risk score, the cardiac-related components of the physical examination, electrocardiographic and chest radiographic findings, medications administered during ED stay, and preliminary ED diagnosis. The electrocardiograms were interpreted according to the 7 categories outlined in the standardized reporting guidelines, as well as ST-segment deviation, T- and Q-wave morphology, and the presence of right or left bundle branch block. The initial electrocardiographic findings and cardiac biomarkers were used to calculate the given variables in the TIMI score. If marker evaluation had not been ordered, the values were assumed to be negative, and a score of 0 was assigned to that variable. Because all other clinical information was obtained directly from the patient and physician while the patient was present in the ED, the other TIMI score variables were never unavailable.
The investigators followed the hospital course of the admitted patients. The study participants or proxies were interviewed by telphone at 1 year after index presentation. If the patient was not reached by telephone, we conducted an electronic medical record review that included surveying 3 area hospitals linked to the health system. If the patient had not returned to the health system, the Social Security Death Index was searched. The following information was recorded through telephone interview and chart review: rehospitalization, stress test or cardiac catheterization, percutaneous coronary intervention with or without stent placement, AMI, and death. The main outcomes were all-cause mortality, nonfatal myocardial infarction, and revascularization (percutaneous coronary intervention or coronary artery bypass surgery) within 1 year after ED presentation. The diagnosis of AMI was determined using the criteria determined by the European Society of Cardiology/American College of Cardiology.
The data values were entered into a Microsoft Access 97 database (Microsoft, Redmond, Washington) and imported into Statistical Analysis Systems, version 9.0 (SAS Institute, Cary, North Carolina) for statistical analysis. The analysis of the relation between the 1-year outcome and TIMI score was performed using chi-square testing and the Cochran-Armitage trend test. All continuous data are presented as either the median with the interquartile range or the mean with SD, according to the distribution of the data. The categorical data are presented as the percentage frequency of occurrence.
Results
A total of 3,758 patients were evaluated for enrollment. Of these, 63 were excluded because of age <30 years, 219 because of cocaine use, 5 because of ST-segment elevation myocardial infarction, and 657 because of a history of coronary events (477 AMI, 173 coronary artery bypass grafting, 107 both AMI and coronary artery bypass grafting, and 114 with documented previous abnormal cardiac catheterization findings). Of the remaining 2,819 patients in the study, follow-up data were obtained for 2,683 patients (95%), and a death index search was performed for 136 patients (5%). The mean patient age was 52 ± 14 years; 60% were women, and 64% were black ( Table 1 ). The median time from chest pain onset to presentation was 300 minutes (interquartile range 105 to 1,140), and the median duration of chest pain was 120 minutes (interquartile range 15 to 840).
| Patient Characteristics | Value |
|---|---|
| Age (years) | 52 ± 14 |
| Women | 1,700 (60%) |
| Race | |
| Black | 1,856 (66%) |
| White | 856 (30%) |
| Asian | 49 (2%) |
| Hispanic | 35 (1%) |
| Other | 23 (1%) |
| Hypertension treated by physician | 1,437 (51%) |
| Diabetes mellitus | 492 (17%) |
| Elevated cholesterol (diet or medication use prescribed) | 576 (20%) |
| Family history of first-degree relative with coronary disease <55 years old | 366 (13%) |
| Tobacco use | 1,008 (36%) |
| Previous coronary artery disease | 137 (5%) |
| Previous congestive heart failure | 234 (8%) |
| Previous angina pectoris | 150 (5%) |
| Previous undiagnosed chest pain | 250 (9%) |
| Previous stress testing | 647 (23%) |
| Abnormal findings | 52 (8%) |
| Previous cardiac catheterization | 217 (20%) |
| Initial electrocardiographic impression | |
| Normal | 1,606 (57%) |
| Nonspecific | 748 (27%) |
| Early repolarization | 42 (1%) |
| Abnormal, not diagnostic | 190 (7%) |
| Ischemia, known to be old | 81 (3%) |
| Ischemia not known to be old | 118 (4%) |
| Suggestive of myocardial infarction | 33 (1%) |
| Other electrocardiographic findings | |
| ST-segment elevation | 126 (4%) |
| ST-segment depression | 137 (5%) |
| T wave inversion | 337 (12%) |
| Q waves | 119 (4%) |
| Left bundle branch block | 56 (2%) |
| Right bundle branch block | 117 (4%) |
Of the 2,819 patients, 1,844 were admitted to the hospital and 931 were discharged from the ED. The remaining 44 patients either left against medical advice or before treatment could be completed. Most patients experienced no cardiovascular events within 1 year. Overall, 252 patients (9%) met the composite outcome of death, AMI, or revascularization within 1 year. Of these, 121 (48%) met the outcome during the initial hospitalization and 131 (52%) met the outcome after hospital discharge. Most cardiovascular events (231 [92%] of 252) occurred among the cohort of patients initially admitted to the hospital or observation unit. Of the patients discharged from the ED, 16 (2%) had a cardiovascular event by the 1-year follow-up date. Of the 44 patients who left against medical advice or before treatment was completed, 5 (11%) had cardiovascular event by the 1-year follow-up date. A total of 119 deaths occurred within 1 year, of which 104 (87%) occurred after the initial visit. Also, 95 AMI events occurred within the year, of which 9 (9%) occurred after the initial hospital visit. A total of 90 patients received percutaneous coronary intervention within the 1 year of follow-up, of which 23 (26%) did so after the initial visit. A relation was found between the 1-year cardiovascular event rate and the presenting TIMI risk score (p <0.001; Table 2 ). Sensitivity analysis was performed of the patients lost to follow-up. The inclusion and exclusion of these patients did not alter this relation. All individual components of the TIMI risk score were associated with the 1-year cardiovascular event rate ( Table 3 ).
| TIMI Score | Patients (n) | Composite Outcome (n) |
|---|---|---|
| 0 | 1,162 | 48 (4%) |
| 1 | 901 | 73 (8%) |
| 2 | 495 | 63 (13%) |
| 3 | 193 | 44 (23%) |
| 4 | 60 | 17 (28%) |
| 5–7 | 8 | 7 (88%) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
