Electrocardiographic prediction of mortality in anterior myocardial infarction
ACP J Club. 1994 Jan-Feb;120:20. doi:10.7326/ACPJC-1994-120-1-020
Birnbaum Y, Sclarovsky S, Blum A, Mager A, Gabbay U. Prognostic significance of the initial electrocardiographic pattern in a first acute anterior wall myocardial infarction. Chest. 1993 Jun;103:1681-7.
To determine if admission electrocardiographic (ECG) patterns predict in-hospital mortality in patients with an evolving first anterior acute myocardial infarction (MI).
Inception cohort followed until hospital discharge.
Coronary care unit (CCU) in Israel.
147 consecutive patients (mean age, 64 y; 111 men) with an evolving first acute MI of the anterior wall at admission to the hospital. Patients with a history or ECG evidence of a previous acute MI were excluded. Anterior acute MI was diagnosed when chest pain lasted > 30 minutes, changes occurred in serum creatine kinase levels, and serial ECG changes evolved in ≥ 2 adjacent precordial leads.
Assessment of Prognostic Factors
Patients were divided into 3 groups according to ECG pattern at admission: group A, tall symmetric abnormal T waves in the involved leads without ST elevation or major changes in the terminal portion of the QRS; group B, abnormal T waves and ST elevation (> 0.1 mV) in ≥ 2 adjacent leads without major changes in the morphology of the terminal portion of the QRS; and group C, abnormal T waves and ST elevation with distortion of the terminal portion of the QRS complex in ≥ 2 adjacent leads. Age; sex; history of diabetes mellitus, hypertension, and angina pectoris; Killip functional class at admission; and thrombolytic therapy were extracted from the patients' charts.
Main Outcome Measure
12, 77, and 58 patients were included in initial ECG groups A, B, and C, respectively. 19 patients (13%) died in the hospital. The mortality rates in groups A, B, and C were 0%, 3% (2 patients), and 29% (17 patients), respectively (P < 0.01). Using a logistic regression model, the strongest predictor of in-hospital mortality was the initial ECG pattern. The odds ratio between ECG patterns B and A was 15.6; between C and B, 11.6; and between C and A, 181. The predicted probability of death for a patient with ECG pattern A was 0.002 (95% CI, 0 to 0.027), 0.025 (CI, 0.006 to 0.95) with ECG pattern B, and 0.29 (CI 0.19 to 0.42) with ECG pattern C.
The electrocardiographic pattern at admission, which was used to divide patients with a first anterior acute myocardial infarction into 3 subgroups, was predictive of in-hospital mortality.
Source of funding: Not stated.
For article reprint: Dr. Y. Birnbaum, Cardiac Intensive Care Unit, Beilinson Medical Center, Petah-Tikva, Israel. FAX 972-3-924-9850.
A limitation of the study by Birnbaum and colleagues is that 81 patients were excluded, albeit for appropriate ECG criteria. The results in the small number of patients (n = 147) have ramifications for the practicing physician who sees a patient with a first anterior MI in evolution. If the terminal QRS segment is abnormal, it can be predicted that the mortality rate will be 10 times higher.
Although 99 of the 147 received thrombolytic therapy, only 11 received front-loaded tissue plasminogen activator (tPA). Unfortunately, the timing of thrombolytic therapy after the initial ECG was not reported and could have affected mortality.
2 ECG criteria have been shown to be indicative of prognosis in the patient who has had an MI: arrhythmia (1) and the location and extent of infarction (2). Since the advent of thrombolysis, heart rate variability during deep breathing has been added (3). The authors did not evaluate these.
In the recent GUSTO study (4), 41 000 patients were given either tPA or streptokinase. A mild benefit existed for the more expensive tPA given front-loaded. An option for physicians who see patients with this terminal QRS distortion might be to be more aggressive with use of tPA. Also, the initial use of angioplasty might be considered as a potential primary modality, if it is immediately available (5).
Although a larger prospective study needs to be done to corroborate this study, astute clinicians should be on the lookout for terminal distortion to the QRS on initial ECGs in evolving anterior MIs as an indicator of poor prognosis, and develop treatment plans accordingly.
Len Scarpinato, DO
Medical College of Wisconsin Milwaukee, Wisconsin, USA