Electrocardiographic left ventricular hypertrophy in patients with suspected acute cardiac ischemia-its influence on diagnosis, triage, and short-term prognosis: a multicenter study

• Published in: Journal of general internal medicine : JGIM 1994-12, Vol.9 (12), p.666-673
• Main Authors: LARSEN, G. C, GRIFFITH, J. L, BESHANSKY, J. R, D'AGOSTINO, R. B, SELKER, H. P
• Format: Article
• Language: English
• Subjects: Aged; Biological and medical sciences; Cardiology. Vascular system; Coronary Care Units; Coronary heart disease; Electrocardiography; Emergency Service, Hospital; Female; Heart; Humans; Hypertrophy, Left Ventricular - diagnosis; Hypertrophy, Left Ventricular - epidemiology; Likelihood Functions; Logistic Models; Male; Medical sciences; Middle Aged; Myocardial Ischemia - diagnosis; Myocardial Ischemia - epidemiology; Patient Admission; Prognosis; Prospective Studies; Risk Factors; Time Factors; Triage
• ISSN: 0884-8734; 1525-1497
• DOI: 10.1007/BF02599006

To understand the diagnostic and short-term prognostic significance of electrocardiographic left ventricular hypertrophy (ECG-LVH) for patients who present to the emergency department with symptoms suggesting acute cardiac ischemia, defined as new or unstable angina pectoris or acute myocardial infarction. Subgroup analysis of a multicenter, prospective study of coronary care unit admitting practices in the prethrombolytic era. The emergency departments of six New England hospitals: two urban medical school teaching hospitals, two medical school-affiliated community hospitals in smaller cities, and two rural non-teaching teaching hospitals. 5,768 patients presenting with symptoms suggesting possible acute cardiac ischemia, including 413 patients who had ECG-LVH defined by the Romhilt-Estes point score criteria and 5,355 patients who had other electrocardiogram (ECG) findings. Only 26% of the 413 patients who had ECG-LVH were ultimately judged to have had acute cardiac ischemia, compared with 72% of patients who had primary ST-segment and T-wave abnormalities (p < 0.001) and 36% of those who had other ECG abnormalities (p < 0.001). Overall, the ECG-LVH patients were one-third less likely than the patients who did not have ECG-LVH to have had acute cardiac ischemia, after controlling for other predictors of acute ischemia by logistic regression (relative risk = 0.66, 95% CI 0.46 to 0.94). The patients who had ECG-LVH were only one-fourth as likely to have had acute myocardial infarctions as were the patients presenting with primary ST-segment and T-wave changes (12% vs 48%, p < 0.001). Instead, a much larger proportion had had congestive heart failure or hypertension. The admitting physicians had identified ECG-LVH poorly on the admitting ECGs: only 22% of those who had ECG-LVH had been correctly identified, and for more than 70%, the secondary ST-segment and T-wave changes of ECG-LVH had been read as being primary. The short-term mortality for the patients who had ECG-LVH was 7.5%. This was intermediate between the mortality for patients who had primary ST-segment and T-wave abnormalities (10.6%) and those who had other ECG abnormalities (5.1%). Mortality was not affected by whether the admitting physician had recognized ECG-LVH initially. ECG-LVH was not a benign ECG finding among the patients who had presented with symptoms suggesting an acute cardiac ischemic syndrome: short-term mortality among the patients who had ECG-LVH (7.5%) approached that for the