This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by KINI, P. M. Articles by PIPBERGER, H. V. Search for Related Content PubMed PubMed Citation Articles by KINI, P. M. Articles by PIPBERGER, H. V.

(Circulation. 1970;42:875.)
© 1970 American Heart Association, Inc.

---

Electrocardiographic Differentiation Between Left Ventricular Hypertrophy and Anterior Myocardial Infarction

P. M. KINI M.D.¹; E. E. EDDLEMAN JR. M.D.¹; H. V. PIPBERGER M.D.¹

¹ From the Veterans Administration Cooperative Study on Automatic Cardiovascular Data Processing, Veterans Administration Hospitals, Batavia, New York, Birmingham, Alabama, Durham, North Carolina, Houston, Texas, Minneapolis, Minnesota, San Francisco, California, Washington, D. C., and West Roxbury, Massachusetts, and the Department of Medicine, Georgetown University School of Medicine, Washington, D. C.

Electrocardiograms recorded from patients with uncomplicated left ventricular hypertrophy (LVH) often show loss of initial anteriorly directed QRS forces. An attempt was made to differentiate these records from electrocardiograms showing a similar pattern as a result of anterior myocardial infarction (MI). Orthogonal ECGs (Frank system) obtained from 103 patients with LVH were compared to ECG obtained from 327 patients with MI. The ECG were selected on the basis of absence of Q waves in lead Z which in the conventional ECG corresponds to absence of the R wave in right precordial leads. With two simple scalar measurements, namely R amplitude in lead X 1.2 mv and the sum of amplitudes of R in leads X and Z 2.5 mv, 66% of LVH cases were recognized. With these criteria, 88% of the MI records were also correctly classified. The same measurements identified 59% of an independent sample of 66 cases of LVH with no Q in lead Z. By utilizing linear discriminant function analysis, 75% of the cases of LVH, 80% of the MI, and 70% of the independent group of LVH were correctly classified.

Classification procedures were also tested on 48 autopsy cases with results slightly inferior to those obtained on clinical samples.

Direction of initial instantaneous QRS vectors in the transverse plane proved to be much less efficient than scalar measurements or multivariate analysis in the separation between LVH and MI.

This study confirms the difficulties concerning the correct interpretation of the significance of loss of anteriorly directed forces in the presence of LVH. However, the error rate in diagnosis can be considerably reduced by the use of criteria proposed on the basis of this investigation.

Key Words: ECG classification • Scalar measurements • Multivariate analysis • Computer analysis

Submitted on June 10, 1970
Accepted on June 30, 1970