Hektoen International

A Journal of Medical Humanities

The electrocardiographic diagnosis of myocardial ischemia and infarction: 1917-1942

Philip R. Liebson
Chicago, Illinois, United States


Although myocardial infarction and angina pectoris had been recognized as serious heart conditions associated with sudden death since the 19th century (based primarily on patient symptoms of chest pain and pathologic correlations of involvement primarily of the left ventricle), James B. Herrick’s classic 1912 paper on the association of these conditions with coronary occlusion provided the pathophysiologic basis of myocardial ischemia and infarction.1 At the time, electrocardiography was just being developed as a diagnostic tool, initially by the development of the string galvanometer by Willem Einthoven.

Fig. 1: One cycle of an electrocardiograph

The form of one cycle of an electrocardiograph (ECG) is pictured in Fig. 1. The P wave is produced by electrical depolarization of both atria. Atrial contraction occurs after the P wave when the ECG line is flat (PR segment). QRS represents the beginning of electrical depolarization of both ventricles, which remain depolarized during the ST segment, when the ventricles contract (systole). The T wave indicates electrical repolarization of the ventricles, and the ventricles relax after the T wave (diastole). Myocardial involvement in coronary artery disease is reflected in changes in the QRS, ST segment, and T wave. The U wave is a finding in certain forms of electrolyte imbalance.

Fig. 2: The Three Limb Leads

In 1912, the electrocardiograph demonstrated only three leads, called limb leads (I, II, III). In that year, Einthoven addressed the Chelsea Society in London and described the equilateral triangle formed by these lead produced by electrodes on both arms and both legs (Einthoven triangle). In an article in the Lancet that year, he first used the term “EKG” in an article in English.

It was between 1917 and 1942 that the ECG (to use the English acronym) was developed as a tool to diagnose infarction or ischemia. Although the standard ECG evaluates 12 leads (3 limb, 3 unipolar, 6 chest), it was only in 1932 that the clinical use of chest leads was described by Wolferth and Wood, in 1934 that the unipolar limb leads (VR, VL, VF) were developed by Frank Wilson, and in 1942 that Emanuel Goldberger developed the augmented limb leads (aVR, aVL, aVF) to complete the arrival of the standard electrocardiogram.

In 1917, Bernard Oppenheimer and his colleague Marcus Rothschild reported studies of “electrocardiographic changes associated with myocardial involvement” at the annual meeting of the American Medical Association.2,3 James Herrick, who was present at the meeting, commented on the similarity between ECG findings of one of Oppenheimer and Rothschild’s patients with autopsy proof of coronary thrombosis and findings in dogs in which Herrick had ligated the left coronary artery.4,5 The following year, G. Bousfield described the spontaneous changes in the ECG during angina in an article in the Lancet.6

In 1920, Harold Pardee published the first ECG specifically focused on an acute myocardial infarction in a human in a case report with further results from five patients.7 These findings reflected ECG changes in animal studies of coronary occlusion. Since there were only three leads available at the time, the only diagnosis of myocardial infarction (MI) could be made with an inferior wall infarct. Anterior wall infarcts required chest leads which were not yet available. The ECG of the case report showed ST segment elevation in leads II and III and depression in lead I, classical for inferior MI. Within several days, the ST segment returned to its normal baseline but the initial deflection of the QRS complex went downward (Q wave) and the T wave became symmetrically inverted. Serial ECGs were obtained between March 4 and May 25, 1917, the first within 4 hours after the initial attack. The second, on March 7, already demonstrated the reversion of the ST segment to baseline and the development of the Q wave and the T wave inversion. These findings were reflected in the ECGs of the five other patients with myocardial infarction. Pardee concluded:

This electrocardiographic sign indicates the presence of a rather large area of muscle degeneration, and when obtained from a patient who gives a history of precordial pain . . . will complete the diagnosis of obstruction of a branch of a coronary artery.

At the time, there continued to be confusion between an acute coronary occlusion and angina pectoris. This is not surprising. Even in current guidelines, the distinction among ST elevation, MI, non-ST elevation MI, angina pectoris, and acute coronary syndrome requires meticulous definitions and the use of cardiac enzymes to differentiate infarction from ischemia.

In the early 1920s abnormal forms of the ECG were useful in diagnosing “myocardial involvement” but because of the lack of chest leads, anterior infarctions were often missed. In many primarily non-cardiac conditions on the other hand, abnormal ECGs could develop. This would be the case with electrolyte abnormalities associated with kidney disease, certain types of strokes, and effects of certain drugs.

Further evaluations of ECG changes in myocardial infarction were described by John Parkinson and Evan Bedford of the London Hospital in the Journal Heart in 1928.8 They reported 28 cases of “cardiac infarction” due to “coronary thrombosis.” You will note that both the cause and effect relationship are now recorded prominently. They called the ST elevation early in the infarct period the “injury current,” suggesting a possible transient reversible change due to myocardial injury, but permanent infarction would be reflected by the development of Q waves in the QRS complex and T wave inversions.

Until recently, there have been various changes in nomenclature of myocardial infarction. Currently, as indicated, infarction is subdivided into ST elevation and non-ST elevation infarction on the basis of differences in prognoses and interventions. For a long time before this, infarctions were subdivided into “Q-wave” and “non-Q wave” infarctions, both showing abnormal enzymes to indicate infarct diagnosis. This was based on the assumption that a Q-wave infarction was due to a transmural involvement (entire muscle wall) vs. a non-transmural involvement (just the endocardial part of the wall) for a non-Q-wave infarction.

By this time it was recognized that angina pectoris, reversible myocardial ischemia, might or might not be reflected in the ECG. Although an ECG performed during an angina event could demonstrate changes in the ST segment or T wave (and frequently subclinical myocardial ischemia can be determined by ECG changes), this may not necessarily be the case. Aside from the characteristic chest pain, objective findings during an angina attack could include reversible heart murmur due to transient mitral regurgitation from involvement of a papillary muscle, or a transient S3 gallop due to transient stiffness of the ventricle.

By the early 1930s, Charles Wolferth and Francis Wood, in Philadelphia, were interested in practical ways to assess angina pectoris and evaluate ECGs during rest and with exercise in normal subjects and in patients with angina.9 Others (Goldhammer and Scherf in Germany) also proposed that exercise could be used to elicit ECG changes in angina, but at the time it was felt that it would be dangerous to subject patients to angina attacks by exercise.3

However, Wolferth and Wood contributed a significant advance in ECG diagnosis by use of chest leads, reported in 1932.10 Because of “silent areas” of the heart not reflected in the ECG, they experimented with the use of chest leads, first in animals and then in patients. They provided several case reports to demonstrate that “lead IV,” as they called the relevant chest lead, could elicit changes (primarily of anterior infarction) that could not be produced in the limb leads. In one subject, a depressed ST segment and peaked upright T wave in the lead in the mid chest (now V1), the reciprocal of the changes in most ECGs, produced what is now diagnosed as a posterior infarct (or “true posterior infarct” to differentiate it from an inferior infarct, previous called a posterior infarct). Today, anterior infarctions are sometimes subdivided into antero-septal, antero-apical, and antero-lateral infarcts based on the location of the typical changes of Q waves, elevated ST-segments, and/or inverted T waves based on the location of these abnormalities (V1-V2, V3-V4, or V5-V6) respectively. Since the chest leads, although placed on specific areas of the chest, may differ considerably from the orientation of the heart, use of these terms is fallible.

Fig. 3: Distribution of QRS axes and a normal QRS axis


In 1934, Frank Wilson from the University of Michigan and his colleague described the use of unipolar electrical leads (VR, VL, VF) to further elucidate changes in the ECG.11 In relation to myocardial infarction, changes in VF reflect inferior infarctions and changes in VL reflect high lateral wall infarcts. This advance was also valuable in determining the mean electrical axis of the QRS, allowing evaluation of the excitatory process of depolarization. Thus, now we use right axis deviation, left axis deviation and normal axis, standardized diagnoses of each ECG. Right axis deviation could be associated with abnormalities of the right ventricle reflecting pulmonary hypertension or right ventricular disease. Left axis deviation could reflect various types of left ventricular abnormalities. The axis of the T wave in relation to the QRS axis (wide-angle T wave) allows demonstration of nonspecific myocardial abnormalities reflected in abnormal repolarization patterns.

These developments in implementing new leads entailed some definition of ECG standards. Accordingly in 1938, British and American cardiology societies made recommendations for the positioning of the six precordial leads, which they named V1 through V6 (V stands for voltage).12 In addition, right chest leads have been added (named RV leads) in order to evaluate the possibility of right ventricular infarction.

Several papers had already noted the confusion in interpretation of the electrocardiogram, in particular changes in the ST segment in patients receiving digitalis, pericardial effusions, and myocarditis. However, one paper that reviewed electrocardiography in “acute coronary occlusion” (1939) suggested that “it might be advantageous to defer electrocardiographic study until several days after the acute episode” because of late-appearing changes, although it did recommend serial studies.13 An extensive review that year based on 302 evaluations using chest leads concluded that “multiple chest leads are of value . . . because they may yield diagnostic evidence of myocardial infarction when a single chest lead is barren.”14 At present, anyone entering the emergency department with chest pain or requiring emergency medical treatment gets an ECG immediately with aspirin administration.

The final step in the development of the standard ECG was the development of augmented limb leads (aVR, aVL, aVF) by Emanuel Goldberger in 1942.15 That same year, Arthur Master standardized the two-step exercise test, predecessor the current exercise stress test.16 The current standard ECG was now developed and the availability of a standardized exercise test could now provide some evaluation of underlying myocardial ischemia.

Since then, the performance of the standard ECG has been improved by more portable equipment, simpler leads and computerized ECG recording so that all 12 leads can be recorded in sequence without changing a dial for each lead or placing one precordial lead over the six positions sequentially. Nevertheless, it was the period from 1917 to 1942 in which the evaluation of myocardial infarction and ischemia became standardized by electrocardiographic interpretation.



  1. Herrick JB. Clinical features of sudden obstruction of the coronary arteries. JAMA 1912;59:2015-2020
  2. Oppenheimer BS, Rothschild MA. Electrocardiographic changes associated with myocardial involvement with special reference to prognosis. JAMA 1917;69:429-431.
  3. Fye WB. A history of the origin, evolution, and impact of electrocardiography. Am J cardiol1994;73:937-949.
  4. Herrick JB. Discussion of ibid. JAMA 1917;69:429-431.
  5. Herrick JB, Smith FM. The ligation of the coronary arteries with electrocardiographic study. Arch Int Med 1918;22: 8-27.
  6. Bousfield G. Angina pectoris: changes in electrocardiogram during paroxysm. Lancet 1918;2: 475.
  7. Pardee HEB. An electrocardiographic sign of coronary artery obstruction. Arch Int Med 1920;26:244-257.
  8. Parkinson J, Bedford DE. Successive changes in the electrocardiogram after cardiac infarction (coronary thrombosis). Heart 1928;14:195-239.
  9. Wolferth CC, Wood F. Angina pectoris: the clinical and electrocardiographic phenomena of the attack and their comparison with the effects of experimental coronary occlusion. Arch Intern Med 1931;47: 339-365.
  10. Wolferth CC, Wood F. The electrocardiographic diagnosis of coronary occlusion by the use of chest leads. Am J Med Sci 1932;183:30-35.
  11. Wilson FN, Macleod AG, Barker PS, Johnston FD. The determination and significance of the area of the ventricular deflections of the electrocardiogram. Am heart J 1934;10:46-61.
  12. Barnes AR, Pardee HEB, White PD, et al. Standardization of precordial leads. Supplementary Report Am Heart J 1938;15:235-239.
  13. Geiger AJ. Uses and limitations of electrocardiography in the diagnosis of acute coronary occlusion. Yale J Biol Med 1939; 11: 619-628.
  14. Wood P, Selzer A. Chest leads in clinical electrocardiography. Br Heart J 1939 1: 49-80.
  15. Goldberger E. A simple, indifferent, electrocardiographic electrode of zero potential and a technique of obtaining augmented, unipolar , extremity leads. Am Heart J 1942; 23: 483-492.
  16. Master AM, Friedman R, Dack S. The electrocardiogram after standard exercise as a functional test of the heart. Am Heart J 1942:24:777



PHILIP R. LIEBSON, MD, graduated from Columbia University and the State University of New York Downstate Medical Center. He received his cardiology training at Bellevue Hospital, New York and the New York Hospital Cornell Medical Center, where he also served as faculty for several years. A professor of medicine and preventive medicine, he has been on the faculty of Rush Medical College and Rush University Medical Center since 1972 and holds the McMullan-Eybel Chair of Excellence in Clinical Cardiology.


Highlighted in Frontispiece Spring 2013 – Volume 5, Issue 2

Spring 2013  |  Sections  |  Cardiology

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.