Relationship between regional ventricular asynergy and the anatomic lesion in coronary artery disease

The percentage of left ventricular (LV) asynergy was measured in patients with isolated narrowing or obstruction of the right coronary artery (RCA), the anterior descending branch of the left coronary artery (LAD), or a combination of these lesions.Incomplete obstruction of a vessel was not associated with important asynergy. Isolated obstruction of the LAD caused asynergy of the distal two-thirds of the anterior wall and apex of the LV and 46 per cent asynergy. Isolated obstruction of the RCA caused asynergy of the middle or basal thirds of the diaphragmatic surface and 15 per cent asynergy. Double-vessel disease produced a combination of the individual lesions, and total obstruction of both arteries caused extensive asynergy.In each patient the extent of asynergy was modified by the underlying coronary artery anatomy and the collateral circulation. Ejection fraction was related to the percentage of LV asynergy.     

Noninvasively induced postextrasystolic potentiation of ischemic and infarcted myocardium in patients with coronary artery disease.

To evaluate noninvasively-induced postextrasystolic potentiation (PESP) of ischemic or apparently infarcted regions of myocardium, an external mechanical cardiac stimulator (develoepd by Zoll) was used to induce ventricular extrasystoles during M-mode echocardiography in 29 patients with coronary artery disease and in four control subjects. Twenty-five patients had > 75 per cent stenosis of the left anterior descending artery including 13 with ECG evidence of anteroseptal myocardial infarction; 21 patients had > 75 per cent stenosis of the right coronary and/or left circumflex arteries, including 11 with ECG evidence of inferior and/or posterior myocardial infarction. Twenty-four regions with reduced wall excursion showed varying effects of PESP: eight regions improved to the normal range, while 16 did not. Twelve of the latter had ECG evidence of prior infarction. Similarly, regions of asynergy that did not respond at all to PESP were usually, but not always, seen in patients with infarctions. Based on prior ventriculographic-histopathologic correlates, non-responding regions are probably totally scarred with irreversible contraction abnormalities, whereas regions with evidence of contractile reserve are potentially viable. Because the ECG and resting echocardiogram are not totally accurate predictors of contractile reserve, noninvasively induced PESP may be a useful adjunct technique in delineating local contractile reserve in patients with echocardiographic evidence of hypocontractile myocardium of uncertain viability.     

Clinical and hemodynamic aspects of single vessel coronary artery disease

Two-hundred consecutive patients with arteriosclerotic heart disease underwent complete clinical and hemodynamic evaluation. Fifty-two patients (26 per cent) had significant single vessel coronary artery disease and were compared to 148 patients with more extensive coronary artery disease and to a group of 14 normal patients. The single vessel disease group, when compared to the diffuse disease group, was characterized by a shorter duration of angina pectoris, lower frequency of a history of congestive heart failure or cardiomegaly, and a lower frequency of electrocardiographic (ECG) evidence of a transmural myocardial infarction. The combination of angina pectoris for three or more years with cardiomegaly was the only factor which completely separated the two coronary disease groups. Cardiomegaly, when present in single vessel involvement, was always due to left anterior descending (LAD) disease, together with an anterior infarction on ECG and left ventricular asynergy. The single vessel disease group included 32 patients with LAD disease, 17 with RCA, and 3 with circumflex artery involvement. Resting hemodynamics in these 52 patients (other than a higher left ventricular end-diastolic pressure and wall stress) were not significantly different from hemodynamics in a normal group. Patients with diffuse disease were characterized by many hemodynamic alterations and by left ventricular (LV) asynergy, when compared to the single vessel disease or normal groups. The diffuse disease group had a lower ejection fraction (EF) and an increased frequency of LV asynergy and coronary collateral circulation than did the LAD group. In the single vessel disease group LV asynergy did not correlate with the ECG. LV synergy, however, was not found in any patient in the LAD group with abnormal Q waves on ECG. The single vessel disease group included only five patients with increased end-diastolic volume (EDV) and all had LAD involvement, increased LV end-diastolic pressure, and decreased EF. The remaining 47 patients with normal LV-EDV revealed that the LAD group had abnormal pressure-volume relationships, indicating a decreased compliance of the left ventricle.     

Vectorcardiographic, electrocardiographic, and angiographic correlations in apparently isolated inferior wall myocardial infarction.

Twenty-six patients with ECG evidence of localized inferior myocardial infarction and poor ejection fraction (less than 50 per cent) were compared with 26 patients with similar ECG's, but with normal ejection fraction (over 50 per cent). The poor ejection fraction group had significantly more frequent and more severe disease in left anterior descending artery and a higher incidence of triple coronary obstruction than the normal ejection fraction group. The poor ejection fraction group had a significantly greater incidence of ventricular asynergy in the anterior and apical segments of left ventricle. Vectorcardiography was available in 35 of the 52 patients studied and frequently supplied diagnostic information not available in the scalar ECG's. Of 18 patients with scalar ECG patterns of isols, vectorcardiography identified five cases with anterior infarction, three with left ventricular hypertrophy, and one with left anterior hemiblock. Vectorcardiography is a valuable supplementary tool in the clinical assessment of patients with apparently isolated inferior infarction. When extensive coronary and poor ventricular function exist, VCG clues may be expected in about half the patients.     

Diagnostic and prognostic significance of electrocardiographic and CPK isoenzyme changes following coronary bypass surgery: correlation with findings at one year.

The incidence of ECG (14 per cent) indication of acute myocardial infarction complicating coronary artery bypass surgery is documented, corroborating the findings of prior series. An additional 32 per cent of patients had appearance of myocardial specific CPK-MB in serum during the immediate postoperative period. All patients surviving to 1 year following surgery (93 of 103) were asked to return for repeat cardiac catheterization to determine the presence and extent of interim ventricular contraction abnormalities. Sixty-five (70 per cent) of the group returned for evaluation. Preoperative and 1 year postoperative left ventriculograms were compared to determine if new contraction abnormalities would confirm the specificity of perioperative QRS and isoenzyme changes, and if the absence of new abnormalities would confirm their sensitivity. The majority of patients (65 per cent) had new areas of asynergy. However, 73 per cent of these were confined to the apex and thus could have been produced by the vent employed during cardiopulmonary bypass. QRS changes were 100 per cent specific and CPK-MB appearance was 78 per cent specific but they were only 20 and 54 per cent sensitive, respectively. Indeed, 46 per cent of those with new asynergy which was non apical had neither QRS change nor CPK-MB appearance. Thus QRS changes were always--and CPK-MB appearance was usually--associated with new asynergy but, in addition, many patients with no perioperative indication of infarction developed new areas of left ventricular contraction abnormality within the first postoperative year.     

The possibility of non-invasive identification of occluded coronary artery in acute myocardial infarction. A comparison of ECG and echocardiography with coronary arteriography or autopsy

In 73 patients with acute occlusion of single coronary artery, the authors assessed the possibilities of ECG and echocardiography in determining non-invasively which of the 3 main arteries had been occluded. The sensitivity of ECG for the individual arteries and particular ECG signs ranged between 30-98%, the specificity was between 86-100%. While it was always possible to determine occlusion of the left anterior descending coronary artery (LAD), in several cases it was difficult to distinguish between occlusion of the left circumflex (LCX) and the right coronary artery (RCA). The LCX occlusion is a frequent source of error in interpreting electrocardiograms of patients with fresh myocardial infarction. The sensitivity of echocardiography in identifying the occluded coronary artery ranged between 77-100%, specificity 97-100%. The following ECG and echocardiographic signs of coronary occlusion were determined. The LAD occlusion is indicated by ECG changes in V1-4, and anteroseptal and apica asynergy on echocardiography. LCX occlusion: increased R wave amplitude in V1, and lateral and posterior wall asynergy. The RCA occlusion: ECG changes in II, III, aVF, asynergy of the posterior wall and part of the septum and right ventricle. These combined signs make possible the identification the occluded coronary artery in 95% of patients with myocardial infarction.     

Differentiation of myocardial ischemia and left ventricular aneurysm in the genesis of exercise-induced ST-T changes in previous anterior myocardial infarction

We attempted to differentiate between myocardial ischemia and left ventricular asynergy as the underlying mechanisms of exercise-induced ST-segment elevation in patients with previous myocardial infarction (MI). Sixty patients with previous anterior MI, who underwent stress myocardial scintigraphy (SMS) and coronary angiography (CAG), which revealed a single vessel disease of the left anterior descending artery, were entered in this study. SMS and CAG were performed within 3 months of MI onset, and SMS and ECG were quantitatively analyzed. T wave changes to a complete upright position with concomitant ST-segment elevation (T-dominant ST-elevation) was seen in 56% of the patients with post-MI angina pectoris (N = 16) and in 50% of those with significant redistribution in SMS (n = 20). On the other hand, ST-segment elevation without T wave reversion (ST-dominant ST-elevation) was seen in 43% of patients with severe LV asynergy (akinesis and dyskinesis, n = 39) and in 50% of those with severe scintigraphic defect in delayed images (relative thallium uptake less than or equal to 40%, n = 10). When these findings were combined, T-dominant ST-elevation had sensitivity and specificity of 54% and 78%, respectively, for the diagnosis of myocardial ischemia, while the corresponding values for ST-dominant ST-elevation were 44% and 100%, for the diagnosis of severe ventricular asynergy. We conclude that the two underlying mechanisms, ischemia and asynergy, may produce different changes in ST-T shape in patients with previous myocardial infarction.     

Relation of left ventricular contractile patterns in coronary artery disease to the electrocardiogram and vectorcardiogram

Two hundred thirty patients with coronary artery disease (CAD) were studied with left ventriculography, coronary arteriography, electrocardiography (ECG) and vectorcardiography (VCG) to determine how well left ventricular (LV) contractile defects could be predicted from the ECG-VCG patterns and how this was related to the coronary disease location and severity. Of 124 patients with infarction patterns on ECG-VCG about 50% had LV contractile defects localized to the corresponding ECG-VCG abnormalities, i.e., antero-apical asynergy with anterior infarction patterns, inferior asynergy with inferior infarction patterns, or antero-apical plus inferior asynergy with anterior plus inferior patterns. About 20% in each infarction group had unexpected synergy on ventriculography except for patients with dorsal infarction patterns (synergy in 68%) who are discussed as a special problem. Another 25-30% of patients had more extensive contractile abnormality than indicated by the ECG-VCG patterns. In 106 patients with left ventricular hypertrophy, normal QRS-abnormal T and normal QRS-T on ECG-VCG, 65-70% had synergy. However, 30-35% had asynergy in various combinations not suspected from the ECG or VCG. Coronary artery disease severity was less pronounced in patients with synergy than with asynergy and single vessel disease was more common in the former, 47% versus 18-30% in the latter. However, coronary artery disease severity was the same for all ECG-VCG groups except for anterior plus inferior infarction patterns where it was most severe.     

Problems in echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence of absence of asynergy.

The relation of minor and major axes of the left ventricle was determined in 100 left ventriculograms performed in the right anterior oblique projection. This relation taken over a wide range of volumes was used to derive a theoretically correct equation for determination of ventricular volume by echocardiography. The final equation was: V =[7.0/2.4 +d] (D3), where V = volume and D = the echocardiographically measured internal dimension. In 12 patients without asynergy, this equation accurately and directly calculated end-systolic and end-diastolic volumes whether the left ventricle was small or large. However, in 12 patients exhibiting left ventricular asynergy the correlation between angiographically and echocardiographically determined volumes was poor. Thus, caution is recommended in the use of time-motion echocardiography to calculate ventricular volumes in patients with coronary artery disease and possible left ventricular asynergy.     

Correlation of abnormal Q waves, coronary pathology, and ventricular contractility.

Four hundred and ninety-two patients with coronary artery disease underwent analysis of their electrocardiograms, coronary arteriograms, and ventriculograms. Significant Q-waves were correlated with critical coronary occlusions (greater than or equal 75 per cent obstruction) and ventricular contractility. It was found that Q-waves correlate equally well with ventriculographic abnormalities and critical coronary occlusions. The Q-wave correlation varied from 77 to 87 per cent, depending on the area of myocardium under consideration, except for true posterior myocardial infarction, which correlated 55 per cent with ventriculographic abnormalities and 55 per cent with critical coronary occlusions. Significant Q-waves in Leads II, III, and aVF are better indicators of ventriculographic abnormality than in Leads III and aVF alone, whereas Q-waves in the latter two leads are more definitive than in Lead III alone. Patients who have critical coronary occlusions and normal electrocardiograms have normal ventriculograms in 71 to 78 per cent of the cases, again depending on the area of the myocardium under consideration. Thus, the normal electrocardiogram correlates better with the ventriculogram than with coronary pathology. The abnormal electrocardiogram correlates equally well with both.     

Significance of a terminal R wave in lead V1 of the electrocardiogram.

A terminal r wave in Lead V1 lower than 0.6 mV. was studied in the ECGs of four groups: (1) 104 healthy children, (2) 207 healthy young adults, (3) 171 patients with no autopsy evidence of a cardiopulmonary disease, and (4) 1,078 autopsy patients with a cardiopulmonary disease. Cases with a complete right bundle branch block were excluded. A terminal r wave occurred in 2.9 per cent healthy children, 1.4 per cent of healthy young adults, 0.6 per cent of patients without and in 5.9 per cent of patients with autopsy evidence of a cardiopulmonary disease. The occurrence of a terminal r wave was most common in pulmonary patients (10 per cent). But it was also found in patients with an anterior or a posterior myocardial infarction and in some cases of left ventricular hypertrophy. In the autopsy series RVH occurred in 57 per cent of patients with a Qr pattern, in 30 per cent of patients with a terminal r wave higher than the initial one, and in none of the patients with a terminal r wave lower than the initial one. It is concluded that the height of the terminal r wave has clinical significance. A terminal r wave higher than the initial one in Lead V1 is associated with a cardiopulmonary disease in subjects over 30 years of age, while an r wave lower than the initial one seems to be an innocent finding.     

Reversibility of left ventricular asynergy by nitroglycerin in coronary artery disease

To evaluate the potential reversibility of left ventricular asynergy in patients with coronary artery disease, pre- and postnitroglycerin left ventriculography was performed in 32 subjects. In four other subjects left ventriculography was repeated without intervention of nitroglycerin. Changes in ejection fraction and percentage of systolic shortening of three minor axes from the first to the second angiogram were then calculated. Changes were not significant for the myocardial infarction group or for the control group without the intervention of nitroglycerin. Normal left ventricles showed small but significant changes (p < 0.05). Patients with coronary artery disease but without previous myocardial infarction who demonstrated asynergy in their first angiogram showed three types of response: (1) no significant change (p > 0.05)-irreversible asynergy; (2) significant change (p < 0.025) with residual dysfunction-partially reversible asynergy; (3) significant change (p < 0.001) without residual dysfunction-completely reversible asynergy. It is concluded that postnitroglycerin ventriculography is useful in assessing the reversibility of left ventricular asynergy in patients with coronary artery disease.     

Diagnostic value of Q-waves in inferior myocardial infarction

Forty-eight patients with proved, healed, inferior myocardial infarction were studied to determine the electrocardiographic characteristics of this syndrome, the correlation between electrocardiographic abnormalities and angiographic findings, and to determine the value of recording Lead III during inspiration to identify abnormal Q-waves.The diagnosis of inferior myocardial infarction (IMI) was established by the presence of two of the following three criteria: (1) past history of acute infarction, associated with typical acute electrocardiographic changes and compatible clinical data, (2) total occlusion or more than 80 per cent occlusion of the right coronary artery, and (3) contraction abnormalities of the inferior left ventricular wall.Fifteen per cent of patients with inferior myocardial infarction had diagnostic Q-waves in all the three limb Leads II, III, and aV_F, and 29 per cent of patients had no diagnostic Q-waves in any of the three limb leads. Relative frequency of diagnostic Q-waves in inferior myocardial infarction were found to be 70, 43, and 15 per cent in Leads III, aV_F, and II, respectively.One hundred per cent correlation was noted between left ventricular inferior wall asynergy and presence of diagnostic Q-waves in all the limb Leads II, III, and aV_F, but the correlation was low (54 per cent) when none of the limb Leads II, III, and aV_F revealed diagnostic Q-waves.Obtaining Lead III in deep inspiration to differentiate an abnormal Q-wave due to inferior myocardial infarction from a benign Q-wave was not found to be a reliable measure and could result in false-negative diagnosis of inferior myocardial infarction in a significant number of patients.     

Reliability of standard electrocardiogram in detecting left ventricular asynergy in 315 patients with recent myocardial infarction

The relationship between asynergy of the left ventricular wall detected by two-dimensional echocardiography and ECG signs of necrosis (number of Q waves greater than or equal to 40 ms, Wagner's score) was evaluated in 315 patients (NYHA I-II) 23-90 days after a first Q-wave myocardial infarction (MI). Poor correlations were found between asynergy and ECG parameters. An ECG anterior MI is an apicoseptal MI by echo (independently of the ECG extent of Q waves) and the ECG is of little or no help in predicting the extent of asynergy to the inferior wall and proximal segments of the septum. An ECG inferior MI is inferoposterior by echo and the ECG has very limited value in predicting the extent of asynergy to the apex and septum. Patients with Q waves in leads II, III, and aVF had more extensive asynergy than those with either 2Q or greater than 3Q. R/S greater than or equal to 1 in V1 and/or V2 was present in 44% of patients with inferior MI while asynergy of at least one segment of the posterior wall was observed in 94%. In conclusion, standard ECG is sensitive in identifying anterior versus inferior infarct but it is unreliable in predicting the real extent of asynergy of the left ventricle, particularly in inferior infarcts.     

A natural history study of the prognostic role of coronary arteriography

Coronary cinearteriograms, clinical records, and left ventriculograms of 304 patients studied for evaluation of chest pain were reviewed. Clinical and follow-up data on survival of the normal subjects and the nonoperative group with abnormal arteriograms are presented.Ninety-two per cent of patients with typical angina pectoris had serious coronary occlusive disease. Ninety-eight per cent of patients with relatively normal coronary arteriograms survived for one to 60 or more months (mean follow-up period 24 months).There was a high mortality rate when the left main coronary artery was involved (47 per cent) and when the left coronary anterior descending branch was seriously occluded (28 per cent when arteriographic scores were high and 14 per cent when total scores were low) and a low mortality rate (0 to 7 per cent) when the LAD was normal. Mean follow-up interval in these groups was 19 months.The mortality rate was nearly three times greater when patients had QRS changes on ECG of prior myocardial infarction and six times greater when left ventricular contraction was significantly impaired.     

Left anterior descending coronary artery obstruction. Clinical, electrocardiographic, and angiographic correlates.

Seventy-six patients with severe (greater than 80%) occlusive left anterior descending coronary artery disease by coronary angiography were examined for the electrocardiographic characteristics of this disease in the presence (group A 59 patients) or the absence (group B 17 patients) of anterior wall asynergy (akinesis or dyskinesis). The incidence of clinically documented anterior myocardial infarction in these two groups of patients was examined. The collateral circulation to the left anterior descending coronary artery was also examined in the groups of patients with and without anterior wall asynergy. Thirty-eight of 59 (64%) patients with anterior wall asynergy (group A) showed electrocardiographic signs of anterior myocardial infarction, 17 per cent showed probable electrocardiographic signs of anterior myocardial infarction and 19 per cent showed no electrocardiographic signs. None of the 17 patients without anterior wall asynergy (group B) showed electrocardiographic signs of anterior myocardial infarction. In group A 74.6 per cent had documented clinical evidence of previous anterior myocardial infarction. Collateral filling of the distal left anterior descending coronary artery was seen in 71 per cent of group A and 100 per cent of group B patients. There was a significantly higher incidence (P = 0.02) of collateral filling in the patients without electrocardiographic evidence of definite anterior myocardial infarction (93% of 28 patients), than in those who showed definite electrocardiographic evidence of anterior myocardial infarction (66% of 38 patients).it is concluded that severe occlusive left anterior descending coronary artery disease with anterior wall myocardial asynergy is usually associated with electrocardiographic signs of anterior myocardial infarction, whereas equally severe left anterior descending coronary artery disease without anterior wall asynergy is rarely associated with electrocardiographic abnormalities of anterior myocardial infarction. Severe left anterior descending coronary artery obstruction without electrocardiographic and angiographic evidence of anterior myocardial infarction is usually associated with collateral circulation to the left anterior descending coronary artery and collateral circulation to the left anterior descending coronary artery is present less frequently when obstruction is associated with anterior myocardial infarction.     

Left anterior descending coronary artery obstruction. Clinical, electrocardiographic, and angiographic correlates.

Seventy-six patients with severe (greater than 80%) occlusive left anterior descending coronary artery disease by coronary angiography were examined for the electrocardiographic characteristics of this disease in the presence (group A 59 patients) or the absence (group B 17 patients) of anterior wall asynergy (akinesis or dyskinesis). The incidence of clinically documented anterior myocardial infarction in these two groups of patients was examined. The collateral circulation to the left anterior descending coronary artery was also examined in the groups of patients with and without anterior wall asynergy. Thirty-eight of 59 (64%) patients with anterior wall asynergy (group A) showed electrocardiographic signs of anterior myocardial infarction, 17 per cent showed probable electrocardiographic signs of anterior myocardial infarction and 19 per cent showed no electrocardiographic signs. None of the 17 patients without anterior wall asynergy (group B) showed electrocardiographic signs of anterior myocardial infarction. In group A 74.6 per cent had documented clinical evidence of previous anterior myocardial infarction. Collateral filling of the distal left anterior descending coronary artery was seen in 71 per cent of group A and 100 per cent of group B patients. There was a significantly higher incidence (P = 0.02) of collateral filling in the patients without electrocardiographic evidence of definite anterior myocardial infarction (93% of 28 patients), than in those who showed definite electrocardiographic evidence of anterior myocardial infarction (66% of 38 patients).it is concluded that severe occlusive left anterior descending coronary artery disease with anterior wall myocardial asynergy is usually associated with electrocardiographic signs of anterior myocardial infarction, whereas equally severe left anterior descending coronary artery disease without anterior wall asynergy is rarely associated with electrocardiographic abnormalities of anterior myocardial infarction. Severe left anterior descending coronary artery obstruction without electrocardiographic and angiographic evidence of anterior myocardial infarction is usually associated with collateral circulation to the left anterior descending coronary artery and collateral circulation to the left anterior descending coronary artery is present less frequently when obstruction is associated with anterior myocardial infarction.     

Quantitative axial oblique contrast left ventriculography: Validation of the method by demonstrating improved visualization of regional wall motion and mitral valve function with accurate volume determinations

To compare the relative merits of conventional and axial $\text{RAO}\text{LAO}$ angiography, we performed biplane left ventricular (LV) cineangiograms in 36 patients in both the conventional 30-degree $\text{RAO}\text{60}\text{-}\text{degree}$ LAO projection and in the axial 45-degree $\text{RAO}\text{60}\text{-degree}\text{LAO}\text{25}\text{-degree}$ cranial projection, in random sequence. LV volumes were calculated by a computerized system utilizing modification of the area-length method. End-diastolic volume, end-systolic volume, and ejection fraction correlated closely between the two projections (r = 0.93, 0.95, and 0.86, respectively). The axial 60-degree LAO view projected the LV apex inferiorly, the LV outflow tract superiorly, reduced LV foreshortening, and “uncovered” the LV outflow tract in all cases. Segmental wall motion abnormalities of the ventricular septum, apex, and posterior wall were better evaluated by the axial 60-degree LAO view in patients with regional asynergy in these zones. The 45-degree RAO, compared to the 30-degree RAO view, allowed a true tangential view of the mitral valve and provided a large “clear area” between the mitral valve and descending aorta, which improved the ability to quantify mitral regurgitation. Thus, axial oblique LV angiography improves evaluation of LV regional wall motion and mitral valve function without compromising LV volume quantitation.     

Postextrasystolic T wave changes and angiographic coronary disease.

The significance of postextrasystolic T wave changes in beats following induced extrasystoles was assessed by angiography in 55 patients. These T wave changes were found in 81 per cent of coronary artery disease patients but also in 68 per cent of patients with normal coronary arteries (PNS). All patients with normal baseline electrocardiograms and normal coronary arteries showed postextrasystolic T wave changes. In electrocardiographic leads corresponding to the distribution of major coronary arteries, T wave changes occurred just as frequently when the artery was normal (54%) as when the artery was stenosed (55%). Left ventricular asynergy was not associated with an increased frequency of postextrasystolic T wave changes and in fact ejection fraction was greater end-diastolic pressure lower in patients with T wave changes. Thus, postextrasystolic T wave changes appear not to be useful in diagnosing or localising coronary artery disease.     

The prognostic value of the P wave morphology in the discharge ECG in a 5-year follow-up study after myocardial infarction.

The discharge ECG's of 641 patients with acute myocardial infarction (AMI) (WHO categories "definite" and "possible" AMI) were studied to assess the prognostic value of P wave morphology as an index of left ventricular dysfunction. Of 69 patients with abnormal P terminal force (PTF), i.e., --0.03 mm.sec. or more negative, 53.6 per cent died within the next 5 years of ischemic heart disease, compared with 20.4 per cent of 558 patients with normal PTF. The odds ratio (age-corrected risk to die, Mantel-Haenszel test) was 4.1 (95 per cent confidence limits 2.4 to 7.0). The mortality curve of patients with normal PTF was linear whereas there was an abrupt rise in mortality rate during the first six months if PTF was abnormal. Of a group of 15 patients with the frontal axis of the terminal P wave --30 degrees or more negative, 8 died (Odds ratio 4.7; 1.3 to 17.1). Ten patients had atrial fibrillation, and five of them died (Odds ratio 2.; 0.5 to 12.9). In 14 cases the duration of the P wave in Lead II was 0.12 sec. but it showed no relationship to mortality (p less than 0.10). The significance of the P wave morphology on the discharge ECG to long-term survival after MI has been demonstrated. These simple ECG variables, related to left ventricular failure, can easily be put to clinical use to differentiate MI patients who are in greater risk of dying during the chronic phase.