Kazumasa Kondo

Body surface distribution of abnormally low QRST areas in patients with left ventricular hypertrophy. An index of repolarization abnormalities.

BackgroundQRST isointegral maps (I-maps) have been useful in detecting repolarization abnormalities. We investigated the body surface distribution of abnormally low QkST areas in patients with left ventricular hypertrophy (LVH) and the relation of the abnormalities in I-map to the severity of LVH as assessed by echocardiography. Methods and ResultsQRST area departure maps were constructed from electrocardiographic (ECG) data recorded in patients with LVH and precordial negative T waves resulting from aortic stenosis (AS) (10 patients), aortic regurgitation (AR) (12 patients), or hypertrophic cardiomyopathy (HCM) with asymmetric septal hypertrophy (22 patients). Fifty normal subjects served as controls. The I-map was constructed from 87 body surface electrocardiograms recorded simultaneously at a sampling interval of 1 msec. The area where the QRST area was smaller than normal limits (mean-2 SD) was designated the “-2 SD area.” The echocardiographic left ventricular (LV) mass was calculated by Devereux,s method. Patients with large LV masses due to AS or AR had 2 SD areas located over the left anterior chest or the midanterior chest, respectively. The 2 SD area was located over the left shoulder and left anterior chest and had a lingual shape ih patients with HCM. The sum of QRST area values less than the normal range (IQRST) was significantly correlated with LV mass in patients with AS or AR (r = 0.83 and r = 0.69, p < 0.01 and p < 0.05). However, there was no significant correlation between IQRST and the severity of LVH in patients with HCM. EQRST divided by the number of electrodes in the 2 SD area was significantly greater in patients with HCM than in those with AS or AR. ConclusionsThese findings suggest that abnormalities in patients with HCM are manifest even in mild LVH and that there is a greater disparity of repolarization in hypertrophied left ventricles due to HCM than in LVH due to aortic valve disease. QRST isointegral departure maps may provide ECG evidence of LV mass of patients with AS or AR and of susceptibility to malignant arrhythmias in patients with HCM.

Usefulness of QRST time-integral values of 12-lead electrocardiograms in diagnosing healed myocardial infarction complicated by left bundle branch block

The usefulness of QRST time-integral (IQRST) values of 12-lead electrocardiograms for diagnosing a prior myocardial infarction complicated by left bundle branch block (LBBB) was determined. The study consisted of 25 patients with LBBB (11 with and 14 without myocardial infarction). The IQRST values in each lead point of 12-lead electrocardiograms were calculated. Data from 607 normal subjects were used as controls and mean +/- 2 standard deviations was regarded as the normal range. The following parameters were derived: number of leads less than the normal range of IQRST values (nQRST) and sum of the differences between the normal mean IQRST value and IQRST value of a given patient in leads where this value was less than the normal range (sigma QRST). The criteria of nQRST (12-lead) greater than or equal to 5 and sigma QRST (12-lead) greater than or equal to 500 microV in 12-lead electrocardiograms were selected on a relative cumulative frequency distribution and demonstrated the presence of a myocardial infarction in LBBB with a sensitivity of 82% and a specificity of 100% for each. With regard to the localization of the myocardial infarction, the criterion of sigma QRST (V1-6) greater than or equal to 300 microV in leads V1-6 of 12-lead electrocardiograms demonstrated the presence of an anterior myocardial infarction in the LBBB with a sensitivity of 88% and a specificity of 77%. It was difficult to localize an inferior myocardial infarction in patients with LBBB by using IQRST values of inferior leads.(ABSTRACT TRUNCATED AT 250 WORDS)

Detection of myocardial infarction in the presence of Wolff-Parkinson-White syndrome by QRST isoarea map in dogs

The possibility of detecting myocardial infarction (MI) in the presence of Wolff-Parkinson-White (WPW) syndrome by means of body surface QRST isoarea maps was studied in eight dogs. Eighty-seven body surface ECGs were recorded simultaneously. Recordings were taken during right atrial (RA) and right atrial and right ventricular (RA + RV) sequential pacing, which simulated WPW syndrome, during control periods and at 1-hour intervals for up to 5 hours after occlusion of the left anterior descending coronary artery. In ECGs during the RA drive, diagnostic findings of MI such as abnormal Q waves were observed but became obscure during the RA + RV drive. On the contrary, the QRST values over the anterior chest during both drives were positive soon after coronary occlusion, decreased gradually as time passed, and became abnormally negative after 5 hours. The QRST isoarea maps during RA and RA + RV pacing showed quite similar patterns and were highly correlated with each other throughout this study (r greater than 0.95). These findings demonstrate that localized abnormalities resulting from MI are evident in QRST isoarea maps even in the presence of preexcitation and fusion.

Correlation between various parameters derived from body surface maps and ejection fraction in patients with anterior myocardial infarction

To determine the best map parameter to predict cardiac function, various map parameters were correlated with the left ventricular ejection fraction (EF) in patients with a previous (between 3 months and 1 year) anterior myocardial infarction, but without overt congestive heart failure or ventricular dyssynergy. From 300 consecutive patients with a previous myocardial infarction, 82 patients with only an anterior infarction and who underwent cardiac catheterization and body surface mapping were selected for this study. The maps from 100 healthy subjects were used as normal controls. Body surface maps using 87 unipolar electrodes were recorded and various parameters were derived from the Q map, the QRS departure maps, the QRS isointegral (IQRS) map, and the QRST isointegral (IQRST) maps. They were compared with the angiographically determined EF. The EF was correlated with nQ (r = -0.72), four parameters derived from the QRS departure map (r ranged from -0.73 to -0.79), two parameters derived from the IQRS map (r = -0.90 and -0.86), and two parameters derived from the IQRST map (r = -0.84 and -0.85). Some parameters derived from body surface maps were found to have a very high correlation with the EF in patients who had a previous anterior myocardial infarction.

Small differences among body surface and epicardial QRST integral maps recorded during normal activation and experimentally simulated left bundle branch block or preexcitation in canine hearts

QRST integral maps were constructed from 87-lead body surface electrocardiograms (ECGs) and from 45-lead epicardial electrograms during artificial pacing, which simulated left bundle branch block (LBBB) and Wolff-Parkinson-White syndrome in 12 dogs. Although the ECGs and electrograms differed in configuration for each conduction model, the body surface and the epicardial QRST integral maps showed only small differences. Correlation coefficients (r) and root mean square differences (rms) were calculated to assess quantitatively the similarities in the QRST integral maps among the different conduction models. Mean r values between the normal conduction and the left bundle branch block models were 0.95 in the body surface maps and 0.89 in the epicardial maps. Mean r values between the normal conduction and the Wolff-Parkinson-White ECG models were 0.97 in the body surface maps and 0.91 in the epicardial maps, and rms values were small enough. The small differences were also verified by the difference maps and by paired t tests. QRST integral maps on the epicardium and on the body surface were largely independent of altered activation sequences in both the left bundle branch block and the Wolff-Parkinson-White ECG models.