|
The prevalence of an abnormal Hi-Res ECG
in normal subjects, as well as patients with previous MI with or without sustained
ventricular tachyarrhythmias, is reported in table I4. The wide range in the prevalence
of LPs may be related to differences in the diagnostic criteria for LPs, as well as in the
time of recording of the Hi-Res ECG6-8 and the site
of MI9. El-Sherif et al showed that
the incidence of an abnormal recording varied widely in the first 60 days after
MI8. An
abnormal recording 6 to 30 days after MI had the most significant relation to arrhythmic
events occurring in the first year post-MI8. There is a progressive decline in the
incidence of LPs following hospital discharge6,7. In a study by Gomes et al, LPs were
more frequent in patients with inferior MI compared to patients with anterior MI9. This
could be related to the fact that the inferoposterior segments of the left ventricle
depolarize later than the anteroseptal and anterior segments. Thus, in patients with
inferior MI, delayed regional activation is likely to outlast normal ventricular
depolarization and appear as LPs after QRS offset. On the other hand, in patients with
anterior MI the abnormal myocardial region is activated early during the QRS complex,
resulting in partial obscuring of LPs.
Risk stratification of survivors of MI has been successfully performed with time-domain
analysis of the Hi-Res ECG. Several prospective studies have confirmed the increased
likelihood of malignant ventricular tachyarrhythmias and sudden cardiac death in
post-MI patients with an abnormal Hi-Res ECG8-11. A multicenter NIH-sponsored study
was conducted to define the best predictive criteria of time-domain Hi-Res ECG in the
post-MI period10. A total of 1158 patients were recruited into the study and followed for
10±3 months. Forty-five patients (4%) suffered serious arrhythmic events. The Hi-Res ECG
was found to be independently predictive of arrhythmic events. The most predictive
continuous parameter was QRSD at 40 Hz, with the highest chi-square value of 34.5.
The best cut-off value for QRSD at 40 Hz was 120 msec. An abnormal Hi-Res ECG,
defined as QRSD at 40 Hz³120 msec, was present in 12% of the study population. The
positive, negative and total predictive accuracy of an abnormal Hi-Res ECG was 17, 98,
and 88%, respectively.
The predictive value of the Hi-Res ECG could be increased by combining its results with
other clinical data, such as left ventricular ejection fraction, degree of ventricular
ectopy, heart rate variability, or response to programmed ventricular
stimulation8,10-12. In a report from El-Sherif et al, the arrhythmic event rate in
post-MI patients was 23% when the Hi-Res ECG was abnormal, and increased to
57% when the combination of abnormal Hi-Res ECG, left ventricular dysfunction
and high-grade ectopy was present8. On the other hand, the arrhythmic event rate
was 3% when the Hi-Res ECG was normal, and 2% when all 3 markers were absent8.
Farrell et al showed that the combination of impaired heart rate variability and an
abnormal Hi-Res ECG was able to identify a small group of patients (about 10% of the
study population) who had a high incidence of arrhythmic events, with a sensitivity
of 58%, a positive predictive accuracy of 33% and a relative risk of 18.512. The Cardiac
Arrhythmia Suppression Trial (CAST) substudy also showed improvement in the
predictive accuracy of the SAECG, when it was analyzed together with left ventricular
function studies and the 24-hour ambulatory ECG10.
TABLE I–Prevalence of an abnormal Hi-Res ECG in normal subjects and in survivors of
myocardial infarction with or without malignant ventricular tachyarrhythmias
Study groups
|
Prevalence (%)
|
time-domain
|
frequency-domain
|
Normal subjects
|
0-10
|
4
|
Recent MI (<2 weeks), no VTA
|
14-29
|
26
|
Remote MI (³1 month), no VTA
|
18-33
|
23
|
Remote MI (³1 month), VTA
|
52-90
|
73-92
|
MI=myocardial infarction; VTA=ventricular tachyarrhythmias.
|
