Beat-to-beat electrocardiographic morphology variation in healed myocardial infarction

Shlomo A. Ben-Haim, Bruno Becker, Yeouda Edoute, Mira Kochanovski, Orly Azaria, Elieser Kaplinsky, Yoram Palti

Research output: Contribution to journalArticlepeer-review


Using high-fidelity electrocardiographic (ECG) amplifiers, we measured subtle beat-to-beat ECG morphologic variations at different phases of the ECG complex. The electrocardiograms were recorded from 49 men with a documented Q-wave myocardial infarction and from 30 age-matched normal men. Forty consecutive beats were averaged to achieve an average ECG signal from which variance could be calculated. The relative variance, defined as the ratio between the integrated variance of the examined window and the integrated variance of the ECG signal that was close to full cycle length, was calculated at QRS onset and at offset in 2 frequency bands (4 to 40 and 60 to 120 Hz). Patients with healed infarction had a relative variance of 2.1 ± 0.5 (mean ± standard deviation [SD]) at QRS offset (a window of 40 ms), which was significantly lower than that of the healthy volunteers: 2.5 ± 0.33 (mean ± SD; p < 0.02) at the low-frequency band. At the high-frequency band, patients with healed infarction had a significantly higher relative variance than the control subjects at QRS onset: 1.95 ± 0.58 vs 1.55 ± 0.35 (mean ± SD; p < 0.005). A model based on the numerous minor conduction abnormalities that exist in the chronically ischemic myocardium is presented to explain the changes in variance at the onset and offset of the QRS. The variance changes described can eventually serve as quantitative indexes of myocardial injury and electrical stability in patients with ischemic heart disease.

Original languageEnglish
Pages (from-to)725-728
Number of pages4
JournalAmerican Journal of Cardiology
Issue number8
StatePublished - 15 Sep 1991
Externally publishedYes


Dive into the research topics of 'Beat-to-beat electrocardiographic morphology variation in healed myocardial infarction'. Together they form a unique fingerprint.

Cite this