To investigate the natural history of disturbances in ventricular activation, atrioventricular conduction, and ventricular cavity size, we retrospectively studied 58 patients from a total of 296 patients with dilated cardiomyopathy seen within 4 years. A total of 309 computerised electrocardiograms (ECGs) and 135 M-mode echocardiograms were analysed. In the majority of the patients, PR interval, QRS duration and QT interval prolonged progressively, though heart rate changed little. Their increase was much more striking in patients who died (n = 10) or had a pacemaker inserted (n = 9), compared to that in the clinically stable patients, though at entry all these values, as well as age and left ventricular cavity size, were similar. There were no significant differences between patients who died and those with a pacemaker inserted, except for QRS axis, which had shifted rightwards in 8 out of 10 who died, but only in 3 of 9 who subsequently had a pacemaker inserted and 14 of the 29 stable patients. A QRS duration over 160 ms was found in 8 out of the 10 patients who died, 6 of 9 who had a pacemaker and only in 5 out of the 39 stable patients (P < 0.001). The sum of PR interval and QRS duration over 375 ms was not found in any stable patient but was present in 6 of the 7 patients who were in sinus rhythm and died (P < 0.001). Left ventricular cavity size also increased with time, but did not correlate significantly with ECG progression, nor did it identify patients who subsequently died. Thus, a combination of increasing PR interval and QRS duration, particularly along with rightwards shift of QRS axis, appears to be a marker of high risk in patients with dilated cardiomyopathy.

We studied 31 patients with dilated cardiomyopathy, correlating mitral valve cusp motion with the continuous wave Doppler signal of mitral regurgitation and the signal averaged electrocardiogram (ECG). Sixteen patients had a B point (early partial closure) on the mitral echogram and 15 did not. Fifteen normal cases were used as controls. The duration of ventricular early potentials (< 40 microV) was measured on the signal averaged ECG of the QRS complex. The PR interval was increased in patients with a B point (190 (33) ms vs. 145 (16) ms in normal, P < 0.01) and PR-AC interval was reduced (25 (71) ms vs. 65 (11) ms in normal, P < 0.05). The B point itself was effectively synchronous with the onset of low velocity early systolic mitral regurgitation, and followed the Q wave of the succeeding beat by 20 ms or less. Early low velocity on mitral regurgitation was not present in patients without a B point. The duration of early potentials was greatly increased in patients with a B point (43 (26) ms) compared both to those without (17 (20) ms, P < 0.01) and to normals (12 (7) ms, P < 0.01) and their duration correlated with B-C interval (r = 0.6, P < 0.02). We conclude that a B point on the mitral echogram in patients with left ventricular disease is due to early systolic low velocity mitral regurgitation which itself results from an abnormal pattern of left ventricular activation, probably bilateral bundle branch block. Once established, this low velocity jet delays complete mitral valve closure.

OBJECTIVE:To investigate the nature of ventricular activation and its relation with mechanical events in patients with dilated cardiomyopathy.STUDY DESIGN:Retrospective and prospective study with 12 lead electrocardiograms, signal averaged electrocardiograms, and M mode and Doppler echocardiograms.SETTING:Tertiary cardiac referral centre.PATIENTS:77 patients (mean (SD) age 59(13)) with dilated cardiomyopathy, four after aortic valve replacement and three after coronary bypass surgery, and six patients with a normal sized left ventricle and complete right bundle branch block were studied. 15 normal subjects (age 45(20)) were used as controls.RESULTS:In patients with dilated cardiomyopathy, QRS duration was longer (127(25) ms v 90(10), P < 0.05) than normal and was normally distributed (r = 0.991, P < 0.01) on a normal probability plot. 20 had classic left bundle branch block, 29 intraventricular conduction delay, four right bundle branch block, and one bifascicular block. The PR interval was prolonged (185(30) ms v 150(15), P < 0.05). Electromechanical delay, Q to the onset of thickening of the interventricular septum as seen on the transverse M mode echocardiogram, was 75(15) ms in controls, but reduced to 43(15) ms in the patients (P < 0.01). Q to the onset of mitral regurgitation was also short (50(15)) ms, and correlated inversely with PR interval (r = -0.67, n = 73, P < 0.01). Early potentials (< 40 microV) were recorded on the signal averaged electrocardiogram in 33 representative patients and in all controls. Their overall duration was 30(12) ms in the patients, much longer than normal (12(7), P < 0.01)). Early potential time correlated positively with PR interval (r = 0.75, P < 0.01) and QRS duration (r = 0.60, P < 0.01) on a 12 lead electrocardiogram, and negatively with apparent electromechanical delay (r = -0.71, P < 0.01, n = 33), but not with true electromechanical delay (73(15)ms) or true PR interval (163(30)ms), calculated by correcting apparent values for early potential. The onset of left ventricular free wall motion was delayed with respect to the septum beyond 95% of the upper normal limit in all the patients with classic left bundle branch block and intraventricular conduction defect. Motion in the right ventricular free wall was delayed in 13 of 20 patients with left bundle branch block and 24 of 29 with intraventricular conduction defect by 65(20) ms, similar to that (75(10) ms) in patients with right bundle branch block.CONCLUSION:In most patients with dilated cardiomyopathy and an electrocardiographic pattern of left bundle branch block or intraventricular block, the onset of mechanical systole is strikingly and symmetrically delayed in both ventricles, compatible with bilateral bundle branch block. Complete atrioventricular block does not occur. The ventricle is activated through the upper septum and this activation is detectable only by signal averaged electrocardiography. The anatomical substrates for this abnormal activation could be the high connections described by Mahaim and Winston.