Using 12-lead ECG and synthesized VCG in detection of right ventricular hypertrophy with terminal right conduction delay versus partial right bundle branch block in the pediatric population 26th Annual ISCE Conference Zhou, S. H., Liebman, J., Dubin, A. M., Gillette, P. C., Gregg, R. E., Helfenbein, E. D., Lindauer, J. M. CHURCHILL LIVINGSTONE INC MEDICAL PUBLISHERS. 2001: 249–257

Abstract

In pediatric electrocardiogram (ECG) analysis, mild right ventricular hypertrophy (RVH) and especially mild RVH with terminal right conduction delay (RVHtcd) are often confused with partial right bundle branch block (PRBBB). This is problematic for computer ECG analysis algorithms and even for most experienced pediatric cardiologists. This study was designed to achieve better classification of mild RVHtcd and PRBBB by combining the 12-lead synthesized vectocardiogram (VCG) transverse plane measurements with scalar ECG measurements. Pediatric ECGs used in the study were recorded with 15 leads and a 500 Hz sampling rate at the Lucile Salter Packard Children's Hospital, Stanford University Medical Center. Out of 4,200 ECGs collected consecutively over a period of 18 months, 447 RVH, 335 RBBB and 589 Normal were interpreted by expert pediatric cardiologists, and were included in the study. Statistical comparison of ECG and VCG measurements were done in stratified ECG sets (412) that have a visually indistinguishable waveform pattern, 117 RVHtcd, 96 PRBBB and 199 normal, showed significant differences in initial and terminal vectors in the transverse plane. The mean angle of the initial vector was anterior (57.2 degrees +/- 41.8) in the normal group, left anterior in the PRBBB group (34.4 degrees +/- 39.5) and in the RVHtcd group (31.9 degrees +/- 41.0) and. The mean angle of the terminal vector was right anterior (158.3 degrees +/- 36.8) in the PRBBB group, rightward (179.7 degrees +/- 29.9) in the RVHtcd group and right posterior (212.6 degrees +/- 37.8) in the normal group. These are clearly applicable features for a classification algorithm. Significantly improved classification results were obtained from a new algorithm using combined ECG and VCG measurements versus an existing algorithm. The limitation of this study stems from the unavailability of a more reliable gold standard. It may be necessary to used body surface potentials obtained with a large number of electrodes to accurately differentiate the study groups.

View details for Web of Science ID 000173072300038

View details for PubMedID 11781964