|
|
-Original-
Detection of Arrhythmogenic Substrates in Prior Myocardial Infarction Patients with Complete Right Bundle Branch Block QRS Using Wavelet-Transformed ECG
Division of Cardiology, Hepatology, Geriatrics, and Integrated Medicine, Department of Internal Medicine, Graduate School of Medicine, Nippon Medical School
Background: It is important to follow up patients surviving acute myocardial infarction (MI), to detect the presence of any life-threatening arrhythmias. Various non-invasive examinations, such as signal-averaged ECG (SAECG), have been reported to predict the fatal ventricular tachycardia (VT); however, these conventional methods have limitations in detecting VT occurring in patients with complete right bundle branch block (CRBBB) QRS. Wavelet transform has been increasingly reported as a superior time-frequency analysis on the surface ECG in detecting abnormal high-frequency components (HFCs), thus suggesting abnormal myocardial conductions; however, it remains unclear whether wavelet-transformed ECG (WTECG) is useful in patients with CRBBB.
Objective: The purpose of this study is to assess the predictive value of WTECG for detecting arrhythmogenic substrates in MI patients with CRBBB.
Methods: Both the WTECG and SAECG were evaluated in 22 subjects with CRBBB, including 10 subjects without cardiovascular diseases (control group), 7 prior MI patients without VT (Non-VT group), and 5 prior MI patients with sustained VT (VT group). A 12-lead ECG (10 kHz sampling) was recorded and the representative QRS complex (300 ms) was transformed at a frequency range of 40-280 Hz using the Gabor function as the analyzing wavelet. In the power curve along a time course, the percentages of the peak power values at each frequency (60, 80, 120, 150, and 200 Hz) in the corresponding power values at 40 Hz (P60/40, P80/40, P120/40, P150/40, and P200/40, respectively) were calculated. 'The power percentages (P120/40, P150/40, or P200/40) ≥50%' was defined as an abnormal HFC (AHFC), and the number of the leads in which an AHFC was detected (NL-AHFC) of 8 leads (I, aVF, V1-V6) was counted for comparison of the two MI groups.
Results: There was no significant difference among the three groups in the SAECG recording. The power percentages of HFCs (P120/40, P150/40, and P200/40) in Non-VT group were significantly higher than those in control group (48.2 ± 36.5 vs. 30.6 ± 7.7, P<0.001; 47.8 ± 35.5 vs. 26.9 ± 7.1, P<0.001; 47.3 ± 39.4 vs. 24.9 ± 7.6, P<0.001; respectively). NL-AHFC (P150/40) in VT group significantly increased more than in Non-VT group (3.2 ± 0.4 vs. 1.4 ± 0.8, P=0.001). When 'NL-AHFC (P150/40) ≥3' was defined as abnormal, the sensitivity, specificity, positive and negative predictive values for detection of VT in MI patients with CRBBB was 100, 85.7, 83.3, and 100%, respectively.
Conclusion: WTECG might be a novel non-invasive method to detect arrhythmogenic substrates in MI patients with CRBBB.
J Nippon Med Sch 2009; 76: 291-299
Keywords
wavelet transform, high-frequency component, ventricular tachycardia, right bundle branch block, myocardial infarction
Correspondence to
Hiroshige Murata, Division of Cardiology, Department of Internal Medicine, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo 113-8603, Japan
s7086@nms.ac.jp
Received, April 1, 2009
Accepted, June 26, 2009
PDF (709K)