Approximation of electrocardiogram T wave by using Gaussian mixtures and automatic measurement of QT interval

Abstract

Results: Active fixation leads were used in 120 (90%) patients, and passive fixation leads were used in 14 (10%) patients. Local injury current was observed in 121 events (28.8%) and was detected more frequently after appropriate (43 [60.6%] events) and inappropriate (56 [64.4%] events) ICD shocks, as compared with appropriate (8 [9.2%] events) and inappropriate (3 [4.7%] events) antitachycardia pacing (ATP) or nonsustained ventricular tachycardia (11 [9.9%] events; analysis of variance, P ≤ .0001]. Appearance of LIC was a changeable phenomenon; that is, LIC was frequently observed after one, but not another, appropriate ICD shock in the same patient. Significant LIC was observed after 61% of events with single ICD shock, 83% of events with 2 shocks, 50% of events with 3 shocks, and 100% of events with 4 or more shocks (analysis of variance, P = .658). Type of ICD therapy (ICD shock vs ATP) was the most significant predictor of LIC (ATP:β-coefficient =−0.81; 95%confidence interval [CI], −1.19 to 0.44; P ≤ .0001), along with cycle length of tachycardia (β-coefficient = −0.0117; 95% CI, −0.0167 to −0.0068; P≤; .00001) and shock energy (β-coefficient = 0.024; 95% CI, 0.003–0.045; P = .025). Conclusion: Appropriate and inappropriate ICD shocks are frequently characterized by the development of LIC in patients with structural heart disease. Type of electrical ICD therapy, shock energy, and cycle length of ventricular arrhythmia are important determinants of LIC.