Abnormal heart rate turbulence predicts the initiation of ventricular arrhythmias

Background: Abnormal heart rate turbulence (HRT) reflects autonomic derangements predicting all-cause mortality, yet has riot been shown to predict ventriculor arrhythmias in at-risk patients. We hypothesized that HRT at programmed ventricular stimulation (PVS) would predict arrhythmia initiation in patients with left ventriculor dysfunction. Methods: We studied 27 patients with coronary disease, left ventricular ejection fraction (LVEF) 26.7 +/- 9.1%, and plasma B-type natriuretic peptide (BNP) 461 +/- 561 pg/mL. Prior to arrhythmia induction at PVS, we measured sinus cycles after spontaneous or paced premature ventricular contractions (PVCs) for turbulence onset (TO; % cycle length change following PVC) and slope (TS; greatest slope of return to baseline cycle). T-wave alternans (TWA) was also measured during atrial pacing. Results: At PVS, abnormal TO (>= 0%) predicted inducible ventricular tachycardia (VT; n = 10 patients; P < 0.05). TO was greater in inducible than in noninducible patients (2.3 +/- 3.1% vs -0.02 +/- 2.8%, P < 0.05) and correlated with LVEF (P < 0.05) but not with BNP. TS did not differ between groups. Conversely, ambulatory HRT differed significantly from HRT at PVS (TO -0.55 +/- 1.08% vs 0.85 +/- 3.02%, P < 0.05; TS 2.63 +/- 2.09 ms/RR vs 8.70 +/- 6.56 ms/RR, P < 0.01), and did not predict inducible VT but trended (P = 0.05) to predict sustained VT on 739 179 days follow-up. TWA predicted inducible (P < 0.05) and spontaneous (P = 0.0001) VT but did not co-migrate with HRT. Conclusions: Abnormal HRT measured at PVS predicted the induction of sustained ventricular arrhythmias in patients with ischemic cardiomyopathy. However, HRT at PVS did not correlate with ambulatory HRT nor with TWA, both of which predicted spontaneous ventricular arrhythmias. Thus, HRT may reflect the influence of autonomic milieu on arrhythmic susceptibility and is likely complementary to traditional arrhythmic indices.