Abstract
BACKGROUND: Previous studies have described the clinical usefulness of the presence of nonsustained ventricular tachycardia (NSVT) and defined heart rate turbulence (HRT) in stratifying patients at risk. We prospectively assessed whether HRT can facilitate the predictive accuracy of NSVT for identifying patients at risk for serious arrhythmic events in patients with left ventricular (LV) dysfunction. METHODS: We enrolled 299 consecutive patients with LV dysfunction (ejection fraction ≤ 40%) including ischemic (n = 184) and nonischemic causes (n = 115). The presence of NSVT was assessed on Holter electrocardiograms (ECGs). HRT was simultaneously measured from Holter ECGs, assessing two parameters: turbulence onset (TO) and turbulence slope (TS). HRT was considered positive when both TO and TS were abnormal. The end point was defined as of sudden cardiac death (SCD) and sustained ventricular tachyarrhythmias (VTs). RESULTS: NSVT was documented in 93 patients (32.7%). For HRT assessment, 17 patients (5.6%) were not utilized. Of 282 patients, 68 (24.1%) were HRT positive. During follow-up of 960 ± 444 days, 14 patients (5.0%) reached the end point. NSVT, HRT, and diabetes were significantly associated with the end point. On multivariate analysis, NSVT had the strongest value for the end point, with an HR of 4.4 (95%CI, 1.4-14.3; P = 0.0138). When NSVT combined with HRT, the predictive accuracy is more increased, with an HR of 8.2 (95%CI, 2.9-23.3; P < 0.0001). The predictive values of the combination were higher than single use of NSVT or HRT. CONCLUSIONS: HRT can facilitate the predictive accuracy of NSVT for identifying patients at risk for serious arrhythmic events in patients with LV dysfunction.