Value of serial heart rate variability measurement for prediction of appropriate ICD discharge in patients with heart failure

Introduction: Decreased heart rate variability (HRV) is associated with adverse outcomes in patients with heart failure. Our objective was to examine whether decreased HRV predicts appropriate implantable cardioverter defibrillator (ICD) shocks.

Methods and results: In 105 patients with a Boston Scientific Contak Renewal, Cognis or Energen device implanted for either primary (73.3%) or secondary prevention (26.7%), time domain HRV variables standard deviation of averages of normal beat-to-beat interval (SDANN) and footprint percentage (FPP) were collected at baseline and during follow-up. In case of appropriate shock, HRV before shock was assessed. Using time-dependent Cox regression models, the relation between median-based dichotomized SDANN or FFP and appropriate shock was investigated. Baseline characteristics between patients with or without shocks were similar, with exception of secondary prevention patients using more often antiarrhythmic drugs. During follow-up (median 451, IQR 202-1,460 days), appropriate shocks occurred in 20 (19%) patients. SDANN and FPP did not differ significantly at baseline between patients with or without shocks (respectively, P = 0.18 and P = 0.78). However, time-dependent Cox regression analysis showed a trend that patients were at lower risk for appropriate shock (SDANN: HR 0.43, 95% CI [0.18-1.05], P = 0.06 and FPP: HR 0.49, 95% CI [0.20-1.20], P = 0.12) when HRV values were above median baseline value during follow-up.

Conclusions: These results imply that HRV could be an independent predictor for appropriate shocks. Therefore, low HRV could be of additional use in predicting imminent appropriate shocks and could possibly guide concomitant medical therapy.