Abstract
BACKGROUND: Previous studies have shown conflicting results about the value of heart rate turbulence (HRT) for risk stratification of patients (pts) with chronic heart failure (CHF). We prospectively evaluated the relation between HRT and progression toward end-stage heart failure or all-cause mortality in patients with CHF. METHODS: HRT was assessed from 24-hour Holter recordings in 110 pts with CHF (54 in NYHA class II, 56 in class III-IV; left ventricular ejection fraction (LVEF) 30%+/- 10%) on optimal pharmacotherapy and quantified as turbulence onset (TO,%), turbulence slope (TS, ms/RR interval), and turbulence timing (beginning of RR sequence for calculation of TS, TT). TO > or = 0%, TS < or = 2.5 ms/RR, and TT >10 were considered abnormal. End point was development of end-stage CHF requiring heart transplantation (OHT) or all-cause mortality. RESULTS: During a follow-up of 5.8 +/- 1.3 years, 24 pts died and 10 required OHT. TO, TS, TT, and both (TO and TS) were abnormal in 35%, 50%, 30%, and 25% of all patients, respectively. Patients with at least one relatively preserved HRT parameter (TO, TS, or TT) (n = 98) had 5-year event-free rate of 83% compared to 33% of those in whom all three parameters were abnormal (n = 12). In multivariate Cox regression analysis, the most powerful predictor of end point events was heart rate variability (SDNN < 70 ms, hazard ratio (HR) 9.41, P < 0.001), followed by LVEF < or = 35% (HR 6.23), TT > or = 10 (HR 3.14), and TO > or = 0 (HR 2.54, P < 0.05). CONCLUSION: In patients with CHF on optimal pharmacotherapy, HRT can help to predict those at risk for progression toward OHT or death of all causes.