Background Heart failure is associated with an increased risk of ventricular arrhythmias and sudden death. It is important to identify high risk individuals in order to target therapies effectively. Previous studies have demonstrated that heart rate variability (HRV) is reduced in heart failure, and that this evidence of autonomic dysfunction has prognostic significance. QT variability index (QTVI) measures the dynamic relationship between heart rate and QT interval; an increased QTVI represents increased lability of repolarisation and correlates with sudden death. The relationship between the two measurements is not clear, and we therefore undertook preliminary studies to examine these measures in a sheep model of heart failure with particular reference to the effect of age. Methods All experiments were performed, under license, in accordance with current UK legislation. Heart failure was induced in 4 old and 4 young Welsh Mountain Sheep, by tachycardic pacing at 210 – 240 bpm via the right ventricular apex for 4 -6 weeks. Pacemaker implantation was performed under general anaesthesia (isoflurane 1-4% v/v in oxygen) with postoperative analgesia (Meloxicam 0.5mg/kg IM) and antibiotic prophylaxis (Enrofloxacin 2.5mg/kg SC). 256 second ECG recordings were made in conscious restrained animals at baseline and at experiment end. Heart rate variability was assessed using standard time domain measures and power spectrum analysis. QTVI was calculated using a modification of Berger’s algorithm as a logarithmic ratio of QT to RR variance. Results are expressed as mean change +/- SD, with significant differences detected by 2 way ANOVA at p<0.05. Results Heart failure was associated with a reduction in all time domain measures (SDNN: relative change -50.3% +/- 25.6, Triangular Index: relative change -44.7% +/- 23.3, RMSSD: relative change -60.7% +/- 40.5). Power spectrum analysis demonstrated changes in high and low frequency bands. No significant effect of age was seen for HRV measures. QTVI at baseline was lower in young animals (young: -1.68 +/- 0.60, old: -0.69 +/- 0.33); in these animals QTVI increased significantly with heart failure in the young group resulting in no difference between groups at end-stage (young: -0.77 +/- 0.27, old: -0.95 +/- 0.20). Conclusion In this model, there is heart failure induces a broad spectrum reduction in heart rate variability. This indicates the presence of autonomic dysfunction, which is common to young and old animals. On the other hand, QTVI increases only in young animals; it appears that old animals already have abnormal repolarisation lability prior to the induction of heart failure. Mechanisms for this age-dependent interaction between autonomic function and cardiac repolarisation warrant further investigation.