Background: The impact of the known non-invasive cardiac indices for identification of hemodialysis (HD) patients at high-risk for sudden cardiac death (SCD) has not been evaluated.

Patients: Fifty-nine HD patients were randomized into 2 groups. Thirty of them (group A-aged 54.6±8.9 years) participated in a supervised training program with stationary bicycles during their HD sessions (3/weekly) for 10 months, while the rest (group B-aged 53.2±6.1 years) remained untrained and were used as controls.

Study hypothesis: Exercise training will affect most of the non-invasive cardiac markers of SCD and the total score positively.

Method: At entry, as well as at the end of the study all subjects underwent a cardiopulmonary exercise testing (estimation of aerobic capacity-VO2peak), an echocardiographic study (LV ejection fraction -LVEF evaluation), an ambulatory 24-hour Holter monitoring for cardiac autonomic nervous system activity evaluation by time- and frequency- domain heart rate variability (HRV) calculation (measurement of SDNN, mean RR and low to high frequency ratio-LF/HF), a signal-averaged ECG (late potentials-LP documentation) and a submaximal exercise test for microvolt-T-wave alternans (TWA) detection.

The identification of patients on hemodialysis (HD) at increased cardiac risk with powerful screening tests and the reduction of factors, which increase the probability of malignant arrhythmias are important for primary prevention of sudden cardiac death (SCD).

Study Design: randomized and controlled clinical trial. Setting & Participants: Fifty-nine HD patients were randomized into exercise training group (group A-30 patients) and controls (group B-29 patients).

Intervention: Group A participated in a 10-month supervised exercise training program during the HD sessions (3/weekly).

Outcomes: The effects of exercise on the score of the non-invasive cardiac markers for SCD risk stratification and on the occurrence of each variable separately. Statistical analysis included a two-group comparison of change scores and an analysis of covariance adjusting for baseline.

Measurements: At entry and the end of the study, in all participants aerobic capacity (VO2peak) and left ventricular ejection fraction (LVEF) were estimated, late potentials and T-wave alternans were detected and heart rate variability was calculated. Specifically, from the 24-h Holter monitoring the standard deviation of all the normal R-R intervals (SDNN), the mean R-R intervals (mean RR), the low (LF) and high (HF) frequency components of the autoregressive power spectrum of the NN intervals and their ratio (LF/HF) were recorded.

Inclusion Criteria:

• on HD for at least 6 months
• in sinus rhythm
• able to reach a sufficient workload during ergometry

Exclusion Criteria:

• with bundle branch block
• unstable hypertension
• diabetes mellitus
• severe congestive heart failure
• recent myocardial infarction
• unstable angina