The aim of this study is to evaluate the relationship between two different procedures for potassium removal during hemodialysis (HD) and cardiac arrhythmias. Cell excitability and the transmission of impulses may be influenced by variations of resting membrane potential (RMP). The rapid decrease of plasma potassium during the first two hours of standard HD causes a membrane hyperpolarization. A different K+ kinetic, with a gradual and constant elimination of K+ during HD, may reduce this further unphysiological aspect and its clinical consequences. This can be obtained keeping blood-dialysate K+ gradient as constant as possible with the use of a dialysate K+ concentration (Kd) decreasing during HD. Our experimental studies on various K+ intradialytic gradients seem to indicate as optimal to this purpose K+ gradients of 1.5 mEq/l at the beginning of dialysis, esponentially decreasing during treatment to Kd values of 2.5 mEq/l at the end of dialysis (variable Kd). Patients included in the trial will be submitted to two different methods of treatment with Kd 2 mEq/l and variable Kd, and to a 24 hours ECG the day of dialysis. We will compare the number of intra and interdialytic premature ventricular complexes to evaluate the impact of two different models of potassium removal on arrhythmias.