Background: The heart rate of healthy people is not completely regular, but varies for example with breathing. This respiratory variation is called heart rate variability (HRV) and is determined by the balance between sympatic and parasympatic innervation. HRV is decreased in diabetes, uraemia and in patients with congestive heart failure, and is a prognostic marker in these diseases. The concentration of potassium influences the excitable cells in the body, and the risk of serious tachy- and bradyarrhythmias is increased when potassium is outside the normal levels. Patients with end stage renal disease are not able to get rid of substances created by metabolism. Those substances, among them potassium, will accumulate in the body and must be removed by dialysis. The potassium levels are consequently elevated before and normal after the dialysis. The aim of this study was to evaluate the effect of varying potassium levels on HRV. Our hypothesis was that HRV would increase as the potassium levels normalized. Methods: Among the patients on regular hemodialysis treatment at the Ullevål university hospital, 26 were sequentially recruited. Heart rate (HR) and HRV was measured 3 times during a normal haemodialysis session; at the beginning, after on hour and at the end. The patients breathed deeply 6/minute for one minute and ECG was recorded the last 30 seconds. The HRV and the HR were then calculated from these recordings. For each measurement there was also taken a blood sample for analysis of the serum concentration of potassium and albumin- adjusted calcium. In addition to the patient group we also recruited a control group of 11 people among the employees of the hospital. In the control group HRV and HR was measured once, using the same method as for the patient group. No blood samples were taken from the control group. Results: The HR was significantly higher in the control group than in the patient group before the dialysis (76/min±18 versus 67/min ±24, P=0.029), but not after the dialysis (76/min±18 versus 75/min ±28, P=0.81). The HRV was significantly higher in the control group both before (0.16s±0.20 versus 0.08s±0.16, P=0.0233) and after the dialysis (0.16s±0.20 mot 0.06s±0.12, P=0.001). In the patient group, the serum levels of potassium decreased from moderately elevated levels before the dialysis (4.61mmol/l±1,8) to normal levels after the dialysis (3,56mmol/l ±0,82). There was no significant change in albumin adjusted calcium (2.36mmol/l ±0.13 versus 2.34mmol/l ±0.12, P=0.42). The HR increased significantly from before to after the dialysis (67/min±24 versus 75/min±28, p=0.001). The HRV decreased with borderline significance form before to after the dialysis (0.08s±0.16 versus 0.06s±0.12, P=0.05). However, we found no correlation neither between the baseline values nor the changes of the potassium levels and the changes in the HR or the HRV in the individual person. Conclusion: This pilot study demonstrates a significant increase in HR and a decrease in HRV of borderline significance during a haemodialysis. The observed changes in HR and HRV can be explained on the background of the changes in the extra cellular concentration of potassium during a dialysis. The lack of correlation can possibly be explained as a result of confounders of which there are several in this study.