Background/Aim: Fasting hyperkalemia in patients with end-stage renal failure is a well-documented phenomenon. Both a decreased secretion of insulin and decreased β-adrenergic receptor sensitivity may take part in this effect. Methods: Twelve anuric, long-term (6.4 ± 2.7 years; mean ± SD) hemodialysis patients underwent three periods of 18-hour fasting (from 6 p.m. to 12 a.m.). At the beginning of each fasting period a single dose of the nonselective β-blocker, nadolol (80 mg), or the β<sub>1</sub>-selective blocker, betaxolol (20 mg), or placebo were given in a random order and in blinded fashion. The wash-out period was 7 days. Results: The mean decrease in blood pressure was similar after nadolol and betaxolol (18 ± 10 vs. 19 ± 11 mm Hg) as was a decrease in heart rate (20 ± 3 and 19 ± 6, respectively). Serum potassium was not different before each of the fasting periods. The increase in serum potassium during fasting was highly significant in each case. The mean increase in serum potassium was 1.2 ± 0.4 mmol/l after nadolol, 0.9 ± 0.6 after betaxolol and 0.6 ± 0.6 after placebo. This effect was significantly larger after nadolol than after placebo (p = 0.01), but this relation was not significant with respect to betaxolol (p = 0.30). Serum insulin as well as glucose decreased significantly and to a similar extent during each fasting period. Plasma aldosterone was unchanged. Conclusion: Nonselective β-adrenergic blockade increases the hyperkalemic effect of fasting in hemodialysis patients.