Renal function, plasma renin activity, plasma aldosterone concentration and urine excretion of free norepinephrine were evaluated in 13 cirrhotics without previous or ongoing ascites and in 13 healthy subjects, after 6 days of controlled electrolyte intake (40 mmol of Na and 70 mmol of K per day) and during 24 h of recumbency. Plasma concentrations of the atrial natriuretic peptide (ANP) were also measured in 8 patients and 8 controls. Despite a low-normal filtered load of sodium (14.6 +/- 1.2 vs. 17.1 +/- 1.2 mmol/min), cirrhotic patients showed supernormal natriuresis (141.5 +/- 14.1 vs. 78.8 +/- 8.6 mmol/day; p < 0.001). Whereas the fractional excretion of sodium in these patients was twice that of controls (0.70 +/- 0.05 vs. 0.36 +/- 0.04%; p < 0.001), potassium excretion (42.5 +/- 2.7 vs. 43.1 +/- 2.7 mmol/day) and urine volume (1270 +/- 98 vs. 1452 +/- 148 ml/day) did not differ. In cirrhotics, plasma renin activity was reduced (0.50 +/- 0.12 vs. 1.39 +/- 0.33 ng/ml/h; p < 0.02), and plasma aldosterone concentration tended to be lower (66 +/- 10 vs. 86 +/- 9 pg/ml, p 0.09), while urine norepinephrine excretion did not significantly differ from controls (961 +/- 120 vs. 782 +/- 43 ng/h). ANP was higher in patients than in controls (92 +/- 17 vs. 48 +/- 9 pg/ml; p < 0.05). Natriuresis was directly correlated with AN P (r = 0.69, p < 0.005) and ANP/plasma aldosterone ratio (r = 0.63; p < 0.01) in patients and healthy subjects taken together, No relationships were found between renal function and urine excretion of norepinephrine in either controls or cirrhotic patients. In cirrhotic patients without ascites bed rest thus causes an exaggerated natriuresis, which might be linked to ANP hypersecretion in the presence of low-normal plasma aldosterone concentrations. This phenomenon may be primed by the central redistribution of an expanded blood volume. Bed-rest-induced hypernatriuresis may be the natural event whereby compensated cirrhotic patients succeed in preventing ascites.