Ascitic cirrhotic patients are a heterogenous population with respect to factors that may affect plasma human atrial natriuretic peptide levels (such as degree of plasma volume and plasma levels of angiotensin II, vasopressin and norepinephrine). Thus the proven variability of plasma human atrial natriuretic peptide values in ascitic cirrhotic patients may be due also to the selection of patients, not only to the study conditions. The response to standardized stepped-care medical treatment of ascites makes it possible to characterize ascitic cirrhotic patients with different patterns of renal sodium excretion, intrarenal sodium handling, plasma renin activity, plasma aldosterone and thus, probably, effective circulating volume. Consequently, we evaluated human atrial natriuretic peptide plasma levels in controls (n = 23), in ascitic cirrhotic patients who underwent spontaneous diuresis (group A, n = 7) and in cirrhotic patients who required diuretic treatment (group B, n = 44). The last group was then divided into two subgroups. Subgroup B-R (n = 25) included patients who responded to spironolactone alone, whereas subgroup B-NR (n = 19) included patients who did not respond to 500 mg/day spironolactone. All patients were maintained on identical normocaloric restricted sodium intake (80 mEq/day) throughout the study. Ascitic cirrhotic patients, as a whole, had higher values of human atrial natriuretic peptide than did controls (70.8 +/- 46.6 pg/ml vs. 41.7 +/- 16.3 pg/ml, p < 0.025). No difference was found in human atrial natriuretic peptide/plasma renin activity between the two groups (87 +/- 160 pg/ng/hr vs. 44 +/- 73 pg/ng/hr, p = NS).(ABSTRACT TRUNCATED AT 250 WORDS)