The present study was undertaken to determine whether urinary excretion of aquaporin-2 (UAQP-2) is of value to diagnose the pathological state of water retention and hyponatremia. UAQP-2 under ad libitum water drinking was 429 fmol/mg creatinine in the patients with water retention, a value significantly greater than that of 153 fmol/mg creatinine in the normal subjects. An acute oral water load test (20 mL/kg BW) was performed in 7 normal subjects (22-25 yr old) and 10 patients with water retention and hyponatremia (55-75 yr old). The percent excretion of the water load was only 30% in the patient group compared with 70% in the control group (P < 0.01). In the control group, minimal urinary osmolality was as low as 131 mosmol/kg H2O, which was responsible for the decrease in plasma arginine vasopressin (AVP) levels after the reduction in plasma osmolality. In the patient group, minimal urinary osmolality was 320 mosmol/kg H2O, and free water clearance remained below 0.6 mL/min after the water load. This impaired water excretion was consistent with the nonsuppressible levels of plasma AVP despite hypoosmolality. The nadir of UAQP-2 was obtained at 60-90 min. The minimal UAQP-2 was reduced to 284 fmol/mg creatinine, a value significantly greater than that of 76 fmol/mg creatinine in the control group. Similar results were obtained in the 6 patients with hypopituitarism, who had impaired water excretion and marked hyponatremia. Water excretion was totally normalized after the replacement of hydrocortisone (excretion of water load, 31% vs. 102%; P < 0.01). Hydrocortisone replacement also significantly reduced the minimal UAQP-2 from 225 to 49 fmol/mg creatinine after the acute oral water load, a value comparable to that in the control subjects. These results indicate that UAQP-2 is a potent marker to diagnose the pathological state of impaired water excretion and hyponatremia, dependent upon AVP, in patients with water retention and hypopituitarism.