It has becoming clear that angiotensin receptor blockers (ARBs) show varying levels of angiotensin II type 1 (AT1) receptor blocking activity. Although the duration of activity and the efficacy on blood pressure of ARB are reported to vary, depending on the agents used, it has not been examined whether the effects on proteinuria and urinary nitrite/nitrate (NOx) excretion differ in hypertensive patients with chronic renal disease. In the present study, patients with hypertension (> 140 and/or 90 mmHg) and chronic renal disease (proteinuria > 0.5g/day; serum creatinine < 265 micromol/l or creatinine clearance > 30 ml/min/1.72 m2) were randomly assigned to perindopril- (n = 15), trandolapril- (n = 15), candesartan- (n = 17), and losartan-treated groups (n = 15), and were followed up for 96 weeks. All agents decreased blood pressure to the same level, and none of them had any effect on creatinine clearance. Candesartan, perindopril, and trandolapril reduced proteinuria markedly (from 3.0 +/- 0.6 to 1.8 +/- 0.5 g/day, 2.7 +/- 0.5 to 1.6 +/- 0.4 g/day, and 2.7 +/- 0.5 to 1.7 +/- 0.4 g/day, respectively) at 12 weeks, and the beneficial effect persisted throughout the study. The effect of losartan, however, diminished over the study period. Whereas perindopril, trandolapril, and candesartan markedly increased urinary NOx excretion (from 257 +/- 23 to 1,011 +/- 150 micromol/day, 265 +/- 70 to 986 +/- 130 micromol/day, and 260 +/- 62 to 967 +/- 67 micromol/day at 12 weeks, respectively), a relatively blunted increase was observed with losartan (from 309 +/- 42 to 596 +/- 64 micromol/day). In conclusion, renal action of ARB varies, with relatively less proteinuria-sparing, as well as NOx-enhancing, effects observed with candesartan showing the greatest reduction of proteinuria and greatest enhancement of NOx. Furthermore, renal nitric oxide may contribute to the renal protective action of these agents when administered to patients with chronic renal disease.