Despite saturation of nitric oxide (NO) synthase (NOS) by its substrate l-arginine (Arg), oral and intravenous supplementation of Arg may enhance NO synthesis, a phenomenon known as "The l-arginine paradox". Yet, Arg is not only a source of NO, but is also a source for guanidine-methylated (N (G)) arginine derivatives which are all inhibitors of NOS activity. Therefore, Arg supplementation may not always result in enhanced NO synthesis. Concomitant synthesis of N (G)-monomethyl arginine (MMA), N (G),N (G)-dimethylarginine (asymmetric dimethylarginine, ADMA) and N (G),N (GA ')-dimethylarginine (symmetric dimethylarginine, SDMA) from supplemented Arg may outweigh and even outbalance the positive effects of Arg on NO. Another possible, yet little investigated effect of Arg supplementation may be alteration of renal function, notably the influence on the excretion of nitrite in the urine. Nitrite is the autoxidation product of NO and the major reservoir of NO in the circulation. Nitrite and Arg are reabsorbed in the proximal tubule of the nephron and this reabsorption is coupled, at least in part, to the renal carbonic anhydrase (CA) activity. In the present placebo-controlled studies, we investigated the effect of chronic oral Arg supplementation of 10 g/day for 3 or 6 months in patients suffering from peripheral arterial occlusive disease (PAOD) or coronary artery disease (CAD) on the urinary excretion of nitrite relative to nitrate. We determined the urinary nitrate-to-nitrite molar ratio (UNOxR), which is a measure of nitrite-dependent renal CA activity before and after oral intake of Arg or placebo by the patients. The UNOxR was also determined in 6 children who underwent the Arg test, i.e., intravenous infusion of Arg (0.5 g Arg/kg bodyweight) for 30 min. Arg was well tolerated by the patients of the three studies. Oral Arg supplementation increased Arg (plasma and urine) and ADMA (urine) concentrations. No appreciable changes were seen in NO (in PAOD and CAD) and prostacyclin and thromboxane synthesis (in PAOD). In the PAOD study, UNOxR did not change in the Arginine group (480 +/- A 51 vs 486 +/- A 50), but fell in the Placebo group (422 +/- A 67 vs 332 +/- A 42, P = 0.025). In the CAD study, UNOxR did not change significantly in the Arginine group (518 +/- A 77 at start vs 422 +/- A 40 after 3 months vs 399 +/- A 66 after 6 months), but fell in the Placebo group (524 +/- A 69 vs 302 +/- A 36 vs 285 +/- A 31; P = 0.025 for 0 vs 3 months). Infusion of Arg tended to decrease the UNOxR in the children (317 +/- A 41 vs 208 +/- A 16, P = 0.06). We propose that oral long-term Arg supplementation prevents loss of NO bioactivity by saving nitrite. The optimum Arg dose needs to be elaborated and is likely to be less than 10 g per day in adults. Orally and intravenously administered arginine was well tolerated by the elderly patients and young children, respectively.