The interest of ascorbic acid - through urinary acidification - for the prevention of urinary infections in intensive care units was assessed within one year in a prospective, double blind and randomized study. Patients without urinary tract infection on admission received orally 1 g/d of ascorbic acid, versus placebo. Urinary infection criteria were: with indwelling catheter bacteriuria greater-than-or-equal-to 10(4)/ml, without indwelling catheter leukocyturia greater-than-or-equal-to 10(4)/ml and bacteriuria greater-than-or-equal-to 10(4)/ml, or bacteriuria greater-than-or-equal-to 10(5)/ml. The overall incidence was 19 infections in 104 patients (18,3 %). All patients were comparable for age (62 +/- 14 yrs) or SAPS (14 +/- 4). The incidence of urinary infection, acid urine (pH less-than-or-equal-to 5,5) and alkaline urine (pH greater-than-or-equal-to 7) were respectively 17,3 %, 51,3 % and 12,8 % for the treated patients, versus 19,3 % (not significant), 43,7 % (not significant) and 31,3 % (p < 0,05) for controls. Neither female gender, diabetes nor urethral catheterization increased the incidence of urinary infection. The incidence was lower in patients receiving antibiotics (p < 0,01). In patients undergoing indwelling catheter > 5 days and broad spectrum antibiotherapy, the incidence of urinary infection was 2,9 % in the acid urine group versus 25,7 % in controls (p < 0,005). This study showed a decresed incidence of urinary infection in patients with acid urine. Ascorbic acid prevented urine alkalinization, but did not reach sufficient acidification in this study.