Randomised controlled trial of nitrofurantoin versus placebo in the treatment of uncomplicated urinary tract infection in adult women.

BACKGROUND Urinary tract infections (UTIs) are very common and have been treated with apparent success with antimicrobials for many years. However, there is a paucity of placebo-controlled clinical trials. AIM To measure the symptomatic and bacteriological short-term effect of nitrofurantoin treatment versus placebo, in the treatment of uncomplicated UTI in adult non-pregnant women. DESIGN OF STUDY Randomised placebo-controlled trial in general practice. SETTING Non-pregnant women, aged between 15 and 54 years old, consulting a general practitioner for symtoms suggestive of uncomplicated lower UTI and with pyuria (positive for leucocyte esterase test). METHOD A dipslide was inoculated in first-void midstream urine and sent for examinion. The patients were randomised to receive nitrofurantoin 100 mg or placebo four times daily for three days. After three, seven, and 14 days a new dipslide was inoculated and symptoms of UTI were checked or improvement of symptoms and bacteriuria. RESULTS Of 166 women consulting with symptoms suggestive for UTI, 78 had pyuia and agreed to participate in the study (the clinically suspected UTI group); of these, 40 received nitrofurantoin and 38 received placebo. The result for combined symptomatic improvement and cure after three days was 27/35 in the nitrofurantoin group and 19/35 in the placebo group (c2 with Yates' correction P = 0.008; number needed to treat [NNT] = 4.4, 95% confidence interval [CI] = 2.3 to 79). After seven days, combined improvement and cure was observed in 30/34 and 17/33 respectively (P = 0.003, NNT = 2.7, 95% CI = 1.8 to 6.0). At inclusion, 56 women had bacteriuria of > or = 10(5) CFU/ml (the bacteriologically proven UTI group). Of these, 29 received nitrofurantoin and 27 received placebo. After three days the bacteriological cure was 21/26 in the treatment group, compared with 5/25 in the placebo group (P < 0.001; NNT = 1.6, 95% CI= 1.2 to 2.6). After seven days the bacteriological cure rate was 17/23 in the intervention group and 9/22 in the placebo group (P = 0.05, NNT = 3, 95% CI = 1.7 to 17). CONCLUSION In women with bacteriologically proven UTI, nitrofurantoin was significantly more effective than placebo in achieving bacteriological cure and symptomatic relief in just three days; this was still present after seven days. In patients with clinically suspected UTI the symptomatic effect was statistically significant after