Pivmecillinam is a unique beta-lactam antimicrobial that has been used for the treatment of acute uncomplicated urinary infection for > 20 y. Since this agent was introduced, the quinolone antimicrobials have become widely used for the same indication. This study compared the efficacy of a 3-d regimen of pivmecillinam 400 mg b.i.d. with norfloxacin 400 mg b.i.d. Women aged between 18 and 65 y presenting with symptoms of acute cystitis of < 7 d duration were eligible for enrollment; 483 were randomized to receive pivmecillinam and 471 to receive norfloxacin. In each group, 30% of women had negative urine cultures prior to therapy. Bacteriologic cure at early post-therapy follow-up was achieved in 222/298 (75%) pivmecillinam patients and 276/302 (91%) norfloxacin patients [p < 0.001; 95% confidence interval (CI) 12.0-21.8]. Clinical cure/improvement at Day 4 following initiation of therapy was observed in 434/457 (95%) women who received pivmecillinam and 425/442 (96%) who received norfloxacin (p = 0.39; 95% CI 1.5-3.9). Early post-therapy (11 +/- 2 d) clinical cure was achieved in 360/437 women (82%) who received pivmecillinam and 381/433 (88%) who received norfloxacin (p = 0.019; 95% CI 0.9-10.3). In women aged < or = 50 y, early clinical cure rates were 294/351 (84%) for pivmecillinam and 299/340 (88%) for norfloxacin (p = 0.11; 95% CI 1.0-9.4). Adverse effects were similar for both regimens, and there was no evidence of the emergence of organisms of increasing resistance with therapy. Short-course therapy with norfloxacin was superior to that with pivmecillinam in terms of bacteriologic outcome, although differences in clinical outcome were less marked. In conclusion, short-course therapy with pivmecillinam is an effective empirical treatment for pre-menopausal women.