In this randomized, multicentered study, 157 males and 130 females with laboratory-confirmed, uncomplicated anogenital Neisseria gonorrhoeae infections were evaluated to determine the efficacy and safety of a single 300-mg oral dose of rosoxacin versus 3.5 g of ampicillin plus 1 g of probenecid. A total of 130 males and 101 females were evaluated. Rosoxacin cured 90.3% (P = 0.053) and 94.1% (P = 0.62), respectively, whereas ampicillin was effective in 98.5 and 98% of males and females, respectively. All 39 patients with anorectal infections were cured. One penicillinase-producing N. gonorrhoeae strain was isolated and was eradicated with rosoxacin. Of 212 pretreatment isolates tested, 201 were inhibited by 0.06 ,ug or less of rosoxacin per ml. The MICs of rosoxacin for the remaining 11 isolates ranged up to 0.5 pLg/ml. A computer-generated balance randomization chart for men and women was provided to each center. Patients were assigned, in order of entry, to receive either a single 300-mg oral dose of rosoxacin or 3.5 g of ampicillin plus 1 g of probenecid. The trial was not double blind. Cure was defined as negative cultures for N. gonorrhoeae 3 to 8 days after treatment.To assess drug toxicity pre-and posttreatment, we carried out complete blood counts, platelet counts, a multichannel 12 biochemical screen, and urinalysis. Adverse reactions were determined by direct questioning at the follow-up visit and were classified by their effect on the central nervous system (CNS), gastrointestinal (GI) system, or other body system. CNS dysfunction included dizziness, drowsiness, headache, euphoria, and altered visual perception whereas GI adverse effects included diarrhea, nausea, or vomiting.