Community-acquired pneumonia remains an important infectious disease problem, with more than 4 million cases occurring in the United States annually. Although Streptococcus pneumoniae remains the most commonly identified organism, a variety of bacterial and nonbacterial pathogens may be involved. Hospitalization is unnecessary in most cases, and oral antibiotic therapy is common. In the majority of cases, the etiology of pneumonia is unknown at the time of presentation, necessitating the use of empiric therapy. Quinolones have not been utilized in this setting in the past because of their inconsistent coverage of S pneumoniae. Sparfloxacin (RP 64206) is a broad-spectrum fluoroquinolone with excellent activity in vitro against the majority of bacteria involved in community-acquired pneumonia, including pneumococcus. We therefore studied the efficacy and safety of sparfloxacin compared with the second-generation cephalosporin cefaclor as empiric therapy for patients with community-acquired pneumonia in a double-masked, double-dummy, multicenter trial. Three hundred thirty patients aged 18 years or older with community-acquired pneumonia suspected of being bacterial in etiology were enrolled at 74 centers in the United States from June 1, 1992, to March 4, 1995. Patients meeting the inclusion criteria were randomized to receive 10 days of either sparfloxacin 400 mg orally once followed by sparfloxacin 200 mg orally daily (n = 168), or cefaclor 500 mg orally every 8 hours (n = 162). There were no significant differences between groups with regard to baseline characteristics. Patients were followed up serially at 4 +/- 1 days, 20 +/- 3 days, and 38 +/- 7 days after the beginning of therapy. Patients were evaluated for clinical response, clinical recurrence of infection, and eradication of baseline pathogens. The primary efficacy variable was the clinical response (cured or improved) in the subgroup of patients meeting the definition of clinically assessable. Responses were also evaluated in the intent-to-treat population. In the intent-to-treat population, 35.7% of patients receiving sparfloxacin were clinically cured, compared with 32.1% of patients receiving cefaclor. Clinical successes (patients clinically cured plus improved) were also comparable (72.6% of patients in the sparfloxacin group and 71.0% of patients in the cefaclor group). Similar clinical success rates were noted using only the clinically assessable population (primary efficacy variable). Forty-four percent of patients receiving sparfloxacin and 39.1% of patients receiving cefaclor were clinically cured. In the sparfloxacin group, 86.6% of patients were clinical successes, compared with 84.4% of patients in the cefaclor group. Microbiologic cures were comparable in both groups. There was no difference in the incidence of recurrence of infection or superinfection. Adverse events thought to be due to study drug occurred equally in both groups (14.3% in the sparfloxacin group vs 14.8% in the cefaclor group). Results show that sparfloxacin is a safe and effective empiric therapy for patients with community-acquired pneumonia and is comparable to cefaclor.