Many of the respiratory, urinary, and skin infections that were traditionally treated in the hospital setting are now managed with home-based care or in a skilled nursing facility. One of the most widely used medications for this purpose is intravenously administered ceftriaxone, because of its convenient once-daily dosing and broad coverage of common pathogens. Grayson and associates conducted a randomized, double-blind, home-based study to compare a once-daily first-generation cephalosporin (cefazolin) plus once-daily probenecid with once-daily ceftriaxone plus oral placebo in the treatment of moderate to severe cellulitis.

Patients were enrolled if they had cellulitis that required intravenous antibiotic treatment (severe cellulitis, the presence of sepsis, or failure of previous antibiotic therapy) and who were suitable for home-based therapy. A total of 132 patients were randomized to receive either once-daily cefazolin (2 g, intravenous) plus probenecid (1 g, oral) or once-daily ceftriaxone (1 g, intravenous) plus placebo (oral). Probenecid is known to reduce the excretion of cefazolin.

The median number of daily doses needed to clear the cellulitis was similar (seven doses of cefazolin versus six of ceftriaxone). At the end of therapy, 86 percent of patients receiving cefazolin achieved clinical cure compared with 96 percent of patients treated with ceftriaxone; however, this difference was not statistically significant. Cure rates at one-month follow-up were higher in the cefazolin group than in the ceftriaxone patients.

Eighteen patients were excluded after assessments because their clinical outcomes were indeterminate. More patients in the cefazolin-probenecid arm of the study experienced nausea and vomiting (16 percent) compared with ceftriaxone patients (4 percent).

The authors conclude that once-daily intravenous cefazolin plus oral probenecid is a less expensive, more narrow-spectrum treatment option for cellulitis with similar clinical results when compared with standard therapy using ceftriaxone.