Clostridium difficile--Associated Disease Treatment Response Depends on Definition of Cure

To The Editor—We commend Zar et al. [1] for their recent study comparing oral metronidazole with oral vancomycin for first line therapy of Clostridium difficile–associated disease (CDAD). In this era, in which the incidence and severity of CDAD are increasing [2], the timing could not be better to readdress this important clinical question. Recent observational data have called into question the presumed equivalence of these 2 drugs for the initial treatment of CDAD [3, 4], which was originally established by nonblinded but prospective randomized trials conducted in the 1980s and early 1990s [5, 6]. Zar et al. [1] addressed the lack of prior study blinding and reported on a randomized, double-blinded trial completed in 2002 that involved >150 patients and showed a difference in outcome among patients with severe CDAD. However, we believe that the definition used for cure may compromise the validity of their conclusion that vancomycin is superior to metronidazole in treating individuals with severe disease but not in treating those with mild CDAD. Zar et al. [1] used an aggregate clinical and microbiological end point to assess efficacy, which consisted of diarrhea resolution by the sixth day of treatment, in addition to clearance of C. difficile toxin from stool specimens obtained on the sixth and tenth days of treatment [1]. By this definition, detection of C. difficile toxin in the day 6 stool specimen would constitute evidence of treatment failure, despite diarrhea resolution. However, most clinicians and previous clinical trials have used only clinical criteria for determining the success of anti–C. difficile treatment. Persistence of C. difficile toxin in stool specimens despite diarrhea resolution is a well-documented phenomenon; however, it is of uncertain clinical significance [6]. A previous study also showed that vancomycin was associated with higher rates of toxin clearance than was metronidazole, but vancomycin was not associated with a lower risk of recurrence [7]. Because of its lack of prognostic value, stool toxin testing after resolution of diarrhea is not recommended for a “test of cure” [8]. Expert opinion supports the use of enterally administered vancomycin over the use of metronidazole in treating patients with severe disease on the basis of higher intracolonic drug concentrations, lower risk of bacterial resistance, and, possibly, faster clinical response [9]. Although it is not clear how the addition of toxin testing to the cure definition could selectively favor vancomycin over metronidazole for patients with severe CDAD but not for patients with mild CDAD, we would, nonetheless, be keenly interested in seeing a reanalysis of the data using a definition of cure based on clinical response alone, to make a more uniform comparison with prior CDAD treatment trials. Potential conflicts of interest.E.R.D. is on the speaker's bureau of Viropharma and the advisory board of Genzyme and Salix. S.J. has served as a consultant to Viropharma, Salix, Genzyme, Replidyne, Romark, and Acambis. D.N.G. holds patents for the treatment and prevention of C. difficile disease licensed to ViroPharma; has been a consultant for Salix, Romark, ViroPharma, Genzyme, Astra-Zeneca, Optimer, GOJO, and Oscient; and has received research support from ViroPharma, Genzyme, Optimer, and Massachusetts Biological Laboratories. S.J.L.: no conflicts.