Are cephalosporins superior to penicillin for treatment of acute streptococcal pharyngitis?

Penicillin has been the " gold standard " for treatment of acute pharyngitis due to group A streptococci (GAS) (i.e., " strep throat ") for 5 decades. It is effective not only for treating acute infection but also for preventing acute rheumatic fever. GAS have remained exquisitely and uniformly susceptible to penicillin in vitro, and the drug is notable for its safety, narrow spectrum , and low cost. For these reasons, it is considered to be the agent of choice in guidelines published by the American Heart Association, the Infectious Diseases Society of America (IDSA), the American College of Physicians, and the American Academy of Pediatrics. Starting in the 1980s, there have been reports of increased bacteriologic and clinical failure rates associated with penicillin. In this issue of Clinical Infectious Diseases, Casey and Pichichero [1] present a meta-analysis of treatment of GAS tonsillo-pharyngitis in adults and conclude that the likelihood of bacteriologic and clinical failure associated with oral penicillin therapy is 2 times higher than that for oral ceph-alosporin therapy. Pichichero [2] previously published a similar meta-analysis in a study involving pediatric populations, but that analysis was severely criticized in some quarters because of methodological flaws [3]. Moreover, a careful analysis of older and more recent articles published during 1953–1979 and 1980–1993, respectively , that met scrupulous design criteria failed to find evidence of an increase in penicillin failures among patients with GAS pharyngitis [4]. Although Casey and Pichichero [1] have attempted to address many of the flaws criticized in the pediatric study, their efforts are inevitably deterred by the quality of the trials available for review. Although they excluded most of the published trials on methodological grounds, many of the 9 trials eventually included for analysis had appreciable deficiencies. For example, as they point out, 3 trials did not provide detailed descriptions of signs and symptoms required for enrollment, 3 did not provide adequate methods for monitoring treatment compliance, 2 did not perform follow-up cultures early enough to avoid confusion introduced by reacquisition, 3 did not specifically attempt to define and eliminate GAS carriers from the study, and 5 did not include typing of GAS strains recovered from patients in whom treatment apparently failed. The latter 2 deficiencies leave open the possibility that the patient groups under study may either have contained an increased number of patients with chronic strepto-coccal carriage or that the strains recovered during convalescence were …