Meta-analysis of Cephalosporin Versus Penicillin Treatment of Group A Streptococcal Tonsillopharyngitis in Children

Abstract

Objective. To conduct a meta-analysis of randomized, controlled trials of cephalosporin versus penicillin treatment of group A β-hemolytic streptococcal (GABHS) tonsillopharyngitis in children. Methodology. Medline, Embase, reference lists, and abstract searches were conducted to identify randomized, controlled trials of cephalosporin versus penicillin treatment of GABHS tonsillopharyngitis in children. Trials were included if they met the following criteria: patients <18 years old, bacteriologic confirmation of GABHS tonsillopharyngitis, random assignment to antibiotic therapy of an orally administered cephalosporin or penicillin for 10 days of treatment, and assessment of bacteriologic outcome using a throat culture after therapy. Primary outcomes of interest were bacteriologic and clinical cure rates. Sensitivity analyses were performed to assess the impact of careful clinical illness descriptions, compliance monitoring, GABHS serotyping, exclusion of GABHS carriers, and timing of the test-of-cure visit. Results. Thirty-five trials involving 7125 patients were included in the meta-analysis. The overall summary odds ratio (OR) for the bacteriologic cure rate significantly favored cephalosporins compared with penicillin (OR: 3.02; 95% confidence interval [CI]: 2.49–3.67, with the individual cephalosporins [cephalexin, cefadroxil, cefuroxime, cefpodoxime, cefprozil, cefixime, ceftibuten, and cefdinir] showing superior bacteriologic cure rates). The overall summary OR for clinical cure rate was 2.33 (95% CI: 1.84–2.97), significantly favoring the same individual cephalosporins. There was a trend for diminishing bacterial cure with penicillin over time, comparing the trials published in the 1970s, 1980s, and 1990s. Sensitivity analyses for bacterial cure significantly favored cephalosporin treatment over penicillin treatment when trials were grouped as double-blind (OR: 2.31; 95% CI: 1.39–3.85), high-quality (OR: 2.50; 95% CI: 1.85–3.36) trials with well-defined clinical status (OR: 2.12; 95% CI: 1.54–2.90), with detailed compliance monitoring (OR: 2.85; 95% CI: 2.33–3.47), with GABHS serotyping (OR: 3.10; 95% CI: 2.42–3.98), with carriers eliminated (OR: 2.51; 95% CI: 1.55–4.08), and with test of cure 3 to 14 days posttreatment (OR: 3.53; 95% CI: 2.75–4.54). Analysis of comparative bacteriologic cure rates for the 3 generations of cephalosporins did not show a difference. Conclusions. This meta-analysis indicates that the likelihood of bacteriologic and clinical failure of GABHS tonsillopharyngitis is significantly less if an oral cephalosporin is prescribed, compared with oral penicillin.