Streptococcal pharyngitis – time to move beyond penicillins?

Given the rarity of rheumatic fever in developed countries, it is probably safe to manage group A streptococcal pharyngitis without antibiotics in non-Aboriginal populations. However, until this approach has been studied, the best way to optimize antibiotic use is to swab only patients who have pharyngitis in the absence of cough or laryngitis, and treat only patients with positive results. Oral penicillin two to four times daily for 10 days is the traditional therapy for streptococcal pharyngitis. However, it is tempting to consider other antibiotics because shorter courses may be adequate, suspensions are usually more palatable than penicillin suspension and efficacy may be higher (1). The outcome of greatest relevance for treatment of streptococcal pharyngitis is the prevention of rheumatic fever and of suppurative complications. Procaine penicillin in oil containing aluminum monostearate prevented rheumatic fever in a 1950 study (2) but, given the rarity of severe complications, reported outcomes with more modern regimens are limited to clinical and bacteriological cure rates and adverse events. Drug costs also need to be considered. A 2009 systematic review of short-course antibiotics was published by the Cochrane Library (3). This review included 20 trials in which more than 11,000 patients were randomly assigned to three to six days of one or more regimens with cephalosporins (n=10), macrolides (n=11) or penicillins (n=4) versus 10 days of penicillin. The quality of many of these trials was low, mainly because there was no blinding, randomization methods were not described or patients were lost to follow-up. For clinical failure within 10 days of antibiotic completion, results favoured the short-course antibiotics (OR 0.80, 95% CI 0.67 to 0.94). With such large numbers of patients, statistically significant differences are not always of clinical significance; clinical success rates were 94% with penicillin versus 95% with short-course antibiotics. Comparing the efficacy of different short-course regimens for clinical cure is difficult because most were used in only one study. Short-course cefuroxime (four or five days) or azithromycin (20 mg/kg/day but not 10 mg/kg/day for three days) were the only regimens that had lower clinical failure rates than penicillin when analyzed independently, with ORs of 0.49 (95% CI 0.03 to 0.81) and 0.08 (95% CI 0.01 to 0.64), respectively. For the 13 trials that looked at later clinical recurrence, results were equivocal for penicillin versus short-course antibiotics. For bacteriological cure, early failure to eradicate group A streptococcus (GAS) (at the end of the course of antibiotics) was equivalent in the two groups but, again, short-course cefuroxime or the higher dose azithromycin were superior to penicillin when analyzed independently (ORs of 0.64 [95% CI 0.48 to 0.87] and 0.29 [95% CI 0.14 to 0.61], respectively). Late bacteriological recurrences were more common in the short-course group but occurred primarily with azithromycin 10 mg/kg for three days. Adverse events were mainly gastrointestinal and were almost twice as common in the short-course group (OR 1.85, 95% CI 1.55 to 2.21). Such events were particularly common with azithromycin 20 mg/kg/day (OR 5.13, 95% CI 2.76 to 9.54) and occurred in approximately 5% of the penicillin group versus 10% of the short-course group. Noncompliance was 24% in the penicillin group versus 6% in the short-course group in the six trials reporting this outcome. A meta-analysis of 10 days of a penicillin versus 10 days of a cephalosporin presented outcomes for more than 6000 children from 35 trials from 1970 to 1999 (4). The quality of many of the studies was low and most were sponsored by pharmaceutical companies (5). The OR for clinical cure (OR 2.34, 95% CI 1.85 to 2.97) and early bacteriological cure (OR 3.02, 95% CI 2.49 to 3.67) clearly favoured cephalosporins. Late bacteriological recurrences and adverse events were not compared. In absolute numbers, clinical cure rates were 94% versus 86%, while bacteriological cure rates were 93% versus 81%. When only the six trials that were double-blinded were analyzed, the benefit of cephalosporins persisted for bacteriological but not clinical cure. One criticism of this meta-analysis is that the more favourable bacteriological cure with cephalosporins may occur because they are proven to be superior at eradicating carriage (5), so falsely high cure rates could occur if carriers were inadvertently entered in trials. However, when the seven trials that attempted to exclude carriers were analyzed alone or combined with another 11 trials in which results from carriers (defined as patients with persistent GAS in the absence of symptoms) could be excluded, the benefits of cephalosporins for clinical and early bacteriological cure persisted. The only relevant study published since this meta-analysis had more than 400 patients per arm, with the cephalosporin again providing higher rates of clinical cure (97% versus 88%) and bacteriological cure (94% versus 71%) (6). To summarize, the clinical cure rate for short-course therapy is marginally superior to 10 days of penicillin. Downsides include increased cost for most regimens, approximately a one in 20 risk of nonserious adverse events on top of the risk associated with 10 days of penicillin and, possibly, an increased risk of late bacteriological recurrence. Benefit was more impressive with 10 days of a cephalosporin, yet one would still have to treat 12.5 patients to achieve improved clinical cure and eight patients to achieve improved early bacteriological cure in one patient, keeping in mind that the data are muddied by the lack of high-quality studies (most of which were sponsored by a pharmaceutical company) and by questions about how GAS carriers were excluded from studies, such that the numbers needed to treat may actually be higher than calculated. Despite a relatively high rate of noncompliance, 10 days of penicillin twice daily still results in excellent clinical outcomes more than 90% of the time. I would continue to use this as the regimen of choice given the satisfactory rate of clinical cure, the increased cost of other options and the long-term consequences to society associated with use of broad-spectrum antibiotics. However, it is also acceptable to use a first-generation cephalosporin, especially with frequently recurrent cases in which eradicating GAS might confer benefit. Shorter courses of appropriate antibiotics are attractive if compliance is a major issue, keeping in mind that the cost and risk of adverse events with any given choice are likely to be higher than with 10 days of penicillin, and that there is increasing concern about macrolide resistance in Canada (7). Much of this resistance arose after the trials included in the Cochrane review (1). A compromise may be to use once daily amoxicillin for 10 days as the penicillin of choice when compliance is an issue (8) because it is less broad spectrum and less expensive than most non-penicillin regimens.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.