Efficacy of beta-lactam monotherapy versus clindamycin or addition of metronidazole in peritonsillar abscess - an observational cohort study with a focus on Fusobacterium necrophorum.

OBJECTIVES

To compare beta-lactam monotherapy versus clindamycin or addition of metronidazole in patients with peritonsillar abscess (PTA).

METHODS

This population-based, retrospective cohort study compared treatment with beta-lactam monotherapy (reference) with clindamycin or addition of metronidazole in patients with PTA during 2013-2020 in the Skåne region, Sweden. The primary outcome was a composite of recurrent pharyngotonsillitis or PTA within 30 days. OR and average treatment effects (ATEs) were estimated using multiple logistic regression and inverse probability weighting (IPW), respectively, with 95% CIs. After IPW, patient groups were assessed by standardized mean differences and were found to be well balanced (<0.10). The primary subgroup analysis restricted the analysis to PTA caused by Fusobacterium necrophorum. Sensitivity analyses included comparing penicillin monotherapy with all other regimens and classifying treatment according to the initial empiric antibiotic, irrespective of subsequent changes.

RESULTS

A total of 889 patients were included (median age 32 years [interquartile range 21-45]; 413/889 [47%] female). F. necrophorum was the most frequently identified pathogen, detected in 270 of 889 patients (30%), followed by Streptococcus pyogenes in 229 of 889 (26%) and group C/G streptococci in 48 of 889 (5%). Of 889 patients, 392 (44%) were treated with beta-lactam monotherapy (96% penicillin) and 497 (56%) were treated with clindamycin or combination therapy with metronidazole. Recurrent pharyngotonsillitis or PTA occurred in 57 of 889 (6%) overall and in 39 of 270 (14%) in the F. necrophorum subgroup. No association between antibiotic regimen and recurrence was observed in multivariable analyses (adjusted OR 1.4 [95% CI 0.8-2.5] overall and 1.0 [95% CI 0.5-2.1] in the F. necrophorum subgroup) or IPW analyses (IPW-derived ATEs 0.02 [95% CI -0.01 to 0.05] and 0.01 [95% CI -0.07 to 0.10], respectively. Findings were robust across sensitivity analyses.

CONCLUSION

Beta-lactam monotherapy of PTA was not associated with higher recurrence rates compared with broader antibiotic therapy, including among patients with F. necrophorum infection.