Randomised controlled trials (RCTs) that compared short-course (up to seven days) with long-course treatment with the same antibiotic, at the same daily dose, in the treatment of acute bacterial sinusitis, in patients of all ages, were eligible for inclusion. Acute bacterial sinusitis had to be diagnosed on the basis of clinical criteria, with or without the use of additional imaging, microbiological or laboratory criteria. Detailed definitions of criteria for clinical diagnosis were reported in the paper. There had to be a difference of a minimum of two days difference between the regimes evaluated. Trials were required to evaluate at least 30 patients in each of the trial arms relevant to this review. Included trials had to report data on the following outcomes: clinical cure or improvement of symptoms and signs (primary outcome), microbiological efficacy, relapses, adverse events, or withdrawals due to adverse events. Trials with mixed patient populations were included if data for those with acute bacterial sinusitis were reported separately, or if such patients comprised over 70% of all patients in the trial.
Included trials used the following antibiotics: azithromycin, beta-lactams, fluoroquinolones, telithromycin and trimethoprim/sulfamethoxazole; a range of doses were used. Concomitant therapies (where allowed) included paracetamol, anti-inflammatory agents, cough preparations and decongestants (including antihistamines, local vasoconstrictors, oxymetazoline and prednisolone or methylprednisolone). Short-course therapies lasted between three to seven days, and long-course therapies between six and 10 days; the most common comparison was five day versus 10 day regimens. Test-of-cure visits took place between trial days 10 and and 36. All trials were in adults with uncomplicated sinusitis.
Two reviewers independently assessed the studies for inclusion; differences were resolved by discussion with two other reviewers.