Background: Upper respiratory tract infection is the third leading reason for physician visits in the United States, and about one third of these visits are attributed to acute rhinosinusitis. Antibiotics are prescribed to about 80 percent of patients with rhinosinusitis, despite clinical guidelines that recommend restricting antibiotics to those who have symptoms that persist for seven to 10 days. Young and colleagues investigated the possibility of using clinical signs and symptoms to reliably identify patients with acute rhinosinusitis in whom antibiotic therapy is justified.

The Study: The investigators requested original patient data from all known randomized controlled trials (RCTs) of antibiotic therapy for rhinosinusitis. These RCTs were identified by searching databases of clinical studies, literature review, and examining the references of pertinent clinical articles. Trials were included if patient selection was primarily based on clinical signs and symptoms consistent with rhinosinusitis. The patient data were entered into a meta-analysis, with the primary outcome of proportion of patients cured (as defined by individual RCTs).

Results: Patient data were provided for nine of the 10 identified RCTs. The studies ranged in size from 135 to 503 patients, and the median age was 35 years. The antibiotics used were pivampicillin, doxycycline (Vibramycin), azithromycin (Zithromax), amoxicillin, phenoxymethyl-penicillin, and amoxicillin/clavulanic acid (Augmentin). The percentage cured following antibiotic therapy ranged from 40 to 82 percent compared with 30 to 68 percent with placebo. In the overall analysis of data from 2,547 patients, the odds ratio for treatment effect of antibiotics relative to placebo was 1.35. This did not change significantly in analyses that excluded scores with the highest or lowest reported cure rates. The overall estimated number needed to treat (NNT) for one additional cure was 14, with a range of 9 to 30. In subgroup analyses, patients with purulent pharyngeal discharge had longer time to cure and a lower estimated NNT of eight. Other patient groups with longer time to cure included older patients and those with more severe or more prolonged symptoms; however, no evidence was found of additional benefit from antibiotic therapy (reduced NNT) in these patients.

Conclusion: The authors conclude that antibiotics confer little benefit in the treatment of uncomplicated rhino-sinusitis and are not justified, even if symptoms have persisted for more than seven to 10 days or are severe. Common clinical signs and symptoms do not identify patients in whom antibiotic therapy is justified. The authors recommend watchful waiting and symptomatic relief for adult patients with functioning immune systems, but caution that severe symptoms, such as high fever, periorbital swelling, erythema or intense facial pain, may indicate development of a complication that requires aggressive therapy.