Background: Acute rhinosinusitis is a common illness that is almost always self-limited. Although physicians often treat rhinosinusitis with antibiotics, most patients will not benefit from this treatment. Previous studies have suggested that a few patients with bacterial sinusitis may improve more quickly with antibiotics. Diagnostic tests for bacterial sinusitis, such as sinus puncture and computed tomography, are not routinely performed in primary care practices, so it would be useful to identify clinical findings that would be associated with a positive response to antibiotics. De Sutter and colleagues studied the clinical signs and symptoms of rhino-sinusitis, the usefulness of sinus radiography to predict the duration of illness, and the effect of amoxicillin treatment.

The Study: The authors conducted a secondary analysis of a randomized controlled trial of amoxicillin versus placebo for patients with acute rhinosinusitis at family physician practices in Belgium. The study population consisted of 300 patients who were 12 years or older and presented with a respiratory tract infection, self-reported purulent rhinorrhea, and at least one sign or symptom suggestive of acute rhinosinusitis (seeaccompanying table). Patients were ineligible if they reported a penicillin allergy, recent antibiotic use, or illness lasting more than 30 days; or had signs of complicated sinusitis or immune dysfunction. Enrolled patients were randomized to amoxicillin treatment or placebo for 10 days. Radiography also was performed on 218 of the participants.

All patients completed a baseline questionnaire that included ratings of the severity of their symptoms and general feeling of illness. Patients reported new ratings in a diary for 10 days; those who had not recovered by day 10 were asked to continue their diaries for five more days. Analyses were conducted to identify independent associations between baseline variables and time to clinical recovery or response to amoxicillin treatment.

Results: At the end of the study, 19 patients continued to have symptoms, and 34 patients had discontinued their diaries before reporting recovery. The remaining 247 patients, whose duration of illness was known, reported recovery within the study period. Those who reported a general feeling of illness or reduced productivity at baseline also took longer to recover regardless of treatment with amoxicillin. No clinical sign, symptom, or radiographic finding had any relationship to the patient's duration of illness or to any benefit from amoxicillin.

Conclusion: Initial clinical signs, symptoms, and sinus radiography have no prognostic value for patients with suspected acute rhinosinusitis. This information also cannot be used to guide the selection of patients who would benefit from antibiotic treatment. Consequently, the authors recommend providing symptomatic treatment rather than antibiotics to patients with uncomplicated rhinosinusitis.

Source: De Sutter A, et al. Predicting prognosis and effect of antibiotic treatment in rhinosinusitis. Ann Fam Med. November/December 2006;4:486–93.

editor's note:Although many randomized controlled trials in the primary care setting have found no benefit from the use of antibiotics, a subset of patients may experience a modest reduction in time to symptom resolution.¹ However, there is no way to know who these patients are. It is clear from the study by De Sutter and colleagues that family physicians should abandon the practice of selectively prescribing antibiotics to patients with certain clinical findings in the belief that these patients are more likely to have bacterial sinusitis. Instead, family physicians are left with two possible courses of action: prescribe antibiotics to every patient with suspected acute sinusitis, or do not prescribe antibiotics at all. Given the lack of benefit and the very real possibility of harm, the latter option is the clear choice.—K.W.L.