Background: Patients often present to their primary care physician with symptoms suggestive of acute maxillary sinusitis. Because it can be unclear whether these symptoms have a bacterial etiology, physicians often treat patients empirically with antibiotics. A high degree of overtreatment likely occurs, with concomitant economic costs and increasing bacterial resistance. Although studies have shown moderate effect sizes for penicillin in the treatment of acute sinusitis, the results have not been statistically significant. Community-based studies have shown the smallest effect sizes. Few studies have compared clinical symptom evaluation with culture results from direct sinus aspirates. One study identified clinical criteria (i.e., three predictive symptoms and one sign), and these criteria were used in the current study to diagnose bacterial sinusitis. In patients fulfilling these clinical criteria for acute bacterial sinusitis, Williamson and colleagues compared treatment with amoxicillin (500 mg three times daily for seven days) and a nasal steroid (200 mcg of budesonide [Rhinocort] in each nostril once daily for 10 days), used alone and in combination. The rationale for the nasal steroid is mostly theoretical, but a Cochrane review found that for each 100 patients treated, an additional seven would benefit from using a topical steroid.
The Study: In this randomized, double-blinded, placebo-controlled trial over two and one half years, 74 physicians were recruited from 117 practices, with an additional 18 practices involved afterward to increase participation. Eligible patients were 16 years or older and had to have two or more of the following symptoms of sinusitis: purulent nasal discharge with predominance on one side; localized pain with predominance on one side; bilateral purulent nasal discharge; and nasal pus visible on inspection. Patients with two or more episodes of sinusitis in the previous year were excluded, as were those with medical comorbidities; allergies or adverse reactions to medications; or recent antibiotic or steroid use; and those who were pregnant or breastfeeding. Patients were randomized to one of four groups: antibiotic and nasal steroid; antibiotic and placebo nasal steroid; placebo antibiotic and nasal steroid; or both medications as placebo. Patients completed a 14-day symptom diary, which included 11 variables that the patients rated from ‘normal’ to ‘as bad as it could be.' The data from the symptom diaries were the main outcome measure.
Results: Of 388 eligible patients, 240 participated. Recruitment barriers included patient demand for immediate antibiotics and insufficient time for physicians to recruit patients. Of the recruited patients, 70 percent had two of the entry criteria, and 30 percent had three or four. Twenty-nine percent of patients taking amoxicillin had symptoms lasting 10 days or longer, compared with 33.6 percent of patients not taking amoxicillin (adjusted odds ratio [OR] = 0.99; 95% confidence interval [CI], 0.57 to 1.73), and 31.4 percent using the nasal steroid had symptoms lasting 10 days or longer, compared with 31.4 percent not using the nasal steroid (adjusted OR = 0.93; 95% CI, 0.54 to 1.62). Further analyses confirmed that the differences were nonsignificant between the antibiotic-and-placebo and nasal steroid-and-placebo groups. Time to cure (defined by scoring 0 or 1 on all symptom-diary items) was similar among groups, with 40 percent of all patients cured by one week, regardless of treatment. In subanalyses of symptoms of pain and feeling unwell, no differences were found except in the following: patients with mild symptoms at baseline tended to benefit from nasal steroids, whereas those with severe baseline symptoms tended to do worse with nasal steroids.
Conclusion: The authors conclude that, in a primary care setting, patients with acute sinusitis benefit little, if at all, from antibiotics or nasal steroids. Given that the definition of sinusitis in this trial was narrowly defined, it is even less likely that patients with other, more broadly defined diagnoses of sinusitis would benefit from active treatment. The only exception is a possible benefit from topical steroids in patients with milder symptoms.