*diarrhea vomiting, and skin rash were most common effects, none were life-threatening
Source: Ahovuo-Saloranta A, Borisenko OV, Kovanen N, Varonen H, Rautakorpi UM, Williams JW Jr, Mäkelä M. Antibiotics for acute maxillary sinusitis. Cochrane Database Syst Rev. 2008 Apr 16;(2):CD000243. Review. PubMed PMID: 18425861.
Efficacy Endpoints: Clinical cure or improvement in disease at 7 to 15 days after start of treatment
Harm Endpoints: Medication side effects
Narrative: Acute rhinosinusitis (sinusitis) is a commonly encountered upper respiratory infection, with an estimated 20 million cases occurring annually1. Although the vast majority of sinusitis is viral in origin,1, 2 most patients given a diagnosis of sinusitis in a primary care setting3 or emergency department4 are prescribed antibiotics; it accounts for 1 out of every 5 prescriptions written for antibiotics in the US5. This review focuses on whether antibiotics are effective in the treatment of acute sinusitis suspected to be of bacterial origin.
Five randomized, placebo controlled trial comprising about 500 patients comparing antibiotics to placebo in acute rhinosinusitis. The studies suggest a mild benefit of antibiotics in the treatment of sinusitis with a treatment effect of about 6.8%. This means that for every 15 patients treated with antibiotics, 1 will improve who would not have otherwise. There was 12.5% more side effects in the treatment group with a number needed to harm of 8, though the side effects of antibiotics were mild.
Caveats: The applicability of this review to clinical practice is complicated by three issues. The first, and most important, is that 4 out of 5 patients in the placebo group improved without antibiotics. Therefore the vast majority of patients with suspected bacterial sinusitis will not benefit from treatment (80% will improve regardless, and 13.2% more saw no benefit to antibiotics). The second issue is that none of the studies in the analysis used the gold standard of sinus aspiration to diagnose bacterial sinusitis. Instead, surrogate markers such as clinical symptoms, symptom duration, and radiographic findings were used. Numerous studies have shown that clinical symptoms cannot reliably distinguish between bacterial and non bacterial causes of sinusitis5, 6, 7. Although symptom duration has historically been used to differentiate viral from bacterial sinusitis, recent studies suggest that symptoms lasting > 10 days are not predictive of a positive bacterial aspirate7. Lastly, although sinus radiographs are 90% sensitive, they are 60% specific for bacterial sinusitis8. In fact, CT evidence of sinus disease is seen in 9 out of 10 patients with a cold9. Finally, since by day 7 of a cold, almost 40% of patients have a clinical sinusitis and by day 21, more than 10% have sinusitis10, it is safe to assume that some (perhaps most) of the patients in these studies did not have the disease of interest. Therefore, it is unclear just what — if any — the treatment effect of antibiotics on bacterial sinusitis would actually be.
Author: Koustav Mukherjee, MD
Published/Updated: January 8, 2013
The title bar is color-coded with our overall recommendation.
If you have suggestions, requests, or questions about a particular NNT review, please send us a message and we’ll try to address it as soon as possible.