The article by Alho and colleagues1 and the accompanying editorial by Hickner2 in the January 2001 issue were intriguing and add an important insight to the growing primary care knowledge base about the management of acute rhinosinusitis. Their finding that patients who present with a history of recurring sinus infection early in the course of rhinosinusitis are no more likely than those in the control group to have a bacterial infection (and perhaps less likely) is really not surprising.
Hansen and colleagues3 showed the same thing in a primary care study of acute sinusitis that was validated by maxillary sinus puncture: a statistically significant correlation between previous history of sinusitis and a negative aspiration result. Alho’s group made a significant contribution to understanding this association because their patients with a history of recurrent sinus infection were more likely to have “sinusitis-like changes” on a radiograph.
A number of studies (especially that of Lindbaek and coworkers4) have clearly shown that these changes do not correlate with bacterial sinusitis. Although Alho and colleagues reported no significant correlations between receipt of antibiotic and radiographic or symptomatic improvement at 3 weeks, it seems unlikely that their 48 patients gave the study enough power to draw a negative conclusion. Nonetheless, several other studies make it likely that radiography has little or no role in helping decide whether there is a subgroup of these patients who would benefit from antibiotics. In fact, the data from Alho and coworkers are suspiciously close to those of the computed tomography study of the common cold by Gwaltney and coworkers,5 which showed that self-defined upper respiratory infection (URI) is often accompanied by major radiographic changes.
If patients who feel more uncomfortable with their viral URI have more inflammation in their sinuses, they are both more likely to present to their physicians with symptoms and radiologic changes and more likely to get validation from their physicians that they do indeed need a course of antibiotics. Once this has occurred, their illness has been medicalized into “chronic recurring sinusitis.” This increases the odds that they will present early during their next viral URI believing that they require an antibiotic.
Because cognitive dissonance is as strong a factor for patients and physicians as it is, it will require prodigious efforts to re-educate both ourselves and our patients to the reality of our ignorance about when antibiotics are warranted in acute rhinosinusitis. Although Alho and colleagues had numbers that were small and excluded anyone not falling between the 2 extremes of “never had clinical sinusitis” and those with “at least 2 yearly episodes of acute maxillary sinusitis during the previous 2 years,” I believe that they have convincingly established that this group of patients should not be given antibiotics if they present with 96 hours or less of symptoms. We need to stop searching for an “equally plausible explanation” and adjust our practice both in diagnosis and treatment.
Jay C. Smith, MD, ABFP
University of Cincinnati
Department of Family Medicine
- Alho O, Ylitalo K, Kalevi J, et al. The common cold in patients with a history of recurrent sinusitis: increased symptoms and radiologic sinusitislike findings. J Fam Pract 2001; 50:26-31.
- Hickner JM. Acute rhinosinusitis: a diagnostic and therapeutic challenge. J Fam Pract 2001; 50:38-39.
- Hansen JG, Schmidt H, Rosborg J, Lund E. Predicting acute maxillary sinusitis in a general practice population. BMJ 1995; 311:233-36.
- Lindbaek M, Kaastad E, Dolvik S, Johnsen U, Laerum E, Hjortdahl P. Antibiotic treatment of patients with mucosal thickening in the paranasal sinuses, and validation of cut-off points in sinus CT. Rhinology 1998; 36:7-11.
- Gwaltney JM, Phillips CD, Miller RD, Riker DK. Computed tomographic study of the common cold. N Engl J Med 1994; 330:25-30.
To the editor:
The articles on rhinosinusitis in the January issue make the issue of proper treatment more perplexing. Findings of not being able to rely on clinical symptoms to differentiate between viral and bacterial etiologies and that chronic sufferers who usually are prescribed long-term antibiotics have significantly more viral illnesses do little to help the busy practitioner avoid the overuse of antibiotics.
Members of Alho’s group showed that a 5% solution of xylitol reduced the attachment of Streptococcus pneumoniae to nasal epithelial cells by 68% and that of Haemophilus influenzae by 50%.1 They also showed that oral xylitol in gum or syrup reduced the incidence of otitis media.2,3 Studies at the University of Iowa show that xylitol, when sprayed into the nose, reduced the saline concentration of the airway surface fluid, allowing defensins there to be more effective. Use of this spray for 4 days significantly reduced the colonization of coagulase-negative Staphylococcus in the nose.4