To the Editor:

We read Dr. Mark Ebell’s article “Making Decisions at the Point of Care: Sore Throat” [September 2003, page 68] with great interest. The article encourages the use of the strep score calculator developed by McIsaac.¹ In the example cited, the patient’s score is 5, meaning there is a 52-percent chance that he has streptococcal pharyngitis. The author notes that it would be reasonable to treat him for this, but another interpretation is that the likelihood of the patient having streptococcal infection is no better than chance alone. Accordingly, there is good reason to recommend a specific microbiologic test, such as a rapid diagnostic assay or a conventional throat culture, as a guide to therapeutic decisions. Some of the more sensitive rapid tests can provide an answer with an 80-to 85-percent certainty within 15 minutes. Given that group A streptococcus causes only about 20 percent of pharyngitis episodes in children and 10 percent in adults, this seems like a more prudent approach, especially given concerns about the overuse of antibiotics.

We also have concern about including antibiotics (azithromycin) other than first-line antibiotics in a patient flow sheet for management of pharyngitis. This implies that these are appropriate initial choices for management of pharyngitis due to group A streptococcus. Unless the patient is allergic to penicillin, either amoxicillin or penicillin remains the drug of choice for this condition.

These recommendations reflect the current guidelines endorsed by the Committee on Infectious Diseases of the American Academy of Pediatrics.² It would be in the best interest of all children to establish consistent, high-quality practice recommendations from the AAP and the AAFP.

Author’s response:

As I state in the article, another option for high-risk patients other than empiric antibiotics is “to base treatment on the results of a rapid strep test, or to obtain a throat culture and call the patient in two days with the results. If the culture is negative, you could discontinue antibiotics, recognizing the potential for false-negative culture results when the pretest probability is sufficiently high.”

Given the limited accuracy of rapid antigen tests, if a patient has a high risk of strep based on the McIsaac rule and a negative rapid antigen test, he or she still has approximately a 15-percent risk of streptococcal pharyngitis. I would argue that empiric therapy should be an option, particularly when other factors (recently diagnosed strep infection in the family, for example) increase the risk of strep even further. This recommendation is consistent with an evidence-based guideline that the Centers for Disease Control and Prevention has endorsed.¹

Regarding the choice of antibiotics, the first option was “none needed” and three of the other four choices were first-line antibiotics available in generic form. I agree that these antibiotics should be used in most nonallergic patients. While in an ideal world erythromycin would be prescribed to children and adults who are penicillin-allergic, the higher risk of adverse gastrointestinal effects with erythromycin and the more convenient dosing of azithromycin has made azithromycin the preferred drug for many physicians with penicillin-allergic patients. The encounter form simply acknowledges that reality.