Pharyngitis can be caused by many organisms, but treatment generally is reserved for sore throat associated with group A beta-hemolytic streptococcus and, rarely, Neisseria gonorrhoeae or the acute antiretroviral syndrome. Although group A streptococcal infection is responsible for only 10 percent of pharyngitis cases in adults who are seen in general ambulatory care settings, 75 percent of patients with sore throat are given antibiotics. Penicillin remains the treatment of choice for group A beta-hemolytic streptococcal infection, but many physicians prescribe broad-spectrum antibiotics. The rationale for treating this infection has been questioned, because acute rheumatic fever has become rare, and treatment probably does not prevent glomerulonephritis. The benefit of treatment appears to be a shorter symptomatic period and a decreased risk for infectious sequelae. Neuner and colleagues analyzed the cost-effectiveness of five common strategies for testing and treating adults with sore throat.

The authors assessed the following five strategies: (1) observation only, (2) empiric antibiotic therapy with no testing, (3) throat cultures, with treatment for patients with positive cultures, (4) optical immunoassay followed by culture to confirm negative tests, with antibiotic therapy for patients with positive results on either test, and (5) optical immunoassay alone, with antibiotic therapy for patients with positive test results. Antibiotic therapy consisted of a 10-day course of penicillin (or erythromycin if the patient was allergic to penicillin). The authors evaluated the effect of the strategies on the following outcomes: (1) acute rheumatic fever, (2) peritonsillar abscess, (3) symptom duration, and (4) allergic reaction to antibiotics. Cost estimates represented actual resource costs rather than charges.

With the assumption of a 9.7 percent prevalence of group A beta-hemolytic streptococcal infection in adults with sore throat (or any prevalence rate between 6 and 20 percent), the analysis revealed that the culture strategy was more effective and less expensive. Empiric therapy was the least effective strategy. The optical immunoassay/culture strategy was the most expensive. In populations at higher risk for group A beta-hemolytic streptococcus infection (prevalence greater than 71 percent), optical immunoassay/culture was the most effective strategy. If the prevalence was less than 6 percent, observation was the least expensive approach.

The authors conclude that empiric therapy of pharyngitis is reasonable only when the prevalence of group A beta-hemolytic streptococcal infection approaches 70 percent, as might happen among family members of infected persons or among those with a high probability of infection based on the Centor decision rule. The culture strategy was found to be the most cost-effective.

In an editorial in the same journal, Bisno points out that physicians are unlikely to use diagnostic testing when a widely endorsed guideline, such as the Centor decision rule, is available. The American College of Physicians and the American Academy of Family Physicians have approved a recommendation allowing for use of a rapid reagent antigen diagnostic test or empiric treatment of patients meeting only three or four Centor criteria (i.e., tonsillar exudates, tender anterior cervical adenopathy, absence of cough, history of fever). Bisno also states that newer guidelines by the Infectious Diseases Society of America advocate the use of rapid antigen diagnostic testing alone to confirm the diagnosis of group A beta-hemolytic streptococcal infection in adults before treatment. He supports the idea that, until more trials are performed, relying on clinical data alone for the diagnosis and management of pharyngitis caused by group A beta-hemolytic streptococcus is imprudent.

editor's note: Debate continues about how to best evaluate and treat adults with pharyngitis. Empiric antibiotic therapy is useful only in high-risk patients and those with clear criteria for treatment (i.e., based on the Centor criteria). The usefulness of antibiotic therapy in lower risk patients and in those without characteristic treatment criteria is doubtful. A Cochrane abstract (Cochrane Database of Abstracts and Reviews. CRD database number DARE-20018156) of a quality-assessed systematic review of other treatments for acute sore throat agrees with Thomas and associates (Thomas M, et al. How effective are treatments other than antibiotics for acute sore throat? Br J Gen Pract October 2000;50:817-20) that treatments other than antibiotics may be more effective first-line options to relieve the symptoms of acute sore throat. Some of these options include steroids, nonsteroidal anti-inflammatory drugs, caffeine, paracetamol, supercolonization with benign bacteria, better physician-patient communication, and influenza and pneumococcal vaccines. Although some short-term benefit has been achieved with the first four options, and longer term efficacy occurred with others, further prospective study is needed to define appropriate alternatives that will be better than antibiotics for consistently relieving sore throat pain in adults.—R.S.