However, a total of 10 of the 16 studies included for the outcome of ARF were conducted in the 1950s. The baseline risk was 0.036, or 1 in 28, with only one controlled study performed in military camps in the 1950s demonstrating possible reduction in ARF.¹² In studies published since 1990, there were no cases of rheumatic fever in 2,484 patients in both treatment and control groups.

If data analysis is restricted to the six studies done since 1975, the maximum risk is not more than 0.3%, or approximately 1 in 345.¹³ The systematic review data reveal that to prevent one case of ARF the NNT is 41, while post-1975 studies have an NNT of 494.¹³ Today, the estimated incidence of rheumatic fever in the United States is 1 in 1 million. Therefore, if we assume the risk is 1 in 1 million, the NNT would be 1,430,000 for ARF.¹³ If only one-third of patients with acute rheumatic fever develop cardiac complications, the NNT increases to more than 1 in 3 million to prevent one case of heart disease due to ARF.

Why does this matter? Disadvantages of treating patients with unnecessary antibiotics include additional expense, risks of therapy with limited benefit (drug allergy reactions and side effects), and concern for increased resistance to antibiotics (especially macrolides or fluoroquinolones). With use of similar estimates as presented, for every 1 million antibiotic prescriptions for “strep throat,” there could be as many as 500 severe (potentially fatal) allergic reactions and 100,000 cases of diarrhea and rash.

Overall, the authors of the Cochrane review concluded that:

• Antibiotics confer modest absolute benefits by shortening the duration of symptoms by only about 16 hours.
• Protecting sore throat sufferers against suppurative and nonsuppurative complications in modern Western society can be achieved only by treating many patients with antibiotics, most of whom will derive no benefit; in contrast, in emerging economies (where rates of acute rheumatic fever are high), the NNT may be much lower for antibiotics to be considered effective.
• Antibiotics reduce bacterial infections, but they can cause diarrhea, rash, and other adverse effects, and communities build resistance to certain antibiotics.

Is There a True “Bottom Line”?

Antibiotics are of limited use for most patients with sore throats. The two main benefits of early antibiotic therapy for GABHS pharyngitis are a modest earlier resolution of the patient’s symptoms and a diminished likelihood of spreading the infection to other people. However, recently updated American Heart Association (AHA) guidelines for the prevention of rheumatic fever (endorsed by the AAP) recommend routine testing of all patients with suspected GABHS pharyngitis (based on clinical and epidemiologic findings; see Figure 1) and antibiotic use only for documented GABHS infection.¹⁴ It will be interesting to see if any major shifts in the current standard recommendation for antibiotics for all suspected or documented GABHS infection will be proposed in an expected updated guideline (projected publication, spring 2011) from the Infectious Diseases Society of America on this controversial issue.