For more than 40 years, penicillin has been the drug of choice for the treatment of streptococcal pharyngitis. Penicillin is still considered superior to other antibiotics because it effectively reduces symptoms and suppurative complications with few adverse effects and at low cost. The optimal duration of treatment, however, remains controversial. Symptoms typically resolve within three to four days after treatment is initiated, but short courses of therapy are associated with bacteriologic recurrence. Conversely, longer courses of treatment have led to antibiotic resistance. Zwart and colleagues compared the effectiveness of a three-day course of antibiotics with that of a seven-day course in treatment of a sore throat.

Adult patients who presented to their general practitioner with a history of up to seven days' sore throat were eligible for the randomized controlled trial. To be included in the analysis, patients had to have at least three of the following conditions concomitantly: fever, tonsillar exudate, tender anterior cervical lymph nodes and absence of cough. Those who met the inclusion criteria were randomized to receive penicillin (250 mg three times daily) for seven or three days, or a placebo. All tablets were identical, and patients treated for three days received four days of placebo to optimize comparability between groups. Acetaminophen was provided to all patients for symptom relief, if needed. During treatment, patients kept diaries to record symptoms, possible side effects, use of analgesics, daily temperature and the impact of illness on daily activities. The primary outcome variable was duration of symptoms, defined as the number of days until permanent resolution of pain or impaired daily activities. All patients were reassessed 14 days after beginning treatment. Telephone interviews were conducted two, four and six months after treatment to assess throat, respiratory and ear symptoms. All patient contacts during the six-month follow-up period were recorded by the patients' general practitioners. Compliance was checked by pill counts.

A total of 561 patients was included in the analysis. Of these, 190 completed a seven-day course of penicillin, and 194 completed a three-day course. Patients in the seven-day group showed permanent resolution of sore throat 1.9 days sooner than those in the three-day group and 1.7 days sooner than those taking placebo. Time to initial resolution of symptoms was similar in all patients treated with penicillin, but symptoms recurred within a week in 77 (40 percent) of the patients in the three-day group compared with 10 (5 percent) of the patients treated for seven days. The proportion of patients taking analgesics (61 percent) was similar across groups until day four, after which only 4 percent of patients in the seven-day group required analgesics. The decrease in analgesic use was much smaller in the three-day and placebo groups. Patients in the seven-day group resumed their daily activities approximately two days earlier than patients in the other groups. Recurrence of sore throat within six months was highest in the three-day group. Adverse effects associated with penicillin use included nausea (40 percent of patients) and abdominal pain (26 percent of patients). Six patients in the placebo group had serious streptococcal complications.

The authors concluded that seven-day penicillin therapy was superior to three-day therapy or placebo in treating sore throat. They discourage the use of shorter courses of treatment of this common condition.