Background: More than 800,000 women in the United States are diagnosed with pelvic inflammatory disease (PID) during their reproductive lifetimes. Although the standard therapy of one intramuscular injection of 250 mg of ceftriaxone (Rocephin) plus 14 days of oral doxycycline (Vibramycin) at 200 mg daily is effective, it is inconvenient and associated with poor compliance. Azithromycin (Zithromax) is effective against many genital tract infections and can be given on a once-weekly basis. Savaris and colleagues conducted a randomized controlled study comparing azithromycin with doxycycline for the oral component of therapy.

The Study: Women with symptoms of PID who presented to the emergency department of a large teaching hospital in Brazil were recruited for the study. Inclusion criteria were pelvic discomfort for up to 30 days, pelvic organ tenderness on bimanual examination, mucopurulent cervicitis, and urinalysis and pelvic or abdominal ultrasound findings consistent with PID. Exclusion criteria included pregnancy, recent antibiotic use or gynecologic surgery, rebound tenderness, fever greater than 100.4°F (38°C), allergies to study medications, and homelessness. Patients who vomited after taking the first dose of medication were observed for at least one hour after the first dose to check for intolerance to the oral formulation of the medication.

All study participants received initial therapy with an intramuscular injection of 250 mg of ceftriaxone, and were randomly assigned to two weeks of follow-up therapy with 100 mg of doxycycline twice daily or 1 g of azithromycin per week. Placebo pills were used to avoid bias and ensure each patient had the same daily intake of study pills. For the duration of the study, patients were asked to either abstain from sexual intercourse or use condoms, and not to take any additional analgesics or antibiotics.

The primary outcome was change in tenderness during abdominal and pelvic examination, which was measured by individual scores on visual analog and McCormack scales (0 = no pain to 3 = rebound tenderness) at 12 different abdominal and pelvic regions. Clinical cure was defined as a reduction in the patient's total tenderness score by at least 70 percent on day 14. Secondary outcomes assessed included changes in temperature, white blood cell count, and clinical status.

Results: The 67 women allocated to doxycycline follow-up therapy were comparable with the 66 women in the azithromycin group in age, initial tenderness scores, history of PID, and use of oral contraception. Complete compliance with medication was achieved in 49 doxycycline patients and 57 azithromycin patients. The mean pain scores at day 14 showed no significant difference between the two treatment groups. The clinical cure rates for patients who adhered to treatment protocol were 98.2 percent (56 out of 57) for azithromycin and 85.7 percent (42 out of 49) for doxycycline (P = .02). The authors calculated modified intention-to-treat cure rates of 90.3 percent (56 out of 62) for azithromycin and 72.4 percent (42 out of 58) for doxycycline (P = .01). By this analysis, the number needed to treat for benefit from azithromycin is six.

Conclusion: The authors conclude that after an initial injection of ceftriaxone, 1 g of azithromycin weekly for two weeks provides results comparable with 200 mg of doxycycline daily for two weeks.

editor's note: The authors correctly state that the two follow-up antibiotics were equivalent in this study. In patients who adhered to the treatment protocols, azithromycin provided statistically superior results. These patients were required to take two pills daily to maintain blinding. In practice, when patients are only asked to take one pill per week for two weeks, the adherence rates (and presumably cure rates) could be much higher for azithromycin than for doxycycline.—a.d.w.