Longitudinal investigation of candida vaginitis in pregnancy: role of superimposed antibiotic use.

OBJECTIVE: To evaluate the purported association of antibiotic use and subsequent symptomatic Candida vaginitis among pregnant women.
METHODS: Two hundred fifty obstetric patients were followed in a prospective, culture-based, longitudinal, and observational study from the first antepartum appointment through the postpartum visit at 6 weeks. All patients were cultured for yeast (Nickerson agar) initially. Patients with symptoms and microscopic evidence of vaginitis at the initial visit were followed through pregnancy but were not analyzed with asymptomatic individuals who had vaginal cultures for Candida at the first visit and at subsequent visits if they developed vulvovaginal symptoms. Patients were categorized as colonized or uncolonized on the basis of initial cultures and were evaluated at least monthly for antibiotic use and vaginal complaints. In addition, hospital records were reviewed after the final visit to document antibiotic use or vaginal infection.
RESULTS: Asymptomatically colonized patients were at a threefold greater risk of developing symptoms than were uncolonized patients (P < .001). Among women receiving antibiotics during pregnancy, 6.1% developed symptoms of Candida vaginitis compared with 15.6% of women who did not receive antibiotics. For the entire study population, 46% of the patients received at least one course of antibiotic therapy and 21% had multiple courses. Only three of the seven who became symptomatic with yeast vaginitis did so within 4 weeks of treatment. Many of the antibiotic regimens were prescribed by providers other than the obstetrician.
CONCLUSION: Antibiotic treatment during pregnancy was frequent in the study population, but was not associated with a significant risk of developing Candida vaginitis.