Pj. Meis et al., THE PRETERM PREDICTION STUDY - SIGNIFICANCE OF VAGINAL INFECTIONS, American journal of obstetrics and gynecology, 173(4), 1995, pp. 1231-1235
OBJECTIVE: Our purpose was to evaluate the association oi bacterial va
ginosis, trichomonas vaginitis, and monilial vaginitis with spontaneou
s preterm birth at < 35 weeks 0 days. STUDY DESIGN: A total of 2929 wo
men at in centers were studied at 24 and 28 weeks' gestation by Gram s
tain of vaginal smear, wet mount, and 10% potassium hydroxide preparat
ions to detect vaginal infections. RESULTS: The rates of detected infe
ction at 24 and 28 weeks, respectively, were bacterial vaginosis 23.4%
and 19.4%, trichomonas 3.3% and 2.7%, and monilia 21.1% and 19.5%. Th
e occurrence of bacterial vaginosis at 28 weeks was associated with an
increased risk of spontaneous preterm birth, odds ratio 1.84 (95% con
fidence interval 1.15 to 2.95, p < 0.01). Detection of Trichomonas vag
inalis (by wet mount) or monilia (by potassium hydroxide preparation)
had no significant associations with preterm birth. CONCLUSION: The pr
esence of bacterial vaginosis at 28 weeks' gestation is associated wit
h an increased risk of spontaneous preterm birth.