OBJECTIVE:Our purpose was to test the hypothesis that wedging of the cervical internal os determined by transvaginal ultrasonography is associated with premature labor and delivery. STUDY DESIGN:Seventy patients admitted to the hospital for threatened preterm labor were evaluated by transvaginal ultrasonography before institution of therapy. Bivariate and logistic regression analyses were performed to determine the variables that made a significant contribution to the prediction of preterm delivery. RESULTS:Preterm delivery was significantly associated with the presence of cervical wedging, as noted on cervical scan, and with short cervical length. A history of previous preterm delivery was of marginal significance as a predictor of preterm delivery (p=0.09). Preterm delivery was not significantly correlated with age, previous voluntary termination of pregnancy, gestational age at the time of study, previous normal spontaneous vaginal delivery, or tocolytic therapy. Use of wedging as a diagnostic test for the prediction of preterm delivery yielded a sensitivity of 100% a specificity of 74.5%, a positive predictive value of 59.4%, and a negative predictive value of 100%. CONCLUSION:The presence of wedging and shorter cervical length was suggestive of true preterm labor requiring aggressive management. A transvaginal ultrasonographic "snapshot" view of the cervix seems to be a more reliable method to evaluate the cervix in patients with threatened premature labor than are uterine contractions alone.