Different strategies have been developed to refine the prediction of the risk of preterm delivery in asymptomatic patients. Transvaginal sonography has been used for this reason to measure and examine the length and shape of the cervix. In this review, we focus on clinical studies involving transvaginal sonographic assessment of the cervix in asymptomatic women at high risk of preterm delivery and in the general pregnant population. Three ultrasound signs are suggestive of cervical incompetence, namely, dilatation of the internal os, sacculation or prolapse of the membranes into the cervix (with shortening of the functional cervical length) either spontaneously or induced by transfundal pressure, and short cervix in the absence of uterine contractions. Transvaginal sonography has clearly demonstrated that cerclage leads to a measurable increase in cervical length which may contribute to the success of this procedure in reducing the risk of preterm delivery. Several non-randomized interventional studies among patients with cervical incompetence have been published. They have defined a new group of patients requiring cerclage when the women show progressive cervical modifications on transvaginal sonography, while in other studies, cerclage performed on the basis of cervical changes on transvaginal sonography did not prevent premature delivery. One prospective randomized trial in asymptomatic high-risk women has shown two benefits of cerclage following indications for transvaginal sonography: (1) it would generate fewer prophylactic cerclages in high-risk women; (2) therapeutic cerclage before 27 weeks may reduce the incidence of premature delivery before 34 weeks. The risk of preterm delivery is inversely correlated with cervical length. Routine transvaginal sonography of the cervix performed between 18 and 22 weeks can help identify patients at risk of preterm delivery. However, given the low prevalence of preterm births, screening would generate either a high false-positive rate or a low sensitivity. One non-randomized interventional study among patients with a short cervix on routine ultrasound examination found a lower risk of delivery before 32 weeks in the cerclage group than in the expectant management group. However, to date, there have been no prospective randomized trials in a general population. Although evidence is still lacking, there does appear to be a benefit in performing a cerclage rather than continuing with expectant management in cases with sonographic appearance of cervical incompetence in asymptomatic women at high risk of preterm delivery. Ultrasound can be offered to reduce the indications of cerclage for cases in which the situation is uncertain. Within the general obstetric population, transvaginal sonography might help in the selection of asymptomatic but high-risk women. However, the benefit associated with cerclage for sonographic indication has not been demonstrated.