Rates of preterm birth have increased in recent years, and the focus is now on primary prevention. Both transvaginal (TV) sonography and transabdominal (TA) sonography are used to assess cervical length. Many factors, including the mother’s physical condition and fetal position, influence the practical use of TA sonography. This retrospective study was undertaken to assess the reliability of TA compared with TV sonography for cervical length measurement in midpregnancy. The maternal and fetal situations in which TA measurement is reliable were also determined.
Women at 20 to 29 weeks 6 days’ gestation who presented for a fetal structural abnormality survey were enrolled. Cervical length was measured first using TA sonography, followed immediately by TV sonography to reduce any bias caused by fetal movement or changes in the maternal bladder status. The discrepancies between cervical lengths on TV and TA sonography were determined for (1) vertex versus breech presentation, (2) whether a fetal presenting part overlay the cervical internal os, (3) whether both the internal os and external os were visible or only the internal os was clearly visible, (4) maternal bladder filling status, (5) maternal age, (6) parity, and (7) gestational age. A paired t test, Pearson correlation, and Student t test were used for statistical analyses, performed with SAS Enterprise Guide version 4.1 software. P < 0.05 indicated significance.
Of 475 pregnant women, 307 were enrolled, with the cervix visualized and its length measured by TV sonography in all 307 cases. Measuring the cervical length using TA sonography was not possible in 52 women because the internal os was not seen on the TA scan; therefore, cervical length was measured transabdominally in only 255 cases (83.06%). The mean gestational age of the 52 unmeasurable cases was 23.9 ± 3.02 weeks compared with 22.7 ± 2.33 weeks (P = 0.01) in the 255 measurable cases. The mean cervical lengths were 3.88 ± 0.73 cm and 3.93 ± 0.72 cm on TA and TV sonography, respectively (P = 0.129; Pearson r = 0.75). The discrepancy between TV and TA assessments for fetal presentation did not differ significantly (P = 0.505). In the 78 cases in which a fetal presenting part overlay the internal os on TV sonography, the discrepancy between TV and TA assessments was significantly greater than that in the 177 cases in which a fetal presenting part did not overlay the internal os (0.21 ± 0.61 vs −0.03 ± 0.44 cm; P = 0.002). A significant discrepancy was found between the 64 cases in which only the internal os was visible on TA sonography and the 191 cases in which both the internal os and external os were visible (0.33 ± 0.75 vs −0.05 ± 0.35 cm; P < 0.001). The maximum vertical depth of the bladder was not significantly associated with the discrepancy between TV and TA sonography (P = 0.940). The discrepancy between TV and TA assessments was significantly greater among primiparous women than multiparous women (P = 0.017). An inverse correlation was noted between maternal age and cervical length discrepancy (P = 0.019; Pearson r = −0.147), but no correlation with gestational age was found (P = 0.268).
Overall, no significant difference was found between the mean cervical lengths obtained by TV and TA sonography. A methodical approach to cervical length assessment could include TA sonography for initial cervical length screening, considering various maternal and fetal clinical factors, with TV sonography used only if necessary.
Department of Obstetrics and Gynecology, Ulsan University Hospital, University of Ulsan College of Medicine, Ulsan (H.-J.R.); Department of Obstetrics and Gynecology, Haeundae Paik Hospital, Inje University College of Medicine, Busan (Y.I.J., C.H.J., G.H.J., S.C., H.J.C.), Korea