Use of transvaginal ultrasonography to predict preterm birth in women with a history of preterm birth.

OBJECTIVE: To investigate whether cervical length measured by transvaginal ultrasonography predicts spontaneous preterm birth at < 35 weeks' gestation in women with a history of spontaneous preterm birth, stratified by spontaneous preterm birth history subtype (preterm premature rupture of membranes (PPROM) or preterm labor with intact membranes at onset of labor).

METHODS: This retrospective cohort study included women with a history of spontaneous preterm birth that were subsequently pregnant with singleton gestations, compared with a low-risk control group. Transvaginal ultrasonographic cervical lengths were measured at 24 to 30 weeks of gestation. The primary outcome was spontaneous preterm birth at < 35 weeks. Secondary outcomes included spontaneous preterm birth at < 37 weeks and < 34 weeks, low birth weight, Cesarean delivery and perinatal morbidity and mortality. Multiple logistic regression analysis was used to control for potential confounders and calculate odds ratios and 95% confidence intervals. Receiver-operating characteristics (ROC) curves were used to determine the best cut-off for transvaginal ultrasound cervical length in predicting spontaneous preterm birth at < 35 weeks.

RESULTS: Women with a history of spontaneous preterm birth with intact membranes at onset of labor (n = 42) had a shorter cervical length (3.28 cm) than women with a history of spontaneous preterm birth with PPROM at onset of labor (n = 48, cervical length 3.77 cm; P = 0.019), and both subgroups had shorter cervical lengths than the low-risk control group (n = 103, cervical length 4.30 cm; P < 0.0001). Both subgroups were associated with spontaneous preterm birth at < 35 weeks, < 37 weeks, < 34 weeks and birth weight < 2500 g. ROC curves determined that the best cut-off for cervical length to predict spontaneous preterm birth at < 35 weeks was 3.0 cm. By multiple logistic regression analysis, the only independent predictors of spontaneous preterm birth at < 35 weeks were cervical length < 3.0 cm, a history of spontaneous preterm birth and antepartum bleeding in the current pregnancy. In women with a history of spontaneous preterm birth, a cervical length as measured by transvaginal ultrasonography of < 3.0 cm had a sensitivity of 63.6%, specificity of 77.2%, positive predictive value of 28.0% and negative predictive value of 93.8%, for preterm birth at < 35 weeks.

CONCLUSION: Women with a history of spontaneous preterm birth with preterm labor and intact membranes at the onset of labor have shorter cervices than women with a history of spontaneous preterm birth and PPROM at the onset of labor, and both groups have shorter cervices than a low-risk control group. Both groups of women with a history of spontaneous preterm birth have an increased risk of recurrent spontaneous preterm birth at < 35 weeks, and this is predicted by a transvaginal ultrasound cervical length of < 3.0 cm.