OBJECTIVE: The purpose of this study was to assess the correlation and agreement between transvagi-nal and transabdominal cervical length measurement after bladder emptying as well as the feasibility of transabdominal sonography in cervical length screening. METHODS: This was a prospective cohort study involving 287 participants (14-34 weeks' gestation) from January to December 2003. After voiding, transabdominal and transvaginal cervical length measurements were obtained. The optimal trans-abdominal technique was established during an unblinded series of transabdominal and transvaginal cervical length measurements (n = 96). The same measurements were obtained in 191 participants under a blinded 2-sonographer protocol. The transabdominal cervical length cutoff to ensure 100% sensitivity in detecting a short cervix (<or=2.5 cm) was determined. RESULTS: There was no difference between mean transabdominal and transvaginal cervical lengths +/- SD (3.57 +/- 0.74 versus 3.61 +/- 0.74 cm; P = .20). The Pearson correlation coefficient was 0.824. The 95% tolerance interval for any paired observation (transabdominal minus transvaginal) was -0.92 to 0.84 cm. All transvaginal cervical lengths of 2.5 cm or less were associated with paired transabdominal cervical lengths of 3 cm or less. CONCLUSIONS: With an optimal sonographic technique, postvoid transabdominal cervical measurement shows a close correlation and agreement with transvaginal assessment and is useful for cervical length screening.
OBJECTIVE: The purpose of this study was to assess the correlation and agreement between transvagi-nal and transabdominal cervical length measurement after bladder emptying as well as the feasibility of transabdominal sonography in cervical length screening. METHODS: This was a prospective cohort study involving 287 participants (14-34 weeks' gestation) from January to December 2003. After voiding, transabdominal and transvaginal cervical length measurements were obtained. The optimal trans-abdominal technique was established during an unblinded series of transabdominal and transvaginal cervical length measurements (n = 96). The same measurements were obtained in 191 participants under a blinded 2-sonographer protocol. The transabdominal cervical length cutoff to ensure 100% sensitivity in detecting a short cervix (<or=2.5 cm) was determined. RESULTS: There was no difference between mean transabdominal and transvaginal cervical lengths +/- SD (3.57 +/- 0.74 versus 3.61 +/- 0.74 cm; P = .20). The Pearson correlation coefficient was 0.824. The 95% tolerance interval for any paired observation (transabdominal minus transvaginal) was -0.92 to 0.84 cm. All transvaginal cervical lengths of 2.5 cm or less were associated with paired transabdominal cervical lengths of 3 cm or less. CONCLUSIONS: With an optimal sonographic technique, postvoid transabdominal cervical measurement shows a close correlation and agreement with transvaginal assessment and is useful for cervical length screening.