Longitudinal assessment of endocervical canal length between 15 and 24 weeks' gestation in women at risk for pregnancy loss or preterm birth.

OBJECTIVE: To determine the weekly cervical shortening rates of the endocervical canal between 15 and 24 weeks' gestation in women at risk for pregnancy loss or spontaneous preterm birth.
METHODS: We performed a retrospective cohort study of transvaginal sonographic measurements of the endocervical canal length done at least twice between 15 and 24 weeks' gestation in women at risk for pregnancy loss and spontaneous preterm birth. The ultrasound diagnosis of cervical incompetence was defined as progressive shortening of the endocervical canal length to 2 cm or less either spontaneously or after application of transfundal pressure. Multivariable linear regression models were developed to determine the weekly crude rate of endocervical canal length shortening rates in cases of competent cervices and incompetent cervices, with incompetent cervices further stratified as those diagnosed at 15-19 weeks' and 20-24 weeks' gestation. Comparisons of the models for weekly rate of endocervical canal length shortening were performed.
RESULTS: The endocervical canal lengths were measured in 61 women (180 measurements) who did not develop ultrasound evidence of cervical incompetence and 28 women (103 measurements) who had ultrasound evidence of cervical incompetence. Between 15 and 24 weeks' gestation, competent cervices had a nonsignificant rate of endocervical canal length shortening (-0.03 cm/week). During this period in gestation, incompetent cervices had significantly greater endocervical canal length shortening (-0.41 cm/week, P < .001). The rate of endocervical canal length shortening of incompetent cervices diagnosed between 15 and 19 weeks' gestation was -0.52 cm/week (P < .001). The rate of endocervical canal length shortening in incompetent cervices diagnosed between 20 and 24 weeks' gestation was significant and varied from -0.49 cm/week to -0.80 cm/week at 20 and 24 weeks' gestation, respectively (P < .001). The models describing the rate of cervical shortening in the two groups of incompetent cervices were significantly different (P < .001). The sonographic detection of endocervical canal length shortening in the 28 cases of cervical incompetence was identified at a median (range) gestational age of 20 (16-24) weeks.
CONCLUSION: Weekly rates of endocervical canal length shortening were established, which may be useful for detecting and managing cervical incompetence in high-risk women examined with cervical sonography.