Objective: to evaluate the uterine cervix by digital and transvaginal ultrasound examinations in pregnant women at high risk of having premature delivery. Methods: during the period between February 1995 and September 1997, 38 pregnant women at high risk of having premature delivery between the 20th and 36th week of gestation were examined. These patients were submitted weekly to both digital and transvaginal ultrasound examinations. The digital examination evaluated the uterine cervix using two parameters: length and dilation. The transvaginal ultrasound studied the length and the anteroposterior diameter of the uterine cervix. The behavior of these cervical measurements was analyzed throughout the pregnancies. The two methods were compared regarding cervical evaluation and accuracy of premature birth diagnosis. Results: the rate of premature deliveries was 18.4% (7/38). Digital examination resulted in cervical evaluations with variation coefficients of 30.3% for length and 193% for dilation. Transvaginal ultrasound resulted in cervical evaluations with variation coefficients of 14.7% and 26.5% for the anteroposterior diameter and length, respectively. The cervical length measures obtained on ultrasound were always greater than those obtained on digital examination. Through analysis with the hypothesis test, an indirect relationship was observed between the cervical length and the gestational period for digital examination and ultrasound study (p<0.05 and p<0.01, respectively), and a direct relationship between the cervical dilation and the gestational age observed on the digital examination (p<0.01). Conclusions: among the parameters studied by means of the digital and transvaginal ultrasound examinations, the ultrasound cervical length presented the best accuracy in the diagnosis of premature birth, proving to be more reliable for the evaluation of cervical alterations in pregnant women at high risk of premature delivery.
Ultrasonography; Digital examination; Uterine cervix; Preterm birth