O volume do colo uterino na gravidez: comparação entre aferição sonográfica bidimensional e tridimensional

Objectives: To compare uterine cervix volume measurements by US3D using the VOCAL technique with the manual method using US 2D measurements and geometric formula for a cylinder. Methods: We included 48 asymptomatic pregnant women at 20 to 24 weeks gestation under low risk of preterm birth. They underwent transvaginal ultrasonography to measure the length and acquisition of the 3D volume of the cervix at the time of second trimester fetal malformations screening. The acquired data sets were stored on the ultrasound machine and analyzed by two independent observers later. The volume calculations were performed by the VOCAL software and by the geometric formula for cylinders. The reliability analysis for the two methods was evaluated by the intraclass correlation coefficient (ICC) and Bland-Altman plots were produced to examine intermethod agreement. Results: Inter-method agreement was considered acceptable, with correlation of r = 0.87 for observer 1 and r = 0.71 for observer 2 (p <0.001). The ICC was 0.58 (95% CI 0.25-0.76) for the observer 1 and 0.80 (95% CI 0.64-0.88) for the observer 2. The intraobserver agreement was excellent for both methods, ICC 0.93 (IC92% 0.87-0.96) for the VOCAL technique and 0.92 (95% CI 0.87-0.96) for the manual method. Interobserver agreement was good for the VOCAL technique, ICC 0.81 (95% CI 0.67-0.89), and poor for the manual method, ICC 0.37 (95% CI -0.10-0.65). Conclusion: This method comparison study shows that the geometric formula for a cylinder has acceptable agreement with VOCAL to measurement of uterine cervix volume. The two techniques used have good intraobserver reproducibility. The manual method presented low interobserver reproducibility. The VOCAL technique demonstrated greater reliability because it presented good intra and interobserver reproducibility.