ACCURACY OF 3-DIMENSIONAL ULTRASONOGRAPHY IN VOLUME ESTIMATION OF CERVICAL-CARCINOMA

Objective. To evaluate the accuracy of a three-dimensional (3D) ultras ound system in volume estimation of cervical carcinoma Study design. T ransvaginal 3D and two-dimensional (2D) scans on cervical carcinoma vo lumes were performed 1 day before surgery. The volume of cervical carc inoma measured from each surgical specimen was compared with the corre sponding volume of the cervical tumor measured by a 3D ultrasound and with the conventional 2D ultrasound volume measurement calculated usin g the formula pi/6 x (R-1 x R-2 x R-3), where R-1, R-2, and R-3 were t he maximal transverse, anteroposterior, and longitudinal length of tum or, respectively. Limits of agreement and 95% confidence intervals wer e calculated and systemic bias between the methods was analyzed. The K lotz test was also used to assess the statistical significance of the degree of dispersion, Results. A total of 61 cases, 55 with exophytic tumors and 6 with endocervical tumors, were examined in this study. Th e limits of agreement between the volume measured from specimen and tu mor volume determined by ultrasound were +6.68 to -6.10 mL for 3D meas urements and +12.46 to -10.98 mL for 2D measurements. The Klotz test s howed the discrepancy in the degree of dispersion between 3D and 2D ul trasound measurements was statistically significant (P = 0.01). Conclu sion. The true volume of cervical carcinoma is measured more accuratel y by a 3D ultrasound system than 2D ultrasound. (C) 1997 Academic Pres s.