The reproducibility of VOCAL endometrial volume measurement: importance of the step rotation - Revista Brasileira de Ginecologia e Obstetrícia

PURPOSE: to determine the intraobserver and interobserver reproducibility of endometrial volume measurements using the VOCAL®-imaging program (Virtual Organ Computer-aided AnaLysis). METHODS: one three-dimensional (3-D) ultrasound dataset of the endometrium was obtained from each of five infertile women with different endometrial volumes. For each 3-D dataset, the endometrial volume was calculated by two different observers using the manual mode in four different rotational steps (30º, 15º, 9º and 6º). Ten measurements were obtained with each method and observer from each 3-D dataset. We have used one-way ANOVA and the Tukey post-test to verify the differences among means and the intraclass correlation coefficient to test reliability. RESULTS: rotational methods employing a rotation step of 30º were associated with lower endometrial volume readings in 3 of the 5 patients. There were no significant differences between the means obtained by the 15º, 9º or 6º step rotation. No significant difference was found between the means obtained by the two different observers. The intraclass correlation coefficients were significantly lower with 30º (all under 0.984) than with the other step rotations (all above 0.996). CONCLUSIONS: the use of a rotational step of 15º or less provides reliable readings of endometrial volume: there were no significant differences between the means calculated by the two observers, associated with highintraclass correlation coefficient (>0.996). We recommend the 15º step rotation because it is quicker to be performed than 6º and 9º.