Three dimensional transrectal ultrasound imaging of the prostate: initial experience with an emerging technology

PURPOSE: A three-dimensional ultrasound system (3-D US) was evaluated for its clinical utility in transrectal prostate imaging, in comparison with the current standard 2-dimensional transrectal ultrasound (TRUS) imaging system. METHODS AND MATERIALS: The computer program developed in our laboratory was coupled with a commercially available ultrasound transducer. Geometric validation and volumetric assessment was performed with "stretched-string" wire models and solution-containing balloons respectively. Anatomic correlation of 3-D TRUS images was performed with cadaveric prostates. Intraprostatic lesion localization by 3D-TRUS was assessed clinically by 2 observers in 11 patients prior to radical prostatectomy and the data compared with those yielded by 2-D TRUS. RESULTS: Geometric assessment by 3D TRUS in comparison with the "between strings in the phantom" model (true dimensions) had an error of up to 1.2%. Volume measurement by 3-D TRUS had an error, compared to the true volume, of 0.9%. The correlation coefficient (r) was 0.99985 for the end-firing probe and 0.978 for side firing. The 3-D images provided accurate representation of the true anatomy in the sagittal, transverse and most uniquely, the coronal plane. Two observers achieved better diagnostic accuracies with intraprostatic abnormalities using 3-D instead of standard 2-D TRUS. The negative predictive value and the specificity were improved. CONCLUSION: 3-D TRUS appears to provided accurate representation of the true anatomy with geometric and volumetric validation. Areas of potential clinical application of 3-D TRUS include treatment monitoring with volume measurements and various intervention and therapeutic procedures for both benign and malignant prostatic disorders.