A comparison of CT scan to transrcctal ultrasound measured prostate volume in untreated prostate cancer

Purpose: To compare CT and TRUS measured prostate volumes in patients with untreated prostate cancer. Methods and Materials: From 1995 to 1999, 48 consecutive patients with prostate adenocarcinoma were treated with external beam radiation at the Portland Veterans Administration Hospital. In 36 of these patients. Transrectal Ultrasound (TRUS) and CT measurements of prostate volume were obtained prior to treatment, fewer than 6 months apart. TRUS volume was calculated using the prolate ellipsoid formula. CT volume was calculated from contours of the prostate drawn by one physician on axial CT images obtained at 3 or 5 mm intervals. The physician was blinded to the TRUS volume. Results: TRUS and CT prostate volume measurements strongly correlated (Pearson's correlation coefficient = 0.925, 95%CI 0.856-0.961, P < 0.0001). CT volume was consistently larger than TRUS volume by a factor of approximately 1.5. When grouped into quartiles by TRUS volume, the overestimation of prostate volume by CT was greatest for prostate volumes less than 22.1 cubic centimeters (r = 0.410, 95%CI 0.094-0.651, P = 0.0123). CT volumes were similarly correlated with TRUS volumes regardless of CT slice interval (3mm: r = 0.866, 95% Cl 0.601-0.959, n = 13; 5mm: r = 0.930 95%CI 0.840-0.970, n = 23). The CT slice interval had no effect on the ratio between CT and US volume. Conclusion: There is a strong correlation between CT scan and TRUS measurement of prostate volume. CT scan overestimates prostate volume by approximately fifty percent.