1738 COMPARISON OF PROSTATE VOLUME MEASUREMENTS BY TRANSRECTAL ULTRASONOGRAPHY USING HEIGHTS OBTAINED BY DIFFERENT SCANNINGS: IS′SALAMI EFFECT′ REAL?

Abstract

INTRODUCTION AND OBJECTIVES: When prostate volume is measured, prolate ellipsoid volume calculation (height length width /6) is fast and precise, and its universal availability makes it practical for routine clinical application. But, there are ongoing debates on whether to use maximal height of prostate measured in the axial or mid-sagittal view in measuring prostate volume via transrectal ultrasonography (TRUS). We compared the differences according to the size of prostate in TRUS prostate volumes calculated using the prolate ellipsoid volume formula, with the height from axial and mid-sagittal scannings. METHODS: A total of 527 consecutive patients who underwent TRUS were included for analysis. Using prolate ellipsoid formula, prostate width was measured on an axial scanning, and prostate length on a mid-sagittal scanning. Prostate height was recorded in the axial scanning at a point perpendicular to ptostate width and in the midsagittal scanning at a point perpendicular to prostate length, resulting in two different volumes (V1 and V2, respectively). Another height was recorded as the greatest vertical diameter in the mid-sagittal scanning (V3) for representing salami effect . Salami effect means that a larger slice is obtained when salami is cut with the inclination of the knife. RESULTS: Mean SD of V1, V2 and V3 was 31.19 13.47, 30.98 12.88 and 31.43 13.46 respectively. In the paired t-test, V1 showed no significant difference compared to V2 (p 0.067). V1 and V2 were observed to be significantly smaller than V3 (p 0.005 and p 0.001). For prostate glands 50ml, V1 (27.78 7.70) and V2 (27.84 7.65) were smaller than V3 (28.12 7.92) (p 0.001). There was no significant difference between V1 and V2 (p 0.538). Meanwhile, for larger glands ( 50ml), V1 (63.66 13.45) and V3 (62.96 14.64) were larger than V2 (60.88 14.29) (p 0.001). V1 was not different from V3 (p 0.151). Similarly, for prostate glands with intravesical prostatic protrusion (IPP 5mm), V1 (49.49 17.62) and V3 (49.34 17.93) were larger than V2 (47.54 17.14) (p 0.001). CONCLUSIONS: When prostate glands are large or intravesical protruded, V1 was larger than V2 and showed no significant difference compared to V3. In these cases, prostate volume tended to be overestimated by TRUS using height obtained with axial scanning due to salami effect . For prostate glands 50ml, no difference between V1 and V2 was observed. Therefore, maximal height should be obtained with mid-sagittal scanning to improve the accuracy of volumetric measurements.