Accuracy of ultrasound in estimation of prostate weight: comparison of urologists and radiologists

• Published in: Canadian journal of urology Vol. 17; no. 1; p. 4985
• Main Authors: Nunez-Nateras, Rafael, Andrews, Jack R, Martin, George L, Andrews, Paul E, Humphreys, Mitchell R, Ferrigni, Robert G, Eversman, William G, Castle, Erik P
• Format: Journal Article
• Language: English
• Published: Canada 01.02.2010
• Subjects: Adenocarcinoma - diagnostic imaging; Aged; Aged, 80 and over; Clinical Competence; Humans; Male; Middle Aged; Organ Size; Prostate - diagnostic imaging; Prostate - pathology; Prostatectomy; Prostatic Neoplasms - diagnostic imaging; Prostatic Neoplasms - pathology; Radiology; Ultrasonography; Urology
• ISSN: 1195-9479

Measurements of prostate size are obtained to contribute in the diagnosis and follow up of patients with a variety of diseases. Since its introduction, transrectal ultrasonography (TRUS) of the prostate has become the most common method for assessment of prostate volumes. Ultrasonography, in general, has been associated with concerns of operator dependent variability. Herein, we analyze the accuracy of urologists and radiologists performing TRUS. The accuracy of preoperative TRUS prostate volume estimation was evaluated by comparing it to gross specimen prostate weight following robot-assisted radical prostatectomy (RARP) performed from August 2004 to March 2008 in Mayo Clinic Arizona. A total of 800 RARPs were evaluated retrospectively with 302 patients having a prostate volume measurement with TRUS at our institution followed by RARP being performed within 30 days. The TRUS measurements were divided into two groups: those TRUS measurements performed by urologists (group 1), and those performed by radiologists (group 2). The accuracy of the two groups were compared using a Pearson correlation analysis. The estimated weight by TRUS in the total cohort of patients correlated with the pathological specimen weight at 0.802 with a standard error of 0.90. Group 1 performed a total of 114 ultrasounds with a correlation of 0.835 and a standard error of 1.27. Group 2 performed a total of 188 with a correlation of 0.786 and a standard error of 0.88. Urologists and radiologists are both consistently within 17%-22% of the estimated prostate specimen weight. Urologists appeared to have a slightly higher accuracy in estimation but a higher range of error for the whole group when compared to radiologists. Transrectal ultrasonography is a reliable technique to estimate prostate weight and accuracy to within 20% of the pathological weight. Urologists and radiologists are essentially equally proficient in estimating prostate weight with TRUS. These findings are particularly important with respect to specialty certification and competency/proficiency evaluation, as health care increasingly moves towards outcomes based reimbursement.