Inter- and Intra-Observer Variability in Prostate Definition With Tissue Harmonic and Brightness Mode Imaging

• Published in: International journal of radiation oncology, biology, physics Vol. 82; no. 1; pp. e9 - e16
• Main Authors: Sandhu, Gurpreet Kaur, Ph.D, Dunscombe, Peter, Ph.D, Meyer, Tyler, Ph.D, Pavamani, Simon, M.D, Khan, Rao, Ph.D
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 2012
• Subjects: Aged; BRACHYTHERAPY; BRIGHTNESS; COMPARATIVE EVALUATIONS; Hematology, Oncology and Palliative Medicine; Humans; Image Enhancement - methods; Interobserver; Intraobserver; Male; Middle Aged; Neoplasm Staging; Observer Variation; Organ Size; PATIENTS; PROSTATE; Prostate - diagnostic imaging; Prostate - pathology; Prostate brachytherapy; Prostatic Neoplasms - diagnostic imaging; Prostatic Neoplasms - pathology; Prostatic Neoplasms - radiotherapy; Prostatic transrectal ultrasonography; Radiology; RADIOLOGY AND NUCLEAR MEDICINE; Tissue harmonic imaging; Tumor Burden; ULTRASONOGRAPHY; Ultrasonography - methods; Prostatic transrectal ultrasonography; Intraobserver; Tissue harmonic imaging; Prostate brachytherapy; Interobserver
• ISSN: 0360-3016; 1879-355X
• DOI: 10.1016/j.ijrobp.2011.02.013

Purpose The objective of this study was to compare the relative utility of tissue harmonic (H) and brightness (B) transrectal ultrasound (TRUS) images of the prostate by studying interobserver and intraobserver variation in prostate delineation. Methods and Materials Ten patients with early-stage disease were randomly selected. TRUS images of prostates were acquired using B and H modes. The prostates on all images were contoured by an experienced radiation oncologist (RO) and five equally trained observers. The observers were blinded to information regarding patient and imaging mode. The volumes of prostate glands and areas of midgland slices were calculated. Volumes contoured were compared among the observers and between observer group and RO. Contours on one patient were repeated five times by four observers to evaluate the intraobserver variability. Results A one-sample Student t -test showed the volumes outlined by five observers are in agreement ( p > 0.05) with the RO. Paired Student t -test showed prostate volumes ( p  = 0.008) and midgland areas ( p  = 0.006) with H mode were significantly smaller than that with B mode. Two-factor analysis of variances showed significant interobserver variability ( p < 0.001) in prostate volumes and areas. Inter- and intraobserver consistency was quantified as the standard deviation of mean volumes and areas, and concordance indices. It was found that for small glands (≤35 cc) H mode provided greater interobserver consistency; however, for large glands (≥35 cc), B mode provided more consistent estimates. Conclusions H mode provided superior inter- and intraobserver agreement in prostate volume definition for small to medium prostates. In large glands, H mode does not exhibit any additional advantage. Although harmonic imaging has not proven advantageous for all cases, its utilization seems to be judicious for small prostates.