Accuracy of the standard systematic 12-core transrectal ultrasound-guided biopsy on a prostate phantom model

ABSTRACT

OBJECTIVE@#The detection rate of the current standard systematic 12 core transrectal ultrasound (TRUS)guided prostate biopsy remains low despite numerous modifications of the technique. This non-randomized experimental study evaluated the accuracy of standard TRUS-guided systematic prostatebiopsy as performed by selected urologists in obtaining samples representative of the peripheral zoneof the prostate, by analyzing virtual biopsies performed on a prostate phantom model.@*MATERIALS AND METHODS@#Thirty (30) urologists (26 consultants and 4 senior residents) were invited toperform two consecutive simulation TRUS guided 12-core biopsies on a phantom prostate model.The task was to hit twelve equal sized spherical targets which would correspond to the lateral andextreme lateral areas of the base, mid gland and apex of the peripheral zone of the phantom prostate,which would represent the usual biopsy technique. Degree of agreement (kappa) was computed.Eight (8) operators had below satisfactory kappa values and were excluded from the succeedinganalysis. Accuracy was calculated by dividing the number of accurately hit targets by the number ofvirtual cores (12). Data were encoded in MS Excel and Stata MP v.14 was used for data analysis.@*RESULTS@#Overall, the mean accuracy was 63.17% and median accuracy was 60% (95% CI 49.2-65.15)for the 22 operators included in the study. The lateral regions, particularly the midgland (95.8%-100% accuracy) were the most frequently biopsied areas and were often resampled. The targets at theprostatic base were missed by most operators (36.05% accuracy).@*CONCLUSION@#Systematic TRUS guided prostate biopsy, in the manner that it is performed, has itsinherent flaws, compounded by limitations in imaging capability and intra-operator variabilityresulting in low accuracy rates. A shift to newer prostate biopsy technique and methodologies withsignificantly higher accuracy rates is recommended.