Comparison of prostate biopsy schemes by computer simulation

Citation
Me. Chen et al., Comparison of prostate biopsy schemes by computer simulation, UROLOGY, 53(5), 1999, pp. 951-960
Citations number
18
Categorie Soggetti
Urology & Nephrology
Journal title
UROLOGY
ISSN journal
00904295 → ACNP
Volume
53
Issue
5
Year of publication
1999
Pages
951 - 960
Database
ISI
SICI code
0090-4295(199905)53:5<951:COPBSB>2.0.ZU;2-D
Abstract
Objectives. To compare the ability of different biopsy schemes to detect ca ncer and predict tumor volume using our previously described prostate biops y simulation system. In addition, we used the simulation system to evaluate the optimal location of transition zone biopsies. Methods. Digital reconstructions of 180 radical prostatectomy specimens wer e used. Forty simulations were performed on each prostate for 10 biopsy sch emes, including a previously reported five-region peripheral zone biopsy pa ttern and a new 11-core multisite-directed scheme consisting of sextant, tw o transition zone, one midline, and two anterior horn biopsies. For simulat ion of the transition zone biopsies, paired near-midline biopsies were simu lated, with needle insertion points from the apex to the base of the prosta te and with needle advances of 1 to 4 cm before firing. A total of 1,180,80 0 individual biopsy tracks were simulated. Results, The 11-core multisite-directed biopsy scheme had the highest detec tion rate for cancers greater than 0.5 cc. This scheme reliably detected ca ncer in 94% (138 of 147) of the cases. These results were significantly bet ter than those of the sextant biopsy scheme (P <0.001) and the five-region 18-core peripheral zone scheme (P = 0.03). Compared with other schemes, the re were increases in small-volume (0.5 cc or less) cancer detection by both the 11-core multisite-directed and five-region schemes, but they were not statistically significant. The multisite and the sextant plus four transiti on zone biopsy schemes had the best correlation of mean total core cancer l ength with total cancer volume. In the simulation of the transition zone bi opsies, the highest detection rate was observed when the biopsies were init iated at the most apical section and inserted for a depth of 3 cm before fi ring. Conclusions. Our simulation results suggest that the detection rate of pros tate biopsies is not related solely to the number of cores taken. Core plac ement (the regions of the prostate from which samples are taken) is also im portant. The 11-core multisite-directed biopsy scheme performed the best, w ith improved cancer detection rates and tumor volume correlation over other schemes. On the basis of our simulations, this scheme has been chosen for clinical evaluation. UROLOGY 53: 951-960, 1999. (C) 1999, Elsevier Science Inc, All rights reserved.