OBJECTIVE

To determine the factors contributing to the fragmentation of prostatic core biopsies containing prostatic carcinoma, which might complicate the quantification of the number cores with cancer and/or the amount of tumour in a core.

MATERIALS AND METHODS

Prostate biopsy cases containing fragmented cores with cancer and cancer cases present only in unfragmented cores were sought in records from the consultant service of one of the authors. A 'part' corresponded to the site-specific jar submitted by the urologist (i.e. left or right; left apex, left mid, etc.). Cases of prostatic adenocarcinoma with fragmented cores containing cancer (463) and cases with cancer lacking fragmented cores (200) were compared.

RESULTS

The mean number of parts per case was 8.1 and 7.5 in the unfragmented and fragmented cases, respectively (not significant). The mean number of cores per part was significantly higher in the fragmented group than the unfragmented group (2.6 vs 2.1, P = 0.004). The number of parts containing cancer was higher in cases with fragmented cores than in cases with unfragmented cores (2.8 vs 1.6, P < 0.001). A higher mean Gleason score was associated with the cancer in fragmented cores than in unfragmented cores (6.6 vs 6.2, P < 0.001). Multivariate regression analysis showed that fragmentation was associated with the number of parts with cancer (P < 0.001), cores per part (P < 0.001) and Gleason score (P < 0.04).

CONCLUSIONS

The number of parts per case with cancer, number of cores submitted per part, and Gleason score contribute to the likelihood of fragmentation of cores containing prostatic adenocarcinoma.