Study Type – Diagnosis (exploratory cohort)
Level of Evidence 2b
OBJECTIVES
To investigate whether there are any differences between preoperative magnetic resonance imaging (MRI) and the final pathology results, based on the time between biopsy and preoperative MRI, as there are reports recommending ≥3 weeks after a prostate biopsy, primarily because haemorrhaging interferes with the interpretation of MRI.
PATIENTS AND METHODS
Between December 2007 and December 2008, we retrospectively analysed 69 consecutive patients who had biopsy‐confirmed prostate cancer. The inclusion criteria for the study were a history of MRI investigation (combined T2‐weighted and diffusion‐weighted MRI) and robot‐assisted laparoscopic radical prostatectomy. The exclusion criteria included an MRI investigation‐to‐surgery interval of ≥1 week, a biopsy having been taken in another hospital, or other than 12 biopsy cores. The amount of haemorrhaging, number of haemorrhaging sites, and the location of the cancer were determined. For this, the prostate was divided into 12 segments which anatomically corresponded to the sites where the 12 core biopsies were taken. Each haemorrhagic prostate segment was scored according to its diameter. Pathology results were reviewed in the same manner. Finally, we assessed any discordance between the sets of results according to the period between the biopsy and the MRI. The association between the MRI and pathology results, in relation to the period between the biopsy and MRI, was plotted and tested using Pearson’s correlation coefficient.
RESULTS
Five of the 69 patients were excluded because they had a biopsy at another hospital, 12 were excluded because the period between the MRI and the surgery was >7 days. Suspected prostate haemorrhage was detected in 49 of 52 (94%) patients who had MRI. There was a significantly negative correlation with the period between biopsy and MRI (coefficient – 0.285, P= 0.041). There were no significant differences in cancer localization between MRI and pathology results according to the period between the biopsy and MRI (coefficient 0.028, P= 0.874). The rate of matching between MRI results and pathology results was 74%.
CONCLUSIONS
We found no significant differences in cancer localization between MRI and final pathology according to the period between the biopsy and MRI. Because of this finding, we do not recommend deferring MRI for the purpose of more accurate cancer staging.