The method of using an endorectal coil in prostate Magnetic Resonance Imaging (MRI) has been an area of debate between radiologists for some time now. In this retrospective project our main objective is to identify if the deforming effects caused by the use of the endorectal coil in a prostate MRI procedure may affect accuracy of biopsy targeting using MR-US fusion biopsy technique. We will collect the dimensions measured by MRI and Transrectal Ultrasound (TRUS) and use the difference between the two as proxy for endorectal coil distortion. We will then analyze if target misses occur significantly more when this difference is greatest.

In this study we included 1) patients who had an MRI with endorectal coil and then underwent targeted MR-US fusion biopsy; and 2) had complete measurements on both MRI and TRUS. We excluded 1) all cases when the MR target and TRUS biopsy results in the same TRUS sextants were negative, since we are unable to assume if the MR targeted biopsy was a miss or a hit; 2) patients who underwent only an MR targeted biopsy or only a standard TRUS biopsy; 3) patients who did not have targets identified on MRI; and 4) patients with incomplete measurements in the records.

A target is considered a miss when the Gleason score of the target is less than the Gleason score of the TRUS biopsy from the same sextant. A target is considered a hit if it had a higher or equal Gleason score than the TRUS from the same sextant. In cases where two target lesions are identified in the same TRUS sextant, we excluded the target with the higher Gleason score, so as to assume a worst-case scenario for target accuracy. We calculated the anteroposterior (AP) deformity which is the difference in the measured AP dimension between MRI and TRUS. We compared the AP difference between missed targets and hit targets. A Mann-Whitney U test was used to test for statistical differences.

Note: To verify our assumption that the AP difference in measurement between the MRI and US represents deformity from the coil, we conducted a side analysis comparing the deformity of our study cases (129) which had the endorectal coil to MRI cases that were performed without the endorectal coil and also had MR-US fusion biopsy at our institution in the same time frame (8). This side analysis confirmed that there was statistically significant difference in the AP dimension between the endorectal cases and the non endorectal cases (P<0.05). The difference in the transverse dimension was not statistically significant and therefore was not included in our analysis. The cranial-caudal dimension was also not included based on our observation that US is unreliable in assessing the superior extent of the prostate, as well as the fact that our analysis evaluates the miss rate in the axial plane.

We started with a total of 229 patients, but this shrunk to 129 patients (total targets = 206) after applying our exclusion criteria. 81 targets were considered a hit while 9 targets were considered a miss. Fig. 1 shows a box plot of AP dimension differences for each group. Differences were not considered significant between the miss and hit groups (p= 0.154).It is well observed that an endorectal coil causes deformity in the prostate which we estimated by the change in AP dimension between MRI and US. Our side analysis comparing the AP change in cases using endorectal coil to cases with no endorectal coil suggest that this change is due at least in part to the endorectal coil as opposed to modality difference. Our comparison of “miss” and “hit” targets found no significant differences in the deformity between the 2 groups suggesting that deformity caused by the endorectal coil may not significantly affect the technical success rate of the targeted biopsy. However, our data was limited by the small number of missed targets. It is also possible that the true miss rate was underestimated since we were unable to compare against the true gold standard, whole-mount prostatectomy pathology.We can conclude that the missed target rate is likely not affected by the deformity from the endorectal coil.