Comparison of initial and tertiary centre second opinion reads of multiparametric magnetic resonance imaging of the prostate prior to repeat biopsy

Abstract

Objectives

To investigate the value of second-opinion evaluation of multiparametric prostate magnetic resonance imaging (MRI) by subspecialised uroradiologists at a tertiary centre for the detection of significant cancer in transperineal fusion prostate biopsy.

Methods

Evaluation of prospectively acquired initial and second-opinion radiology reports of 158 patients who underwent MRI at regional hospitals prior to transperineal MR/untrasound fusion biopsy at a tertiary referral centre over a 3-year period. Gleason score (GS) 7-10 cancer, positive predictive value (PPV) and negative (NPV) predictive value (±95 % confidence intervals) were calculated and compared by Fisher’s exact test.

Results

Disagreement between initial and tertiary centre second-opinion reports was observed in 54 % of cases (86/158). MRIs had a higher NPV for GS 7-10 in tertiary centre reads compared to initial reports (0.89 ± 0.08 vs 0.72 ± 0.16; p = 0.04), and a higher PPV in the target area for all cancer (0.61 ± 0.12 vs 0.28 ± 0.10; p = 0.01) and GS 7-10 cancer (0.43 ± 0.12 vs 0.2 3 ± 0.09; p = 0.02). For equivocal suspicion, the PPV for GS 7-10 was 0.12 ± 0.11 for tertiary centre and 0.11 ± 0.09 for initial reads; p = 1.00.

Conclusions

Second readings of prostate MRI by subspecialised uroradiologists at a tertiary centre significantly improved both NPV and PPV. Reporter experience may help to reduce overcalling and avoid overtargeting of lesions.

Key Points

• Multiparametric MRIs were more often called negative in subspecialist reads (41 % vs 20 %) .

• Second readings of prostate mpMRIs by subspecialist uroradiologists significantly improved NPV and PPV .

• Reporter experience may reduce overcalling and avoid overtargeting of lesions.

• Greater education and training of radiologists in prostate MRI interpretation is advised.

Electronic supplementary material

The online version of this article (doi:10.1007/s00330-016-4635-5) contains supplementary material, which is available to authorized users.