Long-distance longitudinal prostate MRI quality assurance: from startup to 12 months

To evaluate a 12-month long-distance prostate MRI quality assurance (QA) program.

The need for IRB approval was waived for this prospective longitudinal QA effort. One academic institution experienced with prostate MRI [~ 1000 examinations/year (Site 2)] partnered with a private institution 240 miles away that was starting a new prostate MRI program (Site 1). Site 1 performed all examinations (N = 249). Four radiologists at Site 1 created finalized reports, then sent images and reports to Site 2 for review on a rolling basis. One radiologist at Site 2 reviewed findings and exam quality and discussed results by phone (~ 2-10 minutes/MRI). In months 1-6 all examinations were reviewed. In months 7-12 only PI-RADS ≤ 2 and 'difficult' cases were reviewed. Repeatability was assessed with intra-class correlation (ICC). 'Clinically significant cancer' was Gleason ≥ 7.

Image quality significantly (p < 0.001) improved after the first three months. Inter-rater agreement also improved in months 3-4 [ICC: 0.849 (95% CI 0.744-0.913)] and 5-6 [ICC: 0.768 (95% CI 0.619-0.864)] compared to months 1-2 [ICC: 0.621 (95% CI 0.436-0.756)]. PI-RADS ≤ 2 examinations were reclassified PI-RADS ≥ 3 in 19% (30/162); of these, 23 had post-MRI histology and 57% (13/23) had clinically significant cancer (5.2% of 249). False-negative examinations [N = 18 (PI-RADS ≤ 2 and Gleason ≥ 7)] were more common at Site 1 during months 1-6 [9% (14/160) vs. 4% (4/89)]. Positive predictive values for PI-RADS ≥ 3 were similar.

Remote quality assurance of prostate MRI is feasible and useful, enabling new programs to gain durable skills with minimal risk to patients.