Comparison of diagnostic performance between two prostate imaging reporting and data system versions: A systematic...

To compare the diagnostic performance between Prostate Imaging Reporting and Data System version 1(PI-RADS v1) and PI-RADS v2 for detection of prostate cancer (PCa).

A systematic literature search was performed from inception to September 31, 2018, in following databases MEDLINE, EMBASE, Cochrane Library, Google Scholar, in addition to Chinese National Knowledge Infrastructure (CNKI) and Wanfang Data database. Sensitivity and specificity of individual studies along with summary estimates were calculated and presented in forest plots. Multiple subgroup analyses and meta-regression were performed to investigate the heterogeneity. Quality of included studies was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool.

14 studies involving head to head comparison between PI-RADS v1 and v2 were included, with a total of 1682 patients. The pooled sensitivity for PI-RADS v1 and PI-RADS v2 were 85.37% (95% CI 79.29%-89.89%) and 87.79% (95% CI 79.84%-92.88%), with pooled specificity of 77.26% (95% CI 57.29%-89.58%) and 76.73% (95% CI 58.69%-88.44%), respectively. There was no substantial difference in diagnostic accuracy between PI-RADS v1 and PI-RADS v2 (P = 0.57 for sensitivity and P = 0.96 for specificity). Multiple subgroup analyses and meta-regression suggested these two scoring systems had comparable diagnostic performance on magnetic field strength, zonal anatomy, and outcome assessment. For the transitional zone, it seemed that PI-RADS v2 had higher sensitivity than PI-RADS v1 (90.1% vs. 80.59%), but the difference was not substantial (P = 0.17).

PI-RADS v2 has slightly higher sensitivity but at the expense of minor decreased specificity. Thus, on the whole PI-RADS v1 and PI-RADS v2 have comparable diagnostic accuracy.