Comparison of Bosniak Classification of cystic renal masses version 2019 assessed by CT and MRI.

2021 
OBJECTIVE To compare imaging features in cystic masses imaged with both CT and MRI using Bosniak Classification version 2019 (Bosniak.v2019) and original Bosniak Classification (Bosniak.original). MATERIALS AND METHODS This IRB-approved, retrospective, cross-sectional study evaluated sixty-five consecutively identified cystic (≤ 25% enhancing) masses imaged by CT and MRI between 2009 and 2019: 35 with histologic diagnosis and 30 Bosniak.v2019 Class 2 and Class 2F cystic masses verified by an expert radiologist (R1) with minimum 5-year stability. Three radiologists (R2, R3, R4) independently evaluated CT, followed by MRI and assigned Bosniak.original and Bosniak.v2019 class in two sessions separated by ≥ 1 month and assessed the following: septa number, septa/wall thickness, and protrusions. Discrepancies were resolved by consensus with R1. RESULTS There was 70.8% agreement (kappa = 0.60, p = 0.0146) in class assigned by CT versus MRI for Bosniak.original and 72.3% agreement (kappa = 0.63, p = 0.006) for Bosniak.v2019. Increased septa number (p < 0.001) and more protrusions (p = 0.034) were identified on MRI, with no differences in septal/wall thickness (p = 0.067, 0.855) or protrusion size (p = 0.467). For both CT and MRI, Bosniak.v2019 improved specificity (79.0% [95% confidence interval 71.0-87.0%] CT, 70% [62.0-77.0%] MRI) compared to Bosniak.original (63.0% [56.0-69.0%] CT, 66.0% [58.0-74.0%] MRI) with maintained sensitivity and higher overall accuracy. Inter-observer agreement was similar-to-slightly higher for Bosniak.v2019 (K = 0.44 CT, 0.39 MRI) versus Bosniak.original (K = 0.35 CT, 0.37 MRI). CONCLUSION Class assignment differs in cystic masses evaluated by CT versus MRI for original and v2019 Bosniak Classification with similar-to-slightly higher agreement and improved specificity and higher overall accuracy on both CT and MRI with Bosniak version 2019.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    21
    References
    0
    Citations
    NaN
    KQI
    []