Diagnostic performance and inter-reader reliability of bone reporting and data system (Bone-RADS) on computed...

Area of Science:

• Radiology
• Oncology
• Orthopedics

Background:

• Solitary bone lesions require accurate characterization for diagnosis and treatment planning.
• The Bone Reporting and Data System (Bone-RADS) is a standardized lexicon for classifying bone lesions on CT.
• Evaluating the diagnostic performance and reliability of Bone-RADS is crucial for its clinical application.

Purpose of the Study:

• To assess the diagnostic performance of the Bone-RADS for solitary bone lesions on CT.
• To determine the inter-reader reliability of the Bone-RADS classification system.
• To compare the diagnostic accuracy of Bone-RADS for lucent versus sclerotic/mixed bone lesions.

Main Methods:

• Retrospective analysis of 179 patients with solitary bone lesions on CT.
• Two radiologists independently categorized lesions using the Bone-RADS (1=benign, 4=malignant).
• Histopathology results served as the reference standard for diagnostic performance calculations; inter-reader reliability was assessed using Cohen's kappa.

Main Results:

• Bone-RADS demonstrated high sensitivity (95% and 82%) for malignant lucent lesions but low specificity (11%) for both readers.
• For sclerotic/mixed lesions, sensitivities were 75% and 68%, with specificities of 27% for both readers.
• Overall inter-reader reliability was good (κ=0.744), with moderate to very good agreement for lucent (κ=0.565) and sclerotic/mixed (κ=0.851) lesions.

Conclusions:

• Bone-RADS exhibits high sensitivity for detecting malignancy in lucent bone lesions and demonstrates good inter-reader reliability.
• The system shows poor specificity and accuracy for both lucent and sclerotic/mixed bone lesions.
• Clinical factors like pain and history of malignancy may influence the algorithms, potentially impacting specificity and accuracy.