Single source dual-energy computed tomography in the diagnosis of gout: Diagnostic reliability in comparison to digital radiography and conventional computed tomography of the feet

Tobias Kiefer,Torsten Diekhoff,Sandra Hermann,Andrea Stroux,Jürgen Mews,Jörg Blobel,Bernd Hamm,Kay-Geert Hermann
DOI: https://doi.org/10.1016/j.ejrad.2016.08.004
Abstract:Objectives: To investigate the diagnostic value of single-source dual-energy computed tomography (SDECT) in gouty arthritis and to compare its capability to detect urate depositions with digital radiography (DR) and conventional computed tomography (CT). Methods: Forty-four patients who underwent SDECT volume scans of the feet for suspected gouty arthritis were retrospectively analyzed. SDECT, CT (both n=44) and DR (n=36) were scored by three blinded readers for presence of osteoarthritis, erosions, and tophi. A diagnosis was made for each imaging modality. Results were compared to the clinical diagnosis using the American College of Rheumatology (ACR) classification criteria. Results: The patient population was divided into a gout (n=21) and control (n=23) group based on final clinical diagnosis. Osteoarthritis was evident in 15 joints using CT and 30 joints using DR (p=0.165). There were 134 erosions detected by CT compared to 38 erosions detected by DR (p<0.001). In total 119 tophi were detected by SDECT, compared to 85 tophi by CT (p=0.182) and 25 tophi by DR (p<0.001). SDECT had best diagnostic value for diagnosis of gout compared to DR and conventional CT (sensitivity and specificity for SDECT: 71.4% and 95.7%, CT: 71.4% and 91.3% and DR: 44.4% and 83.3%, respectively). For all three readers, Cohen's kappa for DR and conventional CT were substantial for all scoring items and ranged from 0.75 to 0.77 and 0.72-0.76, respectively. For SDECT Cohen's kappa was good to almost perfect with 0.77-0.84. Conclusions: SDECT is capable to detect uric acid depositions with good sensitivity and high specificity in feet, therefore diagnostic confidence is improved. Using SDECT, inter-reader variance can be markedly reduced for the detection of gouty tophi.
What problem does this paper attempt to address?