Background/Purpose: Digital tomosynthesis is a recently developed imaging method in which multiple projected images obtained at different angles are collected with a digital detector. These projected images are used to reconstruct multiple thin planar images of the scanned object in a plane parallel to the detector.  Recent reports indicate that this technology allows low-cost, low-radiation detection of bone erosion in patients with inflammatory arthritis.  The aim of this study was to determine the sensitivity and reproducibility of tomosynthesis, compared with conventional computed tomography (CT), for measurement of bone erosion in gout.

Methods: Tomosynthesis and CT scans of the dominant wrist were prospectively obtained in 37 patients with gout (mean disease duration 18 years, 86% with subcutaneous tophi) (Figure). Each scan was scored separately by two radiologists for bone erosion using semi-quantitative volume assessment (sites and 0-10 scoring as for RAMRIS erosion method).

Results: The mean total erosion score in the tomosynthesis scans was lower than CT scans (10.8 vs 13.2, p=0.004).  For many individual sites (radius, scaphoid, triquetrum, trapezium, capitate, MC1 base), tomosynthesis erosion scores were significantly lower than CT scores.  There were no sites where tomosynthesis scores were higher than CT.  The inter-reader intraclass correlation coefficients (ICC) for the total erosion score was 0.64 (0.59-0.89) for tomosynthesis and 0.80 (0.64-0.94) for CT. At all individual sites except MC2 base, ICCs were lower for tomosynthesis than CT.  We were unable to identify a combination of sites for which mean scores or ICCs for tomosynthesis approximated those of CT.  There was a high correlation between tomosynthesis and CT total scores (r=0.95, p<0.0001) and at most individual sites. Total erosion scores for both tomosynthesis and CT correlated highly with hand tophus count, grip strength, and Health Assessment Questionnaire scores (p<0.001 for all). 

Conclusion: Tomosynthesis erosion scores correlate highly with CT erosion scores and clinical measures of disease severity in patients with gout.  However, tomosynthesis has lower reproducibility and detection of erosion compared with CT.  These findings indicate that CT remains the gold standard for measurement of bone erosion in gout.

Figure: A. Tomosynthesis and B. corresponding coronal CT images of the right wrist in a participant with tophaceous gout.