Background/Purpose:
Imaging and pathology studies have established the
close relationship between intraosseous tophus and bone erosion in gout.  The tophus is an organised structure consisting
of urate crystals and a chronic inflammatory soft tissue response.  It is currently unknown whether bone
erosion in gout occurs due to direct effects of urate crystals or indirect
effects of the soft tissue response. 
The aim of this work was to examine the relationship between bone
erosion and each component of the tophus. 

Methods: Plain radiographs and computed tomography (CT) scans of the feet were
prospectively obtained from 92 people with tophaceous gout. The ten metatarsophalangeal joints were scored on XR for Sharp-van
der Heijde erosion score (scale 0-10).  Conventional CT and dual energy CT (DECT)
images were assessed at each joint for the presence and total volume of tophus
(using manual outlining analysis of conventional CT images) and the presence and
volume of tophus urate deposition at the same joints (using automated volume
DECT analysis).  Tophus soft tissue
volume was calculated by subtracting the DECT tophus urate volume from the total
CT tophus volume.  In total, 918
joints were available for analysis.  The readers of the XR and CT/DECT were
blinded to each other’s scores and all clinical characteristics of the
patients.  Data were analysed using
generalized estimating equations to account for repeated measures in
participants, and mediation analysis to examine the direct and indirect effects
of urate volume on bone erosion score. 

Results:   XR erosion was present at
261 (28.4%) joints. Mean (SD) XR erosion score was 1.2 (2.3) in all joints and 4.1
(2.5) in those with XR erosion.  CT
tophus was present at 447 (48.7%) joints and DECT tophus urate was present at
443 (48.3%) joints.  CT tophus was
adjacent to bone in 74.7% of all XR erosions (odds ratio (OR) 10.1 [95% CI 6.8-14.8])
and DECT tophus urate was adjacent to bone in 64.8% of all XR erosions (OR 10.6
[95% CI 7.1-15.8]).  

Mean (SD) total tophus volume was 1.88
(4.25) cm³, urate volume was 0.27 (0.80) cm³, and soft
tissue volume was 1.61 (3.63) cm³.  In mediation analysis, urate volume and
soft tissue volume were directly associated with XR erosion score (Figure).  About a third of the association of urate
volume with XR erosion score was indirectly mediated through the strong
association between urate volume and soft tissue volume.

Conclusion:   Urate and soft tissue components
of the tophus are strongly and independently associated with bone erosion in
gout.  These data suggest that tophi
contribute to bone erosion in gout both through direct urate crystal effects
and through indirect effects of soft tissue on bone. 

Figure:
Summary of the mediation analysis. Values represent
standardized path coefficients (n=92). 
The directions of these associations were dictated by biological
experience.  R² for model
=0.52.