The presence of ultrasound (US)-detected bone erosions is not specific for rheumatoid arthritis (RA), but erosions of RA are generally more extensive than for other diseases.
The specificity of US-detected bone erosions for RA was only 32.9% in a study by investigators from Denmark and the United Kingdom. However, when considering only four suggested target joints, the presence of erosions with a semiquantitative score ≥2 in any one of them improved specificity to 97.9
“The results of this study demonstrate that although the overall presence of US-detected bone erosions are more frequent in RA compared with other common joint diseases, they are not necessarily a disease-specific finding,” the authors wrote in Annals of the Rheumatic Diseases. “However, the presence of erosions with score ≥2 in any of the four suggested target joints (second and third MCP [metacarpophalangeal], fifth MTP [metatarsalphalangeal] joints and distal ulna) was highly specific for RA. The presence of any erosion in the fifth MTP joint was both sensitive and specific for RA.”
A total of 310 subjects were recruited: 70 with RA (28 with early RA, 42 with established RA), 60 with psoriatic arthritis (PsA), 60 with gout, 60 with osteoarthritis (OA) and 60 healthy volunteers.
The following areas were examined bilaterally in each patient by US: distal radius and ulna, second, third, and fifth MCP, second and third proximal interphalangeal, and first and fifth MTP joints. All joints were scanned in four quadrants and scored using both semiquantitative and quantitative (erosion diameter) systems. The semiquantitative system assigned the following values based on erosion size: 0=no erosions, 1=erosions covering less than one third of the surface of the quadrant, 2=erosions covering one third to two thirds of the surface of the quadrant, and 3=erosions covering more than two thirds of the surface of the quadrant. This scoring system was developed by the authors.
RA patients had twice as many joints with erosions than patients with PsA (incident rate ratio 2.50, 95% CI 1.87-3.35, P<0.001) or gout (2.28, 95% CI 1.71-3.04, P<0.001), and five times as many as those with OA (5.41, 95% CI 3.92-7.47, P<0.001). All four disease groups were found to have more affected joints than the healthy volunteers.
RA patients had more joints with extensive erosions (overall score ≥2) than patients with PsA patients (incident rate 4.31, 95% CI 2.08-8.96, P<0.001), gout (6.67, 95% CI 3.23-13.77, P<0.001) and OA (32.61, 95% CI 11.76-90.46, P<0.001). None of the healthy volunteers had any joints with scores ≥2 for erosion.
The median total erosion score was higher in RA than in PsA (adjusted difference 3.65, 95% CI 1.86-5.44, P<0.001), gout (3.90, 95% CI 2.09-5.71, P<0.001) and OA (5.90, 95% CI 4.11 to 7.70, P<0.001). Patients with RA, PsA, or gout had higher total erosion scores than the healthy volunteers, but OA patients did not differ from the healthy volunteers.
The median maximum erosion diameter in any joint was higher in RA patients than in patients with PsA (adjusted difference 1.79 mm, 95% CI 1.13-2.46 mm, P<0.001), gout 0.99, 95% CI 0.12-1.86, P=0.026) or OA (2.84, 95% CI 1.87-3.81, P<0.001). Patients with RA, PsA, or gout had larger erosions than the healthy volunteers (all P<0.001), but OA patients did not differ from healthy volunteers.
Patients with early RA had significantly fewer joints with erosions than established RA, but there was no substantive difference in the median diameter of erosions between these groups.
The presence of any joint with any degree of erosion (score >0) had only a 32.9% specificity for RA but had 91.4% sensitivity. The presence of any joint with more extensive erosions (score ≥2), however, was specific for RA (89.2%) but had only 50% sensitivity. Excluding the established RA patients reduced the sensitivity of an erosion score ≥2 to 25%.
Small erosions had lower specificity for RA. A receiver operating curve curve (ROC) analysis found that the diameter of the largest erosion in any joint predicted RA (area under the ROC curve 0.73, 95% CI 0.65-0.80, P<0.001). The sensitivity and specificity were maximized at a cut-off of ≥2.5 mm (sensitivity 68.6%, specificity 68.3%).
An erosion score ≥2 in the distal ulna, second MCP, fifth MCP or fifth MTP joints was highly specific for RA (97.9%) while being moderately sensitive (41.4%).
Erosions of any size of the fifth MTP joint had high specificity (85.4%) and sensitivity (68.6%) for RA.
The authors suggest that a focused examination of the four suggested target areas may reduce the time required for US examination, and minimize the potential for false-positive results. “This may help in avoiding repeated investigations and preventing unnecessary follow-up when potential false-positive erosions are found in the absence of supporting symptoms and signs,” they wrote.
The study’s limitations include a “mildly lower” specificity of erosions when normal subjects were excluded. There were also marginal differences in age between the groups, but comparisons were adjusted for age. Also, there was mild variance in the duration of symptoms between groups.