COPENHAGEN — MRI, ultrasound, and digital x-ray radiogrammetry may all have a role in identifying which patients with early, nonerosive rheumatoid arthritis will later develop erosive disease, based on results from a pilot study with 27 patients.
Dr. Pernille Bøyesen reported the findings at the annual European Congress of Rheumatology.
“Our findings support that DXR-BMD [digital x-ray bone mineral density] change, MRI, and US [ultrasound] can all be useful in order to identify early RA patients who will develop radiographic erosions. In addition, 3-month DXR-BMD change might be better based on the significant differences between groups in this limited sample,” Dr. Bøyesen, a researcher in the department of rheumatology at Diakonhjemmet Hospital in Oslo, said in an interview. “The findings are in accordance with studies previously published that showed that DXR-BMD change, MRI, and US are associated with subsequent radiographic damage in RA patients.”
The study focused on 27 patients who had had RA for less than a year and had no joint destruction evident on a baseline hand x-ray. Their average age was 52 years (range, 41-61 years). Average disease duration was 100 days, their average disease activity 28 score (DAS28) was 3.49 (range, 2.4-4.8), and 79% were women.
Each patient underwent further assessment of their dominant wrist at baseline by MRI (with an RA MRI score calculated), and by gray-scale US (with their synovitis and tenosynovitis scored at 0-3). The researchers also obtained x-ray images of each patient's hands and wrists at baseline and at 3, 6, and 12 months. They scored the images by the van der Heijde/Sharp criteria, and applied DXR analysis to calculate the 3-month change in BMD of metacarpal cortical bone.
In all, 10 patients developed x-ray indications of erosive disease during the 1 year of follow-up. The 10 patients had a significantly higher DAS28 (average, 4.5) at baseline, compared with the 17 patients who did not progress (average, 2.7). The 10 patients with progressive disease also had significantly greater bone loss at 3 months as measured by the DXR-BMD assessments. Dr. Bøyesen and her associates also saw a numeric trend toward higher MRI and US scores in these 10 patients at baseline, compared with the 17 nonprogressors, but the differences were not statistically significant.
“We found that the 10 RA patients who became erosive had significantly higher DAS28 scores and 3-month DXR-BMD loss. This might point toward the possibility that DXR-BMD change is better at identifying patients at risk of developing x-ray erosions. However, the fact that the DXR-BMD change was larger in the erosive group by a nonparametric comparison does not implicate the superiority of this test in this context. In order to prove the effectiveness of the different methods, further statistical modeling is required, and due to our limited sample size this has not been possible,” according to Dr. Bøyesen.
“Our findings support that DXR-BMD change, MRI, and US can all be useful to identify early RA patients who will develop radiographic erosions. [The 3-month] DXR-BMD change might be the better test, based on the significant differences between groups in this limited sample,” she said.
By combining the various modalities, the investigators hope to “achieve higher accuracy in identifying RA patients who will develop erosive disease at a later time point. In order to appropriately address the roles of DXR, MRI, and US in clinical practice, we estimate that the sample would need [to comprise] approximately 100-150 RA patients.”
In response to a question from the audience after her presentation, Dr. Bøyesen acknowledged that cortical bone is known for being less susceptible to erosion. Why DXR-BMD detected early density loss in metacarpal cortical bone remains to be answered, she said.
Dr. Bøyesen also noted that both US and x-ray studies are relatively quick and inexpensive, and use equipment that is widely available. US has a slight advantage because it doesn't even expose patients to radiation. MRI examinations take longer, use equipment that is more expensive and less widely available, and require treatment with a contrast agent. However, MRI provides information on multiple structures affected by RA, such as bone marrow edema, synovitis, tenosynovitis, and erosions, she said.
Further analysis showed that findings from DXR-BMD change, MRI, and US were as accurate at predicting erosion as established biomarkers of disease intensity, including erythrocyte sedimentation rate, rheumatoid factor positivity, and anti-citrullinated protein antibody, she said.
Dr. Bøyesen reported no conflicts of interest.
MRI shows bone marrow edema (arrows) in a patient with early RA.
This ultrasound image indicates ulnar B-mode synovitis (arrow).