X-ray Interpretation: A Discrepancy?
X-ray Interpretation: A Discrepancy?
van den Berg R, Lenczner G, Feydy A, et al
Arthritis Rheumatol. 2014;66:2403-2411
In this study, van den Berg and colleagues compared local radiologist/rheumatologist with central reader interpretations of plain pelvic x-rays from 688 patients evaluated as part of the French Devenir des Spondylarthropathies Indifférenciées Récentes (DESIR) cohort. All patients had symptoms of inflammatory back pain suggestive of spondyloarthritis (SpA) lasting at least six months but less than three years in duration.
They found that local radiologists/rheumatologists identified sacroiliac disease (defined as at least unilateral obvious sacroiliac disease) in approximately 27% of patients, whereas central reading identified sacroiliac disease in approximately 21%. In 76 of 183 cases (42%) with sacroiliac disease by local readers, the central readers recorded the x-rays as "negative"; conversely, in 38 of 505 cases (8%) without sacroiliac disease identified by local readers, the central readers interpreted the x-rays as "positive." The agreement between the local and central readers did not improve when more severe disease was considered. Furthermore, two central readers disagreed on 108 of 688 (16%) x-rays.
The investigators concluded that in patients with recent onset of back pain, there was significant disagreement about sacroiliac findings between local and central readers. They commented that this may have strong implications for the use of x-rays in the classification of axial SpA.
Plain x-rays are an important part of evaluations for inflammatory back pain, with the finding of sacroiliac disease often leading to classification of disease as an axial SpA. Of importance, emerging evidence suggesting that early diagnosis and treatment of axial SpA improves long-term outcomes has made early, accurate diagnosis even more important. However, the results from this study highlight that interpreting sacroiliac joint disease on plain x-rays in patients with early suspected axial SpA is difficult at best. One assumes that the central readers were more accurate, but as the study authors state, this is hard to assess in the absence of a gold standard for sacroiliac disease. (In these patients, no other imaging such as MRI or CT was available). Because these differences in findings can have a profound impact on the classification and treatment of axial SpA, we need to know more about how to use imaging in early suspected SpA and look forward to additional studies in this area.
Agreement Between Clinical Practice and Trained Central Reading in Reading of Sacroiliac Joints on Plain Pelvic Radiographs: Results From the DESIR Cohort
van den Berg R, Lenczner G, Feydy A, et al
Arthritis Rheumatol. 2014;66:2403-2411
Study Summary
In this study, van den Berg and colleagues compared local radiologist/rheumatologist with central reader interpretations of plain pelvic x-rays from 688 patients evaluated as part of the French Devenir des Spondylarthropathies Indifférenciées Récentes (DESIR) cohort. All patients had symptoms of inflammatory back pain suggestive of spondyloarthritis (SpA) lasting at least six months but less than three years in duration.
They found that local radiologists/rheumatologists identified sacroiliac disease (defined as at least unilateral obvious sacroiliac disease) in approximately 27% of patients, whereas central reading identified sacroiliac disease in approximately 21%. In 76 of 183 cases (42%) with sacroiliac disease by local readers, the central readers recorded the x-rays as "negative"; conversely, in 38 of 505 cases (8%) without sacroiliac disease identified by local readers, the central readers interpreted the x-rays as "positive." The agreement between the local and central readers did not improve when more severe disease was considered. Furthermore, two central readers disagreed on 108 of 688 (16%) x-rays.
The investigators concluded that in patients with recent onset of back pain, there was significant disagreement about sacroiliac findings between local and central readers. They commented that this may have strong implications for the use of x-rays in the classification of axial SpA.
Viewpoint
Plain x-rays are an important part of evaluations for inflammatory back pain, with the finding of sacroiliac disease often leading to classification of disease as an axial SpA. Of importance, emerging evidence suggesting that early diagnosis and treatment of axial SpA improves long-term outcomes has made early, accurate diagnosis even more important. However, the results from this study highlight that interpreting sacroiliac joint disease on plain x-rays in patients with early suspected axial SpA is difficult at best. One assumes that the central readers were more accurate, but as the study authors state, this is hard to assess in the absence of a gold standard for sacroiliac disease. (In these patients, no other imaging such as MRI or CT was available). Because these differences in findings can have a profound impact on the classification and treatment of axial SpA, we need to know more about how to use imaging in early suspected SpA and look forward to additional studies in this area.
Source...