Anteroposterior Lumbar and Pelvic Radiographs Reliable in Assessing Sacroiliitis in Axial Spondyloarthritis

In patients with axial spondyloarthritis (axSpA), both anteroposterior lumbar and conventional pelvic radiographs are reliable in the assessment of radiographic sacroiliitis, according to study results published in Rheumatology.

The standard method for evaluating sacroiliac joint damage and the severity of sacroiliitis in patients with axSpA remains pelvic radiographs. While computed tomography (CT) and MRI are increasingly used for monitoring sacroiliac joints, these imaging methods are associated with high radiation exposure. The reliable use of anteroposterior lumbar radiographs, a lower-radiation method that enables the viewing of sacroiliac joints, has not been clear.

In a study of patients with axSpA, the researchers aimed to determine the reliability of anteroposterior lumbar compared to conventional pelvic radiographs in assessing sacroiliitis.

They collected anteroposterior lumbar and pelvic radiographs at baseline and the 2 year follow-up for 113 patients with axSpA who were eligible for the study. Two trained readers used the modified New York criteria to score the imaged joints from grade 0 (normal) to grade 4 (total ankylosis), and classified patients with axSpA as radiographic (r-axSpA) or nonradiographic (nr-axSpA). The sum of grades for left and right sacroiliac joints were calculated as a sacroiliitis sum score for each patient. For both types of radiographs, reliability and validity were assessed by intraclass correlation coefficients (ICCs), absolute agreement, and 𝜅 statistics.

Results indicated that inter-reader agreement for both types of radiographs was good at both baseline and the 2-year follow-up. At baseline, absolute agreement was 79.7% for conventional pelvic and 77.9% for anteroposterior lumbar radiographs. Values for year 2 showed an absolute agreement of 82.3% for conventional pelvic and 83.2% for anteroposterior lumbar radiographs. Between the 2 radiographic methods, agreement on the sacroiliitis sum score was strong at both time points (ICC, 0.91). In terms of classification of r-axSpA and nr-axSpA, absolute agreement between the 2 methods was 84.9% at baseline and 85.0% at year 2, with a 𝜅 of 0.70 for both. At the 2-year follow-up, absolute agreement between both radiographic methods for assessment of progression/regression was 72.6%.

Researchers noted that anteroposterior lumbar radiographs, often performed for patients with back pain of an unknown origin, were similar in reliability and inter-reader variability to pelvic radiographs for the assessment of the sacroiliac joints. In 85% of the cases, patients received the same classification between both radiographic techniques, and the same progression was indicated in approximately 84% of cases after 2 years.

Study limitations included the fact that sacroiliac joints were missing on some anteroposterior lumbar radiographs, rendering those patients ineligible for study inclusion and indicating that these radiographs may not be used to assess sacroiliac joints in all cases. In addition, the study compared pelvic radiographs and anteroposterior lumbar radiographs, not taking into account CT as the leading technique for assessing structural sacroiliac joint damage.

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.

Reference

Rodriguez VR, Llop M, Protopopov M, et al. Assessment of radiographic sacroiliitis in anteroposterior lumbar vs conventional pelvic radiographs in axial spondyloarthritis. Rheumatology. Published July 25, 2020. doi:10.1093/rheumatology/keaa260