- Design
- modified Delphi consensus, up to three rounds, threshold of >75% agreement and score of 8 or more out of 10
- Population
- 64 members of the European Society of Musculoskeletal Radiology arthritis subcommittee; 42, 44 and 48 responded across rounds
- Primary outcome
- consensus on structural sacroiliac joint lesion definitions for CT
- Effect
- 7 definitions established with 76% to 96% agreement; fat metaplasia excluded
A modified Delphi process within the arthritis subcommittee of the European Society of Musculoskeletal Radiology produced the first expert consensus definitions for structural sacroiliac joint lesions on CT. Twelve experts from nine countries drafted the definitions, which were then refined over up to three rounds with a predefined bar: more than 75% agreement and a median score of at least 8 on a 10-point scale. Response rates were 66%, 69% and 75% of the 64 members.
Seven definitions passed: erosions, pseudo-widening, subchondral sclerosis, backfill, joint space narrowing, bone bud and ankylosis, with agreement between 76% and 96%. Two decisions are worth noting. Fat metaplasia was excluded by agreement because CT detects it poorly — an honest exclusion rather than a definition stretched to cover a lesion the modality cannot see. Backfill, which is primarily an MRI finding, was retained as a structural lesion.
The stated purpose is reporting consistency and interobserver reproducibility, and the longer aim is a CT-based sacroiliac joint scoring system along RADS lines. Nothing here is a diagnostic criterion for axial spondyloarthritis, and the consensus does not change the imaging modality of choice for suspected early sacroiliitis, which remains MRI.
Where this lands practically is the incidental one. Sacroiliac joints appear on every abdominal and pelvic CT, and many of these patients have back pain nobody has attributed. Having agreed terms means a structural lesion seen on a CT done for something else can be described in words a rheumatologist will read the same way — which is the realistic route by which a CT contributes to recognising sacroiliitis at all.
- Use the seven agreed terms — erosion, pseudo-widening, subchondral sclerosis, backfill, joint space narrowing, bone bud, ankylosis — when describing sacroiliac joint lesions on CT
- Look at the sacroiliac joints on abdominal and pelvic CT done for other indications, and describe what you see in these terms
- Do not report fat metaplasia on CT; it was excluded from the definitions for poor detectability
- Continue to recommend MRI where early sacroiliitis is suspected clinically — this consensus does not change the modality of choice
- Avoid implying a diagnosis of axial spondyloarthritis from CT findings alone; these are lesion definitions, not diagnostic criteria
Why it matters
The sacroiliac joints are on every pelvic CT, and until now there was no agreed way to describe what was there.
Don't overread it
Expert agreement on wording is not evidence that the definitions improve reproducibility or detection — that remains to be tested.
The statistics, in plain English
A Delphi consensus measures agreement among experts, not accuracy against a reference standard — 96% agreement on a definition means the panel concurred on wording, not that the lesion is correctly identified 96% of the time. Response rates of 66% to 75% mean a quarter to a third of the subcommittee did not respond in each round, and non-responders may not have agreed. Interobserver reproducibility is the stated aim but has not yet been measured with these definitions in use.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for radiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free