Article Text

SAT0225 Do Bone Marrow Edema (BME) Lesions in the Sacroiliac Joint (SIJ) Change into Fatty Lesions Over A 3-Month Period in Patients with Axial Spondyloarthritis (AXSPA)?
  1. M. De Hooge1,
  2. R. Van Den Berg1,
  3. M. V. Navarro Compán1,
  4. M. Reijnierse2,
  5. F. van Gaalen1,
  6. K. M. Fagerli3,
  7. R. Landewé4,
  8. M. van Oosterhout5,
  9. R. Ramonda6,
  10. T. Huizinga1,
  11. D. Van Der Heijde1
  1. 1Rheumatology
  2. 2Radiology, LUMC, Leiden, Netherlands
  3. 3Rheumatology, Diakonhjemmet Hospital, Oslo, Norway
  4. 4Clinical Immunology and Rheumatology, AMC, Amsterdam
  5. 5Rheumatology, GHZ, Gouda, Netherlands
  6. 6Rheumatology, University of Padova, Padova, Italy


Background BME lesions may change into fatty lesions in the SIJ over time. In one study significant changes of fatty lesions in the SI joints were observed over 1 year in patients (pts) with axSpA1. Are similar results found with a shorter follow-up period?

Objectives To investigate whether BME lesions in the SIJ seen on MRI change into fatty lesions over a 3-month period in pts with axSpA.

Methods Pts with back pain (≥3 months, ≤2 years, onset <45 years) from the 5 participating centres of the SPondyloArthritis Caught Early (SPACE)-cohort were included. All pts fulfilling the ASAS axSpA criteria underwent MRI-SI at baseline (MRI-SI-B) and after 3 months (MRI-SI-3m). MRI-SI-B and MRI-SI-3m were scored by 2 well-calibrated readers independently, blinded for time point, for the presence of BME and fatty lesions on STIR and MRI T1-weighted images viewed simultaneously. BME and fatty lesions were defined present if 1 lesion on ≥2 consecutive slices or >1 lesion on a single slice. The scores of baseline and 3 months were compared on quadrant level (Q) and sum of all quadrants (8 total) to obtain pt score. Scores of the readers are reported separately.

Results Only axSpA pts with MRI-SI-B and MRI-SI-3m were included for this analysis (n=83, number of Q=664). In 62/83 pts (74.7%) reader 1 indicated the presence of BME or fatty lesions at any time point, reader 2 indicated this in 60/83 pts (72.3%). Readers agreed in 73/83 pts (88%) on the presence of lesions at any time point. Reader 1 indicated no lesions at baseline or follow-up in 410/664 Qs (61.7%) and in 131/664 Qs (19.7%) the type of lesions did not change over time. In 123/664 Qs (18.5%) lesions occurred, disappeared or changed from one type to another. In total, there were 34/664 Qs (5.1%) in which fatty lesions occurred, and 29/664 Qs (4.4%) in which fatty lesions disappeared. In only 3/664 Qs (0.5%) BME lesions on MRI-SI-B turned into FAT lesions at MRI-SI-3m. In the same amount of Qs the opposite thing is happening; FAT lesions change into BME lesions over time. The results of both readers show similar trends (see table).

Conclusions We find a lot of volatility of BME and fatty lesions over 3 months time. It is common for both lesions to occur without a previous lesion or disappear completely without leaving a mark, but it is uncommon to see BME lesions turn into fatty lesions over a period of 3 months. This observation casts serious doubts on the value of fatty lesions as a surrogate for previous inflammation.


  1. Song ARD 2011;70:1257-63

Disclosure of Interest None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.