Article Text

AB1293 Is it useful to repeat an MRI in the diagnostic work-up for spondyloarthritis?
  1. R. van den Berg1,
  2. M. de Hooge1,
  3. V. Navarro-Compán1,
  4. F. van Gaalen1,
  5. M. Reijnierse2,
  6. T. Huizinga1,
  7. D. van der Heijde1
  1. 1Rheumatology
  2. 2Radiology, LUMC, Leiden, Netherlands


Background In the diagnostic work-up of spondyloarthritis (SpA), imaging of the SI-joints by MRI (MRI-SIJ) is an important step. One study showed that in HLA-B27+ patients and male patients it might be useful to repeat an MRI after one or two years1. It is not known if repeating an MRI after a shorter period than 1 year is useful.

Objectives To investigate whether it is useful to repeat an MRI after 3 months in the diagnostic process for SpA.

Methods Patients with chronic back pain (n=157) (≥3 months, ≤2 years, onset <45 years) included in the SPondyloArthritis Caught Early (SPACE)-cohort in the Leiden University Medical Center (LUMC) underwent an MRI-SIJ during baseline visit. Patients with SpA and possible SpA (n=90) underwent a follow-up visit after 3 months, again including MRI-SIJ. MRI-SIJs were scored by 3 independent readers “positive” or “negative” according to the ASAS definition2, blinded for the time-sequence. If 2/3 reads were positive, the MRI-SIJ was marked as positive.

Univariate and multivariate regression analyses were performed to investigate which variables (IBP, elevated CRP, MRI-status at baseline, gender, HLA-B27-status and age at onset) could predict a positive MRI-SIJ at 3 months.

Results Only patients with complete MRI data are included in this analysis (n=90). In the univariate analysis, a positive MRI-SIJ at baseline was the strongest predictor of a positive MRI-SIJ over time (OR 49.5, 95% CI 11.9-206.7, p<0.001). Regardless MRI-status, gender and HLA-B27-status (OR 7.7, 95% CI 2.6-23.1, p<0.001 and OR 2.6, 95% CI 0.9-7.0, p=0.07, respectively) are strong predictors of a positive MRI over time. The latter 2 variables were used in a multivariate model. Groups were made based on this model (table). In the majority of the patients (90%), MRI-status, either positive (n=15) or negative (n=66), did not change over time. Of the patients with a negative MRI-SIJ at baseline, 5 (7%) developed a positive MRI-SIJ after 3 months and in 4 out of 19 patients (21.1%) lesions on MRI-SIJ disappeared over time. Two out of 5 patients in whom MRI was positive for the first time at follow-up fulfilled the ASAS axial SpA criteria only at follow-up.

Conclusions We confirmed that a positive MRI-SIJ at baseline is a very strong predictor for a positive MRI-SIJ after 3 months. In patients with a negative MRI-SIJ at baseline, male gender and HLA-B27 are predictive of a later positive MRI-SIJ. In this group of patients with short symptom duration, variation in MRI-SIJ positivity occurred in 10% of the patients over a very short period of only 3 months. A positive change of MRI-SIJ has led to a different classification in 2.2% (n=2) of the patients. More data are needed to decide if it is necessary to repeat MRI, and if so, with what time interval.

  1. Van Onna M. ARD 2011;70:1981-5.

  2. Rudwaleit M. ARD 2009;68:1520-7

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.