Log in

Common incidental findings on sacroiliac joint MRI: Added value of MRI-based synthetic CT

EJRad, vol158, 110651, jan2023 Sept. 28, 2023, 8 a.m.

Presented By

Dr Walter Maksymowych MB ChB, FRCP(C), FACP
Chief Medical Officer, CARE Arthritis; Professor of Medicine, University of Alberta

Guest Commentary By:

Robert Lambert MB BCh, FRCR, FRCPC
University of Alberta

Synopsis

Introduction
  • A novel technique based on MRI for visualizing osseous structures is synthetic CT (sCT), a deep learning-based multiparametric MRI technique that permits the creation of radiograph-like and CT-like images without ionizing radiation.
  • It has been shown to outperform T1-weighted MRI-images for detection of erosions, sclerosis and ankylosis of the SI joint in patients with axSpA (Jans, Chen et al).
  • The aim of this study was to determine the prevalence of incidental findings demonstrated on MRI of the SI joints in patients clinically suspected of axSpA and to examine the added value of the additional sCT.
Methods
  • Retrospective study of 214 patients clinically suspected of axSpA who underwent a conventional diagnostic MRI with STIR and T1W sequences
  • For sCT reconstruction an additional axial three-dimensional T1-weighted radio-frequency-spoiled multiple gradient echo (3DT1MGE) sequence was scanned
  • Two readers independently determined: (1) The lesion is only visible on sCT, (2) the lesion is visible on sCT and the conventional sequence (T1 TSE or T2 STIR), (3) the lesion is only visible on the conventional sequence (T1 TSE or T2 STIR).
Results
  • In 44.7% of patients sacroiliitis was present.
  • 187 (89.0 %) of all patients MRI showed one or more incidental findings.
  • Degeneration of SIJ was most frequent (66.6%)
  • The most frequent incidental findings outside the sacroiliac joint were facet joint degeneration (29.0%), disc degeneration (25.2%), enostosis (19.5%) and lumbosacral transitional vertebrae (14.3%).
  • Synthetic CT was found to be problem solving or necessary for diagnosis in 543 (68.9%) of these lesions. 42.1% of lesions were not visible on conventional MRI (T1 TSE and STIR).

Commentary

The sCT example images of incidental findings in this manuscript are truly impressive for their resemblance to conventional CT. Many of the incidental findings are degenerative and/or innocuous raising questions regarding impact on patient management and whether the additional scan time and costs are justified.

Special Guest Commentary

The Morbée paper is interesting and has been performed diligently although the results need to be interpreted carefully. The data was prospectively gathered with patient consent and the observational exercise performed retrospectively. The two most important things to note are:

  1. "BoneMRI" software produces images that are almost identical to an actual CT scan and the images are very impressive indeed.
  2. The most useful observation is not incidental - 66.7% of patients demonstrated some evidence of degenerative change in the SIJ. My own experience would concur with this observation.

Minor concerns are:

  • Table 1 lists incidental findings and sacroiliitis - but some of the findings in the SIJ (degeneration) or outside the SIJ (osteitis pubis and OCI) are neither incidental nor sacroiliitis.
  • All the findings displayed by using this software must be visible on the original 3DT1MGE images BEFORE the software manipulation.
  • The use of T1-3DGE for observing bone findings on MRI has been known for some time and there are numerous ways of doing it. There are multiple reports of its value (e.g. VIBE (Diekhoff et al.)), the results of which appear to be very similar or identical to BoneMRI.
  • The number of "indicental findings" observed was hugely influenced by the fact that the BoneMRI was done differently from the semicoronal SIJ MRI and included 22 times the volume of anatomy. Many of the incidental findings were not observed on SIJ MRI because the anatomy wasn't included on the SIJ MRI.
  • The BoneMRI software is not provided by the MRI manufacturers. It is commercially available but not freely available.
  • Although most of the figures are good, some illustrations seem to have been chosen to highlight the value of the BoneMRI and the authors may not have chosen the most appropriate conventional MR image for comparison or their interpretation of the conventional MR image is subject to debate.

Your Comments