Three radiologists, all with >10 years of experience in
musculoskeletal MRI, independently evaluated the images. The first
observer to open the MRI examination saved a mark on the lowest lumbar
disc level. All observers reported this level as L5/S1. First, MCs
were rated on T1/T2, blinded to other sequences. Later the observers
rated STIR findings and decided whether any increased STIR signal was
related to an MC visible on T1/T2. The observers were blinded to
clinical outcome but knew that patients were preliminarily eligible
for the trial. To align their understanding of procedures and rating
criteria, the observers rated and discussed MCs and STIR findings in a
pilot study (32 MRIs not included in the main study).
On T1/T2, we defined MCs as signal changes in the vertebral bone marrow
extending from the endplate, and based rating criteria for MC type and
size on prior work (10 ,11 (
link),38 (
link),45 (
link)) (
Table 2). Only T1/T2
findings defined MC types I, II and III, not STIR findings. Not
recorded as MCs were: (i) changes separated from the endplate; (ii)
roundly shaped fatty changes abutting the endplate with a smaller base
than height (more likely focal fatty marrow or hemangiomas); and (iii)
changes extending through the endplate (Schmorl’s hernias).
On STIR, we defined MC-related signal increase as visible increase
compared to normal vertebral bone marrow, formed and located as an MC
and/or located in or abutting a region with MC on T1/T2 (and not
located in a likely hemangioma). MC-related STIR signal increase was
evaluated for presence, height, anteroposterior (AP) extent, volume,
and maximum intensity (
Table 2). STIR signal
decrease was not evaluated. STIR signal intensity was measured in the
region with most intense MC-related STIR signal, in the cerebrospinal
fluid (CSF) and in normal vertebral body marrow (
Table 2,
Fig. 1).
The measurements were made in circular regions of interest available
in our PACS with size 25 mm
2 (used for most intense
MC-related STIR signal and CSF) and 44 mm
2 (used for normal
vertebral body marrow) (
Fig. 1). Care was taken to
avoid surrounding structures, e.g. intervertebral discs, nerve roots,
central vertebral vein. Intensity of CSF varied between levels and was
measured at the same disc level as the MC-related STIR signal. Maximum
intensity of the MC-related STIR signal (“Stir”) in % points on a
scale from normal vertebral body intensity (“Body,” 0%) to CSF
intensity (“CSF,” 100%) was calculated as ((Stir – Body)/(CSF –
Body)) × 100.
Kristoffersen P.M., Vetti N., Storheim K., Bråten L.C., Rolfsen M.P., Assmus J, & Espeland A. (2020). Short tau inversion recovery MRI of Modic changes: a reliability study. Acta Radiologica Open, 9(1), 2058460120902402.