DOI:10.2214/AJR.07.3378
AJR 2008; 191:124-128
© American Roentgen Ray Society
"MR Corner Sign": Value for Predicting Presence of Ankylosing Spondylitis
Na Ra Kim1,
Ja-Young Choi1,
Sung Hwan Hong1,
Woo Sun Jun1,
Joon Woo Lee2,
Jung-Ah Choi2 and
Heung Sik Kang2
1 Department of Radiology and Institute of Radiation Medicine, Seoul National
University College of Medicine, 28, Yongon-dong, Chongno-gu, Seoul 110-744,
Korea.
2 Department of Radiology, Seoul National University Bundang Hospital, Seoul,
Korea.
Received November 2, 2007;
accepted after revision January 19, 2008.
Address correspondence to J.-Y. Choi
(drchoi01{at}gmail.com).
Abstract
OBJECTIVE. The purpose of this study was to evaluate the MRI
features of the "MR corner sign" and to determine its diagnostic
usefulness in ankylosing spondylitis. We reviewed the spinal MR images of 52
patients with ankylosing spondylitis and compared these images with those of
52 age- and sex-matched control subjects.
CONCLUSION. The MR corner sign was defined as a triangular and
sharply marginated corner abnormality in a vertebral body unassociated with
osteophytes or Schmorl's node. MR corner lesions were significantly more
common in the ankylosing spondylitis group than in the control group (Fisher's
exact test, p < 0.001). The sensitivity, specificity, and positive
and negative predictive values of the MR corner sign were 44%, 96%, 92%, and
63%, respectively. The most frequent feature of signal intensity was a Modic
type II change (77%). In patients with ankylosing spondylitis, the MR corner
sign was fre quently seen at the thoracolumbar junction, whereas degenerative
corner lesions were commonly seen in the lower lumbar spine. When the MR
corner sign is detected on spinal MR images in daily practice, it should not
be overlooked because it suggests the possibility of ankylosing spondylitis,
which should then be further evaluated.
Keywords: ankylosing spondylitis • MRI
Introduction
The axial skeletal manifestations of ankylosing spondylitis include
sacroiliitis, spondylitis, spondy lodiscitis, and spondylo arthritis
[1–3].
Focal destructive areas along the anterior margin of the discovertebral
junction at the superior and inferior portions of a vertebral body are early
and significant features of ankylosing spondylitis seen on conventional
radiographs, and have been termed "Romanus lesions"
[4]. If these erosions heal,
reactive sclerosis appears at the edges of vertebral endplates, which has been
described as "shiny corners" on radiographs
[1–3].
However, because conventional radiographs are relatively insensitive, they
do not allow detection of Romanus lesions at an early stage. Moreover, MRI of
the thoracolumbar spine has been proven to be superior to radiography for
depicting Romanus lesions
[5–7].
In addition, when ankylosing spondylitis is being treated with nonsteroidal
antiinflammatory agents, intensive physical therapy, or tumor necrosis factor
(TNF)-
inhibitors, the latter of which have recently been shown to be
highly efficient in this respect
[8], clinical improvements are
correlated with a reduction in the acute spinal changes documented by MRI
[9]. Thus, the role of MRI in
the early diagnosis of ankylosing spondylitis and in the assessment of
treatment response to drugs has become an important issue.
Recently, when we retrospectively re viewed lumbar spine MR images that had
been obtained some years before ankylosing spondylitis was confirmed, we
observed one or more triangular or quadrant-shaped, sharply marginated, and
nonerosive corner abnormalities at discovertebral junctions—the
"MR corner sign"—in several ankylosing spondylitis patients,
but these lesions had been overlooked. On the basis of our ex perience, we
suggested the possibility of ankylosing spondylitis to clinicians when we
noticed the MR corner sign in the lumbar spine without conventional
radiographs of the sacroiliac joint. These cases were diagnosed as ankylosing
spondylitis by the New York criteria
[10]. We had confidence that
the MR corner sign would be valuable for diagnosing ankylosing spondylitis,
but, to the best of our knowledge, no study has shown the extent to which the
MR corner sign suggests a diagnosis of ankylosing spondylitis in the absence
of knowledge of sacroiliac changes. The purpose of this study was twofold: to
evaluate the characteristic findings of the MR corner sign in ankylosing
spondylitis and to determine the usefulness of the MR corner sign for
diagnosing ankylosing spondylitis.
Materials and Methods
Patients
In a computerized search of our hospital's radiologic and clinical files
for the 5-year period from March 2001 through February 2006, we identified 816
patients with an established diagnosis of ankylosing spondylitis based on the
New York criteria [10].
Fifty-two of these 816 patients (42 men, 10 women; mean age, 34 years; age
range, 17–70 years) who underwent lumbar spine MRI were included in the
ankylosing spondylitis group. Thirty-nine (75%) of these 52 patients were
HLA-B27 positive, six were HLA-B27 negative, and HLA-B27 was not known in
seven. Mean (± SD) erythrocyte sedimentation rates and C-reactive
protein were 40.2 (± 36.9) mm/first hand 2.8 (± 3.5) mg/L,
respectively. Sacroiliitis was confirmed radiologically in all 52
patients.
In addition, 52 age- and sex-matched control subjects (42 men, 10 women;
mean age, 34 years; age range, 17–70 years) with available lumbar MR
images within the same 5-year period were also enrolled. The control subjects
were radiologically normal (n = 5), had herniated disk disease
(n = 34), and had spondylosis (n = 13). Approval for this
retrospective study was obtained from our institutional review board, but for
this limited, anonymous retrospective review of patient data, it was not
necessary to obtain informed consent.
MRI Protocol
MRI was performed using two 1.5-T scanners (Genesis Signa, GE Healthcare
and Magnetom, Siemens Medical Solutions) using a spine or body-array coil. The
following imaging parameters and planes were usually used: TR range/TE range,
550–600/12–14 for sagittal T1-weighted spin-echo imaging and
4000–4500/120–126 for sagittal T2-weighted fast spin-echo imaging.
Images were obtained from T11 to the S1 level with 4-mm-thick consecutive
slices.
Image Analysis
Sagittal T1- and T2-weighted MR images were reviewed by consensus by two
musculoskeletal radiologists. The MR corner sign was defined as being a
triangular or quadrant, sharply marginated, corner abnormality of the
discovertebral junction unassociated with endplate erosion, osteophytes, or
Schmorl's node. Other corner lesions with disk herniation, degeneration,
Schmorl's node, or osteophytes were classified as degenerative corner lesions.
Blind of the clinical information, we assessed whether an MR corner sign was
present or, and if a lesion was present, we analyzed its MRI findings in terms
of number, morphology (tri angular or quad rant), level, and signal intensity
using the Modic classification (Modic type I, II, or III)
[11]. These findings of MR
corner lesions were compared with those of degenerative corner lesions. Levels
of corner lesions were converted to vertebral units for the analysis. These
were defined as the regions between two virtual lines drawn through the middle
of each vertebra [12].
Consequently, we investigated seven vertebral units from T11–T12 to
L5–S1 in this study. In addition, if only one of the four corners in one
vertebral unit was deemed MR corner sign–positive, the unit as a whole
was regarded as MR corner sign–positive. Numbers of corner lesions in
vertebral units were counted, and their individual MR signal intensities were
recorded.
Statistical Analysis
Fisher's exact test was used to compare the prevalences of the MR corner
sign in the ankylosing spondylitis and control groups. To determine the
validity of the MR corner sign for a diagnosis of ankylosing spondylitis, its
sensitivity, specificity, and positive and negative predictive values were
calculated. Pearson's chi-square test and the linear-by-linear association
test were used to investigate differences in terms of numbers of MR corner
lesions, levels, and signal intensities (by Modic classification) between age
groups by decades in the ankylosing spondylitis group. Statistical analysis
was performed using SPSS for Windows version 12.0 (Statistical Package for the
Social Sciences), and p values of less than 0.05 were considered to
indicate statistical significance.
Results
MR corner lesions were more commonly seen in the ankylosing spondylitis
group than in the control group, and this was considered extremely significant
by Fisher's exact test (p < 0.001). The sensitivity and
specificity of the MR corner sign were 44% (23 of 52) and 96%, respectively
(the sign was not present in 50 of 52). Its positive and negative predictive
values were 92% and 63%, respectively. Characteristics of the MR corner sign
are summarized in Table 1. The
number of patients showing the MR corner sign was significantly different
between age groups (p < 0.001) and tended to increase with age,
which was also statistically significant (linear-by-linear association,
p = 0.002). No significant difference was observed between age groups
in terms of level (p = 0.974) or signal intensity (p =
0.070) of MR corner lesions (Table
1).
View larger version (124K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —33-year-old man with ankylosing spondylitis. T2-weighted
(A) and T1-weighted (B) sagittal images show multiple-quadrant
MR corner lesions (arrowheads) in anterior corners at L1–L2,
L2–L3, and L3–L4 and triangular MR corner lesions
(arrows) in anterior corners at T12–L1 and L2–L3 and
posterior corner at L2–L3.
|
|
View larger version (115K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —33-year-old man with ankylosing spondylitis. T2-weighted
(A) and T1-weighted (B) sagittal images show multiple-quadrant
MR corner lesions (arrowheads) in anterior corners at L1–L2,
L2–L3, and L3–L4 and triangular MR corner lesions
(arrows) in anterior corners at T12–L1 and L2–L3 and
posterior corner at L2–L3.
|
|
In the ankylosing spondylitis group, the number of corner lesions with the
MR corner sign out of a total of 756 corners from vertebral units
T11–T12 to L5–S1 was 266 (35%). The MR corner lesions were more
common in the anterior corners (61%, 161 of 266 corners) compared with the
posterior corners. In the ankylosing spondylitis group, MR corner lesions
showed triangular shapes in 152 of 266 corners (57%), and quadrant shapes in
114 (43%) (Fig. 1A,
1B).
In the 23 ankylosing spondylitis patients with an MR corner sign, it was
most frequently observed in the thoracolumbar junction. The most commonly
affected vertebral units were T12–L1 (37%, 19 of 52 patients) and
L1–L2 (39%, 20 of 52 patients) (Fig.
2A). Most MR corner lesions showed high signal intensity on
T1-weighted and T2-weighted images of Modic type II changes (77%, 206 of 266
corners) (Fig. 3A,
3B).
View larger version (103K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —53-year-old man with degenerative corner lesions. T2-weighted
(A) and T1-weighted (B) sagittal images show semicircular corner
lesion of Modic type II change with disk degeneration and osteophytes at
L4–L5 (arrows).
|
|
View larger version (123K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —31-year-old man with ankylosing spondylitis. T2-weighted
(A) and T1-weighted (B) sagittal images reveal multiple sharply
marginated triangular MR corner lesions with Modic type II changes
(arrowheads) in anterior and posterior corners from L1 to L4.
|
|
View larger version (113K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —31-year-old man with ankylosing spondylitis. T2-weighted
(A) and T1-weighted (B) sagittal images reveal multiple sharply
marginated triangular MR corner lesions with Modic type II changes
(arrowheads) in anterior and posterior corners from L1 to L4.
|
|
In the control group, five of 52 patients showed 11 degenerative corner
lesions. One corner lesion had a quadrant shape with disk extrusion, and the
other 10 corner lesions had a semicircular shape accompanied by osteophytes or
Schmorl's nodes (Fig. 2A,
2B). The most frequently
affected vertebral units of degenerative corner lesions were L4–L5 (80%,
four of five patients) (Fig.
2B). Most degenerative corner lesions showed Modic type II signal
intensities (64%, seven of 11 corners). However, because two patients had
triangular shaped corner lesions without disk herniation or osteophytes, these
two cases were deemed false-positives
(Table 2).
View larger version (100K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —53-year-old man with degenerative corner lesions. T2-weighted
(A) and T1-weighted (B) sagittal images show semicircular corner
lesion of Modic type II change with disk degeneration and osteophytes at
L4–L5 (arrows).
|
|
Discussion
Romanus lesions are manifested precisely at the site of attachment of the
annulus fibrosus to the vertebral endplate. Because such a junction between
bone and a ligamentous structure is an enthesis by definition, anterior or
posterior spondylitis can be considered as an enthesitis
[1–3].
Jevtic et al. [7] classified
typical Romanus lesions seen on the radio graphs of patients with ankylosing
spondylitis using MRI findings as follows: first type, low signal intensity on
T1-weighted and high signal intensity on T2-weighted and T1-weighted
contrast-enhanced images at verte bral corners consistent with edematous hyper
emic inflammatory tissue; second type, syn desmo phyte formation with
increased signal in tensity on both T1-weighted and T2-weighted images with no
contrast enhancement; and third type, a mixture of radiographic features and
varied high and low signal intensity changes at the vertebral rim on MRI with
rims of enhancement in the vertebral body after contrast administration. We
consider triangular or quadrant, sharply marginated, and nonerosive corner
abnormalities—the MR corner sign—a type of Romanus lesion.
In this study, we found that the MR corner sign in ankylosing spondylitis
differs from degenerative corner change and that the MR corner sign is a
reliable finding for diagnosing ankylosing spondylitis. This sign showed a
very small, sharply marginated, right-angled triangular shape in the corners
of vertebral bodies that was not associated with endplate erosion, whereas
degenerative corner lesions with a semicircular shape were always associated
with adjacent disk degeneration, herniation, or osteophytes. The MR corner
sign was more commonly observed in the thoracolumbar junction, whereas
degenerative corner lesions were more common at lower lumbar levels. The most
frequent type of signal intensity was Modic type II change, even in young
patients. On the basis of our findings, we suggest that MR corner lesions
occur during early stage ankylosing spondylitis.
According to Baraliakos et al.
[12], spinal inflammation is
found more commonly in the thoracic spine than in the cervical or lumbar
spine. When a STIR sequence was used, spinal inflammation of Modic type I
change was detected in 26%, 74%, and 24% of the cervical, thoracic, and lumbar
spine, respectively, and the T7–T8 vertebral unit was revealed to be
most often affected. In the present study, the sensitivity of MR corner
lesions of 44% from T11 to S1 was higher than the 24% reported for the lumbar
spine by Baraliakos et al., presumably because our data included both active
and inactive spinal changes.
In our experience, patients referred by physicians often undergo lumbar
spine MRI without conventional radiography of the sacroiliac joints. Moreover,
even when conventional radiographs, including pelvis anteroposterior or lumbar
spine anteroposterior, are available, radiologic evi dence of sacroiliitis
might be inconclusive. We propose that in these situations the detection of
the MR corner sign in lumbar spine MRI might be very useful in terms of
diagnosing ankylosing spondylitis.
The present study has several limitations. First, the study is inherently
limited by its retrospective nature. Second, no STIR or gadolinium-enhanced
sequences, which sensi tively detect active lesions, were included. Third,
cervical and thoracic MRI were not performed. Fourth, we have not investigated
whether or not the MR corner sign occurs in other spondyloarthropathies (e.g.,
psoriatic arthritis, reactive arthritis, or colitis-related
spondyloarthropathy).
In conclusion, we consider the MR corner sign a reliable finding of
ankylosing spondylitis. Thus, when the MR corner sign is detected on spinal MR
images, it should not be overlooked because it suggests the possibility of
ankylosing spondylitis. Radiologists should pay attention to this sign in
daily practice.
References
- Braun J, Bollow M, Sieper J. Radiologic diagnosis and pathology of
the spondyloarthropathies. Rheum Dis Clin North Am1998; 24:697
–735[CrossRef][Medline]
- Hermann KG, Althoff CE, Schneider U, et al. Spinal changes in
patients with spondyloarthritis: comparison of MR imaging and radiographic
appearances. RadioGraphics 2005;25
: 559–570[Abstract/Free Full Text]
- Levine DS, Forbat SM, Saifuddin A. MRI of the axial skeletal
manifestations of ankylosing spondylitis. Clin Radiol2004; 59:400
–413[CrossRef][Medline]
- Romanus R, Yden S. Destructive and ossifying spondylitic changes in
rheumatoid ankylosing spondylitis (pelvo-spondylitis ossificans).
Acta Orthop Scand 1952;22
: 88–99[Medline]
- Dihlmann W. Current radiodiagnostic concept of ankylosing
spondylitis. Skeletal Radiol 1979;4
: 179–188[CrossRef][Medline]
- Remedios D, Natali C, Saifuddin A. Case report: MRI of vertebral
osteitis in early ankylosing spondylitis. Clin Radiol1998; 53:534
–536[CrossRef][Medline]
- Jevtic V, Kos-Golija M, Rozman B, McCall I. Marginal erosive
discovertebral "Romanus" lesions in ankylosing spondylitis
demonstrated by contrast enhanced Gd-DTPA magnetic resonance imaging.
Skeletal Radiol 2000;29
: 27–33[CrossRef][Medline]
- Brandt J, Haibel H, Cornely D, et al. Successful treatment of
active ankylosing spondylitis with the anti-tumor necrosis factor alpha
monoclonal antibody infliximab. Arthritis Rheum2000; 43:1346
–1352[CrossRef][Medline]
- Braun J, Baraliakos X, Golder W, et al. Magnetic resonance imaging
examinations of the spine in patients with ankylosing spondylitis, before and
after successful therapy with infliximab: evaluation of a new scoring system.
Arthritis Rheum 2003;48
:1126
–1136[CrossRef][Medline]
- van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic
criteria for ankylosing spondylitis: a proposal for modification of the New
York criteria. Arthritis Rheum 1984;27
: 361–368[Medline]
- Modic MT, Masaryk TJ, Ross JS, Carter JR. Imaging of degenerative
disk disease. Radiology 1988;168
: 177–186[Free Full Text]
- Baraliakos X, Landewé R, Hermann KG, et al. Inflammation in
ankylosing spondylitis: a systematic description of the extent and frequency
of acute spinal changes using magnetic resonance imaging. Ann Rheum
Dis 2005; 64:730
–734[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
