MRI of the Sacroiliac Joints in Patients with Moderate to Severe Ankylosing Spondylitis
Abstract
OBJECTIVE. The objectives of our study were to evaluate whether MRI findings of the sacroiliac joints are able to distinguish between active and inactive disease in patients with established ankylosing spondylitis and to determine whether these findings correlate with markers of clinical activity, disease duration, severity, and degree of radiographic damage.
MATERIALS AND METHODS. Eighteen patients with symptomatic moderate to severe ankylosing spondylitis were evaluated. MRI of the sacroiliac joint (1.5 T) was performed using fat-saturated T2-weighted, T1-weighted, STIR, and fat-saturated contrast-enhanced T1-weighted sequences. The sacroiliac joints were evaluated by two radiologists for enhancement, subchondral bone marrow edema, erosions, and subchondral fatty marrow infiltration. Findings on MRI were analyzed for correlation with multiple clinical characteristics and measures of disease activity, including radiographic scoring.
RESULTS. In 17 patients, MRI showed abnormal findings of the sacroiliac joint. Ten patients showed active disease on MRI as measured by abnormal enhancement and subchondral bone marrow edema. Disease activity detected using MRI correlated in a positive fashion with only C-reactive protein (CRP) level. There was no correlation with the other measures of disease activity or with disease duration. In 14 patients, fatty subchondral bone marrow was detected on MRI. These changes were seen in patients with active and chronic disease and correlated with higher radiographic scores but not with disease duration or markers of disease activity.
CONCLUSION. Contrast-enhanced MRI of the sacroiliac joint is sensitive in depicting sacroiliitis in patients with established ankylosing spondylitis. Subchondral edema and enhancement correlate with high CRP levels. Subchondral fatty bone marrow changes were seen in both active and chronic sacroiliitis and are correlated with higher radiographic scores; these changes may be a marker of more advanced disease.
Introduction
Involvement of the sacroiliac joints is a hallmark of ankylosing spondylitis [1-4]. The definite diagnosis of ankylosing spondylitis requires radiographic evidence of sacroiliitis, which is defined as erosions, subchondral sclerosis, and irregular joint spaces, and clinical features, including inflammatory back pain [1]. Conventional radiographs remain the most widely accepted and available screening method for ankylosing spondylitis; however, radiographs often show normal findings during the early stages of the disease, and several years may be required for sacroiliitis to become apparent on radiography [3-6].
MRI of the sacroiliac joints has been shown to be superior to radiography in depicting sacroiliitis, and gadolinium-enhanced MRI has been shown to be useful in the early detection of active sacroiliitis [7, 8]. New treatment options for patients with ankylosing spondylitis require sensitive imaging techniques to not only help diagnose ankylosing spondylitis early, but also determine disease activity and the degree of damage present at diagnosis [9-11]. The purposes of our study were to evaluate MRI of the sacroiliac joint in patients with moderate to severe ankylosing spondylitis and to determine whether MRI findings of enhancement, edema, and fatty marrow changes were correlated with clinical features and measures of disease activity.
Materials and Methods
Over a 1-year period, MR images and radiographs of the sacroiliac joints of 18 patients with ankylosing spondylitis who had been recruited for two clinical trials were evaluated. Patients had moderate to severe disease activity as defined by the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) score of > 40 (range, 0-100) and physician visual analog scale (VAS) global score of > 40 (range, 0-100). Another criterion for inclusion in our study was that patients had symptomatic inflammatory lower back pain. Definite ankylosing spondylitis was diagnosed using the modified New York criteria [1].
Radiography
Oblique and anteroposterior radiographs of both sacroiliac joints were available for all patients. Radiographs were graded according to the Bath Ankylosing Spondylitis Radiology Index (BASRI). Each sacroiliac joint was evaluated retrospectively by two radiologists in consensus who were blinded to laboratory results and clinical status. The sacroiliac joints were evaluated for blurring of joint margins, sclerosis, erosions, and total ankylosis. A BASRI score was then assigned for each patient.
MRI
All patients underwent MRI of the sacroiliac joints with and without administration of IV gadolinium. MRI (1.5 T; Signa, GE Healthcare) was performed using oblique coronal (parallel to the long axis of the sacrum) and axial fast spin-echo (FSE) T1-weighted sequences (TR/TE, 600/minimum); and STIR (4,000/60), fat-saturated FSE T2-weighted (4,000/50), and fat-saturated contrast-enhanced FSE T1-weighted (600/minimum) sequences. The fat-saturated contrast-enhanced T1-weighted images were obtained after IV administration of 0.1 mmol/kg of body weight of gadopentetate dimeglumine. The field of view was 20 cm; slice thickness, 4 mm; and number of excitations, 2. The imaging times were 7 minutes for the STIR sequences, 4 minutes for the fat-saturated FSE T2-weighted sequences, 6 minutes for the FSE T1-weighted sequences, and 6 minutes for the fat-saturated contrast-enhanced FSE T1-weighted sequences.
The sacroiliac joints were retrospectively analyzed by the same two radiologists in consensus who were blinded to laboratory results and clinical status. MR images were evaluated at a different time point than the radiographs, and the results of the radiographs were not available at the time of MR interpretation. The sacroiliac joints were evaluated for signal characteristics involving the joint space, surrounding bone, and bone marrow adjacent to the joint on each imaging sequence. Enhancement, subchondral bone marrow edema, cartilage abnormalities, periarticular erosions, subchondral fatty marrow infiltration, and ankylosis were recorded. Chronic disease was defined as changes that were low signal on T1- and T2-weighted images, subchondral sclerosis, narrowing of the joint spaces, bone bridging, and ankylosis. Erosions that were high signal on STIR or T2-weighted images, subchondral edema, and enhancement within or adjacent to the sacroiliac joint were considered markers for active inflammatory disease. Findings on MRI of the sacroiliac joint were graded as 1 (mild), 2 (moderate), or 3 (severe). An MRI score was obtained for each patient. MRI was performed within 12 weeks of radiography.
Analysis
Findings on MRI and radiography were analyzed for correlation with clinical inflammatory markers: C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), BASDAI, Bath Ankylosing Spondylitis Functional Index (BASFI), back pain scores (total back pain, nocturnal back pain, inflammatory back pain, and physician and patient global scores), disease duration, and BASRI scores. Laboratory markers were obtained within 1 week of MRI. The ESR (Westergren sedimentation rate test) and CRP level were measured using conventional laboratory technology, with normal values being < 10 mm/h and < 6 mg/L, respectively.
The Wilcoxon's rank sum test and Jonckheere-Terpstra test were used to compare associations between different radiographic measures. The Jonckheere-Terpstra test is a nonparametric test that accounts for the ordered nature of the radiographic grading categories. Large values of Z on the Jonckheere-Terpstra test indicate an important association between measures.
Results
Multiple demographic and disease activity measures are listed in Table 1. The patient population included 14 men and four women with a mean age of 37 years. The mean disease duration was 14 years (range, 2-39 years). Findings for human leukocyte antigen (HLA) B-27 were positive in 94.4% of the patients. As expected by the inclusion criterion of definite ankylosing spondylitis by the modified New York criteria, conventional radiographs showed involvement of the sacroiliac joint in all 18 patients. Unilateral involvement was seen in one patient and bilateral involvement was detected in 17 patients. Seven patients were shown to have complete ankylosis of the sacroiliac joint. Moderate findings, corresponding to a BASRI score of 3, were seen in eight patients. One patient had minimal changes of the sacroiliac joint (BASRI score of 2).
Characteristic | Value |
---|---|
Duration of disease (y) | 14 (5, 20) |
HLA B-27 positive, no. (%) of patients | 17 (94.4) |
BASFI score | 25.7 (22.5, 63.5) |
BASDAI score | 60.2 (44.8, 65.8) |
Global score | |
Patient | 72.3 (40.0, 80.0) |
Physician | 55.5 (48.0, 65.0) |
Inflammatory back pain score | 62.2 (50.5, 93.5) |
Nocturnal back pain score | 71.0 (51.0, 82.5) |
Total back pain score | 71.0 (47.0, 78.0) |
Erythrocyte sedimentation rate (mm/h) | 11 (6, 26) |
C-reactive protein value (mg/L) | 1.125 (0.5, 7.3) |
High C-reactive protein value,a no. (%) of patients) | 9 (50) |
BASRI score, no. (%) of patients | |
2 | 2 (11.1) |
3 | 8 (44.4) |
4 | 8 (44.4) |
Subchondral bone marrow edema score, no. (%) of patients | |
0 | 8 (44.4) |
1 | 9 (50) |
2 | 1 (5.6) |
Enhancement score, no. (%) of patients | |
0 | 8 (44.4) |
1 | 6 (33.3) |
2 | 3 (16.7) |
3 | 1 (5.6) |
Fatty subchondral bone marrow change score, no. (%) of patients | |
0 | 2 (11.1) |
1 | 6 (33.3) |
2 | 7 (38.9) |
3 | 3 (16.7) |
Note—Unless indicated otherwise, data are medians (25%, 75%). HLA = human leukocyte antigen, BASFI = Bath Ankylosing Spondylitis Functional Index, BASDI = Bath Ankylosing Spondylitis Disease Activity Index, BASRI = Bath Ankylosing Spondylitis Radiology Index
a
Normal, < 6 mg/L
All 18 patients except one had abnormal MRI findings of the sacroiliac joint: 16 patients had bilateral involvement and one patient had unilateral involvement. Ten patients showed abnormal enhancement and subchondral bone marrow edema, suggestive of active disease (Figs. 1A, 1B, 1C, 1D, 1E, 2A, 2B, 2C, and 2D). These findings were bilateral in all patients. Seven patients were shown to have ankylosis and subchondral sclerosis of the sacroiliac joint. Two of these patients with ankylosis showed edema and enhancement on MRI, indicating active disease (Figs. 3A, 3B, 3C, 3D, and 3E). In 16 patients, fatty subchondral bone marrow was seen. These changes were seen in patients with MRI findings suggestive of active disease and in those with MRI findings suggestive of chronic disease.
In all patients with MRI findings of enhancement, bone marrow edema was seen on STIR and fat-saturated FSE T2-weighted images. No difference in the severity of edema was detected on STIR images and fat-saturated FSE T2-weighted images; however, edema was slightly better seen on the STIR images than the fat-saturated FSE T2-weighted images (Figs. 1A, 1B, 1C, 1D, and 1E). Four of 10 patients with active disease on MRI showed more extensive enhancement on fat-saturated T1-weighted images compared with the edematous changes seen on STIR or fat-saturated FSE T2-weighted images (Figs. 2A, 2B, 2C, 2D, 3A, 3B, 3C, 3D, and 3E). MRI findings of disease activity correlated only with CRP value (Table 2). There was no correlation between MRI findings and ESR, BASRI, BASDAI, BASFI, back pain scores, or disease duration.
Laboratory or Clinical Criterion | Abnormal Enhancement and Subchondral Bone Marrow Edema | ||
---|---|---|---|
Absent on MRI (n = 8) | Present on MRI (n = 10) | pa | |
CRP value | 0.55 (0.18, 1.125) | 6.4 (0.63, 9.1) | 0.04 |
ESR value | 12.5 (6, 25.5) | 9.5 (6, 34) | 0.83 |
BASFI score | 36 (15.2, 79.8) | 41.2 (22.7, 63.5) | 0.48 |
BASDAI score | 61.2 (44.8, 69.8) | 59.1 (41.8, 65.8) | 0.67 |
Disease duration | 14 (12, 18.5) | 12.5 (4, 20) | 0.58 |
Inflammatory back pain score | 62.2 (57, 76.9) | 72.1 (42, 94) | 0.97 |
Nocturnal back pain score | 71 (67.7, 77.7) | 68.5 (32, 83) | 0.70 |
Total back pain score | 75 (53, 78) | 60.5 (47, 77) | 0.37 |
Global score | |||
Patient | 72.2 (54, 79) | 64.5 (40, 80) | 0.90 |
Physician | 68 (49.7, 82.5) | 51.5 (48, 56) | 0.11 |
Note—Excluding p values, data are medians (25%, 75%). CRP = C-reactive protein, ESR = erythrocyte sedimentation rate, BASFI = Bath Ankylosing Spondylitis Functional Index, BASDAI = Bath Ankylosing Spondylitis Disease Activity Index
a
Based on Wilcoxon's rank sum tests
Subchondral fatty changes on MRI correlated with BASRI score but were inversely related to enhancement (Table 3). MRI findings of fatty subchondral bone marrow did not correlate with disease duration or clinical inflammatory markers.
Radiographic Measure | MRI Finding | Z | pa |
---|---|---|---|
BASRI score | Subchondral bone marrow edema | –1.4 | 0.16 |
BASRI score | Enhancement | –1.3 | 0.18 |
BASRI score | Fatty subchondral bone marrow change | 2.0 | 0.05 |
Subchondral bone marrow edema | Enhancement | 4.1 | < 0.0001 |
Subchondral bone marrow edema | Fatty subchondral bone marrow change | 0.0 | 1.0 |
Enhancement | Fatty subchondral bone marrow change | 0.3 | 0.77 |
Note—MRI findings of subchondral bone marrow edema strongly correlate with abnormal enhancement. Fatty subchondral bone marrow changes on MRI show an inverse correlation with subchondral bone marrow edema or abnormal enhancement. Fatty changes on MRI correlate with the BASRI score. BASRI = Bath Ankylosing Spondylitis Radiology Index
a
Wilcoxon's rank sum test
Radiographic findings did not correlate with clinical or laboratory markers of disease activity or of disease duration.
Discussion
Given the development of new promising treatment options for ankylosing spondylitis, such as tumor necrosis factor alpha, it is important to diagnose ankylosing spondylitis early, detect active sacroiliitis and differentiate it from chronic changes, and monitor for potential changes after therapy [10, 12]. In addition, studies have shown a lack of correlation between clinical signs of disease activity and conventional laboratory markers of disease activity such as CRP levels and ESR [13]. Imaging findings play an important role in the diagnosis of sacroiliitis in patients with ankylosing spondylitis, and MRI of the sacroiliac joint has been shown to be able to reveal the anatomy and degree of inflammation without the use of ionizing radiation [6]. Also, MRI has been found to be more sensitive than radiography for detecting sacroiliitis in patients with ankylosing spondylitis, especially in the evaluation of early disease [7, 8, 14-16].
The purpose of our study was to focus on the use of MRI in evaluating disease activity in patients with moderate to severe ankylosing spondylitis and whether it is possible to show active disease in patients with established ankylosing spondylitis. In this select group of patients with moderate to severe ankylosing spondylitis recruited for clinical trials, MRI of the sacroiliac joint was sensitive in depicting sacroiliitis. MRI allowed lesions with higher water content due to inflammatory edema to be differentiated from lesions with reduced water content due to fibrotic tissue or sclerosis. MRI was also able to show inflammatory changes in advanced stages of ankylosing spondylitis, even in patients with ankylosis of the sacroiliac joint. This finding suggests that fusion does not necessarily signal the absence of associated inflammatory activity at the sacroiliac joint and that patients with sacroiliitis need to be treated in the later disease stages. When evaluating the effects of disease-modifying drugs, it may be valuable to follow these different lesions because they may be, in part, reversible.
Fat-saturated T2-weighted and STIR sequences were slightly less sensitive than contrast-enhanced T1-weighted sequences for detecting the extent of acute inflammatory changes. Although all patients with edematous changes on STIR and fat-saturated FSE T2-weighted images also showed abnormal enhancement, we saw more enhancing lesions on the fat-saturated contrast-enhanced T1-weighted images, supporting the use of IV contrast material. The oblique coronal plane was the most useful plane for detecting changes associated with ankylosing spondylitis.
MRI findings of disease activity correlated only with CRP value. There was no correlation between MRI findings and ESR, BASRI, BASDAI, BASFI, back pain scores, or disease duration. These findings are in contrast to the results of a study performed by Jee et al. [2] who found significant correlation with synovial enhancement and ESR value and with the sum of ESR and CRP values but not with the CRP value alone.
The observation that some patients had periarticular fat accumulation in the bone marrow has no apparent pathophysiologic explanation. However, this finding seems to reflect a later stage of disease because it was associated with more severe involvement of the sacroiliac joints (ankylosis), as determined using the BASRI score, and was inversely related to enhancement. This suggests a temporal change from enhancement to fat as sacroiliac fusion progresses. The observed fat accumulation supports the concept that the disease process in sacroiliitis involves subchondral areas and may be due to an ongoing effect of inflammatory products on local fat metabolism [17, 18].
Our study had several limitations. First is the relatively small number of patients and the retrospective nature of the study. Second is the absence of pathologic tissue confirmation of disease activity. Also, reviewers evaluated the MR examinations and radiographs without knowledge of the laboratory data; however, the reviewers were aware that the patients had been diagnosed with ankylosing spondylitis.
In conclusion, contrast-enhanced MRI of the sacroiliac joint is able to depict sacroiliitis in patients with moderate to severe ankylosing spondylitis. MRI findings suggestive of active disease, such as subchondral edema and enhancement, and MRI findings of fatty bone marrow changes show no correlation with clinical parameters except CRP level. This suggests that MRI can reveal changes at a cellular level that are not depicted by clinical parameters.
Acknowledgments
We thank the Rosalind Russell Medical Center for Arthritis Research for its support.
Footnote
Address correspondence to M. A. Bredella ([email protected]).
References
1.
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
2.
Jee WH, McCauley TR, Lee SH, Kim SH, Im SA, Ha KY. Sacroiliitis in patients with ankylosing spondylitis: association of MR findings with disease activity. Magn Reson Imaging 2004; 22:245-250
3.
Braun J, Bollow M, Sieper J. Radiologic diagnosis and pathology of the spondyloarthropathies. Rheum Dis Clin North Am 1998; 24:697-735
4.
Braun J, Sieper J, Bollow M. Imaging of sacroiliitis. Clin Rheumatol 2000; 19:51-57
5.
Braun J, van der Heijde D. Imaging and scoring in ankylosing spondylitis. Best Pract Res Clin Rheumatol 2002; 16:573-604
6.
Muche B, Bollow M, Francois RJ, Sieper J, Hamm B, Braun J. Anatomic structures involved in early- and late-stage sacroiliitis in spondylarthritis: a detailed analysis by contrast-enhanced magnetic resonance imaging. Arthritis Rheum 2003; 48:1374-1384
7.
Bollow M, Braun J, Hamm B, et al. Early sacroiliitis in patients with spondyloarthropathy: evaluation with dynamic gadolinium-enhanced MR imaging. Radiology 1995; 194:529-536
8.
Braun J, Bollow M, Eggens U, Konig H, Distler A, Sieper J. Use of dynamic magnetic resonance imaging with fast imaging in the detection of early and advanced sacroiliitis in spondylarthropathy patients. Arthritis Rheum 1994; 37:1039-1045
9.
Wanders AJ, Landewe RB, Spoorenberg A, et al. What is the most appropriate radiologic scoring method for ankylosing spondylitis? A comparison of the available methods based on the Outcome Measures in Rheumatology Clinical Trials filter. Arthritis Rheum 2004; 50:2622-2632
10.
Tan AL, Marzo-Ortega H, O'Connor P, Fraser A, Emery P, McGonagle D. Efficacy of anakinra in active ankylosing spondylitis: a clinical and magnetic resonance imaging study. Ann Rheum Dis 2004; 63:1041-1045
11.
Stone M, Salonen D, Lax M, Payne U, Lapp V, Inman R. Clinical and imaging correlates of response to treatment with infliximab in patients with ankylosing spondylitis. J Rheumatol 2001; 28:1605-1614
12.
Baraliakos X, Davis J, Tsuji W, Braun J. Magnetic resonance imaging examinations of the spine in patients with ankylosing spondylitis before and after therapy with the tumor necrosis factor alpha receptor fusion protein etanercept. Arthritis Rheum 2005; 52:1216-1223
13.
Spoorenberg A, van der Heijde D, de Klerk E, et al. Relative value of erythrocyte sedimentation rate and C-reactive protein in assessment of disease activity in ankylosing spondylitis. J Rheumatol 1999; 26:980-984
14.
Bollow M, Biedermann T, Kannenberg J, et al. Use of dynamic magnetic resonance imaging to detect sacroiliitis in HLA-B27 positive and negative children with juvenile arthritides. J Rheumatol 1998; 25:556-564
15.
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
16.
Rudwaleit M, van der Heijde D, Khan MA, Braun J, Sieper J. How to diagnose axial spondyloarthritis early. Ann Rheum Dis 2004; 63:535-543
17.
Ahlstrom H, Feltelius N, Nyman R, Hallgren R. Magnetic resonance imaging of sacroiliac joint inflammation. Arthritis Rheum 1990; 33:1763-1769
18.
McCord JM, Fridovich I. The biology and pathology of oxygen radicals. Ann Intern Med 1978; 89:122-127
Information & Authors
Information
Published In
Copyright
© American Roentgen Ray Society.
History
Submitted: August 15, 2005
Accepted: October 21, 2005
First published: November 23, 2012
Keywords
Authors
Metrics & Citations
Metrics
Citations
Export Citations
To download the citation to this article, select your reference manager software.