Validity of magnetic resonance image and HLA-B27 in early detection of sacroiliitis in Egyptian spondyloarthropathic patients

Objective The aim of this study was to compare the validity of MRI in the early detection of sacroiliitis with laboratory findings of human leukocyte antigen-B27 (HLA-B27), conventional radiography, and clinical assessment. Participants and methods Sixty patients with spondyloarthropathy (group II) with duration of illness less than 2 years and 20 healthy controls (group I) were included in this study. Both groups were subjected to assessment of history, clinical examination, and laboratory investigations (erythrocyte sedimentation rate, C-reactive protein titer, rheumatoid factor, HLA-B27). Conventional radiography and MRI of the sacroiliac joints were performed. Spondyloarthropathic patients were divided according to MRI as follows: group IIA, which included patients with sacroiliitis, and group IIB, which included patients without sacroiliitis. Results In our study, ankylosing spondylitis was diagnosed in 22 (36.6%) patients, followed by undifferentiated spondyloarthropathy in 12 (20%) patients, reactive arthritis in 10 (16.7%) patients, psoriatic arthropathy in 10 (16.7%) patients, and enteropathic arthropathy in six (10%) patients. Evidence of sacroiliitis was found in 66.6% (40/60) of patients by MRI, which was higher than the result obtained by plain radiography 20% (12/60). HLA-B27 positivity found in 53.3% (32/60) of patients. There was a significant difference between the two groups in HLA-B27 and radiological sacroiliitis; there was no sacroiliitis in the control group. MRI showed sacroiliitis even in patients with no inflammatory back pain. There was a highly statistically significant difference between patient subgroups in disease duration (P = 0.001) and primary complaints and clinical sacroiliitis (P = 0.001). Conclusion MRI is the preferred modality in the detection of early sacroiliitis in spondyloarthropathy and HLA-B27 positivity is a highly useful predictor of early sacroiliitis


Introduction
Spondyloarthropathy (SPA) are a group of interrelated rheumatic conditions including ankylosing spondylitis (AS), reactive arthritis, psoriatic arthritis, and SPA associated with inflammatory bowel disease (Crohn's disease or ulcerative colitis), undifferentiated spondyloarthropathy (USPA), and juvenile-onset spondyloarthritis [1]. Men, who are slightly more frequently affected than women, have more radiographic progression [2]. Spondyloarthritides are characterized by sacroiliitis with inflammatory back pain, peripheral arthropathy, absence of rheumatoid factor and subcutaneous nodules, enthesitis, and extraarticular or extraspinal involvement, including the eyes, heart, lung, and skin. There is a tendency toward familial aggregation as well as varying associations with human leukocyte antigen-B27 (HLA-B27) [3]. HLA-B27 is a polymorphic form of the HLA-B molecule that is found in only 8% of the general population worldwide [4]. Over 25 molecular subtypes of HLA-B27 have been described thus far [5]. The most common subtypes (HLA-B*2705, HLA-B*2702, HLA-B*2704, HLA-B*2707) are clearly associated with a risk for spondyloarthritis. Two subtypes of HLA-B27, HLA-B*2706 (found in Southeast Asia) and HLA-B*2709 (found in Sardinia), appear not to be associated with spondylitis, possibly because of amino acid differences in the 'B' pocket of the HLA antigen-binding cleft that alter the composition of peptides presented by these HLA-B27 subtypes. The other subtypes of HLA-B27 are too rare to have had disease associations established [6]. The traditional pathophysiological framework for spondyloarthritis is the arthritogenicpeptide theory, which proposes that HLA-B27 presents self-peptides that resemble pathogen-derived peptides to CD8-restricted T lymphocytes [7]. The radiographic hallmark of the group is sacroiliitis, which, when present, is useful in the diagnosis. MRI changes have now been included in the new classification criteria of early axial SPA and are now considered a major tool in the diagnosis. Over the past decade, tumor necrosis factor-α-blocking agents have been investigated extensively and became the main stream of therapy, providing an effective treatment to the patients [8].

Patients
Our study was carried out in Al-Minia University Hospital. All patients were recruited from the Rheumatology outpatient clinic and the private clinics of rheumatology doctors. Our study included the following groups: (1) Group I: 20 apparently healthy volunteers as a control group. (2) Group II: 60 patients were diagnosed with SPA according to European Spondyloarthropathy Study Group (ESSG) [9]. Then, the patients were divided into two groups: those with axial SPA and those with peripheral SPA [according to the assessment in spondyloarthritis international society (ASAS) classification criteria for axial [10] or peripheral [11] SPA, respectively]. The disease duration is less than 2 years (from onset of symptoms or appearance of the first signs attributable to the disease). The patients with SPA were subdivided according to the presence of sacroiliitis by MRI as follows: group IIA, which included patients with sacroiliitis, and group IIB, which included patients without sacroiliitis. A comparison was performed between the two groups in terms of clinical, laboratory, and radiological variables.

Exclusion criteria
Exclusion criteria were as follows: (1) Disease duration more than 2 years.
(2) History of trauma or surgery to the knees, ankles, or elbows.

Ethical considerations
The nature of the present study was explained to all patients. The laboratory and radiological procedures represent standard care and pose no ethical conflicts. A verbal and written consent was obtained from all patients and controls.

Methods
All patients were subjected to a full history taking, full general and musculoskeletal examination including measurements of swelling joint count, total joint count, and all provocative tests of sacroiliitis. All patients were subjected to assessment of all the following indices: Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Bath Ankylosing Spondylitis Functional Index (BASFI), Bath Ankylosing Spondylitis Metrology Index (BASMI), and Maastricht Ankylosing Spondylitis Enthesitis Score (MASES) [12].
All patients and controls were subjected to the following laboratory and radiologic investigations:

Laboratory investigations
Erythrocyte sedimentation rate was determined using the Westergren method. C-reactive protein (CRP) was determined using which included latex agglutination slide test for qualitative and semiquantitative determination of CRP in nondiluted serum. Rheumatoid factor was determined using the latex fixation test. HLA-B27 was determined using the qualitative two-color direct immunofluorescence method for the rapid detection of HLA-B27 antigen expression in erythrocyte-lysed whole blood using the BD FACSCanto (USA) family of flow cytometers [13].

Radiological investigations
One hundred and sixty sacroiliac joints (SIJs) of 60 patients and 20 controls were evaluated in a blinded manner by plain radiography and enhanced MRI. Complete clinical examinations were performed for all patients before the application of radiography and MRI.
Plain X-ray: The SIJs of all patients and controls were evaluated in a blinded manner by a rheumatologist [SIJs (anteroposterior view)].
MRI on sacroiliac joints: The SIJs of all patents and controls were evaluated in a blinded manner by a radiologist. MRI was carried out using a 0.2-T GE machine with a spine phased-array coil. Two sequences were applied to patients: the coronal short tau inversion recovery (STIR) sequence and the axial T 1 -weighted sequence. Definition of sacroiliitis by MRI for application in the new ASAS classification criteria is as follows [10]: Types of findings required for the definition of sacroiliitis by MRI: Active inflammatory lesions of the SIJs (reflecting active sacroiliitis) are required for the definition of 'sacroiliitis on MRI' as one of The two imaging items in the ASAS classification criteria for axial SPA Bone marrow edema (BME) on STIR or osteitis on T 1 postgadolinium highly suggestive of SPA must be clearly present and located in the typical anatomical areas (subchondral or periarticular bone marrow) The sole presence of other active inflammatory lesions such as synovitis, enthesitis, or capsulitis without concomitant BME/ osteitis is not sufficient for the definition of sacroiliitis on MRI Structural lesions such as fat deposition, sclerosis, erosions, or bony ankylosis are likely to reflect previous inflammation. At this time, however, the consensus group believed that the sole presence of structural lesions without concomitant BME/osteitis does not suffice for the fulfillment of sacroiliitis on MRI in the ASAS classification criteria for axial SPA Amount of signal required: If there is only one signal (lesion) per MRI slice suggesting active inflammation, the lesion should be present on at least two consecutive slices. If there is more than one signal (lesion) on a single slice, one slice may be sufficient

Statistical analysis
Analysis of our data was carried out by a personal computer using the statistical package for the social sciences (version 16; SPSS Inc., Chicago, Illinois, USA) as follows: descriptive statistics: quantitative variables were described as mean, SD, and range and qualitative variables were described as number and percentage. Group comparisons were carried out using the c 2 test, which was used to compare qualitative variables. Student's t-test was used to compare two independent groups in terms of quantitative variables. Correlation: Pearson's correlation coefficients (r) were calculated for the detection of parametric correlations, whereas Spearman's correlation coefficients (r) were calculated for the detection of nonparametric correlations between variables in one group. P value less than 0.05 was considered significant; P value less than 0.01was considered highly significant.

Results
Our study included two groups: groups I and II.
Group I included 20 healthy volunteers (10 men and 10 women) were included in the control group. Their age ranged from 18 to 37 years (mean 26.8 ± 6.2 years).
No statistically significant difference was found between the patients and the controls in age and sex. Laboratory and radiological findings in both groups also showed no statistically significant difference in erythrocyte sedimentation rate, CRP positivity, and CRP titer. However, there was a statistically significant difference between our two groups in HLA-B27 (P < 0.001) and there was a highly statistically significant difference in radiological findings of sacroiliitis (P < 0.001); the control group did not have sacroiliitis and were not positive for HLA-B27.

Demographic data and primary complaint in patient subgroups
(1) Group IIA included 40 (66.7%) spondyloarthropathic patients with sacroiliitis, 28 (70%) men and 12 (30%) women. Their age ranged from 18 to 39 years (mean 26.9 ± 4.6 years), and their disease duration ranged from 4 to 22 months No statistically significant difference was found between patient subgroups in age and sex, whereas we found a statistically high significant difference in disease duration (P < 0.001) and primary complaint (P < 0.001) ( Table 2).
The different manifestations and clinical examinations in spondyloarthritis patients are summarized in Tables 3 and 4. We found a highly statistically significant difference between both subgroups in inflammatory back pain (P < 0.001), NSAIDs efficacy (P < 0.001), and enthesitis (P < 0.004). Also, a statistically significant difference was found between the subgroups in swollen joint count (P < 0.01), tender joint count (P < 0.02), and clinical sacroiliitis (P < 0.006), whereas no statistically significant difference was found between the subgroups in other variables.
On comparing the activity indices in patients subgroups, we found a highly statistically significant difference between SPA patients subgroups in BASMI (P < 0.001) ( Table 5).
Laboratory and radiological findings in the patient subgroups are summarized in Table 6; there was a highly statistically significant relationship between patients with positive sacroiliitis and the positivity of HLA-B27. There was no statistically significant difference in other laboratory variables. All patients with radiological sacroiliitis showed sacroiliitis by MRI. Table 7 shows no significant correlation between the MASES and HLA-B27-positive patients, and active sacroiliitis by MRI in SPA patients.

Comparison of sacroiliitis shown by magnetic resonance imaging and plain radiography
Sacroiliitis was detected in 40 patients (30 bilateral, 10 unilateral) by MRI and in 12 patients (eight bilateral,    BME was detected in 48 (68.5%) joints with sacroiliitis by MRI (Figs. 1-4). STIR could demonstrate all of the joints with BME. BME predominantly occurred in the cartilaginous portion and at the iliac side of the joints. Enhancement of bone was observed in 46 (65.7%) joints. Enhancement was observed in both ligamentous and cartilaginous portions of the joints, but most often in the latter and on the iliac sides. BME and contrast enhancement occurred simultaneously in 46 (65.6%) and were absent in 24 (34.3%) of the joints with sacroiliitis. BME was observed in two (2.8%) of the joints with sacroiliitis without signs of enhancement.
Enhancement in the joint space was observed in 28 (40%) of the joints with sacroiliitis by MRI. It occurred in both joint portions, but more frequently in the cartilaginous than in the ligamentous portion of the joints. Also, all these 28 SIJ had an abnormal high signal in the joint space on STIR. BME and enhancement in the joint space were observed simultaneously in 18 (25.7%) joints. In 30 (42.9%) of the joints, BME was observed without enhancement in the joint space. In 10 (14.3%) of the joints, enhancement in the joint space was present without BME. Considering the fact that enhancement in the joint space and BME are findings compatible with active inflammation, inflammatory changes in 58 joints might be significant for active sacroiliitis.

Discussion
The diagnosis of sacroiliitis is important in planning a treatment of the disease and in establishing a followup protocol for patients with SPA. This study showed that MRI had the maximum sensitivity (66.7%) for detecting sacroiliitis. Their utility was most apparent in patients with disease duration less than 2 years. Plain radiographs showed the least sensitivity (20%) in detecting sacroiliitis. MRI was also very useful in the subgroup of patients with USPA where the radiographs were universally negative. This was in agreement with Shanmuganandan and colleagues [15,16], who reported that MRI had the maximum sensitivity (78%) for detecting sacroiliitis, followed closely by bone scan (73%). Their utility was most apparent in patients with disease duration less than 2 years. They were also very useful in the subgroup of patients with USPA.
In our study, we used MRI in the evaluation of SIJs. MRI has excellent soft-tissue contrast resolution, enabling a clear distinction of the two portions of the SIJ on the basis of a well-delineated differentiation between fatty tissue in the ligamentous portion and cartilage in the cartilaginous portion. This was in agreement with other studies [15,17,18].
In our study, MRI visualizes early active inflammatory changes in the form of BME and contrast enhancement, enabling the diagnosis of sacroiliitis by MRI before definite joint destruction is detectable by radiography. This was in agreement with previous authors [15,[19][20][21][22]. MRI is documented in the literature, as in our study, as a unique imaging method for the detection of early and active sacroiliitis [15,23]. Advances in superficial coil technology, invention of new sequences, and fast visualization methods are the factors increasing the sensitivity of MRI in the detection of early sacroiliitis [15,24].
Our findings support the concept that MRI is very useful for the early diagnosis of SPA. It has been reported that the combination of MRI features typical SJC, swelling joint count; TJC, total joint count; *P < 0.05, significant; **P < 0.001, highly significant.  of sacroiliitis with IBP and HLA-B27 positivity has a higher probability of being diagnosed as axial SPA in patients with well-established symptoms and this was in agreement with previous studies [25,26].
Some previous studies have examined the utility of MRI in early SPA. One longitudinal study investigated MRI, computed tomography, and radiographic changes in the SIJs over 1 year in 34 patients with early inflammatory back pain (IBP) [27]. No correlation was found between MRI scores and radiographic changes at follow-up. Oostveen et al. [21] investigated 25 HLA-B27-positive patients to assess the diagnostic value of MRI in the detection of early sacroiliitis. These patients were followed up for 3 years; nine of 14 patients showed structural changes in sacroiliitis by MRI and six out of nine patients showed inflammatory changes on MRI at baseline and subsequently developed AS. Oostveen et al. [21] emphasized the positive predictive value of structural changes observed on MRI for future radiographically evident AS rather than the predictive role of inflammatory changes.
In the present study, we focused on grading acute inflammatory osteitis in the form of BME rather than structural changes since Van der Heijde et al. [28] concluded that scoring inflammation is more important than scoring structural changes on MRI. Also, in the present study of very early disease, we noted that structural changes were uncommon.
In our study, the advantages of MRI compared with plain radiography in evaluating sacroiliitis include detection of BME and detection of enhancement in the joint space, which was in agreement with another study [15]. The capability of MRI to distinguish between acute and chronic changes and estimate the degree of disease activity can be beneficial in monitoring the effect of pharmacological treatment; the same result was reported by other authors [15,29].

Conclusion
MRI is the preferred modality in patients with early SPA and in those with USPA as it is now clear that early diagnosis can enable early treatment, which has been shown to yield significant symptomatic and radiological improvements. MRI can distinguish between acute and chronic changes. This capability would be beneficial for classifying different forms of SPAs and monitoring disease progression before, during, and after any drug treatment. MRI should therefore be used whenever possible to evaluate sacroiliitis. HLA-B27 positivity is a highly useful predictor of early sacroiliitis. CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; HLA-B27, human leukocyte antigen-B27; a Number of positive patients (%); b Suspicious radiological sacroiliitis: radiological sacroiliitis that did not fulfill the modified New York criteria for ankylosing spondylitis, positive active sacroiliitis by MRI: bone marrow edema at the coronal short tau inversion recovery sequence; **P < 0.01, highly significant.