Abstract
Objective. To analyze the clinical effect of enthesitis in a large Brazilian cohort of patients with spondyloarthritis (SpA).
Methods. A common protocol of investigation was prospectively applied to 1505 patients with SpA in 29 centers in Brazil. Clinical and demographic variables and disease indexes were investigated. The Maastricht Ankylosing Spondylitis Enthesitis Score was used to investigate the enthesitis component. Ankylosing spondylitis was the most frequent disease in the group (65.4%). Others were psoriatic arthritis (18.4%), undifferentiated SpA (6.7%), reactive arthritis (3.3%), and enteropathic arthritis (3.2%).
Results. At least 1 affected enthesis was observed in 54% of the patients with SpA, with a mean of 2.12 ± 2.98 entheses affected. According to the clinical presentation, enthesitis was significantly more frequent in patients with axial + peripheral joint involvement compared to isolated axial or peripheral involvement (p < 0.001). There was a statistical association between the presence of enthesites and axial symptoms (buttock pain, cervical pain, and hip pain), and peripheral symptoms (lower limb arthritis, number of painful and swollen joints; p < 0.05). Patients with enthesitis also presented higher mean scores of Bath Ankylosing Spondylitis Functional Index (BASFI; p < 0.001), Bath Ankylosing Spondylitis Disease Activity Index (p < 0.001), and Ankylosing Spondylitis Quality of Life (ASQoL; p < 0.001). Multivariate logistic regression showed that BASFI (p < 0.0001; OR 74.839), ASQoL (p = 0.0001; OR 14.645), and Achilles tendonitis (p = 0.0059; OR 7.593) were associated with work incapacity.
Conclusion. The clinical presence of enthesitis in this large cohort of patients with SpA was frequent and was associated with a significant increase in disease activity and decline in functional capacity and quality of life.
- SPONDYLOARTHRITIS
- ANKYLOSING SPONDYLITIS
- PSORIATIC ARTHRITIS
- ENTHESITIS
- MAASTRICHT ANKYLOSING SPONDYLITIS ENTHESITIS SCORE
The spondyloarthritides (SpA) represent a heterogeneous group of chronic inflammatory diseases with a common clinical and genetic background, affecting around 1% of the population. Ankylosing spondylitis (AS) is the core disease in the group, which includes psoriatic arthritis (PsA), reactive arthritis (ReA), arthritis associated with inflammatory bowel disease (IBD), and undifferentiated SpA (USpA)1. Because all diseases in the SpA group can present enthesitis involvement, enthesitis was included as an important variable in the classification criteria of axial SpA2 and peripheral SpA3.
Since the differentiation of SpA from rheumatoid arthritis (RA) by Moll and Wright, SpA has been characterized by spinal inflammation, peripheral joint oligoarthritis, distinctive radiographic changes, and enthesitis, associated with HLA-B274. Entheses are adapted to resist shear and compression forces and thus minimize joint damage associated with the high levels of mechanical loading that they bear5. Although the traditional concept of enthesitis relates to focal inflammatory changes at the insertion of tendons, ligaments, fascia, or capsule into bones, several entheses comprise more than simply the insertion site. This concept of functional entheses represents a unifying basis for SpA6.
In patients with PsA, the most frequently affected sites of enthesitis are Achilles tendon, plantar fascia, costochondral joints, and elbows. Patients with clinical complaints frequently show abnormalities on ultrasound (US) scans7. Patients with cutaneous psoriasis and no joint symptoms can also present subclinical entheseal inflammation8. Even in normal joints, normal insertions are associated with microdamage and inflammatory changes, suggesting that local tissue-specific (autoinflammatory) factors may dictate disease expression9.
Musculoskeletal complaints are the most common extraintestinal manifestations of IBD, presenting as articular (arthritis) or periarticular (enthesitis, myositis, or soft-tissue rheumatism) inflammation; although its clinical course can be independent, most frequently enthesitis correlates with the intestinal activity10. In ReA, enthesitis can be more frequent than arthritis11. Positron emission tomography/computed tomography can be useful for the early diagnosis of enthesitis in early stages of ReA, even before the possible confirmation of the infectious site12. US is useful for the detection of affected entheses in young patients with poststreptococcal ReA13.
Because enthesitis is a primary clinical feature in SpA, various instruments were proposed to investigate it. In 1987, Mander, et al proposed an index (the Mander Enthesis Index; MEI) analyzing 66 entheses by local pressure, with intensity of pain graded on a 0–3 scale14. This instrument is not practical for daily use because it involves many sites and is time-consuming. In 2003, a group of experts from the Assessment of SpondyloArthritis International Society proposed an alternative to the MEI with much better feasibility, the Maastricht Ankylosing Spondylitis Enthesitis Score (MASES); it analyzes 13 anatomical sites15. In 2009, the Spondyloarthritis Research Consortium of Canada (SPARCC) conducted clinical validation of an enthesitis instrument, for which the selection of entheses sites was based primarily on published sonographic and magnetic resonance imaging (MRI) studies, the SPARCC index16.
The main objective of our study was to analyze the importance of the enthesitis involvement in the different disease patterns in a large cohort of 1505 patients with SpA.
MATERIALS AND METHODS
This was a prospective, observational, multicenter cohort of consecutive patients with SpA recruited from 29 referral centers participating in the Brazilian Registry of Spondyloarthritis (RBE – Registro Brasileiro de Espondiloartrites). All patients, from each of the 5 major geographic areas in Brazil, were classified according to the European Spondylarthropathy Study Group criteria17. The data were collected from June 2006 to December 2009. The RBE is part of the RESPONDIA group, consisting of 9 Latin American countries (Argentina, Brazil, Costa Rica, Chile, Ecuador, México, Perú, Uruguay, and Venezuela) and Spain and Portugal.
In our study, a protocol of investigation was applied to 1505 patients with SpA. The diagnosis of AS was considered if the patients fulfilled the New York modified criteria18, and for PsA if they fulfilled the Moll and Wright criteria19; ReA was considered when asymmetric inflammatory oligoarthritis of lower limbs was present, associated with enthesopathy and/or inflammatory low back pain following enteric or urogenital infections20, and enteropathic arthritis when the patient presented inflammatory axial and/or peripheral joint involvement associated with confirmed IBD (Crohn’s disease or ulcerative colitis).
All patients were interviewed and clinically examined by the study center coordinator. Our protocol of investigation considered as axial the predominant involvement of spinal joints that could also affect girdle (hips and shoulders) and/or anterior chest wall joints; peripheral as the predominant involvement of large and/or small joints of upper and/or lower limbs; enthesitic when predominantly entheses were affected; and “mixed” when there was not a predominance of axial or peripheral involvement.
Demographic and clinical data were collected including time of disease duration, tender and swollen joint count, and visual analog scale (VAS) for pain according to the patient. A VAS for patient and physician global perception of disease activity (patient global and physician global assessment, respectively) was also collected. Peripheral articular involvement was assessed by the 66 tender/swollen joint count.
Patients were asked whether they have a family member (2 generations) with known diagnosis of AS, PsA, ReA, arthritis related to IBD, or USpA. Family members were not assessed.
To analyze physical activity, we asked the patients about the daily mean time that they spent on physical exercises. We considered “total job incapacity” when the patient was completely and definitely unable to do his/her job, and “partial job incapacity” when the patient was able to do his/her activities only during some periods in the year, with some periods of incapacity.
Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) values were also registered.
The index for evaluating enthesis used in this study was the MASES, which evaluated pain at 13 anatomical sites (11 axial and 2 peripheral)15. The enthesitic sites studied were bilateral first and seventh costochondral joints, anterior-superior and posterior spina iliaca, iliac crests, insertion of the Achilles tendon into the posterior surface of the calcaneus, and the fifth lumbar spinous process. Before the start of data collection, the coordinators of all the participating centers attended 2 training meetings, where the MASES was demonstrated.
Disease activity and functional status were evaluated according to the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI)21 and the Bath Ankylosing Spondylitis Functional Index (BASFI)22, respectively. Quality-of-life data were recorded through Ankylosing Spondylitis Quality of Life (ASQoL) questionnaires23. All questionnaires used were previously translated, cross-translated, validated, and culturally adapted to the Portuguese language24. Radiological evaluation was performed by the Bath Ankylosing Spondylitis Radiology Index (BASRI), including BASRI-spine (lumbar and cervical spine and sacroiliac joints) and BASRI-total (BASRI-spine and BASRI-hips)25.
Statistical analysis
After the descriptive analysis of each enthesis and their distribution according to clinical presentation (axial, peripheral, axial + peripheral) and specific disease into the SpA group, the patients were divided into a group “with enthesitis” and another “without enthesitis.” Categorical variables were compared by chi-square and Fisher’s exact test, and continuous variables were compared by ANOVA test. A value of p < 0.05 was considered significant, and 0.05 > p > 0.10 was considered a statistical trend. To analyze the correlation among enthesitis and work capacity, 2 multivariate logistic regressions were performed: the first analyzed the 7 enthesitic sites and work incapacity, and the second analyzed all the demographic and clinical variables involving work incapacity.
RESULTS
Among the 1505 patients with SpA, AS was the most frequent disease (65.4%). The others were PsA (18.4%), USpA (6.7%), ReA (3.3%), and enteropathic arthritis (3.2%). At least 1 affected enthesis was observed in 54% of the patients.
The mean number of entheses affected was 2.12 ± 2.98. The frequency of involvement of each enthesis in the MASES index is shown in Table 1; posterior spina iliaca and Achilles tendon were the most frequently affected entheses. When the degree of interdependency of the 7 different enthesitic sites was analyzed, it was shown that the presence of any of the 7 enthesitic sites was significantly associated with the other sites, with the only exception the fifth lumbar spinous process. Among the 54% of patients with enthesitis, 24.6% had 1 affected enthesis and 29.4% had ≥ 2 enthesites (10.6% two entheses, 7.8% three, 5.7% four, 3.1% five, 2% six, and 0.2% seven). When just 1 enthesis was painful at the physical examination, the most frequently involved entheses were the fifth lumbar spinal process (39.5% of 287 patients) and the Achilles tendon (23.6% of 293 patients). These were less often affected as a single painful enthesis: posterior spina iliaca (14.0%), first costochondral (7%), anterior-superior spina iliaca (6.5%), seventh costochondral (5.5%), and iliac crest (3.9%).
Enthesitic sites, according to the Maastricht Ankylosing Spondylitis Enthesitis Score, in 1505 patients.
Enthesitis was significantly more frequent in patients with axial + peripheral joint involvement (2.44 ± 3.08) compared with isolated axial (1.42 ± 2.43) or peripheral (1.28 ± 2.15) involvement (p < 0.001). Eighty-three patients (6.2%) were considered predominantly enthesitic, with Achilles tendonitis predominating in cases of USpA and ReA. Patients with USpA had enthesitis in 70.4% of cases, followed by PsA (53.8%), AS (53.3%), ReA (49%), and arthritis associated to IBD (47.9%). The frequencies of the affected entheses of the MASES in the different diseases of the SpA group are shown in Table 2. Non-AS diseases more frequently presented the involvement of Achilles enthesitis (p < 0.001).
Enthesitic sites of the Maastricht Ankylosing Spondylitis Enthesitis Score in 1505 patients, according to specific disease in the SpA group. Except for p values, all data are percentages.
Regarding the demographic data, the presence of enthesitis was associated with being female (p = 0.014) and with fewer daily hours of exercise (p = 0.002). Although the patients’ ages were similar, patients with enthesitis presented shorter disease duration (p = 0.001; Table 3).
Comparison of demographic variables in patients with enthesitis versus without enthesitis in 1505 patients with spondyloarthritis. Except for p values, all data are percentages unless otherwise indicated.
Analysis of the clinical data showed that patients with enthesitis had significantly higher frequency of axial pain [buttock pain (p < 0.001), cervical pain (p < 0.001), and hip pain (p < 0.001)] than the group without enthesitis. Patients with enthesitis also had significantly higher frequency of peripheral symptoms [lower limb arthritis (p = 0.010), number of painful (p < 0.001) and swollen (p = 0.05) joints], as well as higher patient (5.53 ± 2.69 vs 4.07 ± 2.88; p < 0,001) and physician (4.64 ± 2.56 vs 2.95 ± 2.45; p < 0,001) global assessments, compared with the group without enthesitis. Specific variables, such as dactylitis, psoriasis, nail involvement, uveitis, urethritis, and IBD were not statistically correlated (Table 4).
Comparison of clinical, laboratory, and disease indices data in patients with enthesitis versus without enthesitis, among 1505 patients with spondyloarthritis.
Laboratory data also showed statistical associations between enthesitis and higher ESR (27.25 ± 23.63 vs 22.20 ± 20.55; p < 0.001) and CRP (11.72 ± 24.77 vs 8.12 ± 16.02; p = 0.004; Table 4). Patients with enthesitis had higher mean scores for BASDAI (4.95 ± 2.22 vs 3.32 ± 2.26; p < 0.001), BASFI (5.17 ± 2.64 vs 3.82 ± 2.76; p < 0.001), and ASQoL (9.08 ± 5.29 vs 6.19 ± 5.08; p < 0.001; Table 4). Values for BASRI were similar among the groups.
As enthesitis was also associated with partial or total work incapacity (p < 0.001), a logistic regression was performed. First, we performed a multivariate logistic regression analyzing the 7 enthesitic sites: iliac crest (p = 0.0064; OR 7.437, 95% CI 1.159–2.461), posterior spina iliaca (p < 0.0001; OR 17.031, 95% CI 1.460–2.894), and Achilles tendon (p = 0.0023; OR 9.303, 95% CI 1.202–2.333) were associated with work incapacity. Another multivariate analysis, including all the demographic and clinical variables, showed that BASFI (p < 0.0001; OR 74.839, 95% CI 1.156–2.359), ASQoL (p = 0.0001; OR 14.645, 95% CI 1.034–1.111), and Achilles tendonitis (p = 0.0059; OR 7.593, 95% CI 1.156–2.359) were associated with work incapacity.
DISCUSSION
Our study analyzed the clinical importance and the correlations of enthesitis in 1505 patients with SpA in Brazil. Enthesitis was confirmed to be a common manifestation of SpA, affecting all the different diseases in the SpA spectrum. In this large SpA cohort, enthesitis was a frequent manifestation (54% of patients) and was strongly associated with a more severe clinical picture (axial and/or peripheral) and more significant pain and functional incapacity. We also found a direct correlation with inflammatory activity represented by higher levels of ESR and CRP.
The choice of the MASES for assessing the entheses was based on its feasibility in daily practice. After 2 training sessions, rheumatologists were able to start searching for the enthesitic component in patients with SpA; it is important to remember that the fourth question of the BASDAI is related to the enthesitic component20. It was also noteworthy that all the axial entheses were described as painful by SpA patients with statistically similar frequencies. The only peripheral enthesis in MASES, the Achilles tendon insertion, was mentioned preferentially by patients with USpA and ReA.
A major finding was the correlation between enthesitis and work incapacity. Iliac crest, posterior spina iliaca, and Achilles tendon were the entheses statistically associated with work incapacity in a logistic regression including the 7 enthesitic sites. When a multivariate logistic regression analysis was performed with all the demographic and clinical variables, BASFI, ASQoL, and Achilles tendon were associated with work incapacity. Because Achilles tendon is evaluated in MASES and SPARCC, these data are important; they characterize enthesitis as a major prognostic factor to be considered in the management of patients with AS and SpA. It is also important that 23.6% of the patients with Achilles enthesitis had just 1 affected enthesis. Because these results were obtained in a cross-sectional study, the longitudinal second phase of RBE (2013–2014), which will analyze these patients after 5 to 7 years of followup, will be vital to confirm these prognostic findings.
Some previous studies have analyzed the relationship between enthesitis and clinical measures in SpA. One study analyzed disease activity in 100 patients with SpA, through laboratory data, VAS for spinal and peripheral pain, and using MEI and MASES (both subdivided into P for patient, and D for physician), as well as BASDAI, BASFI, and BASRI26. In multiple linear regression analysis, BASFI, VAS, and female sex (41.3%) were the best predictors of MEI-D, whereas BASFI, VAS, female sex, and ESR (32.5%) were the best predictors of MASES-D. The study observed statistical correlations between MEI and MASES and BASDAI and BASFI, but did not find correlation with laboratory data26. Another study analyzed enthesitis in a group of 180 patients with SpA using MEI and Medical Outcomes Study Short Form-36 (SF-36) and found a prevalence of 84.9%, showing a striking correlation between MEI, morning stiffness, disease duration, joint pain, and 6 subgroups of SF-36; the higher correlation was found among MEI and the number of painful joints27. A study analyzing the relationship among enthesitis and clinical, laboratory, and QOL measurements in 76 patients with AS, using MASES, MEI, BASDAI, BASFI, and SF-36, found that severity of enthesitis correlated significantly with BASDAI and BASFI, but not with disease duration or laboratory variables28. Another study investigated QOL in 962 patients with the diagnosis of AS according to the modified New York criteria; BASDAI, BASFI, fatigue (question 1 of BASDAI), and pain were the most significant variables associated with QOL. Metrologic and radiographic variables were not significant29. Enthesitis was more frequent in women in our series, as observed in a previous study30. In that English study, 516 patients (344 women, 172 men) were analyzed. SpA progressed differently in women, who had more neck, knee, and hip pain at presentation and neck pain, back pain (lower and upper), fatigue, and enthesitis during the course of the disease than the men30. In our present study, those patients who presented at least 1 affected enthesis in MASES showed a significant statistical association with axial and peripheral symptoms, as well as with higher mean scores of BASDAI, BASFI, and ASQoL, confirming that enthesitis must be considered a factor for worse prognosis in SpA.
We did not use US or MRI to evaluate the affected entheses. Although power Doppler US can confirm the diagnosis of enthesitis31,32,33,34,35,36, the importance of subclinical enthesitis in patients with SpA is not yet completely understood. MRI can be superior to US for the study of entheses because it can show more clearly the initial phases of the bone marrow edema37,38, and it is the best method to analyze sacroiliitis37.
Our study has some potential biases that need to be discussed. The first is the large number of centers involved (29); there are difficulties in the standardization of the evaluation of the enthesitic component in such large groups. We tried to standardize this investigation in 2 previous meetings with the study coordinators. The second is that enthesitis was characterized only clinically, without an imaging method to confirm the diagnosis. Standardized use of US or MRI to analyze the 13 entheses of MASES in 1505 patients in 29 study centers would be difficult and expensive. The third is that because the patients with higher prevalence of enthesitis were women with higher mean ASQoL scores, it was important to assess fibromyalgia (FM) as a differential diagnosis. Some studies have analyzed the presence of FM associated with AS39,40, a clinical situation that can possibly confound the results related to enthesitis. Another study in Brazil, evaluating 71 patients with AS (54.5% women, 45.5% men), observed that FM was more prevalent among women (3.8:1) and may have influenced the higher BASDAI, BASFI, and ASQoL scores in the patients with associated FM41. FM, as well as specific questionnaires for anxiety and depression, was not assessed in our study, but all the participants were instructed to avoid the misclassification of a tender point as an enthesitis.
The enthesitic component has become so important in the context of SpA that specific studies about response to treatment were done for the entheses with the biologic agents used for treatment of SpA42,43,44, and enthesitis is considered separately in the algorithm for the treatment of PsA45.
Enthesitis is a common manifestation in the SpA group, and needs to be systematically investigated in the management of these patients, because it represents a significant increase in disease activity and a decline in functional capacity and QOL. Achilles tendonitis, an easy enthesis to examine, must be considered a prognostic factor in cohorts where patients with SpA frequently present axial and peripheral involvement, as in the Brazilian cohort.
Footnotes
-
The electronic version of the Brazilian Registry of Spondyloarthritis is supported by an unrestricted grant from Wyeth/Pfizer Brazil. Dr. Sampaio-Barros is a recipient of a research grant from Federico Foundation.
- Accepted for publication June 11, 2013.
REFERENCES
In this issue
