Background Inflammatory bowel disease (IBD) is associated with a myriad of extra-intestinal manifestations (EIMs). Musculoskeletal EIMs are the most common EIM in IBD. Previous studies of sacroiliitis (SI) in IBD have a broad prevalence range of 2%>45%1–3. In these patients, HLA-B27 positivity ranged anywhere from 0–50%1,2,4.
Objectives To determine the prevalence of radiographic SI in a large IBD population, and identify risk factors associated with the development of SI.
Methods 1,247 consecutive patients from our IBD clinic, 350 with ulcerative colitis (UC), and 897 with Crohn's disease (CD) were investigated as long as they had a pre-existing computed tomography (CT) of the abdomen and pelvis. These CTs were re-interpreted for evidence of SI. Each sacroiliac joint was graded based on the modified New York Criteria5. Each patient was classified as having no disease, suspected disease, or definite disease. Definite SI was defined as bilateral grade 2, or unilateral grade 3 or 4. Suspicious SI was defined as unilateral or bilateral grade 1 or unilateral grade 2. Gender, age, type of IBD, HLA-B27 status (by next generation sequencing), and IBD serologies (ASCA, antiCBir1, anti-OMPc, ANCA) were also analyzed.
Results Prevalence of either definite or suspicious SI was found in 16.1% and 19.4% in CD and UC respectively. The prevalence of definite SI was 8.7% and 8.1% in CD and UC respectively. In patients with SI (definite and suspicious), 57% were male and 43% were female. When looking only at the definite group, there 68% males and 32% females (p=0.002). This is in contrast to the suspicious SI group, where there were 48% males and 52% females. Of significance, 16% of patients in the definite SI group were positive for HLA-B27, while only 6% of the patients in the negative group were positive for HLA-B27 (OR 4.36, p=0.006). In the suspicious group, 8% of patients were positive for HLAB27. Definite SI was associated with: anti-Cbir1 (p=0.002) in CD, and ASCA (p=0.05) in UC.
Conclusions In this study, the largest of its kind, we observed an overall SI prevalence of 17% in our IBD cohort. The prevalence of SI seen on CT was similar between UC and CD. Historically, UC and CD occur evenly between genders. However, when gender and SI in IBD were compared, there were more males in the definite group, while in the suspicious group there was a greater predominance of females. It remains unclear if the suspicious group represents a subtle phenotype of SI that has not yet progressed. Significant associations with clinically relevant subsets of IBD defined by serological markers were observed. The observed HLAB27 positivity is at the lower end of what has previously been reported, and is not as strong as is seen in ankylosing spondylitis without IBD. This further suggests that SI and spondylitis in the IBD population are different pathologic entities, with unique genetic risks, compared with the pure AS population.
Palm O. et al, Rheumatology,2001
Turkcapar N. et al, Rheumatol Int,2006
Steer S. et al, J Rheumatol,2003
Peeters H. et al, Ann Rheum Dis,2004
van der Linden S. et al, Arthritis Rheum,1984
Disclosure of Interest A. Nangit: None declared, D. Scaramangas: None declared, G. Dejesus: None declared, D. Karayev: None declared, C. Williams: None declared, D. Li: None declared, X. Yan: None declared, S. Targan: None declared, T. Learch: None declared, D. McGovern: None declared, M. Weisman Grant/research support from: UCB, Human Genome Sciences, Sanofi, Eli Lilly and Co, Genentech, Inc., Santarus Inc., EMD Serono, ChemoCentryx, GSK, Immunomedics Inc., Consultant for: Boehringer Ingelheim/Proskauer, Ardea Biosciences, Epirus Biopharmaceuticals, Acerta Pharma