Background Spondyloarthritis (SpA), one of the most common inflammatory joint diseases in France is associated with HLA-B27. However data on its prevalence remain scarce. Moreover, the precise frequency of SpA in HLA-B27 individuals has never been directly determined at the population level.
Objectives to estimate SpA and HLA-B27 prevalence in the large observational French cohort GAZEL.
Methods GAZEL is an ongoing general population cohort of 20,000 subjects that were enrolled in 1989 among employees of the French national gas and electricity company. In 2010, all participants to the study, aged from 56 to 71 years at that time, were screened by a questionnaire previously validated for the detection of SpA (1). Respondents with stored DNA having declared to be affected with SpA or psoriatic arthritis were then interviewed by phone by a rheumatologist (RSN) to verify clinical grounds of the diagnosis. Anteroposterior radiograph of the pelvis was obtained in all patients suspected of having SpA. HLA-B27 typing was performed by PCR in confirmed SpA patients and in healthy controls without any SpA feature. In addition to HLA-B27 typing, single nucleotide polymorphism (SNP) rs4349859, known to strongly tag HLA-B27 allele, was genotyped by Taqman®.
Results 18,757 subjects were screened for participation. The overall response rate was 74.4%. Among the responders, 6,556 had stored DNA; their male:female ratio was 3.6:1 and their mean age was 65.5±3.3 years. Seventy of them (1.1%) declared to have SpA or psoriatic arthritis. A diagnosis of SpA was confirmed in 25 (0.38% [0.25-0.57%]): 16 had ankylosing spondylitis, 4 psoriatic SpA, 1 SpA associated with Crohn’s disease and 4 undifferentiated SpA. Moreover, 5 cases of peripheral psoriatic arthritis without axial involvement were diagnosed. Diagnosis remained doubtful in 6 cases. 76% [54.5-89.8%] of the confirmed SpA were HLA-B27 positive. Among 2,551 individuals without any SpA feature, HLA-B27 positivity rate was 6.9% [5.9-7.9%]. A lower rate of HLA-B27 was observed in subjects older than 70 years (5.1% [3.4-7.4%]), as compared to younger subjects (7.3% [6.2-8.6%]), that was close to significance (p=0.06). The estimated prevalence of SpA in HLA-B27+ individuals was 4.1% [2.5-6.4%] with a higher prevalence in men, as compared to women (Table 1). Finally, SNP rs4349859 showed high sensitivity (97.0%) and high specificity (99.7%) for tagging HLA-B27.
Conclusions In a French population aged of more than 55 years and predominantly male, the estimated prevalence of SpA was 0.38% [0.25-0.57%]. The relative risk of SpA in HLA-B27+ individuals was 43 [17-108]. The estimate prevalence of SpA in HLA-B27+ individuals was 4.1% [2.5-6.4%]. As previously reported, we observed a lower rate of HLA-B27 positivity in older subjects (2).
Said-Nahal R, et al. Arthritis Rheum (Arthritis Care Res) 2001;45:478-84
Reveille JD, et al. Arthritis Rheum 2012;64:1407-11
Disclosure of Interest None Declared