Background Systemic lupus erythematosus (SLE) is an autoimmune disease that displays clinical heterogeneity. The detailed distributions of its symptoms and clinical markers are unknown.
Objectives We performed a nation-wide study to determine the distributions of the signs and clinical markers of SLE and identify any patterns in their distributions to allow patient subclassification.
Methods We obtained a total of 256,999 patient-year records describing the disease statuses of SLE patients, including the diagnostic symptoms that they possessed, from the Japanese national registry system from 2003 to 2010. Of these, 14,779 involved patients who had been diagnosed with SLE within the last year, and 242,220 involved patients who were being followed-up. In addition to calculating basic descriptive statistics, we performed multiple logistic regression analyses to analyze the effects of sex, age, and disease duration on the frequencies of symptoms in the first year and follow-up years. The distributions of SLE complications were also analyzed. The patients and diagnostic symptoms were clustered using the Ward method.
Results The age-standardized incidence and prevalence of SLE were estimated to be 1.94 and 35.21 per 100,000 people, respectively, and the female:male ratio was 8.23. The female patients were significantly younger at onset than the male patients(p<1.0x10-10). The frequency of a relevant family history was associated with a younger age-at-onset of SLE(p=8.0x10-5). Renal involvement and discoid eczema were more frequent in males(p<1.0x10-10), whereas arthritis, photosensitivity, and cytopenia were more common in females(p<1.0x10-10). Autoantibody production and malar rash were positively associated with young age(p<1.0x10-10), and serositis and arthritis were associated with old age(p<1.0x10-10). Photosensitivity was positively associated with a long disease duration(p<1.0x10-10), and autoantibody production, serositis, and cytopenia were associated with a short disease duration(p<1.0x10-10). The SLE patients were clustered into ten and eight sub-groups in the first and follow-up years, respectively. Diagnostic symptoms of SLE were clustered into two subgroups in both first and follow-up years with difference of finer cluster divisions. The first group included hematoserological abnormalities and the second group included organ-specific abnormalities. The age-standardized risks of cerebral and myocardial infarction were 3.79 (2.87-4.72) and 2.44 (2.16-2.72) times higher in the SLE patients compared with the general population, respectively, and the increase in risk was positively associated with the number of positive tests for APS-related markers (p<0.0083 and 0.00020, respectively).
Conclusions Along with the well-known female predominance of SLE, we identified differences in its clinical features between the sexes and patients with different onset ages. Also our large-scale study indicated that subgroups of SLE patients and the diagnostic symptoms of SLE exist.
Disclosure of Interest None Declared