Characteristic features of late-onset systemic lupus erythematosus: An observational study of data from the Lupus Registry of Nationwide Institutions

Objective: Late-onset systemic lupus erythematosus (LoSLE) is known to possess characteristics different from those of early-onset SLE (EoSLE), thereby making their diagnosis difficult. This study aimed to assess the characteristic features of LoSLE in Japan, a model country with a super-aged society.

Methods: Data were obtained from the Lupus Registry of Nationwide Institutions, which includes a multicenter cohort of patients with SLE in Japan who satisfied the 1997 American College of Rheumatology revised classification criteria for SLE. Data were compared between patients with LoSLE (≥50 years old at onset) and EoSLE (<50 years old at onset). To identify factors associated with LoSLE, binary logistic regression was used for the multivariate analysis, and missing values were complemented by multiple imputations. We also conducted a sub-analysis for patients diagnosed within 5 years of onset.

Results: Out of 929 enrolled patients, 34 were excluded owing to a lack of data regarding onset age. Among the 895 remaining patients, 100 had LoSLE, whereas 795 had EoSLE. The male-to-female ratio was significantly higher in the LoSLE group than in the EoSLE group (0.32 vs 0.11, p < 0.001). With respect to SLEDAI components at onset, patients with LoSLE exhibited a higher frequency of myositis (11.9% vs 3.75%, p = 0.031), lower frequency of skin rash (33.3% vs 67.7%, p < 0.001), and lower frequency of alopecia (7.32% vs 24.7%, p = 0.012). No significant differences in overall disease activity at onset were observed between the two groups. Regarding medical history, immunosuppressants were more commonly used in EoSLE. A multivariate analysis revealed that a higher male proportion and a lower proportion of new rash at onset were independent characteristic features of LoSLE. We also identified late onset as an independent risk factor for a high SDI score at enrollment and replicated the result in a sub-analysis for the population with a shorter time since onset.

Conclusions: We clarified that LoSLE was characterized by a higher male proportion, a lower frequency of skin rash and a tendency to organ damage. Now that the world is faced with aging, our results may be helpful at diagnosis of LoSLE.