Associations of clinical features and prognosis with age at disease onset in patients with systemic lupus erythematosus

The objective of this study is to evaluate the association of clinical features and prognosis with age at disease onset in patients with systemic lupus erythematosus (SLE) in a large, multicenter Chinese cohort. Medical records of 1898 SLE inpatients from 15 hospitals were reviewed and classified into three groups according to their ages at disease presentation. Categorical data were analyzed by chi-square test and potentially associated factors were tested by multinomial logistic regression. Among the patients studied, 259 (13.6%) were juvenile onset (≤18 years), 1444 (76.1%) were early onset (>18 and ≤45 years) and 195 (10.3%) were late onset (>45 years). Whenever manifestations occurred, most patients (>80%) were diagnosed within two years. Juvenile-onset patients were more likely to be untreated before admission (p < 0.001) and have mucocutaneous manifestations (p < 0.001), but musculoskeletal symptoms (p < 0.05) and leukopenia (p < 0.05) were less frequent, while comorbidities were much higher in patients with late-onset SLE (p < 0.001). Neuropsychiatric, cardiopulmonary, renal and gastrointestinal involvement, disease activity index and damage scores were similar among three groups. Anti-Sm antibodies were less prevalent in late-onset patients (p < 0.05) and antimalarial drugs were more often applied to juvenile-onset patients (p < 0.001). As expected, mortality was elevated in the late-onset SLE group (p < 0.05), in which nearly half died of infections, which was much higher than those in the other two groups (p < 0.001). Logistic regression confirmed that patients with juvenile- and early-onset disease were associated with high incidence of being untreated prior to admission, and with low incidence of comorbidities as well as deaths caused by infection compared to patients with late-onset lupus. Interestingly, our data showed that more patients with late-onset disease had a SLEDAI score change of >7 at discharge. In conclusion, age at onset has an impact on SLE disease status, and infection is the main cause of death in those with late-onset lupus. Considering that the late-onset patients had simultaneously easily controllable diseases and high incidence of comorbidities, a different treatment strategy from younger patients should be considered.