Age associated clinical manifestations of systemic lupus erythematosus: a multivariate regression analysis

Abstract

The influence of age on the prevalence of individual clinical manifestations of systemic lupus erythematosus (SLE) has not been adequately distinguished from racial or gender influences. Therefore, we examined variations in the clinical manifestations of SLE with age in a group of 361 patients. Multivariate regression techniques, including logistic regression and analysis of covariance, were used to identify clinical features associated with age, while controlling for important confounding factors, including race, gender, duration of followup, and treatment effects. Lymphopenia was found more frequently with increasing age, while malar rash, seizures, false-positive VDRL, thrombocytopenia (in whites), proteinuria (0.5-3.5 g/day), elevated antidouble stranded DNA antibodies, and hypocomplementemia were found less frequently. No age relationship was found for the prevalence of 16 of 24 clinical features examined, including the important disease manifestations of arthritis, serositis, psychosis, nephrotic-range proteinuria, renal failure, autoimmune hemolytic anemia, and leukopenia. The use of regression analysis allows the recognition of similarities and differences in cumulative clinical features of SLE due to age in isolation from the effects of other demographic factors.