Time to diagnosis in systemic lupus erythematosus: Associated factors and its impact on damage accrual and mortality. Data from a multi-ethnic, multinational Latin American lupus cohort.

Abstract

Systemic lupus erythematosus (SLE) often mimics symptoms of other diseases, and the interval between symptom onset and diagnosis may be long in some of these patients. Aims: To describe the characteristics associated with the time to SLE diagnosis and its impact on damage accrual and mortality in patients with SLE from a Latin American inception cohort. Patients were from a multi-ethnic, multi-national Latin-American SLE inception cohort. All participating centers had specialized lupus clinics. Socio-demographic, clinical/laboratory, disease activity, damage, and mortality between those with a longer and a shorter time to diagnosis were compared using descriptive statistical tests. Multivariable Cox regression models with damage accrual and mortality as the end points were performed, adjusting for age at SLE diagnosis, gender, ethnicity, level of education, and highest dose of prednisone for damage accrual, plus highest dose of prednisone, baseline SLEDAI, and baseline SDI for mortality. Of the 1437 included in these analyses, the median time to diagnosis was 6.0months (Q1-Q3 2.4-16.2); in 721 (50.2%) the time to diagnosis was longer than 6months. Patients whose diagnosis took longer than 6months were more frequently female, older at diagnosis, of Mestizo ethnicity, not having medical insurance, and having "non-classic" SLE symptoms. Longer time to diagnosis had no impact on either damage accrual (HR 1.09, 95% CI 0.93-1.28, p = 0.300) or mortality (HR 1.37, 95% CI 0.88-2.12, p = 0.200). In this inception cohort, a maximum time of 24months with a median of 6months to SLE diagnosis had no apparent negative impact on disease outcomes (damage accrual and mortality).