Causes and predictors of death in Brazilian lupus patients

The objective of this study is to determine the causes and predictors of death in systemic lupus erythematosus (SLE) patients. Causes of death were defined based on death certificates, medical records, and information collected from doctors and relatives. Possible variables predicting mortality were assessed by Kaplan-Meier and Cox regression methods. The multivariate model was validated using the bootstrap method, and the hazard ratios were adjusted according to the shrinkage coefficient. One hundred eighty-one patients were included, and two patients were lost to follow-up. The median (IR) age at T (0) and disease duration of the 179 patients were 26.7 (21.8-34.6) and 8.2 (4.3-12.4) years, respectively. After a median (IR) follow-up of 3.3 (3.1-3.5) years, 13 (7.3 %) patients died due to end-organ failure (5), infection (5), disease activity (1), and atherosclerotic cardiovascular disease (CVD) (1). The cause of mesenteric ischemia in one patient could not be determined. Predictors of mortality collected at T(0) were the following: nephritis, chronic kidney disease, antiphospholipid syndrome (APS), higher modified SLEDAI-2k, higher damage index score, intravenous cyclophosphamide use, higher daily dose of prednisone, and higher systolic blood pressure. Independent predictors of mortality were higher damage index score (HR: 1.40; 95 % CI: 1.08-1.82), cyclophosphamide use (HR: 3.80; 95 % CI: 1.13-12.77), and APS diagnosis (HR: 3.82; 95 % CI: 1.07-13.59). This paper presents a high frequency of late mortality in lupus patients due to the SLE itself and infection. This result is not in agreement with the initial proposed bimodal pattern of lupus mortality, nor is it in agreement with the high frequency of CVD as a cause of death in developed countries. The most important predictors of death were related to the lupus itself.