Association of Lupus, Sjogren, Vasculitis, Celiac Disease and Psoriasis with CAC at baseline and incidence/progression CAC, all-cause and CDV mortality in the 8-year follow-up of ELSA-Brasil

Abstract

Studies suggest that Inflammatory Diseases (ID) such as psoriasis, Systemic Lupus Erythematosus (SLE), Celiac Disease, Sjögren's Syndrome and Vasculitis would be associated with cardiovascular (CDV) Risk Factors (RF), subclinical atherosclerosis measured by calcium score (CAC) and Carotid Intima-Media Thickness (CIMT), resulting in an increased risk of fatal and non-fatal CDV outcomes and a higher overall mortality. ELSA-Brasil is a prospective cohort study focused on study of classical and non-classical RF for CDV diseases and their relationship with incidence of CDV disease and diabetes. Baseline data (wave 1) were obtained between 2008-2010, with a follow-up of 4 (wave 2) and 9 years (wave 3). The objectives of this project are: 1) to assess the association of each ID diagnosis listed above with incidence/progression of CAC and CIMT compared to participants without disease; 2) to assess the association of ID with general morbidity and mortality, CDV and other causes; 3) to assess the presence of lipid paradox described in rheumatoid arthritis where studies showed lower LDL-cholesterol levels with increased risk of morbidity and mortality in each ID by dosages of lipids not only with traditional methodology, but also by innovative techniques such as nuclear magnetic resonance and vertical ultracentrifugation. It will be identified the baseline participants with previous medical diagnosis or high probability of ID from the study questionnaires, from annual calls about participants health, and from investigation of previous hospitalization, results of exam and use of specific medications for long-term ID treatment. Biological material samples from the participants who are identified at this stage will be submitted to analysis of specific biomarkers of each ID for diagnosis confirmation. Participants will be contacted to clarify doubts regarding the diagnosis if necessary. As part of the approach to be used to face the scientific and technological challenges of the project and how to overcome these challenges, the ELSA-Brasil has a representative random sample of 1500 participants from the cohort, who would represent the 15105 participants and will be the group without disease in all comparisons. This allows the measurement of biomarkers in a subsample of 3000 participants (cases) with cost savings, but without losing the power of analysis. It will be analyzed biological material stored in the three waves of the study, according to the time of the appearance of ID. The lipid particles were dosed by innovative techniques, allowing to explore the lipid paradox with new analysis strategies. Some analyzes will be constructed as a case-cohort study using as controls participating in the random sample of the cohort or nested case-control using participants matched for age and sex without disease as the controls. Cross-sectional and prospective analyzes will also be used depending on the specific objectives. ID identified at baseline can be analyzed with prospective data from waves 2 (for CAC incidence/progression) or 3 (for CIMT incidence/progression), or cross-sectional analysis by using all ID diagnosis from waves 1, 2, and 3. Prospective results will be presented by Poisson regression or by Cox proportional hazards with the calculation of relative risk and presented unadjusted, adjusted by sociodemographic risk factors, and with multivariate adjustment by probable confounding factors, depending on whether the variable be obtained at presential visits, where the time range is very similar between the participants or at different periods across the time as in case of fatal and non-fatal events, when Cox will be more indicated. Survival analysis with Kaplan-Meier curves compared by the log-rank test will be used. Data analyzes using machine learning techniques are also foreseen in this project aiming to identify sociodemographic and clinical variables which may be predictives in term of prognostic. (AU)