Abstract
PURPOSE: "Statins" have immunomodulatory and anti-inflammatory effects and may attenuate the risk of infections. We sought to determine the association between chronic statin use and long-term rates of sepsis events. MATERIALS AND METHODS: We used data from 30 183 adult (≥45 years old) community-dwelling participants in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) cohort. The primary exposure was statin use. The primary outcome was hospitalization or emergency department treatment for sepsis. Using Cox proportional hazards models, we determined associations between statin use and first sepsis events, adjusting for patients demographics, health behaviors, chronic medical conditions, degree of medication adherence, baseline high-sensitivity C-reactive protein (hsCRP), and propensity for statin use. RESULTS: Approximately one-third of participants reported statin use (n = 9475, 31.4%). During the 10-year follow-up period from 2003 to 2012, there were 1500 incident sepsis events. Statin use was not associated with rates of sepsis after multivariable adjustment for demographics, health behaviors, chronic medical conditions, medication adherence, abnormal hsCRP, and propensity for statin use, hazard ratio 0.93 (95% confidence interval: 0.81-1.06). Statin use was not similarly associated with rates of sepsis when stratified by propensity for statin use or medication adherence. CONCLUSION: In the REGARDS cohort, statin use at baseline was not associated with rates of future sepsis events.