作者: E. T. Overton , D. Kitch , C. A. Benson , P. W. Hunt , J. H. Stein
DOI: 10.1093/CID/CIT053
关键词: Clinical trial 、 Surgery 、 Cohort 、 Statin 、 Medicine 、 Proportional hazards model 、 Internal medicine 、 Confidence interval 、 Hazard ratio 、 Cohort study 、 Randomized controlled trial
摘要: Background. Excessive inflammation persists despite antiretroviral treatment. Statins decrease cardiovascular (CV) disease risk by reducing low-density lipoprotein cholesterol and inflammation. We performed an exploratory analysis to evaluate whether statin therapy decreased of non-AIDS-defining events nonaccidental death. Methods. A total 3601 subjects not on a from the AIDS Clinical Trials Group Longitudinal Linked Randomized cohort were included. Outcome was time first clinical event (CV event, renal or hepatic disease, incident diabetes, thrombotic/embolic nontraumatic fracture, malignancy, serious bacterial infection, death); categories also analyzed separately. Inverse probability treatment censoring weighted Cox proportional hazard models used assess causal effect. Differential effects baseline covariates evaluated. Results. Over 15 135 person-years (PY) follow-up, 484 initiated statins; 616 experienced (crude rate, 4.4/100 PY 4.1/100 statin); unadjusted ratio (HR) 1.17 (95% confidence interval [CI], .91–1.50). In final model, adjusted HR (AHR) 0.81 CI, .53– 1.24). Results for other similar, except malignancies (AHR, 0.43 [95% .19–.94]) infections 1.30 .64–2.65]). No differential detected. Conclusions. Although associated with reduction in all death, it statistically significant 57% malignancies. Confirmatory studies are needed statin-associated cancer non-AIDS-associated morbidities.