Xanthine oxidase inhibitors and risks of myocardial infarction and diabetes

点击翻译按钮获取中文摘要

 DESCRIPTION (provided by applicant): Gout is the most common inflammatory arthritis caused by hyperuricemia. Over the past few decades, growing evidence supports that gout and hyperuricemia are independently associated with an increased risk of hypertension, cardiovascular diseases including myocardial infarction (MI) and stroke, metabolic syndrome including type 2 diabetes mellitus (T2DM), and mortality. Several studies suggest that the excess mortality in gout patients is attributable to cardiovascular disease. Xanthine oxidase inhibitors (XOI), including allopurinol and febuxostat, lower serum uric acids and are common treatments for gout. Allopurinol has been shown to reduce blood pressure in hypertensive adolescents and improve endothelial function. Furthermore, an animal model showed a beneficial effect of allopurinol on metabolic syndrome. However, to date, it remains largely unknown whether XOIs decrease risks of gout-associated conditions such as MI and T2DM. The primary objective of this proposal is to produce high-quality evidence on the effect of XOI in preventing MI and T2DM in gout patients. The specific aims of this proposal are: 1) to examine the effect of XOI on the risk of incident MI in patients with gout, 2) to investigate the effect of XOI on the risk of incident T2DM in patients with gout and 3) to estimate the effect of XOI on MI and T2DM adjusted for treatment adherence in patients with gout. The proposed study will use data from the United HealthCare insurance claims databases (2004-2013) linked to laboratory test results, and utilize a number of rigorous and innovative approaches for pharmacoepidemiologic research and causal inference to study the effects of XOI on MI and T2DM risks. Given the increasing interest in big data and data linkage in clinical research, it is important to understand the value of using a claims database linked to outpatient lab test results for pharmacoepidemiologic research and to determine the best strategy to incorporate lab data for further confounding control. This proposed study will make an important contribution because it will be the first and largest to investigate the effects of XOI on major medical conditions such as MI and T2DM, adjusting for adherence. Furthermore, the proposed, advanced and innovative epidemiologic methods will extend our understanding of how to deal with confounding by indication, time-varying selection bias, and missing data on laboratory test values in pharmacoepidemiologic research.