Predicting Alcohol Withdrawal using DNA Methylation

点击翻译按钮获取中文摘要

The placement of intoxicated patients from the emergency room into inpatient hospital settings to monitor and/treat possible alcohol withdrawal syndrome (AWS) is common occurrence for most urban hospitals. Although the circumstances that lead to this outcome vary, a contributor to many of these hospitalizations is the lack of reliable clinical information to predict whether a patient is likely to suffer medically severe AWS. Because of this, clinicians are often forced to hospitalize patients, often against their will, unnecessarily. A method predicting who will experience AWS could address this predicament, improve outcomes, avoid unnecessary alienation of patients and decrease healthcare costs. Newly developed epigenetic techniques may be able to predict the likelihood of AWS. Over the past 5 years using genome wide approaches, we and other have shown that heavy alcohol consumption is associated with profound changes in DNA methylation status. Furthermore, we have recently refined the signatures obtained using these expensive time-consuming methylation arrays to an easy to perform, potentially clinically employable digital PCR panel that is highly sensitive and specific for heavy alcohol consumption. These panels are now being used commercially for insurance underwriting. However, whether this DNA methylation panel or any other DNA methylation panel could also be useful for determining likelihood of AWS, alone or together with composite self-report/biomarker tools such as the Prediction of Alcohol Withdrawal Scale (PAWSS) is unknown. In this high risk R21 application, we will test whether DNA methylation can aid current schemes for predicting alcohol withdrawal. Specifically, we will solicit 150 subjects admitted to the University of Iowa for alcohol detoxification. We will then characterize each of these subjects with a battery of tools including the PAWSS, phlebotomize them to provide biomaterial for the methylation studies, then follow each of these subjects to determine which of them went onto develop AWS. Finally, we will determine DNA methylation status in each of these subjects. We hypothesize that the DNA methylation will predict AWS and that the PAWS and DNA methylation predict AWS better than either measure alone. It is innovative because generally accepted biomarkers for assessing risk for alcohol withdrawal do not exist and the use of DNA methylation for these purposes has not been tested. The team is well prepared to conduct the research and includes board-certified clinicians, statisticians and a leading expert on DNA methylation. The institution at which the examination will be based admits thousands of intoxicated patients annually. As a direct result of this research we will establish the feasibility of DNA methylation to predict AWS and gather the data to design a well powered R01 investigation that specifically examines this new approach as compared to existing measures.