Background: Acute alcohol withdrawal syndrome (AWS) is a common complication in hospitalized patients, but its occurrence and severity is difficult to predict based on clinical indicators alone. Phosphatidylethanol (PEth) is an abnormal phospholipid incorporated into erythrocyte lipid bilayers in proportion to levels of alcohol exposure, with a half-life of 5 to 10 days. We sought to determine if levels of this aberrant phospholipid could serve as a predictor of AWS severity in at-risk inpatients.
Methods: Medically-hospitalized patients (n=58) with alcohol-related conditions were approached for research participation during their first hospital day. Blood samples were collected for measurement of PEth by LC-MS at a contracted laboratory. Following discharge, the total dose of lorazepam equivalents received during hospitalization was determined from pharmacy records and dichotomized at the median of 20 mg. The distribution of PEth was compared between high and low benzodiazepine groups and the PEth-associated risk for receiving relatively higher doses of benzodiazepines was estimated.
Results: Patients’ ages ranged from 24 to 67 years old (median age 50), and were predominantly male (74%) and non-Hispanic white (82%). The distribution of PEth differed significantly between those receiving relatively lower benzodiazepine dosing compared to those receiving higher benzodiazepine dosing (p= 0.015). Those that were found to have high PEth levels at admission (greater than the median value of 646 ng/ml) had a 4.2 (95% confidence interval 1.40 to 12.66) fold increased odds of requiring higher levels of benzodiazepines.
Conclusions: Results suggest that PEth may be useful in risk-stratifying medical inpatients for AWS. Future work will be required to optimize its detection at the point-of-care, for determining the optimal thresholds for predicting the occurrence of AWS and its severity, and estimating its cost-effectiveness relative to risk stratification using clinical indicators alone.