Prediction of the Development of Alcohol Withdrawal Delirium

Lee, Choong Heon; Lee, Eun; Park, Sung Hyouk; Jung, Seung Yong; Park, Sang Jin; Soek, Jeong Ho; Namkoong, Kee

Abstract

OBJECTIVES
Alcohol withdrawal delirium is a serious clinical condition with high mortality rate if not treated. This study was to examine whether readily available clinical variables can predict the development of alcohol withdrawal delirium.
METHODS
The authors performed a retrospective study by reviewing charts of 566 patients who had been admitted for alcohol dependence. The cases were divided into two groups: delirium group (n=40) and control group (n=40). We compared baseline characteristics and serum analysis data at admission between two groups. We used logistic regression to predict risk factors for alcohol withdrawal delirium among potential risk factors.
RESULTS
The delirium group had significantly lower hemoglobin, hematocrit, platelet counts, and potassium level than countrol group. Presence of previous withdrawal delirium history, AST, GGTP, and bilirubin level of delirium group were significantly higher than those of control group. Among potential risk factors, past history of withdrawal delirium, decreased hemoglobin, elevated bilirubin level, and number of previous detoxification were predictable factors of the development of alcohol withdrawal delirium by 72.5%.
CONCLUSION
Our results suggest that the infomation obtained at admission can be useful to predict the development of alcohol withdrawal delirium. Also, it makes the individualization of detoxification strategies possible.