Impact of an Alcohol Withdrawal Syndrome Practice Guideline on Surgical Patient Outcomes

Karen M. Stanley, M.S., A.P.R.N., Celene M. Amabile, Pharm.D., Kit N. Simpson, Dr.PH, Deborah Couillard, B.S.N., E. Douglas Norcross, M.D., Cathy L. Worrall, B.S.N., Pharm.D.


Pharmacotherapy. 2003;23(7) 

In This Article

Abstract and Introduction

Study Objective: To standardize treatment of alcohol withdrawal syndrome (AWS) in surgical patients using an AWS practice guideline with a symptom-triggered approach.
Design: Prospective interventional (pilot group) and retrospective (comparison group).
Setting: University teaching hospital.
Patients: Thirty-eight trauma, orthopedic, and general surgery patients identified at risk for AWS in the pilot group, and 34 patients who were managed using nonstandardized approaches.
Interventions: At-risk patients in the pilot group were assessed using the AWS Type Indicator. They received lorazepam, clonidine, or haloperidol, based on AWS Type Indicator assessment and AWS practice guideline criteria.
Measurements and Main Results: A standardized symptom-triggered approach to managing AWS was expected to decrease the use of benzodiazepines, avoid undertreatment of adrenergic hyperactivity and delirium, decrease the need for sitters and physical restraints, and reduce hospital length of stay. Pilot patients received a mean of 23 mg less benzodiazepine (p=0.01), 0.1 mg more clonidine (p=0.01), and 20 mg less haloperidol (p=0.06) than comparison patients. Pilot patients also required significantly fewer sitter hours (p=0.04) and hours of restraint use (p=0.09) than comparison patients. No significant differences were found between groups for length of stay (p=0.77).
Conclusions: This pilot project suggests that trauma, orthopedic, and general surgery patients at risk for AWS can be safely and effectively managed with a standardized, symptom-triggered approach. Moreover, this approach decreased the amounts of benzodiazepines and haloperidol administered to patients at risk for AWS.

Alcoholism affects approximately 15 million adults in the United States, is the fourth leading cause of disability and health care burden worldwide, and is a factor in 50% of motor vehicle accidents, burns, and crimes.[1,2] Reports indicate that up to 25% of hospitalized patients are alcohol dependent.[3,4] The estimated economic cost of alcohol abuse was 184.6 billion dollars in 1998.[5] In parallel with the high prevalence of alcohol abuse, alcohol withdrawal syndrome (AWS) is observed in many hospitalized patients. Symptoms range from mild to life-threatening, and approximately 15% of AWS patients in the hospital experience withdrawal seizures. About 5% of all individuals treated for AWS develop the combination of central nervous system excitation, hyperadrenergia, and delirium known as delirium tremens.[2] Given the prevalence of AWS and the significant morbidity and mortality associated with it, a practice guideline should be considered to standardize treatment.[6,7,8]

In hospitalized patients, AWS is complex, difficult to differentiate from other medical comorbidities, and potentially life-threatening. Delayed recognition and treatment can lead to a complicated and extended hospital course. Individualized symptom-triggered therapy for AWS may prevent patients from becoming oversedated or undertreated.[9,10] Benzodiazepines are often used for therapy of AWS symptoms in the United States,[7] whereas in Europe a variety of agents have been employed, including the anticonvulsants chlormethiazole[11] and carbamazepine,[12] and the antipsychotic tiapride.[13] In studies using benzodiazepines, symptom-triggered therapy using the Clinical Institute Withdrawal Assessment for Alcohol (CIWA-Ar) scale[14] to assess AWS severity resulted in decreased amount of drug used in comparison with standard tapering protocols.[9,10,15] Adjunctive agents such as clonidine and haloperidol are also commonly used when symptoms fail to respond to standard benzodiazepine therapy.[7] However, there is a paucity of data to show how the practice of symptom severity scoring triggers administration of these agents. For example, a sympatholytic agent, such as clonidine, may be given to patients who continue to have hypertension or tachycardia and who fail to respond to benzodiazepine monotherapy. Similarly, patients may require neuroleptic therapy in the presence of delirium and self-injurious behavior.

To address these issues, a symptom-triggered approach for AWS treatment was developed and tested. A novel assessment instrument (Figure 1) was used to categorize AWS symptomatology into three symptom clusters.[1] Treatment decisions were guided by knowledge of the underlying physiology of AWS and the pharmacodynamic actions of three different agents. A benzodiazepine (lorazepam), a sympatholytic (clonidine), and a neuroleptic (haloperidol) were used to treat the specific subsets of AWS symptoms described below.

Alcohol withdrawal syndrome assessment instrument.

Central nervous system excitation (Type A symptoms) occurs within 6-12 hours after the blood alcohol level falls to zero.[1] Alcohol potentiates the inhibitory action of -aminobutyric acid (GABA) in the central nervous system. Chronic exposure to alcohol leads to neuroadaptation, with downregulation of GABA and upregulation of N-methyl-D-aspartate (NMDA) excitatory receptors with enhanced sensitivity to glutamate.[16] Therefore, an unmasked hyperexcitatory state in alcoholics manifests in Type A symptoms when the exogenous inhibitory source is withdrawn. Benzodiazepines are the cornerstone of treatment for this cluster of symptoms.[7] Carbamazepine, valproic acid, gabapentin, and phenobarbital have also been studied in the treatment of central nervous system excitation related to alcohol withdrawal.[17,18]

Adrenergic hyperactivity (Type B symptoms) may occur within 6 hours and up to 48 hours after cessation of alcohol intake. Type B symptoms are related to an increase in central nervous system norepinephrine and increased levels of circulating epinephrine. Hypertension in AWS is caused by increases in and enhanced responsiveness to epinephrine. This symptom cluster is associated with increased morbidity and mortality and may warrant transfer to an intensive care unit (ICU) in severe cases. Clonidine[19,20] and -blockers[21,22,23] have been used successfully in the treatment of specific symptoms related to increased catecholamine output seen in AWS,[24,25] such as tremor and tachycardia.

Alcohol withdrawal delirium (Type C symptoms) may occur within 5 days and up to 2 weeks after cessation of alcohol intake. Proposed mechanisms for this symptom cluster include increased dopamine release and enhanced dopamine receptor activity,[26,27] and NMDA receptor hypersensitivity.[28] Delirium in AWS may be further exacerbated by excessive doses of benzodiazepines for treatment of Type A symptoms. Neuroleptic drugs, such as haloperidol, are moderately effective in controlling this symptom cluster, but Type C symptoms require 2-10 days to resolve regardless of treatment.[1]

Outcome data are needed that relate to use of the AWS Type Indicator and treatment that targets the three specific symptom clusters identified by this instrument. A practice guideline was developed using the AWS Typology method of treatment.[1] It was hypothesized that a standardized symptom-triggered approach to managing AWS would decrease the use of benzodiazepines, avoid undertreatment of adrenergic hyperactivity and delirium, decrease the need for sitters and physical restraints, and reduce hospital length of stay (LOS).