Alcohol Withdrawal Syndrome
Alcohol withdrawal syndrome (AWS) consists of a spectrum of clinical manifestations that vary in severity and duration upon cessation of alcohol intake in the alcohol-dependent patient. Presentations to the MICU are part of a clinical continuum, but the sequence of events may be inconsistent, and is dependent on the degree of alcohol abuse. There are four stages of alcohol withdrawal. The first stage is autonomic hyperactivity, in which clinical symptoms appear within hours of the last drink and peak within 24 to 48 hours. There is usually a clear sensorium, but this is often accompanied by tremulousness, sweating, anxiety or agitation, insomnia, and nausea and vomiting. This stage is secondary to sympathetic outflow, documented by increased circulating catecholamine levels in the urine, serum, and the cerebrospinal fluid.[7,8]
Most symptoms resolve within 24 to 36 hours, but approximately 25% of patients progress to a more severe stage. The second stage is hallucinations, which are usually visual and occur in one quarter of all patients presenting with alcohol withdrawal. They occur within 8 to 48 hours after a decrease in alcohol intake and may last from 1 to 6 days.[10,11] The third stage is neuronal excitation, which is accompanied by seizure activity that occurs in up to 10% of patients and within 12 to 48 hours of abstinence or decreased alcohol intake. The seizures are usually single, short, and generalized tonic-clonic. Importantly, status epilepsy lasting more than 6 hours is suggestive of another underlying neurologic disorder such as idiopathic epilepsy, trauma, or other toxic-metabolic causes. The fourth stage is delirium tremens. This occurs in up to 5% of patients and usually begins at 48 to 72 hours but may be delayed up to 4 to 5 days.[14,15] Approximately 30% of alcoholics with seizures progress to this stage. Delirium tremens is characterized by disorientation, confusion, impaired attention, pronounced autonomic hyperactivity, and visual and auditory hallucinations. If left untreated, mortality rates may reach 15%;[14,16] death is usually due to cardiovascular or respiratory collapse. In an effort to objectively assess the severity of AWS, a scale called the Clinical Institute Withdrawal Assessment for Alcohol Scale was developed. This scale is a 10-step assessment (scored from 0 to 67 points) of signs and symptoms of AWS. The initial score can be administered upon admission to the hospital and repeated hourly thereafter for signs of progression. Patients with a score of greater than 20 should be transferred to the MICU immediately, with the goal of reducing their score to less than 10 within the first 24 hours.
The pathophysiology of AWS is not well known. There are various interactions between neurotransmitters and neuromodulators that contribute to this condition. One of the most widely accepted theories is the upregulation of cyclic adenosine 3', 5'-monophosphate that occurs in certain regions of the brain, including the locus ceruleus, nucleus accumbens, and the ventral tegmental area. This receptor upregulation represents a physiologic dependence. With removal of alcohol in acute withdrawal, the cyclic adenosine 3', 5'-monophosphate pathway overshoots, and the clinical signs of withdrawal become evident. A kindling phenomenon may occur whereby patients may become sensitized to this upregulation and subsequent withdrawals become progressively more severe.
The diagnosis of AWS is initially made by a history of heavy drinking. The well-established CAGE (cut down, annoyed, guilty, eye-opener) questionnaire has been shown to correlate with the Diagnostic and Statistical Manual criteria for alcohol dependence[21,22] but has limited utility in the MICU, since many patients are poorly responsive. In addition, the blood-alcohol concentration is often negative in chronic alcoholics. Conventional markers for the detection of chronic alcohol abuse such as the mean corpuscular volume, the γ-glutamyl-transferase, the aspartate aminotransferase, and the alanine transferase are reported to be neither sensitive nor specific. In more recent studies, an elevation in carbohydrate-deficient transferrin level was shown to be a more sensitive and specific biologic marker for detecting alcohol abuse.
Treatment of AWS is challenging and varies among clinicians. Goals of therapy are to alleviate the symptoms, prevent further progression, treat underlying comorbidities, and plan for long-term rehabilitation. Treatment should be started promptly, with intensive care monitoring, intravenous access, hydration, and parental thiamine with dextrose bolus instituted immediately. If the patient has an altered mental status, supplemental oxygen and airway protection is needed. Endotracheal intubation may be necessary in the presence of seizures or risk of pulmonary aspiration. Benzodiazepines are the mainstay of therapy for AWS. They enhance the activity of endogenous neurotransmitter γ-aminobutyric acid, which essentially is a replacement for alcohol. Benzodiazepines alleviate the excitatory manifestations of AWS and reduce the risk of seizures and delirium. The dose required is variable, and frequent adjustment is based on withdrawal severity. With adequate benzodiazepine sedation treatment, the patient should rest comfortably but be easily awakened. Agents such as diazepam 10 to 40 mg, lorazepam1 to 8 mg, or chlordiazepoxide 50 to 100 mg may be used as long as reassessment and titration is performed on a regular basis. None of these drugs have been shown to be superior to others in this setting. Some investigators, however, report a decrease in seizure activity and a smoother withdrawal course with longer-acting benzodiazepines such as diazepam. A fixed schedule regimen may be used and gradually tapered off, once symptoms are suppressed. In addition, beta-blockade may be used adjunctively for excessive autonomic activity and has been shown to decrease the manifestations of alcohol withdrawal. These agents should be used with caution, as they may mask some of the withdrawal symptoms in cases of inadequate treatment, such as tachycardia and hypertension. Beta-blockade has also been reported to precipitate delirium. Central alpha agonists such as clonidine are used to attenuate central sympathetic outflow and reduce plasma catecholamine levels. This class of medication decreases tachycardia, tremor, and diaphoresis but has no effect on delirium or seizures. Thus, they should only be used in conjunction with sedatives. Neuroleptics, such as haloperidol, have been shown to increase mortality rates if used as monotherapy and therefore are used only in synergistic combination with benzodiazapines. Carbamazepine is used in Europe and has been shown to decrease withdrawal symptoms, especially with seizure, but does not prevent delirium. It may also retard the kindling phenomenon described above. Ethyl alcohol, either enteral or through central venous access, may be used, but no controlled trials have compared the safety or efficacy with that of benzodiazepines. Finally, propofol is an aromatic sedative-hypnotic used for severe cases of AWS. It acts on a subunit of the γ-aminobutyric acid receptor-ionophore complex that increases chloride conductance and inhibits the N -methyl-D-aspartate subtype of the glutamate receptor. Its major utility is for refractory agitation seen with benzodiazepine failure. Its rapid penetration into the central nervous system and rapid elimination kinetics make it an ideal drug for short-term therapy. Significant side effects associated with propofol include hypotension, hypertriglyceridemia, and a risk of infection caused by the infusate medium.
Wernicke encephalopathy (WE) is an important cause of mental status change in chronic alcoholics and carries a 10 to 20% mortality rate if left untreated. WE is an important syndrome to recognize because it is fatal but easily reversed. This syndrome is characterized by nystagmus, gaze palsies, gait ataxia, and mental confusion. Korsakoff amnestic state is a unique mental disorder in which retentive memory is impaired in greater proportion compared with other cognitive dysfunction and is closely associated with WE. These two disorders may present together in critically ill alcoholic patients. WE occurs as a result of thiamine deficiency, which is an important cofactor for several enzymatic systems, such as transketolase and pyruvate dehydrogenase,[41,42] that are responsible for cerebral glucose utilization and glutamate elimination. Therefore, thiamine deficiency results in an alternation of cerebral metabolism leading to a diminished nerve impulse transmission at various synapses.
Only one third of patients present with the classic triad described by Wernicke: confusion, ataxia, and ophthalmoplagia. Most of the patients, on the other hand, present with a change in mental status or some degree of Korsakoff psychosis. Other manifestations include meiotic or nonreactive pupils, postural hypotension with accompanying syncope caused by impaired autonomic dysfunction, and hypothermia caused by loss of central thermoregulation. The diagnosis of WE is made by a high index of suspicion. In the appropriate clinical settings, low levels of serum thiamine, pyruvate dehydrogenase, and transketolase activity might confirm the diagnosis. Magnetic resonance imaging is used to support the diagnosis by demonstrating high-intensity lesions in the paraventricular regions of the thalamus, hypothalamus, mamillary bodies, periaqueductal region of the midbrain, fourth ventricular floor, and superior cerebellar vermis.
Treatment of WE in the MICU is a medical emergency. Patients should receive immediate parental thiamine as a 100-mg bolus. However, only 2 to 3 mg is needed to reverse the ocular symptoms. Thereafter, daily intravenous infusion of 100 mg of thiamine is given until the patient resumes a normal diet. Improvement of the ataxia and confusion is usually seen within 1 to 6 hours of treatment.
Portal systemic encephalopathy or hepatic encephalopathy is a potentially reversible decrease in neurologic function seen in critically ill chronic alcoholics. The syndrome is characterized by irritability, impaired attention, poor memory, and even somnolence that may progress to stupor and coma. Precipitating factors unique to alcoholics include gastrointestinal hemorrhage, hypokalemia, metabolic alkalosis, and excessive diuresis. It is also imperative to rule out sources of infection, including meningitis, pneumonia, urinary tract infection, and spontaneous bacterial peritonitis. Furthermore, the use of benzodiazepines in the treatment in AWS may precipitate hepatic encephalopathy in patients with underlying alcoholic liver disease.
The onset is insidious and is marked by subtle changes in memory and concentration. The syndrome is divided into four stages. The first stage involves higher cortical function and may be manifested by a decrease in attention span, depression, tremor, personality changes, incoordination, apraxia, and irritability. The second stage is marked by exaggeration of the first stage and includes poor memory and computation, disordered sleep, slowed speech, ataxia, and even drowsiness. The third and fourth stages include increasing obtundation, amnesia, nystagmus, clonus, and muscular rigidity and may progress to coma, decerebrate posturing, dilated pupils, and poor response to painful stimuli.
The diagnosis of portal systemic encephalopathy is made on a clinical basis but can be supported by laboratory data, electroencephalogram (EEG), and neuropsychiatric testing. The most reliable assessment is the Portal Systemic Encephalopathy Index, which contains five parameters, including (1) a clinical assessment of mental status, (2) trail-making time, (3) EEG, (4) the presence of asterixis on physical examination, and (5) an arterial ammonia level. The trail-making test is a semiquantitative measure whereby the subject connects 25 consecutive numbered circles, and the time (in seconds) is noted. Figures may be used if the patient is not able to recognize numbers. The EEG is sensitive and reliable for all stages of portal systemic encephalopathy and will display subtle slowing in stage 1, slow rhythms and triphasic waves at the frontal regions in stages 2 and 3, and severe slowing with theta and delta waves in stage 4. An elevation of the arterial ammonia level does not confirm or exclude the diagnosis of portal systemic encephalopathy, but increased levels parallel changes in the EEG and can be helpful in guiding the response to treatment in the MICU.
The goals of portal systemic encephalopathy treatment in the MICU include identification and correction of precipitating causes, initiation of ammonia-lowering therapy, and minimizing potential medical complications of cirrhosis and decreased consciousness. Initial management includes elimination of sedatives and tranquilizers that may contribute to the mental status change, adequate volume repletion, correction of metabolic derangements, treating infection, and halting active bleeding. After initial management, it is imperative to lower elevated serum ammonia levels. A dietary protein restriction of 60 g/d of vegetable protein will effectively decrease ammonia levels. Lactulose is given by nasogastric tube (30 to 60 mL every 2 hours or until a bowel movement) or by enema if the patient has an ileus (300 mL of 50% lactulose + 700 mL water) to increase bowel transit time and to decrease the stool pH. Lactulose is a nonabsorbable synthetic disaccharide that is metabolized by colonic bacteria to lactic, acetic, and other organic acids, which stimulate peristalsis and trap ammonia in the gut as an ammonium ion.[55,56] The remainder of the compound acts as an osmotic laxative. Another ammonia-lowering therapy includes antibiotics such as neomycin, rifaximin, ampicillin, and metronidazole.[57,58] Keeping the head of the bed elevated to decrease risk of aspiration, treating agitation, gastrointestinal bleeding prophylaxis, and minimizing medications metabolized by the liver are part of the management of portal systemic encephalopathy in the MICU. Corticosteroids have been suggested in the management of severe alcoholic liver disease; however, this should only be considered after excluding acute infection and confirming the diagnosis by a liver biopsy.
South Med J. 2005;98(3):372-381. © 2005 Lippincott Williams & Wilkins
Cite this: Critical Care Aspects of Alcohol Abuse - Medscape - Mar 01, 2005.