Harrison's Neurology in Clinical Medicine, 3rd Edition


Marc A. Schuckit


Alcohol (beverage ethanol) distributes throughout the body, affecting almost all systems and altering nearly every neurochemical process in the brain. This drug is likely to exacerbate most medical conditions, affect almost any medication metabolized in the liver, and temporarily mimic many medical (e.g., diabetes) and psychiatric (e.g., depression) conditions. Because ~80% of people in Western countries have consumed alcohol, and two-thirds have been drunk in the prior year, the lifetime risk for serious, repetitive alcohol problems is almost 20% for men and 10% for women, regardless of a person’s education or income. While low doses of alcohol have some healthful benefits, the intake of more than three standard drinks per day on a regular basis enhances the risk for cancer and vascular disease, and alcohol use disorders decrease the life span by about 10 years. Unfortunately, most clinicians have had only limited education regarding these conditions. This chapter presents a brief overview of clinically useful information about alcohol use, abuse, and dependence.


Blood levels of ethanol are expressed as milligrams or grams of ethanol per deciliter (e.g., 100 mg/dL = 0.10 g/dL), with values of ~0.02 g/dL resulting from the ingestion of one typical drink. In round figures, 340 mL (12 oz) of beer, 115 mL (4 oz) of nonfortified wine, and 43 mL (1.5 oz) (a shot) of 80-proof beverage such as whisky, gin, or vodka each contain ~10–15 g of ethanol and represent a standard drink; 0.5 L (1 pint) of 80-proof beverage contains ~160 g (about 16 standard drinks), and 750 mL of wine contains ~60 g of ethanol. These beverages also have additional components known as congeners that affect the drink’s taste and might contribute to adverse effects on the body. Congeners include methanol, butanol, acetaldehyde, hista-mine, tannins, iron, and lead. Alcohol acutely decreases neuronal activity and has similar behavioral effects and cross-tolerance with other depressants, including benzodiazepines and barbiturates.

Alcohol is absorbed from mucous membranes of the mouth and esophagus (in small amounts), from the stomach and large bowel (in modest amounts), and from the proximal portion of the small intestine (the major site). The rate of absorption is increased by rapid gastric emptying (as can be induced by carbonated beverages); by the absence of proteins, fats, or carbohydrates (which interfere with absorption); and by dilution to a modest percentage of ethanol (maximum at ~20% by volume).

Between 2% (at low blood alcohol concentrations) and 10% (at high blood alcohol concentrations) of ethanol is excreted directly through the lungs, urine, or sweat, but the greater part is metabolized to acetaldehyde, primarily in the liver. The most important pathway occurs in the cell cytosol where alcohol dehydrogenase (ADH) produces acetaldehyde, which is then rapidly destroyed by aldehyde dehydrogenase (ALDH) in the cytosol and mitochondria (Fig. 56-1). A second pathway in the microsomes of the smooth endoplasmic reticulum (the microsomal ethanol-oxidizing system, or MEOS) is responsible for ≥10% of ethanol oxidation at high blood alcohol concentrations.



The metabolism of alcohol. MEOS, microsomal ethanol-oxidizing system.

While alcohol supplies calories (a drink contains ~300 kJ, or 70–100 kcal), these are devoid of nutrients such as minerals, proteins, and vitamins. In addition, alcohol can also interfere with absorption of vitamins in the small intestine and decreases their storage in the liver with modest effects on folate (folacin or folic acid), pyridoxine (B6), thiamine (B1), nicotinic acid (niacin, B3), and vitamin A.

A heavy ethanol load in a fasting, healthy individual is likely to produce transient hypoglycemia within 6–36 h, secondary to the acute actions of ethanol on gluconeo-genesis. This can result in temporary abnormal glucose tolerance tests (with a resulting erroneous diagnosis of diabetes mellitus) until the alcoholic has abstained for 2–4 weeks. Alcohol ketoacidosis, probably reflecting a decrease in fatty acid oxidation coupled with poor diet or recurrent vomiting, can be misdiagnosed as diabetic ketosis. With the former, patients show an increase in serum ketones along with a mild increase in glucose but a large anion gap, a mild to moderate increase in serum lactate, and a β-hydroxybutyrate/lactate ratio of between 2:1 and 9:1 (with normal being 1:1).

In the brain, alcohol affects almost all neurotransmitter systems, with acute actions that are often the opposite of those seen following desistance after a period of heavy drinking. The most prominent actions relate to boosting gamma aminobutyric acid (GABA) activity, especially in GABAA receptors. Enhancement of this complex chloride channel system contributes to anticonvulsant, sleep-inducing, antianxiety, and muscle relaxation effects of all GABA-boosting drugs. Acutely administered alcohol produces a release of GABA, and continued use of this drug increases density of GABAA receptors, while alcohol withdrawal states are characterized by decreases in GABA-related activity. Equally important is the ability of acute alcohol to inhibit postsynaptic N-methyl-D-aspartate (NMDA) excitatory glutamate receptors, while chronic drinking and desistance are associated with an upregulation of these excitatory receptor subunits. The relationships between greater GABA and diminished NMDA receptor activity during acute intoxication and diminished GABA with enhanced NMDA actions during alcohol withdrawal explain much of intoxication and withdrawal phenomena.

As with all pleasurable activities, drinking alcohol acutely increases dopamine levels in the brain, especially in the ventral tegmentum and related brain regions, and this effect plays an important role in continued alcohol use, craving, and relapse. The changes in dopamine pathways are also linked to increases in “stress hormones,” including cortisol and adrenocorticotropic hormone (ACTH) during intoxication and decreases in these hormones during withdrawal. Such alterations are likely to contribute to both feelings of reward during intoxication and depression during falling blood alcohol concentrations. Also closely linked to alterations in dopamine (especially in the nucleus accumbens) are alcohol-induced changes in opioid receptors, with acute alcohol also causing release of beta endorphins.

Additional important neurochemical changes include increases in synaptic levels of serotonin during acute intoxication, and subsequent upregulation of serotonin receptors. Acute increases in nicotinic acetylcholine systems also contribute to the impact of alcohol in the ventral tegmental region, which occur in concert with enhanced dopamine activity. In the same regions, alcohol impacts on cannabinol receptors, with resulting release of dopamine, GABA, and glutamate as well as subsequent effects on brain reward circuits.


The acute effects of a drug depend on the dose, the rate of increase in plasma, the concomitant presence of other drugs, and the past experience with the agent. “Legal intoxication” with alcohol in most states requires a blood alcohol concentration of 0.08 g/dL, while levels of 0.04 or even lower are cited in other countries. However, behavioral, psychomotor, and cognitive changes are seen at levels as low as 0.02–0.03 g/dL (i.e., after one to two drinks) (Table 56-1). Deep but disturbed sleep can be seen at twice the legal intoxication level, and death can occur with levels between 0.30 and 0.40 g/dL. Beverage alcohol is probably responsible for more overdose deaths than any other drug.

TABLE 56-1



Repeated use of alcohol contributes to acquired tolerance, a complex phenomenon involving at least three types of compensatory mechanisms. (1) After 1–2 weeks of daily drinking, metabolic or pharmacokinetic tolerance can be seen, with up to 30% increase in the rate of hepatic ethanol metabolism. This alteration disappears almost as rapidly as it develops. (2) Cellular or pharmacodynamic tolerance develops through neurochemical changes that maintain relatively normal physiologic functioning despite the presence of alcohol. Subsequent decreases in blood levels contribute to symptoms of withdrawal. (3) Individuals learn to adapt their behavior so that they can function better than expected under influence of the drug (learned or behavioral tolerance).

The cellular changes caused by chronic ethanol exposure may not resolve for several weeks or longer following cessation of drinking. Rapid decreases in blood alcohol levels before that time can result in a withdrawal syndrome, which is most intense during the first 5 days, but some symptoms (e.g., disturbed sleep and anxiety) can take up to 4–6 months to resolve.


Relatively low doses of alcohol (one or two drinks per day) have potential beneficial effects of increasing high-density lipoprotein cholesterol and decreasing aggregation of platelets, with a resulting decrease in risk for occlusive coronary disease and embolic strokes. Red wine has additional potential health-promoting qualities at relatively low doses due to flavinols and related substances, which may work by inhibiting platelet activation. Modest drinking might also decrease the risk for vascular dementia and, possibly, Alzheimer’s disease. However, any potential healthful effects disappear with the regular consumption of three or more drinks per day, and knowledge about the deleterious effects of alcohol can both help the physician to identify patients with alcohol abuse and dependence, and to supply them with information that might help motivate a change in behavior.


Approximately 35% of drinkers (and a much higher proportion of alcoholics) experience a blackout, an episode of temporary anterograde amnesia, in which the person forgets all or part of what occurred during a drinking evening. Another common problem, one seen after as few as one or two drinks shortly before bedtime, is disturbed sleep. Although alcohol might initially help a person to fall asleep, it disrupts sleep throughout the rest of the night. The stages of sleep are also altered, and time spent in rapid eye movement (REM) and deep sleep is reduced. Alcohol relaxes muscles in the pharynx, which can cause snoring and exacerbate sleep apnea; symptoms of the latter occur in 75% of alcoholic men older than age 60 years. Patients may also experience prominent and sometimes disturbing dreams. All of these sleep problems are more pronounced in alcoholics, and their persistence may contribute to relapse.

Another common consequence of alcohol use is impaired judgment and coordination, increasing the risk of accidents and injury. In the United States, ~40% of drinkers have at some time driven while intoxicated. Heavy drinking can also be associated with headache, thirst, nausea, vomiting, and fatigue the following day, a hangover syndrome that is responsible for much missed time and temporary cognitive deficits at work and school.

The effect of alcohol on the nervous system is even more pronounced among alcohol-dependent individuals. Chronic high doses cause peripheral neuropathy in ~10% of alcoholics: similar to diabetes, patients experience bilateral limb numbness, tingling, and paresthesias, all of which are more pronounced distally. Approximately 1% of alcoholics develop cerebellar degeneration or atrophy. This is a syndrome of progressive unsteady stance and gait often accompanied by mild nystagmus; neuroimaging studies reveal atrophy of the cerebellar vermis. Fortunately, very few alcoholics (perhaps as few as 1 in 500 for the full syndrome) develop Wernicke’s (ophthalmoparesis, ataxia, and encephalopathy) and Korsakoff’s (retrograde and anterograde amnesia) syndromes, although a higher proportion have one or more neuropathologic findings related to these syndromes. These occur as the result of low levels of thiamine, especially in predisposed individuals, e.g., those with transketolase deficiency. Alcoholics can manifest cognitive problems and temporary memory impairment lasting for weeks to months after drinking very heavily for days or weeks. Brain atrophy, evident as ventricular enlargement and widened cortical sulci on MRI and CT scans, occurs in ~50% of chronic alcoholics; these changes are usually reversible if abstinence is maintained. There is no single alcoholic dementia syndrome; rather, this label is used to describe patients who have apparently irreversible cognitive changes (possibly from diverse causes) in the context of chronic alcoholism.

Psychiatric comorbidity

As many as two-thirds of alcohol-dependent individuals meet the criteria for a psychiatric syndrome in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) of the American Psychiatric Association (Chap. 54). Half of these relate to a preexisting antisocial personality manifesting as impulsivity and disinhibition that contribute to both alcohol and drug dependence. The lifetime risk is 3% in males, and ≥80% of such individuals demonstrate alcohol and/or drug dependence. Another common comorbidity occurs with dependence on illicit substances. The remainder of alcoholics with psychiatric syndromes have preexisting conditions such as schizophrenia or manic-depressive disease and anxiety disorders such as panic disorder. The comorbidities of alcoholism with independent psychiatric disorders might represent an overlap in genetic vulnerabilities, impaired judgment in the use of alcohol from the independent psychiatric condition, or an attempt to use alcohol to alleviate some of the symptoms of the disorder or side effects of medications.

Many psychiatric syndromes can be seen temporarily during heavy drinking and subsequent withdrawal. These alcohol-induced conditions include an intense sadness lasting for days to weeks in the midst of heavy drinking seen in 40% of alcoholics, which tends to disappear over several weeks of abstinence (alcohol-induced mood disorder); temporary severe anxiety in 10–30% of alcoholics, often beginning during alcohol withdrawal, which can persist for a month or more after cessation of drinking (alcohol-induced anxiety disorder); and auditory hallucinations and/or paranoid delusions in a person who is alert and oriented, seen in 3–5% of alcoholics (alcohol-induced psychotic disorder).

Treatment of all forms of alcohol-induced psycho-pathology includes helping patients achieve abstinence and offering supportive care, as well as reassurance and “talk therapy” such as cognitive-behavioral approaches. However, with the exception of short-term antipsychotics or similar drugs for substance-induced psychoses, substance-induced psychiatric conditions only rarely require medications. Recovery is likely within several days to 4 weeks of abstinence. Conversely, because alcohol-induced conditions are temporary and do not indicate a need for long-term pharmacotherapy, a history of alcohol intake is an important part of the workup for any patient with one of these psychiatric symptoms.


Esophagus and stomach

Alcohol intake can result in inflammation of the esophagus and stomach causing epigastric distress and gastrointestinal bleeding, making alcohol one of the most common causes of hemorrhagic gastritis. Violent vomiting can produce severe bleeding through a Mallory-Weiss lesion, a longitudinal tear in the mucosa at the gastroesophageal junction.

Pancreas and liver

The incidence of acute pancreatitis (~25 per 1000 per year) is almost threefold higher in alcoholics than in the general population, accounting for an estimated 10% or more of the total cases. Alcohol impairs gluconeogenesis in the liver, resulting in a fall in the amount of glucose produced from glycogen, increased lactate production, and decreased oxidation of fatty acids. This contributes to an increase in fat accumulation in liver cells. In healthy individuals these changes are reversible, but with repeated exposure to ethanol, especially daily heavy drinking, more severe changes in the liver occur, including alcohol-induced hepatitis, perivenular sclerosis, and cirrhosis, with the latter observed in an estimated 15% of alcoholics. Perhaps through an enhanced vulnerability to infections, alcoholics have an elevated rate of hepatitis C, and drinking in the context of that disease is associated with more severe liver deterioration.


As few as 1.5 drinks per day increases a woman’s risk of breast cancer 1.4-fold. For both genders, four drinks per day increases the risk for oral and esophageal cancers approximately threefold and rectal cancers by a factor of 1.5; seven to eight or more drinks per day enhances approximately fivefold the risks for many cancers. These consequences may result directly from cancer-promoting effects of alcohol and acetaldehyde or indirectly by interfering with immune homeostasis.


Ethanol causes an increase in red blood cell size (mean corpuscular volume [MCV]), which reflects its effects on stem cells. If heavy drinking is accompanied by folic acid deficiency, there can also be hypersegmented neutrophils, reticulocytopenia, and a hyperplastic bone marrow; if malnutrition is present, sideroblastic changes can be observed. Chronic heavy drinking can decrease production of white blood cells, decrease granulocyte mobility and adherence, and impair delayed-hyper-sensitivity responses to novel antigens (with a possible false-negative tuberculin skin test). Associated immune deficiencies can contribute to vulnerability toward infections, including hepatitis and HIV, and interfere with their treatment. Finally, many alcoholics have mild thrombocytopenia, which usually resolves within a week of abstinence unless there is hepatic cirrhosis or congestive splenomegaly.


Acutely, ethanol decreases myocardial contractility and causes peripheral vasodilation, with a resulting mild decrease in blood pressure and a compensatory increase in cardiac output. Exercise-induced increases in cardiac oxygen consumption are higher after alcohol intake. These acute effects have little clinical significance for the average healthy drinker but can be problematic when persisting cardiac disease is present.

The consumption of three or more drinks per day results in a dose-dependent increase in blood pressure, which returns to normal within weeks of abstinence. Thus, heavy drinking is an important factor in mild to moderate hypertension. Chronic heavy drinkers also have a sixfold increased risk for coronary artery disease, related, in part, to increased low-density lipoprotein cholesterol, and carry an increased risk for cardiomyopathy through direct effects of alcohol on heart muscle. Symptoms of the latter include unexplained arrhythmias in the presence of left ventricular impairment, heart failure, hypocontractility of heart muscle, and dilation of all four heart chambers with associated mural thrombi and mitral valve regurgitation. Atrial or ventricular arrhythmias, especially paroxysmal tachycardia, can also occur temporarily after heavy drinking in individuals showing no other evidence of heart disease—a syndrome known as the “holiday heart.”


Drinking in adolescence can affect normal sexual development and reproductive onset. At any age, modest ethanol doses (e.g., blood alcohol concentrations of 0.06 g/dL) can increase sexual drive but also decrease erectile capacity in men. Even in the absence of liver impairment, a significant minority of chronic alcoholic men show irreversible testicular atrophy with shrinkage of the seminiferous tubules, decreases in ejaculate volume, and a lower sperm count.

The repeated ingestion of high doses of ethanol by women can result in amenorrhea, a decrease in ovarian size, absence of corpora lutea with associated infertility, and an increased risk of spontaneous abortion. Heavy drinking during pregnancy results in the rapid placental transfer of both ethanol and acetaldehyde, which may have serious consequences for fetal development. One severe result is the fetal alcohol syndrome (FAS), seen in ~5% of children born to heavy-drinking mothers, which can include any of the following: facial changes with epicanthal eye folds; poorly formed ear concha; small teeth with faulty enamel; cardiac atrial or ventricular septal defects; an aberrant palmar crease and limitation in joint movement; and microcephaly with mental retardation. A less severe condition is the fetal alcohol spectrum disorder (FASD), which can include low birth weight, a lower IQ, hyperactive behavior, and some modest cognitive deficits. The amount of ethanol required and the time of vulnerability during pregnancy have not been defined, making it advisable for pregnant women to abstain completely.


Between one-half and two-thirds of alcoholics have skeletal muscle weakness caused by acute alcoholic myopathy, a condition that improves but which might not fully remit with abstinence. Effects of repeated heavy drinking on the skeletal system include changes in calcium metabolism, lower bone density, and decreased growth in the epiphyses, leading to an increased risk for fractures and osteonecrosis of the femoral head. Hormonal changes include an increase in cortisol levels, which can remain elevated during heavy drinking; inhibition of vasopressin secretion at rising blood alcohol concentrations and enhanced secretion at falling blood alcohol concentrations (with the final result that most alcoholics are likely to be slightly overhydrated); a modest and reversible decrease in serum thyroxine (T4); and a more marked decrease in serum triiodothyronine (T3). Hormone irregularities should be reevaluated as they may disappear after a month of abstinence.


Because many drinkers occasionally imbibe to excess, temporary alcohol-related pathology is common in nonalcoholics, especially in the late teens to the late twenties. When repeated problems in multiple life areas develop, the individual is likely to meet criteria for alcohol abuse or dependence.


Alcohol dependence is defined in DSM-IV as repeated alcohol-related difficulties in at least three of seven life areas that cluster together at about the same time (e.g., over the same 12-month period). Two of these seven items, tolerance and withdrawal, may have special importance as they are associated with a more severe clinical course. Dependence predicts a course of recurrent problems with the use of alcohol and the consequent shortening of the life span by a decade.

Alcohol abuse is defined as repetitive problems with alcohol in any one of four life areas—social, interpersonal, legal, and occupational—or repeated use in hazardous situations such as driving while intoxicated in an individual who is not alcohol dependent. About 50% of those with alcohol abuse continue to have alcohol problems 2–5 years later, but only ~10% of these patients—including adolescents—go on to develop alcohol dependence.

The lifetime risk for alcohol dependence in most Western countries is about 10–15% for men and 5–8% for women. Rates are generally similar in the United States, Canada, Germany, Australia, and the United Kingdom; tend to be lower in most Mediterranean countries, such as Italy, Greece, and Israel; and may be higher in Ireland, France, and Scandinavia. An even higher lifetime prevalence has been reported for most native cultures, including American Indians, Eskimos, Maori groups, and aboriginal tribes of Australia. These differences reflect both cultural and genetic influences, as described later. In Western countries, the typical alcoholic is more often a blue- or white-collar worker or homemaker. The lifetime risk for alcoholism among physicians is similar to that of the general population.


Approximately 60% of the risk for alcohol use disorders is attributed to genes, as indicated by the fourfold higher risk for alcohol abuse and dependence in children of alcoholics (even if these children were adopted early in life and raised by nonalcoholics) and a higher risk in identical twins as compared to fraternal twins of alcoholics. The genetic variations appear to operate primarily through intermediate characteristics that subsequently relate to the environment in altering the risk for heavy drinking and alcohol problems. These include genes relating to a high risk for all substance use disorders that operate through impulsivity, schizophrenia, and bipolar disorder. Another characteristic, an intense flushing response when drinking, decreases the risk for only alcohol use disorders through gene variations for several alcohol-metabolizing enzymes, especially aldehyde dehydrogenase (a mutation only seen in Asians), and to a lesser extent, variations in alcohol dehydrogenase.

An additional genetically influenced characteristic, a low sensitivity to alcohol, affects the risk for heavy drinking and may operate, in part, through variations in genes relating to potassium channels, GABA, nicotinic, and serotonin systems. A low response per drink is observed early in the drinking career and before alcohol use disorders have developed. All follow-up studies have demonstrated that this need for higher doses of alcohol to achieve desired effects predicts future heavy drinking, alcohol problems, and alcohol use disorders. The impact of a low response to alcohol on adverse drinking outcomes is mediated, at least in part, by a range of environmental influences, including the selection of heavier-drinking friends, the development of more positive expectations of the effects of high doses of alcohol, and suboptimal ways of coping with stress.


Although the age of the first drink (~15 years) is similar in most alcoholics and nonalcoholics, a slightly earlier onset of regular drinking and drunkenness, especially in the context of conduct problems, is associated with a higher risk for later alcohol use disorders. By the early to midtwenties, most nonalcoholic men and women moderate their drinking (perhaps learning from more minor problems), whereas alcoholics are likely to escalate their patterns of drinking despite difficulties. The first major life problem from alcohol often appears in the late teens to early twenties, and a pattern of multiple alcohol difficulties by the midtwenties. Once established, the course of alcoholism is likely to be one of exacerbations and remissions, with little difficulty in temporarily stopping or controlling alcohol use when problems develop, but without help, desistance usually gives way to escalations in alcohol intake and subsequent problems. Following treatment, between half and two-thirds of alcoholics maintain abstinence for years, and often permanently. Even without formal treatment or self-help groups there is also at least a 20% chance of spontaneous remission with long-term abstinence. However, should the alcoholic continue to drink, the life span is shortened by ~10 years on average, with the leading causes of death, heart disease, cancer, accidents, and suicide.


The approach to treating alcohol-related conditions is relatively straightforward: (1) recognize that at least 20% of all patients have alcohol abuse or dependence; (2) learn how to identify and treat acute alcohol-related conditions; (3) know how to help patients begin to address their alcohol problems; and (4) know enough about treating alcoholism to appropriately refer patients for additional help.


Even in affluent locales, ~20% of patients have an alcohol use disorder. These men and women can be identified by asking questions about alcohol problems and noting laboratory test results that are likely to be abnormal in the context of regular consumption of six to eight or more drinks per day. The two blood tests with ≥60% sensitivity and specificity for heavy alcohol consumption are γ-glutamyl transferase (GGT) (>35 U) and carbohydrate-deficient transferrin (CDT) (>20 U/L or >2.6%); the combination of the two is likely to be more accurate than either alone. The values for these serologic markers are likely to return toward normal within several weeks of abstinence. Other useful blood tests include high-normal MCVs (≥91 μm3) and serum uric acid (>416 mol/L, or 7 mg/dL).

The diagnosis of alcohol abuse or dependence ultimately rests on the documentation of a pattern of repeated difficulties associated with alcohol use. Thus, in screening it is important to probe for marital or job problems, legal difficulties, histories of accidents, medical problems, evidence of tolerance, etc., and then attempt to tie in use of alcohol or another substance. Some standardized questionnaires can be helpful, including the 10-item Alcohol Use Disorders Identification Test (AUDIT) (Table 56-2), but these are only screening tools, and a face-to-face interview is still required for a meaningful diagnosis.

TABLE 56-2




TREATMENT Alcohol-Related Conditions

Acute Intoxication The first priority in treating severe intoxication is to assess vital signs and manage respiratory depression, cardiac arrhythmia, or blood pressure instability, if present. The possibility of intoxication with other drugs should be considered by obtaining toxicology screens for opioids or other CNS depressants such as benzodiazepines. Aggressive behavior should be handled by offering reassurance but also by considering the possibility of a show of force with an intervention team. If the aggressive behavior continues, relatively low doses of a short-acting benzodiazepine such as lorazepam (e.g., 1–2 mg PO or IV) may be used and can be repeated as needed, but care must be taken not to destabilize vital signs or worsen confusion. An alternative approach is to use an antipsychotic medication (e.g., 0.5–5 mg of haloperidol PO or IM every 4–8 h as needed, or olanzapine 2.5–10 mg IM repeated at 2 and 6 h, if needed).

Intervention There are two main elements to intervention in a person with alcoholism: motivational interviewing and brief interventions. During motivational interviewing, the clinician helps the patient to think through the assets (e.g., comfort in social situations) and liabilities (e.g., health and interpersonal related problems) of the current pattern of drinking. The patient’s responses are key, and the clinician should listen empathetically, helping to weigh options and encouraging the patient to take responsibility for changes that need to be made. Patients should be reminded that only they can decide to avoid the consequences that will occur without changes in drinking. The process of motivational interviewing has been summarized by the acronym FRAMES: Feedback to the patient; Responsibility to be taken by the patient; Advice, rather than orders, on what needs to be done; Menus of options that might be considered; Empathy for understanding of the patient’s thoughts and feelings; and Self-efficacy, i.e., offering support for the capacity of the patient to succeed in making changes.

Once the patient begins to consider change, the emphasis shifts to brief interventions designed to help the patient understand more about potential action. Discussions focus on consequences of high alcohol consumption, suggested approaches to stopping drinking, and help in recognizing and avoiding situations likely to lead to heavy drinking. Both motivational interviewing and brief interventions can be carried out in 15-min sessions, but because patients do not always change behavior right away, multiple meetings are often required to explain the problem, discuss optimal treatments, and explain the benefits of abstinence.

Alcohol Withdrawal If the patient agrees to stop drinking, sudden decreases in alcohol intake can produce withdrawal symptoms, many of which are the opposite of those produced by intoxication. Features include tremor of the hands (shakes); agitation and anxiety; autonomic nervous system overactivity including an increase in pulse, respiratory rate, and body temperature; and insomnia. These symptoms usually begin within 5–10 h of decreasing ethanol intake, peak on day 2 or 3, and improve by day 4 or 5, although mild levels of these problems may persist for 4–6 months as a protracted abstinence syndrome.

About 2–5% of alcoholics experience a withdrawal seizure, with the risk increasing in the context of concomitant medical problems, misuse of additional drugs, and higher alcohol quantities. The same risk factors also contribute to a similar rate of delirium tremens (DTs), where the withdrawal includes delirium (mental confusion, agitation, and fluctuating levels of consciousness) associated with a tremor and autonomic overactivity (e.g., marked increases in pulse, blood pressure, and respirations). The risks for seizures and DTs can be diminished by identifying and treating any underlying medical conditions early in the course of withdrawal.

The first step in treating withdrawal is to perform a thorough physical examination in all alcoholics who are considering stopping drinking, including a search for evidence of liver failure, gastrointestinal bleeding, cardiac arrhythmia, infection, and glucose or electrolyte imbalance. It is also important to offer adequate nutrition and oral multiple B vitamins, including 50–100 mg of thiamine daily for a week or more. Because most alcoholics who enter withdrawal are either normally hydrated or mildly overhydrated, IV fluids should be avoided unless there is a relevant medical problem or significant recent bleeding, vomiting, or diarrhea.

The next step is to recognize that because withdrawal symptoms reflect the rapid removal of a CNS depressant, alcohol, the symptoms can be controlled by administering any depressant in doses that decrease the agitation and then gradually tapering the dose over 3–5 days. While most CNS depressants are effective, benzodiazepines (Chap. 54) have the highest margin of safety and lowest cost and are, therefore, the preferred class of drugs. Short-half-life benzodiazepines can be considered for patients with serious liver impairment or evidence of brain damage, but they must be given every 4 h to avoid abrupt blood-level fluctuations that may increase the risk for seizures. Therefore, most clinicians use drugs with longer half-lives (e.g., chlordiazepoxide), adjusting the dose if signs of withdrawal escalate, and withholding the drug if the patient is sleeping or has evidence of orthostatic hypotension. The average patient requires doses of 25–50 mg of chlordiazepoxide or 10 mg of diazepam given PO every 4–6 h on the first day, with doses then decreased to zero over the next 5 days. While alcohol withdrawal can be treated in a hospital, patients in good physical condition who demonstrate mild signs of withdrawal despite low blood alcohol concentrations and who have no prior history of DTs or withdrawal seizures can be considered for outpatient detoxification. These patients should return daily for evaluation of vital signs and can be hospitalized if signs and symptoms of withdrawal escalate.

Treatment of the patient with DTs can be challenging, and the condition is likely to run a course of 3–5 days regardless of the therapy employed. The focus of care is to identify and correct medical problems and to control behavior and prevent injuries. Many clinicians recommend the use of high doses of a benzodiazepine (as much as 800 mg/d of chlordiazepoxide has been reported), a treatment that will decrease agitation and raise the seizure threshold but probably does little to improve the confusion. Other clinicians recommend the use of antipsychotic medications, such as haloperidol or olanzapine as discussed earlier, although these drugs have not been directly evaluated for DTs. Antipsychotics are less likely to exacerbate confusion but may increase the risk of seizures; they have no place in the treatment of mild withdrawal symptoms.

Generalized withdrawal seizures rarely require more than giving an adequate dose of benzodiazepines. There is little evidence that anticonvulsants such as phenytoin or gabapentin are more effective in drug-withdrawal seizures, and the risk of seizures has usually passed by the time effective drug levels are reached. The rare patient with status epilepticus must be treated aggressively (Chap. 26).

REHABILITATION OF ALCOHOLICS An Overview After completing alcoholic rehabilitation, ≥60% of alcoholics, especially middle-class patients, maintain abstinence for at least a year, and many achieve lifetime sobriety. The core of treatment uses cognitive-behavioral approaches to help patients recognize the need to change, while working with them to alter their behaviors to enhance compliance. A key step is to optimize motivation toward abstinence through education about alcoholism and instructions to family members to stop protecting the patient from problems caused by alcohol. After years of heavy drinking, patients also need counseling, vocational rehabilitation, and self-help groups such as Alcoholics Anonymous (AA) to help them learn how to deal with life’s stresses while sober. A third component, relapse prevention, helps the patient to identify situations in which a return to drinking is likely, formulate ways of managing these risks, and develop coping strategies that increase the chances of a return to abstinence if a slip occurs.

While many patients can be treated as outpatients, more intense interventions work better, and some alcoholics do not respond to AA or outpatient groups. Whatever the setting, subsequent contact with outpatient treatment staff should be maintained for a minimum of 6 months and preferably a full year after abstinence. Counseling focuses on areas of improved functioning in the absence of alcohol (i.e., why it is a good idea to continue to abstain) and helping the patient to manage free time without alcohol, develop a nondrinking peer group, and handle stresses on the job.

The physician serves an important role in identifying the alcoholic, diagnosing and treating associated medical or psychiatric syndromes, overseeing detoxification, referring the patient to rehabilitation programs, providing counseling, and, if appropriate, selecting which (if any) medication might be needed. For insomnia, patients should be reassured that troubled sleep is normal after alcohol withdrawal and will improve over subsequent weeks. They should be taught the basic elements of “sleep hygiene” including maintaining consistent schedules for bedtime and awakening. Sleep medications have the danger of being misused and of rebound insomnia when stopped. Sedating antidepressants (e.g., trazodone) should not be used as they interfere with cognitive functioning the next morning and disturb the normal sleep architecture, but occasional use of over-the-counter sleeping medications (sedating antihistamines) can be considered. Anxiety can be addressed by helping the person to gain insight into the temporary nature of the symptoms and to develop strategies to achieve relaxation as well as by using forms of cognitive therapy.

Medications for Rehabilitation Several medications have modest benefits when used for the first 6 months of recovery. The opioid-antagonist, naltrexone, 50–150 mg/d orally, appears to shorten subsequent relapses, whether used in the oral form or as a once-per-month 380-mg injection, especially in individuals with the G allele of the AII8G polymorphism of the μ opioid receptor. By blocking opioid receptors, naltrexone may decrease activity in the dopamine-rich ventral tegmental reward system, and decrease the feeling of pleasure or reward if alcohol is imbibed. A second medication, acamprosate (Campral) at ~2 g/d divided into three oral doses, has similar modest effects; acamprosate inhibits NMDA receptors, decreasing mild symptoms of protracted withdrawal. Several trials of combined naltrexone and acamprosate using doses similar to those noted earlier have reported that the combination may be superior to either drug alone, although not all studies agree.

It is more difficult to establish the asset-to-liability ratio of a third drug, disulfiram, an ALDH inhibitor, used at doses of 250 mg/d. This drug produces vomiting and autonomic nervous system instability in the presence of alcohol as a result of rapidly rising blood levels of the first metabolite of alcohol, acetaldehyde. This reaction can be dangerous, especially for patients with heart disease, stroke, diabetes mellitus, or hypertension. The drug itself carries potential risks of depression, psychotic symptoms, peripheral neuropathy, and liver damage. Disulfiram is best given under supervision of another individual (such as a spouse), especially during discrete periods identified as representing high-risk drinking situations (such as the Christmas holiday). Other relevant drugs under investigation include the nicotinic receptor agonist varenicline, the serotonin antagonist ondansetron, the α-adrenergic agonist prazosin, the GABA-B receptor agonist baclofen, the anticonvulsant topiramate, and cannabinol receptor antagonists. At present, there are insufficient data to determine the asset-to-liability ratio for these medications in treating alcoholism and, therefore, no data to offer solid support for their use in clinical settings.