10.1192/apt.2.6.249Access the most recent version at DOI: 1996, 2:249-257.APT

Jonathan ChickMedication in the Treatment of Alcohol Dependence

Referenceshttp://apt.rcpsych.org/content/2/6/249.citation#BIBLThis article cites 0 articles, 0 of which you can access for free at:

permissionsReprints/

permissions@rcpsych.ac.ukto To obtain reprints or permission to reproduce material from this paper, please write

to this article atYou can respond http://apt.rcpsych.org/letters/submit/aptrcpsych;2/6/249

from Downloaded

The Royal College of PsychiatristsPublished by on October 15, 2014http://apt.rcpsych.org/

http://apt.rcpsych.org/site/subscriptions/go to: Adv. Psychiatr. Treat. To subscribe to

Advances in Psychiatric Treatment (1996), vol.2, pp. 249-257

Medication in the treatment ofalcohol dependence

Jonathan Chick

Social,cultural, emotional and biological influencesdetermine whether people drink to excess andwhether they then experience harm or cause harmto others (Cook, 1994).Psychosocial treatments foralcohol dependence are only modestly successful,with most studies finding that at least 50% ofpatients return to harmful drinking in the followingyear. In the past decade there has been newevidence for the role of pharmacological treatmentsin reducing harm from drinking and in preventingrelapse.

However, none of the treatments described hereis recommended as a solo therapy. Social andpsychological factors in treatment for alcoholdependence are crucial. Firstly, the chief patientvariables predicting good outcome are having a joband a supportive relationship. Secondly, the mostpowerful therapy variable to date is therapistempathy. Finally, compliance with medicationhinges on the patient's understanding and motivation, both of which are enhanced by familysupport and attitudes, and the therapist-patientrelationship (Box 1).

Assisted withdrawal

Aims

Electively, the psychiatrist and alcohol-dependentpatient may decide to facilitate commencingabstinence by reducing the short-term discomfortof withdrawal. This can be the beginning ofrestructuring thoughts and lifestyle towards long-term abstinence. Even in elective 'detoxification',

and certainly in the non-elective situation wherethe drinker has suddenly been deprived of alcoholbecause of an accident, illness or police arrest, allcare must be taken to prevent the life-threateningcomplications of convulsions or delirium. Anticipation is the key.

When and how

Some out-patients succeed in reducing theirdrinking in gradual steps. Benzodiazepines areindicated if withdrawal symptoms promise to betoo uncomfortable for the patient to control the urgeto seek alcohol, or there is a risk of delirium orconvulsions indicated by past history, or recentconsumption was more than 15 units per day formore than 10 days.

In-patient detoxification is indicated where thereis a history of convulsions, incipient delirium or aliving situation inimical to abstinence. Whenmedication is not started until the day after the lastdrink in a severely dependent patient, as sometimesis the case in medical and surgical settings, largeor frequent doses will be needed initially to titratesedation against agitation with up to 200 mgchlordiazepoxide or 80 mg diazepam being required in the first 24 hours.

If the patient is vomiting, give metoclopropa-mide, 10mg intramuscularly 30minutes before thefirst of the benzodiazepine tablets. Lorazepam 1 mgis absorbed adequately from the intramuscular site,and diazepam 10 mg can be given intravenously(Shaw, 1995).

A typical out-patient regime would be chlordiazepoxide 20-30 mg q.i.d., or diazepam 10 mgq.i.d. in the first 36 hours, reducing to nil over fivedays, giving the larger doses at night. Medication

Jonathan Chick, FRCP Ed., FRCPsych, is Consultant Psychiatrist, Alcohol Problems Clinic, Royal Edinburgh Hospital, and SeniorLecturer, Department of Psychiatry, Edinburgh University. While working for the Medical Research Council he studied the evolutionof alcohol problems in distillery and brewery workers and company directors. This led to research into early detection and briefintervention for drinking problems and later to the evaluation of treatments for severely dependent drinkers.

APT (1996), vol. 2, p. 250 Chick

Box 1. Factors demonstrated to improvetreatment outcome

Patient is employed and has a supportiveliving arrangement

Therapist empathyCoping skills and social skills trainingSupervised disulfiramNaltrexoneAcamprosate(Alcoholics Anonymous: not shown in

controlled studies, but powerfully effective for some)

is issued on the understanding that the patient doesnot also take alcohol. If there is doubt that thisinstruction will be followed, medication is issueddaily and a check made that drinking has not beenresumed using an alcohol breath test, if abreathalyser is available. As required, additionaldoses may be needed in the first 48 hours (Table 1).

Advice to a patient withdrawing from alcoholat home

(1) If you have been chemically dependent onalcohol, stopping drinking causes you to gettense, edgy, perhaps shaky or sweaty, andunable to sleep. There can be vomiting ordiarrhoea. This 'rebound' of the nervoussystem can be severe. Medication controls thesymptoms while the body adjusts to beingwithout alcohol. This usually takes three toseven days from the time of your last alcoholicdrink. If you didn't take medication, thesymptoms would be worst in the first 48hours, and then gradually disappear. This iswhy the dose starts high and then reduces.

(2) YOU HAVE AGREED NOT TO DRINKALCOHOL. You may get thirsty. Drink fruitjuices and water but do not overdo it. Youdo not have to 'flush' alcohol out of the body.

Table 1. Example of dose regime for alcoholwithdrawal using capsules of chlordiazepoxide

10 mg

First thing 12 noon 6 p.m. Bedtime

Day!Day 2 2Day3 2Day 4 1Day5

32111

33221

More than three litres of fluid could be toomuch. Don't drink more than three cups ofcoffee or five cups of tea. These containcaffeine which disturbs sleep and causesnervousness.

(3) Aim to avoid stress. The important task is notto give in to the urge to take alcohol. Helpyourself relax by going for a walk, listeningto music or taking a bath.

(4) Sleep. You may find that even with the capsules, or as they are reduced, your sleep isdisturbed. You need not worry about this -lack of sleep does not seriously harm you,starting to drink again does. Your sleeppattern will return to normal in a month orso. It is better not to take sleeping pills so thatyour natural sleep rhythm returns. Try goingto bed later. Take a bedtime snack or milkydrink.

(5) The capsules may make you drowsy so youmust not drive or operate machinery. If youget drowsy, miss out a dose.

(6) Meals. Even when you are not hungry, try toeat something. Your appetite will return.

Complications

Wernicke-Korsakoff syndrome

Wernicke-Korsakoff syndrome seems often to beprecipitated by hospital admission and withdrawalfrom alcohol. The physiological stress of withdrawal may contribute; as may resuming intakeof carbohydrates, breakdown of which requiresenzymes dependent on thiamine and uses upremaining stores of that essential component ofneuronal metabolism. It is good preventivepractice to prescribe thiamine 200 mg orally dailyduring the withdrawal phase. If the patient is ataxicor obviously malnourished, give thiamine paren-terally. The currently available vitamin B injection,Tabrinex', has not acquired a reputation for allergicreactions, but administration should be in a settingwhere resuscitation facilities are available in caseof anaphylactic shock. Anaphylaxis is less likelywith intramuscular than intravenous injection, andof the intravenous routes perhaps slow infusionsaline drip is preferable to slow bolus injection.Treating Wernicke-Korsakoff syndrome. If confusion,incontinence, ataxia or strabismus are noted in apatient withdrawing from alcohol, intravenousthiamine must be given immediately and hypogly-caemia considered. Then, consideration may givento other potential causes, such as subdural haema-toma or hepatic encephalopathy.

Alcohol dependence APT (1996), vol. 2, p. 251

Convulsions

Deaths have occurred in hospital, prison and policecells from bursts of alcohol withdrawal fits. Electivewithdrawal from alcohol in patients with a historyof fits of any cause can be made safer by commencing an anticonvulsant (e.g. phenytoin or carba-mazepine) four days before cessation of drinking.This permits a therapeutic serum level to beachieved in good time. Alternatively, larger thannormal doses of long-acting benzodiazepines aregiven in the first 36 hours, and should be startednot after the blood alcohol level has fallen to zerobut before. If the patient is sober enough tocooperate appropriately with admission, thepsychiatrist should commence benzodiazepineswhile the patient still smells of alcohol.

A patient died at 6 p.m. in a convulsion having beenadmitted at 12 noon to a psychiatric ward. The nurses,adhering to the prescribed 6 hourly regime, had waiteduntil 6 p.m. to give him his first dose of sedative.

Treating convulsions. With the aim of preventingfurther convulsions, the patient is given 10 mgdiazepam intravenously or rectally. Give doublethe dose in a patient who has been taking benzodiazepines regularly prior to this event, or is muchabove average weight. A convulsion may presagea severe withdrawal syndrome, and parenteralthiamine should be given. It is illogical tocommence an anticonvulsant which may take 2-3days to reach a therapeutic serum level. Rather,increase the benzodiazepines, which are effectivein controlling alcohol withdrawal fits.

Delirium tremens

If confusion and hallucinations develop, this isoften 48-72 hours after the last drink. Sufficientbenzodiazepine, given soon enough in the withdrawal phase, reduces the risk. Good nursing in a well-lit, calm environment is preventive. Explainingthings and orientating the patient reduces anxiety,paranoia and confusion.

Treating delirium tremens. Increasing the dose of thebenzodiazepine may be sufficient. If not, the slightepileptogenic effect of phenothiazines should notdeter the psychiatrist from prescribing these tocontrol delusions and hallucinations, if anticonvulsant protection by a benzodiazepine is inplace. The right environment and nursing helpgreatly in reducing the risk of aggression.

Alcoholic hallucinosis

Hallucinations, perhaps with secondary delusionalbeliefs, may develop during a period of heavydrinking or of relative reduction in drinking, ormay be found to persist after the acute symptoms

of delirium tremens resolve. Abstinence plus chlor-promazine (e.g. 50-100 mg b.d.) or other majortranquilliser usually results in complete recoveryover the coming months, permitting withdrawalof the medication. However, a proportion of suchillnesses persist and are later diagnosed asschizophrenia or affective psychosis (Cutting, 1978).

Which sedative?

The longer-acting benzodiazepines, diazepam andchlordiazepoxide, are the most successful inreducing anxiety and the risk of convulsions. Bothhave active metabolites, which themselves requireexcretion by the liver and so may cause a cumulative over-sedation in the elderly or those with liverfailure, unless progressive reduction is madeappropriately. Lorazepam and oxazepam haveintermediate half-lives and do not produce activemetabolites, being inactivated and eliminated bysimple glucuronidation. Dosage may be harder totitrate than for diazepam or chlordiazepoxide.

Withdrawal symptoms can be controlled withchlormethiazole. It has the disadvantage that it isrelatively short-acting (elimination half-life 3-6hours). It should not be given to out-patientsbecause of the risk of respiratory depression, iftaken in overdose or with large amounts of alcohol.It has addictive properties, perhaps because of itsrapid action and short half-life, which has led topatients seeking prescriptions after detoxificationwho develop dependence with subsequent withdrawal convulsions. Intravenous infusions,although effective in emergency situations such asdelirium tremens in a postoperative patient, havecaused fatal respiratory depression, and should beused only where intubation and ventilationfacilities are on hand, and with great care in chronicpulmonary insufficiency and advanced liverdisease.

Deterrent medication toreduce relapse

DisulfiramAction

Disulfiram, if taken regularly in a sufficient dose,causes an unpleasant reaction 15-20 minutes afteralcohol enters the body. The reaction is due toaccumulation of the intermediate metabolite ofethanol, acetaldehyde. The patient flushes,

APT (1996), vol. 2, p. 252 Chick

Box 2. Procedures for successful use ofdeterrent drugs

Physical examination, cardiac history andbaseline liver function tests.

Explain actions of the drug.Negotiation proceeds so that the patient not

only accepts the drug but wants it.Information leaflet for patient and relatives.Offer psychosocial support, and help to cope

with thoughts of drinking and handlingprevious triggers, e.g. out-patient group,AA.

Facilitate the patient inviting a third party'to help him/her remember to take thedrug', e.g. partner, family member, clinicstaff, nurse or supervisor at work.

Set up acceptable supervision regime -tablet(s) seen to have the correct markings(i.e. not soluble aspirin) are dispersed inhalf glass of water and taken in view ofsupervisor; frequency agreed (e.g. dailyor thrice weekly).

Agreement that supervisor phones clinicianif patient appears to be changing the plan;and that clinician phones patient to askreasons for change of plan.

When supervision is at the clinic: properlybriefed staff; delays avoided; approach/atmosphere welcoming and reinforcing.

Liver function tests repeated once after 1-2months.

experiences headache, pounding in the chest orhead, tightness in breathing, nausea and perhapsvomiting. Unwanted effects of the drug itself arefew, with drowsiness or headache the only commonones. A taste in the mouth, or an odour on thebreath, are sometimes experienced. Loss of libidois attributed by some patients to disulfiram, butdoes not occur more frequently than in placebo-treated control patients.

The liver acetaldehyde dehydrogenase isoformthat disulfiram blocks seems to have a turnover ofseveral days. The alcohol-disulfiram reaction isusually only experienced in patients who havetaken it for 34days. A loading dose is recommended. The enzyme appears to stay blocked forseveral days, with alcohol-disulfiram reactionsoccurring up to 7 days after the last dose.

It is recognised practice to increase the dose ofdisulfiram to 300 or 400 mg/day, if the patient hastaken alcohol and the reaction has not been severeenough to act as a deterrent. A few tolerate thereaction, do not wish to increase the dose, and use

the drug as a way of controlling their drinkingrather than abstaining (in which case the clinicianrecords that the patient does this at his own wishand against advice).

Mode of use (Box 2)Disulfiram is an aid. It enables the individual toget used to life without alcohol and allows timefor confidence to resume in the family and at work.It acts as a deterrent to temptation and thus helpspostpone further alcohol crises. Damaged organshave time to recuperate. Patients suspendedbecause of drinking might be reinstated if theemployer knows disulfiram is being used.

If drowsiness is a side-effect, it is taken atbedtime. It is common to prescribe it for six months,but many patients want to continue the methodlonger. Alternatively, patients may keep a supplyto use when they feel they are going to be at risk ofdrinking; for example, a business trip away or asocial event.

Patients may object that it is a sign of weaknessto be taking disulfiram, instead of using will-power.Unfortunately, will-power is not always there whenmost needed, so disulfiram acts as a last line ofdefence. With the pills, a decision whether to drinkor not still has to be made, but only once a day.

Tablets implanted subcutaneously sensitise thepatient for about two weeks only, not for months.This, together with the risk of local irritation, hasled to the virtual disappearance of this approach.

RisksThe dangerous component of the disulfiram-ethanol reaction is the hypotension or cardiacdisturbance. Disulfiram is contraindicated aftermyocardial infarction, angina or arrhythmia. Ifsuch patients strongly request to use disulfiram,perhaps because they know that a bout of drinkingcould be as dangerous as an alcohol-disulfiramreaction, a statement of the added danger shouldbe given and acknowledged by the patient. Manydoctors prescribe disulfiram without a screeningelectrocardiogram, but the medical history andpulse must be checked. For many years now, sincedoses of 500 mg or more have ceased to be used,there have been no deaths due to the disulfiram-ethanol reaction, either reported in the literature,or in areas of frequent prescribing such asScandinavia.

Disulfiram can provoke a psychotic state in apredisposed individual and should not be givento someone with a history of paranoid thinking orpsychotic illness.

Peripheral neuropathy is an infrequent toxiceffect in patients who have used disulfiram in high

Alcohol dependence APT (1996), vol. 2, p. 253

doses (usually >400mg) for many months and maybe more common in association with tricyclics. Itis reversible, but contraindicates re-exposure to thecompound.

Drug interactions

Serum levels of anticonvulsants and tricyclics maybe enhanced in patients taking disulfiram. Themetabolism of caffeine,warfarin, chlordiazepoxideand diazepam is delayed by disulfiram. It shouldnot be given with metronidazole, since thecombination has caused a confusional state.

Supervision

Disulfiram works only if taken consistently. Therandomised controlled studies that have demonstrated efficacy (and shown near 100% two-yearabstinence when combined with contingent maritaland community rewards) have entailed recruitinga supervisor. Patients are encouraged to ask theirpartner, a nurse or welfare officer at work, or anurse at the health centre or the clinic to see themtake the disulfiram. This can either be on a dailybasis, or three times a week as long as the totalnumber taken per week is at least seven 200 mgtablets. The product is now sold in a dispersibleform to take in water so that it can be seen to beswallowed. Most find it tasteless (Box3).

Disulfiram and the liver

There have been rare reports of hepatitis in patientsnewly starting disulfiram, mostly in women.Because of this, some doctors recommend thatregular liver function tests are performed. However, since the reactions reported are in the earlyweeks of therapy, a check after one month ratherthan repeated tests is probably all that is necessary.A randomised trial of disulfiram versus vitamin Cas control found those on disulfiram showedgreater improvement in mean serum gammaglutamyl transferase than controls (Chick et al,1992). The disulfiram group drank less, whichexplains why their mean gamma glutamyl transferase improved more, but it substantiated otherwork which had failed to show that disulfiramimpaired liver function.

Because hypersensitivity could be fatal insomeone with a compromised liver, great cautionshould be used in patients with advanced liverdisease and a rule of thumb might be not toprescribe it when the serum bilirubin is >25mmol/1and the serum albumin is below normal, low serumalbumin being an indication of liver failure. Also,disulfiram is metabolised hepatically and mightstrain a diseased liver. However, elevated liver

Box 3. Deterrent medication: information forpatient and partner

A partner is a person who is asked by thepatient to observe the taking of theAntabuse tablets.

So that other tablets cannot be substituted,the genuine Antabuse tablets are markedDumex 110 L (Dumex is the manufacturer).

Toensure that they are not placed under thetongue and removed later, Antabusetablets should be dissolved in half a glassof water (the tablets break up and disperse and the mixture is tasteless).

It does not matter what time of the day thetablet is given. If it is more convenient, itcan be given on three days a week (i.e.instead of one tablet daily it can be takentwo on Monday, two on Wednesday andthree on Friday, for example).

If it is suspected that the patient has decidedto vomit after taking the tablet, thepartner can stay with the patient for upto 30minutes after the tablet is taken (thisis rarely necessary).

If the patient decides to stop taking thetablets, the patient or the partner shouldtelephone the treating doctor or a memberof nursing staff so that the reason for thismay be discussed.

enzymes, which are found in 60- 70% of patientswith alcohol problems in psychiatric practice in theUK, are not a contraindication to disulfiram.

Calcium carbimide

If disulfiram cannot be prescribed because ofcontraindications or unwanted effects, calciumcarbimide is available. In the UK its license is notcurrently retained, apparently for commercialrather than safety reasons. It has the proprietaryname 'Abstem' in the UK ('Temposil' in Canada)and is prescribed on the basis of "named patient,physician's responsibility", that is, the manufacturer (Cyanamid) does not offer indemnity.

Calcium carbimide may block a different isoformof liver acetaldehyde dehydrogenase than disulfiram - the alcohol reaction can occur any time upto 36hours after a dose, which is much shorter thanthe duration of action of disulfiram. The dose ofcalcium carbimide is 100mg/day. Thyroid depres-

APT (1996), vol. 2, p. 254 Chick

sion and reduced white cell count have beenreported.

Motivating patients to use adeterrent

Many patients who decide the advantages ofstopping drinking outweigh the advantages ofcarrying on, will accept a deterrent. If rewards ofabstinence continue, they will persevere. They maybe glad to find the craving for alcohol wanes: "Iknow I dare not drink, so the debate in my headwhether or not to drink evaporates."

Many, however, cease taking the tablets after twoor three weeks, especially if there is no supervisionarrangement. Benefits in the marriage are apowerful reinforcer. The therapist can facilitate thisby:

(1) improving communication;(2) helping each partner to express their needs

with assertion not aggression; and(3) instructing each partner to reward and

encourage positive behaviours in the other.

Keeping their job and getting rewards at workwill boost motivation. For those unemployed andliving alone, the rewards of feeling more competentand autonomous will help, as will membership ofgroups such as Alcoholics Anonymous (AA).

Some repeat offenders ask for deterrent drugsbecause they recognise that relapse coincides withre-offending. Psychiatrists reporting to a court canoffer deterrent drugs to repeat offenders as a partof a treatment package. Sometimes, courts wishingto assist rehabilitation or to avoid a custodialsentence, defer sentence and ask the psychiatristto report in three or six months on whether theoffender has responded to treatment. This invokesan element of coercion in the treatment. However,just as a spouse's threat to leave, or pendingdismissal from work, often correlate with a periodof renewed effort to stop drinking, so also theoffender may assiduously follow a treatment planif he has to return to court in a few months. This isslightly different from a court mandating probationwith disulfiram as a condition, which puts thepsychiatrist in a different relationship to his patient,and one which many would find uncomfortable.

Treatment duration

Initially, suggest at least six months. There is agradual drop-out over the weeks, as some patientsdecide to resume drinking and stop the medication.The supervision arrangement may have becomelax. However, there are patients, and families, who

ask to continue the deterrent method for longer.There is no evidence of long-term toxicity inpatients taking 200 mg disulfiram daily, thoughthere has been a report of cognitive impairment(reversible) in a patient taking a larger dose formany years (Borrettef al, 1985). There are currentlymany around the world who have used disulfiramfor years.

New advances

There are genetic as well as environmentalcontributions to why alcohol problems run infamilies. This has spurred on research into theneurobiology of addictions (see Nutt, 1996). Thereare strains of laboratory animals predisposed totake alcohol that, given the opportunity, work foralcohol and show withdrawal symptoms.

The neurotransmitters involved in reward,repetitive behaviours and drug-seeking behavioursinclude endorphins, dopamine and serotonin. They-aminobutyric acid (GABA)/glutamate systemsare important.

Ethanol substantially reduces activity of voltage-gated calcium channels at the nerve cell membraneand interferes with the N-methyl-D-aspartate(NMDA) control, which leads to depression of theexcitatory glutamate system. Animals dependenton ethanol, but no longer exposed to it, are foundto have increased glutamate activity because ofcompensatory overactivity of calcium channels andreduced NMDA control.

The two drugs below have been tested inrandomised controlled trials and shown modestbut clinically useful efficacy.

Acamprosate (calcium acetylhomotaurinate)This enhances GABA transmission and antagonisesglutamate transmission, without any benzo-diazepine-like anxiolytic action, probably byaffecting calcium channels and NMDA receptors.It reduces drinking in alcohol-dependent animals,and reduces the reinstatement of drinking behaviour and withdrawal symptoms in animals re-exposed to alcohol after a period of abstinence. Itdoes not substitute for ethanol or benzodiazepinesin such animals in the sense that they will seek outacamprosate (reviewed by Littleton, 1995). It has adose-related effect in improving abstinence rates inrecently detoxified patients (Paille et al, 1995). Otherimpressively large randomised controlled studiesof acamprosate have shown an effect, typically of

Alcohol dependence APT (1996), vol. 2, p. 255

enhancing complete abstinence by some 20% forup to one year (e.g. Whitworth et al, 1996; Sass etal, 1996). Other studies have been published butdiffer in the extent to which sedatives werepermitted in the first weeks and in the outcomecriteria (Chick, 1995). Acamprosate is available onprescription in the UK and most Europeancountries.

Naltrexone/nalmefeneThese antagonise the brain's endogenous opiatetransmitters, endorphins, which are released as oneof many acute actions of ethanol on the limbic system.Individuals with a high genetic loading for alcoholdependence may inherit an oversensitive endorphinrelease after ethanol, which might contribute to theloss of control experienced by some drinkers(Gianoulakisef al, 1996).Naltrexone reduces ethanol-seeking in dependent animals.

Two double-blind, randomised controlled studiesof naltrexone in detoxified patients taking part in anout-patient treatment programme have been published and show a reduced risk of relapse, at least forthree months. The same appears to be true fornalmefene (Mason et al, 1994). The effect size ofnaltrexone treatment in reducing the percentage ofdays drinking was 0.42 in one study and 0.60 in theother (reviewed in Volpiceli! et al, 1995). Forcomparison, the mean effect size in meta-analyses ofother studies of fluoxetine in the treatment ofdepression is around 0.4 (Greenberg et al, 1994).Results of longer studies, and larger samples, areawaited. In 1995naltrexone was licensed in the UnitedStates for use as part of a comprehensive treatmentprogramme for alcohol dependence, and in Canadaand Austria the drug was licensed in 1996.At the timeof writing, naltrexone is licensed in the UK for opiateaddiction (because it removes the euphoria of takingheroin) but not in alcohol dependence, for whichindication it is currently prescribed on the principle:"named patient, physician's own responsibility".

Some patients who resume drinking while takingnaltrexone report that they feel less of the ethanol'high'. Perhaps they then experience less impulse tocarry on drinking (Volpiceli!, 1995; Volpicelli et al,1995). However, there is an increase in the number ofpatients who report achieving total abstinence as wellas a reduction in drinking overall. Early speculationthat opiate antagonists might cause dysphoriaseemed to be supported by statements from heroinaddicts given naltrexone to help them abstain fromopiates. However, laboratory studies and randomisedcontrolled trials have not found consistent evidenceof dysphoria or loss of feelings of pleasure in eithernormal volunteers or alcoholics (for example, seeDoty & de Wit, 1995).

Action and use of the new agentsIt is possible that the reduced likelihood of pickingup the first drink, and the reduced craving for alcoholshown in some studies, is because the strength of theprevious triggers - emotional, cognitive or environmental - is attentuated by the drugs' actions in certainpathways in the limbic system. In time these will beelucidated. More information is now needed onwhich patients respond, what is the optimal timingand duration of use, and whether, and if so which,psychological and social interventions are necessarycomplements. Currently, it is reommended that thedrugs are prescribed at the beginning or soon aftercommencement of detoxification. Naltrexone is givenas one tablet once daily, and acamprosate three timesdaily. Neither drug is addictive in the sense that thereis a withdrawal syndrome. Follow-up has notrevealed rapid relapse after cessation of these drugs.Both have a good safety record and it is notable thatacamprosate is not metabolised in the liver. Neitherdrug exacerbates psychomotor impairment causedby alcohol. It is likely that poor compliance, unlessresolved, will limit the effectiveness of these newproducts, although perhaps less than it has with thedeterrent drugs.

Since they are expensive relative to disulfiram, andof course to AA (which costs the National HealthService nothing), the use of naltrexone and acamprosate may initially be cautious. Perhaps they shouldbe used only in patients who have failed to respondto brief psychological intervention. Such patientswould at least see the logic of using an aid toabstinence and be more likely to comply withmedication. However, the costs to individuals,families, the NHS and society of persistent, relapsingalcohol dependence should also be weighed.

Treatment of coexistingaffective disorder

Depression

Depression is common in patients dependent onalcohol. It may be a result of drinking, or loss offriends, family or work, with resulting feelings ofhopelessness, guilt and lack of direction. They mayhave little appetite because they are drinkinginstead, and may have lost energy and sexual drivebecause of lowered serum testosterone. They maywake in the small hours of the night feeling anxiousbecause of the rebound wakefulness of alcohol

APT (1996), vol. 2, p. 256 Chick

withdrawal. Those signs and symptoms of depressive illness commonly clear with abstinence, helpin tackling or tolerating the problems that exist, andgetting relationships on to a better footing

However, in some patients (women more thanmen) a depressive episode preceded the alcoholdependence. Alcohol was taken in part as self-medication. Or, despite abstinence, depressivesymptoms are found to continue. In this case,antidepressants should be offered in the usual way.Relapsing alcoholism, if secondary to depressiveillness, is an indication for long-term antidepressants. Lithium is not a treatment for alcoholdependence itself, but is effective if it is secondaryto manic-depressive disorder.

Anxiety and panic disorder

A patient may have had panic attacks for yearsbefore discovering that alcohol could end orprevent an attack. Other patients may have had thefirst panic attack due to alcohol withdrawal, butthe attacks then continued, even into sustainedperiods of abstinence. In either case, whencognitive-behavioural therapy is not sufficient,medication is indicated, especially if growingphysical or social damage is accruing at everyalcoholic relapse.

Benzodiazepines are immensely helpful to suchpatients. However, their use should normally beonly short-term because there is a risk that tolerancewill develop with aggravation of symptoms dueto increasing partial withdrawal at periods betweendoses. Buspirone has been studied in such patients.One study (Kranzler et al, 1994) found an advantagein terms of drinking and anxiety but an earlierstudy (Malcolm et al, 1992) did not.

Tricyclic antidepressants and SSRIs, such asparoxetine, have a role in panic disorder. However,those anxious patients who have become dependent on alcohol seem to experience a high rate ofunwanted effects. With SSRIs, even titrating upfrom the lowest dose possible, they may need ananti-emetic such as metoclopropamide, 10 mg sixhourly (maximum 30 mg in 24 hours) for a fewdays, and/or a reducing benzodiazepine regimefor three weeks, if they are to be persuaded to givethe drug a chance to show its effect.

There are some patients with long histories ofalcohol dependence, who have failed to respondto numerous psychological and other treatmentventures, and who have pronounced panic disorder. Unless the patient is already a purchaser ofstreet drugs, the risk of complications from repeatedprescribing of a long-acting benzodiazepine are somuch less than the risks from alcohol excess that itcan sometimes be humane to the individual and the

family to prescribe the benzodiazepine. If prescribed(and to do so is controversial), it should be dispensedin limited aliquots or issued daily by a family member.It should be conditional on abstinence from alcohol,which can be aided by disulfiram if necessary. Thepatient should usually be encouraged to use an anti-depressant too.

'As required' use, for example, for travelling onpublic transport, is to be preferred to regular usein order to limit the development of tolerance, eventhough such use perhaps perpetuates the underlying phobic beliefs.

Reducing harm in patientswho cannot or will not stop

drinking heavily

There is research into how, pharmacologically, livercirrhosis might be prevented in those who carryon drinking. Propylthiouracil has been used to limitthe progression of alcoholic cirrhosis once diagnosed.

There has never been a controlled study to showthat regular thiamine supplements in committeddrinkers can prevent the development of Wernicke-Korsakoff syndrome or, indeed, the risk of alcoholicneuropathy or dementia. Nevertheless, it seemsprudent that individuals who drink heavily enoughto reduce their absorption of thiamine, or seldomeat, or have a family history of Wernicke-Korsakoffsyndrome, should be encouraged to take B-vitaminsupplements.

References

Borrett, D., Ashby, P., Bilbao, ].,et al (1985) Reversible late onsetdisulfiram-induced neuropathy and encephalopathy. Annalsof Neurology, 17, 396-399.

Chick, J. (1995) Acamprosate as an aid in the treatment ofalcoholism. Alcoholand Alcoholism,30, 785-787., Gough, K., Falkowski, W.,et al (1992) Disulfiram treatmentof alcoholism. British Journal of Psychiatry, 161, 84-89.

Cook, C. (1994) Aetiology of alcohol misuse. In Seminars inAlcoholand Drug Misuse (eds J.Chick & R. Cantwell), pp. 94-125. London: Gaskell.

Cutting, J. (1978) A re-appraisal of alcoholic psychoses.PsychologicalMedicine, 8, 285-295.

Doty, P. & de Wit, H. (1995) Effects of naltrexone pretreatmenton the subjective and performance effects of ethanol in socialdrinkers. Behavioural Pharmacology, 6, 386-394.

Gianoulakis, C., Krishnan, B. & Thavundayil, J. (1996)Enhancedsensitivity of pituitary -endorphinto ethanol in subjects at highrisk of alcoholism. ArchivesofGeneralPsychiatry, 53,250-257.

Greenberg, R. P.,Bornstein, R. F.,Zborowski, M. J., et al (1994)Ameta-analysis of fluoxetine outcome in the treatment ofdepression. Journal of Nervous and Mental Diseases, 182,547-551.

Alcohol dependence APT (1996), vol. 2, p. 257

Kranzler, H. R., Burleson, J.A., Boca, E K.,et al (1994) Buspironetreatment of anxious alcoholics. ArchivesofGeneral Psychiatry,51,720-731.

Littleton, J. (1995) Acamprosate in alcohol dependence: howdoes it work? Addiction, 90,1179-1188.

Malcolm, R., Anton, R. R, Randall, C. L., et al (1992) A placebo-controlled study of busprione in anxious in-patientalcoholics. Alcoholism: Clinical and Experimental Research, 16,1007-1013.

Mason, B. J., Ritro, F. C, Morgan, R. O., et al (1994) A double-blind placebo-controlled study to evaluate the efficacy andsafety of oral nalmefene HC1 for alcohol dependence.Alcoholism: Clinical and Experimental Researches, 1162-1167.

Nutt, D. (1996) Addiction: brain mechanisms and theirtreatment implications. Lancet, 347, 31-36.

Faille, F.M., Guelfi, J. D., Perkins, A.C., et al (1995) Randomisedmulticentre trial of acamprosate in a maintenance programme of abstinence after alcohol detoxification. AlcoholandAlcoholism, 30, 239-247.

Sass, H., Soyka, M.,Mann, K.,et al (1996) Relapse preventionby acamprosate: results from a placebo controlled studyin alcohol dependence. ArchivesofGeneral Psychiatry, 53,673-680.

Shaw, G. K. (1995) Detoxification: the use of benzodiazepines.Alcohol and Alcoholism, 30, 765-770.

Volpiceli!, J. R. (1995) Naltrexone in alcohol dependence. Lancet,346, 456., Volpiceli!, L. A. & O'Brien, C. P. (1995) Medical management of alcohol dependence: clinical use and limitations ofnaltrexone treatment. Alcoholand Alcoholism, 30, 789-798.

Whitworth, A. B., Fischer, F., Lesch, O., et al (1996) Comparisonof acamprosate and placebo in long-term treatment of alcoholdependence. Lancet, 347,1438-1442.

Multiple choice questions

effects of SSRIsc is a complete contraindication to long-term

benzodiazepine prescriptiond can be helped by tricyclic antidepressantse is a cause of repeated relapse into problematic

drinking.

3. In alcohol withdrawal:a convulsions can be fatalb fits can be best treated by immediate prescri

ption of phenytoinc confusion and hallucinations may take 72

hours to manifestd phenothiazines are contraindicated if

delirium tremens developse sedatives can be started before the blood

alcohol has fallen to zero.

4. It is true of Wernicke-Korsakoff syndrome that:a patients are seldom under the age of 50b malabsorption of thiamine contributesc it is avoided in developed countries because

alcoholics are sufficiently nourishedd intake of carbohydrate may precipitate

symptomse it can be prevented by giving prophylactic

parenteral vitamins in the poorly nourishedpatient undergoing alcohol withdrawal.

1. Disulfiram:a is contraindicated in patients with raised

serum gamma glutamyl transferaseb causes tiredness, an unwanted effectc should be avoided in patients referred by the

courtsd if supervised by the spouse, can improve

outcomee can be taken as a tablet dispersed in water.

2. Panic disorder in alcohol dependence:a may have been caused by alcohol excessb is associated with sensitivity to the side-

MCQanswers1abcdeFTFTT2abCdeTTFTT3abcdeTFTFT4abcdeFTFTT