Psychiatric Service, 12 de Octubre University Hospital, Madrid, Spain
Received 29 December 2000; in revised form 16 March 2001; accepted 2 April 2001
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Naltrexone is an opioid receptor antagonist, with a verified efficacy for the reduction of euphoria, alcohol intake and relapse risk by alcohol-dependent or -misusing individuals (Volpicelli et al., 1992, 1995a
,b
, 1997
; O'Malley et al., 1992
; Anton et al., 1999
; Chick et al., 2000b
). These actions seem to be mediated by the property to block opiate receptors (Ulm et al., 1995
), not least in forebrain areas. This antagonism appears to inhibit the actions of endogenous opioids, released because of alcohol intake, upon the mesolimbic pathway, which would otherwise produce a rise in dopamine (DA) in the accumbens nuclei (Benjamin et al., 1993
; Valenzuela and Harris, 1997
; Catafau et al., 1999
). Naltrexone efficacy has been demonstrated in short-term double-blind studies (612 weeks) (O'Malley et al., 1992
; Volpicelli et al., 1992
, 1995a
, 1997
; Anton et al., 1999
; Chick et al., 2000b
). However, from the available evidence, naltrexone efficacy has not yet been verified in long-term studies.
Long-term efficacy studies (612 months) have been carried out, however, on acamprosate, calcium acetyl homotaurinate, a drug marketed in Europe. This has been shown to increase the time to relapse, to reduce the number of days of consumption and to augment the abstinence period (Pelc et al., 1992; Ladewig et al., 1993
; Paille et al., 1995
; Sass et al., 1996
; Geerlings et al., 1997
; Poldrugo, 1997
; Besson et al., 1998
; Tempesta et al., 2000
). However, not all the studies confirm its efficacy compared to placebo (Chick et al., 2000a
). This compound modulates the GABA-ergic transmission and decreases postsynaptic potentials in the neocortex, possibly via its action on NMDA (N-methyl-D-aspartate) receptors. Hypotheses have been drawn up concerning its actions on calcium channels as well as on the NMDA receptors reducing conditioned alcohol-withdrawal craving (Littleton, 1995
).
The aim of this study was to demonstrate the efficacy and treatment compliance of naltrexone compared to acamprosate in typical treatment conditions for these patients. An open randomized trial has been chosen for two reasons: (1) this is the experimental situation most similar to daily clinical practice; (2) if a double-blind trial had been carried out, both drugs would have to be administered in three doses per day (because of the pharmacokinetics of acamprosate and manufacturer's recommendations). However, taking into account the resistance to treatment compliance in these patients, especially in the medium and long-term, a double-blind trial in which the medication was administered three times a day would place naltrexone at a disadvantage since this drug is usually given in a single daily dose.
![]() |
PATIENTS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The participants were alcohol-dependent males who had requested detoxification in the Addictive Behaviour Unit of Doce de Octubre Hospital. Inclusion criteria were as follows: (1) male gender aged between 18 and 65 years; (2) meeting DSM-III-R criteria for alcohol-dependence (American Psychiatric Association, 1987); (3) having a stable family environment so that the family can help with treatment compliance and provide information during follow-up visits. Exclusion criteria were: (1) presence of another substance use disorder (with the exception of nicotine); (2) presence of another psychiatric disorder diagnosed by SCID for DSM-III-R (SCID); (3) a medical condition which could hinder treatment compliance; (4) impaired liver function [an aspartate aminotransferase (AST) or alanine aminotransferase (ALT) value more than three times normal values]; (5) previous treatment with naltrexone or acamprosate.
After completing detoxification, in the hospital or as an out-patient, the subjects were informed about the study objectives. They were informed about the two pharmacological treatments, naltrexone or acamprosate, elective treatments at the time of the study for the treatment of alcohol-dependence, but were told that the drug they would receive would be chosen at random. They would know which drug they would receive. They were told that relapse, or not taking the prescribed treatment punctually, would not lead to their being asked to leave the trial. However, they would be taken out of the trial if they did not keep in touch with the investigators for more than 15 days (i.e. two consecutive visits). They were also told that they could choose to leave the study at any time.
Procedure and assessments
After signing the informed consent, participants were assessed with the following instruments: a structured clinical interview for DSM-III-R (SCID) (Spitzer et al., 1992); the Addiction Severity Index (ASI) (McLellan et al., 1980
), Severity of Alcohol Dependence Scale (SADS) (Rubio et al., 1998
); three analogue scales to measure craving (frequency, duration and intensity) (Anton et al., 1999
); and a weekly calendar in which participants recorded all alcohol consumed, so that the time-line follow-back method could be used to document the pattern of consumption during follow-up (Miller, 1996
). The following baseline biological parameters were determined: serum aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyltransferase (GGT), bilirubin, and carbohydrate-deficient transferrin (CDT).
After randomizing the patients (using a random numbers table), patients received either one tablet (50 mg) per day of naltrexone, or six tablets (or five if of lower body weight) of acamprosate (i.e. 16651998 mg/day) divided into three doses following the manufacturer's recommendation. Patients visited their psychiatrists every 7 days (± 3 days) over the first 3 months, after which they visited every 15 days, till the end of the study. In the event of relapse, the frequency of visits was increased in order to help curtail the relapse and to offer the patient assistance if required. At each visit, entries in the diary of alcohol consumption were checked, together with craving, and whether the patient continued the treatment. Consumption and compliance data were compared with information given by the family.
Both groups of patients were offered supportive group therapy, once weekly over the entire study period. The groups were open groups. Therapy was less structured than in classical relapse prevention programmes. Basic relapse prevention was tackled (dealing with situations of risk, craving and negative emotional states). Abstinence was positively reinforced. Patients also received symptom-directed pharmacological treatment for complaints, such as anxiety, depression, insomnia, etc., when these symptoms presented during follow-up. If anxiety or depression emerged, sertraline could be prescribed (100200 mg/day), and for insomnia patients were given hydroxyzine, an H1 receptor antagonist of the piperazine family used as a hypnotic (50100 mg/night). In cases of relapses which were difficult to control pharmacologically or psychotherapeutically, disulfiram was added to the treatment until the relapse was fully over (23 weeks).
The blind investigators
Study data on outcome were collected by investigators (at 3, 6 and 12 months) who were blind to the drug taken by the patients. They used the following sources of data: (1) the patient himself, who was asked not to talk about the type of medication he was receiving; (2) the psychiatrist appointed to the case, who provided any data required from the clinical records, including biochemical results, and who was requested not to divulge the treatment prescribed; (3) the patient's family who provided information about drinking and any attempts by the patient to cease the pharmacological treatment. The degree of concordance between data from the family and the psychiatrists increased from 80% in the first few months to 95% in the final 3 months.
It was hoped that asking the family would help reduce the bias, which could occur if the information were obtained only from the psychiatrist who had prescribed the treatment. The investigators never interviewed the same patient at the three time points, since, at the end of an interview, they could have knowledge of the type of treatment the patient was receiving, which could affect future interviews with the same patient. Patients and relatives were asked not to tell the investigator the name of the treatment they were taking, its appearance, or how often per day they were taking it. Information from the psychiatrist was to complement that obtained from patients and their families and consisted mainly of data from clinical records and results of analyses. The main role of the psychiatrists in the study was to encourage patients to take the medication and to attend psychotherapy sessions.
Outcome measures
The primary outcome variables were: days of accumulated abstinence and days to first relapse (relapse is defined as the consumption of more than five drinks or 40 g ethanol per day). Additional outcome variables were number of drinks consumed per week, number of drinks consumed at a time, craving, abandonment of pharmacological treatment, drop-out from the study and 3-monthly serum GGT.
Statistical analysis
Pairwise 2- and t-tests were used to analyse differences between the two therapeutic groups, naltrexone versus acamprosate. All outcome analyses were conducted under an intention-to-treat analysis plan, with drop-outs regarded as relapsed for the abstinence and relapse analyses. Time to relapse and time to first drink were analysed by KaplanMeier survival analysis. The difference in variables, such as number of drinks consumed per day, drinks consumed at one time or percentage of days abstinent, were analysed by analysis of covariance (ANCOVA), taking baseline levels as covariants, and for drop-outs using the last observation carried forward. The biological drinking markers, CDT and GGT levels, were evaluated by both repeated measures and end-point ANCOVA with baseline levels as covariants. A composite craving severity score was created, as the average of the three scale scores (intensity, duration and frequency). Group differences were analysed by repeated measures ANCOVA with baseline values on the respective scales used as covariants.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
A total of 26 patients dropped out during the study (eight naltrexone, 18 acamprosate). In the naltrexone group, two patients dropped out in the 1st month, four in the 3rd month and two in the 4th month. In the acamprosate group, two dropped out in the 1st month, five in the 2nd, five in the 3rd, four in the 4th, one in the 7th and one in the 8th month. The reasons for drop-out are shown Fig. 1.
Efficacy
At the end of the treatment year the number of abstinent patients in the naltrexone group was twice that in the acamprosate group and the accumulated abstinence was significantly greater in the former (Table 2). The survival until the first relapse was longer for naltrexone than acamprosate patients (P = 0.02) (Fig. 2
). At the end of the study, 41% of the naltrexone group had not relapsed and 54% were abstinent since the last assessment (6 months), compared to 17 and 27%, respectively, in the group treated with acamprosate. Table 2
shows further alcohol consumption data, including the drinks consumed in a session, which was less for patients receiving naltrexone than those receiving acamprosate. In the group treated with naltrexone fewer patients used disulfiram. If a patient drank some alcohol, relapse occurred on average 12 days later in the naltrexone group (SD = 16) whereas it occurred in the group treated with acamprosate after 6 days (SD = 8).
|
|
Regarding the composite score, severity of craving, patients receiving naltrexone had significantly lower scores over the entire study period.
Treatment compliance
In the naltrexone group there was a trend towards fewer drop-outs, fewer attempts to abandon pharmacological treatment, more weeks of completed treatment, and greater attendance at psychotherapy support sessions. The latter reached statistical significance. We considered the hypothesis that the number of days of abstinence could be related to attendance at therapy sessions, rather than to the use of naltrexone or acamprosate. To test this, we compared the mean number of days of abstinence at 3, 6 and 12 months follow-up, taking the number of psychotherapy sessions as the covariant (ANCOVA). The results showed that, in the naltrexone group the mean number of days of abstinence remained constant after the 3rd month, whereas in the acamprosate group the mean number of days of abstinence decreased over the follow-up period (F = 8.23, df = 2, 248, P > 0.05).
Sertraline was prescribed for two patients in whom a depressive episode emerged, and hydroxyzine was prescribed to 16 patients because of inability to fall asleep. The distribution between treatment groups was even, although this was not the case with prescriptions for disulfiram, which was prescribed to significantly more patients in the acamprosate group than the naltrexone group (Table 2).
The GGT determinations done at 3, 6 and 12 months were compared with baseline levels and ANCOVA showed significant temporal improvements in the whole sample (F = 52.3, df = 2, P < 0.0001). Table 3 shows the number of days of heavy drinking and the mean values of GGT. There was a non-significant trend for greater improvement in GGT in the naltrexone patients but a significant reduction in percentage of days of heavy drinking.
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
It is difficult to compare our results with those of other studies, since ours is the first published comparative study of these two drugs. With regards to other research on naltrexone, in previous studies abstinence rates after 6 weeks were 2362% (O'Malley et al., 1992; Volpicelli et al., 1992
, 1997
; Anton et al., 1999
; Chick et al., 2000b
). The results of our study, which was four times longer than the aforementioned ones, are within this range. The levels of abstinence with acamprosate in placebo-controlled trials with a 1-year follow-up are between 18 and 35% (Paille et al., 1995
; Sass et al., 1996
; Whitworth et al., 1996
; Besson et al., 1998
). In our study, we recorded a rate of 17%. If we extrapolate these results, it seems that long-term treatment of patients with naltrexone is more beneficial than with acamprosate.
Two hypotheses could explain the benefits of naltrexone seen in our study. First, it may be that naltrexone increases the period elapsed before the subject takes the first drink. Prolonging the abstinence period enables the learning of strategies taught in the support therapy and increases feelings of self-efficacy. Second, it may be that naltrexone has an effect on control of alcohol consumption once this has already begun, resulting in a delay in relapse. This could also increase the patients' faith in the treatment.
The first of these hypotheses was not confirmed, because the survival time to the first drink did not differentiate between treatments. In contrast, the action of naltrexone on control of alcohol consumption is shown in the survival curve to first relapse and in the number of drinks consumed at any one time. This effect has been described in other studies and has been explained by a reduction in the reinforcing effects of ethanol after drinking (O'Malley et al., 1996a,b
), or by an improvement in the ability to resist thoughts or cravings to continue drinking (Anton et al., 1999
). Whichever explanation, the higher degree of control over their drinking achieved by patients treated with naltrexone could explain their lesser use of disulfiram and their achieving more days of abstinence and a greater use of therapy. In our opinion, this effect could be explained as follows: the craving triggered by consumption is slightly less with naltrexone than with acamprosate, which enables those treated with naltrexone to stop drinking earlier. Since relapses are very common in these patients, those treated with naltrexone would be more capable of interrupting the relapse or diminishing its intensity. This would help to prevent progression in alcohol consumption and increase the probability that the patient seeks help from a therapist and, therefore, ultimately, curtail relapse. This is supported by the fewer absences from therapy in the naltrexone-treated group. Since naltrexone reduces the intensity of relapse, patients attend more therapy sessions. Although this latter effect has not been found by other authors (Anton et al., 1999
), this could be due to the shorter duration of their studies. Finally, the increased number of attempts to abandon treatment with acamprosate may relate to the number of doses required daily, and this could contribute to the smaller percentage of days of abstinence achieved by these patients.
Anticraving effects of naltrexone were more important than those of acamprosate, although this difference could be due to their different mechanisms of action and the fact that most patients drank alcohol during the study period. Given that acamprosate probably exerts its anticraving action by reducing the intensity of the symptoms of the conditioned withdrawal syndrome and naltrexone probably reduces the reinforcing effects of the alcohol, this difference would favour the use of naltrexone in patients who are likely to consume some alcohol. This would explain why the patients treated with naltrexone reported less craving than the acamprosate group over the study period (Rubio et al., 1999). It is also possible that naltrexone would be more effective at reducing craving in patients with moderate dependence, in whom craving mechanisms related to positive reinforcement could be over-represented. Since our sample was of patients with moderate dependence, this could explain the results obtained.
With regards to the tolerability of both drugs, although the group treated with naltrexone experienced more side-effects, these only lasted for the first 2 weeks of the study and there was no significant difference in the rate of drop-out due to this.
Limitations of this study
This was an open study, and there is the possibility that the investigators did not remain blinded. We tried to prevent the investigators from gaining direct information about the type of pharmacological treatment taken by the patients, although it is possible that they could have guessed the treatment from the patients' side-effects. However, this can also occur in double-blind trials, except in studies with total integrity of the double-blindedness (Moncrieff and Drummond, 1997). Objective outcome criteria are not subject to bias: in our study, GGT (which is a helpful, but not perfect, marker of drinking) appeared to corroborate a better reported outcome in the naltrexone group, but the advantage failed to reach statistical significance.
Some of the advantages of naltrexone seen in this study could be explained by the fact that the participants were patients with moderate alcohol dependence. Impaired liver function as an exclusion criterion will have ruled out some of the most severe cases. Possibly, the latter would have responded better to acamprosate than to naltrexone.
At the start of the study, the psychiatrists did not know which pharmacological treatment would be most effective and, therefore, had a similar attitude towards encouraging compliance with both treatments. However, as the study progressed and subjects treated with naltrexone appeared to have a better outcome, the psychiatrist may have made more effort to encourage compliance with naltrexone treatment, which could, at least hypothetically, have then introduced a bias.
Our assessment of the degree of compliance to the pharmacological treatment was conducted by questionnaires corroborated by information from the family. It would have been more accurate to use a urinary marker such as riboflavin.
A difficulty in extrapolating the results of this study to other treatment settings could be that, in our study, there was a high level of family support available to patients. If this had not been available, the retention levels, and compliance with medication, might have been lower for both treatments, and there would possibly have been no measurable difference between them. In our opinion, further studies comparing the efficacy of these two drugs are required in varying therapeutic contexts in patients with different severity profiles.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
FOOTNOTES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Anton, R. F., Moak, D. H., Waid, L. R., Latham, P. K., Malcolm, R. J. and Dias, J. K. (1999) Naltrexone and cognitive behavioral therapy for the treatment of outpatient alcoholics: results of a placebo-controlled trial. American Journal of Psychiatry 156, 17581764.
Benjamin, D., Grant, E. R. and Phorecky, L. (1993) Naltrexone reverses ethanol-induced dopamine release in the nucleus accumbens in awake, freely moving rats. Brain Research 621, 137140.[ISI][Medline]
Besson, J., Aeby, F., Kasas, A., Lehert, P. and Potgieter, A. (1998) Combined efficacy of acamprosate and disulfiram in the treatment of alcoholism: a controlled study. Alcoholism: Clinical and Experimental Research 22, 573579.[ISI][Medline]
Catafau, A. M., Etcheberrigaray, A., Pérez de los Cobos, J., Estorch, M., Guardia, J. and Flotats, A. (1999) Regional cerebral blood flow changes in chronic alcoholic patients induced by naltrexone challenge during detoxification. Journal of Nuclear Medicine 40, 1924.[Abstract]
Chick, J., Howlett, H., Morgan, M. Y., Ritson, B. and UKMAS investigators (2000a) United Kingdom Multicentre Acamprosate Study (UKMAS): a 6-month prospective study of acamprosate versus placebo in preventing relapse after withdrawal from alcohol. Alcohol and Alcoholism 35, 176187.
Chick, J., Anton, R., Checinski, K., Croop, R., Drummond, C., Farmer, R., Labriola, D., Marshall, J., Moncrieff, J., Morgan, M., Peters, T. and Ritson, B. (2000b) A multicentre, randomized, double-blind, placebo-controlled trial of naltrexone in the treatment of alcohol dependence or abuse. Alcohol and Alcoholism 35, 587593.
Geerlings, P. J., Ansoms, C. and van den Brink, W. (1997) Acamprosate and prevention of relapse in alcoholics. European Journal of Addiction Research 3, 129137.
Ladewig, D., Knecht, Th., Lehert, P. H. and Fend, A. (1993) Acamprosat ein stabilisierungsfktor in der langzeitentwohnung von alkoholabhangigen. Therapeutische Umschau 50, 182187.[Medline]
Littleton, J. (1995) Acamprosate in alcohol dependence: how does it work? Addiction 90, 11791188.[ISI][Medline]
McLellan, A. T., Luborsky, L., Woody, G. E. and O'Brien, C. P. (1980) An improved diagnostic evaluation instrument for substance abuse patients: the Addiction Severity Index. Journal of Nervous and Mental Disease 168, 2633.[ISI][Medline]
Miller, W. R. (1996) Form 90: A Structured Assessment Interview for Drinking and Related Behaviors Test Manual : NIAAA Project Match Monograph Series, vol. 5, Mattson, M. E. and Marshall, L. A. eds, pp. 1422. National Institute on Alcoholism and Alcohol Abuse, Rockville, MD.
Moncrieff, J. and Drummond, D. C. (1997) New drug treatments for alcohol problems: a critical appraisal. Addiction 92, 939947.[ISI][Medline]
O'Malley, S. S., Jaffe, A. J., Chang, G., Schottenfeld, R S., Meyer, R. E. and Rounsaville, B. (1992) Naltrexone and coping skills therapy for alcohol dependence. Archives of General Psychiatry 49, 881887.[Abstract]
O'Malley, S. S., Jaffe, A. J., Rode, S. and Rounsaville, B. (1996a) Experience of a slip among alcoholics treated with naltrexone or placebo. American Journal of Psychiatry 153, 281283.[Abstract]
O'Malley, S. S., Jaffe, A. J., Chang, G., Rode, S., Schottenfeld, R. S., Meyer, R. E. and Rounsaville, B. (1996b) Six-month follow-up of naltrexone and psychotherapy for alcohol dependence. Archives of General Psychiatry 53, 217224.[Abstract]
Paille, F. M., Guelfi, J. D., Perkins, A. C., Royer, R. J., Steru, L. and Parot, P. (1995) Randomised multicentre trial of acamprosate in a maintenance programme of abstinence after alcohol detoxification. Alcohol and Alcoholism 30, 239247.[Abstract]
Pelc, Y., Le Bon, O., Verbanck, P., Lehert, P. H. and Opsomer, L. (1992) Calcium acetyl homotaurinate for maintaining abstinence in weaned alcoholic patients; a placebo controlled double-blind multicentre study. In Novel Pharmacological Interventions for Alcoholism, Naranjo, C. and Sellers, E. M. (eds), pp. 348352. Springer-Verlag, New York.
Poldrugo, F. (1997) Acamprosate treatment in a long-term community based alcohol rehabilitation programme. Addiction 92, 15371547.[ISI][Medline]
Rubio, G., Urosa, B. and Santo-Domingo, J. (1998) Validación de la escala de la intensidad de la dependencia alcohólica (EIDA). Psiquiatría Biológica 10 (Suppl. 1), 4447.
Rubio, G., Ponce, G., Jiménez-Arriero, M. A. and Santo-Domingo, J. (1999) La pérdida de control en la dependencia alcohólica: conceptualización. Adicciones 11, 143158.
Sass, H., Soyka, M., Mann, K. and Zieglgansberger, W. (1996) Relapse prevention by acamprosate: results from a placebo controlled study on alcohol dependence. Archives of General Psychiatry 53, 673680.[Abstract]
Spitzer, R. L., Williams, J. B. W., Gibbon, M. and First, M. B. (1992) The Structures Clinical Interview for DSM-III-R (SCID), I: History, rationale, and description. Archives of General Psychiatry 49, 624629.[Abstract]
Tempesta, E., Janiri, L., Bignamini, A., Chabac, S. and Potgieter, A. (2000) Acamprosate and relapse prevention in the treatment of alcohol dependence: a placebo-controlled study. Alcohol and Alcoholism 35, 202209.
Ulm, R. R., Volpicelli, J. R. and Volpicelli, L. A. (1995) Opiates and alcohol self-administration in animals. Journal of Clinical Psychiatry 56 (Suppl. 7), 514.[ISI][Medline]
Valenzuela, F. and Harris, A. (1997) Alcohol: neurobiology. In Substance Abuse. A Comprehensive Textbook, 3rd edn, Lowinson, J. H., Ruiz, P., Millman, R. B., Langrod, J. G. eds, pp. 119142. Williams and Wilkins, Baltimore.
Volpicelli, J. R., Alterman, A. I., Hayashida, M. and O'Brien, C. P. (1992) Naltrexone in the treatment of alcohol dependence. Archives of General Psychiatry 49, 876880.[Abstract]
Volpicelli, J. P., Clay, K. L., Watson, N. T. and O'Brien, C. P. (1995a) Naltrexone in the treatment of alcoholism: predicting response to naltrexone. Journal of Clinical Psychiatry 56, 3944.[ISI][Medline]
Volpicelli, J. R., Watson, N. T., King, A. C., Sherman, C. E. and O'Brien, C. P. (1995b) Effect of naltrexone on alcohol high in alcoholics. American Journal of Psychiatry 152, 613615.[Abstract]
Volpicelli, J. R., Rhines, K. C., Rhines, J. S., Volpicelli, L. A., Alterman, A. I. and O'Brien, C. P. (1997) Naltrexone and alcohol dependence. Role of subject compliance. Archives of General Psychiatry 54, 737742.[Abstract]
Whitworth, A. B., Fisher, F., Lesch, O., Nimmerrichter, A., Oberauer, H., Platz, T., Potgieter, A., Walter, H. and Fleischhacker, W. W. (1996) Comparison of acamprosate and placebo in long-term treatment of alcohol dependence. Lancet 347, 14381442.[ISI][Medline]