INTRODUCTION
Alcohol dependence is a cluster of physiological, behavioral, and cognitive phenomenon in which the use of alcohol takes on a much higher priority for a given individual than other behaviors that once had greater value. A central descriptive characteristic of the dependence syndrome is the desire (often strong, sometimes overpowering) to take alcohol. There may be evidence that return to substance use after a period of abstinence leads to a more rapid reappearance of other features of the syndrome that occurs with nondependent individuals.[1
Most of the alcohol-dependent patients require an initial period of detoxification. Benzodiazepines are commonly and very successfully used in the management of the symptoms of acute withdrawal, along with other adjunctive options such as beta-blockers, carbamazepine, clonidine, and antipsychotics occasionally.[2 3
Naltrexone is a competitive nonspecific opioid antagonist with highest affinity for the μ-receptors. In addition to its role in opioid dependence, it was also found to be useful in the treatment of alcohol dependence and was finally approved by the Food and Drug Administration for the same in 1994. Naltrexone, through its opioid receptor antagonism, is thought to help in attenuating the reinforcing effects of alcohol. As a result, alcohol consumption while taking naltrexone leads to lesser intoxicating effects of alcohol and therefore a decrease in desire to drink more. This ultimately leads to decreased craving and control over the drinking habits.[4
Acamprosate has been a proven safe and well-tolerated treatment modality for alcohol dependence since its introduction in Europe in 1989. It is a highly flexible molecule with structural similarities to several amino acid neurotransmitters and neuromodulators such as gamma-amino butyric acid (GABA), glutamate, aspartate, glycine, and taurine. Studies have shown that acamprosate tends to promote abstinence by counteracting or diminishing the physiological response to chronic alcohol exposure.[5
Baclofen is a prototypic GABA-B receptor agonist. This serves as the possible mechanism by which baclofen suppresses alcohol-induced dopamine release and the consequent dopamine-mediated alcoholism reinforcement and other associated alcoholism-related motivated behaviors. In addition, in relation to the suggested efficacy of baclofen in the cases of alcohol withdrawal syndrome (AWS), studies suggest that activation of GABA-B receptors by baclofen is effective in counterbalancing AWS-associated and N-methyl-D-aspartate-mediated glutamate excitatory neurotransmission.[6
Although the pharmacological action of these three drugs is evidently very different and unique in its own, a detailed knowledge about their relative efficacies can facilitate in developing a relapse prevention strategy that would give rise to greater personal and socioeconomic benefits. In this pursuit, several studies have been carried out to compare them in terms of their efficacy and safety. However, there has been a real scarcity of studies aimed at head on comparison of all these three compounds together.
MATERIALS AND METHODS
Study design
The present study was a prospective study carried out at a government-run tertiary care de-addiction center, in which the safety and efficacy of baclofen, acamprosate, and naltrexone in the management of alcohol dependence was assessed. The sample size was calculated using the following formula:[7
n ij = 2 (z α /(2τ ) + z β )2 σ2 /ε2 ij
Ethical clearance was taken from the ethical committee of the institute vide letter no. BFUHS/2K17p-TH/207. Because this study was conducted as a part of thesis by a postgraduate candidate, it was done under the supervision and directions of the ethical committee which did not direct us to get it registered in any clinical trial. The study was started in the year 2016. Because the CTRI only registers clinical trials prospectively w. e. f April 1, 2018, our study could not be registered retrospectively. In light of the above, kindly consider exemption to the mandatory registration under CRTI. We can take it as a limitation of the study if desired.
The sample size was calculated as 30 for each drug under study (including the dropouts). Randomization of the participants was done using a computerized random table. Individuals giving written informed consent and meeting the International Statistical Classification of Diseases-10 (ICD-10) criteria for alcohol dependence were included in the study. However, individuals with comorbid medical or psychiatric illness that could have interfered with the evaluation of the drugs under study were excluded.
Instruments
Psychiatric pro forma
Psychiatric Interview Performa made by the department of psychiatry was applied to document the history, duration of illness, and course of illness of the patients. It included an informed consent, sociodemographic profile, history, and mental state examination of the patients.
Obsessive Compulsive Drinking Scale
Obsessive Compulsive Drinking Scale (OCDS) is a self-reporting questionnaire including 14 questions to be answered by the participant for assessing the obsession and compulsivity related to craving and drinking behavior by using obsessive and compulsive subscales for scoring.[8
Advance Warning of Relapse questionnaire 3.0
Advance Warning of Relapse (AWARE) questionnaire 3.0 was applied to evaluate the risk for relapse. Higher score indicated more warning signs of relapse.[9
Hogan Drug Attitude Inventory
The Hogan Drug Attitude Inventory was applied to study the attitude of individuals toward medication adherence. A positive total score meant a compliant response, whereas a negative total score meant a noncompliant response.[10
An informed written consent for participating in the study was obtained from the patients presenting for treatment in the outpatient department (OPD) of the de-addiction center and meeting ICD-10 criteria for alcohol dependence. The participants were screened for various substances using free kits available through government supply. The participants taking only alcohol were selected. A face-to-face interview of the participants using the Psychiatric Performa was conducted by the postgraduate students pursuing MD Psychiatry, followed by their consultant in-charge. The participants were admitted for detoxification in the de-addiction ward. After detoxification, the treatment under study was initiated among the participants after assessing them for craving, relapse risk, and attitude toward treatment using OCDS and AWARE questionnaire 3.0. Group 1 included participants who were prescribed baclofen, Group 2 included participants who were prescribed acamprosate, and Group 3 included participants those who were prescribed naltrexone.
Naltrexone was available free of cost under government supply and was started at a dose of 50 mg/day. However, owing to the frequent side effects reported during the initial days of administration, the dose had to be reduced to 25 mg and then built up gradually to 50 mg/day after making sure that the patient was tolerating the medication well till the end of the study period. Because the mean weight of the participants in our study was 58.67 kg, the mean dose of acamprosate administered was 1021.20 mg/day. The starting dose was 333 mg thrice a day which was gradually titrated up to 4–6 tablets per day. Acamprosate had to be bought by the patients themselves as it was not available through government supply, unlike naltrexone. The average estimated cost incurred by the patient was around Rs. 1000 per month as acamprosate costs around Rs. 8 per tablet. Baclofen was administered at an average dose of 38.11 mg/day starting with 20 mg tablet once a day. The dose was increased to twice a day after a week after ensuring that the medication was well tolerated. Baclofen also had to be bought by the patients themselves owing to the above-mentioned reason. The average cost incurred by the patients taking baclofen was around Rs 600 per month as each tablet costs around Rs. 10.
During the weekly follow-up visits in the 1st month, and bi-weekly follow-up visits for the next 5 months, the participants were assessed for craving, relapse risk, and medication adherence using the respective scales and questionnaires. Relapse was defined as taking four drinks for women or five drinks for men on an occasion or a single day. In addition, the side effects, if any, during the ongoing treatment were documented through interview and self-reporting, and the necessary modifications in medication were done along with other precautionary measures. Routine investigations (complete blood count, blood urea, serum creatinine, and electrolytes) and biological markers for alcohol abuse (liver function tests) were tested at the time of admission and later at the end of the study or as desired by the clinical status of the participant. The data pertaining to clinical details were entered in the form of data matrix in Microsoft® Excel® and analyzed using IBM® SPSS® v 20.0.0 (Armonk, NY: IBM Corp.).
RESULTS
Sociodemographic profile
Majority of the patients belonged to the age group of 30–40 years in all the groups, with 11 patients each in group [Table 1 ]. The mean age of the patients in Groups 1, 2, and 3 was 42.73 ± 10.62, 39.67 ± 9.88, and 38.47 ± 10.86 years, respectively. Majority of the patients belonged to Sikh religion (>75% in each group). Majority of the patients were married (>85% in every group). Out of the thirty patients in each group, there were 14, 16, and 18 patients from urban background in Groups 1, 2, and 3, respectively. Majority of the patients belonged to the category of earning a monthly income of Rs. 6327–18,949, with 18, 19, and 19 patients in the three groups, respectively. In terms of occupation, two (6.66%) patients in Group 1 and one (3.33%) patient in Group 2 were unemployed. Comparing the marital status, 27 (90%) patients each in Groups 1 and 2 and 26 (86.66%) patients in Group 3 were married. Taking into consideration the education of the enrolled patients, it was seen that majority of the patients had studied till high school. The number of patients with this level of education was 7 (23.33%), 14 (46.66%), and 5 (16.66%) in Groups 1, 2, and 3, respectively. The number of patients with primary education was three (10%), three (10%), and six (20%) in Groups 1, 2, and 3, respectively. Eight (26.66%), two (6.66%), and four (13.33%) patients, respectively, from the three groups had studied till middle school. The number of patients who had studied till higher secondary level was four (13.33%), five (16.66%), and five (16.66%) in Groups 1, 2, and 3, respectively. The number of graduates was three (10%), four (13.33%), and six (20%) in Groups 1, 2, and 3, respectively. One (3.33%) patient in each group was postgraduate.
Table 1: Distribution of sociodemographic characteristics
Considering the family history of substance intake among the test participants, majority of them had a positive history. Nearly 70% of the patients in the baclofen group, 90% in the acamprosate group, and 86.66% in the naltrexone group had a positive family history [Table 2 ]. The mean duration of alcohol intake in Groups 1, 2, and 3 was 18.58 ± 10.62, 15.43 ± 8.76, and 15.75 ± 10 years, respectively. Almost 80% of the patients in the baclofen group, 60% in the acamprosate group, and 70% in the naltrexone group had no history of any attempts of de-addiction in the past.
Table 2: Other relevant baseline characteristics
Treatment outcomes
Table 3 depicts the treatment outcomes across various parameters assessed from the baseline till the end of the study period (6 months), which was analyzed using the independent sample Kruskal–Wallis test. The mean decline in the score by the end of 6-month study period was 23.48 ± 10.07, 20.68 ± 12.37, and 26.72 ± 13.05, respectively, across the baclofen, acamprosate, and naltrexone groups. The decline in AWARE score seen in naltrexone group (64.72 ± 45.65) was maximum, followed by baclofen (57.15 ± 33.10) and acamprosate (48.79 ± 37.00).
Table 3: Changes in scoring of various scales from baseline to end point (6 months)
Figure 1a depicts the pattern of OCDS score decline across the three groups. The OCDS score decline was maximum in the first 2 months over all the three groups. However, the decline was visibly higher in the group taking naltrexone in months 2nd , 3rd , and 4th . In the 1st , 5th , and 6th months, the response with baclofen was more. Acamprosate group showed more decline in OCDS score as compared to baclofen group in the 3rd and 4th months, but it was still less as compared to naltrexone. These were analyzed using the related-samples Friedman's two-way analysis of variance.
Figure 1: (a) Distribution of month-wise Obsessive Compulsive Drinking Scale score decline. (b) Distribution of month-wise Advance Warning of Relapse score decline. (c) Distribution of month-wise Hogan Score decline
Figure 1b shows that the AWARE score decline was maximum in the first 2 months over all the three groups. However, in the 1st and 4th months, acamprosate showed maximum decline. Naltrexone group showed more decline than the other two groups in the 2nd and 3rd months. In the last 2 months, decline seen was maximum in the group taking baclofen. These were analyzed using the related-samples Friedman's two-way analysis of variance.
An overall positive attitude of the patients toward the three drugs was evident from Hogan Drug Attitude Inventory scores, as shown in Figure 1c . Improvement in this positive attitude was maximum in the naltrexone group during the 1st month. Baclofen showed maximum improvement among all the groups in the 2nd , 5th , and 6th months. Acamprosate showed more improvement than the other groups in the 3rd and 4th months. These were analyzed using the related-samples Friedman's two-way analysis of variance.
The distribution of side effects reported or observed during the course of study is summarized in Table 4 . The number of patients reporting side effects was 5, 8, and 12, respectively, in the Groups 1, 2, and 3.
Table 4: Distribution of side effects
Table 5 represents the distribution of dropouts across various groups. The number of patients dropping out during the course of the study was five, nine, and five, respectively in Groups 1, 2, and 3.
Table 5: Distribution of dropouts
Fifteen patients from the baclofen group, 12 patients from the acamprosate group, and six patients from the naltrexone group relapsed during the course of study [Figure 2 ].
Figure 2: Distribution of relapse
DISCUSSION
In our study, taking into account the sociodemographic variables of the three groups, it was seen that the mean age of the patients in Groups 1, 2, and 3 was 42.73 ± 10.62, 39.67 ± 9.88, and 38.47 ± 10.86 years, respectively. In a similar study, the mean age of the group prescribed naltrexone was 43.18 ± 8.3 years.[11 12
All the enrolled patients in the baclofen and naltrexone groups were males, while there was one female patient (3.33%) in the acamprosate group. Shukla et al . in their study had similar findings as all the patients enrolled in the study were males.[13 et al . in their study had 23.6% of females in the acamprosate group, while there were 28.3% of females in the naltrexone group.[14
In terms of occupation, two (6.66%) patients in Group 1 and one (3.33%) patient in Group 2 were unemployed. This was in contrast to the study conducted by Mishra et al ., in which 40% of the patients enrolled in the Baclofen group and 28% of the patients in the acamprosate group were unemployed.[12 et al . in their study found 25% of patients to be unemployed in naltrexone as well as acamprosate groups.[15
Comparing marital status, 27 (90%) patients each in Groups 1 and 2 and 26 (86.66%) patients in Group 3 were married. Rubio et al ., in their study, had similar findings as 95% of the patients in the naltrexone group and 92% in the acamprosate group were married.[15 et al . in their study had 64% of married patients in the baclofen group and 68% of married patients in the acamprosate group.[12
Considering the family history of substance intake among the test patients, majority of them had a positive history. This was higher as compared to the findings in the study conducted by Morley et al .[14 et al ., 51.2% of the patients enrolled in the baclofen group had a positive family history, which was lesser than the finding in our study.[16
The mean duration of alcohol intake in Groups 1, 2, and 3 was 18.58 ± 10.62, 15.43 ± 8.76, and 15.75 ± 10 years, respectively. Similar findings were seen in the study done by Morley et al ., in which the mean age of dependence in the acamprosate group was 13.7 ± 9.4 years, while it was 18 ± 11.3 years for the naltrexone group.[14 et al . which involved baclofen had 23.5 ± 9.9 years as the mean duration of alcohol abuse in the baclofen group, which was more than what was seen in our study.[17
Considering the number of previous de-addiction attempts, 80% of the patients in the baclofen group, 60% in the acamprosate group, and 70% in the naltrexone group had no history of any attempts of de-addiction in the past.
Dosage
In the study done by Chick et al ., the dose of naltrexone administered was 50 mg/day,[11
In the study done by Maisel et al . and Baltieri and de Andrade, acamprosate was administered at doses of 1988 mg/day and 1332 mg/day for patients with weight >60 kg and <60 kg, respectively.[18 19 et al . who had administered 30 mg/day of baclofen to the test patients.[17 et al . and Mishra et al ., the dosage of baclofen administered was 15 mg/day for the first 3 days, followed by 30 mg/day thereafter.[12 20
Treatment outcomes
The mean decline in the OCDS score by the end of 6-month study period was 23.48 ± 10.07, 20.68 ± 12.37, and 26.72 ± 13.05, in Groups 1, 2, and 3, respectively. This shows that the response in terms of OCDS score was maximum with naltrexone, followed by baclofen and acamprosate. The decline in score within each group was statistically significant (P < 0.00 in each group). However, the difference across the groups was not statistically significant (P = 0.191). Gupta et al . in their study reported a mean decline of 16.56 ± 10.5 in OCDS score in baclofen group after 12 weeks, which was lesser as compared to our study.[16 et al . in their study found a statistically significant (P < 0.05) reduction in OCDS score in the naltrexone group.[11 et al . observed a statistically significant decrease in OCDS score (P < 0.05), with the use of baclofen.[20 P < 0.05) in decline in OCDS score was seen in a study done by Mishra et al .[12 et al ., acamprosate had a statistically significant larger effect size as compared to naltrexone (P < 0.001) in terms of maintaining abstinence.[18
Considering the AWARE score, a statistically significant decline was seen in all the groups (P < 0.001, in all the three groups). This decline indicated that all the three drugs decreased the probability of relapse in alcohol-dependent patients. Naltrexone was most effective because the decline in AWARE score seen in the naltrexone group (64.72 ± 45.65) was maximum, followed by baclofen (57.15 ± 33.10) and acamprosate (48.79 ± 37.00). Although there was a clear difference in the decline in mean scores over all the three groups, the difference was not statistically significant (P = 0.411). The AWARE score decline was maximum in the first 2 months over all the three groups. This could be attributed to the concomitant use of adjunctive drugs such as benzodiazepines and beta-blockers, which were used initially to counter the withdrawal signs and symptoms and other related issues such as sleep disturbances in the beginning of treatment.
Patients' attitude toward the three test drugs did not show much difference within each group and even over all the groups during the study period, as was evident from Hogan Drug Attitude Inventory scores. The difference in the scores over all the groups did not show any statistical significance (P = 0.469). Our study showed an overall positive attitude of the patients toward the three drugs. A statistically significant improvement in attitude toward the treatment was seen in all groups during the 1st month. Improvement was maximum in the naltrexone group during the 1st month. Baclofen showed maximum improvement among all the groups in 2nd , 5th , and 6th months. Acamprosate showed more improvement than the other groups in the 3rd and 4th months. In the baclofen group, there was a declining trend in this improvement over the next 3 months. This could be attributed to the reasons such as lack of motivation, lack of need to continue medication, or other personal issues. The response picked up in the 5th month only to decline again in the last. In the acamprosate group, the response declined significantly in the 2nd month. This was probably because of the cost issues and multidose regimen associated with the use of acamprosate. However, an improvement was seen again in the 3rd , 4th , and 5th months. The response declined again in the last month. In the naltrexone group, there was a declining trend in the improvement in the 2nd and 3rd months probably due to the side effects which were frequently reported in this group, along with other reasons such as lack of motivation and personal issues. The response improved in the 4th month, only to decline again in the 5th and 6th months.
The change in liver function tests over all the three groups did not show any statistical significance, as was evident from the P values for total bilirubin, serum glutamic-oxaloacetic transaminase (SGOT), serum glutamic pyruvic transaminase (SGPT), and alkaline phosphatase (ALP) (P = 0.15, 0.18, 0.47, and 0.70, respectively). The mean total bilirubin showed a decline of 0.10 ± 0.50 in the baclofen group. However, in the naltrexone group, it increased by 0.02 ± 0.47 by the end of the study period. The levels increased by 0.22 ± 0.72 in the acamprosate group. The SGOT levels declined the most in the baclofen group (48.59 ± 67.54) followed by naltrexone (27.91 ± 68.84) and acamprosate (14.20 ± 40.92). The SGPT levels declined the most in the baclofen group (14.88 ± 29.29) followed by the naltrexone (14.79 ± 50.17) and acamprosate (6.67 ± 22.64) groups. The mean ALP levels also declined the most in the baclofen group (11.68 ± 27.03) followed by the naltrexone (6.25 ± 48.27) and acamprosate (0.79 ± 42.76) groups. Thus, baclofen was better tolerated, followed by naltrexone and acamprosate. In the study done by Chick et al ., similar to our findings, there was no significant discrimination in the reduction of SGOT and SGPT levels following treatment with naltrexone.[11 et al ., there was no statistically significant difference (P > 0.05) in the change among these biological markers.[15
Looking at the side effects, majority of them were reported by the patients prescribed naltrexone (40%), followed by acamprosate (26.66%) and baclofen (16.66%). Thus, baclofen showed best tolerability with least number of side effects reported. The most frequently noticed side effect was dizziness, noticed in five patients (16.66%) in the naltrexone group, two patients (6.66%) in the baclofen group, and one patient (3.33%) in the acamprosate group.
Similar to this finding, in the study conducted by Chick et al ., nausea was the most commonly reported side effect (32%) in naltrexone-taking group.[11 21 et al . involving the use of baclofen, 42.86% of the patients taking baclofen reported having headache, which was in stark contrast to our study where none of the patients put on baclofen reported of having headache, whereas 6.66% reported of having dizziness.[20
The percentage of dropouts in the baclofen, acamprosate, and naltrexone groups was 16.66%, 30%, and 16.66%, respectively. Thus, acamprosate was associated with maximum number of dropouts. Contrary to these findings, there were 58.29% dropouts in the naltrexone group in the study conducted by Chick et al .[11 et al ., the dropout rate (10.38%) in the naltrexone group was close to the one seen in our study.[15 et al . showed 15% dropout rate in the group taking baclofen.[22 19
The percentage of relapses in the baclofen, acamprosate, and naltrexone groups was 50%, 40%, and 20%, respectively. Thus, naltrexone proved best at relapse prevention with least number of relapses reported. In the study done by Gupta et al ., 36% relapse rate was reported among those who were prescribed baclofen, which was lesser than what was seen in our study.[16 et al . showed a much higher relapse rate of 55% and 83% in patients taking naltrexone and acamprosate, respectively.[15 et al . in their study reported 23% relapse rate among naltrexone-taking group, which was similar to the findings in our study.[23 et al ., relapse rate was 56% and 49% in patients put on acamprosate dose of 1332 and 1998 mg/day mg/day, respectively.[24
CONCLUSION
Looking at the findings of our study, it could be concluded that naltrexone was most effective in terms of decreasing craving as well as decreasing the risk of relapse. In terms of safety profile, baclofen fared better than the other two drugs. Naltrexone was associated with majority of side effects. Acamprosate was associated with majority of dropouts. All these factors should be weighed upon while opting for the anti-craving agent in the treatment of alcohol dependence.
Limitations
Our study was conducted for only 6-month duration. It should have been conducted for longer duration (minimum 1 year) in order to get more reliable data
The sample size was 30 in each group. If the same study would have been conducted on more number of patients, the probability of generalizing the results of study shall increase
Our study was not blinded
There was a lack of information regarding the exact number of drinking days or the number of heavy drinking days.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
1. ICD-10 Classification of Mental and Behavioural Disorder: Clinical Descriptions and Diagnostic Guidelines. 1992 Geneva World Health Organization
2. Mayo-Smith MF. Pharmacological management of alcohol withdrawal. A meta-analysis and evidence-based practice guideline. American Society of Addiction Medicine Working Group on Pharmacological Management of Alcohol Withdrawal JAMA. 1997;278:144–51
3. O'Malley SS, Jaffe AJ, Chang G, Schottenfeld RS, Meyer RE, Rounsaville B. Naltrexone and coping skills therapy for alcohol dependence. A controlled study Arch Gen Psychiatry. 1992;49:881–7
4. Myrick H, Anton RF, Li X, Henderson S, Randall PK, Voronin K. Effect of naltrexone and ondansetron on alcohol cue-induced activation of the ventral striatum in alcohol-dependent people Arch Gen Psychiatry. 2008;65:466–75
5. Popp RL, Lovinger DM. Interaction of acamprosate with ethanol and spermine on NMDA receptors in primary cultured neurons Eur J Pharmacol. 2000;394:221–31
6. Colombo G, Agabio R, Carai MA. Ability of baclofen in reducing alcohol intake and withdrawal severity: II–preclinical evidence Alcohol Clin Exp Res. 2000;24:58–66
7. Wang H, Chow SC. Wiley Encyclopedia of Clinical Trials: Sample Size Calculation for Comparing Time-To-Event Data 2007 New York John Wiley & Sons
8. Anton RF, Moak DH, Latham PK. The obsessive compulsive drinking scale: A new method of assessing outcome in alcoholism treatment studies Arch Gen Psychiatry. 1996;53:225–31
9. Miller WR, Harris RJ. A simple scale of Gorski's warning signs for relapse J Stud Alcohol. 2000;61:759–65
10. Hogan TP, Awad AG, Eastwood K. A self-report scale predictive of drug compliance in schizophrenics: Reliability and discriminative ability Psychol Med. 1983;13:177–83
11. Chick J, Anton R, Checinski K, Croop R, Drummond DC, Farmer R. A multicentre, randomized, double-blind, placebo-controlled trial of Naltrexone in the treatment of alcohol dependence or abuse Alcohol Alcoholism. 2000;35:587–93
12. Mishra SN, Swain SP, Shukla RK, Sarkar P. A study of comparative efficacy of Baclofen vs. acamprosate in reducing alcohol craving and abuse Indian Journal of Psychiatry. 2010;52:S69
13. Shukla L, Shukla T, Bokka S, Kandasamy A, Benegal V, Murthy P, et al Correlates of Baclofen Effectiveness in Alcohol Dependence Indian J Psychol Med. 2015;37:370–3
14. Morley KC, Teesson M, Reid SC, Sannibale C, Thomson C, Phung N, et al Naltrexone versus acamprosate in the treatment of alcohol dependence: A multi-centre, randomized, double-blind, placebo-controlled trial Addiction. 2006;101:1451–62
15. Rubio G, Jiménez-Arriero MA, Ponce G, Palomo T. Naltrexone versus acamprosate: One year follow-up of alcohol dependence treatment Alcohol Alcohol. 2001;36:419–25
16. Gupta M, Verma P, Rastogi R, Arora S, Elwadhi D. Randomized open-label trial of baclofen for relapse prevention in alcohol dependence Am J Drug Alcohol Abuse. 2017;43:324–31
17. Garbutt JC, Kampov-Polevoy AB, Gallop R, Kalka-Juhl L, Flannery BA. Efficacy and safety of baclofen for alcohol dependence: A randomized, double-blind, placebo-controlled trial Alcohol Clin Exp Res. 2010;34:1849–57
18. Maisel NC, Blodgett JC, Wilbourne PL, Humphreys K, Finney JW. Meta-analysis of naltrexone and acamprosate for treating alcohol use disorders: When are these medications most helpful? Addiction. 2013;108:275–93
19. Baltieri DA, de Andrade AG. Efficacy of acamprosate in the treatment of alcohol-dependent outpatients Braz J Psychiatry. 2003;25:156–9
20. Addolorato G, Caputo F, Capristo E, Colombo G, Gessa GL, Gasbarrini G. Ability of baclofen in reducing alcohol craving and intake: II--Preliminary clinical evidence Alcohol Clin Exp Res. 2000;24:67–71
21. Garbutt JC. Efficacy and tolerability of naltrexone in the management of alcohol dependence Curr Pharm Des. 2010;16:2091–7
22. Addolorato G, Leggio L, Ferrulli A, Cardone S, Vonghia L, Mirijello A, et al Effectiveness and safety of baclofen for maintenance of alcohol abstinence in alcohol-dependent patients with liver cirrhosis: Randomised, double-blind controlled study Lancet. 2007;370:1915–22
23. Volpicelli JR, Rhines KC, Rhines JS, Volpicelli LA, Alterman AI, O'Brien CP. Naltrexone and alcohol dependence. Role of subject compliance Arch Gen Psychiatry. 1997;54:737–42
24. Pelc I, Verbanck P, Le Bon O, Gavrilovic M, Lion K, Lehert P. Efficacy and safety of acamprosate in the treatment of detoxified alcohol-dependent patients. A 90-day placebo-controlled dose-finding study Br J Psychiatry. 1997;171:73–7
