Next Article in Journal
Screening for Post-Stroke Cognitive Impairment in Sub-Saharan Africa: A Good IDEA?
Previous Article in Journal
A New Method for Inducing Mental Fatigue: A High Mental Workload Task Paradigm Based on Complex Cognitive Abilities and Time Pressure
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Brain Sciences
Volume 15
Issue 6
10.3390/brainsci15060542
brainsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Systematic Review

Combination of Drugs in the Treatment of Alcohol Use Disorder: A Meta-Analysis and Meta-Regression Study

by
João Vitor Guimarães Mandaji
1,
Maria Olivia Pozzolo Pedro
2,
Kae Leopoldo
3,
João Pini Alemar
1,
Julio Torales
4,
Antonio Ventriglio
5,* and
João Mauricio Castaldelli-Maia
1,2
1
Department of Neuroscience, Medical School, FMABC Health University Center, Santo André 09060-870, Brazil
2
Department of Psychiatry, Medical School, University of São Paulo, São Paulo 05508-220, Brazil
3
Department of Experimental Psychology, Institute of Psychology, University of São Paulo, São Paulo 05508-220, Brazil
4
Department of Medical Psychology, Nacional University of Asunción, Asuncion 111421, Paraguay
5
Department of Clinical and Experimental Medicine, University of Foggia, 71122 Foggia, Italy
*
Author to whom correspondence should be addressed.
Brain Sci. 2025, 15(6), 542; https://doi.org/10.3390/brainsci15060542
Submission received: 9 April 2025 / Revised: 3 May 2025 / Accepted: 12 May 2025 / Published: 22 May 2025
(This article belongs to the Section Neuropsychiatry)
Browse Figures
Versions Notes

Abstract

:
Background: Alcohol Use Disorder (AUD) is highly prevalent among substance use disorders worldwide and is characterized by a multifactorial pathophysiology. AUD treatment is mostly based on combined pharmacotherapy and multidisciplinary clinical approaches. Nonetheless, meta-analytical studies assessing the efficacy of combination therapy are scarcely available. Methods: We searched for randomized clinical trials through PubMed, ClinicalTrials.gov, Cochrane Library, SciELO, Biblioteca Virtual em Saúde, and Google Scholar databases. Original clinical trials published in English and Portuguese were selected. Data collection followed the PRISMA and MOOSE guidelines and was assessed using the Risk of Bias Tool (RoB 2). Heterogeneity was assessed using the Q test. Meta-regression was conducted using Egger’s regression method. Twelve articles were finally included in the analysis, and random-effects models were applied on aggregate trial results. Results: The meta-analysis found that combination therapies led to an average 4.045% increase in abstinence rates (95% CI: 0.415% to 7.675%) compared to monotherapies. Meta-regression showed a strong positive association between the use of naltrexone, acamprosate, and sertraline—either alone or in combination—and treatment success in AUD. The meta-regression also highlighted the impact of patients’ variables, such as gender, age, country, and psychiatric comorbidities, on their treatment outcomes. These findings may identify a potential therapeutic pathway promoting alcohol abstinence, further supported by a Number Needed to Treat (NNT) of 25, as an acceptable value for substance use disorder treatments. Conclusions: Combined pharmacotherapies are more effective than monotherapy in enhancing abstinence rates in AUD treatment, with naltrexone, acamprosate, and sertraline emerging as key adjunctive agents promoting these outcomes. These findings underscore the complexity of AUD as a multifactorial psychiatric condition and highlight the potential of combined pharmacotherapy as a promising strategy for achieving better treatment outcomes, particularly in terms of abstinence rates.

1. Introduction

In 2020, the National Survey on Drug Use and Health reported that more than 28.3 million people aged 12 years or older in the United States (U.S.) met the DSM-5 (Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition) criteria for AUD in the past year [1]. Research evidence confirms that the personal and societal impact of AUD can be mitigated through treatments, which include psychological, social, and pharmaceutical interventions aimed at reducing alcohol consumption and its associated consequences [2]. Despite this evidence, the epidemiological research highlights that 74.5% of individuals with AUD in the U.S. are not receiving or seeking specific treatment, representing a critical gap in health care [3].
AUD is classified as a chronic, relapsing medical condition with a multifactorial etiology including genetic, neurobiological, psychological, and environmental factors [4]. The neurobiological substrate of AUD is characterized by a complex dysregulation across multiple neurotransmitter systems (Gamma-Aminobutyric Acid, Glutamate, Dopamine, Serotonin, Norepinephrine, Endogenous Opioids, Nicotinic Receptors, Endocannabinoid System) mediating the reward effect and abuse liability [5]. Given its complexity, combining medications may enhance treatment outcomes by targeting different neurobiological pathways and addictive behaviors. For instance, combining drugs can simultaneously address positive reinforcement (e.g., the pleasurable effects of alcohol) and negative reinforcement (e.g., relief from withdrawal symptoms), treat comorbid psychiatric and medical conditions, and minimize side effects by using lower doses of each medication [6,7].
The primary goal of AUD treatment should be to support patients in attaining either complete abstinence from alcohol or a substantial reduction in alcohol-related harm through decreased consumption [8]. Although there is an ongoing debate about whether abstinence or harm reduction should be the preferred treatment targets, both approaches have been acknowledged as beneficial [9]. Research suggests that complete abstinence is considered the most favorable outcome; however, even a reduction in alcohol consumption can lead to significant improvements in both general health and quality of life [10,11].
Several pharmaceutical options are available for AUD treatment. For instance, in the U.S., the Food and Drug Administration (FDA) approved the use of disulfiram, naltrexone, and acamprosate, alongside other off-label options, including nalmefene, baclofen, and topiramate, as suggested by successful clinical trials [12]. These trials reported their efficacy in the reduction of craving, prevention of relapses, and mitigation of withdrawal symptoms [12].
While monotherapies, e.g., disulfiram, have shown limited efficacy in preventing relapses beyond short-term abstinence [12,13,14], there is growing interest in combining drugs, such as disulfiram with anti-craving medications like naltrexone and acamprosate, for long-term therapeutic effects [14,15]. Decreasing craving may improve patients’ treatment adherence while also promoting their sustained abstinence [16]. Preclinical studies support the evidence of potential additive effects of drug combinations, as each medication targets distinct neurobiological pathways: naltrexone attenuates the positive reinforcing effects of alcohol, while acamprosate mitigates its negative reinforcing properties [15,16]. Overall, the heterogeneity in methodologies, assessments of AUD clinical severity, outcome measures, and adjunctive psychosocial treatments across studies on combined therapies all represent a significant limitation in the evaluation of their effectiveness [15,16,17].
This meta-analysis and meta-regression study aims to explore the evidence on the impact of combined therapy in the treatment of Alcohol Use Disorder (AUD).

2. Materials and Methods

2.1. Review Guidelines and Registration

This study followed the PRISMA statement for the transparent report of systematic reviews and meta-analysis [18] and the MOOSE guideline for Meta-analysis of Observational Studies [19]. Each clinical trial was evaluated by the Risk of Bias Tool (RoB 2) (Table 1) [20]. Tables S1 and S2 respectively present PRISMA and MOOSE checklists, indicating the manuscript sections where each item was addressed. This study was registered at the Center for Open Science/Open Science Framework: https://osf.io/fny54/?view_only=1c3c62de4e6f44198aafe6f2343b15db, accessed on 2 May 2025.

2.2. Information Sources

We employed the PubMed (U.S. National Library of Medicine), Google Scholar, and ClinicalTrials.gov databases to identify relevant articles on the research topic published in English and Portuguese. The last search considered was performed on 20 December 2024.
The search was based on the following keywords in the title/abstract: “((alcoholism OR alcohol use disorder OR alcohol abuse) combined pharmacotherapy)” filtered by categories “Randomized Controlled Trial” and “Clinical Trial”. Relevant articles were selected through a three-step process. In Step 1, the first author reviewed 991 studies and selected studies focusing on the effects of combined pharmacotherapy on AUD. In Step 2, two authors (the first and last authors) independently evaluated the abstracts selected in Step 1. Finally, in Step 3, articles were included based on consensus between the first and last authors, applying the following criteria: original studies reporting effect size estimates for monotherapy interventions (serving as the control condition) and combined therapy interventions (serving as the experimental condition) in relation to their impact on levels of abstinence; trials reporting data on pharmacological interventions for AUD; trials specifically addressing combined pharmacotherapy; and clinical trials published in English or Portuguese. Also, all studies that did not fulfill the following requirements were excluded: clinical trials conducted on humans, presence of a control group, and inclusion of the clinical parameter ‘percentage of abstinent days’. Additionally, the reference lists of all selected articles were reviewed to identify any missing studies and ensure the completeness of the literature search.
The primary objective of this selective process was to analyze the effect of combined pharmacotherapy for AUD.
Meta-regression analysis was conducted following the systematic tabulation of all included studies. Only variables consistently reported across all studies were considered, ensuring comparability and suitability for inclusion in the regression model.

2.3. Data Extraction

Data was extracted from the full-text articles by the first author and reviewed by the last author. All divergencies between the two authors were resolved through discussion with the senior author.

2.4. Quality Assessment

The methodological quality of all included studies was assessed using the modified Cochrane Collaboration tool for risk of bias in randomized controlled trials version 2 (RoB 2) [20]. Bias was evaluated across five domains: selection, performance, attrition, reporting, and other potential biases. Among the studies, 11 scored “low risk” and 1 as “some concerns”. All were included in the meta-analysis.

2.5. Data Analysis

Data analysis was performed using R software version 3.5.0. The mean difference in the percentage of days abstinent between combined pharmacotherapy and monotherapy was calculated. Heterogeneity was assessed using the Q test and quantified with the I2 statistic, where values below 25% indicated low heterogeneity, 50% indicated moderate heterogeneity, and ≥75% indicated high levels of heterogeneity (Chiang et al.) [21]. The DerSimonian-Laird estimator was used for tau2, and significant heterogeneity was detected.
Subgroup analyses were conducted to explore the influence of variables such as percentage of male participants, mean age, country of trial, psychiatric comorbidities, and drugs used as control or in combination. Publication bias was defined as p-values of 0.05 or less, and the ‘trim-and-fill’ technique was applied to adjust for bias, re-estimating the overall effect size and missing studies in the funnel plot (Idris, 2012; Shi and Lin, 2019) [22,23].
Univariate meta-regression analysis was performed to test the association of individual variables with treatment effects, followed by a random-effects regression model used to account for variability across studies (Foo et al., 2018) [24]. The random-effects model enabled generalization of findings by assuming the included trials represent a broader population. A significant level of 5% was applied in all analyses.
Table 1. Risk of Bias Tool (RoB2). Naltrexone (NTX); Sertraline (SERT); Topiramate (TPM); Quetiapine (QTP); Disulfiram (DSF); Escitalopram (ESC); Acamprosate (ACAM); Citalopram (CIT); Low Risk (LR); Some Concerns (SC); High Risk (HR); D1 = Randomization process; D2 = Deviations from the intended interventions; D3 = Missing outcome data; D4 = Measurement of the outcome; D5 = Selection of the reported result.
Table 1. Risk of Bias Tool (RoB2). Naltrexone (NTX); Sertraline (SERT); Topiramate (TPM); Quetiapine (QTP); Disulfiram (DSF); Escitalopram (ESC); Acamprosate (ACAM); Citalopram (CIT); Low Risk (LR); Some Concerns (SC); High Risk (HR); D1 = Randomization process; D2 = Deviations from the intended interventions; D3 = Missing outcome data; D4 = Measurement of the outcome; D5 = Selection of the reported result.
Unique IDStudy IDExperimentalD1D2D3D4D5Overall
1Helen M Pettinati, 2010 [25]NTX + SERTLRLRLRLRLRLR
2Melissa DelBello, 2010 [26]QTP + TPMLRLRHRLRLRSC
3Ismene Petrakis, 2007 [27]NTX + DSFLRLRLRLRLRLR
4Ismene Petrakis, 2005 [28]NTX + DSFLRLRLRLRLRLR
5Ismene Petrakis, 2006 [29]NTX + DSFLRLRLRLRLRLR
6Conor K Farren 2008 [30]NTX + SERTLRLRLRLRLRLR
7Janet Witte 2013 [31]ESC + ACAMLRLRLRLRLRLR
8O’Malley SS, 2008 [32]NTX + SERTLRLRLRLRLRLR
9Raymond F Anton, 2006 [33]NTX + ACAMLRLRLRLRLRLR
10Adamson SJ, 2015 [34]NTX + CITLRLRLRLRLRLR
11Josep Guardia, 2011 [35]NTX + QTPLRLRLRLRLRLR
12Jacques Besson, 1998 [36]ACAM + DSFLRLRLRLRLRLR

3. Results

Twelve studies were finally included in this review, as illustrated in Table S1 (PRISMA flowchart). From an initial pool of 6540 studies, 45 duplicates were excluded, and 991 titles and abstracts were screened. Following the title and abstract review, 72 studies underwent full-text analysis, of which 60 were excluded for lack of clinical parameters of interest, absence of control groups, lack of recorded results, or trials conducted on animals rather than humans. Table 2 summarizes the population characteristics of each trial, while Table 3 presents the main results, organized into comparisons between combined therapy and monotherapy, subgroups with and without psychiatric comorbidities, and groups receiving combined pharmacotherapy with or without psychotherapy.
The risk of bias assessment using the RoB 2 tool (Table 1) indicated that eleven of the twelve included studies presented a low risk of bias across all domains, suggesting a high level of methodological quality and reinforcing the internal validity of the meta-analysis. This consistency supports the reliability of the pooled effect estimates, as the influence of systematic errors related to study design or execution is likely minimal. One study, however, showed a high risk of bias in a single domain while maintaining a low risk in the others. Although this may introduce some concern, its isolated nature and limited extent reduce the likelihood of substantial impact on the overall findings. Nevertheless, the presence of any high-risk study warrants caution, particularly if it contributes significantly to the overall effect size. In this context, complementary analyses such as sensitivity or subgroup analyses based on the risk of bias are recommended to ensure the robustness of the results and to clarify the influence of methodological variability on the conclusions.
Most trials (7 out of 12) were conducted in North America. Diagnostic criteria for AUD were predominantly based on DSM-IV [37] (11 studies) and DSM-III [38] (1 study). Studies included were published between 1998 and 2015, with one from the 1990s, six from the early 2000s, and five from the 2010s. The meta-analysis included 3203 participants, of whom 875 reported psychiatric disorders (e.g., depression, bipolar mania, post-traumatic stress disorder) and 2429 received adjunctive psychological therapy.
The meta-analysis revealed a mean difference of 4.045% (95% CI: 0.415% to 7.675%) in percentage of days abstinent, in favor of combined therapy over monotherapy, as shown in the forest plot (Figure 1). However, significant heterogeneity was detected (I2 = 99.68%), and subgroup analysis indicated variability in treatment effects based on participant characteristics, location of trial, and study design.
Of 24 comparisons made, 12 involved naltrexone combined with other drugs (e.g., disulfiram, acamprosate, sertraline, quetiapine, or citalopram) versus naltrexone monotherapy. There were 8 comparisons that reported negative mean differences, indicating better outcomes for monotherapy. Only 6 of the 15 comparisons with positive mean differences reported standard deviations entirely within the positive axis, favoring combined therapy.
As depicted in Figure 2, the funnel plot showed a relatively symmetrical distribution of studies around the regression line, with no strong evidence of publication bias according to Egger’s test (p = 0.0823).
Results from the meta-regression (Table 4) identified significant positive associations with the combined use of naltrexone, acamprosate, disulfiram, sertraline, and topiramate, which enhanced patients’ abstinence rates. Conversely, the covariates “percentage male population” and “regions (U.S. vs. other)”—representing the meta-regression of the challenge “trial performed in the U.S.” versus “trial performed in any other country”—reported significant negative associations, suggesting higher effectiveness of combined therapy in populations with a lower proportion of males or in non-U.S. trials. Psychiatric comorbidities and mean age of participants also showed positive associations with improved treatment outcomes, indicating that the mean age of the trial population and the presence of comorbidities in the group of baseline characteristics significantly affect the impact of combined therapy over alcoholism treatment.
The overall analysis indicates that drug combinations yield better outcomes for AUD treatment, measured as percentage of days abstinent. Table 5 reports the Mixed-Effects Model and heterogeneity test (I2 = 99.68%) and indicates a substantial degree of heterogeneity, suggesting that the variability in effect sizes across studies is largely attributable to real differences in study characteristics—such as populations, interventions, or methodologies—rather than to random error. While I2 is informative for quantifying heterogeneity, its interpretability is limited, as it does not elucidate the source of variation. Therefore, the high heterogeneity observed should be interpreted cautiously and complemented by additional measures, such as τ2, alongside an assessment of its clinical relevance. Both combined therapies and monotherapy showed significant associations in improving abstinence rates, reflecting variability in treatment effects.
Groups’ characteristics were identified as significant covariates influencing outcomes in favor of abstinence rates, as covariates ‘mean age’ and ‘patients’ psychiatric comorbidity’, or against as covariates ‘percentage of male patients’ and ‘country in which the trial was conducted (U.S. vs Other)’. Table 4 also presents the residual heterogeneity (τ2 = 0.0298), with an explained variability of 89.12% (R2), indicating that covariates contributed substantially to differences in effect sizes.

Key Metrics

  • Tau-squared (τ2): The residual heterogeneity (τ2 = 0.0298) indicates variability in the effect sizes that remains unexplained by the model.
  • I2 (Residual Heterogeneity/Unaccounted Variability): The I2 value is extremely high (99.68%), suggesting that nearly all the variability in the prevalence estimates is due to heterogeneity between studies, not to sampling error.
  • H2 (Unaccounted Variability/Sampling Variability): H2 = 314.52 is quite large, indicating that there is substantial heterogeneity in the mean difference of the effect of combined therapy over monotherapy in relation to the percentage of abstinence parameter across studies. This suggests that there are unmeasured factors causing differences in this parameter, and the model does not fully account for the variability observed.
  • R2 (Amount of Heterogeneity Accounted for): The R2 value is 89.12%, suggesting that the covariates in the model are meaningful in explaining the differences in effect sizes across studies.
  • Test for Residual Heterogeneity: The very small p-value (<0.0001) indicates there is still significant heterogeneity left unexplained by the covariates, indicating that other factors not included in the model may be contributing to the variability in effect sizes.

4. Discussion

We conducted a systematic review and meta-regression of randomized clinical trials in order to evaluate the effect of combined pharmacotherapy compared to monotherapy in AUD treatment. The meta-analysis provided evidence that drug combination results in a modest but significant improvement in patients’ abstinence rates, with a mean difference of 4.04% (95% CI: 0.415% to 7.675%) in the clinical parameter “% days abstinent”.
Although this effect may appear small, it holds clinical relevance. In fact, AUD treatment may be complex due to its multifactorial etiology, encompassing neurobiological, genetic, psychological, social, and environmental factors [17]. Furthermore, the effect corresponds to a Number Needed to Treat (NNT) of 25, which can be considered acceptable in the context of chronic disease treatment, such as AUD, especially if the benefits of the treatment outweigh the risks and costs and if aligning with patients’ preferences [39]. Comparatively, the meta-analysis by McPheeters et al. [40] indicates that the number needed to treat (NNT) to prevent relapse to alcohol use is 11 for acamprosate and 18 for oral naltrexone at a dose of 50 mg/day. Furthermore, to prevent the return to heavy drinking, the NNT for oral naltrexone is 11. On the other hand, the meta-analysis by Jonas et al. [41] reports an NNT of 12 for acamprosate and 20 for oral naltrexone to prevent relapse to alcohol use. To prevent the return to heavy drinking, the NNT for oral naltrexone is 12. It is also important to highlight that when comparing these two meta-analyses with the one provided in this study, there is a key difference: unlike the previous analyses, which derived from the comparison of studies based on therapy versus no therapy, the present analysis provides an NNT for the comparison between combined therapy and monotherapy. Therefore, the value of this parameter should be considered from a different perspective than those obtained in the other two studies. In this context, an NNT of 25 can be regarded as clinically significant, particularly if the treatment is safe, well-tolerated, and considers the substantial impact of Alcohol Use Disorder (AUD) on quality of life and mortality.
From a neurobiological perspective, AUD involves dysregulation in brain regions associated with motivation, stress regulation, and reward, such as the midbrain, limbic system, prefrontal cortex, and amygdala [42,43]. Both positive and negative reinforcement mechanisms play crucial roles in sustaining drinking behavior. Positive reinforcement refers to the rewarding and desirable effects of alcohol, while negative reinforcement involves the relief of negative emotional and physiological states, such as anxiety, depression, or withdrawal symptoms, through alcohol consumption [44].
Combined therapies may target these mechanisms through complementary actions. For example, naltrexone reduces positive reinforcement by blocking opioid receptor-mediated dopamine release in the nucleus accumbens, amygdala, and forebrain [45,46], while also enhancing alcohol sedative effects and reducing cravings triggered by consumption or alcohol-related cues [47].
Acamprosate, a structural analog of the amino acid taurine, reduces alcohol consumption and alleviates withdrawal symptoms by reducing alcohol cravings by modulating NMDA (N-Methyl-D-Aspartate) receptors and inhibiting excitatory glutamatergic activity [48,49]. This action diminishes withdrawal distress and suppresses brain hyper-excitability associated with ethanol consumption [48]. Additionally, experimental studies have shown that acamprosate enhances GABA (Gamma-Aminobutyric Acid) reuptake and modulates dopaminergic activity in the nucleus accumbens, reducing positive reinforcement from alcohol [50,51,52,53].
Disulfiram deters alcohol consumption by inhibiting acetaldehyde dehydrogenase, leading to the accumulation of acetaldehyde in the bloodstream [53,54]. This causes adverse effects, such as flushing, tachycardia, and nausea, which act as strong deterrents to alcohol consumption [55].
Having delineated certain mechanisms of action underlying the pharmacological agents employed in the treatment of Alcohol Use Disorder (AUD), the central tenet of this study rests upon the premise that drug combinations may enhance therapeutic efficacy through two principal strategies. The first strategy posits that both interventions target the same drinking behavior, thereby amplifying their impact. Given that existing treatments exhibit, at best, only moderate effect sizes, this approach emerges as a logical means to bolster clinical outcomes. Ideally, such combinations would yield additive—or even synergistic—effects, though the possibility of adverse interactions cannot be disregarded. This rationale extends not only to pharmacotherapy coupled with behavioral interventions but also to the concomitant administration of two pharmacological agents. For example, a pharmacological agent reducing the positive reinforcement of drinking (e.g., naltrexone) can be paired with a medication that attenuates protracted alcohol withdrawal symptoms (e.g., acamprosate) [17]. These recommendations align with the findings of our study, which indicate strong positive associations between the use of naltrexone, acamprosate, and sertraline, both when combined with other drugs and when other drugs are added to them in the treatment of AUD [Naltrexone associated (estimate: 2.6221, p ≤ 0.0001), Acamprosate associated (3.0294, p ≤ 0.0001), Sertraline associated (4.2262, p ≤ 0.0001), Naltrexone control (1.8796, p ≤ 0.0001), Acamprosate control (2.2960, p = 0.0002), Sertraline control (3.2263, p ≤ 0.0001)].
It is of note that the combination of medications may also increase the risk of adverse drug interactions. For instance, disulfiram, when administered with other medications, can exacerbate pharmacological interactions, particularly with central nervous system stimulants [56]. Furthermore, some studies have reported the occurrence of adverse cardiac events with the combination of naltrexone and disulfiram [57].
Despite the findings of our analysis, there is no consolidated evidence of significant synergistic or additive effects of certain combined medications. For example, pre-clinical studies have not demonstrated additive effects when semaglutide is combined with varenicline or bupropion [58]. This suggests that, in some cases, the combination of medications does not provide additional benefits compared to monotherapy.
Also, polypharmacy may potentially reduce patients’ adherence to treatment; adherence is a critical factor for the successful treatment of substance use disorders [59]. Thus, combined medication regimens may lead to treatment discontinuation.
Finally, treatment combination safety is not thoroughly investigated, especially in populations with comorbidities [57].
In this context, our meta-regression revealed that the variable “% Male” showed a strong negative association (−8.6545, p = 0.0095) with the percentage of men in the sample, suggesting that groups with a higher proportion of women tend to show better treatment outcomes. The impact of gender on the treatment outcome of AUD is controversial, with evidence indicating that women more frequently seek help and engage in self-care while tending to drop out of treatment [60]. Also, women are more likely to be responsive and committed to the treatment process [61].
Location of trials showed a significantly negative association (−3.9626, p < 0.0001) when comparing studies conducted in the U.S. to those conducted in the other countries (UK, Spain, Switzerland). These differences may reflect the interplay between culturally specific interpretations of mental disorders and the standardized classificatory framework used in the countries (DSM-III and DSM-IV). Current definitions of alcohol dependence and alcohol use disorders are based on culturally specific criteria [62]. Also, the clinical dimensions and subjective experiences of “Alcohol Use Disorders” are often objectives of measurement as well as interpretation [63]. Considering cross-cultural differences may also help in understanding the relationship between alcohol and harm—whether physical, mental, or social—and may suggest different pathways for harm prevention. This perspective helps to partially explain the high heterogeneity observed in the meta-regression (I2 = 99.68%) and the associations with the country where the trial was conducted. It may suggest that there may be greater social support for alcohol use treatment in synergy with pharmacological therapy in “non-USA” countries.
We also found a weak but relevant positive association between the rate of psychiatric comorbidities and treatment outcomes (0.2766, p = 0.0112). The association between the employment of combination treatments for AUD and comorbidities is still controversial. Pettinati et al. [25] have shown that sertraline and naltrexone demonstrated higher efficacy in individuals with Alcohol Use Disorder (AUD) and concomitant depression. In fact, when compared to sertraline, naltrexone, or placebo administered alone, this combined regimen resulted in a significantly higher incidence of alcohol abstinence and an extended duration until relapse occurred.
Moreover, in those patients who were daily smokers and heavy alcohol consumers, the combination of varenicline and naltrexone was effective in reducing the levels of alcohol consumption. This pharmacological combination reduced the risk of alcohol use compared to varenicline administered alone [64]. Castillo-Carniglia et al. [65] argued that the comorbidity between AUD and psychiatric conditions may be controversial, with AUD leading to additional psychopathological issues as well as mental distress leading to AUD. This still confirms that integrated treatments for alcohol use disorder and comorbid psychiatric disorders lead to better outcomes than non-integrated treatments [66,67,68]. These conclusions align with the results of our meta-regression, reporting a significant correlation between combined treatments and psychiatric comorbidities. Nonetheless, evidence in the literature, including our findings, may be limited since the studies available are based on small sample sizes, short follow-up periods, non-experimental designs, and highly heterogeneous and poorly described treatments [65].
Finally, the meta-regression analysis revealed a positive and significant association between the age of study participants and their treatment outcome (0.2519, p = 0.0095). Supporting this finding, a secondary evaluation of the COMBINE (Combined Pharmacotherapies and Behavioral Interventions for Alcohol Dependence) study identified age as one of the primary factors, alongside the number of consecutive days of abstinence prior to randomization, in predicting abstinence from heavy alcohol consumption [69,70]. This suggests that age may influence treatment responses for AUD, with older individuals tending to exhibit better outcomes in terms of sustained abstinence from intensive drinking.
The PREDICT (Predictors of Response to Treatment in Alcoholism) study validated the results of the COMBINE study within a large population in Germany, confirming age as a relevant predictive factor. In fact, the deterministic forest model, based on clinical and statistical criteria, highlighted age as a key factor, alongside the family history of alcoholism and confidence in resisting consumption during periods of abstinence and craving. The external validation of these findings in the PREDICT study further underscores the significance of age as a predictive factor for treatment outcomes across different populations [70].
Although combined pharmacotherapy demonstrated a modest but clinically significant improvement in abstinence rates, it is important to consider pharmacoeconomic implications. Combination therapies may increase direct treatment costs due to the simultaneous use of multiple medications, potentially impacting healthcare budgets and patient adherence. However, these higher initial costs could be offset by long-term benefits, such as reduced relapse rates, fewer hospitalizations, and decreased burden on healthcare services. Evidence from broader substance use treatment literature suggests that interventions improving abstinence are generally cost-effective when accounting for reductions in morbidity, productivity loss, and societal costs [71]. Future studies should directly evaluate the cost-effectiveness of combined pharmacotherapy strategies for Alcohol Use Disorder, considering both direct medical expenses and indirect societal benefits.

5. Conclusions

The treatment of AUD reports higher efficacy when based on combined pharmacotherapy compared to monotherapy, particularly including naltrexone, acamprosate, and sertraline. These findings highlight a promising therapeutic approach that enhances alcohol abstinence, further supported by an NNT of 25, underscoring the potential of combination therapies to significantly improve abstinence rates and yield clinically meaningful benefits. Meta-regression analyses emphasize the necessity of considering individual and population-specific factors—such as gender, age, geographic region of investigation, and patient’s psychiatric comorbidities—when tailoring treatment strategies. Despite these advantages, the observed heterogeneity across studies suggests that unmeasured variables, including psychosocial and cultural influences, may play a pivotal role in treatment outcomes. Future research should focus on addressing these gaps to refine and optimize therapeutic strategies for AUD.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/brainsci15060542/s1, Table S1: PRISMA CHECKLIST; Table S2: MOOSE.

Author Contributions

J.V.G.M.: Conceived and designed the experiments; performed the experiments; analyzed and interpreted the data; contributed reagents, materials, analysis tools, or data; wrote the paper. J.P.A.: Conceived and designed the experiments; performed the experiments; analyzed and interpreted the data; contributed reagents, materials, analysis tools, or data; wrote the paper. J.M.C.-M.: Conceived and designed the experiments; performed the experiments; analyzed and interpreted the data; contributed reagents, materials, analysis tools, or data; wrote the paper. M.O.P.P.: Conceived and designed the experiments; performed the experiments; analyzed and interpreted the data; contributed reagents, materials, analysis tools, or data; wrote the paper. K.L.: Conceived and designed the experiments; performed the experiments; analyzed and interpreted the data; contributed reagents, materials, analysis tools, or data; wrote the paper. A.V.: Conceived and designed the experiments; performed the experiments; analyzed and interpreted the data; contributed reagents, materials, analysis tools, or data; wrote the paper. J.T.: Conceived and designed the experiments; performed the experiments; analyzed and interpreted the data; contributed reagents, materials, analysis tools, or data; wrote the paper. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the São Paulo Research Foundation (FAPESP, Portuguese: Fundação de Amparo à Pesquisa do Estado de São Paulo), a public foundation located in São Paulo, Brazil. Process number: 2024/05632-0.

Data Availability Statement

Data included in article/Supplementary Materials/references.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Substance Abuse and Mental Health Services Administration. Key Substance Use and Mental Health Indicators in the United States: Results From the 2020 National Survey on Drug Use and Health. 2021. Available online: https://www.samhsa.gov/data/ (accessed on 3 March 2023).
  2. Swift, R.M. Drug therapy for alcohol dependence. N. Engl. J. Med. 1999, 340, 1482–1490. [Google Scholar] [CrossRef] [PubMed]
  3. Dawson, D.A.; Grant, B.F.; Stinson, F.S.; Chou, P.S.; Huang, B.; Ruan, W.J. Recovery from DSM-IV alcohol dependence: United States, 2001–2002. Addiction 2005, 100, 281–292. [Google Scholar] [CrossRef] [PubMed]
  4. Koob, G.F. The neurobiology of addiction: A neuroadaptational view relevant for diagnosis. Addiction 2006, 101, 23–30. [Google Scholar] [CrossRef] [PubMed]
  5. O’Malley, S.S.; Froehlich, J.C. Advances in the use of naltrexone: An integration of preclinical and clinical findings. In Recent Advances in Alcoholism, Volume 16: Research on Alcoholism Treatment; Galanter, M., Ed.; Kluwer Academic/Plenum: Nova Iorque, NY, USA, 2003; pp. 217–245. [Google Scholar]
  6. Johnson, B.A.; Ait-Daoud, N.; Prihoda, T.J. Combining ondansetron and naltrexone effectively treats biologically predisposed alcoholics: From hypotheses to preliminary clinical evidence. Alcohol. Clin. Exp. Res. 2000, 24, 737–742. [Google Scholar] [CrossRef]
  7. Witkiewitz, K.; Kranzler, H.R.; Hallgren, K.A.; Hasin, D.S.; Aldridge, A.P.; Zarkin, G.A.; Mann, K.F.; O’malley, S.S.; Anton, R.F. Stability of Drinking Reductions and Long-term Functioning Among Patients with Alcohol Use Disorder. J. General. Intern. Med. 2021, 36, 404–412. [Google Scholar] [CrossRef]
  8. Henssler, J.; Müller, M.; Carreira, H.; Bschor, T.; Heinz, A.; Baethge, C. Controlled drinking-non-abstinent versus abstinent treatment goals in alcohol use disorder: A systematic review, meta-analysis and meta-regression. Addiction 2021, 116, 1973–1987. [Google Scholar] [CrossRef]
  9. Kline-Simon, A.H.; Falk, D.E.; Litten, R.Z.; Mertens, J.R.; Fertig, J.; Ryan, M.; Weisner, C.M. Posttreatment low-risk drinking as a predictor of future drinking and problem outcomes among individuals with alcohol use disorders. Alcohol. Clin. Exp. Res. 2013, 37 (Suppl. S1), E373–E380. [Google Scholar] [CrossRef]
  10. Yang, W.; Singla, R.; Maheshwari, O.; Fontaine, C.J.; Gil-Mohapel, J. Alcohol Use Disorder: Neurobiology and Therapeutics. Biomedicines 2022, 10, 1192. [Google Scholar] [CrossRef]
  11. Hnatiuk, M.S.; Nesteruk, S.O.; Fedoniuk, L.Y.; Yakymchuk, O.A.; Smachylo, I.I.; Tverdochlib, V.V. Quantitative morphological analysis of age structural changes in prostate of experimental animals with ethanol poisoning. Wiadomości Lek. 2024, 77, 268–272. [Google Scholar] [CrossRef]
  12. Azrin, N.H.; Sisson, R.W.; Meyers, R.; Godley, M. Alcoholism treatment by disulfiram and community reinforcement therapy. J. Behav. Ther. Exp. Psychiatry 1982, 13, 105–112. [Google Scholar] [CrossRef]
  13. Fuller, R.K.; Branchey, L.; Brightwell, D.R.; Derman, R.M.; Emrick, C.D.; Iber, F.L.; James, K.E.; Lacoursiere, R.B.; Lee, K.K.; Lowenstam, I.; et al. Disulfiram treatment of alcoholism. A Veterans Administration cooperative study. JAMA 1986, 256, 1449–1455. [Google Scholar] [CrossRef] [PubMed]
  14. Jorgensen, C.H.; Pedersen, B.; Tonnesen, H. The efficacy of disulfiram for the treatment of alcohol use disorder. Alcohol. Clin. Exp. Res. 2011, 35, 1749–1758. [Google Scholar] [CrossRef] [PubMed]
  15. Lee, M.R.; Leggio, L. Combined Pharmacotherapies for the Management of Alcoholism: Rationale and Evidence to Date. CNS Drugs 2014, 28, 107–119. [Google Scholar] [CrossRef] [PubMed]
  16. Kim, S.G.; Han, B.D.; Park, J.M.; Kim, M.; Stromberg, M.F. Effect of the combination of naltrexone and acamprosate on alcohol intake in mice. Psychiatry Clin. Neurosci. 2004, 58, 30–36. [Google Scholar] [CrossRef]
  17. Mattson, M.E.; Litten, R.Z. Combining treatments for alcoholism: Why and how? J. Stud. Alcohol Suppl. 2005, 15, 8–16. [Google Scholar] [CrossRef]
  18. Moher, D.; Liberati, A.; Tetzlaff, J.; Altman, D.G. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Med. 2009, 6, e1000097. [Google Scholar] [CrossRef]
  19. Brooke, B.S.; Schwartz, T.A.; Pawlik, T.M. MOOSE Reporting Guidelines for Meta-analyses of Observational Studies. JAMA Surg. 2021, 156, 787–788. [Google Scholar] [CrossRef]
  20. Higgins, J.P.; Altman, D.G.; Gøtzsche, P.C.; Jüni, P.; Moher, D.; Oxman, A.D.; Savovic, J.; Schulz, K.F.; Weeks, L.; Sterne, J.A.; et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011, 343, d5928. [Google Scholar] [CrossRef]
  21. Chiang, C.L.L.; Zhang, M.W.B.; Ho, R.C.M. Prevalence of Internet Gaming Disorder in Medical Students: A Meta-Analysis. Front. Psychiatry 2022, 12, 760911. [Google Scholar] [CrossRef]
  22. Idris, N.R.N. Performance of the trim and fill method in adjusting for the publication bias in meta-analysis of continuous data. Am. J. Appl. Sci. 2012, 9, 1512. [Google Scholar]
  23. Shi, L.; Lin, L. The trim-and-fill method for publication bias: Practical guidelines and recommendations based on a large database of meta-analyses. Medicine 2019, 98, e15987. [Google Scholar] [CrossRef] [PubMed]
  24. Foo, S.Q.; Tam, W.W.; Ho, C.S. Prevalence of depression among migrants: A systematic review and meta-analysis. Int. J. Environ. Res. Public Health 2018, 15, 1986. [Google Scholar] [CrossRef] [PubMed]
  25. Pettinati, H.M.; Oslin, D.W.; Kampman, K.M.; Dundon, W.D.; Xie, H.; Gallis, T.L.; Dackis, C.A.; O’Brien, C.P. A double-blind, placebo-controlled trial combining sertraline and naltrexone for treating co-occurring depression and alcohol dependence. Am. J. Psychiatry 2010, 167, 668–675. [Google Scholar] [CrossRef]
  26. National Institutes of Health. A Study to Evaluate the Effectiveness of Topiramate and Quetiapine in the Treatment of Alcohol Dependence. ClinicalTrials.gov. 2007. Available online: https://clinicaltrials.gov/study/NCT00550394 (accessed on 25 May 2024).
  27. Petrakis, I.; Ralevski, E.; Nich, C.; Levinson, C.; Carroll, K.; Poling, J.; Rounsaville, B. Naltrexone and disulfiram in patients with alcohol dependence and current depression. J. Clin. Psychopharmacol. 2007, 27, 160–165. [Google Scholar] [CrossRef]
  28. Petrakis, I.L.; Poling, J.; Levinson, C.; Nich, C.; Carroll, K.; Rounsaville, B. Naltrexone and disulfiram in patients with alcohol dependence and comorbid psychiatric disorders. Biol. Psychiatry 2005, 57, 1128–1137. [Google Scholar] [CrossRef]
  29. Petrakis, I.L.; Poling, J.; Levinson, C.; Nich, C.; Carroll, K.; Ralevski, E.; Rounsaville, B. Naltrexone and disulfiram in patients with alcohol dependence and comorbid post-traumatic stress disorder. Biol. Psychiatry 2006, 60, 777–783. [Google Scholar] [CrossRef]
  30. Farren, C.K.; Scimeca, M.; Wu, R.; Malley, S.O. A double-blind, placebo-controlled study of sertraline with naltrexone for alcohol dependence. Drug Alcohol. Depend. 2009, 99, 317–321. [Google Scholar] [CrossRef]
  31. Witte, J.; Bentley, K.; Evins, A.E.; Clain, A.J.; Baer, L.; Pedrelli, P.; Fava, M.; Mischoulon, D. A randomized, controlled, pilot study of acamprosate added to escitalopram in adults with major depressive disorder and alcohol use disorder. J. Clin. Psychopharmacol. 2012, 32, 787–796. [Google Scholar] [CrossRef] [PubMed]
  32. O’Malley, S.S.; Robin, R.W.; Levenson, A.L.; GreyWolf, I.; Chance, L.E.; Hodgkinson, C.A.; Romano, D.; Robinson, J.; Meandzija, B.; Stillner, V.; et al. Naltrexone alone and with sertraline for the treatment of alcohol dependence in Alaska natives and non-natives residing in rural settings: A randomized controlled trial. Alcohol. Clin. Exp. Res. 2008, 32, 1271–1283. [Google Scholar] [CrossRef]
  33. Anton, R.F.; O’Malley, S.S.; Ciraulo, D.A.; Cisler, R.A.; Couper, D.; Donovan, D.M.; Gastfriend, D.R.; Hosking, J.D.; Johnson, B.A.; LoCastro, J.S.; et al. Combined pharmacotherapies and behavioral interventions for alcohol dependence: The COMBINE study: A randomized controlled trial. JAMA 2006, 295, 2003–2017. [Google Scholar] [CrossRef]
  34. Adamson, S.J.; Sellman, J.D.; Foulds, J.A.; Frampton, C.M.; Deering, D.; Dunn, A.; Berks, J.; Nixon, L.; Cape, G. A randomized trial of combined citalopram and naltrexone for nonabstinent outpatients with co-occurring alcohol dependence and major depression. J. Clin. Psychopharmacol. 2015, 35, 143–149. [Google Scholar] [CrossRef] [PubMed]
  35. Guardia, J.; Roncero, C.; Galan, J.; Gonzalvo, B.; Burguete, T.; Casas, M. A double-blind, placebo-controlled, randomized pilot study comparing quetiapine with placebo, associated to naltrexone, in the treatment of alcohol-dependent patients. Addict. Behav. 2011, 36, 265–269. [Google Scholar] [CrossRef] [PubMed]
  36. Besson, J.; Aeby, F.; Kasas, A.; Lehert, P.; Potgieter, A. Combined efficacy of acamprosate and disulfiram in the treatment of alcoholism: A controlled study. Alcohol. Clin. Exp. Res. 1998, 22, 573–579. [Google Scholar] [CrossRef]
  37. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th ed.; American Psychiatric Association: Washington, DC, USA, 1994. [Google Scholar]
  38. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd ed.; American Psychiatric Association: Washington, DC, USA, 1980. [Google Scholar]
  39. Kristiansen, I.S.; Gyrd-Hansen, D.; Nexøe, J.; Nielsen, J.B. Number needed to treat: Easily understood and intuitively meaningful? Theoretical considerations and a randomized trial. J. Clin. Epidemiol. 2002, 55, 888–892. [Google Scholar] [CrossRef]
  40. McPheeters, M.; O’Connor, E.A.; Riley, S.; Kennedy, S.M.; Voisin, C.; Kuznacic, K.; Coffey, C.P.; Edlund, M.D.; Bobashev, G.; Jonas, D.E. Pharmacotherapy for Alcohol Use Disorder: A Systematic Review and Meta-Analysis. JAMA 2023, 330, 1653–1665, https://doi.org/10.1001/jama.2023.19761. Erratum in JAMA 2024, 332, 1397. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
  41. Jonas, D.E.; Amick, H.R.; Feltner, C.; Bobashev, G.; Thomas, K.; Wines, R.; Kim, M.M.; Shanahan, E.; Gass, C.E.; Rowe, C.J.; et al. Pharmacotherapy for Adults With Alcohol Use Disorders in Outpatient Settings: A Systematic Review and Meta-analysis. JAMA 2014, 311, 1889–1900. [Google Scholar] [CrossRef]
  42. Avegno, E.M.; Gilpin, N.W. Reciprocal midbrain-extended amygdala circuit activity in preclinical models of alcohol use and misuse. Neuropharmacology 2022, 202, 108856. [Google Scholar] [CrossRef]
  43. Koob, G.F.; Vendruscolo, L. Theoretical Frameworks and Mechanistic Aspects of Alcohol Addiction: Alcohol Addiction as a Reward Deficit/Stress Surfeit Disorder. Curr. Top. Behav. Neurosci. 2023, 1–61. [Google Scholar] [CrossRef]
  44. Witkiewitz, K.; Litten, R.Z.; Leggio, L. Advances in the science and treatment of alcohol use disorder. Sci. Adv. 2019, 5, eaax4043. [Google Scholar] [CrossRef]
  45. Herz, A. Endogenous opioid systems and alcohol addiction. Psychopharmacology 1997, 129, 99–111. [Google Scholar] [CrossRef]
  46. Spreckelmeyer, K.N.; Paulzen, M.; Raptis, M.; Baltus, T.; Schaffrath, S.; Van Waesberghe, J.; Zalewski, M.M.; Rösch, F.; Vernaleken, I.; Schäfer, W.M.; et al. Opiate-induced dopamine release is modulated by severity of alcohol dependence: An [(18)F]fallypride positron emission tomography study. Biol. Psychiatry 2011, 70, 770–776. [Google Scholar] [CrossRef] [PubMed]
  47. Volpicelli, J.R.; Alterman, A.I.; Hayashida, M.; O’Brien, C.P. Naltrexone in the treatment of alcohol dependence. Arch. Gen. Psychiatry 1992, 49, 876–880. [Google Scholar] [CrossRef] [PubMed]
  48. Littleton, J. Acamprosat? In alcohol dependence: How does it work? Addiction 1995, 90, 1179–1188. [Google Scholar] [CrossRef]
  49. Naassila, M.; Hammoumi, S.; Legrand, E.; Durbin, P.; Daoust, M. Mechanism of action of acamprosate. Part I. Characterization of spermidine-sensitive acamprosate binding site in rat brain. Alcohol. Clin. Exp. Res. 1998, 22, 802–809. [Google Scholar] [PubMed]
  50. Swift, R.M.; Aston, E.R. Pharmacotherapy for alcohol use disorder: Current and emerging therapies. Harv. Rev. Psychiatry 2015, 23, 122–133. [Google Scholar] [CrossRef]
  51. Stromberg, M.F.; Mackler, A.S.; Volpicelli, J.R.; O’Brien, C.P. Effect of acamprosate and naltrexone, alone or in combination, on ethanol consumption. Alcohol. 2001, 23, 109–116. [Google Scholar] [CrossRef]
  52. Castro, L.A.; Baltieri, D.A. Tratamento farmacológico da dependência do álcool. Braz. J. Psychiatry 2004, 26, 43–46. [Google Scholar] [CrossRef]
  53. Wall, T.L. Genetic associations of alcohol and aldehyde dehydrogenase with alcohol dependence and their mechanisms of action. Ther. Drug Monit. 2005, 27, 700–703. [Google Scholar] [CrossRef]
  54. Thomasson, H.R.; Edenberg, H.J.; Crabb, D.W.; Mai, X.L.; Jerome, R.E.; Li, T.K.; Wang, S.P.; Lin, Y.T.; Lu, R.B.; Yin, S.J.; et al. Alcohol and aldehyde dehydrogenase genotypes and alcoholism in Chinese men. Am. J. Hum. Genet. 1991, 48, 677–681. [Google Scholar]
  55. Suh, J.J.; Pettinati, H.M.; Kampman, K.M.; O’Brien, C.P. The status of disulfiram: A half of a century later. J. Clin. Psychopharmacol. 2006, 26, 290–302. [Google Scholar] [CrossRef]
  56. Kleczkowska, P.; Sulejczak, D.; Zaremba, M. Advantages and disadvantages of disulfiram coadministered with popular addictive substances. Eur. J. Pharmacol. 2021, 904, 174143. [Google Scholar] [CrossRef]
  57. Rosenstand, N.J.; Nielsen, A.S.; Skøt, L.; Anhøj, S.; Nielsen, D.G.; Højlund, M.; Mellentin, A.I. Pharmacological Treatment of Alcohol use Disorder in Patients with Psychotic Disorders: A Systematic Review. Curr. Neuropharmacol. 2024, 22, 1129–1143. [Google Scholar] [CrossRef]
  58. Aranäs, C.; Blid Sköldheden, S.; Jerlhag, E. Antismoking agents do not contribute synergistically to semaglutide’s ability to reduce alcohol intake in rats. Front. Pharmacol. 2023, 14, 1180512. [Google Scholar] [CrossRef]
  59. Stout, R.L.; Braciszewski, J.M.; Subbaraman, M.S.; Kranzler, H.R.; O’Malley, S.S.; Falk, D.; ACTIVE Group. What happens when people discontinue taking medications? Lessons from COMBINE. Addiction 2014, 109, 2044–2052. [Google Scholar] [CrossRef]
  60. Greenfield, S.F. Women and Alcohol Use Disorders. Harv. Rev. Psychiatry 2002, 10, 76–85. [Google Scholar] [CrossRef]
  61. Moos, R.H.; Moos, B.S.; Timko, C. Gender, treatment and self-help in remission from alcohol use disorders. Clin. Med. Res. 2006, 4, 163–174. [Google Scholar] [CrossRef]
  62. Last, J.M. Dictionary of Epidemiology. CMAJ 1993, 149, 400. [Google Scholar] [CrossRef]
  63. Rehm, J.; Room, R. O aspecto cultural: Como medir e interpretar dados epidemiológicos sobre transtornos por uso de álcool em diferentes culturas. Estud. Nórdicos Sobre Álcool Drog. 2017, 34, 330–341. [Google Scholar] [CrossRef]
  64. Nieto, S.J.; Enders, C.K.; Witkiewitz, K.; O’Malley, S.S.; Ray, L.A. Combination treatment with varenicline and naltrexone reduces World Health Organization risk drinking levels. Alcohol. Clin. Exp. Res. 2022, 46, 2258–2266. [Google Scholar] [CrossRef]
  65. Castillo-Carniglia, A.; Keyes, K.M.; Hasin, D.S.; Cerdá, M. Psychiatric comorbidities in alcohol use disorder. Lancet Psychiatry 2019, 6, 1068–1080. [Google Scholar] [CrossRef]
  66. Drake, R.E.; O’Neal, E.L.; Wallach, M.A. Uma revisão sistemática da pesquisa psicossocial sobre intervenções psicossociais para pessoas com transtornos mentais e de uso de substâncias graves concomitantes. J. Subst. Abus. Treat. 2008, 34, 123–138. [Google Scholar] [CrossRef]
  67. Drake, R.E.; Mueser, K.T.; Brunette, M.F.; McHugo, G.J. Uma revisão de tratamentos para pessoas com doenças mentais graves e transtornos de uso de substâncias concomitantes. Psychiatr. Rehabil. J. 2004, 27, 360–374. [Google Scholar] [CrossRef] [PubMed]
  68. Horsfall, J.; Cleary, M.; Hunt, G.E.; Walter, G. Tratamentos psicossociais para pessoas com doenças mentais graves concomitantes e transtornos por uso de substâncias (diagnóstico duplo): Uma revisão de evidências empíricas. Harv. Rev. Psychiatry 2009, 17, 24–34. [Google Scholar] [CrossRef] [PubMed]
  69. Wallach, J.D.; Gueorguieva, R.; Phan, H.; Witkiewitz, K.; Wu, R.; O’Malley, S.S. Predictors of abstinence, no heavy drinking days, and a 2-level reduction in World Health Organization drinking levels during treatment for alcohol use disorder in the COMBINE study. Alcohol. Clin. Exp. Res. 2022, 46, 1331–1339. [Google Scholar] [CrossRef] [PubMed]
  70. Gueorguieva, R.; Wu, R.; O’Connor, P.G.; Weisner, C.; Fucito, L.M.; Hoffmann, S.; Mann, K.; O’Malley, S.S. Predictors of abstinence from heavy drinking during treatment in COMBINE and external validation in PREDICT. Alcohol. Clin. Exp. Res. 2014, 38, 2647–2656. [Google Scholar] [CrossRef]
  71. Bardach, A.E.; Alcaraz, A.O.; Ciapponi, A.; Garay, O.U.; Riviere, A.P.; Palacios, A.; Cremonte, M.; Augustovski, F. Alcohol consumption’s attributable disease burden and cost-effectiveness of targeted public health interventions: A systematic review of mathematical models. BMC Public Health 2019, 19, 1378. [Google Scholar] [CrossRef]
Brainsci 15 00542 g001
Figure 1. Forest Plot of the Effect Size of Combined Pharmacotherapy versus Monotherapy. Forest plot displaying the mean difference in percentage of abstinent days across studies comparing combined pharmacotherapy and monotherapy. The overall effect favors combined therapy, with a mean difference of 4.045% (95% CI: 0.415% to 7.675%). High heterogeneity was observed (I2 = 99.68%) [25,26,27,28,29,30,31,32,33,34,35,36].
Figure 1. Forest Plot of the Effect Size of Combined Pharmacotherapy versus Monotherapy. Forest plot displaying the mean difference in percentage of abstinent days across studies comparing combined pharmacotherapy and monotherapy. The overall effect favors combined therapy, with a mean difference of 4.045% (95% CI: 0.415% to 7.675%). High heterogeneity was observed (I2 = 99.68%) [25,26,27,28,29,30,31,32,33,34,35,36].
Brainsci 15 00542 g001
Brainsci 15 00542 g002
Figure 2. Funnel plot for publication bias assessment. Funnel plot evaluating publication bias across included studies measuring the effect of combined pharmacotherapy compared to monotherapy on alcohol abstinence rates. The Egger’s test regression line is plotted, with the p-value suggesting no significant asymmetry (p = 0.0823), indicating absence of strong publication bias. Model: weighted regression with multiplicative dispersion. Predictor: standard error Egger’s regression test of funnel plot asymmetry (t = 1.8208, df = 22, p-value = 0.0823). Limit estimate (as sei → 0): b = −0.0111 (CI: −0.0471, 0.0250).
Figure 2. Funnel plot for publication bias assessment. Funnel plot evaluating publication bias across included studies measuring the effect of combined pharmacotherapy compared to monotherapy on alcohol abstinence rates. The Egger’s test regression line is plotted, with the p-value suggesting no significant asymmetry (p = 0.0823), indicating absence of strong publication bias. Model: weighted regression with multiplicative dispersion. Predictor: standard error Egger’s regression test of funnel plot asymmetry (t = 1.8208, df = 22, p-value = 0.0823). Limit estimate (as sei → 0): b = −0.0111 (CI: −0.0471, 0.0250).
Brainsci 15 00542 g002
Table 2. Characteristics of the study sample. United States of America (USA); United Kingdom (UK); Cognitive Behavioral Therapy (CBT); Relapse Prevention (RP); Post-Traumatic Stress Disorder (PTSD); Diagnostic and Statistical Manual of Mental Disorders (DSM); Naltrexone (NTX); Sertraline (SERT); Topiramate (TPM); Quetiapine (QTP); Disulfiram (DSF); Escitalopram (ESC); Acamprosate (ACAM); Citalopram (CIT).
Table 2. Characteristics of the study sample. United States of America (USA); United Kingdom (UK); Cognitive Behavioral Therapy (CBT); Relapse Prevention (RP); Post-Traumatic Stress Disorder (PTSD); Diagnostic and Statistical Manual of Mental Disorders (DSM); Naltrexone (NTX); Sertraline (SERT); Topiramate (TPM); Quetiapine (QTP); Disulfiram (DSF); Escitalopram (ESC); Acamprosate (ACAM); Citalopram (CIT).
AuthorDrug AssociationCountry% MaleComorbidMean Age (Years)Treatment DurationMeasur EDPsycotherapy Suport (Ps)
Helen M Pettinati (2010) [25]NTX + SERTUSA62.40%Major Depression43.314 weekDSM IVCBT
Melissa DelBello (2010) [26]TPM + QTPUSA38.50%Bipolar Mania17.912 weeksDSM IVNo Ps
Ismene Petrakis (2007) [27]NTX + DSFUK97.20%Depression Group and No Depression Group4712 weeksDSM IVCBT
Ismene Petrakis (2005) [28]NTX + DSFUK97.20%Psychiatric Disorders4712 weeksDSM IVCBT
Ismene Petrakis (2006) [29]NTX + DSFUK97.20%PTSD Group and No PSTD Group4712 weeksDSM IVCBT
Conor K Farren (2008) [30]NTX + SERTUSA81.96%No Associated Comorbidities43.1511 weeksDSM IVRP
Janet Witte (2013) [31]ESC + ACAMUSA57%No Associated Comorbidities46.1312 weeksDSM IVCBT
O’Malley SS (2008) [32]NTX + SERTUSA66%No Associated Comorbidities41.216 weeksDSM IVCBT
Raymond F Anton (2006) [33]NTX + ACAMUSA69.16%No Associated Comorbidities44.216 weeksDSM IVCBT Group and No CBT Group
Adamson SJ (2015) [34]NTX + CITUSA40.60%Depression43.612 weeksDSM IVCBT
Josep Guardia (2011) [35]NTX + QTPSpain80.60%No Associated Comorbidities43.812 weeksDSM IVNo Ps
Jacques Besson (1998) [36]ACAM + DSFSwitzerland80%No Associated Comorbidities42.4360 daysDSM IIINo Ps
Table 3. Main results of the included studies divided for drug combinations and comorbid groups. Naltrexone (NTX); Sertraline (SERT); Topiramate (TPM); Quetiapine (QTP); Disulfiram (DSF); Escitalopram (ESC); Acamprosate (ACAM); Citalopram (CIT); Placebo (PLA).
Table 3. Main results of the included studies divided for drug combinations and comorbid groups. Naltrexone (NTX); Sertraline (SERT); Topiramate (TPM); Quetiapine (QTP); Disulfiram (DSF); Escitalopram (ESC); Acamprosate (ACAM); Citalopram (CIT); Placebo (PLA).
AuthorNo. TotalComorbidAssociation Compairedn Intervention n Control% Days Abstinent (Intervention)% Days Abstinent (Control)
InterventionControl Baseline(SD)Outcome(SD)Baseline(SD)Outcome(SD)
Helen M Pettinati (2010) [25]88Major DepressionNTX + SERTNTX + PLA4147--53.7%22%--21.3%10%
Helen M Pettinati (2010) [25]81Major DepressionNTX + SERTSERT + PLA4140--53.7%22%--27.5%11%
Melissa DelBello (2010) [26]39Bipolar ManiaTPM + QTPQTP + PLA1821--14%23%--15%19%
Ismene Petrakis (2007) [27]139Major DepressionNTX + DSFNTX6559--99%2.6%--95.2%13.9%
Ismene Petrakis (2007) [27]139Major DepressionNTX + DSFDSF + PLA6566--99%2.6%--97.1%10.3%
Ismene Petrakis (2007) [27]115No ComorbidNTX + DSFNTX6559--94.9%11.1%--95.8%8.4%
Ismene Petrakis (2007) [27]115No ComorbidNTX + DSFDSF + PLA6566--94.9%11.1%--95.5%11.1%
Ismene Petrakis (2005) [28]124Psychiatric DisordersNTX + DSFNTX6559--96.6%8.7%--95.4%11.8%
Ismene Petrakis (2005) [28]131Psychiatric DisordersNTX + DSFDSF + PLA6566--96.6%8.7%--96.6%10.5%
Ismene L Petrakis (2006) [29]93PTSDNTX + DSFNTX6559--97.8%5.7%--94.1%15.7%
Ismene L Petrakis (2006) [29]93PTSDNTX + DSFDSF + PLA6566--97.8%5.7%--97.7%10.5%
Ismene L Petrakis (2006) [29]161No ComorbidNTX + DSFNTX6559--96.1%9.8%--96.5%7.5%
Ismene L Petrakis (2006) [29]161No ComorbidNTX + DSFDSF + PLA6566--96.10%9.8%--95.9%10.7%
Conor K Farren (2008) [30]111No ComorbidNTX + SERTNTX + PLA575429.3%19.13%79.2%30.48%25.6%18.66%84.5%17.47%
Janet Witte (2013) [31]23No ComorbidESC + ACAMESC + PLA121130%34%44%45%52%42%62%43%
O’Malley SS (2008) [32]67No ComorbidNTX + SERTNTX + PLA333443.2%25.29%96.3%3.08%40.6%26.86%94.8%3.09%
Raymond F Anton (2006) [33]302No ComorbidNTX + ACAMNTX14815422.9%24.7%80.5%25.91%29.8%24.7%80%26.06%
Raymond F Anton (2006) [33]300No ComorbidNTX + ACAMACAM14815222.9%24.7%80.5%25.91%24.6%24.78%75.6%26.01%
Raymond F Anton (2006) [33]312No ComorbidNTX + ACAMNTX15715526.8%24.68%77.6%25.94%23.7%24.78%75.9%26.02%
Raymond F Anton (2006) [33]308No ComorbidNTX + ACAMACAM15715126.8%24.68%77.6%25.94%25.3%24.7%78.2%25.93%
Adamson SJ (2015) [34]138DepressionNTX + CITNTX + PLA736525.5%28.4%68%32%26.1%26.4%59.9%32.1%
Josep Guardia (2011) [35]62No ComorbidNTX + QTPNTX + PLA3032--96.3%7.1%--97.6%4%
Jacques Besson (1998) [36]55No ComorbidACAM + DSFACAM + PLA2431--55%42%--28%38%
Jacques Besson (1998) [36]46No ComorbidACAM + DSFDSF + PLA2422--55%42%--31%30%
Table 4. Model results of the meta-regression for the mean difference from combined pharmacotherapy to monotherapy for the percentual abstinence on AUD.
Table 4. Model results of the meta-regression for the mean difference from combined pharmacotherapy to monotherapy for the percentual abstinence on AUD.
EstimateS.E.Z-Valuep-Value95% CI
%Male−8.65451.3103−6.6048<0.0001−11.2228−6.0863
Mean Age0.25190.09722.59280.00950.06150.4423
Psychiatric Comorbid 0.27660.10912.53620.01120.06280.4903
Region (USA vs. Other) −3.96260.5960−6.6488<0.0001−5.1307−2.794
Associated Naltrexone 2.62210.47875.4776<0.00011.68393.5603
Associated Acamprosate 3.02940.48876.1988<0.00012.07153.9872
Associated Disulfiram 0.81830.36982.21290.02690.09351.5431
Associated Sertaline 4.22620.65686.4348<0.00012.93895.5135
Associated Topiramate 8.62463.06672.81230.00492.614014.6352
Associated Citalopram 0.75290.52991.42080.1554−0.28571.7915
Naltrexone Control 1.87960.33235.6568<0.00011.22842.5308
Acamprosate Control 2.29600.60853.77330.00021.10343.4886
Sertraline Control 3.22630.76244.2318<0.00011.73204.7206
Table 5. Results of the Mixed-Effects Model and heterogeneity test for the mean difference from combined pharmacotherapy to monotherapy for the percentual abstinence on AUD.
Table 5. Results of the Mixed-Effects Model and heterogeneity test for the mean difference from combined pharmacotherapy to monotherapy for the percentual abstinence on AUD.
Mixed-Effects Model (k = 24; tau2 Estimator: REML)
tau2 (estimated amount of residual heterogeneity): 0.0298 (SE = 0.0134)
tau (square root of estimated tau2 value): 0.1726
I2 (residual heterogeneity/unaccounted variability): 99.68%
H2 (unaccounted variability/sampling variability): 314.52
R2 (amount of heterogeneity accounted for): 89.12%
Test for Residual Heterogeneity: QE (df = 10) = 2528.5889, p-val < 0.0001
Test of Moderators (coefficients 2:14): QM (df = 13) = 196.1926, p-val < 0.0001
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Mandaji, J.V.G.; Pozzolo Pedro, M.O.; Leopoldo, K.; Pini Alemar, J.; Torales, J.; Ventriglio, A.; Castaldelli-Maia, J.M. Combination of Drugs in the Treatment of Alcohol Use Disorder: A Meta-Analysis and Meta-Regression Study. Brain Sci. 2025, 15, 542. https://doi.org/10.3390/brainsci15060542

AMA Style

Mandaji JVG, Pozzolo Pedro MO, Leopoldo K, Pini Alemar J, Torales J, Ventriglio A, Castaldelli-Maia JM. Combination of Drugs in the Treatment of Alcohol Use Disorder: A Meta-Analysis and Meta-Regression Study. Brain Sciences. 2025; 15(6):542. https://doi.org/10.3390/brainsci15060542

Chicago/Turabian Style

Mandaji, João Vitor Guimarães, Maria Olivia Pozzolo Pedro, Kae Leopoldo, João Pini Alemar, Julio Torales, Antonio Ventriglio, and João Mauricio Castaldelli-Maia. 2025. "Combination of Drugs in the Treatment of Alcohol Use Disorder: A Meta-Analysis and Meta-Regression Study" Brain Sciences 15, no. 6: 542. https://doi.org/10.3390/brainsci15060542

APA Style

Mandaji, J. V. G., Pozzolo Pedro, M. O., Leopoldo, K., Pini Alemar, J., Torales, J., Ventriglio, A., & Castaldelli-Maia, J. M. (2025). Combination of Drugs in the Treatment of Alcohol Use Disorder: A Meta-Analysis and Meta-Regression Study. Brain Sciences, 15(6), 542. https://doi.org/10.3390/brainsci15060542

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop