The effects of pharmacological interventions on smoking cessation in people with alcohol dependence: A systematic review and meta-analysis of nine randomized controlled trials.

Background: Pharmacotherapies are widely used for smoking cessation. However, their efficacy for people with alcohol dependence remains unclear.

Objective: This study aimed to explore the effects of pharmacotherapies on smoking cessation for people with alcohol dependence.

Methods: Five electronic databases were searched in January 2021 for randomized controlled trials (RCTs) reporting the use of pharmacotherapies to promote smoking cessation in people with alcohol dependence. The risk of bias was assessed using the Cochrane tool. RevMan version 5.3 was used to perform meta-analyses of the changes in smoking behaviour, and the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) approach was used to assess the certainty of the evidence.

Results: The search identified nine RCTs involving 908 smokers with alcohol dependence; eight were published in the USA and one in Canada. The risk of bias was low in three studies and unclear in the remaining six. The meta-analysis results showed that, compared with the placebo group, Varenicline had a significant effect on short-term smoking cessation (three RCTs, odds ratio [OR] = 6.27, 95% confidence interval [CI]: [2.49, 15.78], P < .05, very low certainty). Naltrexone had no significant effect on smoking cessation in short-term or long-term observations (three RCTs, OR = 0.99, 95% CI: [0.54, 1.81], P = .97, moderate certainty), and Topiramate had no significant effects (two RCTs, OR = 1.56, 95% CI: [0.67, 3.46], P > .05, low certainty). Only one trial reported that Bupropion did not affect smoking cessation.

Conclusion: Varenicline may promote smoking cessation in people with alcohol dependence. However, Naltrexone, Topiramate and Bupropion have no clear effect on increasing smoking abstinence among drinkers. The small number of studies and the low certainty of evidence indicate that the results should be interpreted cautiously.