RÉSUMÉ
Objective:To systematically evaluate the efficacy of different kinds of smoking cessation drugs by network Meta-analysis.Methods:Literature was retrieved from PubMed, Web of Science, Embase, Cochrane Library, CBM, CNKI, VIP, Wan fang database, from the establishment of the database to November 2022, and randomized controlled trials (RCT) about bupropion, varenicline, nicotine replacement therapy (NRT) versus placebo in the treatment of smoking patients were collected. After data extraction from included literature which met inclusion criteria, and quality evaluation with Cochrane 5.1 risk bias evaluation tool, network Meta-analysis was performed by Stata15.1 software.Results:A total of 19 RCTs, involving 6106 patients and three interventions measures (bupropion, varenicline, NRT) and one control measure (placebo) were included. The results of network Meta-analysis showed that in terms of short-term abstinence rate, varenicline [ OR=4.21, 95% CI (2.32, 7.63)], bupropion [ OR=2.81, 95% CI(1.05, 7.54)] were better than placebo ( P<0.05). The surface under the cumulative ranking area (SUCRA): varenicline (90.2%)>bupropion (64.8%)>NRT (41.7%)>placebo (3.2%). In terms of the long-term abstinence rate, varenicline [ OR=3.06, 95% CI (1.59, 5.90)], NRT [ OR=3.39, 95% CI (2.20, 5.21)] were better than placebo ( P<0.05). SUCRA: varenicline (83.8%)>NRT (73.9%)>bupropion (37.2%)>placebo (5.2%). Conclusion:The existing evidence shows that compared with bupropion, NRT, varenicline has the best effect on quitting smoking, but more high-quality randomized trial evidence is needed for verification.