REVIEW OF: Cinciripini PM et al, Depress Anxiety 2022;39(5):429–440

STUDY TYPE: Randomized, double-blind, placebo-controlled trial

Like many addictive and psychiatric disorders, tobacco use disorder (TUD) and depression are comorbid. People with TUD and depression have worse health outcomes and have a tougher time quitting than their counterparts without depression. In addition, smoking itself is associated with more severe depressive symptoms. To better understand the safety and efficacy of medications for smoking cessation among depressed individuals, researchers reexamined a subset of data from the landmark Evaluating Adverse Events in a Global Smoking Cessation Study (EAGLES) trial (Anthenelli RM et al, Lancet 2016;387:10037:2507–2520; see CATR November 2019 for more about the EAGLES trial).

This study included 2,635 participants with clinically stable major depressive disorder (MDD) and 4,028 participants without a psychiatric disorder, dubbed the nonpsychiatric cohort (NPC); all participants in both groups smoked 10+ cigarettes daily. Overall, there were slightly more females than males (56% vs 44%), and the average age was 47. As a group, the MDD cohort had smoked longer, had made more attempts to quit, and had more psychiatric comorbidities such as anxiety or history of suicidal ideation. Subjects were randomized to receive either varenicline, bupropion, nicotine replacement therapy (NRT), or placebo for 12 weeks, along with brief smoking cessation counseling. The primary safety outcome was occurrence of psychiatric symptoms such as mood, anxiety, or psychosis, termed neuropsychiatric adverse events (NAE). The primary efficacy outcome was smoking status during the final four weeks of treatment (weeks 9–12) and at six-month follow-up. 

The safety analysis showed no difference between medications in rates of NAEs but did find that the MDD group had an overall higher rate than the NPC. Mild NAEs were common, with 41.1% of the MDD group and 29.5% of the NPC experiencing any NAEs. But they were generally tolerable, with only 1.8% of the MDD group and 0.7% of the NPC experiencing an event that was classified as serious or that led to medication discontinuation.

You might expect that bupropion, being an antidepressant, would be most effective in the MDD group, but surprisingly, that is not what researchers found. In both cohorts, varenicline worked the best, bupropion and NRT were in the middle, and placebo was the least effective. Abstinence rates were slightly lower in the MDD group vs the NPC, but this difference did not meet statistical significance (p=0.101).

Carlat Take 

This study highlights that treatments for smoking cessation are safe and effective for patients with MDD, but the “two birds with one stone” approach of using bupropion for smoking cessation in your depressed patients is not the best way to achieve abstinence. Varenicline along with behavioral counseling is your best bet, with bupropion and NRT remaining viable second-line options.

Erratum: In our last issue, we reported on recent research examining the relationship between alcohol intake and all-cause mortality. We erroneously stated that underreporting alcohol consumption would underestimate mortality risk. Instead, we should have stated that underreporting alcohol consumption would overestimate mortality risk.