Principles of smoking cessation. Part 2

Roman Jaeschke: I would like now to turn to our psychiatric colleague with interest in smoking cessation, Professor Zainab Samaan. Zainab, you heard what we were talking about with Francois, about the approaches, chances, motivations, different options [See: Principles of smoking cessation, part I]. We were concentrating on people without underlying psychiatric problems. We were reassured that neither varenicline nor bupropion are inclined to do much neuropsychiatric damage to those people, or none at all. Could you comment on all those issues in respect to the second part of this study, a large part of the study, [which involved] another 4,000 patients who had underlying psychiatric problems. The floor is yours.

Zainab Samaan: Hello Roman, thank you for having me. I am Zainab Samaan. I am a psychiatrist and an Associate Professor at McMaster University. We are talking today about smoking cessation and the different treatments for patients with and without psychiatric disorders. The first part covered patients without psychiatric disorders and the neuropsychiatric manifestations. I am here to talk about smoking cessation and the impact of such treatment on patients with psychiatric disorders.

In general, smoking is a very common problem in the psychiatric population. The patients with psychiatric disorders are 2 to 3 times more likely to smoke compared to the general population. They also have a tremendous amount of comorbidities, including cardiometabolic disorders, making them at a higher risk for the impact of smoking. Therefore, it is very important to identify safer treatments and also assist patients to stop smoking and recover from smoking addiction as much as we can, regardless of the underlying conditions.

What we know so far from evidence is that the 3 different options for smoking cessation – bupropion, varenicline, and nicotine replacement therapy (NRT) – are effective when they are compared to placebo, and they are effective in both psychiatric population and nonpsychiatric population. What we do not know, and the evidence is scant on, is the manifestations of neuropsychiatric side effects of these treatments. We know there were some warnings regarding suicidal behavior with varenicline as well as bupropion and there are seizure disorders and so on, but the evidence remains very limited. What the studies we already know about the smoking cessation for these patients [have in common is that] they usually exclude patients with serious mental disorders, and they do exclude patients who receive concomitant medications, such as antipsychotics or antidepressant treatments. Therefore, the evidence of what we know is not generalizable to the current psychiatric population, because the side-effect profile in healthier population might be different.

This more recent study that was published in The Lancet has done a very good attempt at looking at a psychiatric population and neuropsychological manifestations of the treatment, at the adverse side effects. They compared about 4,000 people without psychiatric disorders with 4,000 people with psychiatric disorders. They reported there were no significant increases in neuropsychiatric side effects when they compared bupropion, varenicline, as well as NRT compared to placebo in this population [with underlying psychiatric problems], similar to what they found in people without psychiatric disorders. Overall, they found that these kinds of adverse effects were present in about 1.5% of the people without psychiatric disorders and about 4% in the people with psychiatric disorders, but they were no different than placebo in both groups. The common side effects reported in this study were nausea, headache, abnormal dreams. They concluded that NRT as well as bupropion and varenicline should be safe to use in the population with psychiatric disorders.

Every study has limitations and this one is no exception. The limitation of this study is that they did exclude patients who have what they consider unstable psychiatric disorders, which means people who are actively having psychopathology, who have medication changes in the last 6 months, and they also excluded people with substance use disorders. We know that smoking and other substances tend to co-occur in patients with psychiatric disorders. They also excluded people who have higher risk of suicidal behavior. If you remember, I mentioned earlier that there was a warning about suicidal behavior with varenicline and bupropion. Excluding patients who could be potentially at higher risk is understandable in the context of a randomized trial; however, it makes the results less generalizable to this population.

Overall, though, if you think about the risk-benefit analysis, NRT, varenicline, and bupropion are effective for the population with nicotine addiction, whether they have a psychiatric disorder or not. In the psychiatric disorder population, there are some cautions we have to take into consideration when we are using these medications. For example, bupropion can lead to increased risk of seizures, especially when it is given with other psychotropic medications. It also increases the plasma level of certain antidepressants, such as tricyclic antidepressants and citalopram, and that may also lead to increasing risk for seizure threshold [lowering] and perhaps cardiac side effects. Varenicline in reviews – [examined in] previous populations, including one of the Cochrane [reviews] – although the evidence was very limited, it did show that there were at least 2 cases out of 137 that showed some suicidal behavior. Because the evidence is so limited on this aspect, we do not know if varenicline does increase the risk of suicidal behavior or not, but we do have to use it with caution.

In conclusion, this study did add quite an important advancement to what we already know; at least they did include patients with psychiatric disorders specifically, albeit [the study] had some limitations, and they showed that NRT, bupropion, and varenicline are effective for both populations. My recommendation would be that we should offer these treatments for patients with psychiatric disorders but we have to keep in mind their side effects and risk may be more than expected for the general population.

Thank you for having me again.