What is the next step in treating a patient with bipolar disorder who wants to quit smoking, but has failed to do so with nicotine replacement? I am concerned I will make the person manic with bupropion, and I am nervous about all the behavioral and psychiatric side effects of varenicline.

Aggressively treating smoking in persons with bipolar disorder is important but is often overlooked. The rates of smoking in bipolar disorder are second only to the rates of smoking in schizophrenia. It has been well established that people with severe mental illnesses such as bipolar disorder live, on average, 25 years less than the general population and no doubt this is due in part to modifiable risk factors such as tobacco dependence.

Nicotine replacement therapy (NRT) is often the first-line treatment for people who wish to quit smoking. This can include patches only, gum only, lozenges only, nasal spray, inhaler, or some combination of a patch with an additional as-needed nicotine replacement medication. A nicotine patch combined with as-needed nicotine replacement medication nearly doubles the odds of quitting versus use of a patch alone (1).

One of the most common errors in prescribing NRT is underdosing. This is particularly a problem in heavy smokers such as those with bipolar disorder or schizophrenia who may be smoking 2—3 packs of cigarettes per day. Strategies for these heavier smokers can include the use of multiple nicotine patches or having the person cut back their smoking to 1 pack of cigarettes per day before embarking on cessation.

Another problem occurs in people who choose to quit with gum only or lozenges only if they do not use the products as directed. Both the gum and lozenge are designed to allow nicotine to be absorbed through the mucus membrane, because it is not well absorbed through the gut. Nicotine gum needs to be self-administered by the "chew and park" method, not chewed as one would regular gum. Nicotine lozenges should not be chewed, and liquids, particularly acidic beverages such as coffee, should not be consumed when they are in use, again to prevent all the nicotine from going into the gut rather than the bloodstream. For the patient presented above, it may be worth determining whether the NRT was administered in sufficient quantity and in the correct manner before excluding another trial of NRT (2).

Bupropion, also known by the brand names Wellbutrin and Zyban when it is used for smoking cessation, can be used as monotherapy or in combination for smoking cessation. The combination of an NRT and bupropion for smoking cessation is more effective than either agent alone (1). Bupropion is often used for treatment of depressed mood in persons with bipolar disorder because it is thought to be less likely to increase mood cycling, although it is not Food and Drug Administration-approved for the treatment of bipolar depression. The decision to use bupropion for smoking cessation can be made in much the same way as the decision to use it for mood would be made. In general, it would not be a good choice for people who have bipolar illness characterized by frequent or very severe manias or those with rapid cycling, but it may be a reasonable choice for someone who has bipolar disorder that is characterized by having predominantly depressive episodes and infrequent mood switches. However, it is important to remember that bupropion has a Food and Drug Administration black box warning for mood and behavior changes, although it has not received as much attention as the warning for varenicline.

Varenicline, also known by the brand name Chantix, has received a lot of attention because of potential risks of changes in behavior, hostility, agitation, depressed mood, suicidal thoughts and behavior, and attempted suicide. Certainly the use of varenicline in patients with serious psychiatric illness such as schizophrenia and bipolar disorder may result in worsening of their preexisting psychiatric illness, but varenicline can still be a relatively safe and effective medication in this population when used appropriately. Smoking in patients with bipolar disorder is often highly treatment refractory, and they may have made numerous previous attempts to quit with no success. Varenicline provides new hope for successful smoking cessation.

Of patients with bipolar illness who wish to quit smoking, the best candidates for varenicline are those whose disorder is currently stable—generally that will mean individuals are either euthymic or in a subsyndromal depression of a duration of at least 6 months. These patients are generally able to monitor for mood or other impulse changes, which experience suggests seem to occur in the first week of treatment if they are going to occur. Patients should also be warned about nausea, which is a common reason for stopping varenicline. NRTs are not used in conjunction with varenicline because of the nicotine antagonist properties of varenicline, although it is safe for a individual to continue to smoke while taking varenicline. Many people who are not ready to stop smoking on their quit date find that as they continue to take the varenicline their smoking trails off and stops. This outcome is rewarding for doctor and patient and a good reason to use varenicline where clinically appropriate.