Of the 20 million Americans who try to quit smoking each year, only about 6 percent continue to abstain from cigarettes in the long term. The availability of nicotine replacements in a patch or gum has helped, but only about 20 percent of persons who use these methods are successful. The observation that a depressed affect correlated with nicotine dependence led to the use of antidepressant therapies to help people quit smoking. Jorenby and colleagues compared bupropion, a nicotine patch, a combination of those two agents, and placebo to determine their efficacy in helping people stop smoking cigarettes.

The study enrolled 893 patients who were at least 18 years of age and smoked at least 15 cigarettes per day. Persons were excluded for a variety of reasons, including cardiac, pulmonary or renal disease, seizure disorder, major depression, pregnancy, use of nicotine replacement in the previous six months or previous use of bupropion. Baseline data included serum cotinine and exhaled carbon dioxide levels, results of depression and nicotine dependence measures. The primary outcome to be assessed was the rate of abstinence at six and 12 months of follow-up. Secondary outcomes included weight gain, withdrawal symptoms and Beck Depression Inventory scores.

The subjects were randomly assigned to receive placebo, a nicotine patch, bupropion or both nicotine patch and bupropion. The length of the trial was nine weeks, with a one-week lead-in period and an assigned quit date on day 8. The bupropion patients received two 150-mg tablets daily, and all other persons received placebo tablets. The nicotine patch groups applied a 21-mg patch on day 8 and continued this daily for eight weeks, followed by a 14-mg patch in the eighth week and a 7-mg patch in the ninth week. All other participants were given placebo patches. Additional supports included weekly counseling sessions and telephone follow-up. Smoking cessation was self-reported but expired carbon monoxide was measured weekly during the nine-week treatment period. Further clinic follow-up and relapse-prevention counseling was done at weeks 10, 12, 26 and 52.

The subjects (58.8 percent female, 41.2 percent male) had smoked an average of 26 cigarettes per day for a mean of 26 years. Of the 893 initial subjects, 311 did not complete the study; however, of these, 134 participated in the follow-up assessments. Persons from the placebo groups had the highest rate of discontinuing treatment (48.8 percent) compared with those who received the combined treatment (28.6 percent). At week 4 of the trial, 48 percent of the nicotine patch group, 60.2 percent of the bupropion group and 66.5 percent of those receiving both therapies were abstinent. The number of persons not smoking at six months was 18.8 percent in the placebo group, 21.3 percent in the nicotine patch group, 34.8 percent in the bupropion group and 38.8 percent in the group receiving the combined therapy. By month 12, 15.6 percent of the placebo group, 16.4 percent of the nicotine patch group, 30.3 percent of the bupropion group and 35.5 percent of the combination-treatment group were not smoking.

At the start of the trial, mean weight was comparable among the groups. By week 7 of the study, persons in the placebo group had gained the most weight, 2.1 kg (4.6 lb) compared with the combination-therapy group, who had the lowest average weight gain, 1.1 kg (2.4 lb). Insomnia was the most common adverse event, reported by 48 percent of the combined-therapy group, 42 percent of the bupropion group, 30 percent of the nicotine patch group and 20 percent of the placebo participants. Another finding of note was that all participants had Beck Depression Scores within the normal range at the onset of the study, and none of the treatments affected follow-up depression scores.

The authors conclude that the use of sustained-release bupropion, with or without a nicotine patch, is significantly more effective for long-term smoking cessation than the use of a nicotine patch alone. The other interesting observation derived from this study was that the use of dual placebos produced a 15.6 percent cessation rate, which is comparable to the 16.4 percent cessation rate achieved by patients who used a nicotine patch.