|
A correction has been published: N Engl J Med 1999;341(8):610.
|
| |||||||||||||||||||||||||||||||||||||||||||||||
Background and Methods Use of nicotine-replacement therapies and the antidepressant bupropion helps people stop smoking. We conducted a double-blind, placebo-controlled comparison of sustained-release bupropion (244 subjects), a nicotine patch (244 subjects), bupropion and a nicotine patch (245 subjects), and placebo (160 subjects) for smoking cessation. Smokers with clinical depression were excluded. Treatment consisted of nine weeks of bupropion (150 mg a day for the first three days, and then 150 mg twice daily) or placebo, as well as eight weeks of nicotine-patch therapy (21 mg per day during weeks 2 through 7, 14 mg per day during week 8, and 7 mg per day during week 9) or placebo. The target day for quitting smoking was usually day 8.
Results The abstinence rates at 12 months were 15.6 percent in the placebo group, as compared with 16.4 percent in the nicotine-patch group, 30.3 percent in the bupropion group (P<0.001), and 35.5 percent in the group given bupropion and the nicotine patch (P<0.001). By week 7, subjects in the placebo group had gained an average of 2.1 kg, as compared with a gain of 1.6 kg in the nicotine-patch group, a gain of 1.7 kg in the bupropion group, and a gain of 1.1 kg in the combined-treatment group (P<0.05). Weight gain at seven weeks was significantly less in the combined-treatment group than in the bupropion group and the placebo group (P<0.05 for both comparisons). A total of 311 subjects (34.8 percent) discontinued one or both medications. Seventy-nine subjects stopped treatment because of adverse events: 6 in the placebo group (3.8 percent), 16 in the nicotine-patch group (6.6 percent), 29 in the bupropion group (11.9 percent), and 28 in the combined-treatment group (11.4 percent). The most common adverse events were insomnia and headache.
Conclusions Treatment with sustained-release bupropion alone or in combination with a nicotine patch resulted in significantly higher long-term rates of smoking cessation than use of either the nicotine patch alone or placebo. Abstinence rates were higher with combination therapy than with bupropion alone, but the difference was not statistically significant.
Affect or mood appears to exert potent effects on the motivation to use nicotine.3,4,5 For instance, among smokers, symptoms of nicotine dependence are correlated with the magnitude of affective symptoms of depression.6 In population-based studies, smokers are more likely than nonsmokers to have symptoms of affective disorders.6 Persons with a negative affect are more likely to start smoking and less likely to be able to quit7,8,9 — effects that may be related to changes in dopaminergic activity in the brain.10 Antidepressants or anxiolytics may therefore be efficacious cessation aids.3,11 Hurt and colleagues12 demonstrated that bupropion is an effective smoking-cessation aid: at 12 months, the abstinence rates were 23 percent among subjects assigned to receive 300 mg of bupropion per day for 7 weeks and 12 percent among subjects assigned to receive placebo. We compared bupropion, placebo, a nicotine patch,2 and a combination of bupropion and the nicotine patch with regard to efficacy.13 We also examined whether treatment with bupropion ameliorates nicotine-withdrawal symptoms such as negative mood.
Methods
Subjects, Screening, and Randomization
Subjects were recruited at four study sites by advertisements in the media. The first subject was enrolled in August 1995, and follow-up was completed in March 1997. Of a total of 1182 persons who were screened, 893 met the screening criteria and were enrolled: 218 in Arizona, 227 in California, 220 in Nebraska, and 228 in Wisconsin. The subjects were randomly assigned to one of four treatments with use of an unequal-cell design: 160 subjects were assigned to receive placebo, 244 to receive the nicotine patch, 244 to receive bupropion, and 245 to receive bupropion and the nicotine patch. Randomization was not balanced within sites.
The subjects were screened by means of a telephone interview and a pretreatment session that included a physical examination, electrocardiography, and chest roentgenography. The study protocol was approved by the institutional review board at each site. All participants provided written informed consent.
To be eligible for the study, subjects had to be at least 18 years of age, to smoke at least 15 cigarettes per day, to weigh at least 45.4 kg (100 lb), to be motivated to quit smoking, and to speak English. Only one smoker per household was allowed to enroll in the study. Subjects were excluded for the following reasons: serious or unstable cardiac, renal, hypertensive, pulmonary, endocrine, or neurologic disorders, as assessed by the study-site physician; ulcers; seizure or dermatologic disorders; a current diagnosis of major depressive episode or a history of panic disorder, psychosis, bipolar disorder, or eating disorders; use of a nicotine-replacement therapy within six months before study enrollment; pregnancy or lactation; abuse of alcohol or a non–nicotine-containing drug within the preceding year; use of a psychoactive drug within the week before enrollment; use of an investigational drug within the month before enrollment; prior use of bupropion; current use of other smoking-cessation treatments; and regular use of any noncigarette tobacco product.
Treatment Period
The treatment period was nine weeks. Target quitting dates were set for the second week, usually day 8. Participants were assessed weekly and attended a brief (15 minutes or less) individual counseling session for smoking cessation each week. Counseling topics included motivation, identification of smoking triggers, coping responses, weight management, and use of the medications. The counselors used a standardized treatment developed by Hurt and colleagues.12 The subjects also received a supportive telephone call from a counselor approximately three days after the target quitting date.
Follow-Up Period
Follow-up assessments and relapse-prevention counseling occurred during clinic visits 10, 12, 26, and 52 weeks after the start of the study. In addition to clinic visits, subjects received eight telephone calls from a counselor during this period, one per month in months 3, 4, and 5 and 7 to 11. All follow-up counseling was less than 10 minutes in duration per call.
Medications
Subjects in the two bupropion groups received 150-mg tablets of sustained-release bupropion (Zyban, Glaxo Wellcome), and all other subjects received identical-appearing tablets. In the bupropion groups, subjects received 150 mg of bupropion in the morning and a placebo tablet in the evening on days 1, 2, and 3 of treatment; and one bupropion tablet in the morning and one in the evening on days 4 to 63. All other subjects took placebo tablets twice daily from days 1 to 63. Subjects in the nicotine-patch groups used one patch (Habitrol, Novartis Consumer Health) per day for eight weeks beginning on the quitting day (day 8). All other subjects applied a placebo patch each day for eight weeks. The patches used from weeks 2 to 7 each contained 21 mg of nicotine; those used during week 8 each contained 14 mg, and those used during week 9 each contained 7 mg.
Assessments
At base line, serum cotinine, vital signs, and exhaled carbon monoxide were determined; data on smoking history were obtained; and three questionnaires were administered. The portion of the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV), concerning mood disorders was used to assess whether subjects had mood disorders. The Beck Depression Inventory14 assesses the severity of depression. Scores of 0 to 9 are considered to be normal, scores of 10 to 18 indicate mild-to-moderate depression, scores of 19 to 29 indicate moderate-to-severe depression, and scores of 30 to 63 indicate severe depression. The Fagerström Tolerance Questionnaire15 measures nicotine dependence. Scores can range from 0 to 11, with higher scores indicating more severe dependence.
During the treatment period, vital signs were assessed and the carbon monoxide content of expired air was measured. All subjects were asked to keep a daily diary for the first 12 weeks of the study that included information on smoking status, craving, and withdrawal symptoms. During the follow-up period, the Beck Depression Inventory was given, vital signs and the carbon monoxide content of expired air were measured, and self-reported smoking status was assessed.
Measures of Outcome
All 893 subjects were included in analyses of the primary outcome. The primary outcome variable was the point-prevalence rate of abstinence at 6 and 12 months of follow-up. Subjects were considered to be abstinent if they reported not smoking since the preceding clinic visit and had an expired carbon monoxide concentration of 10 ppm or less. Subjects were considered to be continuously abstinent if they had not smoked after the quitting day, as confirmed by a carbon monoxide concentration of 10 ppm or less at all clinic visits during the 12-month study. Secondary outcome measures included withdrawal symptoms, body weight, and Beck Depression Inventory scores.
Statistical Analysis
Chi-square and analysis of variance were used to test for base-line differences in demographic and smoking-history variables.16 All statistical tests were two-sided and had an alpha level of 0.05. Sample sizes were based on the results of a previous study of bupropion in which the abstinence rates at four weeks were 40 percent in the bupropion group and 24 percent in the placebo group.12 We estimated that 130 subjects were needed in the placebo group and 230 subjects were needed in the treatment groups for the study to have a power of 0.80 to detect such a difference at an alpha level of 0.05. All subjects who discontinued treatment early or who were lost to follow-up were classified as smokers.
Logistic-regression analysis17 was used to determine pairwise differences among groups in the abstinence rates. The Kaplan–Meier method was used to analyze differences in rates of continuous abstinence; homogeneity among treatments and pairwise differences were tested with the log-rank test.16
Withdrawal symptoms were assessed daily with a composite score calculated as the mean of eight items in the daily diary: craving for cigarettes; restlessness; increased appetite; depressed mood; anxiety; difficulty concentrating; irritability, frustration, or anger; and difficulty sleeping (DSM-IV symptoms plus craving).18,19 The severity of each symptom was rated on a five-point scale, as absent (0), slight (1), mild (2), moderate (3), or severe (4). Repeated-measures analysis of variance was used to analyze the change in scores from base line (before smoking cessation) to after smoking cessation. Group coding was used that permitted tests of the independent and interactive effects of the two pharmacotherapies. In one analysis, the changes in scores during the first six days after the quitting date were analyzed; in a second analysis, the changes in scores during each week of the eight-week period after the quitting date were analyzed. To control experiment-wise error, Tukey's studentized range test16 was used for pairwise group comparisons of changes in scores that were found to be significantly different; this same strategy was used to analyze body weight and Beck Depression Inventory scores. Adverse events that began or increased during the treatment phase were coded with COSTART (Coding Symbols for Thesaurus of Adverse Reaction Terms),20 and differences between groups were tested by Fisher's exact test.
Results
Base-Line Characteristics and Rates of Discontinuation
The base-line characteristics of the study subjects are shown in Table 1. There were no significant differences among the groups. There were no significant interactions between site and treatment for the point-prevalence rates of abstinence at 6 and 12 months.
|
Abstinence Rates
Figure 1A shows the point-prevalence rates of abstinence from smoking, as confirmed by biochemical tests. The point-prevalence rates of abstinence at four weeks were significantly higher in all three treatment groups (48.0, 60.2, and 66.5 percent for the nicotine-patch group, bupropion group, and combined-treatment group, respectively) than in the placebo group (33.8 percent; P=0.005, P<0.001, and P<0.001, respectively). Thereafter, only the bupropion group and the combined-treatment group had significantly higher point-prevalence rates of abstinence than the placebo group (Figure 1A and Table 2).
|
|
Symptoms of Withdrawal and Depression
Figure 2A and Figure 2Bshows the mean changes in withdrawal symptoms from base line for all subjects (regardless of whether or not they were smoking). The changes were analyzed daily during the first six days after the quitting date and then weekly until the end of treatment. All four groups had significant increases in withdrawal symptoms during the first week of treatment (P<0.001). However, the changes were smaller in the three active-treatment groups than in the placebo group during the first six days after the quitting date and during the following weeks.
|
Weight Change
At the beginning of treatment, there were no significant differences in mean body weight among the four groups (Table 1). By week 7 (after which the nicotine-patch dose was decreased from 21 to 14 mg per day), subjects in the placebo group had gained an average of 2.1 kg, as compared with a gain of 1.6 kg in the nicotine-patch group, 1.7 kg in the bupropion group, and 1.1 kg in the combined-treatment group. Pairwise group comparisons by Tukey's studentized range test at week 7 indicated that the subjects in the combined-therapy group had gained significantly less weight than those in the placebo group (P<0.05) or the bupropion group (P<0.05). There were no significant differences between groups in mean weight changes after week 7.
Safety
Table 3 shows the adverse events reported by 10 percent or more of the subjects in any of the groups. Insomnia was the most commonly reported adverse event, occurring among 47.5 percent of the subjects in the combined-treatment group, 42.4 percent of those in the bupropion group, 30.0 percent of those in the nicotine-patch group, and 19.5 percent of those in the placebo group. Reactions at the application site and dream abnormalities were most common among the subjects who used the nicotine patch.
|
Five serious adverse events were reported during treatment. Three were dermatologic or allergic reactions in subjects who were taking bupropion, one of whom was also using a nicotine patch. All three had rash and pruritus, and one also had shortness of breath and chest tightness. The symptoms began 14 to 20 days after the start of therapy. Treatment was stopped, and the three subjects received glucocorticoids and antihistamines. All had full resolution of symptoms. These reactions were attributed to bupropion.
The two other serious adverse events consisted of viral spinal meningitis in a 38-year-old woman 60 days after the initiation of nicotine-patch therapy and chest pain in a 46-year-old man who was hospitalized 4 days after beginning bupropion therapy. He was discharged one day later with a diagnosis of gastric reflux, and the symptoms resolved after treatment with omeprazole. The meningitis and gastric reflux were not attributed to the study medications.
Discussion
We found that treatment with bupropion alone or in combination with a nicotine patch resulted in higher long-term abstinence rates than did the use of placebo or a nicotine patch alone. Treatment with both bupropion and the nicotine patch was not significantly better than treatment with bupropion alone either at the end of the treatment period or during follow-up. As compared with the use of placebo, treatment with the nicotine patch, the nicotine patch and bupropion, and bupropion alone all resulted in less severe withdrawal symptoms and less weight gain after smoking cessation. Previous research has also shown that bupropion and nicotine-replacement therapies can reduce weight gain after smoking cessation.12,21,22,23 Although weight gain was lowest in the combined-treatment group, there were no significant differences in weight gain among the groups after week 7 of treatment.
The subjects in our study were all volunteers and thus may not be representative of the majority of smokers.24 Moreover, all subjects underwent weekly biochemical tests to determine whether they were still smoking. Both these factors could have enhanced cessation rates. The fact that 19.8 percent of the subjects dropped out of the study must also be considered. Those who dropped out of the study were assumed to have resumed smoking, but data on factors such as weight, depression, and severity of withdrawal had to be treated as missing for these subjects, so the potential contribution of these factors remains unknown.
Analyses of data on continuous abstinence showed that relative to placebo, use of the nicotine patch was associated with higher abstinence rates during the 12-month follow-up period. The odds ratio for the comparison between the nicotine patch and placebo at one year was 1.1, similar to values reported in previous work.2 However, analyses of point-prevalence data showed no significant differences between these two groups during follow-up. It is unclear why the nicotine patch produced weak effects according to the point-prevalence analysis. One study suggested that the use of two placebos in a control group may produce higher smoking-cessation rates than the use of a single placebo.25 This might account for the smaller difference in the long-term rates of smoking cessation between the placebo group and the nicotine-patch group in our study. The weak effects, however, seem unrelated to prior use of the nicotine patch. The rate of previous use of a nicotine patch was similar among the four groups. In the nicotine-patch group, there was no significant difference in the rates of continuous abstinence at 12 months between subjects who had previously used patches and those who had not (8.6 percent vs. 10.6 percent, P=0.61).
Supported by a grant from Glaxo Wellcome.
Dr. Jorenby has organized medical-education presentations sponsored by Glaxo Wellcome and SmithKline Beecham. Dr. Leischow has served as a consultant for McNeil Consumer Products, Pharmacia & Upjohn, and Glaxo Wellcome and has organized medical-education presentations sponsored by Glaxo Wellcome. Dr. Nides has served as a consultant for Glaxo Wellcome, Novartis, and SmithKline Beecham and has organized medical-education presentations sponsored by Glaxo Wellcome. Dr. Rennard has served as a consultant for Glaxo Wellcome, Novartis, and SmithKline Beecham and has organized medical-education presentations sponsored by Glaxo Wellcome. Dr. Muramoto has organized medical-education presentations sponsored by Glaxo Wellcome. Mr. Daughton has served as a consultant for SmithKline Beecham and Hoechst Marion Roussel and has organized medical-education presentations sponsored by Glaxo Wellcome and Hoechst Marion Roussel. Dr. Fiore has served as a consultant for Novartis, Glaxo Wellcome, SmithKline Beecham, and McNeil Consumer Products and has organized medical-education presentations sponsored by Novartis, Elan Pharma, Lederle Laboratories, Glaxo Wellcome, McNeil Consumer Products, and SmithKline Beecham. Dr. Baker has served as a consultant for SmithKline Beecham and has organized medical-education presentations sponsored by Elan Pharma and Glaxo Wellcome.
Source Information
From the Center for Tobacco Research and Intervention, University of Wisconsin Medical School, Madison (D.E.J., S.S.S., M.C.F., T.B.B.); the Arizona Program for Nicotine and Tobacco Research, University of Arizona, Tucson (S.J.L., M.L.M.); Los Angeles Clinical Trials, Los Angeles (M.A.N., K.D.); the Pulmonary and Critical Care Medicine Section, University of Nebraska Medical Center, Omaha (S.I.R., D.M.D.); and Glaxo Wellcome, Research Triangle Park, N.C. (J.A.J., A.R.H.).
Address reprint requests to Dr. Jorenby at 1300 University Ave., Rm. 7278 MSC, Madison, WI 53706.
References
| |||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Smoking Cessation
Hughes J. R., Jorenby D. E., Fiore M. C., Baker T. B.
Extract |
Full Text
N Engl J Med 1999;
341:610-611, Aug 19, 1999.
Correspondence
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
