Varenicline for Smoking Cessation in Hospitalized Patients With Acute Coronary Syndrome

The EVITA (Evaluation of Varenicline in Smoking Cessation for Patients Post-Acute Coronary Syndrome) trial is a multicenter, double-blind, randomized, placebo-controlled trial examining the efficacy of varenicline for smoking cessation in patients hospitalized with ACS (NCT00794573). Details of the trial study design have been previously described.¹²

In brief, 302 patients who smoked ≥10 cigarettes/d, expressed motivation to quit smoking, and who were hospitalized in either Canada or the United States with an ACS were enrolled at 40 clinical centers (Figure 1). Patient motivation was assessed based on willingness to be randomized in the trial and willingness to attempt to quit smoking. No formal assessment of motivation to quit was performed. An ACS was defined as either a myocardial infarction or unstable angina. Myocardial infarction required the patient to have elevated cardiac enzymes plus ≥1 of the following: 1) ischemic symptoms, 2) electrocardiographic changes consistent with ischemia, or 3) development of pathological Q waves. Unstable angina required the patient to have all of the following: 1) nonelevated cardiac enzymes, 2) ischemic symptoms, and 3) ≥1 coronary lesion of ≥50% severity on a coronary angiogram performed during the hospital admission. Patients were randomized to either varenicline or matching placebo, and treatment was initiated prior to discharge from hospital. Randomization was performed by enrolling center personnel and stratified by center using a computer-generated list of permuted blocks of 2 and 4. The study was approved by an institutional review committee at each clinical center, and all subjects gave informed consent prior to randomization.

Patients were excluded from participating if they were unlikely to be available for follow-up; were pregnant or lactating; had cardiogenic shock or renal impairment at the time of randomization; had hepatic impairment prior to ACS; reported excessive alcohol use; reported current use of marijuana or noncigarette tobacco products; were currently using over-the-counter stimulants or anorectics; had previously used varenicline; or were using a pharmacotherapy for smoking cessation at the time of ACS. Patients were also excluded for a history of neuropsychiatric disorders including suicidal attempts or suicidal ideation, family history of suicide, panic disorder, psychosis, bipolar disorder, dementia, bulimia, anorexia or recent or recurring depression (as indicated by the Patient Health Questionnaire-9 or current antidepressant use). Participants were excluded based on the Patient Health Questionnaire-9 using the diagnostic algorithm for major depressive syndrome: five or more responses (at least one of which included question 1 or 2 of “more than half the days” or “nearly every day” on questions concerning: 1) little interest or pleasure in activities, 2) feeling down or depressed, 3) sleeping too little or too much, 4) fatigue, 5) poor appetite or overeating, 6) feeling bad about yourself, 7) difficulty concentrating, 8) moving too little or too much, and 9) suicidal ideation.¹³

Study-specific screening logs were not mandatory. However, 30 of the 40 centers (responsible for >70% of enrolled participants) contributed logs. Among the 7,768 patients screened at these centers, 7541 were excluded (Figure 1). Half of exclusions (n=3815; 51%) were attributable to patients not meeting the smoking criteria (eg, nonsmoker or smoked < 10 cigarettes/d). The other most common reasons for exclusion were: not meeting ACS criteria (14%), refusal to participate (7%), major depression in the past 5 years, or two or more lifetime episodes of major depression (4%), or not available for follow-up (4%). All other exclusions occurred in less than 5% of participants. Among patients who were approached at these centers and met all eligibility criteria, 31% agreed to participate.

The sample size was estimated assuming a 7 day point prevalence abstinence rate of 24% at 24 weeks in patients receiving placebo. With this assumption, 150 patients per study arm would achieve a >80% power to identify a ≥15% absolute increase in abstinence rates (24% to 39%) using a two-tailed α of 0.05. The EVITA trial was not powered to examine safety end points, but it was designed to describe the occurrence of serious adverse events in this high risk patient population during study treatment and follow-up.

Following randomization, demographic, smoking, and clinical characteristics of patients were collected via a case report form. Severity of dependence on nicotine was assessed by administering the Fagerström Test for Nicotine Dependence.¹⁴ Study medication was begun while patients were hospitalized, with at least the first dose taken prior to discharge. Patients randomized to varenicline received 0.5 mg once/d x 3 days, followed by 0.5 mg twice/d x 4 days, followed by 1.0 mg twice/d for the remainder of the 12 week treatment period. Patients randomized to matching placebo received the same treatment schedule. Patients were provided with bottles of study medication at hospital discharge and at week 4. Patients were instructed to come to follow-up visits with their bottles, and pill counts were conducted at weeks 4 and 12.

Patients in both treatment arms received low-intensity counseling. Counseling involved advice to stop smoking and emphasized the importance of smoking abstinence following ACS. Additional counseling was given during follow-up (telephone contact and clinic visits). Counseling was delivered by research personnel at each of the clinical centers. Counseling scripts were provided as a guide. However, centers with established programs were permitted to use their own counseling techniques. Scripts included discussing smoking history and previous quit attempts, the risks of continued smoking, concerns about quitting, coping with withdrawal symptoms, and providing educational materials. Patients could receive additional counseling outside of the study if they chose to do so. Following completion of the 12-week treatment period, patients who relapsed could use open-label smoking cessation therapies.

Follow-up involved phone calls by research personnel at weeks 1, 2, and 8 as well as clinic visits at weeks 4, 12, and 24. Side effects, symptoms of withdrawal, medication adherence, and smoking status were assessed during follow-up contacts. At follow-up clinic visits, smoking status was assessed by self-report of smoking in the preceding 7 days. Self-reported abstinence was confirmed using biochemical validation via exhaled carbon monoxide levels (Micro 3 and 4 Smokerlyzer, Bedfont Scientific Ltd., Rochester, United Kingdom) with a threshold of exhaled carbon monoxide ≤ 10 ppm.

The primary end point was point prevalence smoking abstinence at 24 weeks following randomization. Point prevalence smoking abstinence was defined by self-report of complete abstinence in the 7 days before the 24 week clinic visit confirmed by a measured exhaled carbon monoxide ≤10 ppm. Participants with self-reported abstinence who had exhaled carbon monoxide values >10 ppm were classified as smoking. Similarly, participants who reported any smoking in the 7 days before the 24 week visit but who had an exhaled carbon monoxide value <10 ppm were classified as smoking. Secondary end points included point prevalence abstinence at other study follow-ups, continuous abstinence, and ≥50% reduction in number of cigarettes smoked/d. Continuous abstinence was defined by self-report of complete abstinence at all follow-up contacts and exhaled carbon monoxide ≤10 ppm at all follow-up visits. In order to be considered continuously abstinent, participants must have completed all follow-up visits. A ≥50% reduction in number of cigarettes smoked/d was defined as reduction of at least 50% in the number of cigarettes smoked/d compared to the patient’s baseline just prior to hospital admission.

Additional secondary end points included measures of adherence to study medication, effectiveness of patient blinding, and use of nonstudy smoking cessation therapy. Other secondary end points included measures of side effects and safety assessments of serious adverse events. Serious adverse events were defined according to the International Conference on Harmonization and Good Clinical Practice guidelines.¹⁵ An adverse event is any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of the trial drug, whether or not considered related to the trial drug. In addition, a serious adverse event is an adverse event that requires inpatient hospitalization or prolongation of existing hospitalization, causes congenital malformation, results in persistent or significant disability or incapacity, is life threatening, or results in death. Any other significant clinical event, not falling into the criteria above, but which in the opinion of the investigator requires reporting, may also be considered a serious adverse event. A major adverse cardiovascular event was defined as death, myocardial infarction, or hospitalization for unstable angina. All serious adverse events were adjudicated by the End Points Evaluation Committee (data provided blinded to treatment assignment), and the trial was monitored by an external Data and Safety Monitoring Board. The Data and Safety Monitoring Board conferred on enrollment of the first participant as well as six additional times during the course of the trial. Safety data were reviewed at each meeting, and a blinded interim analysis was performed after the enrollment of 50% of participants.

An initial descriptive analysis involved an examination of the demographic, smoking, and clinical characteristics of the two treatment groups. The primary data analysis examined point prevalence smoking abstinence at 24 weeks. All analyses were intention-to-treat. Similar to other smoking cessation trials, our analyses assumed that patients who withdrew from the study or who were lost to follow-up returned to smoking at their baseline rate. Secondary data analyses examined other measures of smoking cessation including point prevalence abstinence at 4 and 12 weeks, continuous abstinence, and ≥50% reduction in number of cigarettes smoked/d. Secondary data analyses also examined measures of side effects and adherence to study medication as well as assessments of serious adverse events. Similar to other trials, the primary safety analyses involved measures at 30 days after treatment discontinuation. For all analyses, patients who died were censored at the time of death. In the event of withdrawal from the trial, participants consented for the use of data collected up to the time of withdrawal.

Discrete data are described using proportions, and continuous data are described using means and standard deviations or, in the presence of skewed distributions, medians and 25th to 75th percentiles. Length of stay and length of time from hospital admission to 1^st dose of medication are presented as medians with 25th to 75th percentiles. Rate differences are presented as % differences with corresponding 95% confidence intervals based on the binomial distribution. Fisher’s Exact Test was used for statistical comparisons of rates between groups. Number needed to treat was calculated as the number of ACS patients needed to be treated with varenicline in order to produce 1 nonsmoker at 24 weeks. Statistical analyses were performed using SAS statistical software (Version 9.2).

A total of 302 patients were randomized (mean age 55±9 years; 75% male; 56% ST-segment elevation myocardial infarction; 38% non-ST-segment elevation myocardial infarction; 6% unstable angina; Table 1). Patients smoked a mean of 21±11 cigarettes/d at the time of hospital admission and had been smoking for a mean of 36±12 years. Two-thirds of patients were enrolled in Canada and one-third in the United States. The scores on the Fagerström Test for Nicotine Dependence ranged from 0 to 10, with 80% of patients having scores ≥4 indicating moderate or severe dependence on nicotine. Demographic, smoking, and clinical characteristics were well balanced between the two study arms.

Nearly all patients had cardiac catheterization during the index admission (98%), with 84% undergoing percutaneous coronary intervention and 6% coronary artery bypass graft surgery (Table 1). Comparatively few cases of ventricular arrhythmia, congestive heart failure, or recurrent ischemia occurred during hospitalization. Median length of stay was 3 days, and median time from admission to 1^st dose of study medication was 2 days. Discharge medications were similar between groups.

Over the course of the trial, there was a progressive decline in smoking cessation rates. By 24 weeks, less than one third of patients receiving placebo were abstinent (32.5% point prevalence abstinence; 25.8% continuous abstinence). However, at 4, 12, and 24 weeks, patients randomized to varenicline had significantly higher rates of smoking abstinence and reduction than patients randomized to placebo (Figure 2). For the primary end point of point prevalence smoking cessation at 24 weeks, 47.3% of varenicline patients were abstinent versus 32.5% of placebo patients (% difference 14.8%; 95% CI 3.9, 25.8%; number needed to treat = 6.8). Continuous abstinence rates were significantly different at 4 and 12 weeks, but not at 24 weeks - absolute rates 35.8% and 25.8%, respectively (% difference 10.0%; 95% CI -0.4, 20.4%; number needed to treat = 10.0). Reduction ≥50% in number of cigarettes smoked/d was significantly different at 4, 12, and 24 weeks - absolute rates 67.4% and 55.6%, respectively at week 24 (% difference 11.8%; 95% CI 0.75, 22.7%; number needed to treat = 8.5).

Over the course of the trial, patients received a mean of 46±30 minutes of counseling from study personnel, with an average of 8±4 minutes per follow-up contact. Patients were encouraged to seek supplemental counseling at their study center or elsewhere. However, only 3% of patients (equal in both groups) reported receiving supplementary counseling at any time through week 24.

At the end of the 12-week treatment period, 78% of patients in the varenicline group reported taking ≥1 pill of study medication/d in the previous week compared with 85% of patients in the placebo group (adherence data available for n=218). At all follow-ups, patients who had reduced or discontinued use of the study drug were encouraged to continue participation in the trial. At 12 weeks, patients were asked to guess the treatment group to which they had been assigned. In both the varenicline and placebo groups, 49% of patients guessed correctly. Importantly, patients and study personnel remained blinded to treatment allocation until the conclusion of the trial. A total of 16% of patients reported the use of a nonstudy smoking cessation therapy at any follow-up through week 24 (varenicline 22% versus placebo 11%).

Side effects were common during the 12-week treatment period with the most frequent being insomnia (15.2%) and nausea (11.3%; Table 2). The occurrence of side effects was similar in the two treatment arms with the exception of abnormal dreams (varenicline 15.2% versus placebo 4.6%). Similar numbers of participants in each treatment arm discontinued study treatment because of adverse events (varenicline 11.3% versus placebo 12.8%).

The rates of serious adverse events were similar in the varenicline and placebo arms of the trial (Table 2). There were a total of 39 serious adverse events, occurring in 35 patients, within 30 days of treatment discontinuation. A single neuropsychiatric event occurred involving hospitalization for depression in a patient 25 days after taking one dose of study medication (varenicline). There were no cases of seizure or suicidal ideation.

Among the serious adverse events, 19 events (occurring in 18 patients) were cardiovascular (Tables 2 and 3), with 13 of these events being adjudicated as major adverse cardiovascular events. These events included 2 deaths (does not include a third death which occurred more than 30 days after treatment discontinuation), 6 myocardial infarctions, and 5 cases of unstable angina. Composite major adverse cardiovascular event rates were 4.0% in the varenicline arm and 4.6% in the placebo arm (P>0.99).

Two deaths occurred within 30 days of treatment discontinuation and a third occurred beyond this window, all three in the varenicline arm. The first was attributable to congestive heart failure, the second was attributable to sudden death, and the third was attributable to a perforated ulcer.

In the first patient, study medication was discontinued after 22 days, and death occurred 18 days later. This patient was admitted with an ST-segment elevation myocardial infarction and was found to have an ischemic cardiomyopathy with an ejection fraction of 20%. The patient subsequently developed refractory heart failure and was found not to be a candidate for either heart transplant or a ventricular assist device. Death occurred 40 days after randomization while the patient was still hospitalized.

In the second patient, study medication was presumed to have been taken until the time of death. This patient had a history of chronic lymphocytic leukemia and was admitted with a non-ST-segment elevation myocardial infarction. The patient was found to have an ejection fraction of 35% and underwent a single vessel bypass of the left anterior descending coronary artery using a left internal mammary artery bypass graft. An 80% lesion in an obtuse marginal was not bypassed. No autopsy was performed in this patient.

In addition to these 2 deaths, a third death was reported in a patient in the varenicline arm that occurred outside the 30 day window following treatment discontinuation. This death was attributable to a perforated ulcer and occurred 63 days after study drug discontinuation and 152 days after treatment initiation.

The EVITA trial was designed to assess whether varenicline, begun in-hospital, is efficacious for smoking cessation following ACS. We found that less than a third of smokers hospitalized with an ACS remain abstinent following discharge. However, patients who received varenicline had significantly higher rates of smoking abstinence compared to patients who received placebo. These findings suggest that varenicline, initiated in-hospital following ACS, and in conjunction with low-intensity counseling, is efficacious for smoking cessation. Future studies are needed to establish safety in these patients.