The optimal approach to encourage smoking cessation after acute coronary syndrome (ACS) remains unclear. The safety of nicotine replacement therapy (NRT) after ACS is not well established. The aim of the present study was to determine the relationship between NRT use and adverse cardiovascular outcomes after ACS. Using a pre-existing database, 663 smokers with ACS were identified. The patients were separated into the NRT (n = 184) or control (n = 479) groups according to whether NRT was prescribed on hospital discharge. Multivariate logistic regression analysis was used to account for the baseline differences between the 2 groups. Of the 663 patients, 202 had adverse events in the first year after ACS. No significant differences were seen with NRT use for the 1-year combined end point of death, myocardial infarction), repeat revascularization, or rehospitalization for angina, congestive heart failure or arrhythmia (odds ratio [OR] 0.89, 95% confidence interval [CI] 0.61 to 1.30, p = 0.54). There were no differences in the individual 1-year end points of death (odds ratio 0.80, 95% confidence interval 0.33 to 1.91, p = 0.61), myocardial infarction (odds ratio 0.90, 95% confidence interval 0.40 to 2.06, p = 0.80), repeat revascularization (odds ratio 0.77, 95% confidence interval 0.44 to 1.36, p = 0.37), or rehospitalization for angina, congestive heart failure, or arrhythmia (odds ratio 1.01, 95% confidence interval 0.66 to 1.53, p = 0.97). In conclusion, NRT use was not associated with an increased risk of adverse cardiovascular events in the first year after ACS.
The ideal approach to encouraging smoking cessation in smokers remains unclear. However, some question remains regarding the safety of nicotine replacement therapy (NRT) immediately after acute coronary syndrome (ACS). Recently, the safety of the use of varenicline, an alternative pharmaceutical agent to assist with smoking cessation, in patients with coronary artery disease (CAD) has been called into question, limiting the available smoking cessation aids for this population. Previous studies have demonstrated the safety of the nicotine patch in patients without documented CAD and outpatients with stable CAD ; however, the safety and effectiveness of NRT for patients immediately after discharge from ACS remains unclear. We sought to determine the relation between NRT and adverse cardiovascular outcomes in smokers after ACS.
Methods
Using data from a pre-existing database, all patients who were current tobacco (including cigarettes, cigars, pipes, or smokeless) users who presented with an ACS (including unstable angina pectoris, non–ST-segment elevation myocardial infarction [MI], and ST-segment elevation MI) and underwent cardiac catheterization from January 2006 to June 2010 were identified. The patients who died during the index hospitalization and the patients who were current users of NRT on admission were excluded from the present analysis. The baseline characteristics, including demographic data, co-morbid conditions, revascularization status, and hospital discharge medications, were gathered from the database and confirmed by electronic chart review. The University of Rochester research subjects review board approved the study.
The patients were divided into subgroups depending on whether they were discharged with a prescription for NRT (including nicotine patch, lozenges, gum, nasal spray, and inhalers) or not, as documented in the electronic health record. The outcome measures included all-cause death, MI, repeat revascularization, or rehospitalization for angina, arrhythmia, or congestive heart failure at 1 year. The prespecified primary end point was a composite of these outcomes, and the individual components were analyzed separately as secondary end points. The outcomes were identified using electronic chart review of the patients’ inpatient and outpatient records and a search of publicly available death records. When these sources could not firmly establish whether an outcome had been reached, the patients were interviewed by telephone.
Power calculations were performed based on the nonsignificant relative risk reduction seen previously, as applied to an incidence of adverse events of 23% in the first year after ACS. We estimated a sample size of 918 patients to ensure 90% power to detect a statistically significant benefit with NRT use. Univariate analyses between those with NRT versus those without were performed for all baseline characteristics using the chi-square test for categorical variables and the nonparametric Wilcoxon rank-sum test for continuous variables. Odds ratios for NRT use after ACS were estimated using multivariate logistic regression analysis. The best subsets method was used to select which clinical covariates were entered in the multivariate model. The subsets of patients with various revascularization strategies, (percutaneous intervention vs bypass surgery vs medical management) were analyzed, and adjustments for various medications were tested in the model to evaluate their statistical effect on NRT use. All p values were 2-sided, and p <0.05 was considered significant. The analyses were conducted with SAS software, version 9.2 (SAS Institute, Cary, North Carolina).
Results
A total of 904 patients were identified from the database. Of these 904 patients, 20 died during the index hospitalization and 221 (24.4%) were lost to follow-up or had incomplete data. These patients were excluded from the analysis. Data from the remaining 663 patients were analyzed, and categorized by NRT status on discharge from the index hospitalization. Of the 663 patients, 184 (27.7%) were in the NRT group and 479 (72.2%) in the no-NRT (control) group; 146 patients were interviewed directly because of incomplete electronic data. The baseline characteristics of the 2 groups are listed in Table 1 . The revascularization strategy and discharge medications of the 2 groups are listed in Table 2 .
| Variable | NRT (n = 184) | Control (n = 479) | p Value |
|---|---|---|---|
| Age (years) | 55 ± 10 | 56 ± 11 | 0.148 |
| Women | 62 (34%) | 151 (32%) | 0.592 |
| White | 152 (83%) | 406 (85%) | 0.497 |
| Black | 22 (12%) | 51 (11%) | 0.630 |
| Hypertension (systolic blood pressure >140 mm Hg or therapy) | 125 (68%) | 312 (65%) | 0.496 |
| Dyslipidemia (low-density lipoprotein >160 ng/dl or therapy) | 111 (60%) | 280 (58%) | 0.661 |
| Diabetes mellitus | 41 (22%) | 121 (25%) | 0.424 |
| Heart failure | 7 (4%) | 28 (6%) | 0.290 |
| Previous myocardial infarction | 37 (20%) | 115 (24%) | 0.285 |
| Previous coronary artery bypass | 11 (6%) | 34 (7%) | 0.608 |
| Previous percutaneous intervention | 31 (17%) | 98 (20%) | 0.293 |
| Current dialysis | 2 (1%) | 6 (1%) | 1.000 |
| Peripheral arterial disease | 13 (7%) | 51 (11%) | 0.160 |
| ST-segment elevation myocardial infarction | 81 (44%) | 185 (39%) | 0.204 |
| Non–ST-elevation myocardial infarction | 76 (41%) | 191 (40%) | 0.737 |
| Acute coronary syndrome without myocardial infarction | 27 (15%) | 103 (22%) | 0.047 |
| Variable | NRT (n = 184) | Control (n = 479) | p Value |
|---|---|---|---|
| Percutaneous intervention | 145 (79%) | 305 (64%) | <0.001 ⁎ |
| Coronary artery bypass | 5 (3%) | 67 (14%) | <0.001 ⁎ |
| Aspirin | 178 (97%) | 457 (95%) | 0.445 |
| Clopidogrel | 165 (90%) | 372 (78%) | <0.001 ⁎ |
| β Blockers | 172 (93%) | 443 (92%) | 0.658 |
| Angiotensin-converting enzyme inhibitors or angiotensin receptor blockers | 167 (91%) | 403 (84%) | 0.028 ⁎ |
| Statins | 177 (96%) | 443 (92%) | 0.082 ⁎ |
The results were adjusted for the independent variables of dialysis status, gender, previous coronary artery bypass grafting, and race by multivariate logistic regression analysis using the r-square best subsets method, because these variables showed interaction with the primary end point. The remainder of the measured variables, including age, other baseline characteristics, revascularization strategy, and differences in discharge medications, were not statistically significant predictors in the multivariate regression model. More patients in the NRT group underwent percutaneous intervention, and this group had a greater cardiovascular medication rate; however, these were not statistically significant predictors in the multivariate regression model.
A total of 202 patients had an adverse outcome in the first year after discharge, including 53 (29%) in the NRT group and 149 (31%) in the control group. NRT use was not significantly predictive of the 1-year combined end point of death, MI, repeat revascularization, or rehospitalization for angina, congestive heart failure, or arrhythmia. No differences were seen in the individual 1-year end points of death, MI, repeat revascularization or rehospitalization for angina, congestive heart failure, or arrhythmia. The results are summarized in Table 3 .
