The Impact of Aversive Advice During Percutaneous Coronary Intervention on Smoking Cessation in Patients With Acute Coronary Syndrome

Methods: This study was conducted as a prospective, single-blinded, randomized controlled trial. A total of 66 patients were randomly assigned to an aversive advice group or a control group and instructed to visit the outpatient clinic 1, 4, and 24 weeks after discharge. In the aversive advice group, a physician who did not participate in the patient follow-up said the following 3 sentences to the patients during the PCI procedure: “Smoking caused your chest pain”; “If you do not stop smoking right now, this pain will come again”; and “The next time you feel this pain you will probably die.” All patients received usual advice on the importance of quitting smoking.

Smoking and cardiovascular disease have a causal relationship [1]. Smoking cessation is among the most important preventive measures in preventing the development of cardiovascular disease [2,3]. It is well known that smoking cessation reduces the risk of coronary artery disease (CAD) not only in the general population, but also in patients who survive CAD. Patients who quit smoking after the development of CAD have a significantly lower risk of recurrent myocardial infarction or cardiovascular death than do patients who continue smoking [4][5][6].
There are several well-known smoking cessation treatments, including behavioral and pharmacological treatment [7,8]. With optimal treatment, 25% to 35% of smokers who try to quit can succeed for 6 months or more [9]. However, these data were conducted with a generally healthy population, and the generalizability of these data to patients with CAD is unclear [10]. This unclear generalizability is demonstrated by trials that found that the smoking cessation drug bupropion did not increase the prevalence of abstinence in patients with acute coronary syndrome (ACS) [11,12]. This may be caused by characteristic differences of patients, who tend to have higher quit rates because of an increased motivation to quit and the teachable moment that occurs after an adverse event [13]. So, if a patient undergoes an extremely fearful and painful experience, such as a percutaneous coronary intervention (PCI), the effect of aversive advice on smoking cessation might be increased.
Prior studies have investigated the effect of aversive conditioning on smoking cessation [14][15][16]. However, it is unclear whether it possesses any true value in promoting smoking cessation. The previous investigations included the general population and, as is presently understood, no study has evaluated the outcomes of aversive advice in patients with ACS. In this context, the hypothesis presented here is that aversive advice during a PCI procedure is indeed effective for smoking cessation in patients with ACS. Therefore, a randomized controlled study was conducted to evaluate the impact of aversive advice during a PCI procedure on smoking cessation in patients with ACS.

METHODS
In total, 70 current smokers with ACS were identified for enrollment between September 1, 2014, and September 30, 2015 andata for enrolled patients were collected by May 31, 2016. The eligible criteria for inclusion in this study were patients who smoked cigarettes, had ACS, and were scheduled to undergo PCI with consciousness during the procedure. ACS included ST-segment elevation myocardial infarction (STEMI), non-STEMI, and unstable angina. STEMI was defined as chest pain or discomfort in conjunction with elevation of the ST-segment more than 0.1 mV in 2 or more contiguous electrocardiogram leads or a new left bundle branch block with elevated biomarkers of myocardial necrosis (cardiac troponin > upper normal limit). Non-STEMI was defined as chest pain or discomfort in conjunction with elevated biomarkers of myocardial necrosis and 1 of the following: (1) transient ST-segment elevation, transient ST-segment depression, or T-wave changes consistent with myocardial ischemia; or (2) identification of a culprit lesion at coronary angiography. Unstable angina was defined as (1) chest pain or discomfort occurring at rest or with minimal effort, (2) recent onset of chest pain or discomfort (<1 month), or (3) chest pain or discomfort with a recent increase in intensity, frequency, or duration without STsegment elevation on the electrocardiogram or increased cardiac biomarker concentrations. Of these 70 patients, 4 patients declined to participate. Thus, 66 patients were included in the analysis (Fig. 1). The study was approved by the Institutional Review Board of Hanyang University Hospital (HYUH 2014-08-004-001), and written informed consent was obtained after PCI from each patient enrolled. The trial is registered at https://cris.nih.go.kr: KCT0003109.
This study was conducted as a prospective, singleblinded, randomized controlled trial. Before PCI were performed, patients were randomly assigned in a 1:1 ratio to the aversive advice group or the control group. Randomization was carried out using a computer-generated randomization list. In the aversive advice group, a physician who did not participate in the patient follow-up said the following 3 sentences to the patients during the PCI procedure: "Smoking caused your chest pain"; "If you do not stop smoking right now, this pain will come again"; and "The next time you feel this pain you will probably die." To ensure that the investigator was unaware of study assignments, the investigator did not have any involvement in the aversive advice. During hospitalization, all patients received daily general advice about negative effects of smoking and the benefits of smoking cessation. Patients were also instructed to visit the outpatient clinic 1, 4, and 24 weeks after discharge. At every visit, urine nicotine metabolite (cotinine) levels were measured. Cotinine levels <50 ng/ml were considered to denote no active smoking [17,18]. The final assessment of smoking status was performed at 24 weeks after discharge (Fig. 1).
The sample size was determined using information from a preliminary study in which the smoking cessation

RESULTS
The mean age of the subjects was 55.9 AE 9.0 years, and the number of female subjects was 2 (3.0%). Of the total 66 subjects, 30 subjects (45.5%) presented with STEMI, and 44 subjects (66.6%) had a multivessel disease. Baseline characteristics including smoking quantity, disease severity, education level, and cardiovascular risk factors were not significantly different between the 2 groups. Further descriptive data are displayed in Table 1.
At 24 weeks after discharge, smoking cessation rates were higher in the aversive advice group than in the control group, and there were no significant differences in smoking cessation rate with regard to other clinical variables (Fig. 2). Table 2 summarizes the effect of aversive advice during the PCI procedure on smoking cessation at different time points. The effect of aversive advice was significant and consistent over 24 weeks. Table 3 shows the multivariable logistic regression analysis for the effect of aversive advice on change in smoking status. In a univariable model, aversive advice has a significant and positive effect on smoking cessation. After adjustment for age, smoking quantity as assessed by packyears, and alcohol consumption, aversive advice was determined to have a significant effect on smoking cessation (model 1). When the variables of CAD severity, such as STEMI (yes or no), multivessel disease (yes or no), and hospital days !5 (yes or no), were added to model, the result was maintained (model 2).

DISCUSSION
The work presented here demonstrates that aversive advice during PCI has a significant effect on smoking cessation in patients with ACS. In addition, it appears the effect of the aversive advice was maintained for 24 weeks after discharge.
In patients with ACS, the previously reported effects of pharmacological and behavioral interventions on smoking cessation were 31% to 56%, which is quite higher than in the general population [11,12,19,20]. In addition, usual brief care alone has reported up to a 51% of smoking cessation rate in patients with ACS [21]. The relatively higher smoking cessation rate of our study could be derived from characteristics of the study population. Considering that, our study showed that aversive advice is a simple and effective method to increase the rate of smoking cessation.
Many studies pertaining to smoking cessation did not obtain biochemical confirmation of smoking cessation (e.g., thiocyanate, carbon monoxide, or cotinine measurement). Indeed, there are some limitations in these self-reported measures because some respondents are embarrassed about their smoking habits and respond inaccurately [22][23][24]. Using biologic marker studies, Wilcox et al. [23] showed that the true percentage of previous smokers who actually had quit smoking was between 46% and 53%. Therefore, objective methods to estimate smoking status are needed.
There are conflicting results in the published data concerning the effect of aversive conditioning on smoking cessation. A study by Hymowitz and Eckholdt [15], in which the investigators used silver acetate for the aversive conditioning of smokers, showed that aversive conditioning initially brings about smoking cessation, but has no significant effect on long-term smoking cessation. Arzi et al. [14] showed that olfactory aversive conditioning during sleep reduces smoking in short term. On the other hand, Suedfeld and Baker-Brown [16] showed that aversive conditioning is not effective for smoking cessation. Rapid smoking and other aversive smoking methods provide insufficient evidence to determine the efficacy [25]. The present work has demonstrated that aversive conditioning does seem to be effective at smoking cessation and lasts for at least 24 weeks. The main difference between prior studies and this one is the study population; the participants herein consisted of patients with ACS who experienced a scary medical event, possibly rendering the patients more compliant.
In this study, disease severity was not associated with smoking cessation rate. There was no significant difference in smoking cessation rate according to STEMI (yes or no), multivessel disease (yes or no), or hospital days !5 days (yes or no). In a multivariable logistic regression model, disease severity variables did not alter the effect of aversive advice on smoking cessation. The lack of a relationship was unexpected. It could be hypothesized that disease severity, recognized by patients, is shaped by a physician's advice and not by laboratory finding or diagnosis. Here, aversive conditioning was conducted with the same 3 sentences for all aversive advice group patients. Therefore, the aversive advice group patients might recognize their disease as having the same severity.
A previous study showed that smokers have a blunted response to aversive stimuli. The Dinh-Williams et al. study [26], in which functional magnetic resonance image was utilized, showed that aversive smoking-related stimuli elicited lower activation in the brain region associated with aversive processes (e.g., parahippocampal gyrus, insula, and inferior frontal gyrus) than aversive non-smokingrelated stimuli did. However, in the current study, aversion therapy did have a significant effect on smoking cessation. This may be understood based on the fact that the patients exposed to aversive stimuli during the procedure, which included severe pain and fear of death, may have remembered the aversive stimuli more acutely than in other investigations. These findings might also be explained by the fact that the overwhelming experience could be very traumatic to the patients, so much so that they might remember the experience as a very painful and fearful episode. In other words, when the patients feel a craving for nicotine, they might recall the painful and fearful memory of the events such that the aversive emotion could be reconsolidated in the memory every time they get the urge to smoke. This may just be the mechanism of aversive conditioning during a PCI procedure.
This study had a number of limitations. Primarily, because the sample size was relatively small, subgroup analysis could not be performed; a large population study would be needed to assess the contributions of other patient attributes. Another limitation was that the follow-up period was just 24 weeks. To evaluate the long-term effect of aversive advice on smoking cessation, 24 weeks is jg SCIENCE ultimately insufficient and truly long-term follow-up is needed. In addition, aversive advice during a PCI procedure could be considered rather harsh for patients with ACS, but as of the current moment, no patient enrolled in the study has experienced any psychological trauma.

CONCLUSIONS
Aversive advice during a PCI procedure is effective to promote smoking cessation in patients with ACS. More specifically, in a clinical setting, aversive advice during a PCI procedure may be a successful and easy way to encourage patients with ACS to stop smoking, though a protocol that includes a greater period between patient discharge and follow-up is required for future evaluation. Overall, a physician's attention and involvement during a PCI procedure is imperative for smoking cessation in patients with ACS.