Electrocardiographic Changes Associated with Smoking and Smoking Cessation: Outcomes from a Randomized Controlled Trial

Introduction Cardiovascular disease (CVD) can be detected and quantified by analysis of the electrocardiogram (ECG); however the effects of smoking and smoking cessation on the ECG have not been characterized. Methods Standard 12-lead ECGs were performed at baseline and 3 years after subjects enrolled in a prospective, randomized, placebo-controlled clinical trial of smoking cessation pharmacotherapies. ECGs were interpreted using the Minnesota Code ECG Classification. The effects of (i) smoking burden on the prevalence of ECG findings at baseline, and (ii) smoking and smoking cessation on ECG changes after 3 years were investigated by multivariable and multinomial regression analyses. Results At baseline, 532 smokers were (mean [SD]) 43.3 (11.5) years old, smoked 20.6 (7.9) cigarettes/day, with a smoking burden of 26.7 (18.6) pack-years. Major and minor ECG criteria were identified in 87 (16.4%) and 131 (24.6%) of subjects, respectively. After adjusting for demographic data and known CVD risk factors, higher pack-years was associated with major ECG abnormalities (p = 0.02), but current cigarettes/day (p = 0.23) was not. After 3 years, 42.9% of subjects were abstinent from smoking. New major and minor ECG criteria were observed in 7.2% and 15.6% of subjects respectively, but in similar numbers of abstinent subjects and continuing smokers (p>0.2 for both). Continuing smokers showed significant reduction in current smoking (–8.4 [8.8] cigarettes/day, p<0.001) compared to baseline. Conclusions In conclusion, major ECG abnormalities are independently associated with lifetime smoking burden. After 3 years, smoking cessation was not associated with a decrease in ECG abnormalities, although cigarettes smoked/day decreased among continuing smokers.


Introduction
Cigarette smoking contributes to the loss of over 5 million lifeyears annually in the United States [1]. Nearly 20% of all coronary heart disease deaths can be attributed to smoking [1,2]. Smoking, therefore, is one of the most important modifiable risk factors for cardiovascular disease (CVD) and myocardial infarction [3].
Cigarette smoking leads to hypoxemia and endothelial dysfunction which accelerate atherosclerotic changes, increasing smokers' risk for CVD [4,5]. The 12-lead electrocardiogram (ECG) is a routine, inexpensive tool for assessment of CVD and CVD risk in both clinical and research settings, and ECG changes powerfully predict future CVD events [6,7]. Large, population-based studies have described a higher prevalence of smoking in subjects with ECG abnormalities and demonstrated that ECG abnormalities are associated with increased CVD events and death [8,9]. Although the strong relation between smoking and CVD is well-established [10,11], there are relatively few data that prospectively described ECG changes following a smoking cessation attempt. Furthermore, most of the existing data about CVD and smoking is from older cohorts that are not representative of today's smokers, who tend to be more overweight, are more likely to be female, and to have lower socioeconomic status [12]. To our knowledge, no studies have described the prevalence of ECG abnormalities in a younger group of cigarette smokers or their associations with smoking burden. The primary aim of this study was to describe the prevalence and magnitude of resting ECG abnormalities among smokers and to explore the relation between smoking burden and ECG abnormalities. A secondary aim was to determine if successful smoking cessation reduced, or if continued smoking increased the incidence of ECG abnormalities that indicate structural or conduction system function, after three years.

Ethics Statement
The institutional review board at the University of Wisconsin School of Medicine and Public Health approved this study. It was conducted according to the principles expressed in the Declaration of Helsinki. All subjects provided written informed consent.

Study Participants and Design
All subjects agreed to participate in a randomized, double blind, placebo-controlled trial to evaluate the efficacy of smoking cessation pharmacotherapies and natural history of continued smoking and smoking cessation (clinicaltrials.gov registration no. NCT00332644) [13]. Specific recruitment strategies have been described previously [13]. This article describes a pre-specified secondary analysis of the data from the baseline and year 3 (final) study visits regarding the effects of smoking and smoking cessation on ECG changes at baseline and 3 years after the target quit date.
Major inclusion criteria were: age $18 years, smoking $10 cigarettes/day (cpd), expired carbon monoxide (CO) level .9 ppm, and stated motivation to quit smoking. Exclusion criteria have been reported previously [13]; the major reasons for exclusion were blood pressure (BP) .160/100 mmHg, myocardial infarction within the previous 4 weeks, heavy alcohol use, history of seizure or serious head injury, use of contraindicated medications, and current pregnancy or breast-feeding [13].

Study Procedures
Subjects were recruited from communities in and around Madison and Milwaukee, Wisconsin from January, 2005 to June, 2007 [13]. Only subjects recruited from Madison, Wisconsin had resting ECGs obtained prior to an exercise stress test that was used to evaluate functional capacity [14]. Of those participants, 541 had a resting ECG available for coding; 3 did not include a subject ID and 6 could not be coded due to tracing artifact. Similarly, of the 370 subjects at 3 years of follow up, 355 ECGs were available for coding and 1 could not be coded due to tracing artifact. The baseline clinical trial visits included measurement of anthropometric data, fasting laboratory tests, completion of validated questionnaires, interviews, and resting 12-lead ECGs. Smoking burden was evaluated by current cigarette smoking in cigarettes per day (cpd) and pack-years (current cpd * years smoked). Recent smoke exposure was measured by an exhaled CO level, which reflects smoking efficiency and recent smoke exposure. Smoking status (abstinence or continued smoking) was assessed by selfreported 7-day point-prevalence abstinence using a smoking calendar and the timeline follow-back method, and then confirmed by exhaled CO levels of ,10 ppm (Micro-3 Smokerlyzer, Bedfont Scientific, Williamsburg, VA) [13]. Fasting blood samples were obtained by venipuncture and refrigerated.

Statistical Analysis
All analyses were conducted using SPSS Statistics 20.0 software (SPSS Inc, Chicago, Illinois). Means (standard deviations) and ranges were determined for the subject characteristics and smoking intensity parameters in Table 1. Two-tailed t-tests or chi-square tests were used to compare differences in anthropomorphic data, physiological data, demographics, and smoking parameters between continuing smokers and abstainers at year 3 ( Table 2). Multivariable logistic regression models were constructed to determine associations between ECG abnormalities, at baseline and at 3 years after the target quit date, with measures of smoking burden. These models controlled for baseline age, sex, race, weight, body-mass index, glucose, low-density lipoprotein cholesterol, educational status, use of lipid-lowering medications and use of antihypertensive medications. Multinomial regression models were used to evaluate progression and regression of ECG changes in smokers and abstainers.

Longitudinal Changes in ECG Abnormalities after a Quit Attempt
Of the 354 subjects with ECGs at year 3, abstinence from smoking was confirmed in 153 (43%) participants by CO level ,10 ppm. As is often observed in longitudinal smoking cessations studies [17,18], there was moderate attrition after 3 years with 119 (34%) subjects not returning for that visit. Several important differences between the baseline and year 3 data were observed amongst those completing follow-up. Subjects who followed up at year 3 were more likely to be white (p = 0.004) and on average, were 3.6 kg heavier (p,0.001) and had 2.9 cm larger waist circumferences (p,0.001). Additionally, those who followed up had fasting glucose and hemoglobin A1C levels that were 4.5 mg/ dL and 0.26% higher respectively (both p,0.001). Systolic and diastolic blood pressures were 4.3 and 7.0 mmHg lower in those who followed up (both p,0.001), and a higher percentage were being treated with antihypertensive medications (p,0.001). A higher percentage of subjects who were followed up at year 3 were  on lipid-lowering therapy (p,0.001), although there was not a significant difference in LDL-cholesterol levels ( Table 1). Major Q waves (MC 1-1, 1-2) also were the most frequent major ECG abnormality at year 3 (22 [6.2%] subjects), followed by atrial enlargement (MC 9-3, 9-6; 15 [4.2%] subjects) and deep T wave inversion (MC 5-1, 5-2; 8 [2.3%] subjects) (Figure 2A). Axis deviation (MC 2-1, 2-2; 40 [11.3%] subjects) and moderate twave inversion (MC 5-3; 16 [4.5%] subjects) remained the most common minor ECG changes at year 3 ( Figure 2B). New major and minor ECG criteria were observed in 7.2% and 15.6% of subjects respectively, but in a similar number of abstinent subjects and continuing smokers (p.0.2 for both, Table 2). Abstinent subjects gained significantly more weight than subjects who continued to smoke over the 3 year follow up period (

Discussion
In the general population, major and minor ECG changes predict increased mortality [7,19]. Individuals who smoke are more likely to have ECG findings consistent with ischemic heart disease [20], structural heart disease [21], and cardiac rhythm disorders [22]. Despite, on average, being nearly a decade younger than participants in population-based studies, the smokers in our study had notably more major ECG abnormalities (especially major Q waves) than has been observed in the general population [8,[23][24][25]. Participants in the Coronary Artery Risk Development in Young Adults (CARDIA) study, a large epidemiological study that reported a cross-sectional evaluation of ECG abnormalities in subjects of a comparable age to those in our cohort, had 60% fewer major Q wave findings (MC 1-1 and 1-2) [26]. Only 14% of comparable CARDIA subjects were current smokers, indicating that smokers have a higher prevalence of major Q wave findings [26]. Indeed, in CARDIA, active smokers had a higher prevalence of major and minor Q-waves than non-smokers, though no information regarding the smoking burden of CARDIA subjects was reported [26]. In larger epidemiological studies that evaluated ECG abnormalities, the majority of subjects studied were one to two decades older that those in our study. In those studies, a minority of subjects were active smokers [7,23,25,27,28] or smokers were excluded from the analyses [29]. The ECG abnormalities in younger smokers have not been well-described and are in need of further study. Considering that age is one of the strongest predictors of 10-year CVD risk [30] the age-independent association between baseline major Q-waves and smoking burden in pack-years supports the idea that increased lifetime smoking burden increases CVD risk and that increased CVD risk related to long-term smoking can be detected by the 12-lead ECG. There was no independent relationship between ECG abnormalities and cpd at baseline, however sudden cardiac death in smokers also has been associated with the overall duration of smoking, but not cpd [31]. While there were fewer minor ECG abnormalities in those who quit smoking, we did not observe a significant difference in major or minor ECG findings at year 3. This is not unexpected since Q-waves, our most common major finding, are indicative of myocardial scar which is unlikely to resolve during our 3 year follow up period [32]. It is possible that the length of follow up was inadequate to detect differences in ECG progression or regression. Still, prior studies have shown that with improved blood pressure control, ECG abnormalities can regress within 2-5 years [32,33]. Longitudinal data regarding ECG changes after a quit attempt have not been described previously. Also, continuing smokers in our study were smoking significantly fewer cigarettes than they were at baseline and this group gained less weight then abstainers, which would be expected to attenuate the effects of smoking on ECG abnormalities. Additionally, there was a significant increase in use of antihypertensive medications and lipid-lowering therapy at year 3, compared to baseline. Treatment of known CVD risk factors could decrease the likelihood of developing ECG abnormalities at year 3 regardless of smoking status. For example, the prevalence of ECG abnormalities suggestive of left ventricular strain, such as T-wave inversion and LVH, slightly decreased at year 3 in both groups. This is more likely due to improvements in blood pressure rather than changes in smoking [32,33].

Limitations
Because this was a randomized clinical trial of smoking cessation interventions, there are no nonsmoking controls for comparison. In smoking cessation studies, it is very common for subjects who relapse to drop out or miss follow-up visits [17,18]. In our study, 34% of subjects did not have a 3-year ECG, which is consistent with the 30-43% one year drop-out rates reported in recent clinical trials of smoking cessation pharmacotherapies [17,18]. There were differences in participants that did not return for follow up with regard to age, and race which may introduce selection bias. Though, those who did not attend follow-up smoked a similar number of cigarettes/day at baseline [5,12]. The three year follow up period in this study may be too short to detect progression or regression of some ECG changes [32,33]. Finally, although ECG changes are associated with increased mortality, hard CVD outcomes including death, were not assessed in this study population.

Conclusions
Even in relatively young adults, major ECG abnormalities are independently associated with smoking burden. Minor ECG abnormalities were not associated with current or past smoking burden. After 3 years, quitting smoking was not associated with a decrease in ECG abnormalities; however those who abstained gained significantly more weight and continuing smokers smoked fewer cpd than at the inception of the study. Longer follow-up likely is needed to identify the effects of quitting versus continued smoking on the ECG and CVD risk. The percentage of subjects with major and minor ECG abnormalities at baseline (blue) and at year 3 (red). Year 3 percentages include both continued smokers and those who abstained. All p-values for comparisons between baseline and year 3 ECG abnormalities were non-significant (,0.05). ECG = electrocardiogram. *Subjects may have had more than one major and/or minor ECG abnormalities. Each abnormality was tallied separately. doi:10.1371/journal.pone.0062311.g001  Figure 2A shows percentages of the specific major ECG abnormalities and figure 2B shows percentages of the specific minor ECG abnormalities at baseline (blue) and at year 3 (red). Year 3 percentages include both continued smokers and those who abstained. All p-values for comparisons between baseline and year 3 ECG abnormalities were non-significant (,0.05). BBB = Bundle Branch Block; AV = Atrioventricular. doi:10.1371/journal.pone.0062311.g002