Effect of smoking cessation on cardiac troponin I concentrations

Abstract

Chronic elevation of cardiac troponin I (cTnI) is associated with heart failure and cardiovascular death. Paradoxically, observational studies have indicated that current smokers have lower cTnI concentrations than non-smokers. We examined determinants of cTnI in smokers and the effect of smoking cessation on cTnI. Overweight or obese smokers received motivational support and varenicline to aid cessation and dietary advice to limit weight gain. Quitters were defined according to the Russell standard (≤5 cigarettes after the quit date) and validated with expired breath CO <10 ppm. Of the total 122 participants, 108 completed assessments at 12 weeks and 78 were classified as quitters versus 30 who continued smoking. cTnI was measured with a high-sensitivity assay with a limit of detection of 1.2 ng/L (Abbott Diagnostics), and concentrations (log-transformed) were compared between quitters and continuing smokers. cTnI concentrations were significantly higher in men than women and correlated with age, but not with number of cigarettes/day. Quitters had median baseline and 12-week levels of 1.4 ng/L (interquartile range [IQR] 1.2–2.5) and 1.4 ng/L (IQR 1.2–2.4), respectively, while nonquitters had baseline and 12-week levels of 1.5 ng/L (IQR 1.2–2.9) and 1.8 ng/L (IQR 1.3–3.4), respectively. The change in cTnI concentrations from baseline to 12 weeks did not differ between quitters and continuous smokers (p = .7). The results suggest that smoking cessation does not affect levels of cTnI, a marker of chronic subclinical myocardial injury, in contrast to prior observational data suggesting that tobacco smoking is associated with lower cTn concentrations.

Keywords:

• Tobacco
• smoking cessation
• cardiovascular diseases
• biomarker
• troponin I

Introduction

Smoking is the most important reversible risk factor for atherosclerotic cardiovascular disease [1]. The mechanisms behind the increased risk of smoking are broad, including dyslipidemia, inflammation, endothelial dysfunction, coagulation, platelet adherence and insulin resistance [2]. Smoking cessation induces improvement in many of these pathogenic factors and leads to reduction in risk of cardiovascular disease and subclinical atherosclerosis [1]. Cardiac troponins (cTn) are a group of proteins in cardiac muscle fibers regulating muscular contraction. Both cardiac troponin T (cTnT) and troponin I (cTnI) are sensitive and specific markers of myocardial infarction that are widely used as diagnostic markers. In the chronic setting, circulating cTn concentrations, even within the normal range, are strongly associated with the risk of cardiovascular events, and are considered a marker of chronic subclinical myocardial injury [3–5].

Paradoxically some epidemiological and subgroup studies in trial cohorts have shown that current smoking is associated with lower concentrations of cTnI. Among 8715 men and women from the Norwegian Nord-Trøndelag Health Study (HUNT 2) current smokers had significantly lower cTnI values than never and former smokers [6]. The West of Scotland Coronary Prevention Study (WOSCOPS) measured troponins in 3318 men and found fewer smokers in the higher quintiles of cTnI (p for trend .043) [7]. In the Generation Scotland Scottish Family Health Study including 19,501 men and women both cTnI and cTnT were inversely associated with current smoking [8].

However, some findings indicate higher odds of detectable cTnI in smokers. In 12,951 men and women that participated in the Justification for the Use of Statins in Primary Prevention (JUPITER) trial, current smoking was associated with 15% increased odds for detectable cTnI [9]. Associations of cTnT with smoking have been inconsistent. Analyses in 8121 men and women from the large, community-based Atherosclerosis Risk in Communities (ARIC) Study did not find a difference in hs-TnT between current, former and never-smokers, but heavier smoking (more pack-years) was associated with elevated hs-TnT [10].

Likewise, results are not consistent in patients with CAD. The Prevention of Events with Angiotensin-Converting Enzyme Inhibitor Therapy (PEACE) trial in patients with stable CAD found lower concentrations of cTnT in current smokers compared to non-current smokers, but no differences in cTnI [11].

Given the inconsistencies between observational data, we sought to understand the effect of smoking cessation on cTnI concentrations in a recently published clinical trial of the effect of diet on weight gain during smoking cessation [12].

Materials and methods

Study design and subjects

The study had a randomized, controlled, parallel-group design as described previously [12]. In brief, eligible men and women were overweight or obese (BMI 25–40 kg/m²) aged 20–65 years, smoked at least 10 cigarettes daily, were motivated to quit and were willing to be treated with varenicline to aid cessation. Exclusion criteria were a cardiovascular event within 2 months prior to screening, diabetes mellitus type 1 or type 2 treated with insulin, serious psychiatric disorder, alcohol or drug abuse, pregnancy and lactation, bariatric surgery, medication for weight loss or participation in a weight loss program within the last four weeks, recent change in weight. The Regional Committees for Medical and Health Research Ethics in Norway evaluated the study and the work was conducted in accordance with the Declaration of Helsinki. All subjects signed a written informed consent before screening procedures.

Participants were randomized to a low-carbohydrate or low-fat diet and treated with a standard course of varenicline for 12 weeks. Smoking status was assessed according to the Russell standard. In this study quitters reported a total of ≤5 cigarettes smoked at any time after the quit date, which was day 14 after baseline and 10 days after start of varenicline. Quitting was validated with expired breath carbon monoxide (CO) < 10 ppm. Of 122 randomized participants, 108 (89%) completed clinical and laboratory assessments at 12 weeks and are included in this analysis. We found no differences in cTnI values between the dietary groups (data not shown). Thus we combined the groups to compare concentrations of cTnI in quitters (n = 78) to continuing smokers (n = 30) at 12 weeks.

Laboratory analyses

Blood sampling was performed after an overnight fast of 10–12 h and samples were frozen at −70 °C or below. Concentrations of cTnI were measured with a high-sensitivity assay (Abbott Diagnostics, Abbott Park, IL) as described earlier [13]. The assay coefficient of variation is 20% at 1.3 ng/L, 10% at 4.7 ng/L and 4% at 26.2 ng/L [14]. The limit of detection is 1.2 ng/L and levels below detection limits were assigned a value of 1.2 ng/L. Recalculations including the measured lower values did not change any result except for the lower quartile values.

Statistical analyses

cTnI values were highly skewed and are presented as median and quartiles. Spearman’s rho was used to calculate correlations. cTnI values were log-transformed for further analysis. We compared changes in continuously distributed variables between baseline and 12 weeks follow-up for quitters and continuing smokers using paired t-tests while changes between quitters and smokers were compared using independent t-tests. Statistical analyses were done with SPSS software version 25 (IBM SPSS Statistics, Armonk, NY) and p values <.05 were considered statistically significant.

Results

As shown in Table 1, baseline characteristics of quitters and continuing smokers were similar. Over 73% of participants were women. At baseline, the average number of cigarettes smoked/day was 18. About 10% carried a diagnosis of previous cardiovascular disease in both groups.

Table 1. Baseline characteristics (mean ± SD) of quitters and continuing smokers and change in weight and waist circumference (mean ± SD) after 12 weeks.

	Quitters	Continuing smokers
	n = 78	n = 30	p Value^a
Age, years	51 ± 8	51 ± 8	.9
Women, no. (%)	57 (73.1)	22 (73.3)	–
Use of medication, no. (%)
Antihypertensives	21 (26.9)	10 (33.3)	–
Statin	14 (17.9)	5 (16.7)	–
Aspirin	12 (15.4)	5 (16.7)	–
No. cigarettes/d	17.5 ± 6.1	18.0 ± 4.4	.7
Systolic blood pressure	115 ± 15	119 ± 16	.2
BMI, kg/m^2	30.5 ± 3.6	30.4 ± 3.5	.9
Weight, kg baseline	89.4 ± 14.5	91.2 ± 15.6	.6
Change in weight	−0.1 ± 3.0	0.2 ± 3.1	.7
Waist, cm baseline	101.0 ± 11.0	101.5 ± 13.0	.7
Change in waist	−2.0 ± 3.8	−0.9 ± 3.9	.2

^aIndependent t-test: p value for change between quitters and continuing smokers.

Baseline values of cTnI correlated with age (Spearman’s rho 0.4; p < .001) and waist circumference (Spearman’s rho 0.2; p = .047), but not with weight or BMI (data not shown). There was no correlation between cTnI and number of cigarettes smoked at baseline (data not shown).

At baseline, 32% had cTnI concentrations below the limit of detection. cTnI concentrations were significantly higher in men than women (median 2.2, interquartile range [IQR] 1.4–3.9) ng/L versus 1.4 [IQR 1.2–2.1] ng/L, p = .003). Only two women had concentrations exceeding the 99th percentile used in clinical practice (15.6 pg/mL in women) [13].

We found no significant difference in cTnI concentrations between quitters and continuous smokers at baseline and 12 weeks and no significant change from baseline to 12 weeks (Table 2), and these findings were consistent across genders (data not shown).

Table 2. Baseline values of cTnl (ng/L) median (interquartile range) and p values between quitters and continuing smokers – and change after 12 weeks.

	Quitters	Continuing smokers
	n = 78	n = 30	p Value^a
cTnI baseline	1.4 (1.2–2.5)	1.5 (1.2–2.9)	.3
cTnI week 12	1.4 (1.2–2.4)	1.8 (1.3–3.4)	.1
Change in cTnI	0.0 (−0.3–0.3)	0.0 (−0.4–0.3)	.7

^aIndependent t-test: p value for change between quitters and continuing smokers.

Discussion

The new and salient finding of this study is that cTnI concentrations showed no change within 12 weeks of achieving cessation in smokers participating in a clinical trial. This finding from an interventional trial supplement prior information derived from observational studies. We are not aware of previous clinical studies of the effect of quitting smoking on cTnI concentrations. The longitudinal data from the Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS) study showed no association between change in smoking status and cTnI levels in line with our results. In this study, smoking status was not validated. Furthermore, the prevalence of smoking was low in the seniors investigated and few smokers had quit at the end of follow up [15].

Our participants had in general low concentrations of cTnI, and more than 30% had concentrations below the limit of detection at baseline, which may be partly ascribed to the relatively young age and high proportion of women. All participants were smokers and were not selected for other cardiac risk factors except overweight and obesity. In the recent Norwegian Nord-Trøndelag Health Study (HUNT 3) concentrations of cTnI were detectable (≥ 1.2 ng/L) in 72.2% [16]. The Generation Scotland Scottish Family Health Study recruited in 2006–2010, also used the limit of 1.2 ng/L and reported 74.8% having detectable cTnI values [8]. In the meta-analysis by Sze, mean prevalence of detectable cTnI in the general population is estimated to be around 78% in studies using high sensitive analysis with detection limit 1.5–1.9 ng/L [3].

The mechanisms for change in cTn concentrations are not fully established [3], and why cTnI could be affected by smoking is unclear. One proposed explanation is the difference in lean body mass in smokers and non-smokers [6]. cTnI is positively associated with BMI [8], and BMI-score over time is associated with a linear increase in cTnI [17]. However, the HUNT results showing lower cTnI concentrations in smokers were controlled for BMI [6]. One study comparing bariatric surgery and lifestyle intervention in patients with morbid obesity reported significant reduction in cTnI one year after surgery, but not after more modest weight reduction in the lifestyle group [18], suggesting large weight change or longer time are needed to cause difference in cTnI. Our study participants were counseled to limit weight gain when quitting, and we found no difference in weight, waist, body fat or lean mass between quitters and continuing smokers after 12 weeks [19]. Furthermore, we found no relation between cTnI and body weight in our data.

cTn values are known to increase after exercise, but in observational data better physical fitness is associated with lower concentrations [20]. However, in our study change in physical activity as measured by an Actigraph accelerometer did not differ between groups [19].

In newly published data from the North Sea Race Endurance Exercise Study current use of smokeless tobacco (Swedish snus) was associated with lower concentrations of cTnI and cTnT [21]. More than 50% of snus-users were former smokers, but very few were dual users and sensitivity analysis excluding these made no difference. These results indicate that an effect of tobacco on troponins could be due to nicotine, and probably not cigarette smoke. Varenicline is a selective partial agonist on the α₄β₂ acetylcholine nicotinic receptor [22], and hypothetically could influence troponins, but no data has shown such an association. Furthermore, all participants were treated with varenicline regardless of outcome.

As the potential biological effects or time course of potential effects of tobacco on cTns are unknown, 12 weeks of quitting could be too short to give an increase in cTnI. However, use of snus was associated with lower change in cTn concentrations from baseline to 3 h after a bicycle race [21], a finding that may indicate a very rapid effect of the interaction of snus and exercise.

It has been questioned if the cTns have less prognostic value as a risk marker among smokers compared to non-smokers. According to HUNT data the association between cTn levels and the risk of CVD events is weaker for smokers than for nonsmokers [6]. This observation was not confirmed in the PEACE-study [11]. If a potential effect of smoking on cTn is mediated by nicotine which is of less importance for cardiovascular health than products of cigarette combustion, this might explain that cTnI is less predictable of long-term CAD in smokers than in non-smokers.

Study limitations

This clinical trial followed a somewhat small number of smokers during a quit attempt. The follow-up period of 12 weeks after quitting smoking could be too short to show changes in cTnI. In the current exploratory analysis, we only measured cTnI concentrations. While cTnI and cTnT are both good predictors of cardiovascular disease, the subtypes exhibit genetic and biological differences. The two cTns have been noted to be only moderately correlated and their association with outcomes differs [5,8], and further work will be needed to understand the effect of smoking cessation on cTnT.

Conclusion

In this evaluation of the effect of smoking cessation on cTn concentrations, we found no difference in change in cTnI between quitters and continuous smokers after 12 weeks. The results suggest that smoking cessation does not affect levels of troponin, a marker of chronic subclinical myocardial injury. These data are in contrast to prior observational data suggesting that tobacco smoking is associated with lower cTnI concentrations.

Acknowledgments

The authors thank Mette Svendsen for organizing the main study and giving dietary advice and Jon Norseth for the cTnI analyses.

Disclosure statement

EH and ST have received honorarium for lectures in smoking cessation for Pfizer (the manufacturer of varenicline). TO has served on advisory boards for Abbott Diagnostics, Roche Diagnostics and Bayer and has received research support from Abbott Diagnostics, Novartis, Roche Diagnostics, Singulex and SomaLogic via Akershus University Hospital, and speaker’s or consulting honoraria from Roche Diagnostics, Siemens Healthineers and CardiNor.

EH participated in the study design and coordination, carried out the study, performed the statistical analyses and drafted the manuscript. ST and TO participated in the conception and design of the study and helped to write the manuscript. All authors read and approved the final manuscript.

Additional information

Funding

This work was supported by only local Departemental resources, including provision of varenicline.