Increased concentrations of high-sensitivity cardiac troponin I (hs-cTnI) and high-sensitivity cardiac troponin T (hs-cTnT) reflect subclinical myocardial injury and are strong predictors of incident heart failure and cardiovascular death.^1–3 Although hs-cTnI and hs-cTnT commonly are considered to provide similar diagnostic and prognostic information, recent data have demonstrated important physiological differences.⁴ Recently, we reported that tobacco smoking is associated with lower circulating hs-cTnI in a general population cohort and that the prognostic merit of hs-cTnI is significantly lower in smokers than in nonsmokers.⁵ The observations were unexpected and raise the question whether the association with tobacco smoking can be generalized to populations with established cardiovascular disease. We therefore tested the hypotheses that current smoking is associated with lower concentrations and prognostic value of hs-cTnI and hs-cTnT in patients with stable coronary artery disease (CAD).

We measured hs-cTnT and hs-cTnI concentrations at baseline in 3126 nonsmokers and 550 current smokers with stable CAD enrolled in the PEACE trial (Prevention of Events With Angiotensin-Converting Enzyme Inhibitor Therapy). hs-cTnT and hs-cTnI were measured using high-sensitivity assays as described previously.^2,3 Cigarette smoking status was self-reported as current smoker (≥1 cigarettes/day), former smoker, and never smoker. During a median follow-up of 5.2 years, 204 patients experienced the composite outcome of cardiovascular death or hospitalization for heart failure. All participants provided written informed consent, and the study was conducted after approval of ethics committees from all participating sites.

The associations between log 2–transformed cardiac troponin concentrations and (1) smoking and (2) outcomes were assessed by multivariable linear regression and Cox regression, respectively. We generated both unadjusted models and models adjusting for covariates associated with troponin concentrations by linear regression analysis, ie, age, sex, N-terminal pro-B-type natriuretic peptide, body mass index, previous hypertension, estimated glomerular filtration rate, and use of β-blockers, diuretics, and insulin.

Current smokers had lower concentrations of hs-cTnT (median, 4.9 ng/L [interquartile range, 2.9–7.5 ng/L] versus 6.1 [interquartile range, 4.1–9.4 ng/L], P<0.001), but not hs-cTnI (median, 4.1 ng/L [interquartile range, 2.8–6.9 ng/L] versus 4.4 [interquartile range, 2.9–7.2 ng/L], P=0.09) than noncurrent smokers. In adjusted models, current smoking was associated with 12% (95% CI, 4%–20%) lower hs-cTnT and a nonsignificant 0% (95% CI, –9% to +7%) lower hs-cTnI level in comparison with never-smokers, whereas former smoking was associated with a nonsignificant 1% (95% CI, –7% to +6%) lower hs-cTnT and a nonsignificant 0% (95% CI, –6% to +6%) lower hs-cTnI level in comparison with never-smokers. The percentage of smokers as a function of cardiac troponin T concentrations is shown in Figure A.

Figure. Relation between cardiac troponin T (cTnT) and current smoking, and between cTnT and outcome according to smoking status. A, Proportion of current smokers as a function of high-sensitivity cTnT concentrations in patients with stable coronary artery disease. B, Association between high-sensitivity cTnT concentrations and the incidence of cardiovascular (CV) death or heart failure hospitalization (HF Hosp.) stratified according to smoking status. Models are adjusted for age, sex, N-terminal pro-B-type natriuretic peptide, body mass index, previous hypertension, estimated glomerular filtration rate, and use of β-blockers, diuretics, and insulin. p.y. indicates person-years.

The associations between hs-cTnT (hazard ratio, 1.56 [95% CI, 1.30–1.88]) and hs-cTnI (hazard ratio, 1.29 [95% CI, 1.16–1.43]) and the incidence of cardiovascular death or heart failure hospitalization remained significant in adjusted models for nonsmokers, whereas this association was attenuated in current smokers (hs-cTnT: hazard ratio, 1.23 [95% CI, 0.87–1.72]; hs-cTnI: hazard ratio, 1.17 [95% CI, 0.88–1.56]). The association did not differ significantly between smokers and nonsmokers in unadjusted models (hs-cTnT: P-for-interaction=0.18; hs-cTnI: P-for-interaction=0.82) or adjusted models (hs-cTnT: P-for-interaction=0.27; hs-cTnI: P-for-interaction=0.55) (Figure B). The area under the receiver operating characteristics curve was 0.74 (0.70–0.77) and 0.71 (0.67–0.75) in nonsmokers and 0.68 (0.58–0.78) and 0.69 (0.60–0.79) in smokers for hs-cTnT and hs-cTnI, respectively.

The current findings for hs-cTnT and hs-cTnI in patients with chronic CAD extend previous observations of an inverse association between hs-cTnI concentrations and current smoking in the general population.⁵ The reason smoking was associated with lower hs-cTnT but not hs-cTnI concentrations in stable CAD is unclear, but could theoretically be related to substances in tobacco smoke that interfere differentially with the analysis or, perhaps more likely, the release and kinetic patterns of hs-cTnT and hs-cTnI. The mechanisms responsible for the inverse association between smoking and concentrations of cardiac troponins remain unknown, and the effect of tobacco smoking on troponin release and degradation merits further investigation. A higher proportion of smokers among patients with low cardiac troponin concentrations and CAD could potentially be attributed, in part, to the very low risk of CAD in nonsmokers with low concentrations of troponins. In contrast to observations made for hs-cTnI in a general population cohort, we did not observe a significant interaction by smoking status on the association between hs-cTnT or hs-cTnI and cardiovascular events in the PEACE cohort, but this may be attributable to the smaller sample size and subsequent lower statistical power.

Strengths of the current study include the prospective observational study design, the long-term follow-up, and the use of adjudicated end points. A limitation is that tobacco habits were self-reported. Furthermore, we cannot rule out the presence of survival bias and residual confounding.

In conclusion, our study demonstrates that cigarette smoking is associated with lower concentrations of circulating hs-cTnT, but not hs-cTnI in patients with stable CAD. Moreover, the association between both hs-cTnT and hs-cTnI and the incidence of cardiovascular events in current smokers with stable CAD is attenuated and nonsignificant in adjusted models.

The PEACE trial was sponsored by the National Heart, Lung, and Blood Institute with support from Knoll Pharmaceuticals and Abbott Laboratories. This biomarker substudy was also supported by Roche Diagnostics and Abbott Diagnostics.

Dr Myhre has received consulting fees from Novartis. Dr Sabatine has received research grant support through Brigham and Women’s Hospital from Abbott Laboratories; Amgen; AstraZeneca; Bayer; Daiichi-Sankyo; Eisai; Gilead; GlaxoSmithKline; Intarcia; Janssen Research and Development; The Medicines Company; MedImmune; Merck; Novartis; Poxel; Pfizer; Quark Pharmaceuticals; Roche Diagnostics; and Takeda. Consulting for Alnylam; Amgen; AstraZeneca; Bristol-Myers Squibb; CVS Caremark; DalCor; Dyrnamix; Esperion; IFM Therapeutics; Intarcia; Ionis; Janssen Research and Development; The Medicines Company; MedImmune; Merck; MyoKardia; and Novartis. Dr Pfeffer has received consulting fees from AstraZeneca, DalCor, GlaxoSmithKline, Novartis, Novo Nordisk, Pfizer, Roche, and Sanofi; research grant support from Novartis; and stock Options in DalCor. Dr Omland reports consulting fees and research grant support via Akershus University Hospital from Abbott Laboratories, CardiNor, Novartis, Roche Diagnostics, Singulex, and SomaLogic. The other authors report no conflicts.

https://www.ahajournals.org/journal/circ

Data sharing: The data, analytic methods, and study materials will not be made available to other researchers for purposes of reproducing the results or replicating the procedure.

Guest editor for this article was Allan Jaffe, MD.

高敏感性心脏肌钙蛋白I（hs-cTnI）和高敏感性心脏肌钙蛋白T（hs-cTnT）的浓度增加反映了亚临床心肌损伤，并且是事件性心力衰竭和心血管死亡的有力预测指标。^1-3尽管hs-cTnI和hs-cTnT通常被认为可提供相似的诊断和预后信息，但最近的数据显示出重要的生理差异。⁴最近，我们报道了吸烟与一般人群队列中较低的循环hs-cTnI相关，吸烟者中hs-cTnI的预后价值显着低于非吸烟者。⁵这些观察是出乎意料的，并提出了一个问题，即是否可以将吸烟与吸烟的关联性普遍推广到已建立心血管疾病的人群。因此，我们检验了以下假设，即当前吸烟与稳定冠心病（CAD）患者中hs-cTnI和hs-cTnT的较低浓度和预后价值有关。

我们在PEACE试验（使用血管紧张素转化酶抑制剂治疗预防事件）中，对3126名非吸烟者和550名目前吸烟且基线稳定的CAD人群的基线测量了hs-cTnT和hs-cTnI浓度。如前所述，使用高灵敏度测定法测量了hs-cTnT和hs-cTnI。^2,3吸烟状况被自我报告为当前吸烟者（≥1支/天），曾经吸烟者和从未吸烟者。在5.2年的中位随访期间，有204名患者经历了心血管死亡或因心力衰竭住院的复合结果。所有参与者均提供了书面知情同意书，并且该研究是在所有参与机构的伦理委员会批准后进行的。

通过多变量线性回归和Cox回归分别评估log 2转化的心脏肌钙蛋白浓度与（1）吸烟和（2）结果之间的关联。我们通过线性回归分析生成了未经调整的模型和针对与肌钙蛋白浓度相关的协变量进行调整的模型，即年龄，性别，N末端前B型利尿钠肽，体重指数，以前的高血压，估计的肾小球滤过率和使用β受体阻滞剂，利尿剂和胰岛素。

当前吸烟者的hs-cTnT浓度较低（中位数为4.9 ng / L [四分位数范围，2.9-7.5 ng / L]，而6.1 [四分位数的范围，4.1-9.4 ng / L]，P <0.001），而不是hs-cTnT cTnI（中位数，4.1 ng / L [四分位数范围，2.8-6.9 ng / L]与4.4 [四分位数的范围，2.9-7.2 ng / L]，P= 0.09）。在调整后的模型中，当前吸烟使hs-cTnT降低12％（95％CI，4％–20％），而hs-cTnI降低0％（95％CI，–9％至+ 7％）则无统计学意义。与从未吸烟者进行比较，而以前吸烟与hs-cTnT降低1％（95％CI，–7％至+ 6％）无关，而0％（95％CI，–6％至+ 6％）无关。 ）比不吸烟者降低了hs-cTnI水平。吸烟者百分比与心肌肌钙蛋白T浓度的关系示于图A。

数字。 根据吸烟状况，心肌肌钙蛋白T（cTnT）与当前吸烟之间以及cTnT与结局之间的关系。A，在稳定冠状动脉疾病患者中，当前吸烟者所占比例与高敏感性cTnT浓度的关系。B，根据吸烟状况对高敏cTnT浓度与心血管（CV）死亡或心力衰竭住院（HF Hosp。）的发生率之间的关联进行了分层。调整模型的年龄，性别，N末端前B型利尿钠肽，体重指数，先前的高血压，估计的肾小球滤过率以及使用β受体阻滞剂，利尿剂和胰岛素。py表示人年。

hs-cTnT（危险比1.56 [95％CI，1.30-1.83]）和hs-cTnI（危险比1.29 [95％CI，1.16-1.43]）与心血管死亡或心力衰竭住院的发生率相关调整后的模型中，非吸烟者的吸烟率仍显着，而当前吸烟者的这种关联性有所减弱（hs-cTnT：危险比1.23 [95％CI，0.87–1.72]； hs-cTnI：危险比1.17 [95％CI，0.88– 1.56]）。在未经调整的模型（hs-cTnT：P-互动= 0.18； hs-cTnI：P-互动= 0.82）或经过调整的模型（hs-cTnT：P -for-互动= 0.27； hs-cTnI：P-for- = 0.55）（图B）。对于hs-cTnT和hs-cTnI，非吸烟者的接收器工作特性曲线下的面积分别为0.74（0.70-0.77）和0.71（0.67-0.75），吸烟者为0.68（0.58-0.78）和0.69（0.60-0.79） 。

慢性CAD患者中hs-cTnT和hs-cTnI的当前发现扩展了先前的观察，即一般人群中hs-cTnI浓度与当前吸烟之间呈负相关。⁵尚不清楚吸烟与稳定CAD中较低的hs-cTnT浓度相关但与hs-cTnI浓度无关的原因，但理论上可能与烟草烟雾中的物质有差异，这些物质会差异性地影响分析，或者更可能地是干扰物质的释放和动力学模式。 hs-cTnT和hs-cTnI。吸烟与心脏肌钙蛋白浓度之间负相关的机制尚不清楚，吸烟对肌钙蛋白释放和降解的影响值得进一步研究。低心脏肌钙蛋白浓度和CAD的患者中吸烟者比例较高可能部分归因于肌钙蛋白浓度低的非吸烟者的CAD风险非常低。与一般人群中hs-cTnI的观察结果相反，

当前研究的优势包括前瞻性观察性研究设计，长期随访以及使用裁决的终点。局限性在于烟草习惯是自我报告的。此外，我们不能排除存在生存偏差和残余混淆的可能性。

总之，我们的研究表明，吸烟与稳定的CAD患者中较低的循环hs-cTnT浓度相关，而与hs-cTnI无关。此外，hs-cTnT和hs-cTnI两者与当前具有稳定CAD的吸烟者的心血管事件发生率之间的相关性在调整后的模型中减弱了，并且不显着。

PEACE试验由美国国家心脏，肺和血液研究所赞助，并得到了诺尔制药公司（Knoll Pharmaceuticals）和雅培（Abbott Laboratories）的支持。该生物标记子研究也得到了罗氏诊断公司和雅培诊断公司的支持。

Myhre博士已从诺华获得咨询费。Sabatine博士已从雅培（Abbott Laboratories）的布莱根妇女医院（Brigham and Women's Hospital）获得研究资助；安进 阿斯利康 拜耳 第一三共；卫材; 吉利德 葛兰素史克 嵌花; Janssen研究与开发；药品公司；免疫 默克 诺华 Poxel; 辉瑞 夸克制药；罗氏诊断; 和武田 为Alnylam提供咨询；安进 阿斯利康 百时美施贵宝（Bristol-Myers Squibb）; CVS Caremark；DalCor; Dyrnamix; Esperion; IFM治疗学；嵌花; 爱奥妮丝 Janssen研究与开发；药品公司；免疫 默克 MyoKardia; 和诺华 Pfeffer博士已从阿斯利康，DalCor，葛兰素史克，诺华，诺和诺德，辉瑞，罗氏和赛诺菲获得了咨询费；诺华提供的研究经费支持；和DalCor中的股票期权。Omland博士报告了来自雅培实验室，CardiNor，诺华，罗氏诊断，Singulex和SomaLogic的阿克舒斯大学医院的咨询费和研究补助金支持。其他作者报告没有冲突。

https://www.ahajournals.org/journal/circ

数据共享：数据，分析方法和研究材料将不会提供给其他研究人员，以用于再现结果或复制程序。

这篇文章的客座编辑是医学博士Allan Jaffe。