Smoke over myocardial infarction: cigarettes and reperfusion injury

This editorial refers to ‘Impact of active smoking on myocardial infarction severity in reperfused ST-segment elevation myocardial infarction patients: the smoker's paradox revisited’†, by R. Symons et al., on page 2756. I kissed my first girl and smoked my first cigarette on the same day. I haven't had time for tobacco since. In contrast to the overwhelming data on the harmful health consequences of cigarette smoking, there is conflicting evidence regarding the association of smoking and clinical outcome after infarction. The smoker's paradox, an unexplained lower mortality and morbidity in smokers vs. non-smokers with myocardial infarction, has been investigated for more than three decades.1–4 The vast majority of data favouring this ‘smoker's paradox’, however, did not find an independent association between smoking and clinical events when adjusting for other established clinical risk characteristics (e.g. age).1–5 Recent studies suggested that smoking might be related to increased myocardial salvage, and less myocardial damage, but were hampered by single-centre designs and small sample sizes. More importantly, these studies used angiographic, electrocardiographic, or biochemical approaches to determine infarct size and microvascular injury.6–8 The use of these methods has several limitations, and none of these studies applied cardiovascular magnetic resonance (CMR) imaging, which is nowadays regarded as the reference technique for the quantification of myocardial injury after reperfused ST-elevation myocardial infarction (STEMI).9,10 In the present issue of the journal, Symons and colleagues report an elegant multicentre analysis of 471 patients with acute STEMI investigating the impact of cardiovascular risk factors including smoking on the severity of myocardial injury and post-infarction remodelling assessed by CMR.11 The main findings were a strong association of smoking with intramyocardial haemorrhage (IMH) as a marker of severe ischaemia–reperfusion injury. However, consistent with the smoker's paradox, smoking was an independent predictor of a more favourable post-infarction left ventricular remodelling, even after multivariable adjustment. Interestingly, the results of Symons et al. suggest a different impact of smoking on acute myocardial damage and post-infarction remodelling based on the presence or absence of IMH. We congratulate the authors for conducting this important CMR study that provides unique insights into the relationship of smoking status with myocardial damage and remodelling after mechanical reperfusion for STEMI. However, it is also important to consider several limitations of the study when interpreting the data. First, the assessment of IMH by CMR in the setting of acute myocardial infarction has proven prognostic relevance.10 However, for IMH assessment, T2-weighted imaging with quantification of the hypointense core may not be specific to detect IMH. Severe microvascular injury (late microvascular obstruction), even in the absence of IMH, may also be represented by a signal drop within the area of oedema.12 To assess for the presence of haemorrhage within the infarction zone more comprehensively, a T2* mapping technique, which is more sensitive to detect the paramagnetic effects of iron, and thereby IMH, might be more specific in addition to T2-weighted imaging.12 Secondly, it is somewhat surprising that only IMH, but no other CMR parameter of myocardial damage (infarct size and microvascular obstruction which both are independently associated with prognosis), is different between smokers and non-smokers. Thirdly, there were important differences in baseline risk characteristics between smokers and non-smokers (e.g. age, diabetes), and propensity scoring would be desirable to correct for these confounders to underline the effects of smoking on reperfusion injury. Fourthly, no hard clinical endpoints are reported to underline the clinical relevance of the smoker's paradox. Finally, there were a large number of missing follow-up CMRs in the study [90 patients (19%)] and the authors do not provide information on whether or not the patients effectively stopped smoking following their STEMI which might influence adverse remodelling. Moreover, differences in remodelling in heavy smokers in comparison with moderate smokers are not reported. Despite these limitations, the study adds important information for our understanding of ischaemia–reperfusion injury, and may warrant a more vigorous approach to mitigate reperfusion injury in smokers. Based on data from randomized studies and a large meta-analysis, non-smokers have a poorer response to clopidogrel than smokers.13,14 In contrast, prasugrel and smoking status have no influence in platelet reactivity.13 This may be another explanation for the higher IMH rate in smokers. From a statistical point of view, it is important to note that the multivariable analysis for remodelling in the present analysis included several factors which may influence each other. Microvascular obstruction and IMH as well as peak troponin and infarct size are closely related to each other.12 Therefore, statistical interactions and multicollinearity in these factors cannot be ruled out. A randomized setting for chronic cigarette smoking and acute myocardial infarction is neither possible nor ethically justified since tobacco use has proven its harm in many ways. Therefore, it is difficult to distinguish in observational trials between direct effects of nicotine and other ingredients of cigarettes or if we observe just a bystander effect because of much lower risk profiles of smokers. Even after extensive correction for cofactors in multivariable analyses, as carefully performed in the present analysis, we cannot exclude that smoking as a statistical variable outweighs the sum of the other risk factors (Figure 1). In summary, the interesting results of the study of Symons et al. are an important step toward a better understanding of reperfusion injury in smokers, but should be mainly considered as hypothesis-generating rather than being a proof for differences in IMH and remodelling between smokers and non-smokers. These findings may stimulate new research and insights into the conundrum of the smoker's paradox in myocardial infarction. Conflict of interest: none declared. The schematic figure illustrates that smoking may be the dominant risk factor in the development of myocardial infarction and for severe myocardial damage after reperfusion which is outweighing other established risk factors.