This editorial refers to the article “The smoker's paradox in acute coronary syndromes – is it real?” by Madalena Coutinho Cruz and co-workers, published in this issue of the Journal.1

In order to assess the possible protection provided by smoking during the in-hospital period and in the first year after acute coronary syndrome (ACS), the so-called ‘smoker's paradox’, the authors collected data from all consecutive patients (n=3298) hospitalized due to ACS in their center between 2005 and 2014.

After excluding former smokers (n=368) and patients treated by fibrinolysis (n=203), and taking into consideration potential confounding factors related to smokers’ prothrombotic status, the authors performed a retrospective analysis of the remaining 2727 patients, comparing two groups: current smokers (n=1138, 41.7% of the population) and non-smokers (n=1589, 58.3%). Two endpoints were defined: in-hospital mortality and a composite endpoint including all-cause mortality, rehospitalization for cardiovascular causes, angiography, percutaneous coronary intervention and coronary artery bypass graft surgery at one year after ACS.

They found that current smokers, on average 15 years younger at presentation, showed a higher rate of STEMI than non-smokers (69.8% vs. 56.3%, p<0.001), in whom NSTEMI predominated. Current smokers also had a more benign clinical, hemodynamic and angiographic profile at hospital admission and a better in-hospital course. They also appeared to have a more favorable prognosis at one year, although they were more likely to be revascularized, mainly by PCI (84.1% vs. 67.0% in non-smokers, p<0.001), during hospital stay and in follow-up.

The raw data assessed by univariate statistics suggested that the smoker's paradox would be found in this population, but after multivariate analysis with full adjustment for age, gender, heart rate, systolic blood pressure, Killip class, creatinine at admission and ACS type, all the differences disappeared, supporting the authors’ conclusions that the smoker's paradox was not observed in the study population and that the initial apparent differences in outcome could be explained simply by smokers’ lower baseline risk.

The study's data and conclusions, coming from a tertiary center with a large ACS population, managed by current state-of-the-art drug regimes and interventions, are significant. To the best of our knowledge, this is the first paper published in Portugal to study the smoker's paradox with a sufficiently large population and with sufficient statistical power to support its conclusions.

The topic addressed by the authors is important, since several articles have been published in high-ranking medical journals (although contradicted by many others2,3) that suggest the existence of a paradoxical protection afforded by smoking in patients with coronary artery disease (CAD), heart failure and stroke.4–6

This is cited by many patients as an argument for not quitting smoking, losing weight or strictly controlling their other risk factors, since they believe that they would benefit from the lower residual cardiovascular risk due to the protection conferred by tobacco smoking.

Arguments are also made in favor of electronic cigarettes. There is the perception that e-cigarettes are neutral for cardiovascular health, but this is not supported by epidemiological studies, since none have yet been performed to elucidate the impact of new forms of tobacco smoking on cardiovascular health.7–9

This misperception is clearly contradicted by other trials. The Organization to Assess Strategies in Acute Ischemic Syndromes (OASIS) 5 trial, which included 18809 ACS patients, showed that smoking cessation reduced the risk of a new ACS by 43% at six months.10

The study has some limitations, most of them acknowledged by the authors: its retrospective nature, and the lack of data regarding smoking burden and on compliance with drug regimes and cardiovascular mortality after discharge. None of these should be considered a major challenge to the study's conclusions.

It would also have been interesting to assess the patients’ global cardiovascular risk previous to ACS, by the EuroSCORE or another tool validated for the Portuguese population, to ascertain whether the overall risk of smokers, who also had higher rates of dyslipidemia and family history of CAD, would have been predicted to be similar to the non-smokers group, who were 15 years older and had higher rates of hypertension and diabetes.

Another interesting exercise would be to study the consequences of differences in smoking burden in terms of cardiovascular risk, considering that greater numbers of pack/years are associated with increased risk. With this information it would be possible to quantify the risk of smoking in a different way, demonstrating, according to Bradford Hill's causal criteria, a risk that increases proportionally to the smoking burden.11

If the study had included only patients in whom the index ACS was the first manifestation of CAD, its findings and conclusions would probably be even stronger. The study population would decrease by excluding patients with previously known CAD (35.8%), stroke/transient ischemic attack (5.9%) and peripheral arterial disease (3.1%), but this would probably would strengthen the study's conclusions.