E-cigarette use is associated with cardiovascular disease... in the same way that statins cause myocardial infarction!!



By Dr Farsalinos

A new study was published today by Prof Glantz et al. in the journal New England Journal of Medicine Evidence, claiming that e-cigarettes pose the same risk for disease (mainly cardiovascular disease, stroke and metabolic dysfunction) as smoking while dual use is more harmful than exclusive smoking. The study was a systematic review and metanalysis of publications that compared odds ratios of having disease between e-cigarette users and smokers.


Let's first think of what kind of studies would be needed to address the question of disease risk from e-cigarette use. You would normally need studies of vapers who had never smoked in the past, and then you could compare their risk of disease with smokers. Considering the fact that the development of cardiovascular disease happens after many years (if not decades) of smoking, you would need to examine people who use e-cigarettes for many years. Finally, you would expect either to analyze longitudinal studies or at least cross sectional studies with data on the duration and frequency of e-cigarette use, past and current smoking status, and information on the temporal association between exposure (e-cigarette use) and disease development (disease musty happen long AFTER e-cigarette use initiation, unless someone believes that e-cigarette use initation today can cause a disease that happened months or years before).


Did they do any of the above?


The metanalysis included 107 studies reporting 124 odds ratios (ORs). However, most of the data (97 out of the 124 ORs) were derived from cross sectional studies, that is studies asking in a single time point if you use e-cigarettes and if you have some disease. These studies lack information on temporality of events, and are thus useless. I will remind everyone of the scandal in another study by Prof Glantz, in which he reported that e-cigarette use increases the risk of myocardial infarction from a cross-sectional study. The scandal was NOT that his conclusion was innapropriate for the study design that he analyzed, but that the study itself HAD DATA on the temporality of events and in fact most e-cigarette users had initated use AFTER they developed a myocardial infarction. Strangely, Prof Glantz and his colleagues failed to examine and report this information. Eventually, and after repeated letters to the editors of the journal by many of us, the study was retracted. Additionally, we have no idea for how long people were vaping in the analysis, how much they were vaping (they included people with any use over that past 30 days), if the disease happened before or after e-cigarette use initiation, if and for how long had they quit smoking and if dual users were people who smoke everyday and vape once a week or people who vape every day and smoke once a week (no need to explain the difference betwqeen the two in the risk profile).


The study reminds me of past attempts to discredit the smoking cessation potential of e-cigarettes. In 2016, Kalkoran and Glantz published a metanalysis suggesting that e-cigarettes prevent smoking cessation, by including participants who were failures at baseline and people who were just experimenting with e-cigarettes (image examining the effect of experimentation with once a week administration of a nicotine patch or a varenicline pill on smoking cessation). Today we know not only that e-cigarettes are effective in smoking cessation but that it is currently THE BEST smoking cessation aid available, equally effective to varenicline which, however, has been withdrawn to the market.


The study also reminds me of a letter to the editor I sent together with Ray Niaura from the New York University, commenting on a study by Alzahrani et al. (Prof Glantz was also a co-author) who once again reported that e-cigarettes use was associated with myocardial infarction. In that letter, we criticized the claim about "increased risk" and we presented analysis from the same cross-sectional survey (National Health Interview Survey 2016 and 2017) showing that ever taking prescribed medicine to lower cholesterol (i.e. mainly statins) was independently associated with increased odds of having had a myocardial infarction (OR=2.15, 95% CI=1.65, 2.80, p<0.001) and having coronary heart disease (OR=2.05, 95% CI=1.65, 2.55, p<0.001), a "risk" that was even higher than having hypercholesterolemia!! Obviously such a methodology CANNOT determine any kind of risk.


In conclusion, the study provides zero evidence on any risk associated with e-cigarette use, whether absolute risk or in comparison with smoking. The question that was supposed to be addressed in this metanalysis CANNOT be examined with the studies included in their analysis. This is another example of bad research that not only provides no new knowledge but will probably be used politically to support prohibitive agendas that cause public health harm.




Background Image

Header Color


Content Color