Among the most heated debates on #cardiotwitter is the meaning and value each of us gives to meta-analyses. The detractors of these types of studies are vocal and they don’t like to be contradicted. For them, meta-analyses amplify the noise of each individual trial and – as they say – “take the confusion to a higher level”.
During my career, I’ve been the author of some meta-analyses myself and I can truly say I understand this point of view. However, I believe it is based on the incorrect assumption that performing a meta-analysis simply means using one of the numerous software programmes available with the aim of cannibalising the work of others. Instead, performing a meta-analysis means having a good idea, working on an efficient search string, establishing a protocol with inclusion and exclusion criteria, applying it methodically, often screening thousands of articles, looking for sources of inconsistency and interpreting the data correctly. Thus, as with any study, the solidity of the results depends on that of the methodology.
Even the most avid detractors will admit that there are excellent meta-analyses, examples of profound logical, mathematical and intellectual reasoning. Claiming that meta-analyses connect “apples and oranges” is a hackneyed argument that may have its supporters; however, many others know that the goal is not to study either the “apples” or the “oranges”, but “fruit” in the broader sense – to find a more general meaning of things and understand something that goes beyond individual studies. Clearly, the pitfalls are many, and there are obviously also unnecessary meta-analyses or ones that are wrong and misleading due to their execution.
Because we believe in this tool, if used well, from the very beginning EuroIntervention has had a Section Editor who is our meta-analysis expert. It is their job to prevent the journal from publishing bad quality meta-analyses and help the authors of the most deserving articles improve the clarity of their methods and the depth of their results. Also, for several months we have been making sure that the authors of the meta-analyses who enter the review process understand the reasons for a possible rejection with constructive and specific comments.
In this issue, as I will explain below, we have a good example of a meta-analysis that has matured with this strengthening process and which we hope will pique your interest as it did ours.
First, though, we begin with a EuroIntervention state of the art surveying existing and potential enhancements for the cath lab through advances in imaging, robotics and artificial intelligence. Authors Rafael Beyar, Nico Bruining and colleagues discuss the impact these developments can have in terms of improved treatments for the patient and increased safety for the operator and cath lab staff. The evolution in image-based tools combined with the latest innovations for visualisation and display, the integration of artificial intelligence and the increased sharing of data internationally allow better diagnosis, management and reduction of X-ray exposure. The emerging use of robotics – seen increasingly in peripheral and neurovascular work – promises to become the surrogate hands of the operator, increasing precision, and providing better outcomes and less radiation exposure. The convergence of all these points may lead to future automation and remote “tele” management of distant patients.
We return to a meta-analysis in coronary interventions in which authors Raffaele Piccolo, Peter Jüni and colleagues compare the impact on mortality between bleeding and myocardial infarction in patients with coronary artery disease, further illuminating the importance of bleeding as a predictive factor. Major and late bleeding after percutaneous coronary interventions (PCI) was seen to have the same risk for mortality as myocardial infarction and should be considered prognostically equivalent. The authors go on to stress the importance of bleeding avoidance strategies among patients undergoing PCI, with early bleeding having an even stronger association with mortality than early myocardial infarction. This article is accompanied by an editorial by Michelle L. O’Donoghue and Siddharth M. Patel.
Coronary sinus narrowing therapy for refractory angina is the subject of an article by Stefan Verheye, Maayan Konigstein and colleagues. They report on early interim results from the REDUCER-I trial, examining the safety and efficacy of the coronary sinus reducer offering a possible treatment option for patients suffering from angina refractory to existent medical and interventional therapies. In the treatment of stable refractory angina, results show that Reducer implantation appears feasible and safe with a sustained improvement in angina severity and in quality of life up to two years. This article is accompanied by an editorial by Michael Foley and Rasha Al-Lamee.
Can a measure of coronary microvascular function such as coronary flow reserve or hyperaemic microvascular resistance predict major adverse cardiovascular events or stroke? This is the question posed by Takumi Toya, Amir Lerman and colleagues who examined the outcomes of these measures as continuous variables. They observed that lower coronary microvascular function and higher hyperaemic microvascular resistance were associated with increased risk as continuous variables, thus confirming their prognostic value in evaluating long-term risk for patients with ischaemia and non-obstructive coronary artery disease. This article is accompanied by an editorial by Christos C. Bourantas and Karthik H. Chandrasekharan.
What diagnostic value could coronary computed tomography angiography-derived fractional flow reserve have as a second-line test in symptomatic patients with suspected obstructive coronary artery stenosis? Authors Jelmer Westra, Shengxian Tu and colleagues compared coronary computed tomography angiography-derived fractional flow reserve to myocardial perfusion scintigraphy and cardiovascular magnetic resonance and found the diagnostic performance to be similar among the three measures. With further experience and studies, this could prove useful in reducing additional downstream diagnostic procedures and help to ensure more appropriate patient referrals for invasive coronary angiography aimed at identifying obstructive coronary artery disease. This article is accompanied by an editorial by Daniele Andreini.
In peripheral interventions, certain meta-analyses of randomised trials found an association with a worsening long-term survival using drug-coated devices. In a large cohort of patients who were treated as inpatients and outpatients by femoropopliteal artery revascularisation procedures, Eric A. Secemsky, Robert W. Yeh and colleagues reconsidered the question from the standpoint of long-term mortality. In this cohort where patients were treated with either drug-coated or non-drug-coated devices the authors found no evidence of increased long-term mortality following treatment with the drug-coated devices, regardless of device type, procedure setting or patient characteristics.
Finally, in carotid artery stenting the risk of thromboembolism is of great concern. Authors Ichiro Nakagawa, Hiroyuki Nakase and colleagues used catheter-based near-infrared spectroscopy to study carotid lipid core plaque and determined that high lipid core plaque was associated with cerebral embolism. If high lipid core plaque is detected in this manner, the authors suggest that more recent iterations of stents should be used, and reinforced methods of neuroprotection should be employed.
Now let’s see what you think about these articles.
Supplementary data
To read the full content of this article, please download the PDF.
