Dr Azeem Latib (Montefiore Medical Center, New York, NY, US) discusses predictors and implications of periprocedural myocardial infarction from the Onyx One Global Study.
1. Can you describe the Onyx One Global study and what it showed?
2. Why did you conduct this initial analysis?
3. What were the findings?
4. What conclusions can be drawn from the data and is further research required?
Filmed on location at CRT 2020.
Interviewer: William Cadden
Videographer Charlie McClanahan
Question 1 : Can you describe the Onyx One Global study and what it showed?
[Latib] The Onyx ONE Global study was a randomised 2,000 patient study which was designed to evaluate the efficacy and safety of two modern drug-eluting stents in patients at high bleeding risk. So what we did in the study is we took those 2,000 patients, who identified as having high bleeding risk based on a number of different factors such as age, over 75, using anti-coagulants, being at risk of bleeding, having low platelet counts, needing surgery and these are patients, because they're high bleeding risk where we would give them only one month of dual antiplatelet therapy. And so what we did with these 2,000 patients we randomised them, either to the BioFreedom stent which is a polymer-free drug-eluting stent and in Europe the only stent really approved for this indication of one month of dual antiplatelet therapy, versus a really a modern dual polymer drug-eluting stent which is the Medtronic Onyx stent. So once patients were randomised they were then treated, the unknown PCI were treated with either one of the study's stents, and in 30 days then one of the antiplatelet drugs were stopped and we then, at one year looked at two endpoints the primary endpoint was a safety endpoint really looking at the risk of mortality having an infarct, and stent thrombosis and then there was also an efficacy endpoint looking at target vessel failure. So what the studies showed was that there was really no difference in the primary endpoint between these two stents. That using a modern, new dual polymer stent like Onyx was as safe as the BioFreedom stent, and there's also no difference in efficacy at one year.
Question 2 : Why did you conduct this initial analysis?
[Latib] After reporting the primary endpoint and looking at the outcomes in the primary endpoint one of the things we noticed was that the event rates were higher than we've seen with some of the other drug-eluting stent trials. We saw for the primary endpoint event rates it was high as 17%. When we looked a little bit more carefully at the data we saw that most of these event rates were being explained by myocardial infarction. And so we then did a sub-analysis to really understand why myocardial infarction rates were higher in this study as compared to previous studies.
Question 3 : What were the findings?
[Latib] So what we found were a couple of things, firstly, myocardial infarction in this study was based, the definition was based on the third and fourth universal definition of MI which really requires a raise in troponin more than 5 times the upper limit. Plus, evidence of ischemia such as angiographic occlusion of the vessel, ST changes on ECG, or chest pain. What we found in this study is that if you looked at myocardial infarction rates, most of the increase in myocardial infarction in this study was explained by periprocedural MI, and not by spontaneous MI. So the majority of the MI's were actually just periprocedural. And when we looked at the two stents there was no difference in periprocedural MI between the two stents. We then also did a landmark analysis where we analysed outcomes up to 30 days and then reanalysed them from 30 days to one year and we found there was no difference in MI rates up to 30 days, but from 30 days to one year there was a difference. The Onyx stent was associated with a lower MI rate. Now, what does that mean? We can theorise about it. Could it be related to the dual polymer, could it be related to the stent design and the thinner stress, we don't really know, I mean it's all hypothesis generating because the study was not designed for that endpoint. It did, though, highlight a couple of other another important issue was really how relevant is periprocedural MI rates. So in this study, the Medtronic who sponsored this study and also provided oversight in [inaudible] study were really good at ensuring that patients got their post-procedural troponin so, they will be, I think there is some ascertainment bias. What I mean by that is that because the sites and the study was so well followed that most patients got their troponins measured we picked up more periprocedural MI's. Now, is periprocedural MI's a bad thing for patients? That, we don't really know there's conflicting evidence. Probably if patients have very large periprocedural MI's, those could be significant but probably small troponin rises are not as significant. So when we went into the Onyx 1 and looked a little bit deeper at the data, what we found was that probably just slightly less than 40% were probably what we would call clinically relevant periprocedural MI's Okay. In the sense that they had very large troponin rises. Most patients had these small troponin rises, more than 5 times upper limit of normal. And most of them it was associated with probably the operator noticing a small side branch or small septal branch being occluded. So, without evidence of ECG changes, or chest pain only a minority of patients actually only had ECG changes and chest pain. So it broadened the question really how relevant is the periprocedural MI. We then also looked at, and did a multi-varying analysis to try and understand what were the predictors of mortality in this study. And what we found was that periprocedural MI was not an independent predictor of mortality. What was an independent predictor of mortality was spontaneous MI. So the patients who had an MI after 30 days, those patients had up to three higher risk of dying.
Question 4 : What conclusions can be drawn from the data and is further research required?
[Latib] I think the study, in the end, you know, makes us realise that periprocedural MI sure, it is an endpoint we use in many studies but it's not all periprocedural MI is important. It's probably the ones that are more clinically significant that are important. And secondly, that more important than periprocedural MI is spontaneous MI. So, what we don't know from the study is if we continued dual antiplatelet therapy in these patients beyond one month would that have decreased some of the spontaneous MI rates? And I think, that's an area for more research. You know, would, in high bleeding risk patients is one month of dual antiplatelet therapy as good as three months. And probably that's where the next study should be going.