 Good morning. My name is Alexander Postallian. I am an interventional cardiologist that works here at the Texas Art Institute. First of all, I want to thank Dr. Stephanie Coulter and the entire organizing team for inviting me to speak in this meeting. It is a pleasure, a privilege and an honor. And I'm going to talk about Payton Ferramino Valley or PFO for a short and migraine headaches, which is a bit of a controversial topic for which we have new data that is worth discussing. So let's see where do we stand today in the matter. These are my disclosures. I have a hammer. If you know that saying, if you have a hammer, everything looks like a nail. Well, I am an interventionist. It's a big part of what I do, but I'm also an enthusiast of doing the right thing every time. So I'm going to call it a reasonable hammer. And everything I say today I think is fair game with maybe a little bit of bias, but not too much. And to prove that, I'm going to start with the conclusions and say, don't do it. Don't close PFOs for migraine headaches yet and maybe never, but let's explore the topic a little bit. But this, at first glance, the relationship between PFO and migraines seems a little murky, kind of hard to understand. But we'll get through it and we'll talk about what each of these are, PFOs and migraines, how they are related, why it makes a lot of sense that they are related and why it doesn't make any sense that they are related. Hopefully at the end, we'll figure out what to do. And fair warning, I'm going to be including some memes and gifts in this presentation and that's what happens when you give a millennial cardiologist a platform such as this one. So what is a PFO? A PFO is a communication between the right and the left side of the heart. It is actually necessary for the fetus. And the fetal lungs are collapsed. The fetus is floating an amniotic fluid. It's not breathing. So because the lungs are collapsed, the vasculature of the lungs is also collapsed. And there has to be a way for the blood to go from the right side of the heart to the left side without going through the lungs. And that's when this little communication takes place. And it is important. So you get oxygenated blood from the umbilical cord, goes to the right side of the heart, then that goes to the left side and then goes to the body. Then when the baby's born, takes a breath, the lungs get expanded and the resistance of the pulmonary vasculature drops massively and now blood starts going from the right side to the lungs and then to the left side. And this eventually causes that little valve to approximate and once the two tissues are touching each other, it fibrosis and it closes off in most patients. But there's still a communication at about 25 to 30 percent of adults when looked at in autopsies. And this serves as a channel of communication between the right and the left side. And 92 percent of patients with PFOs have intermittent, at least, at least, communication between the left and the right side. Another important piece of information is that if a patient has a stroke and you look at patients with stroke only, for which we cannot identify a cause, they don't have blockages in the arteries that go to the brain, they don't have arrhythmias that may cause a stroke, then 46 percent of those patients have PFOs, which is more than 25 to 30 percent in the general population. So that tells you that there might be an association between PFO and stroke. When do we close these, quote unquote, defects? Now it is recommended, or reasonable at least, to close them in the case of stroke of undetermined cause and they're most effective if the strokes are recurrent. If there's a lot of shunting of blood from one side to the other in specific situations, for example, positional, you can also close them. We close different types of communication, not PFOs, but ASDs, which more than a valve tissue is more of just a hole between the right and the left side. And if they are causing hemodynamic consequences, we close them. It's worth noting that nowadays this is a pretty safe and straightforward procedure relatively. The 2021 guidelines for prevention of stroke put forth by the American Heart Association. Now they say that it is considered reasonable to percutaneously close a PFO in patients with a stroke of undetermined significance if you meet certain criteria. And this was after a lot of studies and a lot of controversy studying the subject. Well now we talked about PFOs and let's talk about migraines. It is a severe episodic headache that is associated with nausea and aversion to lights and loud sounds. The pathophysiology or how this headache takes place is a little bit murky. And as you can see on the right side of the screen, we have a fairly complex diagram explaining how the headache originates. And I'm not a neurologist, but typically when we make things, we show things to be complex such as this one. It means that we probably don't understand it very well. We used to think that the main phenomena that caused and maintained the headache were vascular phenomena. But now we think it's more a neuronal change. It's a phenomenon called cortical spreading depolarization. That would be the initiating factor. And then you get the vascular changes that are associated with migraine. These are all theories, what established, but like I said, it's hazy. Migraine is very frequent and it's much more frequent than men than in women. So about 17% of women have migraine headaches at least once in their lifetime. And it's most common at the young ages of 30 to 39. And it's a very common cause of disability when these pains, these headaches occur. They can be severe, they can be frequent, and they can be disabling. There's a classic description of migraine includes four phases. The first is a prodrome, which is essentially fatigue and not feeling well before the headache starts. This is followed by aura. And aura only happens in about 25% of patients. And it's a very interesting thing that doesn't happen very often in other types of headache or at all. And it is sensory neural perceptions. Either visual, you see bright lights or blind spots, you hear things, you feel things. And then this is followed by the full blown headache that can last 472 hours. I already described it. It's mostly Positile, some of this unilateral associated with nausea and vomiting. And this is followed by the post-drone, which is also a period of fatigue after the headache that can last days. The pathophysiology is complex, but profusion may indeed play a role in either the initiation or during the headache itself and the aura phenomenon. For example, this is an old study. They looked at cerebral blood flow. And this was in 1990. And they found that when the subject had aura, there was evidence of hyperperfusion in the brain. Then the headache started, and this was followed by hyperperfusion. And after the entire episode is over, then you have normal cerebral blood flow. And on the right, this is the brain of a patient. White means normal blood flow. Black means less blood flow. You can see, for example, in the top at six, before the aura begins, everything is white. And then you start moving down and you see how the backside of the brain starts becoming dark during the aura episode, meaning that the occipital brain lobes were getting lower blood flow than the rest of the brain during aura. And that is the part of the brain that is in charge of processing visual stimuli. So that's something to think about. Now, we talked about each one of the conditions. Now, what is the statistical association between both? There was this systematic review done in 2008. But there were many studies done before this one that suggested an association. And this paper found that if you had a PFO, there was a significantly higher likelihood of you having migraines. And the other way around, if you had migraines, there was a significant likelihood of you having a PFO. And it was pretty significant to simplify approximately three times more than normal. Then there was this paper published in 2018. They looked at patients that already had a stroke of undetermined cause. And they saw how many of those had migraines, how many of those had PFOs. So if you had a stroke of undetermined cause, 59% of those patients had PFOs, which is significantly higher than the incidence in the general population. If you had a stroke and migraines, any type of migraine, then 79% of those patients had PFOs, which is, again, higher. And if you had a cryptogenic stroke, migraine and aura, which is only in 25% of patients with migraine, then 93% of those patients had a PFO, which is a pretty strong numerical association. Now enter white matter lesions. These are little white spots that we see on brain imaging that are changes related to fibrosis and sometimes to ischemia and small infarcts. And we found that these lesions are more common in patients with migraine and incredibly more common in patients that have migraine and aura. So are these many strokes? Something that I haven't talked about yet is that the reason why a PFO might be associated with cryptogenic stroke is because you can have small clots that can travel from the right side of the heart to the left side of the heart. And then on the left side, they can travel to the aorta, the carotids, the brain and cause a stroke. Also, you can have clots that form on the PFO itself break off and cause a stroke. So what this study is suggesting is that maybe these little white matter lesions are mini strokes caused by little pieces of clot that travel because of the existence of a PFO. So a PFO might originate these lesions and originate migraines. So why does it clearly make sense that these two things are associated? And I warned you about the gifts. There is a somewhat strong statistical association. The small emboli that causes ischemia may trigger a migraine episode and you see with the white matter lesions what I just talked about. There's also a theory that possibly substances that are normally metabolized by the lungs can bypass the lungs, go from the right side of the heart to the left side of the heart, travel to the brain and trigger a migraine attack. And in rats, if you inject micro emboli or micro clots in the carotids of the rats, you can trigger corticals spreading depression or depolarization and trigger a migraine attack. But could also not make sense. Why? Because a lot of studies have shown no association and one big one that had over a thousand patients did not show statistical association. It was also in 2008. The white matter lesion distribution is not associated with the aura phenomenon or where it should be at least. Meaning you can have white matter lesions in the anterior side of the brain and the patient can have visual aura which is in the occipital part of the brain so it doesn't really make pathophysiologic sense. The most recent theories that it's a neuronal process that starts in the brain not related to ischemia don't fit entirely with the hypothesis that embolism is contributing to this and the quality of evidence of the studies that have looked at this association is pretty low. So honestly it's hard to say. I will say that it is fair to assume that there is an association based on most of the data that is available but we don't know until we know and how do we know what happens if we fix the PFO. If we close the PFO, do the migraines go away? So enter the clinical trials. The number one was called the MIST trial. This study was done in 2008 in circulation. It was done in Europe. It evaluated PFO closure with the Starflex device versus a sham procedure. So pretty high quality control group and they included patients that had migraines that failed medical therapy and that had significant shunts on imaging so they had PFOs. The primary endpoint was a pretty lofty goal of no headache whatsoever at three to six months during follow-up. That's difficult to achieve. They had many secondary endpoints about migraine frequency, migraine days etc etc. 74 patients underwent the PFO closure procedure and 73 underwent the sham procedure and they found no difference in the primary endpoint or most secondary endpoints except for a small reduction in total migraine days favoring the PFO closure group. So this is Dr. Andrew Dawson, he's a neurologist and it looks like it's a no for that group. He was one of the leaders of this trial. Now that trial had some issues. Like I said it was an unrealistic primary endpoint, very difficult to achieve, complete absence of headache in such a short follow-up period. There were a few patients in the trial but this is expected because not many patients meet inclusion criteria for these sort of study. It's kind of restrictive and this device was an early device, the Starflex. The company actually went bankrupt after this study was polished and then there was another one that showed it to not be effective for cryptogenic stroke either so it's gone and it had the high complication rates at least in this study which is not what we have today. Now comes trial number two, the Prima trial. So this study was polished in 2016. It was done in Europe and Canada and it was polished in the European Heart Journal. They looked at Amplatser PFO occluder device versus medical treatment and this is a device that we use today. Again no sham control group, this was medical therapy. They included patients that had migraine with and without aura that failed prophylactic therapy. The primary endpoint was a reduction in migraine days at 9 to 12 months so something more achievable than the first trial. It had multiple secondary endpoints, 53 patients underwent the procedure and 54 underwent medical therapy. If you look at this table you'll notice for example in line one patients that had PFO closure at baseline they had eight migraine days with and without aura and that went down to 5.1 after PFO closure and the control group on the second line had also at baseline about eight days of migraine days and that went down to 6.5 so it both went down but it went down less than with PFO closure. If you go to the next point, migraine attacks with or without aura, PFO closure group started at 5, mean migraine attacks went down to 3, then the control group started at 5 and went down to 4 so both decreased less as compared to the PFO group. However none of these reached statistical significance. If you look at some of the secondary endpoints for example migraine with aura days, migraine attacks with aura, it went down significantly in the PFO closure group so it seems like if the patients have aura PFO closure is more efficacious. An important point to mention is that about 10% of the patients in the PFO group had complete and total cessation of headache versus none in the medical treatment group which is quite remarkable. It's a signal to keep an eye on at least but this trial also had some issues. It stopped early because of slow enrollment. It had few patients for the same reasons I already explained. It wasn't sham controlled so here the placebo effect starts to kick in and placebo effect can be very significant. It failed to eliminate shunt in about 12% of the patients and there were quite a bit of patients lost to follow-up. So it looks like for this research team, this Dr. Meyer, it's a maybe. Okay so we have a no and a maybe. The first one was a neurologist and this is an interventional cardiologist. Now we have number three, the premium trial. This trial was published in 2017 in the Journal of the American College of Cardiology. It was done in the U.S. It also looked at the M. Platzer PFO occluder versus medical treatment again no sham arm in this study. They had patients with migraine both with and without aura that failed prophylactic therapy and had significant shunt on imaging so a similar inclusion criteria to the previous study. The primary endpoint was a responder rate that was greater in the PFO closure group. They had multiple secondary endpoints. 123 patients underwent PFO closure and 107 had medical therapy so this is a larger study. They found no difference in the primary endpoint but there was a significant reduction in migraine days so it decreased by 3.4 days versus two days at PFO versus medical therapy and it was a higher rate if you only looked at patients that had aura before the migraine attack. So this is Dr. Jonathan Tobis and looks like for him was also a maybe and he is also an interventional cardiologist so two maybes and a no. So what are the newest data? We have three trials without a clear conclusion to be honest. So there was a pooled analysis was done recently and published this year 2021 in February and they pooled the patient level data of both Prima and Premium so the two more recent trials and mist was excluded and the reason this was done is because Prima and Premium were both amplizer PFO occluder devices versus a Starflex which was the mist and when they pooled both analyses together and their intentions did analysis they found a mean reduction in monthly migraine days of 3.1 versus 1.9 in the PFO group favoring the PFO group and it was significant. There was a significant reduction in the amount of migraine attacks. There was an improvement in the responder rate 38% versus 29% but that was not significant and they saw a complete cessation of headaches in 9% of the patients that had PFO closure versus 0.7% of the medical therapy patients which was also significant. Now both these trials are very low serious procedural complications of under 1% which is in line with modern procedural technique. Now the issues with the pooled analysis is that it's a pooled analysis. It's not a new trial so there is inherent bias and possibly spin. This is something that is fairly common in medical literature lately sometimes it is done inadvertently by the investigators but you have an opinion before you even start writing the study so the way the study flows the way the conclusions sound seem to favor one outcome versus the one therapy versus the other and also the the outcomes that you choose can sometimes be that can be a biased decision so it's possible that there's some of that in these pooled analysis on the two most recent trials and it mixed patients that were different patient populations similar but different. There's something else to consider other than PFO closure versus medical therapy so why does the medical therapy has it been the same or has it changed over the last few years and it has changed so there's drugs that have been around for a long time to prevent migraine headaches we have beta blockers calcium channel blockers anti-depressants anti-seizure medications some other drugs that have been used like case inhibitors then there's a new class of agents the calciton and gene related peptide antagonists and then there's other non-drug therapy measures that can improve migraine like aerobic exercise biofeedback relaxation etc etc etc there are more options now than there were before so it doesn't seem like the pathway of a pfo causing or contributing to migraines is a straight line it is more of a complex convoluted route that may not even be there but there is a likely association i would say now does i mean that if you close it will it get better well it looks like we still don't know so for now don't close pfo's to treat migraine headaches yet and maybe never however there is a possible role in a select group of patients namely a patient with severe migraine headaches with aura that is unresponsive to any other type of therapy and has a significant shunt because of a pfo i think that's a reasonable thing to consider but again not yet there is another trial that is happening right now that is a relief trial it's going it's looking at another device so it's the gore occluder device and this is a sham control study so we'll see what that shows that may help break a sort of tie with the currently available evidence thank you very much for having me again for taking the time to listen to this presentation this is my name my email i would love to hear questions comments or concerns so please reach out thank you very much