So, moving forward, Doctor Orbach was um not able to be here in person, unfortunately, but he has recorded a session for us to view. So we will launch directly into that. Afternoon everyone. I'm Darren Orbach from Boston Children's Hospital. I'm a neurointerventionalist. I'm here to talk about fetal embolization for vein of galen malformation. It's a real pleasure to take part in this Frontiers in Fetal Neurology globalcast, and if the timing works out well, I hope to be able to join live for the question and answer period after the talk. I have no financial disclosures or conflicts of interest to report, and the first question is why contemplate something like fetal embolization for vein of galen malformation? And I'd like to illustrate that with a recent case. This is a fetal MRI scan from a patient from back this past March, and on the top row you can see the DWI and ADC maps from this patient. There are no lesions to see. The brain parenchyma is intact as you can see on the T2 weighted scan on the bottom set of images. So this was on March 5th. Here's the echocardiogram from the same day the baby had cardiomegaly, and they all have cardiomegaly because of increased flow, but there was normal valvular function and normal biventricular function at the time. The baby was born 3 days later on the 8th of March, and look what happens to the echo immediately after birth. Now there is moderate right ventricular systolic dysfunction. There's a little bit of valvular dysfunction, of course. There's increased RV pressure. At this time, the LV is still functioning normally. The BNP, which we use as a marker to follow cardiac function after birth, is very elevated. But devastatingly, look at the MRI on the same day. There's been no treatment here. All that happened between the prior MRI and this MRI is birth, and there's almost a hollow hemispheric infarct of the left hemisphere here in this baby who previously had an intact brain. And let's look at how the case continued to evolve. The following echocardiogram showed worsening. So now you have worsening valvular function. You have worsening severity of the RV dysfunction. Now you have some LV dysfunction, and we took the baby to the angio suite and did an embolization the next day, and there was a very nice response of the cardiac pathophysiology to the embolization. You can see both valvular function improved and biventricular function improved. The BNP is starting to come down, which is great. But you can see the follow-up brain MRI which shows inexorable progression. There are now DWI dots in the right hemisphere, and given the baby's overall condition, the parents opted to go to palliative care at this point. Why does this happen? And the answer is that almost uniquely in the world of neuro, vein of galen malformation is a CNS vascular anomaly that has catastrophic systemic impact. The malformation sustains truly massive degrees of flow where you have biventricular output multiple times normal. SVC return is enormous, and you have over 80% of the total cardiac output just passing through the malformation, not supplying the brain, not supplying the rest of the body. And we know from publications that about 2/3 of the babies will develop a crisis in the NICU. They're the ones who get very sick. They get intubated. They have to go have to undergo urgent neonatal embolizations, and even at expert centers they have a high mortality and morbidity rate. The other third of patients, this infantile treatment cohort, do well in the NICU. We typically watch them and then discharge them after several days. We bring them back electively for embolization, somewhere between 4 and 6 months typically. They of course have a much higher survival rate, though I would point out that even in this group, 30% have a poor neurologic outcome. And here are just some examples from the past few years. This is a 2 week old brain MRI scan. You can see already the extensive abnormalities. Here's a 1 day old really with really no cerebral hemispheres to speak of at all already at 1 day. The other contributing factor is the very severe technical challenge and high morbidity associated with the neonatal treatments that we have, and I thought this was a really important paper published out of Toronto Sick Kids, which is one of the world centers where our current transarterial neonatal embolization paradigm was developed. Pierre Lajoia traveled there from Paris and worked with Karl Terbrug, and they published their their experience. And in a cohort of 33 patients, they report a 30% major procedural morbidity. So this is added on to the aggressive natural history, 15% procedural mortality. So half of those treated patients who had a procedural morbidity issue had mortality. So just to return to the case that I opened this talk with, you make an observation that we see frequently in this condition, which is that in utero, the heart and the brain seem to be protected from injury in most cases, not in every case, but postnatally that's where you get high output heart failure. And where you start to get brain injuries accruing. And why is that? Almost certainly the protective effect in utero is due to the presence of the placenta, which of course is a massive, low resistance vascular organ, carries almost 50% of the total fetal cardiac ventricular output, acting as a huge sump. And then when you lose the placenta, the cord is clamped after birth. You have that entire flow burden going onto the newborn's heart and brain, and that's where the pathophysiology starts to aggressively develop. The second point is that we know from many studies that fetal brain blood flow is far lower than neonatal brain blood flow, and in fact even the newborn's brain blood flow after birth undergoes a massive increase in the first postnatal days and hours. And intriguingly this actually corresponds quite well to the time when most newborns with vein. And malformation start to decompensate. So this is known from many studies, as I said, this is one example. This is based on ultrasound, and if you look at the bottom row, which is ACA flow, time 1 hour to 2 hours after birth, and then compare that with 24 to 48 hours after birth and time 3, you can see there's a near doubling of ACA flow. These are normal term newborns. And of course you can look at this with arterial spin labeling perfusion MR, and the newborn increase in blood flow continues on through childhood till about age 7 or 8. And as I said, this has been very well established. So that raises the question of perhaps this paradigm we have of watching the baby with vein of galle malformation be born, experience the loss of the placenta, and undergo that terrible decompensation and then attempt to reverse that is all wrong. And perhaps we can potentially circumvent or at least mitigate the aggressive pathophysiology by working in the fetal environment which offers the protection of the placenta and the lower expected level of cerebral blood flow under normal conditions. So the first thing you have to do is to be able to identify the right cohort. I had mentioned that about 1/3 of the babies will do well in the NICU. They are not an appropriate cohort for any kind of new fetal intervention. So we looked at this systematically at the start of this project, and it turns out you can very well predict how a given fetus is going to do after birth by measuring the width of the thalcine sinus. You can see that measurement with the dotted dotted red line on the right side. This is the mediallateral width, and if you look at the Chart, it shows how beautifully the width of that sinus correlates with the likelihood that that fetus is going to go into crisis in the NICU. By the time you get to 7 millimeters, there's an 80% likelihood that that baby is going to be in crisis in the NICU. So now we have an identified cohort. And then of course what is feasible. Fetal surgery is a very dynamic and growing field, of course. But the least invasive and least morbidity types of procedures for mom and for the pregnancy are ultrasound guided needle-based procedures. As a paradigm, there are the needle-guided cardiac procedures first performed for congenital aortic stenosis that results in hypoplastic left heart syndrome. Where it was found that if you do aortic valve balloon dilatation in utero, you can mitigate significantly the aggressive presentation of hypoplastic left heart syndrome. And here you can see a trochar going through the fetal sternum into the left ventricle. Here you see a wire crossing the aortic valve in preparation for balloon dilatation. So this raised the question, could we do something analogous for vein of galen malformation? Most definitely not with the intention of closing the malformation, but just packing the Markowski varin enough to diminish the flow and mitigate that aggressive presentation after birth. So to design this, we had the simulations group at our hospital build for us simulated fetal brain phantoms built out of cryogel which looks just like ultrasound, just looks like just like brain under ultrasound, and these were based on actual fetal patients of ours with a varinx inside and a falcine sinus taken straight out of the measurements from real patients. And then we did with the phantoms what we were planning to do with the fetuses so you can see the preoperative planning scan. Here are my MFM colleagues' hands doing the actual access under ultrasound, and this is a process that went on over several years. And of course the question was, all of our neurointerventional tools are designed to be used under fluoroscopy. Could we even see what we're doing under ultrasound? And here's what the coiling looks like. You can see the increasing posterior acoustic shadowing as the coil density increases, but you can still see the coils unfurling. Here's the post op. Appearance of the under MRI of the phantom. And then we also had the simulations group designed for us occipital bone phantoms so that we could practice manually boring through with a needle in a very controlled fashion and putting all that together, we were able to get IRB and FDA approval for a clinical trial of fetal embolization of vein of galen malformation. And to this point we've enrolled 7 patients in the trial. We've been successfully able to treat 5 of them. I'm going to go through the first patient and give you an overview of the results. So this first patient was diagnosed back in February of 2023 on ultrasound. Here you can see the progressive enlargement of the falcine sinus on MRI. It went from 11 millimeters to 13 millimeters, and if you look at the chart, it's almost literally off the chart in terms of risk to this fetus. Brain parenchyma was intact in utero, so a great candidate to have fetal embolization. And on March 15th of 2023 we did go to the OR, and here you can see how well the phantoms predicted what the appearance of the actual embolization in the fetal brain would look like. You can see the needle going through the occipital bone on top, the wire entering the Markowski varynxx on the bottom. And here you see the progressive coiling of the varynx under ultrasound with the increasing shadowing, and we were able to convert the appearance of the malformation from this very high flow malformation on the left to this appearance on the right where we lost the color Doppler signal behind the coil mass and we stopped embolizing. There was an immediate impact on the post-op fetal echocardiogram. You can see this dramatic diminution in cardiac output. And there was an immediate impact on the post-op fetal MRI scan. You can see that the falcine sinus width went down to 8 millimeters, and the varynx itself started coming down in size. Much to my chagrin, the mom had premature rupture of membranes, and the baby was born only 2 days after the embolization, but it turns out it was completely fine. This was as if a vein of galen baby had fallen from a parallel universe into our NICU. Basically no cardiovascular support, no medications, no neonatal embolizations. We just essentially watched the baby for several weeks in the NICU. We watched the postnatal echocardiographic parameters improve on their own. Here's the appearance on day 26 just before discharge from the NICUU. This was the appearance at 4 months, and this is at 10 months. We've brought her back since then at 8 months and 10 months by design for near complete closure of the malformation with elective embolization. That's exactly how the trial is designed to work. She's now 1 year and 7 months old on no medication, neurodevelopmentally intact, and we wrote this case up back in June 2023 as a case report in stroke. How are we doing so far overall in the clinical trial? So you'll recall that I mentioned that these wide falcine sinuses allow us to predict with great confidence that the baby is going to get sick after birth. However, these wide falcine sinus cases are the small minority. So most of the babies we're actually treating in the NICU have falcine sinuses of 456 millimeters. That's much more common. And the literature we have that exists about how newborns who have an embolization, a vein of galen malformation fare is based on mostly those moderately sized falcine sinuses. That's a different question from asking how do these this cohort of very wide sinuses do. So we went back. Looked at that systematically in our own experience over the past decade before there was any fetal embolization, and it turns out these are the truly horrible vein of galen malformations with the very wide sinuses, you start approaching 100% mortality despite care at an expert center and you could run a logistic regression and basically see that every single case of survival. Again, with management at an expert center has a moderate or narrow falcine sinus. All the truly wide cases have mortality. So now let's go back and look at the cases that we treated the fetal with fetal embolization and look at the width of those sinuses. Without fetal embolization, this cohort has a 90% mortality. And as I said, we were successfully able to treat 5 of them. So there was 1 case of complete treatment futility. This baby was born, had bad aggressive cardiac presentation on day 1, was intubated, and then just rapid worsening of the cardiac function. We literally couldn't get him to the angio suite on time to even do another embolization. We had another case with an excellent response to embolization, and tragically this baby got gram-negative septicaemia from a urinary tract infection. This was crushing for all of us, for the family as well, but we have 3 babies who have done fantastically well, all on no medications, neurodevelopmentally intact out of the 5 we were able to treat. I just want to say a word about maternal and pregnancy morbidity. One thing that was surprising to us is that we have found a high rate of premature rupture of membranes, chor and amnion separation, worry of potential placental abruption leading to urgent delivery, and this is quite different from our experience with ultrasound with fetal cardiac interventions where at Boston we've done 260 procedures, virtually no placental abruption, very rare to have premature rupture of membranes, one case of card amniotic separation. So why is that? Almost certainly because the gestational age window in which we're working is quite different. You can see that in the cardiac cases, most are done between 23 and 26 weeks. And look at when we're working with vein of galen malformation. This is mid to late 3rd trimester, where we're working in the lower uterine segment, which is already quite thin at that point. The head is already down. And so we need to change the precise gestational age window that we're doing it, and the current paradigm we're working on is to have a two-arm cohort with an early arm where we intervene before 32 weeks, where the outcome for pregnancy should be similar or identical to the fetal cardiac interventions. Of course the main challenge there is that this is often diagnosed after this gestational age window, as you can see here, the age of diagnosis of the cohort that we've treated so far. And so there has to be a late arm as well where we perform the embolization at 37+ weeks, which is when we would be delivering the babies anyway, and we plan to induce delivery within just a day or two after the procedure, just enough lag for the fetal paralysis to wear off, and that way we're not doing the procedure and hoping that the pregnancy continues continues long enough without any issues to get to delivery. We're still able in this way to leverage the benefit of the presence of the placenta while we change the cerebrovascular and systemic blood flow, but we can largely mitigate the risk of threat to the pregnancy. So essentially the plan is not to do any fetal embolization between weeks 32 and 37, which is when we've been doing all of our cases to this point. I just want to briefly mention there have been a few cases attempted outside of Boston Children's Hospital. One has been published. This was at Necker Children's Hospital. AV heard me give a talk about this several years back, and they had an excellent case for this. That baby did fantastically well. There was one round of embolization in the NICU, and then the baby was discharged home, elective embolization at about 2 months, and at 18 months, that baby has normal MRI scan, normal exam. So to conclude, the current paradigm we have of transarterial embolization of newborns who get sick in the NICU has definitely been transformational. This condition used to be nearly universally fatal before modern embolization was developed, but we are still failing a high fraction of patients with still a high severe mortality and morbidity. There are likely significant advantages to reducing the malformation flow in utero before this aggressive pathophysiology starts. The fetal cohort that we've identified for inclusion in the trial have an extremely high mortality rate when treated by standard routine means, far higher than the published values in the literature, which are based on all newborns who need to be treated and undergo embolization. While the safety and efficacy of this approach have yet to be determined based on early results, morbidity appears to be low, and efficacy is potentially excellent. However, fetal embolization between weeks 32 and 37, as we have been practicing it so far seems to result in a high rate of premature rupture of membranes, so a modification of the time window is needed, and I really cannot overstate the importance of completing a rigorous trial before disseminating this technique or recommending it. Thank you so much for your attention. As I said, I hope to be able to join you.
Click "Show Transcript" to view the full transcription (18843 characters)
Comments