It's a pleasure to be here. You know, this course was started 20 years ago as part of the tumor section, and I'm glad to see that it's still continued on. So thanks to Manish and Mike Link for doing this. For you, I have the burden of being the last speaker here at the end of a long day. And so for you existentialists out there who saw the first black hole picture this week, just to kind of put pineal tumors in perspective, you probably have a more likely to get hit by lightning than have a pineal tumor. So we'll keep that in perspective today. Speaking of that, the historical perspective, you know, in the past, it's really been only the last few decades really that pineal surgery has even been attempted. Historically, the surgical results were very poor. This goes back to Dandy and others who were the first to kind of tackle this. So people used to just shunt these patients and then just blindly radiate everybody. And this is an actual picture from a textbook from the 1950s of this surgical technique. And I think Ian still uses this technique for this posterior fossa tumors. But you can understand why some of those results were not so great. And anyway, now we have much better techniques, the operating microscope and all those other conveniences. And the current emphasis really is on getting a histological diagnosis. We used to see this a lot from patients who had been radiated with pineal tumors. And the one thing about pineal tumor patients, they're actually going to do pretty well in the long run. So they're going to, unlike the glial blastoma patients who only survive a year or two, they don't live long enough to get really all the bad side effects of radiation. These are just the number of patients over the years who, you know, 10, 15 years later develop either radiation necrosis or new malignancies or any kind of demyelinating disease related to that. So one of the things to keep in mind is the fact that there's such a wide variety of tumor types that occur in this area. In fact, there's no area maybe in the entire body where you see such a heterogeneous group of tumors. Among the tumor types, you can kind of categorize them to glial cell, pineal cell, or germ cell tumors with a wide variety of other miscellaneous types. And keep in mind, they're malignant versus benign. They're mixed cell tumors. They're other non-neoplastic pathologies. So getting a histological diagnosis in order to optimally manage these patients is critical. I wanted to say one thing about pineal cysts. We're seeing pineal cysts more and more, and there's actually a movement in certain places in the country to think that patients with headache will get better if you take out their pineal cysts. And I'm here to urge you to not take that bait. I think if we take 100 people off the street and do MRI scans, four or five of them have a pineal cyst. If you take 100 crazy people with really severe headaches and do an MRI, four or five of them are going to have a pineal cyst. So unless you want to buy the farm, and keep in mind that although these operations are safe these days, there's still perhaps a 1% or more mortality. And if you do enough of those pineal cysts with the crazy patients, you are eventually going to have some real trouble. Keep in mind that even despite all of the improvements in radiology and imaging techniques, that radiographic variations are still not reliable for predicting histology. Here you can see three tumors that look almost identical on a T1 with contrast, but the management of a pendomoma versus a pineal cytoma versus a germinoma is very different. And so when we look at diagnostic considerations, I think a tissue diagnosis is mandatory. And of course, all patients have to have germ cell markers because if a patient has an elevated beta HCG or alpha-fetoprotein, then by definition they have a malignant germ cell tumor. And as far as we can tell, there's probably no benefit to surgery in those patients debulking. The number of patients that this affects is relatively small, but the idea is if they have positive germ cell markers, then radiation chemotherapy. And in some cases, you may do a second look surgery later on, but probably hard to justify an aggressive resection in that group of patients. We're not going to have time to talk too much about biopsy versus open resection. I'll just make these points. The advantages of an open resection are be able to make a definitive diagnosis. At least a third of these tumors are benign, and surgery alone is going to be curative. Even some non-benign tumors, such as a pendomomas, can be completely resected and potentially cured with surgery alone. Even the highly malignant tumors, with the exception of germ cell tumors, are going to benefit from debulking. The operative risks are acceptable, and many times you can control the hydrocephalus by getting a radical resection. In terms of endoscopy, this can be useful for selected cases, particularly lesions that are cystic. There is a sampling error because you're going to only be taking small amounts of tumor. Clearly bleeding risks, because you're usually operating on the tumor in the CSF space, where there's no tissue to sort of tamponade any bleeding. One advantage is combining it with an endoscopic third ventriculostomy, but you're going to have to use a separate burr hole for that anyway. It's very difficult to do your ETV and biopsy at the same time. Thinking anatomically, one of the reasons why I think pineal surgery can be done so well is that when you look at the pineal gland, it's really essentially an extra axial structure. And if there are a tumor growing in that, it grows, then there's some important real estate around it. But it generally grows with a capsule around it. Even in cases that are malignant, you can benefit from identifying that capsule and separating that glial margin between eloquent tissue and the tumor. When it comes to surgical approaches, I think it all comes down to, it's either from above and below, there are variations on each, but it's essentially one versus the other. And I'm just going to go quickly over some of these different approaches. I'm going to show a quick video for each of these two. I know we're running late on time, so rather than spend more time on some of the surgical techniques, we'll look at that in the video. But take a tumor like this. Here, a fairly large tumor. And you can see that it really extends pretty far inferiorly. And I think a tumor like this is going to be very hard from a super cerebellar approach. I prefer a super tentorial approach, preferably through the occipital transcentorial. And by coming in this direction, through the interhemispheric fissure, you have a very nice approach to these tumors. So the advantages of the occipital transcentorial approach are that you get a broad exposure with a good view of the more caudal part of the brainstem, particularly in the quadrageminal plate. And you can do this in the prone or lateral position. I usually prefer the lateral position, but prone can be fine. The disadvantages are that you encounter the deep venous system overlying the tumor, so you have to work around that. You do have some occipital lobe retraction, which can give you a temporary field cut. You may have to sacrifice some bridging veins between the hemisphere and the sagittal sinus. And it may be difficult to remove tumors that project pretty far anteriorly. This tumor is not so bad, but some tumors that extend way anteriorly may have a little difficulty reaching under the velum interpositum to get to that. Like most of these approaches to the third ventricle, the setup is critical. If you don't set these patients up correctly, if you don't have the right head position, using gravity in your favor, these operations can be horrendous. So as I say, I usually prefer a lateral position, tilting the head, tilting the nose towards the floor, getting a nice wide exposure at the sagittal sinus in the midline. And then once you do your craniotomy, I like to go across the sagittal sinus. So you're looking at this. There's the phalx. Here's the hemisphere. Our rule of thumb is you can take one bridging vein, but I wouldn't take more than one. Maybe you get away with it, but I think limiting it to one is sort of our rule of thumb. And so sometimes we make a larger craniotomy than we otherwise would, so we have the flexibility to work around the veins. Here's what it looks like once you've got the occipital lobe retracted. There's the straight sinus. You're going to make your cut through the tentorium, just parallel to the straight sinus. Here's what it looks like once the straight sinus is cut. You're looking at the tumor, the cerebellum, deep veins, corpus callosum. And this is what it looks like after the tumor is removed. You're looking down at the quadrigeminal plate and into the third ventricle. Here's a quick video to show. So we have a retractor on the occipital lobe. This is from an earlier video that I have with a retractor on the phalx. Anymore, we essentially do these without retractors. We try not to use retractors at all. If the patient has hydrocephalus, then we put in a drain ahead of time. If they already have an ETV or a shunt, then the brain is usually pretty well relaxed. So here, making the cut with a knife. The straight sinus is right along here. We're cutting through the tentorium. I usually use a bovie. I put a cotenoid underneath the tentorium and then just bovie it. It gets good hemostasis. It saves a lot of bleeding. And you see that you have this nice view of the tumor. You're looking at almost the entire surface of the tumor there. And then once you're there, it's just like any other tumor. It's much like a nice meningioma or something where you just go in, debulk it as much as you can. And I like to debulk it first before I start working the margins. And once you've debulked it, essentially then cauterizing the capsule, shrinking it up. Then you can peel it away from the cerebellum, from any vessels along the lateral side and then along the quadrigeminal plate. And here you can take it mostly piecemeal, but if you've done the hard work of shrinking it in, you can remove the tumor in almost an en bloc method. Here you can see looking in. There's the aqueduct into the third ventricle after the tumor's been removed. And here you can see the post-op scan. The supercerebellar infratentorial approach is the one that we use the most, probably more than two-thirds of the time. And the nice thing about it is it's a direct midline approach. You're working underneath the deep veins. And gravity is assisting the exposure if you do this in the sitting position. The disadvantage is the sitting position. If your anesthesiologists aren't used to this, then it can be a problem. We've done this for many years with dozens if not hundreds of patients with no real complication related to the sitting position itself. Occasionally you'll get some air emboli and you simply cauterize or perform a valve salvo, whatever, to find the bleeding spot, and it rarely causes a problem. You do have to sacrifice some of the midline veins, particularly the precentral cerebellar vein, and that can occasionally be a problem. Like the occipital transcentorial approach, the supracerebellar approach must, the setup is everything. If you don't have the trunk and head flexed as much as you're able to, that means make sure that you don't compromise the airway with that. Ideally, the tentorium is about parallel to the floor, but if you don't get all those angles right, this is a really difficult operation. Here's what you're looking at, midline incision, usually do a craniotomy above the transverse sinus. You're looking at the cerebellum, these bridging veins, we take the ones in the midline, we try and leave all the lateral veins, and here's a close-up. This is what it looks like after you've retracted the cerebellum. There's the quadrigeminal plate, there's the tumor, precentral cerebellar vein, thalamus on each side, and after the precentral cerebellar vein is gone, once again, you have a nice view of the whole dorsal surface of the tumor, and I like to, first of all, dissect the tumor while before it's debulked even, just try and establish this plane along the quadrigeminal plate, then along the thalamus. Once you've started that plane and gotten past the initial arachnoid, then you can internally debulk the tumor and then peel it away from the surrounding structures. This is what it looks like when the tumor is out, and here's a video after you've seen the tumor exposed. Here's the cerebellum down here, tentorium. First we're dissecting out the tumor, and then we're taking these small vessels that are overlying the dorsal surface. Some of those, perhaps, are going up towards the choroid plexus, and there's no real danger in taking those. Once we've come around the bottom of the tumor, put a cottonord or gel foam to make sure that there's no blood getting into the aqueduct, and then we're taking all of these little attachments at the vellum interpositum and these little veins along the side. Once you do that, you start to develop that capsule circumferentially around the entire tumor. In this case, this makes for a nice video because this is one of those rare tumors that you simply remove en bloc after you've developed that plane, again, along the brainstem, quadrigeminal plate, thalamus, vellum interpositum up here, and this nice view into the third ventricle. Okay, so that's great. Everything's easy, no problem, right? Well, here is a patient with a hemorrhagic tumor that we did with the supracerebellar approach. Here's the operative photos, again, taking these veins in the midline. Here's what the tumor looked like at surgery. We sacrificed the precentral cerebellar vein. This is her tumor at surgery, nice, seems like it had a really nice plane around it. Here is after the tumor is removed, gross total resection, nice view into the third ventricle, and about three hours later, she woke up fine. About three hours later, complained of a headache, went in about 10, 15 minutes. She was becoming difficult to arouse, quick scan, she's got a huge venous infarct. So venous infarct, we've done about 200 pineal surgeries, maybe 150 some odd supracerebellar approaches, and we've seen this happen now three times. First time I saw it, I thought, well, maybe it's just one of those things, but after seeing this three times, it's pretty clear that somewhere around 1-2% of patients are going to have difficulty tolerating if you take the precentral cerebellar vein and all those veins in the midline. So that is going to be somewhat of your floor on these types of tumors and what the mortality and morbidity is going to be. So we've sort of pivoted a little bit recently, and we're now doing a lateral supracerebellar infratentorial approach. Instead of going in the midline, we're coming laterally, and the benefit to that is we only have to take maybe one bridging vein between the cerebellar hemisphere and the tentorium, but we're able to leave the precentral cerebellar vein and all those other midline veins intact. And so this is what it looks like. We're able to use a much smaller opening. Here's the, we still use the sitting position, turn the head a little bit, come laterally, maybe midway between the midline and the transverse sigmoid junction. Smaller craniotomy extending above the transverse sinus so we can really get right at the tentorium there. And this is what it looks like. Here's the cerebellum. Here's the dura, tented upward. And here's the tentorium. This is the view that you get, again, looking just off center. And the advantage to this is that you're coming at a lateral angle. You can leave the deep venous system intact, and you have to work around these vessels. You want to keep all those vessels intact, but eventually you can open up a plane between all of these vessels and see the tumor. And this is sort of what it looks like at surgery. Here's the tumor, tentorium. This is what it looks like after the tumor's removed. And again, a quick video to show you what this looks like. After the craniotomy, cerebellum, and now we're at an angle looking towards the midline and where the pineal region and the quadrigeminal plate. And you can see we use minimal retraction. This is the only vessel that we needed to take, this little vein here, which is substantially less than we normally take in the midline approach. Here we're dissecting the tumor off of the thick arachnoid and the precentral cerebellar vein here on the right. And you can see the tumor. It's a smaller space. It's a little harder to work in, but it's the same principle. You can see the brainstem down here. We internally debulk the tumor. In this case, you may have to do a little bit more debulking before you can really establish that plane. But by the same token, you should be able to get the same result. It's a little bit harder, the ipsilateral side. The ipsilateral portion of the capsule is a little bit harder to reach. It's a lot easier to go to the contralateral side. But once you've developed that, you can kind of roll the tumor in on itself. Here you can see looking into the third ventricle. This is what it looks like. And you get that same view after the tumor is removed. Again, no retractors and a well-relaxed brain. So the surgical results, I mean, in our series now, it's up to about 200. We find that about a third of these tumors are benign and that you're able to achieve a gross total resection or a radical subtotal resection, meaning radiographic resection, in about 90% of benign tumors and about 60% of the malignant tumors. But overall, about three-quarters of these pineal tumors, we can get a radical resection. We had two operative-related deaths, both related to this venous infarct. And it's not a matter of the mass effect. You can go in and decompress the cerebellum. It's that the venous infarct extends into the brainstem. And that's what causes these deaths in these two patients. There's a lot of transient major morbidity. Particularly these patients need good ICU care. They've got a lot of air in their head. You have to worry about post-op hydrocephalus. But with a good residency team or good ICU team, most of these patients are going to do very well with a little bit of recovery time. And on a long-term basis, then, the major morbidity is about 2%. For benign tumors, in our series, 100% 10-year tumor-free survival with complete resection. And this, in our case, was 43% of all pineal tumors. Even among the malignant tumors, you can see, with a good radical resection, 5-year survivals of 75%, 10-year survivals of 62%, and a radiographic resection at about 60%. Depending on the pathology, some of them are going to need chemotherapy, maybe fractionated radiation. But the overall point is that with good aggressive management, nearly all of these patients are going to do well. So just to summarize, histological diagnosis is mandatory. Aggressive surgical approaches result in good long-term survival. The surgical morbidity is acceptable with important principles of tumor surgery, internal debulking, followed by a careful dissection of the tumor-brain interface. The surgical options will depend on the location. And a venous impact is an unpredictable and devastating complication in at least 1% of patients. And you can avoid that with an occipital transtentorial or a lateral supracerebellar approach. Thank you for your attention. That was great. Yeah, you know, we don't, it's not a bad idea, we've never had a problem with that, but probably not a bad idea. With the lateral approach, anything with a preoperative venous anatomy that helps you choose what side to come to? Yeah, so, you know, we don't routinely do venograms, you know, from a standard MRI we get a view of where all the, you know, any anomalies in the venous anatomy. In the two or three times where we've had these bad complications, we've looked and, again, maybe with a venogram or something more detailed we might have seen something, but it seemed to us we couldn't, there was nothing unusual about those patients. You know, in this day and age, there's probably no limit to the number of, you know, different preoperative tests you can do. I'll never argue against doing more tests, but we don't as a matter of routine. Yeah, I don't know how you measure that. It's like the Supreme Court in porno. I know it when I see it. It's rare that we say we can't do this because it's too steep. I'm mostly looking at the location, you know, the caudal cranial location, because that's to me what usually is going to decide if I'm going to do an occipital transcendentorial approach. I can't think of a time where we did a supracerebellar approach where we said afterwards, boy, that was dumb. We should have done occipital transcendentorial. It probably happened. I just don't remember it. Okay, thanks.

pineal tumor

surgical techniques

germ cell tumors

pineal cyst removal

surgical approaches

long-term outcomes