So, good afternoon, everyone. I was given the task to talk about reconstruction techniques after the spinal tumor resection, mainly metastatic tumors. But I put a couple of primary tumor slides in there as well. So, this is looking at our experience over the last, what is it, 23 years or so, treating patients with spinal neoplasms. And you can see that the majority of the patients that we treated harbored a metastatic spinal tumor, although 1 in 10 patients, on average, had a primary spinal column neoplasm. Now, the role of surgery had been in question for a long time until this study, which is about 15 years old, where this study looked at whether or not the surgical decompression of reconstruction, followed by radiation therapy, in this instance being conventional radiation therapy, is better than the radiation therapy alone, and clearly demonstrated that the surgery is definitely beneficial when the patients present with epidural spinal cord compression in the setting of metastatic disease. This was true as far as their ambulatory status is concerned, bowel-bladder status is concerned. Even the survival was better in patients who underwent surgery. So, surgery definitely has a significant role in treating patients with metastatic disease, particularly when there's a destruction lesion in the spine. How did we decide how we construct the spinal column after a tumor resection? This is really a multifactorial process. There are some elements that come with the patient, so whether or not the patient has, for example, preexisting spinal deformity prior to the surgical procedure, whether the tumor is lytic or plastic in nature, all of these things need to be taken into consideration. The region of the spine that's involved is very important, the cervical spine in terms of its mobility, and the loads it's subjected to are different than the lumbar spine or the thoracic spine, for example, which is quite well supported with the ribs. And then we have to look at what happens with the surgical intervention. Although the patient may not have spinal instability prior to the surgical intervention, when you take bony elements, joints and ribs to decompress the spinal cord, you may render the patient unstable, and hence the patient may require reconstruction and stabilization afterwards. Furthermore, we have to factor in how long the patient is going to be around. A patient with metastatic disease from lung cancer may not live very long. A patient with metastatic disease from a follicular thyroid cancer is likely to be around for many years to come. So that needs to be taken into account as well. And then we also need to think about what the patient is going to be subjected to after the tumor is resected, whether or not the patient is going to get radiation therapy, chemotherapy, which can create quite a hostile environment for bone healing, bone grafting, and so forth. And that needs to be also thought about prior to the surgical reconstruction. And biomechanical load that the spine is seeing in a specific region is important. A lumbar spine, for example, is subjected to very high loads as compared to the cervical spine, which only carries the head around. So it's a very different biomechanics and the loads that they need to resist. This question always comes up, whether or not the spine is stable prior to the surgical intervention. The stability of the spine has been well studied in the trauma setting. But when you're dealing with metastatic tumors that are really replacing the vertebral body or destroying the spine, the situation is very different. In the trauma setting, there's a very high force impacting the spine, which leads to disruption of the ligaments and joints and discs and whatnot, whereas in the setting of metastatic disease, the vertebral body progressively collapses. And typically, the other elements are not really affected from that. But how do we really determine whether or not the spine is stable or unstable in a given setting? The spinal neoplastic scale was developed by the Spinal Oncologist Study Group, which I had the pleasure of chairing for a number of years. And we did a lot of work trying to determine if we can come up with a formula to see if the spine is stable or unstable. As you can see, not so much the scores that are important here, but the factors that go into it. For example, the location of the tumor, whether or not the patient has mechanical pain, whether or not the lesion is lytic or osteoplastic, whether the spinal alignment is compromised as a result of the metastatic tumor involvement, how much of the vertebral body collapsed, less than 50%, more than 50%, and whether or not the posterior elements are involved. All of these things play a role. And the more of these you accumulate, if you have a lesion in the mobile spine, vertebral body collapse, kyphosis, mechanical back pain, progressively the numbers go up. The higher the number is, the more unstable the spine is. In some situations, and also we talked about the fact that the extent of resection can render the spine unstable, although the patient may not have started out with an unstable spine for the resection of the tumor. And these are a variety of methods that we use, depending upon the location of the tumor in the spine, whether or not we can just remove just simply the vertebral body. We sometimes take the pedicle out. Sometimes you have to do bilelical approaches. Sometimes you have to do an extensile paraspinal tumor resection with the rib resection and so forth. All of these things progressively destabilize the spine further. So the surgery that you do to decompress the spinal cord may render the spine highly unstable and hence the patient may require spinal stabilization. So let's look at some of the case examples to bring some of these points home. Here's a patient with an image of a lumbar spine. This patient has metastatic breast cancer presented to the emergence of acute onset back pain and cotic chino syndrome. And you can see the collapsed vertebral body here. I mean, you don't need to be a neurosurgeon to see whether or not the spine is unstable or stable. It's clearly an unstable situation. And also, you can see how it looks like on the MRI scan where we have significant spinal cord compression and cotic chino compression there. And hence, this really explains the patient's clinical picture. What do we do here? So we typically have to do something to resect the tumor. We have to reconstruct the anterior spinal column. And given some elements here, very significant destruction, severe kyphosis, we will probably need to supplement that from posterior. Here is the view after the retroperitoneal approach and resection of the vertebral body here. And there are a variety of ways one can reconstruct the anterior spinal column. Cheapest one is methametacrolate. And we do that by using what we call a chest tube technique. And this is a dental drill with an eggshell drill bit. You can make holes in the vertebral bodies above and below, essentially. And take a chest tube and cut it longer. And then we typically make some air escape holes at the end and a hole in the middle of the chest tube. And this is really embedded into vertebral bodies above and below and span the space. And then you can take the methametacrolate and inject it when the methametacrolate is semi-liquid after it's mixed. And you can see the cement that travels into the vertebral bodies above and below. And then you fill it and let it solidify. And you can get a nice block of cement that replaces that vertebral body very nicely secured to the levels above and below. We typically couple that with a plate fixation because that eliminates the motion across those three segments. When you eliminate the motion, typically you do not get loosening of the construct. And so this then proves to be a very, very durable way of reconstructing the vertebral body. Of course, you can also reconstruct the spine with a metallic device as seen here. It's a cage, a distractible cage. Advantage of that is that you can fill it with bone. And when you fill it with bone, in the long term, there's a potential of that area to form a bone fusion across that segment. We again do the same. Again, we typically couple that with a plate fixation to make sure that motion is eliminated across those three segments to minimize the risk of failure down the line. This patient underwent a two-stage surgical procedure. First, through a retroperitoneal approach, interior column reconstruction, resection of the vertebral body and epidural tumor. And cage reconstruction followed by a fixation posteriorly. And the reason for that is this is a hypermobile portion of the spine. The patient had already preexisting kyphosis here. And hence, that's how we elected to do this. The patient then received postoperative radiation therapy to this area. So if you put just a bone graft in that area, like a femur strut or something of that nature, to reconstitute the anterior column, and then give radiation therapy after that, that may not heal properly. And you can get an anterior column failure later. And hence, some sort of a synthetic reconstruction of that area is important. We either use methametacrylate or cage for that purpose to avoid that kind of a problem. Here's another example where this patient has a upper thoracic lesion involving the T3 vertebral body. This is a 76-year-old, very fragile old lady who has a lung primary, which was treated with radiation therapy locally. Some chemotherapy was attempted, but she couldn't really tolerate that. And then she presents with bilateral lower extremity weakness and interscapular mechanical back pain, quite severe. And it's not really entirely clear if this is stable or unstable. And you can really go through the exercise to see. We can come up with a scoring methodology here. Here's the lesions in the rigid part of the spine. Pain is mechanical. The lesion is lytic. There is some minimal kyposis in that area, where the body is less than 50% collapsed. And there is, you don't appreciate that, but there's also bilateral dorsal involvement here. And we come up with a score of 14, which is pretty high. And that really tells you that from a clinical perspective, this really has, the patient has an unstable spine. Although it sits in the middle of the thoracic region. And this is treated in this case with using a minimal immersive approach with bilateral wheel seat approaches and tubular retractors. And you can see, we can look from one side and see from the other side and get a circumferential decompression of the spinal cord. We were able to tie the nerve root. And then place a small cage that's filled with the bone anteriorly. And then supplemented that with the percutaneous screws under fluoroscopy guidance. And you can see some of the stab wounds here. That allowed us to effectively treat this patient and reconstruct the anterior column. And this patient was discharged several days later, very happy with the results. Eight months later, she died, unfortunately, from progression of her disease. But you could see, otherwise, we would have had this long incision in the back, which would have taken a long time to recover. And not only that, that area had already been subjected to the radiation therapy from the chest radiation before. And that could have been a problem as far as the wound healing is concerned. Let's look at this case, where this patient has a metastatic hepatocellular cancer and presents with severe mechanical thoracic pain. Is it stable? Is it unstable? Looks pretty stable in the middle of the thoracic spine. Not that different than the other patient. But we go through the scoring system again. And you can see that this is the rigid area, a relatively rigid area. And the patient has mechanical pain. We admit that. The patient's lesion appears to be plastic, although it collapsed a little bit. The patient has normal alignment. The vertebral body is less than 50% collapsed. And there's no posterior element involvement. So the score tends to be a little less here, over a scale of 15, just about somewhere in the middle. So the answer is potentially unstable. We reconstruct what we do, an open operation, resect the vertebral bodies, or anything else to do. And we typically reconstruct the vertebra with injection of a cement, as you can see here. A vertebroplasty procedure, put two needles on both sides and inject the cement. And this patient gets traumatic pain relief. There's no epidural core compression. And the spinal column is stabilized, but it's stabilized by using a percutaneous mean to eliminate this instability-related mechanical pain. Is it successful? If you look at the literature, this is the data compiled from the literature, almost 90% improvement of pain. Very, very dramatic response. And the same is true if you use another method, kyphoplasty, rather than a vertebroplasty. Again, you get almost 90-plus percent improvement of pain. So it's a highly effective procedure in reconstituting the stability of the vertebra in a situation where there's a simple pathological compression fracture, and that really works very well. This was really, this really changed the approach entirely for patients with multiple myeloma, because multiple myeloma patients are not really good for surgical intervention for the most part. This is a patient with metastatic breast cancer involving the upper cervical spine, C2. This area has unique biomechanical characteristics. And hence, we approach this very differently. The vertebral body is, or the C2 body is involved anteriorly. But the main issue here is really the deformity of the cranial cervical junction. The patient has severe mechanical neck pain. She's holding her head up in the emergency room to mitigate the pain, and has no neurological deficit, although the MRI scan looks pretty concerning. You can see that there's significant spinal cord compression. But the main issue is really the deformity of the upper cranial cervical junction and instability of that area. And we typically reconstruct this by taking advantage of the modern instrumentation, occipital cervical fixation technique. This is a method that we use for patients with bathroom vagination. It allows us to disengage the dents from the, from a magnum and then realign the spine. And how do we do that? We put a plate in the occiput, and a bunch of lateral vasculars here. And then you attach the rods and apply distraction across that area. And this allows the dents to be pulled down, essentially, out of the form of a magnum. And then we go ahead and translate the cervical spine forward with respect to the skull base. The patient's head is fixed here in a Mayfield head fixator by taking advantage of the rod here. If you apply compression here, for example, these nuts are locked, this is loosened. And then this pushes the cervical spine forward. And then you can realign the spine. And you can see that two-step process. First, distraction across this area to disengage the dents out of the form of a magnum. And then translate the cervical spine forward with respect to the skull and realign the spinal column. And that's exactly what happens here. You can see on the left where we were. And that's where we are. We did multilevel laminectomy here. Not so much for decompression, because as soon as we realigned the spinal column, the decompression has been achieved. But more so, so that we could use the interoperable ultrasound to monitor what we were doing in the OR when we were doing all these maneuvers to realign the spinal column. We didn't touch the tumor ventrally. From a surgical perspective, we just went ahead and treated that patient with cirrhotectomy or surgery after that. Does that really work? And here's our experience many years ago now. Looking at 19 patients, all of them had metastatic disease involving the C1, C2 area. All of the patients were treated with stabilization. And none of the patients had decompression. They were treated with additional radiation or chemotherapy depending upon their primary. And within the lifespan that they had after the presentation, none of the patients had neurological deficit or neurological progression, indicating that this kind of combined strategy approaching this ultra and posteriorly with oxytocin fixation or C1C2 fixation in some cases coupled with adjuvant treatment with chemo or radiation therapy can be a very effective strategy in dealing with these patients. And if you look at the pain scores, the patients started somewhere around 8 visual analog pain scores. And at last follow-up, it was down to 1.5. And so the patients did very well from a palliative perspective. We occasionally do more extensive surgeries for patients with metastatic disease. This is our experience looking at blocker sections of the lesions in the spinal column. As you can see, there's almost 160, 70 patients here. Only 14 patients had metastasis. But we do occasionally do a blocker section of a tumor in the spine when they present with a solitary metastasis with no evidence of systemic disease with the intent to potentially cure that patient or get long-term survival. Does that really work? And here's the experience from Japan looking at, I think, 10 consecutive patients that had more than 10 years survival. So these patients all had solitary metastasis. They had a blocker section of the tumor. And they lived more than 10 years after the original surgical procedure. And if you look at what kind of primary they had, the majority of the patients had kidney primaries, as you can see, and then occasional thyroid. And I think there's a breast and a prostate patient there. So there is a subset of patients. When the patient is young, no evidence of systemic disease. And kidney is taken out many, many years ago sometimes. And the patient presents with a solitary metastasis. In those cases, you can go after them with intent to potentially cure the patient, although it's rare. For primary tumors, we typically do a blocker section. And this is a survey that we did a number of years ago. Sent all the experts around the world and gave specific case scenarios and asked them how they would reconstruct the spinal column. If you had one-level disease and you had a one-level spondylectomy, typically what would you do? Would you put a cage? Would you do a one-level fixation, pulmonal fixation, and so forth? And then some interesting trends have emerged from that. And you can see here, for example, if you did a one-level spondylectomy, majority of the patients around the world did at least two levels of fixation above or two levels of fixation below. Majority of the patients said that if you had an anterior column defect that was more than two levels, potentially using some sort of vascularized grafting strategy makes sense. And almost all of the patients had 360 fixation here. Interesting patient here. This particular patient had a tumor that was ventrally situated. And I don't have the, I couldn't find the axial image in this case, but the spinal cord was completely drained around this ventral mass. And I was convinced that this was some sort of an unusual tumor, maybe a very unusual mini-germ or something because of the way that it looked on the MRI scan. And we ended up doing a two-level anterior surgery. I was concerned about manipulating the spinal cord and so forth. And then opened it up from the front. And then we resected the thing. And it turned out to be a mixed-pupular ependymoma. And here's how it was reconstructed. We put a fibular strut graft here with a plate fixation from the side. I mean, operation turned out to be incredibly simple after we did all that exposure, got in there, opened the door up. And then sure enough, it was a well-circumscribed tumor. With a final terminal, it had just folded behind the spinal cord, the corners. And then the tumor was somehow tucked in ventrally in front of the spinal cord. This is a patient with metastatic renal carcinoma. And I said in rare situations, we would do an enveloper section. And here is the way that we do it. You approach it from posteriorly with a midline approach. We typically do a line of anectomy above and below. And you can really cut the pedicles on both sides and get the dorsal elements out. That's how it looks like in the OR. The head of the patient is here. The feet are down here. Here are the dorsal elements, the line of anectomy above and below. We have some hardware in there already. Head is in this direction. Feet are in this direction. And then we go ahead and cut the pedicles on both sides with an osteotome. And you can see the posterior elements are taken. And then we go ahead and establish a plain ventral to the spinal column, which we do that by just taking the ribs above and below the air level of interest out. Basically, do a costal transversectomy approach. Find the intercostal vessel and follow it. Dorsal to that. And this allows us to be dorsal to the great vessels ventrally. And you can see, and we can put a, you can see a malleable retractor. The retractor goes in front of the spinal column behind the aorta and IVC. And here is a silastic sheet that's underneath the fecal sac. And we cut the nerves on both sides. And then all that's left, really, the discs above and below. If we cut those, then we should be able to deliver the specimen. You can see we are pulling it out now. And here is the whole segment is taken out. And the thing is that Japanese experience has shown that if you take the tumor out like that, even if it's a metastatic tumor, it just doesn't come back. It does not, your local recurrence rate is almost 0%. With no radiation in Japan. We always give radiation therapy postoperatively. But with no radiation in Japan, 0% local recurrence rate. So it can be very successful in controlling the disease locally, even in patients with metastatic tumors. And this defect is then filled with a cage. As you can see, we have a distractible cage deployed here. And then reconstruct the spinal column. And as just most experts agreed on, this was reconstructed with two levels of fixation above and two levels of fixation below in this particular situation. And I have two primary tumor examples I wanted to put in here. These tend to be more complicated. And hence the reconstruction tends to be a little bit more challenging. Here's a patient with a chordoma involving the lumbar spine. And we believe that the best way to treat this is to take it out in an unblocked fashion with negative margins. And we can do that in two stages. First come from posteriorly. Multi-level laminectomy. Everything is taken down all the way down to the pedicles. And we dissect these lumbar roots very extensively down into the retropenial space. And then we do have this construct as illustrated by the artist. This is the thoracolumbar junction. We have very destabilizing operation. We have two rods that are going from the T10 vertebrae all the way down the pelvis, multiple pelvic screws. And here's the patient. The head of the patient is here. The feet are down here. It's a very load-bearing area. So two rods on both sides. And here are the tickle sac. Here are the nerve roots. We have a bunch of wires put in here, which we're going to use to secure the cage that we're going to put from the front. And these are the, I don't know if you can appreciate, there's a little saw here. It's called a thread wire saw, which we use to cut the spinal column. And this is done during the second stage of the operation where we come from the front and then retrieve the ends of the thread wire saw that we tucked in on both sides during the first stage of the procedure, and then cut the spinal column above and below. And then this is typically filled with a cage. The cage needs to fit perfectly here. And then the wires that we had tucked in from the back, we circled them around the cage to make sure that the cage doesn't kick out and the specimen comes out in one piece, as you can see here. Here's how it looks like after it's really resected. And this is how it looks like in the OR after this area is ventrally reconstructed. This patient is in this direction. The head is up here. The feet are down here. This is the S1 end plate. This is the bottom of the cage. This is the aortic bifurcation. IVC is here. You can see the cage in the depth of the incision. And this is the wire that we encircled around the cage. You can appreciate that a little better on the plain x-ray here. Here's the posterior construct quadriple rods or quad rods. We have a cage filling in this defect. And then we actually have a screw placed here to make sure that the top of the cage does not kick out. But we also secured it with the wires here. And the last case example is showing another way of reconstruction. This is a patient with epithelial sarcoma, a young boy from Israel presented with quadriparosis to an emergency room over there. And a laminectomy was done and a biopsy was carried out. And this turned out to be an epithelial sarcoma. And the patient was sent to radiation oncology. Radiation oncologist said that I can't do anything about this. The tumor needs to be at least devolved or removed. And then the patient's mother somehow ended up at Hopkins. But I was there and showed up in the emergency room with the patient from Tel Aviv. And so this is how it looks like. The tumor was encircling the vertebral artery on one side in the spinal canal. It's an intradural, extradural. Very hopeless looking situation. I said, I'm not sure really I can help with this. The mother of the patient said, what would happen if you didn't do anything? I said, I suspect he's going to progress to quadriplegia. And I'm afraid he's going to die from this. And she said, try whatever you can. And so we designed a surgical procedure here which really had several different parts to it. We were planning to come from the back and mobilize everything, address the tumor that's intradural. And then come from the front during a second stage operation to a transmandibular approach and resect the tumor. But we were able to resect this tumor all from posteriorly to my surprise here. And so we have a hockey stick incision here, posterior midline, multilevel laminectomies. And then we open the door and find a tumor intradurally. And that's what the tumor looked like in the OR. This is the artist's illustration. Goes through the nerve root sleeve and then encircles the vertebral artery like this. And so we were able to basically unfold this up and then peel it off the vertebral artery on that side as the artist has shown here. So the key to this operation actually turned out to be our ability to be able to cut the pedicle on the opposite side using a bone scalpel. This allowed us to disconnect the opposite side. On this side we took the vertebral artery. We cut the connections above and below. And the plane actually, and you can see we resected the dura along with the tumor here. And then completed the cuts above and below. The plane between the esophagus and the, between the pharynx and the neurovascular structures in the front was relatively nice. And so we were able to peel that off. And the whole thing came out in one piece. You can see here. Here's the part of the tumor that was intradural encircling the spinal cord. Here's the edge of the dura that we cut and left with the specimen. And this is how the CT scan looked post-op. A large defect here, right? There's nothing anteriorly. And then we have to reconstruct this. And so the way that we reconstructed this. So this area is biomechanically different. The weight transmission occurs from skull base through the occipital condyles down to the cervical lateral masses. And then through the pedicles of the cervical vertebrae to the anterior column. So the cage reconstruction here. We don't put a cage in the front. We put a cage from the occipital condyle down to the lateral mass. Which is really secure to the posterior construct. Which then through the pedicle scurs here in the thoracic spine transmits the load anteriorly. And this is how it looks like in the OR. The head of the patient's on the right side. The feet are down here. Here's the multi-level laminectomy area. You can see that we had to patch three durals here. Three dural patches. One is here. One is there. And one is more ventral. Here's the cage on the right side. Which is really secure to the rod posteriorly. The occipital plate. The patient's post-op MRI scan. We ended up with a large pseudomeningocel. But it was not an issue. It was not really communicating with the mouth. And then eventually resolved. The patient was asymptomatic from this. And here's the patient's reconstruction. Again, lateral mass cage that goes from the occipital condyle down to the cervical lateral mass. Connected with multiple scurrs to the rod. Which then connects to the thoracic spine with the thoracic scurrs. And that transmits the weight down there. And so there are a variety of techniques that we use in reconstructing the spinal column. So in many cases, in patients with metastatic disease, if you have a single-level corpectomy, you have several different options. You can replace that with a cage. You can put a polymethyl methacrylate, as I've shown. And one can use a strut graft with a plate. The thing to keep in mind is that since the patient is going to get typically post-op radiation therapy, if you are relying a bone healing anteriorly to reconstitute the anterior column, you may end up with a failure down the line. So putting something synthetic in there, it makes sense. If you're dealing with a junctional zone, either the thoracic junction, thoracolumbar junction, craniocervical junction is being an exception. If you're dealing with all junctional zones, typically doing a 360 operation makes sense. If you're starting with a severe deformity, the patient is already kyphotic, for example, it's usually not a good idea just doing an anterior operation and reconstructing the area. And then relying on whatever you put in the front to keep the patient together. Because typically when the patient is severe kyphosis, all the dorsal tension then is loosened and doesn't really provide any support. If the patient is going to be around for a long time, particularly true for primary tumors, osteogenic sarcomas, chordomas, chondrosarcomas, the patient were to survive, you need a fusion in order to achieve that vascularized strut graft can sometimes work very well. And then the other thing we learn hard way is that using double rod constructs posteriorly is very important because if the bone does not heal, then you don't get rod fractures down the line. You need to think about the fact that the patients would require post-op treatment, and hence PMMA and CAGE should be incorporated into your construct. And finally, if you are dealing with an area of the spine that sees significant mechanical load, such as lumbar spine, usually it's a good idea not only to use double rods, but also typically do a 360 reconstruction and fusion. Thank you. Any questions? Yes? In the last case, what did you do with the right vertebrae? Right vertebrae? I took it. I took it with a tube. Did you have a pre-op to actually make sure that you were okay? Yeah, so that's a very good question. I stayed away from that for a number of reasons. If you were to do a pre-operative angiogram, you could do the test occlusion. I'm good with that. But the pre-op angiogram and then take the vertebral artery on the right side preemptively, and then you go to the OR, and you're dissecting the vertebral artery, mobilizing everything, and something happens to the good side. Not good. The patient's dead. And so for that reason, I typically do not do a pre-op balloon embolization or coil embolization of the vertebral artery. I expose everything. I keep both vertebral arteries open. And before I take the vertebral artery, I put usually an aneurysm clip both proximally and distally, and run the MEPs and SSCPs. And I let them run for about half an hour or so. And if there are no changes, I go ahead and take the vertebral artery. And that really is the best occlusion test because when you do a balloon occlusion test, you typically float a balloon, and you put it somewhere in the proximal portion of the vertebral artery. And most of the time, the collateral circulation from above and other places can still fill and give you the sense, false sense of security that really the patient can tolerate it and may not be able to do that. Thank you.

reconstruction techniques

spinal tumors

metastatic tumors

surgery

epidural spinal cord compression

surgical decompression

radiation therapy