Okay, hi. Welcome to the 2021 AANS Virtual Annual Meeting. This is Charlie Sansour from the University of Maryland. Both myself and Dr. Mike Groff from the Brigham will be moderating the spine section portion of this session. So this is the section on disorders of spine and peripheral nerves. I want to let you know that this is an interactive event and so we will be able to do question and answers using the question and answer box on the right side of your screen. These questions will go directly to the faculty. If you'd like to engage with your colleagues, you can use the chat box on the right side of your screen. Another thing, it's important to note that you can self-report your CME credits by logging into myAANS. Because this is an interactive session, there is the opportunity for polling and there will be some presentations that involve polling questions and to participate, you can click on the question and answer box on the right side of your screen. So that's just some nuts and bolts for the introduction. And so we can go ahead and move on to our first session, which is entitled Follow the Patient. So, these first couple sessions, they're going to be nine-minute sessions and they're titled Follow the Patient. So, we're just going to go through specific cases involving spinal cord injury, acute spinal cord injury. I myself will be faculty as well as Dr. David Okonkwo from the University of Pittsburgh. And so, we're going to present a couple cases and these are my disclosures. They don't really influence these cases. So, the overall treatment goals in spinal cord injury involves both neurological protection to enhance recovery and reduce the likelihood of deterioration. Every case that we do involves the potential for the need to realign and stabilize and eventually rehabilitate. And so, we all share these treatment goals to immobilize, gain stability, obtain realignment if necessary. There's always a question about how urgent of a surgery is this. And so, during today's cases, we'll describe all of these attributes to spinal cord injury and look at specifics and reasoning behind them. So, we'll start out with this first case. This is a 17-year-old male, status post-ski accident. The motor exam on this patient is 5 out of 5 in the biceps and then he starts to lose function below the level of the biceps. So, he has 3 out of 5 in the triceps and 0 out of 5 distal to the triceps and the rectal tone is absent. So, depicted here, you can appreciate this patient does indeed have a spinal cord injury. There's increased intermediary lesion length signal as evidenced on this MRI. At the University of Maryland Shock Trauma Center, we have relatively easy access to getting MRIs and this patient actually had this MRI immediately upon his arrival. Depicted here, you can see that there is a significant translation of the vertebrae and over here on this CT scan, this is actually a CTA. You can appreciate this is a bilateral jump facet with one facet being completely jumped, the other facet being jumped, but just not as thoroughly jumped. So, this results in approximately a grade 2 spondylolisthesis at this level. One side is semi-perched-slash-jumped, the other side is jumped. So I'm going to move on to demonstrate that the CTA shows that the vertebral artery anatomy is indeed intact. This is a poll, and so the question is, who would classify this patient as a complete spinal cord injury? It's a tricky situation here, and one of the reasons why this may be considered complete versus incomplete is because there is a piece of missing information. This patient, importantly, what I'm going to show you is that when I first got this history, that was the history that I got from my team. The question that they didn't have an answer to at the time when they first presented this case was, what about the sensation of the perineum? Did this patient have any perianal sensation? And when carefully examined, this patient did indeed have some perianal sensation, and the very presence of that perianal sensation is really what constituted the urgency of the treatment. So we can move on to polling other questions. Is this somebody in whom you would consider immediate traction? and I guess maybe the polling isn't working. So in this patient, you could certainly consider immediate traction. Knowing that the patient had, that this was a patient who had preservation of some of the perineal sensation, this changed the case into a very urgent case and essentially turned it into an incomplete spinal cord injury. And so the decision at that point is, do you take the patient immediately to surgery and do traction while under general anesthesia in the operating room, or do you try to get traction done immediately prior to the operating room? And in my opinion, the answer to that question purely is based on OR availability. If the OR is immediately available, then take the patient straight to the operating room and attempt reduction prior to the incision when the patient is relaxed and intubated. And that's basically how I manage these cases with regard to the decision of what surgery to perform. I'm not sure if the polling can work, but if the polling is able to start, the decision is, is this an anterior case where you start out anteriorly? Is this a case where you would prefer to start anteriorly? Is this a case where you would prefer a posterior decompression and fusion with a posterior reduction? Or is this a posterior, anterior, posterior? So very often this is the dilemma in this kind of case. And in a bilateral junk facets, with the appropriate management of the use of traction, as well as the use of interoperative reduction techniques, very often I make an attempt to go anterior and try to perform the reduction anteriorly. And so depicted here is the result. This is a patient in whom we were able to perform an anterior reduction by using bilateral cloud spreaders. This provide enough leverage to be able to distract these segments together, apart from each other and to obtain a reduction. And this patient, as you can see, we got close to bicortical fixation. This was a young, healthy 17 year old patient. This patient had good quality bone and had very solid fixation. And this structural allograft in place allowed for a solid fusion. So follow-up of this patient was really quite remarkable in the sense that he, at six months out from surgery, was able to walk back into clinic. And about one and a half to two years out from surgery, this patient actually is almost entirely normal. So just goes to show you the importance of having a very thorough neurological examination. This patient had immediate intervention after he presented to the University of Maryland. And the distinction between something that might've been delayed to the following day versus immediate was the presence of that perineal sensation. And I think having that good, thorough examination is what prompted us to do this in the middle of the night. And the result was fabulous. So this patient did very well. So that's an example of a patient with a spinal cord injury who, you know, it's quite possible that you might have more of a sure thing by going in posterior first in order to make sure that you get the complete reduction. But I have succeeded in doing interior-only reductions the vast majority of the time. So I think we'll move on to our next speaker, who's gonna be Dr. David Okonkwo. Dr. David Okonkwo. Charlie, it's good to see you virtually. Thank you for that really interesting lecture and case presentation. And thank you for the honor of contributing to this session. I am going to talk about a patient with a thoracic spinal cord injury. And I'm gonna present the case of the last spinal cord injury to occur in the National Football League. And I'm also part of the medical staff for the Pittsburgh Steelers. And I was the person there on the sideline when he was injured and have been a part of his journey following this patient from really the very moment of injury. And we've just published our description of this case in internal neurosurgery case lessons just two weeks ago. So what happened was that in a head first tackle, crown of the helmet, and with the head in a slight amount of forward flexion, the contact to the torso of the opposing player, followed by the immediate loss of all sensory motor function in the lower extremities. And so we drill for this every year in the National Football League as part of the medical staffs across the league, you drill for this. And it's very clear what everybody's role and responsibilities are, loaded onto a backboard and transported to the nearest trauma center, where I was very fortunate to have Joe Chang from the University of Cincinnati be a part of getting this off on the right foot in making sure all the correct things happen for this patient. You may ask, why did a head first tackle lead to loss of movement in the lower extremities, loss of all sensation in the lower extremities, but not in the upper extremities? We'll get to that in just a second. But we had some key management decisions to make. And these are six of the buckets that, this is not an exhaustive list of what we needed to do, but I'll quickly cover each one of these six things. And you should ask yourself if you have a protocol in place to guide this process, not just for the injured athlete, but for anyone who crosses the threshold of your center with an acute traumatic spinal cord injury. So in this artist's illustration, you'll see that what was actually the culprit in this is that the injury site was actually at T9. And there was a fracture and then an associated hematoma inside of the ligamentum flavum and in the epidural space that caused spinal cord compression. And so we had a fracture, relatively subtle fracture in the pedicles and facet joint, followed by progressive development of a neural element compression. And the reason for that, and this is the first known example of this that any of us can identify, is that this particular athlete had an underlying idiopathic scoliosis. And that scoliosis was known, but it was assumed throughout his life that it was a non-issue for the game of football. But it turns out that when you understand the anatomy of idiopathic scoliosis, there is an inherent biomechanical weak spot along the concavity at the apex of the thoracic curve. The pedicles there are hypoplastic and the spinal cord itself is eccentric in the canal closer to the pedicle of the concavity. And so this crown first hit the exit site of the weakest point in the spine was the concavity of his thoracic curve. And that's why the injury was at T9. And he was a T9 AJA on the field. The surgery then ended up being a transpedicular decompression with evacuation of the hematoma and a short segment stabilization. So then we had to ask ourselves, how are we gonna give this person steroids? And I know that this has become an incredibly controversial subject, but the fact of the matter is that if you understand the data from the original NASCS trials, and this was in the days before we used the term AJA, the spinal cord injury data from the NASCS trials clearly demonstrate that in younger patients, so this is a heavily disproportionately young male data set, and this is the patient population. And in this situation, we're talking about a world-class athlete with 0.3% body fat. The risk of a medical complication from steroid administration is incredibly low in that situation, but there is an incremental benefit for segmental improvement in this patient population with steroids. And so for us, this remains an inherent standard part of the protocol that we have in place for treatment of the injured athlete. Young, world-class athlete is going to get steroids in our care. And then I have paid a lot of attention to the data from Mike Wang and Alan Levy and the folks in Miami who have done a great job really leading the effort to deliver objective evidence about the role and potential impact of hypothermia in this patient population. This has not been able to be proven yet with a randomized prospective trial, but there is a clear theoretical physiological basis for the value of hypothermia in this patient population. And there is a mounting dose of clinical evidence that is not randomized that supports the use of hypothermia. And so in this instance, we administered chilled saline in the trauma bay. If the injury had happened in Pittsburgh, we actually carry chilled saline at the stadium. And if an athlete sustains a spinal cord injury, we administer chilled saline in the ambulance. But he received chilled saline in the trauma bay and then underwent hypothermia through the local protocol, which in Cincinnati was through surface cooling. In Pittsburgh, we would use intravascular cooling. The other thing that we've adopted is to pivot our attention from mean arterial pressure management and MAP goals over to spinal cord perfusion pressure management and SCPP goals because of the growing body of evidence that shows that if your SCPPs are greater than 65, you have a higher likelihood of improvement. Spinal cord perfusion pressure management is achieved by placing a lumbar drain and measuring intrathecal pressure. We can drain CSF if the intrathecal pressure is high. If the intrathecal pressure is normal, but the MAPs are low, then we drive MAPs. We don't drive MAPs to a MAP goal. We drive MAPs to a spinal cord perfusion pressure goal. And then we used adjunctive therapies and I have to credit Dr. Maroon for his input into this. Julian Bales has published some really nice data about the potential value of DHA or omega-3 fatty acids for the injured spinal cord or the injured central nervous system. We used omega-3 fatty acids in this case. There's also some very intriguing data that hyperbaric oxygen may create a favorable recovery milieu for the injured spinal cord. And Jason Huang, who is another valued member of the neurotrauma section, has published a really nice review about that. So we used both and this patient underwent 29 dives in the hyperbaric oxygen chamber over the first several weeks following his injury. And then we did intensive rehabilitation. And we were extraordinarily fortunate where there was no such thing as a resource limitation in this case. And so this particular patient underwent six weeks of intensive rehabilitation and followed by 130 outpatient therapy patients. You know, the typical spinal cord injury patient receives 20 and a shorter inpatient rehabilitation course. But in this case, it was six weeks of inpatient rehabilitation, 130 outpatient therapy sessions. And then in fact, when you trace out his motor recovery, what you find is that he went from an ASIA-A at presentation. He was an ASIA-A on post-injury day two. And by six months was getting very close. And by one year was actually had gone from an ASIA-A to an ASIA-E. And so, you know, I can't exactly ascribe any one particular intervention to the course of what we just, you know, followed and saw and transpired with this particular patient. But, you know, we certainly brought to bear everything that we could to maximize the outcome. And then this particular athlete has now converted this into his life's mission that he understands that there were certain therapies and certain options that were available to him that aren't available to the general population. So he has launched his foundation with a focus on being able to bring these options to anyone who sustains a spinal cord injury. And I'll stop there and happy to take any questions. Thank you. In this session, we're going to cover techniques and indications for cervical arthroplasty. I would like to thank my clinical research coordinators, Alma Ben-Natan and Jeremy Huang for their assistance with this talk. I do have some disclosures which are here. They are not related to arthroplasty. When we talk about arthroplasty and fusion, one of the issues which arises is adjacent segment degeneration. Some of the adjacent segment degeneration may happen because of fusion. Some may also happen because of natural history. The studies that have been done with arthroplasty demonstrate some of the issues related to adjacent segment degeneration and what you can expect after arthroplasty. This is a pretty famous article in JBJS back in 1999, quoting the level of adjacent segment degeneration requiring surgery at a roughly 3%. And perhaps that happens because there is abnormal motion at the levels above and below a fusion. Arthroplasty, however, may help by preventing abnormal biomechanical forces at the levels adjacent to the fusion. There are a good number of arthroplasties available now on the market. Here are some of them. When we talk about arthroplasty, we have to understand that they're not all exactly the same. Some are not articulating. Some have one articular surface or unarticulating. Some have two articular surfaces or are biarticulating. They have different types of composite materials which are included in them. Some are metal on metal. Some are ceramic on ceramic. Some are mixed with ceramic and metal or metal mixed with poly. And then the other thing that we consider, are they modular or non-modular? So here's examples of some unconstrained and semi-constrained designs for arthroplasty and several of the arthroplasties which are currently available. If we look at the first USFDA IDE trials in arthroplasty, and we want to see what the findings are, we can look at this paper from 2012. This paper analyzed all three of the USFDA trials and presented the data. So what we see is that there are fewer secondary surgeries with arthroplasty. And this happened across all three of the first three cervical USFDA trials. Also, if we talk about adjacent level disease, that significantly favored arthroplasty when we used a fixed effects model. That again happened with all three cervical arthroplasty trials. There were two cases in some of the early USFDA trials where there was a screw fracture and that needed revision. So it's not that arthroplasty is perfect, it also requires revisions. It just seems that those revision rates are lower than fusion. Arthroplasty does tend to maintain range of motion at many years after its implantation, whereas fusion basically has no motion at the level where fusion happened. So we want to maintain this kind of motion. You can see it maintained with an arthroplasty. When we look at specific USFDA trials and we look at their long-term follow-up results, and here's one example with the ProDisc-C seven-year follow-up, we see that there is decreased secondary surgical procedures with the arthroplasty compared to the fusion. And with the Moby-C trial, similarly we see decreased subsequent surgery rates in the arthroplasty group compared to the single and two-level fusion cohort. And this was again, follow-up at seven years. The Secure-C artificial disc versus ACDF also with roughly 380 patients with more than 80% follow-up at 84 months had fewer surgeries revisions in the arthroplasty group compared to the ACDF group 10 versus 22. In the PRESTIGE-LP study, there has been more than 10 years of post-op follow-up showing that the arthroplasty continues to move very nicely in terms of a segmental range of motion, whereas of course the fusion does not. And the arthroplasty was effective 10 years after implantation. This is another 10-year follow-up with the PRESTIGE-LP showing that re-operation rates were much lower for the arthroplasty compared to the ACDF cohort. And this is for two-level implanted arthroplasty. So let's look at some of the nuances of doing the operation itself. So this article, which was published in 2007, has an online video segment, which I'm going to edit and show you here. And this online video segment is a summary of the operations of the ARTHROPLASTY group. And this video segment is going to reveal some of the nuances of arthroplasty. So first, we do a discectomy. And during the discectomy, you can use caspar distraction pins. However, when you're sizing the arthroplasty, you want to take the distraction off because you don't want to overstuff the disc space with an arthroplasty because that will not allow the facets to move normally. Here we're sizing what size arthroplasty typically we'll be able to get in there. We have a small shim distractor here. Most of the time, I find that the arthroplasty size that I implant is six millimeters. Sometimes it is seven millimeters, but that's rare. Most of them are six millimeters. Some of them are five millimeters. You can see here, the caspar distraction pins have been removed and we are doing some cutting of the rails through this rail cutter guide in order to put channels into the bone above and below the disc space to put this arthroplasty device in. This particular arthroplasty device demonstrated in this video does not have screws to hold it in place. It is relying on little channels cut into the bone with little rails on the arthroplasty slid into those channels to keep the arthroplasty in place. So we're cutting the little channel guide for all the rails here. So that's the third rail going in there. And then we have one more rail cutter that's going to happen now. And basically what we're doing is using small little cuts in the end plate to line the arthroplasty in place. So now once all four little channels have been cut, we know where the arthroplasty needs to go. And it's important to make sure we centered this arthroplasty in the disc space. So checking out where the ankle vertebral joints are is a good way to make sure that you're centered. And if you need to do an AP x-ray to make sure it's centered that's also helpful. And now we have a rail cutter, which is going to go along the channels that I just started with the previous rail cutter guide. And then this is actually going to cut the channel into the bone. And this is the footprint of the arthroplasty that we're choosing. These arthroplasties do come in different lengths. I like to get the arthroplasty to line up at approximately 80 to 90% or even 95% of the depth of the material body so that can fill the entire disc space. Here's the artificial disc going in. As you can see, it's following the guide cut by the rail cutter. In order to slide into place, you can see the little teeth above and below on the arthroplasty. And those little teeth will help hold this in place and those teeth prevent the arthroplasty from backing out. And then the surface of the arthroplasty has a bony ingrowth material so that the end plate will grow into it and completely secure the arthroplasty in position. This particular arthroplasty has no screws in it. These arthroplasties do typically have an up and a down side, meaning that in this particular arthroplasty, there's a ball and socket and they do need to be aligned the correct way. It's not reversible. You can't put it upside down if it works. And once you have the arthroplasty in position, you can do x-rays APM lateral to make sure that it's centered and that it's occupying 80 to 90% of the depth of the disc space. And I also like to, at this point, scrub out and do a flexion extension as well. So in conclusion, cervical arthroplasty is an excellent option for one or two level cervical disc disease in adults who are below the age of 60, who have normal facets. It is not a great option for patients who are age over 60 because typically they will have a lot of facet arthritis or they will have osteopenia or osteoporosis, both of which are relative contraindications to placing an arthroplasty. If you put an arthroplasty in osteoporotic bone, what can happen is that the arthroplasty can subside into the vertebral end plate. If you do an arthroplasty in patients who have a lot of facet arthropathy, what can happen is the facet arthropathy may inhibit the artificial disc from moving normally. So it's good for younger patients who are adults who have one or two level cervical disc disease. That's the US FDA online indication for cervical arthroplasty in patients who have cervical disc disease that is not secondary to trauma. We don't wanna put these into traumatic patients who have had some kind of a ligamentous disruption because arthroplasty can slip out of position. It is also been found now with a follow-up up to 10 years that patients who have had an arthroplasty who have been involved in the US FDA trials have had fewer secondary surgeries compared to patients who have had a fusion. In addition to that, these patients seem to maintain the range of motion at the index level with their arthroplasty over many years, whereas of course a fusion patient does not maintain their index range of motion over time. So that is also an important point to take home. Some of the nuances that I talked about in terms of placement of the arthroplasty are important, centering the arthroplasty, also taking an arthroplasty that is not overstuffing the disc space height by creating too much distraction is important because it can splay the facets. So typical arthroplasty heights are usually for me six millimeters. And then the depth of the arthroplasty typically for me is 15 to 17 millimeters typically. That'll occupy most of the disc space on the adult patients who I operate on. But putting an arthroplasty that is too short and will not center the pivot point of the disc space in the arthroplasty, so that may give you abnormal range of motion. So having one that really fills the disc space front to back is centered from side to side so you don't have abnormal asymmetric wear on the arthroplasty is also important. So these are the nuances I would like to make sure that people who are doing arthroplasties know. These cases at one and two levels have been associated with relatively few complications. They are typical complications associated with ACDF that happens with arthroplasty as well, but they're at a very low rate. And these are relatively safe implants in my experience. And many patients are very happy with the results of their arthroplasty. So I thank you for your attention in this talk. And if you have any questions, feel free to contact me by email, which you can access through the AANS. And I hope to see you all again in person at a future meeting. Thank you. All right. My name is Zoe Gogowalla. It's a great pleasure to speak at the AANS meeting virtually. I'd like to thank the Scientific Program Committee for the opportunity to talk about lumbar spondylolisthesis with a specific focus on surgical decision-making. In terms of disclosures, I have no commercial conflicts of interest. I happen to be the lead author and PI of the SLIP study, which I will cover in this discussion and have independent foundation resources that have assisted me in getting some of this comparative effectiveness research done. I'd like to acknowledge a number of individuals, Ed Benzel, Jim D'Azurio, Dan Resnick, Fred Barker, Steve Glassman, Praveen Moominini, and Chris Schaffrey, who have all influenced and provided guidance and mentorship over some of the work on spondylolisthesis that I've been involved in. So we know that grade one degenerative spondylolisthesis is one of the most common conditions treated in neurosurgery. When associated with spinal stenosis, it is the most frequent indication for spinal surgery in patients over the age of 65 in the United States. It's estimated that 40% of people with degenerative spondy have a SLIP, and 20% of patients are reoperated on within five years, according to state inpatient databases. There've been two randomized control trials on the topic, both of which have been published in the New England Journal of Medicine in 2016 that reached different conclusions, and we'll talk about that in just a few minutes. As you all know, in the last month, there was a third randomized control trial from Norway, which we will also include in our discussion. By way of background, spinal fusion surgery is on the rise in the United States. We spend $86 billion a year on spine care with an estimated 500,000 lumbar spinal infusions done per year at a cost of $12.8 billion in hospital costs alone. Over a five-year period in the early 21st century, it was estimated that complex spinal fusion has increased substantially with increasing reports about major complications and costs. As you look at the mean hospital costs associated with fusion, it's estimated based on sport data to be 35,000. That compares very differently when compared with a decompression alone, which is $12,000. So when one takes a look at our current state of the treatment of lumbar spondylolisthesis in the United States, where from QOD data, we know that we fuse approximately 80% of these patients, one can calculate costs and find that we spent about $5 billion, the majority of which is spent on lumbar spinal fusion surgery when treating patients with grade one spondylolisthesis, if we assume 150,000 patients per year. If we look at a possible future state or a different state using numbers as they relate to treatment in Europe, where there's a very different focus on fusion and roughly a third of patients undergo fusion for this condition, there is a substantial amount of money, nearly $2 billion that can be saved if one took that approach in the United States. So naturally, there are many questions. People ask, what is the evidence for fusion in patients with grade one spondylolisthesis? Is decompression alone a better option for these patients? And ultimately, what is most cost-effective? And I think we all understand that we look at this not only as physicians, but in our society, we are increasingly cognizant of the payer perspective on this. And just to show you the numbers again, $5.3 billion for a model where we fuse the majority of patients, $1.8 billion where we do a decompression for these operations. If there was even a small shift towards decompression, it would save billions of dollars in the United States. And the question that we ask as spine surgeons is at what patient costs would that involve? So let's review the two randomized control trials that were published in 2016 to see if we can shed some greater understanding on the comparative effectiveness of lumbar fusion versus lumbar decompression and fusion. In the European trial, 247 patients with either one or two level spondylolisthesis with spinal stenosis were randomized at seven Swedish hospitals. The primary outcome measure at two years was the Oswestry Disability Index, looking for a difference of 12 points. And the secondary outcomes included the European Quality of Life EQ5D measure, as well as a health economic evaluation over a five-year period. 247 patients were randomized, roughly half to a fusion group and half to a decompression alone group. And roughly half of those patients in each cohort had degenerative spondylolisthesis. So again, the study population here was patients with both one and two level spondylolisthesis and spinal stenosis, of which 91 cases were single level spondylolisthesis. In terms of their comparative outcomes, as we all know from the publication, at two years, the ODI was not significantly different between these two groups. And there was no significant difference even when looking at the patients who had degenerative spondylolisthesis. When we looked at five years, where follow-up was much more limited, 56% follow-up, there was also no significant difference in the ODI scores between patients treated with decompression alone versus decompression and fusion. And so the authors concluded that there's no added value for adding a fusion when performing a decompression surgery for lumbar spinal stenosis, whether or not there's a degenerative spondylolisthesis associated with it. And so in the same issue of the New England Journal, the American study that I was involved in was also published. This was the SLIP study. And the SLIP study asked the same question, but really focused on single level degenerative spondylolisthesis with the hypothesis that fusion would improve outcome if laminectomy alone to treat these patients would destabilize the spine. And so the question asked by the SLIP randomized controlled trial was, is single level grade one degenerative spondylolisthesis unstable following a laminectomy alone? 64 patients were involved in this superiority trial from five clinical sites, patients were randomized, with sites selected for expertise, infrastructure and volume. The primary outcome measure in the SLIP study was different from the Swedish study. We use the SF36 physical component summary score looking for a difference of five points at two years. Our secondary outcomes at four years included the Oswestry Disability Index looking for a difference of 10 points, reoperation rate and cost. The inclusion criteria for the SLIP study were narrower than for the Swiss study. We included patients with a single level degenerative spondylolisthesis measured three to 14 millimeters with symptomatic lumbar spinal stenosis. And we excluded patients with gross instability, three millimeters on flexion and extension or a history of mechanical lower back pain. We also excluded patients with lumbar spinal fusion surgery and serious medical illness, ASA class four or higher. In terms of our design, this is also a randomized controlled study. 66 patients were randomized to either receive a lumbar laminectomy alone or a laminectomy with pedicle screw fixation and posterior lateral fusion. We have roughly half patients in the lumbar laminectomy alone cohort and half of the patients, 31 patients in the fusion cohort. No inner body devices were used in this trial and no bone morphogenetic proteins were used in this trial in an effort to create homogeneous treatment groups so that we can make more accurate comparisons. In terms of our execution, we had 130 patients screened, randomized 66 patients and had excellent follow-up at one year and at two years with greater than 80% follow-up at two years in both of the cohorts. The baseline characteristics were also comparable. Our outcomes showed in a similar way to what the Swedish study found, both laminectomy alone and laminectomy and fusion had substantial reduction in ODI and substantial improvement in SF36 physical component summary scores over time. However, different from what the Swiss trial showed, the physical component summary score as an outcome measure showed that the two-year, three-year and four-year outcomes for patients treated with fusion in addition to decompression were superior. In addition, at four years, the ODI difference between the patients treated with decompression and decompression with fusion showed an improvement after fusion when using the ODI at the four-year time point. What was most interesting with the SLIP study was the rate of re-operation where we found patients treated with laminectomy alone had a 33.5% rate of re-operation, whereas the patients treated with fusion in addition to laminectomy had a 14% rate of re-operation, which was a statistically significant difference. We had follow-up rates of 86% at two years and nearly 70% at four years. And so we feel that these estimates of re-operation are reasonably accurate over time. So how do we make sense of this? We have two randomized control trials, two studies that have been done with the best tools of medical evidence that we have available and reach opposite conclusions. So let's take a moment to take a look at some of these differences. First, the primary study population was different between the Swedish study and the SLIP study. The Swedish study had a heterogeneous population of patients with either one or two levels of stenosis with or without spondylolisthesis and didn't measure whether there was instability or not with flexion and extension x-rays. So we didn't know whether these patients had instability or not preoperatively. In contrast, the SLIP study looked at non-mobile grade one degenerative spondylolisthesis cases only. The primary outcome measure was also different between the trials. In the Swedish study, they use the ODI as their primary outcome measure. In the SLIP study, we use the SF36 physical component summary score as our primary outcome measure. The SLIP study was underpowered to look at differences in the ODI. The Swedish study had 91 single level cases. SLIP study had 66 single level cases. And at four years, we saw or observed a nine point difference favoring fusion in the use of the ODI instrument assuming an MCID of 10. We also, when looking at the percentage of patients who reached the MCID, saw 61% reach the MCID of 10 points with laminectomy alone versus 85% of patients reaching the MCID over the study period when treated with fusion in addition to lumbar laminectomy. If we use these numbers from the trials, the MCID of 10 for the ODI with a standard deviation of 18, we would calculate that we would need 140 patients with 90% power to use ODI to look at degenerative single level spondylolisthesis. So in a sense, both trials were underpowered for this particular primary outcome measure. In terms of the economics, the reoperation rates were also different between these study populations. The Swedish study had a 20% rate of reoperation in both cohorts and the SLIP study saw differences. I mentioned before with a 34% rate of reoperation for laminectomy alone and a 14% rate of reoperation for fusion. And we're in the process of collaborating with CMS for a cost analysis over a five-year time period with the hypothesis being that the reoperations associated with patients who have lumbar laminectomy would likely increase the cost of the overall treatment of lumbar laminectomy resulting in an economic analysis that we would have to study with the data but could show that getting the right operation done for the right patient the first time around may in fact be cost-effective when you factor in the cost of reoperations. So the other opportunity that we had from the SLIP study was to look at radiographic predictors of early failure. And we looked at these radiographic characteristics and identified motion at spondylolisthesis, disc height greater than 6.5 millimeters or a facet angle greater than 50 degrees as being risk factors for reoperation. We published this in JNS Spine in 2013, finding that if you had all three of these risk factors, there was a 75% rate of requiring a reoperation if you were treated with laminectomy alone over a four-year period. Another question that was raised by many of our European colleagues when we looked at our reoperation rate was was the reoperation for fusion indicated in the patients who were treated in the SLIP study where we saw a 33.5% rate of reoperation. One way that we addressed this was to look at the outcomes of these patients after they had reoperation. And what we found was that patients who required reoperation had a dramatic reduction in their ODI as you see on this graph and a dramatic improvement in their SF-36 physical component summary score after they had a reoperation. This was measured one year after their reoperation suggesting that in fact, these patients treated with laminectomy alone had developed instability at the index level of surgery that was improved by performing a fusion later on down the road. So how do we place these randomized control trials into the context of previous literature? Praveen Mummineni and Chris Schaffrey organized us to look at the data from both of the trials and to put that into the perspective of published guidelines. And so what we found is that if we took a look at Dan Resnick's publication from 2014, which showed that there was moderate support for the performance of lumbar spinal fusion for degenerative lumbar spondylolisthesis based on the literature to date, we felt that both studies added to this evidence base with the SLIP study providing level one evidence supporting fusion for a single level grade one spondylolisthesis and the Swedish study providing level two evidence to support decompression alone for a heterogeneous group of patients with lumbar stenosis with or without lumbar spondylolisthesis. So now just in the last month, we have another randomized control trial from Norway, which provides more data from which we can try to understand this problem a little bit better. This was a non-inferiority design randomized controlled study, which addressed some of the weaknesses of the previous studies. In the decompression alone, they use a midline sparing approach and in the decompression and fusion cohort, they used both MIS and open procedures with inner body being an option for patients treated with fusion. Single level spondylolisthesis was the inclusion criterion for patients in this trial. And so they avoided some of the weaknesses of the Swedish study focusing on single level disease, single level grade one spondylolisthesis with spinal stenosis. In their study, they looked at flexion and extension x-rays and observed that 20% of the patients that they included had motion on flexion and extension. Their primary outcome measure was a two year change in the ODI and they placed upfront in a pre-specified manner, a non-inferiority margin of 15 points. In terms of their results, they did not include, and this is important, they did not include re-operations in their analysis. They had 267 patients, so a large size trial. ODI reduction was comparable for decompression alone and decompression with fusion, 26 points for decompression, 21.3 points for decompression and fusion. And both groups saw a substantial reduction in ODI, 75% of patients had an ODI reduction of 30 points. So very good surgical results. Their re-operation rates showed a higher rate in the decompression alone cohort and a smaller rate in the decompression and fusion cohort, but not at the same level as what we observed in the SLIP study. So when we look at these studies now in context, one of the things we have to factor in is that the two trials done in Europe, both in Sweden and in Norway, may be different, and may represent different populations from the American population. For example, in this most recent Norwegian trial the BMI was 27, and is likely that most studies in the United States would show a BMI that is larger than 27. In the Norwegian trial also, the follow-up was at two years and did not include reops in the analysis. And this is important because if the SLIP study were analyzed without the reops in the analysis, this study also would have shown no difference between the two study populations. And in addition, an important point in spine surgery in general, but with degenerative spondylolisthesis in particular, looking beyond two years, looking at four years, five years, and longer. As Dr. Momine mentioned, in the cervical arthroplasty studies looking out at 10 years, it's going to be very important to understand the durability of the operations that we do. And so I think that it would be important to see what this Norwegian data looks like after four or five years. So where do we go from here? We have three randomized control trials, all published in a high profile New England Journal of Medicine venue. And we still need to study this problem in greater detail. And one of the things that we're working on is to see is there a way that we can classify patients with degenerative spondylolisthesis in a way that we can a priori know which patients might do well with a decompression alone, as opposed to a fusion. I think the vast majority of us who treat these patients in the United States, in particular, feel that the majority of these patients benefit from a fusion. But there may be some patients that don't need a fusion, and it would be interesting to identify those patients if we could. One way in which we might do this might be through a registry approach, and we have started an approach called SLIP2, which looks at fusion versus no fusion for grade one degenerative spondylolisthesis, with a goal of accumulating 1000 patients over a three year period with a complete radiographic capture with global expert review to develop a cognitive neural network model that we might use for machine learning. We have 15 sites engaged with the primary outcome measure being ODI and EQ5D with a pre-specified cost analysis over five years. At the present time we have 30 investigators, both United States, as well as international investigators involved in reviewing And the idea of doing this is to leverage global expert crowdsourcing to opine on each of these cases so that we can share these opinions with our patients. And so just to show you how this might work. I show you a case, a classic case from the SLIP study, where there's motion on flexion and extension that's less than three millimeters but motion nonetheless two millimeters of motion. You see in panel C, a grade one degenerative spondylolisthesis, and on the D panel, you see tight canal stenosis, particularly in the lateral recess at the L4 L5 level. And this was submitted to a panel of experts, 100% of surgeons felt that surgery would be appropriate, and 16 out of 16 patients who reviewed this case felt that they would, in their hands best treat this patient with a laminectomy and a fusion. In this case, when you take a look at a more heterogeneous type case a one that came from the Swedish study, a patient with two levels of spondylolisthesis motion at the L3 L4 level. And with stenosis at the L4 L5 level. There was a more nuanced type of opinion, where surgeons felt that surgery would be appropriate of conservative therapies have failed but there was a broad difference of opinion, with a number of surgeons feeling that in laminectomy alone in their hands might be an ideal approach for this type of patient. So it underscores the, the notion that not all of these cases are the same. And we need to apply some individual characteristics and understanding of these cases to give us the best answer in terms of fusion versus no fusion. We at this point have accumulated 500 patients in this slip to effort from 14 sites, and more recently we have added a component of this study looking at wearable technology and motion on the part of patients organized by Mike Wang, to, to get a better understanding of these patients outcome over time. So we're starting to get some data from this approach. As we've looked at the question of does expert panel opinion matter. Most of the time the panel is recommending fusion 75% of the time and in fact that's what we're seeing being done in this trial so far there's a dramatic reduction in ODI 20 points in both groups at one year. But interestingly when decompression is recommended by the panel patients appear to do better when they are treated with decompression alone, as opposed to being treated with fusion. And I show this here in a panel, where when the patient has the surgery as recommended by an expert panel. The outcomes appear to be better, which is somewhat intriguing finding. One of the things that we hypothesize is that when an expert panel overwhelmingly recommends a specific treatment, be it fusion, or be a decompression alone they're seeing some pattern that we think that we can analyze better with modern technologies. So, in summary, what did we learn, we learned that from these three randomized control trials that single level grade one spondylolisthesis is unstable, at least in the American population, following lumbar laminectomy, as we observed in the study one out of three patients developed instability within four years. If treated with laminectomy alone. We I think all recognize that all three of these randomized control trials, including registry data from the Q amp D, and now more recently from the slip to study show that patient symptoms and function improve following decompression, as well as from decompression and fusion, so both of these surgical options, provide very good results for patients slip to will ultimately help us classify whether there is a population of patients with grade one spondylolisthesis that might do well with decompression alone over time, and again long term studies will be necessary for us to understand this in greater detail. So at the present time, I would leave you with the notion that at least in the United States, using the slip study is our reference point decompression and fusion is superior to decompression alone, using validated health related quality of life outcome at two, three and four years for single level grade one spondylolisthesis. Thank you for the opportunity to to run through these trials, and this discussion of surgical decision making for grade one spondylolisthesis, and then I would hope that as we study this problem further that we would recognize that as we take a look at that doctors and patients should retain the right to make individual decisions regarding the utilization of fusion or not, based on both the desires of individual patients, as well as their individual patient anatomy. So again, thank you very much. Thank you very much. I wanted to thank Dr. Momenini and Dr. Wang and the members of the organizing committee for the section and I'm going to speak just briefly on our surgical video and a technique of a specific case for cervical arthroplasty. These are my disclosures, and none are specifically relevant to artificial disc of the cervical spinal arthroplasty in general. So, beginning with the case, this is a 37 year old right handed Asian male who presented with a history of progressive nine months of left upper extremity weakness, some triceps wasting and wasting of hypothenar muscles. On MRI evaluation, you can see a C6-7 disc herniation in the lower frames with bilateral stenosis. He had on EMG testing besides neurological examination, he had active denervation of the left C7 and C8 nerve reticular roots. His neck scores were mainly that of arm pain, more dominant than neck pain, and his ADI score significant from his disability from his arm weakness. On flexion extension films, there was no evidence of significant instability or facet subluxation. So, our surgical plan on this patient, nothing unusual, no hidden zebras, but was to treat the C6-7 pathology and to decompress and remove the osteophytes and the herniated disc fragment. Patient was very adamant about wishing to preserve motion at this segment, and he had a brother and a parent who had cervical fusions, and he was hoping to avoid some of the adjacent level risks and risk of pseudoarthrosis, which his two siblings and his family also had. So, ultimately, the decision was made to pursue a decompression with an artificial disc placement. So, with regards to positioning, this is similar in our OR and in our practice to use a very similar technique for routine ACDF. We typically place a gel roll underneath the neck to maintain a neutral or near lordotic posture. It is important not to over-distract and use holter traction to achieve hyperlordosis, because that will create the segment difficulty when placing the artificial disc in the lordotic segment, because ultimately, the discs have to be placed in essentially parallel or neutral segment anatomy. So, typically, we use holter traction, not garden walls, tape the shoulder, and place about 8 to 10 pounds of holding weight. Here, you can see that lateral fluoroscopy achieved early on. We ensure, for just surgical ergonomic ease, that whatever target level we're working on is relatively straight up and down to make insertion of the device and the approach easier. We will manipulate the bed control with the head down and the trandelion broke to achieve this, such that straight up and down, as you can see in the needle below, is relatively straight up and down to the working angle and to the target level. And then that is simply marked of, as usual, fluoroscopy, like a routine anterior cervical discectomy infusion, and we check an AP to ensure no rotation. We found that the use of a modest weight and holter traction is typically enough to maintain stability of the head and avoid rotation. But in some cases of severe and difficult patient's anatomy with very short necks or short chins, we will sometimes, on occasion, use garden walls tongs. And again, the key is to not avoid making the patient extremely hyperlordotic or, for that matter, kyphotic as well. Then, at this point, it proceeds in a very similar fashion to a typical anterior cervical discectomy infusion approach. We typically obey a left paramedian approach, although the literature is mixed regarding recurrent laryngeal palsies, but our standard of practice is to approach from the left. We like to avoid excessive cautery to avoid injury to the recurrent laryngeal nerves and also to the sympathetic chain. And then we also limit our soft tissue exposure, and we work segmentally one level at a time to reduce this traction injury. Typically, we will use a Zomed ET tube monitor for recurrent laryngeal monitoring, and we deflate the cuff once the retractor blades are set, again, allowing the ET cuff to reposition itself and to prevent intramucosal injury to the recurrent laryngeal nerve or the superior laryngeal nerve as well. At this point in time, caspar pin placement in artificial disc arthroplasty is important. This represents sometimes a deviation or shift from the way we typically place caspar pins. The pins should be inserted perfectly parallel to the end plates with a staggering offset to allow for insertion of the artificial disc with the inserter and its stop, and you can see that there. So we often will stagger the pins in multilevels to ensure that the inserter will not hit the pins upon insertion of the artificial disc, and you avoid interference with instruments as such. It is fundamentally important in artificial disc placement that we maintain the end plates to be parallel at the time of the disc insertion. Also, we have to ensure that it stays centered on midline more than artificial disc cages or more than standard discectomy infusion cages to ensure that the artificial disc is optimally placed at the end. So now a thorough meticulous discectomy in a usual fashion is completed, and here in artificial—in arthroplasty, it is crucial that you maintain the bony end plates if at all possible. This includes sometimes also preserving the anterior osteophytes, which is different when we typically place anterior cervical plates. Avoiding overdrilling of the anterior osteophytes, preserving the bony dome of the inherent upper end plate are all things that are learned in time as you gain experience with arthroplasty to ensure proper seating of the often domed implant and also to avoid subsidence and also heterotopic bone formation. Here you can see the use of a classic, typical parallel distractor device. And with this, we ensure that the end plates remain parallel during the device, often will distract within this parallel type distractor jig as opposed to directly on the bones. And at the same time then, we will click and progressively distract our cast bars. So we use our cast bars in that sense to hold the distraction that the parallel distractor blades achieve. And now we—once we have the height to a proper height, we begin our decompression. We often will do the end plate parallel distraction prior to any significant discectomy or removal of the chondral cartilage end plates to essentially protect, if you will, the end plates so that when we distract the parallel jig, we do not damage the bony end plates. At this point in time, we do a typical balanced foraminal decompression. And in the case that we presented, there was a free fragment on the left that was removed. And again, it's important to make sure that we take out the—all the compressive pathologies. But also, with artificial discs particularly, because movement continues after the segment is—after we treat and operate on the patient. As movement continues, it is very, very important to do a very thorough decompression of the posterior uncofortable joints to ensure that there is not progressive motion and movement there that can lead to hypertrophy of that at a later date and subsequently recurrent stenosis in artificial discs, which unfortunately can happen with subsidence and prolonged motion. Finally, we typically release the posterior longitudinal ligament in these cases. Again, in an effort to maintain parallel architecture of the segment, PLL often has to be released. It is often thickened or hypertrophic or partially calcified. We leave the bony end plates intact if at all possible. We do not burr them flat as we traditionally have with artificial—with anterior fusion cages. And again, always consider the inherent anatomy of the implant itself. As you can see, this is a Zimmer, a very typical common implant on this available for use for one and two levels. You can see, again, the domed end plate, which is intended to match the anatomic doming often of the inferior aspect of the superior body. As such, we now try to preserve that as much as possible. We do still, unfortunately, have to manually remove posterior osteoarthritis to allow for posterior replacement of the device—of the implants. But it is important to…and to avoid bleeding or significant bleeding, as that can be a…bone formation. And we use copious irrigation in that respect to remove the bone… That implant thing is very key. Over the years, we've learned that the key is essentially simply to pick the largest implant possible. So, you go from onchus to onchus and find the initial measuring tool that matches that so you can achieve the largest possible footprint, as you can see. Sometimes, the particular size of the patient will be between two. We default to trimming the onchover pituitary joints to get air to the larger implant, if possible. Again, I remind everyone to please wax the edumation early on. So, here's a typical…based on the initial… Oops. Oh, I'm sorry. We've switched over. Can you still hear me? Oh, good. Sorry about that. A little bit of a technical difficulty. But here, again, the implant trial is this size. Next slide. And here, we can see that the trial has been selected for the appropriate end plates and the appropriate sizing. We ensure the segment is not rotated. So, this is the time where AP and lateral fluoroscopy is critical to ensure midline placement as well as that the implant is perfectly aligned. Often, there are visual cues on the trials, like holes, where we line them up and make sure that we are not rotated. We want to avoid overstuffing the end plates in these cases. And we always want to, also on the lateral fluoroscopy, look at the facets themselves to ensure that they're not being over-distracted or, for that matter, being rotated and impinged. Next slide. Here is some positioning tips from our experience. And you can see the optimal centering of the dome. It is important to ignore osteophytes. You want to look at the true anatomical center of the vertebral bodies and avoid being fooled by anterior or posterior osteophytes. Essentially, you want to put the object or the implant in the center of load bearing, as can show in there. Unlike some other implants, it is important to understand the biomechanics of the implant you have chosen because some are better suited placed most posteriorly, whereas other implants are anatomically designed to sit in the center of the vertebral body, as in this one. Next slide. So final implant insertion is critical. As you insert the implant, there are often visual cues on the various devices to ensure that you're not rotated, as you can see in the pictures below. Also, having someone stand at the foot of the bed, ensuring that they're vertical and properly placed is important. Gradually tap the device in and put it into the desired center. In this case, the dome should match the anatomic center. Ensure that the level is parallel to the operating room. We confirm disc rotations with philosophy before insertion, as you can see there. Next slide. Ultimately, the implant inserter is then removed. Oftentimes, because of the nature of the implants and the two parallel end plates, it is necessary to adjust one or the other. There are often tools for this, and it is very important that this be done very gradually, as sometimes excessive force or moving one end plate excessively relative to the other can result in dislodgement of the overall assembly, and often at that point requires it to be removed. In our experience, most of the time that we prefer to put the implant in as one unit, get it into the ideal trajectory, and then leave it there. You can see on the bottom slide that, again, we ignore the anterior osteophytes and place the implants in the anatomic center, if at all possible. Next slide. And then typical wound closure is very stereotyped at like a typical anterior cervical discectomy infusion. We will be meticulous in sealing the caspar pin distraction holes, as they can again be a source of bleeding and anterior hydrotopic bone formation. We wax these holes typically, and although controversial, we ourselves over the last few years have injected steroids into the foraminal areas as if we're doing a transforaminal injection from the front, and then seal the space with a bit of tissue to, again, avoid migration of that. Typically, we do not place a drain, as these are relatively not very bloody procedures, and subcuticular stitches and dermabond are then placed to allow the patient to shower that day. Next slide. Post-op management for us typically is such that we allow the patients to mobilize very early, because this is an artificial disc. We allow them to move. Sometimes we give them a soft collar for muscle spasms for comfort, but typically this is not necessary. We typically do give them a course of two weeks of NSAIDs to three weeks of NSAIDs, again, to decrease inflammation and hopefully, in theory, decrease the incidence of hydrotopic bone formation. But typically, this is a very short rehabilitation course, and activity is allowed for the patient early on. Next slide. So ultimately, we can see that, next slide, the post-operative outcomes, we can see that the patient did relatively well, 45 minutes of surgery, typical course, nothing unusual. And typically, this can be done on a near ambulatory basis, this patient particularly left at eight hours. Next slide. Long-term outcome, which is my last slide, at one year, NDI scores reduced to seven, left arm pain and radicular symptoms resolved. There was, of course, some residual neck pain, a score of one out of 10 or 10 out of 100. Motion was preserved in the segment, and no adjacent segment changes were detected at the adjacent levels above or below at one year. Next slide. Thank you very much, and I appreciate your attention. Hi, welcome back to AANS and the spine section. I've been tasked with spending the next 30 minutes talking to you guys about MIS-T Lift. It's kind of a daunting task because I always like to say no two surgeons do the MIS-T Lift exactly the same way. So I'm gonna try to offer a little bit for everybody. So here are my disclosures. And I would first start by saying that evolution is inevitable. Larry Koo just spoke and Larry and I were residents together. He was a year ahead of me. I remember doing one of the early MIS-T Lifts following in his footsteps. And it was a big deal back then, took forever to do and it was very complicated. But along with the evolution comes something called obsolescence, which is the idea that the world is gonna change. And we see that. And that's one of the things that makes spine surgery so engaging and fascinating. Now, this is a picture of Greg Basil. He's one of our chief residents, hopefully gonna join our faculty. He and I talk about this a lot. And I say, look, you know you're ready when you're starting to dream about your spine surgeries. And I usually do the night before the surgery, I think about him and here's him dreaming about the spine. So if you think about what we're doing here in the evolution, if you will, or changing of MIS and T Lift in two separate tracks, you can think about it like a tree of evolution, right? So you would say, well, maybe early spinal fusions, like a Hibbs fusion was a starting point. And then we got to something like instrumented fusion. We heard Zogo Gawala talking about instrumented fusion, how much better it is for certain diseases like a slip. And then we can move on to something like interbody fusion, right? Interbody fusion lets you take the disc out higher, higher union rates, maybe less destructive procedure because you expose less to the spine. And then you could end up with things like posterolateral fusion along with that, right? But that might be something more of a dead end, whereas MIS interbody fusion survives and cortical fixation maybe doesn't, right? So you can start to think about the evolution of our procedures in this way. But if we go back in time, we can also look back to the origins of the concept of T Lift. And T Lift is an interbody fusion that HARMS had innovated and people were already starting to do because of the idea of the neuropraxia rates from a bilateral pliff, like a classic Paul Lin clouded pliff were not low. They were about 3% because of all the nerve root retraction. So coming from one side lets you do that well. Certainly open T Lifts were being done and it led to the rebirth, the rejuvenation of this idea of interbody fusion that was robust and easily taught. But again, you still had neuropraxia and I'm sorry, you had lower rates of neuropraxia, but then you still have the morbidity of the posterior approach, pain, CSF leak and all that. Now, why does this matter? Because if we look at the trends in surgical intervention, we see that the trends are towards sort of more complex types of surgeries, but also for more complex patients. So here's one of those papers looking at that. And if you look at instrumented fusion, the inner body fusion, the rates and percentages increasing significantly over year after year after year in this one decade period. Now, MIST Lift of course, was sort of the culmination if you will, or the stopping point of this, because we found that we can do a fusion, we can do an inner body fusion and we can do it without being so destructive. So the idea of working through the tube made it a workhorse procedure and it allowed us to do a lot of basic, simple, but necessary traditional things like decompression, cage placement, anterior column support, distraction and pedicle screw fixation. And now with less pain, blood loss and complications, right? So I'm gonna work through this. This is a little simpler, right? These are sort of a classic MIST Lift of how you would do it in these nine steps, if you will. And so patient positioning, for those of you who don't do a lot of this yet, obviously a lot of things get in the way. So when you're doing MIS procedures, you may be more using navigation or robotics or fluoroscopy. So you gotta be aware of all these things that can get into your way, like retractor arms and the microscope and things like that and setting up your room. So thinking about this a little bit different than a traditional surgery, this is my classic OR setup where I have the patient in a certain position and I have anesthesia and monitoring, all this stuff laid out nicely in my head. And for our wake procedures, which I'll show you later, we actually will change the room based on whether it's right or left-sided. And then, of course, the second step is where are you gonna make the incision, right? So if you look at this type of AP view, we've always typically stated that the MIST Lift takes a slightly more lateralized approach than either a microdiscectomy or a stenosis decompression. An idea being that you're gonna take a route more directly into the disc and there's probably gonna be potentially less bone removal or there's gonna be an ipsi contra component of the decompression, right? And so people say, well, how do I do that? Well, you can measure it. You can go to the MRI and say, well, look, I want this angulation. I wanna take it this far out. And based on the size of the patient, the amount of soft tissue, you might be able to actually determine the number of inches or centimeters lateral to the midline. And of course, that's gonna vary on your surgical purpose, your goal, what you're trying to get to, and the patient's particular anatomy and body habitus. So I really like to use the Wiltzi plane. This allows you to do a WMIS procedure. In other words, not only are you using a tube, but you're also gonna dissect between the muscles as opposed to just putting a K wire in and distracting and expanding the muscles. You're gonna dissect this plane through a mini open sort of approach and be able to go between the longitimus and multifidus muscles that minimizes bleeding. It minimizes retractor creep and gets everything done very well. Leon Wiltzi demonstrated this nicely. And if you like doing open surgery, I would encourage you to revisit the Wiltzi approach because this is a great way to get right on to the lateral facet, TP junction to put in pedicle screws and get a postural lateral fusion with minimal muscular dissection. Because again, we're working between muscle bellies. Then of course you place a retractor, right? And so there's so many retractor systems out there. Some are pedicle based, some are based off of a retractor arm attached to the bed, and then you can get your retractor in under fluoroscopy. Here we're targeting L4-5 and a lot of them are expandable now. So you can expand in bidirectional expansion independently. So you can say, I need more cranial caudal exposure or I need more medial lateral exposure, right? And this gives us our access. So the access allows us to then very quickly work. And here's another slide showing some people do screws first. So screws on the opposite side to allow for stabilization or distraction when aging. I tend to not do this as much because the workflow is a little trickier getting fluoro in and back out and having to come back to screws at the end. Then the steps of decompression and fascitectomy. This is critical. This enlarges Camden's triangle. It lets you decompress the nerve root on the ipsilateral side and potentially on the contralateral side. And so again, here, we tend to tailor our approaches so that we remove enough bone so that we can clean the nerve root. Certainly the traversing nerve root is usually the one offended. The exiting nerve root will often leave covered, as you see in this diagram. That way, when we apply our cage, we don't get any kind of impingement of the DRG or dorsal root ganglion or the exiting nerve root. And this shows you in a live picture, you see the traversing root nicely decompressed and then the DRG being protected in this upper area. Sorry, that arrow was a little bit off. And so in some cases, for example, you're just doing a fusion. You don't even need to see the traversing root. You can leave the facet partially intact there to protect that nerve root to minimize retraction, right? And then of course you've got to get in a disc. And this is a picture showing how not to do it, right? So here's a cage that went into the vertebral body at L5 when you're trying to do an L5 S1 disc space. It's important just to remember the trajectories when you're doing MIS surgery. It's different from open surgery where you don't have all the orienting anatomy around you so that you can work directly. And some people like to use the fluoroscope. As you get better and better and better, you can just kind of use your own cerebellum to figure out, well, this looks like the right alignment and I'm gonna distract. I'm gonna maintain that trajectory. Some people like to cut end plates. When you're using non-expandable cages, this might be critical because that overhang of the posterior lip of the vertebral body can be one of the impediments to getting a big enough cage in. Just be very careful when you're using box chisels or using osteotomes. It does create bleeding. You can hurt nerve roots. It can lead to subsidence and also placement of the cage, as you saw in the last couple of slides, into the vertebral body itself. And so cutting those little edges or just burying them off with a drill might be very helpful. There's lots of ways to get different kinds of cages in. Some of the systems rely on this, which is inserting and rotating. Because if you look at your decompression and your approach, you have a very, very wide medial to lateral confine so long as you take enough facet joint off. But you're trying to usually lift cranial caudal. So a lot of folks like to insert cages, and I don't wanna use any brand names, where they insert it and then rotate it because you insert the cage being wide, meaning medial lateral, and you flip it up so that it gets inner body distraction, correcting a listhesis or correcting top-down stenosis. Other cages are more bulleted that allow for automated distraction. So having a bulleted cage allows you to go through, and you can see the lifting of the intervertebral space, and then it closes back down around it. Getting bulleted cages have the advantage of doing this. Disadvantage is less contact area, right? So less contact area up front in terms of the anterior aspect of the vertebral body, right? And of course, this cage correction issue is very important. If you look at this case where the patient's cathodic and collapsed beforehand, so you're not only reducing the slip, but you're creating inner body height and lordosis, this relies heavily upon the cage when you're doing MIS surgery because it's hard to manipulate these screw rod systems with the MIS approaches, right? Remember, you want your cages to cross the midline, ideally to get the maximum anterior load sharing, to support the anterior column, so you have less cases of cage retropulsion and subsidence as well, right? And then onto the pedicle screw fixation. I know some people are gonna start with screws on the contralateral side. There's lots of ways to put in screws. I personally like to do something called AP-only cannulation because we have all forms of navigation, and we have robots too, but AP-only lets you get started. And the idea here is that you line up the end plate here, this is the upper end plate, line it up so it's collinear, so it's like one line instead of an oval, and spinous process in the middle. With that, you can see your pedicles, and you can see the medial walls of the pedicles. When you have that view, you can do everything else within this. You start your docking onto the bone, whatever system you use, and then you go in two centimeters into the bone, as long as you don't pass the medial wall of the pedicle here, you're good. And that's really the safest and easiest way to use fluoroscopy. You can use biplanar, you can use an Alzi and Foss approach. There's lots of ways to do this. This is the most efficient, most radio-reduced way to do it without using navigation or robotics. And it lets you get into any type of case, scoliosis, rotational aspects, atrophic pedicles, and all of those types of scenarios. And then the rod insertion, right? So all different ways, different companies have different ways for rod insertion. They're all pretty straightforward in terms of how you do that, right? So does this work? Does all of this stuff I'm telling you about lead to better outcomes? For example, like reduced pain. So here's a nice paper from Kevin Foley, who was one of the first people in, one of the first surgeons to do MIS-T Lift in the world. This is a force plot looking at multiple studies, looking at MIS on the left. So force plot has the zero line here in the middle. On the left favors MIS, on the right favors open surgery. When you talk about reduced pain, the size of these rhomboids indicate the size of the study. The distance from the zero indicates the effect. So effect and population size. You can say all the studies, of course, there's probably publication bias, but all the studies, as expected, show reduced pain with the MIS surgery. What about earlier normalization? So you can get less pain, but how quickly are people getting back to their daily lives? So this is a review of a couple of papers. A lot of the work came out of Vanderbilt, where Joe Chang, Scott Parker, and Matt McGirt were. And it shows, of course, that if you look at things like reduced narcotic use, right, MIS in blue, open surgery in orange, you're seeing a lot less narcotic use, a lot less issues with that. And that's important. We heard about the opiate pandemic earlier today on the, in the plenary session, number three. Infections. This is a paper I wrote looking at the reduced infection rates using Premier database and thousands and thousands of cases and reduced costs because less of infections, right? Less infections because you're opening less of the soft tissue. There's less of a hematoma or a hollow viscous space, if you will, left behind after your surgical cavity has been closed, right? CSF leakage. This is a paper also by Kevin Foley, looking at that again, force plot, looking at the rates of CSF leak. You're not seeing as much through it. You're not usually looking as much at the central fecal sac. So the rates of CSF leakage are less and probably the effects of CSF leakage are less. In other words, the need to close it as tightly is probably less. The narcotic consumption and the return to life are all related to the return to normal pain levels and improvement rates. So in other words, people have a surgery. They not only want to get better from their preoperative symptoms, they also have to recover from your surgical insult. So if you look at things like back and leg pain and ODI, of course, the MIS approaches here show a quicker return, normalization to life in terms of the dashed line as opposed to the solid line, right? So that's all good and well in terms of the effects. Return to work, MIS here in the dotted line showing the return to work being significantly faster than open surgery. So here I've shown you sort of the tubular approach for MIS and MIST lift. And the question is, is the tube the same thing as MIS? And let me just add that we're not talking about some of the other access corridors, whether it's to an anterior lateral or anterolateral space, lots of different approaches are being innovated. And I'm focusing just on the TLIF approach. You know, things like ALIF are great, there's pros and cons, but here I show some of the complications like vascular injury or screw extrusion up against a vessel. And lateral approaches have negatives too, like having, you know, thigh pain, lumbosacral plexus injury, but they have true virtues as well. And I use all those techniques in my armamentaria, right? So MIS TLIF, the advantages of course, it's widely applicable, it's well understood, it's a natural technique, it's demonstrated effective, it's reimbursed well, it allows for direct, indirect decompression, both fusion and instrumentation. But of course, some people still have not adopted. If you look at all the TLIFs that are done, probably less than 20% are still being, are being done currently using MIS techniques or right around 20%. And that's because it is a little bit harder to learn and replicate. The advantages are incremental. It's not transformative, it's not standardized. And some of the traditional complications are still seen, and maybe most importantly, we're not reimbursed anymore for doing an MIS, right? So I struggled with this, this issue of the cost of surgery and what that really means. And so here's a patient of mine very early on with an inciting event, having increased pain, and then they have a surgery and guess what? They have more pain. They're in the hospital, they're recovering, and then eventually they get back to healthy state. And this cost of surgery in an open TLIF could be reduced with an MIS approach, right? But can we get even less than that? Can we get even better than that? So we started to innovate our awake TLIF, and there's an NS focus that Praveen Moumanini is editor on, and Zach Wilson coming up this fall or this winter, right? And so I started getting interested in this because of the existential threats to spinal surgery. I mean, people are really afraid of what we do. They're afraid of the pain, the recovery, the variable outcomes, the negative sequelae, but yet we're dealing with an impressive demand for our services and the economic costs for our society. So I always say, well, you know, the way around all these existential threats is to get smaller, right? So I've argued for a long time that rather than getting bigger and bigger and bigger with surgeries, more rods, more approaches, you know, multiple approaches in the same patient, more product, let's try to get smaller. So we innovated our awake anesthesia surgery that is an awake fusion, ERAS fusion, endoscopic fusion, percutaneous fusion. And we started with the endoscope, not using general anesthesia, percutaneous fixation, long acting sodium channel blockers, expandable inner bodies and osteobiologics to enhance fusion. And we had our, we called it ERAS version 1.0 surgery, right? And, you know, people say, well, aren't you just doing percutaneous fusion? And these are some pictures through the endoscope showing what you can see, like the exiting nerve root here with the fat around it, how well-protected it is, the end plate being prepared, the bony end plate being ready for accepting a fusion with bleeding, the cartilaginous end plate stripped essentially off of it, right? And this shows our data, just like Zoe said, spondylolisthesis, that's the indication, right? So in other words, mostly we're doing spondylolisthesis and top-down stenosis, you know, recurrent disc herniations and deformity actually. So here we are doing these surgeries. This shows you the outcome. It's already been published in multiple venues. So I'll spare you a recitation of all the data that we have because everybody's on Zoom and probably falling asleep like me. So we've done this over 320 times, lots of experience with this. We're able to do surgery in no-fly zones like this gentleman with an EF of 11% and in all kinds of different surgical settings. And I would encourage you to look at endoscopy, working channel endoscopy, if you have not already. It allows you to leverage visualization in the depths of the body without opening things up. So probably when you look at a spine surgery patient and you see a BMI of 50, you go, wow, even if I thought the outcomes were just as good, I would be a little reluctant because that's going to make life hard on me. But actually with the endoscope, it really makes minimal difference. So that's one of the powers of endoscopy and we can be sure that the BMI of our population is increasing. Here's another example. This is a patient who had a lumbar stenosis problem, went and saw a very reputable neurosurgeon, had a decompression for neurogenic claudication. Here's the surgery performed very well, expertly, no problems, but had a dural tear and CSF leak, right? So now this patient is found post-op day one to be abtunded, has a subdural hematoma, secondary to the decreased pressure in the fecal sac and ends up in the ICU for months, okay? Recovers eventually with multiple neurological issues, but makes a remarkable recovery. And then ends up coming to see me, needing surgery because of intractable pain and sort of like a post-laminectomy instability and collapse, top-down stenosis, recurrent stenosis. What are we gonna do with this? Who is gonna venture in and do a surgery posteriorly, risking CSF leak now in the presence of scar? Now, for those of you who would like to do anterior to psoas or lateral, or maybe A-lift, I can totally buy that. You're gonna come to a different route and lift this patient up and then proc screws, but that's a pretty sizable operation. So what are we gonna do? We're gonna take this patient and we're gonna do the awake fusion. So here we're controlling for everything, no general anesthesia, no exposure to the fecal sac, inner body fusion, whole case takes 100 minutes, skin to skin, she's in for less than a day, in and out, you can see the effect on the ODI. So it's not perfect for everybody, but you can see special circumstances and I can show you dozens and dozens of circumstances where I would say a open surgery would be potentially a no-fly zone. But we're coming off coronavirus. In fact, I didn't think I'd be here 18 months out, still talking to you on Zoom. I thought we'd be in Orlando having a drink together. And here's what was going on in April of 2020. This is the phased approach to reopening, whether you like the politics or not, this is what was going on. Here's phase one. Phase one was proposed to deal with elective surgery. I mean, I've said it before, I apologize if you heard me say this, but I don't think I'll ever live to see another president of the US talk about elective surgery on any major platform besides maybe the WNS. Look at what it says. It says, we can resume as clinically appropriate on an outpatient basis if you follow the guidelines. This is April, 2020, and we're not even in Orlando together. So this makes a huge difference. We're through wave five now here in Miami. So patients are afraid. They still are. Economic collapse. We're not sure still what's gonna happen economically to America. The push for single payer, it's coming. This is an accelerant on that. The number of privately insured patients decreasing. People having less private employment, right? The insurance companies can say whatever they want now. They're dealing with coronavirus and tremendous economic pressure. Who out there on this Zoom has not been told, aren't you gonna do more surgery? What happened? We're dying here in administration, right? Go out there and make some money for us, right? So which of these effects are gonna be long lasting? So let me show you that in spring of 2020 is when doing surgery MIS really matters because it didn't really matter before. Everybody said, well, I'll just do it open. I get a higher fusion rate. Why would I stress myself out? I don't like radiation. Well, guess what? We have been able to do surgery through coronavirus because of this. So here's an example of this. It's a lady now during April of last year, May, I was probably like you more at home, traveling less certainly. And they were saying, well, you know, the hospital's full. There's not enough nurses to do care for patients after surgery. So, you know, stop your elective cases unless it's myelopathy, a tumor, a fracture or whatever. Fine, okay. So I'm looking at the U-chart, the Epic system. And you know, you get those messages. I don't really look at those because who has time for the thousands of messages coming through with all your patients? And I actually see this message that says, Dr. Wang, I'm gonna kill myself. And I'm like, huh? I guess I never paid attention to this stuff. So I called this lady at home because I'm just sitting at home too. I said, what's going on, ma'am? She goes, well, you know, when I saw you in February before pandemic, I was being managed by shots and I was kind of getting along okay. But now, you know, not only can I not get the shots, they stopped working. And I said, well, what's going on? She goes, I can't even walk to the bathroom. And, you know, I have kids and, you know, and I will not go on like this. I'll live, and she said to me, I'll literally take my husband's shotgun and I'm gonna kill myself because I can't go on like this. So I call the hospital admin. I'm like, listen, you know, I get it. Everybody's hyper about coronavirus, but are we just gonna let other people just suffer? They said, well, how are you gonna do this? I said, well, we'll do the awake fusion. So sure enough, we take her to the hospital, do this, outpatient, get her better. And the hospital administration goes, wow. So you did that in like 47 minutes. A patient didn't stay overnight and we're getting the same thing we normally do. And I'm like, you haven't been paying attention, right? And they're like, please do more, right? So I've never been busier than during pandemic. Never in my life have I been this busy doing surgery. So questions are, well, I mean, how do you access Camden's triangle? Maybe you don't do endoscopy. Maybe you're worried about radiation. Maybe you can't get BMP, right? Who knows, right? Who knows what's going on, right? So, you know, we've entered the age of robotics. So I don't wanna end the talk of this length without talking a little about robotics. This is from CNS 2019. These are robots, neurosurgical or spine robots out there already. And so we said, well, why don't we do this robotically? This is right at the beginning of pandemic. So let's take away the anesthetic part and just do it under general and then replace it with a robot. So, you know, now you get the endoscope for miniaturization, you get localization with the robot, you get stabilization and all the other pieces being the same, right? And this kind of shows you how it goes. This is put together by Jason Leonakis, one of our chief residents who was a fellow two years ago. And this video just kind of shows it. And what I'm really demonstrating here is that we're gonna leverage, in this case, it's a Mazor X, right? And we're gonna leverage the robot to allow just about anybody who wants to do what I do, which is a percutaneous endoscopic ERAS fusion, right? We're not gonna do it awake because we don't want the patient to move through all these parts. But you can see we get the registration with the OR, we get the robot. And here we're gonna put the screws in first because once the case distracts, we're gonna move the body, right? So we don't wanna spin twice. So we're just gonna drop the screws. We're gonna use the long-acting sodium channel blockers to get long-acting analgesia. And then we're gonna place our screws using the robot, right? So people are like, well, wow. So now you don't have to get all that floral exposure and you don't have to worry about your hand shaking when you're using that. Yeah, just use the robot, right? And the robot's gonna let us put in our screws for sponding, in this case, it's gonna be four screws. And we're just gonna use our K wires at first because we're just gonna make things simpler here. And you don't have to do that. You can just put the screws directly if you want. And then we're gonna drop those over them, right? So here go our screws. And then we're gonna get into Kambin's triangle. So here again, we're gonna leverage the robot. The robot's gonna say, okay, well, spine hasn't moved. So let me find for you Kambin's triangle, right? And here we are going through Kambin's and we're gonna use electrophysiological monitoring this time because the patient's asleep, right? So we get no stimulation in the patient. There's the stem. And now we're gonna go through Kambin's and we're gonna go into the disk space and we're gonna get all this figured out. So this allows us to start to do things in an automated stereotype way with control, using the robot to figure out a lot of the pieces that we didn't want to do with, you know, with a, how do you say, with just using your own personal talent. I spun the arm again. This is the first case we ever did, by the way. So I spun it again just to check everything, make sure we're still good and actually to get a great picture of what's going on. So you can see now we're working with the endoscope. Now we're doing that disk cleaning clearance and we're gonna put in our cage or expandable cage. And all those pieces are gonna descend just like the other. This is actually quite simple. These steps are really straightforward in terms of inner body cage placement, right? So in terms of the future of medicine, this is where it's at. It doesn't matter, especially high-tech, biologics and MIS are gonna be the future. MIST lift is a highly versatile technique. It comes in many forms. You can treat all kinds of problems. There are reduced problems with MIST lift over open surgery. That's been well demonstrated in numerous large studies. And with advances in technology, we're gonna get smaller and smaller, whether it be the endoscope, whether it be the robot, whether it be navigation or biologics, right? All of these are being developed to reduce the burden, not only to the patient, but to the surgeon now, the occupational health burden we suffer when we try to get through these surgeries. But always remember, we're measuring the results against the gold standard. In this case, it's gonna be the open for our even less ultra MIS surgeries against traditional MIS surgery. So that's a very good way of looking at things. So with that, I'm gonna finish with a little bit time left. I know everybody probably wants to hurry up and get to drink or something like that, but I also wanna give a shameless plug to our neurosurgery podcast. We just passed a quarter million listens, which is not so bad for a venue that is really confined largely to surgeons. So thank you to the AANS. Thank you for tuning in. Hopefully we'll see you next year in Philadelphia, live and maybe virtually too, but certainly hopefully live as well. Take care, be safe and God bless. Hi, everyone. It's Mike Groff from Boston. Thanks to everyone who's sticking it, hanging in there and sticking it out to the end of the meeting. Mike, that was phenomenal. And to all the speakers, those were all great talks and very informative. Appreciate all the insight on the clinical management, the technique and decision-making. So now we're gonna pivot slightly and move into the abstract presentations. And we'll lead off with Dr. Yolku, who's gonna talk to us about the outcomes and complications for patients with sacral chordoma. Thank you, Dr. Groff for the introduction. I'm Yaz Yolju, one of the prelim surgery residents at Mayo Clinic, formerly working with Dr. Biden at Mayo Neuroinformatics Lab. I'm gonna present our study, comparing different treatment modalities to carbonide radiotherapy for sacral chordomas. No disclosures. So carbonide treatment is one of the alternative approaches to conventional radiotherapy for treatment of various lesions. And it's been gaining interest in chordoma treatment as well, especially centers in Japan and Germany extensively published results on this approach with some promising results. So as we are implementing this technology in some of the facilities in the United States, we wanted to take a look at the efficiency and safety of carbonide treatment in management of sacral chordomas when we compared to our experience in the United States. So this study has sort of two arms, two separate sets of analyses. One, looking at the Mayo Clinic experience, which was heavily based on on-block resection and compared it to carbonide treatment. And then we also looked at different treatment combinations like surgery alone, positive margin resection, negative margin resection, with or without adjuvant therapy and compared these to national carbonide treatment data. So we use our institutional cohort, which was based on on-block resection. We use an NCDB cohort treatment approaches from the United States with different treatment groups. And then we compared them to QSD Institute in Japan and how they treat the sacral chordomas with carbonides. So outcomes of interest. This list is mostly for the institutional comparison to carbonide, where primary outcomes were overall survival, progression-free survival in this metastasis. And we also looked at urinary retention, rates of colostomy during or after treatment and change in functional mobility scale and peripheral neuropathy. For NCDB part of this study, we could only look at the overall survival because only overall survival was available in NCDB. So the second thing was this technology in Japan is mostly reserved for older patients with lower functional status who are likely benefit from alternative treatments rather than surgery. So to balance this with our cohort, we did the matching prior to the study, which left us with 47 patients in each group on-block resection and carbonide treatment and matching was done based on age, sex, baseline functional mobility scale, highest sacral level of the tumor and tumor volume. So after matching, we saw that ECOG performance status was still significantly different between two groups. So this was later also controlled for in our regression analysis. And if you look at the outcomes, only difference we actually saw was patients who underwent carbonide irradiation having lower rates of peripheral neuropathy, which was determined by the adverse event scale. So if we move to regression models, looking at all of our outcomes, again, after controlling for ECOG performance status, and of course, after matching, peripheral neuropathy rates were lower in carbonide treatment when compared to on-block resection. So these are our survival curves for institutional cohort with on-block resection in comparison to carbonide treatment. We saw similar rates of overall survival, similar rates of progression-free survival, and similar rates of this metastasis. If you move to our NCDB comparison with the carbonide treatment, we see that when patients undergo surgery with negative margins, regardless of radiotherapy status, it did not have significant difference in terms of overall survival when compared to carbonide. But if you look at positive margin resection, we see that when patients did not receive adjuvant radiotherapy, they actually had lower survival compared to undergoing carbonide therapy alone. And additionally, when we looked at patients who did not undergo any surgery and only received radiotherapy, again, they had lower overall survival compared to patients undergoing carbonide treatment alone. So there are not many studies that we can actually compare carbonide treatment with or without surgery, so we can assess the role of carbonide treatment as an adjuvant therapy. But if you look at some of the studies that were done in Europe, we actually see that there are similar rates of overall survival and progression-free survival. However, higher metastasis-free survival rates for both unblocked resection, male cohort and also carbonide treatment from Japan when we compare to the Europe study. So we had certain limitations, of course, with this study. We had more granular data. We could look at metastases and progression-free survival with our institutional cohort, which was limited to unblocked resection. We could not look at the role of adjuvant radiotherapy here. We had this opportunity in NCDB. However, in that set of analyses, we had less granular data. We could only look at the overall survival. And the second big limitation for us was carbonide data being only available for patients who did not undergo surgery as they prefer to use this as a single treatment approach for patients who cannot undergo surgery or very large tumors who they call inoperable. So we could not really make an assessment for potential value of carbonide as an adjuvant therapy. But as a summary, carbonide treatment has similar outcomes with unblocked resection in an older age group with lower performance status. So this brought us to a few questions, whether patients who are poor surgical candidates or have inoperable tumors can be treated with carbonide technology instead of conventional radiotherapy. Second thing is, if we can use carbonide instead of conventional radiotherapy as an adjuvant therapy following a subtotal resection or even a gross total resection, depending on the preference. And when we have a recurrence after unblocked resection, is this technology suitable to use, again, as a secondary therapy? So I would like to thank Neuroinformatics Lab and our collaborators from Radiation Oncology Department and QSC Institute from Japan. And again, thanks to AANS Scientific Comedy for the opportunity to present here. Thank you. Afternoon, I'm Michael Phelan from the University of Toronto. It's a pleasure to be involved with this AANS meeting, albeit virtually, and hopefully we'll be able to get together next year in person. So I'm going to be talking about a propensity-based analysis of a large data set examining the ongoing controversial issue of how best to manage central cord syndrome. So no commercial disclosures, I'm on the author team, and in particular our neurosurgery resident Jetan Badiwala, who received his PhD during this work and was supported by NREF. So the epidemiology of acute spinal cord injury is changing, as we all know, and while the overall incidence is stable, patients are getting older, there are more fall mechanisms and there are more cervical injuries. And as a result of this, central cord syndrome is increasingly becoming a public health priority. And I think we need to revisit this issue and we need to optimize our management protocols. And we need to consider whether surgery influences outcome and is timing important. So the aim of this work was to evaluate the efficacy of early surgery defined as within 24 hours in patients with central cord syndrome. To undertake this, we used a large combined data set we've previously published on this earlier this year in Lancet Neurology to assess the overall timing of surgery. Here, we're focusing on central cord syndrome, ASIA impairment scale C or D, levels C1 to C8, and we define this using a difference in the motor scores between the upper and lower extremities of five or more motor points. And we looked at the impact of surgical decompression. We calculated propensity scores using the logit technique and we aligned these data sets on a number of covariates that you can read on this slide. So this is the CONSORT, a diagram, it's a large data set, and then this was honed down using propensity scored matching to have 93 patients in the early surgery group and 93 patients in the late surgery group. And then this shows the raw and then the matched data using the propensity scores. And you can see from the graphical analysis that these data sets are very well matched. And here we have a table showing the baseline characteristics for the groups between early and late surgery, so age, sex, and so on are very well matched. And it's widely recognized that propensity-based matching is as good as you're going to get with regard to analysis of large data sets other than a randomized trial, which is going to be logistically very difficult to do in this patient population. So here's the overall analysis of motor and functional recovery one year, and we see that for this overall data set that there is overall a significant improvement in the upper extremity motor scores. Importantly, this includes both Asia C and D patients, but there is heterogeneity in these groups. So this is an analysis of the functional independence measure, motor domains, the various subscores, and there are important trends favoring early surgery, but none of these actually met the significance level. So if we now stratify based on age impairment scale grade C or D, the results become more interesting, and to some degree this makes sense. And we see that for the Asia C patients, there are very clear effects, including the overall Asian motor scores as well as the subscores, but we do not see significant effects using these outcome measures, using the analysis for Asia D. And this likely is influenced by ceiling effects and the relative insensitivity of the Asia scoring system for this subset of patients. So in conclusion, it would appear that early surgical decompression is associated with overall improved motor recovery in central cord syndrome, but particularly for the more severe AIS grade C patients, and there does appear to be a ceiling effect with the AIS grade D patients, at least using these outcome measures. And I think increasingly we need to be thinking about central cord syndrome as an incomplete cervical cord injury that should be treated actually fairly similarly to other forms of an incomplete cervical cord injury. Clearly, further research is required with more sensitive outcome measures to more precisely delineate the impact on AIS grade D patients with central cord syndrome, and this continues to represent an important knowledge gap. And I just want to thank, in closing, the various funding sources and the many sites and investigators that have contributed to this combined data set, and thank you again to the AANS. ♪ Excellent. Hey, everyone. Thanks for attending the talk. My name is Ali. I'm a third-year medical student at the Icahn School of Medicine in Mount Sinai, and today I'm going to be talking to you about some of our team's work, titled Reliable Prediction of Extended Length of Stay Following Non-Fusion Spine Procedures Using Machine Learning Validated on Nearly One Million Cases in the United States. So I have no disclosures. My mentor has the disclosures indicated. So just as a brief introduction, bundle payment models are increasingly used for reimbursement of spine surgeries to combat rising healthcare costs in the United States. Hospitals and providers have become increasingly incentivized to reduce costs while maintaining or improving standards of care, and excessive lengths of stay are associated with numerous clinical risks and economic burden. So the automated prediction of prolonged length of stay can inform resource allocation to improve clinical outcomes and reduce costs. I'm not entirely sure why this is going forward on its own, so I apologize. So the goal was to use machine learning to predict excessive length of stay after non-fusion spine surgery on a national scale. So in regards to the methods, we used two separate data sources in order to conduct this analysis. So we started off with the Mount Sinai Data Warehouse, which contains about a little under 2,700 spine surgery cases between 2008 and 2016. And we leveraged the perioperative features in that dataset. We fed that to a gradient-boosted decision tree algorithm that was trained to predict excessive length of stay. And then that algorithm was validated on the national inpatient sample, which contains almost a million non-fusion spine surgery cases between 2009 and 2017. And again, it was used to evaluate length of stay, which was defined as the 75th percentile of length of stay in the Mount Sinai Data Warehouse, which came out to be about three days. So in regards to the results, we evaluated, as I mentioned, the algorithm on the Mount Sinai Data Warehouse test set, in addition to the NIS test sets. You can see that the algorithm performs pretty well on both test sets, and it generalizes well to the larger, more sort of national database as well. And you can also see on the right, the calibration curve indicates that the algorithm is able to predict risk of extended length of stay appropriately as the true length of stay. So on the y-axis is the true probability of increased length of stay, and on the x-axis is the predicted risk, and it's a pretty positive relationship there. So it's doing fairly well across many different quartiles of risk. And so we wanted to visualize with the algorithm can be interpreted, because that's an important aspect of machine learning in clinical medicine. And so this plot is derived from Shapley scores, which essentially is a game theoretic measure of understanding how different variables affect model performance. And so from top to bottom is the list of the top 20 perioperative features, which are the most common. And then on the left, the list of the top 20 perioperative features and characteristics that the algorithm uses in order to predict length of stay. And so elective admission is sort of the top predictor, and so that kind of makes sense. So if a patient has come as an elective surgery, then the likelihood that they have extended length of stay is probably low, because they're probably generally healthier. And a loss of function, minor loss of function, is also predictive of a shorter length of stay, and sort of older age, as indicated by the red, is predictive of a longer length of stay, which again makes intuitive sense. And so the other interesting thing that we can do with our analysis is not only look sort of as a bird's eye view to see how the algorithm does and some of the characteristics that are most predictive, but we can also look at the patient level. And so this is a force plot, again using those same Shapley values, to see which patients the algorithm is correctly predicting to have excessive length of stay or not. So this plot shows that for this patient, who the algorithm correctly predicted to have extended length of stay, you can see some of the variables it used that pushes it in that direction are the fact that this was a non-elective admission, that this patient did not have private insurance, and something that pushes it in the opposite direction is that the patient was a male. And so the likelihood in this data set that a male is able to have excessive length of stay is pretty high. In regards to true negatives, this was a patient that the algorithm correctly predicted would not require extended length of stay. And so again, some of the characteristics are that this patient was an elective admission, they have private insurance, and some of the other characteristics that maybe push it in the other direction is that this was a female, had a minor loss of function. But overall, taking into consideration all the relevant features, the algorithm correctly predicted that this patient would not require an extended length of stay. And similarly, we can do the same type of analysis to determine false positives and false negatives. So we can kind of hone in to see how the algorithm is making decisions and maybe analyze areas that it needs improvement in. So in this case, this patient actually did not have an extended length of stay, but the algorithm incorrectly predicted that they would. And some of the features that it was looking at was the fact that this patient was a non-elective admission, was an older patient, did not have private insurance, and was a male. And so those are typically characteristics in the data set that are indicative of a longer length of stay. But in this case, this seemed to be an outlier patient, and so the algorithm was incorrectly predicted for them. And similarly, on the other side of the coin, this was a false negative, and this was a young patient, and had minor loss of function, and was an elective admission. So you would think that this patient would be generally healthy and would not require extended hospitalization, but that wasn't the case in this case. So we can see when the algorithm is making mistakes, it's making them in a way that we wouldn't find unreasonable. But these patients are sort of corner cases that could still train the algorithm further to improve its predictive ability. So in conclusion, we show that machine learning is able to predict extended hospitalization stays after non-fusion spine surgery with fairly high efficacy. The predictions are generalizable to a large nationwide cohort of patients, and the algorithm predictions can be assessed to identify features of importance and examine predictions at a per patient level, which is very valuable from both a clinical perspective and from a computer science perspective, because it allows us to hone in on areas that need further improvement. So I'd just like to thank my research collaborators, Dr. Karidi and Dr. Zorman, my colleagues at Mount Sinai and OLAB for providing support and guidance, and I'd just like to thank you for observing. I'd like to thank the Scientific Committee for allowing us to present our work. I'd also like to thank the UT Southwestern for institutional funding, and some of the funding was also obtained from the National Institute of Health. Adult spinal deformity, as we know, is increasing in prevalence. This is a leading cause of disability among older adults. In fact, 60 to 75 percent of older adults the age of 65 would have some problem with adult spinal deformity. It's a top five cause of disability in this patient population, and when you look at the health-related quality of life metrics, adult spinal deformity really performs worse than patients with blindness, arthritis, diabetes, heart failure, and lung disease, truly making this a public health concern. When you look at the treatment options for patients with adult spinal deformity, we've clearly shown in spine surgery that surgery helps. Surgery increases the quality of life that these patients have, and in many ways, the quantity of life that they have. It's also associated with very high complication rates, and these patients have to make a trade-off between the complication rates associated with an operation versus conservative care. The ISSG has demonstrated that three in four adults who undergo adult spinal deformity surgery will have a complication, and those complications can range from wound infection, cardiopulmonary complications, neurological deficits, and even death. Now, when we think about some of the data that we have collected historically in spine surgery, we've collected data on patient-reported outcomes and patient satisfaction, but we wanted to explore this concept of decisional regret. Decisional regret, we felt, was a more important psychosocial measure compared to patient satisfaction or some of the patient-reported outcomes metrics that we currently collect. Decisional regret is a negative patient outcome wherein patients feel that he or she would have had a better outcome with an alternative treatment. We looked at our patient population that had lower thoracic to upper fusions all the way to the sacrum with iliac fixation and comprehensive follow-up. We had 155 patients that met this criteria. Ninety-one patients agreed to participate in the study. I should say that of the participants who declined participation, there were no differences in the baseline comorbid conditions and baseline presentations. We then categorized patients into medium versus high and low decisional regret. The primary outcome were rates of decisional regret after surgery, and then the secondary outcomes were factors associated with decisional regret. Now, decisional regret has been studied in many other subspecialties, oncology and so on, but we're hoping to really bring this to the forefront of neurosurgeons or spine surgeons. We used the Ottawa Decision Regret Questionnaire. It's a Likert scale and scored out 0 to 100. The score is less than 40 associated with little or no regret, 40 to 60 medium regret, and the score is greater than 60 associated with high regret. Here's what we found, which is quite interesting. Despite a very robust preoperative optimization strategy, one in five patients felt that deformity correction surgery harmed them. One in five patients regretted their decision to undergo surgery, which I thought was quite alarming. When you look at the associations with decisional regret, there was no correlation with baseline demographics. There was no correlation with invasiveness of surgery. There was no correlation with length of hospital stay, discharge disposition, or even complications. What was most interesting to us, the investigators in this study, was the extent of functional improvement. Despite patients meeting MCID, one in five still regretted the decision to undergo surgery. The question becomes, why? Why do they regret their decision? To answer this question, we turned to qualitative methods to interview patients and surgeons to understand what those decisional gaps might be that we might be able to take advantage of and decrease decisional regret rates in patients undergoing surgery. We utilized qualitative methods. This study was conducted and administered by investigators with expertise in qualitative research. We tried to get the patient perspectives as well as the surgeon perspectives. We interviewed both patients and surgeons who had extensive experience treating patients with deformity to really understand what the nuances are in how they make their decisions. We can help identify areas that we can help improve those decision algorithms. We utilized decision guides that were created specifically for surgeons and specifically for patients. Here you can see broad open-ended questions followed by more direct probing questions. These questions were really based on a theoretical framework of health belief models and patient-centered care. If you look at the patient interview, a sample question might be, tell me about any benefits or upsides to this procedure that your doctor might have discussed with you. Then honing down some more, can you talk me through what you understood about these conversations and can you tell me about your understanding of these downsides? From the surgeon interview guys, sample questions could be open-ended questions. How do you explain the risk and benefits of spine surgery to your patients? When you hone in some more, what do you believe is the greatest risk and benefits of this surgery to your patients? Here's some of the comments. These are comments that are really verbatim from patients. The process of qualitative interviews, I should say, after the interviews, they are transcribed verbatim and then they are coded with multiple investigators using both inductive and deductive coding strategies, as well as comparative methods with a co-coder to really understand what these recurring themes are. Some of the patient voices, comments here, you can see, we were told by numerous people that the longer I waited, the less likely it was to find a doctor who would do anything for me. I made my decision to have surgery even before I met my surgeon. When you go through, you've transcribed it, some of the common themes that rose to the top, patients felt they were running out of time. Patients felt that the only choice was to undergo spinal deformity correction surgery because you're running out of time to undergo invasive procedures. Theme number two, patients mentally committed to surgery even before the initial visit with their surgeon and contextualize the benefits while minimizing the risks. As a surgeon, what this means is while you're having these conversations about all the things that can potentially go wrong with the operation and really educating these patients, they tend to only hear the benefits and really minimize the risk. Theme number three, patients felt that the current decision support tools were ineffective. As surgeons, we tend to focus more on tools that educate patients about the surgery. We have surgical brochures and videos. Patients felt in general that these tools were ineffective. In theme number four, patients felt that pain management was the most difficult part of recovery which wasn't discussed comprehensively by the care team. I think most of you can commiserate with this when you spend hours upon hours speaking with the patients about possible pain management strategies and still we're not doing a good enough job with pain management. From the surgeon perspective, there was comments about frailty. Physiologic age is more important than co-morbidities and chronologic age and really setting expectations and expectation management really being key to the perceived success of the operation. Some of the surgeon themes that rose to the top, the surgeons did not consider chronological age as a major contraindication to surgery. However, the extent of functional improvement was considered almost a contraindication. Theme number two, while the interpretation of surgeons generally understood shared decision-making intuitively, they really varied in the understanding and the interpretation. When the surgeons were interviewed, most of them understood shared decision-making to really be a conversation about risk and benefits. It's obviously much more nuanced than that. It was clearly a gold mismatch. It was a driver of decisional regret between the patients and the surgeons in the desired outcomes where patients prioritized complete pain relief and surgeons prioritized functional improvement. You tell a patient you're going to T10 the pelvis or T4 the pelvis and the patient expects that after the operation in six months, they will be completely pain-free, which is not what you communicated to them, but that's what they were hearing. Then expectations. Surgeons felt that they frequently had to recalibrate patient expectations because patients came in with this preconceived notion that the surgeon was going to be curative. It was clearly a misalignment between surgeons and patients. In conclusion, I'd like to just leave you with a few statements here. Patients focused on chronological age and by that they felt like the older they got, they were running out of time to have an operation and there were fewer surgeons who were willing to perform the operations, while we as surgeons focused on physiological age, this concept of frailty. One of the things I do in clinic now is I really speak with patients about frailty and physiological aging and I try to put them at ease so they can truly hear and listen to some of the things I'm saying about and discussing about the operation. Patients tend to define success by absence of pain and from a surgeon's standpoint, we define success as functional improvement. You were unable to ambulate half a block, now you're ambulating two and a half miles, that's by all definition a successful operation, but you might still have some pain. However, from the patient's perspective, they define success really as the absence of pain. It's very important when you speak to patients to calibrate their expectations. This I thought was one of the most important findings that we're working on and we've just been funded from the NIH to do a subsequent study on this, that patients, the tools that they value really in the shared decision-making realm are tools that really highlight patients' experience with surgery, both positive and negative. The more you can align these tools with, match these tools with patient age demographics in which you'll be operating on, the more likely the patients are to listen and hear what the message that you are trying to convey. We focus as surgeons more on operative techniques and education and less so on these videos or these different communication strategies that use patients to communicate with other patients about the value of the operation or the perceived value and challenges that they had with the operation. This was a big gap that we identified that we are actively studying now. I'd like to thank the scientific committee again. Thank you all for listening. And I'll be happy to take any questions. Okay. Go ahead, Charlie. Okay, thanks, Mike. So we have a couple of questions for Dr. Phalings based on the study with regard to central cord syndrome. And one was, do we see any significant evidence in terms of a difference in patients with ASIA-B with central cord syndrome before or after 24 hours? Charlie, that was ASIA-B? Correct. Yeah, so, yeah. So there are patients, so traditionally the patient with central cord syndrome would be strictly either ASIA-C or D because they're incomplete motor. But I'll kind of interpret the question to mean kind of an ASIA-B type of a picture with a cervical injury without a fracture and sort of like a stenotic canal. And we published on this earlier this year in Lancet Neurology. And the bottom line is that those patients do improve better with early surgical intervention. All right, thanks, Mike. I guess there's actually a bevy of questions for you, but they kind of circulate around the concept. And I think no one studied this topic more than you have. Are there any patients that you see in your practice with central cord syndrome that you think wouldn't benefit from surgical decompression? Yeah, so Michael, I think the question here becomes how do we balance the benefits of early surgery versus the challenges of managing an aging population with medical comorbidities? And so this is a challenge that we're facing. I think the evidence that early surgical intervention benefits people with a spinal cord injury is now pretty much nailed. The evidence is really very strong. But now how do we translate this? And so what do you do, let's say, with a patient, particularly, let's say, with a milder injury, ASIA-D, who has got a whole host of medical comorbidities. In my own practice, I'm not gonna force the issue. I'm gonna stabilize that patient. If a patient is too frail medically and too high risk, there are a small number of patients where you may decline surgical intervention. But usually in those patients, I'll optimize the medical situation and then operate in a more sort of elective fashion, if you will, usually on the first hospitalization prior to rehab. All right, thanks, that's very, very helpful. Dr. Villiani, you spoke in your talk at some length of the variables that you chose to train on which ones had the most impact. But how did you decide a priori which variables to include in the training sets? Yeah, that's a good question. So, you know, we were limited by the variables that were in the national inpatient sample. So because we were trying to build an algorithm on one data set and see if it generalizes well to a larger sort of national cohort, the data set that we had internally at Mount Sinai had both sort of preoperative, perioperative and postoperative features, whereas we were limited to some of the perioperative features in the NIST data set. So we were limited to what we had with NIST and that was the limitation there. Now, the other thing that we did was we sort of algorithmically determined which variables had the most importance and those are also then, those are also fed in order to prevent confounding and other things like that. Okay, thanks. Charlie, did you have another question? You know, one quick question was what, this is for Dr. Vigliani as well. What was the cutoff in terms of what percent of the data did you use to test the, the algorithms and to test the machine learning algorithms? Did you use, do you use half of the data and test it in half the data? Maybe you answered that already. Yeah, so that's a good question as well. It typically varies. So it's a function, it's typically a function of the data set size. So if I have like a hundred thousand cases, then I feel comfortable splitting the data into 80% test, 80% train and 20% test. Given the data set that we had internally was 2,600, we, you know, and counter and balanced by the fact that we also had an external data set, the NIST data set that was much larger. We kept the train test split to 75% train, 25% test. And then we had that larger, about 950,000 cases from the nationally inpatient sample that were also used for external validation. Okay. So this is sort of a question slash comment to Dr. Odagwa, but I think a lot of your talk was centered around the idea of what patients are, what doctors are saying is not exactly what patients are hearing. And there was a line that was taught to me when I was very young that can be useful in that context, which is that there's no back pain that's so severe that it can't be made worse with surgery. I find if you say that in the office, it really gets people's attention. And then some of your study can be maybe more applicable. We're moving into tumors next. So I thought I'd give the last question to Dr. Yolku. Do you feel like the research that you're doing is building a case that radiation therapy might be first-line treatment for Cardoma rather than unblocked resection? Oh, no, definitely not. So our aim was sort of to show it can be helpful if patient cannot undergo unblocked resection or if it can be an adjuvant therapy or any option for a recurrence, for example, as a supportive measure. Got it. All right, well, my congratulations and thanks to all the speakers. That was really a very, very educational session. Thanks to everyone virtually who's still with us and sticking it out. And I think we have, in fact, saved the best for last with really a stellar lineup talking about metastatic spinal cord compression. So why don't we get those speakers up next? Hi, folks, can you hear me? I think I'm gonna go ahead. Thank you so much. So on behalf of the double NSCNS spine section, thank you so much for including me. And it's gonna be me and Dean and John and the three of us partners in crime are gonna talk about some challenges in metastatic disease management. And I'm gonna approach this from an instability point of view. And what I'd like to avoid in this talk is showing some crazy case where we did some heroic thing and that's how you should do it. But actually saying the majority of what we do is like seems simple, but then we look at it and we're like, it's not so simple. And I hope this is helpful, at least in terms of some of the things I've done to help other people out there. So here's some disclosures I don't think are super relevant here. You know, what guides us, whether you have gnomes or some other framework, there's all these things that guide us, neurological deficit, instability, local tumor control, deformity and satisfaction. We just heard a little bit about satisfaction and what the patient hears and what the patient wants. And that's a little bit beyond the scope of this talk. Really, I'm just gonna talk about instability today. I know it seems like a gestalt, but I'll tell you, it's not as simple as we think. And I'd love to talk about how my framework has changed and maybe it helps some people out there. So I think the first question and maybe the only question for a lot of people is, is the spine inherently unstable? Is it unstable? Do I need to instrument? Do I need to brace? Do I need to put cement in, what have you? Percutaneous, open, doesn't matter. And that's a question we've been challenged with for years, but I would argue there's actually more nuance to this. And the nuance is, will the stability worsen? And can we predict that? And will our treatment affect it? In other words, you can tell me some advice you have today with my relationship or my job, but really what I wanna know is, what about tomorrow and the next day? And I think patients wanna know that too. And the SIN score was created years ago with this overarching comment saying, hey, loss of spinal integrity as a result of tumor leads to either severe pain, deformity or neurological compromise. And no one's gonna disagree with that. No one's gonna say, of course, of course. If it's unstable, at least one of these three, we think it's unstable. But that doesn't necessarily give us the answer for that patient in clinic. And so this was created, the SIN category was created to think about, or the criteria, the score was created, I should say, based on just a bunch of people getting together and thinking this is what's important. And it was location, pain, all these different things to try to give you at least a sense that you're thinking about everything. And then add up the points, more points is worse. Unfortunately, when we created this, we created this potentially unstable area. And that's, I'm gonna be honest, is where I'm challenged. And there's been papers coming out that if you get 10 or above, you're probably unstable. If you're nine or below, you're probably unstable. But even then, what if you're nine and a half? I mean, so it becomes a little bit that that middle gray area, just like in life, is hard. And this was authorized. I mean, this was validated by radiologists, radiation oncologists and such. So I wanted to show you a couple of cases. Here's a lady, 60-year-old with cancer, minimal non-mechanical pain, a lesion of S1. And it looks mostly intact. In her transit from the place she was seeing me, seeing someone else who said just radiate, to seeing me, she said, all of a sudden I got really bad pain. And she had mechanical radiculopathy, meaning that when she moved, she had terrible leg pain. It wasn't back pain, it was terrible leg pain. She's got this old spondy. And the question is, is this unstable? Severe mechanical radiculopathy. Every time she moves, it hurts. And if you're not sure, if you say, well, you know, sacrum, I don't know, the pelvis is there, the alis seem intact. I don't know. This is, I thought, a more stable area. You can use this criteria and say, well, let's start adding up the points. When you look at these points, this patient's unstable. And so this is the first thing. This is, if you think you need help in any way, other than talking to each other and talking to friends, you can use this. This is on apps now. You can look up this on the internet and you can say, you know what, this helped me a lot. Now I feel like I've got some objective data, maybe some objective data saying it's unstable. And this lady underwent radiation after she had this stabilized. And she's had multiple metastatic lesions in her body, but this has not been an issue and that has not recurred. However, let's now think about number two. Let's think about, can we predict the future? Here's a 35-year-old history of breast cancer, some upper non-mechanical thoracic pain. So she's got tumor pain. She's got tumor pain in between her shoulder blades that nags her at night. Intact, only bony lesions. And this is her image. Excuse me one second. And so she's got this MRI scan with this lesion here. And you say, okay. This patient is not seen by a surgeon, but seen by a radiation oncologist. And they look at this and they say, you know, this looks pretty stable. We don't hear about this to surgeons. And you know, one side of the pedicle is fractured, but it's in the thoracic spine. And if you sit there and say, okay, Dan, this is exactly what we're talking about. Is this unstable? This patient's got no mechanical pain. I'm gonna operate on this patient. Maybe they can zap this. They think they can zap this with the radiation alone. Maybe I don't need to say, but these are the ones that I'm challenged by. Now, if you use the SIN score, you might actually say, lo and behold, potentially unstable. So this is where I think we live, right? This is this idea of going, thanks so much for the cute little scoring system, Shuba, for giving this to me and then showing me today. I still am confused because the question mark is in the middle. And so I think what I've learned and what I've been struggling with for that matter, I'd love to hear your input from everyone else, is how do we make this less gray in the middle? And one thing I think I've started asking is, what is this patient gonna go through in the next several months? So this patient underwent high dose radiation, no cement, no surgery, and about two months later presents with this. And you can sit there and say, was this predictive? I mean, well, the whole body was gone. She was a little osteoporotic to begin with. Then they zapped high dose to that body. And now she's got severe mechanical neck pain, maybe some pain to the upper arm or upper chest, and she's in excruciating pain. And you sit there and say, well, we thought it was a potential unstable. I guess we're not wrong. But then we went and zapped it, and she was a vigorous, active person. So when you sit there and say, well, is this unstable? Yeah, it is, based on this scoring system. So what I'm leading to is the fact that we can't look at this as its own picture in itself. We have to look at this in the big scheme of things, saying, what are we gonna do? And I think the next scoring system is gonna start putting this into place. Are they stable now? Yeah, but you're gonna do a laminectomy, or you're gonna do ablation, or you're gonna let the patient go downhill ski. And so this is gonna start being more dynamic. And so this patient had this operation. We have studies, and this big study's showing that up to 15% almost of patients who get radiated for meds can have a vertebral column fracture. So does that mean maybe this last lady, maybe we don't operate in the first place, but maybe for this lesion, or the lesion before, I should say, we put cement in, maybe this is a cement in radiation. So just thinking about what to do is helpful because we wanna avoid these. The final thing I'll say in the last couple of minutes here is this case. This is a thoracic case, a patient with known myeloma, okay, myeloma, and I don't have the first MRI that shows some in the body, some in the body in the upper thoracic, I'll show it to you in a second what I do have, but there's epidural disease. And this patient comes in with some high-grade compression and a little bit of numbness in the legs, and the surgeons who see her say, you know, this looks terrible, and let's take you to the operating room, let's do decompression, and we'll take care of it. And it's a myeloma, and we'll radiate it afterwards, but, you know, we'll save this patient. And maybe there was some high fives, you know, slapping some high fives in the case going, man, another job done well where we do a decompressive surgery. But remember, myeloma patients can sometimes be put in remission for years and years. So now you've got a patient with probably some osteoporosis, some little bit of tumor involved, and this patient undergoes a laminectomy over a kyphotic part of the spine and undergoes radiation. So I don't have the sins of the original MRI, and I apologize for that, but the idea is is that, yeah, this patient probably can benefit from a laminectomy, or definitely will benefit from a laminectomy, but are we putting them at risk? Are they unstable at that time? No, but once you do a laminectomy and radiate someone with osteoporosis, maybe it's different. So this patient, after about two months of radiation, presented like this, and the colleague who did the surgery said, Dan, you're the tumor guy, here you go. And I went, no, this is a deformity operation. There's no tumor left, but now it's a much harder operation, right? So can we think about what we're gonna do? And of course, what I ended up doing, here's the CT scan, is I ended up doing a full chropectomy. You say, holy smokes, Shiva, you make everything bigger than it has to be. And I go, no, no, no, but this is a patient who might live a long, long time and is gonna have potentially long-term control of her tumor, and she's gonna have a deformity and gonna have compression. So we went big, and this was part of the idea. Maybe we didn't have to do all this the first time, but the idea of when you start looking at sins or something like it, think about not just what the stability is at that time, but where that patient is gonna be or what we're gonna do. So if you're gonna do a laminectomy, if you're gonna do a minimally, I'm only gonna take one facet, but meanwhile, the other facet is destroyed with tumor, maybe they look stable on that sin score with only one facet gone. But if you go in and do a small little thing and you take the other facet or other post-treatments, think about that. I think that's coming down the pike. A lot of us are looking at that, saying we need to help this gray area going forward. So in summary, we have to decide if things are inherently unstable. You have your colleagues, clinical, both radiographic people and mechanical pain, getting input from your radiologist and maybe doing this sin score. The second thing is, will it worsen? Is this patient gonna challenge this? And do I have to be a little more aggressive than just what this one point in time shows? And then finally, are we intervening and will our intervention, a decompression, a cement or something, make it actually worse? And so add that in, dial that into your recommendation. And so with that, I'll end and thank you for your time. Great, thank you everyone. And I'd like to thank the section, Mike and Charlie for having me here today to be part of this session. So that was a great talk by Dan as always. And what I'm gonna do is launch into talking about cord compression and really focusing on this topic of frailty. I know earlier in this afternoon, we heard a really great discussion on frailty and scoliosis and deformity surgery, and we'll launch into that here. Here are my disclosures, which are not related. So when we think about spine metastases, I think Dan really laid down the foundation here in that it's really hard to decide who to operate on. I think we have better tools now to think of surgery and that surgery decision-making framework, whether you use gnomes or some other paradigm. And these can be very helpful, but there's clearly a complex interplay of prognostic factors that impact survival and also complications after surgery. And unfortunately, in most medical oncology randomized controlled trials of investigational drugs or new immunotherapies, patients with bony and spine metastases are often excluded or underrepresented in these trials. So we don't really have a good sense of what their response is gonna be with those type of treatments. And when we look at gnomes, I think now we have really good ways to think about the neurologic status of the patient, the oncologic status, as well as mechanical with sins and instability, as Dan just mentioned. But what about that systemic portion, that S part of the gnomes paradigm? And, you know, should we consider things like sarcopenia, frailty, malnutrition and the impact of systemic therapies in our surgical decision-making process? This is a challenge that we're faced with today. Now with PATCHEL, when this came out in 2005, this was really a game changer, I think, for many surgeons because at that time, radiation therapy was really the tool used to treat metastases, whether you had cord compression, pathological fracture or something in the spinal column. And it really gave surgeons some evidence that there is a role with modern instrumentation techniques to stabilize and decompress the spine. And we saw that in the study that even patients who are randomized to radiation, those who crossed over to surgery, three actually regained ambulation. So this armed us with some data to justify intervening as surgeons. But, and this predated sins, but following PATCHEL alone is not sufficient for surgical decision-making as I'll show in a case example. When I was a resident and fellow, I used to, you know, going to meetings like this, this is a slide I'd see up on the screen and demonstrating what the survival is typically in these situations, depending on histology. And I would argue that now a lot of these survival curves and timelines have changed, especially with new molecular targets and immunotherapies specifically for lung, melanoma and renal cell. We know that that survival has changed and that makes it harder for us to try to decide on who to operate and when. So we have trouble with the yes part. Can we predict survival complications, the benefit of surgery? Is it worth it to these patients who we know have terminal illnesses and can we improve their quality of life? As Dan mentioned, the SIN score is very helpful. There are some limitations, as he mentioned, perhaps in the intermediate state, intermediate zone and maybe for cervical spine, potentially modifiers are needed. And we know that surgery is not so easy for patients. Complications are high, readmissions are high and a number of studies have demonstrated this. So we know that we can, we have all these tools now with navigation, fenestrated screws, a lot of different technologies to intervene, but the major consideration is morbidity. So here's a case. This is a 67 year old who I treated with a new diagnosis of lung cancer, metastatic to the spine with a T12 pathological fracture, as you can see here, high grade epidural spinal cord compression, essentially bed bound because of severe pain. She could not even turn side to side or get up. And neurologically, the legs were very strong, but she couldn't ambulate because she couldn't simply get up from the supine position. Had no previous radiation. She had EGFR positive and had molecular markers as noted here with PD-L1. And when we think of stability, this is a high, highly unstable pathological fracture, score of 14. And as you can see here on the axial sections, this is high-grade circumferential epidural spinal cord compression with a lytic component and destruction of the posterior elements. But the question really was, what is this patient's survival and is it worth to intervene? So based on the N, the O, and the M, we have justification to operate. Based on the PATCHEL data, we have justification to operate. But what about this S component? And do we just defer to our medical oncologist to ask them, what's the survival and what's your cutoff? Is it the dogma of three months, six months, a year? I think everyone watching this or thinking about this has some likely arbitrary number that they think about as their cutoff. And when we talked to the medical oncologist, they said this patient can live months to years if we can improve their functional status, address their severe mechanical pain, because this patient has a favorable molecular status. And why is that important? This is just one example of how in oncology, this is not something that I think most of us pay that much attention to, but in oncology, it's a changing landscape. All these different molecular targets, receptors, it's really changing how patients are being treated. But as of today, we're not really integrating these molecular subtypes or this data in our surgical decision-making process. And it's a good question to think about and also to study, will this, can we stratify outcomes based on these receptor types? And we've done some work on that with renal cell, but with lung, it's tough. So here's this patient. So preoperatively, we had this multi-ID discussion with a number of different teams in the hospital. And you can see this is an x-ray, preoperative x-ray demonstrating that the right lung field is essentially washed out. This patient's on six liters of oxygen, saturating above 90%, but on six liters, is a lifelong smoker, has really minimal other sites of disease other than this primary lung lesion and the metastases at T12. And so the question is, what do we do here? Well, I ended up taking this patient for separation surgery, and in 90 minutes or so took this patient posterior approach decompression stabilization. Everything went fine, about 200, 300 CCs of blood loss. But then within 30 minutes of extubation, that patient needed to be re-intubated again because she could not maintain her saturations. Patient went to the intensive care unit, and before surgery, we had discussed, and the patient had made it very clear they did not want to be resuscitated or intubated if it was deemed that they could not be successfully extubated. So the intensive care unit tried their best to extubate and maximize the patient's pulmonary function status over the course of two days, but eventually the patient was not weanable from the ventilator, extubated, and passed away. So this is the patient that we thought that we would give every chance for to get to immunotherapy as an outpatient, but eventually succumbed because of the respiratory compromise. So in the last minute or so, I'm just gonna talk about how we study this and what we're gonna learn and apply this to the future. So we studied this. Several years ago, Dan Shuba published a really great study in world neurosurgery coming up with a metastatic frailty index. And we studied this with our cohort of patients at the MGH looking at what does frailty mean? We took our study population of almost 500 patients treated for metastases, applied logistic regression, as well as machine learning analytics to this cohort, and basically found that we did not see really a strong association between the frailty status or determination as defined by that paper and the actual outcome. And so the question really is how are we defining frailty and what are those components that we need to include in that definition? But it turns out that no matter how you look at it, oftentimes malnutrition and chronic lung disease went to the top in terms of factors that were really driving survival. And in my last several slides, I'm just going to point out why is that important? Well, it's important because most of the cited prognostic algorithms in spine oncology are the Tokuhashi and Tomita scores. And when you look at a lot of these historical prognostic algorithms, they don't include things like laboratory results. They're not considering the status of their systemic therapies. They're really looking at the number of bone metastases, visceral metastases. And there are a lot of these algorithms. There's Tokuhashi, modified Bauer, Tomita, Katagiri, the list goes on and on and we've developed some here in Boston as well. But the bottom line is as we look forward and we think about other biomarkers and other ways to predict survival to help us make decision-making for these patients, should we be looking closer at frailty, sarcopenia, body composition, visceral fat indices, the nutritional status of that patient and also integrating molecular markers for these patients. So to wrap up, I think this really highlights the challenges of studying, defining and quantifying what frailty is in this patient population. And we should be considering these other factors that are objective and reproducible. And we really need to study this further with collaborative data going forward. Thank you very much. Great. I'd like to thank the NNS and the spine section for giving me this opportunity to speak. I'm going to change gears a little bit and talk about a case and complications specifically to the case. This is again, not a PACTL criteria case. These are my disclosures. This is a 55-year-old male who presents with a history of thyroid cancer, known thyroid cancer for seven years. He's been treated with radioactive iodine, SBRT to a lung med. A PET-CT showed he had a T7 lesion, but he also had some smaller cervical lumbar lesions. He's very high functioning, still works as an executive. The MRI was obtained and the oncologist and radiation oncologist sent the patient to the emergency room based upon the MRI findings. When you see the patient downstairs in the ER, he's got five out of five strength. He has no sensory problems, normal reflexes, and he's got a little bit of pain over T7. Here you go. Take a look at the MRI. You can see the MRI, he doesn't really have any cord compression. He's got epidural disease and he's got canal compromise, but he's got no cord compression by T2. You can see the T1 with GAD, again, consistent with metastatic tumor at T7. Then you order a CT scan and you can see on sagittal and coronal reconstruction, he's got significant lytic component here, and you can see this in the coronal scans that significant amount of his bone is also destroyed here. You can see the axial cut. You can see the much more pronounced to the right side compared to the left side. Then you get upright films to look at his alignment to make sure he doesn't have a severe cathartic deformity or coronal malalignment, and he does not, he looks really good. Again, this patient does not meet the Patrick criteria. He does not have cord compression, he has pain. But again, the question at this point becomes, well, should we operate on this patient? In these situations, the decision-making really comes down to a multidisciplinary approach. In this case, it was really important to talk with his oncologist, to talk to his radiation oncologist. Number 1, survival with the oncologist. Number 2, any type of systemic therapy, whether it's immunotherapy, chemotherapy, whatever it is, is there systemic therapy that will take care of this tumor? Then talk to the radiation oncologist. Is this a tumor that you can deliver enough dose to with either external beam or SBRT? If the answer to both of these is no, and the patient really does not want to go paralyzed and has pretty good survival, then I think surgery is a very reasonable option, especially in this case. The patient is not grossly unstable, he does not meet Patrick criteria. But again, with a multidisciplinary approach, talking about the tumor, talking about all the options to treat this thing, surgery is probably a reasonable thing. Once you have that discussion with oncology and radiation oncology, then you talk to the patient and say, this is what we all suggest, this is the recommendation and these are your options. Once you do that, then this patient wanted to go ahead and proceed with surgery. We did this all posterior, T7 costo-transversectomy all posterior to above, three below the T8 vertebral body looked a little questionable, so we did three below. Let's talk about the complications. As we do the screws with navigation, everything goes great, everything's fine, motors are stable, and then we scan the patient. You look at this screw here, T5, it's not totally in, and you can see laterally it's out. Now, some people probably would say, well, why don't you just leave it? It's probably not going to be a big deal, but we repositioned it. We repositioned it with a freehand pedicle probe. After you pull out the pedicle probe, you see CSS. You can see the red arrow pointing to the track where that freehanded pedicle probe went, trying to medialize this screw. Again, what was the complication? What do we do? We basically tampened out of this with putting the screw in. There was discussion of doing a full laminectomy, identifying the leak, get the microscope out, try to find it and close it, which is not an unreasonable thing to do. But we said, let's see what happens if we put the screw in. If we really have to do that, and if it continues to leak and we see CSF coming out, then we can do that second step. But again, this is basically repositioning the screw. So again, CSF leak, complication happens, didn't really think much of it, and we carried on with the surgery. So now we're going into the tumor, and again, the tumor is bleeding, massive, massive branch laterally, which is clearly a branch off the segmental. So we all put patties on it, we can't bipolar, you can't see the vessel. When you tampen out, it's one of those tumors where you tampen out it, you've got three sets of hands in the wound, and everybody's tampen outing it, but the blood still keeps coming out, the wound keeps swelling up. So then what we do is we go in and we take out the patty, and we take out the patty, try to bipolar, we can't bipolar, you can't see it. Again, the patient is bleeding even with the tamping out. So at this point, here's the strategy that we use to try to stop this bleeding. If you can't get that segmental, because sometimes you can get the segmental with the bipolar, if you take out the rib and go lateral to the vertebral body, you can actually find that segmental and bipolar the source of the bleeding. In this case, we couldn't. So our strategy to get this bleeding under control was everybody held the patties there, and we called oxidized regenerated cellulose. It's a very common brand name, which I wouldn't name here, but it's been used for a long time. And have the nurse cut them up into multiple, multiple pieces, and you get ready. And so you sequentially will get these pieces ready. And what we did was we pulled off all the patties, two suckers going, find the source of the bleeding, get the cellulose in one at a time, one at a time, one at a time. You put multiple, multiple patties on top, and then you put the cellulose pieces on top, and then you put the patties on top of that, and you hold for five minutes. So what happens is that as you do that, not only does the cellulose create a clot, but it also allows you to pull off the patties and leave the cellulose on top of the bleeding point. And that's what we did. We put the cellulose on and the bleeding stopped. So we keep going, do the surgery. And then during the costotransversectomy, we said, oh, it looks like we may have gotten into the pleura. We flooded the field with irrigation. Valsalva was performed. No, there are no bubbles. We thought it was fine. And then when everything is ready, we're starting to close, and all of a sudden we see bubbles. So when you see the bubbles, there's gonna be a pleura wreck. The minute you see that, there's a few things you wanna do. So one is you wanna put a drain in there if you can, and usually you can. We just use a 19 French channel drain thread all the way to the apex to get rid of the pneumothorax. Anesthesia performs positive pressure ventilation till the fascia's closed. It's really important to drain this, number one, to avoid a pneumothorax. Then you have to have the patient have a chest tube put in postoperatively, which is very painful, or even a hemothorax from all the bleeding that goes into the chest. So if you do know you've got a pleura wreck, you can just take care of it right there. You don't need a formal 28 French chest tube or 32 French chest tube to go in through the intercostal space. You can put a drain right up into the apex and thread it all the way into the apex of the lung. And you can see right here, this is the drain we threaded all the way up, and you can see right here, and you can see that the lung is nice and re-expanded after the surgery. Give it about two or three days when the drainage output comes down, pull it out, and there's no pneumothorax. So in this guy, patient was great. He did great. He was super happy, five out of five, no headache, no shortness of breath, and he went home. So the important thing is that we manage these complications. We all get complications, right? The important thing is that we try to get through these cases and we manage each one, and we have to have strategies for each one. CSF leak, put the screw into the thing, into the hole, observe. Excessive bleeding. Have strategies for tamponading if you can't bipolar. Pleural violation, you see it, you've identified it, make sure you address it. Don't just close and say, let's just get a chest x-ray and see what happens. Address it if you're right there. So finally, in conclusion, as we all know the old adage, if a surgeon says they have no complications, they're either not doing surgery or they're lying. We all get complications. So the key is to manage it, address it, explain the risks preoperatively to the patient. Thank you very much. So Dan, Dean, and John, thank you so much for those very illustrative cases. I think these cases really shed light on the fact that spine surgery, and especially in the cases of tumors, is not straightforward and a lot of extra thought needs to go into each individual case. It's hard to standardize and generalize and have each case fit into certain algorithms. So you really have to look at each one individually and troubleshoot things individually. And I greatly appreciate your presentations. So all in all, I wanna thank all of our speakers for this tremendous session. I know this was a virtual session, so it wasn't as easy to interact, but we did our best using some of the question and answer features through Slido. And I wish you all a great rest of the summer, and we'll hope to see you in person next year in Philadelphia.

neurosurgery

AANS Virtual Annual Meeting

interactive event

Continuing Medical Education

acute spinal cord injury

neurological protection

rehabilitation

cervical arthroplasty

degenerative spondylolisthesis

randomized controlled trials

machine learning

prolonged length of stay

non-fusion spine surgery

spinal metastatic tumors

interdisciplinary collaboration