So, for this breakfast seminar, we're going to discuss MIS approach selection in spinal deformity surgery. I'd like to thank my fellow, Brenton Pennycook, for his help putting these slides together. I do have disclosures, which are here. I've worked with several companies on spinal implants, notably Depew, Globus, and Stryker as consultant, and with Depew on some MIS platform issues. So why do we want to do less surgery for adult spinal deformity? The complication rates are high. The pseudarthrosis rates are problematic for these patients. So in order to try to figure out which ones can be done MIS, we created this algorithm back in 2014 through the study group, the ISSG MIS study group. And I'd like to thank all the co-authors who are listed here who helped us put together this first version. And what we see in this first version of this algorithm is that we were looking at 36-inch long cassette x-rays, and we were trying to identify if the patients were in class one, which means they could have an MIS decompression or just a one-level fusion, in class two, where they get some inner body fusion at the apex of the curve, or in class three, where they have a big open surgery. And I'm not going to spend too much time on this algorithm because this algorithm was redone last year. So MIS techniques are advancing. The algorithm needed to evolve. So we put this algorithm into JNS. And essentially, if I take you through this updated MIS deformity two, version two algorithm, what we see is that we, first we want to figure out if the patient has a fused or rigid spine. If they have a non-fused spine, there's no hardware in there from before, then we go look at their x-ray. If they have an SBA of less than six centimeters, a relatively normal pelvic tilt of under 25, LLPI mismatch that's very small, less than 10 degrees, coronal cob angle less than 20 degrees, these patients can end up in a class one MIS surgery where they just get a decompression only, or if they have spinal stasis, they just get a one-level fusion, and that's the far left side of this algorithm. And then there are more patients that we treat in the class two and class three, and how we end up in class two and three. It means that those patients have an LLPI mismatch between 10 and 30 degrees. They don't have too much thoracic kyphosis, and they don't have any thoracolumbar junctional kyphosis. So in those patients, they can have a multilevel MIS surgery, which is a class two, or if they need more lordosis correction, they might have something like an anterior column release or a mini open pedicle subtraction osteotomy, which was popularized by Michael Wang. And we can use hybrid open approaches in the back after doing an ACR in the front to try to avoid a PSO in some of these cases, so that's a class three. And that's new technology and new techniques that we're using in 2019 and 2020 that was not available back in 2014 when we made the original algorithm. But there are still cases on the far right of this algorithm where the patients have more than five levels of fusion involving L5S1, or more than 10 segments needing treatment, and those patients end up in the class four, meaning they need an open surgery with an open osteotomy just because they have to have that hardware revised in the dorsal area from their prior surgery. Now the question next becomes, if you're going to have to do these operations MIS, what kind of inner body approach are you going to use? MIS tends to use more inner body approaches than open surgery because we get the correction and the foraminal distraction, indirect stenosis correction from foraminal distraction using inner body techniques in MIS. So we did an analysis of a couple hundred cases from the minimally invasive subgroup of the ISSG and we found out that at L1-2 and L2-4, there was a preponderance of lateral approaches, especially if you wanted to get some lordosis. At L4-5, a lot of people were using T-LIF, but some people still did use a lateral. The T-LIF did not give very much lordosis, the lateral gave a bit more. At L5S1, if you really wanted lordosis, you went with an A-LIF, and if you didn't really need much lordosis, you could do a T-LIF and that's basically the approach selection algorithm that we came up with. I'd like to take you through some case examples to see how these things work in practice because algorithms are nice to talk about, but if we don't put them into real life practice, we may or may not figure out how to get them to work for us. So here's a first case example. So in this case, the patient was a class 2 in the new algorithm, and let's take you through this. This patient had this kind of an x-ray. The patient's LOPI is 12, so it's a little bit off. Sacral slope and pelvic tilt numbers are shown. The pelvic tilt is over 25. The patient does have a thoracolumbar scoliosis here, but primarily this patient has back pain and sciatica due to the L4-5 lobe from up down, L4-5 lobe from up down foraminal stenosis, and failed a lamino-foraminotomy already. So this patient doesn't have a global pain picture problem. He has a focal pain picture problem, referable to L4-5, where he's failed a decompression. We decided he doesn't need too much lordosis. He only needs a couple degrees, so we were able to do a T-LIF for him, and we decided to do that T-LIF awake for this patient with no general anesthetic. So the patient was brought to the operating room, was given a spinal anesthetic. We then put in a navigation arc into the pelvis, and this is not that particular patient, but it's a similar case. And once the navigation arc is placed, we don't have to do very much fluoro, so it makes the case move along very quickly. And we bring in the Intraoperative CT Scan and Register Nav, and then this is our room setup. So basically, the nav is at the bottom of the foot, and the surgeons are working looking at the nav, and we don't have to have fluoro. We don't really have to wear lead. So this is my partner, Catherine Miller, who helped me with this case. And so we're just trying to plot out where do we want to put the screws for this patient, and we can plot that out a lot with navigation. Again, the patient had a spinal anesthetic at L3 and is awake for the case. We plot where we want to position the screws. We use Liposomal Bupivacaine for this case in order to give long-acting analgesia. So we can limit the amount of post-op narcotics. And we then go ahead and put in the screw fixation, MIS, using navigation. And when we do that, we use a screw system that doesn't have any K-wire, so it goes fairly quickly. The Jamshidi is basically built into the screw head, and the screw goes over the small awl into the spine. So we put in our screws, and then we put in a tubular retractor, bring in the microscope, and do a T-lift and take off that whole facet joint. I use an expandable cage in this case, so we can get indirect foraminal decompression by expanding the cage. And then we put in the rod and lock it down in order to secure the T-lift. And so this patient had a one-level MIS awake T-lift using the T-lift at L4-5 based on the approach selection algorithm. And you can see the LPI is corrected to 7, it was 12, with just a one single level surgery. And the pelvic tilt is now at 26, it came down as well. The patient did really quite well, was discharged home in a day. So if we pick the right kind of patient, we pick the right kind of surgery with the algorithm, and you pick the right kind of inner body, we can have very nice results without too much complication. This is a next case. This patient had a class 2 surgery also. This patient is obese, has a weight of 300 pounds, and had had multiple prior spine operations in the past with an L4-5 LAMY in the past. And now is having back and leg pain and failed PT and steroid injections and is on narcotics. You can see that she has a 22 degree lumbar cob. She doesn't have an LLPI mismatch, but what she does have is a lot of up-down for amylostenosis in spite of having a laminectomy in the back at L4-L5. And you can see a pretty severe stenosis there at L2-3. She also has stenosis at L3-4. She has stenosis at L4-5 with a facet cyst on the left side. And then she has a CT scan where she has partial autofusion of L5-S1, a partial autofusion of L1-L2. And the intervening levels are all stenotic. So the question becomes, what do we do with this patient? Do we do an open posterior vision surgery to fixate and fuse? You could do that, but that's going to create quite a bit of blood loss. It's a redo, and the patient is pretty deep at 300 pounds. I had to put this patient through the algorithm too. And what we can see is we really want to open up L2-5, and we want to get a bit of lordosis, but we don't need a lot. What we really need is indirect for amyloid decompression because she has up-down for amylostenosis. So we chose to do laterals on this patient. And the beauty of doing a patient like this in a lateral position is the panus falls away. We did a lateral MIS surgery, and we did a presoas technique for her using navigation. My partner, Dean Chow, is very good at that maneuver, and we did that case together. So by putting three inner bodies in L2-L5, we were able to get for amyloid distraction with a fairly small incision in spite of the patient's size and stay out of the scar tissue in the back from her prior laminectomy. And then we did a posterior stage percutaneous screw rod fixation for her using navigation L2-S1, and then we did some MIS foraminotomies as well for additional foraminal decompression. We're able to get all the screws in MIS, staying out of the midline incision by making small paramedian incisions so we can stay away from her prior scar tissue. And you can see even iliac fixation may be placed MIS, and Mike Wang and I had a paper on that topic some years back showing how to do that particular technique. And then my partner, Sig Berven, and I have figured out that many of these cases, especially if they already have an autofusion L5-S1, don't need dual iliac screws. We got away with just one, which lowers the dissection, again, at the lumbosacral junction. And we percutaneously passed a rod, and you can see we corrected her scoliosis from 20 down to 2, and we got up-down foraminal distraction to correct her stenosis. And we didn't lose very much blood, and the patient was up and around fairly quickly after this kind of an operation. So there are some cases where we cannot do MIS surgery. These are cases like this one here. The patient has a multilevel posterior fusion down to the sacrum. We have to revise the hardware. I'm not going to be able to get this case done MIS. This patient is going to need an open PSO surgery to treat the stenosis and scoliosis and kyphosis, and we need to extend the prior hardware. So these kinds of cases, this is a class IV, cannot be done MIS. Trying to do this MIS predicts failure. So we do cases like this open, and we did a VCR for this patient. So over the years, there has been an increasing number of papers that my colleagues and I from the Minimally Invasive ISSG have put together. These are several of the key ones, which are looking at which patients are good candidates, what the complication rates are, when it works, when it doesn't, and we did try to analyze our failures as well, because we learned from them, and we learned from the re-ops, and we learned from the issues like having osteoporosis and subsidence and losing some of our correction over time. And what I can tell you is, having learned from all of that, is that there are some patients who can have minimally invasive correction of their deformity. To restore their sagittal balance is the key. You want to match their LLPI within 10 degrees and reduce their pelvic tilt. Many of these cases, which we used to have to do open, can now be salvaged MIS if we use new techniques like anterior column release, or release the anterior longitudinal ligament, or if we do an MIS PSO, or we use hyperlordotic ALIPS in the front, and then PERC screws in the back, we can avoid some of the big open PSOs that we've done in the past, which have had quite a bit of blood loss and morbidity. But those are not applicable to all patients. There are some patients who have dorsal hardware extending, especially down to L5-S1, who need that hardware revised. In those cases, we still can't get them done MIS. We have to open them up and do open osteotomies for them, or arterial column resection. So it's not a panacea for all. They're appropriate for selected patients, these kinds of MIS cases. And the algorithm to choose which approach to use is also quite very helpful in trying to figure out how you're going to plan the campaign to fix their deformity. Thank you very much for your attention. If you have any questions, I know because this is an online webinar, you can't ask me directly this year. So you can feel free to email me or contact AANS staff, and they will get the message to me. And I'd be happy to respond to you and help you through any of these cases where you might have questions or concerns before you might do it yourself. Thanks so much.

minimally invasive surgery

spinal deformity surgeries

algorithm

MIS techniques

case examples