I'm very grateful for the invitation to speak to you today in such a wonderful cause. My practice is based at the National Hospital for Neurology and Neurosurgery at Queens Square, which was established more than 155 years ago. Under its roof, Sir David Ferrier was one of the first to introduce the concepts of cortical localization of cerebral functions, and Sir Victor Horsley performed the first stimulation on non-human primates. Harvey Cushing, after spending some time with Charles Sherrington in Oxford, performed stimulation in patients, but as we heard yesterday, the modern era starts with Penfield and with George Hodgman, and what a great honor it was to meet him yesterday for the first time. We heard many techniques today, and I would like to make the point that not all of these techniques will be applicable to you. That depends on your skills, your training, your personality, the local setup, the equipment you have. So it is important to be selective on what you will take away from you, from this course. This is going to be a practical session, and I would like to explain to you how I do it, the way I learn it with trial, and fortunately not too many errors. First of all, my anesthetic technique. I started using a sleep-awake, a sleep or a sleep-awake-awake, because that was the way everybody else was doing it, but I found out that there was a number of problems. First of all, the emergence from anesthesia was very variable. The literature says it's between six and 13 minutes, but I found out that the patients can wake up, and this can be in four or five minutes, but up to the point that you can do a very detailed neuropsychological testing, could be anything between 20 minutes to half an hour. I was fortunate enough to have a very experienced neuroanesthetist, possibly the most experienced in the UK. He's now retired, and we've tried many ways to stop sedation sooner, but it was very unpredictable. Also, we had a number of very athletic patients that while waking up, they just took off the pins, and we lost registration. So I know this was the previous timeline of my anesthetic protocol, and now I have switched to awake throughout. The way we do it is we just use a little bit of propofol and remifentanil. This is called target controls infusion, and my anesthetist is using a formula based on the patient's age, sex, et cetera, and we use this only during local anesthesia. I use a circumferential scalp block. I don't use targeted block, and I'll show you in a minute how, and I think this is pretty standard regime, Marcane plus Xylocaine that pretty much all of us are using. I use 80 mLs roughly. That depends on the patient's weight, of course, but on average, it's about 80 mLs that go around the scalp, and about 20 mLs that go around the pin sides. This is a video that one of my residents took with his iPhone a few weeks ago. So the patient is right now is only on a small amount of remifentanil and propofol. I raise a little bit of a bleb with my needle. This is a small gauge needle, and then go circumferential around it using this bleb to advance the needle. So the actual pain the patient feels is very, very minimal. I put a lot of anesthetic around the pin sides, and this is important. When you start the operation and everything is draped, if the patient has pain, A, this is going to be very difficult to find out where it is, and B, most definitely you're going to desterilize the field. So this is the time to do it. And once you put the pins, then we stop all sedation, remifentanil. My equipment is pretty standard. I use the Ultima Stimulator, and the only dial I'm modifying is only the current output. The rest is always the same. So this is what I usually change, and I go between 2 and 5 mA. I very rarely go above 5. The difficulty some people have is they don't stimulate the correct part of the brain, and they keep cracking up the mA, and this is when we develop seizures. The only time I had seizures was when I first started doing this on an asleep patient trying to have a motor response, and the patient had generalized seizure. That was six years ago. I never had seizures since, but I never go above 4, very, very rarely above 5 mA. I use very few instruments. The ultrasonic aspirator, and it's very important for surgeons to familiarize themselves with the settings. I'm always surprised at how so many residents, and sometimes even qualified neurosurgeons, are not familiar with what tissue selector vibration means. This is paramount. I use always the suction at 20%, and I'll tell you in a minute why. And from all micro-instruments, I use only roton number 8. So this is all the instruments I use for gliomary section. My techniques are pretty standard. I always skeletonize large breathing veins. That goes without saying. And similar to mid-peri-care, I tend to put a surgical around the vein as a scaffold and put a bit of TCL. If you have a big resection, you can see that the vein starts with a large caliber, and as you proceed, this vein becomes smaller and smaller and stretch, and you think it's going to burst, but it doesn't. I found this way very reassuring. We never had any hematomas so far. I don't know if it is working or not, but the results possibly say it does work. I never diathermize any cortical arteries or veins. It's very tempting whenever you have a small cortical vessel as a reflex reaction to diathermize this. I would urge you to resist this temptation. Put a bit of Cezacel, Patien, work in another territory, or put a bit of Flossiel, and I promise you it will stop. Many of the deficits we have is not because of direct neural injury, but because we diathermize these arteries and veins, and it causes small cortical infarction that sometimes not even the MR can pick up. It's important to know your tracts and your anatomy. Imagine a cross-section through the body of the lateral ventricle. Are you confident to point where the corticospinal tract is in this section, or imagine a cross-section through the insula. Can you confidently point towards the IFOV? So it's very important to study the tracts and know exactly what you're going to do before you start the operation. A couple of categories of disease that are challenging for surgeons are the insula gliomas and the gliomas affecting the genu of the corpus callosum. And we know there are a couple of ways to attack these gliomas. The first one is to go transylvian. The other one is to go sapele. And I've tried both of them over the last few years, and I found that what works best for me in my hands is to go sapele, but also split the fissure so I know where the vessels are and how to protect them. And again, I've tried different ways to split the fissure from proximal to distal and from distal to proximal, but I realize that there's no need to go down to the sphenoid wing. There's no need to expose the ICA. So I start now from distal to proximal. And you can identify the MCA branches, M2 or M3, very early. And you can see how the PA just covers the vessels. Once you know where the vessel is, you can be quite radical with your ultrasonic aspirator. Of course, you have to check for function, but this is rather too complex to discuss now. I also use my suction always 20%, and you can remove all the tumor just gently touching the vessel. With 20%, it's virtually impossible to damage any vessels. Also, whenever you do this, be very careful of small quantities of CSF. Anytime you see a tiny bit of CSF, a big vessel is nearby. Just slow down and inspect the area before you induce a vascular injury. And this is the post-op image. We routinely remove the broca area. I don't think this is big news. So I think the fact that the scan preoperatively shows that the broca area is affected should never stop you at an attempt cortical resection. Another area that I sort of battle a little bit with the idea is to remove the gene of the corpus callosum. And we know in a vast majority of tumors affecting the frontal lobe, the gene is invariably affected. There isn't a huge amount in the literature. So I started doing this in a rather timid way. In the first few cases, I started removing a little bit. Then half of it, nothing happens. We always test these patients neurophysiologically before and after. And then I remove all of it, and again, nothing happened. And some of these patients are very high-performing individuals. This is from an international lawyer who is traveling all over the world for high-profile cases. So it's something you have to think very carefully before you do. This is an example of the same case. We are looking now from the left. You can see this is the frontal. This is the parietal. This is superior. This is the bottom. And you can see the corpus callosum. The pericallosal arteries, again, are approached sub-peerly. I cannot emphasize this enough. And once you know, you can see there the frontal horn. In these cases, always the frontal horn is much deeper and much smaller than you initially think. You can see the curve of the pericallosal arteries. The corpus callosum is a bit more white, although when infiltrated, it is hard to distinguish. And I would like to show you how I'm using the number eight. You just peel off the cortex from the peer matter. It's like peeling the skin of a banana. And I'm sure that Rotund didn't have this in mind when designing these instruments, but I found them ideal for low-grade gliomas. And you can see this is the free edge of the phalx. And you expose completely the peer with frontopolar and orbitofrontal arteries, pericallosal, callosomarginal. And once you know where these vessels are, it's very straightforward to remove the rest of the tumor and achieve complete resection. And this is the post-op scan with removal of the genome. And again, the patient had no deficits. Not all low-grade gliomas are the same. And some are quite easy to spot. They're quite hard, rubbery, darker side. But some are quite similar to the brain. So it's important to pay attention to the color, to the texture. I always use suction 20% because this gives me a reference of how normal and abnormal behaves. And I often say to my residents, I trust my hands more than I trust my eyes. Because the way the tissue feels in your suction or your instrument is really important, much more important than how it looks like. Sometimes cortex, which is enfolded, can be deceiving as a low-grade tumor. I tried to think of a way to convey this idea of how a tumor responds to suction. This is the best I came up with. So normal brain is usually aspirated just underneath your suction. But when you have a low-grade tumor, it just raises a small area of just a few millimeters. And this can be enough to tell you that this is abnormal or this is normal. Now I do a number of cases on intraoperative MRI scan. And I want to say that I'm not suggesting that this is the best way to do it. Actually, I do this in a very small number of my cases. I do this once or twice a month. But as these are new technologies we're learning about, I just thought to share with you some of these techniques. So we're using three components. We're using fMRI, DTI, and TMS that can be incorporated into the neuronavigation. And of course, we do it awake on iMRI. This is the iMRI at Queens Square. You can see the five Gauss line. And tomorrow at the practical seminar, Adabo Linners will have a chance to discuss a bit more about the iMRI. Outside the five Gauss line, we can use normal instruments. This is a case from a professional classic guitar player. We did mapping. This is a case from, I'll just show you this as an example, how the iMRI technologies come together. This is a case from a young lady with a small tumor. And I know it's usual in conference to show some massive tumors that occupy half of the brain. But sometimes small, deep-seated lesions also can be quite difficult to remove. This lady went to another neurosurgical unit in London. They said this is not probable and were referred to my service for second opinion. So this is how the TMS works. So the patient is seeing a number of images. You can see the camera. This is all stereotactic. You can see the camera array and the reflecting spheres. I don't think I actually have a sound here. Oh, there it is. And now you can see this is a figure of eight called for configuration. And then the computer can match the images before and after. This is how the computer matches before and after. And this is depicted as a color-coded heat map. And you can see this is the magnetic dipole that can be manipulated in both intensity and orientation. And this can go as a DICOM images to the neuro-navigation system. So this is how it works on the iMRI environment. So this is the same patient. I like to use clear steroid drapes because I want to see what the patient is doing and not just rely on communication from my team. And this is much less claustrophobic for the patient. I also like dim lights in theatre because it just helps me focus a bit better. You can see the five gauss line, microscope. We normally have some screens there, but whenever the patient is awake, we turn them off. The bed of the MRI scanner is on the right. And this is a screenshot from the neuro-navigation. Now in this particular area, I wanted to stimulate the posterior parietal area, the gas wound area. Whenever I stimulate, I actually stimulate with a microscope because this is how I'm going to be removing the tumour. So I want to have a very fixed mental image of where the positive and negative parts of the mapping are. And we now, I would like, I wanted to test repetition. And my team is just reading a line and the patient is repeating. So the first line is the patient is my team. The second line is the patient. And I just would like to show you the Catanese model of the Archaic Fasciculus, and this is the gas wound area which I'm now testing. And you see how the patient just stops. I test every area at least three times to have positive mapping. Always cold saline irrigation, and I like to use electrocorticography, but it's not always possible because of some difficulties you have with the national healthcare system. So if I have it available, I will use it, but I don't rely on this. And you see that I again hesitate. This was a very fluent young lady. And these cortical areas of positive mapping, these numbers can be pinned and can be projected into the neuronavigation. So although I'm looking at the screen here, I never rely on fMRI or DTI or TMS. I rely only on direct cortical stimulation. This is a screenshot of this, and you can see the blue line, this is the TMS, Archaic Fasciculus, the yellow, red is cortical spinal tract, and the numbers are the positive sides. So the patient is having a postoperative scan that showed complete resection, and she was completely intact. Now for the last couple of minutes, I would like to raise a question on two points. We all feel that there is a standard model of tracts that we stimulate, but actually is not that concrete. We have seven language tracts, superior, middle, inferior, longitudinal fasciculite, Archaic, ancenate, IFAW, and extreme capsule. And there's a huge discrepancy in the literature as to what is the exact course or function of these tracts. We have some models that do work well, but you have to be aware that these are not final. We have blunt fiber tractography, we have autoradiography, DTI, and I have to say I use DTI routinely, but we should not think that because we see a tract with a beautiful color, start, stop, and finish, that this tract actually exists. I won't go into this right now. And of course you have the direct cortical stimulation. Models of perisylvan connectivity different according to the group. The model of Archaic fasciculus, the most well-studied fasciculus in language, is still not, there's no agreement as to what is the exact shape or structure, and I don't have the time to go into that. The second caution is the plasticity, and we know that, we feel that because the lesion is a low-grade glioma, this induces plasticity through collateral networks. And this is true in low-grade gliomas, this is one of my cases where we remove the motor and sensory cords with no deficits. But look at this, this is a high-grade glioma, and again we resected the arm and hand area with no deficits. This is another high-grade glioma where we remove the broca area with no deficits. This is an area of metastasis where we remove the metastasis and a layer of abnormal tissue around it with no deficits. And we have a large series which are going to be submitted soon for publication. The last point I want to make is that one of the difficulties I had when I started doing this is that there's no agreement with regards to what tracts to stimulate and how to stimulate them. So for the last year, my team and I have been working to segment the brain topographically into approximately 60 parts, cortical and subcortical. I'll be discussing some of this on one of the breakfast seminars on Wednesday on low-grade glioma session. And what I would like to do, we're now working on a theoretical model, and over the next few months I would like to send this to both faculty and delegates for your input. And so once we finalize this model, then we have to validate this prospectively. And last, yesterday, there was a big publication in Nature where they published a semantic map of the brain in a very clever way, which we don't have the time to go through this now, but it turns out that there's a very predictable semantic map that is grouped according to category. For example, structures or houses, or if someone says words like wife, children, house, family, all these go to a very specific, all these words to the very specific part of the brain. And look at that. This is the right side of the brain. So there's a very extensive semantic map on the right-hand side. So this was published yesterday. Have a look at this publication, I think it's fantastic. And this gives us new insights into the brain function. So in conclusion, keep things simple, use a few instruments, know how they work, use technology very wisely, and contribute to standardized mapping. I would like to thank my team. We run a very successful course in London every year. It's for five days, and many of the faculty here are actually coming there. And in this course, you can see people like Mitch Berger, Lorenzo Bello, operate live. Hugh DeFolk show us many videos. It's a five-day course. Professor Ribas give us a whole day of anatomy lectures. I think they might send you an email with a special discount code as a thank you. And I would like to thank you for your attention.

neurosurgeon

anesthetic technique

circumferential scalp block

intraoperative MRI scans

neuro-navigation systems

segmenting the brain