Computer generated transcript

Warning!
The following transcript was generated automatically from the content and has not been checked or corrected manually.

Klen Grover, I'm a healthcare scientist in neurophysiology. Um and one of my special interests is sort of doing the neuro monitoring during the spinal deformity surgery. So, who are the neuromonitoring team? What do we do? Um So we sit um in the corner of theaters staring at the screen um for the whole operation. Um So what we doing and why? So the idea is to provide an early warning system. Um So we can alert the surgeon. So remedial action could be taken before the spinal cord is irreversibly damaged. Neuro monitoring was actually first available back in the late seventies. Um But at that time, it was only his sensory evoked potential known as the SCP. Um But that was quite a step up from when the surgeons had to rely on using a Stagno or a wake up test. Um The SCP monitoring can provide a more timely and less traumatic way of, of of checking spinal cord function. Um I couldn't find a little bit cartoon or something prone to. I had to just change it around a little bit. Um So over the years, monitoring has evolved um into multimodal monitoring and I'm gonna talk about a bit, little bit about each modality, our alert criteria and a few example cases. So let's first look at the SEP um that allows us to monitor sensory pathway function along the dorsal columns. So to record the SEP, we use an electrical stimulator alternately stimulate the peripheral nerve below the level of surgery and then we record the response above the level of surgery. Um The stimulus initiates nerve action potential along the peripheral nerve travels up the spinal thalamic tracts to the sensory cortex. Um And if we place electrodes at various points along that pathway, we can record the evoked potential. So we record sensory evoke potential is a sub cortically generated um response. Uh And we usually as a technical um sort of check, we also recall from the peripheral nerve. Now, all these responses are quite small. Um they're generally not visible amongst the background electrical activity from the brain. So we have to use a process called computer averaging and that averages out the response from the background noise. Um It's actually quite challenging to record the tiny potentials in, in the electrically hostile environment they operate in theater. Um So careful set up is required. I'm just gonna talk a little bit about the pros and cons of uh sensory evoke potential monitoring. So the PROzac, they're relatively straightforward to monitor um particularly the score of 8% is is is is reasonably easy to get the anesthetic regime only requires minimal alteration. Uh, the SCP stimulus, it only causes a foot twitch so it doesn't interfere with, with surgery. Um, and any global spinal cord injuries are, are likely to show or expected to show changes. Um, but the cons are, I've already mentioned, they're quite small. So they require the computer averaging and that takes around about a minute to record the SCP from both legs. Um If the patient is too deeply anesthetized, it does still have an effect on the cortical S CPS, not as pronounced as for meps, which I'll talk about in a minute. Um but to help differentiate systemic changes from cord injury, we simultaneously monitor the pathway above the level of surgery. So in scoliosis surgery, that normally means recording an upper limb sep um with ischemic cord injury, um SC PS might not change immediately. So you may get a delay of, of uh 20 minutes um between injury and the SCP change. And it's really important to realize the SP is only directly monitoring dorsal colon function. So, when S CPS were first used to monitor the scoliosis surgery, it was assumed that a scoliosis correction um would be associated with a global spine injury. Um But by the 19 nineties reports were coming in and being published that patients had um were developing new postoperative motor deficits without any SCP change. And those reports were the main drivers to, to monitor um motor pathways. Um So how do we monitor motor pathways. Um So the other way around, we stimulate the brain um using a train of electrical pulses and then record the evoked muscle twitch um in various muscle groups. So, targeting the leg area, um but we also record an upper limb response um above the level of surgery as a control. Um these responses are a lot bigger in magnitude than the SCP. So they are easier to record. So just run through the pros and cons of mep monitoring. Um So we are uh monitoring the motor pathways propagated by the corticospinal tracts. They're large. So they're quick to obtain, we don't need the computer averaging. Um they can provide an earlier warning than scps of impending spinal cord injury proven to be quite responsive to ischemic cord injury. Um And if as long as you're recording the O PS, frequently enough, it helps you identify which surgical maneuver has caused the change. So because they're so quick to respond, you get a change in the mes. So just after you put a screw in, you say, OK, we'll take that screw out and I'll reverse that little bit of correction and see what happens. So the cons of uh me ep um monitoring. So they're particularly sensitive to changes in the depths of anesthesia. You have to use um a total intravenous um anesthesia propofol and that has to be carefully managed because even too deeper propofol will cause me suppression. Um too low BP can also affect the me. So we need the anesthetist to keep the patient as stable as possible. Um We should avoid using muscle relaxants, so we generally avoid them after initial intubation. Um The electrical stimulus is quite a large sort of um stimulus. So it has to stimulate the, the cortex but um through the skull. So you don't just get a limb twitch. Um you usually get a whole body twitch and that could interfere with surgery. Um The other issues that um MP S um under anesthesia generally quite variable. Um They can vary as much as 1000% from run to run. So that would make deciding upon the alert criterion um difficult. Um We use various techniques to optimize the stimulus um and having the patient hemodynamically and aesthetically stable, reduces that me p variability. Um Just briefly consider why meps can be more sensitive than DPS to ischemic treat. Um So you've got your anterior cord supplied by one major vessel, your anterior spinal artery. Um and any compromise to that one artery can cause ischemic damage to the anterior two thirds of the cord. And of course, that's where our motor tracks lie. Um infarction of the posterior artery is less common um because they have better collateral circulation. Um So show a few wavy lines. Um I mentioned how variable the meps can be. Um So this is a, a demonstration demonstration of that. So we've got me ps we've got tibialis anterior, we've got a response here. We hardly got anything from the abductor hallucis or the arm. So that's the um abductor digit mini. Um But because we're also recording the eeg, we can see that eeg has showed a burst suppression pattern. So you got period here, it's flat and then a burst. Now that would indicate to us that the patient is too deep. So we um speak to the knee, just see if they can lighten them up. And in this case, um they did indeed lighten them up. So the eeg is now continuous and our mes have popped back up to, to normal. Um So we, we're usually able to have an ongoing conversation with our Anestis um to prevent us getting to this stage. Um But our alert criteria must rule out any anesthetic causes. Um um for any change we get in the meps. So this is our typical monitoring screen. Um It's quite busy. We've got the meps at the top from the right and left legs. Um So the leg muscles here, we've also got the hand muscle at the top, the AD M um we're recording the eg continuously and then we've got the sensory um low lower limb sensories, right and left and upper limb sensories, right and left. Now, this set up follows national guidelines um for neuro monitoring during spinal deforms deformity surgery, of which I was a coauthor. So, let's consider the alert criteria for SCPS. So it's universally accepted for spinal deformity surgery. A 50% drop in amplitude of the responses is another that could be bilateral or unilateral. Um In this example, we've got faint lines here. So they were our baseline. So they were pre incision baselines. Um And then you, the side lines are the current trace and you can see that that's considerably less big than the uh original baseline. It's down a bit on the right as well, but the left is down and that's all channels apart from the um knee response. So this is just response from the peripheral nerve and that tells us that stimulator is OK. So we've got an inbuilt technical check there. Um The EEG is OK. So there's no technical or anesthetic change that could cause this abrupt um unilateral change. So, immediate alert could be issued. Um What about the MVP criteria? So we've already discussed me EP amplitude can be um inherently quite variable. So an exact alert criteria is difficult to define and it's multifactorial. Um but a universally accepted criterion would be disappearance or responses in one or more lasts muscles below the level of surgery. Another uh criteria would be abrupt reduction in amplitude. So typically quoted as 80%. Although we have to take other factors into consideration like depth of anesthesia, low BP because they can sometimes cause such a reduction. So the sort of lesser alert criteria more moderate alert criterion is only valid if the meps have been stable before the event. Er er, and that the upper limb mep control is unvaried. So this is uh fairly clear cut example of that. Um It's actually the same patient as the previous slide. So faint lines were a baseline. Um And then you can see here that the hand muscle response is fine, the right me ps are fine but from tips, anterior and abductor hallucis muscles just got a little blip. Um So they gone from really good size responses into virtually nothing. Um So the upper limb is OK. Um There was a um at the same time as this happened, it, it, it occurred during the correction. Um and in this case, the um correction was reversed and the response is recovered. And this is uh another example, um this time you've got a cascade of responses. So time is running down the screen here. Um You can see the responses from the me ep are OK here. Um The opposite side and left me ps are OK from arms and legs. These responses were OK. And then they suddenly disappear. So we've got loss of the right tibialis anterior mep and the abductor hallucis me ep it's a little blip every now and again from the H but it's, that's effectively absent. Um So that's a clearer work. It's unilateral, we've actually got an SCP change as well with this one. So the SEP is on the, on the right are down a little bit so 65%. So that reaches a lut criteria. Um Of note, in this case, there was a, a spike in the patient's BP at the same time as the neurophysiology alert. And we do sometimes see an autonomic response such as that um or a period of tachycardia. Uh And we consider it another indication that we're dealing with a real alert. So this one wasn't during the correction, it was um actually a slip with an osteotome which would breach the uh flavum liga ligamentum. So, what action would the, the surgeons take? So there's nothing to reverse in this one. So we wait. Um So we waited 10 minutes. There was no recovery responses. Um So you see, they're still virtually flat from T A and virtually nothing from, from A H. Um So they decided to abandon. Um So all the instrumentation was removed. Um And there was no recovery responses in sort of 30 minutes after the event. Um This patient um was quite interesting because clinically POSTOP, she was able to walk. Um She had some pain in the groin, in the right groin. She had um upgoing plantar reflex and brisk knee jerk reflex. But because she was able to walk, she was actually rescheduled for um a correction five days after the initial incident. Um But when we went to do her mes, we run these in the anesthetic room before they, they take her into the operating theater. Um, the meps from the right leg muscles are virtually absent. So we've got gastro tibialis, anterior and then A H we've got a little blip on a H, nothing on gastro or t we've got good hand muscle responses bilaterally and the left meps are ok. The SC PS recover. Um, but we've got nothing to monitor here. So, um, they had to abandon um that operation as well. Um This particular case, the ME P has remained absence four weeks later and only partially recovered sort of seven weeks later. Um So we assume that the contusion type injury had disrupted um conduction along the motor pathways, right? So sometimes the uh neuro monitoring change is not so straightforward. Um We developed a checklist to help us decide the appropriate course of action and this case demonstrates its use. So this was uh an A is correction. Um Me es had been stable. Um Correction was almost complete. Um but there was a sudden dip in the left abductor hallucis me ep so it's much more subtle than the previous examples. Um It didn't quite reach the alert level. I think it was 76% reduction, but it was a clear reduction. Mps on the other side were OK and the SES were ok. But because it occurred um during the correction up limbs, OK. Right side. OK. You've got still a cause of concern. Um So the surgeons were informed of, of this and, and we issued a yellow alert. So it's not a clear alert at this stage. Um But crucially, we want the uh surgeons to, to stop operating while we investigate possible causes. Um And see if there's any further deterioration. So, with surgery having been halted, we can run through the check. Um and we have a uncertain cause. We have an MP only change. Um And then we initiate our technical checks. Everything was ok. Um We can uh increase the stimulus to overcome any sort of anesthetic suppression. And we also ask the Anestis to, to make sure the patient is um stable uh with enough, high, enough BP and, and low enough uh propofol level. In this particular case, the meps continue to deteriorate. So we've got further reduction in meps here. So we're now into a territory of an orange alert. So we issue, issue an orange alert. So we run through our orange alert checklist. Um So has there been any other indicator of event? So in this case, there wasn't, if had it been, we'd have gone straight to a red alert checklist. So we there isn't. So we run through our orange alert checklist. Um We want the uh BP raised, so we want pre induction levels plus 10% to ensure that the spinal cord is, is well perfused. Um with there's various other other checks we're going through. Um and technical tricks as well, just increasing the, the stimulus a little bit more. Um And then we get to have, have the, uh, have the response is recovered to baseline. In this case, they didn't, they hadn't recovered to baseline. So we would have gone to the red alert, but in fact, they went down even more. So they deteriorate further. We go, go to red alert. Um, so the red alert is our highest um level of alert and a clear cut alert at this stage, we look through the checklist. We we've already covered the uh first section um because it was covered in the yellow and orange list, but sometimes we go straight to red. Um And in those cases, we want to document and we want the document to check. Our Anestis has the patient stable with the BP higher than normal to ensure the cords refused and mitigate for any cord injury. Um So was there a sa a reversible surgical maneuver in this case? Yes, because they were doing the correction. Um So the next action is to reverse um the correction. Um Three minutes later, the response is starting to improve and then a few minutes after that, they recovered back to baseline levels. So that's the next question is have, have the responses recovered to baseline pre alert levels. Yes. Um And then the surgeons need to decide what, what they're going to do. Should they consider um abandoning or or proceed with caution. Um In this particular case, the meps reversed quickly and had recovered to pre alert. So the surgeons did decide to continue with caution. So they did some of the correction applied it more slowly, didn't push for the perfect correction. Um And this patient was fine postoperative totally and the mes remained nice and stable to through to the end. Ok. So what I've talked about so far is monitoring spinal cord function. Um What about using the meps to monitor lumbar nerve roots? So you have to consider that each muscle we can recall from is innervated by more than one ovary. And there's considerable overall inter individ variability in the myotonic innervation that's shown in this slide. So if you take the L5 nerve root, for example, so it looks like tibialis anterior would be a good muscle to record from, to monitor the L5 nerve root. Um But if the L5 nerve root was severed, you could still record a fairly large motor rate potential propagated via L4 nerve root with some coming through the S one as well. So you, you, you can't rely on the me eps to identify a single nerve injury but saying that they do, they sometimes do show significant changes. So whilst we can't rely on them, it's still worth using them. So the percentage me ep drop will depend on the percentage innovation in that individual from that nerve root. So you can't use a rigid percentage amplitude change as a criteria and you're looking for a focal change um with partial loss of the me EP signal in the muscle groups which may be invited by that nerve root being monitored. Uh and then preservation of the NP amplitude in, in other muscles. Um This was an example of that. Um So spondylolisthesis um reduction, um L5 S one, our responses are nice and stable here and then we get a reduction in tibia and gastro um focal reduction. The other ones are all fine. So the um the the the change is abrupt, it's relevant to the situation. In this case, the reduction was partially released and you can see the responses pop back up um to pre alert levels. Um and this patient had no new POSTOP neurological deficit. So another modality we can consider is to record the ongoing electromyographic signal from the, the muscle and the nerve root to be monitored. Uh irritation of the nerve root triggers emg firing. Um Some types of, of firing patterns are more significant than others. So fast, rapid firing, known as neuro to emg firing is the most significant and most likely to be associated with an impending nerve root injury. Um EMG monitoring can be quite o overly sensitive and poorly specific for, for an over injury. Um But if you're monitoring both the ME P and the EMG, we increase the likelihood of picking up a change which could identify that impending neurological injury. Um This is an example of another spondylosis reduction. These um didn't change in this case. Um We've got rapid bursts of er m from the right gastro mu muscle here. Um The MG firing stopped, but unfortunately, this patient did have a, a right foot drop postoperatively. So that was another spondylolisthesis. Um There's one more mo modality we routinely use in our it And I just want to mention that um we call it the electrophysiology, pedicle integrity test um or short that to it. Um So that's a method where we can help identify medial breaches of the pedicle wall when you're inserting pedicle screws into thoracic and lumbar levels. Um We know a medial breach of the pedicle wall can have serious implications. So you're trying to avoid it. Um When the screws are inserted at cord levels, you may think our mep and sep monitoring would pick it up. Um And if you can choose the cord, you would indeed expect the meps to change quite abruptly. Um But if you take a look at this image, you've got the left T seven screw here which has breached seven millimeters into the canal. And in this case, there was no ME P or or SCP signal change. So luckily for the patient, the screw hadn't traumatized the cord. So conduction pathways were still intact and seps and meps were fine, but clearly that screw leaves the patient at risk. So once they're mobilizing, that malposition screw can easily damage the cord. Um And there are cases in the literature of delayed onset myelopathy even two years after uh scoliosis surgery. Um There's also evidence of reduced um pullout strength with, with malposition screws. Um So we want to try and avoid it. So, the technique um that we use is based on um published data from a guy called Professor Clancy. Um And we've been using this um in Norwich since 2014. So it's a, it's an adjunct to our surgeon's expertise and x-ray screening. Um but the, the method is the the surgeon, once he's formed, the tract holds a sterile ball tip probe against the medial wall of the pedicle. Whilst we incrementally increase the electrical current which passes along the co probe and stimulates any adjacent neuro neurological structures. So that could be the cord itself thoracic levels or at lower levels, lumbar lumbar roots as we increase the stimulus current, we're looking for a compound muscle action potential to be triggered in the relevant um leg muscles. So if the pedicle walls intact, the high impedance, cortical bone of the pedicle resists the passage of the current. Um and no response is recorded. So that would be a pass at low levels of current. If there's a breach, you actually have a window for the current to, to spread. Um And that will initiate response when we recall the leg muscle twitch. So we'll see that here. So the lower the, the mili aps at which you get a response, the more likelihood that there's a breach. Um This is unpublished data from our thoracic pedicle tests. We've performed over 3100 tests so far. Um The columns in this chart represent all of the thoracic pedicle tests with a threshold of less than 10 million. So 10 million A is what we use as a cut off to say less than 10 million out is a fail. Um So there's 245 tracks, just over 9% of all the tests. The orange sections represent the number of times a medial breach was confirmed and that was either by um palpation x-rays or a complications of postoperative ct. And there were 56 breaches in total, that's 19% of all the files. Um But if you notice that the lower the mili our threshold response was obtained that the, the percentage of confirmed breaches is much higher. So that um also confirms um the, the, the fact that the, the mili APTH the, the the more likelihood it is a breach. So a few considerations for, for, for monitoring, um just upper limb control is important for both SEP and me p avoiding the use of muscle relaxants. After during intubation is important, we can actually sort of check with a, a peripheral train of four on our system as well. To make sure that's worn off before we get baselines. Uh Our baselines are set with the responses you obtained in all four limbs prior to the first surgical incisions and as surgeons, you, you got to allow time for that. Ok. So there might be technical issues that need to be overcome. You just need to wait and let us let us get those baselines that's very important. Um The patient you're asking us to monitor, must have an intact conducting pathway between the sites of stimulation and recording. So, if spinal cord function is already compromised, then the signals could be of too poor quality to monitor under anesthesia. Um But in a patient who's stable aesthetically and hemodynamically, the combined use of SEP and me P monitoring during spine surgery has a high sensitivity and high specificity for detecting intra spinal cord injury. Um So, neuro monitoring isn't just a statement staring at the screen. Um interpretation of of our signals requires communication and collaboration between the surgeons and is tests and the neuro monitor team. Um We need the stable patient to be able to get stable monitoring signals. Um so we can increase the chances of reliably detecting changes. Um So bilateral communications with the surgery with the surgeons during the surgery is essential to the success of the monitoring um uh of the spinal cord and nerve roots. Thank you very much. That was great. Thank you very much. He and um quite enlightening of what's actually going on the other side of the, of the table while we're poking around in the spine. Um, uh, there's no questions, um, pending for that. Um, so, yeah, thank you very much for that. Ok. Nor, uh, just pause the recording, right.