 I'd like to start by just thanking Kathy for inviting me today and congratulating the Texas Heart Institute in the profusion school for 50 years. I actually remember, like it was yesterday, setting up the bypass room when I was a resident here and handing Kathy the bottles of amicare juice using a prime to pump. We were here probably at five in the morning, you know, it's plus or minus, but thank you again for having me. I have no financial disclosures today. What we'll do today is actually pick up where Ken left off yesterday and continue the story of cerebral auto regulation. I get to, I get the pleasure of doing the fun part which is the clinical outcomes and kind of the stuff we're doing today. So, we'll start out by talking about cerebral auto regulation and I'll actually frame it in terms of the clinical burden of delirium. Next, we'll talk about how we monitor cerebral auto regulation some of the neonatal monitoring that we've been doing and the current studies of outcomes and finally, we'll touch on a little bit of the next steps in a real time cerebral auto regulation monitor and some future perspective studies. So, it goes without saying that delirium has a large burden on our patient population in terms of morbidity and mortality studies have shown that it increases ICU length of stay. And it increases ICU class, but it's important to remember that delirium affects her patients beyond the operating room and beyond the hospital stay. It affects patients for years to come in terms of cognitive and functional decline. It was so important that the European Society of Anesthesiology created a task force in 2017, looking at preoperative risk factors of delirium. This included things such as advanced age, comorbidities, advanced ASA physical status, preoperative fasting and dehydration, electrolyte abnormalities and alcohol related disorders or drugs with anticholinergic effects. And the postoperative risk factors that this task force identified were side of surgery, so major surgery such as abdominal or cardiothoracic surgery, large intraoperative bleeding and prolonged duration of surgery. And finally postoperative risk factors included pain which is very near and dear to our hearts as anesthesiologists. The same task force then broke apart the components of a patient's hospital state, looking at preoperative, intraoperative and postoperative factors, and looking at the various stages of patients as they progress through their hospitalization. What catches my eye about their algorithm is that there's not much that we can do as anesthesiologists. What they talk about in terms of monitoring are monitoring the depth of anesthesia and avoiding too deep of anesthesia. And under the therapy options for anesthesiologists there's actually nothing there so it's an empty box. When we look at different ways to monitor the depth of anesthesia I'll touch on one kind of a large paper that was published in 2012. And this was a paper by Sester that described a concept called the triple low. The triple low was a phenomenon he described where there was low blood pressure, a low bis value and low minimum alveolar concentration of all tall anesthesia so a low Mac level. He found that with two components of the triple low here on the right, there was a two fold relative risk index and mortality. And when all three components of the triple low were identified in a patient there was a four fold risk of mortality, and this led to increase length of mortality and increased mortality. This was further studied in a prospective randomized and full trial where there was an alert system that would alert the care the care team when the triple low occurred. What was interesting in this study was that there was actually minimal to no intervention by the care team whenever they received alerts of the triple low, and these alerts did not actually affect any patient care or mortality. So I think this study kind of gives us an important lesson in terms of using clinical decision support tools and sometimes they may not lead to the outcomes we expect and perhaps there's things that we can pick up and learn so that we can design a better clinical support tools for us in the operating or the perioperative arena. So now I'm going to shift gears and talk a little bit about auto regulation. In case you miss Ken's talk yesterday, I'll give just a brief one or two minute of background on it and I'm not as smart as Ken so I'll probably do an elementary school level. But if you look at the graph here on the right, you can see that it represents cerebral blood flow as a function of cerebral profusion pressure. And this is a graph of our plot of the typical typical auto regulation curve. The center of this auto regulation curve highlighted in green is an area where there's constant cerebral blood flow. This is the area that we want to maintain our patients in because blood flow is kept constant. There are various mechanisms as to why this happens, such as myogenic control neurovascular coupling and pressure auto regulation. What we'll focus on today is that last component of pressure auto regulation. The left and right of that green flat horizontal line. Those are the, those are the parts of the graph that represent the outside limits of auto regulation outside of these, that green area highlighted in red, it's where the cerebral blood flow is passively dependent on cerebral confusion pressure. This is the area that we don't want our patients to be in. In this area small changes in blood pressure will be to cerebral schema on the left side of that graph or cerebral edema on the right side of that graph. Moreover, it's important to realize that the limits of auto regulation are quite variable among individuals. This is a study in adult congenital or no I'm sorry adult cardiac bypass patients, looking at the lower limit of auto regulation. In this study of over 200 patients, the lower limit of auto regulation was identified to be 66 millimeters of mercury, but you can see that there's a wide interportal range of 43 to 90 millimeters of mercury. Now yesterday, Ken talked about measuring cerebral auto regulation using the pressure reactivity index. This is essentially a correlation coefficient moving correlation coefficient that correlates the relationship between arterial blood pressure and intracernial pressure. In that bottom graph is one patients cerebral auto regulation graph where you've been blood pressures and as a relationship to their pressure reactivity index which is that moving correlation coefficient. So when auto regulation is intact, the PRX value is negative indicating intact auto regulation meaning that the arterial blood pressure and the intracranial pressure are moving in opposite directions. When auto regulation is impaired, the correlation coefficient becomes positive indicating that arterial blood pressure and intracranial pressure are moving in the same direction, meaning the vasculature is now reactive. Or I'm sorry, not reactive. Sorry, Ken's going to kill me. What Ken and Kathy did in the early 2000s was to create a way to measure cerebral auto rate auto regulation noninvasively using the hemoglobin volume index. So instead of inserting ICP catheters, you could actually use the nearest to as a surrogate of cerebral blood flow. So the nearest are used at a wavelength of 805 nanometers and this is a measure of the relative total hemoglobin in the brain. In the brain math, you can actually delineate cerebral auto regulation curves without or basically in noninvasive measures. So the question then becomes, we can now measure cerebral auto regulation or we believe we can. Does the interoperative management really matter. So this is a diagram indicating the various stages of a patient with congenital heart disease and you can see that from the very beginning. There are several factors that play a role in terms of neurodevelopmental outcomes, and I'd like to kind of consent to you that we actually have a very important role in defending the patients to have their maximal neurodevelopmental outcome. And they're essentially maximal potential. And one of the main factors that we can do is by controlling blood pressure. And I think we can. So, when we are below the lower limit of auto regulation, we basically expose the brain to ischemia, leading to bear a pair of intricately go Malaysia. And when we are above the upper limit of auto regulation we are exposing the brain to hyper profusion or so I'd like to start by presenting a study we actually did at Texas Children's Hospital, and we wanted to ask the question if we can measure auto regulation in neonatal heart surgery. And so this is a retrospective study we performed using our data collection system. We essentially harvested the system and collected all physical waveform data in neonates undergoing congenital heart surgery at Texas Children's Hospital from 2013 to 2018. The number of patients was 563 in this period. Now, as I mentioned earlier to delay to delineate cerebral auto regulation you need to signals the arterial blood pressure and near infrared suetroscope at the 805 nanometer wavelength. When we collected these data, we ended up with 169 patients that met our criteria to calculate cerebral auto regulation. We were able to delineate cerebral auto regulation in 145 of these patients so about 87 to 88% of patients. This was kind of the background of our patients, their mean age was 12.7 days, and you can see the rest of their diagnosis category and surgeries that they had in the cardiopipest room. Essentially what we did was calculate hemoglobin volume index using the methods that Ken and Kathy described in their paper. We utilized MATLAB to do it on a larger scale and stratified the data by the patient in their age by days. We then characterize the lower limits of auto regulation and then perform linear regression to look at how the relationship between the lower limit of auto regulation changes with age. So this is kind of applied of our results. Here on the X axis you can see blood pressure so it goes from 0 to 100 so that's the mean arterial blood pressure. On this vertical axis is the hemoglobin volume index so as you remember that's the measure of auto regulation where positive values indicate impaired auto regulation and we've shaded these values in red and negative values indicate intact auto regulation and we've shaded these as we shift the graph that you'll see a third axis which is the patient's age. So in the far back moving towards the back of the screen it's day of life zero all the way up to day of life 30. As we begin to retain this graph, you can see that we were able to create a cohort representation of auto regulation by age. So the trough that you see in blue is where the patients have intact auto regulation, and then the bars that you see in yellow and red are where blood pressures were the patients experience impaired auto regulation. And I know we had a question yesterday asking about how we do any gold blood pressures for patients in terms of their neonatal bypass goals and their map. And I think this graph kind of highlights two important concepts that I want to describe. So as we finish looking at the plot, I can now represent the graph in terms of their lower limit of auto regulation by day. And you can see that in this cohort from zero to 30 days of life, the lower limit of auto regulation ranges from 29 to 40 millimeters of mercury. We also see that each one day increase in age corresponded to about a point 32 millimeter mercury increase in their lower limit of auto regulation. But it's important to see that or to realize that the lower limit of auto regulation is quite variable. So when I look at this, this plot, it's very difficult to discern a basically a gold blood pressure for each patient. And when you factor in their congenital heart lesion, I think that adds additional complexity that needs to be considered when defining blood pressure goals and targets. When we look at the percent time below the lower limit of auto regulation, this is broken up by age on the x axis and then on the y axis you see the average time that each patient spent below their lower limit of auto regulation. You can see that overall in our patient cohort the average time sent below that lower limit of auto regulation was about 17%. So if you look at the severity of time below the lower limit of auto regulation, this is a measure in the literature, kind of described as the dose of impaired auto regulation and it's calculated by how many millimeters mercury below the lower limit of auto regulation you are multiplied by the amount of time you are below the lower limit of auto regulation. And what we see here when I look at when I draw a dotted line, which represents the mean dose of time of time below the lower limit of auto regulation. We can see that the patients who are younger, basically 15 days and less have basically more severe dose of time below the lower limit of auto regulation. So what it tells me is that the patients most vulnerable are those 15 days and less. So now let's shift gears and look at some adult studies. This is a randomized controlled trial in adult patients undergoing cardiopulmonary bypass and these are the results that study. This was conducted by Chuck Hogue and Charlie Brown, and they looked at a four to five year period and randomized patients undergoing adult heart surgery, basically cabbages and valve repairs into basically an auto regulation targeted group where they were able to visualize a real time monitor of auto regulation and treat the patient based on that monitor versus a standard of care group which is basically no auto regulation monitor. In their two groups, they had about 100 patients in each and as expected, the patients who had the auto regulation monitor. The patients experienced more doses of phenolaphrin. So essentially the perfusionist had phenolaphrin and was able to give phenolaphrin whenever the blood pressure was below that lower limit of auto regulation. And as expected, in this bottom row, you can see that the dose or the product of duration of time and map below the lower limit of auto really auto regulation was less and statistically significant in that auto regulation targeted group, meaning that they were capable in having those patients experience less time below the lower limit of auto regulation. So what did they find in their study. Well, in the standard of care group there was about a 50% incident rate of delirium which is about the typical finding that you see in the literature, and the auto regulation targeted group there was a 30% reduction in the incidence of delirium which is quite significant in this study. With the 50 monitoring data, they actually find that there's less end organ injury in the patient and these were not primary outcomes but very interesting to us, but essentially there was less acute kidney injury and less death in the hospital. And I'll skip the auto regulation monitoring in the ICU and this in the ICU, but now we'll move on to the kind of final part of this presentation which is kind of real time monitors of auto regulation which is really fun for us. So this is one of the monitors that we've designed to look at auto regulation in real time, and it basically takes the input signal of arterial blood pressure and nears. What I want to highlight in this is that it's a six hour window which shows the six hour historical trend of your mean arterial blood pressure. And here on the right is the lower limit of auto regulation which is 52 that the monitor has a delineated and an upper limit of auto regulation of 95. The circle in yellow is the patient's current mean arterial blood pressure and you can see a tiny warning sign because you're essentially below the lower limit of auto regulation of 52. And to the left here is a histogram kind of helping you see how well you've done in terms of keeping the patient within those limits of auto regulation. Essentially, we believe that such a monitor will provide the clinical team with a real time understanding of the limits of auto regulation and targeted blood pressure management in the OR. Next I'm going to move to a video of the monitor in action so this was a video I took a couple months ago. This is a six to eight month patient undergoing a bidirectional blend in our operating room. And you can see that they're actually about to come up by fastest when I turn on the monitor, and you can see that the current blood pressure is about 47 to 49 and it was below the lower limit and now the anesthesia team has basically increased the blood pressure so that it's above the upper limit or above the lower limit of auto regulation. So we think that such a tool can be very useful in terms of helping us basically navigate blood pressure targets in a in a domain right now where I feel like we're essentially making efforts based on historical basically experience of institutions. So when we go back to the European society and what we know about postoperative delirium I think that with targeted blood pressure management and better real time monitoring of auto regulation, perhaps that we can fill the void of therapeutic options to treat postoperative delirium and improve neurodevelopmental outcomes in our patients. So in summary, I think that optimizing blood pressure goals may reduce the incidence of postoperative delirium. There's no agreement right now on the appropriate map for individual patients undergoing cardiopulmonary bypass. And I think that real time monitors of cerebral auto regulation can identify optimal blood pressure targets retrospective studies have demonstrated serrated lower limits of cerebral auto regulation 66 in adults and on average 34 neonase but I'd like to highlight that the range is quite variable. I'd also like to highlight that the bypass strategies for adult versus neonates are quite different and that's important to consider in light of all of these discussions. Optimal blood pressure targets may also mitigate neurologic injury and kidney injury. And so, to conclude, I targeted control of blood pressure and cerebral hemodynamics may play a very essential role in the future management of patients undergoing surgery and ICU care. I'd like to thank Kathy and Ken who are here today and individuals below. And I guess we have some time for questions.