Hello and welcome to the 51st Critical Care Congress. My name is Miriam Weiss. I'm a neurosurgery resident from Germany, currently work in Switzerland, but the data that this study was made from were made at Aachen University in Germany. I am particularly interested in neurocritical care and neuromonitoring, and if you have any questions about my presentation, please feel free to send me an email. What I'm presenting today is a study on optimal cerebral perfusion pressure during delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage. This study is primarily from Aachen University in Germany, but we've also had people participate from Columbia University in New York and from Maastricht University in the Netherlands. After aneurysmal subarachnoid hemorrhage, about a third of patients can develop delayed cerebral ischemia or DCI. This can be a neurological deterioration or in sedated patients, it can be a hyperperfusion, which you can measure in CT imaging, and if you don't detect it, if you don't treat it, it can result in permanent cerebral infarction. Blood pressure management is one of the ways to treat DCI. When the aneurysm is occluded and the patient is in the ICU, you typically have quite liberal blood pressure management initially. Many clinics have a lower CPP limit of 60 or 70 mmHg, but not really an upper limit because you assume that there is a certain demand for high cerebral perfusion at this time. When a patient develops DCI, inducing hypertension is typically one of the ways to treat it first. It used to be part of triple H therapy of hypertension, hemodilution and hypervolemia, but hypertension is really what seems to be most beneficial. Everybody does it a bit differently. You can have a systolic pressure target, an MAP target, indication, duration, all depends a bit on where you work because we don't have the evidence. There were several smaller uncontrolled clinical studies, and it was the observation that about two-thirds of patients have a clinical improvement, and you can also have elevations in CBF or cerebral oxygenation. There was one clinical trial, the HIMALAYA trial, which was a prospective randomized controlled multicenter trial, and they intended to measure CBF and clinical outcome in patients with and without induced hypertension in 240 patients. But they had some recruitment issues and also complications, so they ended up with few patients, and they didn't find a significant difference in cerebral blood flow and in clinical outcome. So we still don't really know what to do with blood pressure during DCI. Ideally, you would want an individual target parameter that is objectively measured and continuously updated. There is a new concept that was derived in traumatic brain injury, where you don't have a fixed blood pressure target, but rather you try to change your blood pressure in a way that your autoregulation, cerebral autoregulation, becomes optimal. Autoregulation is the process where cerebral blood vessels react to blood pressure, and they try to keep the blood flow in the brain steady. This works within certain thresholds, but the process is often disturbed in patients with TBI or aneurysmal subarachnoid hemorrhage. You can measure and quantify cerebral autoregulation, and you do this by correlating blood pressure and ICP. If you have a blood pressure change and your ICP remains stable, you have no correlation or a low correlation, and that means that your autoregulation is intact. But if your blood pressure increases and your ICP increases as well in a kind of passive reaction, you assume that autoregulation is disturbed. High correlation values mean that you have disturbed autoregulation, and low correlation values mean that you have good autoregulation. Now an observation was made that the functionality of autoregulation changes depending on blood pressure. If you have very low blood pressure, you have poor autoregulation, and the same is true if you have very high blood pressure, but there is a sweet spot somewhere in the middle where your autoregulation becomes ideal, and the CPP that is measured at this time is called optimal cerebral perfusion pressure or CPP-OPT. There is currently only a handful of studies on CPP-OPT after subarachnoid hemorrhage, so we wanted to know what is CPP-OPT at time of delayed cerebral ischemia. And then you can calculate the difference of your actual or your spontaneous CPP and the optimal target. And we wanted to know, are the patients with DCI somewhere close to it? Are they far away from it? And then also, how does this relationship change when you induce hypertension? Are we bringing patients closer to CPP-OPT? Are we driving them away from it? We have included our subarachnoid hemorrhage patients into a prospective observational study, and we've included patients for this particular question who had invasive neural monitoring because you need an ICP value to calculate CPP-OPT. We've also noted the exact time point of DCI, which for us is the time point of clinical deterioration or the time point where we have a significant hyperperfusion in CT perfusion. And our first-year treatment is inducing hypertension, and our target is a systolic target of above 180 mmHg. There were 240 patients with subarachnoid hemorrhage in the respective time frame, about a third developed DCI, but we had to exclude most of the patients because either they didn't have an invasive probe or they didn't have data at the right time or CPP-OPT couldn't be calculated. So we had 20 patients with DCI and induced hypertension with good data at the right time, and we also included 19 patients without DCI who had good data in a comparable time frame. We looked at patients with DCI first. This was a typical subarachnoid hemorrhage cohort with primarily female patients around the age of 55, and about half of the patients were heavily affected with a high clinical or radiographic grade. What we did was we looked at 48 hours of data before the diagnosis of DCI, and what we saw was that CPP-OPT increases slowly over time. 48 hours before DCI, we have a little bit over 80 mmHg, increasing to a mean of 86 mmHg, and this increases further in the few hours before DCI to around 90 mmHg. For a while, you can see a similar development in the actual CPP values. You can see it increasing, but then about 12 hours before DCI, the CPP decreases to about 81 mmHg. What you see now is the difference of your actual CPP and of your calculated optimal CPP, and you want this value to be around zero, because that means that your actual CPP is very close to CPP-OPT. You can see that the patients with DCI are around zero for a long time, but something happens a couple of hours before DCI where your difference becomes really negative to about negative 8 mmHg. This is a combination of the slowly increasing CPP-OPT but the decrease of CPP a couple of hours before DCI. This dynamic of delta CPP is confirmed in the PRX values, which show you the functionality of autoregulation. Negative values or values around zero are intact autoregulation, and an increase would mean a worsening of autoregulation. The patients with DCI are relatively stable for a long time, but a couple of hours before DCI, when you have this drop in delta CPP, you also have a worsening of autoregulation. As CPP is the combination of MAP and ICP, we wanted to know which one of these components is it that drives the delta CPP towards such negative values. But there wasn't any significant change point in both of them. It seems to be a slight increase of ICP and a slight decrease of MAP that together lead to a decrease of CPP and a very negative delta CPP. We wanted to know if there were any identifiable reasons for a blood pressure decrease that you can maybe identify in the clinical course and try to avoid it to maybe help the occurrence of DCI. So we compared the patients with DCI with the patients without DCI. The demography is relatively comparable, except that the patients without DCI were significantly older. We didn't find a difference in noradrenaline demand, in hemoglobin values, in analgosedation or in nimidipine dose, but we did find a difference in the proportion of patients who currently had an infection at time of DCI who had antibiotic treatment or were just sampled for a focus of infection. This was a lot higher in DCI patients. And then there was a statistical trend, if not a statistical significant difference in the fluid balance. DCI patients had a negative fluid balance while patients without DCI had a positive fluid balance. Now there wasn't this one reason that can explain a relative hypotension. Relative hypotension because it's not a real hypotension according to definition, but it's a hypotension according to CPP op values. And it's difficult to identify the one reason that can explain this difference. Finally, we looked at how this dynamic changes with induced hypertension. And we compared the delta CPP, which again is supposed to be as close to zero as possible before and after induced hypertension. And we already saw that most patients have negative delta CPP to begin with, so a relative hypoperfusion. But we saw that the majority of patients really go beyond CPP opt with our induced hypertension up to 12 millimeters above CPP opt. So we probably have some kind of overtreatment actually according to CPP opt and auto-regulation. These results have to be interpreted very carefully. First of all, because of the case number, it is difficult to acquire this kind of data. Also, subarachnoid hemorrhage patients can be very heterogeneous. So these results must be viewed as exploratory findings. And the other limitation that was important to us is that CPP opt gives you a global target parameter for the whole brain because it's based on ICP. But DCI is a local phenomenon, which is usually territorial hyperperfusion. And we don't yet know if CPP opt, the way it is calculated now, is a good target parameter also for this tissue that is most at risk. Our conclusions are that CPP opt could be around 90 millimeters mercury at time of DCI. This value is a lot higher than the minimum recommendation of 60 to 70 millimeters mercury that many clinics currently have. If you fail to meet CPP opt, this could be associated with the development of DCI because you have this worsening of auto-regulation, which could be the acute trigger of clinical deterioration or hyperperfusion in CT. But then if you induced hypertension the way we do it with a systolic blood pressure target of 180 millimeters mercury, you may overtreat to a certain extent. These were exploratory findings, but we feel like it is useful to keep investigating CPP opt as an individual blood pressure target parameter. And we are actively working on that together with our colleagues from Columbia and from Maastricht to work out if CPP opt is a good target parameter for blood pressure management after subarachnoid hemorrhage. Thank you for your attention and please feel free to contact me if you have any questions.

Dr. Miriam Weiss

optimal cerebral perfusion pressure

delayed cerebral ischemia

aneurysmal subarachnoid hemorrhage

induced hypertension