Hello, I am Aarti Sarwal and in the next 25 minutes I will be doing a very fast paced review of Cerebral Hemodynamic Monitoring. We will cover the physiological concepts of intracranial pressure, cerebral perfusion pressure and compliance. We will review neuromonitoring techniques for evaluation of various hemodynamic parameters with regards to intracranial pressure, blood flow, oxygenation and some other parameters and will describe the poles and pitfalls of using various targets for therapy to optimize management of acute brain injury. A caveat to remember is that in patients who do have an exam, clinical examination and lesion localization remains the go-to surveillance tool for detecting neurological deterioration but sometimes occurs too late in the course or cannot be detected in critically ill patients like comatose patients or patients needing sedation for medical indications. These are specifically the patients which become the focus of neuromonitoring where this can help in patients challenged with clinical exam or where clinical exam has subtle changes. The whole focus of hemodynamic monitoring for the brain is to identify salvageable tissue from the lost tissue which obviously is the holy grail and we don't have a single test that can really give us that answer and the knowledge and clinical evidence about cerebral hemodynamic physiology is evolving. We all are familiar with the Monroe Kelly doctrine which talks about skull as a box which contains three major compartments brain, blood and CSF. Until there is compliance of the brain within the compartment, a change in volume of one constituent causes a corresponding change in the volume of either of the other two constituents keeping the intracranial pressure almost normal. But once a critical threshold is reached where a significant change in volume of one compartment has significantly compromised another compartment, say brain has swollen to the point where all CSF has gone out of the cranium and now blood is getting compromised, that's when pressures exponentially increase and the pressure volume response curve becomes extremely steep. This is where we say brain has become non-compliant and is very prone to causing herniation syndromes. So we know about the cerebral herniation syndromes clinically and radiologically. So why do we need to measure cerebral hemodynamics or intracranial pressure? Well, we know that our interventions in general that are targeting patients over empiric systemic parameters in clinical examination do not have the best accuracy even for the extremes of clinical deterioration and we do really need high suspicion in catching the patients on the spectrum of deterioration where physiological changes in the brain may precede the clinical deterioration that could be a terminal event like the cerebral herniation syndromes. Having said that, neither parameters that we discuss in the next 20 minutes are not the final answer. A multi-modality approach understanding the patient's individual physiology might be the answer for future research. We will start with discussions on the intracranial pressure. So normal intracranial pressure inside the cranium is 0-10 mmHg. Anything above 15 mmHg is considered abnormal and the normal ICP waveform is pulsatile due to intracranial arterial pulsations that bounce against the skull. Please pay attention to the conversion factor. ICP can be measured both by units of CSF of water or mmHg and the conversion factor is 10 mmHg is equivalent to about 13.6 cmHg of water. The most common way we measure the intracranial pressure in a critically ill patient is by an external ventricular drainage catheter which is essentially placed through right frontal area and is a simple tubing that connects the ventricular system to an outside manometer that measures the column of CSF when the circuit is closed in either cmHg or mmHg. Your typical drainage system will show either of these on each side of the manometer. I have listed a great review on this topic that can go over the evidence based management of external ventricular drainage in different pathological processes. You can also measure intracranial pressure by a fiber optic strain gauge monitor that is inserted invasively as well but compared to an external ventricular drainage catheter it measures the pressure in the brain parenchyma not the ventricular system and as opposed to an external ventricular drainage catheter it cannot drain CSF while it's measuring it. Another difference is that the external ventricular drainage catheter can be zeroed or calibrated depending on patient motion or if it needs to be revisited for a placement multiple times but the fiber optic intracranial pressure monitor also called the BOLT cannot be recalibrated after insertion. The values also drift over time depending on the vendor 5-7 days and although trends still matter but the absolute values get an error of about 6-10 mmHg a week after placement so typically cannot be placed or used meaningfully after a week. The global severe brain injury patients like TBI patients that have collapsed ventricles typically can be monitored with fiber optic BOLTs. Patients who have CSF drainage as a part of their management strategy typically should be monitored with an external ventricular drainage catheter. There has been a wide variability of practices when ICP is used as a target for therapy in critical yield patients. A number of 15 mmHg to 25 mmHg has been used in various studies. There's also variability in the units like I mentioned 10 mmHg is equivalent to 13.6 cm of CSF or water. Trends do matter so a patient who was in low teens is now keeping high teens ICP should have further evaluation of the underlying change in cerebral hemodynamics. The duration of ICP crisis whether it's 5 minutes or 25 minutes depends on the underlying pathology and the duration of ICP crisis that dictates escalation of therapy has also been very variable from one trial to the other. Classically in a patient with low brain compliance at risk of herniation will escalate therapy when the ICP has been more than say 20 mmHg for more than 5 minutes but patients who have a stable intracranial pathology that doesn't have incipient increased risk of herniation they may tolerate longer ICP crisis. So say if we were weaning a patient's EVD in that particular case we may tolerate an ICP of more than 25 say for 25 minutes before we open the EVD to call it a failure of wean. Underlying brain compliance matter so if the brain is compliant patients may tolerate fluctuations in ICP numbers but the brain is non-compliant even small changes in ICP can be on the steep part of the curve like we mentioned before and cause significant alteration in cerebral perfusion pressure. There also has been a lack of consensus whether prevention of ICP crisis is as important and impactful as a reactive treatment of ICP crisis. You can assess patient's brain compliance by looking at the ICP waveforms. The typical waveform for intracranial pressure as measured by a bolt or an external ventricular drainage catheter is basically produced by the pulsations of the intracranial arteries transmitted to the parenchyma bouncing against the skull. So you have rapid upstroke, you have stepwise deacceleration with three components P1, P2, P3. P1 reflects the cerebral blood flow at the end of systole, P2 is called the tidal wave and reflects the brain compliance if the brain is compliant it will decay and P3 is the diacritic notch that reflects the aortic valve closure and the blood flow back to the heart. In a non-compliant brain which has had ICP crisis, so this is Lundberg's manifestation of continuous ICP monitoring. In this particular case you're seeing normal compliance, rapid upstroke, P1, P2, P3 are decaying with de-escalating numbers but this patient has an ICP crisis and after the ICP crisis say brain had a new hemorrhagic conversion or a new hemorrhage, the compliance is low. P2 becomes higher than P1 means after the first systolic impulse the rebound is higher than the first impulse and even if this patient's ICP decreases because of a new parenchymal pathology and reduced compliance the waveform will stay the same. In this particular physiological monitor you can see the ICP waveform P1, P2, P3 all three are decaying after the initial systolic end but in this particular case P2 is bigger than P1 almost like brain showing you the finger. The concept of cerebral autoregulation is also important to recognize. Cerebral autoregulation is the inherent capability of brain to change the diameter of its cerebral blood vessels to alter cerebral blood flow in response to inherent metabolic needs, change in substrate or other physiological changes in the brain. So cerebral blood flow is altered to serve the physiology of the brain and that's cerebral autoregulation. Everybody has a zone of autoregulation in which the blood vessel diameters will change actively to preserve cerebral blood flow but below the zone of autoregulation the inherent capacity of the vessels to dilate or constrict will collapse causing decreased cerebral perfusion and above the zone of autoregulation inherent capacity of the arterial diameter to change in response to physiological changes will disappear as well causing passive vasodilatation. The cerebral autoregulation curve is different in younger and older people and does shift to the right side in older people, also shifts to the right in chronically hypertensive patients. So in these patients the lower zone of autoregulation may be much higher because of chronic hypertension or atherosclerotic changes in the cerebral blood vessels. The concept of neurovascular coupling is an extension of the autoregulatory phenomena which is basically regulation of cerebral blood flow to match the demat of cerebral metabolism. If this coupling mechanism is intact then any change in the oxygen supply of the brain say increased seizure is increasing the oxygen needs of the brain and that will increase oxygen demand and the oxygen demand will be counteracted by adequate increase in arterial blood flow to maintain the neuronal metabolic rate called CMRO2 and the residual deoxygenated hemoglobin will be passed on to the venous blood. So increased metabolic rates of the brain are matched by increased blood flow or increased oxygen extraction. So measurement of blood flow or measurement of the mixed venous sats outgoing from the brain venous drainage are both good ways of measuring the oxygen extraction demand or the metabolic rate. This would be an example of a flow sheet that shows simultaneous measurements of the patient's intracranial pressure as seen here and the arterial pressure measured invasively. So in these three highlighted segments you can see the patient's ICP goes up up to 50s and in response to that the patient has a spontaneous non-induced change in arterial blood pressure with a goal to maintain CPP so patient is doing that spontaneously in all three of these segments where the ICP goes up and the arterial pressure goes up in response. That is intact autoregulation. Now autoregulation tends to maintain cerebral perfusion pressure. Traditionally we've thought of CPP as an absolute number of MAP minus ICP and the number normative value has been said to be around 70 to 90 millimetres of mercury. Couple of challenges there when people have reported CPP in different literature the CPP is traditionally measured at the tragus because ICP measurement with EVD is zeroed at the tragus. The MAPs are typically measured at the level of the left atrium or the fourth intercostal space which is where the blood pressure sphygmomanometer is zeroed. So making sure that the zero reference point is the same both for ICP and arterial pressure measurements has not been consistently done across literature. In general we know CPP is the driving pressure for cerebral blood flow across the microvascular capillary bed but now we know that CPP numbers are not absolute in every patient. In patients that have intact cerebral autoregulation if you artificially increase their CPP because they were low they can actually facilitate management by compensatory vasoconstriction and reducing ICPs to a certain extent. If the vasoconstriction goes on for a prolonged time or to a severe degree it can promote cerebral ischemia starting a paradoxical cascade. On the other hand if we had a patient with impaired autoregulation say a severe TBI patient increasing their CPPs to an artificial predetermined threshold like 70 to 90 may actually be harmful because these patients may have an altered zone of autoregulation and this high CPP may cause reperfusion injury. In addition, it has been known to cause increased systemic complications. So there are quite a few studies that have looked at CPP both from the arterial input and venous output perspective and have shown that CPP thresholds depend on the degree of auto-regulation and intracranial compliance and an impaired auto-regulation can be harmful from the perspective of increased reperfusion injury as well as systemic complications. The Brain Trauma Foundation guidelines have incorporated auto-regulation in choosing patients' cerebral perfusion pressure targets where they recommend an avoidance of CPP less than 50 in general and if the auto-regulation is intact, induce CPPs to normal values of 70 to 90 but if the CPP is impaired, if the auto-regulation is impaired, to keep the CPPs 50 to 60 because higher CPPs have shown worse functional outcomes. An emerging trial to know would be finding CPP targets that are specific to a particular patient which are calculated based on the PRX values. PRX is the linear correlation coefficient between patient's ICP as measured by an invasive ICP monitor and the patient's systemic blood pressure and this linear correlation coefficient basically represents auto-regulation and based on the patient's physiology, the PRX value changes over time and finding the optimal PRX values where the patient has the best outcomes has been evaluated in a feasibility trial called Ocogitate. This is not a trial that was designed for changing outcomes but has shown early signals that finding patient-specific CPPs may be feasible. Why would you want to measure cerebral blood flow or cerebral perfusion when you can calculate it? Well, this is a classical example of a patient that had invasive cerebral blood flow monitor and had ICP crisis. Initially, patient's MAPs and ICPs were being measured with a drop in MAP. Patient probably lost auto-regulation and this drop in CPP was still above the threshold of 50 but if you look at, as calculated by MAP minus ICP, but if you look at the patient's measured cerebral blood flow, it's actually pretty compromised. So the measured cerebral blood flow may be different than the calculated cerebral perfusion pressure. In this particular case, pressors were used to increase the MAP to 112. That restored the patient's CPP but that did cause transient ICP crisis that needed mannitol but once that ICP crisis settled down, the measured cerebral blood flow restored as well. You can measure cerebral blood flow by brain tissue perfusion monitor that is similar to a bolt and basically measure the blood flow fiber optically. It's only used mostly in research settings at this point and tertiary care centers and has shown normative values that differ in gray matter and white matter. So the placement of the blood flow decides what normative values that you will use. If it's the end of the brain tissue flow monitor is placed in the gray matter, you will use the gray matter values but typically it's placed in the right frontal region in the white matter. The thresholds for oligemia, ischemia, in fact, has also been investigated and obviously the blood flow thresholds change with age. You can also assess and measure CPP directly non-invasively. There are devices called NIRS that can measure superficially placed skull-based probes to look at the CPP. You can also use trans-cranial dopplers both qualitatively and quantitatively to look at cerebral perfusion pressure. On the left side, you're looking at normal middle cerebral flow velocities with good blood flow throughout the cardiac cycle and in this, you're seeing compromised diastolic flow and in this particular image, oscillating flow with forward flow during systole and backwards flow during diastole. So this would be a patient with normal CPP. This patient has impaired CPP decreased area under the curve but forward flow throughout the cardiac cycle and in this particular case, patient CPP is severely compromised because the forward flow and the backward flow are equating each other so CPP is zero and this can progress to systolic humps and cerebral circulatory arrest if the underlying injury is not salvageable and the oscillating waveforms persist. Attempts have been made to calculate CPP quantitatively from TCD as well. There are several formulas out there. A recent trial called IMPRESSET trial looked at this particular formula proposed by Schmidt and showed significant sensitivity and specificity of non-invasive CPP calculated from TCDs in assessing patient's ICP. You basically measure the CPP based on TCD parameters and then you use the patient's maps to calculate the ICP. You can also measure the brain tissue oxygenation of the brain parenchyma itself by placing an invasive probe. This is placed through the same entry as a parenchymal intracranial pressure monitor or a blood flow monitor is placed and this measures the oxygen concentration that the brain parenchyma is exposed to. There have been anecdotal studies that have shown that the tissue oxygenation gets significantly compromised and can be a target for therapy in patients, especially with severe TBI. And there are three ongoing trials, BOOST-3, OxyTC, and Bonanza that you should be aware of that are using oxygenation targeted protocols to see if that improves patient's outcomes compared to a current care that is usually ICP management or CPP management targeted. Jugular venous oxygen saturation is another index that is used in cerebral hemodynamics guiding the management. Just like we use mixed venous SATs in looking at oxygen extraction and patient's volume status in sepsis, you can use the same catheter either by doing routine sampling of the jugular venous bulb in the patient's brain. Jugular venous blood gases, or you can place a fiber optic monitor that measures continuous jugular venous oxygen saturation and decreased venous oxygen saturation basically gives you an index that either oxygen extraction is increased from the brain because of increased metabolic demands or the blood flow to the brain is compromised because of decreased arterial inflow. Again, it's one of the parameters that can be used to look at global venous oxygen extraction, especially on the ipsilateral hemisphere and different patterns have been described. You can also look for metabolic substrates in the brain parenchyma by measuring lactate, pyruvate and glutamate. A small catheter is placed that has a micro tube that aspirates a small amount of interstitial fluid and this is aspirated and then tested on a periodic basis, typically hourly through an external microdialysis machine that gives you the lactate, pyruvate and glutamate concentrations in addition to other neurotransmitters. Typically lactate pyruvate ratios that become high denote cell death and therapies can be targeted to optimize metabolic substrates but it is typically a marker of brain injury that has already occurred and is not a good marker for preventing further acute brain injury. People have tested this in TBI to optimize clinical protocols as a clinical outcome. The other factor in consideration is the placement, whether it is placed next to the highest risk of secondary brain injury or placed in the normal brain significantly affects how you use this parameter. So these were a few of the goal directed therapies based on specific physiological variables that have been used. When we talk about clinical algorithms in general, there have been several clinical algorithms used. Typically we use the ICP targeted therapies in most ICUs where we pick up an ICP target of 20 to 25 millimeters of mercury and implement ICP loading therapies without regard to CPPs. Then the CPP targeted protocol also called the Rossner protocol takes into account either calculated CPPs from patient maps and ICPs or directly measure the CPP invasively through an invasive blood flow monitor or non-invasively through transcranial Doppler or NERS to look for CPP targets. The combined ICP-PCP target has been evaluated by the Ehrenbruch protocol also called the Cambridge protocol. The other protocol that is quite different from the above protocols is the Lund protocol which basically targets perfusion but tends to improve microcirculation by reducing catecholamine concentrations and would typically allow way more lower CPPs than other protocols by using active cooling and sympathetic blockers. Then the Brain Trauma Foundation guidelines in particular talk about auto-regulation as a target that should be considered when deciding which CPP targets should be practiced in a patient. And the other clinical protocol that we now use is the exam targeted based on the study by Chestnut and recently there has been an interest in calculating patient-specific CPP targets using the CPP-opt approach by calculating the physiological range of PRX at which the CPP is optimized but this needs further validation in clinical studies. In general, basic management of ICP involves optimizing ABCs, airway breathing circulation, keeping the head of the bed elevated more than 30 degrees, keeping the neck mid-lined to optimize venous drainage, hyperventilate the patient as a temporary measure till definitive ICP lowering therapies have been implemented. Prolonged hyperventilation can cause cerebral ischemia and rebound cerebral vasodilatation hence is typically avoided. Targeting normothermia to optimize the CMRO2 for neuronal metabolism which by the principle of neurovascular coupling optimizes the blood flow going to the brain and keeps the blood and brain compartment in equilibrium inside the cranium. Implementing osmotic therapy for various ICP crisis, mannitol, 3% hypertonic saline or the other various concentrations like 13.4 or 23.4% using sedation or suppression of metabolism of the brain by using barbiturates, heavy sedation with benzodiazepines again using the principle of neurovascular coupling and going to surgical decompression and decreasing the intracranial pressure by changing volume of at least one of the compartments whether it's CSF drainage, vasoconstriction by temporary hyperventilation or reducing the brain parenchyma volume by osmotic therapy or decompression. Chestnut study is the one that we need to be aware of which was done in patients older than 13 with a GCS of 3 to 8 evaluating whether an ICP targeted therapy where ICP of 20 millibars of mercury or less was used as a target in addition to clinical care was superior to just care based on imaging and clinical examination where there was no ICP monitor and it did not show superiority of either arm doesn't mean that the patients don't need ICP monitor it just means that the patient's ICP of 20 millimeters mercury is not a magic number and patients require a much more complicated method of treatment hence need to pay attention to other physiological variables that we just talked about. This has also led to an interest in multi-modality monitoring in this particular case we're measuring the mean arterial blood pressure, we're measuring intracranial pressure, we're directly measuring the cerebral blood flow rather than just calculating it, brain tissue oxygen therapy and in this case brain temperature as well. PRX as I told before is a linear correlation coefficient between patients measured ICP and measured MAP and you can see the PRX fluctuating over time and CPP-OPT is the methodology where the optimal PRX is calculated over time typically at the bottom of the u-shaped curve and the CPP-OPT based therapy has been tested in the cogitive trial we talked about. Multi-modality monitoring is typically done through the same entry point to minimize trauma to the brain and up to three to five monitors typically can be placed like COX, Hemidex, Bolt and obviously does increase the chance of local trauma. There is value in multi-modality monitoring as evidenced in this particular case published which basically had an anecdotal case of a patient with invasive multi-modality monitoring. The patient CPP and ICP were being monitored and the ICP showed periodic crisis. Each time the ICP crisis happened patient had increased brain temperature and decreased oxygenation and the blood flow typically increased concordant with the ICP crisis. Now one possible explanation is that the cerebral blood flow is increasing in response to ICP crisis impairing CPP so this patient is intact out of regulation but when the blood flow was actually measured the blood flow is actually going up prior to the ICP crisis. The patient was having seizures which were electrographic and each time patient seized because of neurovascular coupling the increased blood flow was the trigger for ICP crisis and not the other way around. So it's important to recognize in patients whether the blood flow compartment is the one that is driving the patient's intracranial pressure, hyperemia, vasodilatation, hypercarbia could be such causes versus patients with brain swelling, brain parenchymal pathology where the CPP is actually being compromised. So a perfusion driven ICP crisis needs different management and perfusion limiting ICP crisis hence it's extremely important to measure cerebral perfusion rather than just calculating it. You can also do surgical decompression and different trials to know about that would be Destiny Decimal Hamlet and Hemicraniectomy for ischemic stroke for malignant cerebral edema. Rescue ICP and DECRA would be two trials to be familiar with in hemicranie indications in severe TBI. For refractory ICP crisis not responding to advanced tears and in ICH Stitch and MISTI are two trials that had different phases that would be great to be with. Thank you for spending time. Happy to take questions.

cerebral hemodynamic monitoring

intracranial pressure

cerebral perfusion pressure

compliance

neuromonitoring techniques

Monroe Kelly doctrine

cerebral blood flow