 I would like to take this time to acknowledge the profusion students of the Texas Heart Institute of Cardiovascular Profusion Program, and especially like to thank Debbie Adams, Program Director for the invitation to speak. I've always been particularly interested in research of the brain and cerebral blood flow. And now that has become a regular regular part of the profusion life. I think it is important to understand its relationship to neurologic injury. Extracorporeal life, extracorporeal membrane oxygenation may be a life saving treatment for patients with severe refractory lung and or heart failure when conventional intensive care fails. However, in addition to the critical condition of the patient, ECMO treatment itself is associated with significant morbidity and mortality. And as you can see from this pic, the neurological complications are among the leading causes of death and disability and ECMO patients. This insult here is a result of a middle cerebral infarct. This is a brief outline of what I'd like to cover in the next 20 minutes. So cerebral blood flow and cardiopulmonary bypass, talking about altered collateral blood flow, regulation of cerebral blood flow, neurologic injury, infarct and hemorrhage, alterations in hemostasis, brain injury and BV versus VA, the ECMO circuit itself, and of course CO2. In 1959, the closest parallel that can be drawn for inside into auto-regulation during ECMO is from cardiopulmonary bypass. Loss of cerebral auto-regulation can result in sphenia or edema and hemorrhage, even with slate changes in cerebral profusion pressure. You can see from this classic Lawson curve that was reported in 1959 and debated and then brought back again as sort of a key standard. The Lawson curve depicts auto-regulation of cerebral blood flow over a range of cerebral profusion pressures. Point A is the lower limit of the curve after which the decrease in cerebral profusion pressure will lead to reductions in cerebral blood flow. Point B is the higher limit of the curve at which an increase in cerebral profusion pressure will increase cerebral blood flow. And the range of cerebral profusion pressure depicted by X is the zone of auto-regulation where the cerebral blood flow remains constant over changes in cerebral profusion pressure. And it is regulated by vasoconstriction and vasotilidation of cerebral arterials. In 1995, Dr. Arthur Schwartz from New York Presbyterian Hospital and anesthesiologist wrote this paper entitled, Cerebral Blood Flows Determined by Arterial Pressure and Not Cardiopulmonary Bypass Flow Rate. Global hypoprofusion of the brain during cardiopulmonary bypass leads to ischemic insult and neurologic injury. Cerebral ischemia depends on collateral circulation. They did independent manipulations of arterial blood pressure and pump flow measured during cardiopulmonary bypass procedures. And their findings were that cerebral blood flow was greater at high blood pressure than low pressure during bypass. An alteration of pump flow rate produced no changes in cerebral blood flow. So they concluded that cerebral blood flow during cardiopulmonary bypass is regulated by the arterial blood pressure and not pump flow rate. And then in 1994, this original article, current profusion techniques for repair of giant cerebral aneurysms, was something that we created this circuit for clipping of these giant cerebral aneurysms that were unable to be clipped under the conventional method. Femoral cannulation was used, a closed chest approach was used, and the patient was cooled to a brain temperature of 16 to 18 before circulatory arrest. During arrest, blood is then drained into the venous and cardiotomy reservoirs, enabling us to shrink the cerebral vessels and aneurysm for direct bloodless surgical repair under the neurosurgical microscope. In 1996, a doctor Schwarz wrote another paper using a baboon model and the right common carotid artery and the ipsilateral femoral cannulated and joined to a centrifugal pump. Close circuit system of blood was withdrawn from the femoral artery, cooled and re infused into the right common carotid with the external branches occluded. The pump flow was regulated to keep the right common carotid pressure equal to the systemic blood pressure. The cerebral temperature was decreased to less than 25 degrees for three hours. The results for that global brain hypothermia resulted from profound altered collateral cerebral circulation during artificial hypothermic perfusion. Cerebral auto regulation is the ability of the brain to maintain relatively constant blood flow despite changes in perfusion pressure. On cardiopulmonary bypass, it's very similar to VA ECMO and a comparison may be drawn for similar changes in cerebral and systemic chemodynamics leading to changes in cerebral blood flow and auto regulation. There's a lot more data in the pediatric literature for cerebral blood flow changes on ECMO than there is for adults. So more studies are needed to assess cerebral blood flow change in adults. You can see here from the also registry. This is a graft of the annual respiratory adult runs, which happened to be the smallest but most growing population of ECMO patients. This is through 2020. Also from also depicting neurological injury ranges. We see that there are subtle neurocognitive deficits, including intracranial hemorrhage, seizure, ischemic stroke, and brain death. And this injury may be sustained ECLS during ECLS or post ECLS. And you can see from the registry in the year of 2020, a total of 7,161 runs reported of that 51% survived, but 49% had mortality. Now there are about five randomized controlled trials that have been reported for the use of ECMO. I'm just going to talk about four here just to review this evidence. And you can see that the first paper in 1979 was Zepal and ECMO can support gas exchange, but does not increase long term survival was reported. Then the second paper is the Morris paper. It looked at CO2 removal only. And then finally you see the Cesar and the Iolia trial in the Cesar trial, peak studied mechanical ventilation as VA or VV ECMO at a single center. And Elaine Coombs found no statistical difference in mortality at 60 day compared to mechanical ventilation. But what is of note here is there were none of these trials measured neurologic outcome. You can see from the Cesar trial, the exclusion criteria prior to entry was mechanical ventilation greater than seven days, intracranial bleeding, any other contraindication to limited heparinization for patients who were moribund. So no neuro patient was involved and the same in the Iolia trial moribund condition. According to a steps to score current non drug induced coma after cardiac arrest irreversible neurologic injury and decisions to withhold or withdraw life sustaining therapies all these patients were not included. If you look at the factors affecting cerebral blood flow on VV ECMO versus VA you can see that abrupt O2 and CO2 changes during initiation can result in constriction or dilation producing a reduction in sympathetic nerve activity that may lead to brain injury. And incidentally cerebral blood flow changes 4% for each millimeter of mercury changing in CO2. On VA ECMO outflow cannulation sites in the femoral artery flows into the descending aorta and can result in limb ischemia. Also retrograde flow creates afterload on the left ventricle causing distension reduced coronary blood flow or pulmonary edema or hypoxemia. And then we always have the opportunity once the heart begins to beat again a dual circulation which provides deoxygenated blood to the head vessels. This paper is a retrospective analysis of the also database. It's the LaRusso paper utilizing 6,834 patients, and they compare dual lumen and single lumen cannulas for VV ECMO. Dual lumen cannulas are larger with potential to increase cerebral venous congestion and that could be a problem. So their hypothesis was is the dual lumen in VV associated with higher rates of neurologic injury. The dual cannulation has several potential advantages over traditional cannulation, which is why many physicians choose to use it, including easier ambulation and reduced recirculation. The occurrence of neurological complications in this study was not related to the type of cannulation in patients undergoing VV ECMO. And as you can see from the chart on the right hand corner, intracerebral hemorrhage, acute ischemic stroke, seizure and brain death were nearly the same in both groups. Now we move to 2018 where we talk about neurological events are frequent in VA ECMO treated patients ischemic stroke is the most frequent in this particular paper after one week on ECMO support had no specific risk risk factor and is not associated with high higher mortality. Intracranial bleeding occurred earlier and is associated with female sex, central VA ECMO, low platelet count and rapid CO2 change at ECMO start and high mortality. And you can see the common theme here in most of these papers is changes in CO2. This paper also was a comparison of peripheral VA ECMO versus central VA cannulation. Here we see 275 patients, 15 of whom developed a brain injury. This is higher than a 1.7 to 8% incidence reported on other studies and may be explained by their generous CT examination policy. 37% of the lesions were diagnosed from a CT scan performed in the absence of neurologic symptoms. VA ECMO showed sole independent risk association with brain injury development. Pre ECMO cardiac arrest and conversion between ECMO modalities conversions from BV to VA carried a relative risk for development of injury at 7.6 while the relative risk for conversion from VA to VV was 0.66. If you remember the relative risk risk ratio is the ratio of probability of an outcome in an exposed group to the probability of an outcome in an unexposed group. Further explanations on VA ECMO blood is returned directly to the arteries without the lungs as a filter. So thrombi from the ECMO circuit may reach the cerebral circulation. Loss of natural pulsatile flow seen on VA ECMO may also affect cerebral vessels leading to hyper or hypo reactivity which has been suggested to increase the risk for neurologic injury. And then again we have differential hypoxemia or hypo perfusion to the cerebral vessels. Patients who suffered from brain injury had higher APTT at ECMO initiation. The limitations to the study was that it was retrospective and had its inherent limitations. 37% of the patients diagnosed with a brain infarct did not show any neurologic symptoms and the clinical significance predicting these infarcts could be debated. In this paper the frequency of neurologic events shown on VA ECMO are stated as relatively unknown. The study found 135 consecutive VA patients, 7.5% had cerebral bleeding, 2% ischemic stroke and 19% experienced some kind of neurologic event. Intracranial bleed was independently associated with renal failure. And again we see PA CO2 decrease at ECMO initiation was also associated with brain injury. Hemostasis disorders were not associated with intracranial bleed. Whereas in this paper talking about ischemic and hemorrhagic brain injury during VA ECMO. It's multifactorial process with multiple vents may injure the brain and vessels leading to bleeding and hemostasis disorders before and at the start of ECMO and rapid CO2 changes being among them. This author recommends to keep a platelet count above 100,000 during VA ECMO insertion and to infuse low dose heparin. Unless the patient's profusely bleeding to avoid circuit clotting that may injure and may induce itself from the side of Pena. Here you can see regulation of cerebral auto regulation by carbon dioxide because of the rapid decrease in CO2 leading to vasoconstriction. The relationships between PA CO2 change and cerebral bleeding is difficult to understand. You can see you have maximum dilation at point A and maximal constriction at point C. Now we have to just briefly touch on disorders of the ECMO circuit. Initiation of ECMO is similar to systemic inflammatory response syndrome. First we have contact activation and coagulation and inflammatory cascades begin to activate and cytokines rise rapidly. Then we have complement and contact activation producing leukocyte activation. All can lead to endothelial injury, disrupted microcirculation and end organ function. Many studies have discussed stairs during cardiopulmonary bypass, but less work has been done for ECMO. So some future research there. And then in addition, hemolysis, thrombocytopenia, fibrinolysis and acquired von Willebrand syndrome were found to be the same in VA and VD ECMO. There have been several successful case reports of VD and VA ECMO treatment without systemic anticoagulation in hemorrhagic patients. Other studies have looked at ACT levels at 180 to 220 versus 140 to 160 and reported no difference in oxygenator clotting. Subcutaneous anoxaprin was used for another study of 61 patients and no interest cerebral hemorrhage or clotted oxygenators was found. In the help ECMO study by McQuilton, normal versus low-dose heparin, decreasing PTT and anti-10A levels showed no difference in thromboembodic or bleeding events. ECMO causes platelet dysfunction, so it should be monitored during ECMO if possible. Consider using platelet agrigometry or TAG. The platelet aggregation blood test checks how well platelets in part clump together and cause blood to clot. One of the important things that we need to think about during ECMO that is accomplished in some institution and then not others. Using a near-infrared spectroscopy measures regional cerebral oxygen saturation. It's also used now on the limb, on the leg that is femurally cannulated either on the calf or the thigh to test for regional oxygenation there. Possibly showing the clinician the need for using an additional cannula to provide circulation to the lower limb. Imaging preferred cranial computed tomography in the paper by Marika, 37% of those patients who underwent cranial CT scan during ECMO were found to have either intracranial hemorrhage, ischemic stroke or generalized DEMA. Somatosensory of both potentials are used mainly in, right now I have never seen them used in ECMO but in cranial surgery. The cortical generators of the N2 component are located in the territory of the middle cerebral artery and various studies have correlated decreases in N2 and N20 amplitude by more than 50% with cerebral hyperperfusion. EEG, the presence of EEG background abnormality and certain electro-graphic patterns can aid in the prediction of neurologic outcome after ECMO support. And then there are lab results, biomarkers as predictors of neuronal injury. So for glial activation the S100B protein and adhesion molecules measuring the ICAM5. So just to recap there's a lot of information, hopefully I've sparked your imagination somewhat to what's happening on ECMO and neuro injury. Cerebral blood flow is necessary to maintain auto-regulation, consider monitoring during the ECMO run. Altered collateral blood flow we talked about, cannulation techniques and venous congestion is something that needs to be considered as the physician is cannulating the patient, choosing the right size cannula, placing it in the correct location. Neurologic injury can occur at any time during the ECMO run. Brain infarction, intracranial hemorrhage both occur on VED and VA ECMO. And alterations in hemostasis, so we need to consider circuit components, we need to keep our circuits simple. The more you add to the circuit, the more opportunity you have for clotting in different areas of the circuit. And then monitoring, possibly CT, EEG, nears, evoked potentials in lab tests may alert clinicians to evolving brain injuries during ECMO. And again CO2, avoid large fluctuations upon initiation and during ECMO. And in summary, interest cerebral hemorrhage, ischemic stroke, seizure, cerebral edema, intracranial hypertension, global cerebral hypoxia and anoxia and brain death seem to be the most common neurological injury on ECMO. They may result prior to initiation of ECMO during or post ECMO. And failure of ECMO to provide adequate oxygen delivery or abrupt changes in CO2 levels can cause neuro injury. Early diagnosis and intervention are crucial to limit morbidity and mortality from neurologic injury during ECMO. Thank you for your time.