For this next talk, I'm going to be speaking about cardiovascular surgery complications in mechanical devices. My objectives for this next lecture first include the discussion of differential diagnosis and management for the decompensated patient after cardiac surgery, and then I'll review the function, management, and troubleshooting of pacemakers, ventricular assist devices, intra-aortic balloon pumps, and extra-corporeal membrane oxygenators. The differential diagnosis for hemodynamic instability after cardiopulmonary bypass is shown here on this slide, and it's important to commit to memory and to work through in this clinical situation as time is of the essence to respond to this situation. After we go through that differential diagnosis, we'll then turn to devices and functional troubleshooting. But first, let's start with a question. A 62-year-old female with insulin-dependent diabetes, hypertension, and unstable angina, now post-operative at age zero, status post-coronary bypass grafting, and aortic valve replacement is hypotensive in the cardiac surgery ICU and unresponsive to volume resuscitation, norepinephrine, and epinephrine. Epidemiologically, what is the most likely cause? Of the options given, vasoplegia is the correct answer. However, it's not really a fair question because, as you can see, the other options are also highly incident after cardiac surgery. And from the clinical history I gave, you could make the argument that one of these other diagnoses was actually present in this case. Now let's dive a little bit deeper into each of these potential diagnoses in the setting of hemodynamic instability after cardiopulmonary bypass, first starting with vasoplegic syndrome, which is characterized in the early post-operative period by severe hypotension, normal or elevated cardiac output decreased feeling pressure, and low systemic vascular resistance. So vasoplegia is a state after cardiopulmonary bypass where afterload is drastically low and unresponsive to medical therapy, resulting in severe hypotension. Its incidence and implications in prognosis are of the following. So the incidence is somewhere between 5 and 40%, so happens quite regularly, but this reported rate does show that it's difficult to diagnose. Severely norepinephrine refractory vasoplegic shock is associated with a high mortality increase. Catecholamine refractory vasoplegic shock can increase mortality as high as 25%. The pathogenesis behind this severe vasodilation is not completely understood, but it's likely multifactorial due to systemic inflammation, ischemia reperfusion, and the surfaces of the cardiopulmonary bypass machine. Drugs and maybe a relative deficiency of vasopressin as well could be operating in an increased production of nitric oxide. The biochemical mechanism behind this severe vasodilation is proposed here and is related to other processes such as septic shock leading to severe vasodilation, however, there aren't, as of this point, any biochemical markers in the setting of this presentation that are diagnostic. There are increased risk factors for the development of vasodilation and vasoplegia after cardiopulmonary bypass and they are reported here. For the treatment, first, the other aspects of the differential diagnosis need to be ruled out. However, once the diagnosis of vasoplegia is made, treatment is mostly supportive. Goal-directed fluid resuscitation, supporting with vasopressors, and there are multiple unproven but proposed treatments listed here. Next on the differential diagnosis for hemodynamic instability after cardiopulmonary bypass is low cardiac output syndrome, defined as cardiac index less than 2, systolic blood pressure less than 90, and signs of tissue hyperperfusion. This entity is associated with poor outcomes including morbidity in short and long term mortality and increases operative mortality to around 4%. Low cardiac output syndrome seems a bit confusing because it is somewhat vague and actually can result from multiple different etiologies, but it is an important diagnosis epidemiologically because it is so significantly related to outcomes. So when one suspects low cardiac output syndrome, then the differential diagnosis further expands to this following list of potentiating factors. You have to think about the effectiveness of the cardioplegic arrest on bypass and whether or not the myocardium is stunned, or frankly ischemic for that matter. But this might require significant amount of support in the immediate postoperative period or reduction of preload and vasomotor tone to support myocardial recovery. Time is of the essence working through this whole differential diagnosis during the hemodynamic instability of postcardiopulmonary bypass, but in particular with low cardiac output syndrome, the decision to move quickly to mechanical support so as to prevent further injury to the heart and enable it to recover seems imperative. Moving beyond low cardiac output syndrome, what about some more specific diagnosis of hemodynamic instability? And the next is perioperative myocardial infarction. And around cardiac surgery, that's defined as a type 5 MI by the 2018 consensus universal definition of MI. For that definition, troponin values within 48 hours following CABG are of the following criteria. If there was a normal preoperative troponin, that patient must have 10 times the 99th percentile upper limit. If stable but elevated preoperative cardiac troponin, the patient must have an increase in 20% from that value. In addition to either of those, depending on that situation, the patient must have one of the following new Q waves, angiographic documentation of the new graft or new native coronary artery occlusion, or thirdly, imaging evidence of loss of viable myocardium that was previously not there. Perioperative MI is actually quite common after coronary artery bypass surgery and estimated to be around 9%. There are subclasses of this type 5 MI, and it's important to work through this because the treatment might be different. There's a failure in new graft function, which occurs in around 1% of all coronary bypass surgeries, an acute coronary event involving the native coronary arteries that were not manipulated, or more global type 5 MI that was due to inadequate cardiac protection. So if you suspect perioperative MI, what can you do about it? There's really limited evidence comparing management strategies, but I'm going to report a little bit of what has been done. The key issue is to identify patients with actionable lesions, because this group may benefit from urgent revascularization. So if you suspect perioperative MI due to a new lesion, what treatment options are available? There are a couple small single center studies that demonstrated that emergent percutaneous catheter intervention limited myocardial damage more so than a surgical-based strategy. So you would be thinking, should we go back to the operating room? Should we go to the cath lab? Obviously going to the cath lab for a stent intervention that would require antiplatelet therapy in an immediately post-cardiac surgery patient would be difficult due to concerns of bleeding. Then the data bears that out, that emergent percutaneous catheter intervention is better with low-pressure balloon angioplasty preferred over stenting. This is actually in these studies that randomized patients found that 75% of symptomatic patients have early graft occlusion. So if you see chest pain and you identify these changes on EKG and echo, as mentioned, this is more common than many people originally considered. Moving to the next potential diagnosis, bleeding after cardiac surgery. Now all patients bleed after cardiac surgery. Bleeding for significant bleeding happens quite frequently, even 2% to 10% depending on the operation, and this increases morbidity and mortality. What's the cause of bleeding requiring re-operation? Typically technical factors, but coagulopathy is related and sometimes both. Now you might wonder, well, we have chest tubes in, bleeding is going to be obvious in cardiac surgery. That's not always the case, as the chest tubes might not always be draining, and so bleeding itself can be a difficult diagnosis to make. There are multiple causes of bleeding during cardiac surgery and after it. As mentioned, technical factors are usually what leads to bleeding and need for re-operation, but the sites of bleeding are as follows in causes. Incomplete surgical hemostasis, now some of these areas will have more robust bleeding rates than others, an anastomotic site, for instance, will have significant amount of bleeding, mediastinal vessels, chest wall, particularly with harvesting for coronary bypass grafting, and cannulation sites. There are also coagulation factors, so preoperative anticoagulants and then pre-existing coagulation factor deficiencies, heparin that's still present from surgery, hypothermia or acidosis, platelet abnormalities, or fibrolysis. Everyone bleeds after cardiac surgery, but what is considered above normal bleeding that should be considered significant bleeding and potentially needing re-operation? There's no consensus over these numbers, but as a general guideline, this is typically how many people think about it. So 150 in the first 30 minutes, 250 in the first hour, 150 in the second hour, 100 subsequently. And if that doesn't slowly taper off, then there's an issue that needs to be addressed either surgically or medically. Certain procedures place patients at higher risk of bleeding, and these are the more invasive longer bypass runs shown here. Treating coagulopathy is imperative, correcting hypothermia, acidosis, hypertension, and hypocalcemia, and guided by labs and the thromboelastogram to guide component therapy. Next on the differential diagnosis for hemodynamic instability is cardiac tamponade, and maybe the most important diagnosis for clinicians to recognize and treat, because it can lead to almost full recovery if dealt with promptly. And cardiac tamponade is quite common. Effusions after cardiac surgery are estimated to be incident in greater than 60% of patients. Tamponade is a clinical diagnosis if you have a PA catheter in, you'd see equalization of pressures and a setting of hemodynamic instability. Tamponade occurs 2% to 4% of all cardiac surgical procedures, and it can occur throughout the perioperative period, early after surgery, or within 72 hours, or even up to two weeks after surgery. Transthoracic echo is useful, but sensitivity is only estimated at around 60%, because effusions after cardiac surgery might be small, posterior, and difficult to image, but hemodynamically very impactful. In this slide, I share a video of an anteriorly located effusion, bloody effusion, that is causing tamponade physiology by echo, which is reducing right ventricular filling in diastole. So let's break for a quick question. This is the same patient we mentioned earlier, 60-year-old female with type 2 diabetes, hypertension, exertional angina, post-op day zero, after CAB AVR. She is still intubated from the operating room. En route to evaluating her at the bedside for hypertension, she arrests. You diagnose pulseless ventricular tachycardia. According to STS, Society of Thoracic Surgery guidelines, you direct your ICU team to do the following. The Society of Thoracic Surgeons in 2017 published guidelines stating that the post-cardiac surgery patient has different needs in ACLS, and likely because tamponade is so much higher on the differential diagnosis, this algorithm is meant to reduce the overall amount of epinephrine given to a struggling heart after cardiac surgery, treat reversible arrhythmia, which is also common after cardiac surgery, which we'll talk about in a minute, but in particular move to emergency re-sternotomy sooner than previously did to recover this high incidence of patients that have cardiac tamponade after cardiac surgery. So our last possible diagnosis for hemodynamic instability after cardiac surgery is arrhythmia. You might be thinking, well, this is an easy diagnosis because we have the EKG, but I'll say that it's not always so simple, particularly when you can have other things going on at the same time that we've mentioned, bleeding, for instance. And it turns out that arrhythmia, not just related to these other pathologic conditions, is highly common after cardiac surgery. Intravenous ventricular tachyarrhythmia occurs 15-40% of patients. Sustained ventricular tachyarrhythmia is less common but more devastating around 1% of patients. Bradyarrhythmia is also common. It's usually transient and potentially recoverable, but permanent pacing is required in a large amount of cardiac surgery patients, about 2-4%. Sick Sinus Syndrome or Atrium Ventricular Notal Injury are common pathologies. Now let's transition into Part 2 of this lecture focusing on mechanical devices after cardiac surgery. Because arrhythmia, and in particular Bradyarrhythmia, is common after cardiac surgery, temporary epicardial pacing wires are often placed. Let's first talk about some diagnostic uses of these epicardial pacing wires. I'll show in a second how atrial wires can be used to create an atrial electrogram to identify P waves when they are not so clearly visible on the EKG. In this way they can also differentiate junctional tachycardia from SVT. You can also identify various degrees of heart block and sinus node dysfunction. This is a common rhythm strip that you might see taking care of a patient in the cardiac surgery ICU. Is this sinus tachycardia? Can you easily see P waves? In this same patient you can use a pacemaker to track atrial beats and diagnose atrial flutter with 2-1 AV conduction. But as I mentioned, these epicardial pacing wires are not placed for diagnostic purposes but for therapeutic uses after cardiac surgery. And here is just a list to reiterate those functions. Pacing for sinus node dysfunction that might occur after surgery. Temporary treatment for AV conduction defects. Hopefully that will get better. Sometimes temporary overdrive pacing is required to terminate reentrant tachyarrhythmias. You might attempt to establish AV synchrony. Or some might argue that you could prevent postoperative atrial arrhythmia. But this last point is a bit controversial. Daily maintenance of temporary epicardial pacing leads is necessary and should be done thoughtfully and cautiously. Typically you assess three different things every day. The underlying rhythm, the sensing threshold, and the capture threshold. However, this is to be done with caution. In someone who is known to be pacemaker dependent and with bradyarrhythmia below the pacing wires, checking for an underlying rhythm could be dangerous. So this should be done with caution. But knowing the sensing threshold and the capture threshold is critical over time. In particular, in patients that might have underlying bradyarrhythmia and be pacemaker dependent. As thresholds increase, i.e. more voltage is required, this might be an early sign of pacemaker failure that you can anticipate and then bring someone to the EP lab or cath lab for pacemaker placement if they are going to be requiring this long term. On that last slide I said to consider checking the sensing and the capture threshold every day. What is the difference between those? Let's do a question to demonstrate the difference between sensing and capturing. You have this rhythm strip. There is a problem with the pacemaker. Diagnose it. Is it lead malfunction, over-sensing, endless loop tachycardia, or crosstalk? The correct answer is C, endless loop tachycardia. This is an example of a sensing problem. Most pacemakers are set to inhibit. In an inhibit function, the pacemaker has to sense the native beat. If it is sensed, then the pacemaker will be inhibited. Here, this is a sensing issue. The sensing problem with endless loop tachycardia is that the pacemaker is interpreting the native ventricular pacing spike as an atrial depolarization and then triggers another ventricular impulse. This creates a tremendous problem with repeat ventricular excitation and can put a patient into a worse ventricular dysrhythmia and ventricular tachycardia or ventricular fibrillation. It is a failure of sensing. Timing needs to be adjusted or switching entirely to a different mode. Troubleshooting these pacemaker problems is an incredibly important part of post-cardiac surgery care as these problems end up being pretty common. Pacemaker problems break up into two large groups, output dyssynchrony and output failure. Output dyssynchrony is similar to endless loop tachycardia where there is an undersensing or inappropriate sensing pathology. Typically, this pacemaker is not recognizing the underlying native beat and then competes with the patient's own underlying conduction system creating more beats than is necessary in setting the patient up for decreased cardiac output, RNT phenomenon, and increased cardiac work when you're hoping that the heart is able to recover after surgery. This needs to be constantly addressed. Output failure is where there are no appropriate pacing spikes on EKG and for a pacemaker-dependent person can certainly be quite dangerous. The etiology of this can be due to lead malfunction, unstable connection, insufficient power, or the opposite of undersensing which is oversensing where the pacemaker is interpreting native conduction when there really isn't any and thus inhibiting the pacemaker output. So that's really important to diagnose and overcome. There's also phenomenon of crosstalk inhibition. At the end of this slide set, you'll find an appendix with a little bit more discussion on some of these problems for you to look through on your own time. Moving from pacemaker devices in the ICU, we'll talk now about these mechanical assist devices. So we'll start talking about an intra-aortic balloon pump that's placed typically in the femoral artery or sometimes in the subclavian artery and sits proximal to the left subclavian takeoff to help deliver counterpulsation to assist a struggling heart. Here's a demonstration of where an intra-aortic balloon pump sits, as I mentioned, distal to the left subclavian in the aorta, usually placed through femoral but also has been placed through a right subclavian. And you see here during inflation, usually helium gas, this is importantly timed throughout diastole and then rapidly deflates during systole, creating counterpulsation. Physiologically, this increases diastolic pressure for coronary filling and coronary flow and right before diastole, with deflation, it decreases afterload and decreases myocardial oxygen demand to overall increase cardiac output, decrease cardiac work, and improve O2 supply demand of the heart. The balloon pump was given a Class II reclassification by the FDA in 2013 for the indications of cardiogenic shock after MI or post-cardiopulmonary bypass and cardiomyopathy, that's severe unstable ischemic heart disease as a bridge to surgical intervention. Notably, it's not currently indicated in sepsis with depressed cardiac function, requires further study and actually submission to the FDA if used in that setting. It has been used to decompress the left ventricle in LVAD patients, although this is controversial due to concerns that it might interfere with forward flow function from that device, as the LVAD is a continuous flow device without systole or diastole. There are important contraindications to the placement of an intraaortic balloon pump and they are seen here. The critical care team plays an immensely important role in the management of aortic balloon pumps. The first major role is identifying problems with timing and optimizing them. As timing is critical to the proper function of the intraaortic balloon pump, The trigger is the event set on the pump using to synchronize inflation and deflation within the cardiac cycle. The R wave on the EKG is typical. Systolic pressure upstroke from the arterial waveform is also used, particularly if there's problems with the EKG. Pacer wires and pacer spikes have also been used to trigger the balloon pump. Pacer wires and pacer spikes have also been used to trigger the balloon pump. But importantly, as one might imagine, incorrect timing of the balloon pump ultimately increases cardiac work, rather than the counterpulsation and physiologic goals required. Timing is critical and first assessed just using the arterial blood pressure waveform. Ideally, you will have a waveform that looks like this in red, where that first peak is the systolic pressure peak, and then you see diastole. Once the aortic valve is closed, then the balloon pump inflates and augments the diastolic blood pressure. The second peak here on the slide is the counterpulsation diastole, as opposed to what it would look like in purple here if you did not have the balloon pump. So there is a second counterpulsation of flow in diastole provided by the balloon pump. And then appropriately timed deflation occurs rapidly to then cause a very low valley here before the next systolic ejection and peak to effectively decrease afterload, as I previously mentioned. With inflation timing, the goal is to produce a rapid rise in aortic pressure once the aortic valve is closed. It should be just prior to the dicrotic notch. This results in an augmented peak diastolic pressure that would be certainly greater than the unassisted peak systolic pressure. With deflation timing, the goal is to reduce aortic end diastolic pressure right before the heart ejects to decrease afterload. So the assisted end diastolic pressure should be less than the unassisted end diastolic pressure, hopefully about by 15 mmHg. The peak systolic pressure following an assisted beat should be less than the unassisted systolic pressure by around 5 mmHg. So let's move now into talking about some timing errors, how we diagnose them, and how we fix them. The first error is early inflation, and this was one of the most problematic timing errors. Here, the balloon pump engages, it inflates prior to the end of systole. Thus, it causes the premature closing of the aortic valve, and ultimately this decreases cardiac output and increases cardiac work. Here, you would want to alter what the trigger is, or try to cause the trigger to occur slightly later, closer to the natural end of systole. Second error, late inflation. This isn't quite as problematic as the first problem of early inflation, but late inflation is also an issue. So here, diastole has already started, and then the balloon inflates. Essentially, we're losing a lot of potential benefit. Here, you see in the grayed out red line, we're losing what could be much more effective counterpulsation and higher blood pressure in diastole. This will make the balloon pump much less effective. However, it won't worsen what would have otherwise been baseline cardiac output, as we saw with early inflation. Late inflation, therefore, also a problem. Perhaps not as consequent as early inflation. Early deflation is the third potential timing error, where the balloon deflates too early in diastole, resulting in two problems. First, that the diastolic coronary blood flow does not maintain as high of pressure as it could in flow. And then secondly, because it deflates too early, the end diastolic pressure returns to normal prior to systolic ejection, so there is less benefit for afterload reduction prior to cardiac ejection. This is an issue, ultimately, whereby the interiortic balloon pump is just really not providing the benefit one thinks it is, but easy to fix. Late deflation is one of the worst timing errors, whereby you might be inflating appropriately at the beginning of diastole, but by remaining inflated when systole is starting, this tremendously increases afterload and will worsen cardiac output. So here, it is very important to make sure that deflation is happening earlier and truly within diastole, not during systole. Beyond timing, however, there are other things that are important to recognize when you have an interior balloon pump. Heart rate is often abnormal in these patients, and tachycardia, one might not be surprised, reduces the benefit of an interior balloon pump. It also increases timing problems, as one might imagine as well. However, heart rate is very impactful. Gas can be lost from the balloon over time, and though even you believe that you're inflating, you might not be inflating as effectively as you think. Balloon positioning is critical and can be optimized via TEE as well as chest x-ray. And then balloon volume can also be adjusted, typically on most machines. And then notably, intrinsic cardiac function is necessary for function of an intra-aortic balloon pump. And not to totally replace inherent myocardial function. So managing an intra-aortic balloon pump and expecting some of the complications that might occur. Anticoagulation is controversial, usually considered, but not always required. There's concerns about peripheral embolization and end-organ ischemia. End-organ ischemia in particular, perforation and balloon rupture, hemolysis and thrombocytopenia also can occur. But probably the most significant problem is due to vascular injuries. There are multiple strategies for weaning. Typically inotropic requirements should be reduced while on balloon pumps and should be reduced to almost minimal as you move to wean the augmentation, which you can augment the amount of air volume that goes into the balloon pump, but also then alter the beats that are augmented, 1 to 1, 1 to 2, 1 to 3. Typically as you move to less than 1 to 1 beats, you worry more about coagulation problems and embolisms developing. Moving on from intra-aortic balloon pumps, impella pumps that can be placed percutaneously across the aortic valve into the left ventricle have become more common. They are temporary in nature, typically used around a week, and essentially have an axial flow pump within this catheter. There are various models, but some can deliver up to 5 liters per minute of forward flow. Here, I share a mid-esophageal, long-axis view of the heart with an impella device through the aortic valve situated in the mid-LV cavity, emptying the LV and supporting forward flow. You see here in the left atrium what we refer to as smoke that is indicative of low blood flow state from the LA to the LV, but the LV here is decompressed with forward flow through the impella. Impella pump speeds are set from P0 to P9, and have an interface that allows typically for the device to detect what is the optimal flow to unload in a struggling left ventricle. Alarms on the device manifold include low flow, suction likely, or thought to be wrong pump position, and technical alarms include a higher low purge pressure. The device itself has a self-cleaning mechanism to decrease the risk of thrombus, although it still remains high. Some places advocate strongly for anticoagulation during impella use. Moving on to more durable devices, we'll talk now about left ventricular assist devices. We're now in the third generation of LVADs, however, you still will likely see patients with HeartMate 2 devices. So I'll start there. So this has been the most implanted device with third generation LVADs now creeping up to that number. These second generation devices inflow blood from the left LV apex, and then with axial flow accelerate blood out through the outflow cannula into the aorta with a magnetic field rotating the axial impeller at about 6,000 to 15,000 RPM. Third generation LVADs now have a fairly different design with a rotary pump and centrifugal flow. They're magnetically levitated heartware HVADs and HeartMate 3 devices, where essentially the blood is accelerated via this centrifugal flow from inflow from the LV out the aorta, but at lower RPMs, hopefully with less blood cell trauma. And this is a picture that depicts how this smaller device, too, in its third generation fits within the chest, and the driveline that is still required to externally connect to the controller device with batteries that the patient wears. Moving on, the Centromag is much like an LVAD device. It can be inserted into the RV or the LV, and it's also a magnetically levitated impeller that operates at about 5,500 RPM depending on the desired flow. However, this is not a durable long-term device. This is also a bridge device, bridge to transplant, bridge to LVAD, bridge to recovery. This is what the Centromag console looks like, both for setting the device flow, but also for feedback on physiology. So a couple of quick points on pump physiology that apply to Centromags and LVADs. They're continuous flow devices throughout the cardiac cycle, and flow will be reported on the pump, but it's estimated. And it's essentially determined by the pump speed and power changes over time. These devices are preload dependent. They are also afterload sensitive. It's important to see if it's a left ventricular device, how often the uric valve opens. It might not always open. When it opens, that's indicative of native ejection. This may or may not have a palpable pulse that would be related to left ventricular function. Hemodynamics are best and most accurately measured by invasive arterial lines. They also can be done occlusively using Doppler at the brachial artery, as those correlate with MAP. Echocardiography can be incredibly important to determine preload, and most importantly, right ventricular performance when there's a presence of a left ventricular device. ECHO can also tell you about device position, assess for tamponade, LV size, septal position, and the aortic valve opening. The goal is 70 to 90 millimeters of mercury for a mean arterial pressure, and to preserve RV perfusion. Ultimately, the RV ejection is going to provide the preload to a left ventricular assist device. How do you manage hypotension in a patient that has one of these devices? First, think about all the other common post-cardiac surgery problems that we discussed in the first part of this lecture. Vasoplegia, tamponade, bleeding arrhythmia, they are often also active here and make this management particularly difficult. But in addition, these patients are at risk of RV dysfunction. Blood regurgitation can also cause major issues because it can create a ghost circuit of blood that does not ultimately go to the body. And then there are also other VAD-related hypotensive problems that we'll talk about a little bit more in detail, such as suction events, if the pump speed is set too high, or conditions exist to affect preload, mechanical pump failure, where there's an actual failure in the pump, or malposition of the VAD. This is what a suction event looks like, where either preload is dropped, bleeding is occurring, where the LVAD setting is too high, or a combination thereof, of where you can go, for instance, from a picture on the left, increase the RPMs and the speed of the LVAD, and empty the LV too much, therefore resulting in a suction event. Here's a second suction event here, seen on the bottom. There can be hematologic problems with LVADs, procoagulant and anticoagulant problems. Long-term anticoagulation is indicated for these patients. They also typically acquire von Wildebrand factor deficiency. Pulsatility bleeding can be a huge problem, as the device itself may be causing hemolysis in a major thrombotic event. Ultimately, that might also be a pump thrombus in arterial thromboembolism, resulting in requirement to change out the device. These are the key parameters to follow on the LVAD consoles, pump speed, pump power, pump flow, and the pulsatility index. Pump flow, as I mentioned above, is an estimated value of volume running through the pump calculated from the power RPM and the hematocrit. The pulsatility index is calculated over time with changes in flow and can help one determine what volume management needs to be done. Here's one of the old HeartMate 2 modules. These are the typical operating ranges in some of the major devices currently being used in patients. So if you have a pump power problem, how do you troubleshoot it? First, you have to think, is it decreased power or is it increased power? And they yield these differential diagnoses that you need to address almost immediately, as this can be a life-threatening situation. Outflow alarms and troubleshooting them are the following. If you have a low flow rate, you have to look at the RPMs. If you have low RPM as well, you have to worry about decreased preload due to hypovolemia, RV failure, tamponade, or an inflow cannula problem. You could also have increased afterload with this or an outflow cannula obstruction. As you can imagine, these are potentially life-threatening issues that have to be immediately dealt with. The pulsatility index, as I mentioned above, is a representation of change in flow over time and is calculated by the machine in multiple different ways in a proprietary way. As the left ventricle contracts and relaxes, the pump increases and decreases, adding that degree of pulsatility. And this is often related to volume status. By seeing changes in the pulsatility index, it can help understand trends that then explain the diagnoses leading to high-power, low-power alarm problems. So last mechanical support device I'll briefly discuss is ECMO, which is used as a bridge to recovery, transplant, or another more durable device. The most common configuration for peripheral VA ECMO involves drainage from usually the femoral artery to the oxygenator and pump and back into a femoral artery. Depending on the clinical situation, however, you might also see central VA ECMO cannulation, where the heart directly is cannulated for venous drainage and there's direct arterial flow in the ascending aorta. The typical components of a VA ECMO circuit are shown here. Flow rates of VA ECMO may be anywhere from 2 to 3 or 4 to 6 liters per minute, up to totally replacing the patient's cardiac output. All gas exchange is delivered to the arterial circulation in this configuration. In peripheral cannulation, one has to also make sure that the right hand and the right side of the brain appropriately being oxygenated, depending on a potential mixing point from the peripheral ECMO flow and the ejection from the native heart. Oxygen flow rates as well as the sweep flow that determines CO2 levels need to be constantly addressed and manipulated. Experienced perfusionists are required at the bedside for ECMO support. Flow rates shouldn't fall below 2 liters for any prolonged amount of time because clot formation can occur, and typically heparin drips are considered for VA ECMO to maintain ACTs of 160 to 200. Oxygenation management should involve examination of the cannulated leg, and oftentimes there's an antigrade catheter to make sure that the cannulated leg gets appropriate blood flow. The transmembrane gradient across the oxygenator should not be near 150, which would be concerning for clot formation. If the heart is dysfunctional, the left ventricle must be supported in some way. Otherwise, the LV will dilate and the heart can be injured within the retrograde flow of peripheral ECMO and not create an environment for appropriate cardiac recovery. How do you troubleshoot ECMO? There are many nuances to this management, but a couple of important considerations. ECMO flow is decreased, but pump speed unchanged or increased? You have to first think impaired venous drainage, which could be due to many different problems – hypovolemia, bleeding, a poorly positioned or kinked drainage cannula, or even maybe excess pump suction, or maybe the pump speed itself is too high, or there's increased intra-abdominal pressure because the patient is coughing. Suck down can be a really big problem and lead to flow that someone is dependent on for perfusion. The basics of VA ECMO troubleshooting include checking the catheter, maybe repositioning the patient, giving volume, reducing pump speed, and maybe even the addition of another cannula. Another common concern is ECMO flow decreased, but pump speed unchanged or increased. In this situation, you have to worry about thrombus in the circuit, you want to look at membrane pressure, and look at the return line to make sure there's no kinking. Patients placed on VA ECMO also typically have multiple other concerns. Bleeding is often an issue. circuit, but also the patient. Limb ischemia is a typical and can be a potentially devastating problem. Hemolysis, motor failure, disconnections, and then also a very scary air embolus. Anytime there's anything connected to the circuit, if air is introduced into the ECMO circuit, it can be disastrous. Thank you so much for your time and attention.

cardiovascular surgery complications

mechanical devices

differential diagnosis

decompensated patients

pacemakers

ventricular assist devices

intra-aortic balloon pumps

extracorporeal membrane oxygenators

timing errors