Mechanical Circulatory Support
===

[00:00:00] Behind the knife, surgical critical care. Dominate the ICU

Welcome back to Behind the Knife's Surgical Critical Care Review, where we discuss essential critical care know-how in a case-based format. I'm Patrick Georgoff, and I'm joined today by Dr. Nick Tieman. He is an associate professor of surgery, co-director of the Cardiothoracic ICU, and director of the ECMO program at the University of Colorado.

Nick, today we are covering mechanical support in the ICU, a topic that doesn't come naturally to most general surgeons. I know for me, the cardiac ICU was a big black box until I was able to spend some time there, and that's why I think this is such a great episode

I, I agree. This really appeals to my inner tinkerer because these devices are, are so much fun to think about and play with.

Now, importantly, we are not gonna talk about ECMO today. That deserves its own scenario and, and is included separately in this review series. So why don't [00:01:00] we start with the least invasive and hot take, the least effective, the intra-aortic balloon pump?

I can't wait to hear why this is a hot take. So let's frame the case.

We have a 55-year-old male. He's on the golf course when he has an episode of crushing chest pain, and he collapses. His golfing buddies perform CPR, and they call 911. Upon their arrival, he has had return of spontaneous circulation but continues to have chest pain. He is emergently transported to your hospital and brought to the cardiac cath lab where he is found to have a ST elevation myocardial infarction due to a proximal LAD artery occlusion.

He undergoes successful stenting, but has persistent hypotension and a lactic acid of 4.2 after it's all said and done. So we're concerned about cardiogenic shock. The interventional cardiologist places a balloon pump and transfers him to the ICU for further management. Nick, what do you think about this?

Well, I would say that this cardiologist didn't read the IABP Shock 2 trial. This landmark study was published in the New England Journal in 2012, and they randomized 600 patients with [00:02:00] cardiogenic shock complicating an acute myocardial infarction to either balloon pump or no balloon pump. Importantly, all of these patients had planned early revascularization with PCI or CABG.

The investigators found no difference in the primary outcome of 30-day all-cause mortality or any of the secondary outcomes that they looked at. Despite this, balloon pumps are still frequently used, so all of us need to know how they work. Yeah. So Patrick, give us a rundown.

Yeah, here we go. Okay. Uh, balloon pumps are placed percutaneously, usually through the femoral artery, although occasionally they can be placed through the axillary, uh, artery.

This is usually for patients who are awaiting heart transplant. And importantly, the tip of the catheter should sit two to three centimeters distal or below the left subclavian artery. So this can be measured on a chest X-ray, and you should confirm its position every day. Now, the balloon is triggered most commonly by the EKG or the waveform on the arterial line, but it can also be triggered by pacing spikes if the patient is being paced or time-based if the patient is asystolic and arresting.

The balloon inflates during diastole to displace the [00:03:00] blood in the aorta and augment blood flow to the coronary arteries, which as we know are perfused during diastole. But it's not like the blood is only displaced into the coronaries. The entire column of blood is displaced proximally and distally, augmenting the cardiac output to other important structures like your brain, the kidneys, and the rest of your body.

That's right. And then when the balloon deflates during systole, this results in decreased afterload for the struggling left ventricle. This all makes intuitive sense, but sadly the data hasn't demonstrated a clinical benefits to its use. In any case, here we are, and our patient has arrived in the cardiology ICU with his balloon pump.

Despite the augmentation of his cardiac output, he is still requiring inotropes and vasopressors, and his urine output has been pretty minimal. So going back to our original case, what should our cardiologist have done if they were more up-to-date on the literature?

Sure. So the DANGER-SHOCK trial showed us that in patients with STEMI and cardiogenic shock, the placement of an Impella reduced mortality in patients compared to standard care.

So whether we placed an Impella as the initial form of temporary [00:04:00] support, or we escalated due to the inadequate support from the balloon pump, let's say we got there now, we have the Impella in place. So what is an Impella and how does it work?

Yeah, so an Impella is a percutaneous microaxial flow pump, which is also the language you will see in the literature when journals want to avoid using the trade names.

But there really are no other percutaneous microaxial flow pumps besides the Impella, so really it's an issue of semantics.

Yeah. And the Impella is placed via the femoral, uh, artery or axial artery and sits across the aortic valve. And the inlet to the pump sits in the left ventricular cavity, and the outlet is in the ascending aorta.

So Greek history nerds love to think about this 'cause all of it is based on Archimedes' screw. And the last time, Nick, I saw the Archimedes' screw was at a children's museum where we were cranking it around and the water was getting cranked up to the top. And, you know- It's the same

thing. It's the exact same device

it rotates within a tube propelling water, or in, in historical case grain, uh, as well, or in this case blood. And the blood is pumped out of the LV into the ascending [00:05:00] aorta. And unlike the balloon pump, which provides indirect unloading of the LV via a reduction in afterload, the Impella provides direct unloading of the left ventricle.

So that's a key difference between the two devices. So in what scenarios might you take care of a patient with Impella, Nick?

Yeah. So, so as a surgical critical care fellow or surgical intensivist, you'll see these patients in, in several different situations. The, the first is our patient, which is cardiogenic shock following an acute MI.

Uh, the Impella, like we talked about, may be an escalation from a balloon pump or it could be the first-line device placed in the cardiac cath lab. Uh, another cath lab use of Impellas is to provide temporary hemodynamic support during high-risk PCI. In most of those cases, however, the Impella is removed at the conclusion of the procedure and the patient won't have it still in place when they hit the ICU.

Uh, another common use of Impellas is as a left ventricular vent, uh, in patients who are on veno-arterial ECMO, which the combination of the two is also called EC-Pella. [00:06:00] The nice thing about that configuration is that the Impella offloads the left ventricle to promote recovery, but can also support the LV as part of a de-escalation from ECMO.

Uh, some centers are pretty aggressive about using Impellas intraoperatively for high-risk cardiac surgery to aid in separation from cardiopulmonary bypass and to minimize the vasoactive support that the patient needs in the early postoperative period. Finally, uh, many ICUs at transplant centers will take care of patients with chronic heart failure who have an Impella in place to achieve a higher priority on the transplant list awaiting a heart transplant.

Now, there are a couple of different Impellas that you might encounter in the ICU. The first is the Impella CP. Nick, what does CP stand for?

Uh, it stands for cardiac power, so just a totally nonsensical trade name.

Uh, I actually kind of, I love it, cardiac power. All right, so this is most commonly placed by interventional cardiology in the cath lab.

So you may receive that patient in the ICU who rolled out of the lab with an Impella CP. It's usually placed percutaneously via the femoral artery, and this can [00:07:00] achieve flows up to three to four liters per minute. The downside of this device is that due to its small size and the speed which is required to achieve adequate flow, it's common for patients to experience he- hemolysis, which can manifest as anemia, thrombosis, and/or renal failure.

So the other Impella device that's important to know about is called Impella 5.5, uh, which is a direct shout-out to its flow rates of 5.5 liters per minute. This is surgically placed most commonly into the axillary artery via surgical cutdown and, uh, oftentimes sewing onto the graft itself. So this is a bigger device and it can achieve flow rates, again, you guessed up to 5.5 liters per minute.

And hemolysis and other complications are more rare, uh, due to its bigger size and lower s- pump speeds.

Yeah, and in terms of management of these pumps, uh, if you're taking care of somebody in the ICU, there's, there are 10 speed levels ranging from P0, which is the slowest, to P9, which is the fastest. The exact flow at each P level is a, is a calculated value from the pump, so it's gonna vary based on the patient's volume status and wh- how the pump is [00:08:00] positioned.

Uh, most of the time P2 is the lowest value that you're gonna run the pump at unless you're actively trying to remove the pump.

Right. So these, these numbers are not actual, uh, amounts of flow. They're all relative to each patient, right, Nick?

Yeah, that's correct. Yeah. You, you can't, uh, directly relate one, uh, patient to another.

So what do we need to keep an eye on, on these patients who have an Impella in the unit?

So it's really important that the device is properly positioned. Uh, echocardiography, just a surface echo, POCUS is the best way to assess this. You want the pump to be angled so that the tip is pointed down towards the apex of the left ventricle as opposed to towards the ventricular septum or in any way interacting with the mitral valve apparatus.

The tip of the CP should be about three point five centimeters from the aortic valve, and the tip of the Impella five point five should be about five centimeters from the aortic valve. Fortunately, the device is fairly stable, and so once it's in a good place, it doesn't usually move too much.

All right.

A couple of other rapid-fire items about Impella. So the console of the Impella [00:09:00] itself displays waveforms which help with determining that the pump itself is properly positioned, functioning correctly, and set in an appropriate speed. Now, the details of this are probably beyond the scope of this episode, but there are resources available online if you want to better understand the waveforms to help you manage the patient that may show up in your ICU tonight.

There's a purge line as well. This line running into the Impella delivers pressurized fluid that prevents blood from entering the motor housing. Historically, this contained heparin but is now sodium bicarbonate-based. And patients with Impellas should ideally be anticoagulated. The official recommendation of the company is to target an ACT of a hundred and sixty to a hundred and eighty seconds.

Perfect. And so switching gears now, there's another temporary left ventricular assist device that's worth mentioning, and that's the tandem heart device. Now, this is, uh, really just a cannula that's placed in the femoral vein. It's placed across the right atrium into the left atrium via a transseptal puncture.

The blood from the left atrium is then drained out via a centrifugal pump and then [00:10:00] returned into the femoral artery. Because of this design, it's particularly useful for cardiogenic shock that's due to mitral valve pathology. So if you have severe mitral stenosis or a thrombosed prosthetic mitral valve, you can drain the blood that's directly backed up into the left atrium and then return it into the systemic circulation.

Yeah. This kind of blows my mind thinking about this, and it's not as common as some of the other things we are gonna talk about in this episode. So what's the history of this and where do you see it?

Yeah, so y- so y- you're right. I mean, they're not used as frequently, uh, as the other devices or even as they used to be.

Tandem hearts definitely were used more frequently prior to the advancements in Impella that we talked about, and also prior to expansion of ECMO programs. You know, the, uh, many places n- that now have ECMO programs, before that they used to use tandem heart, uh, pumps very often. And, uh, and it used to be kind of a fully self-contained system with a pump and everything else, very easy, you know, take out of the box.

Uh, but the issue is now the company that manufactured that whole system is, it basically closed down that part of their portfolio. So you can still find the cannula and you can still cobble this together, [00:11:00] but you kind of have to have your own different pump and circuit to make that work.

All right, back to the patient.

Uh, despite several attempts to wean him off the Impella, he continues to require, uh, a pretty significant support. His LV's failing. Uh, farther workup has revealed that he is unfortunately not a candidate for heart transplant. He's had a recent history of colon cancer that we didn't mention earlier, so he's off the list.

So Nick, what are our options at this point?

So if he's not a transplant candidate and he can't be liberated from a temporary LVAD, then he really only has one option besides palliative care, and that's the, uh, durable LVAD. And so currently, the only device on the market for durable LVAD support is a HeartMate 3, so he should get a, he should get one of those.

So Patrick, talk us through what that consists of.

Yeah, certainly. The HeartMate 3, much more common. This is absolutely important to know. And it's placed via a median sternotomy or a minimally invasive approach with a left anterior mini thoracotomy and either a hemisternotomy or right anterior mini thoracotomy.

And this is a magnetically levitated centrifugal pump which [00:12:00] is attached to the LV, uh, specifically the LV apex, and it sits within the pericardium. So blood is pumped out of the LV into a graft, which is anastomosed to the ascending aorta. So a driveline or a power cord is tunneled out of the body from the abdomen or subcostal margin and connected to the power source.

Most of these operations can be performed on a cardiopulmonary bypass, but with the heart still beating, amazingly, unless there's a valvular abnormality that needs to be addressed at the same time. That's really important for the postoperative management of these patients, since the RV is often not great and really hates being arrested.

Yeah. It's really important to remember that these are left ventricular assist devices. And so while our patient with an LAD STEMI may have isolated LV dysfunction, many patients who get LVADs also have some degree of RV dysfunction. So most of the art of caring for LVAD patients is focused on supporting the RV because the LV dysfunction is now fixed, but the RV dysfunction obviously isn't.

Uh, these patients will often be on inotropes, and they'll occasionally be on [00:13:00] inhaled vasodilators. And finding the right pump speed is important. Ultimately, the average speed for the HeartMate 3 pump is going to be about 5,400 RPMs, but that is n- really not often the case initially when it's placed in the operating room.

Uh, if you- the pump is too slow, the, the cardiac output from the LV won't be enough to support the body. But if you have the speed set too fast, you can cause suction events where the LV cavity collapses around the pump inlet, and that can pull the septum over, and that can cause more RV dysfunction. So Patrick, talk us through that.

How do you diagnose and manage these suction events?

Yeah. I can only imagine what a suction event feels like when your own LV gets sucked in from the inside. It'd have to feel terrible. So a, a quick and easy way to diagnose a suction event is to look at the arterial line waveform. Oftentimes, there will still be some pulsatility, uh, even with the continuous flow LVAD functioning normally.

Now, if the patient arrived from the OR with 10 to 20 points of pulsatility but is now hypotensive with a flat waveform, it's a safe bet that the patient is having a suction event.

It's worth [00:14:00] pointing out here that the HeartMate 3 device actually has a built-in, quote-unquote, artificial pulse, uh, which is created every two seconds by fluctuations in the RPMs to basically promote washing of the pump.

So you will see some deviation in the arterial line waveform, but that's not exactly what we're talking about here.

Right. So the other way to identify a suction event is to perform an echo, uh, to assess the size of the LV cavity. Now, the patient probably left the OR with the pump speed well-balanced to ensure that the LV is decompressed but not completely empty.

Uh, but things often change once you get to the ICU, and suction events are usually caused by hypovolemia, RV failure, as you'd mentioned, Nick, cardiac tamponade, uh, uh, even a speed that's set too high or some combination of these things. A lot of them can happen at once. The first thing I usually do is drop the speed by a couple hundred RPMs to see if that allows the LV to fill, to break that suction event, and to see if that LV returns to normal function, at least grossly, uh, on the ultrasound.

And if that doesn't work, then additional investigation needs to be done to determine whether the patient needs [00:15:00] volume, additional RV support, or even operative exploration in the worst-case scenario.

Got it. So now what do we do with all these numbers that we see on the LVAD console?

Yeah. And this can be overwhelming when you're looking at this for the first time.

The first is speed and RPMs, which is the only parameter you can actually change on the pump, right? So that's a bit easier, simpler. Uh, the power is an indicator of how hard the pump is working. Changes in the pump power can reflect changes in afterload or can be a sign of pump thrombosis or other serious issues.

Flow is displayed in liters per minute, but it is important to remember that this is also a calculated value, and in particular, it depends on the viscosity of the blood as it is determined by the patient's hematocrit. So if the patient's current crit is not entered into the con-controller appropriately, uh, the flow displayed will not be accurate, which is really important not to forget.

Uh, the final parameter is the pulsatility index, which is a marker of how much native LV contractility is present. So a change in pulsatility index [00:16:00] should alert you that something, uh, may be wrong. But unfortunately, this is not always predictable because the same physiologic derangements can cause either an increased or decreased pulsatility index, depending on what else is going on with the patient.

So a little more nuance there.

Okay, so putting this all together, this is a pretty good summary of temporary and durable mechanical circulatory support in the ICU. Anything else we need to talk about?

So we didn't talk about RV support at all. Uh, since we're running out of time, I'll just mention that there may be two devices that you see in regards to temporary RV support, and that's the Impella RP, which functions similarly to the left-sided Impella to propel blood from the RV across the pulmonic valve.

The other option is a dual-lumen cannula. It drains blood from the right atrium or ventricle to an external pump, which then returns the blood through the distal lumen, which is located in the main pulmonary artery. So the advantage of this approach is that since the blood is extracorporeal, an oxygenator can be incorporated into the circuit to assist with gas exchange as well.

But, uh, y- that's just [00:17:00] a, a very, very high-level review on, on the RV side. Those are less common devices compared to LV support. All right, Nick, let's wrap this up with some quick hits.

Perfect. So first, the management of cardiogenic shock is often stepwise. Depending on the severity, medical management with inotropes is often attempted initially, followed by intra-aortic balloon pump or Impella, followed by ECMO.

Great. Next, intra-aortic balloon pumps augment cardiac output and lower LV afterload. They inflate during diastole, which increases coronary blood flow and augments flow to other organs, and during systole, they deflate to lower the LV afterload.

However, clinical trials have shown no benefit to intra-aortic balloon pumps in cardiogenic shock.

Now, you still have to know about them because despite that, they're still used frequently for that purpose.

However, Impellas have been shown to improve outcome in the setting of cardiogenic shock. They can also be used for high-risk PCI as an LV vent in patients on VA ECMO to aid in perioperative management of high-risk cardiac surgical patients [00:18:00] and as a bridge to heart transplant.

Next, the HeartMate 3 durable LVAD can be used for patients who can't wean off of temporary mechanical circulatory support.

And last, temporary right-sided mechanical circulatory support devices include an Impella and a dual-lumen veno-pulmonary arterial cannula. Both of these devices effectively shunt blood from the right side of the heart across the pulmonic valve.

Until next time, dominate the day