Hi, I'm Steve Hollenberg, and I'm going to talk to you today about intraortic balloon pumps. This is a roadmap of where we're going today. We're going to talk about history, we'll talk about physiology, and then I'm going to show you some data. Let's start with a case. A 59-year-old woman with a history of coronary artery disease presents with an acute anterior wall myocardial infarction. Her blood pressure is 80 over 60, her heart rate is 132, and she is intubated for pulmonary edema. So this is cardiogenic shock in the setting of an acute myocardial infarction. What now? At this point, most people would think about percutaneous mechanical support options, and the available devices are listed on this slide. There are intracorporeal pumps, the balloon pump and the impella, and there are extracorporeal pumps, the tandem heart and VA ECMO. Of these pumps, some are continuous, either axial or centrifugal, but the intraortic balloon pump is a pulsatile intracorporeal pump, and that's what we're going to talk about today. This is Dr. Adrian Kantrowicz, the inventor of the intraortic balloon pump. He's not as well known as he should be. He was born in New York City. He went to New York University for his undergraduate training and the State University of New York downstate for medical training. He did an internship in neurosurgery, and then he went off to the war to be a medical officer. He wanted to come back to SUNY downstate after the war, but there were no residency openings, so he was shunted over to Mount Sinai in New York for general and cardiac surgery. He developed an LV bypass pump for mitral valve surgery in the late 1940s, but it was never used. He did a U.S. Public Health Service fellowship with Carl Wiggers, the great physiologist at Case Western in 1951 and 1952, and at the time they did aortic counterpulsation during diastole in animal models. In 1955, again in animal models, he wrapped the diaphragm around the aorta and he paced it to assist the left ventricle. He actually designed a mechanical left ventricular assist device in the 1960s, but it didn't really work very well, so he designed an intraortic balloon pump instead. He was still a clinician, and he did the world's second heart transplant three days after the first heart transplant was done by Christian Bernard in South Africa. In December 1967. This is a slide showing the physiology of intraortic balloon counterpulsation. The tracing with the intraortic balloon pump is shown here in red, and what would happen without the balloon pump is shown in this blue dotted line. The balloon inflates in diastole, timed to the dichrotic notch of the arterial waveform, and then it deflates in systole. So this is the unassisted aortic end diastolic pressure, and you can see that was what the tracing would look like, but the balloon inflates in diastole, it augments diastolic pressure, and then it deflates, that causes a vacuum, and so it reduces aortic end diastolic pressure and reduces systolic pressure of the next beat. So this increased pressure in the aortic root and the coronary artery increases myocardial oxygen supply by increasing the driving pressure for coronary perfusion. And the decreased systolic pressure and end diastolic pressure reduces the afterload and decreases left ventricular work and wall stress. This increases cardiac output a little bit, about 0.5 to 1.0 liters. This slide shows the physiology of intraortic balloon counterpulsation. The balloon inflates in diastole, timed to the dichrotic notch of the arterial waveform, and deflates in systole. So here you see the arterial waveform in red, augmented and unaugmented. So the non-augmented systolic pressure would look like this. When the balloon inflates, it augments diastolic pressure, and when it deflates, it causes a vacuum and decreases the afterload, reducing aortic end diastolic pressure and reducing systolic pressure of the subsequent beat. So the net effect is increased pressure in the aortic root and coronary artery, thus increasing supply by increasing the driving pressure for coronary flow. In addition, afterload reduction and decreased systolic pressure decrease the work of the left ventricular wall and also its stress. Overall this increases cardiac output a little bit, about 0.5 to 1.0 liters. This slide shows normal and abnormal intraortic balloon pump waveforms. The normal waveform is on the left, and I just showed you that. Ideally, the balloon pump inflates just at the dichrotic notch so as to maximize diastolic augmentation. The abnormal waveforms are shown on the right. If the balloon inflates early, the balloon gets in the way of systole, and you can see that you have a decreased augmentation, but also you're increasing left ventricular afterload. If the balloon inflates late, you don't get as much augmentation as you might otherwise get, and the balloon isn't as effective as it might be. If the balloon deflates too early, again, you lose a little bit of diastolic augmentation. And if the balloon deflates too late, then you don't get as much afterload reduction as you might otherwise have gotten. So now, let's talk a little bit about left ventricular physiology. This is a pressure-volume loop for a single cardiac cycle, and you may remember this from medical school. It goes around counterclockwise. It starts with diastolic filling. The mitral valve opens, and the ventricle fills. When the mitral valve closes, the ventricle, the mitral valve is closed, the aortic valve is closed, the ventricle begins to contract, and you have isovolumic contraction. The pressure goes up, but the volume doesn't change until the time that the aortic valve opens. Then you have systolic ejection, decreasing the volume of the ventricle until the time the aortic valve closes. Now both valves are closed, and the ventricle relaxes. You have isovolumic relaxation until the mitral valve opens and the cycle starts again. This slide shows some hemodynamic indices based on pressure-volume loops. The green line represents end-systolic elastance, which is an index of contractility. And the orange line represents arterial elastance, which is an index of afterload. Left ventricular wall stress equals peak systolic pressure multiplied by end-diastolic volume. And myocardial oxygen consumption is determined by the stroke work, that is the area inside the left ventricular pressure-volume curve, and the potential energy that it takes to get up to end-diastolic volumes, and that's this triangular area. And as you can see, some of these are changed by mechanical support. This slide shows pressure-volume loops in heart failure. The normal pressure-volume loop that I just showed you is on the left, with normal contractility shown by the red line and normal afterload as shown by the blue line. With heart failure, the ventricle dilates. You can see increased end-diastolic pressure and increased end-systolic pressure, and thus a lower ejection fraction. The contractility is decreased as shown by a shift of the curve downwards and to the right. And in compensation, afterload increases a little bit to try to maintain systolic pressure. As contractility goes down even further, even an increase in afterload does not restore systolic blood pressure. The end-systolic volume and end-diastolic volumes are very high with a very low ejection fraction, and stroke volume is reduced. The effect of the intra-aortic balloon pump on pressure-volume curves is shown in this slide. So we start here in the light blue, and we go to this darker blue. It's a volume displacement pump that provides, as I mentioned, about half a liter per minute of cardiac output. So with intra-aortic balloon pumping, the left ventricular pressure goes down. The afterload shifts to the left, that is, afterload decreases. And this causes the stroke volume to increase just a little bit. So you decrease left ventricular pressure. You decrease afterload and arterial elastance. You decrease end-diastolic volume and stroke work because the pressure, the left ventricular pressure is lower, so stroke work is somewhat lower. And oxygen consumption, again, the stroke work may be bigger, but the area of this triangle, the potential energy, is lower with intra-aortic balloon pumping than it was. And there's no change in contractility. Here are the indications for intra-aortic balloon pumping. The primary indication is cardiogenic shock, secondary to acute myocardial infarction, and 75% of patients improve hemodynamically. We're going to talk a little bit later about comparisons of balloon pumping to other methods of mechanical support. Post-cardiotomy shock after cardiac surgery is another indication. And some patients with refractory stage D heart failure may be considered for intra-aortic balloon counterpulsation. Contraindications to intra-aortic balloon pumping are shown on this slide. Significant aortic regurgitation is a contraindication because you wouldn't want to increase diastolic pressure and thus increased regurgitation. Multiple vascular issues and aortic aneurysms are contraindications, as is bleeding and preexisting infection. This slide will give you a rough sense of the incidence of complications. Ischemia occurs in about 8% to 10%, with about half of that being limb ischemia, but other ischemia having to do with thrombosis and cholesterol embolization. Most intra-aortic balloon pumps are inserted in the setting of anticoagulation. Here's the 3% to 5% incidence of bleeding, either due to thrombocytopenia or anemia due to hemolysis. You can have obstruction of arterial flow due to malposition. You can have emboli leading to stroke, sepsis from infection, and or local access site infection. The balloon can leak or it can rupture, and you can wind up with a peripheral neuropathy. I alluded to malposition on the last slide. You want the tip of the intra-aortic balloon pump to be just distal to the takeoff of the left subclavian artery. That gives you maximum diastolic augmentation, but also doesn't obstruct any of the important vessels going to the brain. It's usually implanted under fluoroscopic guidance, but it can be done at the bedside by estimating the distance from the groin site to the sternal notch, and then you evaluate the position by radiography. As I mentioned, if you have it too high, you have the potential to occlude the subclavian artery, and if it's too low, the bottom of the pump can potentially occlude the renal arteries. I alluded to the potential for malpositioning in the previous slide. Ideally, you would like the tip of the catheter to be just distal to the subclavian artery. It's usually implanted under fluoroscopic guidance, but this can also be done at the bedside by estimating the distance from the groin site to the sternal notch, external to the patient. The ultimate position of the intra-aortic balloon pump is evaluated by radiography, either by fluoroscopy in the catheterization laboratory or by a simple chest X-ray. You don't want the pump to be too high because of the potential to occlude the subclavian artery, or too low because of the potential to occlude the renal arteries. So the takeoff of the subclavian artery is usually somewhere around the carina, and here's an example of a pump that's too high. The carina is here in blue, and the tip of the catheter with a radio-opaque marker, as seen in red, is up too high. Here's a pump that's too low. I'm indebted to my colleague, Dr. Benjamin Koenigsberg, for the loan of this slide. It's so low that you almost can't see the tip of the balloon pump in the chest, but it should be up here where the carina is, and it's all the way down here. And here's one that's just right with the radio-opaque tip of the balloon pump right at the carina. Of course, you can also evaluate the position of the pump by augmentation on the waveform, as I showed you before. Sometimes you can see the intra-aortic balloon pump on echocardiography. Here it is in the descending aorta, but the echo doesn't really tell you precisely where it is. Echocardiography is not a useful technique for positioning an intra-aortic balloon pump. So now let's talk a little bit about data. Here are data on the time to intra-aortic balloon placement and survival, showing the unsurprising finding that if you place the balloon earlier, with time of placement less than an hour, results are better than when you take an hour or longer. So the balloon pump makes sense. It has a hemodynamic rationale. So what are the data that suggest that it's actually effective for patient outcomes? I'm going to show you that on the next slide. This is the IABP2 shock trial done by Holger Thiele and his colleagues out in Germany. This trial was completed in 2012 and published in the New England Journal of Medicine. 600 patients with cardiogenic shock and acute MI who were undergoing percutaneous coronary intervention were randomized to intra-aortic balloon pump or not. The mean age was 70 and two-thirds of them were men. The blood pressures were means 90 over 55 with a mean of 69 and 90% of that was supported by catecholamines with a heart rate of 92. The primary endpoint was 30-day mortality powered for a 12% decrease from a 56% mortality to a 44% mortality. And as you can see, this trial was resoundingly negative with a P of 0.92. Now there are a couple of caveats with this trial. The intra-aortic balloon pump was placed after PCI and 87% of the time. So if the PCI made you better, then you might not have needed the balloon pump. 10% of the controls crossed over to intra-aortic balloon pumping, perhaps some of the sicker of those controls and assist devices were placed in a smaller percentage of patients in the balloon pump group than in controls. Now if you look at the subgroup analysis in this group, you find maybe some subgroups where there is suggestion of benefit. So I don't want to oversell subgroup analysis. It should be taken with caution. However, if you look at the groups, if you're less than 50, there was a suggestion that maybe the balloon pump benefit. And if you had an anterior wall MI as opposed to a non-anterior wall MI, maybe intra-aortic balloon pump was a little bit better. So it's possible that younger patients with larger MIs might have seen some benefit, but really the trial has to be regarded as fully negative. Now one of the advantages of other forms of temporary mechanical support over intra-aortic balloon pumping is that counter-pulsation requires pulsation. So over here we have two groups. We have a group with lower baseline power, power essentially being the work that the heart is doing, and higher baseline power. And this is a retrospective study, but in any case, if you had lower baseline power, only 4 out of 13 patients stabilized with intra-aortic balloon pumping, whereas if you had higher baseline power, 10 out of 11 patients stabilized and only 1 decompensated. So take these data with a grain of salt, but it does suggest that the more dysfunctional the ventricle, the less the intra-aortic balloon pump helps. So this slide shows comparison with hemodynamics with left ventricular assist devices compared to intra-aortic balloon pumping in randomized trials. Here are the cardiac index data from three studies of LVADs showing that the cardiac index augmentation is greater with assist devices than with balloon pumping. Here are data showing the mean arterial pressure, again showing that assist devices have a greater augmentation of mean arterial pressure. And finally, the assist devices have a greater decreased decrement in pulmonary capillary wedge pressure. So the hemodynamic effects of other sorts of assist devices are more potent than those of intra-aortic balloon pumping. However, as of yet, randomized control trials have not shown improved survival with mechanical circulatory support. On the top is a meta-analysis from those trials I just showed you, along with another trial called IMPRESS and septic shock, and there has been no demonstration of an augmentation of an improvement in 30-day mortality. If you look at the numbers on the left, the trials are fairly small, but at least so far there are no randomized control trials that show that. And on the bottom is a very interesting study comparing patients in the European Impella Registry where the outcomes were pretty good with a mortality of about 40% with propensity matched pairs from the SHOCK2 trial that I showed you. And again, this shows no difference between balloon pumping in the SHOCK2 trial and those patients in that Impella Registry. Again, not overwhelming evidence that mechanical circulatory support is better than intra-aortic balloon pumping. This is another study looking at 1,680 propensity matched pairs from the NCDR database from 2015 to 2017 comparing intra-aortic balloon pumping to a microaxial left ventricular assist device also known as an impella. Now this is a retrospective study with pairs and should be taken with some caution. However, it's hard to make this out showing superiority for the impella. The intra-aortic balloon pump was favored in mortality and major bleeding, and those results held up whether the device placement was performed before initiation of percutaneous coronary intervention or afterwards. So despite the data that don't say that intra-aortic balloon pumping improves mortality, there are some advantages. It's relatively easy to insert them rapidly. They're not that expensive compared to other mechanical support devices. Intra-aortic balloon pump will run you about $500 or $600. Some of the other devices upwards of $20,000. It augments diastolic pressures and thus coronary perfusion pressures and thus in patients with active ischemia, there's a pretty good physiologic rationale for their use. The augmentation of cardiac output as I mentioned is relatively small and so not really that much of a rationale for patients with low cardiac output absence ischemia. It does reduce afterload a little and sometimes that may be enough. So I mentioned this, but in terms of the level of support offered by different devices, they're listed on this slide. And again, the intra-aortic balloon pump is over here on the left. So if what you need is more cardiac output, you're better off going with one of these devices out there. The most is VA ECMO, the various Impella and Tandem heart devices are in the middle. So if you use intra-aortic balloon pumps to augment cardiac output, your results may not be as good as some of these other devices. Three take-home points. The intra-aortic balloon pump can be useful in some settings, but you want to match the device to the indication. As with all devices, patient selection and timing and patient size, the presence or absence of structural heart disease and comorbidities are the key to success. And you want to make sure that you have a plan about where you are intending to go after placement of the device. Thank you.

intra-aortic balloon pumps

cardiac conditions

mechanical support

history

physiology

effectiveness