Initial experiments aimed at altering the timing of phases of the cardiac cycle originated with animal experiments conducted by Adrian Kantrowitz in the early 1950s in the laboratory of Carl Wiggers at Western Reserve Medical School.⁵ The focus was initially oriented toward improving coronary blood flow rather than augmentation of cardiac output. Appreciating that coronary flow (particularly in the left coronary circulation) occurs primarily in diastole, the concept was to delay the peak systolic pressure pulse to the diastolic phase of the cardiac cycle.⁶ During the subsequent decade, this was followed by animal experiments attempting to use the diaphragm to provide the power for diastolic augmentation by wrapping it around the distal thoracic aorta.⁷ Clauss and colleagues effectuated counterpulsation by withdrawing blood from the circulation during systole and restoring it during diastole,⁸ but it was the work of Moulopoulos and associates⁹ that introduced counterpulsation with a carbon dioxide–filled tube synchronized to the electrocardiogram (ECG) in canines. Kantrowitz subsequently changed the gas to helium, used in modern IABPs because it has only 5% of the density of CO₂ and allows faster inflation-deflation cycles and greater precision in timing. In addition, helium is inert, although it is less soluble and potentially more toxic in case of gas leak in the circulation. The IABP was introduced in humans in 1967 (Fig. 7.1). Initial favorable experience with intra-aortic counterpulsation in critically ill patients in cardiogenic shock (CS)^10,11 led to the first major multicenter trial. This trial demonstrated significant hemodynamic benefit but only a 17% survival to discharge rate.¹² A number of incremental improvements over the next 4 decades resulted in introduction of a percutaneous approach,¹³ a second lumen for guidewire support of balloon advancement through the circulation,¹⁴ increasing automation of the control consoles, and prefolded and progressively lower-profile balloons.¹⁵

Figure 7.1 Adrian Kantrowitz with an early model of a working intra-aortic balloon pump used in patients circa 1967, the year of the initial human experience with the device.

Physiology

The classic concept of intra-aortic balloon counterpulsation involves inflation in synchrony with aortic valve closure, at the onset of isovolumic diastole and the appearance of the dicrotic notch, displacing blood comparable to the balloon’s volume into the peripheral circulation during diastole. To accomplish further unloading, and to prevent interference with left ventricular (LV) ejection, balloon deflation has traditionally begun before opening of the aortic valve and the beginning of LV ejection, although as discussed subsequently, this may not be the optimal algorithm. An example of the effects of balloon counterpulsation on systolic and diastolic pressure is seen in Figure 7.2. The hemodynamic response to institution of IABP is quite variable, and depends on a complex array consisting of the patient’s intrinsic blood pressure, heart rate, heart rhythm, aortic compliance, overall peripheral vascular resistance, intravascular volume status, cardiac function, adjunctive pharmacotherapy, disease state of the coronary vasculature, and degree of preservation of coronary flow autoregulation. In addition, the exact location of the IABP in the vasculature, the volume of the balloon, the frequency of inflation (frequencies from 1 : 1 to 1 : 8 are available depending on manufacturer), and timing of inflation and deflation all play important roles. Thus, the “classic” response consisting of lowering the systolic blood pressure and augmentation of diastolic pressure may not be seen; this classic response is based on the initial experience in CS, which led to a 20% drop in systolic pressure and a 30% rise in diastolic pressure.¹² In fact, systolic pressure can increase secondary to improved cardiac output, can decrease, or can be unchanged, as can coronary blood flow. An important characteristic of the IABP in contrast to most mechanical assist devices is that it contributes to pulsatile flow, with theoretical benefits to end-organ perfusion beyond any actual change in mean flow or pressures.

Figure 7.2 Systemic pressure response to an intra-aortic balloon pump. The console is set to trigger at 1 : 2, hence both unassisted and assisted systole and diastole are seen. Note decrease in systolic pressure with balloon counterpulsation (double arrow) and augmentation of diastolic pressure (area under curve in blue), thus decreasing afterload and increasing inflow pressure to the coronary and other vascular beds during diastole. EDP, end-diastolic pressure.

Diastole and Coronary Blood Flow

Because the majority of coronary flow occurs in diastole, an increase in diastolic pressure has the potential to augment coronary flow as well as flow to other end organs. Diastolic pressure may in fact be augmented substantially, in part because balloon inflation is rapid, yielding an abrupt increase in volume and effecting a rapid rise on the pressure-volume curve. A variety of physical and biologic parameters, including compliance characteristics of the aorta and vascular bed, affect the degree of augmentation.¹⁶ The extent of peak pressure rise has been reported in a range from minimal to near doubling of diastolic pressure.^17,18 However, increase in diastolic pressure and hence coronary perfusion pressure may not result in an increase in coronary flow because autoregulation modulates this potential and because in normal patients, end organs capable of autoregulation maintain flow without significant change.

Despite both several decades of experimentation in animal models and attempts to assess coronary flow in a variety of clinical settings in humans, the effect of IABP on coronary flow remains incompletely defined. Experimental and clinical data fail to show change in flow in native coronary arteries 19,20 or in bypass grafts²¹ regardless of the severity of coronary stenosis.²² Although total coronary blood flow may not increase, coronary blood flow velocity has consistently been demonstrated to rise.^17,23,24 In turn, data from a thrombolysis in myocardial infarction (MI) model suggest that the pulsatile waveform generated by IABP during diastole may contribute to improved time to coronary reperfusion without concomitant increase in coronary blood flow;²⁵ in all likelihood this improvement reflects the enhancement in peak flow velocity that may also help prevent reocclusion.²³ However, pulsatile flow may have additional benefit: in a randomized study comparing cardiopulmonary bypass with or without concomitant IABP, multiple parameters of whole-body perfusion were superior in the pulsatile flow setting.²⁶ In general, augmentation of blood flow appears most likely to occur in patients in profound shock.^17,24,27 Thus, most of the benefits of IABP are related to improvement in hemodynamics with secondary relief of ischemia. Amelioration of ischemia may in turn result in improved LV function and additional improvement in hemodynamics.

Systole and Myocardial Mechanics

Although the IABP is generally described as both improving myocardial oxygen supply and lowering myocardial oxygen demand, the predominant effect is the latter. This is based on decreased afterload and wall stress with increased stroke volume and cardiac output, in particular in patients with low-output states. In general, the magnitude of hemodynamic response is proportional to the extent of depression of cardiac function.¹⁶ Because stroke volume improves, heart rate stays the same or tends to decrease even as cardiac output rises. Reduction in LV end-diastolic pressure and improved cardiac output is associated with lower left ventricular heart filling and pulmonary arterial pressures, whereas systolic pressure typically decreases and mean blood pressure stays the same in hemodynamically stable patients. In patients in shock, mean blood pressure rises and systolic pressure may increase as well. When IABP placement results in improved end-organ perfusion, as evidenced by signs such as better urine output, the overall prognosis is generally improved.¹² Reduction in left-sided heart filling and pulmonary artery pressures also leads to reduced right ventricular (RV) afterload; thus, IABP insertion may be helpful in some patients with right-sided heart failure.²⁸ Although ejection phase indices tend to improve with unloading, the extent of augmentation of cardiac output is limited by the overall reserve in LV systolic function. Thus, at the extremes of ventricular dysfunction, even with excellent positioning of the balloon and timing as well as large IABP volumes, the IABP may not provide sufficient augmentation of cardiac output in some patients. A variety of other parameters have been used to assess the effect of IABP in shock, most prominently the cardiac power index (mean systemic pressure times cardiac index), which is the strongest hemodynamic correlate of outcomes.²⁹

Effects on Organs Other Than the Heart

The effect of IABP on flow to end organs other than the heart is also incompletely characterized. In the cranial circulation, as in the heart, net flow does not appear to increase in patients with a stable hemodynamic state. Further, with balloon deflation, there is some evidence for transient reversal of blood flow as well,³⁰ although this may be a consequence of early deflation (see discussion under “Timing”). Similarly, overall carotid flow is not changed in this setting.³¹ During hypothermia while patients are under cardiopulmonary bypass, pulsatile augmentation as obtained by the IABP does appear to improve cerebral oxygenation.³² Indirect evidence again suggests additional benefit in the setting of shock, with better cerebral blood flow, a setting in which there is modest clinical experience with using IABP in patients with refractory cerebral vasospasm.³³ Despite improvement in cardiac output with IABP, gastric tonometry has not shown improvement in the splanchnic circulation in CS patients,³⁴ and renal vein thermodilution has failed to show improvement in overall renal blood flow.³⁵ In cardiac surgery patients who subsequently developed low-output syndrome, however, IABP did have a positive effect on gastric tonometery.³⁶

The Intra-Aortic Balloon Pump Apparatus

With the introduction of a separate guidewire lumen in the early 1980s,¹⁴ the safety of IABP placement improved significantly. Previously, bulky 12 F or larger devices were placed via a surgical approach through an end-to-side graft to the femoral artery, with substantial associated morbidity. The device was then advanced through the femoral and iliac systems without a guidewire, frequently at the bedside with no fluoroscopic guidance. This led to a high rate of major vascular complications, in particular, iliac artery dissection but also aortic perforation and IABP malposition. The introduction of the percutaneous approach did not alleviate these problems until a guidewire lumen was added.¹⁴

There were conflicting demands on balloon design: first, to minimize the central lumen size as part of reducing overall profile, and second, at the same time to maintain a large enough lumen both for guidewires that could provide safe passage through the circulation as well as sufficient diameter for high-fidelity hemodynamic recordings. One solution to the latter has been the introduction of a line of catheters with a fiber-optic pressure measurement sensor.³⁷ In general, the guidewire lumens are designed to accommodate guidewires in the 0.018-inch to 0.030-inch range. The gas exchange lumen is concentrically placed around the guidewire lumen. Improvements in technology allowed for the introduction of tightly folded balloons prewrapped around the central lumen (Fig. 7.3); previously, a cumbersome process that sometimes led to device failure required the operator to wrap the balloon just before insertion.

Figure 7.3 Magnified cross-sectional view of prefolded intra-aortic balloon (a 7.5 French wrap). The prefolding was a significant advance in technology in the early 1980s because it eliminated folding at the bedside, a process that occasionally led to tearing of the balloon membrane, and substantially reduced the overall profile of the device at insertion. (Courtesy of Datascope, Montvale, NJ.)

Balloon Dimensions

The available IABP catheters range in gas volume from 25 to 50 mL and in shaft lengths from approximately 60 to 72 cm. The actual length of the balloon segment varies by manufacturer and balloon volume: Commercial balloon lengths are in the range of 16.5 to 26 cm. Although balloon volume does affect the degree of diastolic augmentation ³⁸ and cardiac output³⁹and may be particularly important in patients with severe, refractory CS, little difference was found in IABP effectiveness between 32-mL and 40-mL balloons in one study,⁴⁰ and risk to the patient is significantly higher if inflated balloon diameter approximates or exceeds aortic diameter (see “Complications”). Thus the decision for IABP size is typically based on patient size and severity of hemodynamic compromise, with the 40-mL balloon used in approximately three fourths of patients.⁴¹ A special consideration for balloon volumes is patient air transport; rapid changes in altitude will lead to increased (during ascent) or decreased (with descent) volume, requiring monitoring to ensure that appropriate balloon volume is maintained.⁴²

Because it is essential to locate the balloon below the origin of the left subclavian artery at the upper margin and above the renal arteries at the lower (a “safe zone”; Fig. 7.4), the tolerances are relatively low. Studies looking at the length of this segment in Japanese patients found a range of 21 to 25 cm, with good correlation between patient height and length of this segment (although in the relatively short Japanese population, the balloon lengths frequently exceeded the “safe zone” length for individual patients).⁴³ Recommendations vary by manufacturer: A 50-mL size I is generally recommended for patients taller than 6 feet (183 cm), and the 25-mL is used for patients shorter than 5 feet (152 cm). Recent changes in technology have allowed for larger diameter balloons to be available for shorter patients (e.g., a 50-mL IABP on an 8 F shaft is now available for patients 5′4″ [162 cm] and taller). In general, larger balloon sizes lead to improved unloading and augmentation with greater blood volume displacement. Early balloon models were designed to be occlusive during inflation, and experimental data have demonstrated optimal augmentation with 100% occlusion of the aorta.^38,39 Such full occlusion is generally avoided because of potential trauma to the aorta, ischemia to the spinal circulation, or abrasion of the balloon, which might contribute to a risk of rupture. The generally accepted value for ratio of balloon to aorta size is 80% to 90%; in the study by Igari⁴³ in Japanese patients, the range of commercially available balloon diameters was noted to be 14 to 15 mm, whereas aortic diameter mean at the level of the renal arteries was 17.5 ± 3.2 mm, within the accepted 80% to 90% range, although at least one 50-mL balloon has an expanded diameter of 18 mm (Arrow International, Inc., Reading, PA). In an earlier study in Americans, 90% of midthoracic aortas were larger than 19 mm.⁴⁴ Balloons are typically made of polyurethane or polyethylene, with materials chosen to allow rapid inflation-deflation cycles and tolerate an average of 100,000 to 150,000 cycles per day. Experimental materials with heparin and hydrophilic coatings have been developed to address thrombosis risk and trauma to the vasculature during device passage^45,46 but are not generally available.

Figure 7.4 Location of intra-aortic balloon during inflation and deflation. Ideally the tip should be a few centimeters distal to the origin of the left subclavian artery and the proximal portion above the origins of the renal arteries, the “safe zone.” (Adapted from figure provided by Teleflex Medical, Limerick, PA.)

Balloon Console

IABPs are driven by complexly engineered consoles with extensive artificial intelligence designed to recognize electrocardiographic rhythms and hemodynamics, with the ability to trigger balloon inflation either from the ECG (including paced rhythms) or from pressure waveforms, although in some settings, such as during cardiopulmonary bypass, an automatic trigger at a preset rate can be used. Tachyarrhythmias and irregular rhythms have substantial influence on the effectiveness of the IABP, the former in part because of the disproportionately shorter diastolic filling periods, and the latter because of difficulties predicting the timing of occurrence of the dicrotic notch, although improved algorithms have been incorporated.⁴⁷ The central lumen of the balloon is typically connected to a transducer and separate ECG leads are connected to the console. A series of alarms identify leaks in the IABP circuit, high or low pressure, loss of trigger signals, blood in the gas line, low battery, and other anomalies that foreshadow or indicate impending or existing system failures. (Figure 7.5 demonstrates the control panel of a modern IABP console.) Modern IABPs using fiber-optic technology permit automatic calibration in patients after insertion and automatic recalibration on a periodic basis or whenever the algorithms detect a change in patient condition.

Figure 7.5 Intra-aortic balloon console showing continuous display of electrocardiogram, systemic pressure (note diastolic augmentation), and balloon pressure. (Courtesy of Datascope, Montvale, NJ.)

Preparation

Balloon preparation requires establishing a vacuum in the gas lumen by drawing back on a large-volume syringe, and also flushing the central guidewire lumen (Fig. 7.6). If the procedure is to be done at the bedside without fluoroscopy, measuring the approximate distance along the course of the femoral and iliac arteries and then up the descending aorta is essential prior to IABP placement. The approximate depth to which it needs to be inserted should be noted on the shaft prior to beginning insertion. The ECG leads should be connected to the IABP console prior to introducing the catheter if ECG triggering is to be performed.

Figure 7.6 A modern intra-aortic balloon. A minimal size central lumen allows advancement over a guidewire. Low-profile balloons such as this one use fiber-optic cable for pressure monitoring. (Courtesy of Datascope, Montvale, NJ.)

Balloon Insertion

Based on data from the Benchmark Registry, nearly two thirds of IABP insertions in the United States occur in the cardiac catheterization laboratory, one fourth in the operating room, and the remainder at the bedside in a variety of hospital locations. In contrast, outside the United States, the operating room and the catheterization laboratory each account for approximately 40% of placements, with the remaining 20% inserted at the bedside.⁴

The modern IABP is designed to be introduced percutaneously through the common femoral artery, which is the method used in 95% to 98% of cases, two thirds via the right common femoral artery.41,48 A variety of disease states affect the ability to freely pass the device to the central aorta, in particular atherosclerosis, but also spasm and congenitally small vessels. Preprocedure assessment of the vascular tree is important: Diabetics, women, and patients with small body surface area in particular have small femoral arteries,⁴⁹ and there is a corresponding higher rate of vascular complications in these patients.⁵⁰ Pulses at the common femoral artery and below must be carefully documented before catheter insertion, and if there is suspected or confirmed peripheral vascular disease, angiography of the abdominal aorta, iliac artery, and lower extremities should be considered. This step is frequently not practical in the emergency setting or in patients with renal failure; in elective situations noninvasive evaluation should be performed, including ankle-brachial indices with pulse volume recordings, computed tomography, angiography, or magnetic resonance angiography.

Femoral Access

Every effort should be made to ensure that femoral puncture is above the femoral bifurcation and below the inguinal ligament (Fig. 7.7). An excellent practice is to place a short 5 F or 6 F pilot sheath in the femoral artery and to perform angiography of the common femoral artery to confirm sheath entry below the inferior excursion of the inferior epigastric artery; punctures above this landmark correlate strongly with retroperitoneal hemorrhage.⁵¹ Puncture below the femoral head drastically increases the likelihood of puncture into the femoral bifurcation vessels (77% of femoral bifurcations are at or below the inferior margin of the femoral head⁴⁹), which in turn is associated with acute leg ischemia due to vascular obstruction, and pseudoaneurysm formation upon balloon removal is more likely because of lack of the anvil of the femoral head against which to perform manual compression. If the pilot sheath is found to have entered outside the common femoral artery on femoral angiography, consideration should be given to switching to the contralateral side. Using fluoroscopy to aid in puncturing the common femoral artery at a point over the lower half of the femoral head is a recommended technique to ensure proper sheath placement.

Figure 7.7 Optimal access for placement of an intra-aortic balloon pump, right femoral artery, left anterior oblique view. The sheath entry is above the femoral bifurcation near the center line of the femoral head (circle). Note the location of the inferior epigastric artery (IEA; arrow). The lowest inflection of the IEA lies just above the inguinal ligament; punctures below this point have a lower likelihood of subsequent retroperitoneal hemorrhage.

Sheathless Insertion

The use of a sheathless insertion technique has been recommended to reduce complications.⁵² Sheaths typically add between 0.6 and 0.8 mm to the overall diameter required to place a device, so the sheath for an 8 F balloon approximates 10 F in outer diameter. “Going sheathless” therefore has the advantage of having the device occupy less space in the common femoral artery and has been described as reducing vascular complications with an odds ratio greater than 2 : 1.^53,54 Although retrospective analysis of large patient subsets presents compelling data that the sheathless approach is superior for reducing complications,⁵⁵ this finding has not been universally confirmed.^50,56 The latest data suggest that about 80% of IABPs are inserted with a sheath.⁴¹ The sheathless approach is most compelling in diabetic patients, women, and patients with known peripheral vascular disease or small body surface area. Sheathless insertion of an IABP requires careful preparation of the tissue track to allow atraumatic entry of the balloon tip into the artery directly through the skin over a wire. Spreading of tissue and predilatation with a dilator are essential. Fibrosed tissue tracks or thickened/calcified arterial walls frequently resist sheathless entry. In some cases, a stiff guidewire provides an adequate rail along which to slide the naked balloon catheter into the vessel.

Balloon Advancement and Positioning

Once suitable access is gained, guidewire passage to the aortic arch is ideally performed under fluoroscopic guidance. Without fluoroscopy at the bedside, approximating the distance from the femoral puncture to a point near the top of the descending aorta is imprecise and adds to the risk of trauma/ischemia to head and neck vessels, the aorta, or the renal arteries. In the critical care setting, transesophageal echocardiography is a suitable alternative to fluoroscopy for enabling precise balloon placement.⁵⁷ If fluoroscopy is not used, prompt postprocedure radiography to confirm location is imperative.

If guidewire passage meets resistance, alternative approaches include use of guidewires that are hydrophilic, steerable, or both.⁵⁸ If significant iliac stenosis is noted, balloon dilatation or stenting of the iliac artery is an accepted practice with high success rate both for achieving passage of the IABP and as an adjunct to preventing distal ischemia.^59,60 As more cardiologists become skilled in endovascular intervention, the ability to perform combined iliac stenting and IABP placement is expanding.⁶¹

Once the balloon has been advanced to a point 1 to 2 cm inferior to the left subclavian artery origin (see Fig. 7.4)—typically near the top of the descending aorta—the guidewire is withdrawn and the central lumen flushed. The vacuum port (minus its one-way valve) is connected to gas line tubing that in turn is connected to the console, and the dead space is purged and then filled with helium. The central catheter lumen is connected to a transducer on the console. After triggering is initiated, typically on every other beat, fluoroscopy should be used to confirm balloon location and filling, and pressure contours should be evaluated for appropriateness of timing (see later discussion under “Timing”). Once the timing is considered satisfactory, continuous pumping can be initiated. If necessary, the balloon should be repositioned with the console turned to standby to avoid trauma to the aorta.

The distal circulation should be assessed carefully after IABP placement. Distal ischemia is relatively common. The most likely cause of acute ischemia is obstruction of the artery by the catheter shaft itself. If ischemia occurs hours or days after insertion, the possibility of thrombus, typically secondary to stagnant blood in the confined space of a small or diseased vessel, should be considered as well. A technique for addressing distal limb ischemia—placement of a small sheath retrograde in the contralateral femoral artery and antegrade in the ipsilateral common femoral or superficial femoral artery—if performed by expert hands, can occasionally salvage an ischemic limb without forcing removal of the IABP ⁶² (Fig. 7.8). There is some evidence that IABP inflation properties and hemodynamic effects may be superior in the patient positioned horizontally than with the patient tilted at a 30-degree angle.⁶³ Postprocedure and subsequent monitoring of left arm pressures can lead to early diagnosis of inadvertent balloon advancement obstructing left subclavian artery inflow, a phenomenon to which the restless patient who flexes the thigh is predisposed.

Figure 7.8 Percutaneous external femoral-femoral shunt for a patient with critical leg ischemia after placement of an intra-aortic balloon pump (IABP) in the right common femoral artery. The patient has an occluded right superficial femoral artery (SFA), so distal circulation depends on profunda femoris to superficial femoral collateral vessels. The antegrade sheath is placed directly into the profunda femoris. MTM, male-to-male adapter; PFA, profunda femoris artery. (From Merhi WM, Turi ZG, Dixon S, et al: Percutaneous ex-vivo femoral arterial bypass: A novel approach for treatment of acute limb ischemia as a complication of femoral arterial catheterization. Catheter Cardiovasc Intervent 2006;68:435-440.)

Alternate Access Routes

Multiple alternatives to percutaneous femoral access have been described, all associated with higher complication rates. In part, this difference in rates is selection related: Patients with severe peripheral vascular disease have multiple comorbid conditions that affect IABP complications. In addition, introduction of a balloon into a peripheral vessel smaller than the common femoral artery or into a large central vessel through a surgical approach of necessity creates hazards associated with the introduction, maintenance, and withdrawal of the IABP.

Approaches described to date include the brachial, subclavian, axillary, iliac, transthoracic, and translumbar arteries. Although the brachial approach has been successful in isolated cases,64,65 and sheathless entry reduces the overall diameter of lumen encroachment to less than 3 mm, the potential for complications is substantial. They include not only vascular injury and ischemia of the hand but also potential neurologic consequences from formation of thrombus on an indwelling catheter underneath the origin of the right common carotid artery (as the shaft traverses from the right subclavian artery to the innominate artery) as well as under the left common carotid artery and subclavian artery origin.⁶⁶

Iliac insertion through a conduit has been used for patients in whom femoral access is not adequate or who are not candidates for ventricular assist devices: With retroperitoneal placement, patients can be at least partially ambulatory during prolonged counterpulsation.⁶⁷ Other routes of access that have been described are via the subclavian⁶⁸ and axillary arteries,^69,70 with or without conduits, and generally with an eye toward allowing modest ambulation during prolonged IABP use. A transthoracic approach has been described, with placement via the ascending aorta into the standard descending aortic location.^71,72 The morbidity rate associated with these surgical placements is significantly higher,⁷³ in part because they typically involve longer IABP indwelling times as well as the comorbidity issues already mentioned.

Timing

The importance of IABP timing was understood from the time of the original Moulopolous study in 1962.⁹ The timing of balloon inflation and deflation is designed to optimize afterload reduction and enhancement of diastolic pressure without interfering with ventricular ejection. Classically, it was considered optimal to inflate at the dicrotic notch, as soon as aortic valve closure occurred, and to deflate near the onset of ventricular depolarization, anticipating the beginning of mechanical systole (Fig. 7.9). In fact, significant enhancements to IABP efficiency can be obtained with refinements to these concepts.

Figure 7.9 Examples of timing of intra-aortic balloon pump inflation and deflation with 1 : 2 balloon pumping. A, Optimal timing results in mild lowering of systolic pressure with diastolic augmentation. Also shown are examples of early inflation before the dicrotic notch (B), late inflation (C), early deflation (D), and late deflation (E).

Late Inflation and Early Deflation

Gross errors in inflation and deflation lead to failure to obtain benefit from IABP and occasionally to significant hemodynamic compromise. Early in the history of IABP deployment, it was appreciated that inflation throughout the period from closure of the aortic valve to its subsequent opening was necessary for optimal hemodynamic effect.¹⁶ Both late inflation and early deflation reduce augmented LV stroke volume and result in decreased peak diastolic coronary velocity;²³ the latter is demonstrated with transthoracic Doppler ultrasound, a tool that can potentially be used to optimize timing. An additional concern exists with early deflation: The abrupt decrease in IABP volume during diastole can lead to reversal of both coronary and other end organ (e.g., cerebral) flow back into the aorta, shunting blood from vital end organs.³⁰

Early Inflation

Early inflation results in increased afterload late in LV ejection with consequent impairment of LV systolic function. The ejection phase is shortened, LV end-systolic pressure rises, and stroke volume decreases; inflation, in the range of 130 to 190 ms before the dicrotic notch, results in a 20% decrease in stroke volume.⁷⁴ This effect may not be seen if early balloon inflation is less pronounced; no hemodynamic effect was noted when IABP inflation occurred 50 ms before the dicrotic notch. Regardless, although early inflation theoretically lengthens diastole and allows for a longer period of diastolic augmentation, the net effect does not appear to be salutary. In addition, early inflation carries theoretical risks associated with the increased afterload and wall stress, including aneurysm formation and rupture in the peri-MI period.

Late Deflation

Similarly, late balloon deflation could be expected to interfere with ventricular ejection and decrease stroke volume. In fact, somewhat counterintuitively, a similar 110- to 180-ms delay in IABP deflation appears to have salutary effects and is associated with a stroke volume increase of 18%,⁷⁴ which apparently results from both an increase in diastolic filling period and an augmented decrease in afterload later in the cycle. This finding confirms a prior observation by Kern and associates⁷⁵ and suggests that, in general, most operators have timed deflation too early. The beneficial effect occurs with timing deflation to be simultaneous with LV ejection; delaying deflation beyond the range described, however, raises concerns similar to those described for early inflation.

Electrocardiogram Triggering

When the ECG is used as the trigger, the descending slope of the T wave correlates best with the onset of diastole ³⁸ and is the usual timing for balloon inflation. Deflation is typically timed to the R wave, which denotes a short time delay after the onset of electrical systole. Algorithms were developed early to effect prompt deflation and prevent inflation in the setting of ectopy, and IABP software recognizes pacemaker spikes in contrast to QRS complexes as well. As in the findings described previously,^74,75 deflation timed to the J point, with adjustment for the delay between onset of isovolumic systole and aortic valve opening, and perhaps slightly later, was shown early in the IABP literature to improve stroke volume.³⁸

Other Considerations

A common problem has been proper timing in patients with underlying arrhythmias. Atrial fibrillation has been particularly vexing, with unpredictable beat-to-beat intervals. An algorithm to predict the occurrence of the distance between the QRS and the dicrotic notch was developed in the mid-1990s.⁷⁶ Newer dicrotic notch prediction algorithms using high-fidelity micromanometer pressure sensors have been described.⁴⁷ However, atrial fibrillation poses difficulties beyond timing alone; rapid ventricular rates result in a disproportionate decrease in the diastolic interval and limit the effectiveness of IABP because of the inherently short period of counterpulsation,⁷⁷ including subtraction of the fixed time interval required for shuttling helium into and out of the balloon. Operating at 1 : 2 rates may be required in this setting.

Lack of augmentation despite proper timing should result in troubleshooting the IABP console, checking with fluoroscopy to visualize inflation of the balloon and confirm the level of balloon placement, and excluding kinking of the gas line or other mechanical failures. Apparently normal IABP function with absence of hemodynamic improvement should raise a suspicion that the patient has a baseline hemodynamic state that will not benefit from LV unloading, in particular hypovolemia, sepsis, or profound hemodynamic collapse, as well as a number of conditions described subsequently (see “Contraindications”).

Overall, timing of IABP should result in diastolic pressure augmentation and lowering of assisted peak systolic pressure when possible, with the former a generally used end point for patients with shock and the latter the primary goal in patients with more stable hemodynamics at the time of IABP insertion.

Adjunctive Pharmacotherapy

Adjunctive pharmacotherapy typically includes heparinization. There are two theoretical reasons for anticoagulation: First, in patients receiving less than 1 : 1 counterpulsation, there is concern regarding clot formation on the balloon apparatus, and second, stagnant blood around the catheter shaft, especially in the common femoral artery, has been thought to raise the risk of thrombosis. The general consensus has been that anticoagulation should be administered if not contraindicated. The balloons themselves have a thrombogenic surface,⁴⁵ and occlusion of vessels requiring thromboembolectomy has been reported to be one of the most common complications, reaching nearly 3% in one series of 911 patients undergoing coronary artery bypass grafting (CABG).⁵³ Nevertheless, the thinking on this is in flux, with a growing evidence base that routine anticoagulation does not prevent thrombosis or thromboembolism but does increase bleeding risk. Thus, a recent study⁷⁸ compared routine heparin use versus selective anticoagulation only in those patients with an indication for heparin use other than IABP insertion. There was no difference in the rates of IABP-related complications including major limb ischemia, but anticoagulated patients had a higher incidence of non-access-site bleeding, predominantly gastrointestinal. A randomized trial had similar conclusions, although the course of counterpulsation was relatively brief.⁷⁹ Review of the best evidence to date leads to the somewhat controversial conclusion that anticoagulation may be best targeted to selected patients who are at risk of thrombosis or thromboembolism for reasons other than IABP use alone.⁸⁰

A second adjunctive pharmacotherapy issue is the use of prophylactic antibiotics. Although fever, bacteremia, and sepsis were reported to be common (occurring at rates of 47%, 15%, and 12%, respectively) in one small study,⁸¹ the infection rate in larger series has been less than 1%,⁵³ and the consensus is that the evidence base is too thin and the public health implications too unfavorable to recommend routine antibiotic use unless otherwise clinically warranted.⁸² A third medication issue relates to sedation. Continuous bed rest in a critical care setting is frequently associated with disorientation or agitation. Balloon migration from leg bending and patient movement risks significant trauma to the aorta, renal arteries, and head and neck vessels. Careful sedation is essential. Finally, antiarrhythmic agents to slow and regularize the heart rate can have important benefits for lengthening diastole and optimizing balloon timing, both of which in turn enhance IABP augmentation.

Balloon Removal

Removal of the IABP poses several challenges. First, the ability to maintain hemodynamic stability without counterpulsation must be confirmed as part of the weaning process. Although cardiac output is disproportionately decreased with lowering of pumping ratios, such lowering appears superior to decreasing volumes as a weaning method, an approach confirmed by one small retrospective evaluation.⁸³ The balloon should not be turned off completely until the activated clotting time value confirms that anticoagulation has been effectively discontinued and the patient is ready for balloon removal, because thrombosis on the balloon surface occurs rapidly.⁴⁵ Most operators run the IABP at a low cycle rate, 1 : 3 to as low as 1 : 8 (depending on manufacturer), until anticoagulation has worn off sufficiently and the balloon can be removed. Nonfunctioning or stopped IABPs must be removed promptly, preferably within 20 minutes or less. A second challenge relates to the size of the deflated balloon. Because the balloon is delivered prefolded by the manufacturer (see Fig. 7.3), it passes readily through its delivery sheath during insertion. Once inflated, it will not refold when vacuum is applied, and the profile is too large for retrieval without bringing both the sheath and the balloon out of the body together.

Removing the balloon, especially after long indwelling times (mean indwelling time is 53 to 77 hours 41,48), requires meticulous attention to several details. First, the balloon gas line should be aspirated to reduce the balloon profile. Second, it is essential to allow some bleeding after catheter removal and prior to compression to avoid stripping any clot off the balloon inside the common femoral artery. Because these are relatively large catheters, frequently deployed in patients with vascular disease who have a tenuous hemodynamic status even at the time of IABP removal, patients may not tolerate prolonged and aggressive compression of the groin. Recent discontinuation of heparin combined with a large arteriotomy size can result in significant difficulty in controlling hemorrhage and achieving hemostasis.

Surgical closure is sometimes preferable, particularly with severely obese patients, after very long indwelling times, with uncorrectable anticoagulation status, or with low or high punctures.⁸⁴ Vascular closure devices have been used successfully in small series^85,86 but require extreme caution; significant mortality risk is associated with infections related to vascular closure devices,⁸⁷ and assuring sterility at the time of IABP removal is difficult. When use of a closure device is required, the authors typically place a 0.018-inch stiff guidewire inside the lumen after extensive efforts to achieve sterility of both the field and the device, withdraw the IABP, and place the closure device over the wire; we cannot recommend this off-label approach, however, until a better evidence base is available. Finally, it is important to continue to monitor patients for adverse events because nearly 25% of IABP-related complications have been reported to occur after IABP removal.⁸⁸

Indications

As with the original patients in 1968,¹⁰ CS remained the most common indication for several decades, although later data suggest that circulatory support for percutaneous intervention has replaced hemodynamic instability in the acute MI setting as the primary indication for IABP insertion.⁴¹ Other common indications are perioperative support for patients undergoing cardiac surgery, weaning from cardiopulmonary bypass, management of unstable angina, severe congestive heart failure, and with less evidence base, refractory ventricular arrhythmias or angina after MI as well as a host of miscellaneous settings largely defined by case reports (Box 7.1).

Box 7.1 Indications for Use of an Intra-aortic Balloon Pump

Acute ST-Segment Elevation Myocardial Infarction (STEMI)*

Class I Indications