The incidence and prevalence of heart failure continue to increase, affecting more than 5.8 million individuals in the United States and over 26 million individuals worldwide.1,2 Advances in medical management and adjunctive therapy have led to significant strides in improving both quality of life and survival of afflicted patients. However, therapeutic options for those with refractory, advanced American Heart Association (AHA) Stage D heart failure with reduced ejection fraction (HFrEF) remain limited to continuous inotropic therapy, with a small subset of patients referred for cardiac transplantation or the implantation of temporary or durable mechanical circulatory support (MCS) devices.3,4 Although transplantation represents definitive therapy for HFrEF, the limited availability of suitable donor allografts has led to an increased utilization of durable, implantable ventricular assist devices (VADs), which are either used to bridge eligible patients to transplantation (BTT) or increasingly as destination therapy (DT) in those with circumstances that preclude transplantation.5 According to the Eighth Annual INTERMACS report, nearly 23,000 durable device implantations have been performed since 2006, and projections suggest that this number will continue to increase.6 Improvements in device technology and management strategies have led to more robust survival rates and improvement in heart failure symptoms.5
Although published estimates vary, up to one third of patients supported with VADs will go on to develop noncardiac pathology that will require a surgical intervention under anesthesia.7–10 Although a wide range of procedures have been reported in the literature, the more common include endoscopy and gastrointestinal procedures.8,11 Ideally, these noncardiac surgeries (NCS) would be performed electively at tertiary care centers where VAD implantation occurred. However, given the burgeoning number of patients supported with durable VADs coupled with circumstances that may preclude transfer to experienced MCS centers, it is vital for anesthesiologists in all clinical settings to develop familiarity with these devices and review basic management paradigms.
Left ventricular assist devices (LVADs) are designed to supplement or replace the function of a failing heart. Although device technology has evolved since its inception, the same general functional principle applies to all LVADs. Deoxygenated blood from systemic venous return is directed into the right atrium and enters the right ventricle (RV), which is responsible for pumping it through the pulmonary circulation and into the left atrium. When blood enters the left ventricle (LV), it is drawn through an implanted inflow cannula to a pump, which directs blood to the systemic circulation through an outflow cannula implanted in the aorta (Fig. 1). Some blood flow is still pumped across the aortic valve by the native LV, but the patient is no longer wholly reliant on a failing heart for systemic perfusion. The vast majority of assist devices are implanted to assist a failing LV.6 However, in some instances, isolated RV or biventricular support may be required.
First-generation VADs were designed with the goal of replacing LV pump function. These devices, such as the HeartMate XVE (Thoratec Corporation, Pleasanton, CA), were pulsatile volume displacement pumps, which were found to lead to a survival benefit compared with optimal medical therapy for advanced-stage heart failure.12 However, they were notably large, limiting the patient population in whom they could be used, loud, and had limited durability.13,14 First-generation devices have been largely supplanted by second-generation and third-generation continuous-flow (CF) devices, which now account for >95% of all implanted LVADs.6 Continuous-flow LVADs (CF LVADs) drain and unload the LV throughout the cardiac cycle, reducing or sometimes eliminating pulsatility, allowing them to be smaller, quieter, and more durable.
The most widely implanted second-generation device is the HeartMate II (HM II; St. Jude Medical, St Paul, MN), which is a small axial-flow device that received FDA approval for both BTT and DT (Fig. 2). This device consists of an inflow cannula inserted into the LV apex, pump housing positioned in the preperitoneal space that contains a motor and a rotor, and an outflow cannula that is inserted into the ascending aorta, just distal to the aortic valve.15,16 The motor generates a magnetic field that spins the rotor, which draws blood out of the poorly functioning LV and deposits it into the ascending aorta, effectively bypassing the ventricle. A percutaneous lead connects the motor to an external system controller, which is connected to a power supply, specifically an alternating current source or battery packs. This device can also be connected to a system monitor, which can be used to monitor LVAD parameters in the clinical setting.
The Heartware HVAD (Medtronic, Minneapolis, MN) is a third-generation CF device that has some similarities in structure to the HeartMate II, consisting of an inflow cannula in the LV apex, a pump housing unit with a rotating impeller, and an outflow graft to the ascending aorta (Fig. 3). A percutaneous driveline also connects the pump housing unit to a system controller and external power supply, and a system monitor is available for use. In contrast to the HeartMate II, the HVAD uses centrifugal-flow technology with a hybrid hydrodynamic/magnetic suspension of the impeller,18 which eliminates the need for mechanical bearings and promotes hemocompatibility. The impeller spins at high speeds, creating suction that pulls blood into the pump housing, alters the direction of flow, and pushes it out of the pump into the aorta.17 This device is implanted entirely within the pericardial space, obviating the need for an abdominal pocket.
The HeartMate III (HM III; St. Jude Medical) is a third-generation CF LVAD that has recently received FDA approval to be implanted in eligible patients as BTT or bridge to myocardial recovery (Fig. 4). This device was designed with the goal of enhanced hemocompatibility, and uses full magnetic levitation technology, which offers wider blood-flow paths to mitigate shear forces. This device is also unique in that it generates an artificial pulse to eliminate areas of pump stasis and potentially mitigate the physiological disadvantages of nonpulsatile flow.19,20 It is important to note that this pulsatility does not coincide with the patient’s native pulse; rather, it is a timed algorithm in which speeds are transiently increased and decreased at specified intervals, generating an artificial pulse 30 times per minute.
Physiology and Device Function of CF LVADs
Perioperative management of patients with CF LVADs requires an understanding of the principles that govern flow through these devices. Both physiological parameters and device settings impact flow, and therefore, the amount of cardiac output support.
Pump Differential Pressure (PDP)
The PDP, also referred to as pump head pressure or the pressure gradient across the pump, refers to the difference between the pressure in the aorta and LV pressure. The flow produced by CF LVADs is inversely proportional to this pressure differential.21 A higher PDP, which may occur in scenarios of high systemic pressure or low ventricular pressure, results in less flow and less cardiac output support provided by the VAD. Conversely, conditions that generate lower PDPs result in greater flow. As a result, it is essential to consider physiological parameters, such as LV preload and afterload, when managing patients with CF LVADs.22
CF LVADs are “preload dependent” because of the contribution of LV preload to the PDP. In other words, the volume of blood contained in the LV will limit the amount of volume that can flow through the pump. An increase in LV preload also increases LV end diastolic pressure, which decreases the pump pressure gradient and results in increased flow. The converse is also true: decreased LV preload increases the PDP and decreases flow through the pump. Inadequate LV preload can result in leftward shift of the interventricular septum because of negative pressure generated at the inlet of the inflow cannula and cause LV collapse. This event, known as a suction or suck-down event, can result in severe hemodynamic compromise, dysrhythmia, and hypoperfusion. In addition, when the interventricular septum is pulled leftward, normal RV architecture and spatial geometry is disrupted, which compromises RV function.21,23,24
Perioperative management of patients with CF LVADs requires careful review and consideration of the many parameters that affect LV preload (Table 1), including intravascular volume, venous return, RV function, and obstructive lesions. Low intravascular volume may result from bleeding or diuretic therapy, both of which may be exacerbated by a patient’s NPO status before surgical intervention. Patient positioning (such as reverse Trendelenberg) and aspects of surgical technique (ie, abdominal insufflation) may also alter venous return, as can arrhythmias, tamponade, constrictive pericarditis, and tension pneumothorax. Obstructive lesions, such as pulmonic stenosis and mitral stenosis, may also impair LV filling and compromise LV preload.
LVADs are completely reliant on the RV to deliver adequate preload to the LV, and optimization of RV function is of paramount importance.25 RV dysfunction, or any conditions that compromise RV function, including arrhythmias or physiological factors that elevate pulmonary vascular resistance (PVR), must be avoided.
Patients with CF LVADs require careful afterload management to optimize both flow and systemic perfusion pressure. Given that PDP is determined in part by aortic pressure, CF LVADs have been deemed “afterload sensitive.”21 There are some data to suggest that CF LVADs may actually be 3 to 4 times more sensitive to increase in afterload than the native heart.26 Systemic hypertension elevates aortic pressure, which increases PDP and decreases flow and cardiac output support. Conversely, hypotension decreases PDP and results in increased VAD flow, but low systemic pressures may compromise end-organ perfusion, particularly when they persist for a prolonged duration. Current guidelines recommend an optimal mean arterial pressure of 70 to 90 mm Hg.23,24
The pump speed, which is manually set externally on the CF LVAD controller, directly impacts flow.27 At lower speeds, the VAD provides less of a contribution to LV unloading, and LV systolic pressure may exceed aortic pressure. In this scenario, some blood will flow through the pump while some is ejected across the aortic valve. As the speed increases, there is increased LV unloading by the pump throughout the cardiac cycle, which decreases LV systolic pressure and increases aortic diastolic pressure. The result is a narrowing of the pulse pressure and diminished pulsatility. In some patients, perioperative hemodynamic monitoring strategies must be considered carefully because of difficulty in palpating peripheral pulses or obtaining noninvasive blood pressure measurements. Decreased pulsatility also carries concerns beyond the challenges and inconvenience associated with monitoring. Long-term nonpulsatile flow associated with CF LVADs has been associated with the development of intestinal arteriovenous malformation and an increased risk of gastrointestinal bleeding.28,29
Optimal speed selection is a vitally important process, and should be performed in expert consultation with the patient’s heart failure team, when possible, and ideally with the use of echocardiography. Low pump speeds may result in suboptimal LV unloading, rightward shift of the interventricular septum, and compromise of RV function. Conversely, high speeds may aggressively unload the LV, cause leftward shift of the interventricular septum, and potentially result in a suction event with subsequent hemodynamic compromise. Optimal speed, at a basic level, is obtained when a patient receives an adequate cardiac index, the interventricular septum is in the midline position, and there is intermittent opening of the aortic valve with preservation of pulsatility.23
It is important to note that speed ranges vary for each CF LVAD device. HVAD speeds can range from 1800 to 4000 RPM, with manufacturer-specified clinical operating ranges of 2400 to 3200 RPM.30 Pump speeds for HM II can range from 6000 to 15000 RPM, with most patients set between 8000 and 10000 RPM.31 The third-generation HM III device has a set speed range of 3000 to 9000 RPM with a clinical operating range largely between 4800 and 6800 RPM.32
Pulse Index (PI) and Flow Pulsatility
LV systolic contraction increases the ventricular pressure, decreasing PDP, and generating a burst of increased pump flow in systole. In HM II and HM III devices, the magnitudes of these systolic flow bursts are averaged over 15-second intervals to generate a PI value, which is displayed on the monitor screen (Fig. 5).24,27 The PI value changes with adjustments in the set pump speed. As the pump speed is increased, the pump provides more cardiac output support, resulting in more ventricular unloading and less pressure generated in the LV. This results in lower magnitude bursts in flow, and the PI value will be lower, indicating that the PI is inversely related to the degree of support provided by the pump. The PI can also provide other clues about changes in a patient’s physiology, including LV preload, afterload, and contractility. A downward trend in PI that occurs in the absence of changes to pump speed may suggest inadequate preload, which could be the result of hypovolemia or RV dysfunction, decreased afterload, or diminished intrinsic contractility. Increases in pulsatility may indicate hypervolemia to the extent that it does not cause RV strain. Although trends in PI are valuable in identifying changing physiology, these values are nonspecific and do not themselves indicate an etiology.21 In the absence of pulmonary artery or central venous catheters, other diagnostic modalities such as echocardiography and clinical status may be required to identify the underlying cause.
The analogous parameter seen on the HVAD clinical monitoring screen is the pump flow pulsatility waveform (Fig. 6).33 This waveform represents estimated instantaneous flow, visually displaying the beat-by-beat difference between the peak flow during systole and the trough flow during diastole. The magnitude of the pulsatility waveform is affected by the same factors that impact the PI, including LV preload, afterload, and contractility. According to the manufacturer’s specification, the trough flow should always exceed 2 L/min and the waveform should demonstrate at least 2 L/min of pulsatility.30
Pump power is a direct measure of the motor’s current and voltage and has a direct relationship with pump speed and flow.23 Power is displayed on the clinical monitoring screens for both the HeartMate II and HVAD (Figs. 5, 6). Power values increase with speed escalation, and under normal physiological conditions, these values fall within specified ranges. Increases in pump power that are not associated with increases in speed are suggestive of possible pump thrombus formation.24,34 Conversely, a fully occlusive pump thrombus would result in decreased flow and pump power.
It is important to note that the flow value displayed on the CF LVAD monitor screens is not a measured number (Figs. 5, 6).23 Rather, it is estimated on the basis of set pump speed, measured pump power, and in the case of the HVAD, the patient’s hematocrit. These displayed flows should not be considered measured values and must be corroborated by other clinical data points, including physical examination and laboratory markers of end-organ perfusion. For example, increases in power are observed in the absence of speed adjustments during the development of a pump thrombus. Given the direct relationship between pump power and flow, the displayed values of estimated flow will increase, sometimes markedly, whereas actual pump flow does not. In other words, the presence of a pump thrombus requires an increase in power to maintain rotor speed. The measured pump power is used to calculate an estimated flow. With an increase in pump power, the device estimates a higher pump flow, which is not reflective of actual flow through the device. It is also important to note that when flow estimates fall outside of operational range, the device may display symbols instead of numbers, such as “−−−” or “+++” in HM or “−−−” or “>10 L/min”, in HW devices.
In addition to the standard pre-anesthetic interview and physical examination, patients supported with CF LVADs will require a more detailed preoperative evaluation when circumstances permit. A thorough review of previous medical records and consultation with the institution’s VAD team, if one is available, is advisable to ensure that the patient is as optimized as possible and appropriate perioperative planning has occurred. If no institutional VAD team is available, consultation with the patient’s heart failure cardiologist or expert physicians at an experienced MCS center is advisable.
Heart Failure Status and Device Details
Patients with CF LVADs often experience improvement in heart failure symptoms and may achieve a lower acuity New York Heart Association (NYHA) classification. Knowledge of current functional and exercise capacity is also useful in determining a patient’s physiological reserve. End-organ function, specifically hepatic and renal function, should be assessed carefully using laboratory evaluation and imaging when necessary.24,35
Perioperative physicians should be familiar with the type of device implanted so that an appropriate clinical monitoring screen may be obtained. The patient’s standard controller will not provide adequate, easily visualized information in the intraoperative setting. The date of CF LVAD implantation is important; end-organ function, such as renal and hepatic function, may still be recovering if in close time proximity to the date of implant. If possible, perioperative physicians should also be aware of the patient’s most recent LVAD assessment, and aberrancies such as device alarms or suction events should be investigated. Previous echocardiographic data should be obtained and reviewed to evaluate the position of the interventricular septum and the development of valvulopathies, particularly aortic insufficiency, which, when severe, can result in significant recirculation through the LVAD and compromise forward flow.
Both hypovolemia and hypervolemia are detrimental in this patient population. In the absence of recent echocardiographic or right heart catheterization data, volume status may be determined by a combination of physical examination, trend of daily body weight, adherence to diuretic therapy, and data from point-of-care ultrasound.
In addition, detailed information on the presence of adjunctive cardiac devices, such as defibrillators and pacemakers, should be ascertained so that these devices may be appropriately managed intraoperatively, particularly in procedures requiring the use of electrocautery. The patient’s heart failure medication regimen and most recent administrations should be reviewed carefully; certain drugs, such as angiotensin-converting enzyme inhibitors, represent a cornerstone of heart failure management, but may contribute toward intraoperative and postoperative vasoplegia.
Right Ventricular Assessment
CF LVADs only function effectively when the LV is filled adequately by the RV. Preoperative assessment of the RV is crucial as the presence of preexisting dysfunction will help identify patients at higher risk of intraoperative RV compromise, guide decision-making on the optimal invasive monitoring strategy (such as the need for central venous or pulmonary artery catheters), and highlight the need for early initiation of preventative rather than reactive RV protection strategies in more susceptible patients.21 Data on RV dysfunction, specifically chamber size, degree of dysfunction, and quantification of tricuspid regurgitation,36 are often obtained using echocardiography. This information is particularly important in cases where significant fluid shifts are anticipated, given the preload dependence of CF LVADs. In nonemergent cases, preoperative RV optimization maneuvers may be required, such as in the initiation of inotropic therapy to support RV contractility, the administration of inhaled pulmonary vasodilators for RV afterload reduction, or the use of fluids or diuretics to optimize intravascular volume status.23,36–38
Anticoagulation, Bleeding, and Thromboembolism
Perioperative physicians must be familiar with a patient’s anticoagulation regimen and history of thrombotic or hemorrhagic events. Anticoagulation guidelines have been proposed to mitigate the risk of thromboembolic complications in patients with CF LVADs; however, practices vary widely.39,40 Many patients are managed using a combination of warfarin titrated to an international normalized ratio of 1.5 to 2.5 and antiplatelet therapy, most commonly aspirin.23,41 Other patients may be managed with less aggressive antithrombotic regimens because of previous bleeding complications or on the basis of provider preference. Irrespective of the regimen utilized, anticoagulation therapy predisposes patients to bleeding complications, with some estimates indicating that nonsurgical bleeding occurs in up to 40% of all patients with CF LVADs.42 The published literature on anticoagulation strategies suggests that medications may not be the sole culprits for bleeding complications. Both the European and the US arms of the TRACE study show that even on reduced antithrombotic regimens, patients with CF LVADs may still develop bleeding complications, albeit at a decreased rate.43,44 Another possible etiology for the development of bleeding complications is the development of acquired type 2A von Willebrand disease, which occurs when pump-associated shear forces cause proteolysis of high molecular weight multimers of von Willebrand factor.45,46 Abnormalities of these multimers occur in all patients after device implantation,47 and may impair platelet-mediated hemostasis, thus predisposing patients to bleeding complications.
There are no formal guidelines to optimally manage perioperative anticoagulation. Some centers have developed institution-specific protocols, which may be adjusted to balance bleeding and thromboembolic risks. Published data for patients with CF LVADs undergoing NCS suggest that bleeding is the greater risk,48 with the largest published series noting that 6.4% of patients developed hemorrhagic complications requiring transfusion compared with a 0.6% rate of pump thrombus.40 However, thorough consideration is required of the type of surgery, associated bleeding risk, elective or emergent nature of the procedure, and any history of previous hemorrhagic or thromboembolic complications. Variable practices in managing anticoagulation therapy before elective surgery have been described; medications may be discontinued completely several days before surgery or be continued in varying permutations, such as holding either antiplatelet therapy or warfarin, or bridging with an intravenous heparin infusion.21 Emergent surgical procedures, or those performed in particularly sensitive areas, such as neurosurgery, may require full reversal of anticoagulation with transfusion of blood products and/or the administration of hemostatic adjuncts.
Preoperative review of anticoagulation should also prompt perioperative physicians to review transfusion management strategies for elective procedures. It is also important to determine the purpose of LVAD implantation as patients who have CF LVADs as BTT may benefit from a more conservative transfusion strategy to mitigate alloimmunization.
Personnel and Planning
Thorough preoperative planning is a crucial aspect of care for patients with CF LVADs, particularly when an NCS intervention takes place in institutions with limited MCS experience. This process should include communication with specialized support teams, anesthesia provider selection, and determination of the appropriate clinical setting in which to perform the procedure.
Perioperative physicians should closely collaborate with institutional MCS specialists35 to assist with intraoperative device management and monitoring, including cardiac surgeons, cardiac anesthesiologists, cardiologists, perfusionists, and experienced MCS nurses. If such a team is unavailable and circumstances preclude patient transfer to an experienced institution, attempts should be made to consult with clinicians at the nearest MCS center or involve clinical device specialists from the corresponding CF LVAD device manufacturer. Necessary equipment should also be obtained, such as batteries, power supply, and clinical monitoring screens.
In the early years of CF LVAD implantation, all cases were largely managed by fellowship-trained cardiac anesthesiologists because of low case volume and the lack of widespread familiarity with these devices. However, with the growing number of implanted devices and broad educational efforts, general anesthesiologists are increasingly assuming responsibility for the management of CF LVAD patients undergoing NCS procedures without significant differences in perioperative complication rates.11,40,49
Although robust evidence on the optimal clinical setting for NCS is lacking, procedures should be performed in locations with the capabilities and resources to adequately support higher acuity patients. Given bleeding and hemodynamic risks, careful consideration must be given to managing this patient population in locations outside of the operating room. There have been published reports suggesting that patients with CF LVADs may be safely managed outside of an operating room setting,11 but no data on the safety of surgical interventions in stand-alone ambulatory surgery centers.
When possible, the CF LVAD should be connected to the appropriate clinical monitoring screen to ensure visibility of device parameters to all care providers in the room. Although practice models and policies may vary by institution, device parameters should be closely monitored intraoperatively by a dedicated clinician when possible, such as a trained nurse specialist or perfusionist. Adequate power supply should also be ensured, and the device transitioned from battery power to an alternating current source.21
The presence of a CF LVAD does not preclude the use of any anesthetic technique or medication. However, management strategy may require modification on the basis of a patient’s physiology and functional status and the potential physiological strain that the anesthetic technique and surgical procedure may cause. As an example, endoscopic procedures are commonly performed using moderate sedation without the use of an airway device. A patient with a CF LVAD who has pulmonary hypertension may poorly tolerate the consequences of hypoventilation and a general anesthetic may be required. In addition, anticoagulant therapy may limit the feasibility of neuraxial and regional techniques, particularly in more urgent cases in which medication cannot be discontinued for the necessary duration.
Standard monitors should be used in the intraoperative management of all patients with CF LVADs. However, the lack of palpable peripheral pulses in many of these patients will limit the utility of modalities with technology reliant on pulsatility, such as pulse oximetry and various noninvasive blood pressure (NIBP) monitoring techniques. Periodically increasing pulsatility by decreasing LVAD speed and/or administering inotropic medications may facilitate the use of these monitoring devices, but frequent speed adjustment is neither practical nor advisable outside of experienced centers.
In recent years, cerebral oximetry has been used successfully as an adjunct to pulse oximetry to provide information about arterial oxygen saturation. This technology, which uses near-infrared spectroscopy to monitor regional oxygen saturation, does not rely on pulsatility and may be used in patients with CF devices in conjunction with other monitoring modalities.50
Noninvasive intraoperative blood pressure monitoring techniques include automatic oscillometric blood pressure measurements using a cuff and Doppler assessment of arterial pressure. The data on the use of automatic NIBP monitoring are variable. Some studies have noted adequate correlation of NIBP monitoring with invasive intra-arterial catheters, suggesting adequacy of NIBP for intraoperative hemodynamic monitoring without the use of adjunctive modalities. However, other published reports suggest that NIBP does not consistently render readings when measurements are attempted.40 Separate investigations performed by Bennett and colleagues and Coyle and colleagues reported measurement failures in approximately half of the cases, which may prove problematic because of prolonged interruptions in hemodynamic monitoring.51,52 Bennet and colleagues found that measurements could be obtained using Doppler assessment in 94% of patients, which largely correlated with the measurements obtained using intra-arterial catheters.51 Doppler measurements are of limited utility intraoperatively as repetitive measurements are labor intensive and may not be possible in cases where patient positioning prevents ready access to an extremity. In addition, even when measurements are consistently obtainable, distinguishing between the mean arterial pressure and systolic blood pressure may prove to be challenging.53
Invasive arterial pressure monitoring using intra-arterial catheters remains the gold standard for hemodynamic monitoring in patients with CF LVADs.54 However, the number of noncardiac cases in which arterial catheters are used seems to be decreasing in recent years, likely owing to growing familiarity and comfort in managing patients with these devices and a desire to preserve arterial access points for future procedures or transplant.40 Determination of an optimal blood pressure-monitoring strategy requires careful consideration of both physiological and procedural risks. Invasive monitoring should be utilized in patients with poor physiological reserve (those with higher NYHA classifications) and in those undergoing higher risk surgical interventions.
Other invasive hemodynamic monitoring devices, such as central venous and pulmonary artery catheters and transesophageal echocardiography (TEE), may be of benefit in higher risk patients or in those undergoing more complex procedures. These modalities may offer utility in establishing volume status, detecting RV dysfunction, and monitoring patients with pulmonary hypertension. Although no formal guidelines are currently in place for patients with CF LVADs undergoing NCS, utilization of these monitoring modalities requires careful consideration of overall physiological and surgical risks and benefits. Invasive catheters offer the benefits of detecting and diagnosing the etiology of intraoperative and postoperative hemodynamic changes, monitoring volume status and guiding resuscitation, and facilitating remote consultation with CF LVAD experts. However, they carry the risks of bleeding, arrhythmia, and potential scarring of venous access sites. The use of TEE may obviate the need for invasive catheters by similarly offering real-time monitoring of myocardial function, septal positioning, and volume status, but is typically limited to the intraoperative setting and generally cannot be used for prolonged periods of time in the postoperative setting. Furthermore, TEE may not be available widely in every operative setting and its use is often restricted to cardiac anesthesiologists with expertise in echocardiography.
As mentioned previously, the presence and type of adjunctive cardiac devices such as pacemakers and defibrillators must be ascertained preoperatively so that appropriate precautions may be taken to manage electrocautery and arrhythmias. Pacemaker interrogation reports should be obtained or the device interrogated preoperatively when possible. Magnet settings should be clarified and reprogramming performed when necessary. Implanted defibrillators are likely to be triggered by the use of electrocautery and should be turned off, either by an electrophysiology specialist or by the use of an external magnet. In all cases, external defibrillator pads should be applied as arrhythmias may be poorly tolerated by patients with CF LVADs. Whenever possible, bipolar cautery should be used.
Positioning and Surgical Technique
Although NCS interventions have been performed successfully in patients with CF LVADs in a variety of physical positions, perioperative physicians must be aware and vigilant of the impact that these positions may have on physiology and LVAD parameters. Trendelenberg positioning augments venous return and may precipitate RV overload and dysfunction. Reverse Trendelenberg, lateral decubitus, and prone positioning all diminish preload and may impair adequate LV filling and compromise the cardiac index. Depending on functional status and degree of volume loading, individual patients may have varying degrees of tolerance for a particular position, and care must be taken to ensure hemodynamic stability before making an incision.7
Any alteration to the position of the patient or the bed should be performed gradually, with close attention to vital signs and LVAD parameters. If necessary, the surgical technique may need to be altered if a patient is unable to tolerate a particular position. As an example, interventions that would normally be performed in the prone position may need to be performed in a modified lateral position. In all cases, care should be taken to appropriately position external CF LVAD equipment to avoid impingement upon the driveline, damage to the controller, or pressure against the patient’s skin.
The physiological impact of surgical technique on CF LVADs must also be considered. This is perhaps most notable in laparoscopic procedures requiring insufflation, which can augment afterload and compromise preload. Patients who experience hemodynamic compromise may require lower abdominal insufflation pressures.55 Other interventions, such as thoracic procedures requiring one lung ventilation, should also be carefully considered, given the particularly serious consequences of hypoventilation and hypoxia in this patient population.
Intraoperative management of patients with CF LVADs should focus on protection of the RV and maintenance of adequate preload and afterload.21,24 As mentioned previously, a CF LVAD must receive adequate preload from the RV to function effectively. RV protection is therefore of paramount importance in managing this patient population. Protection strategies require consideration of baseline RV function and avoidance of iatrogenic RV strain associated with volume overload and increases in PVR. Hypervolemia from inadequate preoperative optimization or overaggressive fluid administration may cause RV distention and compromise function. Physiological factors that increase PVR, including hypoxia, hypercarbia, and acidosis, should be strictly avoided. Spontaneous ventilation is generally preferred because of its favorable effect on PVR and venous return. However, the risk associated with positive pressure ventilation may be preferable to the adverse hemodynamic consequences associated with hypoventilation-associated increases in PVR in susceptible patients, specifically those with poor baseline RV function or pulmonary hypertension.37,56
Data from pulmonary artery catheters and TEE are most useful in determining the presence of RV dysfunction. However, elevations in central venous pressure combined with a low-flow pulsatility or PI, a decrease in LVAD flows, or evidence of low cardiac output (such as increases in lactate production or low urine output) may also be suggestive of a declining RV.57 Supportive measures include initiation of inotropic support to aid in RV contractility, vasopressors to augment afterload and ensure adequate RV perfusion, inhaled pulmonary vasodilators to reduce RV afterload and promote forward flow, and diuretic therapy to mitigate the detrimental effect of hypervolemia on the RV. Heart rate augmentation may also be of benefit by limiting the RV filling time and increasing right-sided output. Chronotropic agents such as dobutamine may be beneficial, with careful attention to any vasodilatory effects. Patients with implanted pacemakers may also benefit from intraoperative reprogramming to a higher pacemaker rate.
Volume status is an important consideration, given the preload dependence of CF LVADs. Both hypovolemia and hypervolemia are poorly tolerated, and judicious administration of intravenous fluids is of paramount importance. Low intravascular volume may be present preoperatively secondary to NPO status, and these patients may derive hemodynamic benefit from modest volume loading before an operative intervention.7 Hypovolemia may also occur intraoperatively as a result of bleeding or inadequate resuscitation. Hypervolemia may exist preoperatively in patients who have not been sufficiently diuresed or optimized for surgery or intraoperatively as an iatrogenic occurrence secondary to aggressive volume resuscitation.
In the absence of invasive monitors, perioperative physicians must utilize multiple sources of data to help guide volume management and resuscitation. These include preoperative volume status, intraoperative blood and volume loss, hemodynamics, and knowledge of the patient’s native RV function. In addition, laboratory evaluations such as arterial blood gases may be useful in identifying acidosis or evidence of inadequate perfusion. In situations where volume loss is significant, aggressive resuscitation is required to prevent suction events from both hypovolemia or acute RV overload and strain. A transient decrease in LVAD speed may also be required to prevent suction events until the patient has been adequately resuscitated. In these instances, the above-mentioned invasive monitoring modalities are useful in guiding fluid administration. TEE would allow for the evaluation of septal positioning and RV and LV chamber size during resuscitation, and optimal reductions in device speed if necessary. Central venous and pulmonary artery catheters also yield data on intravascular volume status and RV function. Inotropes, vasopressors, or inhaled pulmonary vasodilator therapy may be required to mitigate RV strain and provide hemodynamic support during active resuscitation.
As mentioned previously, CF LVAD devices are afterload sensitive, and maintenance of adequate systemic perfusion pressure is required for both RV and end-organ protection. By way of review, hypertension compromises the systemic cardiac output delivered by the LVAD, exacerbating heart failure. Hypotension augments CF LVAD flow to the detriment of end-organ perfusion. Current guidelines recommend MAP targets of 70 to 90 mm Hg to optimize cardiac output and perfusion,23 which may require the use of systemic vasoconstrictors or vasodilator therapy.
Trends in displayed LVAD parameters and CF LVAD alarms also provide an insight into changes in physiology (Table 2), but must be considered within the context of other clinical data points, such as estimated blood loss or urine output. Decreases from baseline flow pulsatility or PI (depending on the device) and estimated flow are suggestive of inadequate LV preload, which may occur as a result of hypovolemia, or from RV dysfunction from hypervolemia or elevated PVR.57 Therefore, trends in LVAD parameters must not be considered in isolation, but rather in conjunction with other clinical data points as the etiologies of these changes may have diametrically opposite treatment paradigms. When changes are noted but the cause is ambiguous, invasive monitoring modalities such as invasive arterial pressure monitoring, TEE, or central venous or pulmonary artery catheters may be required.
All CF LVAD devices have visual and auditory alarms to signal concerns associated with device components.58 Red alarms in both HeartMate and HeartWare devices are high-priority alarms and may indicate device stoppage, controller failure, disconnection of power supplies or driveline, or critical battery depletion. Yellow alarms are of medium priority and signal low battery, single power line or power source disconnection, low-flow states, or hardware faults.30–32 Alarms that are triggered intraoperatively during NCS should be addressed immediately with institutional MCS clinicians or with clinical specialists from the device manufacturer.
Published data show that atrial and ventricular arrhythmias are common among heart failure patients supported with CF LVAD devices.59 Although rhythm disturbances may be reasonably well tolerated in supported patients, they can lead to worsening heart failure symptoms, hemodynamic instability, and poor outcomes. Perioperative physicians must ascertain the presence of previous arrhythmia burden, closely monitor for the development of intraoperative rhythm disturbances, and correct underlying causes.
Atrial arrhythmias are associated with advanced heart failure and have been reported in over 50% of patients with CF LVADs.60 Many of these patients receive diuretic therapy, which also predisposes them to electrolyte abnormalities. Atrial arrhythmias with associated hemodynamic stability may be monitored or conservatively managed while underlying causes are corrected; caution must be exercised with the administration of any antiarrhythmic medication with a negative inotropic effect. More emergent intervention, such as cardioversion, is required if these arrhythmias compromise right-sided cardiac output, which can impair LV filling.59
Ventricular arrhythmias have also been reported in up to 59% of patients with CF LVADs.61 Predisposing factors include medication-induced electrolyte disturbances, preexisting myocardial scarring associated with underlying cardiac pathology, use of inotropic agents, and the development of suction events in which the inflow cannula comes into contact with the septal wall.62 Ventricular arrhythmias that develop in the first week after CF LVAD implantation are associated with a marked increase in mortality. With progressive remodeling that occurs over time, these arrhythmias seem to be better tolerated,59 with reports of patients living with sustained rhythm disturbances for extended durations without adverse consequences. However, there is concern that prolonged ventricular arrhythmias may result in LV thrombus formation.
If patients remain hemodynamically stable in the setting of a perioperative ventricular arrhythmia, a more conservative strategy may be followed, including identification and correction of the underlying etiology, a decrease in inotropic therapy when possible, and the intravenous administration of antiarrhythmic agents, such as amiodarone and lidocaine. Echocardiography may be utilized to diagnose a suction event and optimize speed or intravascular volume. When ventricular arrhythmias compromise right heart function and cause hemodynamic compromise, more aggressive measures are required. These include cardioversion or defibrillation and adjustments in LVAD speed. Many, but not all, patients with CF LVADs have implanted defibrillators as part of primary and secondary prevention paradigms, which, if not deactivated, may be triggered to treat these arrhythmias. Perioperative physicians must therefore be aware of the presence and type of implanted device.
As discussed previously, suction events occur when there is inadequate LV preload, resulting in leftward shift of the interventricular septum, collapse of the LV cavity, and contact between the inflow cannula and the ventricular wall. Although echocardiography offers a clear diagnosis of a suction event (Fig. 7), impending suck-down can be identified by downward trends in LVAD pulsatility and flow on the HeartMate II, and flattening or negative deflections in the flow pulsatility waveform on the HVAD clinical monitoring screen (Fig. 8). Although often tolerated, these events are potentially catastrophic and may lead to refractory ventricular arrhythmias, severe hemodynamic compromise, and profound hypoperfusion.
CF devices will respond to suction events by lowering pump speed to a set lower limit and then gradually attempting to ramp the speed back up to the set speed. When this algorithmic response cannot relieve a suction event, LVAD speed will need to be decreased manually.63 Cardioversion or defibrillation may be required for ventricular arrhythmias, and management should be directed at the underlying cause of compromised LV preload. If the cause is hypovolemia, fluids should be administered. In cases of RV dysfunction, RV-protective measures using inotropic support and/or inhaled pulmonary vasodilator therapy may be required. In either circumstance, a careful echocardiographic evaluation of the RV after a hemodynamically significant suction event should be performed to evaluate for any sequelae associated with hypoperfusion and to optimize speed on the basis of RV function and septal position.
Cardiac arrest is particularly problematic in patients supported with CF LVADs because of the baseline lack of palpable pulses, which can impede the functionality of standard monitoring devices (NIBP and pulse oximetry). In the absence of invasive monitoring modalities, a physical examination should be performed to assess for perfusion, and the use of adjunctive monitoring devices, such as waveform capnography and cerebral oximetry, may be required. The AHA recommends using PETCO2 <20 mm Hg as a threshold for the presence of hypoperfusion in CF LVAD patients.64
The use of chest compressions in this patient population has been of considerable debate. Device manufacturers recommend against external compressions because of the risk of cannula dislodgement. Other parties have suggested a range of plausible options, including abdominal compressions,65 standard compressions,66,67 or the immediate insertion of adjunctive MCS devices.24 Ultimately, the decision to initiate advanced cardiac life support is made on the basis of provider judgment and available institutional resources.
Device failure has long been considered a rare occurrence,21 with published reports focusing primarily on the occurrence of pump thrombus, although perioperative physicians should be aware that malfunction can also occur in the controller and peripheral components.68 In the event of a catastrophic device failure, aggressive pharmacologic support should be instituted including inotropic and vasopressor support until troubleshooting or surgical intervention can be performed. In cases of extreme physiological decompensation, the initiation of extracorporeal membrane oxygenation may be considered.
When NCS is uncomplicated, the postoperative care of patients with CF LVADs should not differ markedly from routine postoperative management for a given procedure. However, perioperative providers must pay particular attention to hemostasis, volume status, fluid shifts, electrolyte levels, cardiac rhythm, and adequacy of oxygenation and ventilation. The postprocedure recovery location (postanesthesia care unit vs ICU) will likely depend on institutional practice and familiarity with CF LVADs, but care has been safely provided in the routine postanesthesia care unit.7,8 More monitored settings may be desired for higher risk patients and those with ongoing fluid shifts or persistent need for vasoactive support.
Determination of timing to resume anticoagulation will require multidisciplinary discussion between the perioperative team and the patient’s LVAD physician as the risk of hemorrhagic complications must be balanced with the risk of pump thrombus. Various anticoagulation strategies have been utilized in the perioperative period, ranging from complete discontinuation of oral anticoagulants and antiplatelet therapy, immediate reinitiation postoperatively, or therapeutic bridging with heparin after a prespecified duration or after confirmation of hemostasis.7
Similar multidisciplinary discussion may be required for resumption of the patient’s preoperative heart failure regimen. Depending on the intraoperative course and details of the surgical procedure, patients may fully return to their preoperative regimen or require gradual reinitiation.
With the rise in the use of durable, implanted CF LVADs for end-stage heart failure and increasing postimplant longevity, perioperative care for supported patients will continue to expand to non-MCS centers and be increasingly managed by non–cardiac-trained anesthesiologists. Thus, perioperative physicians in all clinical settings are likely to encounter patients with these devices, both for routine and for more emergent surgical interventions. Therefore, familiarity with CF LVAD-supported physiology, hemodynamic management goals, and emergency treatment paradigms is essential.
1. Savarese G, Lund LH. Global public health burden of heart failure. Card Fail Rev. 2017;3:7–11.
2. Roger VL. Epidemiology of heart failure. Circ Res. 2013;113:646–659.
3. Fang JC, Ewald GA, Allen LA, et al. Advanced (stage D) heart failure: a statement from the Heart Failure Society of America Guidelines Committee. J Card Fail. 2015;21:519–534.
4. Peura JL, Colvin-Adams M, Francis GS, et al. Recommendations for the use of mechanical circulatory support: device strategies and patient selection: a scientific statement from the American Heart Association. Circulation. 2012;126:2648–2667.
5. Pinney SP, Anyanwu AC, Lala A, et al. Left ventricular assist devices for lifelong support. J Am Coll Cardiol. 2017;69:2845–2861.
6. Kirklin JK, Pagani FD, Kormos RL, et al. Eighth annual INTERMACS report: special focus on framing the impact of adverse events. J Heart Lung Transplant. 2017;36:1080–1086.
7. Roberts SM, Hovord DG, Kodavatiganti R, et al. Ventricular assist devices and non-cardiac surgery. BMC Anesthesiol. 2015;15:185.
8. Barbara DW, Wetzel DR, Pulido JN, et al. The perioperative management of patients with left ventricular assist devices undergoing noncardiac surgery. Mayo Clin Proc. 2013;88:674–682.
9. Morgan JA, Paone G, Nemeh HW, et al. Non-cardiac surgery in patients on long-term left ventricular assist device support. J Heart Lung Transplant. 2012;31:757–763.
10. McKellar SH, Morris DS, Mauermann WJ, et al. Evolution of general surgical problems in patients with left ventricular assist devices. Surgery. 2012;152:896–902.
11. Stone M, Hinchey J, Sattler C, et al. Trends in the management of patients with left ventricular assist devices presenting for noncardiac surgery: a 10-year institutional experience. Semin Cardiothorac Vasc Anesth. 2016;20:197–204.
12. Rose EA, Gelijns AC, Moskowitz AJ, et al. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med. 2001;345:1435–1443.
13. Prinzing A, Herold U, Berkefeld A, et al. Left ventricular assist devices-current state and perspectives. J Thorac Dis. 2016;8:E660–E666.
14. Schumer EM, Black MC, Monreal G, et al. Left ventricular assist devices: current controversies and future directions. Eur Heart J. 2016;37:3434–3439.
15. Slaughter MS, Rogers JG, Milano CA, et al. Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med. 2009;361:2241–2251.
16. Sheikh FH, Russell SD. HeartMate(R) II continuous-flow left ventricular assist system. Expert Rev Med Devices. 2011;8:11–21.
17. Aaronson KD, Slaughter MS, Miller LW, et al. Use of an intrapericardial, continuous-flow, centrifugal pump in patients awaiting heart transplantation. Circulation. 2012;125:3191–3200.
18. Slaughter MS, Pagani FD, McGee EC, et al. HeartWare ventricular assist system for bridge to transplant: combined results of the bridge to transplant and continued access protocol trial. J Heart Lung Transplant. 2013;32:675–683.
19. Mehra MR, Naka Y, Uriel N, et al. A fully magnetically levitated circulatory pump for advanced heart failure. N Engl J Med. 2017;376:440–450.
20. Zhigalov K, Mashhour A, Szczechowicz M, et al. Clinical outcome and comparison of three different left ventricular assist devices in a high-risk cohort. Artif Organs. 2018. [Epub ahead of print].
21. Chung M. Perioperative management of the patient with a left ventricular assist device for noncardiac surgery. Anesth Analg. 2018;126:1839–1850.
22. Levine BD, Cornwell WK 3rd, Drazner MH. Factors influencing the rate of flow through continuous-flow left ventricular assist devices at rest and with exercise. JACC Heart Fail. 2014;2:331–334.
23. Slaughter MS, Pagani FD, Rogers JG, et al. Clinical management of continuous-flow left ventricular assist devices in advanced heart failure. J Heart Lung Transplant. 2010;29:S1–S39.
24. Hessel EA 2nd. Management of patients with implanted ventricular assist devices for noncardiac surgery: a clinical review. Semin Cardiothorac Vasc Anesth. 2014;18:57–70.
25. Riha H, Netuka I, Kotulak T, et al. Anesthesia management of a patient with a ventricular assist device for noncardiac surgery. Semin Cardiothorac Vasc Anesth. 2010;14:29–31.
26. Salamonsen RF, Mason DG, Ayre PJ. Response of rotary blood pumps to changes in preload and afterload at a fixed speed setting are unphysiological when compared with the natural heart. Artif Organs. 2011;35:E47–E53.
27. Sen A, Larson JS, Kashani KB, et al. Mechanical circulatory assist devices: a primer for critical care and emergency physicians. Crit Care. 2016;20:153.
28. Guha A, Eshelbrenner CL, Richards DM, et al. Gastrointestinal bleeding after continuous-flow left ventricular device implantation: review of pathophysiology and management. Methodist Debakey Cardiovasc J. 2015;11:24–27.
29. Cushing K, Kushnir V. Gastrointestinal bleeding following LVAD placement from top to bottom. Dig Dis Sci. 2016;61:1440–1447.
33. Rich JD, Burkhoff D. HVAD flow waveform morphologies: theoretical foundation and implications for clinical practice. ASAIO J. 2017;63:526–535.
34. Jorde UP, Aaronson KD, Najjar SS, et al. Identification and management of pump thrombus in the HeartWare left ventricular assist device system: a novel approach using log file analysis. JACC Heart Fail. 2015;3:849–856.
35. Khoo KA. Ventricular assist devices and anesthetic implications for noncardiac procedures. AANA J. 2010;78:483–488.
36. Lampert BC, Teuteberg JJ. Right ventricular failure after left ventricular assist devices. J Heart Lung Transplant. 2015;34:1123–1130.
37. Haddad F, Couture P, Tousignant C, et al. The right ventricle in cardiac surgery, a perioperative perspective: II. Pathophysiology, clinical importance, and management. Anesth Analg. 2009;108:422–433.
38. Haddad F, Couture P, Tousignant C, et al. The right ventricle in cardiac surgery, a perioperative perspective: I. Anatomy, physiology, and assessment. Anesth Analg. 2009;108:407–421.
39. Maltais S, Kilic A, Nathan S, et al. PREVENtion of HeartMate II pump thrombosis through clinical management: the PREVENT multi-center study. J Heart Lung Transplant. 2017;36:1–12.
40. Mathis MR, Sathishkumar S, Kheterpal S, et al. Complications, risk factors, and staffing patterns for noncardiac surgery in patients with left ventricular assist devices. Anesthesiology. 2017;126:450–460.
41. Boyle AJ, Russell SD, Teuteberg JJ, et al. Low thromboembolism and pump thrombosis with the HeartMate II left ventricular assist device: analysis of outpatient anti-coagulation. J Heart Lung Transplant. 2009;28:881–887.
42. Stern DR, Kazam J, Edwards P, et al. Increased incidence of gastrointestinal bleeding following implantation of the HeartMate II LVAD. J Card Surg. 2010;25:352–356.
43. Katz JN, Adamson RM, John R, et al. Safety of reduced anti-thrombotic strategies in HeartMate II patients: a one-year analysis of the US-TRACE study. J Heart Lung Transplant. 2015;34:1542–1548.
44. Netuka I, Litzler PY, Berchtold-Herz M, et al. Outcomes in HeartMate II patients with no antiplatelet therapy: 2-year results from the European TRACE study. Ann Thorac Surg. 2017;103:1262–1268.
45. Suarez J, Patel CB, Felker GM, et al. Mechanisms of bleeding and approach to patients with axial-flow left ventricular assist devices. Circ Heart Fail. 2011;4:779–784.
46. Nascimbene A, Neelamegham S, Frazier OH, et al. Acquired von Willebrand syndrome associated with left ventricular assist device. Blood. 2016;127:3133–3141.
47. Crow S, Chen D, Milano C, et al. Acquired von Willebrand syndrome in continuous-flow ventricular assist device recipients. Ann Thorac Surg. 2010;90:1263–1269; discussion 1269.
48. Boyle AJ, Jorde UP, Sun B, et al. Pre-operative risk factors of bleeding and stroke during left ventricular assist device support: an analysis of more than 900 HeartMate II outpatients. J Am Coll Cardiol. 2014;63:880–888.
49. Degnan M, Brodt J, Rodriguez-Blanco Y. Perioperative management of patients with left ventricular assist devices undergoing noncardiac surgery. Ann Card Anaesth. 2016;19:676–686.
50. Oleyar M, Stone M, Neustein SM. Perioperative management of a patient with a nonpulsatile left ventricular-assist device presenting for noncardiac surgery. J Cardiothorac Vasc Anesth. 2010;24:820–823.
51. Bennett MK, Roberts CA, Dordunoo D, et al. Ideal methodology to assess systemic blood pressure in patients with continuous-flow left ventricular assist devices. J Heart Lung Transplant. 2010;29:593–594.
52. Coyle LA, Gallagher C, Maier J, et al. Measurement of blood pressure during support with a continuous-flow left ventricular assist device in the outpatient setting. J Heart Lung Transplant. 2013;32:S235.
53. Lanier GM, Orlanes K, Hayashi Y, et al. Validity and reliability of a novel slow cuff-deflation system for noninvasive blood pressure monitoring in patients with continuous-flow left ventricular assist device. Circ Heart Fail. 2013;6:1005–1012.
54. Castagna F, Stohr EJ, Pinsino A, et al. The unique blood pressures and pulsatility of LVAD patients: current challenges and future opportunities. Curr Hypertens Rep. 2017;19:85.
55. Eck DL, Belli EV, Smith CD, et al. Laparoscopic cholecystectomy in patients with HeartMate II left ventricular assist devices. J Laparoendosc Adv Surg Tech A. 2014;24:100–103.
56. Haddad F, Elmi-Sarabi M, Fadel E, et al. Pearls and pitfalls in managing right heart failure in cardiac surgery. Curr Opin Anaesthesiol. 2016;29:68–79.
57. Devore AD, Mentz RJ, Patel CB. Medical management of patients with continuous-flow left ventricular assist devices. Curr Treat Options Cardiovasc Med. 2014;16:283.
58. Estep JD, Trachtenberg BH, Loza LP, et al. Continuous flow left ventricular assist devices: shared care goals of monitoring and treating patients. Methodist Debakey Cardiovasc J. 2015;11:33–44.
59. Kadado AJ, Akar JG, Hummel JP. Arrhythmias after left ventricular assist device implantation: incidence and management. Trends Cardiovasc Med. 2018;28:41–50.
60. Enriquez AD, Calenda B, Gandhi PU, et al. Clinical impact of atrial fibrillation in patients with the HeartMate II left ventricular assist device. J Am Coll Cardiol. 2014;64:1883–1890.
61. Nakahara S, Chien C, Gelow J, et al. Ventricular arrhythmias after left ventricular assist device. Circ Arrhythm Electrophysiol. 2013;6:648–654.
62. Griffin JM, Katz JN. The burden of ventricular arrhythmias following left ventricular assist device implantation. Arrhythm Electrophysiol Rev. 2014;3:145–148.
63. Stainback RF, Estep JD, Agler DA, et al. Echocardiography in the Management of patients with left ventricular assist devices: recommendations from the American Society of Echocardiography. J Am Soc Echocardiogr. 2015;28:853–909.
64. Peberdy MA, Gluck JA, Ornato JP, et al. Cardiopulmonary resuscitation in adults and children with mechanical circulatory support: a scientific statement from the American Heart Association. Circulation. 2017;135:e1115–e1134.
65. Rottenberg EM, Heard J, Hamlin R, et al. Abdominal only CPR during cardiac arrest for a patient with an LVAD during resternotomy: a case report. J Cardiothorac Surg. 2011;6:91.
66. Shinar Z, Bellezzo J, Stahovich M, et al. Chest compressions may be safe in arresting patients with left ventricular assist devices (LVADs). Resuscitation. 2014;85:702–704.
67. Mabvuure NT, Rodrigues JN. External cardiac compression during cardiopulmonary resuscitation of patients with left ventricular assist devices. Interact Cardiovasc Thorac Surg. 2014;19:286–289.
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
68. Kormos RL, McCall M, Althouse A, et al. Left ventricular assist device malfunctions: it is more than just the pump. Circulation. 2017;136:1714–1725.