Left Ventricular Assist Device–Supported Patient Presenting for Noncardiac Surgery


Key Points

  • 1.

    Regardless of the level of complexity or invasiveness of the planned procedure, the perioperative considerations and the anesthetic approach to left ventricular assist device (LVAD)–supported patients are the same because the removal of sympathetic tone by sedation or induction of general anesthesia should be expected to initially exert the same effect on the physiology of ventricular assist device (VAD)-supported patients regardless of the planned procedure.

  • 2.

    A team-based approach and preoperative planning regarding intraoperative management and postoperative recovery location are key to the successful perioperative management of VAD-supported patients presenting for noncardiac surgery.

  • 3.

    An understanding of the physiology of the VAD-supported state is the key to safe intraoperative management.

  • 4.

    No specific sedatives or anesthetic agents are contraindicated because of the presence of a VAD, but the required anticoagulation often precludes major regional techniques.

  • 5.

    Most patients with a modern nonpulsatile left VAD (LVAD) do exhibit pulsatility of their circulation; however, they can lose this pulsatility after induction because of the relative hypovolemia and vasodilation that accompany an anesthetic, bringing considerations of appropriate monitoring.

  • 6.

    Optimization of volume status will help maintain pulsatility of the circulation in a VAD-supported patient.

  • 7.

    Intraoperative changes to baseline VAD settings are rarely (if ever) needed in a VAD-supported patient who was optimized on these settings when not anesthetized.

Role of Ventricular Assist Devices in the Management of Heart Failure

The prevalence of heart failure (HF) worldwide is estimated to be about 26 million people. In the United States alone, there are approximately 5.7 million adults with HF, and this number is projected to increase to approximately 8 million by the year 2030. Mechanical circulatory support (MCS) with a left ventricular assist device (LVAD) is now the standard management for patients with chronic refractory HF. The goals of LVAD support are twofold: (1) to decompress the failing left ventricle, thus dramatically reducing left ventricular (LV) myocardial oxygen demand (which, in certain circumstances, may promote recovery of the failing myocardium), and (2) to maintain adequate systemic perfusion to avert cardiogenic shock. The pump itself is attached to the heart and great vessels by cannulae that allow continuous collection of blood returning to the left side of the heart and ejection of that blood into the aorta.

According to the latest data from the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS), there are currently 2000 to 3000 LVAD implantations annually at approximately 160 centers in the United States alone. Table 5.1 outlines the current indications for long-term LVAD support, as well as the current frequency and current success of each indication in the United States.

Table 5.1
Indications, Explanations, Current Frequency, and Current Success Rates for Implantations of Durable LVADs in the United States
Indication Explanation Current U.S. Frequency (%) Current U.S. Success
Bridge to transplantation The LVAD is used to bridge the patient with chronic, progressive heart failure to transplantation. This includes patients with an acute exacerbation of chronic heart failure. 26 86% alive at 1 year
31% transplanted
55% still supported
Bridge to candidacy The LVAD is used to restore systemic perfusion to an adequate level and thus improve multisystem organ failure such that the patient might be an acceptable transplant candidate. 37 84% alive at 1 year
20% transplanted
64% still supported
Destination therapy The LVAD is used as a final, permanent management strategy for end-stage, refractory heart failure in a transplant-ineligible patient. 46 >75% alive at 1 year
>50% alive at 3 years
LVAD, Left ventricular assist device.

Until 2009, bridge to transplantation (BTT) was the most common indication for implantation of a durable LVAD, but the approval of the HeartMate II for destination therapy (DT) in 2010 heralded a new era of MCS because before that, a durable device that could provide years of support did not exist. Continuous-flow (CF) devices (e.g., the HeartMate II) have now been used to provide support for 100% of patients implanted for DT since 2010, as well as for more than 95% of all other LVAD indications. The first generation of pulsatile, implantable devices is essentially no longer in use.

The most common indication for LVAD implantation is now DT (see Table 5.1 ), with BTC the second most common indication and BTT (the traditional indication before 2010) now third most common. Overall, all-comer survival with a durable LVAD now approaches 80% at 1 year, and the 4-year survival rate now approaches 50%. As the survival rate has increased, the number of patients supported by LVADs requiring interventional and diagnostic procedures and noncardiac surgery (NCS) procedures has increased. The volume of NCS in LVAD-supported patients varies from institution to institution and practice to practice, but current trends indicate that the vast majority of NCS procedures performed in this population are now diagnostic and therapeutic endoscopies. Although supported patients still tend to receive their care in the academic VAD centers, there has been some expansion into the private practice settings and even some endoscopy centers.

INTERMACS

INTERMACS is a North American registry database sponsored by the National Heart, Lung and Blood Institute; the Food and Drug Administration (FDA); and the Centers for Medicare and Medicaid Services (CMS). Centered at the University of Alabama at Birmingham, INTERMACS was established in 2005 for patients receiving long-term MCS therapy with implantable, durable devices to treat advanced HF. Essentially, INTERMACS collects clinical data about VAD patients as it happens. Postimplant follow-up data are collected at 1 week, 1 month, 3 months, and 6 months and every 6 months thereafter. Major outcomes after implant (e.g., death, transplant, explant, rehospitalization, and adverse events) are entered by implanting centers as such events occur and at defined follow-up time points, along with more “complex” endpoints (e.g., the patient's level of function and quality of life), which are critical to the evaluation of current MCS therapy, for which improvements in both survival and function have been compelling. These indices are becoming increasingly important as survival improves, and new devices will be compared for outcomes beyond simple survival. A similar European-based database called EuroMACS exists in Europe, and there is also a database of pediatric MCS called PEDIMACS. A new international database maintained by the International Society for Heart and Lung Transplantation (ISHLT) called IMACS now exists, and reports of the international experience will soon provide data regarding international outcomes.

Regarding LVAD implantation by indication, the most recent INTERMACS data available at the time of this writing report that DT continues to be the most prevalent indication for LVAD implantation, having increased to 45.7% of all implants in 2014 (compared with 14.7% in 2006 and 2007, and 28.6% between 2008 and 2011). In the sixth annual report (released in 2014), BTC was the second most common modern indication for VADs, with BTT in third place, but in the seventh annual report (released in 2015), 30% of patients were already listed for transplantation at the time of implantation, with an additional 23% implanted as a BTC. “Bridge to recovery” with short-term VADs continues to constitute only a very small percentage of the usage of this technology in the most current report (0.2% in 2014). Additional data available from INTERMACS regard survival by both timing of implantation and by type of device .

The INTERMACS profile (also called the INTERMACS level) describes the clinical condition of the patient on a scale from 1 to 7, with a numerically lower profile indicating more severe illness. A level 7 patient is simply in the advanced stages of HF (e.g., New York Heart Association class III), and the clinical condition of the patient gets worse as the INTERMACS profile number gets lower. For example, a level 4 patient has symptoms at rest, a level 3 patient is essentially hemodynamically stable but inotrope dependent, a level 2 patient is deteriorating despite inotropes, and a level 1 patient is essentially in cardiogenic shock despite maximal therapy.

The experience has been that if a durable LVAD is implanted too early (at numerically higher INTERMACS levels), the risks of adverse events outweigh the benefits. Conversely, if the VAD is not implanted until the patient is already likely developing multisystem organ failure (e.g., level 1), the likelihood of ultimate rescue is low, and the survival rate is poor. Survival data suggest that implantation of durable LVADs when the patient is level 3 or 4 would be ideal to balance the risks and benefits . Large multicenter head-to-head trials conducted in the modern era with modern devices (e.g., Momentum 3, Endurance) have reported the profile of risks and benefits associated with each of the modern devices (see Suggested Reading ).

Specific Devices in Current Use

The two most commonly implanted FDA-approved durable devices in the United States are the HeartMate II (Abbott) and the HeartWare HVAD (Medtronic). The Heartmate 3 is a relatively recently introduced implantable, durable device that has received FDA approval for certain indications, although approval of other indications is still pending at the time of this writing.

HeartMate II

The HeartMate II (HM II; Fig. 5.1 ) is currently the most commonly implanted durable LVAD in the United States and in many countries around the world. The HM II is a miniaturized “second-generation” continuous axial flow pump that was FDA approved as a BTT in 2008 and as DT in 2010. According to the manufacturer, more than 16,000 patients worldwide have received the HM II, with the longest duration of support more than 8 years. Although the impeller is the only moving part, it is stabilized at both ends by bearings. Current postimplantation protocols call for warfarin anticoagulation to an international normalized ratio (INR) of 2.5 to 3.5 plus aspirin. The currently reported rate of successful BTT with the HM II is approximately 86%. Fig. 5.2 shows and discusses details regarding parameters displayed on the HM II clinical control screen.

Fig. 5.1, HeartMate II.

Fig. 5.2, Clinical control screen of the HeartMate II (HM II). Pump flow is continuous estimate of the output from the device (derived from the speed of the impeller and the power it takes to achieve that speed). Flows encountered clinically usually range from 4 to 6 L/min, but the device is capable of flowing up to 10 L/min. If the outflow is less than the lower limit set as the alarm condition, three dashes (—) will be displayed in this box instead of a number. This does not necessarily mean there is no outflow. It only means there is less flow than the lower limit set for the alarm. There is a very loud screeching alarm annunciated from the controller if there is no outflow. This is an exceedingly rare thing to encounter. The pump speed is the number of revolutions per minute (rpm) at which the impeller is rotating. In most situations, this is a set and fixed value. Speeds encountered clinically are usually in the range of 9000 to 10,000 rpm, but some centers run the ventricular assist device (VAD) at lower rotational speeds to allow the left ventricle (LV) to do more work. Increases in speed will facilitate ventricular unloading by increasing flow through the pump. If the amount of flow exceeds the available volume in the ventricle, a “suckdown” will occur. Decreasing the speed can potentially increase the volume in the LV, although initial steps to increase LV volume would ideally involve infusing volume or supporting right ventricle (RV) function as needed. The pulsatility index (PI) is a unitless index of how much pulsatility the device senses as a result of ventricular contractions. Initially, the failed ventricle contributes very little (which is why a VAD was needed) but as the excessive wall tension is decreased in the failing ventricle as a result of VAD action, the ventricle begins to recover, and as long as volume in the LV is optimized, the ventricle will again begin to contract, forcing little pulses through the VAD, as well as through the aortic valve. The PI can be used as a trend to assist with optimization of volume status. PI values around 2 to 3 are typical when there is little pulsatility and the VAD is doing most or all of the work. PI values of 4 to 6 are typical when the partially decompressed ventricle recovers. The PI will decrease with hypovolemia and will increase with myocardial recovery. Thus a low (or falling) PI likely indicates the need to increase the volume status or possibly to increase contractility. RV dysfunction can lead to a decreased filling of the LV. Pump power is the energy required to spin the impeller at the set speed and is partially determined by flow. Increases in speed or flow or resistance to flow will require increased power. Power is generally in the range of 5 to 7 W. A sudden increase in the power requirement may suggest significantly increased afterload, but it can also suggest thrombus or other obstruction to rotor rotation. These will be exceedingly rare events. Abrupt increases in power not explainable by an increase in pump speed should always be investigated. A gradual increase in power to high levels over time suggests developing thrombus in the pump.

HeartWare HVAD

The HeartWare HVAD ( Fig. 5.3 ) is a miniaturized CF centrifugal pump with a magnetically driven, hydrodynamically suspended impeller (the impeller floats in the blood without any bearings). This device is implanted within the pericardium without any significant intervening “inflow cannula”; it directly abuts the LV apex. This design provides for potential use in patients with smaller body surface areas and ostensibly results in shorter surgical implantation times. The HVAD was approved as a BTT in 2012. According to the manufacturer, more than 10,000 patients worldwide have received the HVAD, with the longest duration of support more than 7 years. Current postimplantation protocols call for warfarin anticoagulation to an INR of 2.0 to 3.0 plus aspirin. The manufacturer also recommends testing for aspirin resistance and, if detected, the adjunctive use of clopidogrel, dipyridamole, or both. The currently reported rate of successful BTT with the HVAD is 88% to 90%. The HVAD was recently approved as a DT device in the United States as a result of the ENDURANCE trial and the ENDURANCE supplemental trial. Experience with the HVAD as an implantable right ventricular assist device (RVAD) is accruing. Fig. 5.4 shows and discusses details regarding parameters displayed on the HeartWare clinical control screen.

Fig. 5.3, HeartWare HVAD.

Fig. 5.4, Clinical control screen of the HVAD. The left side of the HVAD control screen shows a continuous estimate of the output from the device in liters per minute (top left) , the speed at which the impeller is rotating in revolutions per minute (rpm) (below the output), a readout of the power consumption in watts (below the pump speed), the mode of operation (in this case, a “fixed” speed) and the status of “the suction alarm.” In the panel to the right are the power and flow waveforms. At the bottom of the screen are indicators of A/C mains power and a battery status meter. According to the manufacturer, the flow estimation (top left) should be used as a trending tool only. The readout of device flow is derived from the speed of the impeller, the power it takes to achieve that speed, and the blood viscosity. The viscosity is calculated from the patient's hematocrit, so to obtain the most accurate estimate of flows with this device, the patient's hematocrit must be input into the monitor and the hematocrit updated whenever it changes by 5% or more in either direction. Flows encountered clinically usually range from 4 to 6 L/min, but the device is capable of flowing up to 10 L/min. The amount of flow a centrifugal pump can generate is dependent on a number of factors to do with the diameter and geometry of the impeller, the capacity of the motor, and so on. However, of great importance is the pressure differential across the pump, visualized on the flow rpm at which the impeller is rotating. In most situations, this is a set and fixed value. Speeds encountered clinically are usually in the range of 2400 to 3200 rpm, but the device range is from 1800 to 4000 rpm. Increases in speed will facilitate ventricular unloading by increasing flow through the pump. If the amount of flow exceeds the available volume in the ventricle, a “suckdown” will occur. Infusing volume or decreasing the speed will increase the volume in the left ventricle. Power is the power required to spin the impeller at the set speed and is partially determined by flow. Increases in speed or flow or resistance to flow will require increased power. Power is generally in the range of 5 to 7 W. A sudden increase in the power requirement may suggest significantly increased afterload, but it can also suggest thrombus or other obstruction to rotor rotation. These will be exceedingly rare events. Abrupt increases in power not explainable by an increase in pump speed should always be investigated. A gradual increase in power to high levels over time suggests developing thrombus in the pump. The HVAD provides no numeric readout of the pulsatility, but one can physically see the pulse pressure on the flow waveform . The peaks are the flow during systole and the troughs during diastole, so the difference, in effect, reflects the “pulse pressure” or “pulsatility” of the patient during support. Of course, the difference in velocity is coming from LV contraction, forcing blood through the pump at a higher velocity during systole. This waveform can help greatly with fluid management in real time because just as in a patient without a VAD, one can increase the pulse pressure by administering fluid to optimize volume status. Maintenance of a pulse pressure is also important to prevent retrograde flow through the pump, as well as to prevent suction events. In general, the diastolic flows should be kept greater than 2 L/min, and there should be at least 2 L/min difference between systolic and diastolic flows. Even though suckdown events are rare, one nice feature of the HVAD in this regard is the “suckdown” detection and alarm. The HVAD controller establishes a diastolic flow baseline. If the diastolic flow falls to less than 40% of the established baseline for more than 10 seconds, the suckdown detection alarm will be annunciated. It would be optimal, however, to observe that the diastolic flow is decreasing and proactively prevent suckdown events from occurring in the first place. For example, volume status might be augmented if hypovolemia or vasodilation is believed to be the problem. If right ventricular (RV) dysfunction results in underfilling of the left ventricle, then RV function would be supported with inotropes, decrease the PVR, or both.

HeartMate 3

The HeartMate 3 (HM 3, Thoratec, Pleasanton, CA; see Fig. 5.5 ) is a miniaturized CF centrifugal pump with a magnetically driven, magnetically suspended impeller. It is implanted within the pericardium and thus shares some of the potential advantages of the HVAD. Design features ostensibly improve hemocompatibility and reduce the risk of thrombus formation. Similar to the HM II and the HVAD, the HM 3 can reportedly produce 10 L/min of flow. The HM 3 was demonstrated to be noninferior to the HM II in the MOMENTUM 3 trial regarding survival free from either disabling stroke or reoperation for device malfunction at 6 months after implantation. This third-generation device was FDA approved for “short-term indications” in 2017, and its evaluation for “long-term indications” (e.g., DT) is ongoing.

Fig. 5.5, HeartMate 3.

Perioperative Management

The perioperative management of an LVAD-supported patient can be divided into preoperative assessment and planning for the case, intraoperative management, and postoperative considerations.

Overview of the Preoperative Assessment

Regardless of the venue or level of complexity or invasiveness of the planned procedure, the perioperative considerations and the anesthetic approach to the LVAD-supported patient are the same because the removal of sympathetic tone by sedation or induction of general anesthesia should be expected to exert the same initial effect on the physiology of the VAD-supported patient regardless of the planned procedure. Thus a thorough, thoughtful assessment of the VAD-supported patient is mandatory, even for what appear to be the most minor of cases, because (1) even an ambulatory and seemingly uncompromised VAD-supported patient may have some level of underlying renal, hepatic, pulmonary, or central nervous system insufficiency, and (2) the physiology of the VAD-supported state can be adversely affected by inadequate optimization before and during the anesthetic. It should also be appreciated that deterioration in the perioperative period may preclude full recovery or may disqualify a patient from later heart transplantation.

If the clinician has questions or concerns, the importance of communicating in advance whenever possible about key issues with a knowledgeable colleague, the physician managing the VAD, the surgeon, and dedicated VAD staff cannot be overemphasized. Fortunately, experience has shown that the anesthetic management of a VAD-supported patient is not so different from that for a nonsupported patient, but an additional level of advanced planning is required. In addition to the usual areas of anesthetic inquiry at the preanesthetic assessment (e.g., airway, dentition, functional status, allergies), Table 5.2 outlines specific areas of focus and consideration during the preanesthetic assessment of a VAD-supported patient, and key areas are discussed in more detail later.

Table 5.2
Specific Areas of Preanesthetic Inquiry and Consideration for Ventricular Assist Device–Supported Patients
Area of Focus Rationale
End-organ insufficiency Even seemingly uncompromised VAD-supported patients may exist with varying degrees of renal, hepatic, pulmonary, or CNS insufficiency. The pathophysiology of the current surgical disease and any coexisting disease states must be taken into account when planning the optimization of the VAD-supported patient for surgery.
Presence of a CIED It is common for LVAD-supported patients to have an ICD or a pacemaker. Perioperative management of pacemakers and ICDs is the same as for any other patient undergoing the same procedure (discussed further in the text).
Anticoagulation Preoperative discussions about the appropriate level of anticoagulation for the case must take place in advance with the physician managing the VAD-supported patient and the surgeon (discussed further in the text).
Type of LVAD present The name of the LVAD present must be known, especially if seeking advice from knowledgeable colleagues about planned management.
Baseline LVAD settings and parameters of function Perioperative changes to VAD settings are rarely needed in a VAD-supported patient who was optimized on these settings when not anesthetized, so it is helpful to make note of the stable baseline settings and parameters of VAD function before altering the sympathetic tone and volume status with the delivery of an anesthetic because some of the baseline parameters potentially serve as targets during optimization. The clinical control screens of the HM II and the HVAD are depicted in Figs. 5.2 and 5.4 .
Staffing Appropriate anesthesia staffing for these procedures (e.g., cardiac vs. noncardiac trained personnel) is based on the status of the patient, the nature of the procedure, and the culture and resources of the institution and surgical venue (discussed further in the text).
CIED, Cardiac implantable electronic device; CNS, central nervous system; HM, HeartMate; ICD, implantable cardioverter-defibrillator; LVAD, left ventricular assist device; VAD, ventricular assist device.

Planning Appropriate Perioperative Anticoagulation

Preoperative planning by the anesthesiologist, surgeon, and cardiologist managing the VAD must determine how anticoagulation will be managed for the perioperative period. An INR of approximately two to three times normal is required for both the HM II and the HVAD to prevent thrombus formation and potential thromboembolism. Maintenance is usually with warfarin and aspirin (and antiplatelet agents in some patients). In elective cases in which bleeding risk is substantial, warfarin can be discontinued or the patient bridged to surgery with heparin, but it would be imprudent to automatically “discontinue heparin on call to the operating room (OR)” or advise a patient to stop warfarin without preoperative discussion with the physician managing the VAD. In general, the amount of anticoagulation can be safely reduced for the immediate perioperative period to the lower limits of manufacturers’ recommendations (which may allow for brief periods without any), but most semi-invasive procedures (e.g., endoscopies) and many general surgical procedures can be safely performed with mild levels of anticoagulation (exceptions include ophthalmologic procedures, neurosurgery, and spine surgery). When needed, infusions of fresh-frozen plasma (FFP), cryoprecipitate, or platelets may be guided by point-of-care (POC) tests (e.g., partial thromboplastin time, INR, thromboelastography, rotational thromboelastometry) to achieve goals. The administration of vitamin K or factor concentrates to reverse anticoagulation is not recommended.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here