Article, Cardiology

The emergency management of ventricular assist devices

a b s t r a c t

Background: Heart failure is a common condition in the United States. When medical therapy fails, ventricular device (LVAD) therapy may be required. With increasing use of LVADs, emergency physicians should understand how to manage problems that may arise with these devices.

Objective: The objective of this review is to familiarize physicians with LVAD components and LVAD physiology, and discuss the evaluation and management of LVAD complications.

Discussion: The LVAD contains numerous components, but the most important include the pump, inflow and outflow cannulas, and driveline. Initial assessment of perfusion is vital, as hemodynamic instability may be due to decreased preload, increased afterload, mechanical failure, dysrhythmias, infection, or bleeding. Assessment of hemodynamic status is required, and utilization of Doppler for measurement of mean arterial pressure is warranted. This review provides recommendations for the evaluation and management of the LVAD patient in heart failure, the unstable patient with decreased preload, the unstable patient with increased afterload, thrombosis of the LVAD, mechanical failure, dysrhythmias and cardiac arrest, infections and sepsis, right ventricular failure, aortic insufficiency, and bleeding. Patients with LVAD require consultation with the LVAD coordinator and cardiothoracic surgeon. By understanding these aspects, physicians can provide optimal management for these complicated patients.

Conclusion: With an increasing number of LVADs, emergency physicians should expect to see patients with complications directly or indirectly related to LVADs. This review provides physicians with an extensive review of LVAD physiology and the evaluation and management of potential complications related to the device.

(C) 2016

Case 1

A 69-year-old woman with a history of end-stage heart failure with left ventricular assist device therapy presents with a chief concern of an LVAD alarm persistently beeping. She is alert, her skin is warm, and she appears comfortable while speaking. However, your nurse states he is unable to obtain a blood pressure. What does the emergency physician do next?

? Conflicts of interest: none.

?? No funding was used for this review.

??? This clinical review has not been published; it is not under consideration for publication elsewhere; its publication is approved by all authors and tacitly or explicitly by the responsible authorities where the work was carried out; and, if accepted, it will not be published elsewhere in the same form, in English or in any other language, including electronically without the written consent of the copyright holder.

* Corresponding author at: 506 Dakota St APT 1, San Antonio, TX, 78203. Tel.: +1 719 339 5510.

E-mail addresses: [email protected] (J. Robertson), [email protected] (B. Long), [email protected] (A. Koyfman).

Case 2

A 55-year-old man with a recent LVAD placement presents with signs of poor perfusion. His blood pressure cannot be obtained per his nurse, but he is oxygenating normally and is afebrile. Two large- bore intravenous lines are placed. What should the clinician do next?

Case 3

An 80-year-old man with an LVAD presents to the emergency department (ED) unresponsive. He is placed on the monitor and ventricular fibrillation is seen. He has a mean arterial pressure (MAP) of zero. What should the ED team do next?


Heart failure is a common condition, affecting an estimated 5.1 mil- lion adults 20 years and older in the United States [1]. Many patients will be well managed with medications and/or surgery, but some pa- tients may develop heart failure that is refractory to Standard therapy

(2). Heart transplant is considered the criterion standard for treating these patients; however, given the paucity of donor hearts and the

0735-6757/(C) 2016

fact that some the patients are not candidates for transplant surgery, VADs may allow for improved quality of life and extension of life in select patients [2,3]. A VAD is a mechanical device that maintains blood flow to organs by lessening the load of a failing heart [2]. It was initially created to serve as a temporary bridge to heart recovery and then later as a bridge to heart transplant (BTT). More recently, VADs have been approved by the US Food and Drug Administration as permanent or lifetime support for those with end-stage heart failure. This is known as Destination therapy [2].

The VAD is different than a mechanical heart in that total artificial heart completely replaces the heart with 2 pumps and 4 disk mechani- cal valves [4]. The VAD simply augments the function of the patient’s own native heart [5]. A VAD may include a LVAD, a right ventricular assist device (RVAD), or a biventricular assist device. The majority of patients will have a sole LVAD, as RV dysfunction tends to be temporary. No standalone RVAD is currently approved for long-term support of RV failure, and typically biventricular assist device therapy is required for biventricular failure [6-8].

LVAD use is increasing, and patients will frequently present to the ED with concerns that are directly or indirectly related to their LVADs [9]. Thus, it is important for the emergency physician to understand the physiology of the LVAD to approach the complications that may arise.

An implantable LVAD contains several components, but the main component is a surgically implanted pump located at the apex of the LV [10,11]. The pump receives blood from an internal cannula located in the LV and then propels blood through a tube called the outflow cannula through the Ascending aorta [12]. Pump types may produce centrifugal or axial flow and may be continuous or pulsatile. There is also a percutaneous driveline with wires connecting the pump to an external driver containing a system controller. For power, the system controller is connected to batteries or a power-based unit [13]. The system controller regulates the motor power and speed, provides re- dundant system operation, performs diagnostic monitoring, displays any alarms, and allows data to be downloaded [11,14]. The pulsatile pump LVADs may also have a channel connected to the driveline wires that provides access to the pump. This channel allows the LVAD to be hand pumped should the mechanical pump completely fail. However, continuous-flow LVADs do not have hand pumps [5].

LVAD development has progressed significantly over the years [7]. The first-generation LVAD pumps were pulsatile, positive displacement pumps that had higher complication rates such as infection and me- chanical failure. The first-generation pumps have since been replaced by the second-generation LVADs which have been more successful in terms of better outcomes and fewer complications [10,15]. Second- generation devices consist of axial pumps that use continuous blood flow. The continuous flow is well tolerated and optimally supports organ perfusion [15], but does have some adverse effects. Third- generation LVAD pumps are in existence that use continuous blood flow but contain an axial or centrifugal rotor. Although similar to the second-generation devices, they are smaller and suspected to be more reliable and stable in the long term. None are approved in the United States for use at this time [15]. To date, there are 2 implantable LVADs that have been approved by the Food and Drug Administration. One of these includes the HeartMate II Left Ventricular Assist System that has been approved for BTT and destination therapy. The other is the HeartWare Ventricular Assist System HVAD that is approved only for BTT.

Acute complications“>Discussion

Managing acute complications

Many patients with LVADs will present to the ED with complaints unrelated to their LVADs. However, some patients will present with complications, likely heart failure and/or arrhythmias, directly related

to their devices [16]. Thus, it is imperative that emergency physicians understand how to manage these complications [5].

Heart failure is a common presentation, and one important aspect of evaluating LVAD patients is determining their adequacy of systemic perfusion [5,16]. A large number of conditions can cause hemodynamic instability in LVAD patients including heart failure, thrombosis, infec- tion, bleeding, and device failure. Hypotension in LVAD patients can be categorized into 3 main physiologic categories including decreased pump preload, increased pump afterload, and intrinsic pump malfunc- tion [5].

Decreased preload in the LVAD chamber can be directly related to hypovolemia, septic shock, RV failure, or device failure [5,16,17]. Increased afterload can be due systemic hypertension or obstruction of the outflow cannula from kinking, thrombosis, or infection [5,12,16,18]. Third, mechanical failure and pump malfunction can also cause decreased cardiac output and poor systemic perfusion [5]. Fourth, patients with LVADS are also at high risk for bleeding due to the need for anticoagulation, the development of acquired von Willebrand Syndrome, and the development of arteriovenous malformations [2,19-22]. Patients may bleed at several sites including the gastrointes- tinal (GI) tract, nares, thorax, mediastinum, and central nervous system [23-25]. LVAD patients may also present with dysrhythmias and cardiac arrest just as any non-VAD patient. Management of these conditions in the LVAD patient can be challenging and will be discussed in subsequent sections. Finally, it should be noted that each device will be slightly dif- ferent, but all Continuous-flow devices are similar in basic features [16]. However, the ED physician should attempt to immediately determine which device is implanted and contact the patient’s LVAD coordinator and, if possible, the surgical team to make sure that he or she under- stands any unique device components and in the event of mechanical failure [16].

Troubleshooting the LVAD: general concepts

The emergency physician should have basic knowledge regarding the components of the LVAD as well as alarms on the device. Two types of alarms are present: hazard alarms and advisory alarms. Hazard alarms are concerning, as activation of these alarms may indicate poor circulatory support [13]. Hazard alarms indicate low flow, pump turn- off or disconnection, low voltage requiring immediate battery replace- ment or alternate power source, or complete loss of power [13]. Alarms can be activated because of the patient’s physiology such as low blood volume, hypertension, or thrombosis or be machine related such as battery failure or lead disconnection [5,16].

The power and flow of the LVAD can be affected by changes in pre- load or afterload [13]. These physiologic aspects are intertwined with power and flow. For example, an occlusion of the flow pathway, such as an inflow obstruction causing decreased preload, will decrease power and flow [26]. Of note, any significant increase in power greater than 10 to 12 W should alert the physician to a possible occlusion inside the pump, such as thrombus [13]. The pulsatility index is related to the amount of LVAD assistance to intrinsic contractility and should be assessed.

Ultimately, the emergency provider is responsible for the evaluation and treatment of any physiologic condition that could possibly be causing LVAD alarm activation. Importantly, physicians should always clinically evaluate each patient’s hemodynamic status and clinical examination even if no alarms are activated [12].

Assessment of the LVAD patient

Assessing blood pressure in a patient with a continuous-flow LVAD can be difficult, which makes evaluating the hypotensive LVAD patient challenging [16]. The major difference between non-LVAD and LVAD patients is that the LVAD pumps constantly through the cardiac cycle, and thus aortic flow will be present during diastole when pulsatile

flow is normally absent [25]. As the pump speed of the LVAD increases, diastolic pressure increases but systolic pressure remains stable, reducing Pulse pressure.

Because of these changes in pulse pressure, it is common to not palpate a pulse in a patient with a normally functioning LVAD. Blood pressure in the LVAD patient should be measured with a Doppler and a sphygmomanometer [16,25]. When listening with the blood pressure cuff, the start of the Korotkoff sound is a pressure that is in a range of systolic and diastolic blood pressures [25]. To obtain a MAP, the brachial or Radial artery should be located using the Doppler. The sphygmoma- nometer cuff should be inflated until flow is no longer heard. The cuff should then slowly be released, and the MAP is determined at the pres- sure where arterial flow is once again heard [16].

Pump speed should not be adjusted to achieve any particular arterial pressure [25], but blood pressure should be managed with appropriate medications [11,25]. Aggressively treating hypertension is important to reduce the risk of stroke, end-organ damage, and LVAD dysfunction [11,25]. The goal is to maintain a MAP in the range of 70 to 80 mm Hg and not exceed 90 mm Hg [25,27]. Finally, because of the narrower pulse pressure, pulse oximetry measurements may not be accurate. Therefore, more invasive monitoring such as obtaining arterial blood gas measurements may be necessary in the very ill LVAD patient [16,25]. Figure takes into account these measures in algorithm form.

The hemodynamically stable LVAD patient in heart failure

In the hemodynamically stable LVAD patient with heart failure, several etiologies should be considered including LVAD thrombosis, RV failure, bleeding, anemia, or valvular disease [28]. With the exception of LVAD thrombosis as discussed below, patients should be evaluated and managed as any other non-LVAD patient [28]. Patients with valvular disease should receive an echocardiogram and be diuresed as needed, and the LVAD team and surgical team should be consulted. In a stable patient with RV failure, secondary causes should be considered along with diuresis, LVAD speed adjustment, and possible inotropic therapy [28]. Finally, any stable patient with bleeding or anemia should be evaluated for the cause of bleeding, anticoagulation should be suspended, and transfusions should be administered as needed [28].

The hemodynamically unstable LVAD patient: decreased preload

As previously mentioned, there are 3 main reasons why an LVAD pa- tient may be hypotensive, including decreased preload, increased afterload, or intrinsic pump malfunction [5]. LVADs are preload depen- dent, and for proper LVAD function, patients require adequate preload

Figure. Evaluation of the LVAD patient.

[16]. Decreased preload can be due to hypovolemia from bleeding, dehydration, septic shock, or any other cause of hypovolemia in a non-LVAD patient [5,17,25]. Decreased preload can also be due to RV failure or actual device failure such as inflow cannula or mechanical obstruction [5,17,29].

Decreased preload is sometimes referred to as a suction event, and these patients typically present with hypotension [16]. One may see a decreased pulsatility index in this situation. Dehydration is a common cause, as many LVAD patients are on diuretic therapy. fluid status should be assessed and echocardiography used to evaluate for a suction event [17]. In the absence of RV failure, patients should be given an intravenous fluid bolus and reevaluated [16,17,25,30]. Note that a suction event will lower the LVAD speed, which will not increase back to baseline until the suction event has been treated [13]. Occasionally, decreasing the pump speed may help decrease the incidence of suction events [17].

Other causes of decreased preload that would alarm the low-flow in- dicator include inflow cannula obstruction and mechanical obstruction [17]. An outflow cannula obstruction, while causing increased afterload, will also cause the low-flow indicator to alarm. LVAD patients are at high risk for thrombosis, and any patient with a continuous-flow LVAD will be anticoagulated [25].

The hemodynamically unstable LVAD patient: increased afterload

Increased afterload occurs when the LVAD chamber cannot be completely emptied [5]. Causes of increased afterload may include hypertension; outflow cannula obstruction, either via thrombosis or mechanical kinks; or infection [5,18]. Managing blood pressure is essential in the LVAD patient, as keeping the MAP at approximately 70-80 mm Hg is crucial to maintaining adequate LVAD blood flow [5,16,25]. Elevated MAP greater than 90 mm Hg may reduce the LVAD’s power, flow, cardiac output, and perfusion. In addition, patients with a MAP N 80 mm Hg are at an increased risk of cerebral vascular ischemia or hemorrhage [31]. Patients should be promptly treated with afterload-reducing, Antihypertensive medications to maintain an appro- priate MAP [16,25].

Thrombosis in the LVAD

Patients with continuous-flow LVADs are at risk for thrombosis, and a thrombus can originate in the actual pump or travel from the left heart or right heart through a septal defect [32]. The thrombus can also im- plant in the pump components including the inflow and outflow cannu- las [17,32,33]. Thrombosis places these patients at risk for LVAD dysfunction, hemolysis, peripheral emboli, stroke, hemodynamic insta- bility, and even death [17,33-35]. Thus, patients with continuous-flow LVADs are placed on anticoagulant and antiplatelet regimens [15]. How- ever, even with proper anticoagulation, patients still are at risk for clots, and thus, thrombosis should be considered in LVAD patients with concerning signs and symptoms.

Signs and symptoms of thrombosis can vary from asymptomatic to heart failure or even cardiac arrest [32]. Pump thrombosis should be suspected if there is an increase in pump power, decrease in pump flow, or a change in the pulsatility index or if the LVAD is alarming in an otherwise asymptomatic patient [16,32,35]. A patient with LVAD thrombosis may also have signs of hemolysis on both clinical examination and laboratory work including dark urine, scleral icterus, fatigue, an elevated serum lactate dehydrogenase of 2.5 times the Upper limit of normal, or decreased hemoglobin on complete blood count [32,35].

If pump thrombosis is suspected, laboratory work including a serum lactate dehydrogenase, CBC, haptoglobin, and Coagulation studies should be drawn [32,35]. Imaging is warranted, including a chest radio- graph to evaluate for any changes in pump position or pulmonary edema. A transesophageal echocardiogram with a ramp study is the

imaging test of choice for diagnosing a thrombus [32,35]. The diagnosis is made when the LV is seen to be unloading improperly on echo with no other reason for the abnormality [32]. Cardiac computed tomogra- phy (CT) angiography has been shown to have the ability to diagnose thrombus and thus may be considered in stable patients [36,37]. Mag- netic resonance imaging should never be conducted, as the LVAD is not compatible with Magnetic fields [12].

The ideal Treatment regimen for LVAD pump thrombosis has not

been established. However, some regimens have been suggested in- cluding attempting medical therapy first in stable patients and then sur- gical therapy if medications are unsuccessful or the patient is unstable [32,38-40]. Essentially, stable patients with hemolysis should be started on Intravenous heparin and admitted to the hospital for further testing [32]. All long-acting anticoagulants should be discontinued, and pa- tients should be given intravenous hydration if no signs of fulminant heart failure are present [35]. Thrombolysis can be considered, as stud- ies have shown lysis to be efficacious in pump thrombosis [38-40]. De- pending on the level of heart failure symptoms, some patients may also require inotropic and diuretic therapy [32]. Emergent surgical pump ex- change should occur if apparent pump thrombosis with alarms is pres- ent, the pump stops, or the patient is hemodynamically unstable and unresponsive to a battery change [32]. Some patients may require extra- corporeal membrane oxygenation before pump exchange if they are in cardiogenic shock with signs of multiple-organ dysfunction [35]. Thrombosis of the inflow or outflow grafts is treated similarly with medical therapy, although endovascular stenting has been shown to be a possible treatment [40-42].

Mechanical failure

Hemodynamic instability and fulminant heart failure can be due to mechanical failure, and this should also be considered in the LVAD pa- tient with concerning symptoms [40]. In the Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure trial, the second most common cause of death in LVAD patients was de- vice failure [43]. As mentioned above, the concerning hazard alarms on the HeartMate II consist of the low-flow alarm as discussed above, dis- connected percutaneous leads, low voltage, and loss of power.

With a low-flow alarm activated, physicians should manage blood pressure as per above, evaluate for any disconnected percutaneous leads, and evaluate for cannula kinking or obstruction. Cannula kinking or obstruction can be assessed with echocardiogram and, if patients are hemodynamically stable, CT angiography [17,32]. Thrombosis of the cannulas can be evaluated for and managed as per above. Importantly, if the low-flow alarm is activated, physicians should always auscultate over the pump to ascertain that it is actually running [13]. If a lead be- comes disconnected from the controller, the pump will stop, and usual- ly, a lead can be reconnected [44]. If the leads are in place and a continuous-flow LVAD pump has stopped, then the power and power leads should be immediately assessed. Signs of pump failure include in- ability to auscultate the motor, undetectable blood pressure via Doppler, or the absence of the power light on the controller [16].

Besides low flow and lead disconnection, hazard alarms also include complete power loss or low voltage requiring battery replacement or a substitute power source [13]. If one of these alarms is activated, the phy- sician should resuscitate the patient but attempt to immediately regain power by changing the battery, reattaching a power cable, or finding an- other power source [13]. On the HeartMate II, a continuous alarm will sound if the pump stops from complete loss of power [13,17]. If the pump is stopped but not caused by a loss of power, an alarm will sound and a red heart will light up on the controller [13,17]. If all the leads are connected, then the Test Select or Alarm Reset button should be activated to reset the pump [17]. If a system controller power lead is not mated to the batteries or the power-based unit cable, a cable dis- connect advisory will be activated, indicated by a flashing green power

symbol and an alarm. The disconnected lead can be reattached, and the alarm should stop [13].

No matter what the cause of pump malfunction is, if the pump stops working, even for a Short period of time, noncirculating blood in the LVAD can place the patient at risk for stroke or thromboembolism once the device is restarted [13]. Thus, the pump should be restarted as soon as possible [13]. If the pump has failed and cannot be restarted, the patient’s surgeon, LVAD engineer, and on-call nurse should be contacted immediately [16]. The patient may require emergent pump exchange at this point. First-generation pulsatile flow pumps do have a hand pumping mechanism that can be temporarily used until pump exchange can occur, but this is not an option for second-generation devices [5].

If a patient arrives to the ED with hemodynamic instability without an alarm activated, then medical management should be attempted, and the physician should attempt to determine the cause of hemody- namic instability [16]. As previously mentioned, perfusion should be assessed and supported. If a patient appears to have hypoperfusion with a MAP of b 40 mm Hg, loss of consciousness, or cyanosis, standard resuscitation should ensue [16]. Symptoms, laboratory values, valve function, and device parameters should all be taken into consideration in the hemodynamically unstable LVAD patient [12]. If there are signs of inadequate perfusion, preload expansion with fluids should be con- sidered along with afterload reduction with nitrates or other antihyper- tensive medications [5,25]. Any nonperfusing cardiac rhythms should be managed via Advanced Cardiac Life Support protocols [16]. Causes of the hemodynamic instability may include hypovolemia, bleeding, thrombosis, arrhythmia, and infection [5]. Patients may require more invasive pump evaluation if medical therapies fail, and the surgical team should be notified immediately [16].

Dysrhythmias and cardiac arrest in LVAD patients

Patients with LVADs are more prone to ventricular dysrhythmias [45]. Although the actual etiology is unknown, it is suspected that electrolyte abnormalities, ischemia, RV failure, and changes in cardiac modeling and electrophysiology occur more frequently after LVAD placement [16,17,45-47]. The incidence of dysrhythmias in LVAD patients is highest in the first 30 days after device insertion [47,48]. ventricular dysrhythmias are also seen in patients with no prior dysrhythmias before LVAD implantation [46,49].

Symptoms of dysrhythmias in LVAD patients may vary [47]. Many patients can tolerate dysrhythmias with minimal symptoms because the LVAD helps maintain cardiac output despite a rapid heart rate and potential arteriovenous asynchrony [16,47,50-52]. However, some LVAD patients with dysrhythmias may present with hemodynamic instability, heart failure, and even cardiac arrest [47]. Of note, the LVAD may function regardless of the dysrhythmia as long as blood is flowing into the pump. However, eventually flow is compromised. Dysrhythmias can lead to RV dysfunction, thrombus, suction events, and poor perfusion [17,30,46,53].

Evaluating for the underlying cause (s) of the dysrhythmia is impor- tant, although it may be difficult to determine in the ED. A dysrhythmia in the LVAD patient can be primary or secondary. A primary dysrhyth- mia is due to a patient’s intrinsic Cardiac electrophysiology, whereas a secondary dysrhythmia is due to the LVAD itself [16]. Regardless of the cause of the dysrhythmia, any unstable patient should undergo direct current cardioversion as per Advanced Cardiac Life Support protocols [11]. If the patient is stable, there is no need for acute cardioversion [11]. Hypovolemia and inadequate venous return are the most common causes of secondary dysrhythmias in LVAD patients, and thus, management of the stable LVAD patient with an abnormal rhythm should include a fluid bolus and emergent echocardiogram [16,54]. Electrolytes should be evaluated, and any abnormalities should be treated. Coronary angiography should be considered in patients who have concerning symptoms for acute coronary syndrome [17].

Antiarrhythmic medications may be considered for primary dysrhyth- mias, but there is limited evidence on the type of medication that should be administered [16]. Patients in complete cardiac arrest may be defibrillated. The LVAD manufacturers do not recommend chest com- pressions, as this can dislodge the device, but if a patient is in fulminant cardiac arrest, then compressions may be necessary until the LVAD team can be consulted [11,13]. A recent cohort study by Shinar et al

[55] noted chest compressions to be safe and effective, but further re- search is required.

Most patients with LVADs also have implantable cardioverter/defi- brillators. If a patient arrives with a complaint of an implantable cardioverter/defibrillator shock, physicians should determine whether the shock was appropriate or inappropriate. If inappropriate, then other pathologies should be ruled out such as supraventricular tachy- cardia, lead problem, oversensing, or electromagnetic interference [47]. If the patient sustained an appropriate shock, then the LVAD needs to be closely evaluated, the patient’s LVAD coordinator should be contacted, and ischemia and RV dysfunction should be ruled out [47].

Infections and sepsis

A major complication and a possible cause of hemodynamic instabil- ity in the LVAD patient are infection and septic shock [5]. Infection is the most common complication of long-term LVAD use [5,56]. In addition to presence of a foreign body, patients are immunosuppressed, which con- tributes to infections related to the device itself and infections from other sources [5,14,57,58]. In 2010, the International Society for Heart and lung transplantation developed criteria that define infections in patients with VADs. The definitions are organized into 3 main categories including (1) VAD-specific infections, (2) VAD-related infections, and

(3) non-VAD infections [14]. These are not standard international defi- nitions but can be helpful when evaluating VAD patients who have concerning signs and symptoms of infection.

VAD-specific infections include infections of the hardware or the body surfaces that contain them including the pump, the cannula, any anastomosis, the pocket, and/or the percutaneous driveline or tunnel [14]. Multiple sites are often infected, although the percutaneous drive- line is the most commonly infected [14,56,58]. Driveline infection (DLI) is the most common VAD-specific infection in VAD patients, followed by Bloodstream infection [56,59]. DLIs can extend into the pump pocket and also cause BSI, which makes diagnosing DLI imperative [59]. Driveline trauma is the most common reason why patients devel- op infection in this area [56].

VAD-related infections are those infections that can occur in patients without VADs; however, there are unique characteristics of VAD pa- tients that make these infections more common in this population. These infections include endocarditis, BSIs, central venous catheter in- fections, and mediastinitis [14]. Finally, a non-VAD infection is one that can occur in any patient including pneumonia, urinary tract infec- tion, cholecystitis, and Clostridium Difficile infection [14]. Non-VAD- related infections are managed as any other non-VAD patient with a similar infection.

The clinical presentation of VAD-specific infections varies. A 2013 retrospective review of 247 LVAD patients demonstrated that only half of patients present with systemic manifestations of infection such as fever, leukocytosis, or systemic inflammatory response criteria [56]. However, some patients may have pain, fever, drainage, and warmth of the driveline exit site and/or leukocytosis on a CBC [60,61]. Similarly, manifestations of a VAD-specific pocket infection may include fever, malaise, weight loss, new local erythema over the pocket site, Local pain and/or tenderness, induration, and swelling [14]. Usually, elevated pump flows due to vasodilation are seen [5].

DLIs typically occur much later after device insertion, with approxi-

mately 80% occurring at least 30 days after implantation. The average time to DLI is 6 months [60]. DLIs can lead to pocket infection, as well as BSI and endocarditis [60,61]. Patients who have BSIs or endocarditis

will usually present similarly to those who do not have LVADs, including fever, sepsis, and the external signs of endocarditis such as Janeway le- sions and Osler nodes [61]. However, it should be noted that patients with endocarditis may present with nonspecific signs and symptoms such as septic shock, weight loss, and malaise [60].

There is no standardized approach toward diagnosing and managing VAD-specific infections, BSI, and endocarditis, but some studies have made general recommendations toward their diagnosis and treatment [14,56,59,61]. If any VAD-specific or VAD-related infection is suspected, 3 sets of blood cultures should be obtained, even if patients lack under- lying signs and symptoms of Systemic Infection. A pump or cannula in- fection is probable when positive blood cultures result in the absence of any other source of infection [60]. However, CT and/or ultrasonographic imaging may be helpful if an underlying abscess is suspected [56,60]. Nuclear imaging may also be helpful, but this remains to be validated [14]. Also, it is recommended that all patients receive basic laboratory tests including a white blood cell count and serial C-reactive proteins or erythrocyte sedimentation rates. All patients should receive a chest radiograph and aspirate for Gram stain, KOH, and bacterial and fungal cultures of the driveline at the exit site [14]. With specialist assistance such as interventional radiology, deeper fluid collections may be obtain- ed by imaging guidance [14]. Finally, other causes of any sepsis should be ruled out in all LVAD patients, including the non-VAD-related sources such as the urinary tract and the lungs [14].

The diagnosis of a BSI requires that the same organism is isolated from at least 1 blood culture and 1 culture from the catheter tip [60]. The diagnosis of endocarditis does require transesophageal echocardi- ography; however, visualization of vegetations may be limited by the metal surface of the VAD. Thus, treatment should be considered if pa- tients have persistent signs and symptoms of endocarditis, regardless of echocardiographic findings [59,60].

The majority of patients with VAD-specific and VAD-related infec-

tions have wound and/or blood cultures positive for coagulase- negative Staphylococcus, S aureus, Pseudomonas species, and nosocomial gram-negative bacilli [56,58,59,61,62]. Candida species and Enterococcus are also commonly found in culture results [60,63]. In any suspected in- fection, early broad-spectrum antibiotic therapy that covers the above organisms should be administered [59,61,62]. The optimum antibiotic regimen is not well defined, but cefazolin, vancomycin, and linezolid cover the above pathogens [61]. Not all patients will have an invasive in- fection, and local infections may only require oral antibiotics [56]. Final- ly, not all patients will require surgery; however, cardiothoracic surgery consultation should be obtained to evaluate for further aspiration and culture of any infected fluid, possible removal of the VAD, and/or surgi- cal therapy to drain infected fluid [14,56,61].

RV failure

RV failure is a significant cause of morbidity and death after LVAD implantation, and it may be extrinsic or intrinsic [25]. Extrinsic causes are due to the LVAD itself, whereas intrinsic causes may be caused by pulmo- nary embolism, pulmonary hypertension, dysrhythmias, tricuspid regurgi- tation, or intrinsic RV dysfunction [28]. These pathologies should be considered in the unstable and/or heart failure LVAD patient with concerning clinical signs and symptoms [28]. RV failure is the most com- mon cause of low pump flows and is usually managed by inotropic therapy and RV afterload reduction. Extrinsic RV failure can also be due to excessive pump speed or cardiac tamponade. Cardiac tamponade should be consid- ered if there are concerning clinical signs such as hypotension and jugular venous distension. Bedside echocardiography should be obtained. Also, all anticoagulation should be discontinued and surgery consulted [28].


Bleeding, especially in the GI tract and at the site of LVAD implanta- tion, is common. In fact, it is the second most common complication

after infection in patients with assist devices [19,64]. In addition, patients with continuous-flow LVADs appear to have higher rates of GI bleeding events than patients with pulsatile LVADs [65]. Although the majority of bleeding is not severe, significant anemia can develop, worsening heart failure [28]. Other areas of bleeding include the nares, thorax, mediastinum, and brain [23-25]. Independent predictors of GI bleeding include a previous history of GI bleeding, elevated interna- tional normalized ratio, and thrombocytopenia [64]. In patients with continuous-flow LVADs, an international normalized ratio of 1.7 to 2.3 is recommended, but a higher target may be needed if patients have a history of Thromboembolic events [11].

Not only does anticoagulation place LVAD patients at risk for bleed- ing, but the assist device itself predisposes patients to bleeding due to formation of arteriovenous malformations along the GI mucosa and the breakdown and deformation of von Willebrand multimers or factors [15,21,22,24]. It is thought the LVAD results in mechanical destruction of the von Willebrand multimers due to the shearing stress through a ste- notic aortic valve. Although the sheer number of von Willebrand multimers or factors may be normal, the destruction of the multimers leads to their poor capability to bind to collagen and platelets [20,21]. Thus, patients with LVADs can develop acquired von Willebrand disease simply because of the presence of the LVAD [19-21].

If a patient sustains a bleeding event, hemodynamic support should be provided [28]. Anticoagulation and antiplatelet medications should also be held, and any coagulopathies should be corrected if indicated [21]. specialty services, such as GI, should be consulted to identify and treat the source of bleeding. Further management of anticoagulation should be later managed by inpatient teams [21].


The LVAD is a potentially life-saving device in patients with end- stage heart failure. Although LVAD patients may present to the ED with complaints unrelated to their devices, there are many complica- tions that are related to the LVAD, and emergency physicians should be familiar with these complications. Assessing these patients includes evaluating perfusion and hemodynamic stability and looking for poten- tial causes, including decreased preload from dehydration, bleeding or thrombosis, or increased afterload from outflow obstruction or systemic hypertension. A MAP goal of 70 mm Hg is advised. Patients with con- cerns for suction events should be given intravenous fluids, and throm- bosis should be considered in those patients with signs of worsening heart failure, cardiac arrest, increases in pump power, or an alarming LVAD in an otherwise asymptomatic patient. Anticoagulation and even thrombolysis should be administered in patients who have con- cerns for thrombosis. Dysrhythmias may be treated just as any patient; however, at this time, chest compressions should be avoided until fur- ther research is conducted. LVAD patients with fever, purulence from the pocket site, or any other signs of infection should be evaluated and treated with antibiotic therapy. Bleeding is a risk of anticoagulation, and management is similar to the non-LVAD patient. Importantly, any LVAD patient should have his or her team contacted no matter the concern, as these are complicated patients who require a team-based approach to evaluation and both acute and long-term management.

Case resolutions

Case 1

The physician instructs the nurse to measure the patient’s blood pressure with a Doppler and a sphygmomanometer, resulting in a MAP of 70 mm Hg. The patient’s pump is auscultated and seems to be working normally. The hazard alarm notes that the battery is extremely low. The patient’s LVAD coordinator is contacted and a replacement battery is acquired. The alarm stops and the patient is discharged in good condition.

Case 2

The physician instructs the nurse to measure the patient’s blood pressure with a Doppler and a sphygmomanometer, and a MAP of 75 mm Hg is obtained. The LVAD pump seems to be working normally via auscultation, although the low-flow alarm on the LVAD is beeping. A liter of normal saline is given to the patient. In the meantime, an elec- trocardiogram is obtained, and no ST-elevation myocardial infarction is seen. The patient shows no signs of infection or bleeding. An echocar- diogram demonstrates a collapsible inferior vena cava and a small right ventricle. The patient’s vital signs improve with another liter of normal saline, and the low-flow alarm stops. Laboratory work reveals signs of mild dehydration. A CT angiography is ordered to rule out thrombosis, which is negative. The patient is admitted for observation, and the patient’s LVAD team follows the patient while he is hospitalized.

Case 3

The patient’s MAP is zero. The pump is not working via auscultation. Chest compressions, per the patient’s LVAD manufacturer, are not advised, but the patient has no perfusion, so compressions are initiated while defibrillation pads are placed. The patient is manually defibrillated and a perfusing rhythm is obtained. He is intubated for airway protection and supportive care is initiated. The patient’s LVAD team evaluates the patient for pump replacement while supportive care is continued. Fortunately, the chest compressions did not dislodge the patient’s pump, but the LVAD will need to be replaced by the team because of complete mechanical failure.


  1. Go AS, Mozaffarian D, Roger VL, et al. Heart disease and stroke statistics–2014 up- date. Circulation 2013;129(3):e28-292.
  2. Givertz MM. Ventricular assist devices: important information for patients and fam- ilies. Circulation 2011;124(12):e305-11.
  3. Daneshmand MA, Rajagopal K, Lima B, et al. Left ventricular assist device destination therapy versus extended criteria cardiac transplant. Ann Thorac Surg 2010;89(4): 1205-10.
  4. Mankad AK, Tang DG, Clark WB, et al. Persistent anemia after implantation of the total artificial heart. J Card Fail 2012;18(6):433-8.
  5. Sayer G, Naka Y, Jorde UP. Ventricular assist device therapy. Cardiovasc Ther 2009; 27:140-50.
  6. Fukamachi K, Shiose A, Massiello AL, et al. Implantable continuous-flow right ven- tricular assist device: lessons learned in the development of a Cleveland Clinic de- vice. Ann Thorac Surg 2012;93(5):1746-52.
  7. Kirklin JK, Naftel DC, Pagani FD, et al. Sixth INTERMACS annual report: a 10,000-pa- tient database. J Heart Lung Transplant 2014;33(6):555-64.
  8. Fitzpatrick JR, Frederick JR, Hiesinger W, et al. Early planned institution of biventricular mechanical circulatory support results in improved outcomes com- pared with delayed conversion of a left ventricular assist device to a biventricular as- sist device. J Thorac Cardiovasc Surg 2009;137(4):971-7.
  9. Devine A, Knapp B, Harbin E, et al. Characteristics and frequency of emergency de- partment visits of patients with continuous flow left ventricular assist devices. Ann Emerg Med 2012;60(45):S142.
  10. Mallidi HR, Anand J, Cohn WE. State of the art mechanical circulatory support. Tex Heart Inst J 2014;41(2):115-20.
  11. Wilson SR, Givertz MM, Steward GC, et al. Ventricular assist devices: the challenges of outpatient management. J Am Coll Cardiol 2009;54(18):1647-59.
  12. Kapur NK, Jmean MF. Management of continuous flow left ventricular assist device patients in the cardiac catheterization laboratory. Cath Lab Digest 2014;22(3).
  13. HeartMate II(R) LVAS operating manual. [Available at] dockets/ac/07/briefing/2007-4333b2-18-%209_2%20HM%20II%20Operating% 20Manual.pdf.
  14. Hannan MM, Husain S, Mattner F, et al. Working formulation for the standardization of definitions of infections in patients using ventricular assist devices. J Heart Lung Transplant 2011;30(4):375-84.
  15. Spiopoulos K, Giamouzis G, Karvannis G, et al. Current status of mechanical circula- tory support: a systematic review. Cardiol Res Pract 2012.
  16. Greenwood JC, Herr DL. Mechanical circulatory support. Emerg Med Clin North Am 2014;32(4):851-69.
  17. Klein T, Jacob MS. Management of implantable assisted circulation devices: emer- gency issues. Cardiol Clin 2012;30(4):673-82.
  18. Weitzel N, Puskas F, Cleveland J, et al. Left ventricular assist device outflow cannula obstruction by the rare environmental fungus Myceliophthora thermophila. Anesth Analg 2009;108(1):73-5.
  19. Meyer AL, Malehsa D, Budde U, et al. Acquired von Willebrand syndrome in patients with a centrifugal or axial continuous flow left ventricular assist device. J Am Coll Cardiol 2014;2:141-5.
  20. Crow S, Chen D, Milano C, et al. Acquired von Willebrand syndrome in continuous-

    flow ventricular assist device recipients. Ann Thorac Surg 2010;90:1263-9.

    Suarez J, Patel CB, Felker M, et al. Mechanisms of bleeding and approach to patients with axial-flow left ventricular assist devices. Circ Heart Fail 2011;4:779-84.

  21. Eckman PM, Ranjit J. Bleeding and thrombosis in patients with continuous-flow ven-

    tricular assist devices. Circulation 2012;125:3038-47.

    Pagani FD, Miller LW, Russell SD, et al. Extended mechanical circulatory support with a continuous-flow rotary left ventricular assist device. J Am Coll Cardiol 2009;54:312-21.

  22. Geisen U, Heilmann C, Beyersdorf F, et al. Non-surgical bleeding in patients with ventricular assist devices could be explained by acquired von Willebrand diseases. Eur J Cardiothorac Surg 2008;33:679-84.
  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(4):S1-S39.
  24. Topilsky Y, Maltais S, Oh JK, et al.. Focused review on transthoracic echocardiograph- ic assessment of patients with continuous axial left ventricular assist devices. Cardiol Res Pract 2011.
  25. Feldman D, Pamboukian SV, Teurteberg JJ, et al. The 2013 International Society for Heart and Lung Transplantation guidelines for mechanical circulatory support: exec- utive summary. J Heart Lung Transplant 2013;32(2):157-87.
  26. Burke MA, Givertz MM. Assessment and management of heart failure after left ven- tricular assist device implantation. Circulation 2014;129(10):1161-6.
  27. Argiriou M, Kolokotron SM, Sakellaridis T, et al. Right heart failure post left ventric- ular assist device implantation. J Thorac Dis 2014;6(1):S52-9.
  28. O’Shea G. Ventricular assist devices: what intensive care unit nurses need to know about postoperative management. AACN Adv Crit Care 2012;23(1):69-83.
  29. Salamonsen RF, Mason DG, Ayre PJ, et al. Response of rotary blood pumps to changes in preload and afterload at a fixed speed are unphysiological when compared with the natural heart. Artif Organs 2011;35(3):E47-53.
  30. Goldstein DJ, Ranjit J, Salerno C, et al. Algorithm for the diagnosis and management of suspected pump thrombus. J Heart Lung Transplant 2013;32:667-70.
  31. Uriel N, Han J, Morrison KA, et al. Device thrombosis in HeartMate II continuous- flow left ventricular assist devices: a multifactorial phenomenon. J Heart Lung Trans- plant 2014;33:51-9.
  32. Stulak JM, Cowger J, Haft JW, et al. Device exchange after primary left ventricular as- sist device implantation: indications and outcomes. Ann Thorac Surg 2013;95: 1262-8.
  33. Tchantchaleishvili V, Sagebin F, Ross RE, et al. Evaluation and treatment of pump thrombosis and hemolysis. Ann Cardiothorac Surg 2014;3(5):490-5.
  34. Raman SV, Sahu A, Merchant AZ, et al. Noninvasive assessment of left ventricular as- sist devices with cardiovascular computed tomography and impact on management. J Heart Lung Transplant 2010;29(1):79.
  35. Mishkin JD, Enriquez JR, Meyer DM, et al. Utilization of cardiac computed tomogra- phy angiography for the diagnosis of left ventricular assist device thrombosis. Circ Heart Fail 2012;5:e27-9.
  36. Lenneman AJ, Combs P, Rhode S, et al. Management and outcomes of ventricular as- sist device patients with suspected pump thrombosis. J Heart Lung Transplant 2013; 32(4):S186-7.
  37. Webber BT, Panos AL. Rodriguez-Blanco. Intravenous thrombolytic therapy for pa- tients with ventricular assist device thrombosis: an attempt to avoid reoperation. Ann Card Anaesth 2016;19(1):192-6.
  38. Birks EJ. Left ventricular assist devices. Heart 2010;96:63-71.
  39. Abraham J, Remick JD, Caulfield T, et al. Left ventricular assist device outflow cannula obstruction treated with percutaneous endovascular stenting. Circ Heart Fail 2015; 8:229-30.
  40. Kamouh A, John R, Eckman P. Successful treatment of early thrombosis of HeartWare left ventricular device with intraventricular thrombolytics. Ann Thorac Surg 2012; 94(1):281-3.
  41. Rose E, Gelijns AC, Moskowitz AJ, et al. Long term use of a left ventricular assist de- vice for end stage heart failure. N Engl J Med 2001;345(20):1435-43.
  42. Cubillo EI, Weis RA, Ramakrishna H. Emergent reconnection of a transected left ven- tricular assist device driveline. J Emerg Med 2014;47(5):546-51.
  43. Cesario DA, Saxon LA, Cao MK, et al. Ventricular tachycardia in the era of ventricular assist devices. J Cardiovasc Electrophysiol 2011;22:359-63.
  44. Ziv O, Dizon J, Thosani A, et al. Effects of left ventricular assist device therapy on ven- tricular arrhythmias. J Am Coll Cardiol 2005;45(9):1428-34.
  45. Nakahara S, Chien C, Gelow J, et al. Ventricular arrhythmias after left ventricular as- sist device. Circ Arrhythm Electrophysiol 2013;6:648-54.
  46. Miller L, Pagani FD, Russell SD, et al. Use of a continuous flow device in patients awaiting Heart transplantation. N Engl J Med 2007;357(9):885-96.
  47. Harding JD, Piacentino V, Gaughan JP, et al. Electrophysiological alterations after me- chanical circulatory support in patients with advanced cardiac failure. Circulation 2001;104:1241-7.
  48. Busch MC, Haap M, Kristen A, et al. Asymptomatic sustained ventricular fibrillation in a patient with left ventricular assist device. Ann Emerg Med 2011;57:25-8.
  49. Fasseas P, Kutalek SP, Samuels FL, et al. Ventricular assist device support for manage- ment of sustained ventricular arrhythmias. Tex Heart Inst J 2001;29:33-6.
  50. Patel P, Williams JG, Brice JH. Sustained ventricular fibrillation in an alert patient: preserved hemodynamics with a left ventricular assist device. Prehosp Emerg Care 2011;15:533-6.
  51. Oz M, Ros EA, Slater J, et al. malignant ventricular arrhythmias are well tolerated in patients receiving long-term left ventricular assist devices. J Am Coll Cardiol 1994; 24(7):1688-91.
  52. Vollkron M, Voitl P, Ta J, et al. Suction events during left ventricular support and ven- tricular arrhythmias. J Heart Lung Transplant 2007;26(8):819-25.
  53. Shinar Z, Bellezzo J, Sahovich M, et al. Chest compressions may be safe in arresting patients with left ventricular assist devices (LVADS). Resuscitation 2014;85:702-4.
  54. Nienaber JJC, Kusne S, Rias T, et al. Clinical manifestations and management of left ventricular assist device-associated infections. Clin Infect Dis 2013;57(10):1438-48.
  55. Goldstein DJ, Naftel D, Holam W, et al. Continuous-flow devices and percutaneous site infections: clinical outcomes. J Heart Lung Transplant 2012;31:1151-7.
  56. Gordon RJ, Weinberg AD, Pagani FD, et al. A prospective, multicenter study of ven- tricular assist device infections. Circulation 2013;127(6):691-702.
  57. Maniar S, Kondareddy S, Topkara VK. Left ventricular assist device-related infec- tions: past, present and future. Expert Rev Med Devices 2011;8(5):627-34.
  58. Nienaber J, Wilhelm MP, Sohail MR. Current concepts in the diagnosis and manage- ment of left ventricular assist device infections. Expert Rev Anti Infect Ther 2013; 11(2):201-10.
  59. Hieda M, Sata M, Nakatani T. The importance of the management of infectious com- plications for patients with left ventricular assist device. Healthcare 2015;3:750-6.
  60. Topkara VK, Kondareddy S, Malik F, et al. infectious complications in patients with left ventricular assist device: etiology and outcomes in the continuous-flow era. Ann Thorac Surg 2010;90:1270-7.
  61. Schaffer JM, Allen JG, Weiss ES, et al. Infectious complications after pulsatile-flow and continuous-flow left ventricular assist device implantation. J Heart Lung Trans- plant 2011;30:164-74.
  62. Kirklin JK, Naftel DC, Kormos RL, et al. Second intermacs annual report: more than 1000 primary LVAD implants. J Heart Lung Transplant 2010;29(1):1-10.
  63. Crow S, Ranjit J, Boyle A, et al. Gastrointestinal bleeding rates in recipients of nonpulsatile and pulsatile left ventricular assist devices. J Thorac Cardiovasc Surg 2009;137:208-15.

Leave a Reply

Your email address will not be published. Required fields are marked *