J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7

ª 2 0 1 7 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N

P U B L I S H E D B Y E L S E V I E R

I S S N 2 2 1 3 - 1 7 7 9 / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j c h f . 2 0 1 7 . 0 6 . 0 1 2

MINI-FOCUS ISSUE: SPECIAL POPULATIONS

STATE-OF-THE-ART REVIEW

Medical Management of PatientsWitha Left Ventricular Assist Device forthe Non-Left Ventricular AssistDevice Specialist

Adam D. DeVore, MD, MHS,a,b Priyesh A. Patel, MD,a Chetan B. Patel, MDa,b

JACC: HEART FAILURE CME/MOC

This article has been selected as the month’s JACC: Heart Failure

CME/MOCactivity, available onlineathttp://www.acc.org/jacc-journals-cme

by selecting the JACC Journals CME/MOC tab.

Accreditation and Designation Statement

The American College of Cardiology Foundation (ACCF) is accredited by

the Accreditation Council for Continuing Medical Education (ACCME) to

provide continuing medical education for physicians.

The ACCF designates this Journal-based CME/MOC activity for a

maximum of 1 AMA PRA Category 1 Credit(s). Physicians should only

claim credit commensurate with the extent of their participation in

the activity.

Method of Participation and Receipt of CME/MOC Certificate

To obtain credit for JACC: Heart Failure CME/MOC, you must:

1. Be an ACC member or JACC subscriber.

2. Carefully read the CME/MOC-designated article available online and

in this issue of the journal.

3. Answer the post-test questions. At least 2 out of the 3 questions

provided must be answered correctly to obtain CME/MOC credit.

4. Complete a brief evaluation.

5. Claim your CME/MOC credit and receive your certificate electronically

by following the instructions given at the conclusion of the activity.

CME/MOC Objective for This Article: After reading this article, the reader

should be able to: 1) identify an approach to possible driveline site

infection in patients with a durable mechanical circulatory support de-

vice; 2) recognize neurological complications and initiate management in

From the aDepartment of Medicine, Duke University School of Medicine, Durh

Institute, Duke University School of Medicine, Durham, North Carolina. Th

Institute. Dr. DeVore has received research funding from American Heart

Novartis. Dr. C.B. Patel consults for Heartware/Medtronic and Thoratec/A

relationships relevant to the contents of this paper to disclose.

Manuscript received April 4, 2017; revised manuscript received June 11, 201

a patient with a durable mechanical circulatory support device; and 3)

initiate the appropriate therapies for the management of acute atrial ar-

rhythmias in a patient with a durable mechanical circulatory support

device.

CME/MOC Editor Disclosure: Editor-in-Chief Christopher M. O’Connor,

MD, FACC, has received consultant fees/honoraria from AbbVie, Inc.,

Actelion Pharmaceuticals Ltd., Bayer, Bristol-Myers Squibb, Cardio-

rentis, Merco & Co., Inc., ResMed, and Roche Diagnostics; and has

ownership interest in Biscardia, LLC. Executive Editor Mona Fiuzat,

PharmD, FACC, has received research support from ResMed, Gilead,

Critical Diagnostics, Otsuka, and Roche Diagnostics. Tariq Ahmad, MD,

MPH, has received a travel scholarship from Thoratec. Abhinav

Sharma, MD, has received support from Bayer-Canadian Cardiovas-

cular Society, Alberta Innovates Health Solution, Roche Diagnostics,

and Takeda. Mitchell Psotka, MD, PhD, and Kishan Parikh, MD, have

has reported that he has no relationships relevant to the contents of

this paper to disclose.

Author Disclosures: This paper was funded by the Duke Clinical Research

Institute. Dr. DeVore has received research funding from American Heart

Association, Amgen, and Novartis; and consults for Novartis. Dr. C.B.

Patel consults for Heartware/Medtronic and Thoratec/Abbott. Dr. P.A.

Patel has reported that he has no relationships relevant to the contents of

this paper to disclose.

Medium of Participation: Print (article only); online (article and quiz).

CME/MOC Term of Approval

Issue date: September 2017

Expiration date: August 31, 2018

am, North Carolina; and the bDuke Clinical Research

is paper was funded by the Duke Clinical Research

Association, Amgen, and Novartis; and consults for

bbott. Dr. P.A. Patel has reported that he has no

7, accepted June 11, 2017.

DeVore et al. J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7

Management of LVAD for the Non-LVAD Specialist S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1

622

Medical Management of Patients With aLeft Ventricular Assist Device for theNon-Left Ventricular Assist Device Specialist

Adam D. DeVore, MD, MHS,a,b Priyesh A. Patel, MD,a Chetan B. Patel, MDa,b

ABSTRACT

More than 2,400 continuous-flow left ventricular assist devices (LVADs) are implanted each year in the United States alone.

Both the number of patients living with LVADs and the life expectancy of these patients are increasing. As a result,

patients with LVADs are increasingly encountered by non-LVAD specialists who do not have training in managing advanced

heart failure for general medical care, cardiovascular procedures, and other subspecialty care. An understanding of the

initial evaluation and management of patients with LVADs is now an essential skill for many health care providers. In this

State-of-the-Art Review, we discuss current LVAD technology, summarize our clinical experience with LVADs, and

review the current data for the medical management of patients living with LVADs. (J Am Coll Cardiol HF 2017;5:621–31)

© 2017 by the American College of Cardiology Foundation.

DURABLE LEFT VENTRICULAR ASSIST

DEVICE USE IN 2017

Over the past decade, the use of continuous-flow leftventricular assist devices (LVADs) has increasedsignificantly. In the United States alone, approxi-mately 2,400 LVADs are implanted annuallycompared with only 459 in 2008 (1). More than 16,000patients have received a continuous-flow LVAD in theUnited States (2). These devices are a life-saving op-tion for advanced heart failure (HF) patients who areeither not eligible for a heart transplant or too ill tosafely wait for a transplant on medical therapy alone.Survival post-LVAD implantation also continues toimprove with 1-month survival estimates at 95%, and1- and 2-year survival estimates at 80% and 70%,respectively (1).

As the LVAD patient population increases, so doesthe need for access to general medical and cardio-vascular care. Providing care for patients withcontinuous-flow LVADs is no longer restricted toadvanced HF specialists; however, there are limiteddata and guidelines to assist non-LVAD specialists incaring for these patients. In this State-of-the-ArtReview, we discuss current LVAD devices andequipment and summarize initial management stra-tegies to assist non-LVAD specialists in providing carefor these complex patients.

CURRENT DURABLE LVADs AND EQUIPMENT

There are currently 2 durable (i.e., able to be dis-charged home) LVADs approved for adults by the U.S.

Food and Drug Administration: the HeartMate II leftventricular assist system (St. Jude Medical, St. Paul,Minnesota) and the HeartWare ventricular assist de-vice system (HVAD) (HeartWare, Framingham, Mas-sachusetts). HeartMate 3 (St. Jude Medical) wasrecently compared to HeartMate II in a large, ran-domized clinical trial and found to have improvedoutcomes at 6 months, although the device currentlyremains investigational (3).

This review considers the management of Heart-Mate II, HeartWare, and HeartMate 3. These pumpsdiffer in technical design, but all are continuous-flowpumps and have the following analogous compo-nents: 1) an inflow cannula that is surgically implan-ted into the left ventricular (LV) apex, serving as aconduit for blood from the LV to the pump; 2) a pumpenclosure which houses an impeller that circulatesblood; 3) an outflow graft that carries blood from thepump to the systemic circulation; and 4) a surgicallytunneled driveline that connects the pump to anexternal controller that operates and monitors thepump function. The external controller is connectedby 2 power cables to a battery powered source or apower module connected to an AC source while bat-teries are charging. Batteries can last up to 12 h, andremaining battery life is displayed on the externalcontroller. The pump will function properly if only 1power cable is connected to the external controller,but an alarm will sound. Additional device periph-erals include a backup controller, spare batteries,chargers, and brand-specific system monitors that areused to program, interrogate, and troubleshootLVADs (Online Figures 1 to 5).

TABLE 1 Typical LVAD Operating Parameters

HeartMate II HeartMate 3 HVAD

Typical speed, rpm* 8,000–10,000 5,000–6,000 2,400–3,200

Speed adjustment increment,rpm/increment

200 100 20

Flow, l/min 4–7 4–6 4–6

Power, W 5–8 4.5–6.5 3–7

Pulsatility index (or HVAD,peak to trough)

5–8 3.5–5.5 2–4 l/min/beat

*Speeds at the lower range or below the clinical ranges shown above indicate a low level ofsupport is needed and should prompt investigation of native contractility. Speeds above or nearthe high range described above should prompt investigation for adequate left ventricularunloading, including the possibilities of LVAD dysfunction or native valvular disease that could beaffecting unloading.

HVAD ¼ HeartWare ventricular assist device (system); LVAD ¼ left ventricular assist device;rpm ¼ revolutions per minute.

AB BR E V I A T I O N S

AND ACRONYM S

AV = aortic valve

BP = blood pressure

CT = computed tomography

HF = heart failure

LV = left ventricle

LVAD = left ventricular assist

device

= mean arterial pressure

J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7 DeVore et al.S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1 Management of LVAD for the Non-LVAD Specialist

623

BASIC LVAD FUNCTION. LVADs move blood from theLV apex to systemic circulation in a continuous(nonpulsatile) manner. Pump speed is the funda-mental parameter that the provider can alter. Aspump speed increases, the impeller within the pumphousing spins more rapidly and circulates a greatervolume of blood, thereby increasing LV unloadingand cardiac output. Contemporary pump design andoperating software are unable to automaticallymodulate speed or cardiac output based on physio-logic demand; therefore, they will operate at thespeed the provider sets, consuming as much power asneeded to maintain that speed.

The system controller monitors power consump-tion and estimates cardiac output based on speed andpower consumption. This estimated flow is reportedon the system controller in liters per minute. Notably,the flow is an estimated value based on power con-sumption at given speeds and may not be accurate inthe case of physiologic derangements (e.g., aorticinsufficiency or LVAD dysfunction including pumpthrombosis). Additionally, the controller measurestemporal power fluctuations to give an estimate ofpulsatility through the pump, termed pulsatility in-dex for HeartMate devices and represented graphi-cally as the HVAD pump flow waveform for theHeartWare device. Table 1 describes typical values forpump parameters informed by clinical trial data andexperience at our institution.

KEY DIFFERERNCES AMONG CURRENT LVADs. Cur-rent generation LVADs have analogous components,but there are clinically important differences amongdevices. HeartMate II is an axial flow pump with alarger profile than HVAD and HeartMate 3, conse-quently implantation in a surgically created pre-peritoneal pocket is required (Online Figure 1).HeartMate 3 and HVAD are centrifugal pumps withsmaller profiles and are implanted directly opposingthe heart (Online Figures 2 and 3). HVAD and Heart-Mate 3 provide more accurate cardiac output esti-mates during normal operation, given the centrifugaldesign, and by using the patient’s manually enteredhematocrit to estimate the serum viscosity. Impor-tantly, centrifugal flow pumps (i.e., HVAD andHeartMate 3) have a larger change in flow for a givenpressure gradient change across the pump (i.e.,afterload minus preload) than the axial flow Heart-Mate II. Therefore, centrifugal flow pumps are moresensitive to changes in preload and afterload, andcontrolling hypertension is a key element of normalcentrifugal flow pump function.

Both the HVAD and HeartMate 3 have automatedspeed modulation capabilities to enhance washing of

the pump and allow possible intermittentejection through the native aortic valve (AV).These rapid speed modulations, termed“Artificial Pulse” for HeartMate 3 and “LavareCycle” for HVAD (available through softwareupdate in Europe), entail slowing theimpeller and then accelerating the pumpspeed to the set speed or higher. Automaticspeed modulation may have benefits in pre-venting pump thrombosis for HeartMate 3, aswell as reducing AV insufficiency and stroke

for HVAD (3,4).

PATIENT ASSESSMENT

Most clinical encounters will require an assessment fornormal LVAD function and an assessment of heart rateand blood pressure (BP). Later, we discuss unique as-pects of a patient assessment and common and/orserious LVAD complications. Each complicationshould also be evaluated with a targeted patient his-tory, physical examination, and initial evaluation.

HISTORY AND PHYSICAL EXAMINATION. Uniqueaspects of the patient assessment include evaluationfor normal LVAD function and for common (some-times occult) complications. The history shouldinclude recent device parameters and alarms (CentralIllustration), symptoms of infection (including of thedriveline) or HF, and signs of hemoglobinuria (e.g.,dark urine) that could be a harbinger of LVADthrombosis. We also evaluate tolerability to antith-rombotic medications (most commonly aspirin andwarfarin) and ask about signs and symptoms sug-gestive of melena, because chronic bleeding fromarteriovenous malformations in the gastrointestinaltract is a common complication.

MAP

CENTRAL ILLUSTRATION Unique Aspects of Evaluating a Patient With an LVAD

DeVore, A.D. et al. J Am Coll Cardiol HF. 2017;5(9):621–31.

Key aspects of evaluating a patient with an LVAD include assessing recent device parameters and alarms and for common and/or serious complications such as infection,

heart failure, LVADthrombosis, andgastrointestinal bleeding. Thefigure alsodisplays the impact of continuous-flowLVADspeedonbloodflowpulsatility. As LVADspeeds

increase, more blood flows though the LVAD instead of being ejected through the aortic valve. Therefore, at higher pump speeds, there is a lower pulse pressure. Lower

pulsatility has an impact on the assessment of pulse and blood pressure on physical examination. SeeOnline Figures 1 to 5 for brand-specific images of device components.

GI ¼ gastrointestinal; HF ¼ heart failure; HVAD ¼ HeartWare ventricular assist device; LVAD ¼ left ventricular assist device; RPM ¼ revolutions per minute.

DeVore et al. J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7

Management of LVAD for the Non-LVAD Specialist S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1

624

TABLE 2 Suggested Approach to Suspected LVAD Infection

Initial Evaluation Initial Management

Obtain local wound culture Remove driveline dressing. Do not expresswound immediately and clean exit site(typically with chlorhexidine soap). Irrigatethe area until clean and then dry the areabefore expressing drainage and collectingspecimen.

Obtain blood cultures. Consider otherevaluations (e.g., urine culture ifnon-LVAD-related infection is alsosuspected).

Review prior culture data and considerempirical antibiotic therapy after bloodcultures are obtained. The most commonpathogens are Gram-positive organisms(e.g., coagulase-negative staphylococci andS. aureus), although Gram-negativeorganisms, fungal infections, andpolymicrobial infections also occur.

Many patients will also need warfarin held andbridging with unfractionated heparin ifsurgical intervention is necessary.

If percutaneous driveline infection issuspected, obtain a chest and abdomenultrasonogram or CT to assess forabscess and/or driveline stranding.

If bacteremia or sepsis is present, assess for thesource of bacteremia by using a chest andabdomen CT and transthoracic ortransesophageal echocardiogram.

Depending on results, surgical evaluation fordriveline debridement and infectiousdiseases consultation may be needed.

CT ¼ computed tomography; LVAD ¼ left ventricular assist device.

J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7 DeVore et al.S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1 Management of LVAD for the Non-LVAD Specialist

625

Unique aspects of the physical examinationinclude assessments of pulse, BP, auscultation of theLVAD, examination of the driveline and device con-nections, and device interrogation for device param-eters and alarms. A key aspect of measuring pulse andBP is understanding that the degree of arterial pul-satility depends on multiple factors, as follows: 1)underlying LV contractility; 2) AV function (i.e., theAV can sometimes be intentionally oversewn formanagement of aortic insufficiency); 3) LVAD pumpspeed; and 4) LVAD preload and afterload. Forexample, pulsatility decreases at increased pumpspeeds (Central Illustration), consequently, assess-ments of heart rate often require telemetry or elec-trocardiography instead of palpation of a pulse, whichis often absent. Similarly, current technology fornoninvasive BP measurement can be unreliable whenthere is decreased pulsatility, so Doppler ultraso-nography is often necessary to measure BP. To mea-sure, a manual BP cuff is inflated to occlude thebrachial artery, and a Doppler ultrasound probe isused to auscultate the brachial artery on the medialaspect of the antecubital fossa as the cuff is deflated.The pressure at which the sound of blood flow returnsto the brachial artery is recorded and best describedas an opening or Doppler pressure, although in prac-tice it is often called a mean arterial pressure (MAP). Ifa patient has significant pulsatility, then the openingpressure likely represents systolic BP. If there is lowpulsatility, then the opening pressure gives areasonable estimate of the MAP. The typical MAPtarget is #80 mm Hg to reduce stroke risk and mini-mize LVAD afterload (5).

Cardiac auscultation can reveal the “hum” of anLVAD, which is important if there is any concern forpump dysfunction, as the hum can vary depending onpump stress or pump stoppage. If there is any concernfor infection in patients, careful attention should begiven to the driveline (Online Figures 1 to 3), as mostLVAD infections involve the percutaneous driveline(6). Typical examination findings of a drivelineinfection include drainage or pus, presence of an ab-scess, or cellulitis.

ROUTINE LABORATORY TESTING AND IMAGING

STUDIES. In addition to usual laboratory testing,most patients with LVADs require screening evalua-tion for anemia and subclinical hemolysis. Screeningfor hemolysis may be done by either plasma-freehemoglobin (hemolysis defined by concentration of>40 mg/dl) or lactate dehydrogenase (hemolysis istypically defined as values 2.5� the upper limit ofnormal, >600 IU, or significantly above baseline)(5,7).

Multimodal imaging capabilities are essential forthe initial evaluation of LVAD patients, especiallyechocardiography due to its ease of access andportability. Echocardiography can provide importantinsights regarding acquired valvular disease, inflowcannula position and orientation, right and left ven-tricular function, filling pressures, and effectivenessof LV unloading. Real-time feedback from echocar-diography is key for diagnosing and troubleshootingLVAD dysfunction, including imaging the response tospeed adjustments, provocative maneuvers (e.g.,Valsalva and positional changes), and pacing adjust-ments. For example, newly identified AV opening orsevere mitral regurgitation is suggestive of inade-quate LV unloading from the pump, potentially froma speed that is set too low or from LVAD dysfunction.Adjustment to a higher LVAD speed under echocar-diographic guidance can provide diagnostic informa-tion regarding pump dysfunction while optimizingunloading (8). Importantly, echocardiography hasinherent limitations due to acoustic shadowing fromLVAD components, particularly for patients withapically-positioned centrifugal flow pumps (i.e.,HVAD or HeartMate 3), which can significantly limit2-dimensional and spectral Doppler analysis fromapical imaging windows. Comprehensive guidelinesfor echocardiographic imaging of LVAD patients arealso available (9).

Electrocardiogram-gated computed tomography(CT) angiography and nuclear imaging also play

FIGURE 1 Evaluation and Management of Gastrointestinal Bleeding

Proposed approach to the initial evaluation and management of gastrointestinal bleeding. CT ¼ computed tomography; GI ¼ gastrointestinal;

LVAD ¼ left ventricular assist device; RBC ¼ red blood cell.

DeVore et al. J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7

Management of LVAD for the Non-LVAD Specialist S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1

626

important roles in the evaluation of LVAD patients.Contrast-enhanced gated CT scans timed for opacifi-cation of the LVAD inflow and outflow cannulas allowexcellent visualization of LVAD components outsideof the metallic pump housing, particularly theoutflow graft, and can help with the diagnosis ofcannula malposition and outflow graft narrowing,kinking, or thrombosis (10). In cases of suspectedinfection, nuclear imaging with radioisotope-taggedwhite blood cells can help identify the presence andextent of infection.

COMMON AND/OR SERIOUS

LVAD COMPLICATIONS

Adverse events with contemporary, continuous-flowLVADs are lower than older, pulsatile LVADs,although the rate of adverse events remains unac-ceptably high (11). More than 50% of patients arereadmitted for adverse events in the first 6 monthspost-LVAD implant, and patients experience anaverage of 3.5 adverse events (most commonly

bleeding, infection, and/or arrhythmia) in the firstyear post-implantation (1). By 2 years post-implant,approximately 80% of patients have experienced amajor adverse event (1). We focus here on commonand/or serious complications that occur outside of theperioperative period. The reader may refer to otherreviews and guidelines that discuss perioperativemedical management and complications (5,12). Themanagement of LVAD complications should occur inconsultation with advanced HF providers, althoughthe following discussion may assist with initial eval-uation and management.

LVAD INFECTIONS. Most LVAD infections involvethe percutaneous driveline (6) and can range inseverity from a local skin infection to a systemicinfection that involves the LVAD pump. Patients andcaregivers are educated about specific instructions forregular dressing changes and monitoring for signsand symptoms of infection at the driveline exit site.Nevertheless, data suggest that many driveline in-fections are unavoidable, resulting from trauma to

FIGURE 2 Neurological Emergencies

Proposed approach to neurologic emergencies. aPTT ¼ activated partial thromboplastin time; ICH ¼ ischemic hemorrhage; LDH ¼ lactate

dehydrogenase; SDH ¼ subdural hematoma; other abbreviations as in Figure 1.

J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7 DeVore et al.S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1 Management of LVAD for the Non-LVAD Specialist

627

the driveline exit site (e.g., accidentally dropping acontroller or battery pack) (13).

The evaluation for infection includes drivelinedrainage culture, blood cultures, and imaging(Table 2). In particular, blood cultures are important toevaluate for occult bloodstream infections, becauseLVAD patients can present with atypical signs andsymptoms. Imaging the internal course of the drivelineup to the pump can be accomplished with ultraso-nography or CT. We prefer CT scans with or withoutcontrast to evaluate the extent of driveline and/orpump pocket infections. The most common pathogensare Gram-positive skin flora (e.g., coagulase-negativestaphylococci and Staphylococcus aureus), althoughGram-negative organism (e.g., Pseudomonas spp.and Enterobacteriaceae), fungal, and polymicrobialinfections also can occur (6,14,15).

NONSURGICAL BLEEDING. Nonsurgical bleeding is acommon complication after LVAD implantation and a

familiar cause for hospital readmission (16). There aremultiple reasons for nonsurgical bleeding, forexample: 1) use of antithrombotic therapy; 2) acquiredcoagulopathy, especially von Willebrand factor defi-ciency fromdegradation of high-molecular-weight vonWillebrand factormultimers as theymove through andare sheared by the pump; and 3) formation of arterio-venous malformations in the gastrointestinal tract,nasopharynx, brain, and other tissues, which seemsrelated to continuous blood flow and an associatedabnormal regulation of angiogenic factors (17).

Our general approach to treatment is to holdantithrombotic therapy, control the source ofbleeding, and transfuse blood products as needed(while recognizing the potential for antibody sensiti-zation in patients considered for heart trans-plantation). Patients rarely need active reversal ofantithrombotic therapies, unless there is life-threatening hemorrhage such as intracranialhemorrhage. Gastrointestinal bleeding related to

FIGURE 3 HF Profiles

Proposed heart failure profiles to guide initial evaluation and management. For refractory cases, consider extracardiac causes of volume

overload such as a peripheral shunt, cirrhosis, or nephrotic syndrome. AI ¼ aortic valve insufficiency; BP ¼ blood pressure; DOE ¼ dyspnea on

exertion; HF ¼ heart failure; IVC ¼ inferior vena cava; JVP ¼ jugular venous pressure; LV ¼ left ventricular; LVEDD ¼ left ventricular

end-diastolic dimension; MCS ¼ mechanical circulatory support; MR ¼ mitral regurgitation; PCWP ¼ pulmonary capillary wedge pressure;

PI ¼ pulsatility index; PND ¼ paroxysmal nocturnal dyspnea; RHC ¼ right heart cardiac catheterization; RV ¼ right ventricular.

DeVore et al. J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7

Management of LVAD for the Non-LVAD Specialist S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1

628

arteriovenous malformations is the most commonpresentation for nonsurgical bleeding and can occur atany time after implantation (18). Figure 1 outlines ourgeneral approach to evaluation and management.Notably, many patients develop recurrent gastroin-testinal bleeding, and a review of prior endoscopicprocedures is an important aspect of the initial eval-uation, although the diagnostic and therapeutic yieldof endoscopy remains high with repeated in-terventions (19).

ATRIAL AND VENTRICULAR ARRHYTHMIAS. Bothatrial and ventricular arrhythmias are common post-LVAD implantation. Approximately 20% of patients inthe HeartMate II Destination Therapy trial presentedwith atrial arrhythmias (20,21). Approximately one-half of all LVAD patients have atrial fibrillationin observational studies (22), and as many as 22% to59% have reported ventricular arrhythmias (23).

Importantly, many LVAD patients may tolerate ven-tricular tachycardia for hours due to continuous he-modynamic support provided by the LVAD. Ourinitial evaluation of LVAD patients with ventriculartachycardia is similar to that of patients without anLVAD, including an assessment of hemodynamicstability with vital signs and obtaining an electrocar-diogram if able. Cardioversion can be safely per-formed for patients requiring urgent therapy byplacing external defibrillator pads in the usual loca-tions, although pads should not be placed over theLVAD pump. There is no need to stop or disconnectthe LVAD before external cardioversion.

When evaluating a patient with ventricular tachy-cardia, the mechanism of arrhythmia is important toconsider. For example, contact from the inflow can-nula to the LV can occur when the LV is completelydecompressed by continuous inflow, possibly fromdehydration or an excessively high speed (24).

J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7 DeVore et al.S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1 Management of LVAD for the Non-LVAD Specialist

629

This scenario can be clarified by a history suggestiveof hypovolemia and confirmed by echocardiogramand/or LVAD interrogation, which may demonstratelow pulsatility index (HeartMate devices) or lowHVAD flow waveform (HeartWare device). Incontrast, another possible mechanism is scar relatedto myocardial fibrosis. The initial treatment strategiesfor these 2 scenarios can be quite different, rangingfrom optimization of fluid status to initiation ofantiarrhythmic medications.

LVAD MALFUNCTION OR FAILURE. LVAD malfunc-tion or failure may result from a variety of causes,including electrical malfunction and thrombosis. Me-chanical pump failure is less common in continuous-flow devices than previous generation LVADs.Electrical malfunction typically presents with LVADalarms or pump stoppage, and initial managementinvolves checking to ensure device connections aresecure, reviewing LVAD alarms, and evaluating LVADflow and patient stability. Management of suspectedelectrical malfunction typically requires consultationwith an LVAD specialist. If the provider and patientare put in the position of attempting to change acontroller, which results in an obligatory temporaryLVAD stoppage, then one must first determine thepatient’s hemodynamic dependence on LVAD support(e.g., patients with an oversewn AV are more depen-dent on LVAD flow than patients with a normal func-tioning AV). Controller exchange should be performedquickly by well-trained personnel to prevent compli-cations related to lack of native ejection for thesepatients during temporary pump stoppage.

LVAD thrombosis can occur on the inflow cannula,pump, or outflow graft. HeartMate II was noted tohave higher-than-expected rates of thrombosis in alarge observational study (25) and higher rates ofdevice malfunction requiring surgical replacement inclinical trials compared with HVAD (16.2% vs. 8.8%,respectively) (26) or HeartMate 3 (7.7% vs. 0.7%,respectively) (3). Thrombosis presents in various de-grees of severity, including asymptomatic elevationsin plasma-free hemoglobin or lactate dehydrogenase,clinical evidence of hemolysis, LVAD alarms forchanges in power or flow, and isolated left-sided orbiventricular HF. Initial management includes eval-uating current antithrombotic strategies includinginternational normalized ratio and antiplatelet ther-apy dose, stabilizing the patient; and in select cases,planning for emergent surgical interventions orthrombolytic agents. Echocardiographic ramp studies(8) may also be used to evaluate LVAD dysfunctionrelated to thrombosis, and CT angiography may beperformed to evaluate the outflow graft.

NEUROLOGICAL EMERGENCIES. Ischemic and hem-orrhagic strokes remain the most dreaded adverseevents following LVAD implantation, occurring withan annual incidence of approximately 9% (27).Ischemic and hemorrhagic strokes cause significantmorbidity and are associated with high rates of mor-tality (28). The risk of stroke and death among patientsappears to be bimodal with highest risks in the peri-operative period and increasing approximately 1 yearlater (29). Figure 2 displays our approach to managingneurological emergencies. Similar to stroke in non-LVAD patients, early patient recognition and urgentmedical evaluation, including brain and vascular im-aging, are key steps in the evaluation. One must alsoconsider whether LVAD thrombosis (i.e., evaluateLVAD pump parameters, evaluate for hemolysis, andobtain echocardiography) could be an embolic sourcein patients with ischemic or hemorrhagic stroke.

HEART FAILURE. Acute right ventricular failure post-LVAD implantation is common and well described,including associated risk factors and outcomes(30–32). Heart failure that occurs late after LVADimplantation is a distinct entity with an emergingevidence base, especially for late right HF (33–35).In an analysis of data from the HeartMate II Destina-tion Therapy trial, patients with late right HF hadworse outcomes at 2 years than those without,including worse quality of life, poorer functionalcapacity, increased rehospitalizations (median: 6[range 2 to 19] vs. 3 [range 0 to 27], respectively), anddecreased rate of survival (58 � 8% vs. 71 � 2%,respectively) (35).

We propose 3 patient profiles as a conceptualframework to guide initial evaluation and manage-ment of LVAD patients presenting with HF (Figure 3).We use the patient’s history and physical examina-tion, an echocardiogram, and/or an invasive hemo-dynamic assessment to determine if HF isbiventricular, isolated left-sided, or isolated right-sided. We also ensure accurate assessment of theMAP and evaluate for LVAD dysfunction.

When evaluating HF, one must understand the roleof continuous AV insufficiency, which can developde novo or from exacerbation of underlying AV pa-thology and is more common with contemporary,continuous-flow LVADs than older, pulsatile LVADs(36). Aortic insufficiency most likely occurs due tocommissural fusion of the AV leaflets from immobilitywith associated remodeling and/or trauma from high-pressure continuous flow from the outflow cannula(37), resulting in a futile circuit from valvular insuf-ficiency that reduces LVAD efficiency and leads to HFsymptoms. That is, blood travels from the LV through

DeVore et al. J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7

Management of LVAD for the Non-LVAD Specialist S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1

630

the pump, into the aorta, back through the AV, andagain into the LV. Due to the continuous nature of AVinsufficiency, even a small regurgitant orifice can leadto a large volume of regurgitant blood flow. Aorticinsufficiency should be assessed by echocardiogra-phy; management typically involves BP control andevaluation for AV intervention (transcatheter or sur-gical AV replacement or surgical oversewing).

UNRESPONSIVE PATIENTS AND

CARDIOPULMONARY RESUSCITATION

Care for an unresponsive LVADpatient requires uniqueconsiderations compared with the standard approachto advanced cardiovascular life support: 1) assessmentfor normal LVAD power and function is essential(evaluate device connections, check device parame-ters and alarms, and auscultate for LVAD hum) (38);2) assessment of pulse and BP is limited by continuous-flow physiology; 3) there are limited data for the safetyand efficacy of chest compressions in LVAD patients,although device manufacturers caution about poten-tial dislodgement of the inflow tract/outflow graftduring compressions; and 4) the initial patient surveyshould consider the previously-mentioned complica-tions (e.g., infection, bleeding, neurologicalemergencies).

PROCEDURES FOR PATIENTS ON AN LVAD

Because patients with LVADs live longer, more LVADpatients are requiring minor procedures and noncar-diac surgery. These procedures should be coordinatedwith an LVAD specialist. Specific attention shouldbe given to the perioperative management ofantithrombotic therapy and BP monitoring as theavailable data highlight an increased risk of bleeding,higher frequency of intraoperative hypotension, and

a risk for acute kidney injury when these patientsundergo noncardiac surgery (39,40). As mentionedpreviously, there are multiple reasons for theincreased risk of bleeding beyond the risks ofantithrombotic therapy, and the risk may be under-estimated by providers who are unfamiliar withLVADs. For example, in a study of patients whounderwent noncardiac operations, including abdom-inal surgeries, thoracic surgeries, and endoscopicprocedures at the Mayo Clinic, 15% of operationsrequired red blood cell transfusions, 9% requiredfresh frozen plasma, and 6% required platelets (39).

CONCLUSIONS

The number of ambulatory LVAD patients isincreasing, as is their life expectancy. Non-LVADspecialists will increasingly encounter LVAD pa-tients and should be armed with the tools to provideinitial assessment and management for these com-plex patients. Co-management of these patients willalso become increasingly important as research anddevice innovations allow us to overcome the chal-lenges preventing expansion of LVAD therapy tomore patients such as the high rate of adverse events,the care required for devices with external batterysources, and cost.

ACKNOWLEDGMENTS The authors thank ErinCampbell, MS, for editorial contributions and Jon-athon Cook for assistance with images. Ms. Campbelland Mr. Cook did not receive compensation for theircontributions. Ms. Campbell is an employee at theinstitution where this study was conducted.

ADDRESS FOR CORRESPONDENCE: Dr. Adam D.DeVore, Department of Medicine, Duke Clinical ResearchInstitute, 400 Pratt Street, NP-8064, Durham, NorthCarolina 27705. E-mail: adam.devore@duke.edu.

RE F E RENCE S

1. Kirklin JK, Naftel DC, Pagani FD, et al. SeventhINTERMACS annual report: 15,000 patients andcounting. J Heart Lung Transplant 2015;34:1495–504.

2. Kirklin JK, Cantor RS, Myers SL, et al. Intermacsinteragency registry for mechanically assisted cir-culatory support: quarterly statistical report 2016Q3: implant and event dates: June 23, 2006 toSeptember 30, 2016. Birmingham, AL: DataCollection and Analysis Center, University of Al.Available at: www.uab.edu/medicine/intermacs/images/Federal_Quarterly_Report/Federal_Partners_Report_2016_Q3.pdf. Accessed March 20, 2017.

3. Mehra MR, Naka Y, Uriel N, et al. A Fully magnet-ically levitated circulatory pump for advanced heartfailure. N Engl J Med 2017;376:440–50.

4. Saeed D, Westenfeld R, Maxhera B, et al.Prevalence of de novo aortic valve insufficiency inpatients after HeartWare VAD implantation withan intermittent low-speed algorithm. ASAIO J2016;62:565–70.

5. Feldman D, Pamboukian SV, Teuteberg JJ, et al.The 2013 International Society for Heart and LungTransplantation guidelines for mechanical circu-latory support: executive summary. J Heart LungTransplant 2013;32:157–87.

6. Gordon RJ, Weinberg AD, Pagani FD, et al.Prospective, multicenter study of ventricular assistdevice infections. Circulation 2013;127:691–702.

7. Shah P, Mehta VM, Cowger JA, Aaronson KD,Pagani FD. Diagnosis of hemolysis and devicethrombosis with lactate dehydrogenase during left

ventricular assist device support. J Heart LungTransplant 2014;33:102–4.

8. Uriel N, Morrison KA, Garan AR, et al. Devel-opment of a novel echocardiography ramp test forspeed optimization and diagnosis of devicethrombosis in continuous-flow left ventricularassist devices: the Columbia ramp study. J Am CollCardiol 2012;60:1764–75.

9. Stainback RF, Estep JD, Agler DA, et al. Echo-cardiography in the management of patients withleft ventricular assist devices: recommendationsfrom the American Society of Echocardiography.J Am Soc Echocardiogr 2015;28:853–909.

10. Vivo RP, Kassi M, Estep JD, et al. MDCT assess-ment of mechanical circulatory support devicecomplications. J Am Coll Cardiol Img 2015;8:100–2.

J A C C : H E A R T F A I L U R E V O L . 5 , N O . 9 , 2 0 1 7 DeVore et al.S E P T E M B E R 2 0 1 7 : 6 2 1 – 3 1 Management of LVAD for the Non-LVAD Specialist

631

11. Kirklin JK, Naftel DC, Kormos RL, et al. FifthINTERMACS annual report: risk factor analysis frommore than 6,000 mechanical circulatory supportpatients. J Heart Lung Transplant 2013;32:141–56.

12. DeVore AD, Mentz RJ, Patel CB. Medicalmanagement of patients with continuous-flow leftventricular assist devices. Curr Treat Options Car-diovasc Med 2014;16:283.

13. Zierer A, Melby SJ, Voeller RK, et al. Late-onset driveline infections: the Achilles’ heel ofprolonged left ventricular assist device support.Ann Thorac Surg 2007;84:515–20.

14. Topkara VK, Kondareddy S, Malik F, et al. In-fectious complications in patients with left ven-tricular assist device: etiology and outcomes in thecontinuous-flow era. Ann Thorac Surg 2010;90:1270–7.

15. Simeon S, Flecher E, Revest M, et al. Leftventricular assist device-related infections: amulticentric study. Clin Microbiol Infect 2017Mar 18 [E-pub ahead of print].

16. Bonde P, Dew MA, Meyer D, et al. 4 Nationaltrends in readmission (REA) rates following leftventricular assist device (LVAD) therapy. J HeartLung Transplant 2011;30:S9.

17. Tabit CE, Chen P, Kim GH, et al. Elevatedangiopoietin-2 level in patients with continuous-flow left ventricular assist devices leads toaltered angiogenesis and is associated with highernonsurgical bleeding. Circulation 2016;134:141–52.

18. Suarez J, Patel CB, Felker GM, Becker R,Hernandez AF, Rogers JG. Mechanisms of bleedingand approach to patients with axial-flow left ven-tricular assist devices. Circ Heart Fail 2011;4:779–84.

19. Dakik HK, McGhan AA, Chiu ST, et al. Thediagnostic yield of repeated endoscopic evaluationin patients with gastrointestinal bleeding and leftventricular assist devices. Dig Dis Sci 2016;61:1603–10.

20. Maury P, Delmas C, Trouillet C, et al. Firstexperience of percutaneous radio-frequencyablation for atrial flutter and atrial fibrillation ina patient with HeartMate II left ventricular assistdevice. J Interv Card Electrophysiol 2010;29:63–7.

21. Slaughter MS, Rogers JG, Milano CA, et al.Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med2009;361:2241–51.

22. Enriquez AD, Calenda B, Gandhi PU, Nair AP,Anyanwu AC, Pinney SP. Clinical impact of atrial

fibrillation in patients with the HeartMate II leftventricular assist device. J Am Coll Cardiol 2014;64:1883–90.

23. Nakahara S, Chien C, Gelow J, Dalouk K,Henrikson CA, Mudd J, et al. Ventricular arrhyth-mias after left ventricular assist device. CircArrhythm Electrophysiol 2013;6:648–54.

24. Vollkron M, Voitl P, Ta J, Wieselthaler G,Schima H. Suction events during left ventricularsupport and ventricular arrhythmias. J Heart LungTransplant 2007;26:819–25.

25. Starling RC, Moazami N, Silvestry SC, et al.Unexpected abrupt increase in left ventricularassist device thrombosis. N Engl J Med 2014;370:33–40.

26. Rogers JG, Pagani FD, Tatooles AJ, et al.Intrapericardial left ventricular assist device foradvanced heart failure. N Engl J Med 2017;376:451–60.

27. Parikh NS, Cool J, Karas MG, Boehme AK,Kamel H. Stroke risk and mortality in patients withventricular assist devices. Stroke 2016;47:2702–6.

28. Cho SM, Moazami N, Frontera JA. Stroke andintracranial hemorrhage in HeartMate II and HeartWareleft ventricular assist devices: a systematic review.Neurocrit Care 2017;27:17–25.

29. Frontera JA, Starling R, Cho SM, et al. Riskfactors, mortality, and timing of ischemic andhemorrhagic stroke with left ventricular assistdevices. J Heart Lung Transplant 2017;36:673–83.

30. Matthews JC, Koelling TM, Pagani FD,Aaronson KD. The right ventricular failure risk scorea pre-operative tool for assessing the risk of rightventricular failure in left ventricular assist devicecandidates. J Am Coll Cardiol 2008;51:2163–72.

31. Kormos RL, Teuteberg JJ, Pagani FD, et al. Rightventricular failure in patients with the HeartMate IIcontinuous-flow left ventricular assist device: inci-dence, risk factors, and effect on outcomes.J Thorac Cardiovasc Surg 2010;139:1316–24.

32. Grant AD, Smedira NG, Starling RC,Marwick TH. Independent and incremental role ofquantitative right ventricular evaluation for theprediction of right ventricular failure after leftventricular assist device implantation. J Am CollCardiol 2012;60:521–8.

33. Takeda K, Takayama H, Colombo PC, et al.Incidence and clinical significance of late rightheart failure during continuous-flow left

ventricular assist device support. J Heart LungTransplant 2015;34:1024–32.

34. Kapelios CJ, Charitos C, Kaldara E, et al.Late-onset right ventricular dysfunction aftermechanical support by a continuous-flow leftventricular assist device. J Heart Lung Transplant2015;34:1604–10.

35. Rich JD, Gosev I, Patel CB, et al. The incidence,risk factors, and outcomes associated with lateright-sided heart failure in patients supported withan axial-flow left ventricular assist device. J HeartLung Transplant 2017;36:50–8.

36. Rajagopal K, Daneshmand MA, Patel CB, et al.Natural history and clinical effect of aortic valveregurgitation after left ventricular assist deviceimplantation. J Thorac Cardiovasc Surg 2013;145:1373–9.

37. Wang TS, Hernandez AF, Felker GM,Milano CA, Rogers JG, Patel CB. Valvular heartdisease in patients supported with left ventricularassist devices. Circ Heart Fail 2014;7:215–22.

38. Yuzefpolskaya M, Uriel N, Flannery M, et al.Advanced cardiovascular life support algorithm forthe management of the hospitalized unresponsivepatient on continuous flow left ventricular assistdevice support outside the intensive care unit. EurHeart J 2016;5:522–6.

39. Barbara DW, Wetzel DR, Pulido JN, et al. Theperioperative management of patients with leftventricular assist devices undergoing noncardiacsurgery. Mayo Clinic Proc 2013;88:674–82.

40. Mathis MR, Sathishkumar S, Kheterpal S, et al.Complications, risk factors, and staffing patternsfor noncardiac surgery in patients with left ven-tricular assist devices. Anesthesiology 2017;126:450–60.

KEY WORDS heart failure, left ventricularassist devices, mechanical circulatorysupport

APPENDIX For supplemental figures, pleasesee the online version of this paper.

Go to http://www.acc.org/jacc-journals-cme totake the CME/MOC quizfor this article.