novembro2008_transplante

8/8/2019 Novembro2008_Transplante

1/9

Improved Survival and Decreasing Incidence ofAdverse Events With the HeartMate II Left

Ventricular Assist Device as Bridge-to-TransplantTherapyRanjit John, MD, Forum Kamdar, BS, Kenneth Liao, MD, Monica Colvin-Adams, MD,Andrew Boyle, MD, and Lyle Joyce, MD

Division of Cardiothoracic Surgery, Department of Surgery, and Division of Cardiology, University of Minnesota, Minneapolis,Minnesota

Background. Pulsatile left ventricular assist devices(LVADs) are effective as bridge-to-transplant therapy,but they are limited by their large size and lack of

durability. Smaller, more durable, continuous flow de-vices such as the HeartMate II LVAD are increasinglybeing used. The aim of this study is to report oursingle-center experience with this device as bridge-to-transplant therapy.Methods. Overall, 47 patients received HeartMate II

LVADs at our center from June 2005 to July 2007; 32 asbridge to transplant, 7 as destination therapy, and 8 asexchange therapy for a failed HeartMate XVE. We re-viewed our experience with the device as bridge-to-transplant therapy and report on patient survival andadverse events.Results. The mean age of the bridge-to-transplant pa-

tients was 50.75 13.78 years; 10 (31.3%) were female.The cause of the underlying disease was ischemic in 18patients (56.3%), idiopathic in 11 (34.4%), myocarditis in1 (3.1%), postpartum cardiomyopathy in 1 (3.1%), andcongenital heart disease in 1 (3.1%). The mean durationof HeartMate II support was 193.2 139.9 days. At 30days after HeartMate II placement, the patient survival

was 96.9% by Kaplan-Meier analysis; at 6 months (aliveor transplanted), 86.9%. Major adverse events includedbleeding requiring reexploration in 5 patients (15.6%),

right ventricular failure requiring right ventricular assistdevice support in 2 (6.3%), LVAD-related infections in 4(12.5%), neurologic or thromboembolic events in 2 (6.3%),and gastrointestinal bleeding in 5 (15.6%). We noted oneserious device malfunction (3.1%) resulting in the pa-tients death; in addition, 2 patients experienced pumpthrombosis (6.3%).Conclusions. Despite morbidity, use of the HeartMate

II LVAD as bridge-to-transplant therapy is associatedwith excellent survival and low mortality rates. Wefound a marked decrease in morbidity related to rightventricular failure, to device-related infections, and tothromboembolic events. However, the requirementsfor anticoagulation therapy may be associated withincreased mediastinal and gastrointestinal bleeding.Strategies to optimize anticoagulation therapy may fur-ther improve results for these critically ill patients.

(Ann Thorac Surg 2008;86:122735) 2008 by The Society of Thoracic Surgeons

Left ventricular assist device (LVAD) placement is anaccepted treatment for patients with end-stage heartfailure [1]. The increased applicability and excellent re-sults with LVADs has revolutionized the treatment op-

tions available for the patient with end-stage heart fail-ure. Most patients who have undergone LVADimplantation as bridge-to-transplant (BTT) therapy havebeen supported by pulsatile, volume-displacement de-vices such as the HeartMate XVE (Thoratec, Pleasanton,California), the Novacor LVAD (WorldHeart, Oakland,California), or the Thoratec VAD (Thoratec) [26]. Suc-

cess with those devices for BTT therapy has led to theirsuccessful use as an alternative altogether to a transplant,namely, as destination therapy [7]. While tremendoussuccess has been achieved with these devices, their use is

associated with significant comorbidity, which may berelated to several factors including the need for extensivesurgical dissection, a large pump, and a large-diameterpercutaneous lead. Even more important, their long-termdurability is limited, frequently requiring reoperationsfor device exchange, which often result in significantmorbidity and mortality.

The new HeartMate II LVAD, which incorporates con-tinuous flow, rotary pump technology, represents theAccepted for publication June 2, 2008.

Presented at the Forty-fourth Annual Meeting of The Society of Thoracic

Surgeons, Fort Lauderdale, FL, Jan 2830, 2008.

Address correspondence to Dr John, Division of Cardiothoracic Surgery,University of Minnesota, 420 Delaware St SE, MMC 207, Minneapolis,MN 55455; e-mail:[email protected].

Dr Boyle discloses that he has a financial relationshipwith Thoratec.

2008 by The Society of Thoracic Surgeons 0003-4975/08/$34.00Published by Elsevier Inc doi:10.1016/j.athoracsur.2008.06.030
mailto:[email protected]:[email protected]:[email protected]


2/9

next generation of devices [8, 9]. The simpler design ofcontinuous flow, rotary pump technology promises greaterlong-term mechanical reliability, with a single moving part:the internal rotor (Figs 1 and 2). The HeartMate II LVAD isone seventh the size and one fourth the weight of theprevious HeartMate XVE. We report our single-centerexperience, discussing survival and adverse events withthe HeartMate II LVAD as BTT therapy in 32 patientswith end-stage heart failure at the University of Minne-sota Medical Center.

Material and MethodsPatients

From June 2005 through July 31, 2007, a total of 47patients underwent HeartMate II placement at the Uni-versity of Minnesota Medical Center: 32 as BTT therapy,7 as destination therapy, and 8 as exchange therapy for a

failed XVE. The mean duration of HeartMate II supportwas 193.2 139.9 days.

Our study focuses on the 32 BTT patients. All 32 studypatients were part of a prospective, multicenter studyevaluating the use of the HeartMate II LVAD as BTTtherapy. Patients with end-stage heart failure who were

on our transplant waiting list were eligible for studyenrollment. The protocol for the study was approved bythe Food and Drug Administration and by our owncenters Institutional Review Board. Patient consent fordata collection and for reporting was obtained by astandard informed consent process. (Detailed inclusionand exclusion criteria are listed in the SupplementaryAppendix, available with the referenced article at www.nejm.org [10].)

HeartMate II

The HeartMate II consists of an internal blood pump witha percutaneous lead that connects the pump to an exter-

nal system driver and power source. The pump has animplant volume of 63 mL and generates up to 10 L/min offlow at a mean pressure of 100 mm Hg. Details ofHeartMate II function and the implantation techniquehave been described elsewhere [8, 9].

The implantation technique at the University ofMinnesota is briefly summarized here. To implant theHeartMate II, a median sternotomy is performed. Thepatient is placed on cardiopulmonary bypass, with aorticand right atrial cannulation. The inflow cannula is in-serted into an opening made in the apex of the leftventricle with a coring knife. A circular Teflon (Impra,subsidiary of L.R. Bard, Tempe, Arizona) pledget in theshape of a donut is placed around the ventricular apical

core. Then, 2-0 Tevdek mattress sutures are used tosecure the inflow cuff to the ventricle using the circularTeflon strip. After a side-biting clamp is placed in themidportion of the ascending aorta, the outflow graft issewn to a longitudinal aortotomy. The pump is surgicallyplaced in a preperitoneal pocket at the level of the

Fig 1. External view of HeartMate II left ventricular assist device.(Illustration Thoratec Corporation. Reproduced with permission.)

Fig 2. Cross-sectional internal view of Heart-Mate II left ventricular assist device. (Illustra-tion Thoratec Corporation. Reproducedwith permission.)

1228 JOHN ET AL Ann Thorac SurgHEARTMATE II LVAD AS BRIDGE-TO-TRANSPLANT 2008;86:122735
http://www.nejm.org/http://www.nejm.org/http://www.nejm.org/http://www.nejm.org/


3/9


4/9

23-year-old male who was transferred from an outsidehospital with acute cardiogenic shock with multisystemorgan failure and required urgent placement of biven-

tricular support with CentriMag Levitronix devices (Lev-itronix LLC, Waltham, MA). After failure to be weanedfrom temporary biventricular support, he underwentplacement of a HeartMate II LVAD; he required rightventricular assist device (RVAD) support with theCentriMag device postoperatively.

Hemodynamic Data

Hemodynamic data at baseline for the BTT patients areshown in Table 2 (excluding the hemodynamic data of1 patient who received biventricular support beforeHeartMate II placement).

End-Organ Function Data

Serum markers of end-organ function before HeartMateII placement and at 3-month follow-up for the BTTpatients are shown in Table 3.

Outcomes

The 30-day operative survival rate by Kaplan-Meier anal-ysis was 95.6% (Fig 3). Of the 32 BTT patients, 1 patientdied 10 days after HeartMate II placement from exsan-guination related to subclavian vein hemorrhage aftercentral venous line removal. The survival rate at 6months (alive or transplanted) was 86.9%. After the initialpostoperative 1-month period, 2 patients died: 1 was

doing well but died at home from sudden unexpectedpump malfunction; 1 died in the hospital from septiccomplications with multisystem organ failure.

Of the 32 BTT patients, 19 (59.4%) have now undergonea heart transplant after an average of 193 93.4 days;

another 9 are doing well at home and awaiting a hearttransplant. All 19 BTT patients who underwent a trans-plant were discharged to home. The 1 remaining patientwho was alive and discharged to home was not eligiblefor a transplant because of postoperative paraplegia.Outcomes are detailed in Table 4. Adverse events occur-ring during HeartMate II support are outlined in Table 5.

Neurologic or Thromboembolic Events

During the entire period of HeartMate II support, wenoted only 1 thromboembolic event (2.2%). This patientsubsequently underwent a heart transplant and, as of theend of this study, is doing well. In addition, 1 patient hadpostoperative paraplegia, without any significant recov-

ery. The cause of her paraplegia is unknown. It may be

Table 2. Hemodynamic Variables at Baseline (n 31)

Data Mean SD

Systolic arterial pressure (mm Hg) 103.9 14.7

Diastolic arterial pressure (mm Hg) 68.7 10.5

Systolic PAP (mm Hg) 56.52 13.41

Diastolic PAP (mm Hg) 28.19 6.23Right atrial mean (mm Hg) 14.17 5.27

PCWP (mm Hg) 24.45 5.90

PVR (Woods units) 3.69 1.98

Cardiac output (L/min) 3.96 1.20

Cardiac index 1.99 0.47

PAP pulmonary artery pressure; PCWP pulmonary capillarywedge pressure; PVR pulmonary vascular resistance.

Table 3. Serum Indicators of End-Organ Function atBaseline and at 3 Months (n 32)

Data Baseline 3 Months p Value

Renal

Creatinine 1.55 0.56 1.23 0.5 0.031

Blood urea nitrogen 34.81 18.70 22.6 13.1 0.001

Liver

Alanineaminotransferase

68.22 100.87 36.2 19.3 0.05

Aspartateaminotransferase

70.38 118.25 53.5 30 0.158

Total bilirubin 1.31 0.83 0.78 0.5 0.001

Table 4. Outcomes on Left Ventricular Assist Device Support(n 32)

Outcomes Number

Transplanted 19

On the transplant waiting list 9

Death 3Early deaths (30 days)

Subclavian vein hemorrhage 1 (day 10)

Late deaths (30 days)

Device failure 1 (day 73)

Multisystem organ failure 1 (day 42)

Table 5. Adverse Events on Left Ventricular Assist DeviceSupport (n 32)

Adverse Event Number

Neurologic or thromboembolic 2 (6.25%)

Stroke 1 (3.12%)

Paraplegia 1 (3.12%)

Hemorrhagic

Mediastinal bleeding requiring reexploration 5 (15.6%)

Gastrointestinal bleeding 5 (15.6%)

Infectious

Driveline infection 4 (12.5%)

Pump pocket infection 0

Right ventricular failure 2 (6.25%)

Requiring RVAD support 2 (6.25%)

Requiring prolonged inotropic support 0

Pump thrombosis 2 (6.25%)

Device malfunction 1 (3.12%)

Device replacement 0

Renal failure 1 (3.12%)

Respiratory failure 2 (6.25%)

Hemolysis 1 (3.12%)

RVAD right ventricular assist device.



5/9

related to the perioperative hypotension that she had.She also had immediate right ventricular failure re-quiring temporary RVAD support for 1 week afterHeartMate II placement with a significant pressorrequirement. Thus, the overall incidence of adverse

neurologic events was 2 of 32 (6.25%). Note that notransient ischemic attacks occurred in our group of 32HeartMate II patients. Also, none of the patients whodiscontinued warfarin experienced any thromboem-bolic events during HeartMate II support.

Infections

During the entire period of HeartMate II support, 4patients (12.5%) had driveline infections that requiredlong-term intravenous and oral antibiotic therapy. Staph-

ylococcus aureus was identified in 3 patients and Enter-obacterspecies in 1 patient. The mean duration to onset ofinfection was 50.5 days. The infections resolved withoutany evidence of recurrence in all 4 patients. There wereno LVAD-pocket infections in our experience.

Right Ventricular Failure

Severe right ventricular failure requiring placement of anRVAD occurred in 2 (6%) patients. Both patients requiredCentriMag Levitronix RVAD placement for immediateright ventricular failure (which occurred immediatelyafter HeartMate II placement). In both patients, theRVAD was explanted within 1 week after placement.Both patients survived more than 6 months; 1 is awaitinga heart transplant. The other patient survived almost 2years after LVAD placement, but was not transplanteligible owing to the development of postoperative para-

plegia. It should be noted that no additional patientsrequired prolonged inotropic support. Thus, the overallincidence of right ventricular failure (as defined byRVAD requirement or intropic support 14 days) was6%. An additional patient who required temporaryRVAD support was not included here as he had right

ventricular failure requiring RVAD support even beforeHeartMate II placement (discussed earlier).

Hemorrhagic Complications

Of the first 14 patients on the HeartMate II in our serieswho received heparin followed by warfarin, 3 had GIbleeding and 1 suffered delayed pericardial tamponade(4 of 14, 28.5%). Of the remaining 18 patients whoreceived only warfarin and no heparin, 2 had GI bleed-ing, and none suffered pericardial tamponade (2 of 18,11.1%). Note that this part of the analysis excludedpatients who underwent mediastinal exploration forbleeding on the day of surgery, because that was unre-lated to postoperative anticoagulation therapy. In addi-tion, 5 patients (15.6%) had postoperative bleeding re-quiring mediastinal reexploration, with no resultingmortality.

Device Thrombus

Of our 32 BTT patients, 1 had a suspected pump thrombusthat resolved with a high-intensity heparin anticoagulationprotocol (the INR was 1.3 at the time of presentation). Thispatient had gone to an outside hospital with HeartMate IIflows in the mid3 L/min range (in contrast to this patientsusual flows 4 L/min) and with increased HeartMate IIpower. After 12 hours of high-intensity anticoagulationheparin therapy (target PTT, 60 s to 80 s), this patients

hemodynamic indicators returned to normal. Heparin ther-apy was continued for 48 hours. Warfarin therapy wascontinued to achieve a goal INR of 1.5 to 2.0. No thrombo-lytic therapy was used. As of the end of this study, thispatient is doing well on the heart transplant waiting list,without any further similar episodes 6 months after thissuspected pump thrombus. Thus, our overall incidence ofdevice thrombus was 2 of 32 (6 %), including 1 explantedpump that received a thrombus score of 3, as reported inthe next section.

Explanted Pump Analysis

Fourteen patients in our study group whose HeartMate IIwas explanted (at the time of their heart transplant)underwent examination of their explanted pumps forthrombus deposition. All pumps except one received athrombus score of 0 (none) or 1 (minimal). The oneexplanted pump that received a high score of 3 ( 50%obstruction with thrombus) had been in the patient whodied after prolonged sepsis. He had several days of lowpump flow preceding his death, which may have beenassociated with the development of pump thrombus.

Comment

The discrepancy between the limited availability of do-nor organs and the ever-increasing number of patients

Fig 3. Actuarial survival (Kaplan-Meier analysis) of patients under-going HeartMate II placement as bridge-to-transplant therapy.

1231Ann Thorac Surg JOHN ET AL2008;86:122735 HEARTMATE II LVAD AS BRIDGE-TO-TRANSPLANT


6/9

with heart failure has led to the increasing use of LVADs[11, 12]. Further, the excellent medium-term results withLVADs has led to the use of permanent LVAD supportfor patients with end-stage heart failure [7]. The findingsfrom our current study extend our and others previousobservations that continuous-flow LVADs such as the

HeartMate II can be safely used as BTT therapy inpatients with end-stage heart failure [10, 13].

Recently, several single-center and multicenter studieshave shown improved outcomes with the newer contin-uous flow devices [10, 13, 14]. Frazier and colleagues [14]reported a 80% 1-year survival in a series of 43 patientsreceiving the HeartMate II both as BTT and destinationtherapy, with markedly improved functional status andquality of life. An actuarial survival of 89% at 1 monthand 75% at 6 months was seen in the HeartMate II BTTmulticenter study. The incidence of several adverseevents in the multicenter study, including LVAD-relatedinfections, need for RVAD support, and pump thrombo-sis, was similar to that seen in our experience. However,the incidence of other adverse events such as bleedingrequiring reexploration and stroke was higher than thatreported in this study. It is possible that some of thesedifferences in adverse events could be related to differ-ences in individual management practices among differ-ent centers, for example, anticoagulation strategy, as willbe discussed later. Nevertheless, an overall improvementin outcomes has been clearly demonstrated both with theHeartMate II as well as with other newer continuous flowdevices as compared with that obtained with pulsatiledevices. Esmore and coworkers [15] reported a 86.7%6-month survival in 30 patients receiving the VentrAssistLVAD in an international multicenter trial. These im-

provements are due primarily to the many advantages ofthese newer devices as we will discuss later in this article.However, improvement in the results with the newerdevices in the current era may also be secondarily relatedto lessons learned from earlier experiences (with pulsa-tile devices) that have led to stepwise and systematicimprovements in patient selection, better preoperativeoptimization, improved operative techniques, and betterpostoperative management such as improved optimiza-tion of right ventricular function in the postoperativeperiod.

The decreased morbidity associated with the HeartMate II,especially the lower incidence of postoperative bleeding anddevice-related infections, may be due to its smaller size,the lack of need for a large pocket to house its pump, andits smaller driveline. The absence of a large preperitonealpocket (which was required with the larger pulsatiledevices) has reduced the need for extensive dissection,reduced postoperative bleeding, and reduced LVADpocket hematomas and the development of pocket infec-tions. Device-related infections still occur with the Heart-Mate II LVAD, but they are often treatable with antibiotictherapy; they do not result in significantly delayed trans-plants or in pretransplant or posttransplant morbidityand mortality (unlike with the earlier generation ofLVADs). To further reduce and possibly even eliminateLVAD-related infections, we are currently evaluating

additional therapeutic modalities, such as the use ofroutine antibiotic-impregnated beads around the Heart-Mate II and the use of platelet gel application to itsdriveline exit site. We have preliminary evidence thatthis strategy has reduced the incidence of LVAD-relatedinfections [16].

Importantly, the smaller size of the HeartMate II hasmade it more applicable for women with heart failure. Inour study, female patients accounted for more than 30%of our BTT group. In contrast, in several studies in theliterature involving the larger pulsatile VE or XVE device,both as BTT and as destination therapy, fewer than 20%of the patients were female [1, 7, 17]. Clearly, the newgeneration of smaller continuous-flow devices, such asthe HeartMate II, will extend this life-saving technologyto a previously underserved population of female pa-tients with end-stage heart failure.

Again, the HeartMate II is a continuous flow, rotaryLVAD composed of a blood pump, a percutaneous lead,and an external power source and system driver. Theinlet and outlet cannulas include woven polyester grafts(Dacron; C.R. Bard, Haverhill, Pennsylvania) that requirepreclotting. The pump motor and associated bloodtube have smooth titanium surfaces; in an effort toduplicate the excellent biocompatibility of the originalpulsatile HeartMate XVE, the inlet and outlet elbows andthe intraventricular cannula are textured with titaniummicrosphere coatings [18]. The rationale for using tex-tured materials for the original HeartMate XVE was thatthey would absorb and entrap elements from the pa-tients blood to form a stable, densely adherent biologiclining on the inside of the device [19]. The resultant tissuelining, rather than the underlying biomaterials, would

then form the long-term blood-contacting interface, thusobviating the need for systemic anticoagulation. Thestandard strategy to reduce the risk of thromboembolismwith continuous-flow LVADs has been systemic antico-agulation therapy. Despite this strategy, the risk ofthromboembolism with those earlier LVADs has beenreported in other studies to be as high as 30% [20]. Therisk of bleeding, even with the HeartMate II, is exacer-bated with anticoagulation treatment. With increasingexperience in this study, we successfully reduced theintensity of anticoagulation in our HeartMate II patientswithout any increased risk of thromboembolic events.

Several previous studies identified multiple risk factorsfor poor heart transplant survival rates, including theneed for pretransplant mechanical circulatory support,pulmonary hypertension, a prior heart transplant, immu-nologic sensitization, and prolonged donor ischemictimes [21]. Outcomes after mechanical circulatory sup-port have vastly improved. Better immunomodulatoryregimens to treat sensitized patients are now available,yet pulmonary hypertension remains a relative contrain-dication to a heart transplant. The issue of pulmonaryhypertension assumes importance when evaluating theefficacy of continuous-flow devices. Previous studiesshowed a lesser degree of left ventricular unloading withcontinuous-flow (versus pulsatile) devices but a similardegree of pressure unloading under resting conditions



7/9

[22, 23]. Other endpoints (such as exercise performanceand cellular recovery) have been shown to be similar forthe two types of devices [2426]. However, concerns haveremained about the ability of partial unloading of the leftventricle to favorably influence altered pulmonary hemo-dynamics in end-stage heart failure patients. As a result

of this lack of definitive evidence (at least until recently)[27], concerns have lingered about the efficacy of circula-tory support provided by continuous-flow (versus pulsa-tile) devices. In particular, despite the well-documentedefficacy of pulsatile LVADs in ameliorating pulmonaryhypertension, it remains uncertain whether or not con-tinuous-flow devices similarly improve the alteredpulmonary hemodynamics of end-stage heart failurepatients [28]. However, recent reports using continuous-flow devices other than the HeartMate II have demon-strated their efficacy in ameliorating pulmonary hyper-tension [29].

The increased incidence of GI bleeding in HeartMate IIpatients is of concern. An increased incidence of GIbleeding with continuous flow devices was first reportedby Letsou and colleagues [30]. They observed a similarlyhigh incidence of GI bleeding of 14% in a series of heartfailure patients supported by the Jarvik 2000 axial-flowLVAD [30]. The cause of bleeding in all 3 of their patientswas due to arteriovenous (AV) malformations. WhetherGI bleeding is related to the need for anticoagulation orwhether it is linked to continuous flow effects or torelatively lower pulsatility is unclear. A similar physio-logic state that occurs with continuous-flow devices isalso seen in patients with aortic stenosis; such patients,like those implanted with continuous flow devices, expe-rience narrow pulse pressure. It is interesting that aortic

stenosis is also associated with GI bleeding [31, 32]. Aphenomenon of acquired von Willebrand disease maybe responsible for the occurrence of GI bleeding inpatients with aortic stenosis [33]. Whether this phenom-enon of acquired von Willebrand disease also occurs inpatients with continuous-flow devices remains to beseen. Our strategy with HeartMate II patients with GIbleeding has been to temporarily discontinue anticoagu-lation therapy as well as to reduce the pump flow speeds.The rational of reducing pump flow speeds is to increasethe pulse pressure in these patients, thereby reducing apotential risk factor for the occurrence of GI bleeding.Using this strategy, we have not had any recurrence of GIbleeding, and the discontinuation of anticoagulation hasnot resulted in the occurrence of thromboembolism.Clearly, further investigation is essential to definitivelycharacterize GI bleeding in all continuous flow devicepatients, including HeartMate II patients.

Our single-center study was limited by its relativelysmall number of patients. In addition, we did not have acomparison group of patients treated with pulsatile de-vices. However, all of our HeartMate II patients under-went LVAD placement over a relatively short period(approximately 2 years), so the surgical techniques andperioperative treatment protocols were consistent for thisgroup of HeartMate II patients (except for the changesthat we described earlier in the anticoagulation protocol).

In conclusion, the extremely low postoperative mortal-ity rate and the low incidence of adverse events makesthe HeartMate II LVAD an ideal device to be used as BTTtherapy. In addition, the HeartMate II LVAD has bene-fited from duplicating several features of the originalHeartMate XVE, especially in that both confer a low

thromboembolic risk. The favorably low thrombogenec-ity and low thromboembolic risk associated with theHeartMate II makes it ideal as destination therapy aswell. With an increasing focus on destination therapy asa real alternative for heart failure patients, the featuresand outcomes associated with the HeartMate II LVADdeserve attention [34]. Clearly, the increased incidence ofGI bleeding is of concern; strategies to optimize antico-agulation as well as understanding the exact relationshipbetween GI bleeding and continuous-flow devices willallow for improved outcomes in this population of end-stage heart failure patients.

We thank Mary E. Knatterud, PhD, for editorial assistance.

References

1. Frazier OH, Rose EA, Oz MC, et al. Multicenter clinicalevaluation of the HeartMate vented electric left ventricularassist system in patients awaiting heart transplantation.J Thorac Cardiovasc Surg 2001;122:118695.

2. Deng MC, Edwards LB, Hertz MI, et al. International Societyfor Heart and Lung Transplantation. Mechanical circulatorysupport device database of the International Society forHeart and Lung Transplantation: third annual report-2005.J Heart Lung Transplant 2005;24:1182.

3. Frazier OH, Rose EA, McCarthy P, et al. Improved mortality

and rehabilitation of transplant candidates treated with along-term implantable left ventricular assist system. AnnSurg 1995;222:32736.

4. Farrar DJ, Hill JD, Pennington DG, et al. Preoperative andpostoperative comparison of patients with univentricularand biventricular support with the Thoratec ventricularassist device as a bridge to heart transplantation. J ThoracCardiovasc Surg 1997;113:2029.

5. Portner PM, Jansen PG, Oyer PE, Wheldon DR, RamasamyN. Improved outcomes with an implantable left ventricularassist system: a multicenter study. Ann Thorac Surg 2001;71:2059.

6. Slaughter MS, Tsui SS, El-Banayosy A, et al. Results ofmulticenter clinical trial with the Thoratec implantable ven-tricular assist device. J Thorac Cardiovasc Surg 2007;133:157380.

7. Rose EA, Gelijns AC, Moskowitz AJ, et al. Long-term me-

chanical left ventricular assistance for end-stage heart fail-ure. N Engl J Med 2001;345:143543.

8. Griffith BP, Kormos RL, Borovetz HS, et al. HeartMate II leftventricular assist system: from concept to first clinical use.Ann Thorac Surg 2001;71:11620.

9. Frazier OH, Myers TJ, Westaby S, Gregoric ID. Clinicalexperience with an implantable, intracardiac, continuousflow circulatory support device: physiologic implicationsand their relationship to patient selection. Ann Thorac Surg2004;77:13342.

10. Miller LW, Pagani FD, Russell SD, et al. Use of a continuous-flow device in patients awaiting heart transplantation.N Engl J Med 2007;357:88596.

11. Aaronson KD, Eppinger MJ, Dyke DB, Wright S, Pagani FD.Left ventricular assist device therapy improves utilization ofdonor hearts. Ann Thorac Surg 2004;39:124754.



8/9

12. Morgan JA, John R, Rao V, et al. Bridging to transplant withthe HeartMate left ventricular assist device: the ColumbiaPresbyterian 12-year experience. J Thorac Cardiovasc Surg2004;127:130916.

13. Feller ED, Sorensen EN, Haddad M, et al. Clinical outcomesare similar in pulsatile and nonpulsatile left ventricularassist device recipients. Ann Thorac Surg 2007;83:10828.

14. Frazier OH, Gemmato C, Myers TJ, et al. Initial clinicalexperience with the HeartMate II axial-flow left ventricularassist device. Tex Heart Inst J 2007;34:27581.

15. Esmore D, Spratt P, Larbalestier R, et al. VentrAssist leftventricular assist device: clinical trial results and clinicaldevelopment plan update. Eur J Cardio Surg 2007;32:73544.

16. Jweied E, Joyce L, Liao K, John R. Strategies to reduce theincidence of LVAD-related infections. J Heart Lung Trans-plant 2008;27(Suppl):132.

17. Lietz KL, Long JW, Kfoury AG, et al. Outcomes of leftventricular assist device implantation as destination therapyin the post-REMATCH era: implications for patient selec-tion. Circulation 2007;116:497505.

18. Rose EA, Levin HR, Oz MC, et al. Artificial circulatorysupport with textured interior surfaces: a counterintuitiveapproach to minimize thromboembolism. Circulation 1994;90:II8791.

19. Graham TR, Dasse K, Coumbe A, et al. Neointimal devel-opment on textured biomaterial surfaces during clinical useof an implantable left ventricular assist device. Eur J Cardio-thoracic Surg 1990;4:18290.

20. Thoennissen NH, Schneider M, Allroggen A, et al. Highlevel of cerebral microembolization in patients supportedwith the DeBakey left ventricular assist device. J ThoracCardiovasc Surg 2005;130:115966.

21. John R, Rajasinghe H, Chen JM, et al. Impact of currentmanagement practices on early and late death in more than500 consecutive cardiac transplant recipients. Ann Surg2000;232:30211.

22. Klotz S, Deng MC, Stymann J, et al. Left ventricular pressureand volume unloading during pulsatile versus nonpulsatileleft ventricular assist device support. Ann Thorac Surg2004;77:14350.

23. Haft J, Armstrong W, Dyke DB, et al. Hemodynamic andexercise performance with pulsatile and continuous-flow leftventricular assist devices. Circulation 2007;116:I815.

24. Radovancevic B, Vrtovec B, de Kort E, Radovancevic R,Gregoric ID, Frazier OH. End-organ function in patients onlong-term circulatory support with continuous- or pulsatile-flow assist devices. J Heart Lung Transplant 2007;26:8258.

25. Amir O, Radovancevic B, Delgado RM, et al. Peripheralvascular reactivity in patients with pulsatile vs. axial flow leftventricular assist device support. J Heart Lung Transplant

2067;25:3914.26. Thohan V, Stetson SJ, Nagueh SF, et al. Cellular and hemo-

dynamic responses of failing myocardium to continuousflow mechanical circulatory support using the DeBakey-Noon left ventricular assist device: a comparative analysiswith pulsatile-type devices. J Heart Lung Transplant 2005;24:56675.

27. Garcia S, Kamdar F, Boyle A, et al. Effects of pulsatile- andcontinuous-flow left ventricular assist devices on leftventricular unloading. J Heart Lung Transplant 2008;27:2617.

28. Martin J, Siegenthaler MP, Friesewinkel O, et al. Implantableleft ventricular assist device for treatment of pulmonary hyper-tension in candidates for orthotopic heart transplantation - apreliminary study. Eur J Cardiothorac Surg 2004;25:9717.

29. Etz CD, Welp HA, Tjan TD, et al. Medically refractory

pulmonary hypertension: treatment with nonpulsatile leftventricular assist devices. Ann Thorac Surg 2007;83:1697706.

30. Letsou GV, Shah N, Gregoric ID, Myers TJ, Delgado R,Frazier OH. Gastrointestinal bleeding from arteriovenousmalformations in patients supported by the Jarvik 2000axial-flow left ventricular assist device. J Heart Lung Trans-plant 2005;24:1059.

31. Heyde EC. Gastrointestinal bleeding in aortic stenosis [let-ter]. N Engl J Med 1958;259:196.

32. Cody MC, ODonovan PB, Hughes RW Jr. Idiopathic gastro-intestinal bleeding and aortic stenosis. Am J Dig Dis 1974;19:3938.

33. Warkentin TE, Moore JC, Morgan DG. Aortic stenosis andbleeding gastrointestinal angiodysplasia: is acquired vonWillebrands disease the link? Lancet 1992;340:357.

34. Lietz KL, Miller LW. Will left-ventricular assist device ther-apy replace heart transplantation in the foreseeable future?Current Op Cardiol 2005;20:132.

DISCUSSION

DR MICHAEL A. ACKER (Philadelphia, PA): I want to thank

and congratulate Dr John and his colleagues from the University

of Minnesota for this very important paper. I think it is veryimportant for not only this audience but also the audience of

heart failure cardiologists out there to realize how phenomenal

these results are and how it probably represents a real stepforward over our first-generation pulsatile devices. Importantly,

these results that Dr John showed at the University of Minnesota

have largely been reproduced, though not quite as good, in therecently completed prospective multicenter trial of HeartMate II

for bridge to transplantation. The FDA panel has approved this

device, and it is expected this approval is forthcoming for bridgeto transplantation. I might note that of all new LVADs in trial

right now, they are all continuous flow. There are approximatelyfive or six in human trial, and what is exciting is that the early

results for each of these devices mirror the excellent results that

we are seeing with HeartMate II.This is a comparison of the HeartMate II pivotal trial as a

bridge to transplantation, to the HeartMate XVE, which is theonly approved device that has saved many patients. The dra-

matic decrease in adverse events is very striking: Bleeding

requiring reoperative surgery is down by a half; most impor-

tantly, percutaneous lead infections is about a tenth, and I might

add that in Dr Johns series, there was no pocket infection; stroke

was decreased by half; and RV failure was also decreased.Importantly, though, there was a device malfunction in this

group the rate of device malfunction is dramatically less than

seen with HeartMate I.Despite these excellent results, questions linger on continuous

flow devices and the long-term effects of decreased pulsatility,

such as why is GI bleeding increased? Is LV unloading sufficient,especially when one compares it with pulsatile devices? And do

we still need an automatic flow algorithm for automatic control?

I have several questions. The first question is this: continuousflow devices are believed by many not to unload the left

ventricle as well as pulsatile devices and as such they will chooseHeartMate I for patients with severe pulmonary hypertension

maybe who are not transplant candidates to promote resolution

of pulmonary hypertension in later transplant. Do you believethis is true, and if so, is there a group of patients where

HeartMate I is still indicated?Second question: HeartMate I is associated in some centers

with a high incidence of sensitization and high PRA, which can

delay or preclude safe transplantation. Have you seen this or



9/9

other study centers seen this in HeartMate II, and if not, whynot?

Third question: do you feel there is a need for an algorithmthat will allow automatic adjustment of flows both to maximizeflows and avoid suckdown for any given situation?

Fourth question: can you conjecture on the long-term effectsof decreased pulsatile pressure such as the GI bleeding inci-

dence you saw?And finally, and most importantly, have these excellent results

been reproduced in your older destination patients, and if so,can you conjecture on whether HeartMate II should be movedinto less sick patients and whether you will be able to convinceheart failure cardiologists to refer these patients so as to improvetheir quality of life?

I want to thank the Association for the privilege of discussingthis important paper.

DR JOHN: Thank you, Dr Acker, for those kind comments. Inresponse to the first question as to whether the HeartMate I is stillindicated, I would like to make this remark. In this current era ofincreasing use of continuous flow devices, there have been threegroups of patients who are believed to still benefit from theHeartMate I. One is the very large patient, the second group is thepatient who cannot tolerate anticoagulation, and the third waswhat Dr Acker alluded to, patients with severe pulmonaryhypertension. With increasing experience, we and others haveshown that the HeartMate II can be safely used in all thesegroups. We have shown that patients with the HeartMate II cantolerate long periods without anticoagulation, without experi-encing an increased incidence of thromboembolic events. Wehave also seen that in the very large patient as well as the patientwith severe pulmonary hypertension that the HeartMate IIsatisfactorily reverses the adverse pulmonary hemodynamics,making them eligible for transplantation. The one group inwhich the HeartMate XVE may be superior is in the very sickcohort of patients and pulsatile flow may be better for this

extremely sick group of patients with multisystem organ failure.The lower rates of allosensitization that are being seen withthe HeartMate II may reflect the current era in which we havelearned to significantly reduce the incidence of bleeding, both byimproved surgical techniques as well as better preoperativeoptimization. The smaller size of the current HeartMate II device

reduces the need for extensive dissection as the pump pocketsize is significantly smaller. As a result, these patients receivesignificantly less blood and blood products, which has contrib-uted to the lower levels of sensitization in this era.

The third question was regarding the suckdown effect. Ifyou do increase the speeds in the HeartMate II to very highlevels, especially in the face of inadequate preload, these pa-

tients will experience a suckdown effect on the ventricle. Thiscan precipitate ventricular arrhythmias and even death. It maybe favorable to incorporate an algorithm in continuous flowpumps to prevent the suckdown effect that may be seen withcontinuous flow pumps. In the absence of that, it should bestressed that it is important to balance speed at which the pumpis run to achieve flows that are adequate for the patients bodysize and to maintain adequate end-organ perfusion (in additionto optimizing preload).

I think the effects of long-term reduced pulsatile pressurehave not translated into adverse end-organ function. We haveshown reversal of end-organ dysfunction, both renal and he-patic function, in patients supported with these pumps. Therehave been several other reports in the literature that have shownresolution and maintenance of end organ perfusion with con-tinuous flow devices. Clearly, the increased incidence of GIbleeding with continuous flow devices is of concern. Is it thereduced pulse pressure or the need for anticoagulation that iscontributing to the bleeding in these patients remains to beseen? An increased incidence of GI bleeding is also seen patientswith aortic stenosis, a condition that shares the low pulsepressure state with patients with continuous flow devices. Weare studying this phenomenon in greater detail to identify riskfactors for GI bleeding in patients with continuous flow devices.

I certainly agree with Dr Acker that these results and thoseduplicated by many other centers have set the stage for the useof these devices in less sick patients. Data from the DT trial arestill ongoing, although anecdotally what we are seeing are betterresults in patients using the HeartMate II as compared with the

HeartMate XVE. These improved results not only primarilyreflect the fact that we have a better pump in the HeartMate II,but secondarily, also reflect on lessons that we have learnedfrom years of experience with the HeartMate XVE as well asother devices while caring for patients with end-stage heartfailure requiring circulatory support. Thank you.


novembro2008_transplante

Documents