The Journal of Heart and Lung Transplantation Volume 18, Number 1 114 115 resolution of non-ischemic heart failure at 12 medical centers, All patients were in imminent risk of dying and were selected for VAD support using standard bridge-to- transplant requirements. There were 8 females and 9 males with an average age of 31 (range: 12 to 60). The etiologies were 6 with myocarditis, and 11 with cardiomyopathi~ [2 post-partum (PPCM), 4 viral (VCM), and 5 idiopathic (IDCM)]. BVADs were used in 10 patients and isolated LVADs in 7 patients, for an average duration of 44 days (range 11 - 97 days), before return of ventricular function and successful weaning from the device. Results Four patients (2 IDCM, 1 VCM and 1 myocarditis) died in the first 7 mon=t-VAD removal. The remaining 13 , patients are alive after an average of 2.0 years (range: 4 months to 8 years). ‘Ibe actuarial survival post-VAD support atIer weaning is 75% at 1 and at 5 years, with 7 patients with more than 1 year survival and 2 with more than 5 years survival. Only 1 patient required a heart transplant, a female with PPCM successfully weaned from VAD support after 97 days, who underwent transplantation 1 year later. Conclusion: These results support the concept that ventricular assist devices can be used to promote or allow the recovery of certain forms of heart disease previously only treatable by heart transplantation, with encouraging long-term survival. MECHANICAL UNLOADING RESTORES BETA-ADRRNERGIC RESPONSIVENESS IN THE FAILING HUMAN HEART. ML Ogletree- Hughes, L Barrett-Stull, NG Smedira, PM McCarthy and CS Moravec. Center for Anesthesiology Research and Department of Thoracic and Cardiovascular Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio A hallmark of the end-stage failing human heart is a decrease in the inotropic response of ventricular muscle to stimulation by g-adrenergic agonists along with changes in 8-adrenergic receptor (P-AR) density. We and others have previously shown a reduction in the isoproterenol response of isolated contracting trabecular muscles removed from failing human hearts when compared to nonfailing muscles. Further, this difference has been associated with a reduction in P-AR density in failing myocardium. The question of whether this characteristic decrease in 8-adrenergic responsiveness and receptor density can be reversed remains unanswered. We hypothesized that mechanical unloading of the left ventricle with a left ventricular assist device (LVAD) would normalize B-AR density and increase the ability of cardiac muscle to respond to g-adrenergic stimulation. Trabecular muscles were removed from the let? ventricle of three groups of patients: I) end- stage heart failure patients at the time of transplantation; 2) end-stage heart failure patients at the time of transplantation who had been bridged with an LVAD for a variable duration of time (3-12 weeks); and 3) non-failing organ donors who could not be matched for transplantation. In muscles from non-failing hearts, the increase in developed tension following a maximal dose of isoproterenol averaged 264.5 +!- 55% of baseline tension. In muscles from end-stage failing human hearts, the increase in developed tension averaged 71.3 +I- 19% of baseline tension. In muscles removed from failing hearts which had been bridged by an LVAD, the increase in developed tension averaged 148.1 +/- 37% of baseline tension. Thus, although the muscles removed from hearts which had been mechanically unloaded did not respond as well as muscles from non-failing hearts, they clearly exhibited an increased response to isoproterenol as compared to muscles taken from failing hearts which had not been bridged. P-AR density increased by 50% in patients supported by LVAD prior to transplant as compared to patients not supported by LVAD. Together, these data suggest that mechanical unloading of the heart can improve the contractile response of cardiac muscle to stimulation via the P-AR system, reversing one of the major functional changes associated with end-stage heart failure. BRIDGE-TO-TRANSPLANT (BTT) EXPERIENCE WITH THE NOVACOR LEFT VENTRlCULAR ASSIST SYSTEM (LVAS): RESULTS OF A MULTICENTER US STUDY R.L. Kormos’. N. Ramasamy’. S. Sit’. A.D. Cleeland’. J.S. Jassawala‘. and P.M. Portnerc, For US Multicenter Investigators. ’ University of Pittsburgh Medical Center, 2 Baxter Healthcare From March 1996 to June 1998 one hundred tifty six recipients at 22 US centers received Novacor NIOOPC LVAS implants as a BTT. Average implant duration was 97d (longest 780+ days: cumulative 41 yrs). Concurrently, 35 Status I patients waiting for transplant (ts) at eight centers were controls. One hundred twenty nine LVAS recipients satisfied all entq criteria (core group). and 14 are still on LVAS support. Of the II5 who completed the study. 87 (76%) were transplanted compared to I3 (37%) controls (p<O.OOS). Intercenter variability of complications may suggest management or patient selection differences. The linearized rate for total complications decreased by more than an order of magnitude with support duration (13.9 /pt mo 0-2d. 4.8 /pt mo 3-30 d. 0.49 /pt mo. >lEOd). Six (4%) recipients required transient mechanical circulatory support of the right ventricle (RVAD). One device was replaced (without impact on outcome) at 49ld secondat) to a stress-related external cable fault. Forty-five (29”/.) 116 117 Abstracts 63 recipients were discharged ulth their implant, the longest for >I yr. A total of 4 patients have been supported at home for over 6 months. Home discharged recipients had a higher tx rate (88% vs 66% in hospital). Linearized complication rates were significantly lOWr for home discharged recipients than those remaining in hospital. The results demonstrate efficacy of univentricular Novacor (LVAS) support as a bridge to transplant. Hotne discharge experience provides an improved quality of life with LVAS support and highlights device safety and reliability for long-term use. DOES LVAD BRIDGINGTO-TRANSPLANTATION INCREASE GRAFT VASCULOPATHY? RJ Kaplan MD, SR Kqadia, NG bedim MD, T Buda RN, M @ormastic MIJH, EMTuzcuMD,PMMcCarthy?4D. WehaveprevMudyshowntharuccesafulIdtvmtriarlararriatdevice(LVAD) bridgingdoesnotincxase the-ofpost-tmmplMtvasallarrejeetial; however, SevenI cent= report that LVADS incmnse pratmn@antT-cdlPRA levels. As early as 1 year postamnspbt, at least 50% ofptialts (pa) have dembnstrablcgraRvaulqmthybyintmvaaadar uhmsmmd (IWS); baaed on this, wewueamcunalthatLVADqprtmi@tinaxase the development of vanrplantcomauyartaydiseure(CAD).WereviewedtheIWSfindingsfromthc unonary~~oflO5~~ps,l7ofwhich~bridged-to- ttmqhtion with M LVAD. Mean ages of mcipians were s+niIar tbr both the bridged and mm-bidgal patients, as were the donor ages for both groups; similatiy, therewasnodl&reuceidorreclpientgenderincidahxforeitltergmup. Maximalplaquethi&tmaa(Pumx)atbaaegneandatlyearwerethesameinboth @cqs(O.27nm1?O.26vs. O.U~.lS, p-NS;O.38mn@.28vs. 0.3lm.22. p=Ns)~wasthea~~inRnurduringthttime(O.l1~.l9vs. 0.08+0.09, p= NS). 47.1% of pts h&gal with an LVAD received donor hearts with CAD already present (PmaxX.5mm at basdine) as did 46.6% of pts not bridged (p=NS).Nnuonr*TCAD(RnaxQ).5rt~ud>o.5atlyar)occumdin 35.3%ofbridgedad4%of~psbylyerrpost-tnnrpkntation (p=NS), and prqmaion of dlaaae @at&m Pmax 3.5, inaeaae inPmaxatlyear 9.3) occurmd in 11.8% and 19.3% ofpta, mapactively @=NS). IWS evahmtion of corOnaryartsysqgnmtSlikCWi~-llO-baw&nt&~pj withngrrdtothe~ofdonorCADortheinci~ofnnuTCAD. MUhiMMtCloBirticrqparion~srhowtdrbrcnceofanLVADbridgctobca risktactorforthe dedqmnt ofTCAD (p<O.O3). Futther eval&hm is naxmaty tobetterunQssundtheet?kctofLVADaonthedm&pmentofgmitvascuk@hy, howeva,it~tht,despitetheiri~ogic~LVADsdonatincrease the incideaoe of TCAD. IS THE USE OF LEFT VENTRICULAR ASSIST DEVICE AS A BRIDGE TO HEART TRANSPLANTATION A RISK FACTOR FOR DEVELOPMENT OF TRANSPLANT CORONARY ARTERY DISEASE? B. Radovancevic, A. Golino, C. Thomas, P. Odegaard, R. Radovancevic, O.H. Frazier. Texas Heart Institute/St. Luke’s Episcopal Hospital, Houston, Texas. Immunologic (rejections), metabolic (insulin resistance, hyperlipidemia) and cytomegalovirus (CMV) infection have been identified as risks for transplant coronary artery disease (TxCAD). The apparent periopemtive bleeding diathesis with the increased transfusion rates that accompanies implantation of a left ventricular assist device (LVAD) may adversely affect the immune system. Thus, we assessed whether using an LVAD as a bridge to heart transplautation presents an increased risk for TxCAD. We reviewed the patient data of 16 consecutive patients (all were men) who had been bridged to transplant with an LVAD (mean, 120t65 days) and survived more than 3 years atIer transplaut. The control group included 46 men (UNOS status I), who underwent transplantation alone over the same period. The presence of T&AD for both groups was defined as any change on coronary angiography. There were no significant differences between the two groups regarding: etiology, age, donor age and sex, panel reactive antibody testing, crossmatching, pre- and posttrausplantation cholesterol levels, incidences of diabetes mellitus or treated hypertension, number of infections, use of calcium chmel blockers, and number or severity of rejection episodes. Significant differences between the bridged patients and control group patients were found in ischemic time (2lOt61 mm. vs. 165+61 min [p=O.O2], CMV infection (69% [I l/16] vs. 17% [S/46] [p-0.0001]), and transfusion incidence before transplant (100% vs. 37% [17/46][p<O.O01]). The incidence of TxCAD was 31% (5116) in the bridged patients and 13% (6/46) in the control group, but the difference was not significant (p=O.O93). It appears that the use of an LVAD as bridge to transplantation does not present as an increased risk for TxCAD, despite significant differences in predisposing factors such as ischemic time and CMV infection. There was, however, a trend toward significance, which might have been stronger had the groups been larger. Because of this trend and the risk associated with TxCAD, aggressive prophylactic therapy may be warranted in this group of patients.