471 The Versatility and Outcomes for Patients with TandemHeart® Temporary Ventricular Assist Device Support

S160

The Journal of Heart and Lung Transplantation, Vol 30, No 4S, April 2011

Conclusions: Implantation of an LVAD in pts with high risk for RVF is a feasible and effective therapy option if RV function is closely monitored. Once RVF is identified, immediate temporary RVAD placement improves outcome. Weaning from and removal of the RVAD allows pts to continue with LVAD support alone, thus providing good quality of life and longterm mechanical support. 469 The Natural History of Tricuspid Valve and Right Ventricular Function Following Left Ventricular Assist Device Implantation J.M. Stulak, J. Cowger, J.W. Haft, M.A. Romano, K.D. Aaronson, F.D. Pagani. Cardiac Surgery, University of Michigan Health System, Ann Arbor, MI. Purpose: Regression of functional tricuspid valve regurgitation (TR) and improvement in right ventricular dysfunction (RVD) has been shown to improve after corection of left-sided valvular pathology. There are few data examining this natural history in pt after VAD implant. Methods and Materials: Between Oct. 1996 and Oct. 2010, 149 pt underwent VAD implant without concomitant tricuspid valve repair and had echocardiographic and right-heart catheterization follow-up available. There were 128 males (86%) with a median age of 55 yr (range, 17-78 yr). Preop TR was ⬍ mild in 141 pt (95%) and RVD was ⬎ mild in 114 pt (77%). Mean preop central venous pressure (CVP) was 10⫾2.4 mmHg, mean pulmonary artery pressure (MPAP) was 33⫾3.1 mmHg, pulmonary vascular resistance (PVR) was 2.5⫾1.3 WU, and wedge pressure (WP) was 22⫾2.8 mmHg. Results: After VAD implant (median follow-up 7.4 months), TR progression occurred in 72 pt (48%); TR progressed 1 grade (0.9 to 1.9, p⬍0.001) in 52 pt (35%) and ⬎ 2 grades (0.6 to 2.8, p⬍0.001) in 20 pt (13%). Overall, postop RVD remained stable compared to preop (grade 2.3 vs. 2, p⫽0.09). RVD was also unchanged in pt with TR progression (2.1 vs. 2.2, p⫽0.84), while reduction in RVD was seen in pt without TR progression (2.4 vs. 1.7, p⬍0.001). While postop MPAP and PVR were similar, compared to pt with no or 1 grade TR progression, postop CVP (7⫾2 vs. 7⫾2.2 vs. 12⫾2.3, p⫽0.014) and WP (11.7⫾2.5 and 11.5⫾2.5 vs. 14.8⫾2.6, p⫽0.046) were significantly higher in pt with ⬎2 grade TR progression. Overall, there was no difference in hospital readmission (66% vs. 60%. vs. 70%, p⫽0.36) or survival to transplant (82% vs. 92% vs. 88%, p⫽0.6), respectively. Conclusions: In this select pt population, progression of TR was observed in almost half of pt, but a minority had TR progression ⬎1 grade. RVD remained stable in the face of TR progression and significantly improved in pt who had no TR progression. Furthermore, TR progression after VAD implant did not portend increased hospital admission or lower survival to transplant. 470 Implantable LVADs as Initial Therapy for Refractory Post Myocardial Infarction Cardiogenic Shock – A Challenge to the ‘Bridge to Bridge’ Paradigm A. Pawale, S. Pinney, K. Ashley, R. Flynn, F. Milla, A.C. Anyanwu. Department of Cardiothoracic Surgery, Mount Sinai Medical Center, New York, NY. Purpose: Initial therapy for refractory cardiogenic shock has largely been based on use of short-term mechanical devices with later conversion to durable options. The premise is that such patients cannot tolerate cardiopulmonary bypass (CPB) and extended surgery needed for implantable left ventricular assist device (LVAD) placement. We have adopted an alternate strategy to implant long term LVADs as initial device therapy in such patients. Methods and Materials: Over last 3-years, we used implantable LVADs (Jarvik 2000, 1; Ventrassist, 1; HM XVE, 2; HM II, 9) in 13 patients (M:F 11:2; mean age 54 y) with post myocardial infarction (MI) shock without prior use of short term LVAD. Median time interval from MI to LVAD was 3.5 days. Eight patients were on a ventilator, 2 had unknown neurological status and 4 had suffered cardiac arrest in preceding 24 hrs. Two had prior

CABG. Nine had received potent antiplatelet agents post-MI. Mean lab values include: creatinine 1.5 mg/dl, ALT 748 U/l, INR 1.5, lactate 3.2 mmol/l. One was done off pump; for others mean CPB time was 72 min. RVADs were used in 2 cases and were later explanted. Results: One patient died of progressive multiorgan failure. All others survived to hospital discharge. There were no re-explorations for bleeding or major infective complications; 2 patients had perioperative stroke. Median duration of mechanical ventilation, ICU stay, and hospital stay were 3, 9 and 18 days respectively. Of the 12 survivors, 7 have since had heart transplant, 1 patient underwent device explant, 3 remain alive on support and 1 died 7 months post-LVAD. Conclusions: Our data challenge the notion that patients in refractory cardiogenic shock are too ill to tolerate immediate placement of implantable LVADs. Despite surgical challenges, a one-stop implantable LVAD approach for cardiogenic shock is feasible and safe, and may offer unique advantage over the bridge to bridge approach as it avoids incremental costs, hospitalization and morbidity associated with repeat interventions. 471 The Versatility and Outcomes for Patients with TandemHeart® Temporary Ventricular Assist Device Support S.P. McCandless, J.R. Revenaugh, A.G. Kfoury, B.B. Reid, D. Budge, J.R. Doty, S. Stoker, S.E. Clayson, R.A. Merchel, H.K. Smith, I.D. Ledford, M. Goddard, P.A. Ventura, W.T. Caine, R. Alharethi. Utah Artificial Heart Program, Intermountain Healthcare, Murray, UT. Purpose: The TandemHeart® temporary ventricular assist device (TVAD) has traditionally been used as support for patients undergoing high risk cardiovascular interventions and experiencing cardiogenic shock. Our goal in this study was to evaluate and weigh the outcomes for patients who underwent TVAD implantation at our institution. Methods and Materials: The Utah Artificial Heart Program database was queried for information regarding patients who were implanted with TVAD between 2002-2010. Patients were retrospectively analyzed for cardiovascular interventions: Percutaneous Coronary Intervention (PCI), ventricular tachycardia (VT) ablations (VTA), and circulatory support only. Clinical presentation data which includes elective PCI, cardiogenic shock (CS), acute myocardial infarction (AMI) without CS, and VT was compiled. Survival at discharge, one month, six months, and one year was performed by a Kaplan-Meier survival analysis. Results: Forty-eight patients qualified for the study. Clinical presentations were AMI without CS (19%), CS (46%), VT (21%), and planned elective PCI (14%). Patients’ demographics, characteristics, and outcomes are below. Table 1

●●●

Demographics (ⴞ) Age (years) Support Time (hrs) Implant-Discharge (days) Pre-Implant EF CI Survival (%) Discharge 1 month 6 months 1 year

All (n⫽48)

Support Only PCI (n⫽21) VTA (n⫽10) (n⫽18)

58 ⫾ 17 59 ⫾ 72 19 ⫾ 17

65 ⫾ 13 39 ⫾ 53 16 ⫾ 12

59 ⫾ 15 4⫾2 6⫾7

51 ⫾ 20 107 ⫾ 80 35 ⫾ 19

24 ⫾ 10 24 ⫾ 8 2.4 ⫾ 0.8 2.3 ⫾ 0.7

29 ⫾ 12 N/A⌿

22 ⫾ 10 2.2 ⫾ 0.6

77 76 71 69

100 100 100 100

56 59␤ 53 53

85 81 76 71

EF: Ejection Fraction, CI: Cardiac Index (L/min/m2). ⌿

Not enough data was obtained to report.

␤

Patients still in hospital at 1 month.

Conclusions: In this study, the TVAD support allowed for improved survival in a population of patients with excessively high expected mortality and proved reliable and effective in a variety of clinical situations. Further,

Abstracts it is encouraging that the observed early survival benefit appears to be extended.

472 Mechanical Circulatory Support in Restrictive Cardiomyopathy F. Hennig, A. Stepanenko, T. Krabatsch, E.V. Potapov, R. Hetzer. Cardiothoracic and Vascular Surgery, Deutsches Herzzentrum Berlin, Berlin, Germany. Purpose: Restrictive cardiomyopathy (RCM) is a rare disease leading to terminal heart failure and consequently need for mechanical circulatory support (MCS). Currently little information is available concerning postoperative results or the influence of cannulation technique on survival. Methods and Materials: A retrospective analysis was performed of patients who received MCS for RCM. Results: Between January 1986 and November 2010 1489 patients received ventricular assist devices at our institution. Of these, 17 patients (10 female, age range 12-70 years) suffered from RCM. At the time of presentation four patients suffered from amyloidosis; in one patient a desmin mutation was found. In two patients RCM was most probably of toxic origin, and one patient had status post myocarditis. In nine patients no underlying disease was identified and RCM was considered to be idiopathic in origin. Nine patients received biventricular and eight left ventricular MCS. Left ventricular drainage was achieved by apical cannulation in nine patients; left atrial cannulation was performed in eight patients. The following MCS devices were implanted: BerlinHeart Excor in 10 patients, HeartMate II in two, HeartWare in two, BerlinHeart Incor in two and MicoMed DeBakey in one. Duration of support ranged from 2 to 553 days. Thirty-day survival was 53% and 100-day survival 38%. One-year survival was 19%. For patients with atrial cannulation 30-day survival was 50%; for those with apical cannulation it was 56%. In patients with an LVEDD ⬍50mm, 30-day survival was 50% with atrial cannulation and 50% with apical cannulation. Thirty-day survival was better in patients with biventricular MCS than in those with isolated left ventricular support (67% vs. 38%; p⫽0.03). Two patients remained on MCS until heart transplantation, one is still supported and 14 died. Conclusions: No evidence was found for superior performance of atrial cannulation as compared to apical cannulation for left ventricular drainage. Patients with biventricular support showed better 30-day survival than those with left ventricular support only. 473 LVAD Therapy in Patients with Restrictive and Hypertrophic Cardiomyopathy Y. Topilsky,1 N.L. Pereira,1 D.K. Shah,2 B. Boilson,1 J.A. Schirger,1 S.S. Kushwaha,1 S.J. Park.2 1Cardiology, Mayo Clinic, Rochester, MN; 2 Cardiovascular Surgery, Mayo Clinic, Rochester, MN. Purpose: Left ventricular assist device (LVAD) is being increasingly used in patients with end stage dilated and ischemic cardiomyopathy. There have been no clinical trials addressing the use of LVAD therapy in patients with end stage heart failure due to restrictive (RCM) or hypertrophic cardiomyopathy (HCM). The purpose of this study was therefore to analyze the outcome of LVAD therapy in these patients. Methods and Materials: 83 patients received continuous axial flow LVAD (Heart mate II) at our institution. We analyzed the baseline characteristics, surgical and long term impact of LVAD therapy in 8 patients with restrictive cardiomyopathy(RCM) or hypertrophic cardiomyopathy(HCM), and compared their outcome to the remaining 75 patients. Results: Patients with RCM and HCM have significantly smaller left ventricular end diastolic dimensions (52.5⫾6mm vs. 68.6⫾8mm; p⬍0.0001), and higher ejection fraction (33.5⫾17.9% vs. 19.3⫾6.9%; p⫽0.0003). We found no difference in early mortality (12.5% vs. 9.3%, p⫽0.8), or length of hospital stay (22.4⫾20.5 days vs. 21.1⫾12.6 days; p⫽0.8) between the two groups. The duration of inotropic support was significantly prolonged (452.8⫾501hours vs. 172.1⫾227 hours; p⫽0.006), right atrial pressure post surgery was higher (16.8⫾6.2mmHg vs. 12.3⫾5.0mmHg, p⫽0.03), pump flow was lower (4.3⫾0.5 liter vs. 5.2⫾0.6 liter, p⫽0.0004), and infections were more common in patients with RCM and HCM (87.5% vs. 44.5%, p⫽0.01) but bleeding events were less

S161 frequent (37.5% vs. 71.6%, p⫽0.05). The one-year Actuarial survival rate was not different between the groups (80.0⫾17.8% vs. 77.0⫾5.8%; p⫽0.9). Conclusions: Patients with end stage heart failure due to RCM or HCM can be treated with continuous axial flow LVAD. The hospital length of stay and mortality is comparable with patients who have dilated or ischemic cardiomyopathy, but infections, right heart failure and prolonged inotropic use are more common in patients with RCM and HCM. Clinical trials would be required in a larger number of patients to recommend this therapy for these patients. 474 Improvements in Neurocognitive Function Are Delayed after LVAD Implantation R.J. Petrucci,1 B.L. Carrow,2 A. Lagore,3 C. Cassano,4 E.J. Gracely,5 J.W. Entwistle.4 1Psychiatry, Drexel University College of Medicine, Philadelphia, PA; 2Thoratec Corporation, Pleasanton, CA; 3LaSalle University, Philadelphia, PA; 4Cardiothoracic Surgery, Drexel University College of Medicine, Philadelphia, PA; 5Family, Community and Peventive Medicine, Drexel University College of Medicine, Philadelphia, PA. Purpose: Neurocognitive (NC) function improves in selected domains from month 1 to month 6 after LVAD implantation.1 Whether early improvements occur is unclear because differences between pre- and postimplant measures have not been assessed. Methods and Materials: Pre- and postoperative NC evaluations were conducted prospectively in 29 LVAD patients at 1 center. The group was 72% male with a mean age of 51 (23-76) years. Forty-one percent were ischemic and 45% had a prior stroke. Eight standardized tests representing 6 domains (visual-spatial perception, auditory and visual memory, executive functions, language, processing speed, motor speed/grip strength) were done at baseline and 1, 3 and 6 months postimplant with 29, 28, 24 and 16 patients available, respectively, at each time point. Results: No statistically significant improvements were seen in NC function between baseline and 1 month. At 3 months improvements were seen in immediate auditory memory and processing speed (each with p⬍.001), visual memory delay, visual-spatial perception, upper extremity motor speed (each with p⬍.01), auditory memory delay and executive functions (each with p⬍.05). The same 5 domains showed continued improvement at 6 months compared to baseline; only the language domain showed no improvement over time. While grip strength declined at 1 month, it recovered to baseline by 3 months. No other significant decline was seen in any cognitive domain over this period. Conclusions: Although the cognitive performance of heart failure patients receiving LVAD support showed no improvement from baseline to 1 month, selective improvement was seen at 3 months in visual-spatial perception, processing speed, auditory and visual memory, executive functions and motor speed. While extended hypoperfusion and deconditioning may explain poor baseline NC performance, improved perfusion does not lead to rapid improvement in NC function. Earlier implantation may help to minimize preoperative NC decline and result in better postoperative NC function. 1. Petrucci et al. J Heart Lung Transplant 2009;28:542. 475 Ventricular Assist Device Related Infections Increase One-Year Mortality S.V. Pamboukian,1 T. Saavedra,2 J.A. Tallaj,1 R.N. Brown,3 J.F. George,3 M. Patel,2 W.L. Holman,3 J.K. Kirklin,3 J. Baddley.2 1 Cardiovascular Disease, University of Alabama at Birmingham, Birmingham, AL; 2Infectious Disease, University of Alabama at Birmingham, Birmingham, AL; 3Cardiothoracic Surgery, University of Alabama at Birmingham, Birmingham, AL. Purpose: The impact of left ventricular assist device (LVAD) related infections on outcome is not clearly defined and poses management challenges. Methods and Materials: Data were collected regarding the occurrence of infections and mortality among patients who received an LVAD from 2004 to 2009. Statistical comparisons were performed using Kaplan Meier analysis. Results: 87 patients (68 males; mean age 52 years) underwent LVAD implant with the majority receiving HeartMate II (44%) or HeartMate XVE (36%). Of these, 31 died, 21 were transplanted and 3 were explanted. The rest remained on