S146

Abstracts

Infection remains a major problem following insertion of Left Ventricular Assist Device(LVAD)s. Driveline and pocket infections can lead to multiple readmissions for antibiotics, morbidity and even mortality. Continuous flow pumps have smaller more flexible drivelines and are generally less mobile than previous pumps (due to lack of pulsatility) and hence may be associated with a low rate of driveline infections. In an attempt to clarify this we retrospectively analysed the frequency and impact of driveline and pocket infections in 20 patients who have received the Jarvik 2000 axial flow pump since May 2003 at our institution cumulating in a total of 5052 days experience. Seventeen were males and 3 were female, mean age 40.6yrs, 14 had dilated cardiomyopathy, 3 post partum cardiomyopathy, 2 ischaemic heart disease and 1 transplant allograft failure. In 17 patients implantation was through a median sternotomy and in 3 through a lateral thoracotomy, all had an abdominal driveline. Our institution dressing protocol was used, consisting of strict aseptic technique and Allevyn™ dressings. Twelve of the patients were discharged home. No pocket infections occurred. Fifteen driveline infections occurred in 10 of the patients. Seven patients had 1 infection, 2 had 2 infections and 1 had 4 infections. The first infection occurred at 133.5 (17–299) days post-operatively. None required readmission or IV antibiotics for driveline infections. Of the 15 infections, the organism cultured was Staphylococcus aureus (6), Pseudomonas aeruginosa (2), Coliform (3), Enterococcus faecalis (1), Serratia marcesens (1), mixed anaerobes (1) and Candida (1). In conclusion, the abdominal cable Jarvik 2000 axial flow pump was associated with no pocket infections and a low rate of driveline infections of which none required readmission and all could be treated with oral antibiotics. This may have implications for the cost effectiveness and longer term use of these devices. 295 END-ORGAN FUNCTION IN PATIENTS ON LONG-TERM CIRCULATORY SUPPORT WITH CONTINUOUS OR PULSATILE BLOOD FLOW ASSIST DEVICES B. Radovancevic,1 E. de Kort,1 R. Radovancevic,1 I.D. Gregoric,1 O.H. Frazier,1 1Center for Cardiac Support, Texas Heart Institute, Houston, TX Objectives: Limited data exists on long term effects of left ventricular assist devices (LVADs) on end organ function, specifically in regard to continuous vs. pulsatile types of LVADs. Methods: We reviewed data of patients who underwent LVAD implantation at our institution between 1989 and 2004 and who were supported ⬎6 months after implantation. Continuous (C)-LVAD group included 12 patients bridged to transplant with a Jarvik 2000 or Thoratec HeartMate II LVAD. Pulsatile (P)-LVAD group was comprised of 58 patients who underwent LVAD support with the Thoratec HeartMate I. Follow-up was up to 15 months after LVAD implantation. Average duration of LVAD support was 370 ⫾ 182 days (range, 180 –754) for the C-LVAD group and 315 ⫾ 111 days (range, 180 –1334) for the P-LVAD group. Results: The groups were comparable in regard to age, gender, cardiac index, ejection fraction, creatinine and total bilirubin levels before implantation. C-LVAD group patients had a lower preoperative hemoglobin 10.5⫾1.7 g/dL than patients in P-LVAD group 12.2⫾1.9 g/dL, P⫽ 0.01. Four out of 12 patients in C-LVAD group and 28/58 in the P-LVAD group were transplanted. Actuarial survival, censored for transplant, at 9, 12 and 15 months was 90% for the C-LVAD group and 88%, 78% and 74% for the P-LVAD group; P⫽NS. In both groups, creatinine and total bilirubin levels improved or stayed within normal range. Conclusions: These data indicate that both LVAD types provide adequate blood flow to maintain proper end organ function. It

The Journal of Heart and Lung Transplantation February 2006

seems that there is no negative effect of prolonged continuous blood flow. 296 PROVIDING LONG-TERM CARE TO DESTINATION THERAPY LVAD RECIPIENTS IN DISTANT COMMUNITIES: PARTNERING WITH LOCAL TEAMS K.D. Rasmusson,1 J.W. Long,1 K. Nelson,1 A. Healy,1 D.U. Blank,2 D.G. Renlund,1 B.Y. Rasmusson,1 H. Smith,1 S. Clayson,1 A.G. Kfoury,1 1Utah Artificial Heart Program, Salt Lake City, UT; 2 Idaho Heart Institute, Idaho Falls, ID Background: Long-term “destination” therapy (DT) with LVADs has proven beneficial in selected patients (Pts) with Stage D heart failure. As DT becomes widely disseminated, long-term care of distant Pts will likely transition from the implanting center (IC) to their local community (LC). Regional geographic considerations prompted us to partner with physicians in other states to allow distant Pts equal access to DT. Such experience is reported here from the largest volume DT program in the U.S. Methods: All DT Pts living more than 150 miles away from the IC were included in the analysis. Community physicians and emergency medical personnel were trained by IC personnel. Outcomes were retrospectively analyzed. Results: 17 Pts from 5 distant communities underwent LVAD implantation. All Pts were successfully discharged and returned to their LC with 24/7 coverage by the IC, along with the LC extended team. See Table 1 for details. Outpatient experience totaled 17.1 patient-years. Most clinical issues were cared for at the local sites. Causes of death included cerebral vascular accidents (3), multi-system organ failure (2), power failure (1), and right heart failure (1). None of the morbidity or mortality resulted from failures of the extended careteam. The power failure death was likely due to lack of social support in the LC. Training of local teams was conducted in 87 individual sites including fire stations and ambulance companies. Gender

Age

2 Female 15 Male

66 years (42–78)

Average # days: implant to discharge 24.4 (9–50)

Distance from local site 694 miles (168–1307)

Average # days outpatient support 368 (37–1100)

Conclusion: Pts referred from distant sites for DT can be successfully returned to their LC with good outcomes, provided local physicians and emergency services are adequately trained and able to provide 24/7 coverage. Developing strong partnerships between implanting and local teams is key to successful management of Pts in their own communities. 297 SENSITIZATION IN AXIAL FLOW DEVICES VERSUS PULSATILE LEFT VENTRICULAR ASSIST DEVICES I. George,1 P. Colley,1 E. Burke,1 T. Martens,1 B. Morrow,1 M. Russo,1 C. Soneru,1 M. Tabaksblat,1 C. Spadaccio,1 Z. Lill,1 F. Cheema,1 S. Itescu,1 M. Deng,2 D. Mancini,2 M.C. Oz,1 Y. Naka,1 1 Department of Surgery, Columbia University College of Physicians and Surgeons, New York, NY; 2Department of Medicine, Columbia Univeristy College of Physicians and Surgeons, New York, NY Background: Biomaterials in pulsatile left ventricular assist devices (pLVAD) are thought to increase immunologic risk through allosensitization. We investigated the hypothesis that axial flow devices (AxF) without biologic membranes result in decreased allosensitization compared to pLVADs with biologic membranes.

The Journal of Heart and Lung Transplantation Volume 25, Number 2S

Abstracts

Methods: AxF (n⫽16) and pLVAD (Heartmate I) (n⫽36) were implanted from 2000 –2005 into patients with severe heart failure and matched for age, etiology, and duration of LVAD support. Sera reacting with more than 10% of the human leukocyte antigen (HLA) reference panel were considered positive for anti-HLA antibodies (Ab). Tissue from endomyocardial biopsy was collected after transplantation. Results: There were no significant differences in age, etiology, sex, or duration of support between cohorts (see table). Anti-HLA Ab were not detected at device implantation in AxF nor pLVAD; however, significant increases in anti-HLA Ab were present within 1 and 3 months after device support in pLVAD versus AxF. Overall, fewer AxF demonstrated positive anti-HLA Ab during support, and a trend towards less rejection was seen after transplant compared to pLVAD.

S147

patients. The hazard of drive-line infections dramatically increased with the duration of support (Figure). Conclusions: Duration of LVAD support can be limited by drive-line infections initiated by traumatic events. Once they occur, they are difficult to control and may lead to pump pocket infections. Efforts to improve the design of the device accessories to minimize torsion and protect the drive-line are essential.

Study Characteristics N Age (yrs) Sex Duration of Support (days) Ischemic Non-Ischemic ⫹ anti-HLA Ab: pre-LVAD ⫹ anti-HLA Ab: 1 month ⫹ anti-HLA Ab: 3 months ⫹ anti-HLA Ab: Txp ⫹ anti-HLA Ab: Overall ISHLT Grade 2 or higher

AxF

pLVAD

16 51.4 ⫾ 12.7 12 Male, 4 Female 117.1 ⫾ 126.5 43.8% (7) 56% (9) 0% (0) 7.7% (1)ⴱ 0% (0)ⴱ 9% (1) 6.7% (1)ⴱ 33% (3)

36 52.0 ⫾ 12.5 30 Male, 6 Female 107.3 ⫾ 112.2 50% (18) 50% (18) 0% (0) 35.2% (6) 70% (7) 25% (4) 42.9% (9) 58% (21)

*p ⬍ 0.05 vs pLVAD; anti-HLA Ab, Antibodies against Human Leukocyte Antigen; Txp, Time of Transplantation; Overall-Positive anti-HLA Ab during support.

Conclusion: In this cohort, hemodynamic support with AxF produced less allosensitization compared to pLVAD. Further study will determine if the absence of biomaterials in AxF improves long-term clinical outcomes after cardiac transplantation.

298 LATE DRIVE-LINE INFECTIONS: THE ACHILLES’ HEEL OF PROLONGED LVAD SUPPORT A. Zierer,1 E. Milner,1 M.K. Pasque,1 G.A. Ewald,2 A.S. Al-Dadah,1 S.J. Melby,1 K. Skelton,1 N. Moazami,1 1Division of Cardiothoracic Surgery, Washington University School of Medicine, St. Louis, MO; 2 Division of Cardiology, Washington University School of Medicine, St. Louis, MO Background: Infectious complications can significantly increase the morbidity of LVAD destination therapy. We evaluated the incidence, cause and clinical course of late-onset drive-line infections. Methods: Between 1996 and 2005, 35 patients were implanted with the Novacor and 38 patients with the HeartMate LVAD as either bridge to transplantation or destination therapy. Our analysis focuses on patients with late-onset infection (ⱖ 30 days) with prior clinical healing of all incisions. Results: Median duration of LVAD support was 234 (IQR: 43– 405) days for Novacor-, and 57 (IQR: 16 –208) days for HeartMate-patients (p⫽0.015). Late localized exit site infections developed in 15 (20%) patients (11 Novacor, 4 HeartMate, p⫽0.041). Eleven could be attributed to documented trauma to the drive-line. The median onset of late infection was 192 (IQR: 135–237) days with the Novacor and 112 (IQR: 42–199) days with the HeartMate device. Treatment included IV antibiotics, excisional debridement and wide drainage in 12 patients. Repeat revision was needed in 9 of them (75%), up to 6 times in 2 individuals. Despite aggressive treatment, 4 progressed to pump pocket infections which led to urgent heart transplantation (n⫽3) or explantation (n⫽1). Complete healing of the infected exit site could only be achieved in 3

299 OUTCOMES OF NON-CARDIAC SURGERY IN PATIENTS WITH MECHANICAL CIRCULATORY SUPPORT D.M. Nelson,1 J. Stehlik,2 B.J. Christensen,3 A.G. Kfoury,1 M.A. Movsesian,2 D.G. Renlund,1 C.G. Cowley,1 H.K. Smith,1 B.Y. Rasumusson,1 J.W. Long,1 1Utah Artificial Heart Program, LDS Hospital, Salt Lake City, UT; 2Cadiology, Salt Lake City VAMC, University of Utah, Salt Lake City, UT; 3Surgery, LDS Hospital, Salt Lake City, UT Introduction: Increasing numbers of patients are living with ventricular assist devices (VADs). Our aim was to assess the risks of non-cardiac surgery in these patients. Methods: We reviewed our experience with non-cardiac surgery in patients who received VADs for intermediate- and long-term support. Results: Since 1993, 174 devices have been implanted in 147 patients. Thirty patients (20%) subsequently underwent 52 noncardiac surgeries—table 1. Average patient age at surgery was 55 years, 83% were male, and 53% had ischemic cardiomyopathy. The mean duration of VAD support before surgery was 220 days (median 89 days). Systemic anticoagulation was administered perioperatively in 14 (27%) procedures. Intraoperative adjustment of the VAD function was needed in three instances (6%). Bleeding was the most common postsurgical complication (11%). 17% of abdominal surgeries required reexploration for bleeding. 30-day mortality was 13%. No deaths resulted from complications of surgery. 72% of the bridge to transplant patients were later transplanted. The average duration of VAD support following non-cardiac surgery for destination therapy