Eur J Cardiothorac Surg 2008;34:295-300. doi:10.1016/j.ejcts.2008.04.027
Copyright © 2008, European Association for Cardio-thoracic Surgery. Published by Elsevier. All rights reserved.
One hundred days or more bridged on a ventricular assist device and effects on outcomes following heart transplantation
Jason O. Robertsona,b,
Cheryl Loberc,
Nicholas G. Smediraa,
Jose L. Naviaa,
Nikolai Sopkoa,
Gonzalo V. Gonzalez-Stawinskia,*
a Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, OH, United States
b Cleveland Clinic Lerner College of Medicine, Cleveland Clinic, Cleveland, OH, United States
c Department of Pharmacy, Cleveland Clinic, Cleveland, OH, United States
Received 4 December 2007;
received in revised form 20 April 2008;
accepted 23 April 2008.
* Corresponding author. Address: Department of Thoracic and Cardiovascular Surgery/F24, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, United States. Tel.: +1 216 444 6708; fax: +1 216 445 3294. (Email: gonzalg{at}ccf.org).
 |
Abstract
|
|---|
Background: Successful bridging to transplantation (BTT) with ventricular assist devices (VAD) is an alternative to mitigate the effects of end-stage heart failure on organ function while awaiting a heart. The effects of long-term VAD BTT on patient outcomes following transplantation are poorly studied. Methods: A retrospective chart review identified 145 patients BTT with a VAD between November of 1996 and June of 2005 at the Cleveland Clinic. Patients were divided into two groups and outcomes were compared: group 1 was supported for <100 days (median = 44 days) and group 2 was supported for 
100 days (median = 161 days). Results: Patients in group 1 were less likely to be blood type O (33% vs 68%, p
< 0.0001). BTT <100 days trended towards independently predicting improved survival by multivariate proportional hazards analysis (risk ratio = 0.75, 95% CI = 0.52–1.08, p
= 0.12), largely due to reduced in-hospital mortality in this group (2% vs 11%, p
= 0.055); however, no significant difference with respect to long-term survival was observed by Kaplan–Meier analysis (p
= 0.14). Furthermore, causes of death differed between groups: group 1 more commonly died of coronary artery vasculopathy (26% vs 0%, p
= 0.022) and group 2 more commonly died of sepsis (60% vs 26%, p
=
0.026). Ultimately, 21% of all group 2 patients died from sepsis (compared to 7% of group 1 patients, p
= 0.018). Conclusions: This study suggests that prolonged BTT with a VAD is a viable treatment strategy but may lead to significantly more post-transplant deaths from sepsis and higher in-hospital mortality. These data may inform management of this high-risk patient population.
Key Words: Ventricular assist device • Transplant • Survival • Sepsis
|
1. Introduction
|
|---|
Ventricular assist devices (VAD) are effective in bridging end-stage heart failure patients to transplantation [1,2]. Among the 2833 patients registered on the heart transplant waiting list in 2005, the 25% who qualified as status 1A had a median time to transplant of 57 days [3]. However, waiting times for cardiac transplantation vary by region and with individual risks. Large body habitus, blood type O, and elevated panel-reactive antibodies may extend time to transplantation [4]. Indeed, the median overall wait time to receive a heart was 130 days, and it was prolonged to a median of 316 days in patients with blood type O [3]. The time-related effects of VAD support on post-heart transplantation outcomes have been seldom studied. Herein we present our experience with heart transplant recipients bridged with a VAD and report how time on an assist device may impact outcomes.
|
2. Materials and methods
|
|---|
2.1 Patients
Between November of 1996 and June of 2005, 145 of 661 heart transplant patients were bridged to heart transplantation using a VAD at the Cleveland Clinic. Of these, 89 individuals were supported by a VAD for <100 days (group 1) and 56 were supported by a VAD for 100 days (group 2). Retrospective chart review provided variables such as donor and recipient demographics, previous medical history, type of device support, degree of HLA mismatch, panel-reactive antibodies (PRA), time to transplant, time to discharge, history of infections post-VAD implantation, history of infections at transplant, causes of death and date of last follow-up. The inpatient pharmacy database was queried to determine transplant relevant treatment regimens for two periods: 10 days pre-transplant and 30 days post-transplant. Information detailing the number of doses dispensed, drug form, drug strength and duration of treatment were collected for each patient for prednisone, methylprednisolone, muromonab-CD3, daclizumab, IVIG, cyclosporine, mycophenolate, basiliximab, anti-thymocyte globulin, and tacrolimus. Actual patient survival was determined using the Social Security Death Index, and a patient was assumed alive if the search was negative as of September 2007. This study was conducted following written approval from the Cleveland Clinic's institutional review board.
2.2 Ventricular assist devices
Of the 89 patients with VAD implanted for less than 100 days, 59 (66%) of the devices were HeartMate (Thoratec, Pleasanton, CA), 25 (28%) were Novacor (WorldHeart, Inc, Oakland, CA), 3 (3%) were DeBakey (MicroMed Technology, Houston, TX), and 2 (2%) were Jarvik (Jarvik Heart, Inc., New York, NY). Of the 56 patients with VAD implanted for greater than or equal to 100 days, 43 (77%) of the devices were HeartMate (Thoratec), and 13 (23%) were Novacor (WorldHeart, Inc.). HeartMate XVE device recipients received aspirin (325 mg) alone for anti-coagulation, unless otherwise indicated, and all other devices were managed with heparin, aspirin and, frequently, other anti-platelet agents before transition to long-term warfarin, with a target international normalized ratio of 2.5–3.5.
2.3 Immunosuppressive regimens
The protocol for pre- and post-transplant immunosuppressive regimens used by our center when transplanting VAD patients has been recently described [5], including regimens for sensitized patients. No special immunosuppressive treatment protocol is followed for patients with an infection at the time of transplant.
2.4 Panel of reactive antibodies
PRA against B and T lymphocytes was assessed prior to transplantation. Recipient sera were heat-treated to remove immunoglobulin M reactivity and tested by complement-dependent lymphocytotoxicity against a comprehensive 25- to 50-member cell panel of HLA-typed donors selected to represent most-defined HLA specificities. In this report, the screening test was considered positive when 10% of the cells showed cytotoxicity by standard dye-exclusion assay.
2.5 Culture and treatment of infections
Cultures of the VAD pump pocket, driveline, and inflow/outflow tracts were systematically undertaken in all patients at the time of transplant. Other specimens were obtained when clinically indicated. All available cultures were reviewed for this study, and location, culture date and organism were recorded. Contaminants grown by culture were excluded. Positive cultures were aggressively treated with full courses of appropriate antibiotics. All patients received perioperative cefazolin at the time of transplant.
2.6 Definitions
Infection at time of transplant was defined as any infection confirmed by a culture taken within a range of one day before or after transplant. Post-VAD infection was defined as any infection confirmed by a culture taken from when the VAD was implanted up until 1 day before transplantation. Overall mortality was defined as the total number of deaths upon final query of the Social Security Death Index.
2.7 Statistical analysis
Univariate analyses were performed using the Fisher's exact test for categorical variables and either the unpaired t-test or the non-parametric Mann–Whitney test for continuous variables, depending on whether the data was normally distributed. Missing or unknown data values were not imputed and are reflected in the denominator for categorical variables. The log-rank test was used for Kaplan–Meier survival analysis. A multivariable proportional hazards model was developed for actual survival. The following baseline factors were added to the model: BTT group, age, BMI, smoking status, medical history (prior cardiac surgery, dyslipidemia, hypertension, and type I and type II diabetes), HLA mismatches and PRA measurements. A p
< 0.05 was considered significant. Prism 4.0 (GraphPad Software, Inc., San Diego, CA) and JMP IN 5.1.2 (SAS Institute, Cary, NC) software were used for statistical analyses.
|
3. Results
|
|---|
3.1 Comparison of patient characteristics
Demographic information from the time of transplant for both donors and recipients is presented in Table 1
. The median (Q1–Q3) times bridged to transplant with a VAD were 44 (25–63) days and 161 (128.3–201.5) days for groups 1 and 2, respectively, with an overall range of 3–1030 days. Patients bridged for <100 days were more likely to be blood type A (p
= 0.014), and patients bridged for 100 days were more likely to be blood type O (p
< 0.0001). Additionally, patients BTT for 100 days were more likely to be discharged from the hospital after VAD implantation and readmitted expressly for their heart transplant (82% vs 42%, p
< 0.0001). There were no statistically significant differences between groups with respect to age, race, medical history, number of HLA mismatches, degree of sensitization, transplant status, or stage of heart failure.
3.2 Comparison of immunosuppressive treatment strategies
Immunosuppressive therapies were compared between groups in order to determine if patients BTT with a VAD were managed differently perioperatively (one measure of possible sensitization) depending on duration of time on a VAD and to control for how drug therapy may affect post-transplant outcomes. There were no significant differences with respect to the number of patients treated or doses dispensed for any of the drugs analyzed, either 10 days prior to transplant or 30 days post-transplant (data not shown). Methylprednisolone was commonly administered pre-transplant (26% vs 22%). Post-transplant, most patients in groups 1 and 2 were placed on maintenance immunosuppression consisting of cyclosporine (87% vs 89%) or tacrolimus (21% vs 20%), mycophenolate (87% vs 80%) and prednisone (97% vs 93%). Methylprednisolone was also commonly used post-transplant (99% vs 100%). Duration of immunosuppressive treatment with each drug was also analyzed for both groups, pre- and post-transplant, and no significant differences were observed (data not shown).
3.3 Comparison of infections
Table 2
compares post-VAD infections and infection at the time of transplant between groups. Patients BTT for 100 days were significantly more likely to have a post-VAD infection (93% vs 68%, p
= 0.0003), multiple post-VAD infections (43% vs 14%, p
= 0.0003) and post-VAD infections generally related to prolonged hospital stays, including wound/ulcer, Clostridium difficile and invasive line infections (27% vs 9%, p
= 0.0082). There were trends for patients BTT for 100 days to have higher incidences of septicemia and VAD related infections, including those of the driveline. The groups were equivalent with respect to infections at the time of transplant.
3.4 Mortality and duration of BTT
At a median follow-up of 3.93 years for group 1 and 3.62 years for group 2 (Table 1), group 2 demonstrated a 10% higher overall mortality (26% for group 1 vs 36% for group 2, p
= 0.26, Table 3
). The post-transplant in-hospital mortality accounted for most of this difference and was approximately five times greater for patients bridged to transplant for 100 days compared to those bridged for <100 days (11% vs 2%, respectively); however, this narrowly missed achieving statistical significance (p
= 0.055, Table 3). The two patients who died in-hospital in group 1 both died of sepsis, and, of the six patients who died in-hospital in group 2, three died of sepsis, one died of antibody mediated rejection, one died of ICH and one died of non-septic MSOF.
Fig. 1
shows Kaplan–Meier survival curves comparing groups 1 and 2. There was no significant difference in survival over time when times bridged to transplant were compared (p
= 0.14).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1. Kaplan–Meier estimate of post-transplant survival by duration of BTT with a VAD. No significant difference in survival is observed when group 1 is compared to group 2 (n
= 145, p
= 0.14).
|
|
A multivariate proportional hazards model was constructed for patients BTT with a VAD. VAD BTT <100 days demonstrated a trend towards predicting survival (risk ratio = 0.75, 95% CI = 0.53–1.08, p
= 0.12). This did not achieve statistical significance in part due to sample size: complete historical data only existed for 121 of the 145 patients, thus reducing the number of patients included in the model. When only variables with complete information were added to the model the confidence interval was tighter and the result was closer to statistical significance (risk ratio = 0.76, 95% CI = 0.55–1.03, p
= 0.082).
3.5 Causes of death differ between BTT groups
The causes of death for each group are presented (Table 4
). Sepsis was the predominant cause of death for group 2, whereas it accounted for a significantly lesser proportion of deaths for group 1 (60% vs 26%, respectively; p
= 0.033). In contrast, coronary artery vasculopathy was associated with 26% of the deaths for group 1 and 0% of the deaths for group 2 (p
= 0.023). Other leading causes of death included cardiogenic shock (13%) and antibody mediated rejection (8.7%) for group 1 and non-septic multisystem organ failure (15%) for group 2, although no further significant differences were observed. These data show that the difference between groups 1 and 2 with respect to percentages of the populations that succumbed to sepsis is marked (7% vs 21%, p
= 0.018), whereas, the percentages of the populations that died from coronary artery vasculopathy are less drastically different (7% vs 0%, p
= 0.082).
Infections were scored from the time of VAD placement to transplant and at transplant for the group of patients who died of sepsis, and it was noted if infections were VAD related. Among group 1, 67% of the individuals that died from sepsis had a post-VAD infection and 0% of these patients had multiple infections. In contrast, 92% of group 2 patients that died from sepsis had post-VAD infections and 42% of those individuals had a history of multiple post-VAD infections. Moreover, 50% of those patients from group 1 had an infection at the time of transplant (67% device related), whereas 64% of patients who died from sepsis in group 2 had an infection at the time of transplant (71% device related). None of these comparisons reached statistical significance due to the small sample size of patients who died from sepsis. Among the group 1 VAD patients who died from coronary artery vasculopathy, 67% had 4 HLA mismatches, but only 17% were sensitized to PRA against either T or B-lymphocytes at the time of transplant.
|
4. Discussion
|
|---|
The efficacy of bridging patients with end-stage heart failure to transplantation using an LVAD has been proven and shown to improve survival following transplantation when compared to inotrope therapy alone [1,2]. However, rarely have studies examined long-term survival of heart transplant patients with respect to time bridged on a VAD. The major findings of our report are that BTT 100 days is associated with significant mortality from sepsis and a trend towards higher in-hospital mortality, but there is no significant difference with respect to overall survival when patients are bridged to transplant for <100 or 100 days with a VAD. Management options for transplant patients are restricted by the availability of suitable donor hearts; although, these data should prompt further investigation into the effects of prolonged bridge to transplant.
In a previous study of 162 patients with a mechanical bridge to transplant lasting at least 60 days, Griffith et al. reported that patients BTT between 101 and 200 days faced higher pre-transplant mortality than those BTT between 60 and 100 days (25% vs 10%) [6]. While patients bridged for longer than 100 days faced a higher incidence of serious infectious and cerebrovascular complications, they indicated that successful transplantation, defined as a patient being discharged from the hospital, was observed in greater than 90% of patients, regardless of duration of mechanical support. Our study shows similar overall post-transplant in-hospital mortality but demonstrates a trend towards higher post-transplant in-hospital mortality for patients bridged 100 days (Table 3). The difference between these studies likely results from differences in our patient populations. Group 1 of our study included a majority of patients bridged for a shorter period of time than the inclusion criteria for the Griffith study allowed, and we had several patients bridged longer than 340 days (their upper limit) in group 2, thereby demarcating a sharper contrast between our two patient populations (Table 1).
Nevertheless, our study demonstrates that long-term, post-transplant survival is not significantly affected by duration of BTT (Fig. 1). The trends towards longer survival time and risk reduction for group 1 patients appear to be largely resultant from the lower in-hospital mortality observed in this group. These data support current practices but in total suggest that the days following transplant are critical for those patients with extended bridge to transplant.
The rates of infection in our study were comparable to those in several other reports [7]. It is well known that duration of LVAD support predicts infection [6,8–11] and that infectious complications of VAD support are serious causes of morbidity and mortality [6–9,12–15] both pre- and post-operatively, with most serious cases resulting from device-related BSI and drive-line infections [6,7,9]. This study demonstrates that post-VAD infections (Table 2) and deaths from sepsis (21% vs 7%) occur with much greater frequency in patients bridged to transplant for 100 days, and sepsis accounts for a greater proportion of deaths in patients BTT 100 days when compared to those BTT for <100 days (Table 4). Among patients that died from sepsis, those BTT 100 days more frequently had infections at the time of transplant and post-VAD implantation and more commonly had histories of multiple post-LVAD infections; although, deaths from recurrence of past infections were infrequently observed. Therefore, this is compatible with data that show pre-transplant infections do not increase likelihood of post-transplant infections or overall post-transplant survival [16]. The study by Sinha et al. does, however, demonstrate that LVAD recipients have higher rates of post-transplant infections than non-LVAD recipients because LVAD have the potential to carry infections into the post-transplant period. To our knowledge, the relationship between duration of VAD support and incidence of post-transplant deaths from sepsis has not been previously studied.
While in our experience post-transplant deaths from CAV accounted for a higher proportion of deaths when patients were BTT for <100 days (Table 4), these data should be interpreted with caution in lieu of a sound medical explanation because, although a trend was present, percentages of the populations that died from CAV did not significantly differ amongst groups 1 and 2 (7% vs 0%, p
= 0.082). The incidence of HLA mismatch 4 antigens was increased in the patients that died from CAV when compared to the general population, but the degree of sensitization at the time of transplant, as indicated by PRA measurements and desensitization strategies (Table 1 and data not shown), was comparable. It is known that implantation of an LVAD increases humoral sensitization; however, there appears to be no association between LVAD implantation and negative events such as rejection and CAV [17–22], perhaps due to stringent immunosuppression regimens [5].
4.1 Study limitations
This was a retrospective, observational, single center study and was beset by all of the limitations and biases inherent to such reports. Furthermore, in an attempt to increase the sample size for our survival analysis, some patients from whom we were unable to obtain complete historical information were included. With larger cohorts of patients, our results may have differed. Finally, we are unable to make any comment on how pre-transplant mortality may relate to time on a VAD because patients that did not reach transplantation were excluded.
|
5. Conclusions
|
|---|
These data support the viability of bridging patients to transplant for extended periods of time, but they draw attention to unique risks that these patients may face prior to and following transplant. The possibility that prolonged BTT on a VAD negatively affects post-transplant, in-hospital survival and the incidence of death from sepsis cannot be underestimated and should be substantiated by further, large studies. These outcomes may warrant consideration when contemplating use of protracted VAD support, and the data reiterate the importance of continued aggressive preventive and interventional strategies to deal with infection in VAD patients [16].
|
Footnotes
|
|---|
Presented at the American Society of Transplant Surgeons (ASTS) Meeting, Marco Island, FL, USA, January 25–27, 2008.
|
References
|
|---|
- Bank AJ, Mir SH, Nguyen DQ, Bolman III RM, Shumway SJ, Miller LW, Kaiser DR, Ormaza SM, Park SJ. Effects of left ventricular assist devices on outcomes in patients undergoing heart transplantation. Ann Thorac Surg 2000;69:1369-1374[Discussion 1375].[Abstract/Free Full Text]
- Quaini E, Pavie A, Chieco S, Mambrito B. The concerted action ‘Heart’ European registry on clinical application of mechanical circulatory support systems: bridge to transplant. The Registry Scientific Committee. Eur J Cardiothorac Surg 1997;11:182-188.[Abstract]
- The U.S. Organ Procurement and Transplantation Network and the Scientific Registry of Transplant Recipients (OPTN/SRTR) Annual Report: Transplant Data 1996–2005; 2006.
- Sun BC. Indications for long-term assist device placement as bridge to transplantation. Cardiol Clin 2003;21:51-55.[CrossRef][Medline]
- Gonzalez-Stawinski GV, Cook DJ, Chang AS, Banbury MK, Navia JL, Hoercher K, Lober C, Atik FA, Taylor DO, Yamani MH, Young JB, Starling RC, Smedira NG. Ventricular assist devices and aggressive immunosuppression: looking beyond overall survival. J Heart Lung Transplant 2006;25:613-618.[CrossRef][Medline]
- Griffith BP, Kormos RL, Nastala CJ, Winowich S, Pristas JM. Results of extended bridge to transplantation: window into the future of permanent ventricular assist devices. Ann Thorac Surg 1996;61:396-398[Discussion 407].[Abstract/Free Full Text]
- Gordon SM, Schmitt SK, Jacobs M, Smedira NM, Goormastic M, Banbury MK, Yeager M, Serkey J, Hoercher K, McCarthy PM. Nosocomial bloodstream infections in patients with implantable left ventricular assist devices. Ann Thorac Surg 2001;72:725-730.[Abstract/Free Full Text]
- Argenziano M, Catanese KA, Moazami N, Gardocki MT, Weinberg AD, Clavenna MW, Rose EA, Scully BE, Levin HR, Oz MC. The influence of infection on survival and successful transplantation in patients with left ventricular assist devices. J Heart Lung Transplant 1997;16:822-831.[Medline]
- Poston RS, Husain S, Sorce D, Stanford E, Kusne S, Wagener M, Griffith BP, Kormos RL. LVAD bloodstream infections: therapeutic rationale for transplantation after LVAD infection. J Heart Lung Transplant 2003;22:914-921.[CrossRef][Medline]
- Ankersmit HJ, Edwards NM, Schuster M, John R, Kocher A, Rose EA, Oz M, Itescu S. Quantitative changes in T-cell populations after left ventricular assist device implantation: relationship to T-cell apoptosis and soluble CD95. Circulation 1999;100:II211-II215.[Medline]
- Ankersmit HJ, Tugulea S, Spanier T, Weinberg AD, Artrip JH, Burke EM, Flannery M, Mancini D, Rose EA, Edwards NM, Oz MC, Itescu S. Activation-induced T-cell death and immune dysfunction after implantation of left-ventricular assist device. Lancet 1999;354:550-555.[CrossRef][Medline]
- Herrmann M, Weyand M, Greshake B, von Eiff C, Proctor RA, Scheld HH, Peters G. Left ventricular assist device infection is associated with increased mortality but is not a contraindication to transplantation. Circulation 1997;95:814-817.[Abstract/Free Full Text]
- Holman WL, Murrah CP, Ferguson ER, Bourge RC, McGiffin DC, Kirklin JK. Infections during extended circulatory support: University of Alabama at Birmingham experience 1989 to 1994. Ann Thorac Surg 1996;61:366-371[Discussion 372–363].[Abstract/Free Full Text]
- McCarthy PM, Schmitt SK, Vargo RL, Gordon S, Keys TF, Hobbs RE. Implantable LVAD infections: implications for permanent use of the device. Ann Thorac Surg 1996;61:359-365[Discussion 372–353].[Abstract/Free Full Text]
- Vilchez RA, McEllistrem MC, Harrison LH, McCurry KR, Kormos RL, Kusne S. Relapsing bacteremia in patients with ventricular assist device: an emergent complication of extended circulatory support. Ann Thorac Surg 2001;72:96-101.[Abstract/Free Full Text]
- Sinha P, Chen JM, Flannery M, Scully BE, Oz MC, Edwards NM. Infections during left ventricular assist device support do not affect posttransplant outcomes. Circulation 2000;102:III194-III199.[Medline]
- Joyce DL, Southard RE, Torre-Amione G, Noon GP, Land GA, Loebe M. Impact of left ventricular assist device (LVAD)-mediated humoral sensitization on post-transplant outcomes. J Heart Lung Transplant 2005;24:2054-2059.[CrossRef][Medline]
- Pagani FD, Dyke DB, Wright S, Cody R, Aaronson KD. Development of anti-major histocompatibility complex class I or II antibodies following left ventricular assist device implantation: effects on subsequent allograft rejection and survival. J Heart Lung Transplant 2001;20:646-653.[CrossRef][Medline]
- Pamboukian SV, Costanzo MR, Dunlap S, Rayburn B, Westfall AO, You ZY, Hung E, McLeod M, Heroux A. Relationship between bridging with ventricular assist device on rejection after heart transplantation. J Heart Lung Transplant 2005;24:310-315.[CrossRef][Medline]
- Radovancevic B, Golino A, Vrtovec B, Thomas CD, Radovancevic R, Odegaard P, van Rossem CC, Gaemers SJ, Vaughn WK, Smart FW, Frazier OH. Is bridging to transplantation with a left ventricular assist device a risk factor for transplant coronary artery disease?. J Heart Lung Transplant 2005;24:703-707.[CrossRef][Medline]
- Weis M, von Scheidt W. Cardiac allograft vasculopathy: a review. Circulation 1997;96:2069-2077.[Abstract/Free Full Text]
- Gonzalez-Stawinski GV, Cook DJ, Chang AS, Atik F, Navia JL, Banbury M, Roselli E, Young JB, Starling RA, Smedira NG. Early and midterm risk of coronary allograft vasculopathy in patients bridged to orthotopic heart transplantation with ventricular assist devices. Transplantation 2005;79:1175-1179.[CrossRef][Medline]
Top
Abstract
1. Introduction
