Retransplant and Medical Therapy for Cardiac Allograft Vasculopathy: International Society for Heart and Lung Transplantation Registry Analysis

Transcription

1 American Journal of Transplantation 2016; 16: Wiley Periodicals Inc. Brief Communication C Copyright 2015 The American Society of Transplantation and the American Society of Transplant Surgeons doi: /ajt Retransplant and Medical Therapy for Cardiac Allograft Vasculopathy: International Society for Heart and Lung Transplantation Registry Analysis L. A. Goldraich 1, J. Stehlik 2, A. Y. Kucheryavaya 3, L. B. Edwards 3 and H. J. Ross 1, * 1 Cardiac Transplant Program, Peter Munk Cardiac Center, University of Toronto, Toronto, Ontario, Canada 2 University of Utah School of Medicine, Salt Lake City, UT 3 International Society for Heart and Lung Transplantation and United Network for Organ Sharing, Richmond, VA Corresponding author: Heather J. Ross, Heather.Ross@uhn.ca Cardiac retransplantation for heart transplant recipients with advanced cardiac allograft vasculopathy (CAV) remains controversial. The International Society for Heart and Lung Transplantation Registry was used to examine survival in adult heart recipients with CAV who were retransplanted (ReTx) or managed medically (MM). Recipients transplanted between 1995 and 2010 who developed CAV and were either retransplanted within 2 years of CAV diagnosis (ReTx) or alive at 2 years after CAV diagnosis, managed medically (MM), without retransplant, constituted the study groups. Donor, recipient, transplant characteristics and longterm survival were compared. The population included 65 patients in ReTx and 4530 in MM. During a median follow-up of 4 years, there were 24 deaths in ReTx, and 1466 in MM. Survival was comparable at 9 years (55% in ReTx and 51% in MM; p ¼ 0.88). Subgroup comparison suggested survival benefit for retransplant versus MM in patients who developed systolic graft dysfunction. Adjusted predictors for 2-year mortality were diagnosis of CAV in the early era and longer time since CAV diagnosis following primary transplant. Retransplant was not an independent predictor in the model. Challenges associated with retransplantation as well as improved CAV treatment options support the current consensus recommendation limiting retransplant to highly selected patients with CAV. Abbreviations: ACS, acute coronary syndrome; BMI, body mass index; CAV, cardiac allograft vasculopathy; CMV, cytomegalovirus; D, donor; EF, ejection fraction; HF, heart failure; HLA, human leukocyte antigen; ICDs, implantable cardioverter-defibrillators; ISHLT, International Society for Heart and Lung Transplantation; NA, not applicable; PSIs, proliferation signal inhibitors; R, recipient; SGD, systolic graft dysfunction Received 28 January 2015, revised 03 June 2015 and accepted for publication 07 June 2015 Introduction Cardiac retransplantation is the accepted therapy for selected heart transplant recipients with advanced cardiac allograft vasculopathy (CAV) (1). However, the role of retransplantation remains controversial because of suboptimal long-term outcomes compared to de novo transplant (2). Given the scarcity of donor hearts, many believe that there is a moral obligation to distribute hearts to those individuals most likely to benefit (3). The long-term prognosis of heart transplant recipients with CAV has notably improved. Advances in the understanding of immunological and nonimmunological pathophysiologic mechanisms have led to progresses in both pharmacological and percutaneous therapeutic options for prevention and treatment of established CAV. Routine use of statin therapy and aspirin, aggressive percutaneous interventions and changes in immunotherapy (i.e. introduction of proliferation signal inhibitors [PSIs]) have reduced the incidence and improved long-term survival of heart transplant recipients with CAV in the current era (2). In the context of improved prognosis for CAV, and concerns about outcomes and ethics related to retransplantation, there is a need for prospective studies comparing medical management to retransplant for CAV. In the absence of such clinical trials, comparisons of risk profile and outcomes of heart transplant recipients with CAV who are treated medically versus those who are retransplanted using the International Society for Heart and Lung Transplantation (ISHLT) Registry might provide meaningful information to help guide clinical decisions. The objectives of the present study were to evaluate mortality rates and risk factors for mortality, using the ISHLT Registry, in adult heart transplant recipients with CAV, managed medically or retransplanted. In addition, the effect of different transplant eras on the risk of adverse outcomes following the diagnosis of CAV was explored. 301

2 Goldraich et al Methods Study design Retrospective cohort analysis of the ISHLT Registry, which includes longitudinal data collected annually from heart transplant recipients from participating institutions worldwide. Study population The study group consisted of all primary adult heart transplant recipients, age 18 years at first transplant, transplanted between 1995 and 2010 and who developed CAV. Multi-organ recipients, living donor transplants (domino heart transplants), and retransplant patients without information pertaining to the reason for retransplant were excluded. The study cohort was divided in two groups: patients retransplanted within 2 years of CAV diagnosis (retransplant group) and those alive without a retransplant at least 2 years after CAV diagnosis (medical management group) (Figure 1). For unbiased comparisons regarding follow-up time in the medical management group, the period of 2 years was selected as it represented the median time (724 days) from CAV to retransplant among all identified retransplants. In order to avoid a confounding diagnosis of primary graftfailure, which accounts for the majority of early retransplants, only patients retransplanted at least 1 year after their first transplant were included. Recipient, donor and transplant characteristics and posttransplant complications associated with increased mortality were compared between groups. CAV diagnosis and outcome definitions The presence of CAV was based on annual follow-up reports from participating centers, documenting newly diagnosed CAV, CAV as the cause of retransplant or CAV as a cause of death. As an exact date of new CAV diagnosis is not reported, for purposes of this study the annual date the event was reported was used as the date of diagnosis. The cause of retransplant was determined using the diagnosis reported for retransplant, or the cause of graft loss reported. Causes of death were classified as: CAV, acute coronary syndrome and sudden cardiac death; graft dysfunction and/or heart failure; other cardiovascular causes; cancer; infection; renal failure; multi-organ failure; other noncardiovascular causes. Exact date of outcomes was used when available (e.g. date of death), or assigned to the date of annual follow-up as explained above. Duration of follow-up was censored at the time of death, retransplant (i.e. third transplant in retransplant group) or lost to follow-up date. The era was determined based on when the diagnosis of CAV was reported. The following Registry eras which represent differences in clinical practice in the management of CAV were compared: (bare-metal stents); (drug-eluting stents and limited use of PSIs); and (drug-eluting stents and increasing use of PSIs). We did not examine the era before 1995, as use of statins and drug-eluting stents were not routine. The study is limited to those recipients with CAV diagnosed prior to 2010 to allow all patients included in the analyses to have a minimum of 2- year follow-up reported after CAV diagnosis. Statistical analyses Survival rates were examined with time-to-event analysis in which the main outcome was time to death or retransplant (i.e. third transplant in retransplant group). To avoid bias, survival was computed from the date of retransplant for retransplant group and started at 2 years after CAV diagnosis for medical management group (Figure 1). In this way, patients with CAV who were managed medically and patients who were retransplanted had approximately the same survival time after diagnosis of CAV, allowing group comparisons. Hence, both cohorts had similar survival prior to time zero of the present analysis. By selecting the 2-year conditional survival for the medical management group, deaths within the first 2 years after diagnosis of CAV were missed. There were no missing deaths after retransplant. In survival analysis, censoring occurred at time of loss to follow-up. Cox proportional hazard models were used to evaluate the independent association of recipient, donor and transplant characteristics with mortality for each study group. Adjustment variables were pre-specified by prognostic relevance in CAV and retransplant among clinical data available in the Registry. Variables with significant missing data (>20%) were not considered for adjustment. Multivariate analysis included era of CAV diagnosis, age at diagnosis of CAV, time of CAV diagnosis posttransplant, donor/recipient gender, donor/recipient cytomegalovirus status at transplant, histocompatibility antigen mismatch (A, B and DR), donor age, recipient body mass index and ischemia time. For the retransplant group, risk factors were based on the retransplant, while for the medical management arm, risk factors were based on the primary transplant. Continuous variables were analyzed by Wilcoxon Mann Whitney test with Bonferroni adjustment for multiple comparisons. Chi-square statistic (or Fisher s exact test when appropriate) was used to compare categorical variables. Survival analyses were performed by Kaplan Meier method, which incorporated information from all transplants for whom any follow-up had been provided. Survival rates were compared using log-rank test statistic. Multiple imputations were used to handle missing information for human leukocyte antigen (HLA) mismatch and continuous data fields, such as ischemia time and recipient body mass index (4). The proportion of missing values for variables where multiple imputation was used varied from 2.8% to 16%. Models were fit on each imputed data set and then combined Figure 1: Flow diagram of group characterization and study time-points. Baseline variables were collected at the time of primary transplant (T) and at the time of retransplant (for the retransplant group). CAV, cardiac allograft vasculopathy. 302 American Journal of Transplantation 2016; 16:

3 Retransplant for Cardiac Allograft Vasculopathy to produce a final set of estimates from which the relative hazard ratio, 95% confidence intervals, and p-values were obtained. Continuous factors were fit using a restricted cubic spline. All analyses were performed using SAS statistical program (SAS Institute, USA). A p-value lower than 0.05 was considered as statistically significant. Ethical considerations This is a retrospective review of ISHLT Registry consisting of de-identified data that has been approved by the participating institutions to be used for scientific purposes. Therefore, additional research ethics board approval was not sought. Results Patient demographics The study population included 4595 patients with a diagnosis of CAV (65 in the retransplant group and 4530 in medical management group). Median times from primary transplant to CAV diagnosis were 2.9 years in the retransplant group and 2.1 years in the medical management group (p ¼ 0.25) (Figure 2). Donor, recipient and transplant characteristics are listed in Table 1. Data for the patients who received a retransplant are listed in 2 columns, to show their baseline characteristics at time of the primary transplant and again at time of retransplant. For the group of patients managed medically, data were collected at time of the primary transplant. Retransplant patients were younger at time of CAV diagnosis compared to medically managed patients (51 vs. 58 years; p < ) and more likely to be cytomegalovirus seropositive (89.3% vs. 62.7%; p ¼ ). Recipient sex was similar in both groups (78.5% males vs. 79.6%; p ¼ 0.82). Prevalence of left ventricular systolic dysfunction defined as left ventricular systolic function less than 45% at 1 year following diagnosis of CAV was 32.0% for retransplant group versus 4.0% for medical management group (p < ). Number of cases/recipients did not differ across eras of CAV diagnosis. Of the 4530 patients in medical management group, 67 (1.5%) eventually also underwent retransplant, though more than 2 years following the CAV diagnosis. One patient in retransplant group underwent a third transplant, which occurred between 2 and 3 years of CAV diagnosis. From 149 participating centers that had patients included in the study, 27.5% had patients in retransplant group. Of those, most centers (68%) had only one retransplanted individual. Posttransplant immunosuppression therapy Immunosuppression at 1 year differed with higher use of PSIs in retransplant group (p ¼ 0.049) and higher use of mycophenolate mofetil/mycophenolic acid over azathioprine (p ¼ 0.01), but use of immunosuppressive therapies was similar between the two groups at 5 years (Figure 3). Survival rates and causes of death The median follow-up time was 5 years for retransplant and 3.8 years for medical management. There were 24 (36.9%) deaths in retransplant group, and 1466 (32.3%) in medical management group. Causes of death were similar between study groups (Table 2). Data on cause of death was missing for 13% of the retransplant group and 22% of the medical management group. Deaths due to graft dysfunction and heart failure were similar between groups. There was no difference in survival between those managed medically versus those retransplanted at 9 years (55% in retransplant vs. 51% in medical management, p ¼ 0.88) (Figure 4). As two-thirds of the patients were transplanted within 1 year of CAV diagnosis, we did not have complete information on the proportion of patients with systolic dysfunction in this group, a data field that is collected annually. Therefore, we were not able to perform survival analysis according to the presence of systolic dysfunction in each study group. However, overall the retransplant group (n ¼ 65) had better survival versus medically managed patients with systolic graft dysfunction at 1 year after development of CAV (n ¼ 124) (p ¼ 0.02) (Figure 5). Survival was similar between retransplanted and medically managed recipients when considering the era of CAV diagnosis (p ¼ 0.54 for ; p ¼ 0.24 for ; p ¼ 0.25 for ). Figure 6 illustrates survival curves for recipients with a diagnosis of CAV in the recent era. Figure 2: Time of cardiac allograft vasculopathy (CAV) diagnosis following primary transplant for study groups. p- value of 0.62 by chi-square statistic. Factors associated with overall mortality Table 3 describes independent predictors of all-cause mortality within 2 years in the overall cohort. Diagnosis of CAV in the early era ( ) and longer time of CAV diagnosis following primary transplant were risk factors for mortality. Importantly, retransplant was not associated with a significant increase or decrease in mortality. Figure 7 American Journal of Transplantation 2016; 16:

4 Goldraich et al Table 1: Recipient, donor, and transplant characteristics according to study group Retransplant arm Variable Primary transplant Retransplant Medical management p-values N NA Age at diagnosis of CAV, years 51.0 ( ) 58.0 ( ) < Age at transplant, years 47.0 ( ) 51.0 ( ) 55.0 ( ) < / Recipient gender, male/female 78.5/21.5% 79.6/20.4% Recipient BMI at transplant 24.5 ( ) 26.5 ( ) 26.1 ( ) / Etiology of heart failure 3 Coronary artery disease 40.6% NA 50.4% /< Cardiomyopathy 48.4% NA 45.2% Valvular 0% NA 2.7% Congenital 9.4% NA 1.4% Retransplant NA 100% 0% Other causes 1.6% NA 0.3% Donor age, years at transplant 29.0 ( ) 29.0 ( ) 35.0 ( ) / Donor gender, male/female 76.9/23.1% 66.2/33.8% 71.2/28.8% / Donor-recipient gender mismatch at transplant / D female/r male 12.3% 20.0% 18.6% D male/r female 10.8% 7.7% 10.1% HLA-A mismatch at transplant / % 8.3% 8.1% % 46.7% 44.6% % 45.0% 47.4% HLA-B mismatch at transplant / % 0% 2.9% % 30.0% 27.7% % 70.0% 69.4% HLA-DR mismatch at transplant / % 3.3% 4.5% % 31.7% 39.4% % 65.0% 56.1% Allograft ischemic time at transplant (hours) 3.0 ( ) 3.1 ( ) 3.1 ( ) / Time of CAV diagnosis post transplant (years) 2.9 ( ) 2.1 ( ) Time of CAV diagnosis post transplant <1 year 10.8% 12.8% 1 to <2 years 26.2% 25.9% 2 to <5 years 35.4% 39.7% 5 to <10 years 24.6% 19.0% 10þ years 3.1% 2.7% Donor/recipient CMV status at transplant / (þ)/(þ) 33.3% 60.0% 36.6% (þ)/( ) 33.3% 5.5% 21.7% ( )/(þ) 22.8% 29.1% 25.9% ( )/( ) 10.5% 5.5% 15.8% 304 American Journal of Transplantation 2016; 16:

5 Retransplant for Cardiac Allograft Vasculopathy Table 1: Continued Retransplant arm Variable Primary transplant Retransplant Medical management p-values Recipient CMV status at transplant / (þ) 56.1% 89.3% 62.7% ( ) 43.9% 10.7% 37.3% Systolic graft dysfunction (%EF < 45) at 1 year after development of CAV % 4.0% < Hyperlipidemia at 1 year after development of CAV % 72.0% Continuous factors are expressed as median (5th 95th percentiles). Characteristics that did not change between primary transplant and retransplant in the retransplant group are described only once. None of pair-wise comparisons between primary transplant versus retransplant for retransplant group were significant at p < 0.05, except etiology of heart failure, donor/recipient CMV status at transplant and recipient CMV status at transplant. NA, not applicable; CAV, cardiac allograft vasculopathy; D, donor; R, recipient; HLA, human leukocyte antigen; CMV, cytomegalovirus; EF, ejection fraction. 1 p-value comparing primary transplant for retransplant group versus medical management group. 2 p-value comparing retransplant versus medical management group. 3 Comparisons were done using the Fisher s exact test. 4 For retransplant group, data are based on 25 recipients retransplanted at or after 1 year following primary transplant and were limited to data reported prior to retransplant. Data are not collected on post 5 years follow-ups in U.S. 5 For retransplant group, data are based on 28 recipients retransplanted at or after 1 year following primary transplant and were limited to data reported prior to retransplant. Data are not collected on post 5 years follow-ups in U.S. Figure 3: Immunosuppression therapy at 1 and 5 years posttransplant for the study cohorts. Rates represent usage of immunosuppressive agents after retransplant for the retransplant group. p < depicts the time effect on hazard ratio for 2-year mortality according to time of CAV diagnosis. Discussion This study was conducted to assess survival in heart transplant recipients diagnosed with CAV. We explored survival in patients who underwent retransplantation for CAV, and compared this group to patients who were managed medically, without retransplant, within 2 years of CAV diagnosis. We show that survival was overall similar between these two groups of patients. However, we identified a subgroup of patients that may have derived a survival benefit from retransplant, specifically, those in the retransplant group who had systolic graft dysfunction. While survival after transplant and after retransplant improved over time, we did not find between group survival differences between the two cohorts of interest in any of the three eras. Cardiac allograft vasculopathy is a complex condition with different clinical presentations, coronary vessel involvement and degrees of graft dysfunction in which factors associated with increased mortality require more standardized and contemporary characterization (5). Time of occurrence after transplant, speed of development, angiographic extension, and widespread myocardial fibrosis causing restrictive cardiac physiology irrespective of the presence of systolic dysfunction have been described as markers of adverse prognosis (6 8). Significant changes occurred over the past two decades in CAV incidence, management options, and survival in transplant recipients with CAV. The cumulative incidence of CAV at 8 years after transplant has decreased from 46% to 40% in patients transplanted in versus American Journal of Transplantation 2016; 16:

6 Goldraich et al Figure 4: Kaplan Meier overall survival curves for both study groups Statistics are based on survival since retransplant for retransplant group and starting at 2 years after the cardiac allograft vasculopathy (CAV) diagnosis for medical management group. Number at risk at the last follow-up: 11 in retransplant group, 572 in medical management group Survival in these patients has improved also among patients who had CAV reported within the first 3 years posttransplant, 5-year mortality has decreased from 29% in the first era to 25% in the more recent era (2,9). Among factors that have modified the natural history of CAV is the routine use of statins, preferential use of mycophenolic acid derivatives as the antiproliferative agent of choice and incorporation of PSIs in the management of patients with CAV (10 17). We observed that rates of mycophenolic acid derivative were elevated in both groups and that PSI use increased over time in medically managed patients. High proportions of both mycophenolic Figure 5: Kaplan Meier survival curves for overall retransplant group versus medical management patients who had systolic graft dysfunction (SGD) at 1 year after development of cardiac allograft vasculopathy (CAV). Statistics are based on survival since retransplant for retransplant group and starting at 2 years after the CAV diagnosis for medical management group. Number at risk at the last follow-up: 11 in retransplant group, 18 in medical management group. acid derivative and PSI use observed at 5 years in both study groups likely reflect contemporary clinical practice, supported by the latest data on immunosupression from the ISHLT Registry reporting that 20% of patients are on a PSI at 5 years after heart transplant, which represents a nearly Table 2: Overall deaths and causes of death per each study group Variable Retransplant Medical management p-values Total number of deaths Total number of deaths NA with known cause of death Cause of death 1 CAV, ACS and sudden 9.5% 8.6% 0.62 cardiac death Graft dysfunction/hf 14.3% 12.2% Infection 19.0% 12.4% Multiple organ failure 4.8% 7.4% Cancer 9.5% 21.4% Other cardiovascular 28.6% 17.7% Renal related 0% 7.2% Other 14.3% 13.1% NA, not applicable; CAV, cardiac allograft vasculopathy; ACS, acute coronary syndrome; HF, heart failure. 1 Data are expressed as percentages. Figure 6: Kaplan Meier survival curves for retransplant and medical management group recipients in whom cardiac allograft vasculopathy (CAV) was diagnosed in the recent era ( ). Statistics are based on survival since retransplant for retransplant group and starting at 2 years after the CAV diagnosis for medical management group. Number at risk at the last follow-up: 15 in retransplant group, 69 in medical management group. 306 American Journal of Transplantation 2016; 16:

7 Table 3: Multivariable model of risk factors for 2-year mortality for both study groups Retransplant for Cardiac Allograft Vasculopathy Variable N Hazard ratio p-value 95% confidence interval Retransplant group vs. medical management group , 2.84 Era of CAV diagnosis (reference , 1.82 group ¼ ) , 1.32 Donor/recipient gender (reference Female/female , 1.26 group ¼ male/male) Female/male , 1.29 Male/female , 1.41 Donor/recipient CMV status at transplant (reference group ¼ (þ)/(þ)) ( )/( ) , 1.08 ( )/(þ) , 1.11 (þ)/( ) , 1.19 HLA mismatch HLA-A 0 ¼ , ¼ ¼ 1793 HLA-B 0 ¼ , ¼ ¼ 2625 HLA-DR 0 ¼ , ¼ ¼ 2110 Continuous variables Age at diagnosis of CAV 0.45 Donor age at first transplant 0.45 Recipient BMI 0.94 Ischemia time (hours) 0.52 Time of CAV diagnosis posttransplant, years < For the retransplant group, the risk factors were based on the retransplant. For the medical management arm, the risk factors were based on the primary transplant. CAV, cardiac allograft vasculopathy; CMV, cytomegalovirus; HLA, histocompatibility antigen; BMI, body mass index. 100% increase from the first year, and that more than 80% are on mycophenolic acid derivative (2,21). In addition to pharmacological management, percutaneous coronary intervention strategies have improved. The adoption of drug-eluting stents in heart transplant patients has reduced in-stent restenosis to rates comparable to the general population with coronary artery disease (18,19). Furthermore, implantable cardioverter-defibrillators may be an effective strategy to lower sudden cardiac death in the setting of CAV (20). Over the same period, survival after retransplant has also improved (2,21). Whether or how should the above described changes influence selection of patients with CAV for retransplantation is not readily apparent. In any case, retransplant rates have gradually increased, currently approaching 3% of all heart transplants performed (21). Retransplant for CAV continues to be the most frequent reason after 1 year in both adults and children worldwide (22,23). While survival after retransplant in the setting of CAV is favorable, identification of patients that would derive survival benefit compared to continued medical management continues to present a significant challenge (22 25). This is underscored by the fact that systolic ejection fraction remains preserved in many transplant recipients despite significant CAV. The results of our study suggest that even in the current era retransplantation should be limited to highly selected patients with CAV, a finding consistent with the most recent consensus recommendations (1). Overall, we found no survival benefit in transplant recipients with CAV treated with retransplant compared to CAV management without retransplantation, as long as the transplant recipients left ventricular ejection Figure 7: Hazard ratio for 2-year mortality according to time of cardiac allograft vasculopathy (CAV) diagnosis. Statistics are based on survival since retransplant for retransplant group and starting at 2 years after the CAV diagnosis for medical management group. American Journal of Transplantation 2016; 16:

8 Goldraich et al fraction remained in the normal range. While long-term survival after retransplantation for CAV appeared favorable, the elevated mortality hazard in the first year after retransplant was probably responsible for the lack of significant survival benefit compared to continued CAV management without retransplant. The similarities in causes of death particularly of cardiovascular deaths observed between retransplant and medical management also raise concerns about potential accelerated recurrence of CAV and increased cardiovascular risk following retransplant. Limitations The ISHLT Registry report is voluntary and limited to participating centers. The definitions of CAV have evolved over time and may have influenced the reported causes of graft failure, retransplant and death. Indeed, the main limitation is that CAV is graded in the Registry as present or absent on data collection forms annually, and the recently adopted nomenclature which includes anatomical and functional severity is unavailable (5). In addition, we did not have a specific date of diagnosis, however we know CAV developed in the preceding 12 months. We therefore believe that, with a follow-up time of up to 9 years, this limitation would be unlikely to affect the overall results of our study. Due to more selective data collection at later points after transplant, we were not able to examine some potential late factors associated with mortality risk following the CAV diagnosis, such as new development of systolic dysfunction, rejection or additional comorbidities. The size of the retransplant group was small as our inclusion criteria were fairly narrow, in an attempt to prevent selection bias. Despite that, some non-random differences between the retransplant group and the medically managed group were present. In our opinion, the difference in the prevalence of primary coronary artery disease is unlikely to significantly influence outcomes post-retransplant. Multivariable analysis was not adjusted for systolic dysfunction after CAV diagnosis due to the number of missing data points. In addition, some patients in the medically managed group later received a retransplant. Finally, differences among participating centers in volume, diagnosis and management including differences in adoption of clinical practice changes over time are an intrinsic limitation of registries and might have been clustered between study cohorts. The implications of our findings are provocative and timely. Ongoing discussions in our field explore the ethics/ appropriateness of retransplantation at the time of continued donor shortage and ever longer waiting times for transplant. In light the design and dataset limitations described above, we admit that definitive conclusions for individual patient management cannot be made based on our study. However, in the absence of prospective randomized data, we believe that our analysis provides important insights into contemporary management and natural history of CAV and can be used in conjunction with other data to optimize selection and timing of CAV therapies. Conclusions Following the diagnosis of CAV, overall long-term survival is comparable with either continued CAV management or with retransplantation in most patients. Survival may be improved with retransplant in selected individuals, particularly those with systolic dysfunction. Challenges associated with retransplantation as well as improved CAV treatment options support the current consensus recommendation limiting re-transplant to highly selected patients with CAV. Disclosure The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation. References 1. Johnson MR, Aaronson KD, Canter CE, et al. Heart retransplantation. Am J Transplant 2007; 7: Stehlik J, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplantation: 29th official adult heart transplant report J Heart Lung Transplant 2012; 31: Ubel PA, Arnold RM, Caplan AL. Rationing failure. The ethical lessons of the retransplantation of scarce vital organs. JAMA 1993; 270: Harrell F, Jr. Regression modeling strategies. Berlin: Springer; Mehra MR, Crespo-Leiro MG, Dipchand A, et al. International Society for Heart and Lung Transplantation working formulation of a standardized nomenclature for cardiac allograft vasculopathy J Heart Lung Transplant 2010; 29: Gao SZ, Hunt SA, Schroeder JS, Alderman EL, Hill IR, Stinson EB. Early development of accelerated graft coronary artery disease: Risk factors and course. J Am Coll Cardiol 1996; 28: Costanzo MR, Naftel DC, Pritzker MR, et al. Heart transplant coronary artery disease detected by coronary angiography: A multiinstitutional study of preoperative donor and recipient risk factors. Cardiac transplant research database. J Heart Lung Transplant 1998; 17: Kobashigawa JA, Itagaki BK, Razi RR, et al. Correlation between myocardial fibrosis and restrictive cardiac physiology in patients undergoing retransplantation. Clin Transplant 2013; 27: E679 E The International Society for Heart and Lung Transplantation. Adult Heart Transplantation Statistics [cited 2014 Sept 2]. Available from: heartlungregistry. 10. Costanzo MR, Dipchand A, Starling R, et al. The International Society of Heart and Lung Transplantation guidelines for the care of heart transplant recipients. J Heart Lung Transplant 2010; 29: American Journal of Transplantation 2016; 16:

9 Retransplant for Cardiac Allograft Vasculopathy 11. Wenke K, Meiser B, Thiery J, et al. Simvastatin initiated early after heart transplantation: 8-year prospective experience. Circulation 2003; 107: Kobashigawa JA, Moriguchi JD, Laks H, et al. Ten-year follow-up of a randomized trial of pravastatin in heart transplant patients. J Heart Lung Transplant 2005; 24: Kobashigawa JA, Tobis JM, Mentzer RM, et al. Mycophenolate mofetil reduces intimal thickness by intravascular ultrasound after heart transplant: Reanalysis of the multicenter trial. Am J Transplant 2006; 6: Eisen HJ, Tuzcu EM, Dorent R, et al. Everolimus for the prevention of allograft rejection and vasculopathy in cardiac-transplant recipients. N Engl J Med 2003; 349: Mancini D, Pinney S, Burkhoff D, et al. Use of rapamycin slows progression of cardiac transplantation vasculopathy. Circulation 2003; 108: Eisen HJ, Kobashigawa J, Starling RC, et al. Everolimus versus mycophenolate mofetil in heart transplantation: A randomized, multicenter trial. Am J Transplant 2013; 13: Topilsky Y, Hasin T, Raichlin E, et al. Sirolimus as primary immunosuppression attenuates allograft vasculopathy with improved late survival and decreased cardiac events after cardiac transplantation. Circulation 2012; 125: Zakliczynski M, Lekston A, Osuch M, et al. Comparison of longterm results of drug-eluting stent and bare metal stent implantation in heart transplant recipients with coronary artery disease. Transplant Proc 2007; 39: Dasari TW, Hennebry TA, Hanna EB, Saucedo JF. Drug eluting versus bare metal stents in cardiac allograft vasculopathy: A systematic review of literature. Catheter Cardiovasc Interv 2011; 77: Tsai VW, Cooper J, Garan H, et al. The efficacy of implantable cardioverter-defibrillators in heart transplant recipients: Results from a multicenter registry. Circ Heart Fail 2009; 2: Lund LH, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplantation: Thirtieth official adult heart transplant report-2013; focus theme: Age. J Heart Lung Transplant 2013; 32: Tjang YS, Tenderich G, Hornik L, Korfer R. Cardiac retransplantation in adults: An evidence-based systematic review. Thorac Cardiovasc Surg 2008; 56: Conway J, Manlhiot C, Kirk R, Edwards LB, McCrindle BW, Dipchand AI. Mortality and morbidity after retransplantation after primary heart transplant in childhood: An analysis from the registry of the International Society for Heart and Lung Transplantation. J Heart Lung Transplant 2013; 33: Tsao L, Uriel N, Leitz K, Naka Y, Mancini D. Higher rate of comorbidities after cardiac retransplantation contributes to decreased survival. J Heart Lung Transplant 2009; 28: Radovancevic B, McGiffin DC, Kobashigawa JA, et al. Retransplantation in 7,290 primary transplant patients: A 10-year multi-institutional study. J Heart Lung Transplant 2003; 22: American Journal of Transplantation 2016; 16: