Coronary artery vasculopathy (CAV) is a late complication in heart transplant (HT) recipients and limits long-term survival. It is thought to result from a fibro-proliferative response to chronic inflammation in the coronary vessels.1,2 Although cytomegalovirus (CMV) infection has also been associated with CAV,3 CAV is often described as chronic rejection and its association with acute rejection is generally accepted.4 Previous studies in pediatric HT recipients have shown that children with ≥2 rejection episodes during the first posttransplant year or a rejection episode >1 year posttransplant (late acute rejection) are at higher risk of developing CAV.5,6 It is notable, however, that study centers contributing to these reports varied widely in their rejection surveillance practices, specifically with respect to the frequency of routine, protocol-driven biopsies for rejection surveillance. Whether the clinical severity of acute rejection (diagnosed on protocol biopsy versus clinical rejection) is related to the risk of developing CAV has not been characterized. Furthermore, although children presenting with “rejection with hemodynamic compromise” have been shown to be at higher risk of death or graft loss,7,8 it is unclear whether some of this risk is mediated by accelerated CAV.
We hypothesized that clinical rejection—defined as a rejection episode accompanied by new onset heart failure or left ventricular systolic dysfunction—is more strongly associated with development of CAV than rejection diagnosed on protocol biopsy (2R/3A cellular rejection or antibody-mediated rejection [AMR]). The specific aims of this single institution study were as follows: (1) to compare patients with no history of rejection (rejection-free subjects) to those with rejection diagnosed only during protocol biopsy for risk of developing CAV and (2) to assess whether the occurrence of a clinical rejection episode or first late acute rejection (>1 y posttransplant), assessed as a time-dependent variable, is associated with development of CAV. We applied the 2010 International Society for Heart and Lung Transplantation (ISHLT) guidelines9 for diagnosing and grading CAV.
MATERIALS AND METHODS
We identified all subjects who received their first HT at Boston Children’s Hospital between January 1, 1986 and December 31, 2015 and were <21 years old at HT. Subjects were included if they had at least 1 cardiac catheterization with coronary artery angiogram posttransplant. The exclusion criteria were as follows: (1) death before first coronary angiogram (usually performed 1 y posttransplant), (2) multiorgan transplantation, and (3) heart retransplantation. Subjects were followed from the time of HT until death, retransplantation, or the last day of observation of December 31, 2016.
Study Design and Definitions
This was a retrospective cohort study to assess the incidence of CAV in pediatric HT recipients at our institution, its association with clinical rejection, and survival after the diagnosis of CAV. The study was approved by the institutional review board with waiver of patient/parental consent.
Clinical rejection was defined as a rejection episode associated with new-onset heart failure or when associated with left ventricular ejection fraction <50% with at least 10% decline compared to baseline. Our institutional protocol for protocol biopsies consists of studies performed approximately 2, 4, 6, 9, 12, 16, and 24 weeks, 9 months, and 1 year after HT. The protocol is modified for children <1 year old who undergo 3–4 biopsies during the first posttransplant year. After 1 year, cardiac biopsies are performed every 6 months for 2 years and annually thereafter. Venous access for cardiac biopsy is obtained via the femoral vein in infants and toddlers and via the internal jugular vein in older children. Biopsy is also performed in patients clinically suspected to have rejection based on symptoms/signs of heart failure (such as poor feeding, fatigue, tachypnea, gallop rhythm, hepatomegaly) and/or echocardiographic findings of reduced function, usually before any treatment. Rejection on protocol biopsy was defined by an endomyocardial biopsy interpretation of Grade 2R cellular rejection or higher for biopsies graded by ISHLT 2004 criteria10 and Grade 3A or higher for biopsies graded by ISHLT 1990 criteria.11 Grade 1R cellular rejection on biopsy was not considered as rejection for the study. AMR was diagnosed when at least 3 of the 4 ISHLT criteria (histopathology, immunofluorescence, graft dysfunction, and donor-specific antibodies) were met.10,12 All rejection episodes, as defined above (Grade 2R or higher cellular rejection, AMR, and all clinical rejections), are treated with 3–4 days of high-dose steroids (intravenous methylprednisolone 15–20 mg/kg/d, maximum 1 gm/d) at our center. Patients with AMR are also treated with plasmapheresis if donor-specific antibodies are identified in high titers. All patients with AMR receive a 6-month course of intravenous immune globulin, given every 3–4 weeks. All patients undergo annual coronary angiogram starting 1 year posttransplant. Angiographic CAV was diagnosed using the ISHLT 2010 guidelines for CAV diagnosis9 with upgrading for severity of CAV as proposed by Kindel et al.13 Coronary angiograms are not performed at our center during clinical rejection and are usually deferred for 6 months.
Patient data were collected by reviewing the institutional transplant database, clinic notes, and reports of the diagnostic studies in the electronic medical records. All coronary angiography reports were reviewed in study subjects. The primary outcome variable was time to CAV diagnosis. Angiograms reported as showing CAV were re-examined by a single reviewer (K.D.) to grade the severity of CAV using the 2010 ISHLT criteria.9 Concurrent hemodynamic data were used to functionally upgrade to severe CAV in the presence of left ventricular ejection fraction <45%, right atrial pressure >12 mm Hg, or pulmonary capillary wedge pressure >15 mm Hg.13 Patients were classified as being free from CAV (CAV 0), having mild disease (CAV 1), or having moderate-severe disease (CAV 2 and 3, respectively).9 The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Baseline variables for study subjects transplanted in the 3 eras were compared using the χ2 test and the Kruskal–Wallis test, as appropriate. Cumulative incidence of any CAV and moderate/severe CAV were assessed at 3, 5, and 10 years posttransplant. Freedom from rejection (diagnosed on protocol biopsy versus clinical rejection) and freedom from CAV (any CAV versus moderate/severe CAV) were assessed using the Kaplan–Meier curves and the log-rank test; subjects who died before developing CAV were censored at the time of death. Freedom from any CAV between eras, between patients stratified by age, and by the number of treated rejection during the first posttransplant year was similarly compared. Association of baseline patient characteristics with time to CAV diagnosis was assessed using a Cox model. A Cox model was also used to compare time to CAV diagnosis between patients who remained rejection-free to those with rejection diagnosed only on protocol biopsy. The associations of clinical rejection and late rejection with time to CAV were analyzed by treating the first occurrence of clinical or late rejection, respectively, as a time-varying covariate; the time-dependent variable remained “yes” for the subsequent clinical course for assessing these associations with CAV.
Post-CAV survival was assessed using Kaplan–Meier analysis with time 0 as the time of CAV diagnosis for each subject. Sample size was limited by the number of patients transplanted at Boston Children’s Hospital.
During the study period, 293 patients underwent HT at Boston Children’s Hospital. Of these, 228 patients met the inclusion criteria and formed the study cohort. Their median age at HT was 8.8 years (range 6 d to 20.8 y), 32 (14%) were infants <1 year, and 104 (46%) had congenital heart disease. Table 1 summarizes baseline characteristics of study subjects by date/era of HT. Patients in earlier years received triple immune suppression with cyclosporine, azathioprine, and prednisone. Prednisone was slowly weaned during the first posttransplant year. Over time, tacrolimus became the preferred calcineurin inhibitor and mycophenolate mofetil was used instead of azathioprine. In 2006, a new immune suppression protocol was introduced, which includes thymoglobulin induction and steroids for the first 5 days followed by steroid avoidance and 2-drug maintenance regimen (tacrolimus and mycophenolate mofetil).14 Only highly sensitized patients received maintenance steroids.
Diagnosis of Rejection
During the first posttransplant year, 55 subjects were treated for 1 rejection episode, 25 for 2 or more rejection episodes, and the rest remained rejection-free. During the entire study period, 106 subjects (46%) remained rejection-free, whereas 77 (34%) had rejection diagnosed only on protocol biopsy. The first rejection on protocol biopsy was cellular in 62 (grade 2R in 60, grade 3R in 2) and antibody-mediated in 15 subjects. Figure 1A illustrates freedom from rejection diagnosed by protocol biopsy and freedom from clinical rejection in the study cohort. Clinical rejection was diagnosed in 45 subjects (20%); 56% of these were treated with inotropes and were therefore rejections with severe hemodynamic compromise. There was a concurrent biopsy in 44 of the 45 clinical rejections: 11 (25%) were cellular, 26 (59%) were AMR, and 7 (16%) were mixed. Median left ventricular ejection fraction during clinical rejection was 0.44 (25th–75th percentile: 0.37, 0.48). Late acute rejection >1 year post-HT was diagnosed in 79 (35%) patients. Of these, 53 (67%) were diagnosed on protocol biopsy and 26 (33%) were clinical rejections.
Incidence of CAV
During the study period, 72 patients (32%) were diagnosed with CAV. Of these, 33 patients were diagnosed with mild CAV and 39 patients with moderate or severe CAV. Five patients progressed from mild to more advanced CAV during the study period. Twenty patients were upgraded to severe CAV based on hemodynamic criteria.
Figure 1B illustrates freedom from any CAV and freedom from moderate-severe CAV. The cumulative incidence of any CAV at 3, 5, and 10 years post-HT was 0.054, 0.12, and 0.35, respectively. The cumulative incidence of moderate-severe CAV at 3, 5, and 10 years post-HT was 0.03, 0.06, and 0.20, respectively. Time to CAV diagnosis was not significantly different between eras for any CAV (P = 0.52) or for moderate-severe CAV (P = 0.24).
In univariate analysis, there was no association of time to CAV with recipient or donor age, recipient gender, immune suppression early posttransplant (tacrolimus versus cyclosporine, azathioprine versus mycophenolate mofetil, induction therapy, and steroids at discharge), CMV status, pretransplant sensitization, or renal function. None of these variables were significant in a multivariable model.
Association of Acute Rejection With CAV
Figure 2A illustrates freedom from CAV in patients stratified by number of treated rejections during the first year post-HT and was not significantly different among patients who had 0, 1, or ≥2 treated early rejections (P = 0.14, log-rank test).
Compared to subjects who remained rejection-free during the study period, those with rejection diagnosed only on protocol biopsy were not at higher risk of CAV (hazard ratio [HR] 1.09, 95% confidence interval [CI]: 0.54-2.09; P = 0.79). Clinical rejection was significantly associated with a higher risk of CAV (HR 4.84, 95% CI: 2.99-7.83) with no difference among subgroups stratified by concurrent biopsy findings (cellular versus AMR; P = 0.68). The association remained significant (HR 5.12, 95% CI: 3.13-8.39) after adjusting for patient and donor age, CMV status at transplant, and posttransplant immune suppression (tacrolimus versus cyclosporine). Figure 2B illustrates a significantly higher risk of CAV following clinical rejection versus following rejection diagnosed on protocol biopsy (P < 0.001). Of 23 subjects with CAV with history of rejection diagnosed only on protocol biopsy, CAV was moderate or severe in 10 (43%). In contrast, of 29 subjects with CAV diagnosed following clinical rejection, CAV was moderate or severe in 24 (83%). Overall, late acute rejection was not associated with a risk of developing CAV (HR 1.33, 95% CI: 0.88-2.01). In subgroup analysis, however, clinical late rejection was associated with a higher risk of developing CAV (HR 4.27, 95% CI: 2.42-7.51), whereas late rejection diagnosed on protocol biopsy was not (HR 0.83, 95% CI: 0.51-1.37).
Graft Loss (Death or Retransplant) After CAV Diagnosis
Graft loss (death or retransplant) after CAV diagnosis was significantly associated with the severity of CAV at first diagnosis (Figure 3). Graft loss (was accelerated in patients with an initial diagnosis of moderate/severe CAV compared to those with mild CAV (P = 0.01). Two-year graft survival was 80% after diagnosis of mild CAV and 36% after diagnosis of moderate/severe CAV (P < 0.01). There was no difference in graft survival between patients whose moderate/severe CAV was angiographic versus those whose CAV severity was based on hemodynamic upgrade (P = 0.69).
In this single-center study where rejection surveillance has been guided by protocol biopsies for 3 decades and most patients have undergone annual coronary angiography to evaluate for CAV, we applied a uniform ISHLT definition of CAV to all angiograms and found that compared to subjects who have remained rejection-free, patients with a diagnosis of moderate rejection detected only during protocol biopsy were not at higher risk of subsequent diagnosis of angiographic CAV. In contrast, not only was the onset of angiographic CAV accelerated following an episode of clinical rejection but the grade/severity of CAV was also likely to be worse. Furthermore, patients with late acute rejection were at higher risk of CAV if their rejection was clinical rejection but not if it was detected during protocol biopsy. Pediatric HT centers are known to vary in their approach to rejection surveillance (biopsy-based versus noninvasive), yet no differences in posttransplant outcomes have been described.15 The strong association of clinical rejection with subsequent CAV (versus no association of rejection on protocol biopsy with CAV) may explain similar posttransplant outcomes at centers with biopsy-based and noninvasive rejection surveillance.
The 2010 revisions to the ISHLT guidelines for CAV considered the degree of stenosis, the number and type of vessels affected, and graft dysfunction in grading CAV.9 Kindel et al have proposed modification of hemodynamic criteria for upgrading CAV in children by demonstrating that adult criteria are too insensitive to detect restrictive physiology in children.13 Applying the current ISHLT definition of CAV to all HT recipients at our institution (with hemodynamic criteria proposed by Kindel et al) allowed us to study CAV trends over time as well as their association with severity of rejection. We did not find an era effect with regards to time to CAV diagnosis. Furthermore, as expected, the cumulative incidence of any CAV and that of moderate-severe CAV as well as survival following CAV were within the range previously reported in multicenter analyses.13
We defined clinical rejection based on clinical presentation (symptoms and signs of heart failure) and/or echocardiography findings of sufficient concern to lead to an urgent biopsy in our practice. We did not include softer findings (such as fever, low immune suppression levels, or ectopy) in the definition because we have rarely, if ever, seen these to be associated with rejection in the presence of preserved graft function. Although the terms “hemodynamically significant rejection” or “rejection with hemodynamic compromise” have been used to describe severe rejection, we chose not to use these terms, first because different authors have used a different definition and, second, because those definitions have included patient treatment modality (ie, use of inotropes). In 2001, Pahl et al defined “rejection with severe hemodynamic compromise” to include use of inotrope therapy for treatment. In 2011, Everitt et al added the term “rejection with mild hemodynamic compromise” to include rejection events associated with low cardiac output where intravenous inotropes were not used.8 In a different 2011 study, Phelps et al classified “hemodynamically significant rejection” as requiring “low-dose support” when ≤2 inotropes were used at doses below a specified threshold and “high-dose support” as when either epinephrine or vasopressin was used, and >2 inotropes were used simultaneously and/or dopamine/dobutamine infusion ≥10 µg/kg/minute was used.16 These definitions are subject to institutional and provider practice variation with respect to the threshold for initiating inotropes, and if so, which ones and how many. While our definition of clinical rejection has overlap with these definitions, it is independent of the treatment employed. Our analysis shows that the clinical severity of rejection may be more important than the number and timing of rejection for the risk of developing CAV as also shown in children following rejection with hemodynamic compromise.17
We did not find age at HT to be associated with the risk of developing CAV. This is in contrast to large multicenter studies and annual ISHLT reports that have consistently demonstrated that children 11–17 years of age are at higher risk CAV compared to younger children.13,18 Proposed explanations for this association include inherent immunologic advantage in younger recipients, particularly infants, higher rates of rejection and noncompliance in teenage recipients, and higher donor age in teenage recipients. Because our study sample is small, our inability to replicate age-CAV association is likely to be a type 2 error. Furthermore, in contrast to prior studies, we did not find an association between the number of treated rejection episodes in the first year post-HT and time to CAV. While this could be due to small sample size, it may also be explained by the high frequency of protocol biopsies at our center. Because many treated rejections during the first posttransplant year are based on biopsy results, children with ≥2 episodes of rejection at our center—by potentially receiving treatment for rejection episodes that centers performing fewer protocol biopsies would not have detected/treated—may be at less risk of CAV from those previously reported.
This study has several clinical implications. First, our findings are relevant in examining the role of protocol biopsies for rejection surveillance in pediatric HT recipients. Some centers have used echocardiography-based surveillance since the early days of HT, whereas others employ protocol biopsies to avoid missing rejection in patients without clinical or echocardiographic changes.19-22 In a recent multicenter analysis, centers with high protocol biopsy intensity (≥8 in the first y post-HT) did not detect grade 2R/3A cellular rejection earlier and did not have lower 4-year mortality compared to centers with low or moderate protocol biopsy intensity.15 Our results expand on these observations and suggest that pediatric centers performing frequent protocol biopsies may reduce biopsy frequency in their recipients without significant detriment to patient outcomes. Incorporating elements of noninvasive surveillance suited to individual center’s expertise and resources may facilitate this. The goal of noninvasive monitoring should be to diagnose nonrejection with high confidence such that a biopsy can be deferred/avoided. Potential tools for such an approach include advanced echocardiography imaging using tissue Doppler or strain and circulating biomarkers, such as high-sensitivity troponin, cell-free DNA, or microRNA.23-25
Second, our findings underscore a need for developing clinical tools to reduce the risk of developing clinical rejection. Despite advances in immune suppression and reduction in first-year rejection during the last 2 decades, the incidence of severe rejection, CAV, and conditional survival in short-term survivors of HT have not improved.8,18 Innovative approaches, such as digital applications to help patients track and improve their compliance, implantable cardiac monitors to detect subclinical decline in graft function, and more frequent monitoring of immune suppression similar to such practice in diabetics and patients on chronic anticoagulation therapy, should be investigated for their potential in reducing the risk of clinical rejection and other long-term morbidities after HT.
Third, the strength of association of CAV with clinical rejection suggests that preventing CAV should become the focus of medical management in HT recipients after initial treatment and resolution of clinical rejection. Evidence-based interventions, such as proliferation-signal inhibitors, statins, and aspirin, should be added or optimized for preventing CAV and thereby improving long-term survival.26-28
This study has several limitations. First, it was a single-center study and included patients over a long time frame. Such studies can provide more granular information than is feasible from registry data but, despite considering all patients transplanted over 3 decades, the sample size was relatively small. Furthermore, differences in clinical practice related to biopsy frequency and reporting may limit generalization of our results to other institutions. Second, the quality of clinical notes was less robust in the earliest era and required review of stored paper records. We did have ready access to the results of all diagnostic studies, however, as they have all been uploaded to become part of the electronic medical records. Third, given the very large number of angiograms performed over time, we decided a priori to review all reports but only reexamine those angiograms previously reported to have CAV to reassign CAV grade based on 2010 ISHLT guidelines. While this implies trusting that all angiograms previously reported as normal did not have any CAV, it was necessary for making the study feasible.
In conclusion, an episode of clinical rejection is strongly associated with subsequent diagnosis of angiographic CAV in pediatric HT recipients, whereas this association is not present in patients with rejection diagnosed only on protocol biopsy. Because more frequent protocol biopsies may not prevent clinical rejection, our results may explain why posttransplant outcomes at centers that perform frequent protocol biopsies and those that utilize noninvasive rejection surveillance have been similar. Preventing CAV should become the focus of medical management in patients after acute treatment of clinical rejection.
1. Chih S, Chong AY, Mielniczuk LM, et al. Allograft vasculopathy: the Achilles’ heel of heart transplantation. J Am Coll Cardiol. 2016; 68:80–91
2. Valantine HA. Cardiac allograft vasculopathy: central role of endothelial injury leading to transplant “atheroma”. Transplantation. 2003; 76:891–899
3. Delgado JF, Reyne AG, de Dios S, et al. Influence of cytomegalovirus infection in the development of cardiac allograft vasculopathy after heart transplantation. J Heart Lung Transplant. 2015; 34:1112–1119
4. Seki A, Fishbein MC. Predicting the development of cardiac allograft vasculopathy. Cardiovasc Pathol. 2014; 23:253–260
5. Pahl E, Naftel DC, Kuhn MA, et al; Pediatric Heart Transplant Study. The impact and outcome of transplant coronary artery disease in a pediatric population: a 9-year multi-institutional study. J Heart Lung Transplant. 2005; 24:645–651
6. Mulla NF, Johnston JK, Vander Dussen L, et al. Late rejection is a predictor of transplant coronary artery disease in children. J Am Coll Cardiol. 2001; 37:243–250
7. Pahl E, Naftel DC, Canter CE, et al; Pediatric Heart Transplant Study. Death after rejection with severe hemodynamic compromise in pediatric heart transplant recipients: a multi-institutional study. J Heart Lung Transplant. 2001; 20:279–287
8. Everitt MD, Pahl E, Schechtman KB, et al; Pediatric Heart Transplant Study Investigators. Rejection with hemodynamic compromise in the current era of pediatric heart transplantation: a multi-institutional study. J Heart Lung Transplant. 2011; 30:282–288
9. 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-2010. J Heart Lung Transplant. 2010; 29:717–727
10. Stewart S, Winters GL, Fishbein MC, et al. Revision of the 1990 working formulation for the standardization of nomenclature in the diagnosis of heart rejection. J Heart Lung Transplant. 2005; 24:1710–1720
11. Billingham ME, Cary NR, Hammond ME, et al. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: heart rejection study group. The International Society for Heart Transplantation. J Heart Transplant. 1990; 9:587–593
12. Berry GJ, Burke MM, Andersen C, et al. The 2013 International Society for Heart and Lung Transplantation working formulation for the standardization of nomenclature in the pathologic diagnosis of antibody-mediated rejection in heart transplantation. J Heart Lung Transplant. 2013; 32:1147–1162
13. Kindel SJ, Law YM, Chin C, et al. Improved detection of cardiac allograft vasculopathy: a multi-institutional analysis of functional parameters in pediatric heart transplant recipients. J Am Coll Cardiol. 2015; 66:547–557
14. Singh TP, Faber C, Blume ED, et al. Safety and early outcomes using a corticosteroid-avoidance immunosuppression protocol in pediatric heart transplant recipients. J Heart Lung Transplant. 2010; 29:517–522
15. Zinn MD, Wallendorf MJ, Simpson KE, et al. Impact of routine surveillance biopsy intensity on the diagnosis of moderate to severe cellular rejection and survival after pediatric heart transplantation. Pediatr Transplant. 2018; 22:e13131
16. Phelps CM, Tissot C, Buckvold S, et al. Outcome of acute graft rejection associated with hemodynamic compromise in pediatric heart transplant recipients. Pediatr Cardiol. 2011; 32:1–7
17. Kleinmahon JA, Gralla J, Kirk R, et al. Cardiac allograft vasculopathy and graft failure in pediatric heart transplant recipients after rejection with severe hemodynamic compromise. J Heart Lung Transplant. 2019; 38:277–284
18. Rossano JW, Dipchand AI, Edwards LB, et al; International Society for Heart and Lung Transplantation. The registry of the International Society for Heart and Lung Transplantation: nineteenth pediatric heart transplantation report-2016; focus theme: primary diagnostic indications for transplant. J Heart Lung Transplant. 2016; 35:1185–1195
19. Putzer GJ, Cooper D, Keehn C, et al. An improved echocardiographic rejection-surveillance strategy following pediatric heart transplantation. J Heart Lung Transplant. 2000; 19:1166–1174
20. Boucek MM, Mathis CM, Kanakriyeh MS, et al. Serial echocardiographic evaluation of cardiac graft rejection after infant heart transplantation. J Heart Lung Transplant. 1993; 12:824–831
21. Hsu DT. Can non-invasive methodology predict rejection and either dictate or obviate the need for an endomyocardial biopsy in pediatric heart transplant recipients? Pediatr Transplant. 2005; 9:697–699
22. Godown J, Harris MT, Burger J, et al. Variation in the use of surveillance endomyocardial biopsy among pediatric heart transplant centers over time. Pediatr Transplant. 2015; 19:612–617
23. Lunze FI, Colan SD, Gauvreau K, et al. Tissue Doppler imaging for rejection surveillance in pediatric heart transplant recipients. J Heart Lung Transplant. 2013; 32:1027–1033
24. De Vlaminck I, Valantine HA, Snyder TM, et al. Circulating cell-free DNA enables noninvasive diagnosis of heart transplant rejection. Sci Transl Med. 2014; 6:241ra77
25. Duong Van Huyen JP, Tible M, Gay A, et al. Micrornas as non-invasive biomarkers of heart transplant rejection. Eur Heart J. 2014; 35:3194–3202
26. Asleh R, Briasoulis A, Kremers WK, et al. Long-term sirolimus for primary immunosuppression in heart transplant recipients. J Am Coll Cardiol. 2018; 71:636–650
27. Kobashigawa JA, Katznelson S, Laks H, et al. Effect of pravastatin on outcomes after cardiac transplantation. N Engl J Med. 1995; 333:621–627
28. Kim M, Bergmark BA, Zelniker TA, et al. Early aspirin use and the development of cardiac allograft vasculopathy. J Heart Lung Transplant. 2017; 36:1344–1349