Kirk R. Kanter, MDa,*, Alexandria M. Berg, MSNb, William T. Mahle, MDb, Robert N. Vincent, MDb, Patrick D. Kilgo, MSc, Brian E. Kogon, MDa, Paul M. Kirshbom, MDa a Division of Cardiothoracic Surgery, Department of Surgery, Emory University School of Medicine, Atlanta, Georgia
b Emory University School of Medicine, Atlanta, Georgia
c Department of Biostatistics, Rollins School of Public Health, Emory University, Atlanta, Georgia
Accepted for publication September 29, 2008.
* Address correspondence to Dr Kanter, Pediatric Cardiac Surgery, Emory University School of Medicine, 1405 Clifton Rd, Atlanta, GA 30322 (Email: kkanter@emory.edu).
Presented at the Forty-fourth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2008.
http://ats.ctsnetjournals.org/cgi/content/full/87/1/204
Abstract
Background: Black recipient race has been shown to predict poorer graft survival after pediatric heart transplantation. We analyzed our single-center experience comparing graft survival by race and the impact of donor–recipient race mismatch.
Methods: One hundred sixty-nine consecutive primary pediatric heart transplant patients were analyzed by donor and recipient race (white recipient, 99; black recipient, 60; other, 10). The groups were similar in preoperative characteristics. There were fewer donor–recipient race matches in blacks compared with whites (10 versus 71; p <>
Results: Although 30-day and 6-month graft survival was similar for black and white recipients (93.9% and 85.8% versus 93.3% and 83.3%, respectively), overall actuarial graft survival was significantly lower in blacks (p <> incidence of positive retrospective crossmatch (n = 26, 43%) than whites (n = 29, 29%), but this was not statistically significant (p = 0.053). The median graft survival for black recipients was 5.5 years compared with 11.6 years for whites. Donor–recipient race mismatch predicted poorer graft survival (5-year graft survival 48.9% versus 72.3%; p = 0.0032). The median graft survival for donor–recipient race-matched patients was more than twice that for mismatched patients (11.6 years versus 4.4 years). Cox proportional hazard analysis showed that donor–recipient race mismatch neutralized the effect of race on graft survival.
Conclusions: Graft survival after pediatric heart transplantation is inferior for black recipients compared with white recipients. These differences may be explained by a high incidence of donor–recipient race mismatch, which also predicts poorer outcome for all racial groups with pediatric heart transplantation. These data may have implications for future donor allocation schemes.Introduction
There are many factors that influence patient and graft survival after cardiac transplantation in children. It has been reported that black race is a risk factor for increased mortality after pediatric cardiac transplantation [1]. In adult heart transplant recipients, there are several reports that suggest that black recipient race is a predictor not only for increased mortality after cardiac transplantation [2–8] but also for increased risk of rejection [4, 9, 10] and increased incidence of transplant coronary artery disease [11]. On the other hand, others have shown no statistical difference in outcomes between black heart transplant recipients compared with nonblacks [12–15]. To address this question, we analyzed our single-institution experience with pediatric heart transplantation from 1988 through 2007 examining 169 consecutive primary pediatric heart transplant patients to determine the influence of race on graft survival.Patients and Method
Approval for this retrospective study was obtained from the Emory University School of Medicine Human Investigation Committee, which waived the need for patient consent. Patient Population
From 1988 through December 2007, 201 cardiac transplants were performed at Children's Healthcare of Atlanta at Egleston, the pediatric hospital affiliated with Emory University School of Medicine. We excluded 29 repeat transplants and 3 primary transplants in patients older than the age of 18. This left 169 primary heart transplants in children younger than 18 years of age for analysis. Of the 169 pediatric primary heart transplant recipients, 99 patients were white, 60 were black, and 10 were from other ethnic and racial groups (5 Hispanic, 4 Asian, and 1 of mixed race). The preoperative characteristics of the three groups are depicted in Table 1. The groups were similar in age, congenital diagnosis (as contrasted with idiopathic cardiomyopathy or acquired cardiomyopathy), incidence of prior operation, United Network for Organ Sharing (UNOS) status at the time of transplantation, and days on the heart transplant waiting list. The incidence of class I and class II preformed antibodies were also compared and found to be not different among the groups (Table 1).
The race of the donor for each recipient was also categorized as described above as black, white, or other. There were 121 white donors, 30 black donors, and 18 donors from other ethnic and racial groups (14 Hispanic, 3 Asian, and 1 mixed race). White recipients had significantly fewer donor–recipient race mismatches (28%) compared with blacks (83%) and others (100%) as shown in Table 1 (p <> Immunosuppression and Rejection Surveillance
All patients had identical immunotherapy regimens regardless of race consisting of triple-drug immunosuppression. Early in this series, the patients were treated with cyclosporine, azathioprine, and a weaning dose of steroids. Efforts were made to wean off steroids entirely within a year of transplantation based on negative endomyocardial biopsies. After 1992, mycophenolate was used routinely in place of azathioprine. Induction immunotherapy was not used until 2005 when we routinely added daclizumab to the standard triple-drug therapy.
Periodic surveillance endomyocardial biopsies were performed on all patients, as were routine annual coronary arteriograms. Rejection episodes were treated with pulsed steroids. Only recurrent, refractory, or hemodynamically compromising rejection episodes were treated with monoclonal antibodies such as OKT-3. Diagnostic intravascular ultrasound of the coronary arteries was not used routinely. Some patients who had unacceptable side effects from cyclosporine or who had persistent rejection episodes were converted to tacrolimus. Calcineurin antagonist dosage therapy was adjusted according to blood levels. There were no differences by recipient race in immunosuppressant treatment strategies among the patients.
Statistics Continuous variables were compared by analysis of variance and are presented as mean ± standard deviation. Nominal variables were compared by 2 analysis with Fisher's exact test. Life-table analysis was done by the Kaplan–Meier survival method with significance determined by log-rank analysis. To determine the effect of race or donor–recipient mismatching on long-term graft or patient survival, a Cox proportional hazards regression model was constructed that modeled the hazard of graft failure (death or retransplantation) as a function of these variables adjusted for other covariates such as age, congenital heart disease, prior operation, UNOS status at time of transplant, and preoperative antibody levels. Hazard ratios, along with 95% confidence intervals, were computed for each variable in the model. The proportional hazards assumption was tested and verified for all variables using Schoenfeld residual correlation analysis. All statistical tests were considered significant if the probability value was less than 0.05.
Results
Black recipients were more likely to have a positive retrospective crossmatch (n = 26, 43%) than white recipients (n = 29, 29%) and other recipients (n = 1, 10%), but this did not achieve statistical significance (p = 0.053). Kaplan–Meier actuarial survival curves stratified by recipient race are shown in Figure 1. White recipients tended to have better graft survival than black or other recipients, which approached statistical significance (p = 0.056). When one compares only white and black recipients, the survival differences do become statistically significant (p = 0.019 by log-rank test). Early graft survival was similar for all three groups at 30 days (white recipients 93.9%, black recipients 93.3%, other racial groups 90.0%) and at 6 months (white recipients 85.8%, black recipients 83.3%, other racial groups 90.0%), indicating that the differences in rates of graft or patient loss associated with recipient race occurred after the immediate posttransplant surgical period. The median graft survival for black recipients was 5.5 years compared with a median graft survival for white recipients of 11.6 years and a median graft survival for other recipients of 2.9 years.
Kaplan–Meier actuarial survival curves for donor–recipient race match and mismatch are shown in Figure 2. Freedom from death or retransplantation was significantly lower for patients who had a donor–recipient mismatch compared with those with donor–recipient match by race (p = 0.003). After excluding the 10 recipients who were neither black nor white (other racial recipient group), there were no statistically significant differences in graft survival (p = 0.33) comparing black matched recipients (n = 10) with white matched recipients (n = 71) as shown in Figure 3. Examining freedom from death or retransplantation for white and black recipients with a donor–recipient race mismatch, there was no significant difference (p = 0.69) for white mismatched recipients (n = 28) compared with black mismatched recipients (n = 50; Fig 4). The median graft survival was also similar for these donor–recipient mismatched groups (white = 4.9 years, black = 4.3 years).
Looking at only white recipients, graft survival (freedom from death or retransplantation) tended to be worse in white recipients who received a heart from a racially mismatched donor (n = 28) compared with those who had a match between the donor race and the recipient race (n = 71; Fig 5), but this did not achieve statistical significance (p = 0.064). Analyzing only black recipients for donor–recipient match (Fig 6), those with matched donors (n = 10) did not have a statistically different survival compared with black recipients who received a mismatched donor (n = 50), although these curves may not have achieved statistical difference because of the small numbers of black matched patients.
A multivariable Cox proportional hazards regression analysis was performed examining the white and black recipients (excluding the 10 recipients in the other race category). Two variables reached statistical significance (p <>
race mismatch and older recipient age as a continuous variable (Table 2). Recipient race (black versus nonblack) did not achieve statistical significance in the Cox proportional hazards model (p = 0.297), suggesting that the differences in graft survival seen between blacks and whites were related to donor–recipient race mismatch rather than race per se because so few blacks received matched donors (n = 10; 17%).
View this table:
[in a new window]
| Table 2 Multivariable Cox Proportional Hazards Regression Estimates | |
Although the Cox proportional hazards model suggested that increasing age was a risk factor for decreased graft survival after pediatric heart transplantation, the association was fairly weak with a hazard ratio of only 1.049 (Table 2). On further analysis, we were unable to find a statistical difference in graft survival by recipient age. The patients were divided into quartiles by age at time of transplant. The four quartiles were then compared using a Kaplan–Meier survival analysis. No significant differences in graft survival between age groups were demonstrated (log-rank p value = 0.41). Ten of the 169 children in this study were of a race other than black or white (5 Hispanic, 4 Asian, 1 mixed race). We further analyzed the results of this study by adding these patients to the black group and comparing the groups of white recipients versus nonwhite recipients. We also grouped the nonwhite and nonblack recipients with the white recipients and analyzed black recipients versus nonblack recipients. Neither of these analyses gave statistical results that were much different from the results described above for white recipient versus black recipient.
Comment
Since the report of Opelz and colleagues [16] in 1977 showing different graft survival after kidney transplantation in patients of dissimilar ethnic background, there has been ongoing interest in the effect of race on graft and patient outcomes after transplantation. Although there are a handful of single-center experiences with adult heart transplant recipients showing no significant effect of race on patient survival [12, 14, 15], there are multiple reports from the adult multiinstitutional Cardiac Transplant Research Database that do show a significant impact of black race on mortality after cardiac transplantation [3, 6]. There are also some single-institution studies that have shown a detrimental effect of black race on graft survival in adults after cardiac transplantation [2, 7]. Other studies from the Cardiac Transplant Research Database show that black race is a risk factor for earlier time to first rejection [4], incidence of late rejection [10], and rejection with hemodynamic compromise [9]. Further reports from the Cardiac Transplant Research Database have shown that blacks are more prone to develop obesity after cardiac transplantation, which has been shown to be a risk factor for late mortality [17], as well as being more likely to experience angiographically significant coronary artery disease, which is also a risk factor for mortality [11].
Looking at the pediatric heart transplant population, Leman and colleagues [13] reported no association of race with mortality in their series of 152 pediatric transplants. On the other hand, we previously have reported a significant disparity in outcomes of black pediatric heart transplant recipients compared with nonblack recipients, when looking at the UNOS database [1]. In this paper, we were unable to show an unfavorable association with lower income or socioeconomic status. Lee and associates [18] also looked at the UNOS database and found that blacks had a higher incidence of renal failure after pediatric cardiac transplantation. This has been associated with decreased survival. This was corroborated by a recent single-institution report from the University of Arkansas that also showed that blacks were more likely to develop renal failure after pediatric heart transplantation [19].
Rejection also seems to be more common in black pediatric heart transplant recipients. Studies from the Pediatric Heart Transplant Study group have shown that blacks are at greater risk for recurrent rejection [20], late rejection [21], and severe rejection associated with hemodynamic compromise [22]. Rejection with heart failure after pediatric cardiac transplantation has also been shown to be more common in blacks as reported by the St. Louis group [23].
It is not entirely clear why blacks fare more poorly after cardiac transplantation, although there are multiple theories. Although it has been suggested that socioeconomic factors play a role, our review of the UNOS database did not demonstrate that this was statistically significant [1]. Experience with adult patients has shown that black heart transplant recipients are more likely to have a high donor–recipient HLA mismatch [4, 7, 8]. Interestingly, in the adult Cardiac Transplant Research Database analysis, although white patients with fewer HLA mismatches had a longer time to first rejection, the black patients had a shorter time to first rejection compared with white recipients regardless of whether they had few or many HLA mismatches [4]. It has been suggested that blacks have a greater heterogeneity of their major HLA antigens to account for this difference [24].
Along these lines, a recent multicenter study looking at 364 pediatric heart transplant recipients at six centers showed that African Americans have a genetic background that may predispose to proinflammatory or a lower regulatory environment as measured by the presence of single-nucleotide polymorphisms [25]. These authors suggested that blacks may have a genetic predisposition to an unfavorable transplant environment, which can also affect the pharmacokinetics of immunosuppressive drug therapy. This has also been suggested by Mehra and colleagues [26, 27] from the Ochsner Clinic, who showed that adult black heart transplant recipients treated with cyclosporine fared worse than whites. When the immunosuppressive regimen in the blacks was changed to tacrolimus and mycophenolate, the survivals between blacks and whites were equivalent.
In the current study, we too showed unfavorable graft survival in black pediatric heart transplant recipients. On further analysis, this difference was neutralized when one takes into consideration whether or not the race of the donor and the race of the recipient were matched or mismatched (Fig 2). On Cox proportional hazards testing, the effect of recipient race was completely overshadowed by donor–recipient mismatching. Importantly, even whites who received donor–recipient mismatched organs had inferior graft survival compared with whites who received matched organs (Fig 5), although this did not achieve statistical significance (p = 0.064) in our patient set.
The data from this study cannot with certainty explain this large difference in graft survival based on donor–recipient racial matching. One can certainly speculate as to the cause. Certainly, as has been mentioned above, HLA matching does appear to influence survival after cardiac transplantation [4, 7, 8]. It makes sense that donors and recipients from the same racial group would more likely have more similar HLA antigens than those from mismatched groups and therefore experience less rejection, which could translate into better graft and patient survival. In our series, black recipients tended to have a higher incidence of a positive retrospective crossmatch compared with white recipients (43% versus 29%; p = 0.053), supporting the theory that the graft survival differences can be attributed to immunologic factors.
When one looks at donor–recipient racial matching, the difference between black recipients and white recipients is even more striking (10% versus 72%; p <> part, this is related to the fact that although blacks constituted 36% of our recipients, only 18% of the donors were black, making it much more likely that a black would receive a mismatched heart. One could speculate that the reason for the negative effect of donor–recipient racial mismatching has to do with increased likelihood of rejection and consequently graft loss. An early paper by Park and colleagues [7] from the Medical College of Virginia looking at adult male cardiac transplant recipients reported that blacks had much worse survival than whites. It is worth noting that only one of their 76 black recipients had a black donor. In contrast, other studies (including our review of the UNOS database) have shown no significant effect of donor–recipient race mismatch on pediatric heart transplant survival [1, 12].
In conclusion, this retrospective review of our experience with 169 pediatric primary heart transplant recipients has shown a significant decrease in graft survival in black patients. This effect seems to be explainable mostly by donor–recipient racial mismatches. These data may have implications for future donor allocation schemes. Certainly, it gives impetus to efforts to expand black organ donation in hopes of reducing the incidence of donor–recipient racial mismatching.Discussion
DR JAMES K. KIRKLIN (Birmingham, AL): I would like to congratulate Dr Kanter and his colleagues on a very elegant presentation and a nice analysis that further elucidates the potential impact of race on outcome after cardiac transplantation. Their finding of race mismatch as a risk factor is provocative in terms of potential allocation algorithms, but I would like to suggest at least a cautionary note based on our own analysis of a large multiinstitutional database. Although recipient black race has consistently been identified as a risk factor for both overall mortality and fatal rejection, we have not been able to identify either donor race or donor–recipient race mismatch as a risk factor. (Slide) This is a depiction of freedom from late rejection in a large database, and one can see that the greatest freedom from late rejection is in nonblack recipients, irrespective of the race of the donor, and the worst freedom from late rejection, at least in this analysis, was in black recipients, again, irrespective of donor race.
Examination of data on over 2,000 patients over a 13-year period indicates that the best survival was in white recipients, 79% and 77%, irrespective of the donor race, and the worst survival at 5 years was among black recipients, again, irrespective of donor race.
Furthermore, the impact of black race as a risk factor has gradually decreased over the years. Solution of a multivariable equation for fatal rejection indicates that in 1990 there was a great impact of black recipient race, which has gradually decreased over the following decade.
So for Dr Kanter I have three questions. In view of the differences between your findings, with a high proportion of African American recipients and the overall PHTS (Pediatric Heart Transplant Study) database, did you examine the possibility of an era effect, that is, could the effect of racial matching be less apparent in the current era?
Secondly, did you find any differences in the causes of death in your four groups, which could shed light on possible mechanisms?
And third, given the likely immunologic basis, as you stated, for the decrease in black recipient survival, have you adopted any specific immunosuppression strategies to improve their survival?
I would like to thank the Society very much for discussing this excellent paper.
DR KANTER: Thank you, Jim. I certainly cannot argue with the larger multiinstitutional database from the Pediatric Heart Transplant Study Group. However, I cannot resolve the differences between those findings and the findings that we had, which were particularly striking. One thought that I may have is that perhaps the genetic milieu of the African American population in the South is perhaps more homogeneous and different from the white patients than it is in other parts of the country. Certainly the UCLA group has shown no effect of donor or recipient race after pediatric transplantation, and one wonders if we are dealing with more heterogeneous HLA antigens.
As far as era effect is concerned, we did look at the early era versus later era divided by 50% and saw no effect on the analysis.
We did not analyze the causes of death or graft failure specifically, but the vast majority of the patients died either of acute or chronic rejection or required retransplantation for transplant coronary artery disease or graft dysfunction. Remember that 29 of our patients had retransplantation. These retransplants were excluded from the analysis since we looked at only primary transplantation.
Finally, have we modified our immunosuppressive regimen in these patients? Actually this analysis is only 6 months old and we are still trying to determine how to change our therapy. There is a series of papers from the Ochsner Clinic in Louisiana that suggests that blacks do better with tacrolimus as compared with cyclosporine, and perhaps that is a strategy that we should adopt.References
- Mahle WT, Kanter KR, Vincent RN. Disparities in outcome for black patients after pediatric heart transplantation J Pediatr 2005;147:739-743.[Medline]
- Felkel TO, Smith AL, Reichenspurner HC, et al. Survival and incidence of acute rejection in heart transplant recipients undergoing successful withdrawal from steroid therapy J Heart Lung Transplant 2002;21:530-539.[Medline]
- Higgins R, Kirklin JK, Brown RN, et al. To induce or not to induce: do patients at greatest risk for fatal rejection benefit from cytolytic induction therapy? J Heart Lung Transplant 2005;24:392-400.[Medline]
- Jarcho J, Naftel DC, Shroyer TW, et al. Influence of HLA mismatch on rejection after heart transplantation: a multiinstitutional study. The Cardiac Transplant Research Database Group. J Heart Lung Transplant 1994;13:583-595.[Medline]
- Kirklin JK, Naftel DC, Bourge RC, et al. Evolving trends in risk profiles and causes of death after heart transplantation: a ten-year multi-institutional study J Thorac Cardiovasc Surg 2003;125:881-890.[Abstract/Free Full Text]
- Lubitz SA, Baran DA, Alwarshetty MM, et al. Improved survival with statins, angiotensin receptor blockers, and steroid weaning after heart transplantation Transplant Proc 2006;38:1501-1506.[Medline]
- Park MH, Tolman DE, Kimball PM. Disproportionate HLA matching may contribute to racial disparity in patient survival following cardiac transplantation Clin Transplant 1996;10:625-628.[Medline]
- Park MH, Tolman DE, Kimball PM. The impact of race and HLA matching on long-term survival following cardiac transplantation Transplantation Proc 1997;29:1460-1463.[Medline]
- Mills RM, Naftel DC, Kirklin JK, et al. Heart transplant rejection with hemodynamic compromise: a multiinstitutional study of the role of endomyocardial cellular infiltrate. Cardiac Transplant Research Database. J Heart Lung Transplant 1997;16:813-821.[Medline]
- Stehlik J, Starling RC, Movsesian MA, et al. Utility of long-term surveillance endomyocardial biopsy: a multi-institutional analysis J Heart Lung Transplant 2006;25:1402-1409.[Medline]
- 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;17:744-753.[Medline]
- Cohen O, De La ZD, Beygui RE, Hekmat D, Laks H. Ethnicity as a predictor of graft longevity and recipient mortality in heart transplantation Transplant Proc 2007;39:3297-3302.[Medline]
- Leman NR, Levi DS, Alejos JC, Wetzel GT. Predictors of graft longevity in pediatric heart transplantation Pediatr Cardiol 2005;26:762-767.[Medline]
- Moore DE, Feurer ID, Rodgers Jr S, et al. Is there racial disparity in outcomes after solid organ transplantation? Am J Surg 2004;188:571-574.[Medline]
- Radovancevic B, Konuralp C, Vrtovec B, et al. Factors predicting 10-year survival after heart transplantation J Heart Lung Transplant 2005;24:156-159.[Medline]
- Opelz G, Mickey MR, Terasaki PI. Influence of race on kidney transplant survival Transplant Proc 1977;9:137-142.[Medline]
- Grady KL, Naftel D, Pamboukian SV, et al. Post-operative obesity and cachexia are risk factors for morbidity and mortality after heart transplant: multi-institutional study of post-operative weight change J Heart Lung Transplant 2005;24:1424-1430.[Medline]
- Lee CK, Christensen LL, Magee JC, Ojo AO, Harmon WE, Bridges ND. Pre-transplant risk factors for chronic renal dysfunction after pediatric heart transplantation: a 10-year national cohort study J Heart Lung Transplant 2007;26:458-465.[Medline]
- Sachdeva R, Blaszak RT, Ainley KA, Parker JG, Morrow WR, Frazier EA. Determinants of renal function in pediatric heart transplant recipients: long-term follow-up study J Heart Lung Transplant 2007;26:108-113.[Medline]
- Chin C, Naftel DC, Singh TP, et al. Risk factors for recurrent rejection in pediatric heart transplantation: a multicenter experience J Heart Lung Transplant 2004;23:178-185.[Medline]
- Webber SA, Naftel DC, Parker J, et al. Late rejection episodes more than 1 year after pediatric heart transplantation: risk factors and outcomes J Heart Lung Transplant 2003;22:869-875.[Medline]
- Pahl E, Naftel DC, Canter CE, Frazier EA, Kirklin JK, Morrow WR. Death after rejection with severe hemodynamic compromise in pediatric heart transplant recipients: a multi-institutional study J Heart Lung Transplant 2001;20:279-287.[Medline]
- Flippin MJ, Balzer DT, Murphy PR, et al. Rejection with heart failure after pediatric cardiac transplantation Ann Thorac Surg 1999;68:176-180.[Abstract/Free Full Text]
- Lazda VA. The impact of HLA frequency differences in races on the access to optimally HLA-matched cadaver renal transplants. The Medical Advisory Committee. Transplantation 1992;53:352-357.[Medline]
- Girnita DM, Webber SA, Ferrell R, et al. Disparate distribution of 16 candidate single nucleotide polymorphisms among racial and ethnic groups of pediatric heart transplant patients Transplantation 2006;82:1774-1780.[Medline]
- Mehra MR, Uber PA, Scott RL, Prasad AK, Park MH. Racial differences in clinical outcome using tacrolimus and mycophenolate mofetil immunosuppression in heart transplantation Transplant Proc 2001;33:1613-1614.[Medline]
- Mehra MR, Uber PA, Scott RL, Park MH. Ethnic disparity in clinical outcome after heart transplantation is abrogated using tacrolimus and mycophenolate mofetil-based immunosuppression Transplantation 2002;74:1568-1573.[Medline]
CLICK HERE TO GO BACK TO MAIN PAGE................or HERE TO GO HOME