| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;78:890-899
© 2004 The Society of Thoracic Surgeons
a Division of Cardiac and Thoracic Surgery, Philadelphia, PennsylvaniaUSA
b Department of Biostatistics, Philadelphia, PennsylvaniaUSA
c Department of Pathology, Philadelphia, PennsylvaniaUSA
d Division of Cardiology, Temple University School of Medicine, Philadelphia, Pennsylvania, USA
Accepted for publication February 3, 2004.
* Address reprint requests to Dr Furukawa, Cardiac and Thoracic Surgery, Temple University School of Medicine, 3401 N Broad St, Suite 300, Parkinson Pavilion, Philadelphia, PA 19140, USA.
satoshi.furukawa{at}temple.edu
Presented at the Thirty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 31–Feb 2, 2003.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: We retrospectively studied records of all primary heart transplants performed between January 1990 and July 2002. Fifty-six patients who had received donor hearts at least 50 years of age were compared with 611 recipients of donor hearts less than 50 years of age. Clinicopathologic parameters were analyzed for their effect on mortality using the Cox proportional hazard model with calculation of hazard ratios (HR). Cut-point analysis of donor age was used to determine which donor age is associated with the greatest risk of mortality after transplant.
RESULTS: Recipients of donor hearts at least 50 years of age were older (58.5 years ± 7.0 vs 53.2 ± 11.6; mean ± standard deviation [SD]; p < 0.0001), suffered more often from ischemic cardiomyopathy (69% vs 50%, p = 0.01), and experienced a longer waiting time (192.2 days ± 301.0 vs 138.6 ± 190.8, p < 0.0001). Donor hearts at least 50 years of age (age 54.1 ± 3.5 years) were more often female (50% vs 34%, p = 0.03), died less often of "head trauma" (9% vs 42%, p < 0.0001), and exhibited fewer cytomegalovirus (CMV) mismatches (29% vs 39%, p = 0.04) than donor hearts less than 50 years of age (age 26.8 ± 12.3 years). Multivariate predictors of mortality were rejection index (HR 1.90 per unit [rejections/100 survival days], p < 0.0001), donor age (HR 1.16 per 10-year increment, p = 0.002), and recipient age (HR 1.24 per 10-year increment, p = 0.04). Recipients of donor hearts at least 50 years of age had reduced 1-year and 5-year survival ([65.7% vs 81.7%, p < 0.05] and [48.3% vs 68.4%, p < 0.05], respectively), as well as a higher proportion of deaths occurring within 1 month of transplant (41% of total deaths vs 23%, p = 0.06). Cut-point analysis indicated the characteristic of donor age of at least 40 years (categorical variable) to predict mortality with the same degree of fit as age used as a continuous variable.
CONCLUSIONS: Although we observed a substantial reduction in survival among patients who were allocated donor hearts at least 50 years of age, this difference was not solely attributable to the categorical variable of donor age 50 in this group. Donor age as a continuous variable, however, was determined to be a notable predictor of survival and use of the donor age cut-point of 40 years (categorical variable) allowed risk stratification with similar accuracy. The use of a donor age cut-point of 40 years may be a useful clinical criterion for graft-related risk assessment.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
Evidence of this change in graft usage is an increase in mean donor age for heart transplantation from approximately 27–30 years between 1990 and 2000 [2]. Between 1999 and 2001 mean donor age was 33.4 years [3]. The use of donor hearts at least 50 years of age, in particular, accounted for a large portion of this change as more than 10% of the transplants performed in 2000 were from this population compared with only 2% in 1990. Despite the increased use of this particular population of donors, some controversy exists as to the long-term outcome of this practice. Multicenter registries have consistently illustrated a linear relationship between increasing donor age and increasing mortality [3, 4]. The use of donor hearts older than the ages of 35 and 45 years has been supported by multiple large single-center reports [5–14] demonstrating equivalent survival compared with the use of younger donor organs, but a much smaller experience currently supports the use of donor hearts at least 50 years of age [15–18].
The present study was undertaken to evaluate a relatively large single-institution experience regarding the use of older donor grafts for heart transplantation. The arbitrary division of our cohort at the donor age of at least 50 years was based on the increasing use of this population of donors as well as a relative paucity of previous single-center reviews on the considerable relevance of this age cut-point. Our specific goals were to identify donor age-related differences in outcome and donor- and recipient-related risk factors that predict increased mortality. In addition we sought to delineate whether particular donor age groups are associated with an increased risk of mortality after heart transplantation.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
A total of 739 heart transplants were performed during the period of January 1990 through July 2002. Patients undergoing transplantation of multiple organs simultaneously (heart/lung [N = 14] and heart/kidney [N = 16]) and patients in whom donor age could not be ascertained (N = 15, 2.0%) were excluded from the analysis. In addition patients requiring repeat transplantation (N = 27) were excluded.
The remaining 667 patients were analyzed for multiple explanatory variables derived from the International Society for Heart and Lung Transplantation (ISHLT) registry [3]. Recipient data included demographics and the following pretransplant clinical characteristics: etiology of heart failure (ischemic, idiopathic, or other), presence of diabetes mellitus, pulmonary arterial pressures, cardiac output, necessity of pretransplant ventricular assist device (VAD) support, duration of VAD support, necessity of intraaortic balloon pump (IAPB) counter pulsation, necessity of inotrope infusion, history of previous sternotomy, waiting time, list status at transplantation, necessity of mechanical ventilatory support, abnormal renal function (defined as serum creatinine 
1.5 gm/dL or creatinine clearance 
75), panel reactive antibody level (B-lymphocyte and T-lymphocyte), and donor–recipient cytomegalovirus (CMV) mismatch. Donor data included demographics, cause of death (head trauma, anoxia, or other), ejection fraction before explantation, coronary angiographic anatomy (if available), necessity of inotrope infusion, cold ischemic time, positivity of serum hepatitis B or C serology, history of cardiopulmonary resuscitation (CPR), and culture positivity (blood or sputum).
Outcome variables analyzed included rejection ( 2/4 ISHLT grade) and survival. Because of the large variation in survival between patients (and the possible subsequent differences in the number of biopsies indicating rejection), a rejection "index" was used to relate the number of biopsies indicating rejection and survival (defined as the number of rejections divided by survival days multiplied by 100). Hospital charts and autopsy data were used to determine the recipient cause of death (divided into early graft failure [defined as cardiac death within 1 week of transplant], delayed cardiac death [death due to any cardiac-related cause later than 1 week after transplant], infection, malignancy, acute rejection, transplant coronary arteriopathic disease [TCAD], and other). Autopsy was requested in all cases of death and granted in approximately 50% of these cases.
Statistical analysis
All data are presented as mean ± standard deviation. Proportions and hazard ratios (HR) are presented with 95% confidence intervals (CI). Between-group differences in frequency data were analyzed using the Fisher exact test. Differences in continuous variables were analyzed using independent t tests. Stratified survival was analyzed using the Kaplan–Meier product-limit method followed by the log-rank test for differences between strata. Risk factors and survival times were analyzed using the Cox proportional hazards model. Explanatory variables identified by univariate analysis to be associated with mortality were subsequently studied in the multivariate model. To determine which specific donor ages were most associated with the increased risk of mortality, a cut-point analysis of donor age was carried out after the Cox proportional hazard analysis. Martingale residuals of the final multivariate model were analyzed using donor age as a predictor. SAS v8.1 software (SAS Institute Inc, Cary, NC) was used for statistical analysis. A p value of less than or equal to 0.05 was considered to be statistically significant.
Donor procurement
Donor assessment was based on a complete clinical and laboratory evaluation and transthoracic echocardiography. Angiography was generally requested in males with an age greater than 45 years, females with an age greater than 50 years, or in younger donors with a heavy smoking history. Angioscopy was occasionally used to assess left main and/or right coronary atherosclerotic disease. All grafts were procured in standard fashion with unmodified cold (4°C) University of Wisconsin solution (UWS) (University of Wisconsin, Madison, WI) which was used for antegrade cardioplegia and storage. Organs with myocardial dysfunction were given triidothyronine (Triostat; Smith-Kline Beecham, Philadelphia, PA) and used only if function improved. Donor–recipient weight mismatch above 30% was generally considered prohibitive, but a greater degree of weight mismatch was occasionally tolerated in the absence of recipient pulmonary artery hypertension.
Surgical technique and postoperative care
The biatrial anastomotic technique described by Lower and Shumway [19] was used until 1993. The majority of transplants performed after this employed bicaval anastomoses. During implantation cold blood antegrade cardioplegia was infused into the graft at completion of each anastomosis and a final warm substrate-enhanced blood cardioplegia dose was used. Concomitant coronary artery bypass was performed in 1 patient and tricuspid valve annuloplasty was performed in 17 other patients.
All patients received cyclosporine and corticosteroids after transplantation. The cyclosporine dose was assessed by radio-immunoassay and was adjusted to the patient's renal function to maintain a blood trough level of 250–300 ng/mL. Additional immunosuppressive agents were used in the majority of patients and use was often protocol-driven. Immunosuppression was not intentionally altered based on donor age. The azathioprine dose was adjusted to a total white blood cell count of at least 4000 cells/µL. Patients undergoing transplant before 1994 received antithymocyte globulin or OKT3 for 2–5 days for early prophylaxis of rejection. After this time period induction therapy was not used routinely unless patients demonstrated intrinsic renal insufficiency before transplant. Endomyocardial biopsies were performed on a routine schedule or when it was considered to be clinically necessary. Acute graft rejection was treated with pulse doses of methylprednisolone. Resistant rejections were treated with cytolytic therapy consisting of OKT3 (5 mg [intravenously] IV daily) or antithymocyte globulin (5–10 mg/kg daily) for 3–7 days.
Early postoperative graft performance was evaluated by echocardiography and right heart catheterization at 1 week. Testing was repeated thereafter as clinically indicated. Coronary angiograms were performed at yearly intervals and substituted with noninvasive testing if renal dysfunction was judged to be prohibitive.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
|
|
|
Survival
Mean follow-up of all patients was 4.5 years. Survival of patients receiving donor hearts at least 50 years of age was significantly reduced (log-rank test, p < 0.001) compared with recipients of donor hearts less than 50 years of age. One-year survival (65.7% [95% CI: 53.2, 78.2] vs 81.7% [78.6, 84.8], p < 0.05] and 5-year survival (48.3% [34.4, 62.3] vs 68.4% [64.5, 72.3], p < 0.05) was lower in patients receiving hearts at least 50 years of age (Fig 2).
|
Recipient cause of death among patients in each group is detailed in Figure 3. Recipients of donor hearts at least 50 years of age died more often of early graft failure (19% vs 7%, p = 0.05) and malignancy (19% vs 5%, p = 0.01) compared with recipients of donor hearts less than 50 years of age. Infection was the most common cause of death in both groups.
|
|
Donor age cut-point analysis
To further characterize the risk associated with donor age, we analyzed the Martingale residuals of the final multivariate model using donor age as a predictor (Fig 5a) and using a nonparametric regression to plot the predicted residuals versus donor age (Fig 5A, 5B) [20]. Specific cut-points ( 15 years, 20 years, 25 years, 30 years, etc) were used to test inflection points after visual inspection of the regression by rerunning the model with each one substituted for donor age (Fig 5C). This validated the most appropriate inflection point as donor age of 40 years. Other donor age cut-points lacked a similar degree of fit. When a donor age of at least 40 years (categorical variable) was subsequently entered into the multivariate equation in place of donor age (continuous variable), it was a significant independent predictor of mortality (HR 1.57 [1.10, 2.24], p = 0.01), with a similar fit to the continuous variable. The cut-point of donor age at least 40 years was a significant predictor of posttransplant mortality, independent of recipient age or rejection index. A Kaplan–Meier survival curve for survival based on subdividing the patients by donor age at least 40 years is illustrated in Figure 6.
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
Despite the increasingly common use of older donor hearts and the consensus conference recommendation that donors more than 55 years may be used in selected recipients, controversy still exists regarding the long-term results of this practice. To date single-center studies [5–8, 10–18] have illustrated equivalent survival with the use of older donor hearts of various ages (35, 40, 45, 50, and 63 years) compared with the use of younger organs whereas multicenter studies [3, 4, 9] have indicated a clear relationship between increasing donor age and increasing mortality. Our relatively large population of transplant recipients of hearts from a wide age range of donors allowed us to independently investigate the contribution of donor age to recipient mortality. We initially chose to specifically address the issue of transplantation of donor hearts at least 50 years of age because of a paucity of previous reports on this age cut-point as well as the increasingly common use of this particular population of organs.
Our 5-year survival rate of 48.3% in patients receiving donor hearts at least 50 years of age was similar to previously published reports by other single centers. Loebe and associates reported their single-institution experience with a large population of donor hearts older than 50 years (n = 167) in which they observed approximately 55% 5-year survival [16]. Blanche and associates reported similar results in their series of 20 recipients of donor hearts at least 50 years of age and observed a 53.1% 5-year survival. Without transplantation it is estimated that the status 1A and 1B wait list patients may otherwise suffer a 58% and 20% annual mortality, respectively [21].
Our survival numbers may have been adversely influenced by the inclusion of the time period 1990–1994 when urgent heart transplantation was the only means for treating acute heart failure refractory to medical therapy. With the Food and Drug Administration (FDA) approval of VADs for "bridge" therapy and most recently for "destination" use, many institutions have changed their patterns of patient selection for transplantation. Moreover as mechanical ventricular assistance becomes an increasingly common method of stabilization for acute cardiac decompensation, survival after heart transplantation with donor hearts of all ages may improve. The beneficial effect of clinical stabilization with VAD before transplantation was seen by Aaronson and associates [22] who reported a 95% survival 3 years after transplantation among patients bridged with LVAD compared with a 65% survival among patients bridged with inotropic agents. Bank and associates [23] performed a similar analysis and determined that patients bridged with VADs experienced improved clinical and metabolic status (as estimated by blood pressure, serum sodium, blood urea nitrogen, and creatinine) at time of transplant.
We observed that recipients of donor hearts at least 50 years of age manifested a higher rate of death in the initial 3 months after transplantation and thereafter suffered approximately the same rate of death (slope) as patients receiving younger donor hearts (Fig 2). Recipients of organs at least 50 years of age were also more likely to die of early graft failure than recipients of donors less than 50 years of age. Although such a pattern of early posttransplant death often suggests a recipient-related effect, we did not find a difference in "severity" of heart failure (as estimated by list status, cardiac output, inotrope requirement, VAD or IABP requirement, ventilator requirement, pulmonary artery pressure, and renal disease) to explain this finding. This early discrepancy in survival may suggest the possibility that older donor hearts may have inherent properties that make them less likely to survive in the early period after transplantation. In an editorial, Young [24] speculated that young age protects the donor heart during the "ravages of catecholamine shower" that accompany brain death. In contrast to our findings, Blanche and associates [17] did not find a higher early rate of death in recipients of donor hearts at least 50 years of age compared with recipients of younger donor hearts. Their results were likely influenced by a higher percentage of status 2 recipients than in our patient population.
When our entire patient population was arbitrarily divided at the donor age of 50 years, we determined that the recipients of donor hearts at least 50 years of age were older than the recipients of younger organs. Both donor age and recipient age were identified by multivariate analysis to be factors considerably associated with diminished survival after transplantation. The calculated rejection index also differed between groups and was determined to be substantially related to survival. Interestingly the survival difference between groups was actually reduced as a result of this particular factor, because recipients of older donor hearts exhibited a lower rejection index. Other intergroup differences existed as well, including donor gender and etiology of heart failure, but there was no marked relationship between these latter factors and survival. This initial analysis, based on the categorical variable of donor age 50 years, was associated with substantial concomitant differences between recipient and donor groups that precluded further characterization of risk incurred at this particular donor age. Although a relationship undoubtedly exists between the age of donors and recipient survival, the inability to quantify the incremental risk with the use of a particular donor population makes the risk–benefit analysis problematic.
We observed a notable correlation between increasing donor age when assessed as a continuous variable and we also observed increased mortality. Therefore we subsequently used a cut-point analysis to determine the donor age at which the greatest incremental risk is incurred. Using this technique after Cox proportional hazard analysis, we determined that the use of a heart from a donor aged 40 years or above was associated with increased mortality, independent of other risk factors. Moreover use of a donor age cut-point of 40 years allowed risk stratification with a fit similar to the use of donor age as a continuous variable. This finding suggests that transplant physicians and surgeons should consider the characteristic of donor age at least 40 years of age when stratifying donor-associated risk.
Although the rate of death after the early posttransplant period was similar between recipient groups, the percentage of survivors receiving older donor hearts was considerably less than those receiving younger hearts. Despite this difference we believe that these older organs should continue to be a viable option for treating heart failure patients. For example, the concept of the "alternate" recipient list proposed by Laks and associates [25, 26] has provided a useful method of transplanting recipients over the age of 70 years or those in need of a third heart transplant. These patients are offered "high-risk" hearts only if the grafts are not able to be placed into any patient on the standard list. They used 62 donor hearts of either older age (median age 45 years) requiring higher levels of inotrope infusions or hearts with regional wall motion abnormalities. Recipient survival was equivalent to those patients transplanted from the standard list. This effort clearly uses organs that would otherwise be rejected and our experience also supports the concept that the use of selected older donor hearts may be justified. However, the risk of transplanting an older donor heart must be balanced with the risk of dying while on the waiting list. Bennett and associates [27] investigated the relative risk of transplanting older hearts into stable patients and demonstrated that after 64 days, patients receiving donor hearts at least 50 years of age exhibited a distinct survival advantage compared with those on the waiting list. Clearly a recipient population in whom placement of older donor hearts will yield the greatest survival benefit must be better defined.
As waiting lists for heart transplantation continue to grow and as VADs and nonsurgical therapies allow for the clinical stabilization of potential recipients with acute decompensated heart failure, changes in practice patterns of donor heart usage are likely. The use of older donor hearts will possibly be necessary as the numbers of available organs continue to decline. Future studies should focus on further classification of the risks and benefits of the use of particular populations of donor hearts in particular recipient populations. It is possible that with continued improvements in donor organ preservation techniques, donor hearts may one day be classified and used according to specific recipient profiles, much as artificial valves and VADs are used today.
In summary in our population of recipients the use of donor hearts at least 50 years of age for transplantation was associated with decreased survival. Because recipients of these hearts exhibited notable differences when compared with recipients of younger donor hearts, the exact contribution of the donor age at least 50 years to reduced survival could not be determined. The use of donor hearts at least 40 years of age predicted survival with the same degree of accuracy as the use of donor age as a continuous variable. These findings suggest that in addition to other common recipient- and donor-related risk factors (VAD dependence, diabetes, repeat sternotomy, ischemic time, etc), the criterion of donor age at least 40 years may be a useful clinical guideline in the risk stratification of donor hearts used for transplantation.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. J. Shea, N. A Sopko, K. Ludrosky, K. Hoercher, N. G. Smedira, D. O. Taylor, R. C. Starling, and G. V. Gonzalez-Stawinski The effect of a donor's history of active substance on outcomes following orthotopic heart transplantation Eur. J. Cardiothorac. Surg., March 1, 2007; 31(3): 452 - 456. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Russo, R. R. Davies, R. A. Sorabella, T. P. Martens, I. George, F. H. Cheema, S. Mital, R. S. Mosca, and J. M. Chen Adult-age donors offer acceptable long-term survival to pediatric heart transplant recipients: An analysis of the United Network of Organ Sharing database. J. Thorac. Cardiovasc. Surg., November 1, 2006; 132(5): 1208 - 1212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Li, K. Tanaka, H. Anzai, B. Oeser, D. Lai, J. A. Kobashigawa, and J. M. Tobis Influence of Pre-Existing Donor Atherosclerosis on the Development of Cardiac Allograft Vasculopathy and Outcomes in Heart Transplant Recipients J. Am. Coll. Cardiol., June 20, 2006; 47(12): 2470 - 2476. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
