|Home | About | Journals | Submit | Contact Us | Français|
Left ventricular hypertrophy (LVH) of the donor heart is believed to increase the risk of allograft failure after transplant. However this effect is not well quantified, with variable findings from single-center studies. The United Network for Organ Sharing database was used to analyze the effect of donor LVH on recipient survival. Three cohorts, selected in accordance with the American Society of Echocardiography guidelines, were examined: recipients of allografts without LVH (<1.1 cm), with mild LVH (1.1–1.3 cm) and with moderate–severe LVH (≥1.4 cm). The study group included 2626 patients with follow-up of up to 3.3 years. Mild LVH was present in 38% and moderate-severe LVH in 5.6% of allografts. Predictors of mortality included a number of donor and recipient characteristics, but not LVH. However, a subgroup analysis showed an increased risk of death in recipients of allografts with LVH and donor age >55 years, and in recipients of allografts with LVH and ischemic time ≥4 h. In the contemporary era, close to half of all transplanted allografts demonstrate LVH, and survival of these recipients is similar to those without LVH. However, the use of allografts with LVH in association with other high-risk characteristics may result in increased mortality.
Heart transplantation provides a remarkable improvement in quality of life and survival in patients with advanced heart failure (1). The selection of appropriate donors for heart transplantation has evolved over the past several decades, guided by our improved understanding of the impact of donor characteristics on recipient survival. Initially, the criteria for an acceptable donor heart allograft had been fairly strict (2), but have expanded over the years to include donors with higher risk characteristics such as higher donor age, mild coronary artery disease, mild left ventricular (LV) dysfunction and left ventricular hypertrophy (LVH) (3-5). Further, as a result of strategies targeted at increasing the availability and utilization of donor organs, individuals of higher age and those with medical comorbidities are now also more likely to become organ donors (5-8). It remains difficult, however, to quantify the possible negative impact of some donor characteristics on recipient survival. With improvements in ventricular assist device (VAD) technology which now provides more stability to individuals awaiting transplantation, or even a possible alternative to transplant through permanent (destination) VAD therapy, accurate understanding of the impact of donor characteristics on the expected post-transplant survival remains very important.
Clinical experience indicates that donor allografts with LVH are more susceptible to ischemic graft injury, and that recipients of these allografts may be at higher risk of mortality. Accurate data regarding this risk are limited; however, some single center studies showed worse outcomes in recipients of donor hearts with LVH, including higher incidence of early graft failure, cardiac allograft vasculopathy and mortality (9,10). Other reports showed no difference in short- and long-term survival between recipients of grafts without LVH and those with mild and moderate LVH (11). The inconsistency of the data is one of the reasons why significant variation, as far as the use of donor hearts with LVH, exists among transplant centers as well as among individual transplant physicians.
To assist in better determination of the effect of donor LVH on post-transplant survival, starting in 2006, the United Network for Organ Sharing (UNOS) has required documentation of LV wall thickness (LVWT) in all heart donors. The aim of our study was to use this newly available information to examine how donor LVH affects survival in heart transplant recipients.
Data from the Organ Procurement and Transplantation Network (OPTN) were used for this analysis. The study population included donor and recipient data for adult heart transplants performed between December 2006 and August 2010, for whom donor echocardiographic data were available. In accordance with the American Society of Echocardiography guidelines (12), we used the greater of two measurements recorded in the database—interventricular septum and posterior wall thickness obtained by echocardiography—to classify the allograft as having no LVH (<1.1 cm), mild LVH (1.1–1.3 cm), moderate LVH (1.4–1.6 cm) and severe LVH (≥1.7 cm).
The study groups of interest were recipients of allografts with no LVH, mild LVH and moderate or severe LVH. Data were summarized using standard statistical descriptors such as means, standard deviations, frequencies and percentages. Comparisons between the study groups were performed using one-way analysis of variance for continuous variables and the Pearson’s chi-square test for categorical variables. The association of the different risk factors with hazards rate of mortality following heart transplantation was assessed separately using a univariable Cox proportional hazards model (13). A multivariable Cox proportional hazards model was used to assess simultaneously the effect of different risk factors on the hazards rate of mortality following heart transplantation. Factors were included in the multivariable model if significance level based on the univariable analyses were p < 0.10. Hazards ratios (HRs) and 95% confidence intervals (CIs) are provided for both univariable and multivariable analyses as measures of strength of association and precision, respectively. Patient’s survival curves were estimated using the Kaplan–Meier method and compared between groups using the log-rank test (14, 15). The survival was censored at the time of death or retransplantation. For patients alive, survival was censored at date of last known follow-up. Twelve patients received heart retransplant during the study period and, for the purpose of this analysis, these events were counted as mortality. Two-tailed p < 0.05 was considered statistically significant. All analyses were performed using STATA software, version 11 (StataCorp LP, College Station, TX, USA).
There were 2910 adult patients who received heart transplantation between December 2006 and August 2010, and who had donor echocardiographic data available. We excluded recipients of donor hearts with left ventricular ejection fraction (LVEF) <50% (n = 38), recipients of multiple organ transplants (n = 114) and patients without follow-up after transplant (n = 132). A total of 2626 patients satisfied our inclusion criteria and formed the study population.
The mean age of the recipients was 52 ± 12 years and 2045 (78%) were males. The mean age of the donors was 33 ± 11 years and 1900 (72%) were males. LVH was present in 1150 (44%) allografts. Mild LVH was present in 1002 (38%), moderate in 125 (4.7%) and severe in 23 (0.9%) of allografts. Due to a small number of patients in the latter cohort, recipients of allografts with moderate and severe LVH were analyzed as one group of 148 patients (5.6%). Detailed baseline characteristics of the three groups of interest are shown in Table 1. The proportion of patients receiving hearts from donors with LVH remained constant during the study period (p = 0.56 across different study periods).
Donors of allografts with LVH were older, were more likely to be of male gender and of a larger body size than donors without LVH. Donors with LVH died more frequently from cerebrovascular accident and required less inotropic support than those without LVH. History of hypertension and diabetes mellitus was present more frequently in donors with LVH. There were no significant differences between the groups in tobacco use, cytomegalovirus (CMV) serologic status and serum creatinine level. In terms of echocardiographic characteristics, donors with moderate-severe LVH had marginally lower ejection fraction than donors with no or mild LVH.
There were no differences in age or gender between the groups. Recipients of allografts with LVH had a larger body size and a higher proportion of diabetes mellitus. There were no differences in history of tobacco use, baseline renal function, CMV serology, history of prior cardiac surgery, mean pulmonary artery pressure or support with a left ventricular assist device (LVAD), extracorporeal membrane oxygenator (ECMO), intra-aortic balloon pump (IABP) or mechanical ventilation. Recipients of allografts with LVH spent shorter time on the waiting list; however there were no significant differences in UNOS status or level of care at the time of transplant (intensive care units: 30%, regular wards: 19% or outpatient: 51%). The donor/recipient ratios for height, weight and body mass index (BMI) were higher for recipients of allografts with LVH and there were no differences between the groups in regards to ischemic time.
During a follow-up of up to 3.3 years (mean follow-up 1.1 ± 0.86 years), 342 deaths (including 12 retransplants) occurred, for a total event rate of 11.8%/year. Recipients of allografts with no LVH, mild LVH and moderate-severe LVH had similar 30-day mortality—4.4%, 5.3% and 4.9%, respectively (p = 0.43) and similar 1-year mortality—12.4%, 13.2% and 13.2%, respectively (p = 0.60). Figure 1 shows Kaplan–Meier survival curves for the three groups. At 3 years after transplant, there was no difference in mortality in recipients of donors with no LVH (21%), mild LVH (26%) and moderate-severe LVH (18%), p = 0.60.
In univariable analyses, donor characteristics associated with all-cause mortality included increasing donor age and history of tobacco use. Donor LVWT (HR 0.95, 95% CI 0.60–1.49, p = 0.83) and the presence of mild (HR 1.07, 95% CI 0.86–1.34, p = 0.52) or moderate-severe LVH (HR 1.03, 95% CI 0.64–1.68, p = 0.89) were not associated with recipient mortality. Recipient predictors of mortality included age >55 years, higher serum creatinine, higher mean pulmonary artery pressure, history of prior cardiac surgery and the need of support with inotropes, LVAD, ECMO or mechanical ventilation (Table 2). To adjust for possible confounders, we constructed a multivariable Cox proportional hazards regression model using a number of donor and recipient characteristics. The donor characteristics that were associated with mortality in this multivariable model were donor age and donor history of tobacco use. Additional recipient characteristics associated with post-transplant mortality are listed in Table 3. Similar to the results of the univariable analyses, the presence of allograft LVH was not associated with higher risk of mortality in the multivariable model (Table 3).
Next we explored whether a combination of specific donor characteristics could have a differential effect on post-transplant survival. Sensitivity analyses for subgroups classified based on the presence of LVH and concomitant history of donor hypertension, or examined by donor gender, were consistent with the main results. However, a subgroup analysis of recipients of allografts with LVH and donor age >55 years showed markedly increased risk of death (mild LVH: HR 6.66, 95% CI 1.43–30.91, p = 0.01; and moderate-severe LVH: HR 6.47, 95% CI 0.58–71.37, p = 0.12). This effect was not found in recipients of allografts with LVH and donor age ≤55 years (mild LVH: HR 1.01, 95% CI 0.81–1.27, p = 0.90; moderate-severe LVH: HR 0.98, 95% CI 0.60–1.62, p = 0.96) (Figures 2A and B). Kaplan–Meier plots shown in Figure 3 demonstrate differences in survival of recipients of allografts with and without LVH, for donors ≤55 years of age and for donors >55 years of age. Allograft ischemic time also showed interaction with donor LVH as far as effect on mortality after transplant. Recipients of allografts with moderate-severe LVH and allograft ischemic time ≥ 4 h were at increased risk of death (moderate-severe LVH: HR 2.23, 95% CI 1.01–4.93, p = 0.04, mild LVH: HR 1.16, 95% CI 0.77–1.73, p = 0.47)—Figures 2C and D. The differences in mortality among the different LVH groups stratified by allograft ischemic time are illustrated through a Kaplan–Meier plot in Figure 4.
Previous investigations that explored the effect of donor LVH on post-transplant survival provided conflicting results (9-11,16-19). Therefore, it remains controversial whether donor cardiac allografts that demonstrate evidence of LVH on echocardiography can be used safely for transplantation. The current International Society for Heart and Lung Transplantation guidelines cite level of evidence C for a class IIa recommendation stating that “it would seem appropriate to use hearts from donors with LVH and LV wall thickness <14 mm provided that it is not associated with ECG findings of LVH” (20). The cautious tone of this recommendation is reflection of the limited body of evidence to guide clinical decisions. Our study addresses this issue in a large study group from a multi-institutional national registry. The inclusion of LVWT as a required donor data field in the OPTN database starting in 2006 allowed us to explore the effect of donor LVH on posttransplant survival in a significantly more detailed fashion than was previously possible. The main findings of our study show that LVH is highly prevalent in the contemporary donors, with 44% of donor allografts showing at least a mild degree of LVH. Our data further suggest that, with current approaches to donor selection and allocation, post-transplant survival is favorable in most recipients of allografts with LVH. However, we also show that when donor LVH is combined with additional high risk characteristics—older donor age or prolonged ischemic time, donor LVH can have a profound effect on post-transplant mortality.
Historically, donor LVH has been associated with primary graft failure and increased mortality in the first 30 days after transplant (1). This was noted in a study by Aziz et al. who documented the use of allografts with LVH in 8% of transplants and reported on an increased incidence of early allograft dysfunction compared with patients receiving allografts without LVH (33% vs. 3%) (9). Similarly, Marelli et al. documented the use of allografts with LVH in 14% of transplants and found decreased early and mid-term survival in recipients of allografts with LVH (16). In contrast with these reports, a more contemporary single-center study by Goland et al. did not find differences in early survival of recipients of allografts with mild or moderate LVH (11). Our study expands on this knowledge by documenting a substantial increase in the use of cardiac allografts with LVH (44%). We also found no evidence of early graft failure in recipients of allografts with and without LVH and no difference in the 30-day mortality rates in these patients.
The examination of the impact of allograft LVH on intermediate and long-term post-transplant outcomes has also lead to varying results. Kuppahally et al. found that the presence of donor LVH (LVWT >1.4 cm) was associated with reduced survival (50% vs. 82%) and higher incidence of allograft vasculopathy (50% vs. 22%) at 3 years after transplant. Goland et al., on the other hand, reported similar 1-year and 5-year survival rates in recipients of allografts with no LVH, mild LVH and moderate LVH. Our results support the findings of this latter study, demonstrating a similar 1- and 3-year survival in recipients of allografts with LVH.
In an attempt to determine the effect of donor LVH in the context of various other clinical characteristics, we explored possible interactions between donor LVH and additional characteristics that could negatively impact post-transplant survival. We found that the use of allografts with LVH from older donors was associated with a six-fold increase in risk of post-transplant mortality. We also show that recipients of allografts with LVH where allograft ischemic time was 4 h or more had a twofold increase in the risk of mortality compared with recipients of allografts without LVH.
How do we explain improved survival in recipients of donor hearts with LVH in this contemporary cohort? It is likely that advances in surgical, technical and preservation techniques that have resulted in improved survival in the first year after transplant have had a large impact on the expected survival of recipients receiving allografts with LVH (1). Careful selection of donors and meticulous matching of donors and recipients has also likely played an important role. Our finding of favorable posttransplant survival in the recipients of allografts with LVH is therefore an important validation of contemporary approaches in donor organ selection and allocation.
We recognize that there are limitations to our study. This was a retrospective analysis of a nationwide clinical registry. Although the UNOS data collection is rigorous and undergoes periodic audits, some errors in data entry may be present. On the other hand, the fact that our multivariable model identified accepted mortality risk factors is reassuring and supports the validity of our findings. The decision to use a particular organ for a patient was not random. Donor organs with moderate-severe LVH represented only 5.6% of total donors, most certainly a result of a higher rate of decline of organs with LVH offered for transplantation.
This selection bias cannot be obviated and needs to be considered when interpreting our results. Even though there were no differences in UNOS status and physical location of the recipients between the groups, it remains difficult to determine whether those recipients of allografts with LVH and other higher risk characteristics were sicker than recipients without those higher risk characteristics, thus affecting our results. The echocardiographic measurements of wall thickness were performed by cardiologists at different donor institutions and not by a single echocardiographer or a group of echocardiographers.
In the contemporary era, nearly half of all heart donors have LVH. The overall survival of recipients of donor hearts with LVH is similar to those without LVH. This would suggest that current donor selection and allocation algorithms successfully mitigate the risk donor LVH could pose to recipient survival. However, the combination of donor LVH with certain high-risk characteristics such as higher donor age and prolonged ischemic time is still likely to result in excess mortality.
This work was supported in part by Health Resources and Services Administration contract 231-00-0115. The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
Funding source: None.
The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.