Author + information
- Received July 25, 2003
- Accepted August 13, 2003
- Published online March 3, 2004.
- Mandeep R. Mehra, MD, FACC*,* (, )
- Patricia A. Uber, pharmD*,
- Hector O. Ventura, MD, FACC*,
- Robert L. Scott, MD, FACC* and
- Myung H. Park, MD*
- ↵*Reprint requests and correspondence:
Dr. Mandeep R. Mehra, Cardiomyopathy and Heart Transplantation Center, Ochsner Clinic Foundation, 1514 Jefferson Highway, New Orleans, Louisiana 70121, USA.
Objectives We evaluated the association of mode of brain death with cardiac allograft vasculopathy.
Background Explosive brain death (EBD) is accompanied by a sudden increase in intracranial pressure, with recruitment of pro-inflammatory cytokines, as well as adhesion cell and co-stimulatory molecules. Whether these early events influence the later development of cardiac allograft vasculopathy following heart transplantation remains unknown.
Methods An inception cohort of 61 consecutive heart transplant recipients between 1993 and 1995 who underwent intravascular ultrasound examination of the coronary arteries were evaluated. Based on the mode of donor brain death, this cohort was divided into either an EBD group (n = 27) or non-EBD (n = 34), and the development of intimal thickness and cardiac events (sudden cardiac death, myocardial infarction, and need for coronary revascularization via percutaneous techniques or surgical bypass) was assessed.
Results Despite similar posttransplant survival and distribution of nonimmunological and immunological variables, heart transplant recipients with EBD demonstrated greater intimal thickening (0.59 ± 0.1 vs. 0.32 ± 0.2 mm; p = 0.02) and higher cardiac events (37% vs. 12%; p = 0.01) when compared to those with non-EBD donors. Hearts from donors with EBD had lower survival (63 ± 19 vs. 72 ± 17 months) than with non-EBD donors (p = 0.04).
Conclusions Explosive brain death is a significant determinant for the late development of cardiac allograft vasculopathy and influences long-term allograft survival. Thus, strategies focusing on limitation of vascular allograft injury in the pre-engraftment phase of cardiac transplantation are warranted.
Investigations in experimental models of brain death and in potential donors for cardiac transplantation have demonstrated that brain death is associated with pathological effects on the myocardium that adversely influence cardiac performance (1). Indeed, some investigators have postulated that the process of brain death is accompanied by inflammatory processes that contribute to diffuse endothelial dysfunction and that this sentinel event may adversely influence the outcome of engraftment (2). Other animal studies have pointed to the fact that the pathological consequences as a result of mode of brain death might adversely influence cardiac function. Shivalkar and colleagues (3)have demonstrated in canine experiments that dogs with a nonexplosive rise in intracranial pressure showed fewer cardiac hemodynamic perturbations and less myocardial injury than did those animals subjected to a rapid and more explosive rise in intracranial pressures.
Cardiac allograft vasculopathy is associated with a relentless course after heart transplantation, and both nonimmunological and immunological interventions have only partially modified its natural history (4,5). Recent investigations have alluded to early posttransplant rejection-independent microvascular aberrations that predict the late development of posttransplant coronary artery disease (6,7). The purpose of our investigation was to evaluate the relationship of mode of donor brain death with incidence and severity of cardiac allograft vasculopathy in heart transplantation.
The study population consisted of a cohort of 61 consecutive heart transplant recipients between 1993 and 1995 who underwent examination of the coronary arteries using intravascular ultrasound. These patients were divided by mode of donor brain death into either group I: non-explosive brain death (EBD) (n = 34), or group II: EBD (n = 27). Explosive brain death was defined as a traumatic brain death that resulted from a gunshot wound to the head (n = 9), accidental head trauma (n = 5), or a clearly definable intracranial bleed that progressed rapidly to brain death (n = 13). All others were classified as “non-EBD” cases. Adjudication of the mode of brain death was performed by an investigator blinded to all clinical and ultrasound examination data. All transplant recipients studied provided informed consent, and the institutional review board approved the study.
Nonimmunological, metabolic, and infection variables
Fasting lipid profile (triglyceride, total cholesterol, low-density lipoprotein-cholesterol, and high-density lipoprotein-cholesterol levels) at the time of ultrasound examination, percentage weight gain post-transplant, hypertension (defined according to the Joint National Commission [JNC] VI criteria), diabetes mellitus, donor age and gender, cold ischemic time, time since transplantation, and cytomegalovirus infection (defined as a clinical syndrome in association with either new seroconversion, positive blood cultures, or evidence of tissue invasion) requiring therapeutic intervention with intravenous gancyclovir were assessed in all cardiac transplant recipients.
Immunological risk factors
Rejection surveillance during the first year consisted of weekly endomyocardial biopsies for the first month following transplantation, bi-weekly until the third month, monthly until the sixth month, and bi-monthly thereafter until completion of the first year. Thus, the routine protocol mandates performance of 13 to 14 biopsies during the first year. All biopsies were graded on the basis of the standard biopsy-grading scheme developed by the International Society for Heart and Lung Transplantation (ISHLT). The criteria for treatment of cellular rejection included the finding of a biopsy grade of 3A or higher. If treatment was initiated for a lower biopsy grade, it was done only in the presence of hemodynamic compromise as evidenced by one or more of the following: a 20% decrease in the ventricular ejection fraction, increase in the pulmonary capillary wedge pressure >25%, and a decrease by 25% in the cardiac index or an absolute value of <2.0 1/min/m2. Furthermore, episodes of treated rejection mandated the performance of a follow-up biopsy within one to two weeks for documentation of resolution or regression to a lower rejection grade.
Additionally, a first-year mean biopsy rejection score as previously described by us was determined for all patients (8). This was adapted from the standardized ISHLT criteria, and individual scores were assigned to each defined biopsy grade as follows: ISHLT grade 0 = 0; grade IA = 1; grade IB = 2; grade 2 = 3; grade 3A = 4; grade 3B = 5; and grade 4 = 6. The mean biopsy score was computed as the average of all biopsy scores during the first year following transplantation.
In addition, episodes of vascular (humoral) rejection and the number of HLA-A, -B, or -DR matches between the donor and recipient were assessed in all patients. Acute vascular rejection was defined by the presence of hemodynamic compromise along with histological evidence of endothelial cell activation and a positive immunofluorescence on biopsy. More importantly, we also assessed detailed immunosuppressant regimens in all patients and calculated the cumulative prednisone dose (total amount of IV and oral prednisone consumed in grams [g]), average daily prednisone dose (mg/kg/day), mean cyclosporine dose (mg/kg/day), and mean cyclosporine levels (pg/ml), as well as mean azathioprine dose (mg/kg/day). Use of cytolytic therapy (OKT3 or anti-thymocyte globulin) and use of methotrexate were also assessed for each study patient.
Coronary ultrasound procedure
Coronary ultrasound was performed as we have previously reported (8). After administration of 200 μg of intracoronary nitroglycerin, the ultrasound catheter was positioned in the distal segment of the target vessel over a 0.014-inch (0.036-cm) guide wire. The target vessel was selected by the presence of at least three easily definable and reproducible branch points to assist in the accurate and serial assessment of three regions of interest. The guide wire was removed, and the ultrasound catheter was advanced to the distal end of the ultrasound sheath under fluoroscopic guidance. The drive module was then engaged, and continuous images of the coronary artery were obtained as the ultrasound transducer was slowly withdrawn. Additionally, fluoroscopic pictures and audio annotations were used to ensure the correct localization of the artery segment for subsequent off-line analysis. Guiding catheter pressure, ST-segment changes, and cardiac rhythm were continuously monitored during the procedure. After ultrasound images were obtained, both the transducer and sheath were removed, and a final angiogram was obtained to confirm the patency of the coronary vessel. No complications were encountered except for coronary artery vasospasm readily reversed by intracoronary infusion of nitrates or verapamil.
Coronary ultrasound assessment
Intravascular ultrasound measurements were performed by one of the investigators who had no knowledge of the prior assessment of risk factors. Three coronary sites per vessel (proximal, mid, and distal), for a total of 183 coronary artery segments, were evaluated. The left anterior descending artery was examined in 54 patients, circumflex artery in 2, and right coronary artery in the remaining 5 subjects. Maximal intimal thickness (measured as the mean of the maximal thickness at the three sites evaluated) was obtained by tracing the lumen vessel wall interface and the external border of the intimal layer. The definition of severe intimal thickness was based upon the scheme previously described by St. Goar et al. (9), wherein severe intimal thickening is represented by >0.5 mm of intimal proliferation involving >180° of the vessel circumference or any intimal layer >1.0 mm in any one area of the vessel circumference. This threshold of severity denotes “unequivocal” presence of intimal thickening and is beyond the 98th percentile of any published “normal” values. Moreover, we have previously demonstrated that this threshold of severe intimal proliferation by intravascular ultrasound is predictive of cardiac events even in the absence of angiographic abnormalities, and therefore denotes a “prognostically relevant” threshold of abnormality (10).
Cardiac events were defined as the occurrence of sudden cardiac death, myocardial infarction (MI), and need for coronary revascularization via percutaneous techniques (angioplasty, atherectomy, stent implantation) or surgical bypass.
Normally distributed data are reported as mean ± SD. Differences between categorical variables were assessed using either the chi-square or the Fisher exact test. An unpaired Student ttest was used to define differences between continuous variables in subgroups. Kaplan-Meier actuarial analysis was used to assess event-free survival for the study cohort as a function of mode of brain death. All analyses were performed using Statview 4.5 (Abacus Concepts, Berkeley, California) statistical software, and statistical significance was set at a p value <0.05.
The study population included 61 consecutive heart transplant recipients (49 men and 12 women) with a mean age of 53 ± 8 years (range: 21 to 69 years) who were followed for an average of four years after the index ultrasound (range of one to eight years of follow-up after index intravascular ultrasound). The index intravascular ultrasound was performed at an average of 2.4 years' posttransplantation (1 to 3.4 years' posttransplantation). All subsequent data are presented between the two groups identified as group I: non-EBD (n = 34) or group II: EBD (n = 27).
Differences in nonimmunological and infection variables
Table 1lists comparative nonimmunological characteristics of the non-EBD (group I) and EBD (group II) cohorts with regard to recipient and donor age, recipient and donor gender, cold ischemic time, lipid characteristics, weight gain, cytomegalovirus (CMV) infection, and diabetes mellitus. Specifically, no significant clinical differences existed between the two groups with reference to any of these nonimmunological variables except for nonsignificant trends of an older recipient age and more CMV infections in the EBD group. More importantly, no differences in donor and recipient CMV serological mismatches were noted.
Differences in immunological variables
Table 2lists a comparison of immunological risk factors including episodes of allograft rejection, numbers of human leukocyte antigen (HLA) matches, and mean first-year biopsy rejection score. No significant differences were observed among clinical episodes of allograft rejection, either cellular or humoral, hemodynamically compromising rejection, or number of HLA matches. Similarly, no significant differences existed between the two groups with regard to average daily cyclosporine dose, mean azathioprine daily dose, as well as cumulative exposure to corticosteroids for the life of the allograft until the index intravascular ultrasound. No significant differences existed between cytolytic antibody exposures between the two groups over the transplant duration.
Intravascular ultrasound and survival analysis
Heart transplant recipients with allografts from EBD donors demonstrated greater degrees of maximal intimal thickness (0.59 ± 0.1 mm vs. 0.32 ± 0.2 mm, p = 0.02) and more cardiac events (37 vs. 12%, p = 0.01) in late follow-up when compared to those with non-EBD donors (cardiac events described below) (Fig. 1). The predominant morphologic characteristics of the vessels studied revealed diffuse and concentric lesions (at least 2 or 3 segments) in 82%, whereas only 18% of vessels studied revealed proximal, focal, and eccentric lesions. The intraobserver (4.1%) and interobserver (5.4%) variability was assessed as previously reported, and demonstrated good test reproducibility (8,10).
Ten cardiac events occurred in the EBD group (two percutaneous coronary interventions, three MIs, three sudden deaths, and two deaths due to allograft failure). The non-EBD group suffered four events (one percutaneous coronary intervention, one sudden death, and two deaths related to late allograft failure). Allografts from EBD donors demonstrated lower survival (63 ± 19 months vs. 72 ± 17 months) compared with non-EBD donors (p = 0.04). Actuarial survival analysis confirmed the significant predictive impact of EBD and severe intimal thickening on cardiac events (log rank, p = 0.01) (Fig. 2).
In a multivariate model, EBD (r = 0.28; p = 0.02) persisted as an independent predictor of intimal thickness in concert with donor age (r = 0.42; p = 0.008) and lower average daily cyclosporine dose (r = −0.25; p = 0.05).
The principal findings of this investigation demonstrate that cardiac allograft vasculopathy is influenced by donor events that precede the actual engraftment process. Indeed, the notion that allograft vasculopathy is predominantly determined by immunological interactions is definitively challenged by these observations, which suggest that “pre-engraftment” factors such as the mode of brain death likely lay the groundwork for the later development of cardiac allograft vasculopathy. Furthermore, this investigation points out that not only is the mode of donor brain death associated with an increase in the development of severe intimal proliferation but also that this finding is predictive of adverse late clinical outcome from cardiac events.
Early events and cardiac allograft vasculopathy
The recognition that rejection-independent events could influence the development of cardiac allograft vasculopathy became obvious by investigations that pointed to the role of ischemia-reperfusion injury surrounding engraftment, the contribution of advanced donor age, and aberrations within the microvasculature, including such metabolic derangements as hyperlipidemia (6–8,11). Several important investigations have linked microvascular fibrin, depletion of arteriolar tissue plasminogen activator factor, and even systemic inflammation, all rejection-independent events, as significant correlates of the future risk for cardiac allograft vasculopathy and allograft loss (6,7,12). Does the process of brain death and its severity play an integral role in determining these early allograft findings that influence the development of cardiac allograft vasculopathy?
Brain death and coronary effects
The contribution of brain death to the eventual development of cardiac allograft vasculopathy was pointed out recently by Anyanwu et al. (13), who studied the incidence and severity of transplant-associated coronary artery disease (CAD) in recipients of domino hearts (living-related heart transplantation from recipients who require heart-lung transplantation). These investigators suggested that such allograft recipients suffered less development of angiographic CAD compared to cadaveric heart transplantation. It has been demonstrated that rapid brain death induces a calcium overflow injury that affects conduction tissue, coronary artery smooth muscle, and myocardial cells (14). Others have shown that the event of EBD is associated with a generalized activation of macrophage-associated cytokines in all peripheral organs including the heart. An investigation by Takada et al. (15)also demonstrated an up-regulation of immunoregulatory and adhesion cell molecules, arguing that the event of brain death sets up the allograft for adverse donor-host interactions following engraftment. More recently, Segel and colleagues (2)evaluated the impact of brain death on measures of endothelial dysfunction and showed the up-regulation of adhesion cell molecules and pro-inflammatory cytokines, suggesting a state of endothelial inflammation in brain death even if the circulation remains perceptibly stable. Szabo et al. (16,17)have demonstrated similar events in a canine model of brain death wherein they found that brain death affects coronary circulation by impairment of coronary blood flow and by the induction of severe endothelial dysfunction.
Mode of brain death and outcomes
It has been known that the severity of brain death is related to hemodynamic instability in the donor and can be reversible. Thus, it is unlikely that permanent myocardial injury is a conclusive event, and Szabo et al. (16,17)have suggested that the hemodynamic instability likely reflects altered loading conditions and impaired coronary perfusion than neurohormonally mediated myocardial injury. A provocative investigation by Birks et al. (18)demonstrated an elevation in pro-apoptotic capsases in dysfunctional donor hearts, which they linked to inflammatory activation, thereby alluding to a potential cellular mechanism of subclinical cardiac pathology.
Comparison with other published studies
Clinical confirmation of these pathological observations has only been sparsely forthcoming. Tsai et al. (19)hypothesized that atraumatic intracranial bleeds, typically associated with older donors, might be associated with increased risk of graft failure in heart transplantation. These investigators suggested that an intracranial bleed might represent an independent risk for poor long-term outcomes; furthermore, they demonstrated that such donor allografts have a higher risk for development of angiographic CAD than those without. More recently, Yamani et al. (20)examined the risk of CAD in relation to the presence of spontaneous intracranial bleeds in the donor. These researchers found a lower freedom from CAD at five years coupled with worse survival in such individuals. Our investigation further amplifies these data because we have been able to demonstrate that it is an explosive mode of brain death that likely arbitrates this adverse outcome rather than a specific attribute confined to intracranial bleeding. Of note, the degree of intimal proliferation observed by intravascular ultrasound crossed the threshold of prognostically relevant intimal thickening, as in previous studies (4,5,8). Furthermore, our data establishes the independent impact of mode of EBD on late outcomes from cardiac allograft vasculopathy, independent of other clinical variables, since most immunological and nonimmunological variables were similarly spread across the two groups evaluated. Interestingly, our data once again, as in the past, confirmed the importance of not only older donor age but also lower exposure to calcineurin inhibitors as other important factors in determining late prognosis from cardiac allograft vasculopathy (21).
Our investigation enrolled those patients who underwent a one-year intravascular study resulting in an exclusion of early deaths. This culminated in an exclusion of six patients (allowing only 61 evaluable patients) who were unable to achieve survival to intravascular ultrasound. Interestingly, four of these six patients were classified as allografts from donors with an EBD. Thus, inclusion of these patients would only have strengthened the adverse prognostic implication of mode of brain death. Another potential limitation pertains to the low use of lipid-lowering therapy in our two groups, but this reflects the time of patient recruitment when the value of lipid-lowering therapy in heart transplantation was just emerging.
Finally, this investigation suggests that explosive mode of brain death, a pre-engraftment factor, is a significant determinant for the late development of cardiac allograft vasculopathy and adversely influences long-term allograft survival. Thus, strategies that focus on limitation of vascular allograft injury in the pre-engraftment phase of cardiac transplantation are warranted and should be developed.
- coronary artery disease
- explosive brain death
- human leukocyte antigen
- International Society for Heart and Lung Transplantation
- Joint National Commission
- Received July 25, 2003.
- Accepted August 13, 2003.
- American College of Cardiology Foundation
- Shivalkar B.,
- Van Loon J.,
- Wieland W.,
- et al.
- Rickenbacher P.R.,
- Pinto F.J.,
- Lewis N.P.,
- et al.
- Mehra M.R.,
- Ventura H.O.,
- Chambers R.,
- et al.
- St. Goar F.G.,
- Pinto F.J.,
- Alderman E.L.,
- et al.
- Yamani M.H.,
- Tuzcu E.M.,
- Starling R.C.,
- et al.
- Pethig K, Heublein B, Kutschka, Haverich A. Systemic inflammatory response in cardiac allograft vasculopathy: high sensitivity C-reactive protein is associated with progressive luminal obstruction. Circulation 2000;102 Suppl III:233–6.
- ↵Yamani MH, Starling RC, Tuzcu EM, et al. Donor intracranial bleeding is associated with advanced transplant coronary vasculopathy: evidence from intravascular ultrasound (abstr). J Heart Lung Transplant 2003;22:S76.