Author + information
- Received August 9, 2000
- Revision received November 14, 2000
- Accepted December 15, 2000
- Published online April 1, 2001.
- ↵*Reprint requests and correspondence: Dr. Paul J. Hauptman, Division of Cardiology, St. Louis University Health Sciences Center, 3635 Vista Avenue at Grand Boulevard, St. Louis, Missouri 63110
This retrospective review of organ donor records was designed to evaluate the practice of donor angiography in one organ procurement organization and determine the outcomes of angiography and its impact on the timing of the organ donation process.
Concerns about transmission of atherosclerosis from donor to recipient have been heightened by the increasing prevalence of older donors. Guidelines that advocate the use of angiography in specific settings have been published, but no formal large-scale review has been performed.
For the period January 1993 through June 1997, we reviewed all New England Organ Bank records of donors between the ages of 40 and 65 including any from whom at least one solid organ was procured. Data abstracted included the presence of risk factors, timing of the evaluation process and angiographic findings.
Coronary angiography was performed in 119 donors aged 40 and older; 64.7% of these hearts were transplanted. Thirty-eight hearts were transplanted from donors not subjected to angiography and outcomes were poorer compared with donors who underwent angiography. Advanced donor age was the only significant predictor of coronary artery disease. The duration of the procurement process was not prolonged by the performance of angiography.
Donor coronary angiography does not complicate the donation process. Older donor age is the most powerful predictor of coronary artery disease and may explain prior observations of poorer outcome with older donor hearts. These factors should be considered when angiography is performed as part of the heart donor evaluation.
“It makes little sense to replace one diseased heart with another.” —DePasquale and Burch, Am Heart J1969;77:719.
The success of heart transplantation (1)and the widening gap between the number of patients waiting and the number of hearts available for transplantation (2)have focused attention on the need for an expanded supply of donor hearts. In addition, the demographic profile of potential donors has changed over the past decade (3,4), with an increasing proportion of donors dying from nontraumatic causes. These donors tend to be older than those who die in motor vehicular and other traumatic accidents. Although even young donors can have occult coronary atherosclerosis (5), concern about transmission of atherosclerosis from donor to recipient is heightened with increasing donor age (6–8).
Some (1,9–15), but not all (6,16–20), studies have suggested that advanced donor age is a risk factor for posttransplant atherosclerosis and/or poorer posttransplant outcome. Thus, the standard evaluation of the potential cardiac donor—cardiac history, electrocardiogram and echocardiogram—is focused, to a great extent, on determining whether significant coronary disease is likely to be present (10,21). The evaluation process can be made more challenging by the presence of (often transient) abnormalities on donor electrocardiograms and echocardiograms that may occur secondary to the effects of brain death but mimic findings associated with acute or chronic ischemic heart disease (22).
In an effort to minimize the risk of transmitting atherosclerosis, potential donors sometimes undergo coronary angiography if a catheterization laboratory and personnel are available. Several groups have advocated the use of angiography in older donors (10,23). A task force of the American College of Cardiology (24)recommended coronary arteriography for male donors over the age of 45 and for female donors over the age of 50. It was also suggested that this age threshold be lowered by 5 to 10 years in the presence of cardiac risk factors. Despite these recommendations, there is considerable uncertainty about the role for angiography in potential cardiac donors. To help resolve this uncertainty, we analyzed the experience of the New England Organ Bank, the organ procurement organization for much of the New England area (25). Beginning in January 1993, the adult heart transplant programs in New England worked in collaboration with the organ bank to request coronary angiography in donors felt to be at risk for significant coronary artery disease (CAD).
The goal of this study was to describe and evaluate the actual experience with angiography of potential donors in one organ procurement organization. Specifically, we aimed to: 1) describe the characteristics of the donors undergoing angiography; 2) describe the results of angiography, the predictors of finding coronary disease and the predictors of performing angiography; 3) assess the impact of angiography on the duration of the donor evaluation process; 4) compare current practice with the recommendations of the published guideline and determine whether logistics with, or availability of, angiography limited its use; and 5) determine whether performing angiography impacts early posttransplant survival.
Subjects were potential heart donors evaluated by the New England Organ Bank between January 1, 1993, and June 30, 1997. The service area encompassed Maine, New Hampshire, Rhode Island, most of Massachusetts and Vermont and a large section of Connecticut, with a total population of approximately 11.1 million people. During the study period, there were five heart transplant programs (four adult and one pediatric) averaging 79 transplants per year. Donor hearts imported into the region were not considered in this analysis.
We restricted the review to donors aged 40 to 65 years inclusive who donated at least one solid organ (kidney, liver, pancreas, lung and/or heart). This age range was used in order to include donors who would be considered angiography candidates according to the American College of Cardiology (ACC) Guideline. Potential organ donors who were evaluated but not used were not considered in this analysis. The donors were assigned to four subgroups according to whether an angiogram was obtained (angiography performed = YA; angiography not performed = NA) and whether the heart was procured and transplanted (heart transplanted = YH; heart not transplanted = NH) as follows: 1) donors who had coronary angiography performed as part of the donor evaluation and whose hearts were transplanted (YA/YH); 2) donors who had coronary angiography performed but whose hearts were not subsequently transplanted (YA/NH); 3) donors who did not undergo angiography and whose hearts were not transplanted (NA/NH); and 4) donors who were not subjected to angiography but whose hearts were transplanted (NA/YH).
A potential heart donor was classified as lacking consent if there were either family refusal to provide consent or a formal determination at the time of donor evaluation by the transplant coordinator and medical director of the New England Organ Bank that heart donation was not medically or logistically feasible (in which case, consent was not requested).
In addition, we abstracted the charts from 15 donors <40 years of age who had an angiogram performed in order to summarize the entire experience with angiography regardless of donor age.
Permission to review the charts of the above patients was granted by the Medical Oversight Board of the New England Organ Bank.
Data from medical records of potential donors
All donor charts were reviewed for information in several areas: clinical characteristics and medical history; time of brain death declaration, angiography (if performed) and aortic cross-clamp; results of laboratory and noninvasive evaluation (including electrocardiogram and echocardiogram); medications used during donor management; cause of donor death and the stated reason (if any) for donor unsuitability.
Hypertension in the donor was defined as a history of, or treatment for, high blood pressure for at least one year preceding brain death. Diabetes was defined as hyperglycemia requiring treatment with oral hypoglycemic agents or insulin prior to the terminal admission. A history of past or ongoing nicotine abuse qualified the donor as a cigarette smoker. In most cases, we did not have information about the presence or absence of a family history of CAD. Cardiopulmonary resuscitation (CPR) was recorded if chest compressions were required during the initial phase of care or during subsequent donor management. Arrhythmias detected during donor management were classified and coded as either supraventricular or ventricular. Use of a vasopressor or inotrope was defined as the infusion of an intravenous agent other than low dose dopamine (<5 μg/kg/min) to support donor blood pressure or cardiac output. Cause of donor death was classified as either from central nervous system causes (cerebrovascular accident, intracranial hemorrhage of any type or primary brain tumor), traumatic brain injury or miscellaneous (asphyxiation, poisoning, other).
With regard to the results of coronary angiography, obstructive CAD was deemed to be present if at least one main artery or branch was reported to have a ≥50% stenosis; nonobstructive disease was present if all lesions were <50%. A qualitative report of “significant,” or “moderate” or greater disease severity was coded as obstructive CAD.
Duration of the evaluation process was obtained from transplant coordinator records detailing the time from brain death declaration to angiography and to aortic cross-clamp during cardiectomy.
Availability of cardiac catheterization facilities
All donor hospitals were contacted to confirm the presence or absence of catheterization facilities during the study period.
Evaluation of survival was limited to the recipients transplanted at programs within the region of the New England Organ Bank. The outcomes of donor hearts exported to transplant centers outside the region were not included. Cause of death was provided by each participating transplant program according to standard definitions used by the United Network for Organ Sharing.
Descriptive statistics were first calculated for the four donor groups. The chi-square and Fisher exact tests were used to determine whether significant associations existed between the donor subgroups for selected donor characteristics. Analysis of variance was used to detect differences in age between the donor groups. The Cochran-Armitrage trend test for linear proportions was calculated to detect time trends in the utilization of angiography and in the rate of procurement of hearts in donors undergoing angiography. Second, logistic regression models were developed to measure the relation of selected factors with the presence of CAD and with the decision to perform angiography. Deviance and Person residuals, and the Hosmer-Lemeshow statistic, were all examined to assess adequacy of model fit. Third, analysis of variance and ttests were utilized to detect differences between elapsed time comparisons and donor group; Tukey’s studentized range test was employed to isolate differences resulting in the analysis of variance. Fourth, survival analysis of recipients of hearts from donors who did and did not have angiography was performed using the Kaplan-Meier method and the Wilcoxon test. Differences in cause of death between groups were analyzed with the Fisher exact test. All analyses were executed with SAS software (Release 6.12, SAS Institute, Cary, North Carolina). Statistical significance was defined as p < 0.05.
Descriptive characteristics of potential donors
There were 364 potential heart donors between the ages of 40 and 65 inclusive during the 4.5-year interval under study (Fig. 1). The medical record for one non-heart-beating donor was not abstracted. Of the remaining 363 donors, 119 underwent coronary angiography; results were available for 118. Seventy-seven hearts (or 64.7%) were subsequently procured and transplanted (YA/YH); 42 were not procured (YA/NH). Hearts were transplanted without angiography from 38 donors (NA/YH). Of the remaining 206 donors, consent for heart donation was not obtained in 67; none of the remaining 139 underwent angiography, nor were their hearts transplanted (NA/NH).
Excluding the group without consent, the donors in this series were evaluated at 74 different hospitals (range 1 to 24 donors per hospital). Fifty-four of these hospitals had angiographic facilities at the time of donor evaluation. The percentage of donors at hospitals with catheterization facilities was 88.8%.
Baseline characteristics of the donors are shown in Table 1. There were significant differences in age, performance of CPR and smoking history but not cause of death or other cardiac risk factors.
The use of angiography in donors 40 years of age and older increased over the study period when analyzed as a percentage of donors (1993, 30.0%; 1994, 36.1%; 1995, 40.9%; 1996, 48.5%; 1997, 46.3%; p = 0.024). However, among those donors who were studied by angiography, the percentage whose hearts were procured did not increase significantly (1993, 61.1%; 1994, 63.6%; 1995, 66.7%; 1996, 63.6%; 1997, 68.4%; p = NS).
Findings on donor angiography
More donors were found to have CAD of any severity in the YA/NH group (28/42, 66.7%) compared with the YA/YH group (22/76, 28.9%). Obstructive disease, defined by a finding of a lesion ≥50% in a main coronary artery, was present in 33.3% (n = 14) and 0% of donors in the two groups, respectively. Among the donors in the YA/NH group who did not have obstructive CAD, six had nonobstructive CAD in two or more vessels; two had coronary artery spasm in the absence of obstructive disease; and one had multiple coronary arteriovenous fistulae. Other reasons for unsuitability of YA/NH donors included abnormal wall motion on ventriculography (n = 11), abnormalities noted on open visual inspection in the operating room (n = 5), abnormal hepatitis serology (n = 1) and progressive hemodynamic instability (n = 1). We could not determine the reason in one donor.
A small coronary-to-pulmonary artery fistula was noted in one donor on angiography; this heart was transplanted. Four hearts from potential donors who did not undergo angiography were rejected intraoperatively on the basis of surgical palpation of the coronary arteries.
No vascular or other complications of angiography were recorded during the subsequent final stages of the procurement process.
Predictors of finding donor CAD
The sole predictor of finding obstructive or nonobstructive CAD in donors undergoing angiography was increasing donor age. Using a donor age of 40 to 44 years as the reference group, the odds ratio for finding disease in donors aged 45 to 49 years was 4.22 (95% confidence interval [CI] of 1.35 to 13.2). For donors aged 50 years and older, the corresponding odds ratio (OR) was 4.01 (CI 1.35–11.95). Histories of smoking (OR 2.47, CI 0.86 to 7.09), donor gender (OR 1.99, CI 0.86 to 4.63) and hypertension (OR 1.51, CI 0.66 to 3.45) were also in the model but did not reach statistical significance.
Predictors of performing angiography
Potential factors influencing the decision to proceed with angiography in donors 40 years of age and older are a history of cigarette smoking or hypertension and increasing donor age. Conversely, the need for CPR during donor evaluation and management may be a marker for an unstable donor and hence procurement without angiography. By logistic regression, limiting the analysis to donors at hospitals with angiography facilities, only CPR (OR 0.39; 95%CI 0.16–0.94, p = 0.037) was a statistically important factor (i.e., donors who underwent CPR were less likely to have an angiogram performed).
Duration of the evaluation and procurement process
Elapsed times from brain death declaration to angiography and from brain death declaration to aortic cross-clamp during procurement are shown in Table 2. The times were shortest for the NA/NH group, reflecting, at least in part, the 7.2% (10/139) of donors in this category who were rushed to the operating room because of hemodynamic instability. The other groups showed no overall difference in the total time, suggesting that angiography did not delay the procurement process. However, in one case, the family of a potential donor with cardiac risk factors completely withdrew consent after they expressed concern that the angiography would prolong the process of organ procurement.
Actual practice compared with published guidelines and reasons for unsuitability of donor hearts
The number of donors meeting ACC Guideline indication for angiography in the YA/NH group was 37 (or 88.1% of the total). The corresponding number in the YA/YH group was 66 (or 85.7%).
A significant proportion (118 of 139, or 84.9%) of donors in the NA/NH group would have met ACC criteria. Among this group of donors, the number of males ≥45 years of age and females ≥50 with no history of hypertension, smoking or diabetes was 13. The number of males ≥40 and females ≥45 with one or more of those risk factors was 105 (63 with one risk factor, 42 with two risk factors). The number of NA/NH donors who were in noncatheterization facilities was 24 of 139 (or 19 of the 118 meeting criteria for angiography), but lack of availability of catheterization or logistics was cited in organ bank records in only 12 cases. Echocardiographic findings, mostly abnormal wall motion, were cited in 36 cases and served as the only reason for excluding the potential heart donor in 26. Age was listed as the only reason for nonutilization in 14. Among the other reasons for nonutilization of donor hearts, medical comorbidities such as chronic renal failure were cited in 16 cases, hemodynamic instability in 10, medical examiner limitations in seven, ventricular or atrial arrhythmias in five, serology in four, social history in four, poor oxygenation in three, prolonged CPR in three, use of high dose vasopressors in two, and lack of an appropriate recipient or revocation of consent in one each. Documented or suspected myocardial infarction on presentation or a history of CAD was listed in 26 prospective donors.
Of note, 21 of the 38 (55.3%) donors in the NA/YH group were actually in hospitals with the capability to perform cardiac catheterization, but no study was performed.
Outcome following transplantation
Graft survival was lower for the select group of donors ≥40 years of age who were not subjected to angiography, compared with donors who underwent angiography (Fig. 2)(Wilcoxon test, p = 0.017). This difference is attributable to early graft failure, which was more common among recipients of hearts from donors who did not undergo angiography (7/35 or 20%) versus those who did (3/70, 4%, p = 0.015). There were no other differences in the cause of death (rejection, infection, graft CAD, malignancy or multisystem organ failure) between the groups.
Donors under 40 years of age subjected to angiography
An additional 15 potential donors (10 male) who were <40 years of age had angiography. Four of these hearts were not transplanted because of abnormal wall motion (n = 2), elevated end-diastolic pressure with mitral regurgitation (n = 1) and a finding of diffuse coronary spasm versus diffuse coronary disease (n = 1). One of the donors had an atrial septal defect, which was repaired intra-operatively at the time of transplant.
Angiography of the potential donor has been advocated as the preferred method to screen donors who carry risk factors for CAD (10,23,24). In this study using data from the New England Organ Bank, we found that coronary angiography was performed in 40.2% (119 of 296) of donors for whom consent was obtained for heart donation over a period of 4.5 years. The use of angiography has increased over time. If the donor underwent CPR during management, angiography was less frequently performed, likely reflecting instability in the donor. From the 119 potential donors who underwent angiography, 77 hearts were deemed to be acceptable and transplanted. Not surprisingly, the strongest predictor of donor coronary disease was advanced donor age. Donors aged 45 to 49 years were 4.22 times more likely to have CAD than donors aged 40 to 44 years. Donors aged 50 years and older had a corresponding odds ratio of 4.01; the lack of further increase in risk likely represents selection bias. A history of smoking in the donor was the only other factor to approach statistical significance and could be an important factor in a larger sample size.
There were no complications of angiography recorded and, in contrast to the report by Jacobbi et al. (26), the procedure did not appear to prolong the procurement process.
Donor angiography and outcomes
Use of angiography was not associated with poorer subsequent outcome. In fact, by Kaplan-Meier analysis, survival of recipients of hearts from donors greater than or equal to 40 years of age was lowerin the early posttransplant period if the donor was notsubjected to angiography. Early graft failure as a cause of death was more common in this group, raising the possibility of poorer preservation secondary to undiagnosed donor coronary disease in these hearts. Therefore, we think that the published data suggesting a worse outcome for recipients of older donor hearts may be related to undiagnosed pre-existing CAD. Performance of angiography and the subsequent removal of a potential donor with CAD from the donor pool may significantly reduce the risk of death following transplant. Alternatively, some investigators have advocated the use of coronary artery bypass grafting at the time of transplant and/or the use of an alternative recipient list when donor disease is detected (27,28).
Donor angiography and guidelines
The majority of donors undergoing angiography clearly met criteria for the procedure according to published guidelines. Among donors who did not undergo angiography, hemodynamic instability and abnormal wall motion on echocardiography were two common reasons cited for unsuitability of the donor heart. Whether or not donors in the latter category should be more strongly considered for angiography in order to rule out pre-existing coronary disease as the culprit cannot be determined from these data. However, if it can be shown that the abnormal wall motion is not due to CAD, the likelihood is greater that brain death alone is responsible (22,29). Given the possibly reversible nature of this abnormality, the supply of donor hearts could be increased by a more aggressive evaluation of these potential donors (30).
Alternatives to angiography are limited. Several investigators have suggested that bench angiography (31,32)at the time of organ recovery can help to determine whether significant proximal coronary disease is present. However, the safety and utility of this technique in human heart transplantation have not been proven.
Appropriate use of donor angiography
Given the aging of the donor population (3,4), the finding that older donors have more CAD than younger donors by intravascular ultrasound (7,8)and our results confirming that CAD is more likely with increasing age of the donor, we suggest that a conservative approach (33)toward procurement may be reasonable if angiography is logistically difficult or if the donor appears to be hemodynamically unstable.
Conversely, it is also possible that some overuse of angiography exists. In this study, 57.6% of angiograms revealed minimal or no CAD (either obstructive or nonobstructive). However, this percentage is not unreasonably high in light of the risks associated with the transplantation a diseased donor heart.
Limitations of the study
We did not assess the financial impact of angiography on the procurement process. It has been suggested that, in some instances, the expansion of donor criteria increases costs (34). During the period under study, donor hospitals were reimbursed for angiography, but the charges varied widely from $1,000 to $5,000, and not all donor hospitals requested payment for this service.
We lacked access to complete hospital records and adequate information on the status of potential recipients at the time of donor identification. Hence, it was not possible to ascertain the exact decision making at the time of organ procurement. Further, no direct data are available to measure the extent to which exposure to intravenous radiographic contrast may affect early renal function following subsequent kidney transplant (35), though avoidance of left ventriculography during donor catheterization seems reasonable.
The study reflects the experience of one organ procurement organization in one region of the country. Practices may differ in other parts of the country. It has already been documented that use and availability of catheterization facilities vary considerably (36)across the United States. However, given that the lack of availability of angiography was cited in only a small minority of cases in this study, it is unlikely to be a major issue limiting heart procurement in the United States.
In conclusion, we believe that donor angiography can be widely applied without complicating or significantly prolonging the organ procurement process. Older donors comprise the highest risk group for finding CAD. Increased use of angiography in donors who have abnormal wall motion on echocardiography may increase yield but further study is indicated. A systematic approach to invasive donor evaluation encompassing current guidelines and clinical criteria should be developed in order to optimize the use of donor angiography, thereby minimizing the loss of potentially viable donor hearts while decreasing the risk of transmission of occult CAD.
The authors thank Dr. Francis L. Delmonico for his encouragement and thoughtful review of an earlier version of the manuscript, and Sui Tsang for help in the analysis of recipient survival.
☆ Dr. Hauptman was supported in part by a grant from the Charles A. King Trust.
- American College of Cardiology
- coronary artery disease
- confidence interval
- cardiopulmonary resuscitation
- donor angiography not performed
- donor heart not transplanted
- donor angiography performed
- donor heart transplanted
- Received August 9, 2000.
- Revision received November 14, 2000.
- Accepted December 15, 2000.
- American College of Cardiology
- Hosenpud J.D.,
- Bennett L.E.,
- Keck B.M.,
- Fiol B.,
- Boucek M.M.,
- Novick R.J.
- ↵Randall T. Too few human organs for transplantation, too many in need … and the gap widens. JAMA 1991;265:1223, 1227.
- Himes L.,
- van Buren C.,
- Young J.B.,
- Stanbridge R.D.,
- Durand R.,
- Nelson K.
- Botas J.,
- Pinto F.J.,
- Chenzbraun A.,
- et al.
- Tuzcu E.M.,
- Hobbs R.E.,
- Rincon G.,
- et al.
- Gao H.Z.,
- Hunt S.A.,
- Alderman E.L.,
- Liang D.,
- Yeung A.C.,
- Schroeder J.S.
- De Angelis E.,
- Keogh A.,
- Anderson P.,
- et al.
- Mehra M.R.,
- Ventura H.O.,
- Chambers R.,
- et al.
- Hoercher K.,
- Young J.B.,
- Stewart R.W.,
- McCarthy P.M.,
- Smedira N.,
- Kendall K.
- Hauptman P.J.,
- Mudge G.H. Jr.
- Kono T.,
- Morita H.,
- Kuroiwa T.,
- Onaka H.,
- Takatsuka H.,
- Fujiwara A.
- Baldwin J.C.,
- Anderson J.L.,
- Boucek M.M.,
- et al.
- Kirkman R.L.,
- Milford E.L.,
- Luskin R.S.
- Pollick C.,
- Cujec B.,
- Parker S.,
- Tator C.