Pediatric heart transplantation for anthracycline cardiomyopathy: Cancer recurrence is rare

PEDIATRIC TRANSPLANTATION

Pediatric Heart Transplantation for Anthracycline Cardiomyopathy: Cancer Recurrence Is Rare Kendra M. Ward, MD,a Helen Binns, MD,a Clifford Chin, MD,b Steve A. Webber, MD,c Charles E. Canter, MD,d and Elfriede Pahl, MDa Background: Although anthracycline therapy is invaluable for treating neoplastic disorders, morbidity includes severe cardiomyopathy that leads to heart transplantation. This multicenter study describes the course of children who experienced anthracycline cardiomyopathy (ACM) and who subsequently required heart transplantation. Methods: We reviewed transplant databases/registries at 4 pediatric heart transplant centers to identify children with ACM who were listed for heart transplantation. We reviewed medical records to determine cancer therapy, clinical course, and outcome. Results: Eighteen patients were listed, and 17 underwent transplantation. Mean age at cancer diagnosis was 6.0 years (SD, 3.7). The mean anthracycline dose was 361 mg/m2 (SD, 110). The median time from cancer diagnosis to listing for heart transplantation was 9.2 years (range, 0.4 –15.2 years). Six transplantations were performed in patients who had disease-free intervals of ⬍5 years. Two patients were lost to follow-up, and 8 are alive at 4.9 years (SD, 2.0; range, 1.3–7.4 years) after transplantation. Seven patients died at 4.7 years (SD, 2.0; range, 1.2–7.1 years) after transplantation. One patient had recurrent cancer. One-, 2- and 5-year survivals were 100%, 92%, and 60%, respectively. Conclusions: Cardiomyopathy that progresses to the need for heart transplantation occurs in patients receiving a wide range of cumulative anthracycline doses. The time from chemotherapy to ACM varies. Outcomes after transplantation are acceptable, and cancer recurrence is rare. Reconsideration of the 5-year disease-free wait period is warranted. J Heart Lung Transplant 2004;23:1040 – 45.

Anthracyclines are valuable agents in treating various neoplastic disorders; however, toxicity can include cardiomyopathy.1–7 The progression of cardiomyopathy varies, and recommendations for long-term monitoring of these patients have been proposed.8 Case reports of heart transplantation as therapy for end-stage anthracycline cardiomyopathy (ACM) suggest favorable outcomes, and cardiac transplantation is becoming an acceptable treatment option.9 –14 A 5-year cancer-free period before transplantation is a routinely accepted guideline.15 We describe a series of pediatric patients with ACM from 4 institutions who required cardiac transplantation. Furthermore, we demonstrate accept-

From the aDepartment of Pediatrics, Children’s Memorial Hospital, Chicago, Illinois; bStanford University, Palo Alto, California; cChildren’s Hospital of Pittsburgh, Pittsburgh, Pennsylvania; and dSt. Louis Children’s Hospital, St. Louis, Missouri. Submitted March 12, 2003; revised August 2, 2003; accepted August 2, 2003. Reprint requests: Elfriede Pahl, MD, Children’s Memorial Hospital, Division of Cardiology, 2300 Children’s Hospital Plaza, Box 21, Chicago, Illinois 60614. Telephone: 773-880-4553. Fax: 773-880-8111. E-mail: [email protected] Copyright © 2004 by the International Society for Heart and Lung Transplantation. 1053-2498/04/$–see front matter. doi:10.1016/ j.healun.2003.08.014

1040

able survival and rare cancer recurrence even in recipients who undergo early transplantation. METHODS The study design is a 4-center, retrospective medical record review. We identified 18 subjects. We reviewed cardiology, oncology, and general medical records for information concerning cancer treatments, clinical course of cardiomyopathy, and transplant outcomes. Data gathered from oncology records included type of cancer, date of diagnosis, cumulative dose of anthracycline, and additional therapy (other chemotherapy, surgery, or radiation). Cumulative doses of doxorubicin and daunorubicin were considered equivalent for this analysis as reported by others.8 If radiation therapy was used, we recorded location and total dose. We reviewed cardiology records for date of initial referral, reason for referral, presence of clinical congestive heart failure (CHF), and echocardiographic and electrocardiographic abnormalities at referral. We measured left ventricular (LV) function with multigated image acquisition analysis at 1 center (Pittsburgh) and with echocardiography at the 3 others. We recorded LV function as mildly, moderately, or severely depressed based on the following criteria: 1) mild depression ⫽ shortening

The Journal of Heart and Lung Transplantation Volume 23, Number 9

Ward et al.

1041

Table 1. Anthracycline/Chemotherapy Regimens

Subject 1 2 3 4 5 6 7 8

Cancer Non-hodgkin’s lymphoma Ewing’s sarcoma Ewing’s sarcoma Acute lymphocytic leukemia Ewing’s sarcoma Primary Neuroectodermal Tumor Ewing’s sarcoma Non-Hodgkin’s lymphoma

Age at diagnosis 6.25 14 4.33 2.92 11.92 7.41 10.5 3.75

9 10 11 12 13 14 15 16 17 18

Neuroblastoma Rhabdomyo sarcoma Ewing’s sarcoma Musculoaponeurotic fibromatosis Wilm’s Non-Hodgkin’s lymphoma Wilm’s Rhabdomyosarcoma Burkitt’s lymphoma Retinoblastoma

2.92 2.25 11 2.58/5/6** 7.25 3 3.08 1.5 7.83 N/A

Cumulative anthracycline dose/m2 240 480 293 458 300 355 438 300

Chest irradiation No No Yes No No No No Yes

540 480 375 360 370 400 278 400 75 N/A

Yes No* Yes Yes Yes N/A No No* No N/A

Other chemotherapy V, 6, P V, C, I, 16 V, A, C V, itM V, A, 16, I V, C, I, 16 V, A, C V, C, Tg, M, H, ara-C, B V, C, 16, D, CP V, A, C V, A, C D V, A, C N/A V, A V, A V, C, it ara-c, M N/A

*Radiation to other location. **Anthracycline used on second recurrence only. A, Actinomycin; B, bis-chloronitrosoure-carmustine; C, Cytoxan; CP, Cis-platinum diamminedichloride; D, Darcarbazine; H, hydroxyurea; I, Ifosphamide; it, intrathecal administration; M, methotrexate; n/a, not available; P, Prednisone; tg, thioguanine; V, Vincristine; 6, 6mp; 16, VP-16.

fraction (SF) at 21% to 27% or ejection fraction (EF) ⬎45%, 2) moderate depression ⫽ SF at 11% to 20% and EF at 30% to 45%, 3) severe depression ⫽ SF ⱕ 10% or EF ⱕ 30%. We reviewed all electrocardiographic, 24hour Holter monitor, echocardiographic, cardiac catheterization, and endomyocardial biopsy data available from referral to the time of listing. Post-transplant data collected included cancer recurrence, post-transplant lymphoproliferative disease or new tumors detected, duration of follow-up, and mortality. We determined time intervals for each patient as follows: cancer diagnosis to onset of CHF, cancer diagnosis to listing for cardiac transplantation, onset of CHF to listing for heart transplantation, listing to transplantation, and duration of follow-up after transplantation or time from transplantation to death. Analyses We analyzed data using SPSS for Windows, version 11.0.1 (SPSS; Chicago, IL). Significance was set at p ⬍ 0.05. We present normally distributed data as means and standard deviations. Medians and ranges are used for data with non-Gaussian distribution. We evaluated associations between variables using Spearman’s correlation, the Mann-Whitney U test, and the Kruskal-Wallis test.

RESULTS We identified 18 subjects from 4 centers who were listed for cardiac transplantation; 7, 4, 4, and 3 subjects were listed per center. Ten were male and 8 were female patients. Anthracycline cardiomyopathy was diagnosed between June 1983 and September 1997. Seventeen of the 18 subjects underwent cardiac transplantation between May 1987 and January 2000. One subject underwent a second transplantation for transplant vasculopathy. Cancer Therapy Cancer diagnoses included sarcoma (7), lymphoma (4), Wilm’s tumor (2), leukemia (1), neuroblastoma (1), musculoaponeurotic fibromatosis (1), primitive neuroectodermal tumor (1), and retinoblastoma (1). Seventeen subjects received doxorubicin, and 1 received daunorubicin. The mean age at cancer diagnosis was 6.0 years (n ⫽ 17; SD, 3.7; range, 1.5–11.9 years). The mean anthracycline dose was 361 mg/m2 (SD, 110 mg/m2). Total cumulative doses of anthracycline ranged from 75 to 540 mg/m2. Six subjects (33%) received chest radiation. Six other subjects received radiation that did not involve the chest. Table 1 shows details regarding cancer therapy for each patient.

1042

Ward et al.

The Journal of Heart and Lung Transplantation September 2004

Figure 1. This figure delineates the total time from initial cancer diagnosis to the time of cardiac transplant listing for each subject and demonstrates the widely variable courses of the individuals. The open bars represent the time in years from cancer diagnosis to congestive heart failure (CHF) diagnosis. The shaded bars represent the time in years from CHF diagnosis (dx) to listing for cardiac transplantation.

Cardiac Presentation Time from cancer diagnosis to initial cardiac evaluation was a median of 1.4 years (n ⫽ 16; range, 0.3–13 years). Six patients experienced cardiac failure within 1 year of cancer diagnosis. Another 3 had cardiac failure between 1 and 2 years from initial cancer diagnosis, and the remaining 7 patients had failure more than 4 years after cancer diagnosis. We could not ascertain the time of presentation in 2 patients (Figure 1). Information about the initial reason for seeing a cardiologist was available for 16 of the 18 subjects. Seven subjects (44%) sought treatment for complaints suggestive of heart failure; their symptoms were dyspnea on exertion/respiratory complaints (6), fatigue (2), exercise intolerance (2), syncope (2), and shock (1). Eight patients had no complaints at the time of initial cardiac evaluation and were identified by screening echocardiograms. One patient was referred to rule out endocarditis. We found no significant association between time to cardiac presentation and symptomatic CHF at that time (p ⫽ 0.2). Left Ventricular Function Initial imaging studies were available for 15 of the 18 subjects. At the time of initial diagnosis of ACM, depres-

sion of LV function was mild in 5 subjects, moderate in 6, and severe in 4. We found no association between cumulative dose and initial severity of ventricular function (p ⫽ 0.68). At initial echocardiogram, 28% (4/14) of subjects with comprehensive data available had moderate or severe atrioventricular valvar regurgitation. At the time of listing, depression of LV function was mild in 4 subjects, moderate in 9, and severe in 4. At the time of listing, moderate or severe atrioventricular valvar regurgitation was present in 8 of 15 subjects (53%). At listing, dilated cardiomyopathy was documented in 15 patients and restrictive cardiomyopathy in 2. Although both patients with restrictive findings had received radiation to the chest, 4 other patients who also had received chest radiation had systolic dysfunction and LV dilatation. Additional Cardiac Evaluation Sixteen subjects (89%) had electrocardiography performed at the time of listing that was available for review. One subject had completely normal electrocardiography results, all others had abnormal results. Findings included ST segment or T wave changes (11), LV hypertrophy (7), atrial enlargement (8), premature ven-

The Journal of Heart and Lung Transplantation Volume 23, Number 9

tricular contractions (6), conduction delay (7), sinus tachycardia (6), right ventricular hypertrophy (2), biventricular hypertrophy (1), and other findings in 4. Seventeen patients underwent cardiac catheterization before listing for transplantation. As expected, a low cardiac index was demonstrated (n ⫽ 15; mean, 2.25 liter/min/m2; SD, 0.8) as well as an increased pulmonary vascular resistance index (4.8 IU; SD, 2.4). Fifteen patients underwent endomyocardial biopsy: histopathologic evaluation was diagnostic for anthracycline toxicity in 12 (80%), 1 subject had radiation fibrosis, and 1 had inflammation. One biopsy specimen was inadequate for evaluation. Progression to Transplantation Listing for heart transplantation occurred at a median of 1.3 years (range, 0.1–13.8 years) after the diagnosis of ACM and at a mean of 7.7 years (SD, 4.8; range, 0.4 –15.2 years) after cancer diagnosis (Figure 1). Two patients were listed within 1 year of cancer diagnosis. An additional 4 patients had disease-free intervals of ⬍5 years at the time of transplantation. Four subjects who came for cardiac evaluation ⱖ10 years after cancer diagnosis required listing within ⱕ18 months of that initial cardiac evaluation. Age at transplantation ranged from 6.1 to 18.7 years (mean, 13.9 years; SD, 3.5). One patient died of multi-organ failure while awaiting transplantation, and that subject had been listed for 45 days at the time of death. The other 17 ultimately underwent transplantation at a median of 53 days (range, 9 –377 days) after being listed. Thirteen subjects underwent transplantation within 2 months of listing. Figure 1 displays time from cancer diagnosis to first cardiac evaluation and then to subsequent listing for transplantation for the 16 subjects for whom this information is known. We gathered data describing the medical management of each subject. Below, we present the clinical courses of 2 subjects to further demonstrate the variability in course to transplantation. Figure 2 graphically presents the data for 1 of these subjects. Indolent course. Patient 10 was diagnosed with rhabdomyosarcoma of the liver at age 2.3 years (Figure 2). She completed a treatment regimen that consisted of 480 mg/m2 doxorubicin, vincristine, actinomycin D, and cyclophosphamide as well as radiation to her liver and overlying diaphragm (total, 4,400 rads). At age 3.8 years, she experienced tachypnea. An echocardiogram demonstrated an SF of 21%, and she began receiving digoxin. By age 5 years, her SF had improved to 25%, and she was asymptomatic. Digoxin therapy was stopped. At age 12.7 years, she had pneumonia with a pleural effusion, and an echocardiogram showed an SF of 13%. Digoxin and captopril therapy was initiated,

Ward et al.

1043

Figure 2. This figure displays the treatment and left ventricular function during 1 subject’s course from cancer diagnosis to cardiac transplantation (Ca tx). The change in shortening fraction (SF) is displayed with the timing of adding therapy.

and she remained stable until age 14.3 when she had chest pain. Her SF was 9%, and she was referred for transplantation. She was admitted to the hospital, and her condition deteriorated rapidly, despite inotropic support. She required paralysis and intubation 6 weeks after listing. She underwent transplantation 53 days after listing and is currently alive and well 8 years after transplantation. Rapid fulminate course. Patient 1 was diagnosed with Stage 3, large B-cell lymphoma at age 6.25. During the remission phase of her treatment, 14 days after the last dose of doxorubicin (cumulative dose, 240 mg/m2) and 4 months after her cancer diagnosis, she experienced cardiogenic shock. While receiving inotrope therapy, the patient had an echocardiogram that showed an SF of 9% . The decision was made to list her for cardiac transplantation, despite proximity to cancer presentation, because the risk of recurrence was deemed small and the risk of death without transplantation inevitable. She underwent transplantation 11 days after listing, which was 5 months after her initial cancer diagnosis. Currently, she is alive and well 6 years after transplantation without recurrence of her malignancy.16 Outcomes One-year actuarial survival after transplantation was 100%, and the 2- and 5-year survivals were 92% and 60%, respectively. Two of the subjects were lost to follow-up. Among the 15 with long-term follow up available, 8 currently are alive at a mean of 4.4 ⫾ 2.0 years after cardiac transplantation. One patient underwent a second transplantation for allograft coronary disease. Seven patients died at a mean of 4.7 years (⫾ 2.0; range, 1.2–7.1 years) after transplantation. Causes of death included rejection,3 graft coronary disease,1 pulmonary embolus,1 motor vehicle accident,1 and unknown causes.1 One patient (Patient 2) had recurrence of Ewing’s sarcoma with pulmonary metastases

1044

Ward et al.

15 months after transplantation and was treated successfully. She subsequently died of pulmonary embolus 57 months after transplantation. Post-transplant lymphoproliferative disease did not occur in any subjects in this cohort. DISCUSSION We describe the clinical course of 18 pediatric patients in the United States listed for heart transplantation for ACM and describe their outcomes. Several case reports and 2 European series (pediatric and adult) have described transplantation for ACM.9 –14 These 2 series described 14 and 9 patients, respectively,11,12 with acceptable results; however, 1 patient experienced post-transplant lymphoproliferative disease.11 Cardiac toxicity caused by anthracycline agents was reported first during early drug trials in the late 1960s.17 Congestive heart failure was described originally during anthracycline therapy or within the 1st year of completing anthracycline chemotherapy.4,10 Short-term follow-up of these patients indicated stabilization of CHF and potential improvement of cardiac function with time.2 Subsequently, late cardiotoxicity, occurring 4 to 20 years after completing anthracycline therapy, was reported.17 The population study by Levitt et al12 revealed a bimodal distribution of presentation with CHF; however, other studies have not found a consistent pattern.8 Morbid outcomes including sudden death and death from progressive CHF were reported.7,18,19 Our study confirms that timing and progression of anthracycline-related cardiac complications vary. We found wide variation in the progression to severe cardiac failure, with a few patients having CHF during chemotherapy and others having failure ⬎10 years later. Although multiple studies have demonstrated a dose-related risk for cardiac toxicity,4,7,19 subjects in the current study did not receive excessive cumulative doses of anthracycline. One patient received only 75 mg/m2, and the majority of our patients (76%) received ⬍450 mg/m2, which is in keeping with current oncologic practice.20 In addition, we found no association between cumulative anthracycline dose and timing to CHF, nor did we find an association between cumulative dose and initial severity of ventricular dysfunction. The lack of cumulative dose as a major risk factor in this cohort, which progressed to severe cardiac failure, highlights the need for further work to identify mechanisms and risk factors for the development of cardiomyopathy in patients treated with anthracycline. Endomyocardial biopsy specimens showed that most subjects had typical histopathologic changes with anthracycline therapy. One subject had biopsy findings of inflammatory changes consistent with myocarditis;

The Journal of Heart and Lung Transplantation September 2004

however, he had a previous admission for CHF deemed secondary to anthracycline therapy received 12 years previously. Ali et al21 described 5 patients with history of anthracycline treatment who experienced acute cardiac decompensation in the presence of viral infection. Viral infection may have contributed to the acute decompensation in these patients. The effects of radiation to the chest in 6 subjects probably contributed to the development of heart failure; however, biopsy specimen showed that only 1 patient had evidence of radiation changes. Radiation to the chest is a known risk factor for restrictive cardiomyopathy, but no published studies have evaluated the risk of cardiomyopathy for patients exposed to both anthracycline therapy and chest radiation. Subjects tolerated immunosuppression without obvious additional sequelae;22,23 i.e., no deaths occurred because of infectious complications. We found no post-transplant lymphoproliferative disease in this series; however, we found 1 cancer recurrence 15 months after transplantation in a patient who underwent transplantation only 12 months after diagnosis of Ewing’s sarcoma. Some transplant programs require at least 1 disease-free year before listing for transplantation.10 The sub-committee of the Pediatric Committee of the American Society of Transplantation has published recommendations that state that malignancy ⬍5 years before evaluation should be viewed as a relative contraindication for transplantation.15 Six subjects underwent transplantation ⬍5 years after cancer diagnosis, and 2 underwent transplantation ⱕ1year after cancer diagnosis. One of these subjects underwent transplantation during the remission phase of treatment within 5 months of cancer diagnosis. This subject is alive and well 6 years later.16 Reconsideration of a disease-free interval of ⬍5 years may be warranted; the appropriateness of listing patients with recent malignancy must be individualized, with input from the oncologist and consideration of type of tumor, stage, grade, and response to initial therapy. Survival outcomes of this cohort are acceptable. Causes of death and actuarial survival data are similar to those listed in the International Society for Heart and Lung Transplant (ISHLT) Registry.24. Four patients died of chronic rejection/non-compliance or of allograft coronary disease, which are the most common causes of death in pediatric transplant recipients.24 The patient with cancer recurrence died of a pulmonary embolus 3.5 years after successful treatment of recurrence. As with any retrospective study, the availability of information in the medical records limits data. Treatment regimens for malignancies and for heart failure changed during the time evaluated in this study, making comparisons of the various aspects of these patients’ courses difficult.

The Journal of Heart and Lung Transplantation Volume 23, Number 9

Ward et al.

CONCLUSIONS From this series, we conclude that transplantation is an acceptable treatment option for patients with intractable cardiac failure secondary to anthracycline therapy. Survival is comparable with ISHLT Registry data for all pediatric heart recipients. Tumor recurrence after transplantation is rare, and current guidelines regarding the duration of a cancer-free interval before listing should be re-examined. Prospective studies of survivors of ACM will help determine the optimum timing of follow-up as well as clarify risk factors for progression to severe cardiac failure. REFERENCES 1. Bristow MR, Mason JW, Billingham ME, Daniels JR. Doxorubicin cardiomyopathy: evaluation by phonocardiography, endomyocardial biopsy, and cardiac catheterization. Ann Intern Med 1978;88:168 –75. 2. Dresdale A, Bonow RO, Wesley R, et al. Prospective evaluation of doxorubicin-induced cardiomyopathy resulting from postsurgical adjuvant treatment of patients with soft tissue sarcomas. Cancer 1983;52:51–60. 3. Goorin AM, Borrow KM, Goldman A, et al. Natural history in children of congestive heart failure secondary to adriamycin cardiomyopathy. Proc Am Assoc Cancer Res Am Soc Clin Oncol 1979;20:328. 4. Goorin AM, Borow KM, Goldman A, et al. Congestive heart failure due to adriamycin cardiotoxicity: its natural history in children. Cancer 1981;47:2810 –6. 5. Henderson IC. Adriamycin cardiotoxicity. Am Heart J 1980;99:671–4. 6. Singal PK, Iliskovic N. Doxorubicin-induced cardiomyopathy. N Engl J Med 1998;339:900 –5. 7. Von Hoff DD, Layard MW, Basa P, et al. Risk factors for doxorubicin-induced congestive heart failure. Ann Intern Med 1979;91:710 –7. 8. Steinherz LJ, Graham T, Hurwitz R, et al. Guidelines for cardiac monitoring of children during and after anthracycline therapy: report of the Cardiology Committee of the Childrens Cancer Study Group. Pediatrics 1992;89:942–9. 9. Arico MPE, Nespoli L, Vigano M, Porta F, Burgio GR. Long-term survival after heart transplantation for doxorubicin induced cardiomyopathy. Arch Dis Child 1991;66: 985–6. 10. Armitage JM, Kormos RL, Griffith BP, Fricker FJ, Hardesty

11.

12.

13.

14.

15.

16.

17.

18. 19.

20.

21.

22.

23. 24.

1045

RL. Heart transplantation in patients with malignant disease. J Heart Transplant 1990;9:627–30. Dorent R, Pavie A, Nataf P, et al. Heart transplantation is a valid therapeutic option for anthracycline cardiomyopathy. Transplant Proc 1995;27:1683. Levitt G, Bunch K, Rogers CA, Whitehead B. Cardiac transplantation in childhood cancer survivors in Great Britain. Eur J Cancer 1996;32A:826 –30. Luthy A, Furrer M, Waser M, et al. Orthotopic heart transplantation: an efficient treatment in a young boy with doxorubicin-induced cardiomyopathy. J Heart Lung Transplant 1992;11:815–6. McManus RP, O’Hair DP. Pediatric heart transplantation for doxorubicin-induced cardiomyopathy. J Heart Lung Transplant 1992;11:375–6. Fricker FJ, Addonizio L, Bernstein D, et al. Heart transplantation in children: indications. Report of the ad hoc subcommittee of the Pediatric Committee of the American Society of Transplantation (AST). Pediatr Transplant 1999;3:333–42. Morgan E, Pahl E. Early heart transplant in a child with advanced lymphoma. Pediatr Transplant 2002;6:509 – 12. Steinherz LJ, Steinherz PG, Tan CT, Heller G, Murphy ML. Cardiac toxicity 4 to 20 years after completing anthracycline therapy. JAMA 1991;266:1672–7. Steinherz L, Steinherz P. Delayed cardiac toxicity from anthracycline therapy. Pediatrician 1991;18:49 –52. Von Hoff DD, Rozencweig M, Layard M, Slavik M, Muggia FM. Daunomycin-induced cardiotoxicity in children and adults. A review of 110 cases. Am J Med 1977;62:200 –8. Speyer JL, Freedberg RS. Cardiac complications. In: Abeloff, ed. Clinical Oncology, 2nd ed. Florida: Churchill Livingstone, 2000, pp. 1047–54. Ali MK, Ewer MS, Gibbs HR, Swafford J, Graff KL. Late doxorubicin-associated cardiotoxicity in children. The possible role of intercurrent viral infection. Cancer 1994; 74:182–8. Asante-Korang A, Boyle GJ, Webber SA, Miller SA, Fricker FJ. Experience of FK506 immune suppression in pediatric heart transplantation: a study of long-term adverse effects. J Heart Lung Transplant 1996;15:415–22. Addonizio LJ. Current status of cardiac transplantation in children. Curr Opin Pediatr 1996;8:520 –6. Hosenpud JD, Keck BM, Boucek MM, Novick RJ. The Registry of the International Society for Heart and Lung Transplantation: eighteenth official report—2001. J Heart Lung Transplant 2001;20:805.