Does Cytomegalovirus Serology Impact Outcome After Pediatric Heart Transplantation?

Does Cytomegalovirus Serology Impact Outcome After Pediatric Heart Transplantation? William T. Mahle, MD,a Margaret T. Fourshee,b David M. Naftel, PhD,b Juan C. Alejos, MD,c Randall L. Caldwell, MD,d Karen Uzark, PhD,e Alexandria Berg, MSN,a Kirk R. Kanter, MD,f and the Pediatric Heart Transplant Study Group Background: Cytomegalovirus (CMV) infection has been implicated in a number of complications after heart transplantation. A recent study suggested that children with positive CMV serology (CMV⫹) before transplantation are at increased risk of developing coronary allograft vasculopathy (CAV) and death when compared with CMV⫺ recipients. We analyzed data from the Pediatric Heart Transplant Study Group to determine the impact of recipient CMV status and CMV mismatching on outcome. In addition, the use and efficacy of CMV prophylaxis were studied. Methods: Subjects ⬍18 years of age who underwent heart transplantation during the period from 1993 to 2007 were analyzed. Those transplants in which either the recipient or donor were ⬍6 months of age were excluded due to the confounding effects of maternal antibody. The primary outcome variable was freedom from CAV (mild or greater). Secondary outcomes included freedom from death and freedom from clinical CMV infection. Risk factors were assessed using parametric hazard regression. Results: Of the 1,598 subjects included in the analysis, 637 (40%) were CMV⫹ at the time of transplantation. Some form of CMV prophylaxis was administered to 67% of all recipients, most commonly with a CMV mismatch (donor CMV⫹/recipient CMV⫺). Freedom from clinical CMV infection at 5 years was 91%. Pre-transplant CMV serology was not associated with mortality (p ⫽ 0.40) or risk of developing CAV (p ⫽ 0.10). CMV mismatch was associated with increased risk of clinical CMV disease (p ⬍ 0.001). The use of CMV prophylaxis had no association with mortality or development of CAV. There was also no significant association between CMV prophylaxis and the development of clinical CMV infection. Conclusions: CMV⫹ serology at time of pediatric heart transplantation had no demonstrable association with death or development of CAV. CMV⫺ recipients who receive a CMV⫹ organ are at increased risk of clinical CMV disease. CMV prophylaxis was commonly used, although further studies are needed to establish an optimal approach for prevention of CMV disease in this population. J Heart Lung Transplant 2009;28: 1299 –305. Copyright © 2009 by the International Society for Heart and Lung Transplantation.

In immunosuppressed patients, cytomegalovirus (CMV) infection can lead to acute clinical disease that may be life-threatening. Those subjects thought to be at greatest risk are those who are CMV naive at the time of From the aDepartment of Pediatrics, Emory University School of Medicine, Atlanta, Georgia; bDepartment of Cardiothoracic Surgery, University of Alabama, Birmingham, Birmingham, Alabama; cDepartment of Pediatrics, UCLA School of Medicine, Los Angeles, California; d Department of Pediatrics, Riley Children’s Hospital, Indianapolis, Indiana; eDepartment of Pediatrics, and Cincinnati Children’s Medical Center, Cincinnati, Ohio; and fDepartment of Cardiothoracic Surgery, Emory University, Atlanta, Georgia. Submitted May 5, 2009; revised July 6, 2009; accepted July 28, 2009. Reprint requests: William T. Mahle, MD, Department of Pediatrics, Children’s Healthcare of Atlanta, Emory University School of Medicine, 1405 Clifton Road NE, Atlanta, GA 30322-1062. Telephone: 404-315-2672. Fax: 404-325-6021. E-mail: [email protected] Copyright © 2009 by the International Society for Heart and Lung Transplantation. 1053-2498/09/$–see front matter. doi:10.1016/ j.healun.2009.07.011

transplantation and receive an organ from a CMV⫹ donor. In addition to the risk of CMV clinical infection a number of studies have suggested that CMV mismatch predisposes subjects to coronary allograft vasculopathy (CAV), acute rejection and graft loss.1–3 However, these findings have been inconsistent. A recent single-center study in children suggested that those children who were CMV⫹ before transplantation had the greatest risk of developing CAV.4 In that study, the so-called CMV-mismatch patients, CMV⫺ recipient and CMV⫹ donor, did not have an increased risk compared with those having CMV⫺ donor and CMV⫺ recipient. These findings imply that the response to CMV after pediatric heart transplantation might be different in children when compared with adults. In the present study we sought to determine whether the CMV status of pediatric heart transplant recipients is associated with the risk of developing CAV or risk of death. 1299

1300

Mahle et al.

METHODS Patient Population Data were collected on patients who underwent heart transplantation between January 1, 1993 and December 31, 2007 at 31 institutions in the Pediatric Heart Transplant Study. The present study excluded all recipients ⬍6 months of age as well as those children whose donor was ⬍6 months of age. The reason for excluding this population is that maternal antibody is thought to be present through the first several months of life and interpretation of CMV seropositivity of the child is confounded by this phenomenon.5 During the study period, there were 1,598 subjects with complete data with respect to CMV status. Follow-up on all patients was complete through December 31, 2007. Data were collected prospectively for later ad hoc retrospective analysis, as previously described. Data Collection Data and events forms were forwarded to the University of Alabama Medical Center (data analysis center), where they were reviewed and entered into the Pediatric Heart Center Study (PHTS) database. Data were verified manually at the time of data entry and using standard statistical procedures. CMV Serology and Prophylaxis CMV serology is entered on the data collection form for both the donor and recipient. The PHTS data form defined prophylaxis as “started during the first 30 days, not used to treat a known infection.” Ganciclovir and acyclovir prophylaxis is at least 14 days of intended treatment that begins by Day 4 post-operatively. Immunoglobulin prophylaxis is defined as at least 1 day of treatment that begins by Day 4 post-operatively. Outcome Measures The primary outcome measures were freedom from CAV or freedom from death after heart transplantation. In this analysis we considered CAV as grade mild or greater to be clinically significant. A previous analysis of CAV by the PHTS has been reported and the methodology for classification has been described in detail.6 Briefly, severity of CAV seen with angiograms was defined by published criteria. Intracoronary ultrasound is not routine at most PHTS centers and was not assessed in this study. The relationship between CMV status and freedom from death was determined from the time of transplant. However, the analysis of CMV prophylaxis upon freedom from death was performed only among those subjects who survived at least 30 days after transplantation. The rationale for this approach is that patients who die within days of transplant may not have sur-

The Journal of Heart and Lung Transplantation December 2009

vived long enough to complete a CMV prophylaxis regimen and hence this would create a bias in the analysis. A similar approach was previously undertaken in another PHTS investigation analyzing the impact of induction therapy and patient survival.7 Clinical CMV infection is recorded in the annual follow-up form for the study subjects. The PHTS data form defines infection as “infectious process requiring IV therapy or a life threatening infection requiring oral therapy.” Statistical Analysis We examined the data using standard descriptive statistics, including mean and standard deviation. We compared subgroups using contingency tables and t-tests. Time to end-points, such as death, CAV or clinical infection, was characterized by actuarial (Kaplan– Meier) methods and parametric methods. We identified risk of various end-points by multivariate-risk-factor, forward, stepwise analysis in the multiphase hazard domain. All participating centers had approval from their institutional review boards to participate in the study. RESULTS During the study period there were 2,491 subjects who underwent transplantation. Of these subjects, 120 were excluded due to missing data regarding CMV status prior to transplantation. There were 773 subjects who were excluded because they were ⬍6 months of age at the time of transplantation or because the donor organ came from a child ⬍6 months of age. Therefore, 1,598 subjects were included in the analysis. The rate of seropositivity increased with age from 25.4% in those 6 months to 2 years of age, compared with 43.1% in those ⬎15 years of age (Figure 1). The relationship of donor to recipient CMV status is shown in Table 1. There were 502 (31.4%) subjects who were CMV⫺ at transplant, but received an organ from a CMV⫹ donor (CMV mismatch). There were significant associations between CMV seropositivity and a number of important clinical

Figure 1. CMV serology of recipients stratified by age at transplant (n ⫽ 1,598 patients).

The Journal of Heart and Lung Transplantation Volume 28, Number 12

Mahle et al.

Table 3. Presence of Any Cytomegalovirus Prophylaxis Reporteda

Table 1. Relationship of Donor/Recipient Cytomegalovirus Serology Donor serology Recipient serology Positive Negative Total

Positive 361 502 863

Negative 276 459 735

Total 637 961 1,598

Recipients/donors age ⬍6 months and those with incomplete serology were excluded.

variables (Table 2). CMV prophylaxis was administered to 1,040 subjects (71.4%) who survived at least 30 days. Table 3 demonstrates that CMV prophylaxis was administered with equivalent frequency to all subjects, with the exception of those who were seronegative and received a CMV⫺ organ, in which case prophylaxis was given less frequently (p ⬍ 0.001). A number of different CMV prophylaxis regimens were employed (Table 4). The most common prophylactic agent was ganciclovir. The 1- and 5-year freedom from clinical CMV infection was 93% and 91%, respectively (Table 5). The risk of developing a first episode of CMV infection appeared to decrease significantly beyond the second year after transplantation. Those subjects who had a CMV mismatch were at highest risk of developing a clinical CMV infection (p ⬍ 0.001) (Figure 2). Subjects who received an organ from a CMV⫺ donor appeared to be at lowest risk for clinical CMV disease. CMV prophylaxis was not associated with the development of clinical CMV infection (Figure 3). The 5- and 10-year patient survival for the entire cohort was 80% and 62%, respectively. The 5- and 10-year freedom from graft loss was 75% and 59%,

Group 1: D⫹/R⫹ 2: D⫹/R⫺ 3: D⫺/R⫹ 4: D⫺/R⫺ Total

n 328 452 259 417 1,456

Yes 263 (80%) 381 (84%) 186 (72%) 210 (50%) 1,040 (71%)

respectively. Recipient CMV seropositivity was not associated with either freedom from death (p ⫽ 0.40) (Figure 4) or freedom from graft loss (p ⫽ 0.50). In addition, those subjects with a CMV mismatch had equivalent survival when compared with other patients. The administration of CMV prophylaxis was not associated with survival among those subjects who survived at least 30 days after transplant (Figure 5). If one analyzed only those subjects who were CMV mismatched (n ⫽ 469) and presumably at greatest risk, there was again no association between prophylaxis administration and survival (p ⫽ 0.6). The 5-year freedom from CAV was 81%. The majority of subjects with CAV were described as having mild disease. On univariate analysis, those subjects who were CMV⫺ and received a CMV⫺ organ had a lower risk of CAV (p ⫽ 0.03) (Figure 6). When we performed a multivariate analysis that included the four serology groups and patient age, then older age was found to be associated with CAV (p ⫽ 0.002). The p-value for

CMV serology: donor¡recipient ⫺

D ¡R⫹

D⫹¡R⫺

D⫹¡R⫹

p-valuea

7.1 73% 55% 17.0

9.0 64% 52% 18.0

9.4 74% 56% 17.9

10.1 64% 56% 18.2

⬍0.0001 0.0007 0.8 0.003

9.3 216

12.1 211

12.8 214

13.9 214

⬍0.0001 0.9

68% 54% 6% 5% 19

77% 64% 11% 6% 19

74% 57% 9% 5% 20

75% 59% 14% 7% 20

0.02 0.06 0.004 0.3 0.5

22%

11%

16%

11%

⬍0.0001

D ¡R Demographics Agea (years) White race Male BMI Donor Agea (years) Ischemia (min) Status at transplant Status 1 Inotropes VAD ECMO BUN Clinical status at listing Failure to thrive

No 65 (20%) 71 (16%) 73 (28%) 207 (50%)b 416 (29%)

Transplanted patients with survival ⬎30 days (n ⫽ 1,456) and grouped by donor (D)/recipient(R) pre-transplant CMV serology. a Prophylaxis defined as antibiotics started during first 30 days posttransplant not used to treat a known infection. b p ⬍ 0.001.

Table 2. Factor-associated CMV Serology ⫺

1301

⫺

BMI, body mass index; BUN, blood urea nitrogen; CMV, cytomegalovirus; ECMO, extracorporeal membrane oxygenation; VAD, ventricular assist device. a Tests for group comparison.

1302

Mahle et al.

The Journal of Heart and Lung Transplantation December 2009

Table 4. Type of Prophylaxis Reporteda

Group 1: D⫹/R⫹ 2: D⫹/R⫺ 3: D⫺/R⫹ 4: D⫺/R⫺ Total

Acyclovir 21% 19% 33% 51% 28%

Cytogamb 37% 59% 14% 33% 42%

Gancyclovir (or Immune valganciclovir)

globulin

86% 88% 77% 44% 76%

29% 35% 18% 34% 30%

Transplanted patients with survival ⬎30 days (n ⫽ 1,456) and grouped by donor (D)/recipient (R) pre-transplant CMV serology. a Prophylaxis defined as antibiotics started during first 30 days posttransplant not used to treat a known infection. b Complete data collection on Cytogam begins later than other medications.

D⫺/R⫺ became non-significant (p ⫽ 0.12) after adjustment for age. The administration of CMV prophylaxis showed no association with the development of CAV (Figure 7). DISCUSSION This study of a multi-institutional registry of children undergoing heart transplantation has shown that clinical CMV infection is relatively uncommon and that CMV mismatching or CMV seropositivity prior to transplantation was not significantly associated with either death or the development of CAV. The majority of subjects in this series received CMV prophylaxis, although the utility of this practice is unclear. CMV belongs to the group of herpesviruses and about 80% of adults are carriers. In school-aged children seropositivity is present in approximately 50%. CMV may present with mild or severe symptoms about 2 to 3 months after transplantation. Typical findings include Table 5. Outcomes Stratified by CMV Serology CMV serology: donor¡recipient ⫺

D ¡R⫺ D⫺¡R⫹ D⫹¡R⫺ D⫹¡R⫹ Survival 1 month 1 year 5 years Graft survival 1 month 1 year 5 years First CMV 1 month 1 year 5 years CAV 1 month 1 year 5 years

95% 86% 77%

97% 92% 81%

95% 88% 75%

95% 89% 78%

94% 85% 74%

97% 91% 76%

95% 87% 73%

95% 88% 76%

99% 98% 97%

98% 96% 95%

99% 82% 79%

99% 92% 90%

99% 99% 91%

99% 98% 80%

99% 97% 84%

99% 99% 86%

p-valuea 0.4

Figure 2. Kaplan–Meier plot of freedom from CMV infection grouped by donor/recipient CMV serology.

fever, appearance of atypical lymphocytes, leukocytopenia, thrombocytopenia, myalgias, arthralgias and renal impairment. Severe symptoms include interstitial pneumonia, hepatitis, colitis, gastritis, esophagitis, retinitis and encephalitis.8 The greatest risk of clinical CMV infection is in the first several months after transplantation. As one might expect, the risk of CMV is highest among those who have a CMV mismatch; 18% of these subjects developed a clinical CMV infection within 2 years of transplant. However, clinical infection did also occur in some of the subjects who were CMV seropositive at the time of transplant. Twelve percent of subjects who were CMV⫹ and received a CMV⫹ organ developed clinical infection within 5 years of transplantation. Although clinical CMV infection has been of interest to the transplant community, an intriguing feature of CMV is the possibility that the sub-clinical disease might impair graft function. There are a number of studies that have indicated that those heart transplant recipients with a more robust CMV-specific T-cell response have less clinical rejection and less CAV.9 More specifically,

0.5

⬍0.0001

0.1

CAV, coronary allograft vasculopathy, CMV, cytomegalovirus. a Log-rank test.

Figure 3. Kaplan–Meier plot of freedom from CMV stratified by administration of prophylaxis (initial CMV prophylaxis subgroup ⫽ 1,456 patients).

The Journal of Heart and Lung Transplantation Volume 28, Number 12

Mahle et al.

1303

Figure 4. Kaplan–Meier plot of survival grouped by donor/recipient CMV serology.

Figure 6. Kaplan–Meier plot of freedom from CAV grouped by donor/recipient CMV serology.

investigators have shown in animal models that CMV replication induces an inflammatory response in the aortic and cardiac allograft vascular adventitia and intima (endothelialitis).10 This early inflammatory response in turns leads to enhanced intimal thickness in aortic and cardiac allografts. Recently, investigators have suggested that CMV replication results in endothelial dysfunction as measured with brachial artery flowmediated dilation in children after heart transplantation.11 Some studies in adults demonstrated that recipient seropositivity status was correlated with either overt CAV detected by coronary angiography or more rapid lumen loss identified by intravascular ultrasound.12–14 Until recently, little had been reported about the relationship of CMV serology and clinical outcomes in pediatric heart transplantation. A single-institution study suggested that pediatric recipients who were CMV seropositive prior to heart transplantation had an increased risk of both CAV and death when compared with subjects who were CMV seronegative.15 In that

series, where CMV prophylaxis was rarely employed, CMV⫹ recipients had a 3.6-fold greater risk of CAV and a 4-fold greater risk of graft loss, after adjusting for other covariates. Interestingly, the study did not find that the CMV mismatch had an increased risk of either CAV or graft loss. These findings would imply that CMV reactivation is the primary mechanism for graft impairment after transplantation. In the present analysis, in which the study population was more than 10-fold greater than in the aforementioned report, we found no significant association between CMV serologic status and adverse outcomes (CAV or death). In addition, in the most recent analysis by the International Society for Heart and Lung Transplantation (ISHLT) of both adult and pediatric heart transplant recipients, there was no association between recipient CMV status or CMV mismatch and risk of 1- or 5-year mortality.16,17 There are a number of possible explanations for the conflicting findings of these studies. Smaller studies are prone to Type II errors and the association of CMV serology and CAV and graft loss may

Figure 5. Kaplan–Meier plot of survival in the CMV-mismatch group only (donor⫹/recipient⫺) stratified by administration of prophylaxis (initial CMV prophylaxis subgroup ⫽ 1,456 patients; only those with D⫹R⫺ CMV serology ⫽ 452 patients).

Figure 7. Kaplan–Meier plot of freedom from CAV stratified by administration of prophylaxis (initial CMV prophylaxis subgroup ⫽ 1,456 patients).

1304

Mahle et al.

have been a chance finding. The inclusion of young infant donors and recipients in the investigation may have confounded the analysis. There is another important difference in these studies that relates to CMV prophylaxis. The approach to CMV varies considerably among heart transplant programs. Strategies include CMV prophylaxis for all recipients, prophylaxis for high-risk (CMV-mismatch) recipients, or no use of CMV prophylaxis. Importantly, within the PHTS, just over two-thirds of subjects received prophylaxis of some form. It is possible that by the administration of CMV prophylaxis the potential adverse impact of CMV infection on graft function was attenuated or eliminated. Nonetheless, there remained a reasonably sized cohort of patients who did not receive prophylaxis. In this subset of patients we did not detect any relationship between prophylaxis administration CMV status and CAV or mortality. In analyzing the relationship between CMV prophylaxis and survival it is important to recall that we included only those subjects who survived to 30 days. This is because those subjects who died from acute graft failure or technical causes would likely have not survived long enough to receive the intended course of prophylaxis. Indeed, a recent preliminary analysis suggested that CMV conferred a major survival benefit, but this most likely represented a bias toward those children who may have survived long enough to receive prophylaxis.18 Most published studies that have reported an adverse impact of CMV seropositivity on heart transplantation focused on the development of CAV. A number of these studies used intravascular ultrasound (IVUS) to document coronary negative remodeling. In our series, the diagnosis of CAV was limited to reported findings from coronary angiography, which is not well suited to identify “early” or “mild” CAV. Moreover, children appear to be at lower risk for developing CAV than adults. The PHTS data set has reported the 5-year freedom from CAV to be 81%, which is much higher than the 58% reported in adult series. Hence, identifying risk factors for the development of CAV is more challenging in the pediatric heart transplant population.19 In addition, the median length of follow-up for survivors is ⬍6 years, and thus longer follow-up could yet reveal an association. Further, one might believe that CMV could lead to more graft loss through both overt and less obvious forms of CAV such as “nonspecific graft dysfunction.” However, we did not detect any association between graft loss and CMV status. The lack of a significant association between CMV status and important clinical outcome measures suggests that further studies need to be performed prior to making recommendations about appropriate surveillance, prophylaxis and treatment of CMV in children undergoing heart transplantation. Although current

The Journal of Heart and Lung Transplantation December 2009

forms of CMV prophylaxis generally have a favorable side-effect profile, more compelling data in children are needed to justify routine use of this strategy—particularly in those at low risk of clinical CMV infection. Renal impairment, usually transient, may result from the administration of these agents. In addition, the cost of prophylaxis must also be considered. The estimated cost of ganciclovir prophylaxis in adults is as high as $7,500 per patient, which is roughly equivalent to one-third the annual cost of routine immunosuppression.20,21 The present analysis has a number of limitations. Importantly, the approach to CMV prophylaxis varied considerably among centers. Variation was noted both in the type of prophylaxis employed and the patient populations who received prophylaxis. Hence, interpreting the benefit of CMV prophylaxis is challenging. In addition, the criteria for identifying clinical CMV disease were not detailed. As such, findings that might be classified as “CMV clinical infection” likely varied among institutions. In conclusion, this analysis of a large multi-institutional registry of pediatric heart transplant recipients has demonstrated that the risk of clinical CMV infection in this population is lower than that of some other published series in adults, although those recipients with a CMV mismatch appeared to be more susceptible. There was no association between recipient CMV status and the development of CAV, graft loss or patient survival. Additional studies are needed to define the best approach to CMV prophylaxis in children after heart transplantation. DISCLOSURE STATEMENT The authors have no conflicts of interest to disclose.

REFERENCES 1. Sambiase NV, Higuchi ML, Nuovo G, et al. CMV and transplantrelated coronary atherosclerosis: an immunohistochemical, in situ hybridization, and polymerase chain reaction in situ study. Mod Pathol 2000;13:173–9. 2. Valantine HA. Role of CMV in transplant coronary artery disease and survival after heart transplantation. Transplant Infect Dis 1999;1(suppl 1):25–30. 3. Decoene C, Pol A, Dewilde A, et al. Relationship between CMV and graft rejection after heart transplantation. Transplant Int 1996;9(suppl 1):S241–2. 4. Hussain T, Burch M, Fenton MJ, et al. Positive pretransplantation cytomegalovirus serology is a risk factor for cardiac allograft vasculopathy in children. Circulation 2007;115:1798 – 805. 5. Gall SA. Maternal immunization to protect the mother and neonate. Expert Rev Vaccines 2005;4:813– 8. 6. Pahl E, Naftel DC, Kuhn MA, et al. The impact and outcome of transplant coronary artery disease in a pediatric population: a 9-year multi-institutional study. J Heart Lung Transplant 2005;24: 645–51.

The Journal of Heart and Lung Transplantation Volume 28, Number 12

7. Boucek RJ Jr, Naftel D, Boucek MM, et al. Induction immunotherapy in pediatric heart transplant recipients: a multicenter study. J Heart Lung Transplant 1999;18:460 –9. 8. Vozarova L, Huckova D, Noskovicova M, et al. The occurrence of symptomatic CMV infections in heart transplant recipients. Bratisl Lek Listy 2003;104:317–22. 9. Bunde T, Kirchner A, Hoffmeister B, et al. Protection from cytomegalovirus after transplantation is correlated with immediate early 1-specific CD8 T cells. J Exp Med 2005;201:1031– 6. 10. Koskinen P, Lemstrom K, Bruning H, et al. Cytomegalovirus infection induces vascular wall inflammation and doubles arteriosclerotic changes in rat cardiac allografts. Transplant Proc 1995; 27:574 –5. 11. Simmonds J, Fenton M, Dewar C, et al. Endothelial dysfunction and cytomegalovirus replication in pediatric heart transplantation. Circulation 2008;117:2657– 61. 12. Fearon WF, Potena L, Hirohata A, et al. Changes in coronary arterial dimensions early after cardiac transplantation. Transplantation 2007;83:700 –5. 13. Weill D. Role of cytomegalovirus in cardiac allograft vasculopathy. Transplant Infect Dis 2001;3(suppl 2):44 – 8. 14. Potena L, Grigioni F, Ortolani P, et al. Relevance of cytomegalovirus infection and coronary-artery remodeling in the first year after heart transplantation: a prospective three-dimensional intravascular ultrasound study. Transplantation 2003;75:839 – 43.

Mahle et al.

1305

15. Hussain T, Burch M, Fenton MJ, et al. Positive pretransplantation cytomegalovirus serology is a risk factor for cardiac allograft vasculopathy in children. Circulation 2007;115:1798 – 805. 16. Boucek MM, Aurora P, Edwards LB, et al. Registry of the International Society for Heart and Lung Transplantation: tenth official pediatric heart transplantation report—2007. J Heart Lung Transplant 2007;26:796 – 807. 17. Taylor DO, Edwards LB, Boucek MM, et al. Registry of the International Society for Heart and Lung Transplantation: twentyfourth official adult heart transplant report—2007. J Heart Lung Transplant 2007;26:769 – 81. 18. Davies R, Russo MJ, Mital SR, et al. The influence of CMV status on outcomes following heart transplantation in children—whose status matters? J Heart Lung Transplant 2009;28(suppl):2. 19. Costanzo MR, Naftel DC, Pritzker MR, et al. Heart transplant coronary artery disease detected by coronary angiography: a multiinstitutional study of preoperative donor and recipient risk factors. J Heart Lung Transplant 1998;17:744 –53. 20. Dayton JD, Kanter KR, Vincent RN, et al. Cost-effectiveness of pediatric heart transplantation. J Heart Lung Transplant 2006;25: 409 –15. 21. Gerbase MW, Dubois D, Rothmeier C, et al. Costs and outcomes of prolonged cytomegalovirus prophylaxis to cover the enhanced immunosuppression phase following lung transplantation. Chest 1999;116:1265–72.