- Email: [email protected]
Infectious, Malignant, and Autoimmune Complications in Pediatric Heart Transplant Recipients
Infectious, Malignant, and Autoimmune Complications in Pediatric Heart Transplant Recipients AGNIESZKA KULIKOWSKA, MD, SARAH E. BOSLAUGH, PHD, CHARLES B. HUDDLESTON, MD, SANJIV K. GANDHI, MD, CARL GUMBINER, MD, AND CHARLES E. CANTER, MD
Objective To review clinical courses of pediatric heart transplant survivors after 5 years from transplantation for infections, lymphoproliferative, and autoimmune diseases. Study design A total of 71 patients were examined in 2 groups, infant recipients (underwent transplant <1 year of age, n ⴝ 38) and older recipients (underwent transplant >1 year, n ⴝ 33). All patients received comparable immunosuppression. Calculated occurrence rates were reported as means per 10 years of follow-up with SEs. Differences were examined by using Poisson regression. Results Infant recipients had significantly higher (P < .001) occurrence rates of severe (mean, 2.04 ⴞ 0.5) and chronic infections (mean, 4.58 ⴞ 0.67) compared with older recipients (means, 0.37 ⴞ 0.19 and 1.87 ⴞ 0.70, respectively). Types of infections were similar to those in the general population with extremely rare opportunistic infections; however, they were more severe and resistant to treatment. Autoimmune disorders occurred at a frequency comparable with lymphoproliferative diseases and were observed in 7 of 38 infants (18%). Most common were autoimmune cytopenias. Conclusions Infant heart transplant recipients who survive in the long term have higher occurrence rates of infections compared with older recipients. Autoimmune disorders are a previously unrecognized morbidity in pediatric heart transplantation. (J Pediatr 2008;152:671-7) n the last 2 decades, heart transplantation has become an accepted therapy for end-stage heart disease in infants, children, and adolescents. The International Society for Heart and Lung Transplantation database1 demonstrates that ⬎50% of pediatric cardiac allografts survive ⬎10 years. Infant heart transplant recipients have a poorer short-term survival rate than older children because of an increased risk of mortality from primary graft failure.1,2 However, if a pediatric heart transplant recipient survives for 12 months, an infant graft has a superior survival half life of ⬎18 years compared with older grafts. When survival curves are extended to 10 years, there is a greater proportion of 10-year survivors among infant heart transplant recipients than among older children and adolescents.1 This survival advantage is primarily related to the decreased risk of rejection3 and transplant coronary arteriopathy4 in infant recipients compared with older children and adolescents. Because most long-term outcome studies of pediatric heart transplant recipients are composed of patients who underwent transplantation in childhood and adolescence,5-8 relatively little is known about other long-term morbidity in infant heart transplant recipients. Infant recipients may have unique immunologic characteristics, which may increase their long-term risk for infection.9 Age-specific differences in antibody response to pneumococcal antigens have been observed in pediatric heart transplant recipients, which might translate into age-specific differences in From the Department of Pediatrics (A.K., susceptibility to infection.10 The initial objective of this study was to investigate the differences S.B., C.C.), and Department of Surgery in prevalence and types of infections and lymphoproliferative disease in long-term pediatric (C.H., S.G.), Washington University, St. Louis, MO; and Department of Pediatrics, heart transplant survivors who underwent transplantation in infancy versus those who underUniversity of Nebraska, Omaha, NE (C.G.). went transplantation at an older age. In addition to these recognized morbidities, this study Submitted for publication May 13, 2007; reports on the prevalence of autoimmune diseases, a previously unrecognized morbidity in last revision received Aug 13, 2007; accepted Oct 13, 2007. long-term survivors of pediatric heart transplants.
I
METHODS We performed a retrospective review of patients who underwent heart transplantation at St. Louis Children’s Hospital who survived a minimum of 5 years after EBV
Epstein-Barr virus
PTLD
Post-transplant lymphoproliferative disease
Reprint requests: Agnieszka Kulikowska, MD, Dept of Pediatrics, Washington University, Division of Pediatric Cardiology, 4990 Children’s Pl, NW Tower, 8th floor, St. Louis, MO 63110. E-mail: [email protected]. 0022-3476/$ - see front matter Copyright © 2008 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2007.10.018
671
transplantation with the approval of the Washington University Human Studies Committee. The entry point for analysis (time zero) was 5 years after transplantation, a previously used start-point for long-term outcomes in solid organ transplantation,5 with the end-point of patient data collection occurring at death, re-transplantation, or last follow-up ending October 2005. All patients were receiving a similar surveillance program of non-invasive examination every 6 months and invasive examination with cardiac catheterization, coronary angiography, and endomyocardial biopsy every 2 years. All patients were initiated on a previously described11 immunosuppression protocol of cyclosporine, azathioprine, and steroids, with steroid elimination by 6 months after transplant. The target trough cyclosporine levels in the long-term survivors was 110 to 150 ng/mL. The target azathioprine dose was 1 to 2 mg/kg/day, adjusted for neutrophil counts. Prednisone, when continued, was given at a dose of 0.1 to 0.2 mg/kg/day. With time, tacrolimus was substituted for cyclosporine at target trough levels of 5 to 8 ng/mL in patients with recurrent rejection, gingival hyperplasia, hirsutism, or hypertension. Mycophenolate mofetil was substituted for azathioprine, with target trough levels of 2 to 4 ng/mL12 in patients with recurrent rejection or to inhibit B cell proliferation13 after successful treatment of post-transplant lymphoproliferative disease (PTLD). Within the past 5 years, sirolimus at target trough levels of 5 to 10 ng/mL was used as a substitute for calcineurin inhibitors in patients with progressive renal dysfunction14 or as a substitute for azathioprine in patients in whom angiographic evidence of transplant coronary arteriopathy developed.15 All patients received standard immunizations for their ages when killed vaccines could be used. Live virus vaccines were not given. The pneumococcal vaccine was administered through primary care physicians after it became widely available. The specific types of pneumococcal vaccine and number of doses varied as a result of the extended time course of the observation period and the different physician administrators. Our analysis focused on non-cardiac immune complications of long-term immunosuppression that could not be attributed to specific toxicities of any given immunosuppressive drug. Specifically we looked at these morbidities: 1) Serious infections were defined as those requiring hospitalization, intravenous antibiotics, or both. Hospital admissions for observation of varicella infections were excluded from analysis. Patients who were hospitalized were treated by the heart transplant service according to standards of care for their clinical diagnosis, which frequently involved consultation with an infectious disease specialist. 2) Chronic/recurrent infections were defined as ⱖ3 infections of the same type per year or infections that did not respond to standard treatment. Common infections (eg, otitis media, sinusitis, ambulatory pneumonias) were initially treated by the patient’s primary care physician with frequent referral to appropriate medical/surgical subspecialists for consultation and management. 672
Kulikowska et al
3) Post-transplant lymphoproliferative disease was diagnosed on the basis of positive pathology results from a tissue biopsy and were treated with immunosuppression reduction with and without excision or with chemotherapy, depending on pathologic tissue evaluation. 4) Autoimmune diseases were diagnosed on the basis of the presence of specific autoantibodies or positive tissue biopsy results. Occurrence rates were calculated for the number of serious infections, PTLD, or autoimmune disease events divided by years of follow-up from time zero (5 years after transplant) and expressed as number of events per 10 years of follow-up from time zero. For chronic/recurrent infections, a rate of number of years with various chronic/recurrent infection divided by number of years of follow-up from time zero was calculated and expressed as years of recurrent/chronic infection per 10 years of follow-up from time zero. Means and standard errors of these occurrence rates were calculated for recipients who underwent transplantation when they were ⬍1 year of age (infant group) and ⬎1 year of age (older group), with presentation of the data as the mean ⫾ SEM. Proc Genlog in SAS software (version 0.1) was used to perform Poisson regression for comparison of the occurrence rates for the infant and older groups, with a P value ⬍.05 set as the criteria for a significant difference between groups. Odds ratios and 95% CIs were reported.
RESULTS Patient Population Between 1986 and 2000, 163 patients received 170 heart transplantations in our institution (86 infants and 77 older recipients). 65 infant recipients (75.6%) and 52 older recipients (67.5%) survived ⬎5 years after transplantation. There were 14 early (⬍3 months after transplant) and 7 late (between 3 months and 5 years) deaths in the infant group and 12 early and 13 late deaths in the older recipient group. The records of 27 infant and 19 older recipients observed long-term at other institutions were unavailable for review. Thus the study group contained 71 subjects: 38 infant recipients (58% of 5-year survivors and 44% of all infants who underwent transplantation) and 33 older recipients (63% of 5-years survivors and 43% of all older children who underwent transplantation). The infant group underwent transplantation at a mean age of 1.71 months and had a mean follow-up of 6.5 years after time zero or 11.5 years after transplantation. 22 of 38 patients (58%) in the infant group survived ⬎10 year after heart transplantation. The older group underwent transplantation at a mean age of 8.44 years and had a mean follow-up of 4.5 years after time zero or 9.5 years after transplantation. 11 of 33 patients (33%) in the older group survived ⬎10 year after heart transplantation. The groups were not significantly different in their sexual or racial compositions. During the follow-up period, there were 3 deaths in the infant group: 1 from transplant coronary arteriopathy 7 years after transplant, 1 from PTLD 5.5 years after transplant, and 1 from progressive pulmonary veno-occlusive disease 6.5 years The Journal of Pediatrics • May 2008
Infant Recipients At 5 years
5
At last follow up
3 3 10.5
5
18
8 3 3
26
58
10.5 47
CSA/AZO TAC/MMF
TAC/AZO AZO
pred/CSA/AZO MMF
CSA/AZO MMF
TAC/AZO SIR/AZO
TAC/MMF SIR/MMF
AZO
Older Recipients At 5 years
12
At last follow up 3 3 3
3 24
9
6
3 15
12 6 40
CSA/AZO pred/TAC/MMF TAC
43
6
TAC/AZO CSA/MMF
9
pred/CSA/AZO TAC/MMF
CSA/AZO Pred/TAC/MMF CSA/SIR TAC
3
TAC/AZO Cyclo/MMF TAC/SIR
pred/CSA/AZO TAC/MMF SIR/MMF
Figure. Immunosuppression therapy over the observation period of the study. The distribution of immunosuppressant combinations is displayed for the infant and older patients groups at 5 years after transplantation, the starting point of the study, and at last follow-up.
after transplant in a patient initially born with hypoplastic left heart syndrome. There were 2 deaths in the older group: both from transplant coronary arteriopathy, 6.5 years after transplantation in 1 case and 10.5 years after transplantation in the other. In addition, 4 patients in the older group underwent re-transplantation for transplant coronary arteriopathy within the study period. Rejection in both groups during the study period was rare. The infant group had 9 episodes of rejection in 4 patients with a cumulative 249 years of follow-up. 2 of these rejection episodes were associated with reduction of immunosuppression for PTLD. The older group had 5 episodes of rejection in 4 patients with 1 of the episodes associated with immunosuppression reduction for the treatment of PTLD. The Figure demonstrates the various combinations of immunosuppressive drugs used at entry of the study, at 5 years after transplant, and at last follow-up. 5 years after transplant, 63% of
the infant group was taking cyclosporine, 92% was taking azathioprine, and only 5% (2 patients) was taking triple therapy with prednisone. At that time, 67% of the older group was taking tacrolimus, 70% was taking azathioprine, and only 18% (6 patients) was taking triple therapy with prednisone. At last followup, most infant recipients (57.5%) were taking tacrolimus, 76% were taking azathioprine, 18.5% were taking sirolimus, and no patient was taking triple therapy with steroids. In the older group at last follow-up, 73% were taking tacrolimus, 55% were taking azathioprine, 12% were taking sirolimus, and 12% remained receiving triple therapy with prednisone.
Severe Infections Within the study period, there were 49 serious infections in the infant group and 9 serious infections in the older
Infectious, Malignant, and Autoimmune Complications in Pediatric Heart Transplant Recipients
673
Table I. Types and number of severe infections in the infant and older heart transplant recipients groups during the observation period of the study Infant recipients Older recipients Pneumonia Bacteremia Viral respiratory infection Gastroenteritis Sinusitis Parvovirus Osteomyelitis Others Total
23 2 5 3 5 3 2 6 49
1 1 0 3 0 1 0 3 9
group. Serious infection was not associated with mortality in the infant or older groups. The occurrence rate of serious infections after 5-year survival in the infant group was 2.04 ⫾ 0.5 per patient per 10 years of follow-up, compared with 0.37 ⫾ 0.19 per patient per 10 years of follow-up after 5-year survival in the older group. The difference was statistically significant (P ⬍ .001), with the odds ratio 3.3 and 95% CI of 1.6 to 6.7. In the infant group, 18 of 38 patients (47%) had at least 1 serious infection, and 8 of 38 patients (21%) had ⱖ3 serious infections, compared with a prevalence of only 4 of 33 patients (12%) in the older recipient group with at least 1 serious infection and 1 of 33 patients (3%) with ⱖ3 serious infections. Table I summarizes the types and numbers of serious infections encountered in the infant and older groups. The overall types of infection were similar in the 2 groups, with the exception of the occurrence of mononucleosis, osteomyelitis, pyelonephritis, and bronchiectasis in the infant group as opposed to the older group. Pertussis was observed in 1 older recipient but no infants. Specific organisms associated with presumed bacterial infections were usually not recovered, as was observed in nearly half the cases of presumed bacterial pneumonia and 2 cases of osteomyelitis. The success rate of pathogen recovery in severe presumed bacterial pneumonia was 54% (12/22 cases), and most of these patients underwent bronchoalveolar lavage. Pneumococcus was the most frequently documented bacterial pathogen, responsible for 5 cases of pneumonia and 2 cases of bacteremia. In the infant group, bacterial pneumonia accounted for 44.9% of all severe infections, with an occurrence rate of 0.91 per 10 years, and pneumococcal infections (5 pneumonia and 2 bacteriemia) represented 14.3% of severe infections with the occurrence rate of 0.29 per 10 years follow-up. Besides Pneumococcus, the other recovered pathogens included Moraxella catarrhalis, Mycoplasma pneumoniae, Pseudomonas aeruginosa in a patient with a tracheostomy, Staphylococcus aureus, and -hemolytic streptococcus. Opportunistic/unusual infections associated with immunosuppression were uncommon, with only 1 case of Pneumocystis carinii and 4 cases (3 infant, 1 older recipient) of Parvovirus B19 hypoplastic anemia. In patients with recurrent serious infections, evaluation of immunoglobulins revealed 2 instances of hypogammaglobuline674
Kulikowska et al
Table II. Types of chronic/recurrent infections and numbers of years with chronic/recurrent infections in the infant and older heart transplant recipients groups during the observation period of the study
Sinusitis Otitis media Otitis externa Tonsillitis Pneumonia Urinary tract infection Verruca/molluscum Others Total
Infant recipients
Older recipients
34 14 2 3 18 1 23 7 102
4 2 1 1 0 4 6 1 19
mia and 1 case of Ig A deficiency in 3 infant recipients. In 3 cases of recurrent pneumococcal infection in the infant group, immunization with the pneumococcal vaccine was associated with an absence of appropriate antibody response.
Chronic Infections Patients in the infant heart transplant group also experienced a higher occurrence rate of chronic/recurrent infections (mean, 4.58 ⫾ 0.67 years with chronic/recurrent infections per 10 years of follow-up after 5-year survival) than the older recipients (mean, 1.84 ⫾ 0.70 years with chronic/recurrent infections per 10 years of follow-up after 5-year survival). The difference was statistically significant (P ⬍ .001), with an odds ratio 4.5 and 95% CI of 2.7 to 7.6. Table II demonstrates the number and types of chronic/recurrent infections. The most common types of chronic/recurrent infections in the infant group were sinusitis, skin warts (verruca vulgaris and molluscum contagiosum), pneumonia, and otitis media, whereas skin warts, sinusitis, and urinary tract infections were the most common types of chronic/recurrent infections in the older group. Although these chronic infections were similar to those observed in the general population, the infant recipients had very severe forms of these infections. As stated earlier, bronchiectasis developed in 3 infant recipients as a consequence of recurrent pneumonias, and 16 patients required surgical otolaryngological procedures for recurrent infections that were unresponsive to antibiotics. 8 infants underwent myringotomies after 5 years post-transplantation, 2 infants underwent sinus operations, and 5 infants and 1 older recipient underwent tonsillo-adenoidectomies. In addition, 5 infant recipients had diagnostic bronchoscopies for recurrent pneumonias. PTLD Higher, but not statistically significant (P ⫽ .12), occurrence rates of PTLD were observed in the infant group (m ⫽ 0.40 ⫾ 0.27 cases per patient per 10 years of follow-up after 5 years) versus the older group (m ⫽ 0.10 ⫾ 0.10 cases per patient per 10 years of follow-up after 5 years). There The Journal of Pediatrics • May 2008
Table III. Individual cases of autoimmune diseases in the infant and older heart transplant recipients groups during the observation period of the study
Autoimmune diseases in infant recipients Membranoprolipherative glomerulonephritis Hypothyroidism Idiopathic thrombocytopenic purpura Idiopathic thrombocytopenic purpura Autoimmune hemolytic anemia Chronic bullous disease Autoimmune hemolytic anemia Autoimmune hemolytic anemia Autoimmune hepatitis Autoimmune diseases in older recipients Crohn’s disease
Onset after transplant
Immunosuppression
14 years (13 years after heart retransplantation) 9 years 9 years (1 year after renal transplant)
Tacrolimus/Imuran Cyclosporine/Imuran Tacrolimus/Imuran
8 years
Tacrolimus/Imuran
8 years 7 years 6 years
Tacrolimus/Imuran Tacrolimus/Imuran Tacrolimus/Imuran
9 years
Cyclosporine/MMF
were 5 cases of PTLD in the infant group and 1 case in the older group within the study period. Most cases were B cell lymphomas: 2 polymorphic B cell lymphomas, 1 polymorphous hyperplasia, 1 monomorphic B cell lymphoma, and 1 large cell Burkitt’s lymphoma. They were treated successfully with chemotherapy or immunosupression reduction and tonsillo-adenoidectomy, when tonsils and adenoids were involved. The 2 mortalities included 1 infant recipient with a rare type of hepatosplenic gamma delta T cell lymphoma with hemophagocytic syndrome who died of a relapse of PTLD despite intensive chemotherapy and 1 older recipient in whom severe fatal graft coronary vasculopathy developed 1 year after remission of the lymphoma. The patients in whom PTLD developed received a standard immunosupressive therapy with a combination of calcineurin inhibitor (3 patients cyclosporine and 3 patients tacrolimus) and azothioprone. In all 6 cases the onset of PTLD did not correlate with the onset of an acute Epstein-Barr virus (EBV) infection demonstrable with serology or, when available, copies of the viral DNA demonstrable with polymerase chain reactivity of blood samples. Patients with PTLD did not have an increased occurrence rate of serious or chronic/recurrent infections compared with the rest of the group.
Autoimmune Disease We identified an occurrence rate of autoimmune disease in infant recipients that was similar to the occurrence rate of PTLD (mean, 0.53 ⫾ 0.24 cases per patient per 10 years of follow-up after 5 years). There was 1 documented case of autoimmune disease in an older recipient, leading to a mean occurrence rate of 0.04 ⫾ 0.04 cases per patient per 10 years of follow-up after 5 years. This difference between the occurrence rates of the groups was not statistically different (P ⫽ .26). Table III illustrates the individual cases of autoimmune disease. The most common were autoimmune cytopenias, with single cases of autoimmune thyroid, kidney, bowel, and cutaneous bullous disease. All 4 patients in whom autoimmune cytopenias developed received tacrolimus. 1 case of autoimmune hemolytic anemia and 1 case of idiopathic trom-
bocytopenic purpura were associated with an acute infection with Parvovirus B19. None of the autoimmune disorders were fatal, and all responded to treatment with steroids, intravenous gamma globulin, or both. Similar to patients with PTLD, patients with autoimmune disease did not experience increased occurrence rate of serious or chronic/recurrent infections compared with the rest of the group.
DISCUSSION We observed significantly higher occurrence rates of severe and chronic/recurrent but non-fatal infections in infants recipients than in older recipients ⬎5 years after transplantation. Earlier long-term outcome studies5-8 in pediatric heart transplant recipients have reported that infection declines, but is not eliminated, as a cause of morbidity and mortality late after transplantation. Most studies on infectious complications after pediatric heart transplantation have focused on problems in the early post-transplant period, when infection is an important cause of mortality.16-18 Age-specific differences in rates of infections have been reported in liver and kidney recipients,19 and younger recipients (⬍2 years after transplant) had higher occurrence rates of bacterial and viral infection compared with older recipients. In a multicenter study of pediatric heart transplant recipients focusing on infection in the first year after transplant,17 infant recipients had a higher incidence of bacterial infections compared with older recipients. It has been well established in the general population20 that young age is a significant risk factor for increased rates of infections such as influenza or invasive pneumococcal disease, so the age difference in the studied groups could itself account for higher infectious morbidities in infant recipients, irrelevant to the immunosupression status.20 However, the occurrence rates of severe bacterial pneumonia (0.91 per 10 years or 9,100/100,000) and invasive pneumococcal disease (0.29 per 10 years or 2,900/100,000) in our infant recipients group (age 5-20 years during study period) exceeds the reported occurrence rates in general population and high-risk groups. Grijalva et al report an estimated rate of all-cause pneumonia
Infectious, Malignant, and Autoimmune Complications in Pediatric Heart Transplant Recipients
675
hospitalizations in the general population as 74 to 91 per 100,000 for ages 5 to 17 years and 791 to 1,297 per 100,000 for children ⬍2 years, depending on vaccination era, and pneumococcal pneumonia rates as 1.9 to 3.5 per 100,000 and 9.2 to 26.2 per 100,000, respectively.21 Although the greatest risk for the development of PTLD occurs within the first 6 months after pediatric heart transplantation,22 there is an ongoing constant hazard for its development years after transplantation. The exceptionally long follow-up period (58% survived ⬎10 years) in our infant recipients group could have contributed to the accumulation of PTLD cases during the time and as a result and to higher incidence rates in infants. EBV seronegative status before transplantation is known to be an independent risk factor for PTLD, which might also significantly contribute to higher occurrence rates of PTLD in infant group because most of the infant recipients had seronegative status at the time of transplantation (infants who had positive serologies had a passive maternal antibodies with documentation of seronegative status after several months) and seroconverted later after transplantation. The timing of development of PTLD, however, did not correlate with the time of EBV seroconversion, which supports recent data23 that EBV viremia as measured with polymerase chain reactivity is a relatively insensitive predictor for the development of PTLD, especially late after transplantation. Our experience is also consistent with earlier reports24 suggesting late-onset (⬎3 years) PTLD is more likely to be monomorphic and require more aggressive treatment than early onset PTLD, which tends to be polymorphic, EBV and CD 20 positive, and often responsive to immunosuppression reduction alone. The presence of 9 cases of autoimmune disease in 7 primarily infant recipients was an unanticipated morbidity in our infant recipients. There have been isolated case reports of autoimmune disease developing in pediatric solid organ recipients. Crohn’s disease25 and multiple autoantibody syndrome26 have been reported in heart transplant recipients, and immunobullous disease27 has been reported in liver and kidney recipients. Although all of our 4 patients with autoimmune cytopenias were receiving tacrolimus, the number of patients in our study was too small to perform any statistical analysis to determine whether the autoimmune cytopenias were specifically tacrolimus related. Autoimmune disease is a well-known complication of a number of primary immunodeficiency diseases observed in children.28 In these diseases, it is hypothesized that the basis of autoimmunity is the inability of a patient with these diseases to completely and effectively eradicate pathogens and their associated antigens. A compensatory chronic inflammatory response via less-effective alternative pathways, which may be exaggerated, leads to damage to healthy tissue and infected tissue. Autoimmunity in this setting is not felt to be a breakdown of tolerance to selfantigens, but rather part of the attempt of an abnormal immune system to rid itself of foreign antigens. This potential mechanism is supported in our population by the concurrent presence of autoimmune hemolytic anemia or idiopathic 676
Kulikowska et al
thrombocytopenic purpura in the presence of Parvovirus B19 infection, a clinical phenomenon that has been described.29 The long-term survivors of infant heart transplantation reported in this study demonstrate a persistent tendency for severe infections, recurrent/chronic infections, or both, especially pneumococcal disease, and the potential to develop autoimmune disease, most commonly hematologic cytopenias, later in childhood. The presence of humoral immunity dysfunction in the setting of T cell dysfunction in this patient group is similar to the findings in older children with 22q11.2 deletion (DiGeorge syndrome),30,31 whereas older patients have a similar clinical picture felt to reflect an aberrant T cell/B cell interaction. In addition to the differences in occurrence rates of infection observed between infant and older heart transplant recipients, infant heart transplant recipients also are different in that they undergo thymectomy at the time of surgery in addition to receiving immunosuppressant medications. Although removal of the thymus in infancy generally causes modest defects in the T cell compartment structure with little or no impact on long-term morbidity, a recent study in infant heart transplant recipients9 found that these patients demonstrate absence of thymic function, a profound decrease in the diversity of the T cell repertoire, and are less able to mount an antibody response to immunization. These authors speculated that this type of immune dysfunction might lead to a greater risk of infection and manifestations of autoimmunity with age. Further evidence of the potential differences in immune function between infant and older heart transplant recipients was demonstrated in a study from the United Kingdom in which only 25% of the younger recipients (⬍4 years at time of transplant) responded to pneumococcal immunization compared with 82% of the older group.10 This observational study is most limited by the small sample size in each group. The retrospective nature of the study is associated with a lack of control of immunosuppression regimens, which were manipulated according to the individual clinical picture of each subject. Serious infections, chronic/recurrent infections, or both did not appear to be clustered in patients in either group treated for PTLD or autoimmune disease. It appeared that the infant group, with the exception of temporary intensification of steroid therapy or other immunosuppressants for treatment of PTLD and autoimmune disease, had immunosuppression comparable with the older group and that the magnitude of immunosuppression was managed with similar target drug levels for each group. Therapeutic drug level monitoring, however, does not give a picture of overall degree of immunosuppression in a given patient. A recent test developed to assess global assessment of cellular immune function has demonstrated that pediatric transplant recipients ⬍12 years old appear to be more immunosuppressed on comparable immunosuppression than adults or patients ⬎12 years old.32 Although this study did not assess the effects of age at transplantation on global immune status, it is conceivable that infant pediatric heart transplant recipients may be more immunosuppressed than The Journal of Pediatrics • May 2008
older recipients despite the same target drug levels with the same immunosuppressive drug protocols. Our findings suggest that in addition to neoplasm, infection, and immunosuppressant drug toxicity, autoimmune disease should be considered as an etiology for non-cardiac problems in pediatric heart transplant recipients, especially hematologic cytopenias.
REFERENCES 1. Boucek MM, Waltz DA, Edwards LB, Taylor DO, Keck BM, Trulock EP, et al. Registry of the International Society for Heart and Lung Transplantation: ninth official pediatric heart transplantation report—2006. J Heart Lung Transplant 2006;25: 893-903. 2. Canter CE, Naftel D, Caldwell R, Chinnock R, Pahl E, Frazier E, et al. Survival and risk factors for death after cardiac transplantation in infants: a multi-institutional study. Circulation 1997;96:227-31. 3. Ibrahim JE, Sweet SC, Flippin M, Dent C, Mendeloff E, Huddleston CB, Canter CE. Rejection is reduced in thoracic organ recipients when transplanted in the first year of life. J Heart Lung Transplant 2002;21:311-8. 4. Pahl E, Naftel DC, Kuhn MA, Shaddy RE, Morrow WR, Canter CE, 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. 5. Sigfusson G, Fricker JF, Bernstein D, Addonizio LJ, Baum D, Hsu DT, et al. Long-term survivors of pediatric heart transplantation: a multicenter report of sixtyeight children who have survived longer than five years. J Pediatr 1997;130:862-71. 6. Gajarski RJ, Smith EO, Denfield SW, Rosenblatt HM, Kearney D, Frazier OH, et al. Long-term results of triple-drug-based immunosuppression in nonneonatal pediatric heart transplant recipients. Transplantation 1998;65:1470-6. 7. Radley Smith R, Wray J, Khaghani A, Yacoub M. Ten year survival after paediatric heart transplantation: a single center experience. Eur J Cardiothorac Surg 2005;27:790-4. 8. Ross M, Kouretas P, Gamberg P, Miller J, Burge M, Reitz B, et al. Ten- and 20-year survivors of pediatric orthotopic heart transplantation. J Heart Lung Transplant 2006;25:261-70. 9. Ogle BM, West LJ, Driscoll DJ, Strome SE, Razonable RR, Paya CV et al. Effacing of the T cell compartment by cardiac transplantation in infancy. J Immunol 2006;176:1962-7. 10. Gennery AR, Cant AJ, Spickett GP, Walshaw D, Hunter S, Hasan A, et al. Effect of immunosuppression after cardiac transplantation in early childhood on antibody response to polysaccharide antigen. Lancet 1998;351:1778-81. 11. Canter CE, Moorhead S, Saffitz JE, Huddleston CB, Spray TL. Steroid withdrawal in the pediatric heart transplant recipient initially treated with triple immunosuppression. J Heart Lung Transplant 1994;13:74-9. 12. Yamani MH, Starling RC, Goormastic M, Van Lente F, Smedira N, McCarthy P, et al. The impact of routine mycophenolate mofetil drug monitoring on the treatment of cardiac allograft rejection. Transplantation 2000;11:2326-30. 13. Taylor DO, Ensley RD, Olsen SL, Dunn D, Renlund DG. Mycophenolate mofetil (RS-61443): preclinical, clinical, and three-year experience in heart transplantation. J Heart Lung Transplant 1994;13:571-82.
14. Snell GL, Levvey BJ, Chin W, Kotsimbos T, Whitford H, Waters KN, et al. Sirolimus allows renal recovery in lung and heart transplant recipients with chronic renal impairment. J Heart Lung Transplant 2002;21:540-6. 15. Mancini D, Pinney S, Burkhoff D, LaManca J, Itescu S, Burke E, et al. Use of rapamycin slows progression of cardiac transplantation vasculopathy. Circulation 2003; 108:48-53. 16. Keough WL, Michaels MG. Infectious complications in pediatric solid organ transplantation. Pediatr Clin N Am 2003;50:1451-69. 17. Schowengerdt K, Naftel D, Seib P, Pearce B, Addonizio L, Kirklin J, et al. Infection after pediatric heart transplantation: results of a multi-institutional study. J Heart Lung Transplant 1997;16:1207-16. 18. Doelling NR, Kanter KR, Sullivan KM, Winn KJ, Vincent R. Medium-term results of pediatric patients undergoing orthotopic heart transplantation. J Heart Lung Transplant 1997;16:1225-30. 19. Tran L, Hebert D, Dipchand A, Fecteau A, Richard S, Allen U. Invasive pneumococcal disease in pediatric organ transplant recipients: a high risk population. Pediatr Transplantation 2005;9:183-6. 20. Behrman RE, Kliegman RM, Jenson HB. Nelson textbook of pediatrics, 17th ed. Abramson JS, Overtruf GD: Chapter 167: Streptococcus pneumoniae (Pneumococcus) p. 867-9; Wright P: Chapter 237: Influenza viruses, p. 1072-5. 21. Grijalva CA, Nuorti JP, Arbogast PG, Martin SW, Edwards KM, Griffin MR. Decline in pneumonia admissions after routine childhood immunisation with pneumococcal conjugate vaccine in the USA: a time-series analysis. Lancet 2007;369:1179-86. 22. Webber SA, Naftel DC, Fricker FJ, Olesnevich P, Blume ED, Addonizio L, et al. Lymphoproliferative disorders after paediatric heart transplantation: a multi-institutional study. Lancet 2006;367:233-9. 23. Axelrod DA, Holmes R, Thomas SE, Magee JC. Limitations of EBV-PCR monitoring to detect EBV associated post-transplant lymphoproliferative disorder. Pediatr Transplantation 2003;7:223-27. 24. Mayo P, Ghobrial IM, Habermann TM, Macon WR, Ristow KM, Larson TS, et al. Differences between early and late posttransplant lymphoproliferative disorders in solid organ transplant patients: are they two different diseases? Transplantation 2005; 79:244-7. 25. Harms B, Bremner AR, Mulligan J, Fairhurst J, Griffiths DM, Salmon T, et al. Crohn’s disease post-cardiac transplantation presenting with severe growth failure and delayed onset of puberty. Pediatr Allergy Immunol 2004;15:186-9. 26. Rawal A, Sarode R, Curtis B, Karandikar N, Friedman K, Rogers Z. Acquired Glanzmann’s thrombasthenia as part of multiple-autoantibody syndrome in a pediatric heart transplant patient. J Pediatr 2004;144:672-4. 27. Morelli JG, Weston WL. Childhood immunobullous disease following a second organ transplant. Pediatr Dermatol 1999;16:205-7. 28. Arkwright PD, Abinun M, Cant AJ. Autoimmunity in human primary immunodeficiency diseases. Blood 2002;99:2694-702. 29. De la Rubia J, Moscardo F, Arriaga F, Monteagudo E, Carreras C, Marty ML. Acute parvovirus B19 infection as a cause of autoimmune hemolytic anemia. Heamatologica 2000;85:995-7. 30. Jawad AF, McDonald-McGinn DM, Zackai E, Sullivan KE. Immunologic features of chromosome 22q11.2 deletion syndrome (DiGeorge syndrome/velocardiofacial syndrome). J Pediatr 2001;139:715-23. 31. Gennery AR, Barge D, O’Sullivan JJ, Flood TJ, Abinun M, Cant AJ. Antibody deficiency and autoimmunity in 22q11.2 deletion syndrome. Arch Dis Child 2002; 86:422-5. 32. Hooper E, Hawkins DM, Kowalski RJ, Post DR, Britz JA, Brooks KC, et al. Establishing the pediatric immune response zones using the Cylex Immuknow assay. Clin Transplantation 2005;19:834-9.
Infectious, Malignant, and Autoimmune Complications in Pediatric Heart Transplant Recipients
677
Objective To review clinical courses of pediatric heart transplant survivors after 5 years from transplantation for infections, lymphoproliferative, and autoimmune diseases. Study design A total of 71 patients were examined in 2 groups, infant recipients (underwent transplant <1 year of age, n ⴝ 38) and older recipients (underwent transplant >1 year, n ⴝ 33). All patients received comparable immunosuppression. Calculated occurrence rates were reported as means per 10 years of follow-up with SEs. Differences were examined by using Poisson regression. Results Infant recipients had significantly higher (P < .001) occurrence rates of severe (mean, 2.04 ⴞ 0.5) and chronic infections (mean, 4.58 ⴞ 0.67) compared with older recipients (means, 0.37 ⴞ 0.19 and 1.87 ⴞ 0.70, respectively). Types of infections were similar to those in the general population with extremely rare opportunistic infections; however, they were more severe and resistant to treatment. Autoimmune disorders occurred at a frequency comparable with lymphoproliferative diseases and were observed in 7 of 38 infants (18%). Most common were autoimmune cytopenias. Conclusions Infant heart transplant recipients who survive in the long term have higher occurrence rates of infections compared with older recipients. Autoimmune disorders are a previously unrecognized morbidity in pediatric heart transplantation. (J Pediatr 2008;152:671-7) n the last 2 decades, heart transplantation has become an accepted therapy for end-stage heart disease in infants, children, and adolescents. The International Society for Heart and Lung Transplantation database1 demonstrates that ⬎50% of pediatric cardiac allografts survive ⬎10 years. Infant heart transplant recipients have a poorer short-term survival rate than older children because of an increased risk of mortality from primary graft failure.1,2 However, if a pediatric heart transplant recipient survives for 12 months, an infant graft has a superior survival half life of ⬎18 years compared with older grafts. When survival curves are extended to 10 years, there is a greater proportion of 10-year survivors among infant heart transplant recipients than among older children and adolescents.1 This survival advantage is primarily related to the decreased risk of rejection3 and transplant coronary arteriopathy4 in infant recipients compared with older children and adolescents. Because most long-term outcome studies of pediatric heart transplant recipients are composed of patients who underwent transplantation in childhood and adolescence,5-8 relatively little is known about other long-term morbidity in infant heart transplant recipients. Infant recipients may have unique immunologic characteristics, which may increase their long-term risk for infection.9 Age-specific differences in antibody response to pneumococcal antigens have been observed in pediatric heart transplant recipients, which might translate into age-specific differences in From the Department of Pediatrics (A.K., susceptibility to infection.10 The initial objective of this study was to investigate the differences S.B., C.C.), and Department of Surgery in prevalence and types of infections and lymphoproliferative disease in long-term pediatric (C.H., S.G.), Washington University, St. Louis, MO; and Department of Pediatrics, heart transplant survivors who underwent transplantation in infancy versus those who underUniversity of Nebraska, Omaha, NE (C.G.). went transplantation at an older age. In addition to these recognized morbidities, this study Submitted for publication May 13, 2007; reports on the prevalence of autoimmune diseases, a previously unrecognized morbidity in last revision received Aug 13, 2007; accepted Oct 13, 2007. long-term survivors of pediatric heart transplants.
I
METHODS We performed a retrospective review of patients who underwent heart transplantation at St. Louis Children’s Hospital who survived a minimum of 5 years after EBV
Epstein-Barr virus
PTLD
Post-transplant lymphoproliferative disease
Reprint requests: Agnieszka Kulikowska, MD, Dept of Pediatrics, Washington University, Division of Pediatric Cardiology, 4990 Children’s Pl, NW Tower, 8th floor, St. Louis, MO 63110. E-mail: [email protected]. 0022-3476/$ - see front matter Copyright © 2008 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2007.10.018
671
transplantation with the approval of the Washington University Human Studies Committee. The entry point for analysis (time zero) was 5 years after transplantation, a previously used start-point for long-term outcomes in solid organ transplantation,5 with the end-point of patient data collection occurring at death, re-transplantation, or last follow-up ending October 2005. All patients were receiving a similar surveillance program of non-invasive examination every 6 months and invasive examination with cardiac catheterization, coronary angiography, and endomyocardial biopsy every 2 years. All patients were initiated on a previously described11 immunosuppression protocol of cyclosporine, azathioprine, and steroids, with steroid elimination by 6 months after transplant. The target trough cyclosporine levels in the long-term survivors was 110 to 150 ng/mL. The target azathioprine dose was 1 to 2 mg/kg/day, adjusted for neutrophil counts. Prednisone, when continued, was given at a dose of 0.1 to 0.2 mg/kg/day. With time, tacrolimus was substituted for cyclosporine at target trough levels of 5 to 8 ng/mL in patients with recurrent rejection, gingival hyperplasia, hirsutism, or hypertension. Mycophenolate mofetil was substituted for azathioprine, with target trough levels of 2 to 4 ng/mL12 in patients with recurrent rejection or to inhibit B cell proliferation13 after successful treatment of post-transplant lymphoproliferative disease (PTLD). Within the past 5 years, sirolimus at target trough levels of 5 to 10 ng/mL was used as a substitute for calcineurin inhibitors in patients with progressive renal dysfunction14 or as a substitute for azathioprine in patients in whom angiographic evidence of transplant coronary arteriopathy developed.15 All patients received standard immunizations for their ages when killed vaccines could be used. Live virus vaccines were not given. The pneumococcal vaccine was administered through primary care physicians after it became widely available. The specific types of pneumococcal vaccine and number of doses varied as a result of the extended time course of the observation period and the different physician administrators. Our analysis focused on non-cardiac immune complications of long-term immunosuppression that could not be attributed to specific toxicities of any given immunosuppressive drug. Specifically we looked at these morbidities: 1) Serious infections were defined as those requiring hospitalization, intravenous antibiotics, or both. Hospital admissions for observation of varicella infections were excluded from analysis. Patients who were hospitalized were treated by the heart transplant service according to standards of care for their clinical diagnosis, which frequently involved consultation with an infectious disease specialist. 2) Chronic/recurrent infections were defined as ⱖ3 infections of the same type per year or infections that did not respond to standard treatment. Common infections (eg, otitis media, sinusitis, ambulatory pneumonias) were initially treated by the patient’s primary care physician with frequent referral to appropriate medical/surgical subspecialists for consultation and management. 672
Kulikowska et al
3) Post-transplant lymphoproliferative disease was diagnosed on the basis of positive pathology results from a tissue biopsy and were treated with immunosuppression reduction with and without excision or with chemotherapy, depending on pathologic tissue evaluation. 4) Autoimmune diseases were diagnosed on the basis of the presence of specific autoantibodies or positive tissue biopsy results. Occurrence rates were calculated for the number of serious infections, PTLD, or autoimmune disease events divided by years of follow-up from time zero (5 years after transplant) and expressed as number of events per 10 years of follow-up from time zero. For chronic/recurrent infections, a rate of number of years with various chronic/recurrent infection divided by number of years of follow-up from time zero was calculated and expressed as years of recurrent/chronic infection per 10 years of follow-up from time zero. Means and standard errors of these occurrence rates were calculated for recipients who underwent transplantation when they were ⬍1 year of age (infant group) and ⬎1 year of age (older group), with presentation of the data as the mean ⫾ SEM. Proc Genlog in SAS software (version 0.1) was used to perform Poisson regression for comparison of the occurrence rates for the infant and older groups, with a P value ⬍.05 set as the criteria for a significant difference between groups. Odds ratios and 95% CIs were reported.
RESULTS Patient Population Between 1986 and 2000, 163 patients received 170 heart transplantations in our institution (86 infants and 77 older recipients). 65 infant recipients (75.6%) and 52 older recipients (67.5%) survived ⬎5 years after transplantation. There were 14 early (⬍3 months after transplant) and 7 late (between 3 months and 5 years) deaths in the infant group and 12 early and 13 late deaths in the older recipient group. The records of 27 infant and 19 older recipients observed long-term at other institutions were unavailable for review. Thus the study group contained 71 subjects: 38 infant recipients (58% of 5-year survivors and 44% of all infants who underwent transplantation) and 33 older recipients (63% of 5-years survivors and 43% of all older children who underwent transplantation). The infant group underwent transplantation at a mean age of 1.71 months and had a mean follow-up of 6.5 years after time zero or 11.5 years after transplantation. 22 of 38 patients (58%) in the infant group survived ⬎10 year after heart transplantation. The older group underwent transplantation at a mean age of 8.44 years and had a mean follow-up of 4.5 years after time zero or 9.5 years after transplantation. 11 of 33 patients (33%) in the older group survived ⬎10 year after heart transplantation. The groups were not significantly different in their sexual or racial compositions. During the follow-up period, there were 3 deaths in the infant group: 1 from transplant coronary arteriopathy 7 years after transplant, 1 from PTLD 5.5 years after transplant, and 1 from progressive pulmonary veno-occlusive disease 6.5 years The Journal of Pediatrics • May 2008
Infant Recipients At 5 years
5
At last follow up
3 3 10.5
5
18
8 3 3
26
58
10.5 47
CSA/AZO TAC/MMF
TAC/AZO AZO
pred/CSA/AZO MMF
CSA/AZO MMF
TAC/AZO SIR/AZO
TAC/MMF SIR/MMF
AZO
Older Recipients At 5 years
12
At last follow up 3 3 3
3 24
9
6
3 15
12 6 40
CSA/AZO pred/TAC/MMF TAC
43
6
TAC/AZO CSA/MMF
9
pred/CSA/AZO TAC/MMF
CSA/AZO Pred/TAC/MMF CSA/SIR TAC
3
TAC/AZO Cyclo/MMF TAC/SIR
pred/CSA/AZO TAC/MMF SIR/MMF
Figure. Immunosuppression therapy over the observation period of the study. The distribution of immunosuppressant combinations is displayed for the infant and older patients groups at 5 years after transplantation, the starting point of the study, and at last follow-up.
after transplant in a patient initially born with hypoplastic left heart syndrome. There were 2 deaths in the older group: both from transplant coronary arteriopathy, 6.5 years after transplantation in 1 case and 10.5 years after transplantation in the other. In addition, 4 patients in the older group underwent re-transplantation for transplant coronary arteriopathy within the study period. Rejection in both groups during the study period was rare. The infant group had 9 episodes of rejection in 4 patients with a cumulative 249 years of follow-up. 2 of these rejection episodes were associated with reduction of immunosuppression for PTLD. The older group had 5 episodes of rejection in 4 patients with 1 of the episodes associated with immunosuppression reduction for the treatment of PTLD. The Figure demonstrates the various combinations of immunosuppressive drugs used at entry of the study, at 5 years after transplant, and at last follow-up. 5 years after transplant, 63% of
the infant group was taking cyclosporine, 92% was taking azathioprine, and only 5% (2 patients) was taking triple therapy with prednisone. At that time, 67% of the older group was taking tacrolimus, 70% was taking azathioprine, and only 18% (6 patients) was taking triple therapy with prednisone. At last followup, most infant recipients (57.5%) were taking tacrolimus, 76% were taking azathioprine, 18.5% were taking sirolimus, and no patient was taking triple therapy with steroids. In the older group at last follow-up, 73% were taking tacrolimus, 55% were taking azathioprine, 12% were taking sirolimus, and 12% remained receiving triple therapy with prednisone.
Severe Infections Within the study period, there were 49 serious infections in the infant group and 9 serious infections in the older
Infectious, Malignant, and Autoimmune Complications in Pediatric Heart Transplant Recipients
673
Table I. Types and number of severe infections in the infant and older heart transplant recipients groups during the observation period of the study Infant recipients Older recipients Pneumonia Bacteremia Viral respiratory infection Gastroenteritis Sinusitis Parvovirus Osteomyelitis Others Total
23 2 5 3 5 3 2 6 49
1 1 0 3 0 1 0 3 9
group. Serious infection was not associated with mortality in the infant or older groups. The occurrence rate of serious infections after 5-year survival in the infant group was 2.04 ⫾ 0.5 per patient per 10 years of follow-up, compared with 0.37 ⫾ 0.19 per patient per 10 years of follow-up after 5-year survival in the older group. The difference was statistically significant (P ⬍ .001), with the odds ratio 3.3 and 95% CI of 1.6 to 6.7. In the infant group, 18 of 38 patients (47%) had at least 1 serious infection, and 8 of 38 patients (21%) had ⱖ3 serious infections, compared with a prevalence of only 4 of 33 patients (12%) in the older recipient group with at least 1 serious infection and 1 of 33 patients (3%) with ⱖ3 serious infections. Table I summarizes the types and numbers of serious infections encountered in the infant and older groups. The overall types of infection were similar in the 2 groups, with the exception of the occurrence of mononucleosis, osteomyelitis, pyelonephritis, and bronchiectasis in the infant group as opposed to the older group. Pertussis was observed in 1 older recipient but no infants. Specific organisms associated with presumed bacterial infections were usually not recovered, as was observed in nearly half the cases of presumed bacterial pneumonia and 2 cases of osteomyelitis. The success rate of pathogen recovery in severe presumed bacterial pneumonia was 54% (12/22 cases), and most of these patients underwent bronchoalveolar lavage. Pneumococcus was the most frequently documented bacterial pathogen, responsible for 5 cases of pneumonia and 2 cases of bacteremia. In the infant group, bacterial pneumonia accounted for 44.9% of all severe infections, with an occurrence rate of 0.91 per 10 years, and pneumococcal infections (5 pneumonia and 2 bacteriemia) represented 14.3% of severe infections with the occurrence rate of 0.29 per 10 years follow-up. Besides Pneumococcus, the other recovered pathogens included Moraxella catarrhalis, Mycoplasma pneumoniae, Pseudomonas aeruginosa in a patient with a tracheostomy, Staphylococcus aureus, and -hemolytic streptococcus. Opportunistic/unusual infections associated with immunosuppression were uncommon, with only 1 case of Pneumocystis carinii and 4 cases (3 infant, 1 older recipient) of Parvovirus B19 hypoplastic anemia. In patients with recurrent serious infections, evaluation of immunoglobulins revealed 2 instances of hypogammaglobuline674
Kulikowska et al
Table II. Types of chronic/recurrent infections and numbers of years with chronic/recurrent infections in the infant and older heart transplant recipients groups during the observation period of the study
Sinusitis Otitis media Otitis externa Tonsillitis Pneumonia Urinary tract infection Verruca/molluscum Others Total
Infant recipients
Older recipients
34 14 2 3 18 1 23 7 102
4 2 1 1 0 4 6 1 19
mia and 1 case of Ig A deficiency in 3 infant recipients. In 3 cases of recurrent pneumococcal infection in the infant group, immunization with the pneumococcal vaccine was associated with an absence of appropriate antibody response.
Chronic Infections Patients in the infant heart transplant group also experienced a higher occurrence rate of chronic/recurrent infections (mean, 4.58 ⫾ 0.67 years with chronic/recurrent infections per 10 years of follow-up after 5-year survival) than the older recipients (mean, 1.84 ⫾ 0.70 years with chronic/recurrent infections per 10 years of follow-up after 5-year survival). The difference was statistically significant (P ⬍ .001), with an odds ratio 4.5 and 95% CI of 2.7 to 7.6. Table II demonstrates the number and types of chronic/recurrent infections. The most common types of chronic/recurrent infections in the infant group were sinusitis, skin warts (verruca vulgaris and molluscum contagiosum), pneumonia, and otitis media, whereas skin warts, sinusitis, and urinary tract infections were the most common types of chronic/recurrent infections in the older group. Although these chronic infections were similar to those observed in the general population, the infant recipients had very severe forms of these infections. As stated earlier, bronchiectasis developed in 3 infant recipients as a consequence of recurrent pneumonias, and 16 patients required surgical otolaryngological procedures for recurrent infections that were unresponsive to antibiotics. 8 infants underwent myringotomies after 5 years post-transplantation, 2 infants underwent sinus operations, and 5 infants and 1 older recipient underwent tonsillo-adenoidectomies. In addition, 5 infant recipients had diagnostic bronchoscopies for recurrent pneumonias. PTLD Higher, but not statistically significant (P ⫽ .12), occurrence rates of PTLD were observed in the infant group (m ⫽ 0.40 ⫾ 0.27 cases per patient per 10 years of follow-up after 5 years) versus the older group (m ⫽ 0.10 ⫾ 0.10 cases per patient per 10 years of follow-up after 5 years). There The Journal of Pediatrics • May 2008
Table III. Individual cases of autoimmune diseases in the infant and older heart transplant recipients groups during the observation period of the study
Autoimmune diseases in infant recipients Membranoprolipherative glomerulonephritis Hypothyroidism Idiopathic thrombocytopenic purpura Idiopathic thrombocytopenic purpura Autoimmune hemolytic anemia Chronic bullous disease Autoimmune hemolytic anemia Autoimmune hemolytic anemia Autoimmune hepatitis Autoimmune diseases in older recipients Crohn’s disease
Onset after transplant
Immunosuppression
14 years (13 years after heart retransplantation) 9 years 9 years (1 year after renal transplant)
Tacrolimus/Imuran Cyclosporine/Imuran Tacrolimus/Imuran
8 years
Tacrolimus/Imuran
8 years 7 years 6 years
Tacrolimus/Imuran Tacrolimus/Imuran Tacrolimus/Imuran
9 years
Cyclosporine/MMF
were 5 cases of PTLD in the infant group and 1 case in the older group within the study period. Most cases were B cell lymphomas: 2 polymorphic B cell lymphomas, 1 polymorphous hyperplasia, 1 monomorphic B cell lymphoma, and 1 large cell Burkitt’s lymphoma. They were treated successfully with chemotherapy or immunosupression reduction and tonsillo-adenoidectomy, when tonsils and adenoids were involved. The 2 mortalities included 1 infant recipient with a rare type of hepatosplenic gamma delta T cell lymphoma with hemophagocytic syndrome who died of a relapse of PTLD despite intensive chemotherapy and 1 older recipient in whom severe fatal graft coronary vasculopathy developed 1 year after remission of the lymphoma. The patients in whom PTLD developed received a standard immunosupressive therapy with a combination of calcineurin inhibitor (3 patients cyclosporine and 3 patients tacrolimus) and azothioprone. In all 6 cases the onset of PTLD did not correlate with the onset of an acute Epstein-Barr virus (EBV) infection demonstrable with serology or, when available, copies of the viral DNA demonstrable with polymerase chain reactivity of blood samples. Patients with PTLD did not have an increased occurrence rate of serious or chronic/recurrent infections compared with the rest of the group.
Autoimmune Disease We identified an occurrence rate of autoimmune disease in infant recipients that was similar to the occurrence rate of PTLD (mean, 0.53 ⫾ 0.24 cases per patient per 10 years of follow-up after 5 years). There was 1 documented case of autoimmune disease in an older recipient, leading to a mean occurrence rate of 0.04 ⫾ 0.04 cases per patient per 10 years of follow-up after 5 years. This difference between the occurrence rates of the groups was not statistically different (P ⫽ .26). Table III illustrates the individual cases of autoimmune disease. The most common were autoimmune cytopenias, with single cases of autoimmune thyroid, kidney, bowel, and cutaneous bullous disease. All 4 patients in whom autoimmune cytopenias developed received tacrolimus. 1 case of autoimmune hemolytic anemia and 1 case of idiopathic trom-
bocytopenic purpura were associated with an acute infection with Parvovirus B19. None of the autoimmune disorders were fatal, and all responded to treatment with steroids, intravenous gamma globulin, or both. Similar to patients with PTLD, patients with autoimmune disease did not experience increased occurrence rate of serious or chronic/recurrent infections compared with the rest of the group.
DISCUSSION We observed significantly higher occurrence rates of severe and chronic/recurrent but non-fatal infections in infants recipients than in older recipients ⬎5 years after transplantation. Earlier long-term outcome studies5-8 in pediatric heart transplant recipients have reported that infection declines, but is not eliminated, as a cause of morbidity and mortality late after transplantation. Most studies on infectious complications after pediatric heart transplantation have focused on problems in the early post-transplant period, when infection is an important cause of mortality.16-18 Age-specific differences in rates of infections have been reported in liver and kidney recipients,19 and younger recipients (⬍2 years after transplant) had higher occurrence rates of bacterial and viral infection compared with older recipients. In a multicenter study of pediatric heart transplant recipients focusing on infection in the first year after transplant,17 infant recipients had a higher incidence of bacterial infections compared with older recipients. It has been well established in the general population20 that young age is a significant risk factor for increased rates of infections such as influenza or invasive pneumococcal disease, so the age difference in the studied groups could itself account for higher infectious morbidities in infant recipients, irrelevant to the immunosupression status.20 However, the occurrence rates of severe bacterial pneumonia (0.91 per 10 years or 9,100/100,000) and invasive pneumococcal disease (0.29 per 10 years or 2,900/100,000) in our infant recipients group (age 5-20 years during study period) exceeds the reported occurrence rates in general population and high-risk groups. Grijalva et al report an estimated rate of all-cause pneumonia
Infectious, Malignant, and Autoimmune Complications in Pediatric Heart Transplant Recipients
675
hospitalizations in the general population as 74 to 91 per 100,000 for ages 5 to 17 years and 791 to 1,297 per 100,000 for children ⬍2 years, depending on vaccination era, and pneumococcal pneumonia rates as 1.9 to 3.5 per 100,000 and 9.2 to 26.2 per 100,000, respectively.21 Although the greatest risk for the development of PTLD occurs within the first 6 months after pediatric heart transplantation,22 there is an ongoing constant hazard for its development years after transplantation. The exceptionally long follow-up period (58% survived ⬎10 years) in our infant recipients group could have contributed to the accumulation of PTLD cases during the time and as a result and to higher incidence rates in infants. EBV seronegative status before transplantation is known to be an independent risk factor for PTLD, which might also significantly contribute to higher occurrence rates of PTLD in infant group because most of the infant recipients had seronegative status at the time of transplantation (infants who had positive serologies had a passive maternal antibodies with documentation of seronegative status after several months) and seroconverted later after transplantation. The timing of development of PTLD, however, did not correlate with the time of EBV seroconversion, which supports recent data23 that EBV viremia as measured with polymerase chain reactivity is a relatively insensitive predictor for the development of PTLD, especially late after transplantation. Our experience is also consistent with earlier reports24 suggesting late-onset (⬎3 years) PTLD is more likely to be monomorphic and require more aggressive treatment than early onset PTLD, which tends to be polymorphic, EBV and CD 20 positive, and often responsive to immunosuppression reduction alone. The presence of 9 cases of autoimmune disease in 7 primarily infant recipients was an unanticipated morbidity in our infant recipients. There have been isolated case reports of autoimmune disease developing in pediatric solid organ recipients. Crohn’s disease25 and multiple autoantibody syndrome26 have been reported in heart transplant recipients, and immunobullous disease27 has been reported in liver and kidney recipients. Although all of our 4 patients with autoimmune cytopenias were receiving tacrolimus, the number of patients in our study was too small to perform any statistical analysis to determine whether the autoimmune cytopenias were specifically tacrolimus related. Autoimmune disease is a well-known complication of a number of primary immunodeficiency diseases observed in children.28 In these diseases, it is hypothesized that the basis of autoimmunity is the inability of a patient with these diseases to completely and effectively eradicate pathogens and their associated antigens. A compensatory chronic inflammatory response via less-effective alternative pathways, which may be exaggerated, leads to damage to healthy tissue and infected tissue. Autoimmunity in this setting is not felt to be a breakdown of tolerance to selfantigens, but rather part of the attempt of an abnormal immune system to rid itself of foreign antigens. This potential mechanism is supported in our population by the concurrent presence of autoimmune hemolytic anemia or idiopathic 676
Kulikowska et al
thrombocytopenic purpura in the presence of Parvovirus B19 infection, a clinical phenomenon that has been described.29 The long-term survivors of infant heart transplantation reported in this study demonstrate a persistent tendency for severe infections, recurrent/chronic infections, or both, especially pneumococcal disease, and the potential to develop autoimmune disease, most commonly hematologic cytopenias, later in childhood. The presence of humoral immunity dysfunction in the setting of T cell dysfunction in this patient group is similar to the findings in older children with 22q11.2 deletion (DiGeorge syndrome),30,31 whereas older patients have a similar clinical picture felt to reflect an aberrant T cell/B cell interaction. In addition to the differences in occurrence rates of infection observed between infant and older heart transplant recipients, infant heart transplant recipients also are different in that they undergo thymectomy at the time of surgery in addition to receiving immunosuppressant medications. Although removal of the thymus in infancy generally causes modest defects in the T cell compartment structure with little or no impact on long-term morbidity, a recent study in infant heart transplant recipients9 found that these patients demonstrate absence of thymic function, a profound decrease in the diversity of the T cell repertoire, and are less able to mount an antibody response to immunization. These authors speculated that this type of immune dysfunction might lead to a greater risk of infection and manifestations of autoimmunity with age. Further evidence of the potential differences in immune function between infant and older heart transplant recipients was demonstrated in a study from the United Kingdom in which only 25% of the younger recipients (⬍4 years at time of transplant) responded to pneumococcal immunization compared with 82% of the older group.10 This observational study is most limited by the small sample size in each group. The retrospective nature of the study is associated with a lack of control of immunosuppression regimens, which were manipulated according to the individual clinical picture of each subject. Serious infections, chronic/recurrent infections, or both did not appear to be clustered in patients in either group treated for PTLD or autoimmune disease. It appeared that the infant group, with the exception of temporary intensification of steroid therapy or other immunosuppressants for treatment of PTLD and autoimmune disease, had immunosuppression comparable with the older group and that the magnitude of immunosuppression was managed with similar target drug levels for each group. Therapeutic drug level monitoring, however, does not give a picture of overall degree of immunosuppression in a given patient. A recent test developed to assess global assessment of cellular immune function has demonstrated that pediatric transplant recipients ⬍12 years old appear to be more immunosuppressed on comparable immunosuppression than adults or patients ⬎12 years old.32 Although this study did not assess the effects of age at transplantation on global immune status, it is conceivable that infant pediatric heart transplant recipients may be more immunosuppressed than The Journal of Pediatrics • May 2008
older recipients despite the same target drug levels with the same immunosuppressive drug protocols. Our findings suggest that in addition to neoplasm, infection, and immunosuppressant drug toxicity, autoimmune disease should be considered as an etiology for non-cardiac problems in pediatric heart transplant recipients, especially hematologic cytopenias.
REFERENCES 1. Boucek MM, Waltz DA, Edwards LB, Taylor DO, Keck BM, Trulock EP, et al. Registry of the International Society for Heart and Lung Transplantation: ninth official pediatric heart transplantation report—2006. J Heart Lung Transplant 2006;25: 893-903. 2. Canter CE, Naftel D, Caldwell R, Chinnock R, Pahl E, Frazier E, et al. Survival and risk factors for death after cardiac transplantation in infants: a multi-institutional study. Circulation 1997;96:227-31. 3. Ibrahim JE, Sweet SC, Flippin M, Dent C, Mendeloff E, Huddleston CB, Canter CE. Rejection is reduced in thoracic organ recipients when transplanted in the first year of life. J Heart Lung Transplant 2002;21:311-8. 4. Pahl E, Naftel DC, Kuhn MA, Shaddy RE, Morrow WR, Canter CE, 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. 5. Sigfusson G, Fricker JF, Bernstein D, Addonizio LJ, Baum D, Hsu DT, et al. Long-term survivors of pediatric heart transplantation: a multicenter report of sixtyeight children who have survived longer than five years. J Pediatr 1997;130:862-71. 6. Gajarski RJ, Smith EO, Denfield SW, Rosenblatt HM, Kearney D, Frazier OH, et al. Long-term results of triple-drug-based immunosuppression in nonneonatal pediatric heart transplant recipients. Transplantation 1998;65:1470-6. 7. Radley Smith R, Wray J, Khaghani A, Yacoub M. Ten year survival after paediatric heart transplantation: a single center experience. Eur J Cardiothorac Surg 2005;27:790-4. 8. Ross M, Kouretas P, Gamberg P, Miller J, Burge M, Reitz B, et al. Ten- and 20-year survivors of pediatric orthotopic heart transplantation. J Heart Lung Transplant 2006;25:261-70. 9. Ogle BM, West LJ, Driscoll DJ, Strome SE, Razonable RR, Paya CV et al. Effacing of the T cell compartment by cardiac transplantation in infancy. J Immunol 2006;176:1962-7. 10. Gennery AR, Cant AJ, Spickett GP, Walshaw D, Hunter S, Hasan A, et al. Effect of immunosuppression after cardiac transplantation in early childhood on antibody response to polysaccharide antigen. Lancet 1998;351:1778-81. 11. Canter CE, Moorhead S, Saffitz JE, Huddleston CB, Spray TL. Steroid withdrawal in the pediatric heart transplant recipient initially treated with triple immunosuppression. J Heart Lung Transplant 1994;13:74-9. 12. Yamani MH, Starling RC, Goormastic M, Van Lente F, Smedira N, McCarthy P, et al. The impact of routine mycophenolate mofetil drug monitoring on the treatment of cardiac allograft rejection. Transplantation 2000;11:2326-30. 13. Taylor DO, Ensley RD, Olsen SL, Dunn D, Renlund DG. Mycophenolate mofetil (RS-61443): preclinical, clinical, and three-year experience in heart transplantation. J Heart Lung Transplant 1994;13:571-82.
14. Snell GL, Levvey BJ, Chin W, Kotsimbos T, Whitford H, Waters KN, et al. Sirolimus allows renal recovery in lung and heart transplant recipients with chronic renal impairment. J Heart Lung Transplant 2002;21:540-6. 15. Mancini D, Pinney S, Burkhoff D, LaManca J, Itescu S, Burke E, et al. Use of rapamycin slows progression of cardiac transplantation vasculopathy. Circulation 2003; 108:48-53. 16. Keough WL, Michaels MG. Infectious complications in pediatric solid organ transplantation. Pediatr Clin N Am 2003;50:1451-69. 17. Schowengerdt K, Naftel D, Seib P, Pearce B, Addonizio L, Kirklin J, et al. Infection after pediatric heart transplantation: results of a multi-institutional study. J Heart Lung Transplant 1997;16:1207-16. 18. Doelling NR, Kanter KR, Sullivan KM, Winn KJ, Vincent R. Medium-term results of pediatric patients undergoing orthotopic heart transplantation. J Heart Lung Transplant 1997;16:1225-30. 19. Tran L, Hebert D, Dipchand A, Fecteau A, Richard S, Allen U. Invasive pneumococcal disease in pediatric organ transplant recipients: a high risk population. Pediatr Transplantation 2005;9:183-6. 20. Behrman RE, Kliegman RM, Jenson HB. Nelson textbook of pediatrics, 17th ed. Abramson JS, Overtruf GD: Chapter 167: Streptococcus pneumoniae (Pneumococcus) p. 867-9; Wright P: Chapter 237: Influenza viruses, p. 1072-5. 21. Grijalva CA, Nuorti JP, Arbogast PG, Martin SW, Edwards KM, Griffin MR. Decline in pneumonia admissions after routine childhood immunisation with pneumococcal conjugate vaccine in the USA: a time-series analysis. Lancet 2007;369:1179-86. 22. Webber SA, Naftel DC, Fricker FJ, Olesnevich P, Blume ED, Addonizio L, et al. Lymphoproliferative disorders after paediatric heart transplantation: a multi-institutional study. Lancet 2006;367:233-9. 23. Axelrod DA, Holmes R, Thomas SE, Magee JC. Limitations of EBV-PCR monitoring to detect EBV associated post-transplant lymphoproliferative disorder. Pediatr Transplantation 2003;7:223-27. 24. Mayo P, Ghobrial IM, Habermann TM, Macon WR, Ristow KM, Larson TS, et al. Differences between early and late posttransplant lymphoproliferative disorders in solid organ transplant patients: are they two different diseases? Transplantation 2005; 79:244-7. 25. Harms B, Bremner AR, Mulligan J, Fairhurst J, Griffiths DM, Salmon T, et al. Crohn’s disease post-cardiac transplantation presenting with severe growth failure and delayed onset of puberty. Pediatr Allergy Immunol 2004;15:186-9. 26. Rawal A, Sarode R, Curtis B, Karandikar N, Friedman K, Rogers Z. Acquired Glanzmann’s thrombasthenia as part of multiple-autoantibody syndrome in a pediatric heart transplant patient. J Pediatr 2004;144:672-4. 27. Morelli JG, Weston WL. Childhood immunobullous disease following a second organ transplant. Pediatr Dermatol 1999;16:205-7. 28. Arkwright PD, Abinun M, Cant AJ. Autoimmunity in human primary immunodeficiency diseases. Blood 2002;99:2694-702. 29. De la Rubia J, Moscardo F, Arriaga F, Monteagudo E, Carreras C, Marty ML. Acute parvovirus B19 infection as a cause of autoimmune hemolytic anemia. Heamatologica 2000;85:995-7. 30. Jawad AF, McDonald-McGinn DM, Zackai E, Sullivan KE. Immunologic features of chromosome 22q11.2 deletion syndrome (DiGeorge syndrome/velocardiofacial syndrome). J Pediatr 2001;139:715-23. 31. Gennery AR, Barge D, O’Sullivan JJ, Flood TJ, Abinun M, Cant AJ. Antibody deficiency and autoimmunity in 22q11.2 deletion syndrome. Arch Dis Child 2002; 86:422-5. 32. Hooper E, Hawkins DM, Kowalski RJ, Post DR, Britz JA, Brooks KC, et al. Establishing the pediatric immune response zones using the Cylex Immuknow assay. Clin Transplantation 2005;19:834-9.
Infectious, Malignant, and Autoimmune Complications in Pediatric Heart Transplant Recipients
677