Posttransplant lymphoproliferative disease (PTLD) is defined as a diffuse, uncontrolled proliferation of lymphoid cells in immunosuppressed transplant recipients (1). Histologically, the disease comprises a spectrum ranging from reactive-looking proliferation to frank non-Hodgkin's lymphoma morphology (2,3). The reported incidence of PTLD varies from 2% to 28%, depending on the intensity of the immunosuppressive regimen used (1). Epstein-Barr virus (EBV)-infected B lymphocytes that induce either a monoclonal or polyclonal proliferation, are involved in 85% of PTLD cases (4,5). The remainder originate from T (14%) or null (1%) cells (4). The main method used for establishing the diagnosis of PTLD is immunohistochemical tissue analysis (6). Posttransplant lymphoproliferative disease generally assumes one of three gross pathologic appearances: solid tumor, diffuse infiltration of parenchymal organs, or enlargement of native lymphoid tissue (5).
The gastrointestinal tract involvement of posttransplant lymphoproliferative disease (GI-PTLD) has been reported (7-9). Unexplained fevers, decreased appetite, weight loss, diarrhea, protein-losing enteropathy and gastrointestinal bleeding have been associated with GI-PTLD (8,9). However, these signs and symptoms are not specific and occur frequently in immunosuppressed liver transplant recipients, who are prone to infections and are usually treated with multiple medications in the posttransplant period.
In this retrospective study, we compared the clinical features, histological findings, management, and outcome of pediatric patients who underwent liver transplant and in whom GI-PTLD developed with 29 liver transplant recipients in whom PTLD did not develop. By reviewing our experience we attempted to identify important clinical signs that should alert clinicians to patients at risk for development of GI-PTLD.
MATERIALS AND METHODS
We retrospectively reviewed the clinical course of 392 children (less than 18 years of age) who underwent liver transplantation in the period between from March 1984 through June 1997 and who survived at least 6 months after transplant at the University of California Los Angeles Children's Hospital. Sixty patients died of causes unrelated to PTLD during that period and were excluded from the study. The indications for OLT were biliary atresia (n = 205), infectious hepatitis and fulminant liver failure (n = 58), metabolic liver disease (n = 39), cryptogenic cirrhosis (n = 38), tumors (n = 11), and other indications (n = 41).
During this period we used two immunosuppressive protocols for the primary prevention of rejection. Between 1984 and 1994 primary immunosuppression treatment was cyclosporin A (CsA; Sandimmune; Sandoz; East Hanover, NJ, U.S.A.) combined with low-dose prednisone. Doses were adjusted for serum levels between 150 and 350 ng/ml. When the method for determining CsA levels changed to a whole-blood assay (most recently a monoclonal radioimmunoassay technique, CYCLO-Trac, Incstar, Stillwater, MN, U.S.A.) the therapeutic range was 150 to 250 ng/ml. Since 1989, azathioprine was routinely added at 1 to 2 mg/kg per day. In 1994 our program changed to primary therapy with tacrolimus (Prograf; Fujisawa, Deerfield, IL, U.S.A.). Initially, doses were adjusted to keep plasma levels at 0.5 to 1.5 ng/ml, using an enzyme-linked immunosorbent assay. Later target whole-blood levels were between 10 and 15 ng/ml, (most recently, 12-15 ng/ml during the first month, 10-12 ng/ml during months 1-3, and 5-10 ng/ml thereafter), measured by a whole-blood microparticle enzyme immunoassay (Imx; Abbott Labs, Abbott Park, IL, U.S.A.) (8).
Acute allograft rejection was treated with high-dose intravenous methylprednisolone (Solu-Medrol, Upjohn, Kalamazoo, MI, U.S.A.) at 20 mg/kg per day, tapered to 0.3 mg/kg per day over 7 days. Steroid-resistant rejection was treated with monoclonal antibody OKT3 (Orthoclone; Ortho Biotech, Raritan, NJ, U.S.A.) intravenously at a dose of 2.5 mg per day for 10 to 14 days in patients weighing less than 20 kg and 5 mg per day in those weighing 20 kg or more. CD3 lymphocyte counts were measured daily, and if less than 5%, the OKT3 dose was increased by 2.5-mg increments. After 1991, CsA-treated patients with steroid-resistant rejection or chronic rejection were treated by conversion to tacrolimus.
Diagnosis of Gastrointestinal PTLD
Patients in whom three or more of the following characteristics developed after or appeared before initial examination underwent upper and lower gastrointestinal tract endoscopy as part of PTLD evaluation: chronic diarrhea of more than 14 days, with or without heme-positive stools; hypoalbuminemia (serum albumin less than 3.0 g/dl), protein-losing enteropathy (measured by stool α1-antitrypsin level in a spot stool specimen), fever, anorexia, and weight loss. Multiple biopsy specimens from normal appearing, erythematous, or ulcerated lesions were submitted for immunohistochemical lymphoma evaluation. This evaluation included a 3-step peroxidase technique with antibodies to all T- and B-cell types and to κ and λ chains. In situ hybridization with DNA probes for EBV was performed in most cases. Abdominal computed tomographic scan was performed on all patients in which EBV was suspected to detect enlarged (>1 cm) intra-abdominal lymph nodes. If significantly lymphadenopathy was identified, tissue for PTLD confirmation was retrieved by surgical intervention whenever possible.
The occurrence of fever, chronic diarrhea, anemia, hypoalbuminemia, gastrointestinal bleeding, anorexia, abdominal pain, and weight loss in GI-PTLD patients was compared with occurrence of those symptoms a group of 29 randomly selected liver transplant recipients without PTLD.
Diagnosis of Epstein-Barr Infection
Primary EBV infection was diagnosed based on the presence of anti-viral capsid antigen (VCA) IgM titers. Reactivation of EBV was diagnosed by a more than fourfold increase in anti-VCA IgG titers with the presence of elevated EBV nuclear antigen (EBNA) and early antigen (EA) IgG titers or by the reappearance of IgM VCA antibody. Since 1995, EBV polymerase chain reaction (PCR) has been used to document early EBV infection (11). A detailed description of EBV PCR techniques performed in our patients will be reported by McDiarmid et al. (12) Epstein-Barr virus was defined as symptomatic infection (IgM+, PCR+).
After the diagnosis of PTLD was established, with the exception of low-dose prednisone, all immunosuppressive medications were discontinued. Intravenous antiviral treatment, initially with acyclovir and subsequently with ganciclovir, was initiated. Patients were monitored closely for signs of rejection. Graft function was monitored by serial liver function tests, and all rejection episodes were determined by analysis of biopsy specimens. Immunosuppression was not reinstituted unless the PTLD was judged to be resolved. The time to reinstitute immunosuppressive therapy varied from a few months to 2 years in some patients, but fortunately, no graft had to be sacrificed to save a patient's life.
McNemar, chi-square tests, means, and standard deviation of the mean were calculated using conventional formulas.
In the period between March 1984 and June 1997, 392 children underwent liver transplantation in the Dumont-University of California, Los Angeles liver transplantation program and survived longer than 6 months. Forty-three percent of the patients were less than 2 years of age, 32% were between 2 and 5 years of age, and 25% were more than 5 years of age. There were 245 girls. In 30 patients, PTLD developed with an incidence density of 1.8 per 100 patient-years. Sixteen patients (53%) with PTLD were less than 2 years of age, eight (27%) were between the ages of 2 and 5 years, and six (20%) were more than 5 years old.
There were 251 patients treated primarily with CsA, including those 29 patients treated with tacrolimus for intractable and/or chronic rejection. One hundred forty-one patients received primarily tacrolimus immunosuppression. Seventy-five patients received OKT3 for steroid-resistant rejection, 73 in the CsA group and 2 in the tacrolimus group. Initially, 30 randomly chosen liver transplant patients were evaluated for clinical signs and served as the control group. One patient in whom PTLD developed later was excluded. Characteristics of patients with PTLD or GI-PTLD and those without PTLD are shown in Table 1. There was no significant difference in the age of patients with and without PTLD. The only parameter that differentiated between the groups was the higher incidence of primary EBV infection in the PTLD and GI-PTLD groups.
Of the nine patients with GI-PTLD, five received tacrolimus, one CsA, and three tacrolimus for chronic or intractable rejection. The majority (n = 5) of the GI-PTLD patients were younger than 2 years. Two were between 2 and 5 years and two were more than 5 years of age. All patients had evidence of EBV infection at the time of the diagnosis. All except one had polyclonal B-cell proliferation, and the disease went into remission after withdrawal of immunosuppressive medications and initiation of antiviral treatment. The overall case fatality rate was 20%; that of those with GI-PTLD was 11% (p > 0.05). A 20-month-old girl died who had obliterative bronchiolitis and respiratory failure. She received no chemotherapy and had no evidence of PTLD at autopsy. Only one patient had PTLD recur in the gastrointestinal tract. Her condition improved after immunosuppression was decreased.
Epstein-Barr Virus Serology
Pretransplant EBV serology was determined in 314 patients (80%). Thirty-eight percent had primary EBV infection after liver transplantation. The EBV titers before transplantation were determined in 25 of 30 patients with PTLD (83%). Eighteen (72%) had primary EBV infection after transplantation. Primary EBV infection after transplantation was identified in 5 of 7 (71%) who had GI-PTLD exclusively.
PTLD Symptoms and Characteristics
Of the 30 patients with PTLD, 9 (30%) had evidence of gastrointestinal tract involvement, 7 of whom had manifestation only in the gastrointestinal tract. Other sites in which biopsies were obtained to establish PTLD diagnosis were lymph nodes in 23 patients, liver in 4, upper airways in 5, and lungs in 2.
To determine the clinical signs and symptoms that are associated with GI-PTLD, we compared this group of patients with the group of patients with PTLD. Fevers, diarrhea, anemia, and weight loss occurred with almost the same frequency in both groups. However, GI blood loss and hypoalbuminemia occurred more frequently in the former group, whereas enlarged lymph nodes were found more often in the latter group. The clinical comparison of patients with PTLD and those with GI-PTLD is shown in Table 2.
The following clinical symptoms were seen in the patients with GI-PTLD: anemia (100%), gastrointestinal bleeding (100%), weight loss (89%), fevers (89%), hypoalbuminemia (89%), protein-losing enteropathy (89%), anorexia (78%), chronic diarrhea (67%), and abdominal pain (44%). One child with GI-PTLD had a persistent tongue ulceration and lymphadenopathy. Another child had an acute abdomen and was found to have an ileal perforation. When we compared symptoms in patients with GI-PTLD with those in a cohort of transplant patients without PTLD, only hypoalbuminemia, gastrointestinal bleeding, and weight loss were statistically associated with GI-PTLD (Table 3). Although patients with GI-PTLD had recurring fevers, diarrhea, and anemia, these signs were also present in the control subjects.
Upper and lower gastrointestinal tract endoscopy was performed in all GI-PTLD patients. One patient had a small-bowel perforation. Although he underwent endoscopy at a later time, his histologic diagnosis was made at the time of surgery. Endoscopy-retrieved tissue for PTLD confirmation was obtained in 7 of 8 patients, whereas abdominal computed tomographic scan and subsequent lymph node tissue analysis was valuable in establishing the diagnosis in 2 of 7 patients.
The endoscopic findings ranged from normal-appearing mucosa, and erythematous nodular lesions to frank ulcerations in the upper and lower gastrointestinal tract. Lower gastrointestinal tract involvement was more common. Five patients (56%) had PTLD in the lower gastrointestinal tract only, one (11%) in the upper tract only, and three (33%) in both.
The results of this study show that gastrointestinal involvement in PTLD is common. On occasion, it may be the only site involved in the disease. More important, children who undergo solid-organ transplants and who have hypoalbuminemia secondary to protein-losing enteropathy, gastrointestinal bleeding, and weight loss should undergo further evaluation to rule out GI-PTLD.
Posttransplant lymphoproliferative disease is usually a B-cell proliferation that is EBV-driven and may appear as a spectrum of disorders, ranging from a simple mononucleosis-like syndrome to frank lymphoma (13,14). Lymphomas are the most common cause of posttransplant malignant disease in pediatric transplant recipients and occur more often in nonrenal recipients (15). Currently, the known risk factors for development of PTLD include young age, EBV-negative status at transplantation, and primary EBV infection. Given these risk factors, pediatric solid-organ transplant recipients are at particularly high risk for development of PTLD (16,17).
The gastrointestinal tract is not an unexpected site of PTLD, because it represents the largest lymphoid organ. Starzl (18) described increased frequency of lymphoid tumors in immunosuppressed organ transplant recipients in the gastrointestinal tract in the late 1960s. Nalesnik (7) described the clinical findings in the involvement of the gastrointestinal tract by EBV-associated PTLD. In that report of 12 patients who were mostly adults, 6 had kidney grafts, and 6 had liver grafts. The clinical features in those patients were similar to those of gastrointestinal lymphomas in patients who had had no immunosuppression. Four patients in that group had ileal perforation and acute abdomen. The lesions described were most often multiple and predominately involved the distal small bowel.
Usually, the signs and symptoms of GI-PTLD are protean, often making the diagnosis difficult to establish. Although the disease can cause acute perforation and a surgical abdomen, as described earlier, usually the manifestations are insidious. Fevers, lymphadenopathy, anorexia, weight loss, chronic diarrhea, and gastrointestinal blood loss are among the described signs and symptoms associated with PTLD (9).
In our center, serum albumin is routinely measured during the follow-up of all OLT patients. Hypoalbuminemia was present in 89% of those with GI-PTLD but was not identified in the non-PTLD control group. Hypoalbuminemia was presumably related to protein-losing enteropathy, which can be assessed by an elevation of α1-antitrypsin in the stool. In fact, all of our patients with GI-PTLD who had hypoalbuminemia had elevated α1-antitrypsin stool levels. Gross gastrointestinal bleeding was present in 89% of those with GI-PTLD, compared with occurrence in only 7% of the non-PTLD control group (p < 0.05). Although heme-positive stools could be the earliest sign of GI-PTLD, all our patients had melena (19). We found anemia in 100% of our GI-PTLD patients and in 86% in liver transplant patients without PTLD (p = 0.22). Anemia in these patients is multifactorial in origin. Weight loss was present in 89% of patients with GI-PTLD, but in 18% in the control group (p < 0.05). Anorexia and fevers were identified in both groups. This may be secondary to a complicated postoperative course and repeated infections and was not significantly different from anorexia and fevers found in the control group (p > 0.05). Chronic diarrhea and abdominal pain were also found in both groups.
There is no reported difference in the incidence of gastrointestinal symptoms in patients treated with CsA or tacrolimus (20,21). We reported earlier the occurrence of gastrointestinal side effects in patients treated with CsA or tacrolimus (20). Although there was more gastrointestinal hemorrhage reported in those patients treated with tacrolimus, there was no difference in the occurrence of diarrhea or anorexia in patients treated with tacrolimus and occurrence in those treated with CsA. Similarly, Van Thiel et al. (21) reported gastrointestinal symptoms related to immunosuppressive agents. During a follow-up period of 3 weeks, they found transient vomiting more commonly in patients treated with tacrolimus, whereas intermittent abdominal pain occurred more often in patients treated with CsA (21). Both those symptoms were reportedly minor inconveniences rather than problems necessitating change in therapy or other actions.
Upper and lower gastrointestinal tract endoscopy was a useful tool in helping to establish the diagnosis of GI-PTLD. When we compared the contribution of endoscopy with that of abdominal computed tomographic scan through subsequent tissue analysis in establishing the diagnosis of GI-PTLD, we found no statistically significant differences (p = 0.125). This may be because of the small sample size of the patients with GI-PTLD. We found more GI-PTLD involvement in the lower gastrointestinal tract. The presence of more lymphoid tissue in that part of the gastrointestinal tract may explain this finding. However, in the only patient who died in this group, diagnosis was made from an almost normal-appearing, slightly erythematous duodenal lesion.
In all other patients, PTLD regressed after the immunosuppressive agents were discontinued, as described previously (22-24). No graft loss caused by rejection was observed during this period, perhaps because of the immunosuppressive characteristics of EBV (25,26). Immunosuppressive agents were reinstituted at low doses when signs of rejection occurred. During the follow-up period, only one patient had GI-PTLD recurrence; she remains alive.
Gastrointestinal toxicity from tacrolimus was previously described as the chronic presence of at least two of the following: nausea, anorexia, diarrhea, and weight loss not attributable to primary gastrointestinal disease (27). In that report, 7 of the 122 liver transplant recipients treated with tacrolimus had severe gastrointestinal toxicity, and their conditions improved only after the immunosuppressive regimen was changed. None of those patients had histologic gastrointestinal tissue analysis reported.
The possibility that some of those patients may have had GI-PTLD remains unanswered. This, and the fact that almost one quarter of our GI-PTLD patients had the disease occur only in the gastrointestinal tract, makes us think that clinicians treating immunosuppressed patients should be more aware of the possibility of GI-PTLD when the signs and symptoms appear and should pursue evaluation for lymphoma by endoscopy, biopsy, and special immunohistologic techniques. This will enable the early diagnosis of PTLD, in the polymorph-polyclonal stage, which is readily reversible and more responsive to the decrease in immunosuppressive therapy.
In conclusion, this study shows that chronic, persistent gastrointestinal symptoms, particularly hypoalbuminemia and protein-losing enteropathy, gastrointestinal blood loss, and weight loss in children who have undergone liver transplant is an indication for further evaluation, including upper and lower gastrointestinal tract endoscopy, to rule out PTLD. This may establish the diagnosis of GI-PTLD sooner in its course and may decrease morbidity and mortality.
1. Newell KA, Alonso EM, Whitington PF, et al. Posttransplant lymphoproliferative disease
in pediatric liver transplantation. Interplay between primary Epstein-Barr virus infection and immunosuppression. Transplantation
2. Frizzera G, Hanto DW, Gajl-Peczalska KJ, et al. Polymorphic diffuse B-cell hyperplasia and lymphomas in renal transplant recipients. Cancer Res
3. Nalesnik MA, Jaffe R, Starzl TE, et al. The pathology of post-transplant lymphoproliferative disorders occurring in the setting of cyclosporine A - prednisone immunosuppression. Am J Pathol
4. Seiden MV, Sklar J. Molecular genetic analysis of post-transplant lymphoproliferative disorders. Hematol Oncol Clin North Am
5. Rustgi VK. Epstein-Barr viral infection and posttransplantation lymphoproliferative disorders. Liver transplantation and Surgery
6. Hart J, Lewin KJ. Liver allograft pathology. In: Ferrell LD, eds. Pathology: State of the art review
. Philadelphia: Hanley and Belfus, 1994:207.
7. Nalesnik MA. Involvement of the gastrointestinal
tract by Epstein-Barr virus-associated posttransplant lymphoproliferative disorders. Am J Surg Pathol
1990;14 Suppl 1:92-100.
8. Kahn E, Gastrointestinal
manifestations in pediatric AIDS. Pediatr Pathol Lab Med
9. Mc Diarmid SV. Liver transplantation in children. In: Cosimi AB, Ginns L, Morris PJ, eds. Transplantation
. Cambridge, MA: Blackwell Science, Inc. (in press).
10. Esquivel CO, So SK, McDiarmid SV, Andrews WS, Colombani PM. Suggested guidelines for the use of tacrolimus in pediatric liver transplant patients (letter). Transplantation
11. Martinez OM, Villanueva JC, Lawrence-Miyasaki LS, Quinn MB, Cox K, Krams SM. Viral and immunologic aspects of Epstein-Barr virus infection in pediatric liver transplant recipients. Transplantation
12. McDiarmid SV, Jordan S, Geoff L, et al. Prevention and preemptive therapy of post-transplant lymphoproliferative disease in pediatric liver recipients. Transplantation
13. Morrison VA, Dunn DL, Maniyel JC, Gajl-Peczalska KJ, Peterson BA. Clinical characteristics of post-transplant lymphoproliferative disorders. Am J Med
14. Penn I. Posttransplant malignancies in pediatric organ transplant recipients. Transplant Proc
15. Ho M, Jaffe R, Miller G, et al. The frequency of Epstein-Barr virus infection and associated lymphoproliferative syndrome after transplantation and its manifestation in children. Transplantation
16. Lamy M, Favart AM, Cornu C, et al. Epstein-Barr virus infection in 59 orthotopic liver transplant patients. Med Microbiol Immunol
17. Sokal EM, Caragiozoglou T, Lamy M, Reding R, Otte JE. Epstein-Barr virus serology and Epstein-Barr associated lymphoproliferative disorders in pediatric liver transplant patients. Transplantation
18. Starzl TE. Discussion of Murray JE, Wilson RE, Tilney NL, et al. Five years' experience in renal transplantation with immunosuppressive drugs: Survival, function, complications and the role of lymphocyte depletion by thoracic duct fistula. Ann Surg
19. Cao S, Cox K, Esquivel CO, et al. PTLD and gastrointestinal manifestations of EBV infection in children following liver transplantation
. Abstract presented at that meeting of American Society of Transplant Physicians, Dallas, Texas, 1996.
20. McDiarmid SV, Busutill RW, Ascher NL, et al. FK506 (tacrolimus) compared with cyclosporine for primary immunosuppression after pediatric liver transplantation. Transplantation
21. Van Thiel DH, Iqbal M, Jain A, et al. Gastrointestinal
and metabolic problems associated with immunosuppression with either CyA or FK506 in liver transplantation. Transplant Proc
22. Benkerrou M, Durandy A, Fisher A. Therapy for transplant-related lymphoproliferative diseases. Hematol Oncol Clin North Am
23. Frayha HH, Nazer H, Kalloghlian A, Antonius JI, Sabbah R. Lymphoproliferative disorder in a liver transplant patient on FK 506 (letter). Lancet
24. Cox KL, Lawrence LS, Garcia-Kennedy R, et al. An increased incidence of Epstein-Barr virus infection and lymphoproliferative disorder in young children on FK506 after liver transplantation. Transplantation
25. Forman SJ, Sullivan JL, Wright C, et al. Epstein-Barr virus related malignant B cell lymphoplasmacytic lymphoma following allogeneic bone marrow transplantation for aplastic anemia. Transplantation
26. David A. Thorley-Lawson. Basic virological aspects of Epstein-Barr virus infection. Semin Hematol
27. Fisher A, Schwartz M, Mor E, et al. Gastrointestinal
toxicity associated with FK 506 in liver transplant recipients. Transplant Proc