Leukodystrophy

We report 2 cases of adenovirus enterocolitis in pediatric patients who underwent bone marrow transplantation. The first case involved a 17-year-old adolescent boy with combined immunodeficiency and non-Hodgkin lymphoma who developed chronic graft versus host disease and persistent adenovirus duodenitis. Case 2 involved a 3-year-old boy who received a mismatched unrelated bone marrow transplant for metachromatic leukodystrophy; the boy developed severe graft versus host disease and died of multiorgan failure. At autopsy, diffuse hemorrhagic enterocolitis with changes of severe graft versus host disease and extensive mucosal invasion by adenovirus was found. Awareness and early recognition of this uncommon complication of concomitant graft versus host disease and adenovirus infection could impact therapy and outcome of patients with bone marrow transplant.

We describe 2 patients in the pediatric age group who developed adenovirus enterocolitis following bone marrow transplantation (BMT). Adenovirus infection is not uncommon in childhood, and although self-limited, it can cause significant mortality in immunocompromised patients. Despite a higher incidence of adenovirus infection following hematopoietic stem cell transplantation in pediatric versus adult patients, relatively few cases with isolated gastrointestinal tract involvement have been reported.1 In addition, our 2 cases highlight a rare complication of concurrent graft versus host disease (GVHD) and adenovirus enterocolitis.

REPORT OF CASES

Our first patient was a 17-year-old adolescent boy with a history of combined immunodeficiency and Epstein-Barr virus (EBV)-positive Burkitt lymphoma. He was status post matched-sibling donor BMT, with chronic GVHD (gastrointestinal and liver), under treatment with prednisone and cyclosporin. He had a history of multiple hospital admissions for complaints of epigastric pain, nausea, and watery diarrhea. In addition, fever, neutropenia, generalized skin rash, abdominal distension, and tenderness were frequent symptoms on his admissions. His history also included significant anemia, addisonian crisis, and positive stool culture for Clostridium difficile. He was under treatment with acyclovir and cytomegalovirus superimmune globulin (CytoGam) for EBV. His abdominal symptoms temporarily improved on steroids. However, during workup for persistent diarrhea, a stool specimen analysis by electron microscopy was positive for adenovirus, and antiviral treatment (ribavirin) was started. After 10 days of antiviral treatment, subsequent stool samples became negative for adenovirus, and no virus was isolated by cell culture from a rectal biopsy. However, on endoscopy, the duodenal mucosa appeared red, edematous, and granular, suggesting diffuse involvement. Also noted were white plaques suspicious for candida esophagitis and granular inflammatory mucosa of colon and stomach. The clinical impression was acute GVHD. Multiple biopsies were obtained. The patient underwent long-term treatment with fluconazole in addition to cyclosporine and steroids. He showed gradual improvement with antiviral therapy, which was eventually discontinued. On his last follow-up, 11 months post-transplantation, he had maintained good hematological parameters with the engraftment, and we found no evidence of recurrent adenovirus enteritis.

Case 2 involved a 3-year-old boy with an early-stage metachromatic leukodystrophy previously diagnosed in an affected sibling. After induction therapy, he underwent a mismatched unrelated donor BMT complicated by febrile neutropenia the day after the transplant. During the 11-week posttransplantation period until his death, the boy's clinical course was complicated by refractory GVHD (confirmed on colonie biopsies) with severe bloody diarrhea, pancreatitis, and total parenteral nutrition liver disease. The patient received immunosuppressive treatment with tacrolimus, steroids, cyclosporin A, antilymphocytic globulin, daclizumab, and budesonide for GVHD, as well as acyclovir, CytoGam, and ganciclovir for cytomegalovirus and EBV. The rectal biopsy was positive for EBV by polymerase chain reaction. The patient underwent daily plasmapheresis for thrombotic thrombocytopenic purpura; had difficult fluid management; and developed respiratory distress, progressive bone marrow hypocellularity, progressive renal failure, and anasarca with encephalopathy. He eventually died of multiorgan failure.

MATERIALS AND METHODS

Duodenal biopsy from patient 1 and samples of small and large intestine obtained at autopsy from patient 2 were fixed in neutral buffered formalin and embedded in paraffin. The sections cut at 5 µm were used for routine histology. For immunostaining, the sections were mounted on sialinated slides, dewaxed, and hydrated. Slides were barcoded for a Ventana Gen II or Nexes autostainer (Ventana Medical Systems, Inc, Tucson, Ariz). Using the autostainer program, the sections were pretreated with protease I and subjected to avidin-biotin blocking and then monoclonal antibody against adenovirus (clones 20/11 and 2/6, 1:400 dilution; Chemicon, Temecula, Calif) was applied using an avidin-biotin-peroxidase technique with diaminobenzidine as the substrate. Known adenovirus-infected tissue sections were used as the positive control. As a negative control, primary antibody was omitted.

To assess the presence of concomitant GVHD in case 2, apoptotic activity was studied by terminal deoxynucleotidyl transferase-mediated X-deoxyuridine triphosphate (dUTP) nick end-labeling (TUNEL) assay, adapted to an automated in situ hybridization instrument (Discover, Ventana Medical Systems). In this assay, 5-µm-thick deparaffinized sections were digested with protease I. In this procedure, which is template independent, the DNA degradation products are detected by enzymatic labeling of the free 3'-OH termini with modified nucleotides. Biotin 16-dUTP is the label used for this reaction, with hematoxylin used as a counterstain.

For electron microscopy studies, routine procedures were used. Colonie tissue from patient 2 was fixed in universal fixative (equal parts of 4% formaldehyde and 1% glutaraldehyde) and postfixed in 1% osmium tetroxide. The tissue was embedded in epoxy resin, and ultrathin sections from selected blocks were stained with uranyl acetate and lead citrate. Electron microscopic examination was performed using a Philips 201 (NV Philips Gloeilampenfabrieken, Eindhoven, The Netherlands) transmission electron microscope.

PATHOLOGIC FINDINGS

Light microscopic examination of the duodenal biopsies from case 1 showed foci of acute and chronic inflammation, with mild blunting of the villi (Figure 1). In addition, focal changes characteristic of moderately severe, acute GVHD were also present, including numerous apoptotic bodies within the crypts and surface epithelium. This was associated with focal crypt drop-out and mucosal regeneration. Occasional superficial epithelial cells with large, smudged, eosinophilic nuclei were noted (Figure 1). Immunohistochemical studies showed that the nuclei of these cells were labeled positively with the antibody against adenovirus (Figure 1). The foci of epithelial cells immunoreactive for adenovirus were mostly seen in areas of epithelium without changes of active GVHD. Immunostaining with the antibody against cytomegalovirus early antigen was negative. Microscopic examination of esophagus biopsies showed Candida albicans esophagitis. The mucosal biopsies of antrum and esophagus showed no evidence of active GVHD or adenovirus infection.

The autopsy for case 2 confirmed changes consistent with early metachromatic leukodystrophy. Severe GVHD with diffuse hemorrhagic enterocolitis (Figure 2, a), along with hepatomegaly, Candida esophagitis, pancreatitis, pulmonary hemorrhage, and acute tubular necrosis of the kidneys, was identified.

Light microscopy on sections of colon revealed extensive mucosal changes with dropout of glands and reactive regeneration indicative of severe acute GVHD (Figure 2, b). In addition, collections of periodic acid-Schiff-positive macrophages within lamina propria consistent with metachromatic leukodystrophy were present. Immunohistochemical studies using the antibody against adenovirus showed positive immunoreactivity in the nuclei of surface epithelium, crypts, as well as in occasional interstitial fibroblasts and macrophages (Figure 2, c). The TUNEL assay was positive in numerous crypt cells, indicating extensive apoptosis consistent with severe GVHD (Figure 2, d).

Ultrastructural studies of colonie mucosa confirmed the presence of intranuclear adenoviral particles with typical crystalline arrays in epithelial cells, as well as in interstitial fibroblasts and macrophages (Figure 2, e and inset).

COMMENT

Adenovirus is a nonenveloped, double-stranded DNA virus with more than 50 serotypes and 6 subgroups that can cause both active or latent infections.2 The incidence of adenovirus infection in hematopoietic stem cell transplantation recipients has been reported to be between 4.9% and 20.90%,3-5 resulting in a mortality rate of 10% to 60%, depending on the level of posttransplantation immunodeficiency.6 Although still prominent this incidence is lower in recipients of solid organ transplantation.7 In a recent review of 206 patients with BMT performed for hematologic malignancies, solid tumors, or nonmalignant conditions, only patients with hematologic malignancies had adenoviral infections.8 Recently, it has been reported that approximately 65% of pediatric recipients of unrelated hematopoietic stem cell transplantation were positive for adenovirus at some point during their hospitalization.4 This increased rate of infection in susceptible patients may be attributable to more frequent viral culturing, better detection methods, an increase in the number of high-risk transplant patients, or more intensive immunosuppression therapy.9 A study from San Diego reported an increase in the incidence of adenovirus infection from 3.5% recorded during 1976-1978 to 12.4% during 1991-1993.5 In general, the incidence of adenovirus infection in adults is significantly higher among recipients of allogeneic BMT (6%) compared to recipients of autologous BMT (0.92%).2 In one study, pediatric BMT recipients were infected with adenovirus 3.5 times more frequently than adult recipients.4 The median time from the day of stem cell transfusion to the first adenovirus-positive culture has been reported to be 41 days (within 30 days in children and >90 days in adults), and in 64% of the cases, positive cultures were associated with evidence of tissue invasion.7,9

Although adenovirus is responsible for several self-limited diseases in pediatric patients, significant mortality has been associated with adenovirus infection in immunosuppressed patients. The mortality is highest among patients with adenovirus pneumonia and disseminated disease.2

Four clinically significant syndromes occurring in allogeneic stem cell transplant patients who are infected with adenovirus have been described: (a) gastrointestinal infection, manifested as diarrhea, hemorrhagic colitis, and hepatitis; (b) urinary tract infection, associated with hemorrhagic cystitis and possible renal failure and nephritis; (c) pulmonary infection, resulting in potentially fatal interstitial pneumonia; and (d) asymptomatic nasopharyngeal infection.3 The risk of developing disseminated adenovirus infection is similar for both the pediatric and adult populations.9 Significant risk factors for developing disseminated adenovirus infections, associated with poor prognosis, include presence of moderate to severe acute GVHD (noted in both of our patients), immunosuppressive therapy, isolation of adenovirus from multiple sites, HLA mismatched or unrelated transplants, and use of T-cell-depleted allografts.3-5,8,9 Among various risk factors, severe GVHD appears to be the major contributory factor in unmanipulated allograft recipients, underscoring the role of immunosuppression in the genesis of the adenovirus infection in allograft recipients.3,4 It has been noted also that acute GVHD (grade 3 or higher) and a long delay between infection and treatment correlate with a more significant risk of treatment failure.6 The association between acute GVHD and adenovirus infection has been reported in several studies.4,8,9 It has been suggested that an increased risk of adenovirus infection in the setting of acute GVHD is related to treatment with steroids and other immunosuppressive therapy.4 In the gastrointestinal tract, epithelial cell injury (apoptosis) is a characteristic finding of acute GVHD, a well-recognized complication of BMT.10 Many viral pathogens, including adenovirus, can induce apoptosis in different cell types.11 In the case of adenovirus, the basic mechanism involves interactions with p53-dependent and -independent pathways modulated by various viral proteins.12 It is therefore conceivable that the epithelial damage initiated by acute GVHD would be further aggravated by superimposed adenovirus infection, as demonstrated in our 2 cases. Since adenovirus enterocolitis can be overlooked on routine histology,13 immunohistochemical and ultrastructural studies are recommended to establish an accurate diagnosis and institute the appropriate treatment.

The authors thank Vern Edwards for reviewing the manuscript and electron microscopic data, and Wilson Chan for performing immunohistochemical staining.

References

1. Yolken RH, Bishop CA, Townsend TR, el al. Infectious gastroenteritis in bone-marrow-transplant recipients. N Engl J Med. 1982;306:1010-1012.

2. La Rosa AM, Champlin RE, Mirza N, et al. Adenovirus infections in adult recipients of blood and marrow transplants. Clin Infect Dis. 2001;32:871-876.

3. Blanke C, Clark C, Broun ER, et al. Evolving pathogens in allogeneic bone marrow transplantation: increased fatal adenoviral infections. Am J Med. 1995;99:326-328.

4. Flomenberg P, Babbitt J, Drobyski WR, et al. Increasing incidence of adenovirus disease in bone marrow transplant recipients. J Infect Dis. 1994;169:775-781.

5. Bruno B, Boeckh M, Davis C, Gooley T, Hackman RC. Adenovirus infections in patients undergoing bone marrow transplantation. Blood. 1997;90:1867a.

6. Bordigoni P, Carret AS, Venard V, Witz F, Le Faou A. Treatment of adenovirus infections in patients undergoing allogeneic hematopoietic stem cell transplantation. Clin Infect Dis. 2001;32:1290-1297.

7. Carrigan DR. Adenovirus infections in immunocompromised patients [review]. Am J Med. 1997;102(3A):71-74.

8. Hale GA, Heslop HE, Krance RA, et al. Adenovirus infection after pediatric bone marrow transplantation. Bone Marrow Transplant. 1999;23:277-282.

9. Howard DS, Phillips II GL, Reece DE, et al. Adenovirus infections in hematopoietic stem cell transplant recipients. Clin Infect Dis. 1999;29:1494-1501.

10. McDonald GB, Shulman HM, Sullivan KM, Spencer GD. Intestinal and hepatic complications of human bone marrow transplantation: part I. Gastroenterology. 1986;90:460-477.

11. Teodoro JG, Branton PE. Regulation of apoptosis by viral gene products. J Virol. 1997;71:1739-1746.

12. Schmidt M, Afione S, Kotin RM. Adeno-associated virus type 2 Rep 78 induces apoptosis through caspase activation independently of p53. J Virol. 2000;74:9441-9450.

13. Yan Z, Nguyen S, Poles M, Melamed J, Scholes JV. Adenovirus colitis in human immunodeficiency virus infection: an underdiagnosed entity. Am J Surg Pathol. 1998;22:1101-1106.

Katayoon Shayan, MD; Fred Saunders, MD; Eve Roberts, MD; Ernest Cutz, MD

Accepted for publication July 9, 2003.

From the Division of Pathology, Department of Pediatric Laboratory Medicine (Dr Shayan), Division of Hematology/Oncology, Department of Pediatrics (Dr Saunders), Division of Gastroenterology and Nutrition, Department of Pediatrics (Dr Roberts), and the Division of Pathology, Department of Laboratory Medicine and Pathobiology (Dr Cutz), The Hospital for Sick Children and University of Toronto, Toronto, Ontario. Dr Shayan is now with the Department of Pathology and Laboratory Medicine, Kaleida Health, Buffalo General Hospital, Buffalo, NY.

Reprints: Ernest Cutz, MD, The Hospital for Sick Children and University of Toronto, 555 University Ave, Toronto, Ontario, Canada M5G 1×8 (e-mail: cutz@sickkids.ca).

Copyright College of American Pathologists Dec 2003
Provided by ProQuest Information and Learning Company. All rights Reserved