The Clinical Relevance of Gastrointestinal Endoscopy in Pediatric Patients with Suspected Gastrointestinal Graft-Versus-Host Disease after Hematopoietic Stem Cell Transplantation

Background and aim: Graft-versus-host disease (GVHD) affecting visceral organs is a severe complication of hematopoietic stem cell transplantation (HSCT). GVHD in the gastrointestinal tract (GI-GVHD) is primarily a clinical-based diagnosis, but due to non-specific symptoms histopathology verification is recommended. Data in children concerning the efficacy of endoscopic examination performed due to clinically suspected GI-GVHD is scarce. The aim of this study was to investigate the modification of the pharmacological treatment regimen based on the gastrointestinal histopathology report.

Material and Methods: A retrospective clinical study from 3 out of 4 transplantation(s) centers in Sweden including 62 children (40 boys, 22 girls) were enrolled. The patients underwent HSCT during 2000-2012. Endoscopy with biopsies, were performed due to clinically suspected GI-GVHD, <1 year after HSCT. Median age at the time of endoscopy was 4.2 years (0.3-18.6 years).

Results: In total, 76 endoscopies were performed. Out of these, 62 % of them were done ≤100 days post-HSCT. A combined upper gastroscopy and lower endoscopy (ileocolon-, colon- or rectosigmoidoscopy) was performed in 59 %, whereas solely lower or upper endoscopy was performed in 37 % and 4 %, respectively. Symptoms at the time of endoscopy consisted of 64 cases with diarrhea, 35 with nausea or vomiting, 31 with abdominal pain and in 10 cases, failure to thrive.

GVHD at any biopsy site was confirmed by histopathology in 54%. In 5 of these cases, concomitant viral enteritis was reported (4 CMV, 1 Adenovirus). The proportion of GVHD per number of endoscopy occasions before and after 100 days post-HSCT was 64 % (30/47) and 38% (11/29), respectively. CMV enteritis was detected in one case and posttransplant lymphoproliferative disease (PTLD) in 2 cases. Seven out of the 8 cases with verified viral GI infection were observed < 100 days post-HSCT.

53 children had an ongoing additional anti-GVHD treatment at the time of endoscopy, of whom 62 % had histopathology confirmed GVHD, whereas 38 % had normal findings. Among the 23 cases, not treated with additional anti-GVHD treatment pre-endoscopy, 8 (35 %) had histopathological findings consistent with GVHD. Modification of the pharmacological treatment regimen, based on the histopathology reports was done in 38 %. Among the 29 cases with modified treatment regimen, 52% received an initiation or a more intensive anti-GVHD treatment, 28 % a withdrawal of additional anti-GVHD treatment, 7% were subjected to withdrawal of additional anti-GVHD treatment and initiation of anti-viral treatment, and in one patient anti-viral treatment was initiated.

Discussion: The clinical usefulness of endoscopy, leading to pharmacological intervention has not previously been investigated in the pediatric HSCT patients with clinically suspected GI-GVHD. Clinicians may hesitate to perform endoscopy in this population due to risks related to general anesthesia/deep sedation and complications such as bleeding, perforations, or septicemia. Conversely, heavy immunosuppressive treatment in patients with viral GI infection might be fatal. We demonstrated that modification of treatment occurred in 38% of the patients, based on the histopathology report. Interestingly, the highest frequency of positive histopathology reports was observed when endoscopy was performed <100 days post-HSCT. During this time period 7 out of 8 cases with viral GI infection, including the 2 cases with severe PTLD, were diagnosed. Furthermore, we also noted that clinical-based diagnosis of GI-GVHD, where additional anti-GVHD treatment was started pre-endoscopy, was incorrect in approximately every third case.

Conclusion: Our results provide evidence that there is a clinical relevance in performing endoscopy in children with clinically suspected GI-GVHD and should be performed in all of these children. Endoscopy in these cases will allow the clinician to target treatment, by confirming or rule out two clinically indistinguishable diseases, GVHD and viral GI infection, but also to avoid unnecessary immunosuppressive drugs.