Biology of Blood and Marrow Transplantation
Volume 15, Issue 5 , Pages 643-648, May 2009

Seeing What's Out of Sight: Wireless Capsule Endoscopy's Unique Ability to Visualize and Accurately Assess the Severity of Gastrointestinal Graft-versus-Host-Disease

  • Prakash Varadarajan

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Lauren M. Dunford

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Julie A. Thomas

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Karen Brown

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Pamela Paplham

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Margaret Syta

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Michael Schiff

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Swamithan Padmanabhan

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Minoo Battiwalla

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Shannon Smiley

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Theresa Hahn

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • ,
  • Philip L. McCarthy Jr.

      Affiliations

    • Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
    • Corresponding Author InformationCorrespondence and reprint requests: Philip L. McCarthy, Jr., MD, Blood and Marrow Transplantation Program, Roswell Park Cancer Institute, Department of Medicine, Elm and Carlton Streets, Buffalo, NY 14263.

Received 21 January 2009; accepted 3 February 2009.

Article Outline

Abstract 

Early recognition of gastrointestinal graft-versus-host disease (GI GVHD) after allogeneic hematopoietic stem cell transplantation (alloHSCT) is vital to initiation of therapy. However, the most common location, the small bowel (SB), is difficult to access with upper and lower endoscopy (UGE/LGE). Wireless capsule endoscopy (WCE) is a noninvasive technology allowing complete SB evaluation. The capsule location can also be tracked to identify motility derangements. From August 2006 to July 2007, 11 alloHSCT patients with GI symptoms underwent WCE, and visual grading was performed. UGE and LGE with biopsies were done when clinically indicated. All patients had evidence of probable acute GVHD (aGVHD) on WCE. WCE revealed lesions of greater severity than those seen by UGE or LGE in most patients. WCE demonstrated that 45% of patients had delayed gastric transit time. WCE is an excellent, noninvasive method for assessing GI GVHD, with the ability to more accurately assess the severity of GVHD, evaluate clinical symptoms, and follow response to treatment.

Key Words: Wireless capsule endoscopy, WCE, Graft-versus-host disease, GVHD, Gastrointestinal

 

Back to Article Outline

Introduction 

Gastrointestinal graft-versus-host disease (GI GVHD) is a frequent complication of allogeneic hematopoietic stem cell transplantation (alloHSCT), presenting with diarrhea, nausea, vomiting, abdominal pain, anorexia, or hematochezia. One prospective study identified GI GVHD as the sole culprit of persistent GI symptoms in 81% of patients 20 days beyond alloHSCT [1]. However, other conditions may mimic the clinical presentation of GI GVHD. Chemotherapeutic toxicity can account for similar symptoms in the first 2 weeks post-alloHSCT 1, 2. Early recognition is vital to initiating appropriate therapy, yet this can be especially challenging when GVHD affects the GI tract in the absence of skin involvement. Furthermore, the small bowel (SB), an area that is difficult to access with traditional endoscopic methods, is the most common location of GI GVHD [3]. Wireless capsule endoscopy (WCE) is a noninvasive technology allowing for complete SB evaluation, and has been successfully used in the diagnosis of occult bleeding and inflammatory bowel disease 3, 4. More recently, WCE has been used to assess GI GVHD 5, 6, 7. We evaluated the efficacy of WCE in the diagnosis and evaluation of treatment response of GI GVHD in symptomatic patients after alloHSCT.

Back to Article Outline

Materials and Methods 

Patients and Data Collection 

From August 2006 to July 2007, 11 alloHSCT patients of median age 50 years (18–67), with symptoms of diarrhea, nausea, vomiting, abdominal pain, or hematochezia underwent WCE to assist in diagnosis, assess therapy response, or evaluate other GI pathology. Informed consent for WCE was obtained from all patients. Concomitant studies including stool bacterial, fungal, and viral cultures, Clostridium difficile toxin assay. Blood Epstein-Barr virus (EBV) and cytomegalovirus (CMV) antigen testing were also obtained. Upper GI and lower GI endoscopy (UGE, LGE) were performed, and biopsies taken when clinically indicated. Patients 1 to 9 had symptoms typical of acute GI GVHD, and had histology diagnostic of acute GVHD (aGVHD) on the mucosal biopsy specimens. All studies for pathologic infectious agents in mucosal biopsy specimens were negative. Patients 1 to 9 had concomitant UGE/LGE studies and WCE, allowing for comparison of their results. Patients 10 and 11 had WCE alone. This study was approved by the institutional review board of Roswell Park Cancer Institute (RPCI).

Wireless Capsule Endoscopy 

Patients drank up to 1 liter of polythethylene glycol electrolyte solution and/or clear liquids, the day prior to the procedure. Patients were connected to the recording device after a 12-hour overnight fast. Two hours after swallowing the capsule (GIVEN Imaging, Yoqneam, Israel), they were allowed to drink clear liquids. The capsule recorded images at 2 per second for a period of 8 hours, producing greater than 50,000 images, which were analyzed by a single gastroenterologist (M.S.). Visual grading of GI GVHD was determined on a 5-point scale by Brand et al. [8]: grade 0, normal; grade 1, loss of vascular marking and/or mild focal erythema; grade 2, moderate edema and/or erythema; grade 3, edema, erythema, erosions, and/or bleeding; and grade 4, ulceration, exudates, and bleeding. The following motility parameters were documented:

1.Gastric retention is defined as capsule remaining in the stomach for 8 hours.

2.Gastric transit time (GTT) is defined as the time in minutes between the first gastric image and the first duodenal image. GTT was considered delayed if the pylorus was not reached in 120 minutes. The average GTT in healthy patients is 10–60 minutes 9, 10.

3.Small bowel transit time (SBTT) is the time in minutes between the first duodenal image and the first image of the cecum.

Back to Article Outline

Results 

WCE for GVHD Imaging 

The 11 patients underwent WCE at a median day 77 (range: 18-593) after alloHSCT (Table 1). No patient had side effects attributable to WCE, and the capsule was passed in all patients. The longest time to passage was 14 days. No patients had positive CMV antigenemia, EBV serologies, or C. difficile toxin assays at the time of WCE. See Figure 1 for representative WCE images. SB lesions visualized by WCE were more severe than those seen on UGE/LGE in 7 out of 9 of the patients who also had contemporaneous UGE/LGE studies. The WCE studies had better correlation with clinical appearance and severity of diarrhea in 4 of 11 patients (patients 1, 2, 5, and 9) compared to UGE/LGE and biopsy (see Table 1). Several of the WCE studies showed lesions in a discontinuous pattern with relative sparing of other small bowel areas. For example, in patient 1, initial WCE showed severe involvement of the proximal jejunum and distal ileum with relative sparing of intervening areas. Repeat WCE revealed differential healing with marked improvement in the proximal jejunal lesions, whereas the distal ileum continued to have several areas of erosions and ulceration. Nine of the 11 patients had corresponding UGE/LGE. All 9 patients had evidence of GI GVHD present on biopsy.

Table 1. Patient Characteristics, WCE, Endoscopic, and Histologic Results, Treatments, Modifications Based on WCE Results and Outcomes
PtAge/SexDxWCE day post- HSCTWCE Lesions and ExtentEGS Grade WCE∗UGE/LGE (day post HSCT) Lesions and ExtentEGS Grade UGE/LGE∗GVHD Rx prior to WCEChanges post-WCEOutcome
131/MAML161Erosions, ulcer, bleeding (prox/mid jejunum); sparing of mid/distal jejunum; ulcer (distal ileum)4UGE/Flex Sig (160) erythema, edema (stomach); edema (rectum)1-2MPStarted on FK506, MMF, MCSteroids tapered, continued on FK506, sirolimus, infliximab, BDP, MMF, and ATG, and diarrhea subsided on day 186
Repeat WCE on Patient 1195Edema, erythema (prox jejunum); sparing of mid/distal jejunum; erosions, ulcers (distal ileum)2-4Not repeatedNoneMP, sirolimus, FK506, ATG, infliximab, BDP, MMFMP tapered, MC stoppedDiarrhea subsided in a week, MP tapered, and patient discharged on day 213
229/FAML214Edema, erythema (duodenum); patchy erosions, ulcer (mid jejunum)2-4UGE (212) hyperemia, edema (stomach, prox duodenum)2MP, BDP, FK506, MMF, infliximabNoneGI GVHD flare subsided and pt was discharged on day 223 on BDP, MMF, and FK506
355/FALL55Edema, erosions, ulcers (jejunum)2-4UGE/LGE (30) edema (stomach, prox duodenum); erosions, ulcers (hepatic flexure)2-4MP, sirolimus, FK506, MMF, infliximab, daclizumabNoneGI GVHD subsided and pt was discharged on day 60 on FK506, MMF, BDP, MP, and sirolimus
457/FAML99Edema, erosions, ulcers, blood clots (jejunum); edema, erosions (ileum)3-4LGE (68) erythema, edema, erosions, ulcers (ascending colon)1-4MP, sirolimus, FK506, MMF, infliximab, ATGDaclizumab addedPt died on day 127 of refractory GI GVHD
550/MAML593Erosions (stomach, distal jejunum, prox ileum)3UGE/LGE (589) hyperemia, edema (stomach); edema (colon)2BDP, MP, FK506, MMF, infliximab, ECPNo changeAbdominal symptoms resolved and pt discharged on day 614
640/FAML37Erosions, ulcers, blood clots (stomach); capsule retained in stomach4UGE/LGE (37) edema, ulcer, bleeding (stomach, duodenum); patchy ulcers (colon)4MMF, FK506, MPNo changePt died on day 77 of refractory GI GvHD
745/MAML170Ulcers (duodenum, jejunum); erosions, ulcers (distal ileum)3-4UGE, jejunoscopy/LGE (204) edema, patchy erosions (jejunum); ulcer (ileum); edema (colon)2-4MMF, cyclosporine, MP (low dose)MP increasedCystitis and lower GI bleeding resolved. Pt discharged on day 179 with MP taper
858/MNHL18Ulcers, erosions (prox jejunum)3-4LGE (15) edema, erythema (ileum); edema, erythema (sigmoid colon)3-4MMF, FK506, MPNo changeGI GvHD subsided and discharged on day 25
967/FAML32Multiple erosions, ulcers (jejunum)3-4UGE(32) erythema, edema (stomach);edema (rectum)2FK506, MMF, MPStarted on infliximabPt died on day 39 of refractory GVHD
1063/FMDS161Erythema, edema (stomach)2Not doneNot doneBDP, FK506, MMF, MP, ursodiolMP tapered, MC startedSymptoms resolved and pt discharged on day 168
1118/FMDS27Patchy erosions (mid jejunum, prox/mid ileum)2-3Not doneNot donenoneNonePt discharged on day 29 tolerating oral food

WCE indicates wireless capsule endoscopy; Dx, diagnosis; Rx, therapy; GI, gastrointestinal; GVHD, graft-versus-host-disease; Dx, diagnosis; AML, acute myeloid leukemia; ALL, acute lymphocytic leukemia; NHL, Non Hodgkin Lymphoma; MDS, myelodysplastic syndrome; ATG, antithymocyte globulin; BDP, beclomethasone dipropionate; MP, methylprednisolone 2 mg/kg unless marked as low dose, which is 1 mg/kg; MMF, mycophenolate mofetil; FK506, tacrolimus; MC, metoclopramide; EGS, endoscopic grading scale; HSCT, hematopoietic stem cell transplant; LGE, lower GI endoscopy; Flex Sig, flexible sigmoidoscopy; UGE, upper GI endoscopy; prox, proximal.

∗GVHD lesions ranged in severity in different locations A range is given when lesions varied by site.

  • View full-size image.
  • Figure 1 

    WCE imaging in patients with GI GVHD. (A) multiple gastric erosions; (B) erosions with ulcer in antrum; (C) proximal ileal erosion; (D) distal ileal ulcer with deep, large clot.

Gastric Retention, Gastric, and SBTT 

Two patients had gastric capsular retention (patients 2 and 6). Patient 2 had UGE changes of mild GVHD, which was confirmed by a duodenal biopsy. Patient 6 had endoscopic evidence of severe GVHD with UGE revealing severe edema, hyperemia, narrowing of the pyloric lumen, and LGE, revealing an aperistaltic colon. The median GTT of the remaining 9 patients was 81 minutes (range: 4-236). Four patients (1, 4, 5, and 10) had delayed GTT, and 3 of these patients had significant nausea at the time of the WCE study. The cecum was not visualized in 2 of these patients (patients 1 and 4). The capsule was endoscopically placed in the duodenum in patient 2 after gastric capsular retention and in patient 9 because of severe vomiting. SBTT in 9 patients was a median 253 minutes (range: 162-444).

Back to Article Outline

Discussion 

GI GVHD has been considered a hypermotility disorder because of the increasing diarrhea that can accompany disease progression. We observed gastric hypomotility in association with GI GVHD. Our observed prevalence of 18% gastric retention of the WCE is similar to that reported for acutely ill patients, including patients with GVHD (24%) [9]. The average GTT in healthy patients is 10-60 minutes but the clinical significance of delayed GTT on WCE has not been well studied 9, 10, nor have WCE-measured GTT been compared with scintigraphic studies, the standard for GTT measurement and gastroparesis diagnosis [11]. Our findings are consistent with prior reports of gastric dysmotility after HSCT. Eagle et al. [12], performed scintigraphic gastric emptying studies in 18 autologous and alloHSCT patients with gastroparetic symptoms and found delayed GTT in 78%. Wu et al. [1] evaluated persistent nausea and anorexia in alloHSCT patients and found endoscopic evidence of delayed gastric emptying, manifested by retained food and bilious material. Thus, WCE may be a useful tool for prospectively and noninvasively evaluating gastric function during the posttransplant period.

WCE visualization of the entire SB allowed for changes in management. Patient 1 was on MP prior to WCE but after the results of the WCE indicated Endoscopic Grading Scale (EGS) of 4 (versus 1-2 on UGE/LGE), treatment was intensified (Table 1). Repeat WCE indicated a decrease in the severity of GvHD with a range of 2-4, indicating response to adjustment of therapy. In patient 9, WCE more accurately correlated with clinical outcome. Patient 9 had EGS of grade 2 GI GVHD on UGE, but a range of 3 to 4 on WCE and ultimately succumbed to severe GI GVHD (Table 1).

WCE findings may not differentiate mild GVHD from other inflammatory conditions such as viral, bacterial infections. Ten percent of healthy patients may have erythema, erosions, and small ulcerations on WCE studies [13]. Thus, stool cultures for pathogens as well as histopathologic confirmation by UGE/LGE is important for diagnostic confirmation. Another limitation of WCE is the inability to visualize the entire SB when there is delayed capsule passage. In our study, the cecum was not visualized in 2 patients (patients 1 and 4) because of prolonged GTT or SBTT, and distal SB lesions may have been missed. Several trials have shown that administration of pro-kinetic agents such as metoclopramide or erythromycin just before introduction of the capsule or placing the patient in right lateral position enhances the chances of study completion [9]. In rare circumstances, the capsule must be retrieved by endoscopy or surgical intervention. In our patient series we did observe capsule retention for several days limiting the potential use of magnetic resonance imaging during this time period. However, all patients passed the capsule without endoscopic or surgical intervention.

There are other noninvasive methods to assess GI GVHD. High-resolution transabdominal ultrasonography (HRU) and color Doppler imaging (CDI) have been investigated. Patients with confirmed aGVHD had bowel wall thickening primarily in the ileocecal region [14]. CDI indicated an increased arterial perfusion of the bowel wall and superior mesenteric artery in patients with typical, but nonspecific signs inflammation and mucosal erythema and edema on endoscopy. CDI also indicated decreased arterial perfusion in patients whose endoscopy yielded no bleeding after biopsy [14]. Although this assessment method is noninvasive, it does not directly observe the intestinal mucosa. Thickening of the bowel wall is nonspecific, and can be a result of infection [14]. F-fluorodeoxyglucose positron emission tomography (FDG-PET) has also been investigated [15]. An increased intestinal accumulation of FDG was seen in 14 out of 17 patients with GI GVHD compared to 0/13 patients without GVHD. FDG-PET was used to monitor response to treatment [15]. Disadvantages of FDG-PET are similar to the HRU and CDI methods. There is no visualization of the intestinal mucosa. Milligan et al. [16] used the Schilling test, a measure of vitamin B12 absorption, with intrinsic factor to examine the ileal vitamin B12 absorption after alloHSCT. There was abnormal B12 absorption following the HSCT conditioning regimen, and it was not possible to determine if acute GI GVHD was associated with further decrease in absorption of B12. Interestingly, chronic GVHD (cGVHD) patients (not limited to GI GVHD) had decreased vitamin B12 absorption even in the absence of GI symptoms [16].

WCE is a useful tool to diagnose GI GVHD in both the acute and chronic setting, especially for documenting SB involvement. It is well tolerated and less invasive than UGE/LGE. We found WCE to more accurately assess and visualize the severity of GI GVHD and better evaluate clinical symptoms. WCE allowed for visual assessment of the response to therapy. This is consistent with previous reports that endoscopic and histologic evaluations of the upper GI tract can underestimate the severity of disease involvement 17, 18. In patients with widespread severe GI GVHD, WCE findings help establish the diagnosis and extent of disease. WCE findings were able to localize the area of involvement, which is important because of the intermittent nature of GI GVHD where lesions can be scattered throughout the small intestine. Lesions visualized on WCE were similar to those seen on UGE/LGE and included erythema, erosions, ulcers, and blood clots [18]. Being able to visualize the mucosa proves advantageous, especially when using WCE as a method of diagnosis and response to treatment. Another benefit of WCE is that the entire small bowel can be visualized, unlike UGE/LGE. Thus, the clinician often would have to rely solely on clinical manifestations. WCE will not eliminate the need for invasive procedures such as UGE/LGE, as it is still necessary to obtain a biopsy for histopathologic diagnosis, but, it can significantly reduce the number of times a patient is subjected to endoscopic procedures. Utilizing WCE can visualize the intestinal mucosa and observe what was once seen only be invasive methods.

Back to Article Outline

Acknowledgments 

P.V. compiled the initial data, prepared an abstract and poster for ASH 2007, and helped author further drafts. L.D. and J.T. performed literature searches and authored further drafts. K.B., P.P., and M.S. are nurse practitioners who provided daily clinical care for the patients in the series. M.S. is a staff gastroenterologist who analyzed all WCE studies. M.B., P.M., S.P., and S.S. are BMT physicians who oversaw the clinical care for the patients. M.B., S.S., and J.T. contributed to the editing of drafts. T.H. and P.M. were the primary investigators in this project, designed the study, and reviewed multiple drafts.

Financial disclosure: The authors have no grant, financial support, or competing financial interests to disclose.

Back to Article Outline

References 

  1. Wu D, Hockenberry DM, Brentnall TA, et al. Persistent nausea and anorexia after marrow transplantation: a prospective study of 78 patients. Transplantation. 1998;66:1319–1324
  2. DiBaise JK, Brand RE, Lyden ER, Tarantolo SR, Quigley EM. Gastric myoelectrical activity and its relationship to the development of nausea and vomiting after intensive chemotherapy and autologous stem cell transplantation. Am J Gastroenterol. 2001;96:2873–2881
  3. Eliakim R, Adler SN. Capsule video endoscopy in Crohn's disease—the European experience. Gastrointest Endosc Clin N Am. 2004;14:129–137
  4. Delvaux M, Fassler I, Gay G. Clinical usefulness of the endoscopic video capsule as the initial intestinal investigation in patients with obscure digestive bleeding: validation of a diagnostic strategy based on the patient outcome after 12 months. Endoscopy. 2004;36:1067–1073
  5. Yakoub-Agha I, Maunoury V, Wacrenier A, et al. Impact of small bowel exploration using video-capsule endoscopy in the management of acute gastrointestinal graft-versus-host disease. Transplantation. 2004;78:1697–1701
  6. Neumann S, Schoppmeyer K, Lange T, et al. Wireless capsule endoscopy for diagnosis of acute intestinal graft-versus-host disease. Gastrointest Endosc. 2007;65:403–409
  7. Shapira M, Adler SN, Jacob H, Resnick IB, Slavin S, Or R. New insights into the pathophysiology of gastrointestinal graft-versus-host disease using capsule endoscopy. Haematologica. 2005;90:1003–1004
  8. Brand RE, Tarantolo SR, Bishop MR, Pavletic ZS, Lyden ER, Quigley EM. The correlation of endoscopic grading to clinical and pathologic staging of acute graft-versus-host-disease. Blood. 1998;92(suppl 1):45;[abstract]
  9. Ben-Soussan E, Savoye G, Antonietti M, Ramirez S, Lerebours E, Ducrotte P. Factors that affect gastric passage of video capsule. Gastrointest Endosc. 2005;62:785–790
  10. Lewis B. Capsule endoscopy—transit abnormalities. Gastrointest Endosc Clin N Am. 2006;16:221–228
  11. Parkman HP, Hasler WL, Fisher RS. American Gastroenterological Association technical review on the diagnosis and treatment of gastroparesis. Gastroenterology. 2004;127:1592–1622
  12. Eagle DA, Gian V, Lauwers GY, et al. Gastroparesis following bone marrow transplantation. Bone Marrow Transplant. 2001;28:59–62
  13. Goldstein JL, Eisen GM, Lewis B, Gralnek IM, Zlotnick S, Fort JG. Video capsule endoscopy to prospectively assess small bowel injury with celecoxib, naproxen plus omeprazole, and placebo. Clin Gastroenterol Hepatol. 2005;3:133–141
  14. Klein SA, Martin H, Schreiber-Dietrich D, et al. A new approach to evaluating intestinal acute graft-versus-host disease by transabdominal sonography and colour Doppler imaging. Br J Haematol. 2001;115:929–934
  15. Stelljes M, Hermann S, Albring J, et al. Clinical molecular imaging in intestina graft-versus-host disease: mapping of disease activity, prediction, and monitoring of treatment efficiency by positron emission tomography. Blood. 2008;111:2909–2918
  16. Milligan DW, Quick A, Barnard DL. Vitamin B12 absorption after allogeneic bone marrow transplantation. J Clin Pathol. 1987;40:1472–1474
  17. Ponec RJ, Hackman RC, McDonald GB. Endoscopic and histologic diagnosis of intestinal graft-versus-host disease after marrow transplantation. Gastrointest Endosc. 1999;49:612–621
  18. Cheung DY, Kim JI, Kim SS, et al. Endoscopic evaluation in gastrointestinal graft-versus-host disease: comparisons with histological findings. Digest Dis Sci. 2008;53:2947–2954

 Financial disclosure: See Acknowledgments on page 647.

PII: S1083-8791(09)00076-7

doi:10.1016/j.bbmt.2009.02.002

Biology of Blood and Marrow Transplantation
Volume 15, Issue 5 , Pages 643-648, May 2009

Article Tools

* Image Usage & Resolution