Rectal inflammation as first manifestation of graft-vs-host disease: radiologic-pathologic findings

Bowel ultrasonography as an aid for diagnosis of intestinal acute graft-versus-host-disease after allogeneic haematopoietic stem cell transplantation

OBJECTIVE

Aim of our prospective study was to investigate accuracy of bowel ultrasonography in detecting gastrointestinal acute graft versus host disease (GVHD), when using clinical assessment as gold standard. In a subgroup of patients, bowel ultrasonography was compared with colonoscopy and histology in diagnosing of gastrointestinal acute GVHD.

METHODS

Fifty-two patients underwent allogeneic hematopoietic stem cell transplantation and developed gastrointestinal symptoms.

RESULTS

Clinical assessment lead to a diagnosis of gastrointestinal acute GVHD in 17/52 patients, no gastrointestinal acute GVHD was detected in 20/52 patients, while 15 patients were not able to complete the study. Bowel ultrasonography detected either bowel wall thickness of the ileum and the colon or dilation in 16/17 patients and showed 94% sensitivity (95% CI 0.69-0.99), 95% specificity (95% CI 0.73-0.99), and 94.5% accuracy. Colonoscopy was performed in 13/52 patients, showing gastrointestinal acute GVHD in 11/13. In these 11 patients, histology confirmed the diagnosis of gastrointestinal acute GVHD, and bowel ultrasonography detected findings compatible with gastrointestinal acute GVHD in all 11 patients, and was negative in the 2 patients with no gastrointestinal acute GVHD.

CONCLUSION

Bowel ultrasonography can be considered a valuable tool to add to clinical assessment for patients with suspected gastrointestinal acute GVHD for addressing a prompt and appropriate treatment.

CT-analysis of the course of gastrointestinal graft-versus-host disease--patterns of involvement

OBJECTIVE

To describe the main patterns of distribution of gastrointestinal graft-versus-host disease (GVHD) and their chronological course.

METHODS

Twenty-five adult patients (17 men, 8 women, mean age 47 years) were enrolled from 11/2003 to 11/2007. All patients underwent abdominopelvic CT shortly after onset of GVHD-related symptoms and also at follow up. The mean number of CT examinations per patient was 3.2±2.7 with a total of 81 in a median time period of 97 days after HCT. The gastrointestinal tract was divided into 7 segments. Gastrointestinal abnormalities were defined as follows: presence of wall thickening (>4mm), increased mucosal enhancement, bowel dilatation (>3 cm for the small bowel, >8 cm for the colon), fluid-filled loops of the bowel, bowel loop separation and double-halo sign.

RESULTS

40% (10/25) of the patients presented a classical pattern of evolution of involved segments by GI-GVHD. In these cases, especially the small bowel was initially involved showing a retreat with time towards the terminal ileum with longer length of stay in this location. 28% (7/25) of the patients presented with a nonclassical permanently migratory involvement of the GI jumping from one GI segment to another. Other 32% (8/25) of our patients revealed a nonclassical persistent, unchanged involvement pattern of GI involvement by GVHD at time.

CONCLUSION

Contrary to existing reports, our data collected in 25 patients diagnosed with GI-GVHD after allogeneic hematopoietic stem cell transplantation suggest the presence of three different courses (classical, nonclassical migratory and nonclassical persistent) of this disorder. Awareness of this knowledge enables more accurate risk stratification.