Clinicopathologic Threshold of Acute Colorectal Graft-versus-Host Disease

-Colon biopsies are often used to determine the presence and severity of acute gastrointestinal graft-versus-host disease following bone marrow transplant.

-To establish a threshold consensus within our institution on the number of crypt apoptotic bodies (CAB) indicative of grade 1 acute colorectal graft-versus-host disease, we retrospectively reviewed colon biopsies from posttransplant patients and incorporated clinical and endoscopic findings to validate recently proposed minimum criteria for grade 1 graft-versus-host disease as 7 or more CAB per 10 contiguous crypts.

-Eighty-one biopsies performed for suspected graft-versus-host disease from 74 individual patients were initially stratified based on their prior (prestudy) diagnoses: no significant abnormality, grade 1 graft-versus-host disease, and descriptive diagnoses mentioning increased apoptosis. A chart review was performed to assess the clinical and endoscopic impression at the time of biopsy and to determine the subsequent management and outcome.

-Twenty-six biopsies with an average of 3 CAB were considered true-negative cases, and 32 biopsies with an average of 9.75 CAB were considered true-positive cases (t = 3.95999, P < .001). True-negative cases had an average density of 1.36 CAB per crypt, and true-positive cases had an average density of 2.97 CAB per crypt (t = 3.950178, P < .001).

-A threshold of 7 or more CAB per 10 contiguous crypts promotes appropriate treatment of grade 1 acute graft-versus-host disease after other diagnostic entities are excluded. Although this threshold is 100% specific to grade 1 acute colorectal graft-versus-host disease after other histologic mimics are excluded, this threshold has a low sensitivity (59.4%) as patients with less than 7 CAB per 10 contiguous crypts constitute a heterogeneous group.