Citation
Ja. Bombi et al., ASSESSMENT OF HISTOPATHOLOGIC CHANGES IN THE COLONIC BIOPSY IN ACUTE GRAFT-VERSUS-HOST DISEASE, American journal of clinical pathology, 103(6), 1995, pp. 690-695
Abstract
The authors reviewed 78 colonic biopsies from 63 patients undergoing b
one marrow transplantation (BMT), 56 allogeneic and 7 autologous, to a
ssess different morphologic changes related with intestinal acute graf
t-versus-host disease (A-GVHD), Only five biopsies were done before th
e 20th day after BMT. The authors studied the presence of ulcers and a
trophy in the mucosa, the nuclear atypia, and inflammatory infiltrate
in the surface epithelium, and the architectural abnormalities, as wel
l as dilation of the glands, and the presence of crypt abscesses, The
authors explored the glandular epithelium for apoptosis and intraepith
elial lymphocytes. They examined the lamina propria for inflammatory i
nfiltrate, especially for so-called ''focal periglandular infiltrate (
FPGI),'' edema, fibrosis, increase of capillary vessels and presence o
f muciphages and clusters of enterochromaffin cells. Cases were groupe
d according to clinical symptoms and histologic diagnosis of A-GVHD in
skin or liver, Group A had 15 asymptomatic cases (control). Group B h
ad 20 cases from asymptomatic patients with a histologically proven A-
GVHD, Group C had 43 cases with gastrointestinal symptoms and histolog
ically proven A-GVHD. Then, the relative frequency of every histologic
feature previously described were compared in the different groups by
means of an univariate analysis. Apoptosis of the glandular epitheliu
m was found in 48 cases, two of them in patients carrying an auto-BMT.
Focal periglandular infiltrate was found in 11 cases that was associa
ted with apoptosis (P < .03, Fisher's exact test), In this series, the
presence of FPGI was linked with finding apoptosis in a given specime
n with a probability of 98%, Glandular architectural changes and dilat
ion were also more frequent in group C than in groups A and B when com
pared in the univariate analysis.