Table 1.
The patient’s characteristics.
Table 2.
The histologic grading systems for intestinal acute graft-versus-host disease used in this study.
Fig 1.
Histology of acute graft-versus-host disease of the colon.
Crypt cell apoptosis: Apoptotic bodies consist of cell debris visible as dark particle in a lighter background (arrows) (Lerner grade 1, Sales grade 0, Melson grade 1. HE) (A). Crypt destruction: Degeneration of crypt epithelial cells, the lumen is filled with cell detritus and a few granulocytes (black arrows); a neighboring crypt exhibit an apoptotic body (white arrow) (Lerner grade 2, Sales grade 1, Melson grade 1. HE) (B). Individual crypt loss: Gap between preserved crypts the size of one crypt (arrow) (Sale grade 2, Melson grade 2), there is no crypt cell apoptosis or ongoing crypt destruction visible (therefore Lerner grade 0. HE) (C). Contiguous crypt loss: Wide gaps between residual crypts, Paneth cell metaplasia in the remaining crypts indicates chronic mucosa injury (Lerner grade 0 (because there is no ongoing destruction of crypt epithelial cells), Sales grade 3, Melson grade 3. HE) (D). Mucosa denudation: Loss of the superficial epithelial layer and fibrin deposit (arrow) (Lerner grade 3) and contiguous crypt loss (Sales grade 3, Melson grade 3. HE) (E). Diffuse mucosa denudation: Exhaustive mucosa destruction with near complete crypt loss (Lerner grade 4, Sales grade 4, Melson grade 4. HE) (F).
Fig 2.
Correlation between the histological maximum Lerner grade of the samples obtained in the biopsy series with each of the Glucksberg grade of intestinal graft-versus-host disease (GvHD).
Diameters of the dots correlate with number of cases linearly (A). Endoscopic findings in patients with acute GvHD (B).
Fig 3.
Endoscopic imaging of colonic manifestation of GvHD.
Mild colitis with small erosions and very mild localized edema (A, B). Severe changes with generalized, pronounced edema, fibrinous exudation and multifocal granular ulceration (C, D).
Fig 4.
Comparison between the application of steroid therapy and the maximum histologic Lerner grade of a biopsy series.
Numbers in bar charts indicate the count of cases for each group.
Fig 5.
Comparison between the response to steroid therapy and the minimum histologic Lerner grade of a biopsy series.
Numbers in bar charts indicate the count of cases for each group (only patients having received steroid therapy are included (n = 128)).
Fig 6.
Univariate Cox regression analysis of non-relapse mortality and overall survival on histological grades.
Fig 7.
Univariate Cox regression analysis of non-relapse mortality and overall survival on thrombocyte count, Glucksberg grade of the graft-versus-host disease overall, and Glucksberg grade of the intestine.
Fig 8.
Multivariate Cox regression analysis of non-relapse mortality and overall survival on covariates selected using the Akaike information criterion backward stepwise variable method.