A Reappraisal of the Histopathologic Criteria for the Diagnosis of Cutaneous Allogeneic Acute Graft-vs-Host Disease

M Daniela, F Alessandro, P Nicola… - American journal of …, 1999 - academic.oup.com
M Daniela, F Alessandro, P Nicola, L Daniele, B Alberto, S Marco
American journal of clinical pathology, 1999academic.oup.com
To determine the validity of the Lerner grading system and review the histopathologic
findings of cutaneous acute graft-vs-host disease (aGVHD), 78 skin biopsy specimens from
49 bone marrow transplant recipients were evaluated. Histopathologic sections were
independently reviewed twice by 3 pathologists and classified (Lerner system), without
knowledge of the patients' clinical evolution. Intraobserver agreement in grading aGVHD
was substantial to almost perfect. Interobserver agreement between pairs of observers was …
Abstract
To determine the validity of the Lerner grading system and review the histopathologic findings of cutaneous acute graft-vs-host disease (aGVHD), 78 skin biopsy specimens from 49 bone marrow transplant recipients were evaluated. Histopathologic sections were independently reviewed twice by 3 pathologists and classified (Lerner system), without knowledge of the patients’ clinical evolution. Intraobserver agreement in grading aGVHD was substantial to almost perfect. Interobserver agreement between pairs of observers was moderate to substantial onfirst review and substantial on second review. Overall, wefound an almost perfect agreement in diagnosing Lerner grade III, whereas areas of disagreement occurred with Lerner grades 0,1, and II. Histopathologically, specimens of patients who developed aGVHD (positive) showed significantly higher frequency ofepidermal atrophy, spongiosis, diffuse basai vacuolization, more than 3 single necrotic keratinocytes per high-powerfield, satellitosis, inflammatory infiltrate, with a predominantly lichenoid pattern, lymphocytic exocytosis, and dermal melanophages. When considering skin samples classified as grade I and II, wefound statistically significant differences between positive and negative cases onlyfor the presence of inflammatory infiltrate, lymphocytic exocytosis, and satellitosis. Lerner grading is reproducible, although lesser agreement occurred when evaluating grades I and II, and the Lerner grading system should be revised by including the estimate of the inflammatory infiltrate as an additional criterionfor grade II.
Oxford University Press