[Skip to Content]
[Skip to Content Landing]
August 1997

Lichen Planus—Like Histopathologic Characteristics in the Cutaneous Graft-vs-Host ReactionPrognostic Significance Independent of Time Course After Allogeneic Bone Marrow Transplantation

Author Affiliations

From the Departments of Dermatology (Dr Horn and Ms Atkins), Pathology (Dr Horn), Biostatistics (Ms Zahurak), and Oncology (Dr Vogelsang), The Johns Hopkins University School of Medicine, Baltimore, Md; and the Department of Dermatology, Emory University School of Medicine, Atlanta, Ga (Dr Solomon). Dr Horn is now with the Department of Dermatology, University of Arkansas for Medical Sciences, Little Rock.

Arch Dermatol. 1997;133(8):961-965. doi:10.1001/archderm.1997.03890440027003

Background:  The discrimination between acute and chronic graft-vs-host disease (GVHD) after allogeneic bone marrow transplantation (BMT) is important because the treatment regimens and prognosis differ.

Objectives:  To identify whether accepted histopathologic criteria of a graft-vs-host reaction (GVHR) alone or in combination accurately reflect clinical phase of disease, to correlate patterns with clinical outcome, and to identify any concordance between inflammation and epidermal changes of a GVHR.

Design:  Skin biopsy specimens were analyzed according to histologically defined standards.

Settings:  This study was performed in a tertiary care hospital.

Patients:  One hundred seventy-three skin biopsy specimens (10 days before to 1326 days after BMT) from 83 patients undergoing allogeneic BMT for various malignant neoplasms were selected for study. A consecutive 12-month sample was used.

Main Outcome Measures:  The main measures in this study were statistical correlations between histopathologic findings and time after BMT, the outcome of BMT, and the correlations between selected histopathologic criteria.

Results:  Fully evolved histologic features of chronic lichenoid GVHR in the specimens occurred across a wide time range (33-832 days after BMT) and were associated with a 5.6-fold increased risk for death (P=.02) from GVHD. Histologic features of acute GVHR in the specimens also occurred across a wide time range (14-481 days after BMT) and were associated with a 2.2-fold increased risk for death; this finding was not statistically significant (P=.11). Inflammation of the upper dermis was significantly associated with acanthosis and epidermal cell necrosis (P<.001 and P<.001, respectively, for bandlike pattern), confirming the importance of this finding as a criterion for the diagnosis of a GVHR. Blinded evaluation of a subset of specimens for the diagnosis of acute vs chronic GVHR resulted in wide interobserver variation.

Conclusions:  This study demonstrates the following: specific histologic parameters in skin biopsy specimens do not consistently separate acute from chronic GVHD as defined by days after BMT; independent of time course, fully evolved histopathologic characteristics of a lichen planus—like GVHR is associated with a greater likelihood of death from GVHD; and identification of upper dermal inflammation correlates with the epidermal features of GVHR and should be included in the diagnostic scheme.Arch Dermatol. 1997;133:961-965