Direct Immunofluorescence in Dermatomyositis and Lupus Erythematosus: Cross-Sectional Diagnostic Accuracy and Correlation With Serologies and Other Features of Systemic Disease

Abstract

This is a retrospective cross-sectional diagnostic test accuracy study of direct immunofluorescence (DIF) performed on a group of potential lupus erythematosus (LE)/dermatomyositis (DM) skin biopsies from 2015 to 2020 at a large, academic medical center. For purposes of this study, which was focused primarily on detection of LE/DM-related interface dermatitis, DIF was considered positive for a LE/DM pattern if it showed granular deposition of immunoglobulin G, with or without C3, at the basement membrane zone on the final pathology report. Blinded clinicopathologic correlation was the reference standard. One thousand fifty-eight sequential pairs of skin biopsies obtained from adults and submitted for both DIF and routine histology were screened. Cases were excluded if histopathology did not demonstrate features that could be consistent with LE/DM (broadly predefined as interface dermatitis and/or neutrophilic dermatosis), resulting in 254 cases of possible LE/DM eruptions for further analysis. Sensitivity of DIF with an immunoglobulin G granular pattern at the basement membrane zone was 71.4% (confidence interval 55.4%-84.3%) in LE-related and 59.3% (confidence interval 38.8%-77.6%) in DM-related eruptions. No statistically significant difference was found in sensitivities of DIF between LE and DM ( P = 0.3). DIF positivity did not correlate with available key demographic, clinical, and serologic features.