The immunofluorescent profile of dermatomyositis: a comparative study with lupus erythematosus.
Academic Article
Overview
abstract
We have demonstrated a role for microvascular injury mediated by the membrane attack complex of complement (C5b-9) in the genesis of cutaneous lesions of dermatomyositis (DM) (1). The purpose of this study is to revisit the immunofluorescent (IF) profile of DM, to further investigate the role of C5b-9 in the pathogenesis of cutaneous lesions, and to see if any features of the IF profile reliably distinguish DM from LE. Lesional skin biopsies from 24 patients with clinical findings characteristic of DM were received in formalin and in Michel's transport medium. Conventional light microscopy, and IF studies with antibodies monospecific for IgG, IgA, IgM, C3, fibrin and C5b-9 were performed. The control group comprised biopsies from 31 patients with well-documented LE. A positive lupus band test (LBT) correlated highly with a diagnosis of LE, with a sensitivity of 64.5% and a specificity of 95.6% (p=0.001). The LBT was most sensitive in the setting of DLE and SLE and was least sensitive in the setting of SCLE. The finding of vascular C5b-9 deposition correlated with a diagnosis of DM versus LE (p=0.001) although the false positive rate was 21.4%. The false negative rate was reduced when vascular C5b-9 was seen in the absence of antibodies to Ro, La, or RNP. While a negative LBT correlated with a diagnosis of DM (p=0.001), the specificity was only 64.5%. However, when it was seen in concert with C5b-9 along the DEJ, specificity was increased to 80.6% (p=0.001). The presence of C5b-9 in vessels and along the DEJ in concert with a negative LBT was predictive of DM (p=0.001) with a specificity of 93.5%, sensitivity of 78.3%, a false positive rate of 10% and a false negative rate of 14.7%. The combination of a negative LBT, vascular C5b-9 deposition and negative serology for Ro, La, and RNP was a predictor of DM versus LE with a sensitivity of 90.5%, a specificity of 96.8%, a false positive rate of 5% and a false negative rate of 6.2% (p=0.001). The IF profile of DM in lesional skin comprises a negative LBT, deposition of C5b-9 within vessels and along the DEJ, and variable keratinocyte decoration for IgG and C5b-9. The most statistically powerful predictor of DM is the combination of a negative LBT with vascular C5b-9 deposition and negative serology for antibodies to Ro, La, Sm, and RNP. Demonstration of a negative LBT in all but 1 case of DM suggests that the DEJ is not a primary site for antigen-antibody interaction. We postulate that the aforementioned IF findings reflect humorally mediated injury of endothelium and keratinocytes, effected by C5b-9.
