Reply to: Direct Immunofluorescence Findings in Discoid Lupus Erythematosus and Bullous Pemphigoid
We read with great interest the correspondence entitled “Direct immunofluorescence findings in discoid lupus erythematosus and bullous pemphigoid” by Lehman et al,1 which was a reply to our article entitled “Comparative study of direct immunofluorescence in discoid lupus erythematosus and bullous pemphigoid” by Ohata et al.2 We agree that the pattern of immunoreactant deposition is important in direct immunofluorescence (DIF) interpretation. This was mentioned as “The pattern of the deposition of immunoreactants was as continuous granules or closely spaced fibrils in DLE (discoid lupus erythematosus), whereas as linear in BP (bullous pemphigoid). Closer examination of DIF images of BP showed true linear with or without n-serrated pattern.” in the results section of our original article.2 The pattern of immunoreactant deposition in DIF may differentiate DLE from BP, but this pattern may sometimes be difficult to evaluate depending on the condition of the skin section, the DIF system including dilution of immunoreactants, and the experience of the person performing the tests. Recently, Lemcke et al3 reported the difference between DIF staining of BP cases performed manually and that performed with EUROTide/EUROPath in an automated version. Although their study did not show the staining pattern alteration, the pattern may be arbitrarily judged in some ambiguous cases.
In our other study for DIF in 104 BP cases, IgG and C3 deposition on epidermal basement membrane zone (BMZ) was seen in 86 (82.7%) and 100 (96.2%) patients, respectively,4 and we see strong linear deposition of IgG and C3 in approximately 80% and 90%, respectively (unpublished observations). Thus, approximately 66% and 86% of patients with BP showed strong IgG and C3 deposition on BMZ in DIF, respectively. We do not agree that “the vast majority of (BP) cases show strong linear deposition of C3, typically with strong linear IgG, along the BMZ.”
The number of positive immunoreactants is easy to assess and does not easily vary under different conditions. Moreover, even in some cases with ambiguous immunoreactant staining patterns, the number of positive reactants is not judged arbitrarily. We believe that the number of positive reactants is more easily and objectively evaluated than DIF staining patterns. Thus, it is important to be aware of the mean number of positive immunoreactants and how likely each immunoreactant is to be positive in both DLE and BP.
Lehman et al remarked that “the findings in Ohata et al's study are insufficiently compelling to suggest that follicular BMZ DIF findings would trump changes at the epidermal BMZ.” without specific reasons. We mentioned that “In discriminating DLE from BP, the number of positive immunoreactants at the follicular BMZ is likely to be more contributory than that at the epidermal BMZ.” in the discussion section of our original article.2 This is because the difference between the mean numbers of positive immunoreactants at the follicular BMZ in DLE and BP (3.50 and 1.50, respectively) was larger than that at the epidermal BMZ in DLE and BP (3.22 and 2.48, respectively).2 We admit the limitations of our study with regard to the small number of cases studied; however, we do not know that this is the reason for Lehman et al's assertion.