Efficacy of Triaging Direct Immunofluorescence in Intraepidermal Bullous Dermatoses.

BACKGROUND: Direct immunofluorescence (DIF) is considered pivotal in diagnosing autoimmune blistering diseases. Our goal was to examine the necessity of DIF in intraepidermal bullous cases, of which pemphigus vulgaris (PV) is the prototype. METHODS: Sixty-six cases from 2010 to 2014 submitted for DIF with an intraepidermal blistering disease listed in the differential diagnosis were reviewed by 2 board-certified dermatopathologists to see if they would order DIF based on routine histologic findings. If either pathologist requested DIF, it was considered required. RESULTS: DIF was "required" in 29% (19/66) (94% intraobserver concordance) and was positive in 16% (3/19) of those "required," leading to a diagnosis of PV (2/3) or pemphigus foliaceus [(PF) 1/3]. DIF was "not required" in 71% (47/66). Of these, 37/47 had negative/nonspecific DIF findings (79%). Of the 10 DIF+ cases, 8 were accurately diagnosed as PV based solely on the hematoxylin and eosin (H&E) findings. Three of 47 "not required" cases were misdiagnosed. Two (2/10 DIF+) were called "spongiotic dermatitis" by H&E interpretation yet had DIF consistent with PF. One case of acantholytic pityriasis rubra pilaris was diagnosed as PV on H&E, an error that may have been avoided with real-time clinical correlation. CONCLUSIONS: H&E diagnosis was 80% sensitive and 97% specific for intraepidermal blistering diseases. Positive predictive value was 89%; negative predictive value was 95%. H&E triaging could significantly reduce the need for DIF, especially in PV. If PF is in the differential diagnosis, DIF is necessary. In certain settings, triaging by H&E can obviate the need for DIF, resulting in significant savings.

Evaluation and comparison of staining patterns of factor XIIIa (AC-1A1), adipophilin and GATA3 in sebaceous neoplasia.

BACKGROUND: Reliable nuclear immunohistochemical stains for sebaceous neoplasms have not been readily available. Positive nuclear staining has been reported for GATA3 and factor XIIIa (AC-1A1). We sought to determine the diagnostic utility of these nuclear stains by comparing their staining pattern to adipophilin, a consistently positive cytoplasmic stain. METHODS: Cases with the diagnosis of sebaceous hyperplasia, sebaceous adenoma, sebaceous epithelioma/sebaceoma, sebaceous carcinoma, and nonsebaceous neoplasms (basal cell carcinoma and squamous cell carcinoma) were examined. Intensity and extent of staining of the basal cells and mature sebocytes were evaluated for each stain. RESULTS: Factor XIIIa (AC-1A1) was 87.3% sensitive and 95.1% specific for all sebaceous neoplasms sand showed high inter-observer reliability. Adipophilin was 83.2% sensitive and 87.8% specific. GATA3 was the least sensitive (80.9%) and specific (75.6%) marker. When factor XIIIa was compared against composite staining of all three markers its staining was still uniquely significant (P = .0210). CONCLUSION: Factor XIIIa (AC-1A1) is a sensitive and specific nuclear marker for sebaceous differentiation. Its diagnostic utility exceeds that of adipophilin. Factor XIIIa should be included in the expanding group of immunohistochemical and special stains which can be utilized to aid in the diagnosis of sebaceous neoplasms.