King's College London

Research portal

The optimal oral biopsy site for diagnosis of mucous membrane pemphigoid and pemphigus vulgaris

Research output: Contribution to journalArticle

Barbara Carey, Sandeep Joshi, Abedalla Abdelghani, John Mee, Manoharan Andiappan, Jane Setterfield

Original languageEnglish
JournalBritish Journal of Dermatology
Early online date15 Jul 2019
Publication statusE-pub ahead of print - 15 Jul 2019

King's Authors



Accepted ‘standard practice’ for the diagnosis of immunobullous disease is a perilesional sample for direct immunofluorescence (DIF).


To compare diagnostic outcomes of a normal buccal punch biopsy (NBPB) with a perilesional biopsy (PLB) for mucous membrane pemphigoid (MMP) and pemphigus vulgaris (PV).


A retrospective analysis of 251 DIF‐positive patients with MMP and 77 DIF‐positive patients with PV was undertaken. Parameters analysed included the intraoral sites of involvement and histopathological, DIF and indirect immunofluorescence (IIF) findings.


For MMP, PLB was positive in 134 of 143 (93·7%) samples, compared with 129 of 144 (89·6%) by NBPB. The diagnostic sensitivities for PLB (81%, 39 of 48) and NBPB (77%, 37 of 48) among 48 patients who underwent both techniques were not significantly different (P = 0·62). In gingival‐only MMP, PLB was positive in 63 of 69 (91%) and NBPB was positive in 63 of 75 (84%). For multisite MMP, PLB was positive in 71 of 74 (96%) and NBPB was positive in 66 of 69 (96%). In gingival‐only MMP, biopsies from reflected alveolar mucosa in 17 consecutive patients were positive in 17 of 17 cases (100%). For PV, PLB was positive in 42 of 43 (98%), compared with 42 of 42 (100%) by NBPB. Histopathology was diagnostic in 93 of 134 (69·4%) cases of MMP and 38 of 41 (93%) cases of PV. IIF was positive in 126 of 197 (64·0%) MMP and 68 of 74 (92%) PV patient sera.


In the largest series of combined oral DIF results in patients with MMP and PV, we have shown that NBPB is equivalent to PLB for the diagnosis of PV and multisite MMP, and is more sensitive than both histology and IIF.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454