Reliability of histopathological salivary gland biopsy assessment in Sjögren's syndrome: a multicentre cohort study

• Published in: Rheumatology 2015-06, Vol.54 (6), p.1056-1064
• Main Authors: Costa, Sebastian, Quintin-Roué, Isabelle, Lesourd, Agnès, Jousse-Joulin, Sandrine, Berthelot, Jean-Marie, Hachulla, Eric, Hatron, Pierre-Yves, Goeb, Vincent, Vittecoq, Olivier, Pers, Jacques Olivier, Marcorelles, Pascale, Nowak, Emmanuel, Saraux, Alain, Devauchelle-Pensec, Valérie
• Format: Article
• Language: English
• Subjects: Biopsy - statistics & numerical data; Humans; Immunology; Life Sciences; Observer Variation; Reproducibility of Results; Salivary Glands, Minor - pathology; Sjogren's Syndrome - pathology
• ISSN: 1462-0324; 1462-0332; 1460-2172
• DOI: 10.1093/rheumatology/keu453

The aim of this study was to assess intraobserver and interobserver reliability of minor salivary gland biopsy (MSGB) in SS. All MSGBs available from the Tolerance and Efficacy of Rituximab in Primary Sjögren's Syndrome (TEARS) study were subjected to a standardized blinded assessment by a single specifically trained pathologist twice at a 2 month interval; both the Chisholm-Mason (CM) grade and the focus score (FS) were determined. Baseline histopathological reports by local pathologists at each study centre were compared with the first standardized blinded assessment. Agreement was assessed for the dichotomized FS (dFS) and dichotomized CM (dCM) grade, as well as for nine other histopathological features. Eighty-nine MSGBs were studied. Intraobserver κ values were 1 for dFS, 0.80 for dCM, 0.67 for germinal centre-like structures, 0.44 for fibrosis and 0.29 for confluent foci. Most of the local histopathological reports based their diagnosis on the CM grade, although the FS was often reported or the data needed to determine it were provided. Interobserver agreement κ values were 0.71 for dFS, 0.64 for dCM, 0.46 for focal lymphocytic sialadenitis, 0.42 for non-specific chronic inflammation and 0.16 for fibrosis. Although FS reliability was good, disparities were noted in the assessment methods used by local pathologists. The protocol for FS determination was not followed routinely, with the result that the FS was often overestimated. Germinal centre-like structures, which predict lymphoma, showed good reliability but were inconsistently reported.