A study of interobserver reproducibility of morphologic lesions of focal segmental glomerulosclerosis

• Published in: Virchows Archiv : an international journal of pathology Vol. 462; no. 2; pp. 229 - 237
• Main Authors: Meehan, Shane M., Chang, Anthony, Gibson, Ian W., Kim, Lisa, Kambham, Neeraja, Laszik, Zoltan
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer-Verlag 01.02.2013; Springer Nature B.V
• Subjects: Diagnostic Imaging - methods; Glomerulosclerosis, Focal Segmental - classification; Glomerulosclerosis, Focal Segmental - diagnosis; Glomerulosclerosis, Focal Segmental - pathology; Humans; Kidney Glomerulus - pathology; Lesions; Medicine; Medicine & Public Health; Microscopy - methods; Observer Variation; Original Article; Pathology; Reproducibility of Results; Focal segmental glomerulosclerosis; Interobserver reproducibility; Columbia criteria
• ISSN: 0945-6317; 1432-2307
• DOI: 10.1007/s00428-012-1355-3

The morphology of focal segmental glomerulosclerosis (FSGS) includes collapsing, cellular, and sclerosing forms. The Columbia Working Classification of FSGS divides these into collapsing (COLL), cellular (CELL), tip lesion (TIP), perihilar (PH), and not otherwise specified (NOS) morphologic forms. This study examined the ability of renal pathologists to classify FSGS using single light microscopic images of glomeruli as a uniform data set. Sixty-one digital images of individual glomeruli with FSGS, stained by periodic acid-Schiff or Jones methenamine silver methods, were classified independently by six specialist renal pathologists. Diagnostic consistency was quantified using the kappa statistic for nominal categories. Agreement for 366 diagnoses by six observers was 75.2 % with a kappa value of 0.676. Six of six observers agreed in 31 of 61 cases (50.8 %) and four or more in 53 cases (86.9 %). Respective kappa values ranged from moderate to good: COLL 0.77, CELL 0.53, TIP 0.76, PH 0.84, and NOS 0.60. Capillary retraction with lobular expansion, hypercellularity, and sclerosis in the same glomerular segments, and the location of segmental lesions were sources of diagnostic inconsistency. The morphologic forms of FSGS defined by the Columbia system are reproducible between observers and have a low probability of confusion between forms. Individual glomeruli may have overlapping features of more than one form of FSGS.