Secondary mucosal colonisation by a carcinoma originating from a distant site is a pattern of metastasis to the intestines and hepatobiliary tract and a mimic of primary neoplasia. Although endometriosis is considered benign, its ability to spread widely underscores its quasi-neoplastic nature. After noting that endometriotic glands can colonise the colonic mucosa along the basement membrane, mimicking metastatic disease, we conducted an intradepartmental review of intestinal specimens showing endometriosis obtained from 2016 to 2023 to characterise and quantify the incidence of this phenomenon.
Material from 38 lower gastrointestinal specimens with a primary or ancillary diagnosis of endometriosis was identified from our surgical pathology database. Slides were reviewed, documenting the extent and micro-anatomic location affected by endometriosis, with a focus on identifying examples showing mucosal colonisation.
The most common site of involvement was the distal colon (23 cases; 11 of rectum, 9 of sigmoid colon and 3 of rectosigmoid) followed by the appendix (N=10), cecum (N=2), small intestine (N=2) and ‘colon not otherwise specified’ (N=1). Mucosal involvement was identified in eight cases (21%), half of which demonstrated seamless colonisation of the epithelium by endometriotic glands. In two of these, the procedure was prompted by the presence of a rectal mass or stricture with concern for malignancy.
Endometriosis occasionally (4/38; 10.5%) colonises colonic epithelium, potentially mimicking a metastasis or intraepithelial neoplasia/dysplasia. Although unusual, this phenomenon was observed in half of specimens from patients with mucosal involvement in whom a mass or stricture suggested malignancy, a potentially misleading pattern of endometriosis.
There is great variability in the assessment and reporting of fat in frozen sections of donor liver biopsies. The Banff Working Group has proposed a novel method and definition for scoring large droplet fat (LDF) in donor liver biopsies. This study compares the Banff method with a simpler Average of Fields (AF) method and evaluates the impact of different LDF definitions.
Three pathologists assessed percentage of LDF (LDF%) in 10 donor liver biopsies using Banff and AF methods, applying the Banff LDF definition (cell distention with a single droplet larger than adjacent hepatocytes). Additionally, LDF% by the AF method was compared using two LDF definitions: Banff definition versus LDF definition 2 (single fat droplet occupying greater than half of a hepatocyte with nuclear displacement).
Intraobserver concordance between the Banff and AF methods was similar for all three pathologists (kappa 0.76–1). Both methods exhibited 70% interobserver concordance, and there was substantial agreement (kappa 0.68) in the LDF% among the three pathologists for both methods. Comparing the two LDF definitions, results were significantly lower with the Banff definition; LDF >50% was observed in four cases with LDF definition 2 but none of the cases with the Banff definition.
There is high interobserver and intraobserver concordance of LDF% between the Banff and AF methods. LDF% determined by the Banff definition was lower than with LDF definition 2, and needs to be validated based on graft outcome before it can be recommended for clinical use.