Article Text

Download PDFPDF

Crypt dysplasia on Barrett's oesophagus
  1. Francisco Baldaque-Silva1,
  2. Margarida Marques1,
  3. Joanne Lopes2,
  4. Fatima Carneiro3,
  5. Michael Vieth4,
  6. Guilherme Macedo1
  1. 1Gastroenterology Department, Centro Hospitalar São João, Porto, Portugal
  2. 2Pathology Department, Centro Hospitalar São João, Porto, Portugal
  3. 3Institute of Molecular Pathology and Immunology of the University of Porto (IPATIMUP), Porto, Portugal
  4. 4Institute of Pathology, Klinikum Bayreuth, Bayreuth, Germany
  1. Correspondence to Dr Francisco Baldaque-Silva, Gastroenterology Department, Centro Hospitalar São João, Portugal, Alameda Prof. Hernâni Monteiro 34, Porto 4200–319, Portugal; fbaldaquesilva{at}

Statistics from

In a study by Sharma et al1 recently published in Gut, high-definition endoscopy and narrow band imaging (NBI) enabled detection of more areas of dysplasia than Seattle protocol. In addition, regular mucosa did not harbour high-grade dysplasia (HGD)/cancer, leading to the suggestion that biopsies could be avoided in those areas.

However, authors stated in the discussion that the routine use of NBI targeted biopsies for detection of HGD/cancer should not be recommended. To highlight this point, we would like report a case of a C10M12 Barrett's oesophagus (BO) patient with two previous macroscopic lesions with HGD and adenocarcinoma resected using multiband mucosectomy (figure 1). In a careful follow-up using the same endoscope as in the reported study,1 no macroscopic lesions were detected and biopsies were performed according to Seattle protocol. The pathological analysis in consensus by two expert gastrointestinal pathologists revealed, in areas not related to the previous resections, the presence of low-grade and high-grade multifocal crypt dysplasia (CD) with no dysplasia on the epithelial surface. Due to fact that radiofrequency ablation is unavailable in our country, due to the relevant morbidity of oesophagectomy and our experience on endoscopic submucosal dissection (ESD), a decision for endoscopic resection was made. An ESD of the 12 cm BO was performed and en bloc circular specimen was obtained. The histopathological analysis with hematoxylin-eosin, ki67 and p53 immunohistochemistry, in consensus by three expert gastrointestinal pathologists, confirmed the presence of multifocal low-grade and high-grade CD, with no dysplasia at the epithelial surface above the basal changes (figure 2).

Figure 1

Histopathology obtained from endoscopic mucosal resection. (A) Barrett’s metaplasia without neoplasia and numerous goblet cells (H&E 40×). (B) Low-grade dysplasia in Barrett’s with stratified hyperchromatic nuclei and preserved architecture with reduced number of goblet cells (H&E 100×). (C) High-grade dysplasia in Barrett’s with reduced number of goblet cells, hyperchromatic, pleomorphic nuclei and little distorted foveolar architecture (H&E 100×). (D) Barrett’s adenocarcinoma with hyperchromatic, pleomorphic nuclei within a distorted foveolar architecture with glandular fusion and lateral expansion of the glands (H&E 100×). Access the article online to view this figure in colour.

Figure 2

Histopathology obtained from endoscopic submucosal dissection. (A) Barrett’s epithelium with basal low-grade dysplasia with reduced number of goblet cells and maturation towards the surface (H&E 40×). (B) Nuclear immunoexpression of Ki67 within the basal-like dysplasia foci (Immunohistochemistry (IHC) 100×). (C) Nuclear immunoexpression of p53 within the same area (IHC 100×). Access the article online to view this figure in colour.

Evidence suggests that dysplasia begins in the crypt bases and progresses, with time, to involve the full length of the crypts and surface epithelium.2 CD, also called basal CD-like atypia, is characterised by the presence of dysplasia limited to the bases of the crypts, without involvement of the surface epithelium. It was demonstrated that CD harbours proliferative and molecular abnormalities present in conventional dysplasia.3 This condition is rarely addressed in the literature but can be found in up to 7% of BO cases and is associated with increased risk for neoplastic progression. In fact, 87% of these patients have prior or concomitant dysplasia or adenocarcinoma in the remaining BO.3 On the other hand, in such cases, it can never be excluded that adjacent neoplasia grows underneath the surface mimicking such crypt foci.

It is important for gastroenterologists to be aware of the possibility of the presence of dysplasia-like changes limited to the crypts not suitable for detection even with advanced endoscopic tools. Conventional biopsies are required for its detection and in case biopsies show such dysplastic crypt foci, an adjacent neoplastic lesion should be excluded. Therefore, natural history and clinical significance of CD require further investigation.


View Abstract


  • Contributors FB-S and MM: Endoscopic procedures, conception and design, analysis and interpretation of data, drafting of the article and final approval. JL, FC and MV: analysis and interpretation of data, drafting of the article and final approval. GM: drafting of the article and final approval.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Comissao de Etica do Hospital de Sao Joao, Porto, Portugal.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles