The histological diagnosis of Inflammatory Bowel Disease (IBD) is not an easy task for a pathologist. In the modern era, personal pathology experience alone is insufficient to make a diagnosis of IBD. The information that a pathologist must know in order to evaluate IBD samples appropriately and to make a diagnosis is diverse, and for the most part should be provided by the gastroenterologists, surgeons or clinicians responsible for the care of patients. Even the most experienced pathologist cannot report a case without knowing the clinical background of the patient. Obviously, this should be the standard for all samples, not only for IBD.
Samples that arrive in the pathology lab for IBD assessment should always be accompanied by a request form which includes the patient’s details, such as date of birth and family history, and all the relevant clinical information: symptoms, other comorbidities if present, microbiological and serological tests if performed and drug history. It is important that not only IBD-related therapies but also all other drugs that the patient is taking are stated, as many medications may cause mucosal changes that can mimic IBD on histology. If a patient already has an established diagnosis of IBD, it is important to state whether the diagnosis is of Crohn’s Disease (CD) or Ulcerative Colitis (UC), the duration of the disease, whether the patient has had previous biopsies and whether these are available locally to review. These recommendations should also be applied for gross specimens, although for surgical resections, radiological findings additionally need to be provided.
For biopsy samples the endoscopy report should be included and should indicate the macroscopic appearance of the mucosa, including an endoscopic score of the inflammation. The extent of disease is mainly determined endoscopically. In cases of UC it is relatively easy to define the extent of disease, and this is also a useful parameter to predict the risk of dysplasia. For CD, however, it is more difficult to evaluate extent of disease on the basis of endoscopy alone. The endoscopy report should inform the pathologist how many biopsies were taken and the anatomical sites where they were obtained. Ideally, for a patient with IBD or suspected IBD, an ileocolonoscopy should be performed and at least two biopsies taken from each site, namely the terminal ileum, right colon, transverse colon, left colon, sigmoid and rectum.
Oesophagogastroduodenoscopy (OGDS) in suspected IBD is recommended in the paediatric population, in which differentiation between UC and CD can be difficult. In adult IBD, there are no specific recommendations regarding OGDS.
Biopsies should be collected from the most inflamed and affected areas, from the edges of ulcers if present and from segmental macroscopically healthy areas if present. An adequate number of biopsies is important to allow differentiation between the various types of colitis and inflammatory conditions. There is evidence that the quality and accuracy of the diagnosis increases with the examination of multiple samples as IBD histological features can be focal in the mucosa and the distribution of changes can vary within the same sites or between different sites. In addition, the microscopic examination can reveal inflammatory patterns not seen during endoscopy.
As regards surveillance for colorectal cancer in patients with long-standing IBD, ECCO Guidelines advise the performance, if possible, of high-definition chromo-endoscopy with targeted biopsies as this increases the detection of dysplastic areas and flat lesions. If chromo-endoscopy is not available or cannot be performed, it is recommended to take four biopsies from every 10 cm of the entire colon and also targeted biopsies from every visible lesion. For the evaluation of pouchitis in patients with ileo-anal anastomosis, multiple biopsies are required, which should preferably be taken from the posterior and the anterior wall, from the area above the pouch and from the rectal cuff.
Once samples have been collected, they should be immediately placed in a formaldehyde-based fixative or another equivalent solution and labelled appropriately. An essential step is that biopsies from different sites are put in different containers. The orientation of the biopsies is also crucial for the accuracy of the diagnosis. For this reason, pathology lab staff should be trained to embed the samples correctly. There are other methods to orientate the biopsies, for instance mounting a biopsy series on cellulose acetate strips at the time of collection in the endoscopy room. This cost-effective method allows the carriage of fewer pots to the pathology lab. However, it requires training of the endoscopy staff and has some limitations, in that it may limit the number of biopsies that can be taken from each site and the samples can detach from the paper during the transfer of vials.
Once the pathologist knows the clinical picture, he or she should be able to interpret the biopsies correctly. Slides are routinely stained with haematoxylin and eosin; special staining or immunohistochemistry (IIC) is not recommended for the diagnosis of IBD. However, IIC may be useful, for example, to exclude cytomegalovirus infection or Epstein-Barr virus infection.
The pathology report must be clear and understandable for clinicians. The pathologist should reach a comprehensive diagnosis and the report should include all histological features that have led to the diagnosis. Long-standing IBD should be interpreted cautiously as many factors, including treatments, can markedly change the aspect of the disease. Changing an established diagnosis should only be done after careful evaluation of the clinical picture and requires the evaluation of previous histology. Previous biopsies are not always available locally and sometimes it is necessary to request old samples from another pathology lab.
Last but not least, the pathologist needs to talk to their clinicians and vice versa. A dialogue is necessary, and difficult cases require a discussion in a multidisciplinary team (MDT) setting. Also, cases in which dysplasia is detected are always worthy of an MDT discussion.
These recommendations may sound obvious to many readers; however, the majority of pathology centres still receive inadequate samples for the diagnosis of IBD. Many biopsies arrive with incomplete or no clinical details and, even worse, biopsies from different sites may arrive within the same vial. When this happens, the pathologist should contact the clinician to collect information; this is time consuming, leading to delays in diagnosis. Plus, biopsies may be interpreted wrongly when the pathologist ignores the clinical background.
It cannot be stressed enough that pathologists do not have a crystal ball and that they require accurate information to interpret the slides. Behind every slide there is a patient who needs the best care. Please do not keep all your clinical information for yourselves and spoil your pathologist!