Crohn's - Biopsy Diagnosis

Professor Najib Y. Haboubi, Trafford Healthcare NHS Trust

Powerpoint File

Colonic biopsies have become one of the most important and the gold standard modality of diagnosing or excluding inflammatory conditions. From the small often single biopsy the clinicians require accurate diagnosis and that although can be achieved on occasions with the help of clinical information it is often that the request form accompanying the biopsies contains little or no clinical information. That limited clinical information is either not available at the time or simply the clinician does not see the relevance of its presence to the pathologist who he/she may regard as "independent witness"! This may put pressure on the pathologist to either produce a meaningless report or sometimes a forced error. If we want to examine the theory which says "what is the reliability of diagnosing colonic mucosal biopsies in benign conditions with incomplete or no clinical information"? For that a literature search was carried out.

An important study by Lessells et al(1) was published in 1994. The aim was to observe the consistency of reporting rectal biopsies and especially to differentiate inflammatory bowel disease from other causes. In their methodology they used 60 rectal biopsies from patients with bloody diarrhoea with no further clinical information and 41 of those patients followed up were inflammatory bowel disease. These were circulated to 11 Consultant General Histopathologists who filled a proforma with the 22 features and 12 diagnoses. The result was the sub-categorisation of inflammatory bowel disease was good in ulcerative colitis but poor in Crohn's disease and that there is a poor recognition of infective colitis.

The second study from Theodossi et al(2) involved this time 10 experienced gastro-intestinal pathologists interpreting colonic and rectal biopsies with no clinical information from 34 patients of ulcerative colitis, 24 patients with Crohn's disease and 18 normal cases and the result was Crohn's disease was often and consistently thought to be ulcerative colitis and also for normal slides the term non-specific colitis was often applied without any consistency. That stimulated Haboubi & Kamal(3) to look into the justifiability of diagnosing chronic non-specific colitis and to assess its clinical relevance. For that rectal biopsies from 35 patients presented with acute diarrhoea which, were diagnosed in a different department initially as chronic non-specific colitis, were reviewed by one of the pathologists who had no access initially to the clinical information. The reviewer classified the biopsies into category 1 normal, category 2 active inflammation with no features of chronicity, category 3 features of chronicity with or without activity and category 4 other. The results showed 13 cases were histologically normal and the other cases varied from patients with solitary ulcer syndrome to active inflammation with no features of chronicity and others with features of chronicity. When the full clinical information became available it was found that none of the 13 cases which were reclassified as category 1 or normal which initially were diagnosed as chronic non-specific colitis needed any clinical intervention and none subsequently developed colonic disease. The conclusion of that study was that "chronic non-specific colitis was used in that department to cover a heterogeneous group of diseases as well as normal biopsies. The pathologists in that department did not differ from the experts in gastrointestinal pathology".

Most of the studies show that reporting inflammatory colonic biopsies without adequate information can lead to erroneous diagnoses. The causes for that are:

1. Limited morphological response of the colonic mucosa to various injurious agents.

2. Histological mimics of inflammatory bowel disease.

3. Overlap of some of the features.

When it comes to the diagnostic accuracy Bentley et al(4) looked into a cohort of biopsies with the aim to:

1. Determining the effect of single versus multiple biopsies on the accuracy of diagnoses.

2. Study the accuracy and the reproducibility of different criteria used in the diagnosis of multiple biopsies by the experts and the non-experts pathologist.

Crohn's Disease

Experts Non-experts

Rectal biopsies 24% Rectal biopsies 12%

Full colonoscopy 64% Full colonoscopy 60%

Ulcerative Colitis

Rectal biopsies 64% Rectal Biopsies 62%

Full colonoscopy 74% Full colonoscopy 72%

That clearly showed that there is not much difference between the experts and the non-experts in full colonoscopic biopsies in ulcerative colitis and Crohn's disease and it also shows that the diagnosis of rectal biopsies in Crohn's disease can give in the best hands a 24% accuracy rate. Therefore it is much better to use full colonoscopic biopsies.

In our department we recommend that;

1. Endoscopy reports should accompany the specimen and the request forms. In these forms there are usually clinical and topographical information of the disease.

2. We issue usually a pattern based initial report with a working diagnosis and differential diagnosis.

3. There is no allowance in our system for the term 'chronic non-specific colitis'.

4. We firmly believe that the final diagnosis of IBD should always be at the clinicopathological conference, which we find very helpful because we feel that the final responsibility of such a diagnosis is a mutual responsibility between the pathologist and the clinician.

In our experience we confirm McBroom & Ramsey's(5) observation that up to 40% of cases reviewed in gastrointestinal meetings the treatment has changed.

References

1. Lessells A M, Swanson J, Beck R et al. Observe variability in the histological reporting of abnormal rectal biopsy specimens. J. Clin. Path. 1994; 47:48-52.

2. Theodossi A, Spiegelhalter D J, Jass J et al. Observe variation and discriminatory value of biopsy features in inflammatory bowel disease. GUT 1994; 35:961-8.

3. Haboubi N Y and Kamal F, Non-Specific Chronic Colitis; is it a justifiable diagnosis? J. Colorectal Disease 2001;3:4:263-265.

4. Bentley E, Jenkins D, Campbell F et al. How could pathologists improve the initial diagnosis of colitis? Evidence from an international workshop, J. Clin. Path, 2002; 5:12:955-961.

5. McBroom M B & Ramsey A D. The clinical pathological meeting. A means of auditing diagnostic performance. American Journal of Surgical Pathology 1993, 17:75-80.