Article Text


Endoscopy III
PWE-220 How many biopsies at colonoscopy are required to confirm the diagnosis histologically in suspected colorectal cancer?
  1. V Sehgal,
  2. B Krishnan,
  3. K Besherdas
  1. Department of Gastroenterology, Chase Farm Hospital, London, UK


Introduction Colorectal cancer (CRC) is the third most commonly diagnosed cancer worldwide. The prognosis depends to a large extent on the stage of disease at the time of diagnosis such that the early investigation of relevant signs and symptoms is encouraged. At present the gold-standard method for the investigation of CRC is endoscopy as it permits direct visualisation and biopsy of the lesion in question. There is little evidence on the number of biopsies needed to be obtained if a CRC is suspected macroscopically at the time of endoscopy but there is a suggestion that a minimum of six biopsies should be taken to increase the yield of an early positive diagnosis.1 Patients in whom the endoscopy biopsy is non-diagnostic and cancer suspected, a repeat colonoscopy to obtain additional tissue sample is often recommended. The aims of this study were to assess whether the number of biopsies taken of suspected cancers at the time of endoscopy was proportional to the rates of positive diagnoses being made while reducing the need for a repeat endoscopic procedure to confirm or exclude cancer.

Methods A retrospective analysis of all patients with suspected CRC upon endoscopy at Chase Farm Hospital over a 1-year period (2009–2010) was performed. Data were obtained from endoscopy and histopathology reports. Statistical analysis was performed using the Student t test and Fisher's test using SPSS V.20.0 (p<0.05).

Results 80 patients (37 male), median age 71.5 years were investigated over the audit period. Histology revealed adenocarcinoma (ACA) 52 (65%), high-grade dysplasia (HGD) 20 (25%), normal 5 (6.3%) and other 3 (3.7%). The median number of biopsies taken of suspected cancers for the whole cohort was 6 (1–12) and 44 (55%) had six or more biopsies taken. 16 (20%) patients required at least one repeat endoscopic procedure for diagnostic purposes (initial histology was HGD in 10, 62.5%, of these patients) and histology upon repeat endoscopy demonstrated ACA in 14 (87.5%) of these patients. Patients requiring repeat endoscopy had significantly fewer biopsies (median 4.5) taken at the time of initial endoscopy compared to those who did not (median 5.5), t (76)=2.54, p<0.05 using the Student t test. Patients requiring a repeat endoscopic procedure were more likely to have had less than six biopsies taken initially (11, 68.7%) compared to patients who had six or more biopsies taken (5, 31.3%) although the difference was not significant (p=0.05).

Conclusion Patients who have fewer biopsies are more likely to require repeat endoscopy for histological confirmation with subsequent delays in diagnosis. We recommend obtaining a minimum of six endoscopic biopsies in patients with suspected macroscopic CRC to confirm the diagnosis histologically and prevent a repeat endoscopy.

Competing interests None declared.

Statistics from

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.