Article Text


High magnification chromoscopic colonoscopy as a screening tool in acromegaly
  1. D P Hurlstone,
  2. S S Cross,
  3. A J Lobo,
  4. D S Sanders
  1. Halamshire Hospital, Sheffield, UK
  1. Correspondence to:
    Dr D P Hurlstone
    17 Alexandra Gardens, Lyndhurst Rd, Nether Edge, Sheffield S11 9DQ, UK; p.hurlstone{at}

Statistics from

We read with great interest the paper by Jenkins et al (Gut 2002;51:v13–14) regarding screening guidelines for colorectal cancer (CRC) and polyps in patients with acromegaly and the subsequent discussion by Renehan addressing screening inconsistencies compared with other high risk groups.1,2

The optimal colorectal screening modality and frequency in this group however requires clarification. Colonoscopy in this patient group is technically demanding and often complicated by inadequate bowel preparation.3 However, despite current controversies regarding true CRC risk categorisation in acromegaly, previous data from the largest published series showed a trend for adenoma and carcinoma formation in the right hemi colon.4 This is an important observation for many reasons.

Flat adenomas and carcinomas can be difficult to detect by conventional colonoscopy alone, often presenting as subtle mucosal erythema, mucosal pallor, fold convergence, interruption of innominate grooves, air induced deformation, or loss of vascular net pattern.5 The neoplastic risk for this morphologically distinct group has additionally been shown by many authors to be higher when compared with exophytic polypoid lesions and exhibit a propensity for the right colon.6–9 De novo neoplastic lesions and “minute” colorectal cancers are also associated with an increased risk of lymph node metastasis due to early invasion of the submucosal layer.10 Tada et al found extensive submucosal invasion in a cohort of flat colorectal neoplasms,11 with Shimoda’s series corroborating these data with submucosal invasion demonstrable in 69% of flat carcinomas compared with only 35% of sessile and broad based polypoid carcinomas.12

Morphologically flat and depressed lesions are also known to occur in chronic ulcerative colitis13 where the need for CRC screening with total colonoscopy and now adjunctive chromoscopy is adopted by many centres. Failure to detect such lesions may in part account for those cases of CRC which occurred in Winawer’s study, despite clearance of all exophytic polyps, and thus stresses the requirement for accurate diagnosis and definitive treatment of these high risk lesions.14

Given the lack of standardised and uniform reporting regarding the morphology of colorectal lesions in many of the existing prevalence studies of adenomas and CRC in acromegaly however, at present we can only hypothesis that the high incidence of right hemi colonic neoplasia may be an indicator of an alternative morphologically distinct lesion such as the flat adenoma and carcinoma with a trend towards a de novo pathogenic sequence.

In our prospective study, 38 patients with acromegaly underwent total colonoscopy by a single endoscopist using the Olympus C240Z magnifying colonoscope. Preparation was with 4 litres of Kleanprep 24 hours prior to the procedure. Pancolonic chromoscopy using 0.5% indigo carmine sprayed onto the colonic mucosa using an Olympus diffusion catheter (CS12890) was applied. Identified lesions were morphologically grouped according to the Japanese Research Society Classification (JRSC).15,16 A flat lesion was defined as mucosal change with a flat or rounded surface combined with a height of less than half the diameter of the lesion.17 High magnification views of all suspected lesions were then obtained and reported according to the modified Kudo criteria.18 Tissue sampling was performed with cold biopsy or endoscopic mucosal resection following exclusion of a Kudo type V(n)/IIIs invasive crypt pattern which suggests deep submucosal invasion. Mean intubation and extubation times were recorded. Neoplastic change was classified according to the Vienna criteria.19

Caecal intubation was achieved in 37/38 (97%) patients with 36/38 (94%) receiving confirmatory terminal ileal biopsies. Males represented 14/37 (37% of the cohort, mean age 64 years (range 40–75)). The mean duration of intubation to the caecum was 16.5 minutes (range 3–31) and extubation (excluding interventional procedures) was 35 minutes (range 20–55). There were no complications.

A total of 28 lesions were identified in 15 patients. Twenty two hyperplastic lesions were identified (79%) of which 17 (77%) were flat (JRSC II). Twenty (91%) were located in the left colon and rectum. Of the five adenomas identified, four (80%) were present in the right colon with 4/5 (80%) being of JRSC II morphology. A single adenoma with high grade dysplasia was present in the right colon and was flat with a small area of central depression. No invasive carcinomas were diagnosed. Results are summarised in table 1.

Although the numbers entering this study are small, our results show a clear prevalence for JRSC class II lesions in this select patient group. Although only one adenoma with high grade dysplasia was detected, it was small (5 mm) and was not identified prior to chromoscopic and magnification enhancement, and therefore carries major clinical connotations.

We suggest that further large prospective studies are required to establish the true prevalence of flat and depressed colorectal lesions in acromegaly so that the optimal screening modality and frequency can finally be established. Furthermore, colonoscopists require training in chromoscopic techniques if a higher endoscopically “treatable” lesional frequency is to be detected at a screening level, so as to avoid the high apparent incidence of interval neoplasms.

Table 1

Lesion demographics


View Abstract

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.