Article Text



Statistics from

Women really ARE from Venus ▸


While the UK tools up to introduce national colorectal cancer screening in 2006, elsewhere debate continues as to which screening test to use. In the USA, flexible sigmoidoscopy has been described as the equivalent of unilateral mammography while in Europe colonoscopic screening is generally regarded as unsuitable for population screening on the grounds of its acceptability and safety.

This second paper from a group working in four US military hospitals reports the results of colonoscopy screening of 1463 women aged 40–79 years and compares their findings with those from their earlier study of men. All were asymptomatic and had negative faecal occult blood tests and no other indications of possible colon cancer (anaemia, unexplained weight loss, rectal bleeding, etc) in the recent past.

A total of 72 women (4.5%) were found to have advanced neoplasia, defined as either cancer (n = 1), a villous adenoma, or an adenoma either >1 cm or showing high grade dysplasia, almost half the prevalence they had found earlier in men (8.6%). Advanced neoplasia was found in the rectum and sigmoid in 1.7% and more proximally in 3.2%. Even if the flexible sigmoidoscopy had reached the splenic flexure and the finding of an adenoma of any size had triggered a full colonoscopy, they estimated that more than 90% of the advanced lesions in the more proximal colon would have been missed by flexible sigmoidoscopy. Overall, the proportion of advanced neoplasia detected by flexible sigmoidoscopy would have been only 35% compared with 66% in men. Indeed, these estimates may be generous as all were sedated and had undergone a full colon prep.

Given the higher proportion of colon cancer that are right sided in women, their findings are perhaps not surprising. Nevertheless, they do raise concerns as to the suitability of flexible sigmoidoscopy for population screening of women at average risk of colorectal cancer.

Endoscopic haemostasis made EASIE ▸


The need for structured endoscopy training is recognised and this may improve the quality of training but little good evidence exists. Learning skills only needed in emergency situations is even more difficult. The authors have developed a biological simulator (compact EASIE) which utilises a pig stomach with vessels sewn in and a pump to simulate bleeding. In this prospective, randomised, blinded trial, they tested its benefits in teaching haemostatic techniques above and beyond standard endoscopy training.

Twenty eight 28 gastroenterology fellows underwent baseline evaluation of their general skills and ability to control ulcer bleeding by injection, electrocoagulation, and application of haemoclips in addition to variceal ligation. Participants evaluated their own performance, and this was also rated by expert tutors using a visual analogue scale. Fourteen Fellows (group A) then continued standard training in their hospitals while the other 14 (group B) did so after an intensive one day training session using the cEASIE simulator. Performance was then reassessed after seven months. At the end, overall performance in group A had not improved (6.0 to 7.0), the time to complete tests was the same, and although the trainees were more adept at ligation they all struggled using clips. In contrast, overall performance in group B significantly improved (4.5 to 8.0). Time to complete procedures was reduced, trainees managed to use clips satisfactorily, and there was a marked improvement in ligation.

The small numbers probably prevented statistical significance being demonstrated in many aspects of this study but the message comes across loud and clear: structured, intensive, yet brief, training improves technical haemostatic skills in upper gastrointestinal bleeding. Perhaps tellingly, at the end of the study fellows in group A overrated their own skill levels compared with their tutors while those who had used the simulator had a more realistic assessment and this may be one of the strengths of training programmes.

Stent instead of stoma? ▸


For patients who are not suitable for resection of a rectal neoplasm, traditional management has been palliation, usually by a defunctioning colostomy. More recently, expandable metal stents have become available for use in the large bowel. This report analyses the use of these stents in patients with malignant rectal obstruction.

Over a seven year period, 521 patients underwent surgery for rectal neoplasia and during the same time 34 patients underwent insertion of a rectal stent. Early failure of the stent occurred in seven patients because of stent migration, pain, or incontinence. Restenting was required in two patients, giving a long term patency of 79%. Because of further symptoms, a colostomy was require in two further patients at three months and nine months. In 18% of patients, therefore, it was necessary to perform palliative surgery because of early complications or long term failure of the stent. Median survival of patients in the stent group (five months) was comparable with that of patients with stage 4 rectal cancer (usually 3–6 months) who are managed with a diverting colostomy.

Self expanding metal stents are clearly useful to avoid a colostomy in selected patients with unresectable rectal cancer and limited life expectancy, although other tumour related symptoms such as pain or bleeding may not be improved by this treatment. However, a small proportion of patients will require surgical palliation because of failure of the stent treatment.

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.