Following 6 months of initial discussions and successful fund-raising the UK National Barrett's Oesophagus Registry (UKBOR) was set up in June 1996 as a joint initiative between the European Cancer Prevention Organisation (ECP) and the Oesophageal Section of the British Society of Gastroenterology (BSG). A Scientific Advisory Committee was formed composed of gastroenterologists, histopathologists, molecular biologists, epidemiologists and surgeons.
The aims of the Registry are to establish a national database of diagnosed cases of Barrett's oesophagus (BO) from all parts of the United Kingdom in order to provide information on:
- •Prevalence of diagnosed cases, regional variations and variations with time.
- •Natural history and influence of medical, endoscopic and surgical treatment.
- •Incidence of adenocarcinoma (AC) in Barrett's oesophagus (BO).
- •Rate of progression of BO to AC.
- •Factors influencing rate of progression of BO to AC.
Also to provide a central resource for histopathological confirmation of high-grade dysplasia, for molecular genetic studies and for all publications on BO, which may be accessed by BSG and ECP members, and to provide a database and coordinating infrastructure for prospective studies in BO.
Recording of data
The registration form has already been published (Caygill et al., 1998; see Appendix). Hospitals are asked to send data to the Registry either on this form or as a computer printout giving the same information. This form contains data that enable us to calculate M:F ratios, age profile at diagnosis and regional variations. It also contains information that will enable us to access hospital records at a future date and to ‘flag’ the patients with the National Health Register, so that we can obtain death certificates.
We also have a second form designed to give information on symptoms, medication, lifestyle factors such as smoking, alcohol intake, height and weight, information on all endoscopies, details of pathology, H. pylori status and related conditions.
In addition to the data record forms, we have also asked the hospitals for their diagnostic criteria and surveillance programmes. This has highlighted one of the biggest problems faced by the Registry, which needs to be addressed. To date 24 hospitals have sent us their diagnostic criteria. Of these, one stated that they did not have any. The position amongst the other 23 is summarized in Table 1
. There do not appear to be any consistent diagnostic criteria, with the length of columnarized segment varying from any length to >4 cm. Not all patients are biopsied. Although most hospitals biopsy all their Barrett's patients, two biopsy fewer than 20% of those diagnosed endoscopically.
In the UK there is only one hospital with a formal surveillance programme and two hospitals have informal surveillance programmes. One hospital has a surveillance programme for specialized intestinal metaplasia.
A BSG committee has been set up by one of us (Dr Anthony Watson) to address these issues.
To date over 6500 patients from 36 hospitals have been registered. A number of analyses have been performed which have produced three peer reviewed papers and 12 abstracts. A list of these publications is given in the Appendix.
Conclusions from analyses
The analyses emanating from the Registry have so far produced the following conclusions:
- •In spite of marked differences in diagnostic criteria patient characteristics between the different centres are very similar.
- •More BO is detected in males than females (M:F = 1.7).
- •Mean age at diagnosis is higher in females than in males.
- •Peak age at diagnosis is a decade higher in females than in males, but there is some regional variation.
- •In males BO progresses to AC at twice the rate as in females.
- •These patient characteristics have remained consistent throughout the Registry's evolution, whether the analysis was of nine centres with 2103 BO patients, of 20 centres with 4261 BO patients or 27 centres with 5717 BO patients.
- •The rate of new BO cases as a percentage of total endoscopies has risen continuously between 1977 and 1996.
- •In a cohort of 102 British BO patients 31% of men and 71% of women under the age of 50 were obese (BMI >30).
- •Obesity prevalence in those aged >50 years was lower and corresponded to the general UK population.
- •Obesity may be an important co-factor in BO development through its effect on gastro-oesophageal reflux.
BO Registries – the future
National BO Registries are feasible and the lifestyle data gathered could be easily analysed to give valuable information on the aetiology and natural history of BO and the pathological steps that lead to its progression to AC. In future the data stored could be used to generate studies seeking either to prevent BO or prevent its progression to AC.
A European/International network of BO Registries, exchanging and pooling data and information, would be an invaluable medical resource. To date nine European countries have expressed an interest in establishing National Registries based on the UK model. Four centres in the USA and two centres in Canada have expressed an interest in establishing Regional Registries on the same model.
We wish to thank the Childwick Trust, The RL St J Harmsworth Memorial Fund and the David and Frederick Barclay foundation for their financial support for UKBOR.
Caygill CPJ, Reed PI, McIntyre A, Hill MJ (1998). The UK National Barrett's Oesophagus Registry: A study between two centres. Eur J Cancer Prev7: 161–4.
Caygill CPJ, Reed PI, Johnston BJ, et al. (1999). A single centre's 20 years’ experience of columnar-lined (Barrett's) oesophagus diagnosis. Eur J Gastroenterol Hep11: 1355–8.
Caygill CPJ, Reed PI, Hill MJ, Watson A (1999). An initial comparison of nine centres registering patients with the UK National Barrett's Oesophagus Registry (UKBOR). Eur J Cancer Prev8: 539–42.
Fitzgerald R, Caygill CPJ (1999). Treating upper digestive tract cancers as a single entity may be misleading. BMJ318: 1289–90.
Caygill CPJ, Reed PI, Hill MJ, Watson A (1997). The UK National Barrett's Oesophagus Registry. Gut40(Suppl 1): O35-12.
Caygill CPJ, Reed PI, Johnston BJ, Hill MJ, Levi S (1997). A study of Barrett's oesophagus registrations from a single centre over a 20 year period. Gut41(Suppl 3): E43.
Caygill CPJ, Reed PI, Watson A, McIntyre A, Hill MJ (1998). The first year of the UK National Barrett's Esophagus Registry. Gastroenterology114: O87-12.
Caygill CPJ, Reed PI, Hill MJ (1998). UK National Barrett's Oesophagus Registry: a comparison of nine centres. Digestion59(Suppl 3): 580.
Caygill CPJ, Reed PI, Johnston BJ, et al. (1998). UK National Barrett's Oesophagus Registry: role of obesity, tobacco and alcohol in Barrett's oesophagus. Eur J Cancer Prev7: 481.
Caygill CPJ, Jankowski J, Ali M, Reed PI, Hill MJ (1999). Expression of bile acid receptors: a preliminary report. Eur J Cancer Prev8: 354.
Caygill CPJ, Reed PI, Hill MJ, Watson A (1999). The UK National Barrett's Oesophagus Registry: a progress report. Eur J Cancer Prev8: 354.
Caygill CPJ, Reed PI, Johnston DA, Lopez M, Hill MJ (1999). Obesity and Barrett's oesophagus in a UK population. Gut45: O81-12.
Caygill CPJ, Reed PI, Hill MJ, Watson A (2000). The UK National Barrett's Oesophagus Registry: an update. Eur J Cancer Prev9: 456–7.
Reed PI, Caygill CPJ, Hill MJ, Watson A (1998). The UK National Barrett's Oesophagus Registry (UKBOR): the first two years. Eur J Surg Oncol24: 457.
Watson A, Reed PI, Caygill CPJ, et al. (1999). Changing incidence of columnar-lined (Barrett's) oesophagus (CLO) in the UK. Gut44(Suppl 1): O45-12.
Watson A, Reed PI, Caygill CPJ, et al. (1999). Changing incidence of columnar-lined (Barrett's) oesophagus (CLO) in the UK. Gastroenterology116: O351-12.