Search

Introduction: The relationship between Barrett's oesophagus and extra-oesophageal malignancies (EOM) has been a matter of controversy. Some researchers have pointed out that the incidence of colon cancer is higher in Barrett patients while others have found no such association. The current study explores the relationship between Barrett's oesophagus and EOM in general and colon cancer in particular in 27 813 subjects in General Practice Research Database (GPRD). Methods: The mean follow-up experience was about 6 years and the study subjects constitute four groups: Barrett's oesophagus group (1677), oesophagitis group (6392), simple reflux group (6328), and normal group (13 416). Respectively the last three groups were 4/1, 4/1, and 8/1 matched to the Barrett group according to their GP practice, date of birth and sex. All malignancies occurring before or in the first year after the diagnosis of Barrett's, oesophagitis, or reflux were regarded as prevalent and excluded from the analysis. Risk of occurrence of an unrelated condition such as cataract was also explored for comparison. Number of visits to the GP was defined as a date on which there are both medical and drug records available and its confounding effects were control[ed. Hazard ratios were calculated using Cox-proportional hazard regression analysis. Results: A total of 2167 EOM, 260 colorectal cancer, and 1918 cataract diagnoses were identified in study subjects of which 1119, 113, and 853 cases were excluded respectively as being prevalent. The hazard ratios have been presented in the table after adjusting for age, sex, and number of visits to the GP. Conclusion: The risk of colorectal cancer was not specifically higher in any of these groups. There were small increases in risk of EOM in the Barrett's oesophagus, oesophagitis and reflux groups in comparison to the general population. The explanation for these increases is unclear but they may be mediated by smoking if smoking rates are increased in... [ABSTRACT FROM AUTHOR]

Introduction: Barrett's oesophagus is recognised to be a pre-malignant condition and the incidence of oesophogeal adenocarcinoma (OAC) in people with Barrett's is much higher than general population. Nevertheless, it has been suggested that gastro-oesophageal reflux per se is a more important determinant of risk. The current study was set up to explore the risk of oesophageal cancer (OC) and specifically OAC in Borrett's using a total of 166 228 person-years follow-up experience from General Practice Research Database (GPRD). Methods: The 27 813 study subjects constitute four groups: Barrett's oesophagus group (1677), oesophagitis group (6392), simple reflux group (6328), and normal group (13 416). The last three groups were 4/1, 4/1, and 8/1 matched to the Barrett group according to their GP practice, date of birth, and sex, respectively. All prevalent cases of OC, including those diagnosed before or in the first year after the diagnosis of their condition, were excluded from the analysis. Hazard ratios were calculated using Cox-proportional hazard regression analysis. Using the England and Wales cancer registry data, the incidence of non-AC cancer of oesophagus and subsequently the estimated number of observed AC were worked out in study groups and the relative rate were calculated by standardised incidence ratio method. Results: A total of 137 oesophageal cancer cases were identified in study subjects of which 80 cases were excluded as being prevalent. 57 were entered into the analysis. The results have been presented in the table. Conclusion: Barrett's oesophagus increases oesophageal cancer risk about 20 times and oesophageal adenocarcinoma about 50 times in comparison to general population. There is only a modest increase in risk of oesophageal cancer in patients with reflux but no record of Barrett's. Given that the person-years follow-up in Barrett's group is about twice the largest previous study lower relative risk estimates further emphasises the... [ABSTRACT FROM AUTHOR]