Exocrine pancreatic cancer, cigarette smoking, and diabetes mellitus: a case-control study in northern Italy.

The role of cigarette smoking and diabetes mellitus as risk factors for exocrine pancreatic cancer (PC) was investigated in a hospital based case-control study. Current smokers were at increased risk for PC (OR = 2.36, 95% CI 1.53-3.63): the magnitude of the risk was related to the lifetime amount of smoking (chi2(trend) = 17.00; P < 0.0001). Among former smokers, after 15 years from ceasing smoking, the risk for PC dropped to the level of a lifetime non-smoker, whichever the lifetime smoking amount. Diabetes was associated with a 2.89-fold increased risk for PC (95% CI 1.71-4.86): the risk was 4.76 (95% CI 1.99-11.53) for diabetes diagnosed up to 2 years before the diagnosis of PC and dropped to 2.07 (95% CI 1.02-4.20) for diabetes diagnosed more than 5 years before PC. The risk for PC was estimated according to the treatment used to control diabetes: it was 6.49 (95% CI 2.28-18.48) for insulin treated diabetes and 2.12 (95% CI 1.16-3.87) for diabetes treated with oral hypoglycemic drugs. The risk of PC for diabetes treated for more than 5 years before the diagnosis of PC was 6.21 (95% CI 1.61-23.96) for patients treated with insulin and 1.21 (95% CI 0.50-2.92) for those treated with oral hypoglycemic drugs: the type of treatment needed to control the disease may discriminate between the diabetes that represents a consequence of cancer from the diabetes that could represent an etiological co-factor. More studies are needed to clarify whether long-lasting insulin-treated diabetes is an etiological co-factor in PC.

Risk factors for Barrett's esophagus: a case-control study.

Barrett's esophagus (BE) is an acquired disorder due to chronic gastroesophageal reflux. Environmental factors seem to play an important role in the pathogenesis of BE, especially in Western society. A multicenter case-control study was carried out between February 1995 and April 1999 in 8 Italian Departments of Gastroenterology gathered in a study group (GOSPE), in order to analyze the influence of some individual characteristics and life-style habits on the occurrence of BE. Three groups of patients were studied: 149 patients with BE, 143 patients with esophagitis (E) and 308 hospital controls (C) with acute, non-neoplastic, non-gastroenterological conditions. The diagnosis of BE was based on endoscopy and histology. E was defined by the Savary classification (grade I-III). Data collection was performed by using a questionnaire that focused on smoking, coffee and alcohol consumption, medical history, drugs history, gastroesophageal reflux disease (GERD) symptoms (heartburn, regurgitation) and socio-economic status. Multivariate analysis showed that the frequency of weekly GERD symptoms was significantly associated with both BE and E (p<0.0001), such as the presence of hiatal hernia (p< or =0.001). Ulcer was significantly associated with BE (p=0.001). Among patients with E, the risk was directly related to spirits consumption (p=0.03). Patients with GERD symptoms that lasted more than 13 years were more likely to have BE than E (p=0.01). In conclusion, results from our study point out that long-standing GERD symptoms, hiatal hernia and possibly alcohol consumption are risk factors in the development of the BE and E.