Long-term complications to reflux disease in community practice. A 17-year cohort study of 4706 patients.

OBJECTIVE: To examine the risk of developing strictures in patients with erosive and non-erosive gastroesophageal reflux disease in a community-based setting, since controlled trials indicate that the use of proton pump inhibitors renders the risk of strictures insignificant. MATERIAL AND METHODS: A 17-year cohort study of 4706 patients referred to endoscopy due to upper GI symptoms, with a population comparison cohort of 47,060 individuals. All patients were followed and treated according to prevailing guidelines by their usual care provider. Main outcomes were relative risks (RR) and 95% confidence intervals (CI) for incident strictures and dilatations. RESULTS: 776 (16.5%) patients were diagnosed with erosive esophagitis, particularly men (61.2%). Over a period of 1-17 years (mean 10.5), 20 patients (2.6%) in the esophagitis group developed a peptic stricture, necessitating one or more dilatations in 16 patients (2.1%). Among the non-esophagitis patients, the incidences for both outcomes were 1.2%. Male gender doubled the risk of developing strictures, and alcohol abuse raised the risk four folds. Erosive patients had a risk of developing strictures eight times (95% CI: 5.0-13.0) higher than controls, whereas non-erosive patients' risk was 4.0 (95% CI: 2.8-5.7). The majority of strictures developed within the first 10 years after a diagnosis of esophagitis. CONCLUSION: Patients with esophagitis had eight times higher risk of strictures than population controls and two times higher than dyspeptic patients without esophagitis. This indicates that long-term outcomes in general practice are poorer than in controlled trials, most likely due to a lack of compliance with medication.

Unexplained chest/epigastric pain in patients with normal endoscopy as a predictor for ischemic heart disease and mortality: a Danish 10-year cohort study.

BACKGROUND: Normal upper endoscopy may be a marker of ischemic heart disease in patients with unexplained chest/epigastric pain. METHODS: We examined the 10-year risk of ischemic heart disease and mortality in a cohort of 386 Danish patients with chest/epigastric pain, normal upper endoscopy, and no prior hospital discharge diagnosis of ischemic heart disease (defined as patients with unexplained chest/epigastric pain), compared with 3,793 population controls matched by age, gender, and residence. Outcome data were obtained from population-based health registries. Cox regression analysis was used to estimate the relative risk of hospitalization for ischemic heart disease and the adjusted mortality rate ratio (MRR). RESULTS: The 10-year relative risk of hospitalization for ischemic heart disease following a normal upper endoscopy among patients with unexplained chest/epigastric pain was 1.6 (95% CI, 1.1-2.2), compared with controls. The 10-year MRR was 1.1 (95% CI, 0.9-1.5). Within the first year after the upper endoscopy the MRR was 2.4 (95% CI, 1.3-4.5). The cause-specific MRR among patients with unexplained chest/epigastric pain compared with controls was up to threefold higher for deaths related to alcohol dependence, pneumonia, and lung cancer. CONCLUSION: Unexplained chest/epigastric pain in patients with normal endoscopy is a strong marker for ischemic heart disease and increased mortality.

Risk of gastrointestinal cancer in patients with unexplained chest/epigastric pain and normal upper endoscopy: a Danish 10-year follow-up study.

Unexplained chest/epigastric pain is a common symptom in the general population. However, it has not previously been studied whether such pain could be a marker of subsequent gastrointestinal cancer. We aimed to estimate the risk of gastrointestinal cancers in a Danish 10-year follow-up study among patients with chest/epigastric pain, normal upper endoscopy, and no prior discharge diagnosis of ischemic heart disease (N = 386), compared with population controls (N = 3860). The overall 10-year risk of gastrointestinal cancer (stomach, colorectal, liver, and pancreas) was 2.9% for patients with unexplained chest/epigastric pain vs. 1.5% for controls. The adjusted relative risks <1 year and > or =1 year after upper endoscopy were 8.4 (95% confidence interval [CI], 2.6-27.5) and 1.2 (95% CI, 0.5-2.9), respectively. We found that patients with unexplained chest/epigastric pain have an increased risk of gastrointestinal cancer within the first year after upper endoscopy. Consequently, unexplained chest/epigastric pain might be an early gastrointestinal cancer symptom.

Inflammatory bowel diseases, 5-aminosalicylic acid and sulfasalazine treatment and risk of acute pancreatitis: a population-based case-control study.

OBJECTIVES: Patients with inflammatory bowel diseases are suggested to have an increased risk of acute pancreatitis. Although azathioprine and glucocorticoids are risk factors for acute pancreatitis, the relation is poorly understood, in particular the role of 5-aminosalicylic acid and sulfasalazine treatment. To clarify these relations, we conducted a population-based case-control study. METHODS: We identified 1,590 incident cases of acute pancreatitis from the Hospital Discharge Registry of the North Jutland County of Denmark from 1991 to 2002, and selected 10 controls per case (N = 15,913) from the Central Personal Registry, matched by age and gender. Among cases and controls, we identified patients with inflammatory bowel diseases. Data on drug use were extracted from a Pharmaco-epidemiological Prescription Database. RESULTS: Adjusted odds ratios for acute pancreatitis in patients with Crohn's disease and ulcerative colitis were 3.7 (95% confidence interval (CI), 1.9-7.6) and 1.5 (95% CI, 0.7-3.6), respectively. In all patients treated with 5-aminosalicylic acid and sulfasalazine the adjusted odds ratios for acute pancreatitis were 0.7 (95% CI, 0.4-2.2) and 1.5 (95% CI, 0.4-5.2), respectively. Restricted to patients with inflammatory bowel diseases only, the adjusted odds ratios for acute pancreatitis in patients exposed to 5-aminosalicylic acid and sulfasalazine were 0.7 (95% CI, 0.1-3.8) and 0.6 (95% CI, 0.1-6.7), respectively. CONCLUSION: We found a nearly four-fold increased risk of acute pancreatitis in patients with Crohn's disease and a 1.5-fold increased risk for ulcerative colitis. In patients with inflammatory bowel diseases, the use of 5-aminosalicylic acid or sulfasalazine was not associated with increased risk of acute pancreatitis.