Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

  Browse by Author
  Browse by Movies
Volume: Barrett's Esophagus
Chapter: Etiology and origins of Barrett's epithelium

What is the prevalence of long segment Barrett's esophagus?

A.J. Cameron (Rochester)

The incidence of adenocarcinomas of the esophagus and cardia has increased greatly. In our data from Olmsted County, we compared the years 1965 to 1974 with 1990 to 1997. The incidence of esophageal adenocarcinoma per 100,000 population per year increased from 2.2 to 2.1. The incidence of adenocarcinoma of the cardia also increased by about ten times from 0.2 to 2.0.

Most cases of adenocarcinoma of the esophagus and a much smaller proportion of adenocarcinomas of the esophagogastric junction or cardia, arise in a long segment Barrett's esophagus. There are different ways to estimate the prevalence of Barrett's esophagus. For the purpose of this talk, it will be defined as 3 cm or more of columnar epithelium in the lower esophagus.

There is a high prevalence of heartburn in the general adult population. In 1976, Nebel et al. found daily heartburn in 7% and weekly or more frequent heartburn in 19% of adults. This has not changed much over time. In a 1997 report, Locke et al. found weekly heartburn in 18% of adults.

Prospective endoscopic studies have been done to determine the prevalence of Barrett's esophagus in patients with frequent reflux symptoms. Defining Barrett's as 3 cm with proven intestinal metaplasia, in four studies, 3.5% to 7% of patients with reflux had a long Barrett's esophagus. The studies reporting a higher incidence were carried out in males.

If the data given in the last two paragraphs is combined, the prevalence of Barrett's in the general population can be estimated. Assume that 1 in 5 (20%) adults has weekly reflux symptoms, and that 1 in 20 (5%) with reflux symptoms has Barrett's esophagus. Then we might expect 1 in 100 adults to have a Barrett's. This may be an underestimate because some patients with Barrett's do not have any symptoms.

The most frequent presentation of adenocarcinoma of the esophagus is a short history of weeks of tumor-related symptoms such as dysphagia, weight loss, chest pain, or bleeding. Endoscopy is done, and often a Barrett's esophagus is found then for the first time, together with the cancer. In 6 series describing such patients, a median of 60% of patients had a history of chronic reflux symptoms before the tumor, thus 40% of patients with Barrett's may be estimated to have no reflux symptoms.

The prevalence of Barrett's esophagus varies with age. In a large endoscopic series, the prevalence was 0 in early childhood and rose to 1% of all patients having endoscopy after the age of 60. This is consistent with an acquired disorder. Barrett's is also commoner in males, in one study. The ratio was 2 males to 1 female.

We have recently done a prospective study of endoscopy in 200 patients with reflux symptoms and again showed the prevalence to increase with age, about 1/2 the maximum prevalence being reached in the 40 to 49 age group. In this study, patients with reflux symptoms for more than 10 years had a 10.3% prevalence of Barrett's compared to a 2.2% prevalence in patients with reflux symptoms extending for less than 10 years.

The author previously reported an autopsy study. 733 Mayo Clinic autopsies were reviewed consecutively. Seven of 733, about 1% of older people, had a Barrett's esophagus in this study. Therefore, three different methods to estimate the prevalence of Barrett's in older people, 60 years and above, each gives an approximately 1% of all population members of that age having a Barrett's esophagus.

The true population prevalence must be distinguished from the clinically diagnosed prevalence. The author et al. did a study in Olmsted County, Minnesota, where the Mayo Clinic is located. We obtained information from the medical records of all patients living in the county who had a diagnosis of Barrett's esophagus based on endoscopy with biopsy. In January, 1998 the clinically diagnosed prevalence was 80 per 100,000 population. The age and sex adjusted cases per 100,000 based on our previous autopsy study was 376. Even in a County with good medical facilities and frequent use of endoscopy, it is estimated that only 1 in 5 of persons with Barrett's esophagus living in the County has been diagnosed. Therefore, most patients with Barrett's are not being followed for early cancer detection.

Estimated prevalence of Barrett's in the older general population is about 1% or about 0.4% if we include the entire population, with children having a very low prevalence. The prevalence is about 5% in adults with chronic reflux symptoms. There is less data on patients without reflux symptoms, but an estimate would be about 0.5% of such people in the general population having Barrett's esophagus. All of these figures are higher than the clinically diagnosed population prevalence for Olmsted County, 0.08%.

Our data show that between 1965 and 1995, the incidence of new diagnosis of Barrett's per 100,000 population rose from less than 1 per year to the current figure of 11 per year. Over the same years, the number of upper endoscopic examinations per 100,000 population rose from 50 to about 1,400. This suggests that Barrett's may have been present in the population throughout the years of study, and that we now diagnose it more as we do many more endoscopies.

There is some further evidence that Barrett's may not be a new disease. Allison and Johnstone reported in 1953 that 10% of 115 patients with esophageal stricture having endoscopy had a columnar-lined esophagus above a hiatal hernia. Naef et al. did almost 5,000 rigid esophagoscopic examinations between 1963 and 1971. They found a columnarlined esophagus in 1.25%.

Different series showed an increase in the incidence of esophageal adenocarcinoma of 5 to 10 times, in the past 25 years. The data presented suggests the prevalence of reflux symptoms in the population has not changed much in these years, and there is little evidence of a change in the prevalence of Barrett's esophagus. There is no useful data on the prevalence of Barrett's esophagus before the 1950's. We do not know whether it was always present. The increase in adenocarcinoma is clearly a real finding.

Olmsted County, Minnesota has approximately 100,000 population. It is estimated there are about 15,000 people in the county who have reflux symptoms every week. The autopsy estimate is that there are about 400 cases of long Barrett's in the population, but only 80 of these are clinically diagnosed. There are approximately two cases of esophageal adenocarcinoma and two cases of cardia adenocarcinoma per year in the same 100,000 population.

Publication date: August 2003 OESO©2015