From Our Archives
Medical Revising Editor: Jay W. Marks, MD
Esophageal cancer is the eighth most common type of cancer and causes 12,000 deaths per year in the U.S. (2% of all cancer deaths). One type of esophageal cancer, adenocarcinoma, accounts for 50% of esophageal cancers and occurs primarily in Caucasian men. The incidence of adenocarcinoma and deaths from esophageal cancer have been increasing steadily in the U.S. and western Europe.
Gastroesophageal reflux disease (GERD), in which acid refluxes from the stomach into the esophagus (known primarily for causing heartburn) is a condition that afflicts 20% of the populations of the U.S. and western Europe.
It has been hypothesized that GERD is an important cause of adenocarcinoma of the esophagus. Specifically, the hypothesis is that chronic reflux of acid into the esophagus causes changes in the cells lining the lower esophagus--changes that are referred to as Barrett's esophagus--that ultimately lead the cells to become cancerous. It is estimated that ½ to 1% of patients with Barrett's esophagus develop adenocarcinoma each year they are followed. (This means that during 20 years a patient with Barrett's esophagus has a 10 to 20% risk of developing adenocarcinoma.) Therefore, it has been recommended that patients with Barrett's esophagus undergo regular and frequent endoscopy (every year) and biopsy of the esophagus so that early malignant changes can be detected and treated early before cancer spreads.
An important study published in the Mar. 18, 1999, issue of the New England Journal of Medicine examined the links between adenocarcinoma of the esophagus, Barrett's esophagus, and heartburn (GERD) in the entire population of Sweden. The authors found a strong association between heartburn and cancer. Although the risk of cancer was increased even among individuals with mild heartburn, the risk was greater if the heartburn occurred frequently or was present for many years. Specifically, patients with frequent symptoms of heartburn for more than 20 years were 44 times more likely to develop adenocarcinoma than individuals without heartburn. The authors estimated that GERD might be responsible for half of all adenocarcinomas and 87% of the adenocarcinomas among patients with heartburn.
Adenocarcinoma was almost as likely to occur in patients without the changes of Barrett's esophagus as patients with the changes, suggesting that Barrett's esophagus may not be a stronger predictor of malignancy than heartburn alone.
The study raises several important questions but unfortunately does not answer them. Should all patients with heartburn (or at least those patients with frequent and/or prolonged heartburn) see their physicians to begin regular endoscopy to detect early adenocarcinoma? The effort and resources needed to provide regular endoscopy and biopsy for all patients would be great, and there is no evidence yet that regular endoscopy would detect adenocarcinoma early and prevent deaths. Moreover, there still are experts who feel that the changes of Barrett's esophagus always precede adenocarcinoma and provide a satisfactory means to identify patients who are at risk for adenocarcinoma and who need regular endoscopy and biopsy. (Perhaps in the Swedish study the changes of Barrett's esophagus were missed or destroyed by the adenocarcinoma.) Thus, all that is necessary is a single endoscopy with biopsy to determine if a patient has the changes of Barrett's esophagus and needs regular endoscopy. (This is a commonly recommended approach.)
There is a related question that also has no answer. If a patient with frequent and/or prolonged heartburn does not have the changes of Barrett's esophagus at the first endoscopy, should endoscopy be repeated at a later time to detect changes of Barrett's esophagus that may develop after the first endoscopy? If so, how much later should the endoscopy be performed?
Should all patients with heartburn (or at least all patients with frequent and/or prolonged heartburn) be aggressively treated, that is, should they receive acid-suppressing medication indefinitely or undergo surgery to prevent reflux of acid and adenocarcinoma? Again, there is no evidence yet that such treatment will prevent either the changes of Barrett's esophagus or adenocarcinoma.
In the study from Sweden, heartburn was considered synonymous with GERD. It is known, however, that GERD may occur without symptoms or with unusual symptoms that are not generally recognized as being due to GERD. Moreover, it has been suggested that patients with the changes of Barrett's esophagus are less likely to have heartburn than patients without the changes. Thus, it is possible that some patients in the study with adenocarcinoma but without heartburn actually had GERD, and therefore, that more than the estimated 50% of adenocarcinomas were associated with GERD.
Despite the concerns raised by the Swedish study, it is important to remember that the study demonstrated an association between heartburn and adenocarcinoma, not that heartburn (GERD) causes adenocarcinoma. In order to prove that GERD causes adenocarcinoma, it will be necessary to show that effective treatment of GERD prevents adenocarcinoma. Such a study will not be available soon, if ever.
- Lagergren J, Bergstrom R, Lindgren A, Nyren O. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. N Engl J Med 1999;340:825-31. (Original article from Sweden).
- Cohen S, Parkman HP. Heartburn--a serious symptom. N Engl J Med 1999;340:878-9. (Editorial accompanying original article).