Medical and Surgical Therapy for Gastroesophageal Reflux Disease and Barrett Esophagus

Clinical Features of Barrett Esophagus

The mucosa of the esophagus consists of stratified squamous epithelium. Barrett esophagus is the metaplastic replacement of this squamous epithelium with intestinal columnar epithelium, which predisposes to the development of adenocarcinoma. Barrett esophagus has been described across all demographics, but in the Western world, where it is the most prominent, the highest incidence is in middle-aged white males. It is diagnosed by recognition of columnar mucosa on endoscopy and confirmed histologically from tissue obtained from biopsy during the upper endoscopy. The mean age at time of diagnosis is in the 6th decade of life and demonstrates a male-to-female predominance of approximately 2 : 1. Obesity is a risk factor for gastroesophageal reflux disease (GERD) due to its effects on the body's normal antireflux mechanisms and is consequently a likely risk factor for Barrett esophagus. Body mass index (BMI) exists as the most common quantifier of obesity; however, many have suggested that an increased abdominal circumference or waist-to-hip ratio may carry more significance, as it is a better measure of abdominal adiposity. Other independent risk factors that have been suggested include smoking and presence of a hiatal hernia. Conversely, the presence of Helicobacter pylori has demonstrated an inverse association with risk of Barrett esophagus for reasons that are not entirely clear. The majority of available data suggest that the prevalence of Barrett esophagus is significantly lower in African Americans compared with non-Hispanic whites by several times. Although it is not entirely understood whether this imbalance exists due to differences in risk factor profiles or genetic susceptibility, a genetic predisposition seems likely. Variances at specific genetic loci have been implicated in increasing the genetic susceptibility to developing Barrett esophagus. Also, the vast majority of patients are diagnosed on their first endoscopy. This suggests that reflux acts as a trigger in the right genetic environment for the development of Barrett esophagus.

Barrett esophagus has been categorized by its extent either as long (>3 cm) or short (<3 cm) segments. A third, more controversial, description is intestinal metaplasia at the gastroesophageal junction. This likely represents either very short segment Barrett or intestinal metaplasia of the gastric cardia, and the malignant potential is less clear. Short-segment Barrett represents the large majority of cases, while long segments likely represent more severe reflux both in terms of acid exposure and its proximal extent. There has been theoretical concerns that long-segment disease may be associated with a higher incidence of both dysplasia and adenocarcinoma due to disease burdens on the cellular level, and there is some retrospective data to support this. However, the available evidence has not been sufficient to justify incorporating length consistently across current surveillance guidelines. Instead, the presence and degree of dysplasia has provided the marker of risk and has driven the clinical management. Histologically Barrett esophagus is classified as nondysplastic, indefinite for dysplasia, low-grade dysplasia (LGD), or high-grade dysplasia (HGD). A specimen indefinite for dysplasia is often a result of active inflammation, which precludes accurate histologic classification. This is an interim diagnosis only, which requires close follow-up for definitive characterization of the histologic pattern. Acid suppression therapy should be maximized and repeat biopsies obtained after a brief period to allow healing. Regardless of the classification, there remains a subjective element to histologic grading, which has been demonstrated by relatively low interobserver agreement within specimens, particularly for LGD. Consequently, a second opinion from an expert gastrointestinal pathologist is recommended to confirm a diagnosis of LGD or HGD.

Surveillance

The development of esophageal adenocarcinoma represents the most feared complication of Barrett esophagus, with mortality increasing with the stage at diagnosis. The goal of early detection forms the rationale of surveillance. This efficacy of surveillance has been questioned recently, but poor efficacy is likely related to an excessively long interval between surveillance endoscopies and an inadequate number of biopsies to adequately evaluate the Barrett segment. Surveillance for nondysplastic Barrett esophagus should include the use of high-definition endoscopy with systematic four-quadrant biopsies every 2 cm with separate endoscopic mucosal resection (EMR) of mucosal abnormalities at an interval of 1 to 2 years. If there is a history of LGD, biopsies should be obtained every 1 cm with an interval of 6 to 12 months, although increasingly patients with dysplasia are undergoing ablation rather than continued surveillance. There is evidence that surveillance results in the detection of esophageal adenocarcinoma at an earlier stage. For instance, in a retrospective review of 224 patients, Grant et al. found that those who underwent surveillance had significantly lower stage tumors at the time of diagnosis. One would then expect this to translate to a benefit in mortality but concern exists that this may represent lead-time bias instead of a true survival benefit. Disputing this is data from multiple centers showing that with endoscopic resection of early adenocarcinoma patients are cured of their disease and have survival similar to the general population. Corley et al. did not demonstrate an association between surveillance and a decreased risk of death in a case-control study of 70 patients with esophageal adenocarcinoma diagnosed greater than 6 months after Barrett esophagus, although these results have been criticized for high overall mortality rates, not using endoluminal therapy in early disease and not differentiating between adequate surveillance and any surveillance. In contrast, shortly thereafter, Verbeek et al. demonstrated decreased esophageal adenocarcinoma mortality at 2 and 5 years for those adhering to a surveillance program (hazard ratio [HR], 0.79) ( Fig. 33.1 ). Logically, if early-stage cancer is curative in most patients, it makes sense that surveillance at an appropriate interval that allows detection of progression to HGD or early adenocarcinoma will be beneficial.

Treatment of Nondysplastic Barrett Esophagus