Is it possible to precisely evaluate the impairment of esophageal sensitivity of Barrett's mucosa?
A. Watson (London)
It is generally believed that patients with Barrett's columnar-lined esophagus (CLE) have an impairment of mucosal sensitivity to acid exposure [1, 2], although there is a paucity of direct, objective evidence. Patients with Barrett's CLE complain of less heartburn than patients with erosive esophagitis [3, 4], and indeed the majority of patients in whom a diagnosis of Barrett's CLE is made have not been previously documented as suffering from gastroesophageal reflux disease (GERD), and in some patients, the diagnosis is made incidentally at endoscopy in patients who do not have typical reflux symptoms. In one study of 120 patients with reflux stricture, many of whom had Barrett's CLE, only 32% were known to have gastroesophageal reflux disease prior to diagnosis of Barrett's CLE and 24% had no reflux symptoms , despite the fact that reflux stricture and Barrett's CLE represent the extreme end of the pathophysiological spectrum of gastroesophageal reflux disease . Further evidence to suggest a diminished or absent awareness of esophageal mucosal acid exposure in patients with Barrett's CLE came from an epidemiological study performed in Olmsted County in the United States, where the prevalence of Barrett's CLE based on endoscopic diagnosis was 22.6 per 100,000 population, whereas autopsy studies revealed a prevalence of 376 per 100,000 population . Thus, for every patient in whom an endoscopic diagnosis of Barrett's CLE was made in Olmsted County, there were a further 16 people with Barrett's CLE who were either asymptomatic or who had minimal symptoms. This suggests the hypothesis that a sub-set of patients with gastroesophageal reflux disease may be symptomatically unaware of pathological esophageal acid exposure, such that destruction of the squamous epithelium and replacement with columnar-lined epithelium may occur in the absence of the typical symptoms of gastroesophageal reflux disease. Indeed, the existence of a group of patients with a "sensitive" esophagus, in whom small and indeed physiological degrees of acid exposure can produce severe symptoms, suggests that there may be a spectrum of nocioception in patients with gastroesophageal reflux disease, with patients with "sensitive" esophagus at one end and those with complications such as reflux stricture and Barrett's CLE at the other.
In order to test the hypothesis that patients with reflux stricture and Barrett's CLE may have impairment of mucosal sensitivity to acid exposure, Ball and Watson studied
20 patients with reflux stricture (7 of whom had Barrett's CLE) and 20 patients with erosive esophagitis using a modified Bernstein test . The distal esophagus was perfused sequentially with 30 mls of isotonic saline and 30 mls N/10 HCL at a rate of 6 mls per minute with the patient lying supine. Subjective sensitivity to acid perfusion was scored depending on the severity of symptoms experienced and the volume of acid infused at the time symptoms occurred. A score of 0-3 reflected whether symptoms were absent, mild, moderate or severe and a similar score of 0-3 reflected the onset of symptoms after infusion of 30 mls, 20 mls, 10 mls and less than 10 mls of acid respectively. Thus, a range of scores of 0-6 was available for each patient.
All patients with erosive esophagitis scored 3 or more and 14 (70%) scored the maximum of 6. In the Barrett's reflux stricture group, 8 patients (40%) experienced no symptoms during acid perfusion and only 3 (15%) scored the maximum of 6. There was no significant difference in sensitivity between those stricture patients with or without Barrett's CLE.
This study is currently being repeated in patients with Barrett's CLE but without stricture, with slight variations in technique to ensure that only the squamous mucosa is perfused, and incorporation of perfusion with alkali as well as acid. However, the results of this preliminary study suggests firstly that it is indeed possible to evaluate the impairment of esophageal sensitivity in Barrett's mucosa and secondly that in end-stage GERD, namely patients with stricture and Barrett's CLE, there is significant impairment, and frequently absence of subjective sensitivity to acid exposure. This may explain the late presentation of such patients and the difference in prevalence between endoscopic and autopsy studies. It has therapeutic implications in that symptoms are clearly an unreliable guide both to the stage of the disease and the response to treatment in such patients.