What would justify making a positive diagnosis of alkaline reflux?
R.K. Mittal (Charlottesville)
The duodenal secretions have an alkaline pH which is due to the bicarbonate secreted by the duodenal mucosa and secretions from the liver and pancreas. In an experimental situation, perfusion of the esophagus with pancreatico-biliary secretions can induce mucosal inflammation and esophagitis. Based on this observation and case reports of the presence of esophagitis in patients with pernicious anemia, who do not have acid in the stomach, investigators have been led to believe that alkaline pancreatico-duodenal contents can reflux back into the esophagus, and induce esophagitis. Several investigators have indicated that even those patients who have acid in their stomach may reflux pancreatico-biliary secretion along with or in between acid reflux episodes and exacerbate the noxious potential of the refluxate material. There are three important questions that are relevant to this area: 1) can the pancreatico-biliary
secretion injure the esophagus; 2) do pancreatico-biliary secretions reflux all the way back into the esophagus in the human setting; and 3) is the concentration of bile salts and trypsin in the esophageal refluxate high enough to induce mucosal injury?
In the acute experimental animals, it appears that bile salts in concentrations of 3-5 mM range or larger in the presence or absence of HC1 can induce significant esophagitis [1,2]. Taurine and glycine conjugate of bile salts potentiate the HC1 in inducing esophagitis, because they can remain in the solution at low pH. However, most of the animal studies are acute studies and do not provide data on the minimal concentration needed to induce esophagitis, if the esophagus were to be exposed to it on a chronic basis.
Do pancreatico-biliary secretions indeed reflux all the way back into the esophagus in enough concentrations to induce mucosal damage? The investigators have attempted to answer this question with the following methodology: 1) measuring esophageal pH - it is suggested that an esophageal pH of >7 represents reflux of pancreatico-biliary material into the esophagus ; 2) nuclear imaging of the radio labeled bile acid in the stomach ; 3) aspiration of esophageal refluxate material and its analysis for the presence of bile acids; and 4) measurement of bilirubin in the esophagus using a bili-probe, the rationale being that if bilirubin was present in the esophagus, then other pancreatico-biliary secretions must be present as well.
Does an esophageal pH of >7 represent reflux of pancreatico-biliary secretions into the esophagus? The pH of the pancreatico-biliary and duodenal secretions is 7.5-8.0. Therefore, intuitively, it seems correct that if the pH in the esophagus was above 7, it could represent pancreatico-biliary secretions. However, the major problem with this supposition is that it assumes that there are no other secretions in the esophagus that may have a pH of >7. It is clear that the pH of saliva is 7.5-8.0. In addition, esophageal mucosal glands may secrete bicarbonate that could make esophageal pH close to 7.5-8.0. Therefore, an esophageal pH of >7 could represent one of three things: 1) saliva in the esophagus; 2) esophageal secretions; or 3) pancreatico-biliary secretions. It is impossible to differentiate between the three materials just based on the esophageal pH measurement alone . The other argument that makes the pH > 7 as representative of pancreatico-biliary secretions less likely is that pancreatico secretions must pass through the stomach in order to reach the esophagus. Gastric pH is usually 1.0-2.0 and it is more than likely that the duodenal secretions are quickly acidified as they enter into the stomach.
Can radionuclide studies document biliary bile reflux into the esophagus? The bile salt pool can be labeled by ingestion of bile acid isotope, which can subsequently be aspirated from the gastric lumen and measured. Cholescintigraphy, employing intravenous radio isotope that is secreted into the bile, can also demonstrate duodenogas-tric reflux. The isotope can be measured by ע-cameras positioned on the abdomen. Using these techniques, one can document duodenogastric reflux in the postsurgical stomachs. A newly developed technology involves a ע-detector which is small enough to be passed into the gastric lumen for detection of radiolabeled bile acids. These studies can qualitatively document gastroduodenal bile reflux in a variety of pathological states, and in about 10% of the normals. However, none of these techniques are sensitive enough to be able to demonstrate reflux of bile salts in the esophagus.
The direct demonstration of esophageal reflux of bile requires esophageal aspiration and quantification of bile salts through one of the bile acid assays [6,7]. The esophageal refluxate can be collected via a salem sump tube positioned proximal to the gastroesophageal junction. Since the reflux is an intermittent and unpredictable phenomenon, continuous aspiration over prolonged periods is required to capture the esophageal refluxate material. The esophageal refluxate can then be analyzed for the presence of bile acids through either the enzymatic assay (3 α-hydroxy steroid dehydrogenaze), or liquid chromatography, or mass spectrometry. The type of assay used is important because the enzymatic assay, even though sensitive, can produce false positive results. High performance liquid chromatography (HPLC), even though less commonly reported, is fairly sensitive (detection range of 10-20 ÁM) and a specific assay. Similarly, the liquid ion mass spectrometry, again, is fairly sensitive and specific for detection of bile acids.
There are a total of seven studies, including a total of 159 patients, in which the bile salts in the esophageal refluxate were analyzed. The majority of these studies were on patients with an intact stomach. Of the patients, 36% had no bile salts detected in the esophagus, and only 13% of the patients had bile salt concentrations of >200 ÁM. Of the control subjects, 93% had no bile salts detected, and the highest concentration recorded was 40 ÁM. One reason that the concentrations of bile in the esophageal refluxate are small may be because of an inherent problem in the aspiration methodology to collect the esophageal samples. Esophageal refluxate obtained by continuous esophageal aspiration is likely to contain swallowed saliva, which can cause dilution of the refluxed material. Some investigators have made an attempt to aspirate the saliva from the mouth in order to prevent mixing of the saliva with the esophageal refluxate. However, even with the most compulsive oral suctioning, some saliva is bound to trickle down the esophagus and dilute the esophageal samples significantly.
If we assume that the bile salt concentrations in the refluxate, reported in these studies, are true, then these concentrations are similar to plasma concentration of bile acid. These concentrations are generally 100-1,000 times lower than those used in experimental models of bile esophagitis. It is highly unlikely that these small concentrations of bile salts can induce damage, at least in an acute situation. Whether these small concentrations over prolonged periods of time can induce esophagitis or potentiate HC1 acid remains to be seen.
Recently, Champion et al.  have used a bilitec 2000 probe (Synectics, Inc.) to identify the presence of bilirubin as a marker of pancreatico-biliary reflux in the esophagus. These authors found that when using bilirubin as a marker of alkaline reflux, esophageal pH of >7 correlated poorly with duodenogastric reflux. Bilirubin reflux correlated highly with the acid reflux. In other words, no bilirubin was detected in the esophagus when the esophageal pH was >7, suggesting that bile acid reflux occurs in the esophagus in the acidic and not in the alkaline environment. However, this technique cannot provide the quantitative data regarding the concentration of bile acids in the esophagus and cannot answer the questions whether significant concentrations of bile salts capable of inducing mucosal damage do indeed reflux into the esophagus.
I feel that a positive diagnosis of biliary reflux would require direct demonstration of bile acid and trypsin in the esophageal refluxate. Furthermore, these substances should be present in enough concentration to induce mucosal damage. Two things that can unequivocally establish the role of pancreatico-biliary reflux in reflux disease would be: 1) the development of a bile probe, very similar to a pH or bilirubin probe, that can sit in the esophagus and indicate not only the qualitative but quantitative aspects of the pancreatico-biliary reflux; or 2) I think inhibition of bile acids and trypsin in the esophagus with specific pharmaceutical agents, along the lines similar to inhibition of acid, may indirectly indicate the true role of the bile salts and pancreatic enzymes in the pathogenesis of reflux disease. Unfortunately, neither the bile acid probe nor the pharmaceutical substances for neutralization of the bile and trypsin in the esophagus are on the horizon in the immediate future. Therefore, determination of the true role of pancreatico-biliary reflux in reflux esophagitis will have to wait.
4. Eriksson B, Emas S, Jacobsson H, Larsson SA, Samuelsson K. Comparison of gastric aspiration and HIDA scintigraphy in detecting fasting duodenogastric bile reflux. Scand J Gastroenterol 1988:23:607-610.