Refractory GERD: Further Insights Into the Cause of Symptoms

Article Outline

• Comment
• Copyright

Zerbib F, Duriez A, Roman S, et al. (Gastroenterology Department. Saint Andre Hospital. Bordeaux, France). Determinants of gastro-oesophageal reflux perception in patients with persistent symptoms despite proton pump inhibitors. Gut 2008;57:156–160.

The patient with “gastroesophageal reflux disease (GERD) despite therapy” or “refractory GERD” is becoming an increasingly common problem in clinical practice today, both for typical and atypical symptoms. New insights into the cause of symptoms in this challenging group of patients are provided by Zerbib et al (Gut 2008;57:156–160). The authors examined the ambulatory impedance pH recordings of 59 patients with persistent symptoms of typical GERD, namely heartburn and/or regurgitation despite twice daily therapy with proton pump inhibitors (PPIs). Of these 59 patients, 39 were excluded from the study because of either a lack of relation between symptoms and reflux events or absence of symptoms during the 24-hour recording. Impedance pH recordings were examined for reflux events (these were termed acid if pH < 4; weakly acidic if pH between 4 and 7; weakly alkaline if pH did not drop below 7), reflux composition (described as liquid, gas, or mixed liquid and gas), position, and a variety of other variables. Patients were considered to have proximal reflux if reflux reached the impedance site 15 cm above the lower esophageal sphincter (LES) and reflux events were considered to be symptomatic if they occurred within a 2-minute window before the activation of the event marker. Of 20 patients, 9 (45%) had an increase in the number of reflux episodes above normal. The overwhelming majority of reflux episodes were weakly acidic (80%). However, of the 312 symptomatic reflux episodes, the rates of symptomatic acid and weakly acidic episodes were comparable (27.6% vs 23.6%). The most striking finding of the study was that 53.3% of symptomatic reflux events extended proximally to the probe site 15 cm above the LES compared with 33.8% of the asymptomatic events (P = .037). Heartburn symptoms were more likely than regurgitation to be supine, purely liquid, acidic, have a lower nadir pH, be preceded by acid reflux events and have a longer bolus clearance time. Thus, proximal extent, composition of reflux, and sensitization of the esophagus all seemed to contribute to the perception of symptoms in patients with classic GERD symptoms while on BID PPI therapy.

Back to Article Outline

Comment 

GERD despite therapy, be it for typical or atypical symptoms, is a vexing problem encountered in gastroenterology practice today. A variety of causes for persistent symptoms despite twice daily PPI therapy have been described including problems with both compliance and improper dosing of PPIs, weakly acidic reflux, duodenogastric/bile reflux, visceral hypersensitivity, concomitant functional bowel disease, psychological comorbidities, delayed gastric emptying, and eosinophilic esophagitis (Clin Gastroenterol Hepatol 2008;6:393–400). The problem is most notable in patients with nonerosive reflux disease. Although issues such as compliance, incorrect dosing of medications, and eosinophilic esophagitis are fairly simple to sort out, the timing and sequence of diagnostic testing, especially impedance pH studies in these patients, is somewhat more difficult and problematic.

In patients with persistent typical GERD symptoms despite twice daily PPI therapy, standard transnasal esophageal pH monitoring is most likely to be normal (Am J Gastroenterol 2005;100:283–289). Combined 24-hour impedance pH monitoring has demonstrated that persistent symptoms in patients on twice daily PPIs may be related to nonacid reflux in a subset of patients as determined by either the symptom index or the symptom association probability (Gut 2006;55:1398–1402; Am J Gastroenterol 2006;101:1956–1963). But could there be more to this phenomenon than simply looking for symptom associations with either acid or nonacid reflux events?

The issue of spatiotemporal characteristics of reflux and symptom perception has been an area of interest now for some time. Weusten et al, using a 5-channel pH monitoring system, found that symptomatic reflux episodes were longer and the proximal extent higher than that of a symptomatic episodes in a group of 18 individuals with classic reflux symptoms of heartburn and acid regurgitation (Gastroenterology 1995;108:1739–1744). Furthermore, there was a significant correlation between the duration of reflux episodes in the distal most sensor and the extent of the reflux episodes. Symptom perception is increased when acidic reflux reaches the proximal esophagus in patients with both erosive and nonerosive reflux disease, but this seems to be accentuated in patients with nonerosive reflux disease (Aliment Pharmacol Ther 2003;18:605–613). Bredenoord et al used multichannel impedance pH recordings to explore determinants of symptoms in a group of 32 patients with classic reflux symptoms when taken off of PPI therapy (Gut 2006;55:313–318). Several key observations deserve mention. First, only 203 of 1,807 reflux episodes were symptomatic, a finding that is typical in prolonged pH recordings. In contrast with the asymptomatic reflux events, the symptomatic episodes were associated with a larger pH drop, lower nadir pH, higher proximal extent of reflux, longer acid and volume clearance times, and preceded by significantly higher cumulative acid exposure time. Interestingly, in this study of patients who were off of PPI therapy, weakly acidic reflux was responsible for few symptoms.

It is also important to hearken back to an interesting physiologic observation made during the recent era of endoscopic antireflux therapies. The Gatekeeper system, which was removed from the market, involved placement of 3–6 hydrogel prostheses into the submucosal layer of the esophagus at the squamocolumnar junction under endoscopic guidance. Cicala et al found that placement of the Gatekeeper system in 9 GERD patients resulted in a change in the intraesophageal distribution of acid 6 months after therapy (Gut 2005;54:183–186). These investigators found that although LES pressure and distal acid exposure time did not change in this small group of patients, the proximal extent of acid reflux, as measured by multichannel ambulatory pH monitoring, decreased significantly and was in the range of healthy subjects. This was accompanied by a decrease in the symptom score in 8 of the 9 patients. These findings suggested that this technique may have exerted its effect by decreasing the aperture through which reflux may flow, thus resulting in less proximal migration of the refluxate (Gut 2005;54:180–181).

The study by Zerbib et al provides further evidence that proximal spread of reflux is an important cause of reflux perception. Furthermore, this work now extends our appreciation of this concept from erosive GERD to nonerosive GERD and now to persistent GERD symptoms despite twice daily PPI therapy. These findings, in conjunction with recent work by Fass et al, who described enhanced perception of esophageal acid in patients with GERD after sleep deprivation (Gastroenterology 2007;133:1787–1795), give us better insight into potential causes of typical GERD symptoms in refractory patients.

However, it is important to keep in mind some of the limitations of this study. First, this group of 20 patients was a subset of 59 patients referred for testing for persistent GERD symptoms despite therapy. The other 39 patients were excluded because of either absence of symptoms in 4 patients or lack of symptom association with either the symptom index or symptom association probability during the study in the remaining 35 patients. As such, we still do not know the cause of symptoms in these patients. Furthermore, there was considerable overlap in proximal extent of reflux in the symptomatic and asymptomatic reflux episodes. It is also important to remember that there is no “gold standard” for the various symptom association techniques, because each of the techniques has advantages and disadvantages. This study also leaves unanswered the question of how we should decrease the proximal spread of reflux. Most would agree that antireflux surgery in these patients is an overly aggressive approach, especially given the risks of the procedure. Endoscopic approaches to GERD have fallen by the wayside despite the physiologic data obtained with the Gatekeeper device, suggesting that application of such a device can decrease proximal spread of reflux and thereby decrease symptoms. We also have few medical options. Baclofen decreases the frequency of transient LES relaxation, and may improve symptoms in patients with persistent reflux symptoms despite PPI therapy (Gut 2003;52:1397–1402). Newer agents targeting this mechanism are under active development, but are not yet ready for clinical practice.

At the end of the day, it is important to remember that there are a variety of causes of “GERD despite therapy.” The work by Zerbib et al highlights the importance of proximal spread of the reflux column along with sensitization of the esophagus by prior reflux events as a cause of heartburn symptoms. We eagerly await better therapeutic approaches to decrease proximal extent of the reflux in these patients.