Article Text

Gas and liquid reflux during transient lower oesophageal sphincter relaxation
  1. C P BARHAM,
  1. Division of Surgery, Level 7
  2. Bristol Royal Infirmary
  3. Bristol BS2 8HW, UK
  1. D SIFRIM,
  1. J SILNY
  1. Centre for Gastroenterological Research
  2. University of Leuven
  3. Leuven, Belgium
  4. Department of Gastroenterology
  5. Royal Adelaide Hospital
  6. Adelaide, South Australia
  7. Helmoltz-Institute for Biomedical Engineering
  8. Aachen, Germany

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Editor,—We read the paper by Sifrimet al (Gut1999;44:47–54) with interest. Transient lower oesophageal sphincter relaxations (TLOSRs) have become widely accepted as the cause of most acid reflux episodes in health and disease. It is also accepted that TLOSRs are involved in the belch mechanism. This paper refutes our assertion that the majority of acid reflux events occur in healthy individuals in association with the venting of gas, based on pressure measurements in ambulant subjects.1 ,2

The current paper has shown that gas can be detected passing out of the stomach during TLOSRs. The authors found that only 18% of TLOSRs were solely gas reflux and nearly 60% of TLOSRs were associated with liquid reflux alone or neither the reflux of liquid or gas. This implies that spontaneous TLOSR occurs commonly and is not initiated by the fundal stretch “belch” mechanism. We wonder what exact teleological reason there can be for such a phenomenon.

The authors would suggest that in control subjects, acid reflux episodes are more common than simple belching. This seems contrary to most people’s daily experiences and differs notably from our observations in normal subjects, where common cavity events with acid reflux are much less frequent than those without acid reflux. The pressure pattern of acid reflux alone, is quite different from the two types of common cavity event which we identified. Presumably all of this might be accounted for by the relative insensitivity of the authors’ impedance system in the detection of small amounts of refluxed gas.

We remain unconvinced by the assertion that in normal subjects, acid reflux is a primary event, rather than one which accompanies the act of belching.



Editor,—We thank Drs Barham and Alderson for their comments on our paper. The issue of the relation between gas reflux, as in belching, and acid reflux has been a thorny one and the subject of some interest over the past few years. We accept that our findings challenge the popular concept that, in normal subjects, acid reflux commonly occurs in the setting of and perhaps as a consequence of belching. Because transient lower oesophageal sphincter relaxations (TLOSRs) are the basic underlying event for both reflux of gas (as in belching) and liquid (as in acid reflux), and because belching is considered a normal physiological process, it is natural to presume that, on occasions, acid sometimes escapes along with the gas.

The observations that have fuelled this conclusion, however, have been based on some assumptions about the interpretation of pressure tracings. In their paper on the mechanism of reflux in ambulant healthy subjects, Barham and colleagues assumed that the presence of an abrupt rise in intra-oesophageal pressure preceding the drop in pH indicated the reflux of gas. No direct detection of intra-oesophageal gas was undertaken. In our study we have used intraluminal impedance to detect gas, and pH to detect acid. It is clear from the impedance measurements that common cavities can occur in the absence of any detectable gas. Although we acknowledge that technical limitations such as the sampling rate may have limited the sensitivity of detection of rapidly moving small quantities of gas, intraluminal impedance is well capable of detecting substantial volumes of gas such as those that occur during belching. Therefore, we believe that our observations have a distinct methodological advantage over those of Barham and colleagues with regard to the patterns of gas reflux during TLOSR.

It is possible, therefore, that the apparent discrepancy between our findings and those of Barham et al may result from the technical limitations of their study. We found that gas accompanied liquid reflux on almost half (47%) of the acid reflux episodes, compared with 69% of acid reflux episodes that were associated with belching—that is, associated with a common cavity. Firstly, some of the common cavities observed by Barhamet al may have been pure liquid reflux. Secondly, they were unable to determine accurately the timing of the liquid and gas retroflow. It is possible that some of the presumed gas reflux actually started after the onset of the liquid reflux, as we observed in our study.

Drs Barham and Alderson have interpreted our findings as implying that “spontaneous TLOSR occurs commonly and is not initiated by the fundal stretch ‘belch’ mechanism”. Our findings do nothing of the sort. The gastric fundus can be stretched or distended by liquids and solid meals just as well as by gas, and all of these stimuli, as well as intragastric balloons, have been shown to increase the rate of TLOSRs. Intragastric gas is not essential for triggering TLOSRs. There is a low rate of TLOSR during fasting. The mechanisms responsible for this have not been properly defined but may well involve small degrees of fundal distension from the gastric air bubble that is usually present even in the “empty” stomach.

Although we believe that our data and their interpretation are valid, it is also important to acknowledge that there are other differences between our study and that of Barham et althat could have influenced the findings. Firstly, our study was performed in stationary sitting subjects, as opposed to ambulatory subjects. Secondly, the majority of our TLOSRs occurred in the postprandial period whereas those in the study by Barhamet al were gathered from substantial interdigestive as well as postprandial periods. The impact of ambulation on patterns of gas and liquid reflux has still to be studied.