The mechanical model of gastroesophageal reflux:
what are the most important factors in causing gastro-esophageal reflux?
S.S. Kadirkamanathan, D.F. Evans, C.P. Swain (London)
Gastroesophageal reflux (GER) may be regarded as a mechanical process which occurs when the forces promoting reflux overcome those contributing to the reflux barrier. Negative intrathoracic pressure (ITP) and positive elevation of intra-gastric and intra-abdominal pressures (IGP and IAP) tend to promote reflux [1, 2]. Factors contributing to the reflux barrier include the lower esophageal sphincter (LES) , the diaphragm or the crural sling , gastric sling fibres , the angle of His  and the esophageal mucosal "choke" .
Investigations measuring some of these reflux parameters have demonstrated that there is an overlap between patients who reflux and those who do not. For example, patients with gastroesophageal reflux disease (GERD) could not be separated from controls on the basis of their LES pressure . Such discrepancies have made it difficult to study the contribution of each individual factor in the pathogenesis of reflux. Static manometric investigations cannot detect the physiological effects of trans-sphincteric pressure changes (Figure 1) that occur during normal activities such as breathing  and exercise , two factors which can often give rise to GER.
Figure 1. Pressure effects around gastroesophageal junction.
Because of difficulties in studying the role of multitude of factors influencing reflux systematically in patients, physical models have been designed to study reflux under controlled, experimental conditions which would be more difficult to simulate in man. Models offer the ability to perform large numbers of repeatable experiments to illustrate and explain the mechanics of reflux by controlled changes of the parameters known to influence GER.
Mechanical models have been used 1) to develop methods for LES pressure , 2) to test the efficiency of new measuring devices , 3) to investigate the role of different factors that cause or prevent reflux [13,14] and 4) to investigate the physics of closure mechanism of the LES . DeMeester et al.  built an in vitro model of GER to examine the contribution of LES pressure and the intra-abdominal length of esophagus to gastroesophageal competence. Fresh post-mortem human esophageal tissue was used to function as the LES which was placed inside a water filled chamber representing the intra-abdominal compartment. The authors concluded that the presence of an intra-abdominal esophagus was necessary for the competence of the cardia and recommended that any anti-reflux operation should aim to restore 2.5-3.5 cm of intra-abdominal esophageal segment.
A modification of this model  was subsequently used to study the role of total sphincter length in the competence of the cardia. Post-mortem canine esophagus was used as the LES and an electrolytic transducer, positioned across the sphincter was used to measure the sphincter competence. Studies using this model suggested that the competence of the cardia was related to the combined effect of LES pressure and length. It also demonstrated that the diameter of the cardia adversely affected the degree of competence achieved by a given sphincter length suggesting that short sphincter lengths are less likely to resist small degrees of gastric dilation.
Petterson et al.  used two in vitro models of LES to study the opening and closing mechanisms of LES. Canine esophagi and flaccid rubber tubes were used to represent the LES in these models. The first was a modification of that described by DeMeester et al.  while the other used ligatures with graded tension at the GE junction to simulate the sphincter. The first model was used to study the opening and closure mechanisms of LES and the sealing property of the esophageal mucosa. The second model was used to study the effect of gastric wall tension on the competence of LES. The studies showed that gastric distention increased the gastric wall tension which resulted in a decrease in the effective LES length compromising the sphincter competence. All these models suggest that distal esophageal sphincter function is a mechanical rather than a pharmacological process.
Our research group developed a new mechanical model which was designed to allow more flexible simulation of factors (Figure 1) influencing GER . Unlike previous models, this model had the following features. It can simulate the physiological pressure changes in the intrathoracic and intra-abdominal compartments seen during activities such as breathing and exercise and pressures in the intra-abdominal and intragastric compartments can be controlled independently. The sphincter position can also be varied with respect to thoracic and abdominal compartments so that hiatus hernia can be simulated.
Figure 2. Model design.
A model esophagus (Figure 2), stomach and LES were constructed from latex tubing and placed inside a two-chamber container representing the thorax and abdomen. LES length, pressure and position could be changed relative to thorax and abdomen. IAP and ITP could be varied rhythmically with mechanical respirators to simulate respiration and exercise. The model was used to study the mechanism of GER, in particular the role played by LES position and the pressure changes in the compartments surrounding the sphincter during respiration and exercise. The effect of selective increases in IGP and IAP on GER were also studied.
The results using this dynamic model suggested that simulation of both breathing and exercise were associated with more reflux than that observed during the static experiments using identical preset parameters of LES length and pressure and pressures in the intragastric, intra-abdominal and intrathoracic compartments. The reflux rate was inversely proportional to the LES length and this effect was increased during breathing and exercise simulation. These dynamic experiments clearly showed that a short sphincter length remains an important contributory factor to the pathogenesis of reflux especially when the pressure changes of breathing and exercise were applied. Experiments performed to study the effect of sphincter position revealed that an abdominal sphincter afforded more protection against reflux than a thoracic sphincter. A 1 and 2 cm sphincters were more competent when the sphincter was in the abdominal compartment than when it was positioned entirely in the thorax (Figure 3 A, B). Interestingly, there was no measurable difference in the competence of a 3 cm sphincter whether the sphincter compartment was in the abdominal or thoracic compartment (Figure 3 C). This may explain why some patients with large hiatus hernia have been observed to have no reflux as a long intrathoracic sphincter can perform as effectively as a sphincter of the same length in the abdomen. A further finding was that the stepwise increase in IGP caused more reflux when compared with a similar rise in IAP (Figure 4). This was very apparent when the LES was partly or wholly positioned in the abdominal compartment. However, the effects of identical increases in IGP and IAP were similar when the sphincter length was short (1 cm) and LES entirely in the thoracic compartment. These results can be explained by the augmentation of LES pressure by the increase in IAP when the sphincter is partly or wholly in the abdomen. This protective effect is lost when the sphincter is entirely in the thorax.
Figure 3. Effect of sphincter position on reflux. A. Effect of sphincter length and position on flow rate (sphincter length = 1 cm). B. Effect of sphincter length and position on flow rate (sphincter
length = 2 cm). C. Effect of sphincter length and position on flow rate (sphincter
length = 3 cm).
Figure 4. Effects of increases in IGP and IAP.
A. Intra-abdominal sphincter (sphincter length = 1cm, sphincter pressure = 25 mmHg). B. Intrathoracic sphincter (sphincter length = 1cm, sphincter pressure = 25 mmHg).
All models of GER described have some limitations. The diaphragm when modeled is an immobile structure and there is no allowance made for the contraction of the crural sling. The LES whether constructed from cadaveric esophageal tissue or artificial materials cannot relax or contract to mimic normal LES relaxations seen in life. The esophageal body cannot contract and therefore it cannot mimic the esophageal clearance mechanism. It might be feasible to build a more elaborate model to incorporate some of the physiological features listed above.
10. Soffer EE, Merchant RK, Duethman J, Launspach RN, Gisolfi C, Adrian TE. Effect of graded exercise on esophageal motility and gastroesophageal reflux in trained athletes. Dig Dis Sci 1993;38:220-224.
13. DeMeester TR, Wernly JA, Bryant GH, Little AG, Skinner DB. Clinical and in vitro determinants of gastroesophageal competence: a study of the principles of antireflux surgery. Am J Surg 1979;137:39-46.