In healthy subjects and in patients with mild to moderate gastro-oesophageal reflux disease, gastro-oesophageal reflux occurs mainly during transient lower oesophageal sphincter (LOS) relaxations.1-3 Transient LOS relaxations are a neural reflex, organised in the brain stem, with efferent and afferent pathways travelling in the vagus nerve.3 Distention of the proximal stomach is a major trigger for the reflex to occur, although stimulation of the pharynx or the larynx may also contribute.3 It is clear that the initiation of the reflex requires activation of gastric mechanoreceptors.

Because of their pivotal role in the occurrence of gastro-oesophageal reflux, the neurophysiology and pharmacology of transient LOS relaxations are topics of intense ongoing research. Atropine is one of the drugs that were recently shown to inhibit gastro-oesophageal reflux by inhibiting transient LOS relaxations.4 It is unclear if atropine is acting at the level of the stomach, by altering the mechanosensitivity of the proximal stomach, or at the level of the brain stem, by interfering with central integrative processing.

In this issue of Gut, Lidums and colleagues5 used a gastric barostat procedure to study the influence of atropine on proximal gastric tone and on the occurrence of transient LOS relaxations in healthy subjects (see page 30). Atropine caused prolonged relaxation of the proximal stomach after a meal and decreased the rate of transient LOS relaxations. By comparing the effects on proximal gastric tone and on the rate of transient LOS relaxations, the authors concluded that the inhibitory effect of atropine on transient LOS relaxations was most likely at the central level.

The nature of the fundic mechanical receptors involved in triggering postprandial transient LOS relaxations is still poorly understood. Based on animal studies it has been proposed that mechanoreceptors are positioned either in series or in parallel to smooth muscle fibres. In parallel, mechanoreceptors respond to stimuli that elongate the stomach wall while in series, mechanoreceptors respond to stimuli that increase the tension within the stomach wall.6 ,7 Figure 1 illustrates the responses of tension mechanoreceptors (in series) and elongation mechanoreceptors (in parallel) to different stimuli (distention, relaxation, and contraction of smooth muscle). In series, mechanoreceptors are activated during distension and during contraction against a resistance; they are inactivated during relaxation. In parallel, mechanoreceptors are activated during distension and relaxation and are inactivated during contraction. Animal studies suggest that gastric mechanoreceptors with afferents in the vagal pathways are primarily tension receptors.7

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Schematic representation of types of gastric mechanoreceptors (schematically drawn as coils and positioned in reference to gastric smooth muscle cells) and their response to distention, muscular relaxation, and muscular contraction.

Gastric distention, the best studied trigger of transient LOS relaxations in humans, results in elongation as well as increased wall tension of the stomach. Hence it is unclear which of these two types of mechanoreceptors initiates the afferent limb of the reflex. Administration of atropine prolongs relaxation and thus elongation of the proximal stomach after a meal. Hence if the mechanoreceptors involved in triggering transient LOS relaxations were elongation receptors, atropine should have enhanced transient LOS relaxations.

The influence of administration of atropine on activation of tension receptors in the gastric wall after a meal is less clear. Attempts have been made to differentiate tension from elongation following the law of Laplace:

        T=ΔP(r/2) whereT=wall tension,P=pressure, andr=radius.8 In the present study, intragastric pressure was kept constant by the barostat device, and postprandial intrabag volume, and thus the radius, was maintained higher after atropine. If calculated according to the simplified law of Laplace, a higher wall tension after a meal results from administration of atropine. However, use of the Laplace formula requires a number of assumptions that are not necessarily fulfilled under the given circumstances: the wall of the stomach is infinitely thin, the intragastric balloon and stomach are assumed to have a perfect spherical shape, and the pressure external to the stomach is known and is evenly distributed. Most importantly, wall tension in the proximal stomach has both passive and active components. The Laplace formulas, most suitable in a passive tension model, do not take into account the modulatory effects of changes in the contractile state of the proximal stomach which may occur reflexly or in response to neurohumoral or pharmacological modulation. During a process of gastric relaxation, such as that which occurs during accommodation of a meal or after administration of atropine, wall tension is likely to fall because of diminution in the active component of wall tension. In the presence of an intragastric barostat bag, keeping intragastric pressure constant and above the level of intra-abdominal pressure and thus providing a continuous drive for low level distension, it is impossible to estimate the true effect on wall tension. However, it seems reasonable to assume that, under the conditions of the present study, wall tension was not decreased after a meal in the presence of atropine. Consequently, if the mechanoreceptors involved in triggering transient LOS relaxations were tension receptors, atropine should have enhanced transient LOS relaxations in the present study.

Based on these underlying assumptions, the authors correctly state that whatever the mechanoreceptors involved, under the conditions of the present study, relaxation of the proximal stomach would be expected to increase the rate of transient LOS relaxations. In support of this notion, fundic relaxation induced by sumatriptan was associated with an increased rate of transient LOS relaxations and gastro-oesophageal reflux.9 ,10 Furthermore, slow infusion of lipids into the duodenum induces gastric relaxation and increases the rate of transient LOS relaxations, both of which are prevented by pretreatment with a cholecystokinin receptor antagonist.11 Hence atropine induced inhibition of transient LOS relaxations is most likely due to its action at the level of the central nervous system. This does not preclude the possibility of influencing the occurrence of transient LOS relaxations through an effect on proximal gastric tone. A successful peripherally acting approach would, however, require knowledge of the type of mechanoreceptor involved in triggering transient LOS relaxations. In the case of a tension receptor, relaxation of the proximal stomach should decrease the occurrence of transient LOS relaxations. In the case of an elongation receptor, enhanced gastric emptying or redistribution of a meal to the distal stomach should decrease the rate of transient LOS relaxations. Identification of the type of mechanoreceptor involved in triggering transient LOS relaxations in humans is likely to require an innovative approach, avoiding the undistinguishing influence of gastric distention.