Swallowing: neurophysiologic control of the esophageal phase

Coordination of Pharyngeal and Esophageal Phases of Swallowing

Although swallowing has been reviewed extensively, the coordination of the phases of swallowing have not. The phases are controlled by the brainstem, but peripheral factors help coordinate the phases. The occurrence, magnitude, and duration of esophageal phase depends upon peripheral feedback activated by the bolus. The esophageal phase does not occur without peripheral feedback from the esophagus. This feedback is mediated by esophageal slowly-adapting mucosal tension receptors through the recurrent and superior laryngeal nerves. A similar reflex mediated by the same peripheral pathway is the activation of swallowing by stimulation of the cervical esophagus. This reflex occurs primarily in human infants and animals, and this reflex may be important for protecting against aspiration after esophago-pharyngeal reflux. Not only are there inter-phase excitatory processes, but also inhibitory processes. A significant inhibitory process is deglutitive inhibition. When one swallows faster than peristalsis ends, peristalsis is inhibited by the new pharyngeal phase. This process prevents the ongoing esophageal peristaltic wave from blocking the bolus being pushed into the esophagus by the new wave. The esophageal phase returns during the last swallow of the sequence. This process is probably mediated by mucosal tension receptors through the superior laryngeal nerves. A similar reflex exists, the pharyngo-esophageal inhibitory reflex, but studies indicate that it is controlled by a different neural pathway. The pharyngo-esophageal inhibitory reflex is mediated by mucosal tension receptors through the glossopharyngeal nerve. In summary, there are significant peripheral processes that contribute to swallowing, whereby one phase of swallowing significantly affects the other.

The normal swallow: muscular and neurophysiological control

Swallowing is a complex physiologic function that involves precisely coordinated movements within the oral cavity, pharynx, larynx, and esophagus. This article reviews the anatomy, muscular control, and neurophysiological control of normal, healthy swallowing.

The fetal esophagus: anatomical and physiological ultrasonographic characterization using a high-resolution linear transducer

OBJECTIVE

To study the sonographic anatomy and physiology of the human fetal esophagus during the mid-trimester of pregnancy using a high-resolution linear transducer.

METHODS

This was a prospective observational study of the fetal esophagus between 19 and 25 weeks' gestation. The study was performed in 60 consecutive fetuses, after a normal anatomy scan, using a 5-13-MHz matrix array wide-band transducer. During the examination the collapsed esophagus was first visualized, and followed by a 5-min video recording in order to demonstrate luminal patency and peristaltic waves.

RESULTS

Complete anatomical visualization of the esophagus was possible in 52 (86.7%) patients and at least partial visualization in 58 (96.7%) patients. Three different patterns of esophageal motility were observed: a simultaneous and short opening of the whole esophagus was found in 35 (58.3%) fetuses; a segmental, peristalsis-like movement from the pharynx, through the mediastinum, and into the stomach was found in 18 (30%) fetuses; and in one fetus reflux-like passage of solid contents from the stomach was observed. The mean time required for demonstration of esophageal patency was 96.1 (range, 10-300) s.

CONCLUSIONS

Demonstration of normal anatomy and physiological activity of the fetal esophagus is feasible using appropriate transducers. The most commonly observed pattern of esophageal motility in the mid-trimester of pregnancy is the simultaneous relaxation of the upper and lower esophageal sphincters with concurrent opening of the esophageal lumen from the upper thorax to the stomach. Demonstration of a patent esophagus may be helpful in fetuses with suspected esophageal atresia.

A mathematical model for estimating muscle tension in vivo during esophageal bolus transport

We present a model of esophageal wall muscle mechanics during bolus transport with which the active and "passive" components of circular muscle tension are separately extracted from concurrent manometric and videofluoroscopic data. Local differential equations of motion are integrated across the esophageal wall to yield global equations of equilibrium which relate total tension within the esophageal wall to intraluminal pressure and wall geometry. To quantify the "passive" (i.e. inactive) length-tension relationships, the model equations are applied to a region of the esophagus in which active muscle contraction is physiologically inhibited. Combining the global equations with space-time-resolved intraluminal pressure measured manometrically and videofluoroscopic geometry data, the passive model is used to separate active and "passive" components of esophageal muscle tension during bolus transport. The model is of general applicability to probe basic muscle mechanics including the space-time stimulation of circular muscle, the relationship between longitudinal muscle tension and longitudinal muscle shortening, and the contribution of the collagen matrix surrounding muscle fibers to passive tension during normal human esophageal bolus transport and in pathology. Example calculations of normal esophageal function are given where active tone is found to extend only over a short intrabolus segment near the bolus tail and segmental regions of active muscle squeeze are demonstrated.

Brain stem control of swallowing: neuronal network and cellular mechanisms

Swallowing movements are produced by a central pattern generator located in the medulla oblongata. It has been established on the basis of microelectrode recordings that the swallowing network includes two main groups of neurons. One group is located within the dorsal medulla and contains the generator neurons involved in triggering, shaping, and timing the sequential or rhythmic swallowing pattern. Interestingly, these generator neurons are situated within a primary sensory relay, that is, the nucleus tractus solitarii. The second group is located in the ventrolateral medulla and contains switching neurons, which distribute the swallowing drive to the various pools of motoneurons involved in swallowing. This review focuses on the brain stem mechanisms underlying the generation of sequential and rhythmic swallowing movements. It analyzes the neuronal circuitry, the cellular properties of neurons, and the neurotransmitters possibly involved, as well as the peripheral and central inputs which shape the output of the network appropriately so that the swallowing movements correspond to the bolus to be swallowed. The mechanisms possibly involved in pattern generation and the possible flexibility of the swallowing central pattern generator are discussed.

Silent cerebral infarction: a potential risk for pneumonia in the elderly

OBJECTIVE

To determine whether patients who have silent cerebral infarction are more likely to develop pneumonia than are controls without silent cerebral infarction.

DESIGN

We examined 269 community-residing participants of the senior day-care centre without history of previous stroke, and then followed them over a two-year period to assess pneumonia. On the basis of computerized tomography scans, they were divided into two groups: no infarction (n = 102) and cerebral hemispheric infarction (n = 167). Cerebral infarcts were further divided into deep and superficial infarcts.

RESULTS

The incidence of pneumonia was significantly higher in subjects with silent cerebral infarction (19.8%) than in controls (4. 9%) (odds ratio, 4.67 [95% CI, 1.87-11.67]; P < 0.01). Deep infarcts were more closely associated with the incidence of pneumonia (29.1%) than superficial infarcts (7.6%) (odds ratio, 5.00 [CI, 1.91-13.08]; P < 0.01).

CONCLUSIONS

Elderly subjects with silent cerebral infarction were more likely to develop pneumonia than were controls without silent cerebral infarction. Amongst hemispheric silent cerebral infarcts, those located in the deep brain structures may be an important predictor of the development of pneumonia.

Myogenic mechanism for peristalsis in the cat esophagus

A myogenic control system (MCS) is a fundamental determinant of peristalsis in the stomach, small bowel, and colon. In the esophagus, attention has focused on neuronal control, the potential for a MCS receiving less attention. The myogenic properties of the cat esophagus were studied in vitro with and without nerves blocked by 1 microM TTX. Muscle contraction was recorded, while electrical activity was monitored by suction electrodes. Spontaneous, nonperistaltic, electrical, and mechanical activity was seen in the longitudinal muscle and persisted after TTX. Spontaneous circular muscle activity was minimal, and peristalsis was not observed without pharmacological activation. Direct electrical stimulation (ES) in the presence of bethanechol or tetraethylammonium chloride (TEA) produced slow-wave oscillations and spike potentials accompanying smooth muscle contraction that progressed along the esophagus. Increased concentrations of either drug in the presence of TTX produced slow waves and spike discharges, accompanied by peristalsis in 5 of 8 TEA- and 2 of 11 bethanechol-stimulated preparations without ES. Depolarization of the muscle by increasing K(+) concentration also produced slow waves but no peristalsis. We conclude that the MCS in the esophagus requires specific activation and is manifest by slow-wave oscillations of the membrane potential, which appear to be necessary, but are not sufficient for myogenic peristalsis. In vivo, additional control mechanisms are likely supplied by nerves.

Brain stem localization of rodent esophageal premotor neurons revealed by transneuronal passage of pseudorabies virus

BACKGROUND/AIMS

Brain stem premotor neurons control swallowing through contacts with both afferent neurons and motoneurons. The location and connectivity of premotor neurons innervating the esophagus was determined using pseudorabies virus.

METHODS

In 30 rats, viral injections were made into either the cervical or subdiaphragmatic esophagus, cricothyroid muscle, or stomach. After a 48-62-hour survival, brain sections were processed immunocytochemically for the virus.

RESULTS

Neuronal labeling was limited to the compact formation of the nucleus ambiguus for survivals of 48-54 hours. At 57-62-hour survivals, virus-labeled second-order neurons (premotor neurons) were localized to the central subnucleus of nucleus of the solitary tract. Injections in the cricothyroid muscle and stomach resulted in distinct patterns of motoneuronal labeling in the nucleus ambiguus and dorsal motor nucleus and premotor neuronal labeling in the nucleus of the solitary tract.

CONCLUSIONS

Virus-labeled premotor neurons in the nucleus of the solitary tract occurred as a result of retrograde transport of the virus from the nucleus ambiguus because no viral antigen was present in the tractus solitarius. The esophagus is controlled by a central circuit whereby esophageal vagal afferents terminate on premotor neurons in the central subnucleus that in turn innervate esophageal motoneurons in the nucleus ambiguus.

Mechanical studies of the esophageal function

The discussion centers on the use of mechanical principles, mathematical modeling, and concurrent manometric and videofluoroscopic data to study the esophageal function. Basic principles of mechanics indicate that intrabolus pressure must be distinguished from the direct contractile squeeze of the circular muscle on the manometric assembly. Because these two regions are mechanically distinct, pressure amplitude is not a proper indicator of the forces characterizing esophageal bolus transport. In the application of computer simulations to the transport of a fluid bolus through the aortic arch regions, it was discovered that separate contraction waves must exist in the upper and lower esophageal segments when bolus retention occurs. Through detailed analysis of enhanced concurrent manometric and videofluoroscopic data in human volunteers, we have found that a dual-wave characteristic across the transition zone is a normal reflection of the change in muscle types, each muscle type producing a separate contraction wave. In normal transport, these two contraction waves are properly coordinated spatially and temporally. However, during bolus retention, a mismatch in space and time between these two waves takes place. Analysis suggests that this mismatch is neurological rather than histological in origin, and occurs primarily within the lower smooth-muscle segment.

The brainstem esophagomotor network pattern generator: a rodent model

The evidence reviewed in this essay supports the following working model of the central function generator for esophageal peristalsis in the rat: solitarial subnucleus centralis (NTSc) neurons operate in a dual capacity as esophagomotor reflex interneurons and as command neurons programming respective outputs from nucleus ambiguus compact formation (AMBc) motoneurons during secondary and primary peristalsis. In both conditions, there is a critical requirement for cholinergic input which enables NTSc neurons to generate the timed sequence of AMBc motoneuronal activity. In primary peristalsis, the cholinergic coupling mechanism is activated centrally, probably via projections from deglutitive premotor neurons to the parvicellular reticular formation and thence to the NTS. In reflex (or secondary) peristalsis, the cholinergic input could in part be generated by cholinergic vagal viscerosensory fibers innervating the esophagus. Postulated connections between NTS deglutitive neurons and the parvicellular cholinergic neurons of the intermediate reticular formation have yet to be demonstrated. Premotor input from NTSc to AMBc is generated by somatostatinergic and excitatory aminoacidergic neurons. Coactivation of both inputs by cholinergic afferents is necessary to generate esophagomotor output from AMBc neurons. The model under study is derived from investigations into central mechanisms governing striated muscle peristaltic activity. Whether the basic operational principles revealed thus far apply to peristaltic pattern generation in species with a smooth muscle esophagus, requires further investigation.

Oesophageal reflex responses: abnormalities of the enteric nervous system in patients with oesophageal symptoms

An intraluminal balloon was used to study the peristaltic reflex, which is mediated by the intrinsic nerves of the oesophagus. Serial balloon distension was performed in nine asymptomatic volunteers and 133 patients with oesophageal symptoms. Eight of the volunteers had a normal response with proximal stimulation and distal inhibition of motility. Only 42 patients (31.6 per cent) had a normal response. The commonest abnormal response (39.1 per cent) was some form of failure of the distal inhibitory reflex. Other patterns of abnormality were an unresponsive oesophagus (15.8 per cent) with no motility change during balloon inflation, or spasm (13.5 per cent) proximal to the balloon. These alterations of secondary peristaltic activity suggest that there are abnormalities of the intrinsic (enteric) nerves of the oesophagus. Different abnormalities were found in patients with similar symptoms. Awareness of this difference might allow a more rational approach to treatment. This hypothesis was tested in a small pilot study treating functional dysphagia with cisapride. Three of nine patients had marked symptomatic improvement within 4 weeks and all three had an unresponsive oesophagus. The remaining six patients, who had failure of distal inhibition or a normal response, did not improve.

Neuropharmacologic correlates of deglutition: lessons from fictive swallowing

Pharmacologic investigations into the transmission processes underlying fictive swallowing in the rat have disclosed the potential diversity of chemical signals used in central deglutitive pathways. Monoaminergic mechanisms appear to serve as links between subcortical structures and the medullary pattern generator of swallowing (PGS), and may play a critical role in maintaining internal facilitatory drive, required by the PGS for optimal responsivity to peripheral sensory input. Cholinergic bulbar interneurons form an integral component of the PGS subnetwork controlling esophageal peristalsis. Local GABA neurons exert a tonic inhibition of the buccopharyngeal stage, may regulate buccopharyngeal-esophageal coupling, and may contribute to peristaltic rhythmic generation at both the premotoneuronal and motoneuronal level. Receptor subtypes for excitatory amino acids (glutamate, aspartate) are differentially associated with deglutitive premotoneurons for both the buccopharyngeal and esophageal stage, as well as with ambiguus motoneurons. Preliminary evidence suggests the existence of excitatory peptidergic mechanisms involving thyrotropin-releasing hormone, vasopressin, oxytocin, and somatostatin, a probable candidate for excitatory transmitter in the solitarioambigual internuncial projection to motoneurons innervating esophageal striated musculature. Further validation of this experimental model may ultimately help to establish a framework for the clinical recognition, management, and exploitation of drug actions on central deglutitive neuroeffectors.

The otolaryngologic manifestations of gastroesophageal reflux disease (GERD): a clinical investigation of 225 patients using ambulatory 24-hour pH monitoring and an experimental investigation of the role of acid and pepsin in the development of laryngeal injury

Occult (silent) gastroesophageal reflux disease (GER, GERD) is believed to be an important etiologic factor in the development of many inflammatory and neoplastic disorders of the upper aerodigestive tract. In order ot test this hypothesis, a human study and an animal study were performed. The human study consisted primarily of applying a new diagnostic technique (double-probe pH monitoring) to a population of otolaryngology patients with GERD to determine the incidence of overt and occult GERD. The animal study consisted of experiments to evaluate the potential damaging effects of intermittent GER on the larynx. Two hundred twenty-five consecutive patients with otolaryngologic disorders having suspected GERD evaluated from 1985 through 1988 are reported. Ambulatory 24-hour intraesophageal pH monitoring was performed in 197; of those, 81% underwent double-probe pH monitoring, with the second pH probe being placed in the hypopharynx at the laryngeal inlet. Seventy percent of the patients also underwent barium esophagography with videofluoroscopy. The patient population was divided into seven diagnostic subgroups: carcinoma of the larynx (n = 31), laryngeal and tracheal stenosis (n = 33), reflux laryngitis (n = 61), globus pharyngeus (n = 27), dysphagia (n = 25), chronic cough (n = 30), and a group with miscellaneous disorders (n = 18). The most common symptoms were hoarseness (71%), cough (51%), globus (47%), and throat clearing (42%). Only 43% of the patients had gastrointestinal symptoms (heartburn or acid regurgitation). Thus, by traditional symptomatology, GER was occult or silent in the majority of the study population. Twenty-eight patients (12%) refused or could not tolerate pH monitoring. Of the patients undergoing diagnostic pH monitoring, 62% had abnormal esophageal pH studies, and 30% demonstrated reflux into the pharynx. The results of diagnostic pH monitoring for each of the subgroups were as follows (percentage with abnormal studies): carcinoma (71%), stenosis (78%), reflux laryngitis (60%), globus (58%), dysphagia (45%), chronic cough (52%), and miscellaneous (13%). The highest yield of abnormal pharyngeal reflux was in the carcinoma group and the stenosis group (58% and 56%, respectively). By comparison, the diagnostic barium esophagogram with videofluoroscopy was frequently negative. The results were as follows: esophagitis (18%), reflux (9%), esophageal dysmotility (12%), and stricture (3%). All of the study patients were treated with antireflux therapy. Follow-up was available on 68% of the patients and the mean follow-up period was 11.6 +/- 12.7 months.(ABSTRACT TRUNCATED AT 400 WORDS)

Central neural control of esophageal motility: a review

We review recent studies on the central neural control of esophageal motility, emphasizing the anatomy and chemical coding of esophageal pathways in the spinal cord and medulla. Sympathetic innervation of the proximal esophagus is derived primarily from cervical and upper thoracic paravertebral ganglia, whereas that of the lower esophageal sphincter and proximal stomach is derived from the celiac ganglion. In addition to noradrenaline, many sympathetic fibers in the esophagus contain neuropeptide Y (NPY), and both noradrenaline and NPY appear to decrease blood flow and motility. Preganglionic neurons innervating the cervical and upper thoracic ganglia are located at lower cervical and upper thoracic spinal levels. The preganglionic innervation of the celiac ganglion arises from lower thoracic spinal levels. Both acetylcholine (ACh) and enkephalin (ENK) have been localized in sympathetic preganglionic neurons, and it has been suggested that ENK acts to pre-synaptically inhibit ganglionic transmission. Spinal afferents from the esophagus are few, but have been described in lower cervical and thoracic dorsal root ganglia. A significant percentage contain calcitonin gene-related peptide (CGRP) and substance P (SP). The central distribution of spinal afferents, as well as their subsequent processing within the spinal cord, have not been addressed. Medullary afferents arise from the nodose ganglion and terminate peripherally both in myenteric ganglia, where they have been postulated to act as tension receptors, and, to a lesser extent, in more superficial layers. Centrally, these afferents appear to end in a discrete part of the nucleus of the solitary tract (NTS) termed the central subnucleus. The transmitter specificity of the majority of these afferents remains unknown. The central subnucleus, in turn, sends a dense and topographically discrete projection to esophageal motor neurons in the rostral portion of the nucleus ambiguous (NA). Both somatostatin-(SS) and ENK-related peptides have been localized in this pathway. Finally, motor neurons from the rostral NA innervate striated portions of the esophagus. In addition to ACh, these esophageal motor neurons contain CGRP, galanin (GAL), N-acetylaspartylglutamate (NAAG), and brain natriuretic peptide (BNP). The physiological effect of these peptides on esophageal motility remains unclear. Medullary control of smooth muscle portions of the esophagus have not been thoroughly investigated.