Characterization and mechanism of the esophago-esophageal contractile reflex of the striated muscle esophagus

Abstract

An esophago-esophageal contractile reflex (EECR) of the cervical esophagus has been identified in humans. The aim of this study was to characterize and determine the mechanisms of the EECR. Cats (n = 35) were decerebrated, electrodes were placed on pharynx and cervical esophagus, and esophageal motility was recorded using manometry. All areas of esophagus were distended to locate and quantify the EECR. The effects of esophageal perfusion of NaCl or HCl, vagus nerve or pharyngoesophageal nerve (PEN) transection, or hexamethonium administration (5 mg/kg iv) were determined. We found that distension of the esophagus at all locations activated EECR rostral to stimulus only. EECR response was greatest when the esophagus 2.5–11.5 cm from cricopharyngeus (CP) was distended. HCl perfusion activated repetitively an EECR-like response of the proximal esophagus only within 2 min, and after ~20 min EECR was inhibited. Transection of PEN blocked or inhibited EECR 1–7 cm from CP, and vagotomy blocked EECR at all locations. Hexamethonium blocked EECR at 13 and 16 cm from CP but sensitized its activation at 1–7 cm from CP. EECR of the entire esophagus exists, which is directed in the orad direction only. EECR of striated muscle esophagus is mediated by vagus nerve and PEN and inhibited by mechanoreceptors of smooth muscle esophagus. EECR of smooth muscle esophagus is mediated by enteric nervous system and vagus nerve. Activation of EECR of the striated muscle esophagus is initially sensitized by HCl exposure, which may have a role in prevention of supraesophageal reflux.

NEW & NOTEWORTHY An esophago-esophageal contractile reflex (EECR) exists, which is directed in the orad direction only. EECR of the proximal esophagus can appear similar to and be mistaken for secondary peristalsis. The EECR of the striated muscle is mediated by the vagus nerve and pharyngoesophageal nerve and inhibited by mechanoreceptor input from the smooth muscle esophagus. HCl perfusion initially sensitizes activation of the EECR of the striated muscle esophagus, which may participate in prevention of supraesophageal reflux.

INTRODUCTION

The physiology of the striated muscle portion of the esophagus has received little attention in the literature, even though it exists in a very critical location in the esophagus and the striated muscle esophagus has a very different anatomy and physiology from smooth muscle esophagus. This lack of knowledge can lead to misunderstanding of esophageal function and dysfunction. One esophageal response that has received considerable attention in the smooth muscle portion of the esophagus has received almost no attention in the striated muscle portion of the esophagus, and this is the effect of esophageal distension.

Distension of the esophagus has been shown to activate reflex contraction of the esophagus above the point of distension in three ways depending on the site distended and the species. Distension of the smooth muscle portion of the esophagus has two effects on the smooth muscle esophagus: 1) short-term contraction mediated by the enteric nervous system (ENS) (3, 4), referred to as the “on-response,” and 2) a contraction lasting as long as the stimulus mediated by a central nervous system and the vagus nerve (16). The on-response has been demonstrated in a species, i.e., opossum (3, 4), which has a long smooth muscle portion of the esophagus, and it occurs mostly in the most distal portions of the esophagus. The vagally mediated esophageal smooth muscle response occurs in the more rostral portion of the smooth muscle esophagus (2, 16). The third esophageal response to esophageal distension is contraction of the striated muscle portion of the esophagus. This striated muscle response has been observed in two human studies (5, 20), but its characteristics and mechanisms have not been investigated.

Distension of all parts of the esophagus activate a reflex contraction of the upper esophageal sphincter (UES), i.e., the esophago-UES contractile reflex (EUCR). Although the mechanism of the EUCR has been determined (12), the relationship between the EUCR and the striated muscle esophageal response to esophageal distension is unknown. Additionally, studies have found that exposure of the esophagus to acid alters EUCR (13), but the effect on striated muscle esophageal response to esophageal distension is unknown.

The aims of this study were to characterize the nature and function of the esophago-esophageal contractile reflex of the striated muscle esophagus (EECRst), investigate the mechanisms of its initiation, and determine the relationship of EECRst to EUCR and the effects of esophageal acid exposure on the EECRst.

METHODS

These studies were conducted using domestic, short-hair decerebrate cats (n = 35; 17 male and 18 female) weighing 3.3 ± 0.3 kg (means ± SD). The cat is a particularly good model for investigation of human esophageal physiology because it is a mammal that has both smooth and striated muscle esophagus, whereas most other mammalian research animals, e.g., rat and dog, do not. Additionally, the cat is an especially good experimental model for this experiment because it has a long striated muscle portion of the esophagus (2), and the purpose of this study was to investigate a striated muscle esophageal reflex response. All studies were conducted with the approval of the Institutional Animal Care and Use Committee of the Medical College of Wisconsin (MCW).

The cats were housed in the Biomedical Resource Center (BRC) of MCW. The MCW BRC has accreditation of the Assessment and Accreditation of Laboratory Animal Care and an Animal Welfare Assurance number on file with the Office of Laboratory Animal Welfare (OLAW). OLAW is responsible for ensuring that the guidelines of the Health Research Extension Act are followed. The cats were housed by sex in group housing in a large room with multiple enrichment items and fed ad libitum. The cats were given 1 wk acclimatization and housed for 1–4 wk. Other details of acquisition and housing are listed in the Supplemental Data (https://doi.org/10.6084/m9.figshare.8307578).

Animal Preparation

Just as the cat was selected to obtain a model of the esophageal physiology most similar to the human, decerebration was selected as the immobilization method best able to preserve the physiology of the awake state. Anesthetics and analgesics significantly inhibit all bodily functions, and they are eliminated or minimized as part of the decerebration process. In a prior study (12), we provided evidence that the sensitivity of activation of the esophageal reflexes in the decerebrate cat is similar to those of the awake human.

Cats were fasted overnight and decerebrated the next morning. The animals were anesthetized (3% isoflurane), trachea cannulated, and the carotid arteries ligated. The skull was exposed, and a hole over a parietal lobe was made using a trephine. The hole was enlarged using rongeurs, the central sinus was ligated and cut, and the brain was severed midcollicularly. The forebrain was then suctioned out of the skull, and the blood vessels of the Circle of Willis were coagulated by suction through cotton balls soaked in warm saline. The bony sinuses were filled with bone wax, the exposed brain was covered with paraffin oil-soaked cotton balls, and the skin over the skull was sewn closed. The animals were then placed supine on a heating pad (Harvard Homeothermic monitor), and the body temperature maintained between 38 and 40°C.

After decerebration, the femoral vein was cannulated for infusion of saline and the femoral artery cannulated to record arterial blood pressure. The larynx and pharynx were exposed and electromyography (EMG) electrodes were placed on the cricopharyngeus (CP; n = 35), thyropharyngeus (TP; n = 8), and thyrohyoideus (n = 13). The CP is the primary muscle (11) of the UES, and its actions are considered responses of the UES.

Although our primary interest was the EECRst, it is not possible to distinguish smooth from striated muscle esophagus by visual observation, and the change from one type of muscle to the other occurs over a transition zone. Therefore, we investigated the entire esophagus with the expectation that the smooth and striated muscle responses of the esophagus could be distinguished based on the results of the study. Two sets of studies were conducted with different sets of recordings of the esophagus. One set had EMG electrodes on the esophagus at 1 (E1; n = 22), 3 (E3; n = 16), and 5 (E5; n = 15) cm from the CP, and manometry was recorded at 8 (M8; n = 10), 11 (M11; n = 10), and 14 (M14; n = 10) cm from the CP. In the second set, EMG electrodes were placed on the esophagus at 2 (E2; n = 13) and 4 (E4; n = 17) cm from the CP, and manometry was recorded at 1 (M1; n = 13), 4 (M4; n = 13), 7 (M7; n = 13), 10 (M10; n = 12), 13 (M13; n = 12), and 16 (M16; n = 12) cm from the CP. The abdomen was then opened along the ventral midline, and a fistula of the proximal stomach was formed using a 3-mL plastic syringe that exited the abdominal cavity. This fistula was used to insert stimulation devices into the esophagus without disturbing the pharynx or larynx. The pressure transducer was inserted into the esophagus through the gastric fistula and positioned to record from sites listed above. The catheter was held in place by tying a suture attached to the transducer tip to the front teeth. The Supplemental Data has the complete accounting of the role of each cat in this study (https://doi.org/10.6084/m9.figshare.8307578).

Recording Techniques

Electromyography.

Bipolar Teflon-coated stainless steel wires (AS 632, Cooner Wire, Chatsworth, CA) bared for 2–3 mm were placed in each muscle, and the wires were fed into differential amplifiers (A-M Systems 1800). The electrical activity was filtered (bandpass of 0.1–3.0 KHz) and amplified (1,000–10,000 times) before feeding into the computer.

Blood pressure and esophageal manometry.

Blood pressure was recorded using a Statham pressure transducer, and esophageal manometry was recorded using a 3- or 6-channel solid-state pressure transducer with recording sites 3 cm apart (Gaeltec Devices). Both were attached to low-level, direct current preamplifiers (Grass P122) set at 3 Hz high-frequency cutoff filtration.

Computer data acquisition.

All data were acquired and analyzed using DATAQ Instruments data acquisition hardware and software.

Fluid Infusion

A catheter was placed into the esophagus through the gastric fistula such that the tip was at mid-esophagus, 7–9 cm from the UES. Fluid, either 0.9% NaCl (pH = 5.6) or 0.1 N HCl (pH = 1.2) was infused at 1 mL/min for 15–30 min (n = 17).

Neural Blockade

Pharyngoesophageal nerve.

The pharyngoesophageal nerve (PEN) was identified and isolated in deep dissection of the pharyngeal region and observed as it connects the nodose ganglion with the pharynx and esophagus. The PEN was transected bilaterally before it connected to the pharynx (n = 4).

Enteric nervous system.

The neural ganglia of the ENS were blocked (n = 4) by administration of hexamethonium chloride (5 mg/kg iv).

Vagus nerve.

The vagus nerve was identified in the neck at the level of the proximal cervical esophagus and transected bilaterally (n = 6). Three of these cats had previously received hexamethonium, and one had received hexamethonium and transection of the PEN.

Esophageal Mechanical Stimulation Techniques

The esophagus was stimulated by placing a 1-cm long balloon into the esophagus through the gastric fistula. The balloon was placed between the recording sites of the esophageal pressure transducer at 17, 14.5, 11.5, 8.5, 5.5, and 2.5 cm from the CP. The esophagus was stimulated in two ways: ramp pattern to determine the threshold stimulus, and square-wave pattern to determine the magnitudes of responses at different distension diameters and time delays to initiate the response. In the ramp technique, the balloon was inflated by injection of air from a 50-mL syringe attached to a Harvard infusion pump at a rate of 46 mL/min. The threshold pressures needed to activate the CP and esophagus were determined. In the square-wave technique, the balloon was inflated by hand using a 10-mL syringe to cause 1, 1.5, or 2.0 cm diameter distension in a square-wave fashion for 3–5 s. The distending pressure was calibrated to convert balloon pressure to diameter.

Protocol

The esophagus was distended to obtain control measurements at each esophageal location, as described above, and then the effects of fluid infusion or neural blockade were tested. The effects of fluid infusion of different pH on spontaneous esophageal motility during the fluid infusion as well as the effects after a period of fluid administration on the esophageal responses to esophageal distension were tested.

Statistics

The means and SE of the variables were calculated. Differences between control and experimental procedures or between two variables were tested using the Student’s t-test for normally distributed variables and Mann-Whitney U test for those that were not normally distributed. When the same animals were tested before and after a procedure, the paired Student’s t-test was used. For multiple comparisons, ANOVA was used. Linear regression was used to calculate the linear relationship of two variables. A P value of ≤ 0.05 was considered statistically significant.

RESULTS

Responses to Esophageal Distension

We found that distension of the esophagus activated esophageal responses that occurred simultaneously or propagated (Fig. 1). The propagated contractions moved to the most distal esophageal recording site and, as such, were considered secondary peristalsis (SP). The simultaneous contractions occurred at a short time delay from the stimulus and did not propagate. These simultaneous contractions were considered the contractile responses of the EECR. The balloon distension sometimes caused an artifactual decrease in pressure recorded by the manometric device just orad or caudad of the balloon (Fig. 1).

Fig. 1.

Esophago-esophageal contractile reflex (EECR). This figure shows the esophageal effects of distending the esophagus by 2 cm at each level [2.5–17 cm from the cricopharyngeus (CP)] of the esophagus compared with a spontaneous swallow. Distension of the esophagus causes simultaneous contraction of the esophagus above the stimulus only and sometimes also activates secondary peristalsis. The distending balloon sometimes causes an artifactual rapid reduction in pressure in nearby recording sites. Distension at 14.5 and 17 cm from the CP causes little effect on the esophagus from 2.5–10 cm from the CP. E no., electromyography recording no. of cm from the CP; Eso location, location of the balloon from the CP; M no., manometry recording no. of cm from CP.

We found that there were two different EECRs activated by esophageal distension, and they were activated from different esophageal regions (Figs. 1 and 2). When the esophagus was distended at 11.5 cm from the CP and sites proximal to this, the proximal EECR was strongly activated in the entire esophagus above the point of stimulation (Figs. 1 and 2). When the esophagus was distended at 14.5 or 17 cm from the CP, the EECR above these points up to 13 cm from the CP, i.e., distal EECR, were strongly activated, whereas more proximal sites, i.e., proximal EECR, were weakly activated (Figs. 1 and 2). The EECR at the site 10 cm from the CP seemed to respond to both stimulations and be part of both EECRs (Figs. 1 and 2).

Fig. 2.

Graphs of the effects of esophageal distension on the percentage occurrence and magnitude of esophago-esophageal contractile reflex (EECR). This is a graph of the percentage occurrences (A) and magnitudes (B) of EECR at 1, 4, 7, 13, and 16 cm from the cricopharyngeus (CP) caused by esophageal distension of 1, 1.5, and 2 cm at 2.5, 5.5, 8.5, 11.5, 14.5, and 17 cm from the CP; n = 10 cats. Two groups of esophageal responses occur. One set includes primarily activation of the EECR in the esophagus at 13 and 16 cm from the CP, which is caused by distension of the esophagus at 14.5 and 17 cm from the CP. Stimulation at these sites has little effect on activating EECR in the esophagus from 10–1 cm from CP. On the other hand, distension of the esophagus from 11.5–2.5 cm from the CP causes significant increase in activation of the EECR from 10–1 cm from the CP. Regardless of the esophageal location, distension of the esophagus does not activate EECR caudal to the stimulus, except for a few rare cases.

Esophageal distension always activated the EECR but only activated SP sometimes; however, both responses could be activated from all regions of the esophagus (Fig. 1). Activation of SP always occurred after EECR (Fig. 1 and 3). The initial portion of the SP often occurred simultaneously with the EECR (Figs. 1 and 3), which in some cases caused overlap of the contractile responses of the EECR and SP (Fig. 3). The reflex contraction of the CP, i.e., the EUCR, always occurred in association with the proximal or distal EECR (Figs. 1 and 3; Tables 1 and 2). We also found that in 50% (4 of 8) of the animals, the EECR was also accompanied by activation of the TP (Fig. 3) but never observed TP activation during the SP (Fig. 4).

Fig. 3.

Comparison of esophago-esophageal contractile reflex (EECR) with and without secondary peristalsis (SP). This figure shows the comparison of activation of EECR with and without activation of SP. The simultaneous nature of EECR with accompanying activation of the CP and thyropharyngeus (TP) is readily seen when EECR is activated alone. When SP accompanies EECR, the two responses merge in recordings from the cervical esophagus. Therefore, it is sometimes difficult to separate EECR from SP in the cervical esophagus even though they are different reflex responses. E no., electromyography recording no. of cm from the CP; Eso Dist, location of the balloon from the CP; M no., manometry recording no. of cm from CP.

Table 1.

Threshold stimulus at different locations in the esophagus needed to activate EECR at different locations

Stim Loc	CP	M1	M4	M7	M10	M13	M16
2.5	26 ± 2 (10)	27 ± 2 (10)
5.5	22 ± 4 (13)	22 ± 3 (12)	25 ± 3 (12)
8.5	20 ± 2 (14)	23 ± 3 (14)	23 ± 2 (14)	27 ± 3 (12)
11.5	20 ± 3 (12)	22 ± 2 (11)	23 ± 2 (11)	23 ± 2 (12)	27 ± 4 (8)
14.5	24 ± 3 (12)	26 ± 4 (7)	32 ± 4 (6)	31 ± 5 (6)	32 ± 4 (7)	34 ± 4 (9)
17	32 ± 4 (10)	30 ± 3 (4)	28 (1)	26 (1)	30 ± 5 (6)	34 ± 6 (5)	33 ± 6 (10)

Values are mean ± SE (n) in mmHg; n = number of cats. Stimulus and recording locations are cm from cricopharyngeus (CP). EECR, esophago-esophageal contractile reflex; M no., manometric recording site; Stim Loc, stimulus locations in the esophagus.

Table 2.

Time delay from stimulus (1.5 cm diameter) to response at different locations

cm from CP	CP	M1	M4	M7	M10	M13	M16
2.5	0.6 ± 0.1 (9)	0.6 ± 0.1 (11)
5.5	0.9 ± 0.1 (12)	1.0 ± 0.1 (12)	1.1 ± 0.1 (10)
8.5	0.9 ± 0.1 (11)	1.0 ± 0.1 (11)	1.1 ± 0.1 (12)	1.1 ± 0.1 (9)
11.5	1.2 ± 0.1 (12)	1.3 ± 0.1 (12)	1.4 ± 0.1 (14)	1.5 ± 0.1 (13)	1.5 ± 0.3 (4)
Prox Eso	P < 0.003	P < 0.001	P = 0.056	P < 0.02
14.5	1.2 ± 0.2 (10)	1.5 ± 0.2 (8)	1.5 ± 0.2 (6)	1.6 ± 0.1 (7)	1.6 ± 0.1 (4)	1.8 ± 0.2 (5)
17	1.1 ± 0.2 (6)	1.2 ± 0.3 (3)	1.5 ± 0.1 (2)	1.6 (1)	1.7 ± 0.2 (4)	1.8 ± 0.1 (3)	1.9 ± 0.1 (3)

Values are mean ± SE (n); n = number of cats. CP, cricopharyngeus; Eso, esophagus; M no., manometric recording site; P, probability of a relationship between distance from CP of the stimulus (2.5–11.5 cm from CP) and the time delay of the response using ANOVA for the proximal (Prox) esophago-esophageal contractile reflex.

Fig. 4.

Comparison of the effects of infusion of NaCl or HCl on esophageal responses. This figure shows the comparison of the esophageal responses activated by infusion of the esophagus with NaCl or HCl. NaCl infusion often activated repeated secondary peristalsis (SP), and HCl perfusion activated repeated esophago-esophageal contractile reflex of the striated muscle esophagus (EECRst)-like activity. The EECRst-like responses differed from the SP in that the EECRst-like responses did not propagate to or occur in the distal esophagus, and thyropharyngeus (TP) activation only occurred during the EECRst-like responses. E no., electromyography recording no. of cm from the CP; M no., manometry recording no. of cm from CP.

Proximal EECR

We found that when the esophagus was distended by 2 cm at and proximal to 11.5 cm from the CP, the EECRs activated at and proximal to 10 cm from the CP occurred at a 60–100% occurrence rate, whereas when this stimulus was applied at 14.5 and 17 cm from the CP, the percentage occurrence rate of the EECRs at and above 10 cm from the CP was ~10–60% (Fig. 2). The average percentage occurrence rate (Fig. 2A) and magnitude (Fig. 2B) of the EECR at 1, 4, and 7 cm from the CP were significantly higher when the esophagus was distended at and proximal to 11.5 cm from CP than at and distal to 14.5 cm from the CP.

The thresholds for activation of the proximal EECR tended to be higher when the stimulus was at the more distal locations (11.5 vs. more proximal sites), but this effect was not statistically significant (Table 1). The time delay to activation of the proximal EECR was significantly related to the distance of the stimulus away from the CP for all locations (Table 2), except for 4 cm from the UES. The P value for this relationship was just above (P = 0.056) statistical significance. When data were combined for responses that activated the proximal EECR, the relationship between distance of the stimulus from CP and time delay to activation of EECR was linearly related (y = 0.6x + 0.89; R = 0.946; P < 0.03; n = 19).

The increased threshold for activation of the EECR and the increased time delay of stimulus to response of the EECR as the stimulus is moved distally can give the EECR the appearance of propagation similar to SP, especially when the distending stimulus occurs at a slow rate (Fig. 5).

Fig. 5.

Comparison of esophago-esophageal contractile reflex (EECR) activated by ramp distension with secondary peristalsis (SP). This figure shows the effects of EECR activated by a ramp distension of the esophagus comparted with secondary peristalsis. Like the square-wave distension, the ramp distension activated the EECR only above the stimulus, but because of the increased threshold and time delay of occurrence of the EECR at more distal esophageal locations, the EECR exhibited the appearance of propagation. However, the EECR did not occur below the stimulus and the rate of the sequential activation of the EECR was much slower than SP propagation; therefore, the EECR did not propagate as the SP did. This figure shows that slow distension of the esophagus can activate a response above the distension that looks like SP but is not. E no., electromyography recording no. of cm from the CP; M no., manometry recording no. of cm from CP; Stim, stimulus.

Distal EECR

We found that esophageal distension of the esophagus at 14.5 and 17 cm from the CP activated the distal EECR primarily at 13 and 16 cm from the CP (Figs. 1 and 2). We did not observe relationships between stimulus strength or location and response delay or magnitude. However, the length of the esophagus over which the distal EECR occurred is short; therefore, the variability of the stimulus locations may have been too large compared with the difference between the means of responses to observe a relationship.

Effects of Fluid Infusion

We found that during HCl infusion, an EECR-like response was activated in a repetitive fashion (Fig. 4). This response only occurred in the proximal half of the esophagus and did not propagate as the SP did (Fig. 4). The delay to activation of this response with either NaCl or HCl was ~3 min, but HCl infusion had a significantly stronger effect, activating many more such events [18 ± 3 (16) vs. 3.0 ± 2.5 (5), P < 0.01] in many more animals (16 vs. 2). This EECR-like response, but not SP, was accompanied by activation of the TP (Fig. 4).

We found that activation of the EECR was inhibited by prolonged esophageal infusion of HCl but not NaCl (Fig. 6). The threshold stimuli needed to activate the EECR at sites 11.5 and 14.5 cm from the CP were significantly increased after exposure of the esophagus to 23 ± 2 (12) min of HCl but not 20 ± 3 (7) min of NaCl (Fig. 6).

Fig. 6.

Comparison of effects of NaCl or HCl on threshold to activate the esophago-esophageal contractile reflex (EECR). The graph shows that esophageal perfusion with HCl significantly increased the threshold for activation of the EECR at 11.5 and 14.5 cm from the cricopharyngeus (CP), whereas NaCl perfusion had no effect on threshold to activate EECR. E no. (n), electrode no. of cm from CP. *P < 0.05.

Effects of Neural Blockade

Hexamethonium (n = 4) blocked the distal EECR at 13 and 16 cm from the CP to esophageal distension of the esophagus at 14.5 and 17 cm from the CP (Fig. 7). On the other hand, hexamethonium administration significantly increased the response of the proximal EECR at 1 and 4 cm from the CP to distension of the esophagus at 17 cm from the CP (Fig. 7).

Fig. 7.

Effects of pharyngoesophageal nerve (PEN) transection or hexamethonium (Hex) administration on activation of the esophago-esophageal contractile reflex (EECR). This figure shows that transection of the PEN (n = 4 cats) significantly blocked activation of the EECR of the striated muscle esophagus (EECRst) at E1, and Hex (5 mg/kg) administration (n = 4) significantly blocked activation of the EECR mediated by smooth muscle esophagus at E13 and E16. In addition, Hex administration significantly increased the effectiveness of stimulation of the distal esophagus in activating EECRst in the proximal esophagus at E1 and E4. E no., electromyography recording no. of cm from the CP. *P < 0.05.

Transection of the PEN blocked esophageal distension-induced activation of the proximal EECR at 1 cm from the CP in all animals (n = 4), at 4 cm from the CP in 3 of 4 animals, and at 7 cm from the CP in 1–3 of 4 animals but had no effect on the EECR at 10, 13, and 16 cm from the CP (Fig. 7).

Transection of the vagus nerve (n = 6) blocked activation of the EECR at all levels of the esophagus in all animals (Table 3).

Table 3.

Effect of vagotomy on EECR magnitude activated by 2 cm distension

Eso Stim	CP	M1	M4	M7	M10	M13	M16
2.5	13 ± 4*	12 ± 4*
5.5	16 ± 2*	16 ± 3*	11 ± 6
8.5	19 ± 2*	18 ± 3*	15 ± 4*	15 ± 5*
11.5	16 ± 4*	12 ± 2*	11 ± 2*	23 ± 7*	18 ± 5*
14.5	24 ± 5*	11 ± 4*	8 ± 4*	12 ± 9	10 ± 6	15 ± 8*
17	16 ± 3*	4 ± 2*	2 ± 1	2 ± 1	4 ± 1	26 ± 8*	38 ± 10*

Values are means ± SE in mmHg and all are n = 6 cats. EECR, esophago-esophageal contractile reflex; M no., manometric recording site at no. cm from CP. The values are statistically (Mann-Whitney) compared with the effects of vagotomy, which are 0 ± 0 (6) at every esophageal stimulation and recording site.

^*P < 0.05.

DISCUSSION

We characterized a reflex, i.e., the proximal EECR, of the esophagus in the cat, which had previously been observed in humans (5, 20) but was not studied. Distension of the upper two-thirds of the esophagus in the cat activates simultaneous contraction of the esophagus above the stimulus, and this response was not blocked by hexamethonium. Therefore, this response is not mediated by nicotinic cholinergic receptors and thus is not mediated by ganglionic transmission. Ganglionic transmission is not part of the esophageal striated muscle neural pathway (2), and the upper two-thirds of the cat esophagus is striated muscle (2, 10); therefore, the proximal EECR is likely unique to the striated muscle esophagus. This may explain why this reflex has not been investigated in humans, as only the upper third of the human esophagus is striated muscle (7, 9, 15) and most esophageal studies in humans have focused on the smooth muscle portion of the human esophagus (17).

This striated muscle EECR (EECRst) activated from all areas of the esophagus was blocked by transection of the vagus nerves (Fig. 8). Transection of the PEN blocked the EECRst at 1 cm from the CP and inhibited the EECRst at 4 and 7 cm from the CP regardless of the area of esophagus stimulated. Therefore, the afferents for this reflex are probably in the vagus nerves (Fig. 8) and the efferents for the most rostral portions of this reflex are in the PEN because the PEN is a motor nerve of the CP and proximal esophagus in the cat (10) and the vagus nerve (Fig. 8) is the afferent nerve of the esophagus (18). It is likely that the recurrent laryngeal nerve (RLN) provides both afferent and efferent innervation of this reflex for more caudal regions of the cervical esophagus, as the RLN is the primary motor and sensory nerve supplying the cervical esophagus. Humans do not have a PEN (10), and most functions mediated by the PEN in cats are mediated by the RLN in humans (9, 10). Therefore, it is likely that the EECRst is mediated by the RLN in humans.

Fig. 8.

Diagram of neural control of esophago-esophageal contractile reflex (EECR). This diagram depicts the neural control of the EECR defined by studies in this manuscript and the literature. The diagram shows that the EECR is activated by stimulation of esophageal mechanoreceptors (12) of the striated or smooth muscle esophagus, transmission to the central nervous system (CNS) by vagus nerves (12), integration in the CNS, and activation of esophageal striated muscle vagal motor nerves, as defined by vagotomy and pharyngoesophageal nerve (PEN) transection. Smooth muscle activation due to smooth muscle distension includes the afferent CNS pathway, but also includes an efferent motor pathway that includes the enteric nervous system (ENS) (16) as defined by the effects of vagotomy (12, 16) and ganglionic blockade (16). Additionally, as defined by the effects of ENS blockade, distension of the distal smooth muscle activates ENS inhibition of the striated muscle CNS efferent neurons of the proximal striated smooth muscle, which inhibits the EECR.

In contrast to the EECRst, the distal EECR is mediated by smooth muscle esophagus (EECRsm) as it occurs in the smooth muscle section of the feline esophagus, and it is blocked by vagotomy and hexamethonium, i.e., a ganglionic blocker (Fig. 8). Whereas the EECRst is a newly characterized reflex, the EECRsm of the feline is likely the same vagally mediated response of the smooth muscle esophagus observed previously in the opossum (16). This previously defined smooth muscle contractile response occurs orad of the stimulus for a short distance above the point of esophageal distension and is mediated by the vagus nerves and muscarinic cholinergic receptors (16). The previously defined physiological aspects of the esophageal smooth muscle contractile response to esophageal distension in the opossum are similar to those of the EECRsm; therefore, we conclude that the EECRsm we recorded is the feline version of the previously defined smooth muscle contractile response to esophageal distension (16).

The activation pattern of the EECRst may lead to confusion in distinguishing EECRst from SP. The striated muscle portion of the human esophagus is short (15), and its response is rapid (6, 14), which can make it difficult to distinguish a peristaltic contraction from a simultaneous contraction. In humans, where the esophageal response is always recorded by manometry and/or video fluoroscopy, peristalsis is usually defined by observation of sequential activation of the pressure wave and/or movement of the bolus distally (6, 14). Movement of the bolus distally is not a reliable index of peristalsis in the cervical esophagus, i.e., striated muscle esophagus of humans, because even a simultaneous contraction of the cervical esophagus will push the bolus distally if the UES also closes, and the UES usually closes when there is contraction of the cervical striated muscle esophagus (5, 14). Additionally, our current findings also indicate that sequential activation may not be a reliable index of peristalsis. We found that the onset delays of the EECRst may occur in a sequential fashion, which makes them appear as if they propagated. A similar sequential activation pattern of the EECR was observed in a human study (5). Considering that the concept of peristalsis of the esophagus assumes a contraction whose propagation is controlled by the central or enteric nervous systems (8, 17), every contractile response of the cervical esophagus in humans that moves the bolus distally and appears to occur in a sequential fashion may not be peristalsis. Studies are needed in humans to clarify this issue.

The infusion of acid into the human esophagus activates a response (1) very similar to the repeated EECRst-like contractions that we observed in the current study. The nature of this response is unknown, but there are two possibilities. This repetitive EECRst-like response might be repeated secondary peristalses whose propagations are inhibited due to fluid infusion by a previously described esophago-esophageal inhibitory reflex (1). On the other hand, these EECRst-like responses might be repeated activations of the EECRst because this response 1) only occurs in the striated muscle portion of the esophagus, 2) does not propagate, and 3) is associated with activation of the TP. It is unknown why this EECRst-like response is repetitive, but it might be due to repetitive distension of the esophagus by the infusion of fluid, which dissipates as the stimulus, i.e., the infused fluid, is expelled from the esophagus. This process occurs time and time again depending on the rate of fluid infusion. More studies are needed to determine the nature and mechanisms of these repeated EECRst-like responses.

The EECRst, like the EUCR, closes the proximal esophagus and is activated, like the EUCR (12), by esophageal distension. The function of the EUCR is likely prevention of supraesophageal reflux, and this protective effect would be amplified by the EECRst. The repeated EECR-like response more readily occurs in gastroesophageal reflux disease patients than human controls (1). This observation is consistent with our finding in cats that exposure of the esophagus to acid activates the repeated EECRst-like response. Therefore, acid exposure may sensitize the receptors that mediate activation of the EECR. A similar sensitization of receptors that activate the EUCR (19) has also been observed in humans. These observations suggest that a function of both the EUCR and EECR is to prevent supraesophageal reflux of acid.

The EECRst, like the EUCR, is inhibited (13) by longer term exposure of the esophagus to HCl but not NaCl. We found in human studies that supraesophageal reflux disease patients (1) have decreased activation of EUCR as well as decreased esophageal responses to esophageal fluid infusion. Therefore, we hypothesize that the EUCR and EECRst may become desensitized with repeated acid exposure and that desensitization of the receptors that mediate these reflexes may contribute to supraesophageal reflux and supraesophageal reflux disease.

An interesting new finding of this study was the increase in sensitization of the smooth muscle esophagus to activation of the EECRst after hexamethonium administration. This result suggests that there is an inhibitory reflex from smooth to striated muscle portions of the esophagus mediated by the ENS (Fig. 8) that limits the EECR response to smooth muscle stimulation. Such an inhibitory reflex could account for the minimal effect of esophageal smooth muscle distension in activating the EECRst, but the physiological function of such a reflex is unknown. We hypothesize that this reflex inhibits the EECRst that would normally occur during gastroesophageal reflux episodes, which are small or do not extend far proximally, thereby allowing primary or secondary peristalsis to clean out the esophagus without impedance. Further studies are needed to investigate this issue.

Some might consider the number of animals used a limitation of the study. The number of cats for the Hex and PEN studies was only four for each. However, in both experiments the effects of the protocols completely blocked the responses and/or the effects were statistically significant. Therefore, we think both sets of experiments clearly support the mechanistic conclusions provided.

In conclusion, distension of the esophagus activates esophageal contraction (EECR) above the stimulus only. There are two types of EECR, one of the striated muscle esophagus (EECRst) and one of the smooth muscle esophagus (EECRsm). Although both are mediated by the central nervous system and vagus nerve and both are inhibited by long exposure of the esophagus to acid, the EECRst is mediated by the PEN and the EECRsm is mediated by the ENS. The ENS also mediates inhibition of the EECRst due to distension of the smooth muscle esophagus. Short-term exposure of the esophagus to acid causes a very similar response, in appearance and nature, to the EECRst, i.e., EECRst-like response, which is activated in a repetitive fashion. The function of the EECRst is unknown, but we hypothesize that the EECRst, like the EUCR, may provide protection from supraesophageal reflux.

GRANTS

This study was supported in part by the following National Institutes of Health Grants: RO1 DK-25731, RO1 DK-068158, and PO1-068051.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

I.M.L. and R.S. conceived and designed research; I.M.L. and B.K.M. performed experiments; I.M.L. analyzed data; I.M.L. and R.S. interpreted results of experiments; I.M.L. prepared figures; I.M.L. drafted manuscript; I.M.L., B.K.M., and R.S. edited and revised manuscript; I.M.L., B.K.M., and R.S. approved final version of manuscript.