Address for Correspondence Dr. Ravinder K. Mittal, Division of Gastroenterology (111D), San Diego VA Medical Center and University of California San Diego, 3350 La Jolla Village Drive, San Diego, CA 92161, USA. Tel: 858-552-7556; fax: 858-552-4327; e-mail: firstname.lastname@example.org
Background Lower esophageal sphincter (LES) lift seen on high-resolution manometry (HRM) is a possible surrogate marker of the longitudinal muscle contraction of the esophagus. Recent studies suggest that longitudinal muscle contraction of the esophagus induces LES relaxation.
Aim Our goal was to determine: (i) the feasibility of prolonged ambulatory HRM and (ii) to detect LES lift with LES relaxation using ambulatory HRM color isobaric contour plots.
Methods In vitro validation studies were performed to determine the accuracy of HRM technique in detecting axial movement of the LES. Eight healthy normal volunteers were studied using a custom designed HRM catheter and a 16 channel data recorder, in the ambulatory setting of subject’s home environment. Color HRM plots were analyzed to determine the LES lift during swallow-induced LES relaxation as well as during complete and incomplete transient LES relaxations (TLESR).
Key Results Satisfactory recordings were obtained for 16 h in all subjects. LES lift was small (2 mm) in association with swallow-induced LES relaxation. LES lift could not be measured during complete TLESR as the LES is not identified on the HRM color isobaric contour plot once it is fully relaxed. On the other hand, LES lift, mean 8.4 ± 0.6 mm, range: 4–18 mm was seen with incomplete TLESRs (n = 80).
Conclusions & Inferences Our study demonstrates the feasibility of prolonged ambulatory HRM recordings. Similar to a complete TLESR, longitudinal muscle contraction of the distal esophagus occurs during incomplete TLESRs, which can be detected by the HRM. Using prolonged ambulatory HRM, future studies may investigate the temporal correlation between abnormal longitudinal muscle contraction and esophageal symptoms.
Longitudinal muscle contraction of the esophagus plays an important role in the physiology and pathophysiology of esophageal motor function and esophageal symptoms.1 Studies suggest that it may actually be the cause of lower esophageal sphincter (LES) relaxation.2,3 Furthermore, heartburn and esophageal chest pain may be related to sustained contraction of longitudinal muscles of the esophagus.4,5
Several investigators have used catheter based, high frequency intraluminal ultrasound imaging (HFIUS) to study longitudinal muscle contraction in the humans.6–9 The problem with the HFIUS technique is that the US image analysis is extremely time consuming and labor intensive, which has prevented its wide-spread use. High-resolution manometry (HRM) is an improved recording technique to study esophageal motility, but similar to other intraluminal pressure recording techniques, it only measures circular muscle contractions of the esophagus. Several investigators have observed LES lift on routine HRM clinical manometry studies and suspect it to be a surrogate of longitudinal muscle contraction of the esophagus.10,11 However, it remains unclear if the HRM can accurately determine oral–aboral movements of the LES.
Esophageal symptoms occur infrequently and at unpredictable times. The equipment and technique of HRM are only available for recording esophageal motility in the non-ambulatory setting and only for short time intervals, which is not ideal to determine the temporal correlation between symptoms and motor events. Therefore, it would be advantageous to record HRM in the ambulatory setting for symptom correlation. The aim of our study was two-fold: (i) to study the feasibility of ambulatory, long-term HRM recording in humans, (ii) to determine if LES lift, as a marker of longitudinal muscle contraction can be detected during various types of LES relaxation in normal healthy subjects.
In vitro studies
We determined if HRM and color contour plots can measure movement of the high pressure zone (HPZ) along the catheter accurately. A custom-made HRM catheter was used for these studies. The catheter was 3 mm in diameter and equipped with 16 solid states, unidirectional pressure transducers spaced 1 cm apart, starting 5 cm from the tip of the catheter (from Unisensor, NH, USA). A vascular occluder (from In Vivo Metrics, CA, USA) was used to create an artificial HPZ. It was wrapped around a small cylinder made of 1% agar with a 2 mm diameter piezoelectric crystal embedded in it. Another piezoelectric crystal was glued to the HRM catheter (Fig. 1A). The entire assembly was placed inside a tub filled with water so as to allow measurement of distance between the two piezoelectric crystals sonographically. The cuff was inflated with water to a pressure of 30–50 mmHg. To shift the cuff location (artificial HPZ) on the catheter, the cuff was gradually pulled in either direction over various distances. The exact distance moved by the catheter was recorded using piezoelectric crystals (measurement accuracy 0.01 mm). HRM recordings were scored in a blinded fashion to determine the movement of the artificial HPZ on the catheter. The relationship between the distances detected by the piezoelectric crystal and HRM color isobaric contour plot was determined.
In vivo humans studies
The HRM catheter with 16 solid state pressure transducers, used for in vitro studies, was also used for the in vivo human recordings. Pressure transducers were calibrated in a pressure chamber and the catheter was soaked in the water at 37o for one and half hour prior to placement in the subject. This procedure reduces the zero drift of the pressure transducers, which was determined to be less than 5 mmHg during 24 h. The protocol for these studies was approved by the ‘Human Subject Protection Committee of the University of California San Diego’ and each subject signed a consent form prior to the beginning of the study. Eight normal healthy asymptomatic subjects participated in the study protocol. The age of subjects ranged from 22 to 50 years (mean: 34 years). The catheter was placed transnasally and positioned such that the two or three most distal transducers were located in the stomach and the remainder spanned across the LES and distal esophagus. The catheter was secured to the tip of the nose with tape and connected to a 16-channel ambulatory data recorder (from MMS, The Netherlands). The data recorder weighs approximately 1.5 lbs and uses AA batteries as a power source. In addition to pressure signals, it has the ability to record events such as pain, meals etc. Data were captured at a sampling rate of 4 or 8 Hz. At the 8 Hz sampling rate, data could be acquired for 16 h and at 4 Hz for 32 h. After catheter placement, subjects returned home and ate regular meals during the recording period. The subjects documented meal times, time spent in the supine position and other events of interest using the event marker on the recorder.
Stomach, LES, and esophageal pressure recordings were displayed as HRM color isobaric contour plots. Longitudinal muscle contraction occurs in association with swallow-induced peristalsis and transient LES relaxation. Therefore, the following events were identified and marked for analysis in the HRM color isobaric contour plots: 10 randomly selected dry swallows, 10 complete transient LES relaxations (TLESR), i.e., end-expiratory LES pressure of less than 2 mmHg at the peak of relaxation, and 10 incomplete TLESR. LES pressure was measured in reference to gastric pressure. Incomplete TLESRs were recognized when the end-expiratory LES pressure during relaxation was more than 5 mmHg. LES pressure was measured prior to and during LES relaxation to determine percent relaxation during all of the above events. The lower edge of the LES HPZ was also identified, where it transitioned to the stomach pressure. For each event, vertical displacement of the lower edge of the LES was measured prior to and during relaxation. Maximum, minimum, and average esophageal pressures were measured over a 10-s interval before and during TLESRs. All esophageal pressures were measured relative to the atmospheric pressure.
Recordings obtained during a previously published study were also analyzed.12 In that study, we simultaneously recorded esophageal pressures using a stationary, HRM system (Sierra Scientific Inc, Los Angeles, CA, USA) and HFIUS images of the esophagus at 3 cm above the LES (Hewlett Packard, Sonos 100, MA, USA). These recordings were obtained in eight subjects, following ingestion of a 1000 kcal meal. We identified 10 complete and 10 incomplete TLESRs in these studies. Ultrasound images before and during TLESRs were analyzed, as described previously.12 Briefly, from the tomographic ultrasound images, m-mode images were developed using custom software and temporally aligned with the pressure data. Change in muscle thickness, as a marker of longitudinal muscle contraction was determined during complete and incomplete TLESRs. The goal of these measurements was to verify if longitudinal muscle contraction of the esophagus (as detected by HFIUS imaging) occurs during incomplete TLESR.
Data are reported as mean and standard error of mean, unless otherwise stated. Paired and unpaired Student’s t-tests were used for statistical comparison. Pearson’s product-moment correlation coefficient was used to estimate correlation coefficients. P-value < 0.05 was considered statistically significant.
In vitro study
Fig. 1 shows the design of ambulatory HRM catheter and in vitro recording technique to study the effects of movement of the artificial HPZ on the HRM catheter. Fig. 1B shows the effect of movement of different distances of the artificial HPZ as recorded using piezoelectric crystals and HRM color isobaric contour plot. Note that each of the movements of the HPZ is recorded sonographically as well as by the HRM color isobaric contour plot. The graph in Fig. 1C shows the relationship between distances moved, as monitored by the two techniques. There is a strong linear correlation between the HPZ movements recorded sonographically and by the HRM color isobaric contour plot, r value of 0.89. Further analysis revealed that if the distance moved was more than 5 mm, the reliability of HRM to detect axial movement was even better with an r value of 0.99.
Studies were successfully conducted in all eight subjects and the pressure measurements were converted into HRM color isobaric contour plots using MMS software. Adequate recording was obtained for a minimum of 16-h duration in all subjects. Spontaneous swallow-induced movement of the LES is shown in Fig. 2. Note the elevation of the lower border of the LES HPZ (LES lift) in each of the two swallows. The LES lift was small and ranged from 2 to 4 mm (mean: 2.03 ± 0.21 mm). Fig. 3 shows the relationship between percent LES relaxation and LES lift during spontaneous swallows; there is poor correlation between the two values (r2 = 0.001). Similarly, the correlation between LES lift and delta LES pressure was also poor with an r2 of 0.018.
HRM recording during complete TLESR is shown in Fig. 4A. LES relaxation occurs fairly quickly and completely. Once the LES is completely relaxed its location in the HRM color isobaric contour plot cannot be identified. Therefore, LES lift during complete TLESR cannot be measured with HRM. However, in some complete TLESRs, the onset of LES relaxation is relatively slow and during these events LES lift and LES relaxation are seen to occur concurrently at the onset of relaxation (Fig. 4).
Fig. 4B shows LES lifts of 7 and 17 mm during two incomplete TLESRs. The lower edge of LES in these records was marked as the transition zone between the LES HPZ and gastric pressure. A total of 80 incomplete TLESRs were scored (10 per subject in eight subjects). The LES lift ranged from 4 to 18 mm (mean: 8.37 ± 0.60 mm). Fig. 3 shows the relationship between change in LES pressure during LES relaxation, or delta LES pressure, and the amplitude of LES lift. There was poor correlation between the two values (r2 = 0.046). Percent LES relaxation and LES lift also did not show significant correlation (r2 = 0.007), (Fig. 3).
Changes in esophageal pressure during complete and incomplete TLESRs are shown in Fig. 5. The increase in esophageal pressure starts to occur with the onset of LES relaxation, and once the LES is fully relaxed, the esophagus and stomach become one cavity (common cavity). The average increase in esophageal pressure during complete TLESRs is 3.7 ± 0.9 mmHg (n = 8 subjects with 10 complete TLESRs for each subject). Interestingly, the esophageal pressure also increases during incomplete TLESRs (Fig. 5) (mean: 4.7 ± 0.6 mmHg; n = 8 subjects with 10 incomplete TLESRs for each subject). The amplitude of esophageal pressure increase during complete and incomplete TLESR is not different.
Fig. 6A and B show m-mode US images and HRM during incomplete and complete TLESRs respectively. Muscle thickness is shown as an m-mode image at the bottom of the HRM plot. The changes in thickness are demonstrated more clearly when superimposed on the HRM plot (yellow line). During TLESR, an increase in muscle thickness is observed. This muscle thickening, a marker of longitudinal muscle contraction, returns to baseline upon termination of the TLESR. The peak increase in muscle thickness during complete TLESRs, 94 ± 30% tended to be greater but not significantly different than with the incomplete TLESRs, 54 ± 8% (P > 0.05).
In summary, our data show the following: (i) in vitro recordings show that movements of an artificial HPZ along the length of HRM catheter are faithfully recorded using the HRM color isobaric contour plot, (ii) swallow-related LES relaxation is associated with a small LES lift, (iii) LES lift cannot be recognized during complete TLESRs because a fully relaxed LES is not seen on the HRM plot, (iv) incomplete TLESRs are associated with significant LES lift, which is greater than those associated with spontaneous swallow-induced LES relaxation, (v) there is an increase in esophageal pressure with the onset of complete as well as incomplete TLESRs, with no difference in the degree of increase in pressure between the two, and (vi) ultrasound image analysis shows that, similar to complete TLESR, incomplete TLESRs are also associated with sustained longitudinal muscle contraction of the esophagus for the entire duration of the TLESR.
As pressure transducers on the HRM catheter are located 1-cm apart (from center of one transducer to the center of other), one can be certain that LES movement on the HRM catheter of 1 cm or more would be easily and certainly detected. However, our in vitro study shows that even less than 1 cm of HPZ movement can be detected by the HRM catheter. There may be several reasons for the above: (i) HRM color isobaric contour plot uses a computer algorithm (linear interpolation) to generate a seamless color plot between the actual pressure data obtained at every 1 cm. The LES is 3–4 cm in length and pressures along its length are distributed in the shape of a bell. Small movements of the LES shift the location of peak pressure as well as change the full length of the pressure profile. These shifts, in conjunction with the computer algorithm used to generate the HRM color isobaric contour plots, change the location of the LES border on the HRM color isobaric contour plot; (ii) each transducer itself is 4-mm long and if the HPZ encompasses only a part of the length vs the entire length of the transducer it may record different pressure thus allowing for some changes in pressure even when the HPZ movement is less than 1 cm. Irrespective of the explanation, actual measurements using sensitive piezocrystal technique proves our claim. One can observe oral and aboral LES movement of 2–3 mm with tidal respiration (see, Fig. 2) and greater with forced inspiration on HRM recording, proving that the latter can indeed record axial LES movements using HRM.
Cranial movement of the LES with swallows has been observed by radiologists for more than 50 years. Dodds implanted radio-opaque markers along the esophagus and LES, and clearly demonstrated that with swallows and secondary peristalsis, the LES moves orally and results in a relative movement between the side hole of the manometry catheter and LES.13,14 The latter results in a possible artifact of LES pressure recording during LES relaxation if the LES pressure were to be monitered by a side hole sensor. To avoid the above artifact, Dent devised the sleeve sensor.15 Mucosal clips implanted endoscopically in the distal esophagus also show oral movement of the LES with both swallow and peristalsis.16,17 Oral movement of the LES with swallows is observed consistently by HRM, albeit the magnitude of this movement is relatively small as compared with the one observed using flouroscopy imaging studies. The reason may be that we studied spontaneous swallows in which LES relaxation is small as compared with the wet swallows (water or barium) in which the LES relaxation is more complete. The reason for studying spontaneous swallows in our study was that if the LES is fully relaxed, LES lift cannot be measured by the HRM technique. The other reason may be that the HRM technique is relatively less accurate when the LES lift is less than 5 mm. Using water infusion manometry, Bredenoord et al performed high-resolution manometry for 1–2 h in stationary subjects. They observed separation of the LES and crural diaphragm, intermittantly during these recording periods,18 which proves the feasibility of recording axial movements of the HPZ with HRM. Interestingly, those periods when LES and crural diaphragm were separated more likely to be associated with reflux than when they were not.
Pandolfino et al. used endoscopically placed mucosal clips to study esophageal shortening during TLESR and found significant and sustained shortening of the esophagus during the entire period of TLESR.19 Babei et al. found a unique pattern of longitudinal muscle contraction of the esophagus in association with TLESR, which is different from the one associated with swallow-induced peristalsis.12 The longitudinal muscle contraction starts just above the LES and proceeds in an antiperistaltic fashion toward the mouth. On the other hand, swallow-induced peristalsis is associated with simultaneous contraction of the circular and longitudinal muscle that traverses from the oral to the aboral direction in a peristaltic fashion.7,20 Another study by Pandolfino also revealed a greater segmental shortening of the distal as compared with mid esophagus in association with TLESR.19 Once the LES is fully relaxed during complete TLESRs it cannot be recognized on the HRM color isobaric contour plot. Therefore, the LES lift cannot be measured during complete TLESRs. However, with incomplete TLESRs, we observed LES lift during the entire period of TLESR. Usually the LES lift starts before the onset of LES relaxation and lasts througout the TLESR. Ultrasound images during the period of incomplete TLESR show the presence of longitudinal muscle contraction, similar to what we observed in an earlier study.
A mechanical axial stretch on the LES in the cranial direction induces LES relaxation that is blocked by L-NAME (blocker of nitric oxide).2 Axial stretch-induced LES relaxation is caused by activation of inhibitory motor neurons as it is not blocked by any of the neuronal synaptic blockers.3 Fundoplication restricts axial stretch on the LES and prevents LES relaxation.21 On the basis of above studies, we expected a linear relationship between the LES lift and percent LES relaxation, which was not observed in our study. It is possible that the movement of LES was influenced by respiration. Similar to in vivo studies, it is not possible to separate respiration-induced LES movements from those caused by longitudinal muscle contraction.
The increase in intraesophageal pressure during complete TLESRs, referred to as common cavity pressure, is thought to represent reflux of gastric contents into the esophagus.22–24 However, Tipnis et al. reported that even though pressure increases simultaneously throughout the length of the esophagus, the reflux of gastric contents into the esophagus occurs sequentially, i.e., reflux occurs first in the distal esophagus and traverses to the proximal esophagus.25 They found that contractions of the longitudinal muscle of the esophagus occur in association with increases in esophageal pressure and proposed shortening of the esophagus, in accordance with Boyle’s law of physics, is the reason for increase in esophageal pressure. As gastroesophageal reflux does not occur during incomplete TLESRs,26–28 the increase in esophageal pressure during incomplete TLESRs is most likely related to the contraction of longitudinal muscles and esophageal shortening. Recently, we reported that the simultaneous pressure waves in the achalasic esophagus are also related to longitudinal muscle contraction and esophageal shortening.29 We propose that increases in esophageal pressure could be an important marker of longitudinal muscle contraction of the esophagus.
Ultrasound imaging clearly show longitudinal muscle contraction of the distal esophagus during incomplete TLESR. It has been suggested that the LES relaxation during TLESR may begin from the lower edge of LES; such relaxation would appear as a LES lift in the HRM color isobaric contour plot and is an important limitation of the technique. The use of LES lift as a marker of longitudinal muscle contraction has significant clinical implications for the diagnostic evaluation of esophageal symptoms. Sustained contraction of longitudinal muscles of the esophagus may be the cause of heartburn5 and ‘angina like’ esophageal pain.4 Longitudinal muscle contraction was measured using HFIUS imaging in those studies, which is time consuming and not practical. Furthermore, esophageal symptoms occur infrequently and at unpredictable times and to make a temporal association between symptoms and abnormal motor events long-term monitoring is actually required. Using ambulatory HRM and LES lift as a marker of longitudinal muscle contraction, future studies may be able to determine the true correlation between esophageal symptoms and longitudinal muscle spasm.
This study was supported in part by grant NIH-RO1-DK060733.
Conflict of Interest
No conflicts of interest for any of the authors.
Ravinder K. Mittal: design of equipment, design of experiment, supervise data acquisition, write and edit the manuscript; Anna Karstens: initial testing of equipment, recruiting subjects for the study, data acquisition and data analysis; Eric Leslie: recruiting subjects for the study, data acquisition, data analysis and help with writing the manuscript; Arash Babaei: recruiting subjects for the study, data acquisition and analysis; Valmik Bhargava: initial testing of equipment, software development, data analysis, creating figures, and writing of the manuscript.