SEARCH

SEARCH BY CITATION

Keywords:

  • esophageal manometry;
  • gastroesophageal reflux;
  • lower esophageal sphincter

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Author Contribution
  9. Funding
  10. Competing Interests
  11. References

Background  The advent of drugs that inhibit transient lower esophageal sphincter relaxation (TLESR) necessitates accurate identification and scoring. We assessed the intra- and inter-assessor variability of the existing objective criteria for TLESR, improving them where necessary.

Methods  Two 3-h postprandial esophageal manometric and pH recordings were performed in 20 healthy volunteers. Each recording was duplicated. The recordings were analyzed by five experienced observers for TLESRs based on their expert opinion. TLESRs were also analyzed for the presence of the original four criteria as well as inhibition of the crural diaphragm (ID), a prominent after-contraction (AC), acid reflux and an esophageal common cavity.

Key Results  The overall inter- and intra-observer agreements for TLESRs scored, according to observer’s expert opinion, were 59% (range 56–67%) and 74% (60–89%), respectively. When TLESRs were restricted to those fulfilling the original criteria, these agreements fell to 46% (40–53%) and 60% (44–67%), respectively. Cleaning the recordings by removal of technically flawed sections improved agreements by 5%. Inclusion of additional criteria (ID and AC) resulted in inter- and intra-observer agreements of 62% (52–70%) and 69% (53–79%), respectively. A consensus analysis performed collectively by three observers and based on the new criteria (original ± ID and AC) resulted in 84% agreement between the paired recordings.

Conclusions & Inferences  The original criteria for the definition of TLESRs allows for substantial inter- and intra-observer variability, which can be reduced by incorporation of additional objective criteria. However, the highest level of intra-observer agreement can be achieved by consensus analysis.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Author Contribution
  9. Funding
  10. Competing Interests
  11. References

Transient lower esophageal relaxations (TLESRs) are the major mechanism of reflux in healthy subjects1 and the majority of patients with reflux disease.2,3 Their pivotal importance as a mechanism of reflux has generated an enormous amount of interest in their mediation and control. Investigation of these events necessitated the development of objective criteria by which they could be identified. The first such criteria were published in 1995.4 Prior to that, the definition of TLESR had been mostly arbitrary and derived from the visual pattern recognition and approximations rather than from a systematic analysis of the variables involved.1,2,5–14

The advent of pharmacologic agents that inhibit TLESRs15 has brought into even sharper focus the need for accurate and objective criteria by which to identify and score these events as the development of drugs requires that any variability in scoring TLESRs be minimized. The original objective criteria, published in 1995 have never been subjected to rigorous evaluation with regard to inter and intra-subject variability. Such information is crucial to understanding the inherent limitations in the manometric analysis, and the subsequent determination of sample sizes for drug evaluation. The aims of this study, therefore, were to assess the performance of the criteria by which to define TLESRs with regard to intra- and inter-assessor variability, and to improve them where necessary.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Author Contribution
  9. Funding
  10. Competing Interests
  11. References

The study was conducted at two centers, the Royal Adelaide Hospital, Adelaide, South Australia, and the Academic Medical Centre, Utrecht, The Netherlands. Combined manometric and pH monitoring studies were performed in 20 healthy male subjects aged 18–48 years (mean 27 years), weight 65–93 kg (mean 78 kg) and all H pylori negative on urea breath test. All gave written informed consent and the protocol was approved by the Research Ethics Committees at both centers.

Recording methods

Esophageal manometry was performed using a 16-channel perfused assembly, only eight of which were used for the study. A reverse-perfused sleeve sensor16 recorded LES pressure, a side hole 1 cm distal to the sleeve reorder intragastric pressure, five side holes at 3-cm intervals recorded pressure in the esophageal body and a side hole in the pharynx monitored swallowing. The gastric and esophageal side holes were perfused with degassed distilled water at 0.3 mL min−1 and the sleeve perfused at 0.6 mL min−1 by a low-compliance pneumohydraulic capillary infusion system (Dentsleeve, Wayville, SA, Australia). The pharyngeal side hole was perfused with air at a rate of 0.16 mL min−1.

Esophageal pH was monitored 5 cm above the proximal border of the LES with an antimony electrode (Medtronic Functional Diagnostics Zinetics Inc, UT, USA). The electrode was calibrated in buffers of pH 1 and 7, before the study and the calibration checked in the same buffers at the end of the study. No drifts in pH >0.5 pH units were detected during any of the studies. Manometric and pH signals were recorded by a computerized acquisition system (MMS, Enschede, The Netherlands) and displayed and stored on a personal computer for subsequent display and analysis.

Protocol

Two 3-h postprandial manometric and pH recordings were performed at least 1 week apart in each subject. Subjects were studied in the sitting position. After an overnight fast, subjects were intubated and, after positioning of the manometric assembly and pH electrode, allowed to accommodate to the assemblies for 5–10 min. Fasting recordings were made for 45 min, after which the subjects ate within 15 min a standardized meal (mashed potatoes, minced beef, banana and a non-carbonated soft drink, total caloric value 700 kcal, fat content 35%). Recordings were continued for a further 3 h.

Analysis

Paired recordings were made in each of 20 subjects resulting in a total of 40 recordings. Each recording was then duplicated, de-identified and given a new identification number resulting in two duplicate, but uniquely labeled sets of 40 recordings. Inter-assessor agreement was assessed by comparing scoring of recordings by separate assessors. Intra-assessor agreement was assessed by comparing scoring of duplicate versions of the same recording by an individual.

The recordings were analyzed primarily for the occurrence of TLESRs. TLESRs were identified initially according to the assessor’s expert opinion. In addition, events identified as TLESRs were also scored for the presence of the original four criteria as published previously, absence of a swallow for 4 s before, until 2 s after the onset of the TLESR, LES relaxation rate of ≥1 mmHg s−1, a time from the onset to complete relaxation of ≤10 s, and a nadir pressure of <2 mmHg.4 LES relaxations that lasted for more than 10 s and that were associated with a swallow within 5 s before or 2 s after the onset of an LES relaxation were also included as TLESRs.3,4

Events identified as TLESRs were also examined for additional characteristics that have been reported to be associated with TLESR including: inhibition of the crural diaphragm, a prominent LES after-contraction, acid reflux, and the presence of an esophageal common cavity.17 Inhibition of the crural diaphragm was deemed to have occurred if there was at least a 50% drop in the magnitude of the inspiratory pressure transients on the LES pressure recordings.18 An after-contraction was defined as a prominent transient increase in LES pressure after completion of the TLESR that was greater than that occurring after an adjacent swallow-induced LES relaxation. An esophageal common cavity was defined as a simultaneous increase of pressure of at least 5–10 mmHg, in at least two channels above the LES.14,19,20 Acid reflux was deemed to have occurred during a TLESR if there was an abrupt drop in esophageal pH to below pH 4 or, if basal esophageal pH was already <4, there was a further drop in pH of at least 1 pH unit.

Although a prominent after-contraction had been described as a feature of TLESRs since the initial report1,2,17 this feature had not been hitherto defined quantitatively. The amplitude of the after-contraction associated with TLESRs was therefore compared with that associated with the nearest swallow-induced LES relaxation after the TLESR. Amplitude was measured at the maximum point in mmHg above intragastric pressure. Duration was measured in seconds from the onset of the after-contraction, until pressure had resumed a stable level of basal pressure.

Statistical evaluation

A union set of events was created for each subject and study day. This set consisted of all TLESR events recorded by the five assessors from the recordings from both study days. Based on the union set, it was determined which events represented the same TLESR. This was done by first ordering all TLESR events in the union set in time sequence. Events were judged to match if the time difference between the events was ≤15 s. Events were automatically judged to be separate, if an assessor had identified more than one TLESR within the 15-s interval. The level of agreement between two assessors was measured as the proportion of TLESRs that were scored identically by the two assessors. Levels of agreement were calculated for TLESRs identified according to the set criteria as well as according to the assessor’s judgment, based on their expert opinion.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Author Contribution
  9. Funding
  10. Competing Interests
  11. References

For a total of 40 recordings, the mean number of TLESRs based on published criteria was 13.9 (95% CI 10.9–16.5) on Day 1 and 14.7 (11.4–17.9) on Day 2.

Analysis 1

In the initial evaluation, assessors analyzed the 40 recordings (two per subject, each duplicated and all re-coded) from the first 10 subjects. TLESRs were scored on the basis of the assessor’s judgment and also assessed as to whether they satisfied the previously published criteria.4 When TLESRs were scored on the basis of the assessor’s expert opinion, the mean level of agreement was 59.5% (range 50.7–67.1%), with a mean kappa coefficient of 0.45. However, when only TLESRs that satisfied the published criteria were scored, the agreement among the assessors fell to 40.1% (32–47%, P = 0.001, Fig. 1), with mean kappa value of 0.40. Similarly, intra-assessor agreement was higher when TLESRs were scored according to expert opinion 74.4% (60.4–89.6%) with a mean kappa value of 0.67 (range 0.54–0.89) compared with that based on published criteria 53.6% (35–64%, P = 0.0005) with a mean kappa value of 0.60 (0.47–0.81).

image

Figure 1.  Diagrammatic representation of the level of inter- and intra-assessor agreement. For the inter-assessor agreement, the upper value of each pair represents data for the assessors expert opinion; the lower value for data based on the original criteria (ref 4). For the intra-assessor agreement, the values on the left represent data based. on the assessors expert opinion and on the right, based on the original criteria.

Download figure to PowerPoint

During assessment of the recordings, it was evident that some sections were difficult to analyze with confidence because of a number of factors. These factors included: low basal LES pressure, unstable gastric pressure recording, and an unclear swallow signal (Fig. 2). The postprandial recordings were therefore assessed by all assessors in 30-min sections for the quality of the recordings with respect to these three variables, and each section prioritized for exclusion as definite, high, moderate or none. Sections prioritized as either definite or high by at least three of five assessors were then excluded from the analysis up to a maximum of 20% of a single recording (mean 18.5%). ‘Cleaning’ of the recordings in this manner increased inter-assessor agreement according to expert opinion by an average of 5–65% (57–72%, = 0.001) and intra-assessor agreement by a similar degree to 79.4% (68–92%, P = 0.02).

image

Figure 2.  Example of a section of recording that was judged of poor quality. Not the poorly defined swallow signal and irregular gastric baseline.

Download figure to PowerPoint

Establishment of enhanced criteria for identification of TLESRs

An additional four manometric and pH features were noted in the majority of events identified as TLESRs, during the initial analysis: (i) inhibition of the cural diaphragm – 82%, (ii) after-contraction – 80%, (iii) acid reflux – 36%, and (iv) an esophageal common cavity – 70%. In an effort to increase the inter- and intra-assessor agreement, the events for the scored TLESRs in Analysis 1 were re-classified according to various combinations of the original four published criteria along with the additional four characteristics, and the data for each combination re-analyzed for the level of inter- and intra-assessor agreement. This analysis showed that the best combination of criteria by which to identify TLESRs was at least four out of a combination of six criteria; the original four criteria as well as inhibition of the crural diaphragm and a prominent after-contraction. This new set of enhanced criteria was then formally evaluated in a second set of recordings.

Analysis 2

A second analysis was performed, using the enhanced criteria, on a set of 50 recordings: the remaining duplicate studies from subjects 11–20, and a set of five duplicate recordings from the first analysis. The mean inter-assessor agreement for scoring TLESRs based on their expert opinion was 63.7% (range 55.6–70.5%) with a kappa value of 0.42. This was significantly better (P = 0.006) than that during Analysis 1. When only TLESRs that satisfied the new criteria were scored, levels of agreement remained lower (than the ones based on expert opinion) (61.6%, 52–69.5%, P = 0.12) with a kappa value of 0.42, although the difference was smaller and the level of agreement was higher (P = 0.0001) than in Analysis 1. However, intra-assessor agreement was similar when TLESRs were scored according to expert opinion 71.1% (60.8–79.9%) with a mean kappa value of 0.60 (0.49–0.81), compared with that based on the new criteria 68.9% (52.7–78.7%, P = 0.20), with a mean kappa value of 0.60 (0.50–0.81). Cleaning the recordings, as done in Analysis 1, improved inter-assessor agreement by around 5–68.6% (P = 0.07) with a kappa value of 0.42, but had no measurable effect on the level of intra-assessor agreement.

Consensus analysis

The level of inter- and intra-assessor agreement, even with the new criteria, was judged to be suboptimal. Therefore, a consensus analysis was performed. Recordings from subjects 11–20, that had been part of Analysis 2, were used. The dataset consisted of 20 recordings in 10 subjects (two in each subject). Each recording was duplicated and the whole dataset recoded thereby making a complete dataset of 40 recordings. These 40 recordings had been analyzed previously by each of the three assessors involved in the consensus analysis as part of Analysis 2. Each assessor had therefore identified events that he had judged to be a TLESR. All events that had been identified by at least one of the three assessors (RHH, DS and GB) were then marked on the recordings by identification tags. Any included event could therefore have been identified by from one to three assessors. These events were then reassessed by three assessors concurrently and a decision made as to whether the event was or was not a TLESR. At the end of the analysis, the code was broken, duplicate recordings identified, and agreement between the two sets of duplicates assessed.

A comparison of the scoring of the two duplicate sets of recordings revealed a level of agreement of 84% between the analyses with a kappa value of 0.53. Of the 378 TLESRs identified initially, there was disagreement in scoring between the duplicate on 62 (16.4%) of occasions. With 36 of these, the TLESR was identified initially on only one of the duplicate recordings. In all of these instances, the event had been identified as a possible TLESR by only one of the three assessors. The remaining disagreement was due to largely poor quality of that segment of the recording in which the event occurred, and the occurrence of ‘back-to-back’ TLESRs (Fig. 3) in which there was disagreement as to whether these were two separate events or a single event separated by the impact of a distal esophageal contraction on the sleeve sensor. Disagreements regarding five events could not be explained by any particular features of the events or the quality of the recordings.

image

Figure 3.  Example of a back-to-back TLESR. The second labeled TLESR15 is followed by a second apparent LES relaxation (arrow) separated from the first by only a brief transient period of LES pressure. Such events were scored variably as either one or two events among the assessors.

Download figure to PowerPoint

Reproducibility of TLESRs

Given the high level of agreement in scoring TLESRs during the consensus analysis, the paired recordings from the 10 subjects included in the consensus analysis were analyzed for the rate of TLESRs. The mean number of TLESRs scored on day 1 [18 ± 1.8 (SEM)] was not statistically different from that scored on day 2 (17.6 + 1.6, P = 0.56, Fig. 4). The correlation between the number of TLESRs scored on days 1 and 2 was high (r = 0.86, P = 0.0045).

image

Figure 4.  Top panel: correlation during consensus analysis between number of transient lower esophageal sphincter relaxations (TLESRs) scored during the two recording days. Lower panel: individual subject data for the numbers of TLESRs scored on each study day. The horizontal bars indicate median values.

Download figure to PowerPoint

Definition of after-contraction

A total of 112 consecutive TLESRs with an after-contraction were identified. Both the amplitude (47.0 ± 1.7 mmHg vs 19.3 ± 1.0 mmHg, < 0.0001) and the duration (22.1 ± 1.2 s vs 10.3 ± 1.3 s, = 0.0001) of the after-contraction associated with TLESR were significantly higher than that associated with swallow-induced LES relaxation, although substantial overlap existed between the two measurements (Fig. 5).

image

Figure 5.  Amplitude (upper panel) and duration (lower panel) for the after-contraction following transient lower esophageal sphincter relaxation. Each point represents data for an individual event. The horizontal bars indicate median values.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Author Contribution
  9. Funding
  10. Competing Interests
  11. References

The development of drugs that target TLESRs for the control of gastro-esophageal reflux, and thereby the treatment of reflux disease, has created a need for the accurate identification and scoring of these events. The study has assessed the only available objective criteria for the identification of TLESRs in terms of their accuracy and reproducibility. The contributors to the analysis have all had substantial experience in the analysis of these events. The major finding was that the original criteria, as published in 1995, exhibited poor inter- and intra-assessor agreement and performed less well than when the assessors used their own judgment. Performance of the criteria was improved marginally when technically difficult sections of the recordings were removed, and improved further when additional criteria such as inhibition of the crural diaphragm and the after-contraction were added. However, the best agreement was achieved by consensus analysis.

The underperformance of the original criteria compared with the assessors extensive personal experience and expertise can be explained by several factors. First, the original criteria were not designed to identify all TLESRs necessarily, but rather to instill a degree of objectivity to an analysis which previously had been based on varying largely subjective criteria that had differed among studies.1,2,5–14 Additional features commonly recognized to be associated with TLESRs such as inhibition of the crural diaphragm16 and a prominent after-contraction17 were not included at that time, but have become useful adjuncts subsequently. In addition, the original criteria contained rigid parameters for the contributing variables. For example nadir LES pressure, that requires extremely accurate manometric technique that can be difficult to achieve. The assessor’s expertise was able to compensate for such minor deviances from the criteria and score events as TLESRs that did not meet the strict criteria. Adoption of a revised and expanded set of criteria, of which only some had to be met, improved the level of the assessor’s agreement and brought it closer to that of the assessor’s experienced judgment.

Second, the ability to assess the criteria is influenced by the quality of the recording. A good clear swallow signal, adequate basal LES pressure to determine the onset and rate of relaxation and clean gastric pressure as a reference for LES pressure are all important. There were some segments of the recordings that were deemed by the assessors as being too difficult for the reliable identification of TLESRs. Exclusion of the most difficult of these segments improved the reproducibility of the analysis, however, not sufficiently.

Despite all attempts to improve the analysis by removing technically difficult segments and enhancing the criteria, inter- and intra-assessor reproducibility remained at around 70%. The reasons for this were unclear and analysis of events that were not scored reproducibly did not reveal any consistent factors. In an effort to minimize such inconsistencies, a consensus analysis was done with three of the assessors reviewing all events identified by at least one of the assessors. This approach, used in many other analyses to overcome disagreements among assessors, resulted in a more acceptable level of reproducibility and would appear to be the optimal for the analysis of TLESRs. Nevertheless, the kappa value even for this analysis was only moderate, suggesting that there remains an element of identification of TLESRs that may be subjective. It would have been optimal if the consensus analysis could have been performed de-novo on the original and re-coded recordings. However, time constraints of the assessors did not allow them that option.

The analysis also allowed for the objective assessment of the after-contraction. However, despite this feature being identified by all assessors as a useful attribute, and whilst the amplitude and duration were greater than the LES contraction after a swallow-induced LES contraction, there was substantial overlap such that a clear distinction between swallow-induced LES relaxation and TLESR was not evident.

A useful outcome of the study was an analysis of the reproducibility of TLESRs among subjects. This was only possible when acceptable inter- and intra-assessor agreement had been achieved. The consensus analysis revealed a high level of reproducibility of TLESRs within subjects. These data have implications for the numbers of subjects required for testing the effects of pharmacologic agents. However, the analysis was performed in healthy subjects and the applicability to patients with reflux disease21 requires further testing.

A potential limitation of the study is that it relied in the use of a sleeve sensor for measurement of LES pressure. The introduction of high-resolution manometry has dictated a new approach to the analysis of manometric recordings and the development of new variables for measurement of LES relaxation.22 The new approaches to measurement of LES relaxation, however, continue to rely on an electronic equivalent of a sleeve sensor, the so-called ‘e-sleeve’. Thus, the approaches to analysis of TLESRs used in the current study still apply, although the actual parameters might be different. A recent study23 suggests that high-resolution methods might be more accurate than the sleeve for scoring of TLESRs associated with reflux.

In summary, the criteria that are used for the identification of TLESRs perform well, but only in the context of a consensus analysis with more than one observer. The addition of further two criteria to the original four, from which only four of the six need to be satisfied, enhances their applicability. It seems likely that these criteria could also be applied to e-sleeve recordings, although this would require formal testing. Careful attention to the quality of the manometric recording is important. However, a consensus analysis with more than one concurrent assessor is needed for optimal results.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Author Contribution
  9. Funding
  10. Competing Interests
  11. References

The Authors thank Dr. Cecilia Kristensson and Dr. Magnus Ruth of AstraZeneca R&D, Mölndal, Sweden, for their support, guidance in the initiation and conduct of the study.

Author Contribution

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Author Contribution
  9. Funding
  10. Competing Interests
  11. References

All authors, RHH, GEEB, RP, DAS, AJPM, were involved in the study conception, design, data analysis, and drafting of the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Author Contribution
  9. Funding
  10. Competing Interests
  11. References
  • 1
    Dent J, Dodds WJ, Friedman RH et al. Mechanism of gastroesophagal reflux in recumbent asymptomatic subjects. J Clin Invest 1980; 65: 25667.
  • 2
    Dent J, Holloway RH, Toouli J, Dodds WJ. Mechanisms of lower oesophageal sphincter incompetence in patients with symptomatic gastroesophageal reflux. Gut 1988; 29: 10208.
  • 3
    Van Herwaarden MA, Samsom M, Smout AJ. Excess gastroesophageal reflux in patients with hiatus hernia is caused by mechanisms other than transient LES relaxations. Gastroenterology 2000; 119: 143946.
  • 4
    Holloway RH, Penagini R, Ireland AC. Criteria for objective definition of transient lower esophageal sphincter relaxation. Am J Physiol 1995; 268: G12833.
  • 5
    Cucchiara S, Bortolotti M, Minella R, Auricchio S. Fasting and postprandial mechanisms of gastroesophageal reflux in children with gastroesophageal reflux disease. Dig Dis Sci 1993; 38: 8692.
  • 6
    Dodds WJ, Dent J, Hogan WJ et al. Mechanisms of gastroesophageal reflux in patients with reflux esophagitis. N Engl J Med 1982; 307: 154752.
  • 7
    Dodds WJ, Kahrilas PJ, Dent J, Hogan WJ, Kern MK, Arndorfer RC. Analysis of spontaneous gastroesophageal reflux and esophageal acid clearance in patients with reflux esophagitis. J Gastrointest Motil 1990; 2: 7989.
  • 8
    Freidin N, Fisher MJ, Taylor W et al. Sleep and nocturnal acid reflux in normal subjects and patients with reflux oesophagitis. Gut 1991; 32: 12759.
  • 9
    Martin CJ, Patrikios J, Dent J. Abolition of gas reflux and transient lower esophageal sphincter relaxation by vagal blockade in the dog. Gastroenterology 1986; 91: 8906.
  • 10
    Mittal RK, McCallum RW. Characteristics of transient lower esophageal sphincter relaxation in humans. Am J Physiol 1987; 252: G63641.
  • 11
    Mittal RK, McCallum RW. Characteristics and frequency of transient relaxations of the lower esophageal sphincter on patients with reflux esophagitis. Gastroenterology 1988; 95: 5939.
  • 12
    Mittal RK, Stewart WR, Schirmer BD. Effect of a catheter in the pharynx on the frequency of transient lower esophageal sphincter relaxations. Gastroenterology 1992; 103: 123640.
  • 13
    Boulant J, Fioramonti J, Dapoigny M, Bommelaer G, Bueno L. Cholecystokinin and nitric oxide in transient lower esophageal sphincter relaxation to gastric distention in dogs. Gastroenterology 1994; 107: 105966.
  • 14
    Wyman JB, Dent J, Heddle R, Dodds WJ, Toouli J, Downton J. Control of belching by the lower oesophageal sphincter. Gut 1990; 31: 63946.
  • 15
    Hirsch D, Tytgat G, Boeckxstaens G. Transient lower oesophageal sphincter relaxations – a pharmacological target for gastro-oesophageal reflux disease? Aliment Pharmacol Ther 2002; 16: 1726.
  • 16
    Sivri B, Mittal RK. Reverse-perfused sleeve: an improved device for measurement of sphincteric function of the crural diaphragm. Gastroenterology 1991; 101: 9629.
  • 17
    Mittal RK, Holloway RH, Penagini R, Blackshaw LA, Dent J. Transient lower esophageal sphincter relaxation. Gastroenterology 1995; 109: 60110.
  • 18
    Mittal RK, Fisher MJ. Electrical and mechanical inhibition of the crural diaphragm during transient relaxation of the lower esophageal sphincter. Gastroenterology 1990; 99: 12658.
  • 19
    Sifrim D, Silny J, Holloway RH, Janssens JJ. Patterns of gas and liquid reflux during transient lower oesophageal sphincter relaxation: a study using intraluminal electrical impedance. Gut 1999; 44: 4754.
  • 20
    Sifrim D, Holloway R, Silny J, Tack J, Lerut A, Janssens J. Composition of the postprandial refluxate in patients with gastroesophageal reflux disease. Am J Gastroenterol 2001; 96: 64755.
    Direct Link:
  • 21
    Boeckxstaens GE, Hirsch DP, Verkleij CB et al. Reproducibility of meal-induced transient lower oesophageal sphincter relaxations in patients with gastro-oesophageal reflux disease. Neurogastroenterol Motil 2005; 17: 238.
  • 22
    Pandolfino J, Ghosh S, Q Z, Jarosz A, Shah N, Kahrilas P. Quantifying EGJ morphology and relaxation with high-resolution manometry: a study of 75 asymptomatic volunteers. Am J Physiol (Gastrointest Liver Physiol) 2006; 290: G101331040.
  • 23
    Rohof W, Boeckxstaens G, Hirsch D. High resolution esophageal topography is superior to conventional sleeve manometry for the detection of transient lower esophageal sphincter relaxations associated with a reflux event. Neurogastroenterol Motil 2011; 23: 42732.