Gastroenterology
Volume 130, Issue 3 , Pages 639-649, March 2006

Obesity: A Challenge to Esophagogastric Junction Integrity

  • John E. Pandolfino

      Affiliations

    • Department of Medicine, Northwestern University, The Feinberg School of Medicine, Chicago, Illinois
    • Corresponding Author InformationAddress reprint requests to: John E. Pandolfino, MD, Northwestern University, Feinberg School of Medicine, Division of Gastroenterology, Department of Medicine, 676 N. St. Clair Street, Suite 1400, Chicago, Illinois 60611. fax: (312) 695-3999.
    • ,
  • Hashem B. El–Serag

      Affiliations

    • Department of Medicine, Baylor College of Medicine, Houston VA Medical Center, Houston, Texas
    • ,
  • Qing Zhang

      Affiliations

    • Department of Medicine, Northwestern University, The Feinberg School of Medicine, Chicago, Illinois
    • ,
  • Nimeesh Shah

      Affiliations

    • Department of Medicine, Northwestern University, The Feinberg School of Medicine, Chicago, Illinois
    • ,
  • Sudip K. Ghosh

      Affiliations

    • Department of Medicine, Northwestern University, The Feinberg School of Medicine, Chicago, Illinois
    • ,
  • Peter J. Kahrilas

      Affiliations

    • Department of Medicine, Northwestern University, The Feinberg School of Medicine, Chicago, Illinois

Received 2 August 2005; accepted 30 November 2005. published online 13 December 2005.

Article Outline

Background & Aims: The aim of the current study was to analyze the relationship between obesity and the morphology of the esophagogastric junction (EGJ) pressure segment using high-resolution manometry. Methods: Two hundred eighty-five patients (108 men, aged 18–87) were studied. A solid-state manometric assembly with 36 circumferential sensors spaced 1 cm apart was placed transnasally, and simultaneous intra-esophageal and intragastric pressures were measured over 6–8 respiratory cycles. Separation of the lower esophageal sphincter (LES) and crural diaphragm was quantified by measuring the distance between the two EGJ elements during inspiration. The association between anthropometric variables and pressure values were examined using univariate and multivariate analysis. Results: There was a significant correlation of body mass index (BMI) and waist circumference (WC) with intragastric pressure (inspiration, BMI [r = 0.57], WC [r = 0.62] P < .0001; expiration, BMI [r = 0.58], WC [r = 0.64], P < .0001) and gastroesophageal pressure gradient (GEPG) (inspiration, BMI [r = 0.37], WC [r = 0.43], P < .0001; expiration, BMI [r = 0.24], WC [r = 0.26], P < .0001). Multivariate analysis adjusting for age, gender, and patient type did not alter the direction or magnitude of this relationship. In addition, obesity was associated with separation of the EGJ pressure components (BMI, r = 0.17, P < .005; WC, r = 0.21, P < .001). Conclusions: Obese subjects are more likely to have EGJ disruption (leading to hiatal hernia) and an augmented GEPG providing a perfect scenario for reflux to occur. Whether or not weight loss can reverse these abnormalities is unknown.

Abbreviations used in this paper:  CDi, crural diaphragm , EGJ, esophagogastric junction , GEPG, gastroesophageal pressure gradient , IQR, interquartile range , NHANES, National Health and Nutrition Examination Survey , WC, waist circumference.

 

The prevalence of obesity has increased to epidemic proportions in the United States during the past 3 decades. According to data from the National Health and Nutrition Examination Survey (NHANES), the prevalence of obesity (defined as body mass index ≥30) has increased from approximately 15% in 1980 (NHANES II) to almost 31% in 1999 (NHANES Continuous 1999–2000).1 Similarly, the prevalence of gastroesophageal reflux disease (GERD) has been rising in the United States with studies suggesting it may now be as high as 20%.2, 3 Furthermore, a high body mass index (BMI) has been shown to be associated with an elevated risk of GERD4, 5, 6 and a specific dose–response relationship between increasing BMI and prevalence of GERD has been demonstrated.3, 7, 8 Illustrative of that, El-Serag et al showed that obese patients were 2.5 times as likely as those with normal BMI (<25) to have GERD symptoms or esophageal erosions.7 Given these parallel time trends and epidemiologic associations, it is tempting to speculate that obesity may in some way promote the development of GERD.

Of several proposed mechanistic hypotheses to explain how obesity may increase the risk of GERD, the most attractive focus on the mechanical stresses imposed on the esophagogastric junction (EGJ), specifically, increased pressure stress on the EGJ and anatomic disruption of the EGJ. Pressure stress on the EGJ would be manifested by an increased intragastric pressure, this being the driving force behind reflux. Anatomic disruption of the EGJ would result in the formation of a hiatal hernia. However, data showing an association between obesity and either increased intragastric pressure or hiatal hernia are lacking or contradictory. Stene-Larsen et al9 showed that a coexisting hiatus hernia was found in 68% of patients with reflux esophagitis and that obesity was significantly associated with both conditions. Wilson et al10 also showed a significant correlation between obesity and GERD, but only in the presence of hiatal hernia. Data on GEPGs in obese patients and its contribution to GERD are also both rare and conflicting.11, 12, 13, 14 Haslam et al15 reported no correlation between intragastric pressure and BMI.

The difficulty in establishing a relationship between obesity, intragastric pressure, and hiatal hernia is likely attributable to methodologic difficulties in quantifying the latter two. In the case of intragastric pressure, existing studies have largely used water-perfused manometry, which is subject to hydrostatic effects, making it difficult to reference intracorporal pressure to atmospheric pressure. In the case of hiatal hernia, little consistency exists among studies or investigators in the criteria for identifying and measuring hiatal hernia.16 Thus, the aim of the current study was to analyze the relationship between obesity, pressure stresses on the EGJ, and the morphology of the EGJ pressure segment itself using state of the art manometric techniques that are relatively free of the methodologic limitations characteristic of earlier analyses.

Back to Article Outline

Materials and Methods 

Patients 

Three hundred fifty-five consecutive patients were studied prospectively using a standardized manometric protocol. Seventy subjects had previous gastric surgery, achalasia, or mechanical obstruction of the EGJ and were excluded. Manometric data from the remaining 285 patients (108 men, ages 18–87) were analyzed. Patients were enrolled from the gastrointestinal diagnostic laboratory at Northwestern Memorial Hospital without regard to presenting complaint. The subjects were classified into 3 specific categories based on presenting symptoms using a questionnaire incorporating a visual analog scale (0–10) for heartburn, regurgitation, dysphagia, and chest pain. Subjects with predominant complaints of heartburn and/or regurgitation were classified as having GERD, whereas subjects with predominant complaints of dysphagia and no heartburn were classified as dysphagia and no GERD symptoms. Subjects with predominant symptoms of chest pain, globus, or abdominal pain were classified as having atypical symptoms. The study protocol was approved by the Northwestern University Institutional Review Board, and informed consent was obtained from each subject.

High-Resolution Manometry 

A solid-state manometric assembly with 36 circumferential sensors spaced at 1-cm intervals (OD 4.2 mm) was used (Sierra Scientific Instruments, Inc., Los Angeles, CA). This device uses proprietary pressure transduction technology (TactArray) that allows each of the 36 pressure-sensing elements to detect pressure over a length of 2.5 mm in each of 12 radially dispersed sectors (Figure 1). The sector pressures are then averaged, making each of the 36 sensors a circumferential pressure detector with the extended frequency response characteristic of solid-state manometric systems and free of the hydrostatic influences characteristic of water-perfused systems. Before the recording, the transducers were calibrated at 0 and 100 mm Hg using externally applied pressure. The response characteristics of each sensing element were such that they could record pressure transients in excess of 6000 mm Hg/sec and were accurate to within 1 mm Hg of atmospheric pressure for measurements obtained during the final 5 minutes of the study, immediately prior to the time of thermal recalibration (see below).

  • View full-size image.
  • Figure 1. 

    Solid-state manometric assembly with 36 sensors spaced at 1-cm intervals. Each pressure sensor consists of 12 radially dispersed sensing elements that are 2.5 mm in length. Sector pressures are averaged within each sensor making is circumferentially sensitive.

Study Protocol 

Anthropometric measures were obtained by our motility technician on all patients (weight, height, waist circumference [WC]) using a tape measure and scale just before the manometric protocol was undertaken. WC was measured at the greatest abdominal circumference with a tape measure. The patients then underwent transnasal placement of the manometric assembly and the catheter was positioned to record from the hypopharynx to the stomach. Studies were done in a supine position and the manometric assembly was positioned with at least 5 intragastric sensors to optimize EGJ and intragastric recording. The catheter was then taped to the nose. The manometric protocol included at least ten 5-mL swallows and a 5-minute period to assess basal sphincter pressure.

Data Analysis 

Manometric data were initially analyzed using ManoView analysis software (Sierra Scientific Instruments, Inc.). First, the data were corrected for thermal sensitivity of the pressure-sensing elements using the thermal compensation functionality of the ManoView software. This was done by visually identifying the instant in the recording that the assembly was pulled from the nose. Immediately after that instant, the catheter was still at body temperature, but all pressure sensors were exposed to atmospheric pressure. The software routine then sets this pressure as 0 and calculates the magnitude of the pressure correction that was required for each sensing element. Those sensor-specific thermal correction factors are then applied to the entire manometric data set, in essence correcting it for temperature-dependent calibration drift. Note that, although this sensor technology is subject to thermal drift, that effect is almost linear so that the correction factors applied to reestablish the 0 reference will also correct error attributable to thermal drift in the entire data set.

After thermal correction was applied, the EGJ pressure profile was analyzed. The components of the EGJ are easily identified on the isocontour plots as abrupt transitions in the pressure topography (Figure 2). The proximal border of the lower esophageal sphincter (LES) was defined by the abrupt transition to intraesophageal pressure. The axial position of maximal intrinsic sphincter pressure (LESmax) was defined during expiration as the first pressure peak encountered progressing into the EGJ from the esophagus. The crural diaphragmatic (CDi) component of EGJ pressure was defined as the axial level characterized by maximal inspiratory pressure augmentation. Note that, in individuals with normal anatomy, LESmax and CDimax were superimposed and indistinguishable. In such individuals it was simply denoted as EGJmax. Alternatively, LESmax and CDimax were separated but, because the distal border of the LES and the proximal border of the CDi overlapped, they appeared as a double-peaked axial pressure profile with a valley between the peaks that remained greater than intragastric pressure. Only when the axial separation between LESmax and CDimax reached 3–4 cm was there a well-defined hiatal hernia pressure not subject to the influence of the LES or CDi (Figure 2B).

  • View full-size image.
  • Figure 2. 

    Isocontour manometric plots of the EGJ of a patient with normal anatomy (A) and a patient with hiatal hernia (B). The isocontour representation provides an overview of regional differences in intraluminal pressure integrated over time. The on-screen pressure scale was color coded, but in this illustration a gray scale is used with each isocontour level representing a 5-mm Hg increment in pressure. High-pressure zones, such as the EGJ or the separated components of the EGJ, are easily recognized by sharp transitions in pressure. The exact pressure at any point on this spatiotemporal grid can be ascertained either using the isobaric contour tool or the point-and-click smart mouse tool of the ManoView software. The time of peak inspiration is identified by maximal augmentation of EGJ, intragastric, or crural diaphragm (CDi) pressure, depending on an individual’s anatomy. Expiratory pressures were measured midway between adjacent peak inspirations. The distal border of the LES and the proximal border of the CDi typically overlap; however, once separation occurs they can be distinguished by the sharp transitions in pressure levels as in (B). The axial separation between LESmax and CDimax was quantified by taking the midpoint of the 2 EGJ components and determining the distance between them using the manometric ports as a distance reference. When separation exceeded 3–4 cm, a well-defined hiatal hernia pressure not subject to the influence of the LES or crural diaphragm could be ascertained (B).

Simultaneous intraesophageal pressure and intragastric pressure were measured over 6–8 respiratory cycles at the peak of inspiration (inspiratory pressure) and at the midpoint between successive peak inspirations (expiratory pressure) using the isobaric contour tool in ManoView. The isobaric contour tool allows delineation of the anatomic/temporal boundaries of a pressure domain of designated magnitude. Intraesophageal pressure was measured 2 cm above the proximal aspect of the EGJ and intragastric pressure was measured 2 cm below the distal aspect of the CDi. When present, intrahernia pressure was measured at the midpoint between the proximal aspect of the extrinsic sphincter and the distal aspect of the intrinsic sphincter. The gastroesophageal pressure gradient (GEPG) was calculated using the average values of the simultaneous intra-esophageal and intragastric pressure measurements. All pressure measurements were referenced to atmospheric pressure. Analysis of separation of the LES and CDi was performed by measuring the distance between LESmax and CDimax at peak inspiration (Figure 2B).

Characterization of EGJ pressure morphology was performed with a computer program (Matlab) customized for processing binary manometric data into spatial variation plots (The MathWorks Inc., Natick, MA). This was done by exporting the binary manometry data in ASCII text format for processing and storage. These ASCII files were then reconverted into a known binary format for use in Matlab and isocontour or spatial pressure variation plots were generated. For these plots to appear smooth (as opposed to notched), the dataset was enhanced both in the time dimension (between sampling times) and in the spatial dimension (between pressure recording sites). This interpolation was done using a cubic spline algorithm implemented on a finely resolved rectilinear space–time grid to generate intermediate data points, resulting in a virtual increase in the spatial data from 1 to 10 recording sites per centimeter and doubling the temporal sampling from 35 to 70 Hz (Figure 2). Isocontour increments were adjusted to resolve low- or high-pressure activity as appropriate.

Statistical Analysis 

BMI was calculated as weight in kilograms divided by the height in meters squared and was examined as a continuous as well as a categorical variable (normal, <25; overweight, between 25 and 30; obese, >30). The associations between BMI and pressure values across the EGJ were examined in univariate and multivariable analyses. Intraesophageal pressure, intragastric pressure, and the GEPG were analyzed during both inspiration and expiration. Parametric variables were summarized as mean and SD and compared using ANOVA. Nonparametric variables were summarized as median with interquartile range (IQR) and compared using Wilcoxon nonparametric tests (Kruskall–Wallis). A P < .05 was considered significant. Pearson’s correlation coefficient was calculated. Least-square regression analysis was used to examine the magnitude of change in pressure related to various independent variables (BMI and WC) and dependent outcomes (intragastric pressure, intraesophageal pressure, and GEPG). These associations were further examined while adjusting for potential confounders of age and gender in multivariate regression analysis. We added WC to the regression models to examine the possibility of this variable serving as a potential mediator (in the causal pathway) of the effect of BMI on the observed EJG pressures. Logistic regression analysis was also performed to assess the relationship between various independent variables (LES pressure, GEPG, EGJ component separation, BMI, and age) and GERD as a dependent categorical outcome. Last, we conducted several sensitivity analyses to test the robustness of the findings including separate analyses for men and women, for patients with and without GERD symptoms, and an analysis that excludes outlier values for weight and height.

Back to Article Outline

Results 

A total of 285 patients were studied. The mean age of the patients was 51.0 years (SD, 15.7) and their mean height and weight were 169.1 cm (SD, 10.0) and 77.9 kg (SD, 22.0), respectively. Mean BMI and WC of the patients was 27.2 kg/m2 (SD, 7.8) and 94.5 cm (SD, 16.6). Approximately two-thirds were women (62%). There were no significant differences in BMI between men and women, but men had a higher WC than women (98.9 versus 90.9 cm). Using visual analog scale data from the symptom assessment questionnaire to assess severity of heartburn, regurgitation, dysphagia, chest pain, and other miscellaneous symptoms, the subjects were classified as GERD (n = 186), dysphagia without GERD symptoms (n = 67), or atypical symptoms (n = 32). GERD patients had significantly increased mean GERD severity scores (heartburn score + regurgitation score) compared to subjects classified as non-GERD (GERD 10.3, SD 4.6; non-GERD 1.1, SD 2.7, P < .0001).

Intragastric Pressure 

Intragastric pressure was significantly higher in obese and overweight patients compared with those with a normal BMI during both expiration and inspiration (Figure 3). There was also a suggestion of a dose–response relationship as evidenced by the nonoverlapping 95% confidence interval (CI) for the associations between the 3 subject groups. There was a (strong) positive correlation between BMI and intragastric pressure during inspiration (r = 0.57, P < .0001) and expiration (r = 0.58, P < .0001) (Table 1, Figure 4A). Unadjusted linear regression revealed that intragastric pressure would increase by 0.3 mm Hg (SE, 0.03) per unit increase in BMI during both inspiration and expiration. Similar to BMI, there was a strong correlation between WC and intragastric pressure during inspiration (r = 0.62, P < .0001) and expiration (r = 0.64, P < .0001) (Figure 4B). Unadjusted linear regression revealed that pressure would increase by 0.16 mm Hg (SE, 0.01) per centimeter increase in WC during both inspiration and expiration, or 0.4 mm Hg per inch.

  • View full-size image.
  • Figure 3. 

    Mean intragastric and intraesophageal pressures during inspiratory and expiratory phases of respiration. ANOVA was used to compare the pressure values between the 3 subject groups. Gastric expiration pressure: obese subjects had significantly higher gastric pressures compared to overweight subjects (P < .0001) and normal BMI subjects (P < .0001), whereas overweight subjects had a significantly higher gastric pressure than normal subjects (P < .0001). Gastric inspiration pressure: obese subjects had significantly higher gastric pressures compared with overweight subjects (P < .0001) and normal BMI subjects (P < .0001) whereas overweight subjects had a significantly higher gastric pressure than normal subjects (P < .0001). Esophageal expiration pressure: obese subjects had significantly higher intraesophageal pressures compared to overweight subjects (P = .013) and normal BMI subjects (P < .0001), and overweight subjects had a significantly higher intraesophageal pressure than normal subjects (P = .002). Esophageal inspiration pressure: there was no statistically significant difference between the 3 groups.

Table 1. Association Between Pressure Measurements of the Esophagogastric Junction, BMI (kg/m2), and Waist Circumference (cm) During Expiration and Inspiration in 285 patients
Expiratory PressureInspiratory Pressure
GastricEsophagealGEPGGastricEsophagealGEPG
BMI
Correlation coefficient0.570.360.240.550.090.38
Regression coefficient0.300.170.120.300.040.26
P value<.0001<.0001<.0001<.00010.18<.0001
Waist circumference
Correlation coefficient0.660.440.260.640.110.43
Regression coefficient0.160.010.060.160.020.14
P value<.0001<.0001<.0001<.00010.16<.0001

Abbreviation: GEPG, gastroesophageal pressure gradient.

  • View full-size image.
  • Figure 4. 

    (A) Association between intragastric pressure and BMI during both respiratory phases. (B) Association between intragastric pressure and WC during both respiratory phases. *P < .0001 using Pearson correlation calculation.

In a multiple linear regression analysis, the association between BMI and intragastric pressure was not altered in direction or magnitude after controlling for age, gender, and patient type (GERD versus non-GERD; Table 2). The magnitude and direction of the association between WC and intragastric pressure was also not altered by adjustment for these same variables (Table 2). A model was examined to assess the association between gastric pressure (both inspiratory and expiratory) with both BMI and WC. In this model, WC was an independent risk factor for increased intragastric pressure, whereas BMI was no longer significantly associated with pressure. For example, for expiratory intragastric pressure, the parameter estimate for WC in a model that adjusted for age, gender, and patient type, was 0.16 (SE 0.02, P < .0001), whereas that of BMI was 0.05 (SE 0.04, P = .15). Furthermore, we conducted stratified analyses in men and women separately and observed significant associations between intragastric pressure and both BMI and WC. The associations were stronger in men than women (data not shown).

Table 2. Association Between Both BMI and Waist Circumference and the Pressure Measurements Across the Esophagogastric Junction in 285 Patients Using Multiple Linear Regression That Adjusted for Age, Gender, and Patient Type (GERD vs non-GERD)
Expiratory PressureInspiratory Pressure
GastricEsophagealGEPGGastricEsophagealGEPG
BMI
Correlation coefficient0.570.350.250.550.080.37
Regression coefficient0.290.170.130.300.040.26
P value<.0001<.0001<.0001<.00010.20<.0001
Waist circumference
Correlation coefficient0.700.430.260.670.090.47
Regression coefficient0.170.10.060.170.020.15
P value<.0001<.0001<.0001<.00010.15<.0001

Intraesophageal Pressure 

Intraesophageal pressure during both expiration and inspiration was significantly higher in overweight and obese patients as compared with a normal BMI (Figure 3). There was a suggestion of a dose–response relationship as evidenced by the nonoverlapping 95% CI for the associations between overweight and obesity with expiratory but not inspiratory intraesophageal pressure. There was a significant positive correlation between intraesophageal pressure at expiration and both BMI and WC (Table 1). In addition, it was estimated that an increase of 1 BMI unit was associated with a 0.17 mm Hg (SE, 0.03) increase in expiratory intraesophageal pressure and a 1 cm increase in WC was associated with 0.1 mm Hg (SE, 0.01) increase in expiratory intraesophageal pressure. There was a weaker association between inspiratory intraesophageal pressure and BMI or WC (Table 1). Controlling for age, gender, and patient type did not reduce the association of WC and BMI with expiratory intraesophageal pressure (Table 2).

Similar to intragastric pressure, in a model that examined WC and BMI simultaneously, only WC revealed a statistically significant relationship with expiratory intraesophageal pressure. The parameter estimate for WC in a model that adjusted for age, gender, and patient type was 0.09 (SE 0.02, P < .0001) whereas that of BMI was 0.03 (SE 0.04, P = .63).

Gastroesophageal Pressure Gradient 

The mean GEPG during both inspiration and expiration was significantly increased in obese patients compared to overweight or normal subjects (Figure 5). There was no difference in the GEPG between normal and overweight subjects during expiration; however, there was a significant difference during inspiration between overweight subjects and subjects with a BMI <25. Although expiratory intraesophageal and intragastric pressure both increased with higher BMI, the disproportionate increase in gastric pressure resulted in a significant association between expiratory GEPG and BMI (Table 1). There was a significant positive association between BMI and the GEPG during inspiration (Table 1). A similar association was found between WC and GEPG for both respiratory phases (Table 1). Controlling for age, gender, and patient type did not alter the direction or magnitude of the association between BMI and the GEPGs during both expiration and inspiration (Table 2). Similarly, the same adjustments did not alter the direction or magnitude of the association between WC and the GEPG during either respiratory phase (Table 2). A model that examined WC and BMI simultaneously while controlling for age, gender, and patient type showed WC only to be statistically significant. The parameter estimate for WC in this model was 0.07 (SE 0.02, P < .0001) during expiration and 0.14 (SE 0.03, P < .0001) during inspiration. In all the analyses described, older age was associated with increased intragastric and intraesophageal pressures as well as GEPG and component separation.

  • View full-size image.
  • Figure 5. 

    Mean GEPG during both respiratory phases. Expiratory: obese vs normal BMI, P < .001 using ANOVA. Inspiratory: obese vs normal BMI, P < .0001, obese vs overweight, P < .0001, normal BMI vs overweight, P = .01, using ANOVA.

Sensitivity analyses were conducted for men and women separately, for GERD and non-GERD separately, and for the entire group excluding 4 outlier observations for weight and height. In general, there were no differences in the direction or significance of the association observed in these secondary analyses compared to the primary analyses described. However, parameter estimates of BMI were higher in men than women; for example, for gastric pressure the parameter estimate for men was 0.59 (SE 0.08) and 0.27 (SE 0.03) for women. These differences could be explained by differences in WC; BMI analyses conducted in strata of similar WCs were largely similar between men and women.

EGJ Component Separation 

Spatial separation between the EGJ component attributable to the LES and that contributed by the CDi increased significantly in overweight and obese patients compared with patients with a normal BMI (Figure 6). The correlation between anthropometric measurements and separation of EGJ pressure components was statistically significant (BMI, r = 0.17, P < .005; WC, r = 0.21, P < .001). In regression analyses, these associations remained significant after controlling for age, gender, and patient type (GERD versus non-GERD).

  • View full-size image.
  • Figure 6. 

    (A) Mean values for sphincter separation (B) Mean EGJmax pressure during both respiratory phases. Using ANOVA, there was a significant increase in sphincter separation when normal BMI subjects were compared to overweight subjects (P = .005) and obese subjects (P = .001). There was no statistical difference between the sphincter separation when overweight and obese subjects were compared (P = .80). There was also no significant difference in the basal LES pressure between the 3 groups during both respiratory phases.

EGJ Component Separation and EGJmax Pressure 

BMI was not significantly correlated with EGJmax pressure during expiration (r = 0.04, P = .43) or inspiration (r = 0.01, P = .98). However, there was a significant association between EGJ component separation and inspiratory EGJmax pressure with a regression coefficient of −6.7 mm Hg per each 1 cm of separation (Figure 7). The relationship between EGJ component separation and expiratory EGJmax pressure was weaker, but still significant (parameter estimate of −2.0 mm Hg per each 1 cm of separation; see Figure 7). Respiration had a substantial effect on EGJmax pressure (Table 3) and this effect was dependent on the degree of axial separation of the EGJ components rather than BMI or WC. This impact of EGJ component separation on EGJmax pressure is illustrated in the spatial pressure variation plots in Figure 8A, B, and C depicting the 3 possible interrelationships between the LES and the CDi: completely overlapping (Figure 8A), partially overlapping (Figure 8B), and nonoverlapping EGJ pressure components (Figure 8C). Subjects with axial separation greater than 3 cm appeared to lose the respiratory component of EGJmax pressure (Figure 8C). In fact, in such individuals, expiratory EGJmax pressure was greater than inspiratory EGJmax pressure because inspiration had the effect of subtracting from the EGJ component attributable to the intrinsic LES, presumably because the sphincter resided within the thoracic space (Figure 8C, Table 3). Additionally, there was no significant difference in inspiratory EGJmax pressure in subjects with EGJ component separation greater than 3 cm comparing subjects with a BMI of <25 to those with a BMI >30.

Table 3. The Effect of Separation Between the LES and CDi on LES Pressure in Patients With Low and High BMI
Expiratory EGJmax in mm Hg (IQR)Inspiratory EGJmax in mm Hg (IQR)
CD to LES separation (cm)CD to LES separation (cm)
≤2(n=99)>2,<3(n=19)≥3(n=10)≤2(n=88)>2,<3(n=20)≥3(n=13)
BMI ≤2525.4(12.8)a19.3(11.5)ac17.3(9.1)ac40.9(17.5)32.1(16.7)c11.5(8.3)bc
BMI ≥3027.5(10.1)a25.2(14.0)a21.5(9.9)a45.9(16.3)36.6(19.9)c15.3(14.4)bc

a P <.05 using ANOVA vs inspiration.

b P <.05 using ANOVA vs separation of >2, <3 with same BMI.

c P <.05 using ANOVA vs separation ≤2 with same BMI.

  • View full-size image.
  • Figure 8. 

    Spatial pressure variation plots of a patient with (A) normal EGJ anatomy, (B) an obese patient with clear separation but overlap between the LES and crural diaphragm, and (C) a patient with a large hiatal hernia. These plots depict pressure along the length of the EGJ as a function of time. Each vertical line represents the simultaneous pressure values along the EGJ at that specific time and each line is spaced every 0.2 seconds. (A) In the normal weight subject without hiatal hernia the locus of maximal inspiratory pressure augmentation indicative of the CDi is superimposed on the LES making the 2 indistinguishable (EGJmax indicated by dots). Nonetheless, there are 2 distinct zones to the EGJ; a proximal high-pressure zone and a subdiaphragmatic low-pressure zone. The subdiaphragmatic low-pressure zone is most evident during expiration and likely represents the sphincteric contribution of the gastric sling and clasp fibers. (B) In this obese subject, gastric pressure is clearly increased. Additionally, the morphology of the EGJ is altered such that there is a 2-cm separation between LESmax (proximal dots) and CDimax (distal dots). Note that the lumen surrounded by the crural diaphragm pressurizes to intragastric pressure with expiration significantly shortening the EGJ high-pressure zone analogous to what is seen with hiatal hernia. (C) This subject has complete separation of the LES and the CDi (5 cm) with no overlap between the 2 high-pressure zones. Note that expiratory LESmax was greater than inspiratory LESmax because inspiration had the paradoxical effect of subtracting from LES pressure, presumably because the sphincter resided within the thoracic space.

Intrahernia Pressure 

Analysis of the spatial pressure variation plots (Figure 8) revealed that separation of at least 4 cm was required to accurately assess intrahernia pressure without inadvertently measuring superimposed LES or CDi pressures (Figure 8C). Separation <4 cm was associated with overlap of the LES and CDi to the point that a contiguous high-pressure zone was present (Figure 8B). Fifteen subjects had sufficient separation of the 2 EGJ components to allow measurement of intrahernia pressure. The median intrahernia pressure during expiration was significantly increased compared to inspiration (12.0 mm Hg, IQR 3.8 versus 9 mm Hg, IQR 4.5; P < .005). Both expiratory and inspiratory intrahernia pressure were positively correlated with concomitant gastric pressure (r = 0.61 and 0.54, P < .005), but not esophageal pressure. The median intrahernia to esophagus pressure gradient was greater during inspiration compared with expiration (inspiration 10.0 mm Hg, IQR 5.8; expiration, 7.0 mm Hg, IQR 5.8, P < .005).

Effect of EGJ Morphology and Pressure Profile on GERD Symptoms 

Patients classified as GERD had significantly increased GEPG during expiration compared to non-GERD patients (GERD, 3.6 mm Hg, SE 0.32; non-GERD, 2.6 mm Hg, SE 0.35, P = .03), whereas there was a trend toward increased GEPG during inspiration in the GERD patients compared with non-GERD patients (GERD, 6.5 mm Hg, SE 0.32; non-GERD, 5.6 mm Hg, SE 0.38, P = .08). EGJ component separation was also significantly increased in GERD patients compared with non-GERD patients (GERD, 1.8 cm, SE 0.09; non-GERD, 1.4 cm, SE 0.1, P = .02). Logistic regression analysis was performed to assess the relationship between the independent physiologic variables measured (LES pressure, GEPG, EGJ component separation) and GERD as a dependent categorical outcome while controlling for age and BMI. This analysis revealed that separation of the EGJ components (regression coefficient, 0.46; SE, 0.15; P = .002) and GEPG during inspiration (regression coefficient, 0.10, SE 0.04; P = .004) and expiration (regression coefficient, 0.11; SE, 0.04; P = .002) were significantly associated with GERD while controlling for BMI and age. LES pressure during both phases of respiration was not significantly associated with GERD after controlling for age and BMI.

Back to Article Outline

Discussion 

The major findings of this study were that the pressure morphology within and across the EGJ is altered with obesity in a fashion that would augment the flow of gastric juice into the esophagus. Particularly during the inspiratory phase of respiration, increased intragastric pressure and GEPG were strongly correlated with increased BMI. Both factors correlated even more strongly with WC suggesting this to be the mediator in the causal pathway of the BMI effect. There was a dose–response relationship such that the higher the BMI and WC, the greater the intragastric pressure and GEPG. Obesity was also associated with increased axial separation between the LES and the extrinsic crural diaphragm, an objective measure of the anatomic disruption of the EGJ culminating in the development of hiatal hernia. Taken together, these findings offer a physiologic explanation for the association between obesity and GERD.

Although obesity could potentially be associated with a host of dietary and lifestyle factors relevant to GERD pathogenesis, this investigation focused on mechanical challenges to EGJ competence. Broadly considered, these mechanical challenges can be in the form of EGJ disruption causing it to function abnormally or pressure perturbations across the EGJ that adversely affect normal physiologic function. Focusing first on EGJ disruption, it is widely recognized that hiatus hernia has broad pathophysiologic implications in GERD pathogenesis: increased incidence of strain-induced reflux,17 reduced threshold for distention induced transient LES relaxations,18 swallow induced reflux,19, 20 reduced LES pressure,21, 22 and impaired acid clearance.19, 20 By leveraging high-resolution manometry methodology, we were able to parameterize the continuum of anatomic EGJ disruption leading to hiatal hernia with a measurement (LES to CDi separation) not subject to the confounding effect of swallow, gastric distention, or esophageal distention. Evident in Figure 8, LES to CDi separation could be resolved to within 1 cm and increasing LES to CDi separation was associated with decreased EGJmax pressure, particularly during inspiration. This, coupled with the observation that the GEPG is greatest during inspiration, and greater yet in obese individuals during inspiration (Figure 5), strongly supports the relevance of LES to CDi separation as both an important physiologic parameter and as a likely mediator of the obesity effect on GERD.

The most obvious pressure stress leading to LES to CDi separation and eventually hiatal hernia is increased intra-abdominal pressure that, in a fasting state, is equal to intragastric pressure. One of our central aims in this study was to circumvent the limitations of earlier manometric investigations of intragastric pressure using solid-state high-resolution manometry allowing us to obtain accurate pressure measurements across the EGJ with all pressures referenced to atmospheric pressure. Doing so, we could clearly demonstrate that intragastric pressure had a positive correlation with both BMI and WC, supporting the contention that there is increased mechanical stress on hiatal integrity with obesity.

Once flow of gastric juice across the EGJ is initiated, flow rate is directly proportional to the trans-EGJ pressure gradient. Existing literature on this measurement is limited. Mercer et al13 analyzed the GEPG in 8 healthy volunteers, 5 lean reflux patients, and 16 obese subjects and reported that obese subjects had a pressure gradient of 12.3 ± 0.6 mm Hg whereas lean healthy subjects and lean refluxers had pressure gradients of 9.4 ± 0.8 mm Hg and 10.0 ± 1.1 mm Hg, respectively. Comparing those values to our own, they approximate the peak inspiratory gradients that we observed. Another aspect of this measurement accentuated by our data set is the compounding effects of obesity and respiration on the GEPG. Although obesity augments gastric and to a lesser degree esophageal pressure (Figure 3), it also results in increased inspiratory effort resulting in a greater increment of GEPG increase during inspiration (Figure 3, Figure 5).

The findings of this study also provide support for the role of abdominal obesity, reflected as greater WC, as a mediator of the effect of obesity on gastric pressure. Although both BMI and WC were significantly associated with pressures, when examined separately, models that examined both variables simultaneously showed WC to be independently associated with pressures, and the effect of BMI became nonsignificant or greatly reduced. Furthermore, WC seems to explain the finding that for the same changes in BMI, men in this study had approximately double the effect on pressure profiles than women. Although men and women in this study had similar BMI, men had on average 10 cm greater WC and when the effect of BMI on pressure profiles was compared between men and women in strata of similar WC, no differences were seen. Last, abdominal obesity can explain some of the epidemiologic features of severe GERD including erosive esophagitis, Barrett’s esophagus, and esophageal adenocarcinoma. For example, the distribution of body fat tends to be more visceral than truncal in high-risk groups for severe GERD: Caucasians (compared with African Americans) and men (compared with women).23

The direct relationship between GERD and the pressure morphology of the EGJ was assessed by comparing these parameters in patients with and without GERD. Patients classified as having GERD tended to have higher GEPG during both phases of respiration and significantly greater separation of the EGJ components compared with non-GERD patients. A logistic regression model with GERD as the dependent variable also revealed that this relationship was not altered after controlling for age and BMI. However, these results should be interpreted with caution because the method of classifying GERD was based solely on symptoms at presentation. In addition, it is possible that a substantial proportion of the atypical patients may have reflux disease and this could have weakened the relationship. Future studies incorporating objective data, such as 24-hour ambulatory pH data and validated GERD questionnaires are needed to confirm these findings.

In conclusion, attempting to equate obesity with GERD withstands scrutiny no better than equating reflux disease with any single physiologic abnormality such as LES hypotension, ineffective esophageal motility, or hiatal hernia. GERD is clearly a multifactorial disease. However, obese subjects are more likely to have a hiatal hernia, increased intragastric pressure, and an augmented GEPG providing the perfect scenario for reflux to occur. This has clinical significance because weight loss is considered to be an appropriate lifestyle modification to improve GERD symptoms. Unfortunately, it will be difficult to predict what degree of weight loss is required to alter the pressure profile across the EGJ. In addition, given that increased LES to CDi separation was also observed in nonobese patients, this anatomic change may well persist despite weight loss.

Back to Article Outline

 

Contributed devices from Sierra Scientific, Inc.

Back to Article Outline

References 

  1. Flegal KM , Carroll MD , Ogden CL , Johnson CL . Prevalence and trends in obesity among US adults, 1999–2000 . JAMA . 2002;288:1723–1727
  2. Dent J , El-Serag HB , Wallander MA , Johansson S . Epidemiology of gastro-oesophageal reflux disease (a systematic review) . Gut . 2005;54:710–717
  3. Locke GR , Talley NJ , Fett SL , Zinsmeister AR , Melton LJ . Risk factors associated with symptoms of gastroesophageal reflux . Am J Med . 1999;106:642–649
  4. Fisher BL , Pennathur A , Mutnick JL , Little AG . Obesity correlates with gastroesophageal reflux . Dig Dis Sci . 1999;44:2290–2294
  5. Nilsson M , Lundegardh G , Carling L , Ye W , Lagegren J . Body mass and reflux oesophagitis (an oestrogen-dependent association?) . Scand J Gastroenterol . 2002;37:626–630
  6. Diaz-Rubio M , Moreno-Elola-Olaso C , Rey E , Locke GR , Rodriguez-Artalejo F . Symptoms of gastro-oesophageal reflux (prevalence, severity, duration and associated factors in a Spanish population) . Aliment Pharmacol Ther . 2004;19:95–105
  7. El-Serag HB , Graham DY , Satia JA , Rabeneck L . Obesity is an independent risk factor for GERD symptoms and erosive esophagitis . Am J Gastroenterol . 2005;100:1243–1250
  8. Murray L, Johnston B, Lane A, Harvey I, Donovan J, Nair P, et al. Relationship between body mass and gastro-oesophageal reflux symptoms (The Bristol Helicobacter Project) . Int J Epidemiol . 2003;32:645–650
  9. Stene-Larsen G , Weberg R , Froyshov Larsen I , Bjortuft O , Hoel B , Berstad A . Relationship of overweight to hiatal hernia and reflux oesophagitis . Scand J Gastroenterol . 1988;23:427–432
  10. Wilson LJ , Ma W , Hirschowitz BI . Association of obesity with hiatal hernia and esophagitis . Am J Gastroenterol . 1999;94:2840–2844
  11. Sanchez NC , Tenofsky PL , Dort JM , Shen LY , Helmer SD , Smith RS . What is normal intra-abdominal pressure? . Am Surg . 2001;67:243–248
  12. Sugerman H , Windsor A , Bessos M , Wolfe L . Intra-abdominal pressure, sagittal abdominal diameter and obesity comorbidity . J Intern Med . 1997;241:71–79
  13. Mercer CD , Rue C , Hanelin L , Hill LD . Effect of obesity on esophageal transit . Am J Surg . 1985;149:177–181
  14. Zacchi P , Mearin F , Humbert P , Formiguera X , Malagelada JR . Effect of obesity on gastroesophageal resistance to flow in man . Dig Dis Sci . 1991;36:1473–1480
  15. Haslam N , Syndercombe A , Zimmer CR , Edmondson L , Duggan JE . Intragastric pressure and its relevance to protective cricoid force . Anaesthesia . 2003;58:1012–1015
  16. Kahrilas PJ . The role of hiatus hernia in GERD . Yale J Biol Med . 1999;72:101–111
  17. Sloan S , Rademaker AW , Kahrilas PJ . Determinants of gastroesophageal junction incompetence (hiatal hernia, lower esophageal sphincter, or both?) . Ann Intern Med . 1992;117:977–982
  18. Kahrilas PJ , Shi G , Manka M , Joehl RJ . Increased frequency of transient lower esophageal sphincter relaxation induced by gastric distention in reflux patients with hiatal hernia . Gastroenterology . 2000;118:688–695
  19. Mittal RK , Lange RC , McCallum RW . Identification and mechanism of delayed esophageal acid clearance in subjects with hiatus hernia . Gastroenterology . 1987;92:130–135
  20. Sloan S , Kahrilas PJ . Impairment of esophageal emptying with hiatal hernia . Gastroenterology . 1991;100:596–605
  21. Mittal RK , Rochester DF , McCallum RW . Effect of the diaphragmatic contraction on lower oesophageal sphincter pressure in man . Gut . 1987;28:1564–1568
  22. Kahrilas PJ , Lin S , Chen J , Manka M . The effect of hiatus hernia on gastro-oesophageal junction pressure . Gut . 1999;44:476–482
  23. Weinsier RL , Hunter GR , Gower BA , Schutz Y , Darnell BE , Zuckerman PA . Body fat distribution in white and black women (different patterns of intraabdominal and subcutaneous abdominal adipose tissue utilization with weight loss) . Am J Clin Nutr . 2001;74:631–636

 Supported by RO1 DC00646 (PJK) and K23 DK062170-01 (JEP) from the Public Health Service.Contributed devices from Sierra Scienti .c,Inc.

PII: S0016-5085(05)02526-6

doi:10.1053/j.gastro.2005.12.016

Refers to article:

  • Pressure Details From the Weight-Challenged Gastroesophageal Junction: More Than the Usual Suspects

    Radu Tutuian, Donald O. Castell
    Gastroenterology March 2006 (Vol. 130, Issue 3, Pages 988-989)

Gastroenterology
Volume 130, Issue 3 , Pages 639-649, March 2006

Article Tools

* Image Usage & Resolution