The Effects of Obesity on Oesophageal Function, Acid Exposure and the Symptoms of Gastro-Oesophageal Reflux Disease

R. Anggiansah; R. Sweis; A. Anggiansah; T. Wong; D. Cooper; M. Fox


Aliment Pharmacol Ther. 2013;37(5):555-563. 

In This Article



Anthropometric data from consecutive patients referred for oesophageal investigation of suspected GERD were collected prospectively over an 18-month period. The severity of symptoms off acid suppression medication was assessed by a validated questionnaire.[25] Exclusion criteria included intake of medications that affect gastrointestinal function within 1 week of testing, intolerance to the pH monitoring catheter or incomplete (<20 h) measurement, previous oesophageal or gastric surgery and co-morbid conditions that affect oesophageal or gastric motility (e.g. achalasia).

Patients were studied in a single oesophageal laboratory at a tertiary referral unit serving a multi-ethnic community of 6 million in London and South-East England. The severity of mucosal disease was not included in the analysis because endoscopy was not always performed in proton pump inhibitor (PPI) naïve patients and findings were incompletely documented in the referral letter. All tests and questionnaires were completed after at least 7 days off acid suppression. All patients provided written consent before undergoing study procedures.

Anthropometric Measurements

A well-calibrated measuring scale was used to measure body weight (kg) and a wall mounted ruler was used to measure height (cm) with the patient standing barefoot. WC was measured using a hand-held measuring tape 5 cm above the iliac crest. For presentation in figures, BMI (<25, 25–29.9 and >30 kg/m2) and WC (<89, 89–99 and >99 cm) were divided into tertiles as defined by established cut-off values.[16]


A 6-pressure sensor solid state manometric catheter (Gaeltec Ltd, Isle of Skye, UK) was used to measure oesophageal motility and the OGJ. The procedure was performed in a semi-recumbent position. All pressures were measured relative to intra-gastric reference during mid-expiration. The catheter was advanced trans-nasally into the stomach and the intrinsic lower oesophageal sphincter (LOS) position and position of the pressure inversion point (PIP) were determined during a station pull-through procedure with at least three respiratory cycles in each catheter position. Measurements included LOS pressure, total and abdominal LOS length. The abdominal LOS was defined by the position of the PIP. The presence of a hiatus hernia was defined by the absence of an abdominal LOS and the distance between the LOS and PIP described its size (i.e. negative abdominal LOS length). Peristaltic function was assessed by 10 × 5 mL water swallows separated by at least 30 s.

For the purposes of this study, OGJ function and structure were defined by LOS pressure and abdominal LOS length respectively, and peristaltic function was defined by distal contractile pressure 5 cm above the LOS. Other variables, such as total LOS length and peristaltic velocity, showed no independent correlation with obesity or oesophageal acid exposure and were not included in the regression models.

Ambulatory 24 h pH monitoring

Ambulatory oesophageal pH studies were performed by using a single-use catheter with a single antimony electrode (Slimline; Medtronic, Inc., Shoreview, MN, USA). The catheter was calibrated as per the manufacturer's protocol and positioned so that the pH sensor was 5 cm proximal to the superior aspect of the manometrically determined LOS. The catheter was then connected to a portable digital data recorder (Mark III Digitrapper; Medtronic Synectics) where the data were stored. Subjects were encouraged to pursue their usual daily activities and normal diets but to avoid acidic foods and drinks (e.g. citrus fruits, carbonated drinks). All patients were asked to return in 24 h when the catheter was removed and the data downloaded from the data recorder or transposed from the symptom diary. Data were analysed with gastrosoft Software (Medtronic, Inc.).

The severity of reflux disease was defined by oesophageal acid exposure (percentage time <pH 4 during the 24-h monitoring period). A logarithmic transformation of the data was applied to reduce variability in regression models. Other variables derived from ambulatory reflux monitoring, such as the number of reflux events and acid clearance time, were closely associated with oesophageal acid exposure and were not included in the regression models. Similarly, acid exposure in the upright and supine positions was closely associated with total acid exposure and detailed results are not presented.

Statistical Analysis

All baseline demographics are presented as median [interquartile range (IQR)]. The association between reflux symptoms, oesophageal acid exposure, oesophageal function and obesity (measured by BMI and WC) was estimated with the unstandardised regression coefficient and 95% confidence intervals (CIs). The results are presented as the correlation coefficient (R) both unadjusted and adjusted for age and gender (potential confounding factors known to associate with both obesity and reflux). Linear regression models were used to examine the relationship between obesity and oesophageal acid exposure, considering effects of manometric parameters. We examined whether the strength of association between obesity and acid exposure in the regression model (adjusted R 2) was altered by correcting for manometric parameters affecting oesophageal acid exposure. In addition, we examined whether or not any change in the strength of association (adjusted R 2) between obesity and acid exposure was related to the introduction of LOS pressure or intra-abdominal LOS length into the regression analysis. To gain insight into the role of general and abdominal adiposity in GERD, separate regression models were used for BMI and WC. The association between BMI and WC with oesophageal function and acid exposure was also examined by measuring the change in the strength of association (adjusted R 2) that occurred on adding the alternate anthropometric measure of obesity into the regression model. Statistical analysis was performed using spss 16.0 (SPSS Inc., Chicago, IL, USA).