Article Text

Download PDFPDF

Oesophageal pH monitoring in Barrett's oesophagus
  1. C S Neumann1,
  2. B T Cooper
  1. 1City Hospitals NHS Trust, Dudley Road, Birmingham, UK
  1. Correspondence to:
    Dr BT Cooper, Dudley Road Hospital, Birmingham B18 7QH, UK;
  1. R Fass2
  1. 2University of Arizona, GI Motility Laboratories, Southern Arizona VA Health Care System, and University of Arizona Health Sciences Center, 3601 S 6th Avenue (1-111G-1), Tucson, AZ 85723, USA; Ronnie.Fass{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

We wish to comment on an interesting paper published previously in Gut which we inadvertently overlooked at the time. Fass et al (Gut 2001;48:31013) reported that there was a positive correlation between percentage time that oesophageal pH was less than 4 in 24 hours and the length of the columnar lined segment in 15 patients with long segment Barrett's oesophagus. Some years ago, we published data concerning 24 hour ambulatory oesophageal pH monitoring in untreated patients with Barrett's oesophagus and compared the results with those obtained in patients with reflux oesophagitis but no Barrett's oesophagus.1 pH monitoring was performed within one week of endoscopy. We found overlap in the 24 hour pH results between Barrett's patients and patients with reflux oesophagitis. However, among Barrett's patients, those with moderate to severe reflux oesophagitis above Barrett's segment had more acid reflux than those with mild reflux oesophagitis or none. At the time we did not correlate pH monitoring results with the length of the Barrett's segment but have now reviewed our pH data and have been able to correlate these results with the length of the columnar lined segment.

We studied 16 patients with long segment Barrett's oesophagus: seven males and nine females, aged 36–78 years (mean 59). Mean length of the Barrett's segment was 7 cm (range 4–16). Mean percentage time that oesophageal pH was <4 in 24 hours was 19.36%, with a very wide range (1.1–70%) but there was a correlation between length of the Barrett's segment and oesophageal acid exposure (r=0.66; confidence interval 0.2–0.8). Thus our older data support those of Fass et al as well as those of Sontag and colleagues2 and Oberg and colleagues3 in showing a correlation between oesophageal acid exposure and length of Barrett's oesophagus in long segment disease. We found a significant correlation between Barrett's length and supine reflux but unlike Fass et al, we were unable to show a significant correlation with upright reflux.

Many studies, including our own,4 have shown good symptomatic response to proton pump inhibitor (PPI) therapy in patients with Barrett's oesophagus but without significant regression of Barrett's epithelium, although approximately 50% of patients develop squamous islands within the Barrett's segment. Within each study to date, the same dose of PPI has been given to each patient. However, as oesophageal pH monitoring studies show, there is wide variation in acid reflux between patients. Effective control of acid reflux into the oesophagus may be important in preventing dysplasia5 and our study of patients treated with omeprazole for up to six years showed that none developed dysplasia during follow up.4 Therefore, PPI dose should be that which inhibits acid reflux effectively and will vary from patient to patient. Patients may resist frequent pH monitoring to determine the effective PPI dose so we would support the views of Fass et al that consideration should be given to treating patients with longer segments of Barrett's oesophagus with higher doses of PPI. Moreover, Barrett's patients with associated moderate to severe reflux oesophagitis should also be treated with higher PPI doses.


Author’s reply

I would like to thank Drs Neumann and Cooper for their comments on our article on the correlation of oesophageal acid exposure with Barrett’s oesophagus length (Gut 2001;48:31013). In recent years our laboratory has focused on factors that promote the development of Barrett’s oesophagus. Surprisingly, our understanding of the mechanisms that are responsible for the emergence of Barrett’s epithelium remains extremely poor. Despite the tendency in the literature to group Barrett’s patients as those with long and short segment Barrett’s oesophagus, we believe that the specific length of Barrett’s epithelium might be the key for unlocking the mystery of Barrett’s evolution. Consequently, we have initiated several research projects that were designed to assess the role of acid reflux in determining the specific length of Barrett’s oesophagus.

The results of the above mentioned study have been confirmed by other investigators.1,2 Oberg et al have also demonstrated that the extent of Barrett’s mucosa is inversely correlated with lower oesophageal sphincter pressure and length.2 Furthermore, we recently reported that the size of hiatal hernia correlated with the length of Barrett’s oesophagus (r=0.62, p=0.0012).3 The last two studies suggest that the longer the Barrett’s oesophagus the higher the likelihood of finding more severe oesophageal anatomical abnormalities that are strongly associated with increased oesophageal acid exposure.

In another study from our laboratory, Tharalson et al have demonstrated a significant relationship between the rate of change in acid exposure along the oesophagus and the length of Barrett’s oesophagus, using a pH probe with four sensors located 5 cm apart.4 The study investigated the rate at which recorded acid exposure values increase from the proximal to the distal oesophagus. This was the first study to demonstrate a statistically significant relationship (for the per cent total and upright time of pH testing) in which the length of Barrett’s oesophagus increases as the rate of acid exposure increases.

Presently we are not clear if acid reflux is the sole determining factor for Barrett’s appearance. It is likely that other factors, such as bile reflux, might have a synergistic effect. However, the role of bile reflux in determining the length of Barrett’s oesophagus remains to be elucidated.

We agree with Drs Neumann and Cooper that due to the close relationship between length of Barrett’s mucosa and oesophageal acid exposure, patients with longer Barrett’s epithelium may require higher doses of proton pump inhibitors to normalise their oesophageal acid exposure. One should be prepared to increase the dose of proton pump inhibitors in patients with longer segments of Barrett’s oesophagus if normalisation of oesophageal acid exposure is desired (not only symptom control).