SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

Introduction:

The treatment of patients with Barrett’s oesophagus is controversial. Debate exists regarding the use and value of high dose acid suppression as the standard of practice. Despite prolonged use of high dose proton pump inhibitors (40 mg omeprazole, 60 mg lansoprazole), most studies have shown no convincing evidence of significant regression of Barrett’s length. These studies, however, have used fixed doses of proton pump inhibitors and did not regularly document control of oesophageal acid exposure.

Aim:

To determine whether regression of Barrett’s epithelium can be achieved with documented maximal acid suppression.

Methods:

We have prospectively followed nine patients with Barrett’s oesophagus (eight male; mean age 60 years) for more than 1 year. They were all treated using medical therapy with pH monitoring documenting oesophageal acid exposure over 24 h < 1.6% of the time, and with two or more esophagogastroduodenoscopies performed by the same endoscopist.

Results:

Acid control was individually tailored and achieved with proton pump inhibitor b.d. (omeprazole 20 mg or lansoprazole 30 mg) and ranitidine at bedtime (HS) (Ran) if necessary. All nine patients (100%) showed some evidence of regression. All nine patients (100%) showed a decrease in Barrett’s length (mean 2 cm, range 1–3 cm). Six out of nine (66.67%) patients showed evidence of squamous islands on the last oesophagogastroduodenoscopy. The mean total distal oesophageal acid exposure was 0.38% (range: 0–1.5%). The mean follow-up of patients was 54 months (range: 13–118 months).

Conclusions:

Consistent and individually tailored maximal acid suppression documented by pH-metry is achievable and may result in decreased length and development of squamous islands in patients with Barrett’s epithelium. This approach should be further evaluated as potentially the preferred medical treatment for these patients.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

Barrett’s oesophagus is a pre-malignant condition secondary to long standing gastro-oesophageal reflux. It is thought that reflux damages the normal squamous epithelium, which is then replaced by a more acid resistant columnar epithelium. The replacement of the stratified squamous epithelium by a columnar epithelium with goblet cells is known to predispose the patient to the development of oesophageal adenocarcinoma, the most rapidly increasing cancer in the US.1 Currently accepted management of these patients includes: (i) frequent endoscopic surveillance to identify high-grade dysplasia and or adenocarcinoma at a curable stage; and (ii) medical or surgical therapy aimed at symptomatic control of gastro-oesophageal reflux.2

Complete oesophageal acid suppression obtained by aggressive medical therapy allows for more differentiated Barrett’s epithelia, while minimizing cell proliferation.3, 4 Theoretically, regression of Barrett’s epithelium might make progression to dysplasia or adenocarcinoma less likely, although this remains to be proven. Given the potential for a profound decrease in acid secretion with proton pump inhibitors, many investigators have felt that the use of high doses of these agents might be effective because H2-receptor antagonists have failed and only a few patients have shown regression following anti-reflux surgery.5[6][7]–8

The present study was designed to determine whether individually titrated, maximal oesophageal acid suppression in patients with Barrett’s oesophagus without carcinoma can result in a reduction of Barrett’s length.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

Nine patients with Barrett’s oesophagus, confirmed histologically to have intestinal metaplasia, constitute this report. They were prospectively recruited from 43 patients with Barrett’s oesophagus seen in our oesophageal referral clinic since January 1989. All patients (eight male, one female) were Caucasian with a mean age of 60 years (range 44–72 years). Patients were followed for a mean of 54 months (range 12–118 months). Patients were followed on maximal acid suppressive therapy for a mean of 16 months (range 1–52 months).

Criteria for entry included patients: (i) followed on medical therapy for at least 1 year; (ii) with a minimum of two surveillance endoscopies by the same endoscopist performed at our institution; and (iii) with total distal oesophageal acid exposure whilst on therapy of < 1.6% documented by ambulatory pH monitoring. Thirty-four patients were excluded; five had undergone only one endoscopy; six had undergone endoscopy without comment on the extent of Barrett’s length; nine had endoscopies performed by different endoscopists; five had no pH-metry documented on medication; seven opted for anti-reflux surgery; and two were followed for less than 1 year.

Treatment with medication was tailored in a step-up fashion to achieve maximal acid suppression based on 24 h pH-metry studies. The studies to detect titration were performed between 2 and 8 weeks following a change in the medication regimen.

Change in the length of Barrett’s oesophagus was determined by the endoscopist’s estimate of the proximal extent of the continuous Barrett’s oesophagus from the incisors; the endoscopist estimated the overall length of Barrett’s oesophagus in centimetres. Changes in the length of Barrett’s mucosa from initial diagnosis (pre-maximal acid suppression) to the final visit were recorded. All endoscopy reports were blinded to date and sequence and reviewed by an experienced endoscopist (POK) to determine the length of Barrett’s oesophagus. The length of Barrett’s oesophagus was obtained from actual typed reports.

Biopsy specimens of the Barrett’s epithelium were taken every 1-cm with four quadrant samples. All samples were sent to our pathology department where they went through a standard overnight fixation and dehydration sequence. Biopsies were embedded in paraffin and sectioned at 5 μm. All specimens were stained with haematoxylin and eosin or alcian blue at pH 2.7 and read by a trained gastrointestinal pathologist for the presence of specialized intestinal metaplasia and dysplasia.

All patients had 24-h ambulatory pH-metry performed whilst continuing medication to document distal oesophageal and gastric pH exposure. Studies were performed with a 2.1-mm diameter dual-electrode antimony pH catheter (Synectics Medical Inc., Irving, TX) placed transnasally into the stomach after an overnight fast. The proximal electrode was withdrawn to a position 5 cm proximal to the lower oesophageal sphincter, previously defined manometrically. The distal electrode was located 15 cm distal to the proximal electrode, thus 7–10 cm below the lower oesophageal sphincter in the stomach. The pH values from both intra-oesophageal and intragastric electrodes were recorded continuously at the rate of one sampling per 4 s on an ambulatory Mark III Digitrapper (Synectics Medical Inc.). Meal times, recumbent periods, and any symptoms were marked electronically on the digitrapper by each patient. In addition, each patient recorded these events along with the meal contents manually in a diary. The data from the digitrapper were downloaded into an IBM compatible computer for analysis. The pH reports were then reviewed for the percentage time that the intragastric and oesophageal pH was less than 4 in the upright position, recumbent position, and total. The normal percentage total time for which pH < 4 in the distal oesophagus was defined as < 1.6% based on previous studies performed in normal subjects on twice daily proton pump inhibitor.9

Dosage and changes in medications were documented. All treatment regiments consisted of proton pump inhibitor b.d. before meals (AC) (omeprazole 20 mg or lansoprazole 30 mg) with ranitidine at bedtime (HS) if necessary. Ranitidine HS was added if the patient continued to have abnormal overnight oesophageal acid exposure when studied via pH-metry.10, 11

Statistics

Statistical analysis was done by using GraphPad Prism software (version 3.0, GraphPad software, SanDiego, CA). Changes in Barrett’s epithelia were analysed by using the paired t-test to compare pre- and post-treatment change in the length of Barrett’s oesophagus.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

Nine patients met the criteria for entry into the study (Table 1). All patients were Helicobacter pylori-negative by histology. All patients showed evidence of a decrease in length of Barrett’s epithelium (Figure 1). The mean decrease in Barrett’s length from initial endoscopy (pre-maximal acid therapy) to the last endoscopy (maximally acid suppressed) was 2 cm, with a range of 1–3 cm. The length of Barrett’s decreased from an average of 7.22 cm (range 2–16 cm) to an average of 5.22 cm (range 1–14 cm). The majority of patients (six out of nine, 67%) showed evidence of squamous re-epithelialization (squamous islands). None of the patients presented with squamous islands on the first oesophagogastroduodenoscopy before maximal acid suppression.

Table 1.   Summary of results on nine patients with Barrett’s oesophagus receiving maximal acid secretion therapy Thumbnail image of
image

Figure 1.  The length of Barrett’s oesophagus in nine patients shown before and after documented maximal acid secretion therapy. Two patients regressed from 7 to 5 cm of Barrett’s oesophagus, two patients regressed from 8 to 5 cm of Barrett’s oesophagus.

Download figure to PowerPoint

All biopsies confirmed the presence of specialized intestinal metaplasia in the tubular oesophagus. Five patients had no dysplasia, three patients had low-grade dysplasia, and one patient had focal high-grade dysplasia. This patient had flat mucosa with no visible nodules and elected to be followed on continued surveillance. No patients had progression of dysplasia during their follow-up. Three patients reverted from low-grade dysplasia to no dysplasia.

The last pH study performed was considered in calculations. All patients had a documented reduction in oesophageal acid exposure to < 1.6% (maximal acid suppression) on medication. The mean total distal oesophageal acid exposure time for which pH < 4 was 0.38% with a range of 0–1.5%. The mean upright distal oesophageal acid exposure with pH < 4 was 0.58%, with a range of 0–2.4%. The mean recumbent distal oesophageal acid exposure with pH < 4 was 0.09%, with a range of (0–0.8). The mean total time for which intragastric pH < 4 was 24.2%, with a range of 2.2–72.9% (Table 1).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

In vitro experimental data have shown that elimination of acid exposure can reduce proliferation of Barrett’s epithelial cell culture.3 Unfortunately, many studies performed to date have failed to show clinical support for this by failing to document any consistent decrease in Barrett’s length when a fixed high dose proton pump inhibitor is used as treatment.12 Many of these studies, however, document the development of squamous islands in varying frequency.13, 14 Each of these studies have in one way or another failed to give or document sufficient antisecretory therapy to maximally control oesophageal acid exposure. On the other hand, following one preliminary report suggesting that a decrease in length of Barrett’s metaplasia might occur on proton pump inhibitor therapy, prospective studies with long-term treatment have suggested a possible therapeutic role for proton pump inhibitors in inducing regression of the columnar epithelium.16, 17 Thus, debate and controversy regarding possible regression of Barrett’s epithelium from medical therapy continues.

How should patients with Barrett’s oesophagus be managed long-term? Should they be treated as any patient with chronic GERD, using sufficient antisecretory medication to achieve symptom relief, or should they have aggressive acid suppressive therapy aimed at elimination of oesophageal acid exposure? The current standard of practice supports the former approach, citing a large body of medical and surgical data that fail to consistently document a decrease in Barrett’s length, surface area or elimination of cancer risk. Several important observations must be made in evaluating this approach. Recent literature documents that most patients with Barrett’s oesophagus will continue to have abnormal oesophageal acid exposure, despite complete symptom relief, when studied with prolonged pH monitoring.18 Reflux most often occurs in the overnight period when oesophageal clearance is delayed and the potential for damage is greatest. Because patients with Barrett’s have a higher frequency of reflux-related oesophageal motility abnormalities (principally decreased lower oesophageal sphincter pressure and ineffective oesophageal motility), they are at greater risk for nocturnal reflux even on proton pump inhibitor therapy. Recent studies have found that abnormal oesophageal acid exposure continues in these patients despite lansoprazole doses as high as 60 mg per day. This same study emphasizes the marked variability in proton pump inhibitor response because the authors found no correlation between the dose of lansoprazole required for complete symptom relief and the amount of 24 h oesophageal acid exposure.14 Simply stated, a fixed dose of proton pump inhibitor as high as 40 mg omeprazole or 60 mg lansoprazole a day does not assure adequate gastric acid suppression and absence of pulses of acid reflux. This marked inter-subject variability in pH control may explain the failure of previous studies to document consistently positive effects on Barrett’s length, because most have failed to document (or achieve) maximal acid suppression. Despite these limitations many studies have demonstrated the development of squamous islands, indicating that re-epithelialization with squamous tissue is possible.

What is the rational for maximal acid suppression when anti-reflux surgery (which reduces oesophageal acid exposure to normal when effective) has failed to consistently show regression. As with fixed dose proton pump inhibitors, surgical outcome varies, and does not always result in or document total absence of oesophageal acid exposure. Despite this, several surgical reports have suggested that an effective control of acid exposure results in a reduction of Barrett’s length.7, 8 More importantly, recent observations suggest that effective acid suppression will result in a reduction of proliferation of Barrett’s cell culture using PCNA expression as a marker of cell proliferation.3 The first of these studies by Fitzgerald and colleagues studied the effect of pulse acid, continuous acid and alkaline pH (no acid) on proliferation indices in incubated Barrett’s tissue culture. Proliferation was greatest with pulse acid compared to continuous acid, whilst alkaline infusion eliminated cell turnover.4 This same laboratory studied the effect of oesophageal acid suppression on villin expression and PCNA in patients treated with lansoprazole. Patients who normalized oesophageal acid exposure (total time pH < 4 less than 4%) had a decrease in PCNA expression compared to those who did not.3 Both these studies suggest that optimal acid reduction may be beneficial.

If this approach is to be followed, what is the optimal end-point for oesophageal acid suppression? Current studies suggest that a reduction in oesophageal acid exposure time to normal (total time pH < 4, less than 4.2%: the normal value for physiologic reflux in untreated normal subjects) is not optimal. We have demonstrated that when normal subjects are treated with omeprazole 20 mg b.d., total oesophageal acid exposure time is reduced to < 1.6% total.9 We currently use this as our standard for optimal oesophageal acid control.

Using this end-point, our study suggests that endoscopic regression can occur with individually tailored maximal acid suppression. We acknowledge that the small sample size and absence of controls limits definitive conclusions from our study. However, the endoscopies were performed as they are in all Barrett’s patients and the endoscopic reports were blinded and reread in that fashion. Our results are uniform; all our patients demonstrated some decrease in length with follow-up. We have clearly demonstrated that maximal control is achievable in some patients with proton pump inhibitor b.d., and that others will require the addition of an H2-receptor antagonist at bedtime. In addition, our observations support the recently published randomized, blinded, controlled trial of Peters et al., which documented a consistent reduction in Barrett’s length and surface area when patients were treated with omeprazole 40 mg b.d. for a mean follow-up of 24 months compared to ranitidine 150 mg b.d. Excellent oesophageal acid suppression was achieved with mean oesophageal acid exposure time of 0.1% (range 0–1.2%), similar to our control with omeprazole 20 mg or lansoprazole 30 mg b.d. plus ranitidine. This confirmed previous observations of poor oesophageal acid control and no regression in the control group treated with ranitidine 150 mg b.d.19

Two points are important to address. First, the reproducibility of the length of columnar epithelium has been questioned when endoscopy is repeated.20 In fact, these inconsistencies may make endoscopic length a less than optimal end-point for the assessment of therapy. However, until an alternative or better standard is established, we believe that our results are expressed with as little bias as possible. Second, the number of pH studies required to achieve maximal acid suppression varies from patient to patient, although usually less than three studies over an extended period are required. Two of our patients underwent nine pH studies; they had low grade dysplasia and in one case a personal request was made for multiple studies to assure continued maximal acid suppression. They are the exception, not the rule.

In summary we have demonstrated that maximal oesophageal acid suppression is feasible in patients with Barrett’s oesophagus using prolonged pH monitoring to titrate a reduction in oesophageal acid exposure time to less than 1.6%, and that this approach may result in a reduction of Barrett’s length. We believe that our results, and those of Peters et al., suggest this approach may be the preferred treatment of patients with Barrett’s oesophagus.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

This study was supported by the Donald O. Castell, MD, Gastroenterology Research and Education Trust.

Bibliography

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography
  • 1
    Castell DO & Richter JE. The Esophagus, 3rd edn. Philadelphia, PA: Lippincott Williams & Wilkins, 1999: 463–71.
  • 2
    Fennerty MB, Sampliner RE, Garewal HS. Barrett’s oesophagus—cancer risk, biology and therapeutic management. Aliment Pharmacol Ther 1993; 7: 33945.
  • 3
    Ouatu-Lascar R, Fitzgerald RC, Triadofilopoulos G. Differentiation and proliferation in Barrett’s esophagus and the effects of acid suppression. Gastroenterology 1999; 117: 32735.
  • 4
    Fitzgerald RC, Omary MB, Triadafilopoulos G. Dynamic effects of acid on Barrett’s esophagus. An ex vivo proliferation and differentiation model. J Clin Invest 1996; 98: 21208.
  • 5
    Kothari T, Mangla J, Kaira TMS. Barrett’s ulcer and treatment with cimetidine. Arch Int Med 1980; 240: 42578.
  • 6
    Cooper BT & Barbezat GO. Treatment of Barrett’s esophagus with H2 blockers. J Clin Gastroenterol 1987; 9: 2832.
  • 7
    Skinner D, Walther BC, Riddell RH, Schmidt H, Iascone C, DeMeester TR. Barrett’s esophagus: Comparison of benign and malignant cases. Ann Surg 1983; 198: 55466.
  • 8
    Williamson WA, Ellis FH Jr, Gibbs SP, Shuhian DM, Arotz HT. Effect of anti reflux operation on Barrett’s mucosa. Ann Thorac Surg 1990; 49: 53741.
  • 9
    Kuo B & Castell DO. Optimal dosing of omeprazole daily: effects on gastric and esophageal pH and serum gastrin in healthy controls. Am J Gastroenterol 1996; 91: 152731.
  • 10
    Peghini PL, Katz PO, Castell DO. Ranitidine controls nocturnal gastric acid breakthrough on omeprazole: a controlled study in normal subjects. Gastroenterology 1998; 115: 13359.
  • 11
    Katz PO, Anderson C, Khoury R, Castell DO. Gastro-esophageal reflux associated with nocturnal gastric acid breakthrough on proton pump inhibitors. Aliment Pharmacol Ther 1998; 12: 12314.
  • 12
    Sampliner RE. Effect of up to 3 years of high-dose lansoprazole on Barrett’s esophagus. Am J Gastroenterol 1994; 89: 18448.
  • 13
    Cooper BT, Neumann CS, Cox MA, Iqbal TH. Continuous treatment with omeprazole 20 mg daily for up to 6 years in Barrett’s oesophagus. Aliment Pharmacol Ther 1998; 12: 8937.DOI: 10.1046/j.1365-2036.1998.00389.x
  • 14
    Sharma P, Sampliner RE, Camargo E. Normalization of esophageal pH with high dose proton pump inhibitor therapy does not result in regression of Barrett’s esophagus. Am J Gastroenterol 1997; 92: 5825.
  • 15
    Deviere J, Buset M, Dumocean JM, Richaert F, Cremer M. Regression of Barrett’s epithelium with omeprazole. N Engl J Med 1989; 320: 14978.
  • 16
    Gore S, Healey CJ, Sutton R, et al. Regression of columnar lined (Barrett’s) oesophagus with continuous omeprazole therapy. Aliment Pharmacol Ther 1993; 7: 6238.
  • 17
    Malesci A, Savarino V, Zentilin P, et al. Partial regression of Barrett’s esophagus by long-term therapy with high-dose omeprazole. Gastrointest Endosc 1996; 44: 7005.
  • 18
    Katzka DA & Castell DO. Successful elimination of reflux symptoms does not insure adequate control of acid reflux in patients with Barrett’s esophagus. Am J Gastroenterol 1994; 89: 98991.
  • 19
    Peters FTM, Ganesh S, Kuipers EJ, et al. Endoscopic regression of Barrett’s oesophagus during omeprazole treatment; a randomised double blind study. Gut 1999; 45: 48994.
  • 20
    Kim SL, Waring JP, Spechler SJ, et al. Diagnostic inconsistencies in Barrett’s esophagus. Gastroenterology 1994; 107: 9459.