Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

  Browse by Author
  Browse by Movies
Volume: Barrett's Esophagus
Chapter: Screening and surveillance

Is there evidence that Barrett's mucosa regresses spontaneously?

N.A. Shepherd (Gloucester)

Adenocarcinoma complicating Barrett's esophagus (columnar-lined esophagus, CLE) is rapidly becoming a major health problem for western communities. Its almost exponential rise in incidence in the last 20 years has ensured that cancer of the esophagus is now the cancer with the fifth highest mortality in UK males. Therefore there is currently much interest in regression of CLE, either spontaneously or in response to the various therapeutic modalities, in an attempt to halt the disquieting increase in the incidence of esophageal adenocarcinoma in the Western World.

Definitions: regression, reversal or what?

The term regression, when applied to CLE, has undoubtedly become tainted with confusion and misunderstanding. The Oxford Dictionary defines regression as "the action of returning to or toward a place or point of departure". Thus regression does not, necessarily, imply that the CLE has disappeared or been ablated entirely. Nevertheless many have, seemingly, used the term in this way and this has led to confusion. Perhaps it is best defined as "the total loss of columnar mucosa from an area previously occupied by columnar mucosa". However, pathological studies have demonstrated that various therapeutic modalities, including acid suppressive therapy afforded by proton pump inhibitors (PPIs) [1, 2], laser therapy [3, 4], photodynamic therapy (PDT) [5] and various electrocoagulation therapies [6-8] can induce squamous re-epithelialisation on the surface whilst CLE-type epithelium remains below [9, 10]. Some have designated such changes as pseudo-regression (Appelman, personal communication). It is certainly the case that the demonstration of such pseudo-regression requires the histopathologist's interpretation of biopsies as the endoscopist is unable to determine whether such CLE remains below [10].

Regression can be subclassified as complete, partial or focal. The latter two patterns have been described following all the therapeutic manoeuvres previously mentioned. Whilst complete regression is the ultimate goal, evidence of partial regression may also be important for reducing the cancer risk [11, 12]. Some have preferred the term "reversal" rather than regression for the phenomenon of squamous re-epithelialization [6, 13]. This is because of a belief that the term "regression" is both ambiguous and has been tainted with misunderstanding when applied to CLE. Reversal still does not imply that the CLE has entirely been replaced by squamous mucosa or destroyed completely; such remaining CLE epithelium may still have important implications for neoplastic potential [10].

How do we assess regression/reversal?

Regression/reversal can be defined by both endoscopic and histopathological means. Simple endoscopic imaging and measurement of segment length and surface area provide some evidence for regression [2, 14]. Measuring the length of Barrett's mucosa will indicate the amount of squamous over-riding that has occurred at the squamocolumnar junction [14] whilst measurement of the surface area will assess not only the squamous overriding, but also the extent of squamous islands within the Barrett's mucosa [2]. However quantitative endoscopic imaging using computerized image analysis is probably the most accurate means [15]. This will more accurately assess both the length of the segment and the presence and extent of squamous islands [15].

Pathologically, regression (or pseudo-regression) is characterized by the squamous overriding at the squamocolumnar junction and by the presence of macroscopic and microscopic squamous islands [9]. There is good evidence to suggest that these squamous islands arise from native structures within the esophagus, the esophageal gland ducts, and this relationship is often demonstrated in the microscopic squamous islands [9, 16]. Such an association may not be demonstrated in larger macroscopically (and endoscopically) visible squamous islands but it is presumed that such larger squamous islands evolve from these microscopic squamous islands, themselves evolving from metaplasia from native esophageal gland ducts [9].

Evidence for regression: spontaneous and therapeutic

In answer to the question "Is there evidence of spontaneous regression of CLE?", the answer appears to be critically dependent on the CLE type. There is some evidence that short segment CLE (SSCLE) can regress with the total disappearance of all tongues and patches of Barrett's mucosa in conjunction with only squamous mucosa on biopsy [17]. It would appear that the most important determinant of regression is the absence of an associated hiatus hernia rather than concurrent therapy: patients with SSCLE showing complete regression may not be on PPIs [17]. For long segment/traditional/classical CLE, there is no compelling evidence for spontaneous complete regression. Most cases of complete regression of CLE have occurred after therapy particularly aimed at achieving such complete regression, most notably laser and PDT [3-5]. Spontaneous complete regression of long segment CLE has not, to this author's knowledge, been demonstrated in a scientifically rigorous manner. The situation is very much complicated by the fact that many patients, at the time of diagnosis of CLE, are already on acid suppressive therapy because of symptomatology [18]. Any evidence for regression is more likely to be a response to PPI therapy and it is becoming increasingly difficult to demonstrate any evidence of "spontaneous regression".

Microscopic squamous islands have been described in untreated CLE [19]. Could these represent foci of regression? Given that the number, size and extent of squamous islands increases in patients on acid-lowering therapy [1, 14], it could be that these microscopic squamous islands, in untreated CLE patients, may represent areas of reduced refluxate exposure and thus microscopic areas of "regression". There is no compelling evidence for this: it is just as likely that these microscopic squamous islands represent remaining, and undamaged, foci of pre-existing squamous mucosa.

There remains controversy concerning the extent of regression that occurs after PPI therapy: some have shown partial regression with reduction in CLE length [1, 14, 20] and an increase in squamous islands and thus decrease in surface area [1, 14, 21, 22]. One double blind controlled trial reported has also demonstrated regression in both of these parameters [2]. However, other studies have not been able to show evidence of regression in PPI-treated patients [23, 24].

There are also controversies as to whether surgical techniques induce any regression of CLE: most workers believe they do not result in significant regression [11, 25]. However partial regression has been described, especially in children [26, 27]. One conundrum here is the fact that CLE mucosa itself produces acid from the fundic-type mucosa: reducing acid reflux from the stomach may not induce regression if the CLE mucosa itself is maintaining an acidic environment. Many studies of the effects of reflux surgery on CLE have predated the use of powerful acid suppressive drugs. The combination of reflux surgery and PPI therapy could theoretically induce squamous re-epithelialization and partial regression of CLE in a similar way, but perhaps more prominently, than PPI therapy alone. There is as yet no direct evidence that such regression occurs in many patients.

It is not intended to dwell on the various ablative methodologies that have been used to induce "regression" of CLE, as these are described in much more detail elsewhere. Success has been demonstrated in ablation of CLE and replacement by squamous mucosa using lasers [3, 4], PDT [5], and various electrocoagulation methods [6-8]. All of these may be associated with the phenomenon of covert CLE epithelium beneath surface squamous mucosa [9].

Heresy: do we really need complete regression/reversal of columnar-lined esophagus?

The critical question, in terms of reducing the cancer risk, is not whether complete or partial regression is consistently achievable, but what promotes the progression of the dysplasiacarcinoma sequence in CLE. If it is acid, then it may be controlled by acid suppressive therapy. This, of course, assumes that high acid levels in the esophagus are the only important factor in inducing neoplastic change. There is, however, some evidence to implicate duodeno-gastroesophageal reflux (DGER), particularly bile in this sequence [28, 29]. Thus acid-lowering therapy may not be the whole answer: however, there is also evidence to suggest that reduction of acid in the stomach reduces DGER and thus reduces the amount of bile exposure in CLE [30].

Is there current evidence to suggest that acid-lowering drugs, such as PPIs, not only lead to partial regression of CLE [1, 2], but also cause epithelial stability of the remaining CLE mucosa [12]? There is as yet little. We have demonstrated a reduction in the amount of unstable sulphomucin-positive mucopolysaccharide after 5 years of PPI therapy [1] whilst there is also a reduction in cyclin-D1, essentially a proliferation marker, in patients with CLE treated with PPIs [31]. Furthermore there is evidence that effective intraesophageal acid suppression leads to a reduction in CLE proliferation and to differentiation [32]. These data would suggest that effective control of acid in the esophagus may indeed lead to epithelial stability which itself could reduce the neoplastic potential. Thus insistence that the only means of ensuring a reduction in neoplastic potential is the production of complete regression/reversal, as has been implied [33], may not necessarily be essential [12]. An interesting counterbalance of these studies of the remaining CLE mucosa is a recent study demonstrating continued biomarker abnormality in the squamous epithelium in squamous islands in treated CLE [13]. However, it could be that these changes merely reflect the inevitable cellular proliferative activity in this regenerative mucosa.

Covert columnar-lined esophagus mucosa: does it matter?

Given that many therapeutic modalities can induce squamous re-epithelialization both at the squamocolumnar junction and as a result of induction of squamous islands which themselves enlarge and coalesce, what are the implications for any remaining CLE mucosa that remains below [10]? Some have suggested that this retains neoplastic potential and there are cases of neoplasia, both dysplasia and carcinoma, in this concealed mucosa beneath re-epithelialized squamous mucosa after therapeutic ablation of CLE [33-35]. These cases could represent de novo neoplastic change in such clandestine Barrett's mucosa; equally they could represent dysplastic or carcinomatous mucosa that was preexisting before therapy and has simply become buried by squamous re-epithelialization. This author believes that covert but non-neoplastic CLE mucosa, beneath squamous mucosa, may not have the same neoplastic potential as surface CLE mucosa as it is not being exposed to the noxious intraluminal agents, be they acid, bile or other agent [12].


Complete regression of CLE currently remains our goal in attempts to reduce the neoplastic potential of CLE. However, it may be that indicators of partial regression, such as reduction in length and squamous area, squamous re-epithelialization in the form of encroachment at the squamocolumnar junction and squamous islands, may merely be surrogate markers for a profound stabilisation of the mucosa of the CLE when not exposed to the noxious agents, particularly acid and bile, that have been implicated in the neoplastic sequence of CLE. We require further studies of the CLE mucosa, most notably molecular studies of cell cycle and apoptosis genes and proteins and the genetic abnormalities implicated in the neoplastic sequence, to determine whether there is such epithelial stability.

Complete regression of CLE may not be the Holy Grail for the management of this enigmatic disease. We live quite happily with metaplastic epithelia, particularly that with a mature phenotype, at many sites in the body and it is often further injurious agents that induce neoplastic change in such metaplastic epithelium. Could it not be that CLE might survive and yet mature under appropriate conditions and may not necessarily have the neoplastic potential of untreated CLE?


1. Wilkinson SP, Biddlestone L, Gore S, Shepherd NA. Regression of columnar-lined (Barrett's) oesophagus with omeprazole 40 mg daily: results of 5 years of continuous therapy. Aliment Pharmacol Ther 1999;13:1205-1209.

2. Peters FT, Ganesh S, Kuipers EJ, Sluiter WJ, Klinkenberg-Knol EC, Lamers, et al. Endoscopic regression of Barrett's oesophagus during omeprazole treatment;a randomized double blind study. Gut 1999;45:489-494.

3. Sampliner RE, Hixson LJ, Fennerty MB, Garewal HS. Regression of Barrett's esophagus by laser ablation in an anacid environment. Dig Dis Sci 1993;38:365-368.

4. Barham CP, Jones RL, Biddlestone LR, Hardwick RH, Shepherd NA, Barr H. Photothermal laser ablation of Barrett's oesophagus: endoscopic and histological evidence of squamous re-epithelialisation. Gut 1997;41:281-284.

5. Overholt BF, Panjehpour M. Photodynamic therapy in Barrett's esophagus. J Clin Laser Med Surg 1996;14:245-249.

6. Sampliner RE, Fennerty B, Garewal HS. Reversal of Barrett's esophagus with acid suppression and multipolar electrocoagulation: preliminary results. Gastrointest Endosc. 1996;44:532-535.

7. Dumoulin FL, Terjung B, Neubrand M, Scheurlen C, Fischer HP, Sauerbruch, et al. Treatment of Barrett's esophagus by endoscopic argon plasma coagulation. Endoscopy 1997;29:751-753.

8. Grade AJ, Shah IA, Medlin SM, Ramirez FC. The efficacy and safety of argon plasma coagulation therapy in Barrett's esophagus. Gastrointest Endosc 1999;50:18-22.

9. Biddlestone LR, Barham CP, Wilkinson SP, Barr H, Shepherd NA. The histopathology of treated Barrett's esophagus: squamous reepithelialization after acid suppression and laser and photodynamic therapy. Am J Surg Pathol 1998;22:239245.

10. Sharma P, Morales TG, Bhattacharyya A, Garewal HS, Sampliner RE. Squamous islands in Barrett's esophagus: what lies underneath? Am J Gastroenterol 1998;93:332-335.

11. Attwood SE. Barrett's esophagus. Curr Opin Gen Surg 1993;180-188.

12. Shepherd NA. Barrett's oesophagus and proton pump inhibitors: a pathological perspective. Gut 2000;46:147-149.

13. Garewal H, Ramsey L, Sharma P, Kraus K, Sampliner R, Fass R. Biomarker studies in reversed Barrett's esophagus. Am J Gastroenterol 1999;94:2829-2833.

14. Gore S, Healey CJ, Sutton R, Eyre-Brook IA, Gear MW, Shepherd NA, et al. Regression of columnar lined (Barrett's) oesophagus with continuous omeprazole therapy. Aliment Pharmacol Ther 1993;7:623-628.

15. Kim R, Rose S, Shar AO, Weiner M, Reynolds JC. Extent of Barrett's metaplasia: a prospective study of the serial change in area of Barrett's measured by quantitative endoscopic imaging. Gastrointest Endosc 1997;45:456-462.

16. Berenson MM, Johnson TD, Markowitz NR, Buchi KN, Samowitz WS. Restoration of squamous mucosa after ablation of Barrett's esophageal epithelium. Gastroenterology 1993;104:1686-1691.

17. Weston AP, Badr AS, Hassanein RS. Prospective multivariate analysis of factors predictive of complete regression of Barrett's esophagus. Am J Gastroenterol 1999;94:3420-3426.

18. Krueger KJ, DiPalma JA. Omeprazole and Barrett's regression: is asymptomatic good enough? Am J Gastroenterol 1997;92:177-178.

19. Sampliner RE, Steinbronn K, Garewal HS, Riddell RH. Squamous mucosa overlying columnar epithelium in Barrett's esophagus in the absence of antireflux surgery. Am J Gastroenterol 1988;83:510-512.

20. Malesci A, Savarino V, Zentilin P, Belicchi M, Mela GS, Lapertosa G, et al. Partial regression of Barrett's esophagus by long-term therapy with high-dose omeprazole. Gastrointest Endosc 1996;44:700-705.

21. Neumann CS, Iqbal TH, Cooper BT. Long term continuous omeprazole treatment of patients with Barrett's oesophagus. Aliment Pharmacol Ther 1995;9:451-454.

22. 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;12:893-897.

23. Bozymski EM, Shaheen NJ. Barrett's esophagus: acid suppression, but no regression. Am J Gastroenterol 1997;92:556558.

24. 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;92:582-585.

25. Spencer J. Surgery in Barrett's oesophagus. Eur J Cancer Prev 1992;1:331-335.

26. Hassall E, Weinstein WM. Partial regression of childhood Barrett's esophagus after fundoplication.Am J Gastroenterol 1992;87:1506-1512.

27. Sagar PM, Ackroyd R, Hosie KB, Patterson JE, Stoddard CJ, Kingsnorth AN. Regression and progression of Barrett's oesophagus after antireflux surgery. Br J Surg 1995;82:806-810.

28. Vaezi MF, Richter JE. Role of acid and duodenogastroesophageal reflux in gastroesophageal reflux disease. Gastroenterology 1996;111:1192-1199.

29. Stein HJ, Kauer WK, Feussner H, Siewert JR. Bile reflux in benign and malignant Barrett's esophagus: effect of medical acid suppression and Nissen fundoplication. J Gastrointest Surg 1998;2:333-341.

30. Marshall RE, Anggiansah A, Manifold DK, Owen WA, Owen WJ. Effect of omeprazole 20 mg twice daily on duodenogastric and gastro-oesophageal bile reflux in Barrett's oesophagus. Gut 1998;43:603-606.

31. Arber N, Lightdale C, Rotterdam H, Han KH, Sgambato A, Yap E, et al. Increased expression of the cyclin D1 gene in Barrett's esophagus. Cancer Epidemiol Biomark Prev 1996;5:457-459.

32. Ouatu LR, Fitzgerald RC, Triadafilopoulos G. Differentiation and proliferation in Barrett's esophagus and the effects of acid suppression. Gastroenterology 1999;117:327-335.

33. Sampliner RE, Fass R. Partial regression of Barrett's esophagus - an inadequate endpoint. Am J Gastroenterol 1993;88:2092-2094.

34. Barr H, Shepherd NA, Dix A, Roberts DJ, Tan WC, Krasner N. Eradication of high-grade dysplasia in columnar-lined (Barrett's) oesophagus by photodynamic therapy with endogenously generated protoporphyrin IX. Lancet 1996;348:584-585.

35. Van Laethem JL, Cremer M, Deviere J, et al. Intramucosal adenocarcinoma arising under squamous re-epithelialisation of Barrett's oesophagus. Gut 2000;46:574-577.

Publication date: August 2003 OESO©2015