- Email: [email protected]
Does a surgical antireflux procedure decrease the incidence of esophageal adenocarcinoma in Barrett's esophagus? A meta-analysis
THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2003 by Am. Coll. of Gastroenterology Published by Elsevier Inc.
Vol. 98, No. 11, 2003 ISSN 0002-9270/03/$30.00 doi:10.1016/j.amjgastroenterol.2003.08.008
Does a Surgical Antireflux Procedure Decrease the Incidence of Esophageal Adenocarcinoma in Barrett’s Esophagus? A Meta-Analysis Kathleen E. Corey, M.D., M.P.H., Sarah M. Schmitz, B.A., and Nicholas J. Shaheen, M.D., M.P.H. Center for Esophageal Diseases and Swallowing, the Division of Digestive Diseases, and the School of Public Health, University of North Carolina at Chapel Hill
OBJECTIVE: The risk of adenocarcinoma of the esophagus is increased among those with Barrett’s esophagus (BE). Whether the risk of cancer in the setting of BE can be decreased by a surgical antireflux procedure (SARP) is unclear. This study compared the reported incidence of esophageal adenocarcinoma in subjects with BE who underwent SARP with those with BE who had medical management. METHODS: We used MEDLINE to perform a meta-analysis of the English language literature published from 1966 through October 2001. We reviewed abstracts found with the search term “Barrett’s esophagus” and the following: “adenocarcinoma,” “esophageal neoplasm,” “proton pump inhibitor,” “fundoplication,” or “antireflux procedure.” Study entry criteria included 1) trial or cohort study with a report of cancer risk expressible in cancers per patient-year, 2) histologic confirmation of BE and any adenocarcinomas, and 3) adequate description of intervention (medical vs SARP). Data were abstracted by two reviewers using standardized forms. Subgroup comparisons were made using only medical management studies published in the last 5 yr. Multivariable regression controlling for subject age, country of origin, and BE length was performed. RESULTS: We reviewed 1247 abstracts, and 34 met the inclusion criteria. There were a cumulative 4678 patient-years of follow-up in the SARP group and 4906 patient-years in the medical group. The cancer incidence rate in the SARP group was 3.8 cancers/1000 patient-years, compared with 5.3 in the medical group (p ⫽ 0.29). Similarly, there was no significant difference between cancer rates when comparing SARP with medical series reported in the last 5 yr (3.8/1000 patient-years vs 4.2/1000 patient-years, p ⫽ 0.33). Multivariate analysis controlling for subject age, country of origin, and BE length did not alter these findings. CONCLUSION: The reported risk of adenocarcinoma in subjects with BE is low and not significantly decreased by a surgical antireflux procedure. Antireflux surgery in the setting of BE should not be recommended as an antineoplastic measure. (Am J Gastroenterol 2003;98:2390 –2394. © 2003 by Am. Coll. of Gastroenterology)
INTRODUCTION Barrett’s esophagus (BE) is a metaplastic condition resulting from chronic gastroesophageal reflux injury in which the squamous epithelium of the esophagus is replaced by intestinal epithelium (1, 2). With time, patients might develop dysplasia and, eventually, esophageal adenocarcinoma. Thus, BE has been associated with an increased risk of the development of esophageal adenocarcinoma. Estimates have placed the incidence of esophageal adenocarcinoma in patients with BE at approximately 500 cases per 100,000, whereas incidence in the general population is considerably less (3– 6). A randomized, controlled trial of surgical and medical interventions for gastroesophageal reflux disease (GERD) by Spechler et al. (7) reconfirmed the increased risk of esophageal adenocarcinoma in BE patients, with a risk of 0.4% per patient-year compared with 0.07% per patient-year in patients without BE (7). Because of the increased incidence of carcinoma in the setting of BE, longterm endoscopic follow-up and surveillance has been suggested (8). The development of BE has been linked to chronic GERD (9). However, the effect of treatment of chronic reflux disease on the progression of BE remains uncertain. Currently, chronic reflux disease is managed either medically by the long-term use of H2 receptor antagonists or proton pump inhibitors, or with surgical antireflux procedures (SARP). In both treatment types, progression to adenocarcinoma has been demonstrated (10, 11). Some authors have suggested that surgical approaches to reflux might be superior to medical management in arresting the progression of BE to cancer (12). However, neither medical nor surgical intervention has been consistently shown in the literature to preferentially decrease the incidence of adenocarcinoma. Thus, no consensus exists as to which intervention is preferred to decrease the progression to malignancy in the setting of BE. This meta-analysis compared the reported rates of adenocarcinoma of the esophagus in the English language literature in those BE patients undergoing SARP with the rates of those patients managed medically. Additionally, we
AJG – November, 2003
sought to identify any study variables that might influence cancer risk.
MATERIALS AND METHODS
Surgery and Barrett’s Cancer
2391
origin, BE length, and BE definition was performed with the SAS system (SAS Institute, Cary, NC). A value was generated to assess the agreement between the abstracted data of the two reviewers.
Study Selection and Data Abstraction We performed a search of the MEDLINE database for the period 1966 through October 2001. All English-language abstracts with “Barrett’s esophagus” as a text word or subject heading and one of the following were reviewed: “adenocarcinoma,” “esophageal neoplasm,” “proton pump inhibitor,” “fundoplication,” and “antireflux procedure.” Bibliographies were reviewed for additional eligible studies. The following criteria for inclusion into the meta-analysis were used: 1) randomized controlled trial or cohort study with a reported esophageal adenocarcinoma risk expressible in cancers per patient-year, 2) histologic confirmation of BE at cohort inception and histologic confirmation of any subsequent adenocarcinoma, 3) ability to determine the patients’ treatment regimen (SARP vs medical therapy), 4) ability to determine the patients’ cancer outcomes and allcause mortality, and 5) A follow-up period of 12 months or more. Two independent investigators (K.C., S.S.) abstracted the included studies. Variables assessed included year of publication, country of origin, study design, definition of BE, endoscopic surveillance interval, total cancers developed, cancers developed in the first year after start of intervention, and total patient-years. For serial studies reporting the outcomes in the same cohort of patients, the more recent publication was used. Differences between the reviewers in classification were resolved by consensus. The currently accepted definition of BE is any length of endoscopically evident salmon-colored tissue in the tubular esophagus which, on histologic evaluation, is of the “specialized” or “intestinalized” type. However, because past investigators have sometimes used slightly different definitions, we also assessed the reports for the definition of BE used. The reports were classified with respect to BE definition in the following categories: 1) metaplasia not otherwise specified (NOS), 2) metaplasia NOS greater than 3 cm, 3) intestinal metaplasia, 4) intestinal metaplasia greater than 3 cm, and 5) no definition specified.
RESULTS
Data Analysis The crude cancer incidence was obtained by dividing the total number of cancers observed by the total number of patient-years in the study. Patient-years were determined by assessing the number of years from study entry until the development of adenocarcinoma or the end of the study period. If individual patient data were unavailable with which to calculate the total number of patient-years, the mean number of years of patient follow-up was multiplied by the number of subjects in the study to obtain the total patient-years encompassed by the study. Multivariate regression controlling for the mean subject age, country of
DISCUSSION
A total of 1247 abstracts were reviewed. Of these, 49 articles were retrieved, of which 16 publications met inclusion criteria. The remaining 33 articles lacked either adequate description of the intervention, used a follow-up period of less than 12 months, or were serial reports of the same cohort. Review of the bibliographies of the initial 49 retrieved articles led to an additional 18 articles for inclusion. Thus, a total of 34 articles were included for analysis (Table 1). Agreement between the reviewers was high ( ⫽ 0.906). Univariate analysis revealed 4678.3 patient-years among the patients receiving SARP. Medical patient-years totaled 4906.2. Owing to the advances in medical therapy in recent years with the addition of proton pump inhibitors to GERD and BE treatment, medical therapy was analyzed in all studies and in the subgroup of studies conducted only in the most recent 5 yr analyzed (1996 –2001). There were 4205.5 patient-years of medical patient follow-up in studies conducted in those 5 yr. In the surgical intervention group, there were 18 cancers over 4678.3 patient-years, or 3.8 cancers per 1000 patient-years. In the medical group, there were 26 cancers over 4906.2 patient-years, or 5.3 cancers per 1000 patient-years (p ⫽ 0.29, when compared with the SARP group). In medical patients in studies limited to the past 5 yr, there were 18 cancers over 4205.5 patient-years, or 4.3 cancers per 1000 patient-years (p ⫽ 0.33), when compared with the SARP group (Table 2). There was no significant difference in cancers per patient-year between the surgical and medical groups or between the surgical group and the medical therapy group in the last 5 yr. Multivariable analysis controlling for country of origin, age, length of Barrett’s segment, and definition of BE was performed. After controlling for these potentially confounding covariates, no significant differences were noted in the cancer risk between the medical and surgical groups (p ⫽ 0.39, full models not shown).
This analysis compared the published rates of esophageal adenocarcinoma in those with BE treated with surgical antireflux procedures with the rates of those treated medically. No significant difference was seen in the incidence of adenocarcinoma between these groups of patients. These results conflict with some surgical series, which suggested that SARP might be superior to medical management of BE in the prevention of adenocarcinoma. Katz et al.(13) studied 102 patients to determine risk factors for dysplasia in BE patients and found that whereas 23 of the medically treated
2392
Corey et al.
AJG – Vol. 98, No. 11, 2003
Table 1. Studies Reporting Cancer risk in BE Patients With a Specified Medical and/or Surgical Intervention
First Author (reference)
Source and Year
Attwood (19) Brand (20) Cooper (21) Cooper (22) Csendes (23) DeMeester (24) Drewitz (5) Katz (13) Klinkenberg-Knol (25) Klinkenberg-Knol (26) Chen (27) Low (28) Luostrarinien (29) Lundell (30) Malesci (31) McDonald (32) Ortiz (15) Ovaska (33) Patti (34) Peters (35) Sagar (36) Sampliner (37) Sampliner (38) Sharma (39) Sharma (40) Sharma (41) Skinner (42) Spechler (7) Srinivasan (43) Starnes (44) Wesdorp (45) Weston (46) Wilkinson (47) Williamson (10) Total
No. of Patients; Surgical
Total PatientYears; Surgical
Total Cancers; Surgical
No. of Patients; Medical
Total PatientYears; Medical
Total Cancers; Medical
19 10 0 0 152 35 0 16 0 0 45 14 5 22 0 113 32 12 38 0 56 0 9 0 0 0 13 129 0 8 0 0 0 26 754
57 45 0 0 1266 105 0 76.8 0 0 180 29.4 33.3 66 0 734.5 160 80.4 72.96 0 392 0 27 0 0 0 52 1174 0 17.6 0 0 0 109.2 4678.3
1 1 0 0 4 0 0 0 0 0 0 0 0 0 0 3 1 2 0 0 1 0 0 0 0 0 0 1 0 1 0 0 0 3 18
26 0 47 7 0 6 170 0 84 32 0 0 0 16 14 0 27 10 8 53 0 27 58 78 32 13 0 137 9 12 9 108 12 0 918
78 0 182 9.6 0 18 807.5 0 546 128 0 0 0 48 168 0 135 67 15.36 106 0 78.3 174 220.7 98.4 74.1 0 1452 40.5 26.4 17 359.1 57 0 4906.2
1 0 0 0 0 0 9 0 1 0 0 0 0 0 0 0 1 1 0 1 0 0 0 1 1 0 0 4 0 1 0 5 0 0 26
Brit J Surg 1992 New Engl J Med 1980 Aliment Pharmacol Ther 1998 Q J Med 1987 Surgery 1998 Ann Surg 1990 Am J Gastroenterol 1997 Am J Gastroenterol 1998 Gastroenterology 2000 Ann Intern Med 1994 Ann Surg 2001 Am J Gastroenterol 1999 Ann Med 1998 Eur J Gastroenterol Hepatol 2000 Gastrointest Endosc 1996 J Thorac Cardiovasc Surg 1996 Br J Surg 1996 Dig Dis Sci 1989 J Gastrointest Surg 1999 Gut 1999 Br J Surg 1995 Am J Gastroenterol 1994 Dig Dis Sci 1990 Gut 2000 Am J Gastroenterol 1997 Am J Gastroenterol 1997 Ann Surg 1983 JAMA 2001 Aliment Pharmacol Ther 2001 Arch Surg 1984 Gut 1981 Am J Gastroenterol 1999 Aliment Pharm Ther 1999 Ann Thorac Surg 1990
patients developed dysplasia and four developed adenocarcinoma, none of the patients who underwent antireflux procedures developed dysplasia of any type. Also, McCallum et al. (14) demonstrated that 3.4% of BE patients who underwent SARP developed dysplasia, compared with 19.7% of those treated medically. In addition, no patient in the SARP group developed adenocarcinoma, whereas 1.3% in the medical group did. Finally, Ortiz et al. (15) performed a randomized, controlled trial of 59 BE patients assigned to Table 2. Incidence Rate of Esophageal Adenocarcinoma in BE Patients Receiving SARP and Medical Therapy
Incidence rate (Cancer/1000 Patient-Years)
SARP
Medical Therapy
Medical Therapy in Last Five Years
3.8 (2.4–6.1)
5.3 (3.6–7.8)
4.3 (2.6–5.8)
Numbers in parentheses are 95% CIs.
receive either SARP or medical therapy with H2 antagonists. High-grade dysplasia developed in only two patients, one of whom underwent medical therapy and one who had a failed SARP. Although the incidence of cancer was not significantly different between the surgical and medical arms of the Ortiz study, significantly fewer SARP patients than medically managed patients went on to develop any form of dysplasia. In addition, fewer surgical patients experienced upward BE progression relative to those who underwent medical therapy. Although these positive studies seem to promote the value of SARP in BE, all are small studies, adding relatively few patient-years to the literature. When these studies are added to the cumulative reported surgical experience, they are insufficient to cause a significant difference between the medical and surgical arms of the meta-analysis. Previous investigators have attempted to compare the cancer rates in medically and surgically treated subjects with BE. A pooled analysis undertaken by Bammer et al. (12) found that surgery was superior to medical therapy in
AJG – November, 2003
decreasing the incidence of esophageal adenocarcinoma. That study differs from the present meta-analysis in several respects. Our study excluded any report that did not have 1 yr of follow-up after treatment, an exclusion that the study by Bammer et al. did not undertake. In addition, the study by Bammer et al. did not include 14 studies, which our searches found and included. This resulted in considerably more medical patient-years in the present study. These differences between the study by Bammer et al. and the present meta-analysis might have resulted in the previous study overestimating the benefit of SARP in BE patients. Several potential sources of bias exist within this study. First, this meta-analysis includes cohort studies in which the patients were not randomly assigned to an intervention. Thus, the degree of GERD severity might not have been equally distributed between SARP and medical therapy groups. The potential exists that patients with more severe GERD underwent antireflux procedures, whereas those with less severe disease underwent medical therapy. If disease severity correlates with cancer risk, this would have the effect of biasing the results of the relationship against surgery. A second source of bias might be derived from the use of a variety of SARP in the surgically treated patients. Procedures were not uniform across the studies, nor did all studies specify the procedure used, making analysis controlling for the type of SARP performed impossible. Also, most studies did not assess the competency of the wrap in controlling reflux. Certainly, if cancer is promoted by ongoing reflux, the cancer risk in those with ineffective wraps might be expected to be higher than in those with good acid control. Preliminary evidence suggests that ineffective wraps might indeed be associated with an increased risk of the subsequent development of dysplasia (16). Next, most studies of medical therapy made no attempt to assess patient compliance with medications. If subjects in the medical arm were receiving less than optimal doses or were taking these medications with poor compliance, cancer rates might similarly be elevated. Also, information regarding other confounders, such as alcohol use and tobacco use, were not available from most studies and could not be controlled for in this study. Although tobacco and alcohol use have been associated with the development of adenocarcinoma (17), they are more strongly associated with the development of squamous cell carcinoma (18). Finally, although our study encompasses the entire reported English literature, only a small number of subjects in the medically and surgically managed arms developed cancer. These small numbers of cancers increase the likelihood of a type 2 statistical error. Despite the small number of cancer cases, a large number of patient-years were available, strengthening the precision of these estimates. In conclusion, the incidence of adenocarcinoma development in BE patients is low, approximating 5 per 1000 patient-years. Based on the cumulative published literature, surgical antireflux procedures do not seem to provide benefit over medical therapy in BE patients in reducing the inci-
Surgery and Barrett’s Cancer
2393
dence of cancer development. Surgical antireflux procedures should not be recommended over medical therapy to decrease the cancer risk associated with BE.
ACKNOWLEDGMENTS This work was supported in part by National Institutes of Health grant K23DK59311-01. Reprint requests and correspondence: Nicholas Shaheen, M.D., M.P.H., University of North Carolina Hospitals, Division of Digestive Disease and Nutrition, CB #7080, 724 Burnett-Womack Building, Chapel Hill, NC 27599-7080. Received Dec. 13, 2002; accepted Apr. 8, 2003.
REFERENCES 1. Spechler SJ. Clinical practice. Barrett’s esophagus. N Engl J Med 2002;346:836 –42. 2. Shaheen N, Ransohoff DF. Gastroesophageal reflux, Barrett esophagus, and esophageal cancer: Scientific review. JAMA 2002;287:1972–81. 3. Shaheen NJ, Crosby MA, Bozymski EM, Sandler RS. Is there publication bias in the reporting of cancer risk in Barrett’s esophagus? Gastroenterology 2000;119:333–8. 4. Spechler SJ, Robbins AH, Rubins HB, et al. Adenocarcinoma and Barrett’s esophagus. An overrated risk? Gastroenterology 1984;87:927–33. 5. Drewitz DJ, Sampliner RE, Garewal HS. The incidence of adenocarcinoma in Barrett’s esophagus: A prospective study of 170 patients followed 4.8 years. Am J Gastroenterol 1997; 92:212–5. 6. Devesa SS, Blot WJ, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;83:2049 –53. 7. Spechler SJ, Lee E, Ahnen D, et al. Long-term outcome of medical and surgical therapies for gastroesophageal reflux disease: Follow-up of a randomized controlled trial. JAMA 2001;285:2331–8. 8. Sampliner RE, The Practice Parameters Committee of the American College of Gastroenterology. Updated guidelines for the diagnosis, surveillance, and therapy of Barrett’s esophagus. Am J Gastroenterol 2002;97:1888 –95. 9. Winters C Jr, Spurling TJ, Chobanian SJ, et al. Barrett’s esophagus. A prevalent, occult complication of gastroesophageal reflux disease. Gastroenterology 1987;92:118 –24. 10. Williamson WA, Ellis FH Jr, Gibb SP, et al. Effect of antireflux operation on Barrett’s mucosa. Ann Thorac Surg 1990; 49:537–41. 11. Ye W, Wong-Ho C, Lagergren J, et al. Risk of adenocarcinomas of the esophagus and gastric cardia in patients with gastroesophageal reflux diseases and after anti-reflux surgery. Gastroenterology 2001;121:1286 –93. 12. Bammer T, Hinder RA, Klaus A, Trastek VF, Achem SR. Rationale for surgical therapy of Barrett esophagus. Mayo Clinic Proc 2001;76:335–42. 13. Katz D, Rothstein R, Schned A, et al. The development of dysplasia and adenocarcinoma during endoscopic surveillance of Barrett’s esophagus. Am J Gastroenterol 1998;93:536 –41. 14. Role of anti-reflux surgery against dysplasia in Barrett’s esophagus Gastroenterology 1991;100:A21 (abstract). 15. Ortiz A, Martinez de Haro LF, Parrilla P, et al. Conservative treatment versus antireflux surgery in Barrett’s oesophagus:
2394
16. 17.
18.
19.
20. 21.
22. 23.
24. 25.
26. 27. 28. 29. 30.
Corey et al.
Long-term results of a prospective study. Br J Surg 1996;83: 274 –8. Csendes A, Burdiles P, Braghetto I, et al. Dysplasia and adenocarcinoma after classic antireflux surgery in patients with Barrett’s esophagus. Ann Surg 2002;235:178 –85. Zhang ZF, Kurtz RC, Sun M, et al. Adenocarcinomas of the esophagus and gastric cardia: Medical conditions, tobacco, alcohol, and socioeconomic factors. Cancer Epidemiol Biomarkers Prev 1996;5:761–8. Vaughan TL, Davis S, Kristal A, Thomas DB. Obesity, alcohol, and tobacco as risk factors for cancers of the esophagus and gastric cardia: Adenocarcinoma versus squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev 1995;4:85–92. Attwood SE, Barlow AP, Norris TL, Watson A. Barrett’s oesophagus: Effect of antireflux surgery on symptom control and development of complications. Br J Surg 1992;79: 1050 –3. Brand DL, Ylvisaker JT, Gelfand M, Pope CE. Regression of columnar esophageal (Barrett’s) epithelium after anti-reflux surgery. N Engl J Med 1980;302:844 –8. 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; 12:893–7. Cooper BT, Barbezat GO. Barrett’s oesophagus: A clinical study of 52 patients. Q J Med 1987;62:97–108. Csendes A, Braghetto I, Burdiles P, et al. Long-term results of classic antireflux surgery in 152 patients with Barrett’s esophagus: Clinical, radiologic, endoscopic, manometric, and acid reflux test analysis before and late after operation. Surgery 1998;126:645–57. DeMeester TR, Attwood SE, Smyrk TC, et al. Surgical therapy in Barrett’s esophagus. Ann Surg 1990;212:528 –40. Klinkenberg-Knol EC, Nelis F, Dent J, et al. Long-term omeprazole treatment in resistant gastroesophageal reflux disease: Efficacy, safety, and influence on gastric mucosa. Gastroenterology 2000;118:661–9. Klinkenberg-Knol EC, Festen HP, Jansen JB, et al. Long-term treatment with omeprazole for refractory reflux esophagitis: Efficacy and safety. Ann Intern Med 1994;121:161–7. Chen LQ, Hu CY, Gaboury L, et al. Proliferative activity in Barrett’s esophagus before and after antireflux surgery. Ann Surg 2001;234:172–80. Low DE, Levine DS, Dail DH, Kozarek RA. Histological and anatomic changes in Barrett’s esophagus after antireflux surgery. Am J Gastroenterol 1999;94:80 –5. Luostarinen ME, Mattila JJ, Auvinen OL, et al. Histological improvement of oesophagitis after Nissen fundoplication. Ann Med 1998;30:547–52. Lundell L, Miettinen P, Myrvold HE, et al. Long-term management of gastro-oesophageal reflux disease with omeprazole or open antireflux surgery: Results of a prospective, randomized clinical trial. The Nordic GORD Study Group. Eur J Gastroenterol Hepatol 2000;12:879 –87.
AJG – Vol. 98, No. 11, 2003
31. 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;44:700 –5. 32. McDonald ML, Trastek VF, Allen MS, et al. Barrett’s esophagus: Does an antireflux procedure reduce the need for endoscopic surveillance? J Thorac Cardiovasc Surg 1996;111: 1135–8. 33. Ovaska J, Miettinen M, Kivilaakso E. Adenocarcinoma arising in Barrett’s esophagus. Dig Dis Sci 1989;34:1336 –9. 34. Patti MG, Arcerito M, Feo CV, et al. Barrett’s esophagus: A surgical disease. J Gastrointest Surg 1999;3:397–403. 35. Peters FT, Ganesh S, Kuipers EJ, et al. Endoscopic regression of Barrett’s oesophagus during omeprazole treatment: A randomised double blind study. Gut 1999;45:489 –94. 36. Sagar PM, Ackroyd R, Hosie KB, et al. Regression and progression of Barrett’s oesophagus after antireflux surgery. Br J Surg 1995;82:806 –10. 37. Sampliner RE. Effect of up to 3 years of high-dose lansoprazole on Barrett’s esophagus. Am J Gastroenterol 1994;89: 1844 –8. 38. Sampliner RE, Garewal HS, Fennerty MB, Aickin M. Lack of impact of therapy on extent of Barrett’s esophagus in 67 patients. Dig Dis Sci 1990;35:93–6. 39. Sharma P, Weston AP, Morales T, et al. Relative risk of dysplasia for patients with intestinal metaplasia in the distal oesophagus and in the gastric cardia. Gut 2000;46:9 –13. 40. Sharma P, Morales TG, Bhattacharyya A, et al. Dysplasia in short-segment Barrett’s esophagus: A prospective 3-year follow-up. Am J Gastroenterol 1997;92:2012–6. 41. 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;92:582–5. 42. Skinner DB, Walther BC, Riddell RH, et al. Barrett’s esophagus. Comparison of benign and malignant cases. Ann Surg 1983;198:554 –65. 43. Srinivasan R, Katz PO, Ramakrishnan A, et al. Maximal acid reflux control for Barrett’s oesophagus: Feasible and effective. Aliment Pharmacol Ther 2001;15:519 –24. 44. Starnes VA, Adkins RB, Ballinger JF, Sawyers JL. Barrett’s esophagus. A surgical entity. Arch Surg 1984;119:563–7. 45. Wesdorp IC, Bartelsman J, Schipper ME, Tytgat GN. Effect of long-term treatment with cimetidine and antacids in Barrett’s oesophagus. Gut 1981;22:724 –7. 46. Weston AP, Badr AS, Hassanein RS. Prospective multivariate analysis of clinical, endoscopic, and histological factors predictive of the development of Barrett’s multifocal high-grade dysplasia or adenocarcinoma. Am J Gastroenterol 1999;94: 3413–9. 47. 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;13:1205–9.
Vol. 98, No. 11, 2003 ISSN 0002-9270/03/$30.00 doi:10.1016/j.amjgastroenterol.2003.08.008
Does a Surgical Antireflux Procedure Decrease the Incidence of Esophageal Adenocarcinoma in Barrett’s Esophagus? A Meta-Analysis Kathleen E. Corey, M.D., M.P.H., Sarah M. Schmitz, B.A., and Nicholas J. Shaheen, M.D., M.P.H. Center for Esophageal Diseases and Swallowing, the Division of Digestive Diseases, and the School of Public Health, University of North Carolina at Chapel Hill
OBJECTIVE: The risk of adenocarcinoma of the esophagus is increased among those with Barrett’s esophagus (BE). Whether the risk of cancer in the setting of BE can be decreased by a surgical antireflux procedure (SARP) is unclear. This study compared the reported incidence of esophageal adenocarcinoma in subjects with BE who underwent SARP with those with BE who had medical management. METHODS: We used MEDLINE to perform a meta-analysis of the English language literature published from 1966 through October 2001. We reviewed abstracts found with the search term “Barrett’s esophagus” and the following: “adenocarcinoma,” “esophageal neoplasm,” “proton pump inhibitor,” “fundoplication,” or “antireflux procedure.” Study entry criteria included 1) trial or cohort study with a report of cancer risk expressible in cancers per patient-year, 2) histologic confirmation of BE and any adenocarcinomas, and 3) adequate description of intervention (medical vs SARP). Data were abstracted by two reviewers using standardized forms. Subgroup comparisons were made using only medical management studies published in the last 5 yr. Multivariable regression controlling for subject age, country of origin, and BE length was performed. RESULTS: We reviewed 1247 abstracts, and 34 met the inclusion criteria. There were a cumulative 4678 patient-years of follow-up in the SARP group and 4906 patient-years in the medical group. The cancer incidence rate in the SARP group was 3.8 cancers/1000 patient-years, compared with 5.3 in the medical group (p ⫽ 0.29). Similarly, there was no significant difference between cancer rates when comparing SARP with medical series reported in the last 5 yr (3.8/1000 patient-years vs 4.2/1000 patient-years, p ⫽ 0.33). Multivariate analysis controlling for subject age, country of origin, and BE length did not alter these findings. CONCLUSION: The reported risk of adenocarcinoma in subjects with BE is low and not significantly decreased by a surgical antireflux procedure. Antireflux surgery in the setting of BE should not be recommended as an antineoplastic measure. (Am J Gastroenterol 2003;98:2390 –2394. © 2003 by Am. Coll. of Gastroenterology)
INTRODUCTION Barrett’s esophagus (BE) is a metaplastic condition resulting from chronic gastroesophageal reflux injury in which the squamous epithelium of the esophagus is replaced by intestinal epithelium (1, 2). With time, patients might develop dysplasia and, eventually, esophageal adenocarcinoma. Thus, BE has been associated with an increased risk of the development of esophageal adenocarcinoma. Estimates have placed the incidence of esophageal adenocarcinoma in patients with BE at approximately 500 cases per 100,000, whereas incidence in the general population is considerably less (3– 6). A randomized, controlled trial of surgical and medical interventions for gastroesophageal reflux disease (GERD) by Spechler et al. (7) reconfirmed the increased risk of esophageal adenocarcinoma in BE patients, with a risk of 0.4% per patient-year compared with 0.07% per patient-year in patients without BE (7). Because of the increased incidence of carcinoma in the setting of BE, longterm endoscopic follow-up and surveillance has been suggested (8). The development of BE has been linked to chronic GERD (9). However, the effect of treatment of chronic reflux disease on the progression of BE remains uncertain. Currently, chronic reflux disease is managed either medically by the long-term use of H2 receptor antagonists or proton pump inhibitors, or with surgical antireflux procedures (SARP). In both treatment types, progression to adenocarcinoma has been demonstrated (10, 11). Some authors have suggested that surgical approaches to reflux might be superior to medical management in arresting the progression of BE to cancer (12). However, neither medical nor surgical intervention has been consistently shown in the literature to preferentially decrease the incidence of adenocarcinoma. Thus, no consensus exists as to which intervention is preferred to decrease the progression to malignancy in the setting of BE. This meta-analysis compared the reported rates of adenocarcinoma of the esophagus in the English language literature in those BE patients undergoing SARP with the rates of those patients managed medically. Additionally, we
AJG – November, 2003
sought to identify any study variables that might influence cancer risk.
MATERIALS AND METHODS
Surgery and Barrett’s Cancer
2391
origin, BE length, and BE definition was performed with the SAS system (SAS Institute, Cary, NC). A value was generated to assess the agreement between the abstracted data of the two reviewers.
Study Selection and Data Abstraction We performed a search of the MEDLINE database for the period 1966 through October 2001. All English-language abstracts with “Barrett’s esophagus” as a text word or subject heading and one of the following were reviewed: “adenocarcinoma,” “esophageal neoplasm,” “proton pump inhibitor,” “fundoplication,” and “antireflux procedure.” Bibliographies were reviewed for additional eligible studies. The following criteria for inclusion into the meta-analysis were used: 1) randomized controlled trial or cohort study with a reported esophageal adenocarcinoma risk expressible in cancers per patient-year, 2) histologic confirmation of BE at cohort inception and histologic confirmation of any subsequent adenocarcinoma, 3) ability to determine the patients’ treatment regimen (SARP vs medical therapy), 4) ability to determine the patients’ cancer outcomes and allcause mortality, and 5) A follow-up period of 12 months or more. Two independent investigators (K.C., S.S.) abstracted the included studies. Variables assessed included year of publication, country of origin, study design, definition of BE, endoscopic surveillance interval, total cancers developed, cancers developed in the first year after start of intervention, and total patient-years. For serial studies reporting the outcomes in the same cohort of patients, the more recent publication was used. Differences between the reviewers in classification were resolved by consensus. The currently accepted definition of BE is any length of endoscopically evident salmon-colored tissue in the tubular esophagus which, on histologic evaluation, is of the “specialized” or “intestinalized” type. However, because past investigators have sometimes used slightly different definitions, we also assessed the reports for the definition of BE used. The reports were classified with respect to BE definition in the following categories: 1) metaplasia not otherwise specified (NOS), 2) metaplasia NOS greater than 3 cm, 3) intestinal metaplasia, 4) intestinal metaplasia greater than 3 cm, and 5) no definition specified.
RESULTS
Data Analysis The crude cancer incidence was obtained by dividing the total number of cancers observed by the total number of patient-years in the study. Patient-years were determined by assessing the number of years from study entry until the development of adenocarcinoma or the end of the study period. If individual patient data were unavailable with which to calculate the total number of patient-years, the mean number of years of patient follow-up was multiplied by the number of subjects in the study to obtain the total patient-years encompassed by the study. Multivariate regression controlling for the mean subject age, country of
DISCUSSION
A total of 1247 abstracts were reviewed. Of these, 49 articles were retrieved, of which 16 publications met inclusion criteria. The remaining 33 articles lacked either adequate description of the intervention, used a follow-up period of less than 12 months, or were serial reports of the same cohort. Review of the bibliographies of the initial 49 retrieved articles led to an additional 18 articles for inclusion. Thus, a total of 34 articles were included for analysis (Table 1). Agreement between the reviewers was high ( ⫽ 0.906). Univariate analysis revealed 4678.3 patient-years among the patients receiving SARP. Medical patient-years totaled 4906.2. Owing to the advances in medical therapy in recent years with the addition of proton pump inhibitors to GERD and BE treatment, medical therapy was analyzed in all studies and in the subgroup of studies conducted only in the most recent 5 yr analyzed (1996 –2001). There were 4205.5 patient-years of medical patient follow-up in studies conducted in those 5 yr. In the surgical intervention group, there were 18 cancers over 4678.3 patient-years, or 3.8 cancers per 1000 patient-years. In the medical group, there were 26 cancers over 4906.2 patient-years, or 5.3 cancers per 1000 patient-years (p ⫽ 0.29, when compared with the SARP group). In medical patients in studies limited to the past 5 yr, there were 18 cancers over 4205.5 patient-years, or 4.3 cancers per 1000 patient-years (p ⫽ 0.33), when compared with the SARP group (Table 2). There was no significant difference in cancers per patient-year between the surgical and medical groups or between the surgical group and the medical therapy group in the last 5 yr. Multivariable analysis controlling for country of origin, age, length of Barrett’s segment, and definition of BE was performed. After controlling for these potentially confounding covariates, no significant differences were noted in the cancer risk between the medical and surgical groups (p ⫽ 0.39, full models not shown).
This analysis compared the published rates of esophageal adenocarcinoma in those with BE treated with surgical antireflux procedures with the rates of those treated medically. No significant difference was seen in the incidence of adenocarcinoma between these groups of patients. These results conflict with some surgical series, which suggested that SARP might be superior to medical management of BE in the prevention of adenocarcinoma. Katz et al.(13) studied 102 patients to determine risk factors for dysplasia in BE patients and found that whereas 23 of the medically treated
2392
Corey et al.
AJG – Vol. 98, No. 11, 2003
Table 1. Studies Reporting Cancer risk in BE Patients With a Specified Medical and/or Surgical Intervention
First Author (reference)
Source and Year
Attwood (19) Brand (20) Cooper (21) Cooper (22) Csendes (23) DeMeester (24) Drewitz (5) Katz (13) Klinkenberg-Knol (25) Klinkenberg-Knol (26) Chen (27) Low (28) Luostrarinien (29) Lundell (30) Malesci (31) McDonald (32) Ortiz (15) Ovaska (33) Patti (34) Peters (35) Sagar (36) Sampliner (37) Sampliner (38) Sharma (39) Sharma (40) Sharma (41) Skinner (42) Spechler (7) Srinivasan (43) Starnes (44) Wesdorp (45) Weston (46) Wilkinson (47) Williamson (10) Total
No. of Patients; Surgical
Total PatientYears; Surgical
Total Cancers; Surgical
No. of Patients; Medical
Total PatientYears; Medical
Total Cancers; Medical
19 10 0 0 152 35 0 16 0 0 45 14 5 22 0 113 32 12 38 0 56 0 9 0 0 0 13 129 0 8 0 0 0 26 754
57 45 0 0 1266 105 0 76.8 0 0 180 29.4 33.3 66 0 734.5 160 80.4 72.96 0 392 0 27 0 0 0 52 1174 0 17.6 0 0 0 109.2 4678.3
1 1 0 0 4 0 0 0 0 0 0 0 0 0 0 3 1 2 0 0 1 0 0 0 0 0 0 1 0 1 0 0 0 3 18
26 0 47 7 0 6 170 0 84 32 0 0 0 16 14 0 27 10 8 53 0 27 58 78 32 13 0 137 9 12 9 108 12 0 918
78 0 182 9.6 0 18 807.5 0 546 128 0 0 0 48 168 0 135 67 15.36 106 0 78.3 174 220.7 98.4 74.1 0 1452 40.5 26.4 17 359.1 57 0 4906.2
1 0 0 0 0 0 9 0 1 0 0 0 0 0 0 0 1 1 0 1 0 0 0 1 1 0 0 4 0 1 0 5 0 0 26
Brit J Surg 1992 New Engl J Med 1980 Aliment Pharmacol Ther 1998 Q J Med 1987 Surgery 1998 Ann Surg 1990 Am J Gastroenterol 1997 Am J Gastroenterol 1998 Gastroenterology 2000 Ann Intern Med 1994 Ann Surg 2001 Am J Gastroenterol 1999 Ann Med 1998 Eur J Gastroenterol Hepatol 2000 Gastrointest Endosc 1996 J Thorac Cardiovasc Surg 1996 Br J Surg 1996 Dig Dis Sci 1989 J Gastrointest Surg 1999 Gut 1999 Br J Surg 1995 Am J Gastroenterol 1994 Dig Dis Sci 1990 Gut 2000 Am J Gastroenterol 1997 Am J Gastroenterol 1997 Ann Surg 1983 JAMA 2001 Aliment Pharmacol Ther 2001 Arch Surg 1984 Gut 1981 Am J Gastroenterol 1999 Aliment Pharm Ther 1999 Ann Thorac Surg 1990
patients developed dysplasia and four developed adenocarcinoma, none of the patients who underwent antireflux procedures developed dysplasia of any type. Also, McCallum et al. (14) demonstrated that 3.4% of BE patients who underwent SARP developed dysplasia, compared with 19.7% of those treated medically. In addition, no patient in the SARP group developed adenocarcinoma, whereas 1.3% in the medical group did. Finally, Ortiz et al. (15) performed a randomized, controlled trial of 59 BE patients assigned to Table 2. Incidence Rate of Esophageal Adenocarcinoma in BE Patients Receiving SARP and Medical Therapy
Incidence rate (Cancer/1000 Patient-Years)
SARP
Medical Therapy
Medical Therapy in Last Five Years
3.8 (2.4–6.1)
5.3 (3.6–7.8)
4.3 (2.6–5.8)
Numbers in parentheses are 95% CIs.
receive either SARP or medical therapy with H2 antagonists. High-grade dysplasia developed in only two patients, one of whom underwent medical therapy and one who had a failed SARP. Although the incidence of cancer was not significantly different between the surgical and medical arms of the Ortiz study, significantly fewer SARP patients than medically managed patients went on to develop any form of dysplasia. In addition, fewer surgical patients experienced upward BE progression relative to those who underwent medical therapy. Although these positive studies seem to promote the value of SARP in BE, all are small studies, adding relatively few patient-years to the literature. When these studies are added to the cumulative reported surgical experience, they are insufficient to cause a significant difference between the medical and surgical arms of the meta-analysis. Previous investigators have attempted to compare the cancer rates in medically and surgically treated subjects with BE. A pooled analysis undertaken by Bammer et al. (12) found that surgery was superior to medical therapy in
AJG – November, 2003
decreasing the incidence of esophageal adenocarcinoma. That study differs from the present meta-analysis in several respects. Our study excluded any report that did not have 1 yr of follow-up after treatment, an exclusion that the study by Bammer et al. did not undertake. In addition, the study by Bammer et al. did not include 14 studies, which our searches found and included. This resulted in considerably more medical patient-years in the present study. These differences between the study by Bammer et al. and the present meta-analysis might have resulted in the previous study overestimating the benefit of SARP in BE patients. Several potential sources of bias exist within this study. First, this meta-analysis includes cohort studies in which the patients were not randomly assigned to an intervention. Thus, the degree of GERD severity might not have been equally distributed between SARP and medical therapy groups. The potential exists that patients with more severe GERD underwent antireflux procedures, whereas those with less severe disease underwent medical therapy. If disease severity correlates with cancer risk, this would have the effect of biasing the results of the relationship against surgery. A second source of bias might be derived from the use of a variety of SARP in the surgically treated patients. Procedures were not uniform across the studies, nor did all studies specify the procedure used, making analysis controlling for the type of SARP performed impossible. Also, most studies did not assess the competency of the wrap in controlling reflux. Certainly, if cancer is promoted by ongoing reflux, the cancer risk in those with ineffective wraps might be expected to be higher than in those with good acid control. Preliminary evidence suggests that ineffective wraps might indeed be associated with an increased risk of the subsequent development of dysplasia (16). Next, most studies of medical therapy made no attempt to assess patient compliance with medications. If subjects in the medical arm were receiving less than optimal doses or were taking these medications with poor compliance, cancer rates might similarly be elevated. Also, information regarding other confounders, such as alcohol use and tobacco use, were not available from most studies and could not be controlled for in this study. Although tobacco and alcohol use have been associated with the development of adenocarcinoma (17), they are more strongly associated with the development of squamous cell carcinoma (18). Finally, although our study encompasses the entire reported English literature, only a small number of subjects in the medically and surgically managed arms developed cancer. These small numbers of cancers increase the likelihood of a type 2 statistical error. Despite the small number of cancer cases, a large number of patient-years were available, strengthening the precision of these estimates. In conclusion, the incidence of adenocarcinoma development in BE patients is low, approximating 5 per 1000 patient-years. Based on the cumulative published literature, surgical antireflux procedures do not seem to provide benefit over medical therapy in BE patients in reducing the inci-
Surgery and Barrett’s Cancer
2393
dence of cancer development. Surgical antireflux procedures should not be recommended over medical therapy to decrease the cancer risk associated with BE.
ACKNOWLEDGMENTS This work was supported in part by National Institutes of Health grant K23DK59311-01. Reprint requests and correspondence: Nicholas Shaheen, M.D., M.P.H., University of North Carolina Hospitals, Division of Digestive Disease and Nutrition, CB #7080, 724 Burnett-Womack Building, Chapel Hill, NC 27599-7080. Received Dec. 13, 2002; accepted Apr. 8, 2003.
REFERENCES 1. Spechler SJ. Clinical practice. Barrett’s esophagus. N Engl J Med 2002;346:836 –42. 2. Shaheen N, Ransohoff DF. Gastroesophageal reflux, Barrett esophagus, and esophageal cancer: Scientific review. JAMA 2002;287:1972–81. 3. Shaheen NJ, Crosby MA, Bozymski EM, Sandler RS. Is there publication bias in the reporting of cancer risk in Barrett’s esophagus? Gastroenterology 2000;119:333–8. 4. Spechler SJ, Robbins AH, Rubins HB, et al. Adenocarcinoma and Barrett’s esophagus. An overrated risk? Gastroenterology 1984;87:927–33. 5. Drewitz DJ, Sampliner RE, Garewal HS. The incidence of adenocarcinoma in Barrett’s esophagus: A prospective study of 170 patients followed 4.8 years. Am J Gastroenterol 1997; 92:212–5. 6. Devesa SS, Blot WJ, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;83:2049 –53. 7. Spechler SJ, Lee E, Ahnen D, et al. Long-term outcome of medical and surgical therapies for gastroesophageal reflux disease: Follow-up of a randomized controlled trial. JAMA 2001;285:2331–8. 8. Sampliner RE, The Practice Parameters Committee of the American College of Gastroenterology. Updated guidelines for the diagnosis, surveillance, and therapy of Barrett’s esophagus. Am J Gastroenterol 2002;97:1888 –95. 9. Winters C Jr, Spurling TJ, Chobanian SJ, et al. Barrett’s esophagus. A prevalent, occult complication of gastroesophageal reflux disease. Gastroenterology 1987;92:118 –24. 10. Williamson WA, Ellis FH Jr, Gibb SP, et al. Effect of antireflux operation on Barrett’s mucosa. Ann Thorac Surg 1990; 49:537–41. 11. Ye W, Wong-Ho C, Lagergren J, et al. Risk of adenocarcinomas of the esophagus and gastric cardia in patients with gastroesophageal reflux diseases and after anti-reflux surgery. Gastroenterology 2001;121:1286 –93. 12. Bammer T, Hinder RA, Klaus A, Trastek VF, Achem SR. Rationale for surgical therapy of Barrett esophagus. Mayo Clinic Proc 2001;76:335–42. 13. Katz D, Rothstein R, Schned A, et al. The development of dysplasia and adenocarcinoma during endoscopic surveillance of Barrett’s esophagus. Am J Gastroenterol 1998;93:536 –41. 14. Role of anti-reflux surgery against dysplasia in Barrett’s esophagus Gastroenterology 1991;100:A21 (abstract). 15. Ortiz A, Martinez de Haro LF, Parrilla P, et al. Conservative treatment versus antireflux surgery in Barrett’s oesophagus:
2394
16. 17.
18.
19.
20. 21.
22. 23.
24. 25.
26. 27. 28. 29. 30.
Corey et al.
Long-term results of a prospective study. Br J Surg 1996;83: 274 –8. Csendes A, Burdiles P, Braghetto I, et al. Dysplasia and adenocarcinoma after classic antireflux surgery in patients with Barrett’s esophagus. Ann Surg 2002;235:178 –85. Zhang ZF, Kurtz RC, Sun M, et al. Adenocarcinomas of the esophagus and gastric cardia: Medical conditions, tobacco, alcohol, and socioeconomic factors. Cancer Epidemiol Biomarkers Prev 1996;5:761–8. Vaughan TL, Davis S, Kristal A, Thomas DB. Obesity, alcohol, and tobacco as risk factors for cancers of the esophagus and gastric cardia: Adenocarcinoma versus squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev 1995;4:85–92. Attwood SE, Barlow AP, Norris TL, Watson A. Barrett’s oesophagus: Effect of antireflux surgery on symptom control and development of complications. Br J Surg 1992;79: 1050 –3. Brand DL, Ylvisaker JT, Gelfand M, Pope CE. Regression of columnar esophageal (Barrett’s) epithelium after anti-reflux surgery. N Engl J Med 1980;302:844 –8. 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; 12:893–7. Cooper BT, Barbezat GO. Barrett’s oesophagus: A clinical study of 52 patients. Q J Med 1987;62:97–108. Csendes A, Braghetto I, Burdiles P, et al. Long-term results of classic antireflux surgery in 152 patients with Barrett’s esophagus: Clinical, radiologic, endoscopic, manometric, and acid reflux test analysis before and late after operation. Surgery 1998;126:645–57. DeMeester TR, Attwood SE, Smyrk TC, et al. Surgical therapy in Barrett’s esophagus. Ann Surg 1990;212:528 –40. Klinkenberg-Knol EC, Nelis F, Dent J, et al. Long-term omeprazole treatment in resistant gastroesophageal reflux disease: Efficacy, safety, and influence on gastric mucosa. Gastroenterology 2000;118:661–9. Klinkenberg-Knol EC, Festen HP, Jansen JB, et al. Long-term treatment with omeprazole for refractory reflux esophagitis: Efficacy and safety. Ann Intern Med 1994;121:161–7. Chen LQ, Hu CY, Gaboury L, et al. Proliferative activity in Barrett’s esophagus before and after antireflux surgery. Ann Surg 2001;234:172–80. Low DE, Levine DS, Dail DH, Kozarek RA. Histological and anatomic changes in Barrett’s esophagus after antireflux surgery. Am J Gastroenterol 1999;94:80 –5. Luostarinen ME, Mattila JJ, Auvinen OL, et al. Histological improvement of oesophagitis after Nissen fundoplication. Ann Med 1998;30:547–52. Lundell L, Miettinen P, Myrvold HE, et al. Long-term management of gastro-oesophageal reflux disease with omeprazole or open antireflux surgery: Results of a prospective, randomized clinical trial. The Nordic GORD Study Group. Eur J Gastroenterol Hepatol 2000;12:879 –87.
AJG – Vol. 98, No. 11, 2003
31. 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;44:700 –5. 32. McDonald ML, Trastek VF, Allen MS, et al. Barrett’s esophagus: Does an antireflux procedure reduce the need for endoscopic surveillance? J Thorac Cardiovasc Surg 1996;111: 1135–8. 33. Ovaska J, Miettinen M, Kivilaakso E. Adenocarcinoma arising in Barrett’s esophagus. Dig Dis Sci 1989;34:1336 –9. 34. Patti MG, Arcerito M, Feo CV, et al. Barrett’s esophagus: A surgical disease. J Gastrointest Surg 1999;3:397–403. 35. Peters FT, Ganesh S, Kuipers EJ, et al. Endoscopic regression of Barrett’s oesophagus during omeprazole treatment: A randomised double blind study. Gut 1999;45:489 –94. 36. Sagar PM, Ackroyd R, Hosie KB, et al. Regression and progression of Barrett’s oesophagus after antireflux surgery. Br J Surg 1995;82:806 –10. 37. Sampliner RE. Effect of up to 3 years of high-dose lansoprazole on Barrett’s esophagus. Am J Gastroenterol 1994;89: 1844 –8. 38. Sampliner RE, Garewal HS, Fennerty MB, Aickin M. Lack of impact of therapy on extent of Barrett’s esophagus in 67 patients. Dig Dis Sci 1990;35:93–6. 39. Sharma P, Weston AP, Morales T, et al. Relative risk of dysplasia for patients with intestinal metaplasia in the distal oesophagus and in the gastric cardia. Gut 2000;46:9 –13. 40. Sharma P, Morales TG, Bhattacharyya A, et al. Dysplasia in short-segment Barrett’s esophagus: A prospective 3-year follow-up. Am J Gastroenterol 1997;92:2012–6. 41. 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;92:582–5. 42. Skinner DB, Walther BC, Riddell RH, et al. Barrett’s esophagus. Comparison of benign and malignant cases. Ann Surg 1983;198:554 –65. 43. Srinivasan R, Katz PO, Ramakrishnan A, et al. Maximal acid reflux control for Barrett’s oesophagus: Feasible and effective. Aliment Pharmacol Ther 2001;15:519 –24. 44. Starnes VA, Adkins RB, Ballinger JF, Sawyers JL. Barrett’s esophagus. A surgical entity. Arch Surg 1984;119:563–7. 45. Wesdorp IC, Bartelsman J, Schipper ME, Tytgat GN. Effect of long-term treatment with cimetidine and antacids in Barrett’s oesophagus. Gut 1981;22:724 –7. 46. Weston AP, Badr AS, Hassanein RS. Prospective multivariate analysis of clinical, endoscopic, and histological factors predictive of the development of Barrett’s multifocal high-grade dysplasia or adenocarcinoma. Am J Gastroenterol 1999;94: 3413–9. 47. 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;13:1205–9.