Reactive gastropathy is associated with inflammatory conditions throughout the gastrointestinal tract

Transcription

1 Alimentary Pharmacology and Therapeutics Reactive gastropathy is associated with inflammatory conditions throughout the gastrointestinal tract I. Maguilnik*, W. L. Neumann,, A. Sonnenberg, & R. M. Genta,, ** *Universidade do Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil. Miraca Research Institute, Miraca Life Sciences, Irving, TX, USA. University of Texas Southwestern Medical Center, Dallas, TX, USA. Portland VA Medical Center, Portland, OR, USA. Oregon Health & Science University, Portland, OR, USA. **Dallas VA Medical Center, Dallas, TX, USA. Correspondence to: Dr R. M. Genta, Miraca Life Sciences, 6655 North MacArthur Blvd, Irving, TX 75039, USA. Publication data Submitted 10 July 2012 First decision 7 August 2012 Resubmitted 9 August 2012 Accepted 10 August 2012 EV Pub Online 28 August 2012 SUMMARY Background The epidemiology of reactive gastropathy and its relationship with other conditions of the gastrointestinal tract associated with NSAID use have not been evaluated. Aims To test the hypothesis that if reactive gastropathy shares common aetiological factors with these conditions, the analysis of a large cohort would unveil associations. Methods We queried a national pathology database for subjects with a diagnosis of reactive gastropathy; controls were patients with normal gastric biopsies. We also extracted diagnoses of H. pylori infection, intestinal metaplasia, duodenal lymphocytosis, duodenitis, ileitis, microscopic colitis and focal colitis. Results Of patients with gastric biopsies, had oesophageal, duodenal, ileal and colonic biopsies. Reactive gastropathy was diagnosed in 15.6% of patients, H. pylori infection in 10.3% and normal gastric mucosa in 16.3%. Reactive gastropathy was evenly distributed across the US and increased from 2.0% in the first decade of life to >20% in octogenarians. Compared with controls, reactive gastropathy was significantly associated with Barrett s mucosa (OR % CI ); duodenitis (OR 1.36; 95% CI ); duodenal intraepithelial lymphocytosis (OR 1.25; 95% CI ); active ileitis (OR 1.88; 95% CI ); focal active colitis (OR 1.57; 95% CI ); and collagenous colitis (OR 1.50; 95% CI ). Conclusions Reactive gastropathy, a common histopathological feature of the stomach, shows an age-dependent rise and is associated with changes of the digestive tract believed to be caused by NSAID use or duodenogastric reflux. However, a large fraction of reactive gastropathy remains unexplained; its frequent occurrence merits further efforts at elucidating its aetiology. Aliment Pharmacol Ther 2012; 36: doi: /apt.12031

2 Reactive gastropathy: epidemiology and associations INTRODUCTION An association between the presence of bile in the stomach and damage to the gastric mucosa was first mentioned by William Beaumont in 1833 as part of his decade-long observations of Alexis St. Martin s permanent open gastric fistula. 1 The histopathological features of the changes seen in the gastric mucosa exposed to biliary and duodenopancreatic reflux after partial gastrectomy, particularly Billroth II operations, were described in the 1960s and 1970s and generically referred to as post-operative gastritis 2 5 ; their possible premalignant nature was later noted in cohorts of operated patients. 6 9 Between 1983 and 1986, Dixon s group in Liverpool published a series of papers in which they systematically described these changes and proposed bile reflux gastritis as a distinct clinicopathological entity Shortly thereafter, it seemed clear that identical lesions were often found in the gastric mucosa of patients who regularly used nonsteroidal anti-inflammatory drugs (NSAIDs), and the broader term chemical gastritis was introduced and recommended by the Updated Sydney System. 14 An update of the criteria for the differential histopathological diagnosis of this condition was published in In this article, we use the synonym reactive gastropathy, a histopathological diagnosis currently favoured by most pathologists and widely understood by gastroenterologists as an indication of gastric damage associated with either bile reflux or chronic NSAID use. Although the diagnosis of reactive gastropathy in biopsy specimens from the intact stomach is quite common, its clinical and pathological associations remain unclear. Specifically, its distribution in the population and its relationship with other types of gastritis, oesophagitis and certain conditions of the small and large intestine (duodenitis, ileitis, focal colitis and microscopic colitis) believed to be associated with NSAIDs use have not been systematically evaluated. This study was designed to test the hypothesis that if reactive gastropathy shares a common aetiological factor with these conditions, the analysis of a large cohort of patients should unveil associations. We also aimed at determining the epidemiologic characteristics of reactive gastropathy in the United States. PATIENTS AND METHODS Study setting This study was conducted at Miraca Life Sciences, a specialised gastrointestinal laboratory receiving specimens from gastroenterologists operating in private out-patient endoscopy centres across the United States. Biopsies are interpreted by a group of gastrointestinal pathologists who attempt to maintain a high level of diagnostic uniformity through a predetermined approach to specimen handling, diagnostic criteria and terminology. Consensus is maintained and updated through daily multi-headed microscope conferences, monthly didactic and journal review conferences, a terminology review committee and ongoing quality assurance reviews. This study was approved by the Miraca Institutional Review Board. Patients and controls We analysed electronic data from the Miraca database, which includes demographic and clinical information for each patient, a summary of the endoscopic findings or the entire endoscopic report, the site of origin of each specimen and the histopathology report for each biopsy. To identify the records for eligible patients, we extracted data for subjects who had at least one gastric biopsy submitted to Miraca Life Sciences between 1 January 2008 and 31 December Patients with a history or a diagnosis of upper gastrointestinal cancer or surgery were excluded; when an individual had multiple encounters, only the first one was included in this analysis. By electronically analysing the histopathology reports and using additional search terms and Boolean logic in Visual Basic for Applications (VBA), we collected all patients whose gastric biopsies carried the histopathological diagnosis of either reactive or chemical gastropathy. Cases in which these terms were qualified by mild, minimal, consistent with, or suspicious for were excluded. Controls were patients in whom all gastric biopsies were diagnosed as normal. To evaluate for possible pathological associations, we also extracted the diagnoses of H. pylori infection, intestinal metaplasia, duodenal lymphocytosis, peptic duodenitis, active ileitis, lymphocytic colitis, collagenous colitis and focal active colitis. Patients with a diagnosis or known history of inflammatory bowel disease were excluded from the analyses for active ileitis and focal active colitis. To determine whether reactive gastropathy was commonly associated with certain clinical presentations, we also evaluated patients for a history of epigastric or abdominal pain, dyspepsia, anaemia, gastro-oesophageal reflux disease (GERD) and diarrhoea. Histopathological criteria In our laboratory, gastric biopsy specimens are evaluated according to the Updated Sydney System. 14 Specifically, Aliment Pharmacol Ther 2012; 36:

3 I. Maguilnik et al. H. pylori gastritis is diagnosed when H. pylori organisms are demonstrated either by a specific monoclonal immunochemical stain (Cell Marque, Rocklin, CA, USA) or by the HP Blue Stain. Helicobacter-negative chronic active gastritis is defined by focal or diffuse chronic inflammation (lymphocytes and plasma cells) with neutrophils infiltrating surface or foveolar epithelium in the absence of Helicobacter organisms. Chronic inactive gastritis is characterised by focal or diffuse chronic inflammation without neutrophilic granulocytes or Helicobacter organisms. Criteria for reactive gastropathy are based on the 2005 definition, which includes various combinations of foveolar hyperplasia, regenerative changes in the surface epithelium, oedema or hyperaemia of the lamina propria, erosions and smooth muscle proliferation. 15 An illustrative case is depicted in Figure 1. Duodenitis is diagnosed when the duodenal mucosa shows active (neutrophilic) inflammation in the epithelium, which may also be eroded, irrespective of the magnitude of the lymphoplasmacytic infiltrates, and is often accompanied by foci of gastric foveolar metaplasia (the so called peptic duodenitis ). Duodenal 15, 16 lymphocytosis and lymphocytic colitis both require counts of 25 or more intraepithelial lymphocytes per 100 epithelial cells. 17 Statistical analysis The prevalence of each gastric diagnosis was expressed as a fraction of the entire patient population, whereas the prevalence of oesophageal, duodenal, ileal and colonic pathology diagnoses was expressed as a fraction of patients who also had oesophageal, duodenal, ileal or colonic biopsies. The age difference between pairs of patient groups was compared using the Student s t-test. The gender distribution among different patient groups and the associations of reactive gastropathy with various other pathology and clinical findings were expressed as odds ratios and their 95% CI. The prevalence of the gastric diagnoses in each US region was calculated as a fraction of the patient population of the respective region. Geographical distributions were compared using linear regression analysis. RESULTS Patients The study included unique patients (median age 57 years, range 3 months to 101 years; 61.9% female) with gastric biopsies; of these, had also the results of oesophageal biopsies available in the database; Figure 1 Antral mucosa exhibiting the features of reactive gastropathy, including corkscrew-like foveolar hyperplasia, a mucin depleted epithelium and bundles of hyperplastic smooth muscle arranged perpendicular to the surface had results from duodenal biopsies; from ileal biopsies; and from colonic biopsies. Demographics and geographical distribution Reactive gastropathy was diagnosed in 15.6% of the patients; H. pylori infection in 10.3% and a normal gastric mucosa in 16.3%. In contrast to H. pylori gastritis, which reached a plateau in the fourth decade and declined slightly thereafter, the prevalence of reactive gastropathy increased with age, from 2.0% in the first decade of life to a peak of 23.6% in women older than 90 years and to a peak of 17.9% in men aged years (Figure 2). The geographical distribution of reactive gastropathy and H. pylori gastritis among the US census divisions and regions are shown in Table 1. The occurrence of reactive gastropathy was fairly evenly distributed across the United States with a variation from a low of 19.7% in the Mountain region to 23.5% in New England (a 20% variation between the lowest and the highest prevalence rate). In contrast, H. pylori-positive gastritis exhibited a 200% variation (from 6.7% in the West North Central region to 14.2% in the Middle Atlantic). No significant correlation was found between the geographical variations of these two histopathological diagnoses. Correlation with other gastric pathology Chronic inactive gastritis was present in 5.9% of all patients, Helicobacter-negative chronic active gastritis in 3.4%, and intestinal metaplasia in 3.5%. Table 2 shows a stratification of the patient population by gastritis 738 Aliment Pharmacol Ther 2012; 36:

4 Reactive gastropathy: epidemiology and associations Percentage of patients (%) F M F M Age (years) Figure 2 Prevalence of reactive gastropathy (solid lines) and of Helicobacter pylori gastritis (dotted lines) in males and females for each decade of life. Table 1 Percentage of patients with reactive gastropathy and Helicobacter pylori gastritis in different US census regions Region Reactive gastropathy (%) Northeast New England Middle Atlantic Midwest East North Central West North Central South South Atlantic East South Central West South Central West Mountain Pacific Grand total H. pylori gastritis (%) type, sex and age. Compared with controls, patients with reactive gastropathy were slightly older and more likely to be females. In contrast, H. pylori infection was diagnosed more frequently in males and in younger patients. Intestinal metaplasia, which was not exclusive of other pathology, was more frequent in males and in older patients. Of the patients with reactive gastropathy, 6069 (7.7%) also had intestinal metaplasia; in contrast, intestinal metaplasia was detected in 7.1% of patients with H. pylori gastritis (OR 1.09; 95% CI ; P < ) and in 20.0% of patients with chronic inactive gastritis (OR 0.36; 95% CI ; P < ). A concurrent H. pylori infection was present in 329 (0.42%) patients with reactive gastropathy. Associations with other histopathological findings in other parts of the GI tract Table 3 shows that patients with reactive gastropathy were significantly less likely to have a normal mucosa in any segment of the gastrointestinal tract. Reflux oesophagitis and Barrett s oesophagus (particularly with dysplasia) were significantly associated with reactive gastropathy, as was lymphocytic oesophagitis, suggesting that these conditions might share common aetiological factors. In contrast, there was a strong inverse relationship with eosinophilic oesophagitis. In the duodenum, both peptic injury and duodenal intraepithelial lymphocytosis were more common in patients with reactive gastropathy than in those with a normal stomach. A significant association with these two conditions was also noted for H. pylori gastritis: OR 1.47 for peptic duodenitis (95% CI ) and OR 2.25 (95% CI ) for intraepithelial lymphocytosis. Significant associations were also detected between reactive gastropathy and active ileitis, focal active colitis and collagenous colitis. Aliment Pharmacol Ther 2012; 36:

5 I. Maguilnik et al. Table 2 Stratification of patient population by gastritis type, gender and age Gastritis type Total (%) Females (%) Males (%) OR (95% CI) Mean age (s.d.) Total (100) (61.9) (38.1) 55.9 (16.6) Normal gastric mucosa (16.3) (57.6) (42.4) (17.5) Reactive gastropathy (15.6) (65.7) (34.2) 1.41 ( ) 59.4 (15.5) Helicobacter pylori gastritis (10.3) (58.7) (41.3) 1.05 ( ) 55.8 (15.6) H. pylori-negative gastritis (3.4) (60.9) 6764 (39.1) 1.15 ( ) 57.2 (16.4) Chronic inactive gastritis (5.9) (63.1) (36.9) 1.26 ( ) 57.1 (17.3) Atrophic gastritis 2565 (0.5) 1690 (65.8) 875 (34.1) 1.42 ( ) 68.1 (12.8) Intestinal metaplasia (3.5) (58.6) 7259 (41.4) 1.04 ( ) 63.4 (14.5) s.d., standard deviation. Odds ratios (OR) are age-adjusted; values of more than 1 indicate a greater probability that the condition occurs in women. Table 3 Association of reactive gastropathy with selected conditions of the digestive tract Histopathological diagnosis by organ RG % Normal stomach % OR (95% CI) Patients with oesophageal biopsies (n = ) Normal oesophagus ( ) Eosinophilic oesophagitis ( ) Lymphocytic oesophagitis ( ) Reflux oesophagitis ( ) Barrett s metaplasia, No dysplasia ( ) Barrett s metaplasia, dysplasia ( ) Patients with duodenal biopsies (n = ) Normal duodenum ( ) Peptic duodenitis ( ) Intestinal lymphocytosis ( ) Patients with ileal biopsies (n = 4496) Normal ileum ( ) Active ileitis* ( ) Patients with colon biopsies (n = ) Normal colon ( ) Lymphocytic colitis ( ) Collagenous colitis ( ) Focal active colitis* ( ) Ischaemic colitis ( ) An odds ratio (OR) greater than 1 indicates that the condition is more likely to occur in patients with reactive gastropathy than in those with a normal gastric mucosa. * Patients with a history or diagnosis of inflammatory bowel disease were excluded from these analyses. Associations with gastric and duodenal ulcers The relative frequency of endoscopically detected gastric and duodenal ulcers in patients with normal gastric mucosal histology, reactive gastropathy and H. pylori gastritis is depicted in Table 4. Reactive gastropathy and H. pylori gastritis were similarly associated with gastric ulcers. In contrast, duodenal ulcers, while strongly associated with H. pylori gastritis, showed no significant relationship with reactive gastropathy. Clinical manifestations The most common clinical indications for EGD in patients who subsequently had a histopathological diagnosis of reactive gastropathy were GERD (46% of the patients), epigastric pain (40%), dyspepsia (17%) and dysphagia (16%). Less common presenting manifestations included vomiting (10%), anaemia (8%), diarrhoea (5%) and weight loss (5%). None of these indications differed significantly from those of patients without 740 Aliment Pharmacol Ther 2012; 36:

6 Reactive gastropathy: epidemiology and associations Table 4 Endoscopic gastric and duodenal ulcer in reactive gastropathy and Helicobacter pylori gastritis compared to patients with a normal gastric mucosa Gastric condition Total GU (%) OR (95% CI) DU (%) OR (95% CI) Normal gastric mucosa (1.2) (0.7) 1 Reactive gastropathy (4.8) 4.10 ( ) 585 (0.7) 1.08 ( ) H. pylori gastritis (5.1) 4.33 ( ) 1147 (2.2) 3.26 ( ) An odds ratio greater than 1 indicates that the condition is more likely to occur in patients with reactive gastropathy or H. pylori gastritis than in those with a normal gastric mucosa. reactive gastropathy. NSAID intake was rarely mentioned in the endoscopy report or the pathology requisition; however, there was a small but significant difference between patients with and without reactive gastropathy (0.5% vs. 0.4%; OR % CI ). A total of 4959 patients (1.0%) had a history of gastric ulcer listed in the clinical indications for EGD; 1146 of these (23.1%) had reactive gastropathy, 375 (7.6%) had H. pylori gastritis and 411 (8.3%) a normal gastric mucosa. A history of duodenal ulcer was reported in 878 patients: 141 (16.0%) had reactive gastropathy, 95 (10.8%) had H. pylori infection and 121 (13.8%) had a normal gastric mucosa. DISCUSSION This is the first large-scale study addressing the prevalence and associations of reactive gastropathy in the United States. Histopathological changes that met our criteria for the diagnosis of reactive gastropathy were detected in 15% of more than patients who had gastric biopsies from non-operated stomachs in an ambulatory setting during the 4-year period of this study. The prevalence of reactive gastropathy increased steadily with age, from 2% in children younger than 10 years to slightly over 20% in patients older than 80 years. In contrast to Helicobacter pylori infection that showed substantial geographical variations within the United States (Table 1), the prevalence of reactive gastropathy was relatively uniform across the country, suggesting that the causes of these histopathological changes are operative at similar levels in large segments of the population. A causal association between NSAIDs use and gastric ulcer, perforation and bleeding has been unequivocally demonstrated In addition to reactive gastropathy, 12, 15, 21 several other conditions of the gastrointestinal tract have been suspected to be associated with chronic NSAIDs use, but causal relationships have remained elusive. These other conditions include small intestinal and colonic epithelial lymphocytosis, 22, 23 peptic duodenitis, active ileitis, focal active colitis and collagenous colitis. 27, 28 We designed this study to test the hypothesis that if reactive gastropathy and these other conditions share a common cause (the use of nonsteroidal anti-inflammatory drugs), they should show some degree of association. Although the retrospective cross-sectional nature of this study does not allow to draw firm conclusions about causality, the associations (direct and indirect) shown in Tables 3 and 4 provide some support for this hypothesis. First, the presence of reactive gastropathy showed a significant inverse association with normal mucosal histology throughout the gastrointestinal tract outside the stomach. Second, some specific positive associations also emerged. After excluding patients with inflammatory bowel disease, those with reactive gastropathy were half as likely as patients with a normal stomach to have unremarkable ileal mucosa and almost twice as likely to have active ileitis, a proposed manifestation of NSAIDinduced injury A significant association was also present with focal active colitis, which has also been 29, 30 linked to NSAID use. Our data also support an indirect association between NSAID consumption and collagenous colitis, a relationship that has been previously described In contrast, neither lymphocytic nor ischaemic colitis was significantly associated with reactive gastropathy. Patients with reactive gastropathy were also more likely than controls to exhibit peptic duodenitis and duodenal intraepithelial lymphocytosis, two of the duodenal findings purported to be induced by NSAIDs. This association persisted even in the absence of H. pylori, which was also independently associated with these two conditions. The endoscopic presence of a duodenal ulcer, however, was strongly related to H. pylori gastritis, but not to reactive gastropathy. The associations between reactive gastropathy and oesophageal histopathological conditions were perhaps Aliment Pharmacol Ther 2012; 36:

7 I. Maguilnik et al. the most intriguing. Patients with reactive gastropathy were twice as likely as controls to have a diagnosis of lymphocytic oesophagitis, a recently described entity of unknown aetiology associated with dysphagia and characterised histologically by dense peripapillary lymphocytic infiltrates of the squamous epithelium. 34 Eosinophilic oesophagitis was half as common in patients with reactive gastropathy as compared to controls. This inverse relationship was even more pronounced than that reported with H. pylori gastritis. 35 A possible explanation would be that use of NSAID sufficient to cause reactive gastropathy could reduce or suppress the eosinophilic responses that lead to eosinophilic oesophagitis. No data, however, are available on the relationship between NSAIDs and eosinophils. Another, still more speculative explanation, is that the progressive dietary adjustments made by patients with eosinophilic oesophagitis over the course of their condition (including smaller, more frequent meals) could be less likely to induce duodenogastric reflux that would contribute to the development of reactive changes in the gastric mucosa. Both reflux oesophagitis and Barrett s mucosa, with or without dysplasia, were diagnosed more frequently in our patients with reactive gastropathy than in controls. Dixon et al. have suggested that the increased prevalence of reactive gastropathy (indicated by a high bile reflux index) found in patients with Barrett s oesophagus could be interpreted as evidence that bile reflux injury plays a role in the aetiology of columnar metaplasia of the distal oesophagus. 36 As no relationship is known to exist between Barrett s mucosa and NSAIDs, our data provide further support for this hypothesis. We have also analysed the presentations of patients who had a histopathological diagnosis of reactive gastropathy. Although a history of gastric ulcer was rarely reported, it was more common in patients with reactive gastropathy than in those without; a history of duodenal ulcer was similarly uncommon in patients with and without reactive gastropathy. None of the usual symptoms leading to EGD, including reflux, dysphagia, dyspepsia, nausea and vomiting, epigastric pain, weight loss, anaemia or diarrhoea, showed a positive association with reactive gastropathy. These findings are not surprising: pathological changes of the gastric mucosa, even when histologically impressive, tend not to be associated with subjective clinical manifestations. A case in point is represented by H. pylori gastritis, which is not unequivocally 37, 38 associated with dyspepsia. Although the clinical and histopathological associations that emerged from this study must be interpreted cautiously because of a potential selection bias related to the locations of the gastric mucosa that were sampled as well as to the clinicians decisions to take biopsy specimens from other organs (oesophagus and duodenum) and to perform a colonoscopy, our study of a population of more than half a million individual patients shows that reactive gastropathy nowadays constitutes the most commonly found histopathological change in the gastric mucosa. The prevalence of reactive gastropathy is characterised by a clear age-dependent rise, it is significantly higher in women, and shows no associations with geographical residence within the United States. The associations of reactive gastropathy with other histopathological changes of the gastrointestinal tract support the contention that NSAID use and bile reflux may contribute to its occurrence. A large fraction of reactive gastropathy remains unexplained by such known risk factors; however, its frequent occurrence merits further efforts at elucidating its still mysterious aetiology. ACKNOWLEDGEMENTS Declaration of personal interests: Dr Neumann s gastrointestinal pathology fellowship was partially funded by Miraca Life Sciences, Irving TX. Dr Genta is an employee of Miraca Life Sciences, Irving TX. Declaration of funding interests: None. REFERENCES 1. Bensley EH. Alexis St Martin. Can Med Assoc J 1959; 80: Cotton PB, Rosenberg MT, Axon AT, et al. Diagnostic yield of fibre-optic endoscopy in the operated stomach. Br J Surg 1973; 60: Krentz K. Gastroscopic and aspiration biopsy findings on the operated stomach. Dtsch Med Wochenschr 1964; 89: Lechner HJ, Anton D. The histology of gastric mucosa in B II-stomach. Z Gastroenterol 1979; 17: Mosimann F, Sorgi M, Wolverson RL, Alexander-Williams J, Donovan IA, Harding LK. Bile reflux after duodenal ulcer surgery. A study of 114 asymptomatic and symptomatic patients. Scand J Gastroenterol Suppl 1984; 92: Langhans P, Bues M, Bunte H. Morphological changes in the operated stomach under the influence of duodenogastric reflux. Clinical followup over 20 years. Scand J Gastroenterol Suppl 1984; 92: Langhans P, Heger RA, Stegemann B. The cancer risk in the stomach subjected to nonresecting procedures. An experimental long-term study. 742 Aliment Pharmacol Ther 2012; 36:

8 Reactive gastropathy: epidemiology and associations Scand J Gastroenterol Suppl 1984; 92: Offerhaus GJ, Stadt J, Huibregtse K, Tytgat G. Endoscopic screening for malignancy in the gastric remnant: the clinical significance of dysplasia in gastric mucosa. J Clin Pathol 1984; 37: Offerhaus GJ, Huibregtse K, de BJ, et al. The operated stomach: a premalignant condition? A prospective endoscopic follow-up study. Scand J Gastroenterol 1984; 19: Dewar EP, Dixon MF, Johnston D. Bile reflux and degree of gastritis after highly selective vagotomy, truncal vagotomy, and partial gastrectomy for duodenal ulcer. World J Surg 1983; 7: Dewar P, Dixon MF, Johnston D. Bile reflux and degree of gastritis in patients with gastric ulcer: before and after operation. J Surg Res 1984; 37: Dixon MF, O Connor HJ, Axon AT, King RF, Johnston D. Reflux gastritis: distinct histopathological entity? J Clin Pathol 1986; 39: Sobala GM, O Connor HJ, Dewar EP, King RF, Axon AT, Dixon MF. Bile reflux and intestinal metaplasia in gastric mucosa. J Clin Pathol 1993; 46: Dixon MF, Genta RM, Yardley JH, Correa P. Classification and grading of gastritis. The updated Sydney System. International Workshop on the Histopathology of Gastritis, Houston Am J Surg Pathol 1996; 20: Genta RM. Differential diagnosis of reactive gastropathy. Semin Diagn Pathol 2005; 22: Genta RM, Kinsey RS, Singhal A, Suterwala S. Gastric foveolar metaplasia and gastric heterotopia in the duodenum: no evidence of an etiologic role for Helicobacter pylori. Hum Pathol 2010; 41: Carmack SW, Lash RH, Gulizia JM, Genta RM. Lymphocytic disorders of the gastrointestinal tract: a review for the practicing pathologist. Adv Anat Pathol 2009; 16: Cryer B, Feldman M. Peptic ulcer disease in the elderly. Semin Gastrointest Dis 1994; 5: Graham DY. The relationship between nonsteroidal anti-inflammatory drug use and peptic ulcer disease. Gastroenterol Clin North Am 1990; 19: Griffin MR, Ray WA, Schaffner W. Nonsteroidal anti-inflammatory drug use and death from peptic ulcer in elderly persons. Ann Intern Med 1988; 109: El-Zimaity HM, Genta RM, Graham DY. Histological features do not define NSAID-induced gastritis. Hum Pathol 1996; 27: Lazenby AJ. Collagenous and lymphocytic colitis. Semin Diagn Pathol 2005; 22: Pardi DS, Ramnath VR, Loftus EV Jr, Tremaine WJ, Sandborn WJ. Lymphocytic colitis: clinical features, treatment, and outcomes. Am J Gastroenterol 2002; 97: Amano Y, Kushiyama Y, Takahashi Y, Ishihara S, Adachi K, Kinoshita Y. Prevalence of ileo-colonic ulcerative lesions in chronic NSAIDs users. Am J Gastroenterol 2004; 99: Lengeling RW, Mitros FA, Brennan JA, Schulze KS. Ulcerative ileitis encountered at ileo-colonoscopy: likely role of nonsteroidal agents. Clin Gastroenterol Hepatol 2003; 1: Smale S, Tibble J, Sigthorsson G, Bjarnason I. Epidemiology and differential diagnosis of NSAID-induced injury to the mucosa of the small intestine. Best Pract Res Clin Gastroenterol 2001; 15: Giardiello FM, Hansen FC III, Lazenby AJ, et al. Collagenous colitis in setting of nonsteroidal antiinflammatory drugs and antibiotics. Dig Dis Sci 1990; 35: Kakar S, Pardi DS, Burgart LJ. Colonic ulcers accompanying collagenous colitis: implication of nonsteroidal antiinflammatory drugs. Am J Gastroenterol 2003; 98: Greenson JK, Stern RA, Carpenter SL, Barnett JL. The clinical significance of focal active colitis. Hum Pathol 1997; 28: Goldstein NS, Cinenza AN. The histopathology of nonsteroidal anti-inflammatory drug-associated colitis. Am J Clin Pathol 1998; 110: Chande N, Driman DK, Reynolds RP. Collagenous colitis and lymphocytic colitis: patient characteristics and clinical presentation. Scand J Gastroenterol 2005; 40: Filik L. Collagenous colitis and nonsteroidal anti-inflammatory drugs. Eur J Intern Med 2010; 21: e Milman N, Kraag G. NSAID-induced collagenous colitis. J Rheumatol 2010; 37: Haque S, Genta RM. Lymphocytic oesophagitis: clinicopathological aspects of an emerging condition. Gut 2012; 61: Dellon ES, Peery AF, Shaheen NJ, et al. Inverse association of esophageal eosinophilia with Helicobacter pylori based on analysis of a US pathology database. Gastroenterology 2011; 141: Dixon MF, Neville PM, Mapstone NP, Moayyedi P, Axon AT. Bile reflux gastritis and Barrett s oesophagus: further evidence of a role for duodenogastro-oesophageal reflux? Gut 2001; 49: Ford AC, Moayyedi P, Jarbol DE, Logan RF, Delaney BC. Meta-analysis: Helicobacter pylori test and treat compared with empirical acid suppression for managing dyspepsia. Aliment Pharmacol Ther 2008; 28: Moayyedi P, Soo S, Deeks J, et al. Eradication of Helicobacter pylori for non-ulcer dyspepsia. Cochrane Database Syst Rev 2006; 2: CD Aliment Pharmacol Ther 2012; 36: