- Email: [email protected]
Changes in the Gastric Mucosa With Aging
Accepted Manuscript Changes in the Gastric Mucosa with Aging Amnon Sonnenberg, M.D., M.Sc., Robert M. Genta, M.D.
PII: DOI: Reference:
S1542-3565(15)00159-7 10.1016/j.cgh.2015.02.020 YJCGH 54179
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 18 February 2015 Please cite this article as: Sonnenberg A, Genta RM, Changes in the Gastric Mucosa with Aging, Clinical Gastroenterology and Hepatology (2015), doi: 10.1016/j.cgh.2015.02.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. All studies published in Clinical Gastroenterology and Hepatology are embargoed until 3PM ET of the day they are published as corrected proofs on-line. Studies cannot be publicized as accepted manuscripts or uncorrected proofs.
ACCEPTED MANUSCRIPT Sonnenberg 1 CGH-D-14-01737.R1 (marked copy)
RI PT
Changes in the Gastric Mucosa with Aging
M AN U
NUMBER OF WORDS: text 2438; abstract 195 NUMBER OF FIGURES: 3 NUMBER OF TABLES: 2 RUNNING HEAD: Aging Stomach OF INTERESTS:
RM Genta is employed by Miraca Life Sciences, Irving, TX.
TE D
CONFLICTS
SC
Amnon Sonnenberg, M.D., M.Sc. 1,2, Robert M. Genta, M.D. 1,3
A Sonnenberg declares no relevant conflicts of interest. No funding was obtained for this study. AUTHOR CONTRIBUTIONS: Both authors contributed equally to conception and design, analysis
EP
and interpretation of data, and writing of manuscript.
2
1
Miraca Life Sciences, Research Institute, Irving, Texas;
AC C
INSTITUTIONAL AFFILIATIONS:
Department of Medicine, Oregon Health & Sciences University, Portland, Oregon; 3University
of Texas Southwestern Medical Center, Dallas, Texas ADDRESS FOR CORRESPONDENCE: Amnon Sonnenberg, M.D., M.Sc., Gastroenterology, Portland VA Medical Center P3-GI, 3710 SW U.S. Veterans Hospital Road, Portland, Oregon 97239. Phone: 503-220-8262, ext. 56679, Fax: 503-220-3426, E-mail: [email protected]
ACCEPTED MANUSCRIPT Sonnenberg 2 ABSTRACT Background & Aims: We aimed to characterize age-related changes in the gastric mucosa and
RI PT
investigate the contribution of Helicobacter pylori infection to these changes. Methods: We collected data from a large national pathology database of 895,323 subjects who underwent esophago-gastro-duodenoscopy (EGD) with gastric biopsies from January 2008
SC
through December 2013 at endoscopy centers throughout the US. The prevalence of various types of gastric histopathology was expressed as percent of the total study population, stratified
M AN U
by sex, age, and US state.
Results: Over a lifetime, the gastric mucosa became abnormal in 50% of subjects. A higher proportion of people in areas with a high prevalence of H pylori infection acquired gastric abnormalities. H pylori-associated chronic active gastritis and mucosal changes secondary to infection were observed in 22% of biopsies; these were the most common gastric abnormalities
TE D
observed. Reactive gastropathy, detected in 18% of biopsies, increased with age and was the second most common factor of gastric pathology observed. Conclusions: Based on an analysis of biopsies collected by EGD in the US, gastric abnormalities
EP
increase with age. Most pathologies detected by histologic analysis are caused by H pylori
AC C
infection, but the causes of many others are unknown.
Keywords: epidemiology; gastric atrophy; gastritis; Helicobacter pylori; intestinal metaplasia.
ACCEPTED MANUSCRIPT Sonnenberg 3 INTRODUCTION In the past, gastritis was considered a normal development associated with aging of the stomach.
RI PT
The discovery of H. pylori has revealed that a large portion of gastritis was secondary to infection with this specific organism and that some of the gastritis became reversible once the infection was eradicated. [1] Other unknown factors still contribute the age – related changes of
SC
the gastric mucosa. Other micro-organisms besides H. pylori have been implicated as potential causes for the occurrence of H. pylori – negative gastritis. [2] A lifelong exposure to harmful
M AN U
environmental agents may also damage the gastric mucosa. Non-steroidal anti-inflammatory drugs are known to cause erosions and affect the proliferative balance of the gastric epithelium, resulting in the constellation of mucosal changes known as reactive gastropathy. [3-4] Besides its acute effects on mucosal integrity, high consumption of dietary salt has been established as a risk factor for gastric ulcer, as well as gastric cancer. [5-7] Apart from these few well-characterized
TE D
external damaging factors, however, little is known about how other nutritional toxins, allergens, or medications might chronically affect the gastric mucosa. The prevalence of H. pylori in the general population has shown a continuous decline during the past five decades. [8] No recent
EP
study has looked at the age-dependent changes of the gastric mucosa in a large representative
AC C
sample of the US population.
Miraca Life Sciences is a centralized pathology laboratory that serves gastroenterologists
distributed throughout the United States. The laboratory receives and examines gastric specimens from over 200,000 patients annually. The results of histopathological evaluation are stored in an electronic database that provides a unique opportunity to study the epidemiology of gastrointestinal disease based on histopathology. The aim of the present study was to use this
ACCEPTED MANUSCRIPT Sonnenberg 4 large database to characterize the age – related changes of the gastric mucosa and establish the contribution of diagnoses associated with H. pylori to such changes.
RI PT
MATERIAL AND METHODS Data Source
SC
We used a large national pathology database of subjects who underwent esophago-gastroduodenoscopy (EGD) with gastric biopsies between January 2008 and December 2013 in
M AN U
endoscopy centers distributed throughout the US. The mucosal biopsy specimens were evaluated and reported by a single group of histopathologists. The group consists of 35 pathologists with subspecialty training in gastrointestinal pathology, who practice in the same environment, use uniform diagnostic criteria and standardized diagnostic codes, and participate in daily consensus conferences where cases and diagnostic criteria are discussed. All data were derived from pre-
TE D
existing records. No direct contact with either patients or providers was made and no individual patient information was revealed. All patient records were de-identified before being analyzed. For these reasons, the study protocol was exempted by the Institutional Review Board from the
AC C
EP
need for informed consent from its participants.
Patient Selection
All patients in the database who had gastric biopsies during the study period and no history of previous gastric surgery were included in the analysis. Besides demographic characteristics, each patient record also contained the clinical and endoscopic diagnoses, as well as a detailed list of all results of the histopathologic examination. Pathologic diagnoses were coded in a pre-defined,
ACCEPTED MANUSCRIPT Sonnenberg 5 standardized, and searchable fashion. The presence of gastric biopsy specimens was the sole criterion for inclusion.
RI PT
Histopathologic Criteria
More than 90% of gastric biopsies in this laboratory were routinely stained with a specific antiHelicobacter monoclonal immunochemical stain (Cell Marque, Rocklin, California, USA). The
SC
remainder was stained with a modified Methylene Blue stain (HP Blue, Anatech, Ltd., Battle Creek, MI). All specimens were also stained with Alcian Blue – Periodic Acid Schiff to enhance
M AN U
the detection of intestinal metaplasia. Intestinal metaplasia was diagnosed in the presence of goblet cells in the gastric mucosa. Gastric biopsy specimens were diagnosed following the guidelines of the Updated Sydney System. [9] Briefly, Helicobacter infection was diagnosed when the characteristic curved organisms were visualized in a gastric biopsy specimen. Chronic
TE D
active Helicobacter-negative gastritis was diagnosed when the lamina propria contained mixed populations of lymphocytes, plasma cells, and polymorphonuclear cells, and the surface of foveolar epithelium was infiltrated by neutrophilic polymorphonuclear cells, but no Helicobacter
EP
organisms could be visualized by the anti-Helicobacter monoclonal immunochemical stain. Chronic inactive gastritis was characterized by focal or diffuse chronic inflammation without
AC C
neutrophilic granulocytes or Helicobacter organisms. Since antral atrophy in the absence of intestinal metaplasia is rarely diagnosed histologically and its relevance is uncertain [9], only the diagnosis of gastric corpus atrophy (defined as the loss of oxyntic glands) was used in this study. Gastric mucosal atrophy was diagnosed by the loss of oxyntic glands. Criteria for reactive gastropathy were based on the 2005 definition, which included various combinations of foveolar hyperplasia, regenerative changes in the surface epithelium, edema or hyperemia of the lamina propria, erosions, and smooth muscle proliferation. [10] A patient was categorized as having a
ACCEPTED MANUSCRIPT Sonnenberg 6 normal gastric mucosa when all specimens, irrespective of their number and location, consisted exclusively of gastric mucosa with no abnormal histopathologic changes of any kind.
RI PT
Statistical Analyses
The prevalence by sex, age, and state was expressed as percent of the total study population stratified accordingly by sex, age, and state. The study population was also stratified by areas
SC
with low and high prevalence of H. pylori positive gastritis. Based on the mean prevalence of H. pylori-positive gastritis in patients from areas with the same ZIP code, patients were allocated
M AN U
into three separate groups. Group 1 contained all patients from ZIP codes with H. pylori prevalence less than 6%, group 2 contained all patients from ZIP codes with H. pylori prevalence between 6% and 12%, and group 3 contained all patients from ZIP codes with H. pylori prevalence greater than 12%. For statistical comparisons, the 95% confidence interval of each
TE D
prevalence rate was calculated using the Poisson distribution. Each two rates were considered significantly different, if their respective confidence intervals did not overlap. The geographic distribution among different states of each two histopathologic entities was compared using
AC C
RESULTS
EP
linear regression analysis.
General characteristics
A total of 895,323 unique patients with EGD were included in the present analysis. Table 1 contains a stratification of the patient population by histopathology versus gender and H. pylori prevalence. Reactive gastropathy was the most common histopathologic diagnosis, followed by H. pylori – positive chronic active gastritis, and chronic inactive gastritis. The upper panel of
ACCEPTED MANUSCRIPT Sonnenberg 7 Figure 1 shows the sex distribution among various types of gastric histopathology. H. pylori positive chronic active gastritis and intestinal metaplasia were both characterized by significant
were all characterized by significant female predominance.
Distribution by Age
RI PT
male predominance, whereas chronic inactive gastritis, gastric atrophy, and reactive gastropathy
SC
The prevalence of normal gastric mucosa decreased with increasing age (Figure 1, lower panel). Although the prevalence of normal gastric mucosa was significantly higher in the group with low
M AN U
H. pylori prevalence, overall, this age-specific decline occurred similarly in both groups. Figure 2 depicts the age-dependent rise in various types of gastric histopathology among subjects from areas with low (upper panel) and high prevalence of H. pylori (lower panel). In both groups reactive gastropathy was the major abnormality found in the aging gastric mucosa.
TE D
In areas with high prevalence of H. pylori, besides H. pylori positive gastritis, intestinal metaplasia and gastric atrophy constituted the other major contributing factors. Although
EP
relatively common, chronic inactive gastritis showed a less pronounced age-dependent rise.
Geographic Distribution
AC C
Figure 3 depicts the prevalence of various types of gastric histopathology in areas with low and high H. pylori prevalence. H. pylori-positive chronic active gastritis, chronic inactive gastritis, intestinal metaplasia, and gastric atrophy were all significantly more common in areas with high than low H. pylori prevalence. The prevalence of H. pylori negative chronic active gastritis and reactive gastropathy appeared unaffected by the underlying prevalence of H. pylori.
ACCEPTED MANUSCRIPT Sonnenberg 8 Table 2 contains the correlations associated with the geographic distribution of various gastric histopathologies among 41 individual states with more than 3,000 patient in the database. Statistically significant correlations with p < 0.05 are highlighted by bold fonts. The geographic
RI PT
distributions of H. pylori-positive chronic active gastritis, chronic inactive gastritis, intestinal metaplasia, and gastric atrophy were correlated with each other, suggesting that the underlying
SC
distribution in the prevalence of H. pylori may be responsible for their similar variations.
M AN U
DISCUSSION
Using a national database of histopathology reports, the present study analyzed age-related changes of the gastric mucosa in a large patient population from the United States. The fraction of patients with normal gastric mucosa dropped by 50% over lifetime, the drop being more pronounced in areas with a high prevalence of H. pylori. H. pylori – positive chronic active
TE D
gastritis and other mucosal changes secondary to H. pylori infection were the most common reasons responsible for such changes. Besides H. pylori – related changes, an age-dependent rise in the occurrence of reactive gastropathy was the second most common factor to result in
EP
increasing gastric histopathology with rising age.
AC C
H. pylori – positive chronic active gastritis, intestinal metaplasia, and gastric mucosal atrophy showed resembling epidemiologic patterns. All three diagnoses were characterized by a similar geographic variation. Intestinal metaplasia and gastric mucosal atrophy were both more frequent in areas with high H. pylori prevalence. Increasing age and the passage of time seem to be the major factors that promote the transition from chronic active gastritis to intestinal metaplasia and gastric mucosal atrophy, but other risk factors must play a role. Although both conditions showed a similar age-dependent rise, the fraction of patients with non-metaplastic
ACCEPTED MANUSCRIPT Sonnenberg 9 gastric atrophy was much smaller than that of patients with intestinal metaplasia. This may reflect, at least in part, the fact that intestinal metaplasia is readily diagnosed even when minimal and limited to a single biopsy specimen, whereas pathologists tend to be cautious in diagnosing
RI PT
atrophic gastritis unless extensive biopsy sampling from both antrum and corpus is available for examination. Nevertheless, our data suggest that many more patients make the transition from chronic active gastritis to intestinal metaplasia than from intestinal metaplasia to gastric atrophy.
SC
Whereas both H. pylori-positive chronic active gastritis and intestinal metaplasia showed a slight, but statistically significant predominance of men, gastric atrophy was significantly more
M AN U
common in women. This suggests that in the US population autoimmune phenomena (more common in women) rather than H. pylori infection (more common in men) may now play an important role in the development of atrophic gastritis.
Characterized by a strikingly similar geographic distribution as H. pylori-positive chronic
TE D
active gastritis, intestinal metaplasia, and gastric mucosal atrophy, the majority of cases with chronic inactive gastritis may also belong into the aforementioned group of histopathologic changes. Several possibilities exist to explain the absence of H. pylori on gastric sections
EP
showing chronic inactive gastritis. [11] Chronic inactive gastritis may represent a relatively short-lived transition point that occurs after the cure (therapeutic or spontaneous) of H. pylori
AC C
infection. In some patients the gastric mucosa reverts to normal, whereas in others chronic gastritis progresses to the more final (and likely irreversible) expressions of intestinal metaplasia and atrophy. Its age-dependent rise, less pronounced than that of intestinal metaplasia or atrophy, may support such contention. The mucosal changes of chronic inactive gastritis may have been initiated by an immune response to H. pylori, but the actual presence of H. pylori is no longer a prerequisite for its presence.
ACCEPTED MANUSCRIPT Sonnenberg 10 Besides various forms of gastritis associated with H. pylori, reactive gastropathy was the second most common type of gastric lesion to show a clear-cut age-dependent rise. Previous authors have suggested that duodenogastric reflux and non-steroidal anti-inflammatory drugs
RI PT
both constitute risk factors for the occurrence of reactive gastropathy or “chemical” gastritis. [9, 12] In our own population, reactive gastropathy was significantly associated with Barrett’s mucosa, duodenitis, duodenal intraepithelial lymphocytosis, active ileitis, focal active colitis, and
SC
collagenous colitis, but none of these associations was strong or affected a large subgroup of patients. [13] In the majority of cases, the occurrence of reactive gastropathy remains
M AN U
unexplained. It may reflect the effect of cumulative exposure of the gastric mucosa to other yet unidentified nutritional or environmental toxins over a prolonged time period. H. pylori-negative chronic active gastritis is relatively rare in the general population. It was not characterized by any clear-cut association with age. Different from H. pylori – positive
TE D
gastritis, H. pylori – negative active gastritis was significantly more common in women than in men. The absence of any correlation with H. pylori – related forms of gastritis speaks against the assumption that this entity could possibly represent instances of missed H. pylori on gastric
EP
biopsies. Previous studies have shown that H. pylori – negative chronic active gastritis is more common in patients with inflammatory bowel disease, especially children. [14-15] This entity
AC C
may reflect a response of the gastric mucosa to other infectious organisms besides H. pylori. In appreciating the results of the present study, several potential limitations of our epidemiologic approach need to be mentioned. In our study, cross-sectional variations in the agedependent prevalence of various types of gastritis were used to generate hypotheses about the natural history and temporal development of gastric histopathology. Ideally, however, one would rather study a group of same patients over a prolonged time and follow the age-dependent
ACCEPTED MANUSCRIPT Sonnenberg 11 changes of the gastric mucosa in individual patients. For multiple ethical, logistic, and financial, reasons, however, such a study would be extremely difficult to execute. Although we used the Helicobacter monoclonal immunochemical stain to test for the presence of H. pylori, no other
RI PT
immune stains or molecular techniques were available to test for the presence of other bacterial organisms. Because of our reliance on a pathology database, no detailed information about socioeconomic status, clinical presentation, or previous exposure to other risk factors was available.
SC
Because the data were generated in subjects who seek medical attention for some type of illness, our population was biased towards subjects with upper gastrointestinal symptoms. Similarly,
M AN U
endoscopists may have been more inclined to take biopsies in patients whose stomach showed some macroscopic abnormality. Accordingly, the prevalence of gastritis might have been smaller if one could have selected a completely random group of subjects from the general population. In summary, our study shows an age-dependent rise in the occurrence of gastric
TE D
histopathology. The largest fraction of such pathology was associated with sequelae of H. pylori infection. Reactive gastropathy represented the second most common cause for the occurrence of age-dependent mucosal alterations. Such age variations may stem from mucosal transition
EP
through different disease stages over time, as well as the cumulative damage experienced by gastric mucosa exposed to various noxious environmental influences. By stating "senectus ipsa
AC C
est morbus" (age itself is the disease) Publio Terenzio Afro was probably wrong [16]: stomach aging recapitulates past histories of outside diseases, and many of them are still unexplained. Future studies should be aimed at elucidating the nature of such environmental risk factors and the role of other microorganisms besides H. pylori.
ACCEPTED MANUSCRIPT Sonnenberg 12 REFERENCES 1. Saltzman JR, Russell RM. The aging gut. Nutritional issues. Gastroenterol Clin North Am
RI PT
1998;27:309-24. 2. Bik EM, Eckburg PB, Gill SR, et al. Molecular analysis of the bacterial microbiota in the human stomach. Proc Natl Acad Sci USA 2006;103:732-7.
Pract Res Clin Gastroenterol 2009;23:805-19.
SC
3. Salles N. Is stomach spontaneously ageing? Pathophysiology of the ageing stomach. Best
M AN U
4. Gibson JA, Odze RD. Pathology of diseases that cause upper gastrointestinal tract bleeding. Gastrointest Endosc Clin N Am 2011;21:583-96.
5. Coton T, Mallaret C, Coilliot C, Carré D, Guisset M. Severe acute ulcerated gastritis induced by salt. Presse Med 2009;38:499-500.
6. Sonnenberg A. Dietary salt and gastric ulcer. Gut 1986;27:1138-42.
Res 2014;159:83-95.
TE D
7. D'Elia L, Galletti F, Strazzullo P. Dietary salt intake and risk of gastric cancer. Cancer Treat
8. Sonnenberg A. Historic changes of Helicobacter pylori-associated diseases. Aliment
EP
Pharmacol Ther 2013;36:329-42.
AC C
9. 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 1994. Am J Surg Pathol 1996;20:1161-81.
10. Genta RM. Differential diagnosis of reactive gastropathy. Semin Diagn Pathol 2005;22:27383. 11. Genta RM, Lash RH. Editorial: no bugs bugging you? Emerging insights into Helicobacternegative gastritis. Am J Gastroenterol 2013;108:72-4.
ACCEPTED MANUSCRIPT Sonnenberg 13 12. 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;46:235-40. 13. Maguilnik I, Neumann WL, Sonnenberg A, Genta RM. Reactive gastropathy is associated
RI PT
with inflammatory conditions throughout the gastrointestinal tract. Aliment Pharmacol Ther 2012;36:736-43.
14. Sonnenberg A, Genta RM. Low prevalence of Helicobacter pylori infection among patients with
SC
inflammatory bowel disease. Aliment Pharmacol Ther 2012;35:469-76.
15. Genta RM, Sonnenberg A. Non-Helicobacter pylori gastritis is common among pediatric
M AN U
patients with inflammatory bowel disease. Aliment Pharmacol Ther 2012;35:1310-16. 16. Encyclopedia Italiana. Senectus ipsa est morbus. http://www.treccani.it/, last accessed
AC C
EP
TE D
02/02/2015.
ACCEPTED MANUSCRIPT Sonnenberg 14 FIGURE LEGENDS Figure 1. Upper panel: Sex distribution among various types of gastric histopathology. CAG =
RI PT
chronic active gastritis, CIG = chronic inactive gastritis, Hp = H. pylori, IM = intestinal metaplasia, RG = reactive gastropathy. Lower panel: Age dependent decline in the faction of subjects with normal gastric mucosa in areas with low and high prevalence of H. pylori.
SC
Figure 2. Age dependent rise in various types of gastric histopathology in areas with low (upper panel) and high prevalence of H. pylori (lower panel). CAG = chronic active gastritis, CIG =
M AN U
chronic inactive gastritis, Hp = H. pylori, IM = intestinal metaplasia, RG = reactive gastropathy. Figure 3. Prevalence of various types of gastric histopathology in areas with low and high H. pylori prevalence. CAG = chronic active gastritis, CIG = chronic inactive gastritis, Hp = H.
AC C
EP
TE D
pylori, IM = intestinal metaplasia, RG = reactive gastropathy.
ACCEPTED MANUSCRIPT Sonnenberg 15 Table 1. Patient distribution by histopathology, sex, type, areas of H. pylori prevalence Low Hp prevalence
Mid Hp prevalence
High Hp prevalence
Total
55,613 32,663 25,526 3,619 104,771 8,651 554,907
39,388 19,043 18,313 1,746 54,364 5,177 340,378
10,075 8,677 5,862 776 32,728 2,712 185,047
37,414 23,106 18,989 2,317 79,622 6,560 436,899
47,515 19,924 18,989 2,272 46,796 4,557 273,377
95,004 51,707 43,840 5,365 159,146 13,829 895,323
RI PT
Male
SC
Hp-pos CAG Chronic inactive gastritis Intestinal metaplasia Atrophy Reactive gastropathy Hp-neg CAG Total
Female
AC C
EP
TE D
M AN U
Sex unknown in 38 patients. CAG = chronic active gastritis, Hp = H. pylori.
ACCEPTED MANUSCRIPT Sonnenberg 16 Table 2. Correlation matrix of the geographic distribution CIG
IM
Atrophy
1 0.1107 0.103 0.0708 0.2041
1 0.6987 0.0411 -0.2792
1 -0.3783 0.1533
RG
Hp-neg CAG
RI PT
Hp-pos CAG CIG IM Atrophy RG Hp-neg CAG
Hp-pos CAG 1 0.6701 0.4578 0.3535 -0.1599 -0.0037
1 0.1015
1
AC C
EP
TE D
M AN U
SC
For all highlighted correlation coefficients p<0.05. CAG = chronic active gastritis, CIG = chronic inactive gastritis, Hp = H. pylori, IM = intestinal metaplasia, RG = reactive gastropathy.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
PII: DOI: Reference:
S1542-3565(15)00159-7 10.1016/j.cgh.2015.02.020 YJCGH 54179
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 18 February 2015 Please cite this article as: Sonnenberg A, Genta RM, Changes in the Gastric Mucosa with Aging, Clinical Gastroenterology and Hepatology (2015), doi: 10.1016/j.cgh.2015.02.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. All studies published in Clinical Gastroenterology and Hepatology are embargoed until 3PM ET of the day they are published as corrected proofs on-line. Studies cannot be publicized as accepted manuscripts or uncorrected proofs.
ACCEPTED MANUSCRIPT Sonnenberg 1 CGH-D-14-01737.R1 (marked copy)
RI PT
Changes in the Gastric Mucosa with Aging
M AN U
NUMBER OF WORDS: text 2438; abstract 195 NUMBER OF FIGURES: 3 NUMBER OF TABLES: 2 RUNNING HEAD: Aging Stomach OF INTERESTS:
RM Genta is employed by Miraca Life Sciences, Irving, TX.
TE D
CONFLICTS
SC
Amnon Sonnenberg, M.D., M.Sc. 1,2, Robert M. Genta, M.D. 1,3
A Sonnenberg declares no relevant conflicts of interest. No funding was obtained for this study. AUTHOR CONTRIBUTIONS: Both authors contributed equally to conception and design, analysis
EP
and interpretation of data, and writing of manuscript.
2
1
Miraca Life Sciences, Research Institute, Irving, Texas;
AC C
INSTITUTIONAL AFFILIATIONS:
Department of Medicine, Oregon Health & Sciences University, Portland, Oregon; 3University
of Texas Southwestern Medical Center, Dallas, Texas ADDRESS FOR CORRESPONDENCE: Amnon Sonnenberg, M.D., M.Sc., Gastroenterology, Portland VA Medical Center P3-GI, 3710 SW U.S. Veterans Hospital Road, Portland, Oregon 97239. Phone: 503-220-8262, ext. 56679, Fax: 503-220-3426, E-mail: [email protected]
ACCEPTED MANUSCRIPT Sonnenberg 2 ABSTRACT Background & Aims: We aimed to characterize age-related changes in the gastric mucosa and
RI PT
investigate the contribution of Helicobacter pylori infection to these changes. Methods: We collected data from a large national pathology database of 895,323 subjects who underwent esophago-gastro-duodenoscopy (EGD) with gastric biopsies from January 2008
SC
through December 2013 at endoscopy centers throughout the US. The prevalence of various types of gastric histopathology was expressed as percent of the total study population, stratified
M AN U
by sex, age, and US state.
Results: Over a lifetime, the gastric mucosa became abnormal in 50% of subjects. A higher proportion of people in areas with a high prevalence of H pylori infection acquired gastric abnormalities. H pylori-associated chronic active gastritis and mucosal changes secondary to infection were observed in 22% of biopsies; these were the most common gastric abnormalities
TE D
observed. Reactive gastropathy, detected in 18% of biopsies, increased with age and was the second most common factor of gastric pathology observed. Conclusions: Based on an analysis of biopsies collected by EGD in the US, gastric abnormalities
EP
increase with age. Most pathologies detected by histologic analysis are caused by H pylori
AC C
infection, but the causes of many others are unknown.
Keywords: epidemiology; gastric atrophy; gastritis; Helicobacter pylori; intestinal metaplasia.
ACCEPTED MANUSCRIPT Sonnenberg 3 INTRODUCTION In the past, gastritis was considered a normal development associated with aging of the stomach.
RI PT
The discovery of H. pylori has revealed that a large portion of gastritis was secondary to infection with this specific organism and that some of the gastritis became reversible once the infection was eradicated. [1] Other unknown factors still contribute the age – related changes of
SC
the gastric mucosa. Other micro-organisms besides H. pylori have been implicated as potential causes for the occurrence of H. pylori – negative gastritis. [2] A lifelong exposure to harmful
M AN U
environmental agents may also damage the gastric mucosa. Non-steroidal anti-inflammatory drugs are known to cause erosions and affect the proliferative balance of the gastric epithelium, resulting in the constellation of mucosal changes known as reactive gastropathy. [3-4] Besides its acute effects on mucosal integrity, high consumption of dietary salt has been established as a risk factor for gastric ulcer, as well as gastric cancer. [5-7] Apart from these few well-characterized
TE D
external damaging factors, however, little is known about how other nutritional toxins, allergens, or medications might chronically affect the gastric mucosa. The prevalence of H. pylori in the general population has shown a continuous decline during the past five decades. [8] No recent
EP
study has looked at the age-dependent changes of the gastric mucosa in a large representative
AC C
sample of the US population.
Miraca Life Sciences is a centralized pathology laboratory that serves gastroenterologists
distributed throughout the United States. The laboratory receives and examines gastric specimens from over 200,000 patients annually. The results of histopathological evaluation are stored in an electronic database that provides a unique opportunity to study the epidemiology of gastrointestinal disease based on histopathology. The aim of the present study was to use this
ACCEPTED MANUSCRIPT Sonnenberg 4 large database to characterize the age – related changes of the gastric mucosa and establish the contribution of diagnoses associated with H. pylori to such changes.
RI PT
MATERIAL AND METHODS Data Source
SC
We used a large national pathology database of subjects who underwent esophago-gastroduodenoscopy (EGD) with gastric biopsies between January 2008 and December 2013 in
M AN U
endoscopy centers distributed throughout the US. The mucosal biopsy specimens were evaluated and reported by a single group of histopathologists. The group consists of 35 pathologists with subspecialty training in gastrointestinal pathology, who practice in the same environment, use uniform diagnostic criteria and standardized diagnostic codes, and participate in daily consensus conferences where cases and diagnostic criteria are discussed. All data were derived from pre-
TE D
existing records. No direct contact with either patients or providers was made and no individual patient information was revealed. All patient records were de-identified before being analyzed. For these reasons, the study protocol was exempted by the Institutional Review Board from the
AC C
EP
need for informed consent from its participants.
Patient Selection
All patients in the database who had gastric biopsies during the study period and no history of previous gastric surgery were included in the analysis. Besides demographic characteristics, each patient record also contained the clinical and endoscopic diagnoses, as well as a detailed list of all results of the histopathologic examination. Pathologic diagnoses were coded in a pre-defined,
ACCEPTED MANUSCRIPT Sonnenberg 5 standardized, and searchable fashion. The presence of gastric biopsy specimens was the sole criterion for inclusion.
RI PT
Histopathologic Criteria
More than 90% of gastric biopsies in this laboratory were routinely stained with a specific antiHelicobacter monoclonal immunochemical stain (Cell Marque, Rocklin, California, USA). The
SC
remainder was stained with a modified Methylene Blue stain (HP Blue, Anatech, Ltd., Battle Creek, MI). All specimens were also stained with Alcian Blue – Periodic Acid Schiff to enhance
M AN U
the detection of intestinal metaplasia. Intestinal metaplasia was diagnosed in the presence of goblet cells in the gastric mucosa. Gastric biopsy specimens were diagnosed following the guidelines of the Updated Sydney System. [9] Briefly, Helicobacter infection was diagnosed when the characteristic curved organisms were visualized in a gastric biopsy specimen. Chronic
TE D
active Helicobacter-negative gastritis was diagnosed when the lamina propria contained mixed populations of lymphocytes, plasma cells, and polymorphonuclear cells, and the surface of foveolar epithelium was infiltrated by neutrophilic polymorphonuclear cells, but no Helicobacter
EP
organisms could be visualized by the anti-Helicobacter monoclonal immunochemical stain. Chronic inactive gastritis was characterized by focal or diffuse chronic inflammation without
AC C
neutrophilic granulocytes or Helicobacter organisms. Since antral atrophy in the absence of intestinal metaplasia is rarely diagnosed histologically and its relevance is uncertain [9], only the diagnosis of gastric corpus atrophy (defined as the loss of oxyntic glands) was used in this study. Gastric mucosal atrophy was diagnosed by the loss of oxyntic glands. Criteria for reactive gastropathy were based on the 2005 definition, which included various combinations of foveolar hyperplasia, regenerative changes in the surface epithelium, edema or hyperemia of the lamina propria, erosions, and smooth muscle proliferation. [10] A patient was categorized as having a
ACCEPTED MANUSCRIPT Sonnenberg 6 normal gastric mucosa when all specimens, irrespective of their number and location, consisted exclusively of gastric mucosa with no abnormal histopathologic changes of any kind.
RI PT
Statistical Analyses
The prevalence by sex, age, and state was expressed as percent of the total study population stratified accordingly by sex, age, and state. The study population was also stratified by areas
SC
with low and high prevalence of H. pylori positive gastritis. Based on the mean prevalence of H. pylori-positive gastritis in patients from areas with the same ZIP code, patients were allocated
M AN U
into three separate groups. Group 1 contained all patients from ZIP codes with H. pylori prevalence less than 6%, group 2 contained all patients from ZIP codes with H. pylori prevalence between 6% and 12%, and group 3 contained all patients from ZIP codes with H. pylori prevalence greater than 12%. For statistical comparisons, the 95% confidence interval of each
TE D
prevalence rate was calculated using the Poisson distribution. Each two rates were considered significantly different, if their respective confidence intervals did not overlap. The geographic distribution among different states of each two histopathologic entities was compared using
AC C
RESULTS
EP
linear regression analysis.
General characteristics
A total of 895,323 unique patients with EGD were included in the present analysis. Table 1 contains a stratification of the patient population by histopathology versus gender and H. pylori prevalence. Reactive gastropathy was the most common histopathologic diagnosis, followed by H. pylori – positive chronic active gastritis, and chronic inactive gastritis. The upper panel of
ACCEPTED MANUSCRIPT Sonnenberg 7 Figure 1 shows the sex distribution among various types of gastric histopathology. H. pylori positive chronic active gastritis and intestinal metaplasia were both characterized by significant
were all characterized by significant female predominance.
Distribution by Age
RI PT
male predominance, whereas chronic inactive gastritis, gastric atrophy, and reactive gastropathy
SC
The prevalence of normal gastric mucosa decreased with increasing age (Figure 1, lower panel). Although the prevalence of normal gastric mucosa was significantly higher in the group with low
M AN U
H. pylori prevalence, overall, this age-specific decline occurred similarly in both groups. Figure 2 depicts the age-dependent rise in various types of gastric histopathology among subjects from areas with low (upper panel) and high prevalence of H. pylori (lower panel). In both groups reactive gastropathy was the major abnormality found in the aging gastric mucosa.
TE D
In areas with high prevalence of H. pylori, besides H. pylori positive gastritis, intestinal metaplasia and gastric atrophy constituted the other major contributing factors. Although
EP
relatively common, chronic inactive gastritis showed a less pronounced age-dependent rise.
Geographic Distribution
AC C
Figure 3 depicts the prevalence of various types of gastric histopathology in areas with low and high H. pylori prevalence. H. pylori-positive chronic active gastritis, chronic inactive gastritis, intestinal metaplasia, and gastric atrophy were all significantly more common in areas with high than low H. pylori prevalence. The prevalence of H. pylori negative chronic active gastritis and reactive gastropathy appeared unaffected by the underlying prevalence of H. pylori.
ACCEPTED MANUSCRIPT Sonnenberg 8 Table 2 contains the correlations associated with the geographic distribution of various gastric histopathologies among 41 individual states with more than 3,000 patient in the database. Statistically significant correlations with p < 0.05 are highlighted by bold fonts. The geographic
RI PT
distributions of H. pylori-positive chronic active gastritis, chronic inactive gastritis, intestinal metaplasia, and gastric atrophy were correlated with each other, suggesting that the underlying
SC
distribution in the prevalence of H. pylori may be responsible for their similar variations.
M AN U
DISCUSSION
Using a national database of histopathology reports, the present study analyzed age-related changes of the gastric mucosa in a large patient population from the United States. The fraction of patients with normal gastric mucosa dropped by 50% over lifetime, the drop being more pronounced in areas with a high prevalence of H. pylori. H. pylori – positive chronic active
TE D
gastritis and other mucosal changes secondary to H. pylori infection were the most common reasons responsible for such changes. Besides H. pylori – related changes, an age-dependent rise in the occurrence of reactive gastropathy was the second most common factor to result in
EP
increasing gastric histopathology with rising age.
AC C
H. pylori – positive chronic active gastritis, intestinal metaplasia, and gastric mucosal atrophy showed resembling epidemiologic patterns. All three diagnoses were characterized by a similar geographic variation. Intestinal metaplasia and gastric mucosal atrophy were both more frequent in areas with high H. pylori prevalence. Increasing age and the passage of time seem to be the major factors that promote the transition from chronic active gastritis to intestinal metaplasia and gastric mucosal atrophy, but other risk factors must play a role. Although both conditions showed a similar age-dependent rise, the fraction of patients with non-metaplastic
ACCEPTED MANUSCRIPT Sonnenberg 9 gastric atrophy was much smaller than that of patients with intestinal metaplasia. This may reflect, at least in part, the fact that intestinal metaplasia is readily diagnosed even when minimal and limited to a single biopsy specimen, whereas pathologists tend to be cautious in diagnosing
RI PT
atrophic gastritis unless extensive biopsy sampling from both antrum and corpus is available for examination. Nevertheless, our data suggest that many more patients make the transition from chronic active gastritis to intestinal metaplasia than from intestinal metaplasia to gastric atrophy.
SC
Whereas both H. pylori-positive chronic active gastritis and intestinal metaplasia showed a slight, but statistically significant predominance of men, gastric atrophy was significantly more
M AN U
common in women. This suggests that in the US population autoimmune phenomena (more common in women) rather than H. pylori infection (more common in men) may now play an important role in the development of atrophic gastritis.
Characterized by a strikingly similar geographic distribution as H. pylori-positive chronic
TE D
active gastritis, intestinal metaplasia, and gastric mucosal atrophy, the majority of cases with chronic inactive gastritis may also belong into the aforementioned group of histopathologic changes. Several possibilities exist to explain the absence of H. pylori on gastric sections
EP
showing chronic inactive gastritis. [11] Chronic inactive gastritis may represent a relatively short-lived transition point that occurs after the cure (therapeutic or spontaneous) of H. pylori
AC C
infection. In some patients the gastric mucosa reverts to normal, whereas in others chronic gastritis progresses to the more final (and likely irreversible) expressions of intestinal metaplasia and atrophy. Its age-dependent rise, less pronounced than that of intestinal metaplasia or atrophy, may support such contention. The mucosal changes of chronic inactive gastritis may have been initiated by an immune response to H. pylori, but the actual presence of H. pylori is no longer a prerequisite for its presence.
ACCEPTED MANUSCRIPT Sonnenberg 10 Besides various forms of gastritis associated with H. pylori, reactive gastropathy was the second most common type of gastric lesion to show a clear-cut age-dependent rise. Previous authors have suggested that duodenogastric reflux and non-steroidal anti-inflammatory drugs
RI PT
both constitute risk factors for the occurrence of reactive gastropathy or “chemical” gastritis. [9, 12] In our own population, reactive gastropathy was significantly associated with Barrett’s mucosa, duodenitis, duodenal intraepithelial lymphocytosis, active ileitis, focal active colitis, and
SC
collagenous colitis, but none of these associations was strong or affected a large subgroup of patients. [13] In the majority of cases, the occurrence of reactive gastropathy remains
M AN U
unexplained. It may reflect the effect of cumulative exposure of the gastric mucosa to other yet unidentified nutritional or environmental toxins over a prolonged time period. H. pylori-negative chronic active gastritis is relatively rare in the general population. It was not characterized by any clear-cut association with age. Different from H. pylori – positive
TE D
gastritis, H. pylori – negative active gastritis was significantly more common in women than in men. The absence of any correlation with H. pylori – related forms of gastritis speaks against the assumption that this entity could possibly represent instances of missed H. pylori on gastric
EP
biopsies. Previous studies have shown that H. pylori – negative chronic active gastritis is more common in patients with inflammatory bowel disease, especially children. [14-15] This entity
AC C
may reflect a response of the gastric mucosa to other infectious organisms besides H. pylori. In appreciating the results of the present study, several potential limitations of our epidemiologic approach need to be mentioned. In our study, cross-sectional variations in the agedependent prevalence of various types of gastritis were used to generate hypotheses about the natural history and temporal development of gastric histopathology. Ideally, however, one would rather study a group of same patients over a prolonged time and follow the age-dependent
ACCEPTED MANUSCRIPT Sonnenberg 11 changes of the gastric mucosa in individual patients. For multiple ethical, logistic, and financial, reasons, however, such a study would be extremely difficult to execute. Although we used the Helicobacter monoclonal immunochemical stain to test for the presence of H. pylori, no other
RI PT
immune stains or molecular techniques were available to test for the presence of other bacterial organisms. Because of our reliance on a pathology database, no detailed information about socioeconomic status, clinical presentation, or previous exposure to other risk factors was available.
SC
Because the data were generated in subjects who seek medical attention for some type of illness, our population was biased towards subjects with upper gastrointestinal symptoms. Similarly,
M AN U
endoscopists may have been more inclined to take biopsies in patients whose stomach showed some macroscopic abnormality. Accordingly, the prevalence of gastritis might have been smaller if one could have selected a completely random group of subjects from the general population. In summary, our study shows an age-dependent rise in the occurrence of gastric
TE D
histopathology. The largest fraction of such pathology was associated with sequelae of H. pylori infection. Reactive gastropathy represented the second most common cause for the occurrence of age-dependent mucosal alterations. Such age variations may stem from mucosal transition
EP
through different disease stages over time, as well as the cumulative damage experienced by gastric mucosa exposed to various noxious environmental influences. By stating "senectus ipsa
AC C
est morbus" (age itself is the disease) Publio Terenzio Afro was probably wrong [16]: stomach aging recapitulates past histories of outside diseases, and many of them are still unexplained. Future studies should be aimed at elucidating the nature of such environmental risk factors and the role of other microorganisms besides H. pylori.
ACCEPTED MANUSCRIPT Sonnenberg 12 REFERENCES 1. Saltzman JR, Russell RM. The aging gut. Nutritional issues. Gastroenterol Clin North Am
RI PT
1998;27:309-24. 2. Bik EM, Eckburg PB, Gill SR, et al. Molecular analysis of the bacterial microbiota in the human stomach. Proc Natl Acad Sci USA 2006;103:732-7.
Pract Res Clin Gastroenterol 2009;23:805-19.
SC
3. Salles N. Is stomach spontaneously ageing? Pathophysiology of the ageing stomach. Best
M AN U
4. Gibson JA, Odze RD. Pathology of diseases that cause upper gastrointestinal tract bleeding. Gastrointest Endosc Clin N Am 2011;21:583-96.
5. Coton T, Mallaret C, Coilliot C, Carré D, Guisset M. Severe acute ulcerated gastritis induced by salt. Presse Med 2009;38:499-500.
6. Sonnenberg A. Dietary salt and gastric ulcer. Gut 1986;27:1138-42.
Res 2014;159:83-95.
TE D
7. D'Elia L, Galletti F, Strazzullo P. Dietary salt intake and risk of gastric cancer. Cancer Treat
8. Sonnenberg A. Historic changes of Helicobacter pylori-associated diseases. Aliment
EP
Pharmacol Ther 2013;36:329-42.
AC C
9. 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 1994. Am J Surg Pathol 1996;20:1161-81.
10. Genta RM. Differential diagnosis of reactive gastropathy. Semin Diagn Pathol 2005;22:27383. 11. Genta RM, Lash RH. Editorial: no bugs bugging you? Emerging insights into Helicobacternegative gastritis. Am J Gastroenterol 2013;108:72-4.
ACCEPTED MANUSCRIPT Sonnenberg 13 12. 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;46:235-40. 13. Maguilnik I, Neumann WL, Sonnenberg A, Genta RM. Reactive gastropathy is associated
RI PT
with inflammatory conditions throughout the gastrointestinal tract. Aliment Pharmacol Ther 2012;36:736-43.
14. Sonnenberg A, Genta RM. Low prevalence of Helicobacter pylori infection among patients with
SC
inflammatory bowel disease. Aliment Pharmacol Ther 2012;35:469-76.
15. Genta RM, Sonnenberg A. Non-Helicobacter pylori gastritis is common among pediatric
M AN U
patients with inflammatory bowel disease. Aliment Pharmacol Ther 2012;35:1310-16. 16. Encyclopedia Italiana. Senectus ipsa est morbus. http://www.treccani.it/, last accessed
AC C
EP
TE D
02/02/2015.
ACCEPTED MANUSCRIPT Sonnenberg 14 FIGURE LEGENDS Figure 1. Upper panel: Sex distribution among various types of gastric histopathology. CAG =
RI PT
chronic active gastritis, CIG = chronic inactive gastritis, Hp = H. pylori, IM = intestinal metaplasia, RG = reactive gastropathy. Lower panel: Age dependent decline in the faction of subjects with normal gastric mucosa in areas with low and high prevalence of H. pylori.
SC
Figure 2. Age dependent rise in various types of gastric histopathology in areas with low (upper panel) and high prevalence of H. pylori (lower panel). CAG = chronic active gastritis, CIG =
M AN U
chronic inactive gastritis, Hp = H. pylori, IM = intestinal metaplasia, RG = reactive gastropathy. Figure 3. Prevalence of various types of gastric histopathology in areas with low and high H. pylori prevalence. CAG = chronic active gastritis, CIG = chronic inactive gastritis, Hp = H.
AC C
EP
TE D
pylori, IM = intestinal metaplasia, RG = reactive gastropathy.
ACCEPTED MANUSCRIPT Sonnenberg 15 Table 1. Patient distribution by histopathology, sex, type, areas of H. pylori prevalence Low Hp prevalence
Mid Hp prevalence
High Hp prevalence
Total
55,613 32,663 25,526 3,619 104,771 8,651 554,907
39,388 19,043 18,313 1,746 54,364 5,177 340,378
10,075 8,677 5,862 776 32,728 2,712 185,047
37,414 23,106 18,989 2,317 79,622 6,560 436,899
47,515 19,924 18,989 2,272 46,796 4,557 273,377
95,004 51,707 43,840 5,365 159,146 13,829 895,323
RI PT
Male
SC
Hp-pos CAG Chronic inactive gastritis Intestinal metaplasia Atrophy Reactive gastropathy Hp-neg CAG Total
Female
AC C
EP
TE D
M AN U
Sex unknown in 38 patients. CAG = chronic active gastritis, Hp = H. pylori.
ACCEPTED MANUSCRIPT Sonnenberg 16 Table 2. Correlation matrix of the geographic distribution CIG
IM
Atrophy
1 0.1107 0.103 0.0708 0.2041
1 0.6987 0.0411 -0.2792
1 -0.3783 0.1533
RG
Hp-neg CAG
RI PT
Hp-pos CAG CIG IM Atrophy RG Hp-neg CAG
Hp-pos CAG 1 0.6701 0.4578 0.3535 -0.1599 -0.0037
1 0.1015
1
AC C
EP
TE D
M AN U
SC
For all highlighted correlation coefficients p<0.05. CAG = chronic active gastritis, CIG = chronic inactive gastritis, Hp = H. pylori, IM = intestinal metaplasia, RG = reactive gastropathy.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT