Chronic Atrophic Gastritis - Clinical Infectious Diseases

1006

Chronic Atrophic Gastritis: Early Diagnosis in a Population Where Helicobacterpylori Infection Is Frequent

Sixto Recavarren-Arce, Robert H. Gilman, From the Gastrointestinal Physiology Working Group* of theUniversidad Peruana Cayetano Heredia; the Ricardo Palma Clinic; andRaul Leon-Barua, Guillermo Salazar, Jeff McDonald,

A. B. PRISMA, Lima, Peru; The Johns Hopkins University, Baltimore,Roberto Lozano, Fernando Diaz,Maryland; and the University of Utah, Salt Lake City, Utah

Alberto Ramirez-Ramos, and Robert Berendson

Chronic atrophic gastritis (CAG) is a premalignant condition characterized by loss of gastricantral deep glands. The histologic changes in antral gastric biopsy specimens from 54 Peruvianpatients with dyspepsia were studied to detail the development and characteristics of CAG. Ninety-six percent of the biopsies revealed severe superficial mucosal inflammation and 89% showed deepinflammation. Moderate or severe CAG was present in 36 (67%) of the 54 patients. In the earlystages of CAG, a glandular lymphoid adherence lesion was noted in 17 (31%) of the 54 biopsyspecimens. This lesion consisted of lymphocytes adherent to the antral deep gland cells and wasassociated with glandular epithelium alterations. The late stage was characterized by small glands,remnants of glands, and gland replacement with a fibrocellular infiltrate or intestinal metaplasia.We propose that the development of CAG probably proceeds via a stereotyped sequence, with anearly deep inflammatory component that may trigger local gland destruction and eventual perma-nent loss.

The progression of pathological lesions to the intestinal type and not to loss of crypts or foveolae in the superficial portionof the mucosa. There are, however, few detailed pathologicalof gastric cancer is not clear. The currently postulated sequence

is superficial gastritis, deep gastritis, chronic atrophic gastritis, descriptions of CAG in the literature. Furthermore, a detailedpathological description of the progression from early to lateintestinal metaplasia, dysplasia, and then cancer [1, 2]. It has

been postulated that in Third World countries this cascade is CAG has not been published, and the mechanism of this lesionremains unknown.initiated and intensified by Helicobacter pylori infection [1, 2].

This idea comes largely from data that have linked the higher In this article, we describe in detail the lesions seen in CAG,including a lymphocytic adherence lesion. The data presentedprevalence of chronic atrophic gastritis (CAG) with early

H. pylori infection [3]. In turn, other epidemiological studies here suggest that the loss of deep glands in CAG, an end-stagelesion, is preceded by a series of progressive inflammatoryhave shown increased risk of gastric cancer in subjects with

CAG [4, 5]. changes in the gastric mucosa. These stages have not been welldescribed in the literature.In Peru H. pylori infection occurs early in life, and 80% of

people are infected with H. pylori by early adulthood [6–8].There are also high rates and early development of CAG inPeruvians [3]. CAG is a premalignant lesion, and in Peru gastric Materials and Methodscancer is one of the most common causes of cancer in both

We examined a series of 54 antral gastric biopsy specimenssexes [9].from middle-class Peruvians referred for dyspepsia. We ob-CAG is usually defined as loss (atrophy) of gastric propertained the specimens from one endoscopist. We later obtainedor antral deep glands [10]. Most pathologists are aware thatthe clinical histories and endoscopy reports. Special care wasgland loss refers only to loss of antral deep or proper glandstaken to obtain large and deep biopsy specimens, and thosefrom the body and cardia were not included. Shallow biopsyspecimens (those not containing muscularis mucosa) were also

This article is part of a series of papers presented at the 2nd International not included, because they are inadequate for evaluation ofWorkshop on Helicobacter pylori Infections in the Developing World, held in deep-gland status and diagnosis of CAG.Lima, Peru, in January 1996.

Several biopsy fragments were taken from each patient via* Other members of the GPWG include Dr. Jose Watanabe, Dr. CarlosRodriguez, Dr. Jaime Cok, Maritza Alvarez, Jose Palomino, and Janet Valdez. endoscope. Each biopsy fragment was fixed in buffered forma-

Reprints: Dr. Sixto Recavarren-Arce, Universidad Peruana Cayetano, Here- lin (pH, 7.2). All were embedded together in paraffin, sectioneddia, Av. Honorio Delgado 430, Rimac, P.O. Box 4314, Lima, Peru.at 4–5 microns, and stained with hematoxylin/eosin and peri-Correspondence: Dr. Robert H. Gilman, Department of International Health,

The Johns Hopkins School of Hygiene, 615 North Wolfe Street, Baltimore, odic acid–schiff (PAS) stains. Sections from 28 patients wereMaryland 21205. also stained with Warthin-Starry silver stain.Clinical Infectious Diseases 1997;25:1006–12 Adequacy of biopsy sections and final histologic diagnosis.q 1997 by The University of Chicago. All rights reserved.1058–4838/97/2505–0008$03.00 Multiple biopsy fragments from the antrum were included in

Dow

nloaded from https://academ

ic.oup.com/cid/article-abstract/25/5/1006/341238 by guest on 09 April 2019

1007CID 1997;25 (November) Chronic Atrophic Gastritis

each paraffin block. Sections were considered adequate for Deep gland loss. Loss of deep glands is considered theessential component of CAG. Lesions comprising CAG wereevaluation of CAG when they fulfilled two conditions: (1) the

histologic sections were oriented from the surface mucinous classified as either early or late. Early CAG is defined as antraldeep gland loss associated with CDG. The CDG extends in aepithelium to the muscularis mucosae [10] and (2) a portion

of the muscularis mucosae was present. The number of sections columnar form between the glands or surrounds and separatesthe glands into islands (pseudoacinar pattern), often with ameeting these criteria was counted, and each such adequate

histologic section was then evaluated for pathological charac- glandular lymphoid adherence lesion and the presence of glandremnants. When only a few PAS-positive gastric glands areteristics.

All the antral fragments from one patient were contained on present without any lumen, they are defined as remnants.Late CAG was characterized by distension of the deep glandsa single slide. These fragments were defined and analyzed as

a patient’s biopsy specimen. They were categorized for the and accentuation of the pseudoacinar pattern. In addition, inlate CAG there are small glands, remnants of glands, and areastype and extent of the most severe lesion, and that lesion deter-

mined the final diagnosis. In order to prevent overreading, where lost glands are replaced with a fibrocellular substrate.Grade of deep gland loss. The extent of deep gland lossmild lesions were not considered diagnostic for a pathological

characteristic. or replacement due to either CAG or CAG with intestinalmetaplasia was arbitrarily expressed as mild (õ10%), moderateMorphological changes. To study morphological changes,

each evaluable histologic section was assessed with use of a (10%–40%), or severe (ú40%) loss of deep antral glands.Intestinal metaplasia. The extent of intestinal metaplasiaspecially designed format that considered the following charac-

teristics. was noted, but special stains were not performed for typingthe metaplasia.Changes in the mucinous surface epithelium. H. pylori in-

fection is associated with a partial or total disappearance ofthe mucinous apical portion of the epithelial cell and distortion

Resultsof its nuclei and basal cytoplasm. These changes have beentermed the destructive mucin lesion of the covering gastric There were 67 patients originally included in the study, of

whom 13 were excluded because their biopsy specimens wereepithelium [1].Inflammatory infiltrate of the superficial portion of the muco- not satisfactory. The mean ({SD) age of the 54 patients with

readable biopsies was 41 { 14 years (range, 16–74 years).sal lamina propria. This lesion is nearly always coincidentwith damage to the mucinous surface epithelium. Infiltration is Twenty-eight patients were male and 24 were female. Age and

gender status were not available for four and two subjects,usually chronic-active, consisting of polymorphonuclear cells(PMNs) and round cells. Occasionally, there is a chronic in- respectively. Of the 47 patients, 38 had no discrete lesions

other than gastritis, 6 had an active gastric ulcer, and 3 anflammatory infiltrate consisting only of mononuclear cells(lymphocytes, plasma cells, and macrophages). The inflamma- active duodenal ulcer. No patient had gastric cancer. For the 54

patients with readable biopsies, there were 6.8 { 1.9 differenttory infiltrate involves the entire interfoveolar space but doesnot extend deeper than the gland neck. The combination of fragments per slide, of which 4.3 { 3.4 were adequate for

examination.damage to the surface mucinous epithelium and superficialinflammatory infiltrate has been called chronic superficial gas- H. pylori was detected in 36 (67%) of the 54 biopsy speci-

mens stained with hematoxylin and eosin and in 23 (82%) oftritis (CSG) and is patchy when caused by H. pylori infection.In Peru CSG is nearly always associated with H. pylori infec- the 28 stained with silver (figure 1).

Changes in the mucinous surface epithelium. Mucinoustion [1, 2, 11]. The whole biopsy specimen was examined forthe presence of CSG. damage (destructive lesion) was present in 44 (81%) of the 54

biopsies (figure 1).Gland neck lesion. Although the superficial inflammatoryinfiltration with PMNs is usually diffuse, often there are intense Superficial inflammation. Nearly all biopsy specimens (52

of 54; 96%) had some evidence of moderate to severe superfi-foci at the gland necks. A gland neck lesion is defined as aPMN collar in and around the gland neck epithelium [12]. cial inflammatory infiltration. In 44 (85%) of these 52, the

lamina propria infiltrate was chronic active (i.e., with PMNs)Chronic deep gastritis (CDG): inflammatory infiltration ofthe deep portion of the lamina propria. This is defined as (figure 2); in 8 (15%), it was of a chronic type only.

Neck lesion. This lesion (figure 3) was always associatedinflammatory infiltrate in the mucosae below the gland necks,without antral gland loss. The presence, type, and extent of with chronic active infiltration in the lamina propria. Neck

lesions were present in the biopsy specimens of 15 (28%) ofinflammatory cells in the deep portion of the mucosa weredetermined. Care was taken to distinguish deep inflammatory the 54 patients and in 33 (17%) of 185 biopsy fragments.

Deep inflammatory infiltration. Moderate to severe deepinfiltrate from lymphoid follicles. Infiltration of the deep mu-cosa is nearly always associated with a superficial infiltrate and inflammatory infiltration was present in 48 (89%) of the 54

biopsy specimens (figure 4). The inflammatory infiltrate be-mucinous cell damage. Lymphoid follicles were noted sincethey appear to be a response to H. pylori infection [11]. tween the antral glands had PMNs and round cells (chronic

Dow



1008 Recavarren-Arce et al. CID 1997;25 (November)

Figure 1. Warthin-Starry staining reveals severe damage to the gas- Figure 2. Hematoxylin and eosin staining shows severe superficialtric covering mucinous epithelium (destructive lesion), associated inflammation in the antral mucosa. Deep proper glands are spared.with numerous Helicobacter pylori bacteria. (Original magnification, Muscularis mucosae at the bottom. (Original magnification, 1150.)1600.)

and pleomorphic lymphoid elements common in deep gastritisactive) in 21 (44%) of 48 and only mononuclear cells (chronic) are not seen in antral gastric lymph nodes.in 27 (55%) of 48. Occasionally, in a single patient’s biopsy, Glandular lymphoid adherence lesion. At the proper deepthe infiltrate was chronic active in one fragment and chronic glands we noted a lesion that consisted of infiltrating lympho-in another. Similarly, in some cases the inflammatory infiltra- cytes, and occasionally eosinophils, that focally distorted thetion was diffuse while in others it was patchy. Eosinophils glands (figure 5); the infiltrating lymphocytes destroyed thewere present in the majority of infiltrates. Deep inflammatory glandular epithelium (figure 6). In severely affected areas, lym-infiltration either occurred in columns between the deep antral phocytes partially or totally destroyed the glands (figure 7). Inglands or, when more severe, separated the glands into islands these fields apposition of lymphocytes and gland cells wasor pseudoacini (figure 4). In addition, inflammatory cells tended observed (figure 8). At further stages the affected glands be-to accumulate next to the muscularis mucosa in a bandlike came smaller than normal glands (figure 9). The glandularformation. At least one lymphoid follicle was noted in 40 (74%) lymphoid adherence lesion was seen in 17 (32%) of the 54of the 54 biopsies. biopsies and in 32 (15%) of 219 biopsy fragments.

The peripheral border of mucosal lymph nodules can appear CAG. Thirty-six of 54 biopsy specimens (67%) hadsimilar to areas of deep inflammatory infiltrate, but eosinophils changes in at least one fragment showing either moderate or

Figure 4. Hematoxylin and eosin staining shows gastritis of thedeep type. Round inflammatory cells infiltrate antral glands separatingFigure 3. Hematoxylin and eosin staining of the neck lesion. Poly-

morphonuclear cells infiltrate one glandular neck (arrow) with a ‘‘col- them. A pseudoacinar pattern is produced. Muscularis mucosae at thebottom. (Original magnification, 1200.)lar’’ pattern. (Original magnification, 1250.)

Dow




Figure 5. Hematoxylin and eosin staining shows the lymphoid ad- Figure 6. A higher magnification (1600) of the hematoxylin andherence lesion. The antral glandular tubes appear focally distorted by eosin stain in figure 5. Focal, severe lymphocytic infiltration producesthe severe lymphocytic cell infiltration (between the four arrowheads). marked antral gland damage; one gland (arrowhead at left) can hardlyNormal glands (n) are seen at left and right of the damaged area. be identified in the area. A few eosinophils are present. At left and(Original magnification, 1250.) right, normal glands (n) limit the lesion.

lymphocytic adherence lesion indicated the presence of initialsevere CAG. The extent of gland loss, gland distension, small

atrophy.glands, and remnants were best seen with PAS staining (figure

Late stage of CAG. The fully developed lesion has small10). The presence of lymphoid adherence lesion and deep gland

glands, gland remnants, and zones of antral gland loss. In late-remnants was used to distinguish loss of glands from deep

stage CAG, the glands are replaced by a fibrocellular substrategastritis without gland loss.

or fibrosis, which largely replaces the inflammatory infiltrate.Early stage of CAG. The early stage of CAG was charac-

Gland loss is more extensive and the pseudoacinar pattern moreterized by a deep inflammatory infiltrate extending between the

accentuated than in early CAG. The remaining intact antralantral glands. This infiltrate occasionally separates the gastric

glands are small and rounded, with a distended lumen, andglands into pseudoacinae. Distinction between CDG without

gland remnants are common (figure 11).deep gland loss and deep gastritis associated with early-stage

CAG lesions are characteristically patchy in distribution. ItCAG in which there is early gland loss is difficult. We found

was not infrequent to see one portion of the section with normalthat the presence of gland remnants and/or the glandular

gland architecture and another affected by CAG.

Figure 7. Hematoxylin and eosin stain of the lymphoid adherencelesion. Infiltrating lymphocytes are focally distorting antral glands. Figure 8. A higher magnification (1600) of the hematoxylin and

eosin stain in figure 7. The partially destroyed antral gland showsPartial destruction (thin arrows) and probably total destruction (thickarrow) of antral glands by infiltrating lymphocytes are seen. Between apposition of lymphocytes with epithelial antral glands (thin arrows).

The gland with probable total destruction (thick arrows) shows cir-muscularis mucosae (mm) and the glands there is a bandlike (bl)accumulation of lymphocytes, with total absence of glands. (Original cumscribed lymphocytic infiltration that appears to be the size and

shape of a gland.magnification, 1300.)

Dow




Figure 9. Hematoxylin and eosin stain of the lymphoid adherent Figure 10. Para-aminosalicylic staining shows deep gastritis andlesion. The lymphocytic cells are in close relationship to a group of early chronic atrophic gastritis. Focal antral gland loss, gland disten-antral glands (between arrows). The affected glands and their cells tion, small glands, and remnants are illustrated. Lymphocytic infiltrateare small in comparison with normal glands (seen at left and right). is severe. (Original magnification, 1400.)Muscularis mucosae (mm) at the bottom. (Original magnification,1400.)

No relationship between CAG and the number or type oflymph follicles was noted, nor were any dysplastic changesseen.

Band lesions. In both early and late stages CAG had abandlike lesion separating the antral gastric cells from the mus-cularis mucosa. In the early stage, this band was composedmostly of lymphocytes accumulating on top of the muscularismucosa (figure 7). In later stages the band is associated withgland loss and then fibrosis (figure 11).

Intestinal metaplasia. Intestinal metaplasia (IM) was pres-ent in 19 (35%) of the 54 biopsy specimens. In many sections

Figure 11. Para-aminosalicylic staining shows a late chronicatrophic gastritis complex. Severe fibrosis separates the distendedantral glands arranged in a pseudoacinar pattern (muscularis mucosa Figure 12. Schema of gastric changes leading to chronic

atrophic gastritis (CAG) in a developing country where H. pylori[mm] at the bottom). Despite advanced lesions, the total thickness ofthe mucosa is not altered. The late stage of the bandlike lesion (bl) infection is highly prevalent. The plus sign means the changes are

progressive.is seen. (Original magnification, 1150.)

Dow




IM appeared only in the superficial epithelium and foveolar, gastritis were not provided. The investigators concluded thatH. pylori infection was a significant risk factor for the develop-suggesting that it may originate in this area and extend through

the whole gland with time. ment of CAG and intestinal metaplasia.Final histologic diagnosis. The final histologic diagnosis There are at least two possible mechanisms for gastric gland

from the antral biopsies for the 54 evaluable patients were as destruction. First, accumulation of PMNs at the gland neckfollows: 2, normal; 6, CSG; 7, CDG; 8, early CAG; 12, late may destroy the gland’s regenerative potential since the neckCAG without IM; and 19, late CAG with IM. The two normal contains the gland’s stem cells [14]. The superficial mucosa’sbiopsy specimens did not have H. pylori infection. As seen in ability to regenerate completely after successful treatment oftable 1, there was a statistically significant correlation between H. pylori in the presence of multiple neck lesions suggests thatlesion severity and age. There was no significant difference destruction of the neck may not be responsible for CAG. Thebetween early and late lesions and no significant trend in sever- second alternative is the lymphoid adherence lesion, whichity of pathologic lesions when patients with peptic ulcers may cause CAG via an autoimmune mechanism that attacks(n Å 9) were compared with those without (n Å 38). The and destroys the glands.hypothesized sequence of events leading to loss of glands is We noted in the early stage of H. pylori–associated CAGsummarized in figure 12. that a lymphoid adherence lesion, consisting of lymphocytes

and occasionally eosinophils, closely surrounded the mucinousantral glands, with distortion of and progressive atrophic

Discussionchanges in the affected glands. Whitehead et al. noted thislesion’s association with atrophy of the body of the stomachWe have presented what we believe to be one of the firstbut did not suggest that it was important in the atrophy processdetailed pathological descriptions of H. pylori–associated CAG[15]. More recently, this lesion was reported in the body ofand the progression from early to late CAG. In addition, wethe stomach in patients with pernicious anemia and was consid-have described a new lymphoid adherence lesion that may beered to be an autoimmune lesion [16].involved in the pathogenesis of CAG.

In another series of biopsies we noted that the lymphoidCurrently, CAG connotes deep antral gastric gland loss. Weadherence lesion was present not only in the antrum but alsobelieve that gland loss is an end-stage lesion. We propose thatin the body of stomachs of Peruvian patients (S. Recavarren,CAG progresses from inflammatory changes in gastric antralunpublished observation). The morphological features of theglands to their eventual loss and replacement by IM and/orlymphoid adherence lesion suggest that cell-mediated immu-fibrosis. Our scheme is based on transversal data and requiresnity may play an important role in the loss of the gastric antralconfirmation by longitudinal studies of individual patients.glands. To our knowledge, the band lesion on top of the muscu-The absence of CAG in Peruvian children and absence oflaris mucosa has not been previously described with regard toIM in Peruvian adolescents, compared with the high prevalenceH. pylori–associated gastritis. This change is similar to thein adults, support the progressive nature of the lesion. In addi-band atrophy described in association with quiescent stages oftion, we found a statistically significant association betweenulcerative colitis [17].age and increasing lesion severity, but this association was not

In our study not all biopsies were positive for H. pylori. Instriking except for the most severe lesions.contrast, in nearly all biopsies, chronic active superficial gastri-Further evidence of the progressive nature of H. pylori–tis was associated with PMN leukocyte infiltration. This findingassociated CAG was provided by a Dutch study [13]. Fifty-is the most sensitive indicator of H. pylori infection [18], sinceeight infected patients with chronic gastritis underwent endos-the bacterium may not be seen owing to its patchy distribution.copy repeated after a mean interval of 11.5 years and developed

This study also documents that often many endoscopic bi-CAG eight times more frequently than did 49 uninfected pa-tients without gastritis. Unfortunately, the criteria for atrophic opsy fragments are inadequate for evaluation of CAG. It is

Table 1. Gastric pathological diagnosis as related to mean ages of 50 patients* in a population where Helicobacter pylori infection is frequent.

Pathological diagnosis

Variable Normal CSG CDG Early CAG Late CAG CAG with IM

No. of patients 2 5 7 8 12 16Mean age (y) 22.0 45.8 34.0 33.8 43.5 47.7SD 8.5 14.9 8.8 11.4 16.2 15.4

NOTE. Analysis of variance linearity trend for age: P õ .05. CAG Å chronic atrophic gastritis; CDG Å chronic deep gastritis; CSG Å chronic superficialgastritis; IM Å intestinal metaplasia.

* Ages of 4 patients, 3 with CAG and IM and 1 with CSG, were not available.

Dow




5. Correa P, Cuello C, Duque E, et al. Gastric cancer in Columbia. III. Theimportant for pathologists to inform endoscopists concerningnatural history of precursor lesions. J Natl Cancer Inst 1976;57:the adequacy of their biopsy fragments. In adequate biopsies,1027–35.

the muscularis mucosa can be seen as one or two rose-colored 6. Klein PD, The Gastrointestinal Physiology Working Group of Cayetanodots attached to the base of the biopsy tissue. Biopsies that Heredia and the Johns Hopkins University, Graham DY, et al. Water

source as risk factor for Helicobacter pylori infection in Peruvian chil-lack this finding should be repeated. In addition, the adequacydren. Lancet 1991;337:1503–6.of biopsy specimens would be improved by redesigning the

7. Klein PD, Gilman RH, Leon-Barua R, Diaz F, O’Brien-Smith E, Grahambiopsy forceps to provide a slightly deeper biopsy. Often, long,DY. The epidemiology of Helicobacter pylori in Peruvian children

shallow biopsy specimens give a false impression of adequacy. between 6–30 months of age. Am J Gastroenterol 1994;89:2196–200.Finally, correct orientation of the biopsy can be achieved only 8. The Gastrointestinal Physiology Working Group of Cayetano Heredia and

The Johns Hopkins University. Helicobacter pylori and gastritis in Peru-when well-trained and experienced technicians embed the frag-vian patients: relationship to socioeconomic level, age and sex. Am Jment appropriately.Gastroenterol 1990;85:819–23.This study describes the sequence of lesions in the develop-

9. Galvez-Brandon J. La epidemiologia de los tumores malignos. In: Oncol-ment of CAG, emphasizing early markers in the process such ogy. Lima: The National Cancer Institute, 1975.as the glandular lymphocyte adherence lesion and gastric gland 10. Ming S, Goldman H. Endoscopy and endoscopic biopsy. Haggitt RC,

Rubin CE, eds. Philadelphia: WB Saunders/Harcourt, Brace, Jovanov-remnants. Recognition of these early CAG lesions may permitich, 1994:10–2.their treatment and prevent progression to later, less-reversible

11. The Gastrointestinal Physiology Working Group. Rapid identification ofstages of CAG, which are premalignant. Late CAG is associ-pyloric campylobacter in Peruvians with gastritis. Dig Dis Sci 1986;

ated with hypochlorhydria, bacterial overgrowth, and increased 31:1089–94.levels of nitrosamine, a potent carcinogen [18–20]. In underde- 12. Dixon MF. Campylobacter pylori and chronic gastritis. In: Rathbone J,

Heatley RV, eds. Campylobacter pylori and gastroduodenal disease.veloped countries such as Peru, H. pylori–associated gastritisOxford: Blackwell Scientific, 1989:111.and CAG are common and appear to be associated with an

13. Kuipers EJ, Uyterlinde AM, Pena AS, et al. Long-term sequelae of Helico-increased risk of gastric cancer. Interventions that reverse earlybacter pylori gastritis. Lancet 1995;345:1525–8.

CAG may also decrease the risk of gastric cancer. 14. Tsuji M, Kawano S, Shingo T, et al. Cell kinetics of mucosal atrophy inrat stomach induced by long-term administration of ammonia. Gastroen-terology 1993;104:796–801.

References 15. Whitehead R, Trulove SC, Gear WM. The histological diagnosis of chronicgastritis in fibreoptic gastroscope biopsy specimens. J Clin Pathol 1972;1. Recavarren-Arce S, Leon-Barua R, Cok J, et al. The Gastrointestinal Physi-

ology Working Group. Helicobacter pylori and progressive gastric pa- 25:1.16. Stolte M, Baumann K, Bethke B, Ritter M, Lauer E, Eidt H. Activethology that predisposes to gastric cancer. Scand J Gastroenterol 1991;

26:S181, 51–7. autoimmune gastritis without total atrophy of the glands. Gastroenterol1992;30:729–35.2. Leon-Barua R, Recavarren-Arce S, Gilman RH, Berendson R. Can eradica-

tion of Helicobacter pylori prevent gastric cancer? Drugs 1993;46: 17. Whitehead R. Mucosal biopsy of the gastrointestinal tract. In: BenningtonJL, ed. Major problems in pathology. London: WB Saunders, 1973:341–6.

3. Recavarren-Arce S, Leon-Barua R, Rodriguez C, Cok J, Berendson R, 140–4.18. Dixon MF, Genta RM, Yardley JH, et al. Classification and grading ofGilman RH. Helicobacter pylori–associated chronic gastritis in Peru-

vian adolescents is very common and severe. J Clin Gastroenterol 1995; gastritis: the updated Sydney system. Am J Surg Path 1996;20:1161–81.20:335–7.

4. Siurala M, Isokoski M, Varis K, Kekki M. Prevalence of gastritis in a 19. Ivey JK. What’s new in chronic gastritis? Med J Aust 1973;2:857–9.20. Franklin MA, Skoryna SC. Studies on natural gastric flora: survival ofrural population. Bioptic study of subjects selected at random. Scand J

Gastroenterol 1968;3:211–23. bacteria in fasting human subjects. Can Med Assoc 1971;105:380–6.

Dow



Documents

Chronic Atrophic Gastritis - Clinical Infectious Diseases