Objectives. Gastric intestinal metaplasia (IM) is frequently encountered and is considered a precursor of gastric adenocarcinoma. In the Van region of Turkey, gastric adenocarcinoma incidence is high but the prevalence of gastric IM is not known. Helicobacter pylori (H. pylori) infection is a main factor leading to atrophy, IM, and cancer development in the stomach. The aim of the current study was to investigate the prevalence of IM and its subtypes and the prevalence of H. pylori infection, atrophy, dysplasia, and cancer in gastric IM subtypes. Materials and Methods. This retrospective study was conducted on 560 IM among the 4050 consecutive patients who were undergoing esophagogastroduodenoscopy (EGD) with biopsy between June 2010 and October 2014. Clinical records and endoscopic and histopathologic reports of patients with IM were analyzed. Results. The prevalence of gastric IM was 13.8%. The prevalence of incomplete IM was statistically significantly higher than complete IM. Type III IM was the most frequent subtype.
Conclusions. Gastric IM is a common finding in patients undergoing EGD with biopsy in this region. High prevalence of incomplete type IM, especially type III, can be associated with the high prevalence of gastric cancer in our region.
1. Introduction
The incidence of gastric cancer has declined over the past decades but gastric cancer is still the fourth most common cancer and second leading cause of cancer related death worldwide [1, 2]. Gastric cancer has two types: intestinal and diffuse types. The pathogenesis of the intestinal type of gastric cancer is related to precursor lesions such as chronic atrophic gastritis, intestinal metaplasia (IM), and adenoma/dysplasia. However, the pathogenesis of the diffuse type of gastric cancer is not well defined [3].
Helicobacter pylori (H. pylori) infection is associated with severe gastritis, chronic atrophic gastritis, and IM, as well as gastric cancer [4–6]. Severe atrophy accompanying IM is related to a particularly high risk in H. pylori infected patients [7]. Some clinical studies reported that IM improves or did not progress after H. pylori eradication therapy [8–10]. As a result, H. pylori eradication is one of the most promising approaches in gastric cancer prevention [11].
Gastric IM is defined as replacement of gastric mucosa by epithelium of intestinal morphology and is generally considered to be a precursor of gastric cancer [12]. Risk factors of IM are H. pylori infection, high salt intake, smoking, alcohol consumption, and chronic bile reflux [13]. Although there are several classifications of IM types, currently they are generally subclassified as complete (type I) or incomplete (types II and III) [14, 15]. The incomplete type of IM, particularly type III, has a higher gastric cancer risk than the complete type [13, 15].
The incidence of gastric cancer varies in different parts of the world [1, 2]. Although the incidence of gastric cancer in Turkey is not high, its incidence in the Van region is very high and gastric cancer is an important health problem [16, 17]. Gastric IM is known as an important risk factor for gastric cancer. As far as we know, no data is available regarding its prevalence in the Van region of Turkey.
The aim of the current study was to investigate the prevalence of IM and its subtypes and the prevalence of H. pylori infection, atrophy, dysplasia, and cancer in gastric IM. We also compared the relationship between H. pylori infection and IM subtypes.
2. Materials and Methods2.1. Subjects
This retrospective study was conducted with 560 consecutive IM patients among the 4050 patients who were undergoing esophagogastroduodenoscopy (EGD), between June 2010 and October 2014, at the Department of Gastroenterology of the Yuzuncu Yıl University Medical Hospital.
Data including age, gender, indication for EGD, and endoscopic findings and pathology reports (histopathologic features, presence of associated chronic gastritis, H. pylori atrophy, dysplasia, and cancer) were retrieved from electronic medical records.
Indications for EGD are as follows: gastric cancer alarming symptoms or cancer screening, abdominal pain, anemia, dyspeptic symptoms, reflux, and follow-up of Barrett’s esophagus and other premalignant lesions. If more than one EGD was performed with the same patient, only the first examination was included in the study.
The exclusion criteria were patients undergoing gastric surgery and upper gastrointestinal bleeding and patients with lacking data.
Esophagogastroduodenoscopy was performed after local 10% lidocaine sedation with a Fujinon video endoscope (EG 530 WR, Tokyo, Japan). Mean biopsies number was 3.7±1.6 [1–12], and if lesions were suspected to be cancerous, extra biopsies were taken from the suspected area.
Biopsy specimens were evaluated for the presence of atrophic gastritis and IM. H. pylori was determined using histology testing. Pathologists of the Yuzuncu Yıl University Medical Hospital examined all the specimens.
All the endoscopically biopsied materials are fixed with Holland solution, after routine tissue follow-up processes embedded into paraffin blocks and stained with hematoxylin and eosin. Specimens are stained with periodic acid-Schiff/Alcian Blue (PAS/AB) at pH 2.5 stain combination and High Iron Diamine Alcian Blue (HID-AB) at pH 2.5 stain. Subtypes of IM are determined according to Filipe et al. classification [18].
The specimens were fixed in formalin and assessed for H. pylori (by Giemsa staining), the degree of neutrophil infiltration, IM (by staining with hematoxylin and eosin), atrophy, dysplasia, and other lesions.
IM was identified by replacement of glandular epithelium with goblet cells. IM was classified into complete (type I) or incomplete types (types II and III).
Type I includes mature absorptive cells and goblet cells, while the incomplete type secretes sialomucins. In type I IM, in PAS/AB stain at pH 2.5, the goblet cells are positively stained; the cylindrical cells between them show no reaction in both combinations and have markedly brush borders.
Type II has few or absent absorptive cells, presence of columnar “intermediate” cells in various stages of differentiation secreting neutral and acid sialomucins and goblet cells secreting sialomucins or, occasionally, sulfomucins, or both. In type II IM, in PAS/AB stain at pH 2.5, goblet cells are positively stained; cylindrical cells contain PAS positively stained granules; in HID/AB pH 2.5, cylindrical cells show no reaction.
Type III is formed by columnar intermediate cells secreting predominantly sulfomucins and goblet cells secreting sialomucins, sulfomucins, or both. If more than one subtype of IM was present, the specimen was classified based on the predominant IM phenotype. In type III IM, in PAS/AB stain at pH 2.5, goblet cells and cylindrical cells are positively stained with AB and in HID AB stain at pH 2.5 combination is positively stained.
The study protocol was approved by the ethics committee of our university.
2.2. Statistical Analysis
Descriptive statistics for continuous variables (characteristics) were presented as mean, standard deviation, and minimum and maximum values while they were count and percent for categorical variables. Chi-square test was used to determine linear association between categorical variables. In addition, Fisher’s exact test was performed to compare two proportions in groups. Statistical significance level was considered as 5% and SPSS (version 13.0) statistical program was used for all statistical computations.
3. Results
Among the 4050 consecutive patients who had gastric biopsies performed, 560 were found to have gastric IM. Patient mean age was 57±15 years (range: 17–98 years). There were 227 (40.5%) females and 333 (59.5%) males. The most frequent indications for EGD included dyspepsia (41.2%), abdominal pain (25.4%), and reflux symptoms (13.4%). The most frequent gastric endoscopic findings were gastritis (nonerosive, erosive, atrophic, etc.) (75.3%), ulcer (6.1%), and cancer (3.9%). Demographic features, indication for endoscopy, and endoscopic findings of the study population are shown in Table 1. The majority of patients were 41–70 years of age.
Demographic features, indication for EGD, and EGD findings of patients with intestinal metaplasia.
The prevalence of gastric IM was 13.8% at our institution. Eight of the 560 patients did not have an IM type (1.4%); 46 patients (8.2%) had complete type IM (Type 1). Of the 506 (90.4%) incomplete IM patients, 179 (32%) patients had type II, 214 (38%) patients had type III, and 113 (20.2%) patients had an unidentified subtype (Figure 1). The prevalence of incomplete IM was significantly higher than complete IM (P<0.05).
Flow diagram of intestinal metaplasia patients included in the study.
Of the 46 complete IM patients, 6 had atrophy, 8 had dysplasia, and 1 had cancer. Of the 505 incomplete IM patients, 113 had atrophy, 72 had dysplasia, and 17 had cancer (Table 2).
Prevalence of atrophy, dysplasia, and cancer in types of intestinal metaplasia.
Prevalence of H. pylori infection was 38.6% in gastric IM. The relationship between Helicobacter pylori status and types of IM was assessed. No significant difference in prevalence of complete type and incomplete type IM was found with H. pylori status (P>0.05) (Table 3).
Relationship between Helicobacter pylori status and types of intestinal metaplasia.
In the model of gastric carcinogenesis, H. pylori plays a pivotal role in causing chronic active gastritis. Chronic H. pylori induces gastritis and may progress over years through the sequential stages of atrophic gastritis, IM, and dysplasia to gastric adenocarcinoma [19, 20]. An important risk factor for gastric cancer development is the presence of premalignant changes of the gastric mucosa, such as IM, atrophy, and dysplasia [4, 11, 21].
The prevalence of gastric IM and atrophy in the general population is known to vary around the globe, mostly depending on H. pylori status [7, 22]. IM is a well-known risk factor for the development of gastric cancer [21, 22]. Diagnosis of atrophic gastritis, IM, and dysplasia is often ignored in routine clinical practice [23]. There are no widely accepted guidelines for the management of gastric IM [24]. The American Society for Gastrointestinal Endoscopy (ASGE) and The European Society of Gastrointestinal Endoscopy and other European academic societies have developed evidence-based guidelines for the management of patients with gastric IM [11, 25]. ASGE guidelines for endoscopic surveillance for gastric IM cannot be uniformly recommended but surveillance may be beneficial for patients at increased risk of gastric cancer due to ethnic background or family history [25]. Recently, according to the published guidelines for endoscopic management of precancerous conditions and lesions in the stomach, endoscopic surveys should be offered to patients with extensive atrophy and/or IM without subtyping every three years [11]. Several studies demonstrated that endoscopic histological follow-up in patients with IM is able to detect gastric cancer at an early stage with a considerable mortality reduction and scheduled endoscopic control could be cost-effective in IM patients [26–28]. According to some authors, complete IM is associated with a lower risk of gastric cancer; therefore, in the absence of other risk factors for gastric cancer, patients with complete IM do not need long-term endoscopic surveillance [13]. Due to the high prevalence of gastric cancer and H. pylori infection in our region, we strongly recommend endoscopic surveillance of gastric IM without paying attention to subtype.
The prevalence of gastric IM was 13.8% in our study. The prevalence of gastric IM in the general population remains difficult to ascertain due to the asymptomatic nature of the lesion. There is a wide variation in the prevalence of gastric IM depending on the different methods used in studies and particularly the prevalence of H. pylori infection in the region. [22, 28]. Sonnenberg et al. conducted a large retrospective study of 78,985 patients undergoing EGD with biopsy across the United States and found that the prevalence of gastric IM was 7% [29]. Almouradi et al. reported that, among the 437 patients who had gastric biopsies performed, 66 were found to have gastric IM and they observed that the overall prevalence was 15% [30]. In Netherlands, the prevalence of IM in a large histopathology database was reported as 8% [21]. We think that the prevalence of IM may be higher in our region. This low prevalence of IM may derive from the insufficient number and inappropriate localization of biopsies.
In our study, the prevalence of incomplete type of IM (90.4%) was much higher than the complete type (8.2%). The type III IM was the highest subtype. Eriksson et al. reported that among 505 patients the total prevalence of IM was 19% and the prevalence of type III IM was 2.8%, type II IM was 4.4%, and type I IM was 11% [31]. Ozdil et al. reported that of 3301 consecutive dyspeptic patients 17.8% had IM. They observed that 86% had complete and 14% had incomplete IM [32].
The presence of incomplete type IM, especially type III, significantly increases the risk of gastric cancer as compared to type I and type II [28, 33–35]. High detection of incomplete type IM, particularly type III, may shed light on the high prevalence of gastric cancer in our region. The prevalence of incomplete type IM was considerably higher than in previous studies [36, 37].
Evidence on establishing the subtype of IM is limited and conflicting [24]. There are several problems when subtyping IM. There is the need for additional histochemical techniques and these techniques are far from being standardized. The current technique uses reagents such as High Iron Diamine Alcian Blue staining, which are toxic and potentially carcinogenic. Pathologically, subtyping is sometimes difficult because IM have a multifocal distribution within the stomach and IM areas may be small. As a result, sampling errors may be unavoidable. Also, different subtypes of IM frequently coexist. When there are different subtypes, it is unclear which should be considered for categorizing [12, 34]. Perhaps due to these reasons, in our study, a distinction of complete and incomplete types in eight of 560 patients and a distinction of type II and type III in 113 patients could not be taken into account.
In our study, we observed that gastric IM was higher in older than in younger patients. Several studies revealed that age ≥50 years was an independent risk factor for IM, which is consistent with previous studies reporting that the incidence of IM increases with age [25, 29, 36].
In a study regarding H. pylori infection in Turkey, Özden et al. found H. pylori positivity to be 81% in the general population [38]. In our study, the prevalence of H. pylori infection was 38.6% in gastric IM and no significant difference in prevalence of complete type and incomplete type IM was found with H. pylori presence. With and without atrophy, IM may cause lower diagnostic accuracy of H. pylori with histologic examination [13]. In addition, decreased sensitivity of the rapid urease test is observed in subjects taking proton pump inhibitors, so other diagnostic tests, including serology or stool antigen, should be considered [13, 37]. H. pylori eradication is one of the most promising approaches in gastric cancer prevention [11]. However, whether gastric atrophy and IM are reversible after the eradication of H. pylori remains controversial [22, 39], since in some studies eradication of H. pylori has not been effective in IM regressions [8–10, 26]. However, several studies have shown that eradication of H. pylori infection significantly retarded progression of gastric IM and therefore H. pylori eradication has been proposed [40–43].
This retrospective study design has several potential limitations. Firstly, biopsies may not have been taken in adequate number and from the same part of the stomach. Secondly, results of the study may be affected by the lack of medical history such as information about proton pump inhibitors or antibiotic usage in some patients. Finally, diagnosis of H. pylori infection was only made by histological examination.
In conclusion, this study indicated that gastric IM is a common finding in patients undergoing EGD with biopsy. In addition, we found that the prevalence of incomplete IM was statistically significantly higher than complete IM. In our study population, type III IM was the most frequent subtype. High prevalence of the incomplete type and notably type III may be associated with a high prevalence of gastric cancer in our region. H. pylori was detected in 38.6% of gastric IM patients in Turkey. To further establish the relationship between IM and gastric cancer in our region, large prospective, randomized, and multicenter studies are needed.
Conflict of Interests
The authors declared no conflict of interests.
FerlayJ.ShinH.-R.BrayF.FormanD.MathersC.ParkinD. M.Estimates of worldwide burden of cancer in 2008: GLOBOCAN 20082010127122893291710.1002/ijc.255162-s2.0-78049485263FerroA.PeleteiroB.MalvezziM.BosettiC.BertuccioP.LeviF.NegriE.La VecchiaC.LunetN.Worldwide trends in gastric cancer mortality (1980–2011), with predictions to 2015, and incidence by subtype20145071330134410.1016/j.ejca.2014.01.0292-s2.0-84897407598ParkD. Y.SrivastavaA.KimG. H.Mino-KenudsonM.DeshpandeV.ZukerbergL. R.SongG. A.LauwersG. Y.CDX2 expression in the intestinal-type gastric epithelial neoplasia: frequency and significance2010231546110.1038/modpathol.2009.1352-s2.0-73949127446International Agency for Research on Cancer1994Lyon, FranceIACRHoughtonJ.WangT. C.Helicobacter pylori and gastric cancer: a new paradigm for inflammation-associated epithelial cancers200512861567157810.1053/j.gastro.2005.03.0372-s2.0-18844439381CatalanoV.LabiancaR.BerettaG. D.GattaG.de BraudF.Van CutsemE.Gastric cancer200971212716410.1016/j.critrevonc.2009.01.0042-s2.0-67649449102UemuraN.OkamotoS.YamamotoS.MatsumuraN.YamaguchiS.YamakidoM.TaniyamaK.SasakiN.SchlemperR. J.Helicobacter pylori infection and the development of gastric cancer20013451178478910.1056/nejmoa0019992-s2.0-0035856021SatohK.KimuraK.TakimotoT.KihiraK.A follow-up study of atrophic gastritis and intestinal metaplasia after eradication of Helicobacter pylori1998342362402-s2.0-0032239179ZhouL.SungJ. J. Y.LinS.JinZ.DingS.HuangX.XiaZ.GuoH.LiuJ.ChaoW.A five-year follow-up study on the pathological changes of gastric mucosa after H. pylori eradication2003116111142-s2.0-0037253722LeeY.-C.ChenT. H.-H.ChiuH.-M.ShunC.-T.ChiangH.LiuT.-Y.WuM.-S.LinJ.-T.The benefit of mass eradication of helicobacter pylori infection: a community-based study of gastric cancer prevention201362567668210.1136/gutjnl-2012-3022402-s2.0-84875826886Dinis-RibeiroM.AreiaM.de VriesA. C.Marcos-PintoR.Monteiro-SoaresM.O'ConnorA.PereiraC.Pimentel-NunesP.CorreiaR.EnsariA.DumonceauJ. M.MacHadoJ. C.MacEdoG.MalfertheinerP.Matysiak-BudnikT.MegraudF.MikiK.O'MorainC.PeekR. M.PonchonT.RistimakiA.RembackenB.CarneiroF.KuipersE. J.Management of precancerous conditions and lesions in the stomach (MAPS): guideline from the European Society of Gastrointestinal Endoscopy (ESGE), European Helicobacter Study Group (EHSG), European Society of Pathology (ESP), and the Sociedade Portuguesa de Endoscopia Digestiva (SPED)20124601194610.1007/s00428-011-1177-82-s2.0-84857049101AkiyamaJ.UemuraN.Intestinal metaplasia subtype and gastric cancer risk20092414610.1111/j.1440-1746.2008.05736.x2-s2.0-58549092365CorreaP.PiazueloM. B.WilsonK. T.Pathology of gastric intestinal metaplasia: clinical implications2010105349349810.1038/ajg.2009.7282-s2.0-77749251829JassJ. R.FilipeM. I.A variant of intestinal metaplasia associated with gastric carcinoma: a histochemical study19793319119910.1111/j.1365-2559.1979.tb02996.x2-s2.0-0018295614MatsukuraN.SuzukiK.KawachiT.AoyagiM.SugimuraT.KitaokaH.NumajiriH.ShirotaA.ItabashiM.HirotaT.Distribution of marker enzymes and mucin in intestinal metaplasia in human stomach and relation to complete and incomplete types of intestinal metaplasia to minute gastric carcinomas19806522312402-s2.0-0018873607YalcinS.Gastric cancer in Turkey—a bridge between West and East20093129322-s2.0-77951733381TurkdoğanM. K.AkmanN.TuncerI.DilekF. H.AkmanE.MemikF.AksoyH.The high prevalence of esophageal and gastric cancers in Eastern Turkey199826179842-s2.0-0031740372FilipeM. I.PotetF.BogomoletzW. V.DawsonP. A.FabianiB.ChauveincP.FenzyA.GazzardB.GoldfainD.ZeegenR.Incomplete sulphomucin-secreting intestinal metaplasia for gastric cancer. Preliminary data from a prospective study from three centres198526121319132610.1136/gut.26.12.13192-s2.0-0022366048FuccioL.ZagariR. M.MinardiM. E.BazzoliF.Systematic review: Helicobacter pylori eradication for the prevention of gastric cancer200725213314110.1111/j.1365-2036.2006.03183.x2-s2.0-33845999566CorreaP.Human gastric carcinogenesis: a multistep and multifactorial process—first American Cancer Society Award lecture on cancer epidemiology and prevention19925224673567402-s2.0-0027065034de VriesA. C.van GriekenN. C. T.LoomanC. W. N.CasparieM. K.de VriesE.MeijerG. A.KuipersE. J.Gastric cancer risk in patients with premalignant gastric lesions: a nationwide cohort study in the Netherlands2008134494595210.1053/j.gastro.2008.01.0712-s2.0-41349092267de VriesA. C.HaringsmaJ.KuipersE. J.The detection, surveillance and treatment of premalignant gastric lesions related to Helicobacter pylori infection200712111510.1111/j.1523-5378.2007.00475.x2-s2.0-33846211765CapelleL. G.de VriesA. C.HaringsmaJ.BorgF. T.de VriesR. A.BrunoM. J.van DekkenH.MeijerJ.van GriekenN. C. T.KuipersE. J.The staging of gastritis with the OLGA system by using intestinal metaplasia as an accurate alternative for atrophic gastritis20107171150115810.1016/j.gie.2009.12.0292-s2.0-77952951065GonzálezC. A.Sanz-AnquelaJ. M.GisbertJ. P.CorreaP.Utility of subtyping intestinal metaplasia as marker of gastric cancer risk. A review of the evidence201313351023103210.1002/ijc.280032-s2.0-84879201093HirotaW. K.ZuckermanM. J.AdlerD. G.DavilaR. E.EganJ.LeightonJ. A.QureshiW. A.RajanE.FanelliR.Wheeler-HarbaughJ.BaronT. H.FaigelD. O.ASGE guideline: The role of endoscopy in the surveillance of premalignant conditions of the upper GI tract200663457058010.1016/j.gie.2006.02.0042-s2.0-33645240406WhitingJ. L.SigurdssonA.RowlandsD. C.HallisseyM. T.FieldingJ. W. L.The long term results of endoscopic surveillance of premalignant gastric lesions200250337838110.1136/gut.50.3.3782-s2.0-0036181067TavaF.LuinettiO.GhignaM. R.AlvisiC.PeregoM.TrespiE.KlersyC.FrattiC.FioccaR.SolciaE.Type or extension of intestinal metaplasia and immature/atypical ‘indefinite-for-dysplasia’ lesions as predictors of gastric neoplasia200637111489149710.1016/j.humpath.2006.05.0122-s2.0-33750075784ZulloA.HassanC.RomitiA.GiustoM.GuerrieroC.LorenzettiR.CampoS. M.TomaoS.Follow-up of intestinal metaplasia in the stomach: when, how and why201243303610.4251/wjgo.v4.i3.30SonnenbergA.LashR. H.GentaR. M.A national study of Helicobactor pylori infection in gastric biopsy specimens201013961894.e21901.e210.1053/j.gastro.2010.08.0182-s2.0-78649717911AlmouradiT.HiattT.AttarB.Gastric intestinal metaplasia in an underserved population in the USA: prevalence, epidemiologic and clinical features20132013485625610.1155/2013/8562562-s2.0-84887458647ErikssonN. K.KärkkäinenP. A.FärkkiläM. A.ArkkilaP. E. T.Prevalence and distribution of gastric intestinal metaplasia and its subtypes200840535536010.1016/j.dld.2007.12.0122-s2.0-41849120440OzdilK.SahinA.KahramanR.YuzbasiogluB.DemirdagH.CalhanT.YilmazM. S.SokmenH. M.Current prevalence of intestinal metaplasia and Helicobacter pylori infection in dyspeptic adult patients from Turkey201057104156315662-s2.0-79952399933GonzálezC. A.PardoM. L.LisoJ. M. R.AlonsoP.BonetC.GarciaR. M.SalaN.CapellaG.Sanz-AnquelaJ. M.Gastric cancer occurrence in preneoplastic lesions: a long-term follow-up in a high-risk area in Spain2010127112654266010.1002/ijc.252732-s2.0-78049505717FilipeM. I.MuñozN.MatkoI.KatoI.Pompe-KirnV.JutersekA.TeuchmannS.BenzM.PrijonT.Intestinal metaplasia types and the risk of gastric cancer: a cohort study in Slovenia19945733243292-s2.0-002832188710.1002/ijc.2910570306GomezJ. M.WangA. Y.Gastric intestinal metaplasia and early gastric cancer in the west: a changing paradigm2014106369378CorreaP.Human gastric carcinogenesis: a multistep and multifactorial process —First American Cancer Society Award lecture on cancer epidemiology and prevention19925224673567402-s2.0-0027065034GaliatsatosP.WyseJ.SzilagyiA.Accuracy of biopsies for Helicobacter pylori in the presence of intestinal metaplasia of the stomach2014251192310.5152/tjg.2014.6476ÖzdenA.DumluŞ.DöndericiÖ.Helicobacter pylori infeksiyonunun ülkemizde seroepidemiyolojisi19923664668WangJ.XuL.ShiR.HuangX.LiS. W. H.HuangZ.ZhangG.Gastric atrophy and intestinal metaplasia before and after Helicobacter pylori eradication: a meta-analysis201183425326010.1159/0002803182-s2.0-79251592276ItoM.HarumaK.KamadaT.MiharaM.KimS.KitadaiY.SumiiM.TanakaS.YoshiharaM.ChayamaK.Helicobacter pylori eradication therapy improves atrophic gastritis and intestinal metaplasia: a 5-year prospective study of patients with atrophic gastritis20021681449145610.1046/j.1365-2036.2002.01311.x2-s2.0-0036668705CorreaP.FonthamE. T. H.BravoJ. C.BravoL. E.RuizB.ZaramaG.RealpeJ. L.MalcomG. T.LiD.JohnsonW. D.MeraR.Chemoprevention of gastric dysplasia: randomized trial of antioxidant supplements and anti-Helicobacter pylori therapy200092231881188810.1093/jnci/92.23.18812-s2.0-0034614094de VriesA. C.KuipersE. J.Epidemiology of premalignant gastric lesions: Implications for the development of screening and surveillance strategies2007122223110.1111/j.1523-5378.2007.00562.x2-s2.0-35948966815ToyokawaT.SuwakiK.-I.MiyakeY.NakatsuM.AndoM.Eradication of Helicobacter pylori infection improved gastric mucosal atrophy and prevented progression of intestinal metaplasia, especially in the elderly population: a long-term prospective cohort study201025354454710.1111/j.1440-1746.2009.05995.x2-s2.0-77649195494