Advertisement

Use of Proton Pump Inhibitors and Risks of Fundic Gland Polyps and Gastric Cancer: Systematic Review and Meta-analysis

  • An Tran-Duy
    Correspondence
    Reprint requests Address requests for reprints to: An Tran-Duy, PhD, Department of Clinical Epidemiology and Medical Technology Assessment, Maastricht University Medical Center, PO Box 5800, 6202 AZ, Maastricht, The Netherlands. fax: +31 43 38 74419.
    Affiliations
    Department of Internal Medicine, Maastricht University Medical Center, Maastricht, The Netherlands

    Department of Clinical Epidemiology and Medical Technology Assessment, Maastricht University Medical Center, Maastricht, The Netherlands
    Search for articles by this author
  • Bart Spaetgens
    Affiliations
    Department of Internal Medicine, Maastricht University Medical Center, Maastricht, The Netherlands
    Search for articles by this author
  • Arno W. Hoes
    Affiliations
    Julius Center for Health Sciences and Primary Care, UMC Utrecht, Utrecht, The Netherlands
    Search for articles by this author
  • Niek J. de Wit
    Affiliations
    Julius Center for Health Sciences and Primary Care, UMC Utrecht, Utrecht, The Netherlands
    Search for articles by this author
  • Coen D.A. Stehouwer
    Affiliations
    Department of Internal Medicine, Maastricht University Medical Center, Maastricht, The Netherlands
    Search for articles by this author

      Background & Aims

      There have been increasing numbers of case reports and observational studies of adverse events in patients receiving long-term therapy with proton pump inhibitors (PPIs). The effects of PPI therapy on risks of fundic gland polyps (FGPs) and gastric cancer have received considerable attention. We performed a systematic review with a meta-analysis of randomized controlled trials and observational studies that assessed these risks.

      Methods

      We searched the PUBMED, EMBASE, and Cochrane Central Register of Controlled Trials databases for relevant studies published through July 2015. We calculated pooled odds ratio for FGPs and the risk ratio for gastric cancer in PPI users compared with PPI nonusers using fixed- and random-effects models.

      Results

      We analyzed data from 12 studies, comprising more than 87,324 patients: 1 randomized controlled trial reporting the effect of PPIs on gastric polyps (location not specified), 6 cohort and 1 case-control studies on FGPs, and 1 cohort and 3 case-control studies on gastric cancer. Pooled odds ratios for FGPs were 1.43 (95% confidence interval, 1.24–1.64) and 2.45 (95% confidence interval, 1.24–4.83) from fixed- and random-effects models, respectively. The pooled risk ratio for gastric cancer was 1.43 (95% confidence interval, 1.23–1.66) from each model. We observed significant heterogeneity among studies reporting on FGPs, but not among studies reporting on gastric cancer.

      Conclusions

      Based on a systematic review with meta-analysis, long-term use of PPIs (≥12 months) is associated with an increased risk of FGPs. PPI therapy might also increase the risk of gastric cancer, but this association could be biased, because of the limited number of studies and possible confounding factors.

      Keywords

      Abbreviations used in this paper:

      CI (confidence interval), FAP (familial adenomatous polyposis), FGP (fundic gland polyp), GERD (gastroesophageal reflux disease), NOS (Newcastle-Ottawa Scale), OR (odds ratio), PPI (proton pump inhibitor), RCT (randomized controlled trials), RR (risk ratio)
      Since the introduction of the first proton pump inhibitor (PPI) omeprazole in the late 1980s,
      • Shin J.M.
      • Sachs G.
      Pharmacology of proton pump inhibitors.
      PPIs have generally been considered to be safe and have been widely used in clinical practice. However, with an increasing number of case reports and emerging observational studies on the adverse events in patients receiving long-term PPI therapy, questions about the potential risks associated with their use have recently been raised. Currently, the most prominent concerns about long-term PPI use relate to the risks of bone fractures, enteric infection, pneumonia, and vitamin B12 deficiency. To a lesser extent, the effects of PPI therapy on the risks of fundic gland polyps (FGPs) and gastric cancer, of which the mechanisms are not well understood, have also drawn attention.
      FGPs are gastric mucosal lesions of small sizes (typically 2–5 mm) located in the fundus or body of the stomach.
      • Stolte M.
      • Sticht T.
      • Eidt S.
      • et al.
      Frequency, location, and age and sex distribution of various types of gastric polyp.
      • Odze R.D.
      • Marcial M.A.
      • Antonioli D.
      Gastric fundic gland polyps: a morphological study including mucin histochemistry, stereometry, and MIB-1 immunohistochemistry.
      These polyps are composed of cystically dilated fundic glands lined by parietal or chief cells that arrange in a disordered fashion.
      • Sipponen P.
      • Laxen F.
      • Seppala K.
      Cystic 'hamartomatous' gastric polyps: a disorder of oxyntic glands.
      • Lee R.G.
      • Burt R.W.
      The histopathology of fundic gland polyps of the stomach.
      They can be single or multiple,
      • Stolte M.
      • Sticht T.
      • Eidt S.
      • et al.
      Frequency, location, and age and sex distribution of various types of gastric polyp.
      • Odze R.D.
      • Marcial M.A.
      • Antonioli D.
      Gastric fundic gland polyps: a morphological study including mucin histochemistry, stereometry, and MIB-1 immunohistochemistry.
      • Lee R.G.
      • Burt R.W.
      The histopathology of fundic gland polyps of the stomach.
      and be sporadic or associated with an inherited polyposis syndrome, such as familial adenomatous polyposis (FAP).
      • Odze R.D.
      • Marcial M.A.
      • Antonioli D.
      Gastric fundic gland polyps: a morphological study including mucin histochemistry, stereometry, and MIB-1 immunohistochemistry.
      • Kinoshita Y.
      • Tojo M.
      • Yano T.
      • et al.
      Incidence of fundic gland polyps in patients without familial adenomatous polyposis.
      • Iida M.
      • Yao T.
      • Itoh H.
      • et al.
      Natural history of fundic gland polyposis in patients with familial adenomatosis coli/Gardner's syndrome.
      In 1992, Graham
      • Graham J.
      Gastric polyposis: onset during long-term therapy with omeprazole.
      published a letter noting 3 cases of FGPs developed after 1 year of therapy with omeprazole. Since then, case reports describing the occurrence of FGPs in chronic PPI users have steadily emerged over time.
      • Stolte M.
      • Bethke B.
      • Seifert E.
      • et al.
      Observation of gastric glandular cysts in the corpus mucosa of the stomach under omeprazole treatment.
      • Van Vlierberghe H.
      • De Vos M.
      • De Cock G.
      • et al.
      Fundic gland polyps: three other case reports suggesting a possible association with acid suppressing therapy.
      • Kazantsev G.B.
      • Schwesinger W.H.
      • Heim-Hall J.
      Spontaneous resolution of multiple fundic gland polyps after cessation of treatment with lansoprazole and Nissen fundoplication: a case report.
      • Flick A.L.
      Gastric fundal polyps with cellular atypia in a patient using esomeprazole (Nexium).
      • Kim J.S.
      • Chae H.S.
      • Kim H.K.
      • et al.
      Spontaneous resolution of multiple fundic gland polyps after cessation of treatment with omeprazole.
      • Spiegel A.
      • Stein P.
      • Patel M.
      • et al.
      A report of gastric fundic gland polyps.
      • Hegedus I.
      • Csizmadia C.
      • Lomb Z.
      • et al.
      Massive fundic gland polyposis caused by chronic proton pump inhibitor therapy. [Hungarian] Tartos protonpumpagatlo kezeles hatasara kialakulo massziv fundusmirigy-polyposis.
      However, the association of FGPs with the use of PPIs currently remains a topic of debate and the link between FGPs and the risk of gastric cancer is still unclear. It has been suggested that sporadic FGPs arise through activating mutations of the β-catenin gene, and FAP-associated FGPs are caused by alterations of the somatic adenomatous polyposis coli gene.
      • Abraham S.C.
      • Nobukawa B.
      • Giardiello F.M.
      • et al.
      Fundic gland polyps in familial adenomatous polyposis: neoplasms with frequent somatic adenomatous polyposis coli gene alterations.
      • Abraham S.C.
      • Nobukawa B.
      • Giardiello F.M.
      • et al.
      Sporadic fundic gland polyps: common gastric polyps arising through activating mutations in the beta-catenin gene.
      Although FGPs are generally considered to be benign, some reports showed low-grade dysplasia in FGPs,
      • Carmack S.W.
      • Genta R.M.
      • Schuler C.M.
      • et al.
      The current spectrum of gastric polyps: A 1-year national study of over 120,000 patients.
      • Wu T.T.
      • Kornacki S.
      • Rashid A.
      • et al.
      Dysplasia and dysregulation of proliferation in foveolar and surface epithelia of fundic gland polyps from patients with familial adenomatous polyposis.
      and 1 case report describing a chronic PPI user who developed FGPs that contained high-grade dysplasia.
      • Jalving M.
      • Koornstra J.J.
      • Gotz J.M.
      • et al.
      High-grade dysplasia in sporadic fundic gland polyps: a case report and review of the literature.
      Several researchers suggest that PPIs cause gastric cancer because they profoundly reduce gastric acid production and consequently lead to increased secretion of gastrin. Hypergastrinemia as a result of acid suppression causes hyperplasia of enterochromaffin-like cells in rats
      • Havu N.
      Enterochromaffin-like cell carcinoids of gastric mucosa in rats after life-long inhibition of gastric secretion.
      but this phenomenon has never been reported in other species.
      • Ali T.
      • Roberts D.N.
      • Tierney W.M.
      Long-term safety concerns with proton pump inhibitors.
      Although hypergastrinemia occurs in nearly all PPI users,
      • Schenk B.E.
      • Kuipers E.J.
      • Klinkenberg-Knol E.C.
      • et al.
      Hypergastrinaemia during long-term omeprazole therapy: influences of vagal nerve function, gastric emptying and Helicobacter pylori infection.
      whether or not it increases the risk of gastric cancer is controversial. In the last decade, discussion has shifted from the isolated carcinogenic effect of PPIs to the proposed synergistic effect between PPIs and Helicobacter pylori in carcinogenesis; the latter augments the acid-suppressive function of PPIs,
      • Verdu E.F.
      • Armstrong D.
      • Fraser R.
      • et al.
      Effect of Helicobacter pylori status on intragastric pH during treatment with omeprazole.
      • Gillen D.
      • Wirz A.A.
      • Neithercut W.D.
      • et al.
      Helicobacter pylori infection potentiates the inhibition of gastric acid secretion by omeprazole.
      causes non–H pylori bacterial overgrowth,
      • Sanduleanu S.
      • Jonkers D.
      • De Bruine A.
      • et al.
      Non-Helicobacter pylori bacterial flora during acid-suppressive therapy: Differential findings in gastric juice and gastric mucosa.
      and exacerbates gastritis because of double infection with H pylori and non–H pylori bacterial species.
      • Sanduleanu S.
      • Jonkers D.
      • De Bruine A.
      • et al.
      Double gastric infection with Helicobacter pylori and non-Helicobacter pylori bacteria during acid-suppressive therapy: increase of pro-inflammatory cytokines and development of atrophic gastritis.
      H pylori–driven gastric inflammation has been widely accepted as a risk factor of gastric cancer.
      • Song H.
      • Ekheden I.G.
      • Zheng Z.
      • et al.
      Incidence of gastric cancer among patients with gastric precancerous lesions: observational cohort study in a low risk Western population.
      • Wroblewski L.E.
      • Peek R.M.
      • Wilson K.T.
      Helicobacter pylori and Gastric Cancer: Factors That Modulate Disease Risk.
      Given the widespread use of PPIs, it is crucial to determine whether a relationship between PPI use and the risks of FGPs and gastric cancer exists. In this study we performed a systematic review with meta-analyses of the existing randomized controlled trials (RCTs) and observational studies to assess the association between PPI use and the risks of FGPs and gastric cancer.

      Methods

       Search Strategy and Eligibility Criteria

      We searched PUBMED, EMBASE, and Cochrane Central Register of Controlled Trials to identify all English-language studies published from the inception until July 31, 2015 that assessed the association between PPI use and the risk of FGPs or gastric cancer. We used the search terms “proton pump inhibitor,” “omeprazole,” “esomeprazole,” “pantoprazole,” “lansoprazole,” “dexlansoprazole,” “rabeprazole,” “fundic gland polyps,” “fundus polyps” “gastric polyps,” “stomach polyps,” “gastric cancer,” “gastric carcinoma,” “gastric adenocarcinoma,” “gastric neoplasm,” gastric neoplasia,” “stomach cancer,” “stomach carcinoma,” “stomach adenocarcinoma,” “stomach neoplasm,” “stomach neoplasia.” Detailed search queries are provided in Supplementary Tables 1 and 2. We also manually scanned the bibliographies of relevant articles for additional studies.
      We included RCTs and observational studies (ie, cohort or case-control studies) that met the following criteria: (1) studies written in English, (2) studies reported on FGPs or gastric cancer, (3) outcomes of PPI users were compared with those of PPI nonusers, and (4) studies provided adequate data that enabled an estimation of an odds ratio (OR) or a risk ratio (RR).

       Study Selection and Data Extraction

      The studies were independently reviewed by 2 reviewers (AT-D and BS). Where there was any uncertainty about the inclusion of a study, this issue was discussed between the 2 reviewers to achieve a resolution. The following information was independently extracted from the included studies and jointly verified for accuracy: adjusted OR or RR of gastric polyps and cancer (when available), first author’s surname, year of publication, country of study, type of outcome, study design, period of patient recruitment, patient inclusion criteria, number of participants and events, type and dose of PPIs, duration of PPI exposure, and age and gender of participants. We contacted authors when there was unclear information.

       Study Quality Assessment

      We assessed the risk of bias in RCTs using the Cochrane Collaboration’s tool,
      which addresses the following domains: sequence generation, concealment of allocation sequence, blinding, incomplete outcome data, selective outcome reporting, and any other problems that could put the studies at risk of bias. We assessed the methodological quality of the observational studies using the Newcastle-Ottawa Scale (NOS; range, 0–9)

      Wells GA, Shea B, O’Connell D, et al., editors. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. 2012. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed December 25, 2015.

      as recommended by the Cochrane Handbook for Systematic Reviews of Interventions.
      On this scale, a study is judged by giving points (stars in the NOS terminology) to items belonging to 3 categories: selection of the study groups, the comparability of the groups, and the ascertainment of the exposure (for case-control studies) or outcome of interest (for cohort studies) (see Supplementary Tables 3 and 4 for details).

       Outcome Measures, Data Analysis, and Presentation

      ORs and RRs were selected as measures of the effects of PPI use on the risk of FGPs and gastric cancer, respectively, so that the adjusted effect estimates in the included studies could be used in the meta-analyses. Pooled ORs and RRs were computed using both fixed-effect and random-effects models. Heterogeneity across studies was tested using the Q-statistic and quantified using the inconsistency I2.
      • Higgins J.
      • Thompson S.G.
      Quantifying heterogeneity in a meta-analysis.
      When possible, subgroup analyses were performed to assess the potential impact of duration of PPI exposure, quality of observational studies (NOS ≤5 vs NOS >5; cutoff point arbitrarily defined), and study design (RCT vs observational studies) on the pooled effects. In a sensitivity analysis, the influence of individual studies on the summary statistics was examined by omitting 1 study at a time from the meta-analysis. Publication bias was assessed using funnel plots and Egger regression test of funnel plot asymmetry.
      • Egger M.
      • Smith G.D.
      • Schneider M.
      • et al.
      Bias in meta-analysis detected by a simple, graphical test.
      In each funnel plot, the standard errors of the estimates were plotted on a vertical reversed scale against the effect estimates on the horizontal scale, and the triangle was centered on the pooled estimate and extending to 1.96 times the standard errors on either side. In the absence of bias, the scatter of the data points is subject to sampling variation alone and the plot resembles a symmetrical inverted funnel; in this case, the triangle should include about 95% of studies if the fixed-effect assumption (ie, all the studies have the same true treatment effect) is valid.
      • Egger M.
      • Smith G.D.
      • Schneider M.
      • et al.
      Bias in meta-analysis detected by a simple, graphical test.
      The statistical analyses were performed using the R package metaphor.
      • Viechtbauer W.
      Conducting meta-analyses in R with the metafor package.
      A P < .05 was considered statistically significant.
      We followed both the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement
      • Moher D.
      • Liberati A.
      • Tetzlaff J.
      • et al.
      Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.
      and the guidelines for Meta-analysis Of Observational Studies in Epidemiology (MOOSE)
      • Stroup D.F.
      • Berlin J.A.
      • Morton S.C.
      • et al.
      Meta-analysis of observational studies in epidemiology: a proposal for reporting.
      in reporting this study.

      Results

      Figure 1 shows the flow chart of the selection of studies. The search identified 1809 studies. After screening the titles, abstracts, and full text, 12 studies were considered eligible for complete data extraction. Of these, 1 RCT
      • Peura D.A.
      • Metz D.C.
      • Dabholkar A.H.
      • et al.
      Safety profile of dexlansoprazole MR, a proton pump inhibitor with a novel dual delayed release formulation: Global clinical trial experience.
      reported on the effect on gastric polyps (location not specified), 6 cohort
      • Choudhry U.
      • Boyce Jr., H.W.
      • Coppola D.
      Proton pump inhibitor-associated gastric polyps: A retrospective analysis of their frequency, and endoscopic, histologic, and ultrastructural characteristics.
      • Vieth M.
      • Stolte M.
      Fundic gland polyps are not induced by proton pump inhibitor therapy.
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      • Ally M.R.
      • Veerappan G.R.
      • Maydonovitch C.L.
      • et al.
      Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
      • Hsu W.H.
      • Wu I.C.
      • Kuo C.H.
      • et al.
      Influence of Proton Pump Inhibitor Use in Gastrointestinal Polyps.
      • Zelter A.
      • Fernandez J.L.
      • Bilder C.
      • et al.
      Fundic gland polyps and association with proton pump inhibitor intake: A prospective study in 1,780 endoscopies.
      and 1 case-control studies
      • Cao H.L.
      • Qu R.
      • Zhang Z.H.
      • et al.
      Sporadic fundic gland polyps are not associated with proton pump inhibitors therapy but negatively correlate with Helicobacter pylori infection in China.
      reported on FGPs, and 1 cohort
      • Poulsen A.H.
      • Christensen S.
      • McLaughlin J.K.
      • et al.
      Proton pump inhibitors and risk of gastric cancer: A population-based cohort study.
      and 3 case-control studies
      • Garcia Rodriguez L.A.
      • Lagergren J.
      • Lindblad M.
      Gastric acid suppression and risk of oesophageal and gastric adenocarcinoma: A nested case control study in the UK.
      • Crane S.J.
      • Locke I.G.R.
      • Harmsen W.S.
      • et al.
      Subsite-specific risk factors for esophageal and gastric adenocarcinoma.
      • Tamim H.
      • Duranceau A.
      • Chen L.Q.
      • et al.
      Association between use of acid-suppressive drugs and risk of gastric cancer: A nested case-control study.
      reported on gastric cancer. Table 1 shows characteristics of the included studies, outcomes, characteristics of the drug treatments, and study quality (for detailed quality assessment of the observational studies, see Supplementary Tables 3 and 4).
      Table 1Characteristics of the Included Studies on the Use of PPIs and Risks of Gastric Polyps and Cancer
      RCT or cohort study
      Author, yOutcomeStudy design; analysis type; countryPeriod of recruitmentPatient inclusion criteriaHelicobacter pylori infection prevalence in study populationPPI users/PPI nonusers, nEvents in PPI users/events in PPI nonusers, nMean age (SD or range)% MalePPI useAdjustment for covariates and study quality
      Peura et al., 2009
      • Peura D.A.
      • Metz D.C.
      • Dabholkar A.H.
      • et al.
      Safety profile of dexlansoprazole MR, a proton pump inhibitor with a novel dual delayed release formulation: Global clinical trial experience.
      Gastric polyps (location not specified)RCT; prospective; United StatesDec 2005–May 2007No criteria regarding age, gender, lifestyle, socioeconomic status, medication, or comorbidityNot reported5633/89619/1Patients receiving PPIs: 47.7
      Weighted mean.
      (13.7)
      Pooled SD.


      Patients receiving placebo: 48.4 (13.7)
      Patients receiving PPIs: 47.9

      Patients receiving placebo: 33.7
      Type and dose: dexlansoprazole MR, 30, 60 or 90 mg d-1; lansoprazole, 30 mg d-1

      Duration of exposure: ≤12 months (mean, 276 days
      Values available for dexlansoprazole only.
      ; SD, 134 days
      Values available for dexlansoprazole only.
      )
      Sequence generation and concealment of allocation sequence were described in another article
      • Fass R.
      • Chey W.D.
      • Zakko S.F.
      • et al.
      Clinical trial: the effects of the proton pump inhibitor dexlansoprazole MR on daytime and nighttime heartburn in patients with non-erosive reflux disease.
      ; these steps were conducted using the Interactive Voice Response System (ClinPhone Inc., Northbrook, IL), of which the adequacy assessment is beyond the scope of the present study; this was an open-label extension of a triple-blind RCT; incomplete outcome data were not adequately addressed; there was no evidence of selective outcome reporting; important potential confounders (eg, H pylori infection, peptic ulcer) were not taken into account in study design and analysis
      Choudhry et al., 1998
      • Choudhry U.
      • Boyce Jr., H.W.
      • Coppola D.
      Proton pump inhibitor-associated gastric polyps: A retrospective analysis of their frequency, and endoscopic, histologic, and ultrastructural characteristics.
      FGP detected by EGDCohort; retrospective; United StatesMar 1987– Feb 1996Patients undergoing EGDNot reported231/20727/6Not reportedNot reportedType and dose: omeprazole, 20 or 40 mg d-1; lansoprazole, (only 1 patient; dose not reported)

      Exposure time: 3–98 months (mean, 37.4; median, 35.0)
      No adjustment; NOS = 5
      Vieth and Stolte., 2001
      • Vieth M.
      • Stolte M.
      Fundic gland polyps are not induced by proton pump inhibitor therapy.
      FGP detected by histologic examinationCohort; retrospective; GermanyDuring year 1999Patients with no infection of H pylori0%2251/28,096116/1415PPI users: 53.1
      Weighted mean.
      (16.4)
      Pooled SD.


      PPI nonusers: 53.0
      Weighted mean.
      (17.6)
      Pooled SD.
      PPI users: 51.6

      PPI nonusers: 43.1
      Type and dose: omeprazole, 20 mg d-1; lansoprazole, 30 mg d-1; pantoprazole,40 mg d-1); exposure time >4 weeksNo adjustment; NOS = 5
      Jalving et al., 2006
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      FGP detected by EGD, confirmed by histologic examinationCohort; retrospective; the NetherlandsNov 2002– Mar 2005Patients undergoing EGDNot reported322/27775 (23.3%)/32 (11.5%)PPI users: 55.5
      Weighted mean.
      (21–86)

      PPI nonusers: 51 (17–88)
      PPI users: 45.0

      PPI nonusers: 37.9
      Type and dose: not reported

      Exposure time: 2 groups, <1 year, and >1 year
      Logistic regression was used to estimate ORs; covariates included age, reasons for PPI use, and use of other medications
      Information obtained from personal communications with the authors.
      ; NOS = 5
      Ally et al., 2009
      • Ally M.R.
      • Veerappan G.R.
      • Maydonovitch C.L.
      • et al.
      Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
      FGP detected by EGD, confirmed by histologic examinationCohort; retrospective; United StatesMar 2007–Sep 2007Patients undergoing EGD12.2% (none of these developed FGPs)252/13333/1052 (15)50Type and dose: not reported

      Exposure time: 2 groups, 1–48 months and >48 months
      Multiple logistic regression was used to estimate ORs; covariates included duration of PPI use, H pylori infection, and white race; NOS = 7
      Hsu et al., 2010
      • Hsu W.H.
      • Wu I.C.
      • Kuo C.H.
      • et al.
      Influence of Proton Pump Inhibitor Use in Gastrointestinal Polyps.
      FGP detected by EGD, confirmed by histologic examinationCohort;

      retrospective; Taiwan
      Jan 2007–Aug 2007Exclusion: patients with incomplete colonoscopy examination, inadequate bowel preparation, inflammatory bowel disease, newly diagnosed GI malignancy, and familial colon polyposisPPI users: 6.6%

      PPI nonusers: 21.2%
      122/13760/38PPI users: 61.3 (11.2)

      PPI nonusers: 52.5 (12.9)
      PPI users: 44.3%

      PPI nonusers: 45.6%
      Type and dose: not reported

      Duration of exposure: 1–97 months (mean, 21.9 months)
      No adjustment; NOS = 7
      Zelter et al., 2011
      • Zelter A.
      • Fernandez J.L.
      • Bilder C.
      • et al.
      Fundic gland polyps and association with proton pump inhibitor intake: A prospective study in 1,780 endoscopies.
      FGP detected by EGD, confirmed by histologic examinationCohort;

      prospective;

      Argentina
      Jun 2007– Aug 2008Patients undergoing EGDPPI users: 18.2%
      Applicable only for a subgroup of 306 patients whose biopsies were taken; of these patients, 55 received PPIs.


      PPI nonusers: 16.7%
      Applicable only for a subgroup of 306 patients whose biopsies were taken; of these patients, 55 received PPIs.
      313/146249/2850.7
      Weighted mean.
      (14.9)
      Pooled SD.
      41.1Type and dose: not reported

      Duration of exposure: >12 months
      Multiple logistic regression was used to estimate ORs; covariates included age, gender, and H pylori infection; NOS = 7
      Poulsen et al., 2009
      • Poulsen A.H.
      • Christensen S.
      • McLaughlin J.K.
      • et al.
      Proton pump inhibitors and risk of gastric cancer: A population-based cohort study.
      Gastric cancer identified using a modified Danish version of ICD-7 codePopulation-based cohort;

      retrospective; Denmark
      Jan 1990–Dec 2003Patients aged 40–84 years without a history of cancer (except nonmelanoma skin cancer); for patients receiving PPIs, only new users were included (ie, all patients prescribed PPIs during 1989 [the year before the index date] or before 40 years old were excluded)Not available; 13% underwent H pylori eradication therapy18,790/not reported109/not reported62 (SD not reported)47%Type: omeprazole (dominant), lansoprazole, esomeprazole, pantoprazole, rabeprazole

      Dose: not reported

      Duration of exposure: 4 groups, <1 year, 1 year, 2–4 years and >5 years
      Log-linear Poisson was used to estimate incident rate ratios; covariates included calendar period, age, gender, gastroscopy (≥1 year before gastric cancer diagnosis), COPD, alcohol-related admission or therapy, number of NSAID prescription, history of H pylori eradication therapy, smoking, and alcohol-related admission or therapy; NOS = 8
      Case-control study
      Author, yOutcomeStudy design; analysis type; countryPeriod of recruitmentPatient inclusion criteriaCases/control subjects, nPPI users in cases/PPI users in control subjects, nMean age (SD or range)% MalePPI useAdjustment for covariates
      Cao et al., 2014
      • Cao H.L.
      • Qu R.
      • Zhang Z.H.
      • et al.
      Sporadic fundic gland polyps are not associated with proton pump inhibitors therapy but negatively correlate with Helicobacter pylori infection in China.
      FGP detected by EGD, confirmed by histologic examinationCase-control;

      retrospective; China
      Mar 2011–Mar 2012Patients undergoing EGDCases: 22.3%

      Control subjects: 42.3%
      328/65654/136Cases: 55.1 (12.6)

      Control subjects: 54.9 (12.7)
      24.1%Type and dose: not reported

      Duration of exposure: 3 groups, 1–6 months, 6 months–1 year, and >1 year
      No adjustment; NOS = 3
      Garcia Rodriguez et al., 2006
      • Garcia Rodriguez L.A.
      • Lagergren J.
      • Lindblad M.
      Gastric acid suppression and risk of oesophageal and gastric adenocarcinoma: A nested case control study in the UK.
      Gastric adenocarcinoma, identified using medical codes in GPRDNested case-control; retrospective; United KingdomJan 1994–Dec 2001For both cases and control subjects: patients aged 40–84 years, enrolled with a general practitioner for at least 2 years, having at least one year of prescription history recorded in the database, and with no history of cancer.

      For cases: patients with other concurrent cancer
      Not available522/10,00043/399Mean: not reported; range: 40-84Cases: 67.6

      Control subjects: not reported
      Type and dose: not reported

      Exposure time: 3 groups, <year, 1–3 years, and >3 years
      Unconditional logistic regression was used to estimate ORs; covariates included smoking, alcohol consumption, age, sex, calendar year, BMI, gastroesophageal reflux, hiatal hernia, peptic ulcer, and dyspepsia; NOS = 6
      Crane et al., 2007
      • Crane S.J.
      • Locke I.G.R.
      • Harmsen W.S.
      • et al.
      Subsite-specific risk factors for esophageal and gastric adenocarcinoma.
      Gastric adenocarcinomaCase-control; retrospective; United States1971–2000Patients without a history of cancerNot available121/1212/074 (13)53Type and dose: not reported

      Exposure time: >1 year
      Unable to compute an OR because no patient in control subjects was exposed to PPIs; NOS = 5
      Tamim et al., 2008
      • Tamim H.
      • Duranceau A.
      • Chen L.Q.
      • et al.
      Association between use of acid-suppressive drugs and risk of gastric cancer: A nested case-control study.
      Gastric cancer identified using ICD codesNested case-control; retrospective; CanadaJan 1995–Dec 2003All people living in Quebec, eligible for outpatient prescription drug benefits for at least 5 years, and with no history of cancerNot available1598/12,991248/1402Cases: 75.5 (9.3)

      Control subjects: 75.9 (8.8)
      Cases: 52.1

      Control subjects: 51.5
      Type, dose, and exposure time: not reportedConditional logistic regression was used to estimate ORs; covariates included number of prescriptions to any drug, total length of hospitalizations, and number of visits to GPs, specialists, and emergency rooms during the year before the diagnosis; NOS = 4
      BMI, biomass index; COPD, chronic obstructive pulmonary disease; EGD, esophagogastroduodenoscopy; GI, gastrointestinal; GP, general practitioner; GPRD, General Practice Research Database (now CPRD); ICD, International Classification of Diseases; NSAID, nonsteroidal anti-inflammatory drug; SD, standard deviation.
      a Weighted mean.
      b Pooled SD.
      c Values available for dexlansoprazole only.
      d Information obtained from personal communications with the authors.
      e Applicable only for a subgroup of 306 patients whose biopsies were taken; of these patients, 55 received PPIs.

       Proton Pump Inhibitor Use and Risk of Gastric Polyps

       Pooled effect

      Eight studies (1 RCT and 7 observational studies) reporting on the effect on FGPs were included in the analysis. Of the observational studies, 3 studies
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      • Ally M.R.
      • Veerappan G.R.
      • Maydonovitch C.L.
      • et al.
      Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
      • Zelter A.
      • Fernandez J.L.
      • Bilder C.
      • et al.
      Fundic gland polyps and association with proton pump inhibitor intake: A prospective study in 1,780 endoscopies.
      adjusted the effect estimates for potential confounding. H pylori infection status, an important potential confounding factor, was considered in 2 studies only.
      • Ally M.R.
      • Veerappan G.R.
      • Maydonovitch C.L.
      • et al.
      Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
      • Zelter A.
      • Fernandez J.L.
      • Bilder C.
      • et al.
      Fundic gland polyps and association with proton pump inhibitor intake: A prospective study in 1,780 endoscopies.
      The directions of the estimated ORs and their 95% confidence intervals (CIs) were not consistent among the studies (Figure 2A). When all the estimates were pooled, the summary OR (95% CI) in PPI users compared with PPI nonusers from the fixed- and random-effects models were 1.43 (1.24–1.64) and 2.45 (1.24–4.83), respectively. Significant heterogeneity was present across studies (P < .0001 [both models]; I2 = 92.8% and 92.7% in fixed- and random-effects models, respectively).
      Figure thumbnail gr2
      Figure 2Forest plots of (A) odds ratios for gastric polyps and (B) risk ratios of gastric cancer in patients receiving proton pump inhibitors compared with subjects not receiving proton pump inhibitors. CC, case-control study; CH, cohort study; GC, gastric cancer; GCA, gastric cardia adenocarcinoma; GNCA, gastric noncardia adenocarcinoma; GPs, gastric polyps.

       Sensitivity and subgroup analyses

      The sensitivity analysis showed that the direction and 95% CI of the pooled effect of the use of PPIs on gastric polyps and the presence of heterogeneity were not influenced by any single study in both fixed- and random-effect models (see Supplementary Table 5).
      Pooled ORs (95% CI) from the fixed-effect model based on data from groups of patients receiving PPIs ≤12 months (2 studies
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      • Cao H.L.
      • Qu R.
      • Zhang Z.H.
      • et al.
      Sporadic fundic gland polyps are not associated with proton pump inhibitors therapy but negatively correlate with Helicobacter pylori infection in China.
      ), ≥12 months (3 studies
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      • Zelter A.
      • Fernandez J.L.
      • Bilder C.
      • et al.
      Fundic gland polyps and association with proton pump inhibitor intake: A prospective study in 1,780 endoscopies.
      • Cao H.L.
      • Qu R.
      • Zhang Z.H.
      • et al.
      Sporadic fundic gland polyps are not associated with proton pump inhibitors therapy but negatively correlate with Helicobacter pylori infection in China.
      ), and >48 months (2 studies
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      • Ally M.R.
      • Veerappan G.R.
      • Maydonovitch C.L.
      • et al.
      Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
      ) were 0.82 (0.59–1.12), 3.81 (2.78–5.24), and 4.02 (2.49–6.48), respectively. The random-effects model produced the same results, except for the pooled OR (95% CI) in the second subgroup (≥12 months), which was 2.88 (0.97–8.55) (Figure 3A).
      Figure thumbnail gr3
      Figure 3Forest plots of (A) odds ratios for fundic gland polyps and (B) risk ratios of gastric cancer in subgroups of patients receiving proton pump inhibitors for different durations compared with subjects not receiving proton pump inhibitors. GC, gastric cancer; GCA, gastric cardia adenocarcinoma; GNCA, gastric noncardia adenocarcinoma.
      Pooled ORs (95% CI) from studies with NOS ≤5
      • Choudhry U.
      • Boyce Jr., H.W.
      • Coppola D.
      Proton pump inhibitor-associated gastric polyps: A retrospective analysis of their frequency, and endoscopic, histologic, and ultrastructural characteristics.
      • Vieth M.
      • Stolte M.
      Fundic gland polyps are not induced by proton pump inhibitor therapy.
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      • Cao H.L.
      • Qu R.
      • Zhang Z.H.
      • et al.
      Sporadic fundic gland polyps are not associated with proton pump inhibitors therapy but negatively correlate with Helicobacter pylori infection in China.
      and studies with NOS >5
      • Ally M.R.
      • Veerappan G.R.
      • Maydonovitch C.L.
      • et al.
      Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
      • Hsu W.H.
      • Wu I.C.
      • Kuo C.H.
      • et al.
      Influence of Proton Pump Inhibitor Use in Gastrointestinal Polyps.
      • Zelter A.
      • Fernandez J.L.
      • Bilder C.
      • et al.
      Fundic gland polyps and association with proton pump inhibitor intake: A prospective study in 1,780 endoscopies.
      were 1.12 (0.96–1.31) and 4.04 (2.92–5.59), respectively, from the fixed-effect model, and were 1.89 (0.66–5.42) and 3.54 (1.36–9.26), respectively, from the random-effects model. Pooled OR (95% CI) from observation studies only was 1.42 (1.24–1.64; fixed-effect model) and 2.44 (1.18–5.04; random-effects model), against the estimated OR (95% CI) of 2.67 (0.25–28.52) from 1 RCT.

       Publication bias

      The funnel plot (Figure 4) showed that, out of 8 data points, 6 lay outside the triangle, and 6 lay on the right side of the triangle altitude. The Egger regression test of the funnel asymmetry showed highly significant publication bias (P < .0001).
      Figure thumbnail gr4
      Figure 4Funnel plot for detection of possible bias of publication on the effect of proton pump inhibitor use on gastric polyps.

       Proton Pump Inhibitor Use and Risk of Gastric Cancer

       Pooled effect

      Of 4 included studies reporting on gastric cancer, 1 case-control study
      • Crane S.J.
      • Locke I.G.R.
      • Harmsen W.S.
      • et al.
      Subsite-specific risk factors for esophageal and gastric adenocarcinoma.
      reported zero PPI users among control subjects and thus was excluded from the analysis because an OR could not be computed. All 3 remaining studies adjusted the effect estimates for the potential confounders; however, H pylori infection status was not considered for adjustment in any of these studies. One study
      • Garcia Rodriguez L.A.
      • Lagergren J.
      • Lindblad M.
      Gastric acid suppression and risk of oesophageal and gastric adenocarcinoma: A nested case control study in the UK.
      reported ORs for gastric cancer at 2 subsites, namely gastric cardia adenocarcinoma and gastric noncardia adenocarcinoma, and reported no raw data or single OR or RR for all-subsite gastric cancer. Therefore, 4 estimates (all from observational studies) were available to be pooled, which yielded a summary RR (95% CI) of 1.43 (1.23–1.66) from both fixed- and random-effects models (Figure 2B). No significant heterogeneity across studies was present (both models: P = .53; I2 = 0%).

       Sensitivity and subgroup analyses

      The sensitivity analysis showed that the direction of the pooled effect of the use of PPIs on gastric cancer and of 95% CI of the pooled effect, and the absence of heterogeneity, were not influenced by any single study in both fixed- and random-effect models (see Supplementary Table 6).
      Pooled RRs (95% CI) from the fixed-effect model based on data from groups of patients (in 2 studies
      • Poulsen A.H.
      • Christensen S.
      • McLaughlin J.K.
      • et al.
      Proton pump inhibitors and risk of gastric cancer: A population-based cohort study.
      • Garcia Rodriguez L.A.
      • Lagergren J.
      • Lindblad M.
      Gastric acid suppression and risk of oesophageal and gastric adenocarcinoma: A nested case control study in the UK.
      ) who received PPIs <12 months, ≥12 months, and >36 months were 1.76 (1.24–2.52), 1.42 (0.98–2.07), and 2.45 (1.41–4.25), respectively. The random-effects model produced the same results, except for the pooled RR (95% CI) in the second subgroup (≥12 months), which was 1.31 (0.79–2.19) (Figure 3B).

       Publication bias

      When the number of studies is small, the visual interpretation and test for asymmetry of the funnel plot of the publications are not reliable.
      With 3 included studies reporting on gastric cancer, we did not assess the bias of these publications.

      Discussion

       Proton Pump Inhibitor Use and Risk of Gastric Polyps

      The summary effect and its 95% CI based on the estimates from 1 RCT and 7 observational studies showed that the use of PPIs was associated with an increased risk of FGPs. The likelihood of this association was strengthened by the fact that the pooled OR and its lower bound of the 95% CI from studies with higher quality were much larger than those from studies with lower quality and from all studies. However, the magnitude of the mean effect size remains unclear. Of the 8 included studies, only 3
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      • Ally M.R.
      • Veerappan G.R.
      • Maydonovitch C.L.
      • et al.
      Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
      • Zelter A.
      • Fernandez J.L.
      • Bilder C.
      • et al.
      Fundic gland polyps and association with proton pump inhibitor intake: A prospective study in 1,780 endoscopies.
      adjusted the effect estimates for potential confounders. The development of FGPs is known to be suppressed by H pylori.
      • Dickey W.
      • Kenny B.D.
      • McConnell J.B.
      Prevalence of fundic gland polyps in a western European population.
      • Sakai N.
      • Tatsuta M.
      • Hirasawa R.
      • et al.
      Low prevalence of Helicobacter pylori infection in patients with hamartomatous fundic polyps.
      • Watanabe N.
      • Seno H.
      • Nakajima T.
      • et al.
      Regression of fundic gland polyps following acquisition of Helicobacter pylori.
      However, the status of H pylori infection was considered for adjustment only in 2 studies.
      • Ally M.R.
      • Veerappan G.R.
      • Maydonovitch C.L.
      • et al.
      Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
      • Zelter A.
      • Fernandez J.L.
      • Bilder C.
      • et al.
      Fundic gland polyps and association with proton pump inhibitor intake: A prospective study in 1,780 endoscopies.
      Duration of PPI use is an important factor influencing the risk of FGPs,
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      but is not specifically reported in most of the studies. In general, there was diversity in the PPI exposure. Mean (range) of duration of PPI use was 37.4 (3–98) months in the study by Choudhry et al
      • Choudhry U.
      • Boyce Jr., H.W.
      • Coppola D.
      Proton pump inhibitor-associated gastric polyps: A retrospective analysis of their frequency, and endoscopic, histologic, and ultrastructural characteristics.
      and was 21.9 (1–97) months in study by Hsu et al.
      • Hsu W.H.
      • Wu I.C.
      • Kuo C.H.
      • et al.
      Influence of Proton Pump Inhibitor Use in Gastrointestinal Polyps.
      In other studies, mean or median duration of PPI use was not reported and the duration ranges varied widely from 5 to >48 months. In all the included studies, FGPs were found in patients undergoing an esophagogastroduodenoscopy, of whom a large proportion had gastroesophageal reflux disease (GERD). The effect of PPIs in these patients might be confounded by indication, because the patients with GERD are often prescribed PPIs and might have a higher risk of FGPs.
      • Genta R.M.
      • Schuler C.M.
      • Robiou C.I.
      • et al.
      No Association Between Gastric Fundic Gland Polyps and Gastrointestinal Neoplasia in a Study of Over 100,000 Patients.
      Because none of the included studies adjusted the effect estimates for a history or presence of GERD, their findings might be biased.
      The sensitivity analysis showed that the study by Vieth and Stolte
      • Vieth M.
      • Stolte M.
      Fundic gland polyps are not induced by proton pump inhibitor therapy.
      had a considerable impact on the pooled effect from the fixed-effect model; this is in agreement with the observation that this study
      • Vieth M.
      • Stolte M.
      Fundic gland polyps are not induced by proton pump inhibitor therapy.
      had a weight far larger than those of other studies, and its effect estimate was smallest among all studies except 1 (see Figure 2A). However, because the exclusion of the study by Vieth and Stolte
      • Vieth M.
      • Stolte M.
      Fundic gland polyps are not induced by proton pump inhibitor therapy.
      did not change the direction of the 95% CI of the pooled effect from any of the models, the likelihood of the association between PPI use and the risk of FGPs was not influenced by this study.
      • Vieth M.
      • Stolte M.
      Fundic gland polyps are not induced by proton pump inhibitor therapy.
      Because of the presence of a considerable heterogeneity across the included studies, the assumption in the fixed-effect model that the true effect is the same value in every study may not hold.
      Despite this, the pooled ORs and the 95% CIs obtained from both fixed- and random-effects models consistently showed that the use of PPIs was associated with an increased risk of FGPs.
      The results from the analysis of the subgroups by length of PPI exposure suggest that the risk of FGPs increases with increasing duration of PPI use. This subgroup analysis suggests that the use of PPI for at least 12 months is a necessary condition for the development of FGPs. This observation is in line with the results from several studies published in the form of abstract or letter, which showed an increased risk of FGPs in patients receiving PPIs for 12 months or longer.
      • Graham J.R.
      Omeprazole and gastric polyposis in humans.
      • Collins S.
      • Tydd T.
      Gastric polyps and long-term omeprazole.
      • Mogadam M.
      • Houk R.
      Long-term use of omeprazole is associated with development of benign epithelial gastric polyps.
      • Reekmans A.
      • Naegels S.
      • Urbain D.
      Fundic gland polyps and chronic PPI treatment: a prospective study.
      Despite the finding that the use of PPIs was associated with an increased risk of FGPs, the clinical significance of PPI-associated FGPs remains unclear. Benign gastric polyps have been found incidentally in 5–140 out of 1000 people undergoing an esophagogastroduodenoscopy,
      • Carmack S.W.
      • Genta R.M.
      • Schuler C.M.
      • et al.
      The current spectrum of gastric polyps: A 1-year national study of over 120,000 patients.
      • Archimandritis A.
      • Spiliadis C.
      • Tzivras M.
      • et al.
      Gastric epithelial polyps: a retrospective endoscopic study of 12974 symptomatic patients.
      • Marcial M.A.
      • Villafana M.
      • Hernandez-Denton J.
      • et al.
      Fundic gland polyps: prevalence and clinicopathologic features.
      and 3%–77% of these gastric polyps could be FGPs.
      • Carmack S.W.
      • Genta R.M.
      • Schuler C.M.
      • et al.
      The current spectrum of gastric polyps: A 1-year national study of over 120,000 patients.
      • Ally M.R.
      • Veerappan G.R.
      • Maydonovitch C.L.
      • et al.
      Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
      • Archimandritis A.
      • Spiliadis C.
      • Tzivras M.
      • et al.
      Gastric epithelial polyps: a retrospective endoscopic study of 12974 symptomatic patients.
      • Marcial M.A.
      • Villafana M.
      • Hernandez-Denton J.
      • et al.
      Fundic gland polyps: prevalence and clinicopathologic features.
      • Papa A.
      • Cammarota G.
      • Tursi A.
      • et al.
      Histologic types and surveillance of gastric polyps: a seven year clinico-pathological study.
      In the study by Choudhry et al,
      • Choudhry U.
      • Boyce Jr., H.W.
      • Coppola D.
      Proton pump inhibitor-associated gastric polyps: A retrospective analysis of their frequency, and endoscopic, histologic, and ultrastructural characteristics.
      no hyperplasia or dysplasia was found in the FGPs in 7 PPI users. In the study by Jalving et al,
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      1 and 7 patients among 19 patients with FAP-associated FGPs had high- and low-grade dysplasia in FGPs, respectively; among these 8 patients with dysplasia, only 2 patients used PPIs. Based on their data, the authors of the previously mentioned study
      • Jalving M.
      • Koornstra J.J.
      • Wesseling J.
      • et al.
      Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
      suggested that the development of FGPs and/or dysplasia in FGPs in patients with FAP was unlikely to be related to PPI use. In contrast, in 2 studies published in abstract form, Weinstein et al
      • Weinstein W.
      • Ang S.
      • Ippoliti A.
      • et al.
      Fundic gland polyps in patients on long term omeprazole therapy: a light and electron microscopic study of the gastric mucosa.
      noted that the use of omeprazole increased hyperplasia in the mucosa adjacent to the FGPs, and Hirt et al
      • Hirt M.
      • Lee S.
      • Weinstein W.
      Fundic gland polyps: a comparison of the omeprazole-associated and the sporadic types.
      suggested that FGPs associated with PPI use exhibited a histologic difference from the sporadic FGPs that were traditionally described. In another study, among patients with FAP-associated FGPs, 2 patients were on long-term PPI therapy (>6 months) and both had dysplasia; in comparison, only 3 out of 7 patients not receiving PPIs had dysplasia.
      • Attard T.M.
      • Yardley J.H.
      • Cuffari C.
      Gastric polyps in pediatrics: an 18-year hospital-based analysis.
      Apart from these observations, there has been no large, systematic study on the association between the PPI-related FGPs and the development of dysplasia or carcinoma.

       Proton Pump Inhibitor Use and Risk of Gastric Cancer

      The summary effect and its 95% CI based on the estimates from 3 observational studies suggest that the use of PPIs may be associated with an increased risk of gastric cancer. This is in agreement with the results from the study by Ahn et al,
      • Ahn J.S.
      • Eom C.S.
      • Jeon C.Y.
      • et al.
      Acid suppressive drugs and gastric cancer: A meta-analysis of observational studies.
      which examined the association between the use of acid-suppressive drugs, including histamine H2 receptor antagonists and PPIs, and the risk of gastric cancer. In their meta-analysis, the authors of this study
      • Ahn J.S.
      • Eom C.S.
      • Jeon C.Y.
      • et al.
      Acid suppressive drugs and gastric cancer: A meta-analysis of observational studies.
      used the same 3 studies
      • Poulsen A.H.
      • Christensen S.
      • McLaughlin J.K.
      • et al.
      Proton pump inhibitors and risk of gastric cancer: A population-based cohort study.
      • Garcia Rodriguez L.A.
      • Lagergren J.
      • Lindblad M.
      Gastric acid suppression and risk of oesophageal and gastric adenocarcinoma: A nested case control study in the UK.
      • Tamim H.
      • Duranceau A.
      • Chen L.Q.
      • et al.
      Association between use of acid-suppressive drugs and risk of gastric cancer: A nested case-control study.
      as those in our study and reported a very similar summary effect of PPIs and its 95% CI (1.39 [1.19–1.64]). Despite a slight difference in the results between our study and the study by Ahn et al,
      • Ahn J.S.
      • Eom C.S.
      • Jeon C.Y.
      • et al.
      Acid suppressive drugs and gastric cancer: A meta-analysis of observational studies.
      the messages are the same at first glance. However, the magnitude and the statistical significance of the association between PPI use and the risk of gastric cancer remain uncertain. In the study by Ahn et al
      • Ahn J.S.
      • Eom C.S.
      • Jeon C.Y.
      • et al.
      Acid suppressive drugs and gastric cancer: A meta-analysis of observational studies.
      the potentially biased outcomes from the included studies were briefly mentioned, but various sources of bias and uncertainty in the summary effect of PPIs were not discussed in great details. From the included studies, only 4 values of RRs were available for the meta-analysis, of which 2 RRs were not significantly different from 1. H pylori infection is a major confounding factor, because it is associated with both PPI prescription and risk of gastric cancer. The study by Poulsen et al
      • Poulsen A.H.
      • Christensen S.
      • McLaughlin J.K.
      • et al.
      Proton pump inhibitors and risk of gastric cancer: A population-based cohort study.
      included the history of H pylori eradication as a covariate in the statistical model, but this could not exclude the residual confounding by untreated H pylori infection. Except for this study, none of the included studies adjusted the effect estimate for the status of H pylori infection. Other potential confounding factors include early symptoms of gastric cancer similar to benign gastric conditions, such as GERD and peptic ulcers, which can lead to PPI prescriptions. Except for the study by Garcia Rodriguez et al,
      • Garcia Rodriguez L.A.
      • Lagergren J.
      • Lindblad M.
      Gastric acid suppression and risk of oesophageal and gastric adenocarcinoma: A nested case control study in the UK.
      which adjusted the effect estimates for a history of GERD, peptic ulcer, and dyspepsia, none of the included studies attempted to reduce this potential bias. Diet (eg, smoked foods or cured meats), inherited cancer syndromes, and family history of gastric cancer are among other potential confounders, but none of them were considered in the included studies.
      To the best of our knowledge, no meta-analyses of studies examining the association between PPI therapy and gastric cancer, except for those in the study by Ahn et al
      • Ahn J.S.
      • Eom C.S.
      • Jeon C.Y.
      • et al.
      Acid suppressive drugs and gastric cancer: A meta-analysis of observational studies.
      and our study, have been carried out. In our study, a subgroup analysis of the effect of PPIs with regard to the duration of therapy was performed. This added new information to the existing literature, because in the study by Ahn et al
      • Ahn J.S.
      • Eom C.S.
      • Jeon C.Y.
      • et al.
      Acid suppressive drugs and gastric cancer: A meta-analysis of observational studies.
      no such an analysis directed specifically at the PPI use was conducted. Our analysis showed that the pooled RR, and the lower and upper bounds of its 95% CI in patients using PPI for <12 months were larger than those in patients using PPI for 12 months or longer, but smaller than those in patients using PPI for >36 months. A speculative explanation for this observation may be that the PPI effect in the subgroup with PPI use <12 months, compared with the subgroup with PPI use ≥12 months, may be confounded to a larger extent with the effect of H pylori infection, because in the H pylori eradication procedure PPI is prescribed only for a short-term use. However, when PPIs are used for a sufficiently long time, the interaction effect between PPIs and H pylori may substantially increase the risk of gastric cancer as observed in the subgroup with PPI use >36 months. This interaction effect is supported by several studies. In a normal gastric acid environment, H pylori colonizes the gastric antrum,
      • Weeks D.L.
      • Eskandari S.
      • Scott D.R.
      • et al.
      A H+-gated urea channel: the link between Helicobacter pylori urease and gastric colonization.
      causing an antrum-predominant gastritis. Inflammation of the antral mucosa stimulates gastrin secretion, which maintains the normal or high acidic environment. In this case the patterns of H pylori colonization and gastric secretion are stable. When acid secretion is suppressed for whatever reason including the use of PPI, H pylori also colonizes the stomach body, causing a corpus-predominant gastritis,
      • Kuipers E.J.
      • Uyterlinde A.M.
      • Pena A.S.
      • et al.
      Increase of Helicobacter pylori-associated corpus gastritis during acid suppressive therapy: Implications for long-term safety.
      which may impair function of parietal cells. As a consequence, the acid-suppressive effect of PPIs is enhanced
      • Verdu E.F.
      • Armstrong D.
      • Fraser R.
      • et al.
      Effect of Helicobacter pylori status on intragastric pH during treatment with omeprazole.
      • Gillen D.
      • Wirz A.A.
      • Neithercut W.D.
      • et al.
      Helicobacter pylori infection potentiates the inhibition of gastric acid secretion by omeprazole.
      and this leads to overgrowth of non-H pylori bacteria
      • Sanduleanu S.
      • Jonkers D.
      • De Bruine A.
      • et al.
      Non-Helicobacter pylori bacterial flora during acid-suppressive therapy: Differential findings in gastric juice and gastric mucosa.
      ; the double infection with non-H pylori and H pylori bacteria causes more severe gastritis,
      • Sanduleanu S.
      • Jonkers D.
      • De Bruine A.
      • et al.
      Double gastric infection with Helicobacter pylori and non-Helicobacter pylori bacteria during acid-suppressive therapy: increase of pro-inflammatory cytokines and development of atrophic gastritis.
      which can lead to atrophic gastritis following long-term use of PPIs.
      • Kuipers E.J.
      • Lundell L.
      • Klinkenberg-Knol E.C.
      • et al.
      Atrophic gastritis and Helicobacter pylori infection in patients with reflux esophagitis treated with omeprazole or fundoplication.
      • Lundell L.
      • Havu N.
      • Miettinen P.
      • et al.
      Changes of gastric mucosal architecture during long-term omeprazole therapy: results of a randomized clinical trial.
      Corpus gastritis and gastric atrophy are well recognized as important risk factors for gastric cancer.
      • Song H.
      • Ekheden I.G.
      • Zheng Z.
      • et al.
      Incidence of gastric cancer among patients with gastric precancerous lesions: observational cohort study in a low risk Western population.
      • Uemura N.
      • Okamoto S.
      • Yamamoto S.
      • et al.
      Helicobacter pylori infection and the development of gastric cancer.
      It should be noted that the validity of this explanation depends also on the prevalence of H pylori infection in subgroups of patients, but this information was not available in the included studies. Another explanation for the inconsistency in the trend of the relationship between duration of PPI use and the risk of gastric cancer may be that the numbers of data points included in the first subgroup (PPI use <12 months) and third subgroup (PPI use >36 months) were small, which made the estimates of the summary RRs imprecise.

       Study Strengths and Limitations

      In this review, the quality of the included studies, including the bias in their results, was rigorously assessed. We used both fixed- and random-effects model to summarize the effects of PPI use on the risk of gastric polyps and cancer, and compared the results from 2 types of models.
      Our study has some limitations. The number of included studies reporting on gastric cancer was small and thus the pooled effects and the heterogeneity assessment may be imprecise, and the publication bias of these studies could not be adequately assessed.
      Because the status of H pylori infection was only partially controlled for in 1 study, we could not distinguish between an isolated PPI effect and a synergistic PPI–H pylori effect. In the subgroup analysis, we did not have enough information to classify the studies or patients based on mutually exclusive periods of PPI use; the overlap of the periods (eg, ≥12 months and >36 months) and the unequal number of data points between subgroup did not allow a firm conclusion about the trend of change in the risk of FGPs or gastric cancer with increasing lengths of PPI exposure.

      Conclusions

      Long-term use of PPIs (≥12 months) is most likely associated with an increased risk of FGPs, but the clinical significance of this is currently unclear. Therapy with PPIs may also increase the risk of gastric cancer, but this association can be biased because of a limited number of available studies and possible confounding by indication and by other factors. Given the large number of patients receiving PPIs, more and higher quality studies are needed to confirm or repudiate any causal link between PPI use and gastric cancer. In addition, the potential link between PPI-related FGPs and gastric cancer should be studied in more detail.

      Supplementary Material

      Supplementary Table 1Search Queries for MEDLINE and EMBASE (Ovid)
      NumberQuery
      1((‘proton pump’ adj inhibitor?) or PPI? or omeprazole or esomeprazole or pantoprazole or lansoprazole or dexlansoprazole or rabeprazole).ti,ot,sh,ab,kw,tw
      2(((fundic or gastric) adj2 polyps) or ((stomach or fundus) adj3 polyps)).ti,ot,sh,ab,kw,tw
      3((gastric and (cancer or carcinoma? or adenocarcinoma? or neoplasm? or neoplasia)) or (stomach adj3 (cancer or carcinoma? or adenocarcinoma? or neoplasm? or neoplasia))).ti,ot,sh,ab,kw,tw
      42 or 3
      51 and 4
      Supplementary Table 2Search Queries for Cochrane Central Register of Controlled Trials
      NumberQuery
      #1(“proton pump” next inhibitor*) or PPI* or omeprazole or esomeprazole or pantoprazole or lansoprazole or dexlansoprazole or rabeprazole
      #2((fundic or gastric) near/2 polyps) or ((stomach or fundus) near/3 polyps)
      #3(gastric and (cancer or carcinoma* or adenocarcinoma* or neoplasm* or neoplasia)) or (stomach near/3 (cancer or carcinoma* or adenocarcinoma* or neoplasms or neoplasia))
      #4#2 or #3
      #5#1 and #4
      Supplementary Table 3Assessment of the Cohort Studies Using NOS
      OutcomeFundic gland polypsGastric cancer
      CriteriaChoudhry et al, 1998Vieth and Stolte, 2001Jalving et al, 2006Ally et al, 2009Hsu et al, 2010Zelter et al, 2011Poulsen et al, 2009
      Selection
      1. Representativeness of the exposed cohort
       A. Truly representative of the average age and sex in the community
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      1
       B. Somewhat representative of the average age and sex in the community
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      111111
       C. Selected group of users (eg, nurses, volunteers)
       D. No description of the derivation of the cohorts
      2. Selection of the nonexposed cohort
       A. Drawn from the same community as the exposed cohort
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      1111111
       B. Drawn from a different source
       C. No description of the derivation of the nonexposed cohort
      3. Ascertainment of exposure
       A. Secure record (eg, surgical records)
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      11111
       B. Structured interview
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      1
       C. Written self-reportxx
       D. No description
      4. Demonstration that outcome of interest was not present at start of study
       A. Yes
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      1
       B. Noxxxxx
      Comparability
      1. Comparability of cohorts on the basis of the design or analysis
       A. Study control subjects for Helicobacter pylori infection
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      xxx1111
       B. Study control subjects for peptic ulcer, hypergastrinemia, age, smoking, and/or alcohol consumption
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      xx1x111
      Outcome
      1. Assessment of outcome
       A. Independent blind assessment
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      111111
       B. Record linkage
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      1
       C. Self-report
       D. No description
      2. Was follow-up long enough for outcomes to occur
       A. Yes (≥12 months)
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      1111111
       B. No
      3. Adequacy of follow-up of cohorts
       A. Complete follow-up: all subjects accounted for
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
       B. Subjects lost to follow-up unlikely to introduce bias: small number lost
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      1
       C. Follow-up rate <50% and no description of those lost
       D. No statementxxxxxx
      Total score5557778
      a If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Outcome categories, and a maximum of 2 points can be given for Comparability.
      Supplementary Table 4Assessment of the Case-Control Studies Using NOS
      OutcomeFundic gland polypsGastric adenocarcinomaGastric cancer
      CriteriaCao et al, 2014Garcia Rodriguez et al, 2006Crane et al, 2007Tamim et al, 2008
      Selection
      1. Is the case definition adequate?
       A. Yes, with independent validation
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
      1
       B. Yes (eg, record linkage or based on self-reports)xxx
       C. No description
      2. Representativeness of the cases
       A. Consecutive or obviously representative series of cases
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
      1111
       B. Potential for selection biases or not stated
      3. Selection of control subjects
       A. Community control subjects
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
      111
       B. Hospital control subjectsx
       C. No description
      4. Definition of control subjects
       A. No history of disease (endpoint)
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
      1
       B. No description of sourcexxx
      Comparability
      1. Comparability of cases and control subjects on the basis of the design or analysis
       A. Study control subjects for Helicobacter pylori infection
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
      xxxx
       B. Study control subjects for GERD, peptic ulcer, hypergastrinemia, age, smoking, alcohol consumption, diet, inherited cancer syndromes, family history of gastric cancer
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
      x11x
      Exposure
      1. Ascertainment of exposure
       A. Secure record (eg, surgical records)
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
      111
       B. Structured interview where blind to case/control status
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
       C. Interview not blinded to case/control status
       D. Written self-report or medical record only
       E. No descriptionx
      2. Same method of ascertainment for cases and control subjects
       A. Yes
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
      1111
       B. No
      3. Nonresponse rate
       A. Same rate for both groups
      If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
       B. Nonrespondents describedxx
       C. Rate different and no designationx
      Total score3654
      a If the criterion is met, the study is awarded 1 point. A maximum of 1 point can be given for each numbered item within the Selection and Exposure categories, and a maximum of 2 points can be given for Comparability.
      Supplementary Table 5Effect of Omitting Individual Studies on Pooled OR and Heterogeneity of Studies Reporting on Gastric Polyps in Patients Receiving PPIs Compared With Patients Receiving no PPIs
      Omitted study (reported outcome)Pooled OR (% changed)95% CI of pooled ORP value (heterogeneity test)I2
      Fixed-effect modelRandom-effects modelFixed-effect modelRandom-effects modelFixed-effect modelRandom-effects modelFixed-effect modelRandom-effects model
      Peura et al, 2009 (gastric polyps)1.42 (-0.2)2.44 (-0.1)1.24 to 1.641.18 to 5.04.00.0093.7994.88
      Choudhry et al, 1998 (fundic gland polyps)1.38 (-3)2.05 (-16.3)1.2 to 1.591.06 to 3.96.00.0092.9393.25
      Vieth and Stolte, 2001 (fundic gland polyps)2.06 (44.2)2.85 (16.4)1.68 to 2.521.36 to 5.94.00.0091.8590.38
      Jalving et al, 2006 (fundic gland polyps)1.35 (-5.4)2.50 (2.1)1.16 to 1.571.12 to 5.54.00.0093.4694.4
      Ally et al, 2009 (fundic gland polyps)1.41 (-0.9)2.57 (5.2)1.22 to 1.631.18 to 5.63.00.0093.7894.99
      Hsu et al, 2010 (fundic gland polyps)1.36 (-4.4)2.46 (0.6)1.18 to 1.581.11 to 5.45.00.0093.4894.7
      Zelter et al, 2011 (fundic gland polyps)1.22 (-14.5)1.91 (-21.8)1.05 to 1.411.04 to 3.5.00.0085.490.35
      Cao et al, 2014 (fundic gland polyps)1.62 (13.3)2.96 (21.1)1.38 to 1.891.51 to 5.82.00.0092.6590.95
      Supplementary Table 6Effect of Omitting Individual Studies on Pooled RR and Heterogeneity of Studies Reporting on Gastric Cancer in Patients Receiving PPIs Compared With Patients Receiving no PPIs
      Omitted study (reported outcome)Pooled RR (% changed)95% CI of pooled RRP value (heterogeneity test)I2
      Fixed-effect modelRandom-effects modelFixed-effect modelRandom-effects modelFixed-effect modelRandom-effects modelFixed-effect modelRandom-effects model
      Rodriguez et al, 2006 (gastric cardia adenocarcinoma)1.46 (1.9)1.46 (1.9)1.25 to 1.701.25 to 1.70.526.5260.000.00
      Rodriguez et al, 2006 (gastric noncardia adenocarcinoma)1.4 (-2.4)1.4 (-2.4)1.19 to 1.641.19 to 1.64.496.4960.000.00
      Tamim et al, 2008 (gastric cancer)1.35 (-5.4)1.35 (-5.5)1.01 to 1.811.00 to 1.82.364.3640.935.08
      Poulsen et al, 2009 (gastric cancer)1.46 (2.2)1.46 (2.2)1.24 to 1.711.24 to 1.72.453.4530.000.00
      Fixed-effect and random-effects models produced almost identical results.

      References

        • Shin J.M.
        • Sachs G.
        Pharmacology of proton pump inhibitors.
        Curr Gastroenterol Rep. 2008; 10: 528-534
        • Stolte M.
        • Sticht T.
        • Eidt S.
        • et al.
        Frequency, location, and age and sex distribution of various types of gastric polyp.
        Endoscopy. 1994; 26: 659-665
        • Odze R.D.
        • Marcial M.A.
        • Antonioli D.
        Gastric fundic gland polyps: a morphological study including mucin histochemistry, stereometry, and MIB-1 immunohistochemistry.
        Hum Pathol. 1996; 27: 896-903
        • Sipponen P.
        • Laxen F.
        • Seppala K.
        Cystic 'hamartomatous' gastric polyps: a disorder of oxyntic glands.
        Histopathology. 1983; 7: 729-737
        • Lee R.G.
        • Burt R.W.
        The histopathology of fundic gland polyps of the stomach.
        Am J Clin Pathol. 1986; 86: 498-503
        • Kinoshita Y.
        • Tojo M.
        • Yano T.
        • et al.
        Incidence of fundic gland polyps in patients without familial adenomatous polyposis.
        Gastrointest Endosc. 1993; 39: 161-163
        • Iida M.
        • Yao T.
        • Itoh H.
        • et al.
        Natural history of fundic gland polyposis in patients with familial adenomatosis coli/Gardner's syndrome.
        Gastroenterology. 1985; 89: 1021-1025
        • Graham J.
        Gastric polyposis: onset during long-term therapy with omeprazole.
        Med J Aust. 1992; 157: 287-288
        • Stolte M.
        • Bethke B.
        • Seifert E.
        • et al.
        Observation of gastric glandular cysts in the corpus mucosa of the stomach under omeprazole treatment.
        Z Gastroenterol. 1995; 33: 146-149
        • Van Vlierberghe H.
        • De Vos M.
        • De Cock G.
        • et al.
        Fundic gland polyps: three other case reports suggesting a possible association with acid suppressing therapy.
        Acta Gastroenterol Belg. 1997; 60: 240-242
        • Kazantsev G.B.
        • Schwesinger W.H.
        • Heim-Hall J.
        Spontaneous resolution of multiple fundic gland polyps after cessation of treatment with lansoprazole and Nissen fundoplication: a case report.
        Gastrointest Endosc. 2002; 55: 600-602
        • Flick A.L.
        Gastric fundal polyps with cellular atypia in a patient using esomeprazole (Nexium).
        Dig Dis Sci. 2005; 50: 2100-2102
        • Kim J.S.
        • Chae H.S.
        • Kim H.K.
        • et al.
        Spontaneous resolution of multiple fundic gland polyps after cessation of treatment with omeprazole.
        Korean J Gastroenterol. 2008; 51: 305-308
        • Spiegel A.
        • Stein P.
        • Patel M.
        • et al.
        A report of gastric fundic gland polyps.
        Gastroenterol Hepatol. 2010; 6: 45-48
        • Hegedus I.
        • Csizmadia C.
        • Lomb Z.
        • et al.
        Massive fundic gland polyposis caused by chronic proton pump inhibitor therapy. [Hungarian] Tartos protonpumpagatlo kezeles hatasara kialakulo massziv fundusmirigy-polyposis.
        Orv Hetil. 2012; 153: 351-356
        • Abraham S.C.
        • Nobukawa B.
        • Giardiello F.M.
        • et al.
        Fundic gland polyps in familial adenomatous polyposis: neoplasms with frequent somatic adenomatous polyposis coli gene alterations.
        Am J Pathol. 2000; 157: 747-754
        • Abraham S.C.
        • Nobukawa B.
        • Giardiello F.M.
        • et al.
        Sporadic fundic gland polyps: common gastric polyps arising through activating mutations in the beta-catenin gene.
        Am J Pathol. 2001; 158: 1005-1010
        • Carmack S.W.
        • Genta R.M.
        • Schuler C.M.
        • et al.
        The current spectrum of gastric polyps: A 1-year national study of over 120,000 patients.
        Am J Gastroenterol. 2009; 104: 1524-1532
        • Wu T.T.
        • Kornacki S.
        • Rashid A.
        • et al.
        Dysplasia and dysregulation of proliferation in foveolar and surface epithelia of fundic gland polyps from patients with familial adenomatous polyposis.
        Am J Surg Pathol. 1998; 22: 293-298
        • Jalving M.
        • Koornstra J.J.
        • Gotz J.M.
        • et al.
        High-grade dysplasia in sporadic fundic gland polyps: a case report and review of the literature.
        Eur J Gastroenterol Hepatol. 2003; 15: 1229-1233
        • Havu N.
        Enterochromaffin-like cell carcinoids of gastric mucosa in rats after life-long inhibition of gastric secretion.
        Digestion. 1986; 35: 42-55
        • Ali T.
        • Roberts D.N.
        • Tierney W.M.
        Long-term safety concerns with proton pump inhibitors.
        Am J Med. 2009; 122: 896-903
        • Schenk B.E.
        • Kuipers E.J.
        • Klinkenberg-Knol E.C.
        • et al.
        Hypergastrinaemia during long-term omeprazole therapy: influences of vagal nerve function, gastric emptying and Helicobacter pylori infection.
        Aliment Pharmacol Ther. 1998; 12: 605-612
        • Verdu E.F.
        • Armstrong D.
        • Fraser R.
        • et al.
        Effect of Helicobacter pylori status on intragastric pH during treatment with omeprazole.
        Gut. 1995; 36: 539-543
        • Gillen D.
        • Wirz A.A.
        • Neithercut W.D.
        • et al.
        Helicobacter pylori infection potentiates the inhibition of gastric acid secretion by omeprazole.
        Gut. 1999; 44: 468-475
        • Sanduleanu S.
        • Jonkers D.
        • De Bruine A.
        • et al.
        Non-Helicobacter pylori bacterial flora during acid-suppressive therapy: Differential findings in gastric juice and gastric mucosa.
        Aliment Pharmacol Ther. 2001; 15: 379-388
        • Sanduleanu S.
        • Jonkers D.
        • De Bruine A.
        • et al.
        Double gastric infection with Helicobacter pylori and non-Helicobacter pylori bacteria during acid-suppressive therapy: increase of pro-inflammatory cytokines and development of atrophic gastritis.
        Aliment Pharmacol Ther. 2001; 15: 1163-1175
        • Song H.
        • Ekheden I.G.
        • Zheng Z.
        • et al.
        Incidence of gastric cancer among patients with gastric precancerous lesions: observational cohort study in a low risk Western population.
        BMJ. 2015; 351: h3867
        • Wroblewski L.E.
        • Peek R.M.
        • Wilson K.T.
        Helicobacter pylori and Gastric Cancer: Factors That Modulate Disease Risk.
        Clin Microbiol Rev. 2010; 23: 713-739
      1. Higgins J.P.T. Green S. Cochrane Handbook for Systematic Reviews of Interventions: Version 5.1.0. The Cochrane Collaboration, 2011 (Available:) (accessed 25 December 2015)
      2. Wells GA, Shea B, O’Connell D, et al., editors. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. 2012. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed December 25, 2015.

        • Higgins J.
        • Thompson S.G.
        Quantifying heterogeneity in a meta-analysis.
        Stat Med. 2002; 21: 1539-1558
        • Egger M.
        • Smith G.D.
        • Schneider M.
        • et al.
        Bias in meta-analysis detected by a simple, graphical test.
        BMJ. 1997; 315: 629-634
        • Viechtbauer W.
        Conducting meta-analyses in R with the metafor package.
        J Stat Softw. 2010; 36: 1-48
        • Moher D.
        • Liberati A.
        • Tetzlaff J.
        • et al.
        Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.
        Ann Intern Med. 2009; 151: 264-269
        • Stroup D.F.
        • Berlin J.A.
        • Morton S.C.
        • et al.
        Meta-analysis of observational studies in epidemiology: a proposal for reporting.
        JAMA. 2000; 283: 2008-2012
        • Peura D.A.
        • Metz D.C.
        • Dabholkar A.H.
        • et al.
        Safety profile of dexlansoprazole MR, a proton pump inhibitor with a novel dual delayed release formulation: Global clinical trial experience.
        Aliment Pharmacol Ther. 2009; 30: 1010-1021
        • Choudhry U.
        • Boyce Jr., H.W.
        • Coppola D.
        Proton pump inhibitor-associated gastric polyps: A retrospective analysis of their frequency, and endoscopic, histologic, and ultrastructural characteristics.
        Am J Clin Pathol. 1998; 110: 615-621
        • Vieth M.
        • Stolte M.
        Fundic gland polyps are not induced by proton pump inhibitor therapy.
        Am J Clin Pathol. 2001; 116: 716-720
        • Jalving M.
        • Koornstra J.J.
        • Wesseling J.
        • et al.
        Increased risk of fundic gland polyps during long-term proton pump inhibitor therapy.
        Aliment Pharmacol Ther. 2006; 24: 1341-1348
        • Ally M.R.
        • Veerappan G.R.
        • Maydonovitch C.L.
        • et al.
        Chronic proton pump inhibitor therapy associated with increased development of fundic gland polyps.
        Dig Dis Sci. 2009; 54: 2617-2622
        • Hsu W.H.
        • Wu I.C.
        • Kuo C.H.
        • et al.
        Influence of Proton Pump Inhibitor Use in Gastrointestinal Polyps.
        Kaohsiung J Med Sci. 2010; 26: 76-83
        • Zelter A.
        • Fernandez J.L.
        • Bilder C.
        • et al.
        Fundic gland polyps and association with proton pump inhibitor intake: A prospective study in 1,780 endoscopies.
        Dig Dis Sci. 2011; 56: 1743-1748
        • Cao H.L.
        • Qu R.
        • Zhang Z.H.
        • et al.
        Sporadic fundic gland polyps are not associated with proton pump inhibitors therapy but negatively correlate with Helicobacter pylori infection in China.
        Chin Med J (Engl). 2014; 127: 1239-1243
        • Poulsen A.H.
        • Christensen S.
        • McLaughlin J.K.
        • et al.
        Proton pump inhibitors and risk of gastric cancer: A population-based cohort study.
        Br J Cancer. 2009; 100: 1503-1507
        • Garcia Rodriguez L.A.
        • Lagergren J.
        • Lindblad M.
        Gastric acid suppression and risk of oesophageal and gastric adenocarcinoma: A nested case control study in the UK.
        Gut. 2006; 55: 1538-1544
        • Crane S.J.
        • Locke I.G.R.
        • Harmsen W.S.
        • et al.
        Subsite-specific risk factors for esophageal and gastric adenocarcinoma.
        Am J Gastroenterol. 2007; 102: 1596-1602
        • Tamim H.
        • Duranceau A.
        • Chen L.Q.
        • et al.
        Association between use of acid-suppressive drugs and risk of gastric cancer: A nested case-control study.
        Drug Saf. 2008; 31: 675-684
        • Dickey W.
        • Kenny B.D.
        • McConnell J.B.
        Prevalence of fundic gland polyps in a western European population.
        J Clin Gastroenterol. 1996; 23: 73-75
        • Sakai N.
        • Tatsuta M.
        • Hirasawa R.
        • et al.
        Low prevalence of Helicobacter pylori infection in patients with hamartomatous fundic polyps.
        Dig Dis Sci. 1998; 43: 766-772
        • Watanabe N.
        • Seno H.
        • Nakajima T.
        • et al.
        Regression of fundic gland polyps following acquisition of Helicobacter pylori.
        Gut. 2002; 51: 742-745
        • Genta R.M.
        • Schuler C.M.
        • Robiou C.I.
        • et al.
        No Association Between Gastric Fundic Gland Polyps and Gastrointestinal Neoplasia in a Study of Over 100,000 Patients.
        Clin Gastroenterol Hepatol. 2009; 7: 849-854
        • Graham J.R.
        Omeprazole and gastric polyposis in humans.
        Gastroenterology. 1993; 104: 1584
        • Collins S.
        • Tydd T.
        Gastric polyps and long-term omeprazole.
        J Gastroenterol Hepatol. 1998; 13: A149
        • Mogadam M.
        • Houk R.
        Long-term use of omeprazole is associated with development of benign epithelial gastric polyps.
        Am J Gastroenterol. 1996; 91: A151
        • Reekmans A.
        • Naegels S.
        • Urbain D.
        Fundic gland polyps and chronic PPI treatment: a prospective study.
        Gastroenterology. 2001; 120: A244
        • Archimandritis A.
        • Spiliadis C.
        • Tzivras M.
        • et al.
        Gastric epithelial polyps: a retrospective endoscopic study of 12974 symptomatic patients.
        Ital J Gastroenterol. 1996; 28: 387-390
        • Marcial M.A.
        • Villafana M.
        • Hernandez-Denton J.
        • et al.
        Fundic gland polyps: prevalence and clinicopathologic features.
        Am J Gastroenterol. 1993; 88: 1711-1713
        • Papa A.
        • Cammarota G.
        • Tursi A.
        • et al.
        Histologic types and surveillance of gastric polyps: a seven year clinico-pathological study.
        Hepatogastroenterology. 1998; 45: 579-582
        • Weinstein W.
        • Ang S.
        • Ippoliti A.
        • et al.
        Fundic gland polyps in patients on long term omeprazole therapy: a light and electron microscopic study of the gastric mucosa.
        Gastroenterology. 1994; 106: A210
        • Hirt M.
        • Lee S.
        • Weinstein W.
        Fundic gland polyps: a comparison of the omeprazole-associated and the sporadic types.
        Gastroenterology. 1996; 110: A135
        • Attard T.M.
        • Yardley J.H.
        • Cuffari C.
        Gastric polyps in pediatrics: an 18-year hospital-based analysis.
        Am J Gastroenterol. 2002; 97: 298-301
        • Ahn J.S.
        • Eom C.S.
        • Jeon C.Y.
        • et al.
        Acid suppressive drugs and gastric cancer: A meta-analysis of observational studies.
        World J Gastroenterol. 2013; 19: 2560-2568
        • Weeks D.L.
        • Eskandari S.
        • Scott D.R.
        • et al.
        A H+-gated urea channel: the link between Helicobacter pylori urease and gastric colonization.
        Science. 2000; 287: 482-485
        • Kuipers E.J.
        • Uyterlinde A.M.
        • Pena A.S.
        • et al.
        Increase of Helicobacter pylori-associated corpus gastritis during acid suppressive therapy: Implications for long-term safety.
        Am J Gastroenterol. 1995; 90: 1401-1406
        • Kuipers E.J.
        • Lundell L.
        • Klinkenberg-Knol E.C.
        • et al.
        Atrophic gastritis and Helicobacter pylori infection in patients with reflux esophagitis treated with omeprazole or fundoplication.
        N Engl J Med. 1996; 334: 1018-1022
        • Lundell L.
        • Havu N.
        • Miettinen P.
        • et al.
        Changes of gastric mucosal architecture during long-term omeprazole therapy: results of a randomized clinical trial.
        Aliment Pharmacol Ther. 2006; 23: 639-647
        • Uemura N.
        • Okamoto S.
        • Yamamoto S.
        • et al.
        Helicobacter pylori infection and the development of gastric cancer.
        N Engl J Med. 2001; 345: 784-789
        • Fass R.
        • Chey W.D.
        • Zakko S.F.
        • et al.
        Clinical trial: the effects of the proton pump inhibitor dexlansoprazole MR on daytime and nighttime heartburn in patients with non-erosive reflux disease.
        Aliment Pharmacol Ther. 2009; 29: 1261-1272

      Linked Article

      • Use of Proton Pump Inhibitors and Risks of Fundic Gland Polyps and Gastric Cancer: Systematic Review and Meta-analysis
        Clinical Gastroenterology and HepatologyVol. 15Issue 5
        • Preview
          Tran-Duy et al1 suggest that the use of proton pump inhibitors (PPIs) for 12 months or longer is associated with an increased risk of fundic gland polyps and possibly gastric cancer, implying that the PPIs are involved in the pathogenesis of those outcomes.1 However, for such an analysis to be valid, absent prospective randomized trials, it would require that there be no important differences between patients receiving or not receiving PPIs. However, of course, those populations are different by definition, in that one population had underlying conditions that led to prolonged PPI use (eg, gastric hypersecretion) and the other did not.
        • Full-Text
        • PDF