Intervention Review

You have free access to this content

Surveillance of gastric intestinal metaplasia for the prevention of gastric cancer

  1. Anthony O'Connor1,*,
  2. Deirdre McNamara2,
  3. Colm A O'Moráin1

Editorial Group: Cochrane Upper Gastrointestinal and Pancreatic Diseases Group

Published Online: 23 SEP 2013

Assessed as up-to-date: 6 NOV 2012

DOI: 10.1002/14651858.CD009322.pub2


How to Cite

O'Connor A, McNamara D, O'Moráin CA. Surveillance of gastric intestinal metaplasia for the prevention of gastric cancer. Cochrane Database of Systematic Reviews 2013, Issue 9. Art. No.: CD009322. DOI: 10.1002/14651858.CD009322.pub2.

Author Information

  1. 1

    Adelaide and Meath Hospital incorporating the National Children's Hospital/Trinity College, Department of Gastroenterology and Clinical Medicine, Dublin, Ireland

  2. 2

    Adelaide and Meath Hospital/Trinity College, Dublin, Ireland

*Anthony O'Connor, Department of Gastroenterology and Clinical Medicine, Adelaide and Meath Hospital incorporating the National Children's Hospital/Trinity College, Belgard Road, Tallaght, Dublin, D24, Ireland. jpoconno@tcd.ie. jpanthonyoconnor@hotmail.com.

Publication History

  1. Publication Status: New
  2. Published Online: 23 SEP 2013

SEARCH

 

Summary of findings    [Explanations]

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms

 
Summary of findings for the main comparison.

Surveillance compared with no surveillance for gastric intestinal metaplasia

Patient or population: patients with gastric intestinal metaplasia

Settings: primary and secondary care

Intervention: upper gastrointestinal endoscopy, serum pepsinogen testing

Comparison: no surveillance

Comments

No randomised studies met the inclusion criteria.

 

Background

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms
 

Description of the condition

Adenocarcinoma of the stomach is the second leading cause of cancer related death in the world. It causes approximately 750,000 deaths on an annual basis worldwide. Two histologically distinct subtypes of gastric adenocarcinoma with distinct morphological features have been identified, namely diffuse and intestinal types. Diffuse gastric cancer (HDGC) is an inherited condition associated with an increased risk of gastric (stomach) cancer. Diffuse gastric cancer is a specific type of stomach cancer that tends to affect much of the stomach rather than staying in one area of the stomach. The much commoner intestinal type cancers tend to be less heritable and are more associated with dietary factors, smoking and alcohol, having a causal relationship with Helicobacter pylori (H pylori) infection. This pathogen is usually acquired in childhood and has a prolonged latent phase. In this extended premalignant period, a well characterised series of neoplastic changes occur. This was described by Correa as consisting of a progression from normal gastric mucosa to chronic active gastritis, on to atrophic gastritis, and then to intestinal metaplasia, dysplasia and invasive carcinoma (Correa 1975).

The prevalence of intestinal metaplasia (IM) in the community varies depending on the rate of H pylori infection in the population. One northern European group estimates the prevalence as 19% (Eriksson 2008). Other environmental variables exist, such as tobacco smoking. A multicentre international study showed that the prevalence of IM among H pylori-infected participants varied a great deal from region to region, with a rate of 3% in Argentina compared to 55% in New Zealand. An interesting correlation was observed between IM prevalence in infected persons and the availability of tobacco in the community (Peleteiro 2008).

The rate of progression of IM to gastric cancer was reported to be 0.1% to 0.9% in large cohorts (de Vries 2008). This is broadly comparable with the rate of progression of other premalignant conditions of the digestive tract, such as Barrett’s oesophagus, adenomatous colonic polyps and cirrhosis of the liver. In these conditions, however, there are consensus guidelines on surveillance to prevent outright neoplasia.

Early gastric cancer, also called superficial spreading cancer, is defined as adenocarcinoma that is limited to the gastric mucosa or submucosa regardless of whether or not regional lymph nodes are involved (T1) (Murakami 1971). These adenocarcinomas represent a subset of gastric cancers that have a favourable prognosis (Edge 2009). This has led some to argue that structured screening and surveillance programmes could reduce the significant burden of gastric cancer mortality (Tepes 2009). It has been suggested that when premalignant conditions are surveyed earlier then cancerous lesions are found, which translates to a survival benefit (Whiting 2002). The eradication of H pylori after endoscopic resection of early gastric cancer can then be used to prevent the development of metachronous gastric carcinoma (Fukase 2008).

 

Why it is important to do this review

There is no clear consensus on the need for surveillance of gastric intestinal metaplasia nor are there any internationally recognised guidelines. Given the advances that have been made in the minimally-invasive management of dysplastic and early neoplastic stomach cancers in recent years, there is a definite imperative to seek these lesions so that intervention can be initiated. There are opportunities to achieve this endoscopically, radiologically and by use of serum biomarkers.

 

Objectives

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms

To see whether or not endoscopic or biochemical surveillance of patients with gastric intestinal metaplasia (GIM) could result in increased detection of dysplasia and early gastric cancer to decrease gastric cancer mortality.

 

Methods

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms
 

Criteria for considering studies for this review

 

Types of studies

Randomised clinical trials irrespective of blinding, publication status, language or size of the cohort.

 

Types of participants

We included participants with recognised features of GIM in any part of the stomach, both focal and extensive, complete and incomplete. We included people whose IM was diagnosed both opportunistically through a symptomatic service or by population based screening. GIM will be defined by histologic recognition of glands with features of intestinal phenotype. Participants will be included regardless of documented H pylori status as infection may not be detectable once metaplasia has developed. The effect of H pylori eradication is being assessed by another Cochrane review (Moayyedi 2006). We excluded patients with familial syndromes such as autosomal dominant hereditary diffuse gastric cancer.

 

Types of interventions

We compared surveillance versus no surveillance of patients with GIM. Surveillance will have been ongoing for at least five years. Two major surveillance interventions were considered.

  • Oesophago-gastro-duodenoscopy with at least five gastric biopsies.
  • Measurement of serum pepsinogens: using either pepsinogen I to II ratio, applying a cut-off ≤ 3; or serum pepsinogen I level alone with a cut-off ≤ 70 µg/litre.
  • A combination of the above.

The potential for harm from endoscopy and biopsy is low and complications are rare. Potential harmful occurrences would include an abnormal reaction to sedatives, bleeding from the biopsy or accidental puncture of the upper gastrointestinal (GI) tract. More common side effects would include trapped wind, sore throat and transient drowsiness from sedation.

 

Types of outcome measures

 

Primary outcomes

  • Gastric cancer mortality

 

Secondary outcomes

  • Progression of intestinal metaplasia to dysplasia
  • Progression of intestinal metaplasia to intestinal type neoplasia
  • All-cause mortality
  • If cost-effectiveness data can be harvested we will also analyse this using the incremental cost-effectiveness ratio
  • Whether the difference in surveillance modalities affects the advancement of or mortality from gastric cancer
  • Intervention related adverse events cannot be meta-analysed so they cannot be analysed as a secondary outcome, however this is something that will be useful to practitioners and we will comment on it in the final text

 

Search methods for identification of studies

 

Electronic searches

We searched the following electronic databases:

  • Cochrane Central Register of Controlled trials (CENTRAL) in The Cochrane Library (Appendix 1) (October 2012);
  • MEDLINE (Appendix 2) (1946 to October week 4 2012);
  • EMBASE (Appendix 3) (1980 to Week 44 2012);
  • LILACS (Appendix 4) (6 November 2012).

 

Searching other resources

We also searched the abstracts from the following congresses from 2003 to 2012:

  • Digestive Diseases Week (published in Gastroenterology and Gastrointestinal Endoscopy),
  • United European Gastroenterology Week (published in Gut and Endoscopy),
  • Asian Pacific Digestive Disease Week (Abstract book),
  • European Helicobacter Pylori Study Group (published in Helicobacter), and
  • abstracts from the International Gastric Cancer Congress.

 

Data collection and analysis

 

Selection of studies

Two authors (AOC, DMcN) independently and without blinding identified and reviewed potentially relevant articles to determine if they fulfilled the inclusion criteria. A third author (COM) adjudicated on whether or not studies should be included when a consensus could not be reached. Where publications were available in abstract form only, every effort was to be made to contact the authors to determine whether eligibility criteria were met.

 

Data extraction and management

The two unblinded review authors used a specially designed assessment form to extract the following data.

  • General information (title, authors, source, year of publication, full text or abstract, language).
  • Design of trial.
  • Surveillance intervention used (pepsinogen, endoscopy, radiology).
  • Duration of follow-up.
  • Criteria for diagnosis of IM.
  • Study population (number of patients; baseline characteristics; smoking history if available; region i.e. Northern, Eastern, Meditteranean Europe, North America, South America, Africa, Asia).
  • Primary outcomes: gastric cancer mortality.
  • Secondary outcomes: progression of IM to dysplasia, progression of IM to intestinal type neoplasia, all-cause mortality. If cost-effectiveness data can be harvested we will also analyse this using the incremental cost-effectiveness ratio, whether the difference in surveillance modalities affect the advancement or mortality of gastric cancer.

The data were assembled into 2 x 2 tables for each of the measured variables (for example gastric cancer mortality versus survival x surveillance versus control; dysplasia versus no dysplasia x surveillance versus control; progression to neoplasia versus no progression x surveillance versus control all-cause mortality versus survival x surveillance versus control). The presence of significant heterogeneity was tested by the Chi² test with P < 0.1 being considered statistically significant.

 

Assessment of risk of bias in included studies

The two unblinded review authors assessed study quality. If there was disagreement the third review author adjudicated. The quality of the studies were reviewed according to the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). The criteria used included the following.

  • Allocation concealing and blinding.
  • Method of randomisation
    • low risk of bias: computer generated random numbers etc.;
    • high risk of bias: birth date, patient self-selection.
  • Method of tissue sampling at endoscopy
    • low risk of bias: biopsies from at least four of antrum, corpus, incisura, fundus, cardia;
    • unclear: not stated, unspecified;
    • high risk of bias: biopsies only taken from areas suspected of containing IM based on clinicians' impression under white light or narrow band imaging or chromendoscopy.
  • Reference ranges for pepsinogen I and II assay
    • low risk of bias: less than or equal to 3.0.
  • Methods for barium X-ray screening
    • consistent protocol;
    • experienced interpreter.
  • Losses to follow-up
    • proportion of patients lost to follow-up;
    • whether or not explanation offered.

 

Measures of treatment effect

The primary effect measure would have been risk ratio (RR). Odds ratios (OR) were to be used for sensitivity analysis.

 

Unit of analysis issues

Repeated observations on participants were to be dealt with by using only the latest available outcome data.

 

Dealing with missing data

Efforts were to be made to contact the authors of studies with missing data. As rates of gastric cancer and progression of premalignant lesions vary significantly in different regions of the world, we feel it is more appropriate to omit missing data than apply a statistical model to make assumptions as these may be highly unreliable. The potential effect of this will be examined as a sensitivity analysis.

 

Assessment of heterogeneity

Heterogeneity was to be measured using the I² statistic, and we will consider anything more than 50% to represent substantial heterogeneity. A fixed-effect model was to be used for primary analysis and a random-effects model will be used for sensitivity analysis to check that the conclusions are robust.

 

Assessment of reporting biases

The 2 x 2 tables were to be combined into a summary test statistic using the pooled relative risk and 95% confidence interval (CI). We would have performed adjusted rank correlation (Begg 1994) and the regression asymmetry test (Egger 1997) for detection of bias. Funnel plots were to be carried out and trials registers checked to detect publication bias.

 

Data synthesis

We were to perform meta-analysis according to the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). For statistical analysis we will use RevMan 5 (Revman 2011).

 

Subgroup analysis and investigation of heterogeneity

Subgroup analyses were to be carried out on whether or not the intestinal metaplasia was focal (goblet cells within one foveola or gland), moderate (goblet cells within two to five foveola or glands), or extensive (goblet cells within more than five glands). This is to analyse whether or not the surveillance yield is greater from more extensive lesions.

 

Sensitivity analysis

As a sensitivity analysis, where substantial heterogeneity exists a random-effects meta-analysis would have been performed in addition to the fixed-effect analysis to see whether the conclusions are affected.

 

Results

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms
 

Description of studies

See: 'Characteristics of excluded studies' table

 

Results of the search

Electronic searches retrieved 1862 reports of studies. The details of the search results are shown in Figure 1.

 FigureFigure 1. Study flow diagram.

 

Included studies

No studies met the inclusion criteria.

 

Excluded studies

None of the reports of studies retrieved were suitable for inclusion. None of the excluded studies were randomised. None compared the outcomes of a surveillance group versus a non-surveillance group. Two studies compared the cancers found in people in a surveillance group and their survival versus cancers discovered at open access endoscopy (Rokkas 1991; Whiting 2002). A further study examined the biopsy follow-up of patients with intestinal metaplasia of the cardia only (Goldstein 2000).

 

Risk of bias in included studies

There were no included studies.

 

Effects of interventions

See:  Summary of findings for the main comparison

There were no included studies.

 

Discussion

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms
 

Current practices and guidelines

Guidelines for the surveillance of GIM have been published by several international societies. The American Society for Gastrointestinal Endoscopy (ASGE) does not uniformly recommend surveillance but does recommend surveillance of those with a family history or ethnic predisposition to gastric cancer (Hirota 2006). They stop short of suggesting an appropriate surveillance interval. This is based on a review carried out on patients in the US, which argued that there is a low risk of progression of GIM without dysplasia (Fennerty 2003). The ASGE guidelines suggest surveillance for those with low grade dysplasia and that histological mapping of the stomach is necessary for those undergoing surveillance. They advocate resection of high grade dysplastic lesions. Other national guidelines also reflect these, such as those in Israel (Vilkin 2011).

Guidelines from the European Society for Gastrointestinal Endoscopy (ESGE) have only recently been prepared. These recommend that patients with IM or atrophy in both antrum and corpus should be offered surveillance every three years after diagnosis, but that those with antral disease only should not be followed up (Dinis-Ribeiro 2012). A key question relates to how the initial diagnosis of GIM is made. Both sets of guidelines advocate a mapping biopsy when surveillance is undertaken and the ESGE guidelines recommend routine biopsy at all endoscopies. If we are to accept that there is unanimity amongst all guidelines that it is not possible to reliably diagnose GIM endoscopically and that biopsy of both corpus and antrum is necessary to evaluate the metaplastic stomach then it would appear that sampling from both the antrum and corpus is mandatory at all upper GI endoscopies. The Japanese guidelines on H pylori recommend such a strategy (Asaka 2010). This would have an added benefit as it would facilitate the diagnosis of H pylori infection, the treatment of which can half the progression of IM as the distribution of colonisation is very much unequal when GIM has become established (Enomoto 1998; Satoh 1998). In Asia, where gastric cancer is commonest, there are no established international guidelines but surveillance is commonly undertaken with practices varying between regions. However, a recent survey from Korea suggested that 95% of specialists perform annual endoscopic surveillance for patients with IM (Shin 2012).

 

The risk of progression of intestinal metaplasia (IM)

Heretofore, one of the main obstacles to considering surveillance programmes for GIM was the lack of data on the rate of progression of the lesions. However, several studies have emerged over the last few years to address this question. In a low-prevalence western cohort of 92,500 patients with premalignant gastric lesions the annual incidence of gastric cancer is 0.1% for patients with atrophic gastritis, 0.25% for patients with IM, 0.6% for patients with mild to moderate dysplasia, and 6% for patients with severe dysplasia (de Vries 2008). This study also examined the surveillance practices of Dutch doctors. It was found that patients who underwent surveillance were significantly younger and tended to have biannual endoscopic examinations.

A study from a southern European cohort, where rates of gastric cancer are higher, showed the incidence rate of gastric cancer for patients with IM was 16.5 per 1000 person years (Gonzalez 2010). The rate of progression in Asian countries would appear to be greater, with one Japanese study showing that the relative risk of progression from IM to cancer was 6.4 (95% CI 2.6 to 16.1) (Uemera 2001). A major determinant of progression would appear to be the topographic extent of the metaplasia lesion. Lesions extending along the lesser curvature from the cardia to pylorus (the so-called ‘magenstraße’ distribution) and with diffuse patterns are associated with higher risks of gastric cancer, with ORs of 5.7 and 12.2 respectively (Cassaro 2000).

 

Outcomes of endoscopic surveillance of GIM

Two large studies completed in the UK have examined the effect of surveillance for GIM. One study looked at surveillance of type III IM only (intermediate cells secreting predominantly sulphomucins and goblet cells secreting sialo- or sulphomucins, or both). Twenty-six patients in this study underwent follow-up with endoscopy and biopsy at intervals between six and 12 months. Of this cohort 11 developed early gastric cancer (defined as malignancy confined to the mucosa and submucosa) and four were lost to follow-up, with the other 11 remaining under surveillance at the end of the 12-year study period. At an institutional level, in the unit that carried out the study this led to markedly increased rates of detection of early cancers in the second half of the study (Rokkas 1991).

The second study involved annual surveillance of patients with a multiplicity of premalignant conditions including GIM, atrophic gastritis, dysplasia, polyps and foveolar hyperplasia. Within this group there were 185 patients with GIM, of whom 93 opted for surveillance. It was found that the five-year survival of people with cancers detected by surveillance (n = 14) was significantly higher (50% versus 10%) than for patients detected at open access endoscopy. Over the study period of 10 years, 10 of the cancers were detected on follow-up of patients with GIM (10.75%) (Whiting 2002).

The outlook for surveillance may be changing with emerging technologies. Two recent publications have examined the use of narrow-band imaging in a surveillance context. In one study from the Netherlands on 47 patients with GIM who were undergoing surveillance the sensitivity, specificity, positive and negative predictive values for detection of premalignant lesions were 71%, 58%, 65% and 65% for narrow band imaging and 51%, 67%, 62% and 55% for white light endoscopy, respectively (Capelle 2010). The use of pepsinogens as a non-invasive surrogate biopsy has also been shown to predict progression of premalignant lesions. A study of nearly 7000 patients in Japan who received annual endoscopy showed that a serum pepsinogen level of < 70 ng/ml and a pepsinogen I and II ratio of < 3.0 predicted a greater risk of developing gastric cancer regardless of H pylori status (Watabe 2005). Devising strategies for the early detection of gastric cancer is only useful if this substantially alters the outcome of the disease.

If left untreated, early gastric cancer does not usually cause clinical problems for several years and may not cause any symptoms during the patient's natural lifespan (Tsukuma 2000; Ubukata 2004). In such a circumstance, extensive surgery such as gastric resection may not be justifiable, and this again casts doubt over the utility of surveillance. Emerging technologies such as endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) allow for feasible minimally invasive curative interventions for dysplastic and early cancerous lesions. Japanese studies have confirmed these satisfactory cure rates with one 10-year follow-up of a cohort of people with early gastric cancer who were treated with EMR showing overall five- and 10-year survival rates of 84% and 64%, respectively, and disease-specific five- and 10-year survival rates of 99% and 99% (Uedo 2006). It is the advancement of therapeutic endoscopic modalities that makes early gastric cancer diagnosis most desirable.

 

Effect of H pylori eradication in relevant studies

Most authorities suggest that any programmes of surveillance aimed at gastric cancer prevention in patients with premalignant lesions of the stomach should be married to measures aimed at H pylori eradication (Correa 2010; Malfertheiner 2012; Tepes 2009). In a study of 246 H pylori-positive patients who received eradication therapy there was significant reduction of atrophy in the gastric corpus but not in the antrum. There was no progression of IM. In the control group of 306 infected patients who had placebo there was significant progression of IM (Yang 2009; Zhou 2003).

The diminished benefit from eradication when metaplasia has become established is probably due to lower levels of H pylori colonisation by that time. Furthermore, the strain of H pylori harboured by the individual undergoing surveillance may also have implications. In one study of patients with atrophy and metaplasia with a mean follow-up period of 12 years, H pylori strains harbouring cagA, vacA s1, and vacA m1 genotypes were more frequently found in patients with more advanced gastric preneoplastic lesions. In addition, infection with cagA-positive, vacA s1, and vacA m1 strains was associated with progression of gastric preneoplastic lesions (OR 2.28, OR 2.90 and OR 3.38, respectively). Infection with strains that are simultaneously cagA positive and vacA s1/m1 was associated with progression of gastric precancerous lesions with an OR 4.80 (95% CI 1.71 to 13.5) in relation to those infected with cagA-negative and vacA s2 or m2 strains (Gonzalez 2011).

 

Cost-effectiveness of surveillance

The cost-effectiveness of any surveillance is contingent on the prevalence of gastric cancer in the community and is more likely to be economically viable in high-prevalence regions. The cost-effectiveness may also depend on whether or not H pylori eradication is undertaken at the same time in affected patients, as this may decrease the burden of dyspepsia consultations and ulcer complications. A decision-analysis model was conducted based on a 10-year risk of transformation to outright cancer of 1.8% (de Vries 2008) and annual endoscopy. It presupposed a down-staging of gastric cancer by a factor of 58% to 84% (Whiting 2002). It concluded, based on US figures, that the numbers of gastroscopies required to detect one cancer and to prevent one gastric cancer related death in the surveillance arm were 556 and 3738 respectively, and that the incremental cost-effectiveness ratio of endoscopic surveillance as compared to a non-surveillance policy was $72,519 per life-year gained, which was therefore considered cost-effective (Hassan 2010).

A major weakness of this study was the fact that the improved survival and down-staging of early cancers found at surveillance was based on figures from Japanese endoscopists who were more adept at the diagnosis and endoscopic management of these lesions but the other matters were based on western patients and practices. Another decision model based on US figures shows that EMR followed by surveillance of middle-aged males with dysplasia was cost-effective, but that less advanced lesions such as GIM would be less so as 10-yearly surveillance followed by EMR for early cancer for those with IM would reduce lifetime cancer risk by 61% at a cost of $544,500 per quality-adjusted life year (QALY) (Yeh 2009). A European study based on three-yearly surveillance with the more modern technique of magnification chromendoscopy and serum pepsinogen testing showed, for individuals with atrophic chronic gastritis or complete IM, a yearly measurement of pepsinogen levels or an endoscopic examination on a three-yearly basis would cost EUR 455 per QALY gain. Endoscopic examination and pepsinogen serum level measurement on a yearly basis would cost EUR 1868 per QALY for patients with extensive IM (Dinis-Ribeiro 2007).

In reality, given the varying prevalences of H pylori and gastric cancer worldwide, as well as issues around access to and reimbursement for interventions, it will be very difficult to construct a cost-effectiveness model in this condition that will suffice on a global basis.

 

Adverse effects of surveillance

One potential adverse effect of surveillance would be an increase in cancer related anxiety. While data do not exist pertaining to GIM surveillance, one study suggested that patients undergoing endoscopic surveillance for Barrett's oesophagus (which would have a very similar endoscopic surveillance intervention and risk of progression) suffer anxiety and have impaired quality of life (Cooper 2009). This, however, could possibly be overcome with adequate patient education and empowerment. There are also financial imperatives for patients regarding the need to occasionally take time off work for surveillance appointments, although these would be very infrequent. Conceivably there could also be adverse outcomes for patients undergoing surveillance regarding insurance premiums.

Other adverse effects would be specific to the surveillance modality used. Endoscopic surveillance involves regular upper GI endoscopy. While this is usually a very safe procedure it does come with some common side effects that are usually trivial and transient, such as throat discomfort and flatus. More serious adverse events can also occur, including perforation, although these are very rare and complicate only around 0.05% of diagnostic upper GI endoscopies (Quine 1995). As the aim of surveillance is to identify more advanced precancerous lesions that merit excision, this would probably result in an increased number of therapeutic endoscopy procedures such as endoscopic submucosal dissection (ESD). The perforation and delayed bleeding rates for gastric ESD have been reported to range from 1.2% to 5.2% and 0% to 15.6%, respectively (Oda 2013). The risks of surveillance using pepsinogen testing are lower as fewer endoscopies would be carried out and the risks of venepuncture are very small, but include pain, bruising and phlebitis.

 

Summary of discussion

While some centres, usually with a research interest in the topic, have reported the clinical outcomes of their surveillance of GIM, and others have retrospectively analysed outcomes of patients with GIM, the two have very rarely been compared, and never in a randomized manner. Therefore, no evidence from randomised studies exists to support surveillance of gastric intestinal metaplasia (GIM). It is essential to consider the differences in the value of surveillance (and in the practicalities of conducting trials to outline this) by looking at both high and low gastric cancer prevalence regions. In regions with a high prevalence of gastric cancer, such as Japan, endoscopic practices for the detection and management of premalignant lesions of the stomach are sophisticated. In such a setting it stands to reason that surveillance would be beneficial to patients. However, the practicalities of proving this on a randomised basis in these regions would probably be insurmountable. As gastric cancer is such an important cause of mortality in these regions it would be difficult to ethically justify a trial where patients with a known premalignant condition were randomised into a trial which did not include surveillance. On the other hand, in low gastric cancer prevalence regions it is likely that the 'number needed to surveil' to pick up an early cancer is likely to be significantly greater. Once an early cancer is identified, the expertise in the minimally-invasive management of that lesion may not be as sophisticated as it would be in higher-prevalence regions, which would mean that the cost-benefit ratio of surveillance would be entirely different, and it would follow that a randomized trial may yield less impressive results than in a high-prevalence region. In the absence of randomised studies though it is impossible to quantify the benefits and harms of treatment. Therefore clinical practice should be informed by local guidelines, patient preference and clinical expertise. In high-prevalence areas it would be regarded as unethical to withhold surveillance, given the high mortality associated with gastric cancer and the potential for endoscopy to identify early cases, which are known to carry a better prognosis. It may be the case that surveillance of premalignant lesions in isolation may be an inadequate response to the challenge of gastric cancer prevention. Surveillance with endoscopy, pepsinogens, or both may be most useful as part of an overall suite of measures to reduce the risk of gastric cancer in this population, including H pylori screening and eradication, dietary measures or even the use of chemoprophylactic agents. There may be concerns about the acceptability of such measures for patients in low-prevalence countries, to prevent a disease which is perceived as being rare. In addition, the economic burden and opportunity cost for endoscopic and laboratory services may be prohibitive.

 

Summary of main results

There were no included studies.

 

Overall completeness and applicability of evidence

There were no included studies.

 

Quality of the evidence

There were no included studies.

 

Potential biases in the review process

There were no included studies.

 

Agreements and disagreements with other studies or reviews

There were no included studies.

 

Authors' conclusions

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms

 

Implications for practice

No evidence from randomised studies exists to support surveillance of gastric intestinal metaplasia. Therefore, clinical practice should be informed by local guidelines, patient preference and clinical expertise.

 
Implications for research

From a scientific perspective, there is a clear need for a long-term randomised study which compares a surveillance cohort to a non-surveillance cohort for the prevention of cancers in patients with premalignant lesions of the stomach. Such a study would have H pylori eradication at point of entry as the evidence for this measure for the prevention of cancers is now widely accepted. It would need to focus on outcomes such as early cancer diagnosis and survival, as well as adverse events associated with surveillance and a cost-benefit analysis. It would need to identify which patients are most likely to benefit from surveillance and would ideally be carried out in centres with a clinical and basic science research interest in H pylori and gastric cancer, with appropriate endoscopic, surgical and oncological resources to respond to the lesions found so they are treated according to international best practices. However, given the serious implications of a diagnosis of gastric cancer, it would be unfair to approach a patient with a small but definite increased risk of cancer to propose randomization into an arm where surveillance was not offered. It would be anticipated that recruitment and the consent process for such a randomised controlled trial would be an insurmountable challenge. As such, it is difficult to imagine how the question can be answered by that particular study design. Currently, surveillance practices are variable and surveillance is normally only routinely undertaken in centres where a special interest in the condition exists. It may be that prospective cohort studies of patients undergoing such surveillance, powered to detect a variance in tumour nodes metastasis (TNM) status at diagnosis and also of gastric cancer mortality, are needed, which could then be compared to overall population data from cancer registries.

 

Acknowledgements

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms

The authors thank all the staff of the Cochrane Upper Gastrointestinal and Pancreatic Diseases (UGPD) Group for their support and assistance.

 

Data and analyses

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms

This review has no analyses.

 

Appendices

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms
 

Appendix 1. CENTRAL search strategy

  1. (carcin$ or cancer$ or neoplas$ or tumour$ or tumor$ or cyst$ or growth$ or adenocarcin$ or malig$).mp.
  2. (Digest$ or Gastr$ or epigastr$ or stomach$).mp.
  3. 1 and 2
  4. (Metaplasia adj5 (Digest$ or Gastr$ or intestin$ or epigastr$ or stomach$)).mp.
  5. Gastric Mucosa/pa [Pathology]
  6. abdominal neoplasms/ or peritoneal neoplasms/ or stomach neoplasms/
  7. or/3-6
  8. Sentinel Surveillance/
  9. (Metaplasia adj2 surveillance).mp.
  10. surveillance endoscop*.mp.
  11. "Early Detection of Cancer"/mt [Methods]
  12. Precancerous Conditions/
  13. Neoplasm Staging/
  14. Endoscopy, Gastrointestinal/
  15. Gastroscopy/
  16. (Oesophago-gastro-duodenoscop* or esophago-gastro-duodenoscop*).mp.
  17. (Oesophagogastroduodenoscop* or esophagogastroduodenoscop*).mp.
  18. Barium Sulfate/
  19. Barium X-Ray*.mp.
  20. (Pepsinogen adj5 ratio*).mp.
  21. (Pepsinogen I or pepsinogen II).mp.
  22. or/8-21
  23. 7 and 22

 

Appendix 2. MEDLINE search strategy

  1. (carcin$ or cancer$ or neoplas$ or tumour$ or tumor$ or cyst$ or growth$ or adenocarcin$ or malig$).mp.
  2. (Digest$ or Gastr$ or epigastr$ or stomach$).mp.
  3. 1 and 2
  4. (Metaplasia adj5 (Digest$ or Gastr$ or intestin$ or epigastr$ or stomach$)).mp.
  5. Gastric Mucosa/ab, co, pp [Abnormalities, Complications, Physiopathology]
  6. abdominal neoplasms/ or peritoneal neoplasms/ or stomach neoplasms/
  7. or/3-6
  8. Sentinel Surveillance/
  9. (Metaplasia adj2 surveillance).mp.
  10. surveillance endoscop*.mp.
  11. "Early Detection of Cancer"/mt [Methods]
  12. Precancerous Conditions/pc [Prevention & Control]
  13. Neoplasm Staging/mt [Methods]
  14. Endoscopy, Gastrointestinal/mt [Methods]
  15. Gastroscopy/mt [Methods]
  16. (Oesophago-gastro-duodenoscop* or esophago-gastro-duodenoscop*).mp.
  17. (Oesophagogastroduodenoscop* or esophagogastroduodenoscop*).mp.
  18. Barium Sulfate/du [Diagnostic Use]
  19. Barium X-Ray*.mp.
  20. (Pepsinogen adj5 ratio*).mp.
  21. (Pepsinogen I or pepsinogen II).mp.
  22. or/8-21
  23. 7 and 22
  24. randomized controlled trial.pt.
  25. controlled clinical trial.pt.
  26. randomized.ab.
  27. placebo.ab.
  28. drug therapy.fs.
  29. randomly.ab.
  30. trial.ab.
  31. groups.ab.
  32. or/24-31
  33. exp animals/ not humans.sh.
  34. 32 not 33
  35. 23 and 34

 

Appendix 3. EMBASE search strategy

  1. (carcin$ or cancer$ or neoplas$ or tumour$ or tumor$ or cyst$ or growth$ or adenocarcin$ or malig$).mp.
  2. (Digest$ or Gastr$ or epigastr$ or stomach$).mp.
  3. 1 and 2
  4. (Metaplasia adj5 (Digest$ or Gastr$ or intestin$ or epigastr$ or stomach$)).mp.
  5. stomach mucosa/co [Complication]
  6. exp abdominal tumor/ or stomach tumor/
  7. *cancer staging/
  8. or/3-7
  9. Sentinel Surveillance/
  10. (Metaplasia adj2 surveillance).mp.
  11. surveillance endoscop*.mp.
  12. precancer/di, pc, su, th [Diagnosis, Prevention, Surgery, Therapy]
  13. gastrointestinal endoscopy/
  14. gastroscopy/
  15. (Oesophago-gastro-duodenoscop* or esophago-gastro-duodenoscop*).mp.
  16. (Oesophagogastroduodenoscop* or esophagogastroduodenoscop*).mp.
  17. barium sulfate/
  18. Barium X-Ray*.mp.
  19. (Pepsinogen adj5 ratio*).mp.
  20. (Pepsinogen I or pepsinogen II).mp.
  21. or/9-20
  22. 8 and 21
  23. random:.tw. or placebo:.mp. or double-blind:.tw.
  24. 22 and 23

 

Appendix 4. LILACS search strategy

((Metaplasia) AND (intestin$)) [Words] and (Surveillance or Endoscopy or Gastroscopy or Pepsinogen) OR (Endoscopy, Digestive System/) OR (Gastrectomy/methods) [Words]

 

Appendix 5. Glossary of terms

Adenocarcinoma

Cancer that begins in glandular (secretory) cells. Glandular cells are found in tissue that lines certain internal organs and makes and releases substances in the body, such as mucus, digestive juices, or other fluids.

Adenomatous colonic polyps

Small growths in the large bowel that are not cancer but may be associated with an increased risk of developing cancer. Start in gland-like cells of the epithelial tissue (thin layer of tissue that covers organs, glands, and other structures within the body).

Antrum

portion before the outlet, which is lined by mucosa and does not produce acid.

Atrophic

Pertaining to a wasting or decrease in size of a body organ, tissue, or part owing to disease, injury, or lack of use.

Autosomal

Pertaining to a chromosome that is not a sex chromosome.

Biochemical

Pertaining to the science dealing with the chemistry of living matter.

Biomedical

Pertaining to the application of the natural sciences, especially the biological and physiological sciences, to clinical medicine.

Carcinoma

Cancer that begins in the skin or in tissues that line or cover internal organs

Cardia

The cardia is the anatomical term for the part of the stomach attached to the oesophagus.

Cirrhosis

A type of chronic, progressive liver disease in which liver cells are replaced by scar tissue.

Chemoprophylactic

The use of drugs to prevent disease.

Chromendoscopy

Chromoendoscopy involves the topical application of stains or pigments to improve tissue localisation, characterisation, or diagnosis during endoscopy.

Cohort

A group of individuals who share a common trait.

Cost per life-year saved

A number used in risk analysis to compare the cost-benefit ratios of different interventions in health and safety. As death is inevitable, it focusses on the avoidance of premature death. In this context, were a patient with an ordinary life expectancy of 80 years to be diagnosed with early stage gastric cancer at the age of 55 and saved, compared with a late stage cancer that might carry a survival prognosis of 1 year, this would represent 24 life years saved. This would be the denominator in a cost per life-year saved equation. The overall individualised cost of the intervention for that patient, including the cost of screening for the condition and the treatment and rehabilitation involved would represent the numerator.

Corpus

The body of the stomach.

Diffuse

Spread out over a large area; not concentrated.

Dysplasia

An alteration in size, shape, and organization of adult cells.

Endoscopic

Pertaining to a procedure that uses an endoscope to examine the inside of the body. An endoscope is a thin, tube-like instrument with a light and a lens for viewing. It may also have a tool to remove tissue to be checked under a microscope for signs of disease.

Endoscopic mucosal resection

Endoscopic mucosal resection (EMR) is a technique used for the staging and treatment of superficial neoplasms of the gastrointestinal tract.

Endoscopic submucosal dissection

Endoscopic submucosal dissection (ESD) is an advanced technique of therapeutic endoscopy for superficial gastrointestinal tumours involving the mucosa.

Foveolar

Foveolar cells are mucus producing cells which cover the inside of the stomach, protecting it from the corrosive nature of gastric acid.

Gastric

Pertaining to the stomach.

Gastritis

Inflammation of the lining of the stomach.

Goblet cells

A column-shaped cell found in the respiratory and intestinal tracts, which secretes the main component of mucus.

Helicobacter pylori

A type of bacterium that causes inflammation and ulcers in the stomach or small intestine. People with Helicobacter pylori infections may be more likely to develop cancer in the stomach. Also called H pylori.

Histologically

Pertaining to the study of tissues and cells under a microscope.

Metachronous

Two or more cancers appearing at different points in time.

Metaplasia

A change of cells to a form that does not normally occur in the tissue in which it is found.

Morphological

Pertaining to the form and structure of an organism considered as a whole.

Mortality

A term used for death rate, or the number of deaths in a certain group of people in a certain period of time.

Mucosa

The moist, inner lining of some organs and body cavities (such as the nose, mouth, lungs, and stomach).

Neoplastic

Pertaining to neoplasia, which is abnormal and uncontrolled cell growth, commonly associated with cancer.

Pathogen

A bacterium, virus, or other microorganism that can cause disease.

Pepsinogen

A substance made by cells in the stomach. Acid in the stomach changes pepsinogen to pepsin, which breaks down proteins in food during digestion.

Phenotype

The set of observable characteristics of an individual resulting from the interaction of its genotype with the environment.

Premalignant

A term used to describe a condition that may (or is likely to) become cancer. Also called precancerous.

Pylorus

The opening from the stomach into the duodenum (small intestine).

Serum biomarkers

Biological molecules found in blood, other body fluids, or tissues that is a sign of a normal or abnormal process, or of a condition or disease.

Sulphomucins

A mucin containing sulfuric esters in its mucopolysaccharides or glycoproteins.

Surveillance

A treatment plan that involves closely watching a patient’s condition but not giving any treatment unless there are changes in test results that show the condition is getting worse.

Topographic

The description of an anatomic region or a special part.

vacAs1 and vacAm1 genotypes

Key toxins of the H pylori pathogen that help determine the disease outcome for infected patients.

 

Contributions of authors

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms

Conceiving the review: AOC

Designing the review: AOC

Coordinating the review: COM

Data collection for the review: AOC, DMcN

Designing search strategies: AOC, DMcN

Undertaking searches: AOC, DMcN

Screening search results: DMcN

Organizing retrieval of papers: AOC

Screening retrieved papers against inclusion criteria: DMcN

Appraising quality of papers: COM

Extracting data from papers: AOC

Writing to authors of papers for additional information: AOC

Providing additional data about papers: AOC

Obtaining and screening data on unpublished studies: AOC

Data management for the review: AOC

Entering data into RevMan: AOC

Analysis of data: AOC, DMcN

Interpretation of data: DMcN, COM, AOC

Providing a methodological perspective: AOC

Providing a clinical perspective: COM

Providing a policy perspective: DMcN

Providing a consumer perspective: AOC

Writing the review: AOC

Providing general advice on the review: COM

Securing funding for the review: COM

 

Declarations of interest

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms

The authors declare no conflicts of interest.

 

Sources of support

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms
 

Internal sources

  • Adelaide and Meath Hospital incorporating the National Children's Hospital/Trinity College, Ireland.

 

External sources

  • No sources of support supplied

 

Differences between protocol and review

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. Contributions of authors
  13. Declarations of interest
  14. Sources of support
  15. Differences between protocol and review
  16. Index terms

None

References

References to studies excluded from this review

  1. Top of page
  2. AbstractRésumé
  3. Summary of findings
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' conclusions
  10. Acknowledgements
  11. Data and analyses
  12. Appendices
  13. Contributions of authors
  14. Declarations of interest
  15. Sources of support
  16. Differences between protocol and review
  17. Characteristics of studies
  18. References to studies excluded from this review
  19. Additional references
Goldstein 2000 {published data only}
Rokkas 1991 {published data only}
  • Rokkas T, Filipe MI, Sladen GE. Detection of an increased incidence of early gastric cancer in patients with intestinal metaplasia type III who are closely followed up. Gut 1991;32(10):1110-3.
Whiting 2002 {published data only}

Additional references

  1. Top of page
  2. AbstractRésumé
  3. Summary of findings
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' conclusions
  10. Acknowledgements
  11. Data and analyses
  12. Appendices
  13. Contributions of authors
  14. Declarations of interest
  15. Sources of support
  16. Differences between protocol and review
  17. Characteristics of studies
  18. References to studies excluded from this review
  19. Additional references
Asaka 2010
  • Asaka M,  Kato M,  Takahashi S,  Fukuda Y,  Sugiyama T,  Ota H, et al. Guidelines for the management of Helicobacter pylori infection in Japan: 2009 revised edition. Helicobacter 2010;15(1):1-20.
Begg 1994
Capelle 2010
  • Capelle LG,  Haringsma J,  de Vries AC,  Steyerberg EW,  Biermann K,  van Dekken H, et al. Narrow band imaging for the detection of gastric intestinal metaplasia and dysplasia during surveillance endoscopy. Digestive Diseases and Sciences 2010;55(12):3442-8.
Cassaro 2000
  • Cassaro M, Rugge M, Gutierrez O, Leandro G, Graham DY, Genta RM. Topographic patterns of intestinal metaplasia and gastric cancer. American Journal of Gastroenterology 2000;95(6):1431-8.
    Direct Link:
Cooper 2009
  • Cooper SC, El-agib A, Dar S, Mohammed I, Nightingale P, Murray IA, et al. Endoscopic surveillance for Barrett's oesophagus: the patients' perspective. European Journal of Gastroenterology and Hepatology 2009;21(8):850-4.
Correa 1975
  • Correa P, Haenszel W, Cuello C, Tannenbaum S, Archer M. A model for gastric cancer epidemiology. Lancet 1975;2(7924):58-60.
Correa 2010
  • Correa P,  Piazuelo MB,  Wilson KT. Pathology of gastric intestinal metaplasia: clinical implications. American Journal of Gastroenterology 2010;105(3):493-8.
de Vries 2008
  • de Vries AC, van Grieken NC, Looman CW, Casparie MK, de Vries E, Meijer GA, et al. Gastric cancer risk in patients with premalignant gastric lesions: a nationwide cohort study in the Netherlands. Gastroenterology 2008;134(4):945-52.
Dinis-Ribeiro 2007
Dinis-Ribeiro 2012
  • Dinis-Ribeiro M, Areia M, de Vries AC, Marcos-Pinto R, Monteiro-Soares M, O'Connor A, et al. Management of precancerous conditions and lesions in the stomach (MAPS): guideline from the European Society of Gastrointestinal Endoscopy (ESGE), European Helicobacter Study Group (EHSG), European Society of Pathology (ESP), and the Sociedade Portuguesa de Endoscopia Digestiva. Endoscopy 2012;44(1):74-94.
Edge 2009
  • Edge SB, Byrd DR, Compoton CC, Fritz AG, Greene FL, Trotti A (Editors). AJCC Cancer Staging Manual. 7th Edition. New York: Springer, 2009.
Egger 1997
  • Egger M, Smith GD, Phillips AN. Meta-analysis: Principles and procedures. BMJ 1997;315(7121):1533-7.
Enomoto 1998
  • Enomoto H, Watanabe H, Nishikura K, Umezawa H, Asakura H. Topographic distribution of Helicobacter pylori in the resected stomach. European Journal of Gastroenterology and Hepatology 1998;10:473-8.
Eriksson 2008
Fennerty 2003
Fukase 2008
  • Fukase K, Kato M, Kikuchi S, Inoue K, Uemura N, Okamoto S, et al. Effect of eradication of Helicobacter pylori on incidence of metachronous gastric carcinoma after endoscopic resection of early gastric cancer: an open-labelled, randomised control trial. Lancet 2008;372:350-2.
Gonzalez 2010
Gonzalez 2011
  • González CA,  Figueiredo C,  Lic CB,  Ferreira RM,  Pardo ML,  Ruiz Liso JM, et al. Helicobacter pylori cagA and vacA genotypes as predictors of progression of gastric preneoplastic lesions: a long-term follow-up in a high-risk area in Spain. American Journal of Gastroenterology 2011;106(5):867-74.
Hassan 2010
Higgins 2011
  • Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org. Chichester: John Wiley & Sons.
Hirota 2006
  • Hirota WK, Zuckerman MJ, Adler DG, Davila RE, Egan J, Leighton JA, et al. ASGE guideline: the role of endoscopy in the surveillance of premalignant conditions of the upper GI tract. Gastrointestinal Endoscopy 2006;23(4):570-80.
Malfertheiner 2012
  • Malfertheiner P, Megraud F, O'Morain CA, Atherton J, Axon AT, Bazzoli F, et al. Management of Helicobacter pylori infection--the Maastricht IV/ Florence Consensus Report. Gut 2012;61(5):646-64.
Moayyedi 2006
  • Moayyedi P, Hunt R, Forman D. Helicobacter pylori eradication for the prevention of gastric neoplasia. Cochrane Database of Systematic Reviews 2006, Issue 1. [DOI: 10.1002/14651858.CD005583]
Murakami 1971
  • Murakami T. Pathomorphological diagnosis: definition and gross classification of early gastric cancer. In: Murakami T editor(s). Gann Monography on Cancer Research. Vol. 11, Tokyo: University of Tokyo Press, 1971:53-66.
Oda 2013
  • Oda I, Suzuki H, Nonaka S, Yoshinaga S. Complications of gastric endoscopic submucosal dissection. Digestive Endoscopy 2013;25 Suppl 1:71-8.
Peleteiro 2008
  • Peleteiro B, Bastos J, Barros H, Lunet N. Systematic review of the prevalence of gastric intestinal metaplasia and its area-level association with smoking. Gaceta Sanitaria 2008;22(3):236-7.
Quine 1995
Revman 2011
  • The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011.
Satoh 1998
  • Satoh K, Kimura K, Taniguchi Y, Kihira K, Takimoto T, Saifuku K, et al. Biopsy sites suitable for the diagnosis of Helicobacter pylori infection and the assessment of the extent of atrophic gastritis. American Journal of Gastroenterology 1998;93:569-73.
    Direct Link:
Shin 2012
  • Shin WG,  Kim HU,  Song HJ,  Hong SJ,  Shim KN,  Sung IK, et al. The Korean College of Helicobacter and Upper Gastrointestinal Research. Surveillance strategy of atrophic gastritis and intestinal metaplasia in a country with a high prevalence of gastric cancer. Digestive Diseases and Sciences 2012;57(3):746-52.
Tepes 2009
Tsukuma 2000
Ubukata 2004
Uedo 2006
  • Uedo N,  Iishi H,  Tatsuta M,  Ishihara R,  Higashino K,  Takeuchi Y, et al. Longterm outcomes after endoscopic mucosal resection for early gastric cancer. Gastric Cancer 2006;9(2):88-92.
Uemera 2001
  • Uemura N,  Okamoto S,  Yamamoto S,  Matsumura N,  Yamaguchi S,  Yamakido M, et al. Helicobacter pylori infection and the development of gastric cancer. New England Journal of Medicine 2001;345(11):784-9.
Vilkin 2011
  • Vilkin A, Half B, Moshkowitz M, Kariv R, Levi Z. Surveillance of patients with gastric intestinal metaplasia and dysplasia. Harefuah 2011;150(3):266-8.
Watabe 2005
  • Watabe H,  Mitsushima T,  Yamaji Y,  Okamoto M,  Wada R,  Kokubo T, et al. Predicting the development of gastric cancer from combining Helicobacter pylori antibodies and serum pepsinogen status: a prospective endoscopic cohort study. Gut 2005;54(6):764-8.
Yang 2009
  • Yang P, Zhou Y, Chen B, Wan HW, Jia GQ, Bai HL, et al. Overweight, obesity and gastric cancer risk: results from a meta-analysis of cohort studies. European Journal of Cancer 2009;45(16):2867-73.
Yeh 2009
Zhou 2003
  • Zhou L, Sung JJ, Lin S, Jin Z, Ding S, Huang X, et al. A five-year follow-up study on the pathological changes of gastric mucosa after H. pylori eradication. Chinese Medical Journal 2003;116(1):11-4.