NEW ASPECTS OF THE DIAGNOSIS OF

HELICOBACTER PYLORI INFECTION

Lea Veijola

Department of Bacteriology and ImmunologyHaartman InstituteUniversity of Helsinki

Academic Dissertation

To be publicly discussed with the permission of the Medical Faculty of the Universityof Helsinki in the Auditorium XII, third floor in the main building of the University,Unioninkatu 34, Helsinki,on October 13th, 2007, at 12 noon.

2

Supervised by

Docent Hilpi Rautelin, MD, PhD

Department of Bacteriology and ImmunologyHaartman Institute, University of HelsinkiFinland

Reviewed by

Docent Olli Meurman, MD, PhD

Clinical Microbiology LaboratoryTurku University HospitalFinland

Docent Tuomo Karttunen, MD, PhD

Department of PathologyUniversity of OuluFinland

To be discussed with

Docent Katri Kaukinen, MD, PhD

Medical School, University of Tampere andDepartment of Gastroenterology and Alimentary Tract Surgery, TampereUniversity HospitalFinland

ISBN 978-952-92-2660-3 (pbk.)ISBN 978-952-10-4188-4 (PDF)Helsinki University Printing HouseHelsinki 2007

3

TABLE OF CONTENTSList of original publications 5Abbreviations 6Abstract 7Introduction 9

REVIEW OF THE LITERATURE 11

History of Helicobacter pylori 11

Bacteriology and epidemiology of H. pylori 11

1. Bacteriology 111.1. Virulence factors 12

2. Epidemiology 13

H. pylori associated diseases 13

1. H. pylori induced gastritis and atrophic gastritis 141.1. Gastritis 141.2. Atrophic gastritis 15

1.2.1. The role of H. pylori in autoimmune gastritis 151.2.2. The mechanism behind H. pylori induced autoimmunity 161.2.3. Histological assessment of gastritis and atrophy 171.2.4. Serum tests in assessment of atrophy and autoimmunity 17

1.3. Dyspepsia 182. Peptic ulcer 18

2.1. Epidemiology 182.2. Pathogenesis 192.3. Duodenal gastric metaplasia 20

2.3.1. Regression of DGM 212.4. H. pylori eradication in peptic ulcer disease 212.5. Non-steroidal anti-inflammatory drugs and peptic ulcer 222.6. Vagotomy 23

3. Mucosa associated lymphoid tissue lymphoma 234. Gastric cancer 24

4.1. Gastric carcinogenesis 254.2. H. pylori eradication to prevent gastric cancer 25

5. Extragastric diseases 26

4

Diagnostic methods of H. pylori 27

1. Invasive methods 291.1. Rapid urease test 291.2. Histology 301.3. Culture and antimicrobial susceptibility 311.4. Molecular methods 32

2. Noninvasive methods 332.1. Blood-based tests 33

2.1.1. Enzyme immunoassay 332.1.2. Near patient tests 342.1.3. Immunoblot assay 342.1.4. 13C-urea blood test 35

2.2. Urea breath test 352.2.1. Different UBT procedures 352.2.2. The clinical usage of UBT 36

2.3. Stool tests 372.3.1. Stool culture 372.3.2. Stool antigen tests 372.3.3. PCR of stool specimens 38

2.4. Urine tests 382.5. Vomit and saliva 39

3. Diagnosis of H. pylori in special clinical situations 393.1. Bleeding peptic ulcer 393.2. Children 403.3. Elderly 413.4. Atrophic gastritis 41

H. pylori eradication therapy 42

Mucosal recovery after H. pylori eradication 43

Host response to H. pylori 43

PRESENT STUDY 45Aims 45Patients 46Methods 48Results 51Discussion 60Conclusions 67Acknowledgements 68References 70Original publications 95

5

LIST OF ORIGINAL PUBLICATIONS

This thesis is based on the following articles which are referred to in the text by their

Roman numerals, and reprinted by the permission of the copyright holders.

I Veijola L, Oksanen A, Löfgren T, Sipponen P, Karvonen A-L, Rautelin H.

Comparison of three stool antigen tests in confirming Helicobacter pylori

eradication in adults. Scand J Gastroenterol 2005;40:395-401.

II Veijola L, Myllyluoma E, Korpela R, Rautelin H. Stool antigen tests in the

diagnosis of Helicobacter pylori infection before and after eradication

therapy. World J Gastroenterol 2005;11:7340-7344.

III Veijola L, Sankila A, Rautelin H, Kosunen TU, Sipponen P, Hyvärinen H,

Tilvis R, Sarna S, Arkkila PET, Seppälä K. Clinical significance of

widespread gastric metaplasia in the duodenal bulb. J Clin Gastroenterol

2006;40:510-514.

IV Veijola L, Oksanen A, Linnala A, Sipponen P, Rautelin H. Persisting

chronic gastritis and elevated Helicobacter pylori antibodies after

successful eradication therapy. Helicobacter, in press.

6

ABBREVIATIONSbid bis in die (twice a day)

CI confidence interval

DGM duodenal gastric metaplasia

DNA deoxyribonucleic acid

EIA enzyme immunoassay

H. pylori Helicobacter pylori

HSV highly selective vagotomy

IgA Immunoglobulin A

IgG Immunoglobulin G

MALT mucosa associated lymphoid tissue

NSAID non-steroidal anti-inflammatory drug

OD optical density

OR odds ratio

PA pernicious anaemia

PCA parietal cell antibodies

PCR polymerase chain reaction

PG pepsinogen

PPI proton pump inhibitor

qid quater in die (four times a day)

RR risk ratio

RNA ribonucleic acid

RUT rapid urease test

tid ter in die (three times a day)

UBT urea breath test

7

ABSTRACT

Aims: Helicobacter pylori infection, although the prevalence is declining in Western

world, is still responsible for several clinically important diseases. None of the

diagnostic tests is perfect and in this study, the performance of three stool antigen

tests was assessed. In areas of high H. pylori prevalence, the definition of patients

with the greatest benefit from eradication therapy may be a problem; the role of

duodenal gastric metaplasia in categorizing patients at risk for duodenal ulcer was

evaluated in this respect. Whether persistent chronic inflammation and elevated H.

pylori antibodies after successful eradication are associated with each other or with

atrophic gastritis, a long term sequelae of H. pylori infection, were also studied.

Patients and methods: The three stool antigen tests were assessed in pre- and post-

eradication settings among 364 subjects in two studies as compared to the rapid

urease test (RUT), histology, culture, the 13C-urea breath test (UBT) and enzyme

immunoassay (EIA) based H. pylori serology. The association between duodenal

gastric metaplasia with duodenal ulcer was evaluated in a retrospective study

including 1054 patients gastroscopied due to clinical indications and 154 patients

previously operated for duodenal ulcer. The extent of duodenal gastric metaplasia was

assessed from histological specimens in different patient groups formed on the basis

of gastroscopy findings and H. pylori infection. Chronic gastric inflammation (108

patients) and H. pylori antibodies and serum markers for atrophy (77 patients) were

assessed in patients earlier treated for H. pylori.

Results: Of the stool antigen tests studied, the monoclonal antibody-based EIA-test

showed the highest sensitivity and specificity both in the pre-treatment setting (96.9%

and 95.9%) and after therapy (96.9% and 97.8%). The polyclonal stool antigen test

and the in-office test had at baseline a sensitivity of 91% and 94%, and a specificity of

96% and 89%, respectively and in a post-treatment setting, a sensitivity of 78% and

91%, and a specificity of 97%, respectively. Duodenal gastric metaplasia was strongly

associated with H. pylori positive duodenal ulcer (odds ratio 42). Although common

still five years after eradication, persistent chronic gastric inflammation (21%) and

8

elevated H. pylori antibodies (33%) were neither associated with each other nor with

atrophic gastritis.

Conclusions: Current H. pylori infection can feasibly be diagnosed by a monoclonal

antibody-based EIA test with the accuracy comparable to that of reference methods.

The performance of the polyclonal test as compared to the monoclonal test was

inferior especially in the post-treatment setting. The in-office test had a low

specificity for primary diagnosis and hence positive test results should probably be

confirmed with another test before eradication therapy is prescribed. The presence of

widespread duodenal gastric metaplasia showed promising results in detecting

patients who should be treated for H. pylori due to an increased risk of duodenal

ulcer. If serology is used later on in patients with earlier successfully treated for H.

pylori, it should be taken into account that H. pylori antibodies may persist elevated

for years for unknown reason. However, this phenomenon was not found to be

associated with persistent chronic inflammation or atrophic changes.

9

INTRODUCTION

Since the discovery of Helicobacter pylori (H. pylori) in 1982 (Warren and Marshall,

1983) the associated diseases and the benefits of eradication therapy in many of them

have been clarified. Recently, updated guidelines on several topics concerning the

treatment indications and diagnosis have been published by the European

Helicobacter Pylori Study Group as Maastricht III Consensus Report (Malfertheiner et

al., 2007).

There are several methods available for the diagnosis of H. pylori infection, but none

of them is ideal and therefore, in many clinical situations several tests are needed.

Most of the tests originally developed for the diagnosis of H. pylori infection in

patients with peptic ulcer disease are now used in different settings: peptic ulcer

prevalence is rapidly declining and the prevention of gastric cancer at the

precancerous stage has become relatively more important. Also, the prevalence of H.

pylori infection is diminishing (Rehnberg-Laiho et al., 2001) and this has a great

impact even on the performance of those tests with high sensitivity and specificity.

Gastroscopy has become a commonplace diagnostic investigation in western countries

but at the same time the waiting list for this procedure has not necessarily shortened.

On the other hand, the increasing use of proton pump inhibitors (PPIs), although

efficiently alleviating abdominal symptoms hamper the endoscopic appearance and H.

pylori diagnosis (Mégraud and Lehours, 2007). In areas with high H. pylori

prevalence, the indications for eradication therapy must be allocated to patients with

the greatest benefit, one such group being patients with duodenal ulcer. It is easy to

detect an acute ulcer during gastroscopy, however, in patients with ulcer diathesis on

PPI therapy the scars in the duodenal bulb might be minimal and the diagnosis may be

overlooked. Clinically useful tools to identify patients at risk for duodenal ulcer are

needed.

In a subgroup of patients after H. pylori eradication therapy, chronic inflammation of

the gastric mucosa as well as elevated H. pylori antibodies prevail for long periods.

The mechanism behind these phenomena is obscure as is their possible association

10

with each other or with clinical consequences e.g. atrophic gastritis, autoimmune

phenomena induced by H. pylori, or even gastric cancer risk.

Although the past 25 years of the H. pylori era have dramatically changed our

understanding and treatment of several gastric diseases, many questions still remain

open and need to be answered.

11

REVIEW OF THE LITERATURE

HISTORY OF H. PYLORI

Spiral bacteria in gastric scrapings from dogs were described as early as in 1892 by

Bizzozero at a meeting in Turin (Bizzozero, 1893, Egan and O'Morain, 2007). The

first human report published was by Krienitz from a patient with gastric cancer in

1906 (Krienitz, 1906) and at that time spiral bacteria were also considered in the

development of gastric and duodenal ulceration, sometimes patients were even treated

with high doses of bismuth (Kusters et al., 2006). However, it was also suggested that

those bacteria were only contaminants from the oral cavity (Egan and O'Morain,

2007). The modern era began when H. pylori was cultivated from gastric mucosa and

it’s association with gastritis and later with peptic ulcer was demonstrated by Warren

and Marshall, who were awarded with the Nobel Prize in Medicine 2005 (Warren and

Marshall, 1983, Marshall and Warren, 1984).

BACTERIOLOGY AND EPIDEMIOLOGY OF H. PYLORI

1. Bacteriology

Helicobacter and Wolinella comprise the family Helicobacteraceae which in turn

together with Campylobacteraceae belong to the Epsilonproteobacteria, class nov.

(Mégraud and Lehours, 2007). Thirty different Helicobacter species have been

formally named with a growing list of species waiting for formal recognition (List of

Prokaryotic Names with Standing in Nomenclature (http://www.bacterio.cict.fr)

Accessed 18th May 2007).

H. pylori is a Gram negative spiral shaped multi-flagellated bacteria found

almost exclusively on human gastric mucosa. Under unfavourable circumstances it

can become coccoidal, a nonculturable form with debatable viability. The bacterium

is a microaerophilic and capnophilic organism, slowly growing with rigorous culture

demands (Mégraud and Lehours, 2007). The phenotypic identification is based on the

growth of small, circular, smooth colonies observed after 3 to 4 days on selective

12

media plated with gastric biopsy specimens. The identification of the cultured bacteria

is based on positive reactions for oxidase, catalase, and urease. So far, the complete

genomes of three H. pylori strains have been published (Tomb et al., 1997, Alm et al.,

1999, Oh et al., 2006b, Josenhans et al., 2007). H. pylori is genetically very

heterogenous (Kusters et al., 2006), and this genetic diversity of different H. pylori

strains was even exploited in a study suggesting that humans harboured H. pylori as

early as 58,000 years ago before migrations from Africa (Linz et al., 2007).

1.1. Virulence factors

Several virulence factors of H. pylori have been identified. The most studied is CagA,

which according to numerous studies, is associated with peptic ulcers, precancerous

conditions and gastric cancer in the Western world (Kusters et al., 2006). CagA

protein is a highly immunogenic protein encoded by the cagA gene, present in

approximately 50 to 70% of H. pylori strains. It is a marker for the presence of a

genomic pathogenicity island PAI, encoding 27 to 31 proteins, 18 of which serve as

building blocks of the type IV secretory apparatus, fundamental in the aberration of

host immune responses (Odenbreit et al., 2000, Kusters et al., 2006, Robinson et al.,

2007). Another well-known virulence factor, cytotoxin VacA is associated with peptic

ulcer and adenocarcinoma. It is encoded by the gene vacA which is present in all

strains, but because of several alleles various amounts of toxin are produced and

approximately 50% of H. pylori strains secrete the toxin VacA. The most virulent

genotype s1/m1 of vacA is clearly associated with the cagA positive genotype

(Kusters et al., 2006). The gene dupA is associated with duodenal ulceration and has

been associated with the decreased risk for gastric atrophy and cancer (Robinson et

al., 2007). Adhesin BabA is associated with peptic ulcers and gastric cancer (Kusters

et al., 2006), as is outer inflammatory protein OipA (Robinson et al., 2007). SabA

might be crucial in the control of the host immune response by H. pylori (Kusters et

al., 2006). Lipopolysaccharide contains oligosaccharide antigens structurally and

immunologically closely related to human blood group antigens, and have been

suggested to promote autoantibodies and to play a role in gastric autoimmunity

(Kusters et al., 2006). However, the significance of all these virulence factors is

debatable since neither CagA nor VacA seropositivity is a better predictor of

13

increased peptic ulcer or stomach cancer risk than H. pylori antibodies (González et

al., 2003, Mégraud and Lehours, 2007).

2. Epidemiology

It has been estimated that approximately 50% of the world’s population is infected

with H. pylori (Horvitz and Gold, 2006). Transmission of the infection may occur via

gastro-oral, oral-oral, and fecal-oral routes, but it is not clear which is the predominant

mode of transmission (Delport and van der Merwe, 2007, Kusters et al., 2006,

Parsonnet et al., 1999). In general the infection is acquired during childhood

(Rehnberg-Laiho et al., 1998, Rowland et al., 2006, Malaty, 2007) and the source of

the infection is most likely the family members (Konno et al., 2005, Malaty, 2007).

The incidence of new H. pylori infections among adults in the Western world is less

than 0.5% per year (Kusters et al., 2006) and spontaneous disappearance of the

infection is rare (Kosunen et al., 1997, Malaty, 2007). The prevalence of the infection

is correlated with low socioeconomic status during childhood, high density of living,

and low household income (Malaty, 2007). The infection is steadily declining in

developed countries (Rehnberg-Laiho et al., 2001, Malaty, 2007, Kusters et al., 2006)

and its prevalence in children under the age 20 in Finland is only 5.6% (Rehnberg-

Laiho et al., 1998) as compared with 66% in Finnish centenerians (Rehnberg-Laiho et

al., 1999). This decreasing H. pylori incidence in children leads to declining

seropositivity rate in successive birth cohorts, so called cohort phenomenon.

H. PYLORI ASSOCIATED DISEASES

Since the discovery of H. pylori the associated diseases have been under intense

research. It’s now well established that only a minority of H. pylori infected persons

develop peptic ulcer or gastric cancer. The recent assumption of who develops disease

is determined largely by the inflammatory response to infection, which in turn is

determined by the virulence of the infecting strain, host genetic predisposition to

disease and environmental co-factors (Robinson et al., 2007). The negative

association with H. pylori and adenocarcinoma of the cardia has evoked concern of

14

the possible detrimental effect of widespread policy of H. pylori eradication (Blaser,

1999b). There is also a hypothesis of regarding the protective effect of H. pylori

against orally ingested pathogens; and the long symbiosis between humans and H.

pylori has evoked an opinion of H. pylori being a commensal and not causing an

infection at all (Blaser, 1999b). Since atopy and autoimmunity have become more

common in developed countries as the prevalence of H. pylori has decreased, it has

been speculated that the immune response induced by H. pylori could play a general

immunoregulatory role contributing to the hygiene hypothesis (Kosunen et al., 2002,

Robinson et al., 2007).

1. H. pylori induced gastritis and atrophic gastritis

1.1. Gastritis

From the self inoculation studies it can be concluded that acute H. pylori infection in

adults is thought to be symptomatic, and to provoke acute gastritis with

polymorphonuclear infiltration in the gastric mucosa and transient hypochlorhydria

(Sobala et al., 1991, Graham et al., 2004b, Robinson et al., 2007). Since the

spontaneous clearance of the infection is rare (Valle et al., 1996, Kosunen et al.,

1997), the infection proceeds in most subjects to chronic gastritis. This inflammatory

response is mediated by cytokines, and the magnitude of the cytokine response is

influenced by genetic polymorphism. An individual’s ultimate clinical outcome is

dependent on the cytokine response and on the gastric acid secretion. Interleukin 1-

beta is in this context one of the most important pro-inflammatory cytokines and the

most powerful acid inhibitor known (El-Omar et al., 2000, Lochhead and El-Omar,

2007). An increase in stimulated acid production predisposes to duodenal ulceration

and decreased acid production predisposes to corpus gastritis or pangastritis which in

turn predisposes to gastric ulceration, atrophic gastritis, and gastric carcinoma

(Lochhead and El-Omar, 2007, Robinson et al., 2007). The intragastric distribution of

gastritis is thought to be dependent on host genetic factors, bacterial virulence factors

and environmental factors including age at onset of infection (Robinson et al., 2007).

15

1.2. Atrophic gastritis

Atrophic gastritis is defined as loss of appropriate glands (Rugge et al., 2002). It is

often accompanied by intestinal metaplasia, which is recognized morphologically by

the presence of goblet and absorptive cells or cells resembling colonocytes or by its

mucin content depending on the type of intestinal metaplasia (Dixon et al., 1996,

Lauwers, 2003). It is estimated that about 50% of H. pylori infected persons develop

atrophic gastritis during their lifetime (Lauwers, 2003). Atrophic gastritis and

intestinal metaplasia are considered to be precancerous conditions and there is a

significant association between the extent of intestinal metaplasia, its phenotype

(intestinal, incomplete, or colonic), and increased cancer risk (Lauwers, 2003). The

cancer risk is 5 to 90-fold depending on the extent and severity of atrophy (Sipponen

and Graham, 2007).

Chronic atrophic gastritis has been classically divided into type A,

autoimmune corpus-dominated gastric atrophy, and type B H. pylori associated

multifocal atrophic gastritis affecting both antrum and corpus (Dixon et al., 1996,

Pérez-Pérez, 1997). However, the division into two entities has now been impugned

(see below). About 10 to 15% of patients with atrophic gastritis develop pernicious

anaemia (PA), which is a malabsorption of vitamin B12 leading to characteristic

anaemia. There is a clear familial tendency to develop autoimmune gastritis (Baxter et

al., 2005).

1.2.1. The role of H. pylori in autoimmune gastritis

Accumulating data suggest that gastric corpus atrophy is caused by H. pylori driven

autoimmune process (Annibale et al., 2001, Bergman et al., 2005, Oh et al., 2006a),

although the role of H. pylori as a trigger in autoimmune gastritis is debatable (Faller

et al., 1997, Pérez-Pérez, 1997, Hershko et al., 2005, Malfertheiner et al., 2007). A

substantial proportion of patients with autoimmune gastritis has or has had H. pylori

infection (Bergman et al., 2005, Annibale et al., 2007). During long follow-up,

patients with H. pylori gastritis may develop autoimmune type gastritis with elevated

parietal cell antibodies (PCA) and a normal antrum mucosa (Valle et al., 1996). In a

recent population screening study, almost all patients with pre-autoimmune type, but

16

not yet atrophic, gastritis had active H. pylori infection (Storskrubb et al., 2005). With

an increase in the grade of corpus atrophy, H. pylori seropositivity declines but the

prevalence of PCA increases (Oksanen et al., 2000). In patients with atrophic corpus

gastritis the eradication of H. pylori may restore the atrophic changes and function of

the corpus mucosa (Annibale et al., 2002) the longer the follow-up period (Kokkola et

al., 2002). Thus, atrophic corpus gastritis seems not to proceed at least after

eradication therapy. Patients with positive H. pylori serology as the only marker of the

infection had a higher grade of atrophy and more often pernicious anaemia as

compared with patients with current infection detected by other methods (Kokkola et

al., 2000). Seronegative patients had the highest grade of atrophy and PA, supporting

the hypothesis that H. pylori indeed induces autoimmune gastritis but with increasing

grade of atrophy, H. pylori disappears spontaneously (Karttunen et al., 1991, Kokkola

et al., 2003) and PA is associated with a high grade of corpus atrophy (Bergman et al.,

2005). However, among type 1 diabetics and patients with juvenile autoimmune

thyroid disease, there are patients with high titres of PCA associated with gastric

autoimmunity but no signs of H. pylori infection (Segni et al., 2004, De Block et al.,

2002). These diabetics, as compared to the H. pylori infected patients with low titres

of PCA, had an association with different HLA-DQ haplotypes (De Block et al.,

2002, Bergman et al., 2005). Thus, at least among patients with diabetes or

autoimmune thyroid disease, atrophic gastritis induced by H. pylori and autoimmune

gastritis may be separate entities.

1.2.2. The mechanism behind H. pylori induced autoimmunity

H. pylori has been shown to induce antibodies reactive with gastric mucosa in

approximately half of the infected individuals, and gastric H+,K+-ATPase has been

identified as the single major autoantigen in chronic H. pylori gastritis with corpus

atrophy (Claeys et al., 1998, Bergman et al., 2005). The autoantibodies correlate with

the severity of gastritis, atrophy and apoptosis of the corpus mucosa. Possible

mechanisms for H. pylori to induce autoimmune gastritis are the presentation of self

antigens enhanced by inflammation, and the expression of epitopes structurally

similar to those of autoantigens, so called molecular mimicry (D'Elios et al., 2004). At

least 11 epitopes of human H+,K+-ATPase recognized by autoreactive T cells have

been identified (D'Elios et al., 2005).

17

1.2.3. Histological assessment of gastritis and atrophy

The grade of gastritis and atrophy is assessed according to the updated Sydney

classification (Dixon et al., 1996) with the help of the visual analogue scales,

originally presented by Tuomo Karttunen (Smith and Genta, 2000). However, there is

still some interobserver variability in these assessments among pathologists. The

repeatability is poor concerning especially the estimation of chronic residual gastritis

after H. pylori eradication therapy (Tepeš et al., 1999a) and antral atrophy (Rugge et

al., 2002). A few mononuclear cells are always present in the lamina propria of the

gastric mucosa but the criteria for mild chronic inflammation is fulfilled when more

than two to five lymphocytes, plasma cells and macrophages are seen per highpower

(×40 objective) microscopic field or two or three lymphocytes or plasma cells are

seen between the foveolae (Dixon et al., 1996).

1.2.4. Serum tests in assessment of atrophy and autoimmunity

Pepsinogen (PG) I and II are two immunologically distinct hydrolyzing proteinases of

the mucosal lining of the stomach. PG I consists of five isozymogens and PG II of

two isozymogens. Both PG I and II are secreted from the chief and mucous neck cells

of fundic glands, but PG II is also found in the antral and proximal duodenal mucosa.

The predominant PG in atrophic corpus gastritis is PG II since, in parallel with the

decline of fundic glands the secretion of PG I abates, and thus the low PG I and low

PG I/II ratio are good predictors of atrophic corpus gastritis (Samloff and Taggart,

1987). These have been investigated in screening patients with increased risk for

gastric cancer (Varis et al., 2000, Yoshida et al., 2006, Palli et al., 2007, Sipponen

and Graham, 2007, Cao et al., 2007).

Gastrin is secreted by antral G cells and it regulates the gastric acid secretion.

In blood several forms of gastrin exist, the most prevalent being gastrin-34 (this

accounts for the basal secretion) and gastrin-17 (stimulated by food ingestion)

(Kaneko et al., 2002). Low gastrin-17 is a sign of multifocal or antrum-limited

atrophic gastritis in patients with H. pylori infection (Sipponen et al., 2002, Yoshida

et al., 2006). In patients with atrophic corpus gastritis including patients with PA the

serum gastrin levels are elevated (Bergman et al., 2005).

18

PCA were discovered in over 90% of patients with PA as early as in the 1960’s

(Whittingham and Mackay, 2005). Later on the autoantigen was identified as H+/K+

adenosine triphospahatase, the proton pump of the gastric glands (Bergman et al.,

2005). These antibodies are detected in some 60% of patients with atrophic gastritis.

In 50 to 70% of patients with PA there are antibodies against gastric secreted intrinsic

factor, a product of parietal cells, which enhances the absorption of vitamin B12

(Whittingham and Mackay, 2005).

1.3. Dyspepsia

Dyspepsia or indigestion has several definitions. According to the Rome II criteria it

is defined as upper abdominal pain or discomfort (Talley et al., 1999). The prevalence

of dyspepsia in the community is about 40% during six months (Moayyedi et al.,

2006). These patients constitute a significant burden to health care. Around 60% of

them have negative investigations and are labelled as having non-ulcer or functional

dyspepsia (Moayyedi et al., 2006), which remains a poorly defined entity and its

association with H. pylori remains still unknown (Lai and Sung, 2007). In 10% of

dyspepsia patients eradication therapy leads to long term symptom improvement

according to the Cochrane review (Moayyedi et al., 2006) and in addition it reduces

the risk of long term sequelae (atrophic gastritis and gastric cancer) of H. pylori

infection (Malfertheiner et al., 2007). Dyspepsia is an accepted indication for H.

pylori eradication (Malfertheiner et al., 2007).

2. Peptic ulcer

2.1. Epidemiology

H. pylori has apparently colonized human stomach since time immemorial (Linz et

al., 2007), but duodenal ulcer disease has been present essentially only for 200 years

(Blaser, 1998, Blaser, 1999a). The first description of gastric ulceration was made in

1506 by an Italian physician (Egan and O'Morain, 2007), but in mid nineteenth

century physicians began to observe more often peptic ulceration, initially gastric

19

ulcers in young women and then duodenal ulcers in men (Blaser, 1998). The life-time

risk of peptic ulcer in a person with H. pylori infection varies from 3% (USA) to 25%

(Japan) (Suerbaum and Michetti, 2002). In Western countries duodenal ulcers are

approximately four-fold more common than gastric ulcers; elsewhere gastric ulcers

are more common (Kusters et al., 2006). It is widely accepted that the chronic active

gastritis affecting mainly the antrum is associated with duodenal ulcer disease (Dixon,

2000).

2.2. Pathogenesis

The pathogenesis of peptic ulcer disease is not thoroughly understood. The

localisation of the ulcer is determined by acid output; the higher the output, the more

distal the ulcer (Johnson, 1957, Albillos et al., 1990, Savarino et al., 1995). The acid

secretion however, varies widely in peptic ulcer patients (Savarino et al., 1995) and

benign gastric ulcers in a persistently achlorhydric stomach have been described

(Bynum, 1991). The peptic ulcer studies in animals indicate that pepsin in

combination with acid produces much more severe mucosal damage than acid alone,

but there is considerable overlap between ulcer patients and control subjects and no

pepsinogen level is diagnostic for peptic ulcer (Samloff and Taggart, 1987). The

“leaky roof” theory, linking H. pylori and increased acid secretion, presupposes that

acid injury in the duodenum promotes the development of gastric metaplasia,

allowing the H. pylori to attach these areas and this leads to duodenal ulcer (Egan and

O'Morain, 2007). However, attachment of H. pylori to gastric apical cells indeed

induces proinflammatory chemokines and changes core protein expression leading to

changes in the mucous gel layer, but neither these alterations in mucosal protection

(Jass and Walsh, 2001) nor the studies on the host’s genetic polymorphism of

inflammatory and immunoregulatory cytokines (Blanchard et al., 2004) account for

the disease phenotype. The outcome of H. pylori gastritis on an individual basis

cannot be predicted by virulence factors of the bacteria e.g. CagA either (González et

al., 2003, Sgouros and Bergele, 2006). Venerably, we have to admit that we do not

know why some people develop peptic ulcer, but the exogenous risk factors must be

responsible for the rapid changes in the epidemiology of the disease (Kusters et al.,

20

2006). In areas of high prevalence of H. pylori infection it is a challenge to identify

patients at risk of peptic ulcer.

2.3. Duodenal gastric metaplasia (DGM)

DGM indicates the presence of patches of mucus-secreting gastric-type cells along the

villous surfaces in duodenum. It is differentiated from heterotopia, supposed to be a

congenital abnormality, where the fully developed body-type gastric mucosa contains

also parietal and chief cells (Wyatt and Rathbone, 1989). In 1935 artificial defects in

the duodenal mucosa of cats were demonstrated to heal with a simple mucous

epithelium identical with gastric surface epithelium, and in 1964 DGM was first

described in a patient with suspected Zollinger-Ellison syndrome. DGM was shown to

be associated with duodenal ulcer and high acid levels already before the discovery of

H. pylori and it was suggested to be part of the healing process following tissue

damage (Patrick et al., 1974). Steer described the adherence of (at that time unknown)

bacteria to the surface of these gastric type epithelial cells and he noticed that these

bacteria were absent from the biopsies of the patients with normal gastric and

duodenal mucosa (Steer, 1984). Active duodenitis was shown to be associated when

the extension of DGM was more than 5% (Wyatt et al., 1987). DGM was commonly

found in asymptomatic volunteers and small foci were not associated with

inflammation and therefore the clinical importance of gastric metaplasia was

questioned (Fitzgibbons et al., 1988). However, at the same time it was suggested that

the gastric antrum was necessary for H. pylori colonization and the bacterial infection

in the duodenal bulb was made possible by its mucosal change. This change was

originally meant to protect the duodenal mucosa, but H. pylori infection in the bulb

predisposed to ulcerative lesions (Caselli et al., 1988).

After the strong association shown between DGM and duodenal ulcer, DGM if

colonized by H. pylori turned out to be an even stronger risk factor for duodenal ulcer

(Carrick et al., 1989). The inflammatory injury of the duodenal mucosa caused by H.

pylori was supposed to stimulate further gastric metaplasia and as part of the vicious

circle the progressively increasing area led to ulceration (Wyatt et al., 1990). Since

then several studies on the association between DGM and duodenal ulcer have been

published but the significance of DGM in duodenal ulcer pathogenesis is still

21

questionable (Savarino et al., 1997, McColl, 1996, Jonkers et al., 1998). However,

many studies have shown that in patients with duodenal ulcer, the prevalence and

extension of gastric metaplasia in the duodenum are more widespread as compared to

non-ulcer dyspepsia or gastric ulcer patients (Wyatt et al., 1987, Kreuning et al.,

1989, Carrick et al., 1989, Wyatt et al., 1990, Andersen et al., 1991, Amarapurkar et

al., 1993, Yang et al., 1995, Harris et al., 1996, Pan et al., 1996, Jonkers et al., 1998,

Kim et al., 1998, Futami et al., 1999, Heikkinen et al., 2001, Chang et al., 2001).

Recently, in a prospective study positive H. pylori culture from the duodenal bulb was

a strong risk factor for subsequent development of duodenal ulcer (Pietroiusti et al.,

2005). In accordance with the association of DGM with hyperacidity, in atrophic

gastritis DGM is non-existent (Wyatt et al., 1990). For reasons unknown, it has been

shown to be common also in H. pylori negative non-ulcer dyspepsia (Heikkinen et al.,

2001).

2.3.1. Regression of DGM

The effect of H. pylori eradication on the prevalence and extent of DGM is

contradictory (Noach et al., 1993, Harris et al., 1996, Khulusi et al., 1995, Savarino et

al., 2000, Bago et al., 2002). Healing of duodenal ulcer had no effect on the extent of

gastric metaplasia but acid suppressive therapy and H. pylori eradication diminished

the area significantly during a six month period (Khulusi et al., 1996), and a longer

follow-up of four years also showed a declining tendency (Kim et al., 1998).

The prevalence of DGM in vagotomized patients was reported to be lower than in

duodenal ulcer patients (Wyatt et al., 1987), although the 2 to 3 year follow-up of

patients after vagotomy failed to show any change either in the prevalence or in the

extent of DGM (Jönsson et al., 1988).

The mechanism behind DGM is suggested to be the high acid exposure which

impairs the function of Cdx2, considered to be the master regulator of intestinal

differentiation (Faller et al., 2004).

22

2.4. H. pylori eradication in peptic ulcer disease

Of the duodenal ulcers 95%, and of the gastric ulcers, 70% are associated with H.

pylori (Ford et al., 2006). As far as ulcer healing is concerned in duodenal ulcer

patients, H. pylori eradication is superior to ulcer healing drugs, but in gastric ulcers

there is no difference between these two therapies. In both duodenal and gastric ulcers

eradication therapy is clearly effective in preventing the recurrence of ulcers when

compared to patients with no ulcer healing drug therapy: duodenal ulcers recurred in

6% of patients successfully treated for H. pylori compared with 67% of those who

remained positive for H. pylori, whereas in gastric ulcers the response was 4% and

59%, respectively (Hopkins et al., 1996). In duodenal ulcer patients H. pylori

eradication therapy and continuous therapy with modern ulcer healing drugs are

equally effective in preventing recurrence. In patients with gastric ulcer there are no

such studies (Ford et al., 2006). However, eradication therapy is cost effective and

continuous PPI therapy is an option only for patients with contraindications for H.

pylori eradication therapy or several unsuccessful eradication therapies. Patients with

peptic ulcer disease benefit clearly from H. pylori eradication according to the recent

Cochrane Database of Systematic Reviews (Ford et al., 2006) and H. pylori

eradication is strongly recommended for these patients also in the Maastricht II and

III Consensus reports (Malfertheiner et al., 2002) (Malfertheiner et al., 2007).

2.5. Non-steroidal anti-inflammatory drugs (NSAID) and peptic ulcer

NSAIDs or H. pylori alone increase the risk of peptic ulcer about 20-fold, but H.

pylori infected NSAID users are 60 times more likely to develop a peptic ulcer than

uninfected non-NSAID users (Lai and Sung, 2007). NSAIDs and H. pylori infection

independently increase also the risk of peptic ulcer bleeding. In patients with chronic

NSAID use, PPI maintenance treatment is better than H. pylori eradication therapy in

preventing upper gastrointestinal bleeding. However, H. pylori eradication therapy is

of value in chronic NSAID and low dose aspirin users although it is insufficient to

prevent ulcer disease and bleeding completely. H. pylori eradication therapy is

recommended to all naive NSAID users and to chronic NSAID users needing PPI

therapy for NSAID related side-effects, the latter to prevent the development of

23

atrophic gastritis accelerated by PPIs (Malfertheiner et al., 2007). Eradication

treatment is also recommended to patients on low-dose aspirin therapy with a history

of peptic ulcer (Lai and Sung, 2007).

2.6. Vagotomy

Highly selective vagotomy (HSV) was the surgical treatment of choice for patients

with chronic duodenal ulceration before the era of H. pylori eradication therapy. In

the procedure the nerves controlling the gastric emptying were preserved while those

controlling acid secretion were divided (Donahue, Griffith and Richter, 1996). The

operation decreased the peak acid output by approximately 50% (Egan and O'Morain,

2007). The rate of ulcer recurrence in a study of 500 patients who had undergone

HSV was 18.5% in 15 years (Macintyre and Millar, 1991).

3. Mucosa associated lymphoid tissue lymphoma (MALT)

Primary gastric lymphomas comprising 5% of gastric malignancies, are generally

non-Hodgkin lymphomas (NHL) and the most common histological subtypes are

diffuse large B-cell and marginal zone B-cell NHL of the MALT-type (Ferrucci and

Zucca, 2007). In Finland according to the statistics of the Finnish Cancer Registry

(www.cancerregistry.fi), there were 59 NHLs of the stomach in 2003. Less than 1%

of H. pylori infected subjects develop MALT lymphoma (Kusters et al., 2006) but of

the patients with MALT lymphoma 72-98% are infected with H. pylori (Suerbaum

and Michetti, 2002). Twelve months after H. pylori eradication therapy 62% of

patients with low grade gastric MALT lymphoma had complete remission and thus H.

pylori eradication is first line therapy in patients in stage I disease and the indications

for eradication therapy are strong in all patients with MALT lymphoma

(Malfertheiner et al., 2007).

24

4. Gastric cancer

Gastric cancer has been known since 400 BC in writings of Hippocrates (Houghton

and Wang, 2005). Environmental factors modulating the risk of gastric cancer include

certain food items and cigarette smoking (Lochhead and El-Omar, 2007).

Gastric cancer is divided according to the classification presented by Laurén into

intestinal and the diffuse type (Laurén, 1965). The incidence of gastric cancer is

declining in Western countries, especially the intestinal type of distal gastric cancer,

but the diffuse type, although also associated with H. pylori, is not (Houghton and

Wang, 2005). The association of H. pylori infection with both histologic types of

gastric cancer has been equally strong (Lochhead and El-Omar, 2007) and CagA

seropositivity has been associated with gastric cancer (Held et al., 2004, Palli et al.,

2007). In most studies concerning the location of the cancer, the risk of H. pylori and

CagA seropositivity was confined to non-cardia cancer (Lochhead and El-Omar,

2007, Palli et al., 2007). Cardia cancer, formerly thought not to be associated with H.

pylori, is now divided into type A being a consequence of atrophic gastritis due to H.

pylori infection or autoimmune atrophic gastritis, and type B which resembles

oesophageal adenocarcinoma and is likely to be a consequence of short segment

gastro-oesophageal reflux disease (McColl, 2006, Palli et al., 2007). H. pylori

infection is negatively associated with gastro-oesophageal reflux disease, the reason

for this being largely unclear, but H. pylori eradication does not exacerbate symptoms

in these patients and reflux disease is not a contraindication for H. pylori eradication

(Malfertheiner et al., 2007).

The lifetime risk for gastric cancer in H. pylori infected persons is less than

2% (McColl, 2005). H. pylori has been classified as a class I carcinogen by the

International Agency for Research on Cancer 1994 (Anonymous, 1994) and it is

estimated that 60-90% of gastric cancers are associated with H. pylori (Malfertheiner

et al., 2005). In Finland the decreasing prevalence of H. pylori antibodies has been

associated with subsequent decreasing incidence of gastric cancer after a median

latency of 20 years (Rehnberg-Laiho et al., 2001), and this result has also been

confirmed in Japan (Kobayashi et al., 2004).

25

4.1. Gastric carcinogenesis

The concept of the gastric carcinogenesis was introduced by Correa in 1988 as a

sequence of consecutive events starting from chronic superficial gastritis, atrophy and

intestinal metaplasia, dysplasia, and finally leading to gastric adenocarcinoma

(Correa, 1988). However, some recent data suggest that intestinal metaplasia is a

marker for atrophy, a paracancerous rather than precancerous condition, and that

cancer arises from a different cell lineage in gastric glands (Graham and Shiotani,

2005, Tatematsu et al., 2003, Meining et al., 2001). Histogenetically numerous

carcinomas of the stomach are primarily of the gastric type and may secondarily

change into the intestinal type (Meining et al., 2001). In addition, not all gastric

cancers rise from atrophic gastric mucosa (Sipponen et al., 1994), although in the

totally normal stomach the cancer is extremely rare (Graham and Shiotani, 2005,

Sipponen et al., 1992). At present gastric cancer is regarded as a paradigm for

infection-induced chronic inflammation-mediated cancer (Lochhead and El-Omar,

2007). There is experimental evidence that the origin of these cancers are bone

marrow derived stem cells (BMDC) and the role of inflammation is in creating a pro-

carcinogenic environment for engraftment of circulating marrow-derived stem cells

(Li et al., 2006, Cai et al., 2005, Houghton and Wang, 2005). It is suggested that the

bone marrow derived cells take over the function of tissue stem cells and in the

abnormal tissue environment with lack of glandular cells, the BMDC fail to

differentiate properly and progress instead to metaplasia, dysplasia, and cancer

(Houghton and Wang, 2005).

4.2. H. pylori eradication to prevent gastric cancer

After H. pylori eradication therapy acute inflammation disappears in a few weeks, but

chronic inflammation persists longer (Forbes et al., 1996, Tepeš et al., 1999b,

Franceschi et al., 2002, Hojo et al., 2002, Kuipers and Sipponen, 2006). The studies

on the resolution of atrophy and intestinal metaplasia after successful H. pylori

eradication therapy have shown variable results (Kokkola et al., 2002, Kuipers and

Sipponen, 2006, Hojo et al., 2002).

26

The studies on the effect of H. pylori eradication on gastric cancer risk have

been somewhat disappointing (Kuipers and Sipponen, 2006, Malfertheiner et al.,

2006). In some studies, H. pylori eradication has been shown to be beneficial (Saito et

al., 2000, Uemura et al., 2001, Take et al., 2005, Zhou et al., 2005, Nakagawa et al.,

2006), but in others there has been no statistical difference between treated and non-

treated (Wong et al., 2004, Leung et al., 2004, Mera et al., 2005). In a Chinese study

including patients with no precancerous conditions at baseline, no gastric cancers

were evident after eradication therapy (Wong et al., 2004). Eradication therapy

probably does not prevent gastric cancer in subjects with precancerous conditions

(there is a point of no return) (Kuipers and Sipponen, 2006). The sustained chronic

inflammation after successful eradication as a possible explanation for dismal results

in these gastric cancer prevention studies is not thoroughly investigated.

The Maastricht III Consensus Report concluded by recommending H. pylori

eradication to prevent gastric cancer because it has the potential to reduce non cardia

gastric cancer risk and the optimal time to eradicate is before the pre-neoplastic

changes appear. Eradication is strongly indicated in patients with gastric cancer

resection, atrophic gastritis and first degree relatives of patients with gastric cancer

(Malfertheiner et al., 2007).

5. Extragastric diseases

H. pylori has been implicated in several studies as a cause of iron-deficiency anemia

regardless of the presence or absence of peptic ulcer disease (Cardenas et al., 2006). It

is often refractory to oral iron treatment but responds favourably to H. pylori

eradication (Hershko et al., 2005). Unexplained iron deficiency anemia is an

indication for H. pylori eradication (Malfertheiner et al., 2007).

Idiopathic thrombocytopenic purpura (ITP) is also an extragastric disease

associated with chronic H. pylori infection (Franceschi and Gasbarrini, 2007). In a

recent study 58% of ITP patients had H. pylori infection and 62% of patients with

successful eradication therapy showed increased platelet count (Kodama et al., 2007).

ITP is an indication for H. pylori eradication therapy according to the Maastricht III

Consensus Report (Malfertheiner et al., 2007).

27

Other extragastric diseases possibly associated with H. pylori infection are

chronic ischemic heart disease and cerebrovascular disease but despite intensive

studies on the field, the results are contradictory and the benefit of eradication therapy

in these situations is controversial and not recommended for the present (Franceschi

and Gasbarrini, 2007, Goodman et al., 2006).

DIAGNOSTIC METHODS OF H. PYLORI INFECTION

The accuracy of all diagnostic tests described for the detection of H. pylori infection

below depends on the clinical situation in which they are used. Especially in settings

with a low prevalence of H. pylori infection even tests with high sensitivity and

specificity may turn out to give low positive predictive values and this should be

taken into account when these tests are used (Campbell and Machin, 1993). No single

method can be considered to be ideal for the detection of H. pylori infection

(Anonymous, 1997). On the other hand, the cost-effectiveness of the different

diagnostic strategies is based on the prevalence of the infection and on the price and

accuracy of different tests (Vakil et al., 2000).

There are several both invasive and non-invasive methods for the diagnosis of

H. pylori infection, see Table 1. The invasive methods need gastric biopsy material

taken by gastroscopy and non-invasive methods are based on serum, stool, urine, or

saliva specimens. In general, the choice between different methods is dependent on

the clinical situation and if gastroscopy is needed (age over 45 or alarming symptoms)

(Malfertheiner et al., 2002, Pikkarainen et al., 2002, Delaney et al., 2007).

28

Table 1. Special charasteristics of the diagnostic tests for the detection of H. pylori

infection introduced into clinical practice.

Diagnostic method Special charasteristics

Invasive tests Gastroscopy needed

Rapid urease test Quick, cheap, reduced sensitivity if low number of bacteria

Histology Enables the assessment of gastric pathology, expensive

Culture from gastric

biopsies

High specificity, enables the testing of antimicrobial

susceptibility, needs sophisticated laboratory skills

PCR Sophisticated method, enables the testing of antimicrobial

susceptibility, detection of virulence factors, comparison of

clinical isolates

FISH Sophisticated method, enables the testing of antimicrobial

susceptibility

Non-invasive tests Gastroscopy not needed

EIA-based

quantitative antibody

tests

Accurate after validation, low cost, easy to perform, slow

decline in antibody titers after eradication, need for paired

serum samples in posteradication setting

Rapid blood-based

tests

Low accuracy

Immunoblot assay Detects antibodies to specific antigens e.g. CagA

Urea blood test Promising, not well studied

Urea breath test High accuracy, needs sophisticated laboratory equipment

Stool antigen tests Monoclonal EIA-based test accurate, in-office tests have low

accuracy, easy to perform, moderately expensive

29

1. Invasive methods

1.1. Rapid urease test (RUT)

The RUT is based on the strong urease activity of H. pylori. Urea embedded in culture

media generates ammonia when broken down by H. pylori urease and thus produces

an increase in pH usually detected by a change in colour with an indicator. There are

both non-commercial and commercial RUTs available. The sensitivity and specificity

of these tests are dependent on the time elapsed between biopsy and reading of the

result, and the number of biopsies taken (in general only one antral biopsy) (van

IJzendoorn et al., 2005). The reaction time for the newer tests is shorter, up to three

hours compared with the 24 hours of older tests, without affecting their reliability

(Tseng et al., 2005, Malfertheiner et al., 1996). RUTs are considered highly accurate

with a sensitivity and specificity over 90% (van Keeken et al., 2006, Kuo et al., 2002,

Laine et al., 1996, Lin et al., 1992). Gastric bleeding causes low accuracy and up to

54% of H. pylori cases may be missed (Archimandritis et al., 2000b, Tu et al., 1999,

Gisbert and Abraira, 2006a, Güell et al., 2006, Castro-Fernandez et al., 2004, Peitz et

al., 2004, Chung et al., 2001, Griñó et al., 2001, Lee et al., 2000). The low accuracy

of the RUT in gastric bleeding is suggested to be partly due to technical reasons

(insufficient number of good quality biopsies), and due to PPI therapy and naso-

gastric lavage, but several studies seem to exclude the role of blood in the stomach as

a direct cause of false-negative results (Gisbert and Abraira, 2006a). Other clinical

situations with inferior performance of RUT include: post-surgical stomach

(Archimandritis et al., 2000a); gastric ulcer probably due to concomitant atrophic

corpus gastritis and low numbers of H. pylori in antral biopsies (Bermejo et al.,

2002); use of H2-blockers (Koch et al., 1990, Lerang et al., 1998); and PPIs (Yakoob

et al., 2005, Griñó et al., 2001). In addition to low sensitivity, hypochlorhydric

patients harbour other urease-positive bacteria in the stomach possibly leading to false

positive test results (Brandi et al., 2006). The sensitivity of RUT is considered low

also in the post-eradication setting and figures as low as 71% four weeks after therapy

have been described (Laine et al., 1998b). Therefore, in clinical situations when there

is a low density of H. pylori, two biopsies, one from the antrum and one from the

30

corpus are recommended for the RUT (Bermejo et al., 2002, Gisbert and Abraira,

2006a).

1.2. Histology

Histology has long been considered the “gold standard” in H. pylori diagnosis

(Kusters et al., 2006, Mégraud and Lehours, 2007). However, the results of histology

are dependent on the experience of the pathologist (Metz et al., 1998, Maconi et al.,

1999), the number of biopsies taken (van IJzendoorn et al., 2005), the use of special

stains (Laine et al., 1997, Kassa et al., 1996) and the density of H. pylori in the gastric

mucosa (Epple et al., 1997, Testoni et al., 2002). In clinical practice, two biopsies

both from the antrum and the corpus are recommended (Dixon et al., 1996). However,

the recommendation of the updated Sydney System to take one extra biopsy, from the

incisura angularis gives very little additional information (Eriksson et al., 2005). The

need for special stains is especially emphasized in cases with a low density of H.

pylori; the specificity of hematoxylin-eosin stain is clearly lower (89%) when

compared to Giemsa and Genta stains (98%) (Laine et al., 1997) and the use of these

latter stains is strongly recommended (Dixon et al., 1996). In some studies

immunohistochemistry has been superior to the standard stains for H. pylori (Marzio

et al., 1998, Jonkers et al., 1997b, Ashton-Key et al., 1996, Mégraud and Lehours,

2007). In atrophic gastritis (Lahner et al., 2004, Kokkola et al., 2000, Testoni et al.,

2002) and remnant stomach (Matsukura et al., 2004), histology is insensitive. The

performance of histology in patients with gastrointestinal bleeding has been less

consistent than expected with sensitivity between 70 to 86% and affected by

medication, timing of endoscopy and the number of biopsies taken (often restricted to

minimum in unstable bleeding patients) (Laine et al., 2005, Castro-Fernandez et al.,

2004, Castillo-Rojas et al., 2002, Chung et al., 2001, Griñó et al., 2001). In a recent

meta-analysis, the pooled sensitivity and specificity of histology in the diagnosis of

patients with bleeding peptic ulcer was 70% and 90%, respectively (Gisbert and

Abraira, 2006a). After H. pylori eradication therapy, the sensitivity of histology is

lower compared to the primary diagnosis and moreover, increasing the number of

biopsies is of limited value (Laine et al., 2000, Rollán et al., 1997). If histology is

used to confirm the cure of H. pylori infection, the optimal time to test is 4 weeks

31

after antimicrobial therapy (de Boer, 1997). The absence of chronic antral

inflammation is a very reliable parameter to exclude H. pylori infection (Cutler et al.,

1995) and after eradication therapy, any persistent active gastritis may represent an

indirect sign of eradication failure in apparently H. pylori negative biopsy samples

(Anonymous, 1997, Mégraud and Lehours, 2007). PPI therapy with concomitant non-

acidic stomach causes both false positive and negative results on histology (Jonkers et

al., 1997a). However, although histology has its restrictions and is considered

expensive (Hahn et al., 2000), the major advantage of using it is the assessment of

morphological changes of the gastric mucosa (Dixon et al., 1996).

1.3. Culture and antimicrobial susceptibility

Culture of H. pylori from gastric biopsies is considered highly specific but tedious and

insensitive even in experienced laboratories (Grove et al., 1998, Andersen et al.,

1998, Ndip et al., 2003). The sensitivity is influenced by transport medium and time

(Pérez-Pérez, 2000, Ndip et al., 2003). However, in favourable circumstances it can

be very accurate, and sensitivity may reach up to 95%, and specificity is 100%

(Rautelin et al., 1997, Matsukura et al., 2004, Monteiro et al., 2001, Laheij et al.,

2000, Lerang et al., 1998, Thijs et al., 1996, Mégraud and Lehours, 2007). In general,

culture is indicated after one or two failed eradication regimens, and one to two

biopsies each from antrum and corpus are recommended for culture (Ndip et al.,

2003, Mégraud and Lehours, 2007).The sensitivity of culture is affected by

eradication therapy up to 8 weeks, and even more with triple than dual therapy (Laine

et al., 2000). The sensitivity of culture performed from one antral biopsy was 22%

lower in post-eradication setting when compared with sensitivity in primary diagnosis

(Laine et al., 2000). In patients with bleeding peptic ulcer the sensitivity in a recent

meta-analysis was 45%, but the specificity was high, 98% (Gisbert and Abraira,

2006a). Culture enables the possibility to study bacterial virulence factors (for

scientific purposes) and the antimicrobial susceptibility of H. pylori, a debatable

advantage in primary diagnosis (Zullo et al., 2003, Mégraud and Lehours, 2007). In

areas with low prevalence of primary resistance to antimicrobials used in eradication

therapy, culture gives little extra benefit. This is also true for geographical areas with

high metronidazole resistance when its use as first line therapy should be avoided. In

32

Finland as reported in a recent study, the prevalence of metronidazole resistance of H.

pylori was high, 38% (45% in women) but clarithromycin resistance was only 2%

(Koivisto et al., 2004).

1.4. Molecular methods

The polymerase chain reaction (PCR) is an in vitro technique for exponentially

amplifying a predetermined fragment of DNA. PCR tests for H. pylori infection have

been based on several genes of H. pylori and all have shown good performance

(Brooks et al., 2004, Monteiro et al., 2001, Lage et al., 1995, Chisholm et al., 2001,

Pacheco et al., 2001, Peek et al., 1995, Andersen et al., 1998, Thijs et al., 1996, Fabre

et al., 1994). PCR has had high accuracy also when applied to formalin-fixed

paraffin-embedded specimens (Weiss et al., 1994). In patients with bleeding peptic

ulcer, the sensitivity of PCR was shown to be comparable to histology and superior to

RUT and serology (Castillo-Rojas et al., 2002). PCR-based methods also enable

identification of antimicrobial resistance to clarithromycin (Marais et al., 1999, van

Doorn et al., 2001, Oleastro et al., 2003, Chisholm et al., 2001). For scientific

purposes, it is possible to determine virulence factors like cagA, fingerprinting for

epidemiologic surveys (Schwarz et al., 1997) and to differentiate between reinfection

and recrudescence after failed eradication therapy (Jeen et al., 2001).

In situ hybridization is a method that uses labelled complementary DNA or

RNA strands to localize a specific DNA or RNA sequence in a tissue section.

Fluorescent in situ hybridization (FISH) uses fluorescent probes. FISH is a new

method for H. pylori diagnosis (Trebesius et al., 2000, Rüssmann et al., 2001a).

Fluorescent labelled oligonucleotide probes binding to 16S rRNA of H. pylori provide

detection of single bacteria by fluorescence microscopy and probes to 23S rRNA

sequence enable to define clarithromycin resistance (Rüssmann et al., 2001a, Yilmaz

et al., 2007). The method is also applicable to paraffin-blocks (Trebesius et al., 2000,

Rüssmann et al., 2003) (Jüttner et al., 2004, Morris et al., 2005, Yilmaz et al., 2007).

The correlation of this method with histology and conventional chlaritromycin

susceptibility testing has been excellent (Trebesius et al., 2000, Rüssmann et al.,

2001b, Jüttner et al., 2004, Yilmaz and Demiray, 2007).

33

2. Noninvasive methods

2.1. Blood-based tests

2.1.1. Enzyme immunoassay (EIA) for H. pylori antibodies

In EIA-based tests the antibody of interest is captured by an antigen (or antigens of

interest are captured by antibodies) usually on a solid-phase support. These captured

antibodies (or antigens) are detected by an enzyme-labelled antibody, which catalyzes

a color reaction when exposed to a substrate. The intensity of the color reaction

enables the quantification of the material of interest. Several commercial tests for H.

pylori antibodies are available (Wilcox et al., 1996, Meijer et al., 1997, Vaira et al.,

1998). A comparison of several commercial EIA-kits testing the same sera in several

laboratories in Europe indicated that Pyloriset EIA-G (Orion Diagnostics, Espoo,

Finland), one of the most widely used serology kits in Finland, has the best accuracy

(Feldman et al., 1995, Mégraud and Lehours, 2007). However, the wide range in

sensitivities and specificities of the commercial EIA-kits prompted local validation

and adjustment for cut-off values to ensure appropriate accuracy (Malfertheiner et al.,

2002). The differences in the performances of diagnostic kits in different geographic

regions are possibly due to the variation in antigenic properties of local H. pylori

strains and those used to prepare antigen for the test. However, the diagnostic

performance of properly evaluated serological assays is comparable to that of biopsy-

based methods and UBT (Oksanen et al., 1998, Herbrink and van Doorn, 2000). In

some clinical settings, serology can even be the best diagnostic modality. Recently,

the role of EIA serology has been reappraised especially in certain clinical situations

when other tests will give false negative results, e.g. bleeding ulcers, atrophic

gastritis, MALT lymphoma, and recent use of PPIs and antimicrobials (Malfertheiner

et al., 2007).

Patients with positive serology and both negative histology and culture are

overrepresented among patients with atrophic gastritis (Storskrubb et al., 2005), and

the seropositivity may actually represent individuals with an ongoing infection

(Kokkola et al., 1998). In patients with atrophic gastritis, H. pylori EIA results

formerly considered false positive, probably reflect the insensitivity of other tests

34

when H. pylori density is low (Kokkola et al., 1998). The performance of new

commercial EIA tests as compared with invasive tests has been excellent when

patients in older age-groups with atrophic gastritis have been excluded from the

analysis (Salomaa- Räsänen et al., 2004).

Serology is less affected by PPI and antibiotic therapy as compared with RUT,

UBT or histology (Dickey et al., 1996). However, in very young children serology

has low accuracy (Kindermann et al., 2001, Kolho et al., 2002). The disadvantage of

serology is also that it does not necessarily indicate a current infection: one year after

successful H. pylori eradication therapy antibodies decline below the cut-off level

only in 28% of patients (Lerang et al., 1998) and 37% in 4 years (Cutler et al., 1998).

In assessing the success of eradication therapy paired sera are needed (Kosunen et al.,

1992), and the fall in antibody titre by 50% in six months (Kosunen et al., 1992,

Fallone et al., 1998) or 40% in four months (Lerang et al., 1998) is considered a

marker for successful eradication therapy. Locally validated serology with paired sera

after eradication is highly accurate in confirming the cure of H. pylori infection, with

sensitivity of about 95-97% and a specificity of 95-100% after six months of

eradication therapy (Kosunen et al., 1992) (Hirschl and Rotter, 1996).

2.1.2. Near patient tests

These are mostly whole blood tests, but some require serum separation. The accuracy

of these commercial tests is considered poor with a sensitivity as low as 59% (Talley

et al., 1998) and a specificity as low as 54% (Sadowski et al., 1998) and they are not

recommended in clinical practice (Malfertheiner et al., 2007).

2.1.3. Immunoblot assay

In immunoblot or the Western blot, antigens separated by electrophoresis according to

their molecular weight and transferred onto nitrocellulose or other absorbent paper

react with specific antibodies which are then detected by a conjugate (labelled

antibody) and substrate as in EIA. In some studies the immunoblot assay has had

better performance when compared to EIA serology and it enables the study of

antibodies to specific antigens, for instance CagA (Nilsson et al., 1997, Andersen and

Espersen, 1992). The commercial kit Helicoblot 2.1 (Genelabs Diagnostics,

35

Singapore, Singapore) is the most widely studied assay both among European and

Asian populations with good sensitivity for the diagnosis of H. pylori infection

(Hoang et al., 2006). Immunoblot CagA positivity has been used to assess the present

(Monteiro et al., 2001, Yamaoka et al., 1998) and past (Ekström et al., 2001) H.

pylori infection, the latter because these antibodies seem to stay elevated longer than

EIA serology (Ekström et al., 2001, Ekesbo et al., 2006). In some studies, antibodies

to CagA have been strongly associated with atrophic gastritis (Sande et al., 2001) and

with gastric cancer (Palli et al., 2007, Ekesbo et al., 2006, Rudi et al., 1997,

Vilaichone et al., 2003). There are, however, also contradictory results (Sokic-

Milutinovic et al., 2004, Schilling et al., 2000).

2.1.4. 13C-urea blood test

The 13C-urea blood test is based on the same 13C isotope as in UBT but the isotope

ratio mass spectrometry measurement is done from blood instead of the exhalation, 30

minutes after the ingestion of 13C-urea. (Cutler and Toskes, 1999) The performance of

the test in adults when compared with UBT or RUT (up to 97%) (Cutler and Toskes,

1999, Ahmed et al., 2005) or histology (93%) (Chey et al., 1999) has been excellent.

2.2. Urea breath test (UBT)

2.2.1 Different UBT procedures

The UBT is based on the capacity of H. pylori on the gastric mucosa to hydrolyse

orally administered isotope-labelled urea to produce isotopically labelled CO2 which

diffuses into the blood and is excreted by the lungs and can be measured in exhaled

air. UBT has several modifications concerning the use of the radioactive 14C (its use is

contraindicated in children and pregnant women) or non-radioactive 13C- isotope, the

dose (100-15 mg) and the form (fluid or tablets) of the urea given, the test meal (no

test meal, citric acid, commercial special drinks, or fruit juices), measurement of the

isotope (13C can be measured by isotope ratio mass spectrometry IRMS, non-

dispersive isotope-selective infrared spectrometry NDIRS or laser assisted ratio

analyser LARA, which is based on an optogalvanic effect), and the optimisation of

36

the cut-off point after urea ingestion (affected by dose and form of the urea ingested

and the use of test meal) (Gisbert and Pajares, 2004c).

2.2.2. The clinical usage of UBT

In general, the UBT is considered highly accurate both in primary diagnosis and after

H. pylori eradication therapy with almost all protocols having a sensitivity and

specificity over 95% (Gisbert and Pajares, 2004c, Gomollón et al., 2003). Also the

commercial form 13C UBT (50 mg of urea combined with 456 mg citric acid) with

one to two tablets dosage has been studied in the primary diagnosis setting (Ohara et

al., 2004, Gatta et al., 2003, Hamlet et al., 1999, Wong et al., 2003) and in the post-

eradication setting (Gatta et al., 2003, Hamlet et al., 1999, Wong et al., 2003) with

excellent concordance with invasive tests and the standard UBT. However, atrophic

gastritis (Capurso et al., 2006, Lahner et al., 2004), gastric resection, fast gastric

emptying, and antimicrobials and bismuth disturb the performance of UBT (Gisbert

and Pajares, 2004c). Furthermore, PPI therapy decreases the accuracy of UBT

substantially (Mana et al., 2005, Peitz et al., 2004, Gatta et al., 2004, Manes et al.,

2001) and the effect lasts up to two weeks (Laine et al., 1998a, Graham et al., 2004a).

To be fully confident with UBT results, PPI therapy should be withdrawn two weeks

before testing, and although the effect of the H2-receptor antagonists on UBT results

is in most part remote, it would be safe to avoid them one day before testing (Gisbert

and Pajares, 2005). The antacids and sucralfate have no deleterious effect on the

performance of UBT (Murakami et al., 2003). In the oropharynx and in the

hypochlorhydric stomach there might be other bacteria than H. pylori with high

urease activity responsible for false positive UBT results (Brandi et al., 2006)

although the clinical relevance of this seems to be very limited (Gisbert and Pajares,

2004c). Actually, in case of apparently false positive UBT results the more probable

explanation is in fact the insensitivity of the reference method used (Epple et al.,

1997). The UBT has performed well in patients with upper gastrointestinal bleeding

when performed in emergency room before endoscopy (Winiarski et al., 2003).

37

2.3. Stool tests

2.3.1. Stool culture

H. pylori culture of stool specimens has been reported, but it is tedious and not easily

reproducible (Thomas et al., 1992, Parsonnet et al., 1999, Ndip et al., 2003). It has

neither clinical nor scientific use.

2.3.2. Stool antigen tests

These tests are based either on monoclonal or polyclonal antibodies. The laboratory

tests are EIA tests (Apollo, HePy-stool, Hp Ag, Amplified IDEIA HpStAR and

Premier Platinum HpSA) but also commercial near patient tests have been developed

(ImmunoCard STAT! HpSA, MiniLab and Simple H. pyl, also marketed by the name

Stick H. pylori). The most studied tests are the polyclonal antibody-based HpSA, the

monoclonal antibody-based HpStAR, and the immunochromatographic ImmunoCard

based on monoclonal antibodies.

HpSA test detects H. pylori antigens from stool specimens by an EIA

technique where the capture antibody adsorbed to microwells and the antibody, as a

component of the conjugate, are polyclonal. The HpStAR test is a monoclonal

antibody-based EIA. The ImmunoCard is an immunoassay based on a lateral flow

chromatography technique and monoclonal antibodies are used to detect H. pylori

antigens. It is a rapid test with results available within five minutes.

The early studies of the HpSA test both in primary diagnosis and in post-

eradication setting had inconsistent results. In a mini-review among 2879 patients in

pre-treatment setting when compared to UBT or endoscopy with biopsies the HpSA

test showed sensitivity of 93.1% and specificity of 92.8%, respectively and in a post-

treatment setting among 614 patients the figures were 87.8% and 88.4%, respectively

(Vaira and Vakil, 2001b). Another review article from that time included also studies

published in abstract form and including 4769 patients in pre-treatment and 2078

patients in post-treatment analysis with comparable results (Gisbert and Pajares,

2001). However, despite good results as a whole, in individual studies even in pre-

treatment setting the performance varied and sensitivity ranged from 80% to 100%

38

and specificity from 70% to 100% (Vaira and Vakil, 2001b). In addition, more

information of the HpSA test especially in the post-treatment setting was demanded

(Vaira and Vakil, 2001b, Gisbert and Pajares, 2001, Parente et al., 2002). More

recently, a monoclonal antibody-based test HpStAR (also marketed under the name of

FemtoLab) became available with promising results in children with a sensitivity and

specificity of 98% and 96.7% in a pre-treatment setting and 100% and 96.9%,

respectively four weeks after eradication therapy (Makristathis et al., 2000). Soon

thereafter, two studies were published in adults one in a pre-treatment setting with a

sensitivity of 88.5% and specificity of 96.3% with no statistical difference when

compared with the polyclonal HpSA (Agha-Amiri et al., 2001), and another in a post-

treatment setting where the sensitivity of the monoclonal antibody-based test after

adjustment of the cut-off point was better than the sensitivity of the HpSA test with

equal specificity (Leodolter et al., 2002).

2.3.3. PCR of stool specimens

PCR based methods to detect H. pylori and even clarithromycin resistance have been

applied to stool specimens (Rimbara et al., 2005). The performance of the stool PCR

has been variable (Sen et al., 2005, Kabir, 2004) and considered, so far, insufficient

for clinical use (Wi niewska et al., 2002, Watanabe et al., 1998).

2.4. Urine tests

The commercial tests to detect H. pylori antibodies in urine are based either on EIA or

rapid immunochromatographic methods. The EIA based Urinelisa test (Kabir, 2003)

has been evaluated in two European multicenter studies in children (Mégraud et al.,

2005) and in adults (Leodolter et al., 2003) with only modest results (sensitivity 63-

89%, specificity 68-97%). The performance of the immunochromatographic Rapirun

in adults showed a poor sensitivity of 82% and a specificity of 83% (Leodolter et al.,

2003). The disadvantage of both tests is that the sensitivity is even lower when frozen

urine samples are used (Adachi et al., 2002).

39

2.5. Vomit and saliva

H. pylori has been cultivated from vomitus in a clinical experimental study (Parsonnet

et al., 1999). The EIA based tests for detection of H. pylori antibodies in saliva have

no current role in patient management because of less than optimal accuracy

(Malfertheiner et al., 2007, Kabir, 2003). A PCR method has been applied also to

saliva (Tiwari et al., 2005) but there has been considerable variation in the detection

rate (Kabir, 2004).

3. Diagnosis of H. pylori in special clinical situations

3.1. Bleeding peptic ulcer

H. pylori eradication almost eliminates the rebleeding risk in H. pylori infected

subjects; the risk is about 50% within the subsequent 10 years, if patients are left

untreated after an ulcer has healed (Gisbert et al., 2004a). The diagnosis of H. pylori

in patients with a bleeding peptic ulcer is challenging. The number of biopsies that

can be taken in clinically unstable patients is limited, and the regularly used

antisecretory medication impairs the performance of invasive tests, UBT and stool

antigen tests (Gisbert and Abraira, 2006a). Accordingly, during acute bleeding the

sensitivity of RUT was shown to be 67%, histology 70% and culture 45% in a meta-

analysis (Gisbert and Abraira, 2006a) and even the combination of RUT and histology

was unreliable to rule out H. pylori infection (Güell et al., 2006, Chung et al., 2001).

UBT is sensitive if performed during the first 24 hours of admission (Winiarski et al.,

2003, Chung et al., 2001) with a sensitivity of 93% in a meta-analysis (Gisbert and

Abraira, 2006a). PPI therapy, the standard treatment of patients presenting with upper

gastrointestinal bleeding, impairs the UBT result enormously; 92% of patients with

regular dose PPI therapy after one week (Peitz et al., 2004) and 60% of patients with

high dose PPI therapy after three days (Udd et al., 2003) had negative RUT. However,

serology has the advantage of not being affected by PPI therapy (Gisbert and Abraira,

2006a). The results of the polyclonal antibody-based stool antigen test have been

inconsistent: some studies have shown 100% sensitivity with very low specificity

(Griñó et al., 2003, van Leerdam et al., 2003) and in some studies the sensitivity has

40

been very poor, 74% (Gisbert et al., 2004d). After acute bleeding the combination of

several tests and repeating them in negative cases, and with a sufficient interval after

antisecretory medications diminishes the false negative results (Güell et al., 2006,

Epple et al., 1997, Gisbert and Abraira, 2006a).

3.2. Children

In children presenting with abdominal symptoms, H. pylori-associated diseases such

as ulcer disease, which might explain the symptoms, are uncommon and thus

clinically appropriate diagnostic interventions, not only focusing on H. pylori

infection, should be considered (Koletzko and Feydt-Schmidt, 2001, Jones et al.,

2005). The H. pylori screening for gastric cancer in children is warranted only in

families with a history of gastric cancer (Jones et al., 2005). The main H. pylori

associated extragastric conditions in children are iron-deficiency anaemia and short

stature (Horvitz and Gold, 2006). UBT is excellent in all age groups but false positive

test results may occur in young children (Koletzko and Feydt-Schmidt, 2001). It has

been suggested indeed that this may have led to the wrong conclusions in

epidemiological studies which reported high spontaneous clearance rates of H. pylori

during infancy (Koletzko and Feydt-Schmidt, 2001). Also the compliance in children

less than three years of age may be a problem in UBT (Koletzko and Feydt-Schmidt,

2001). It remains to be shown if the urea blood test has less compliance problems; it

has been very accurate in children 2 to 18 years of age (Jolley and Wagner, 2007).

The sensitivity of serology based on IgG antibody detection is low in children under

the age of six years (Kolho et al., 2002, Kindermann et al., 2001) and in a study from

a developing country none of the children under the age of two infected with H. pylori

had positive serology (Casswall et al., 1999). However, the decline in IgG antibody

titre is useful in monitoring the eradication success also in children (Kolho et al.,

2002, Kato et al., 1999). The evidence of stool antigen tests, although the monoclonal

antibody-based test has performed well in recent studies in all age groups (Koletzko et

al., 2003, Kolho et al., 2006), is currently regarded insufficient, and according to the

last consensus conference of the Canadian Helicobacter Study Group in 2005 and the

multicenter European Study by the Pediatric Task Force on Helicobacter pylori

41

infection, the UBT is the best non-invasive tool for H. pylori diagnosis in children

(Jones et al., 2005) (Mégraud et al., 2005).

3.3. The elderly

H. pylori diagnosis in the elderly has also some special characteristics. The diagnostic

strategies of dyspepsia necessitate a full diagnostic approach which must include

endoscopy so as not to ignore serious diseases more common in the old age. The

diagnosis is more challenging because of the high prevalence of atrophic gastritis (the

diagnostic difficulties are discussed below) and abundant use of antimicrobials (which

impairs the performance of many diagnostic tests) for commonplace infections in old

age (Salles-Montaudon et al., 2002). H. pylori diagnosis and therapy is understated in

the elderly hospitalized for peptic ulcer disease, at least in the United States (Pilotto

and Salles, 2002). In Finland also, bleeding peptic ulcers are common in elderly

women (Paimela et al., 2002) and the abundant use of NSAIDs in the elderly

(Klaukka et al., 2005) may lead to overlook the role of H. pylori. The specificity of

serology as a diagnostic method in the elderly is confusing: as discussed above, in the

absence of a real gold standard in H. pylori diagnostics, the patients with positive

serology alone may be false positives or the serology may be the only sensitive

method to detect H. pylori when it appears in low density (Salomaa-Räsänen et al.,

2004).

.

3.4. Atrophic gastritis

The diagnosis of H. pylori in atrophic gastritis is a challenge since all invasive and

almost all non-invasive tests have been shown to have poor sensitivity in this

connection (Ricci et al., 2007, Lahner et al., 2004, Testoni et al., 2002, Kokkola et al.,

2000). Serology is sensitive, but has arguable specificity as it does not differentiate

between current and past infection. When patients with positive H. pylori serology as

the only marker of H. pylori infection received eradication therapy, they showed a fast

decline in IgG antibodies corresponding to the decline known to be diagnostic for

successful H. pylori eradication therapy (Kokkola et al., 1998). In severe atrophic

corpus gastritis the antibodies decline slowly also spontaneously and in almost one

42

fourth of these patients the antibodies decline below the cut-off level within 10 years

(Kokkola et al., 2003). This discrepancy in positive serology is reflected in the use of

serology as a diagnostic method in the elderly: serology has excellent accuracy in the

elderly if the patients with atrophic corpus gastritis are excluded (Salomaa-Räsänen et

al., 2004). In a population with positive H. pylori serology as the only sign of H.

pylori infection, atrophic gastritis and intestinal metaplasia are much more common

than in patients with current infection detectable by histology or culture. Negative

serology indeed can be used to exclude precancerous conditions with a high

probability in a screening situation (Storskrubb et al., 2005). Also in lymphocytic

gastritis, characterized by a marked increase in the number of intraepithelial

lymphocytes, histology has a low sensitivity in the diagnosis of H. pylori infection

although patients are usually seropositive for H. pylori (Mäkinen et al., 2003).The

need for a sensitive test to detect a still ongoing infection in atrophic gastritis is

urgent, especially if H. pylori eradication therapy in the late stages of gastric atrophy

is shown to prevent gastric cancer.

H. PYLORI ERADICATION THERAPY

According to the Maastricht III Consensus Report (Malfertheiner et al., 2007) the first

choice in H. pylori eradication treatment is a standard dose of PPI combined with

clarithromycin 500mg bid and amoxicillin 1000mg bid or metronidazole 400 to

500mg bid for seven days. If continued for 14 days instead of seven this treatment is

12% more effective. This treatment is preferable if the level of local clarithromycin

resistance is less than 15-20%. Metronidazole instead of amoxicillin is recommended

only if the resistance to it is less than 40%. Another first line therapy recommended is

so called quadruple therapy consisting of a standard dose of PPI combined with

tetracycline 500mg qid, metronidazole 400mg tid, and bismuth subsalicylate 120mg

qid. Antimicrobial resistance is the main factor influencing the treatment outcome

(Cavallaro et al., 2006, Mégraud and Lehours, 2007). The eradication rate following a

seven day therapy regimen with clarithromycin, amoxicillin and PPI has been 72-77%

(Cavallaro et al., 2006) but in Finland the same regimen had an eradication rate of

91% (Koivisto et al., 2005).

43

MUCOSAL RECOVERY AFTER H. PYLORI

ERADICATION THERAPY

After H. pylori eradication therapy the acute inflammation disappears in one month,

but chronic inflammation persists longer (Forbes et al., 1996, Tepeš et al., 1999b,

Franceschi et al., 2002, Hojo et al., 2002, Kuipers and Sipponen, 2006): in a meta-

analysis the chronic inflammation improved in almost all studies; the observation

period in most studies, however, was one year or less (Hojo et al., 2002). In one study

with a four-year follow-up including only patients with a duodenal ulcer, the gastric

antral mucosa recovered in half of the patients to normal (defined as a total gastritis

score of 0 or 1) (Zerbib et al., 2000), and in another study among 11 duodenal ulcer

patients after four-year follow-up, the majority of patients had normal antrum and

corpus mucosa (defined as chronic inflammation grade 0) (Tepeš et al., 1999b).

However, the interobserver agreement in the assessment of the reversibility of

gastritis after eradication therapy has been poor (Tepeš et al., 1999a).

The studies on resolution of atrophy after successful H. pylori eradication

therapy have shown restitution of normal architecture of the mucosa with regression

of atrophic changes at least in some patients (Kokkola et al., 2002, Kuipers and

Sipponen, 2006, Hojo et al., 2002). In a few studies the regression of intestinal

metaplasia has been observed (Kuipers and Sipponen, 2006, Hojo et al., 2002). After

spontaneous disappearence of H. pylori with increasing gastric mucosal atrophy, the

normalization of antrum mucosa with disappearence of intestinal mucosa has been

suggested (Valle et al., 1996). However, in this long-term follow-up of 32 years

biopsies of the antrum mucosa at the beginning of the follow-up were not taken.

HOST RESPONSE TO H. PYLORI

The humoral immune system has only marginal relevance for protective immunity in

H. pylori infection. H. pylori induces a Th1-polarized response that unfortunately

does not result in clearance of the infection. H. pylori is thought to manipulate the

host immune response and inflammation (Kusters et al., 2006). The key activator of

the innate immune response is probably intracellular peptidoglycan (Kusters et al.,

44

2006). H. pylori is capable of inhibiting phagocytosis by macrophages by an as-yet-

unknown mechanism. IL-10-producing T cells seem to be crucial in the control of

inflammation and they enable the bacteria to persist in gastric mucosa (Kusters et al.,

2006). Several cytokine genes have stable polymorphisms which are known to affect

the level of cytokine production in response to H. pylori infection (El-Omar, 2006,

Kusters et al., 2006). The best known of these is interleukin 1- , a potent

proinflammatory cytokine and the most potent known inhibitor of acid secretion

(Kusters et al., 2006). These cytokine polymorphisms may contribute to the risk of

gastric adenocarcinoma, but their contribution to the risk of peptic ulceration is

conflicting (Chakravorty et al., 2006, Robinson et al., 2007). Reflecting the

importance of the innate immune response are the studies carried out in families with

a high incidence of gastric cancer where the combination of certain cytokine

polymorphisms confer a 27-fold increased risk of gastric cancer for those infected

with H. pylori (Houghton and Wang, 2005). The combination of high risk bacterial

genotypes elevated further the risk up to 87-fold (Houghton and Wang, 2005). The

modulating role of the immune response is thought to be responsible for the large

variation in the worldwide incidence rates of gastric cancer (Robinson et al., 2007).

45

PRESENT STUDY

AIMS OF THE PRESENT STUDY

1. To evaluate the performance of stool antigen tests in the diagnosis of H. pylori

infection in the pre-treatment and post-treatment settings.

2. To study the association between the presence of widespread DGM and H. pylori

positive peptic ulcer.

3. To determine the persistence of chronic gastritis and elevated H. pylori antibody

levels after successful eradication and their possible association with each other

and with atrophic gastritis.

46

PATIENTS

Study I

In five centres, 82 adult patients who had clinical indication for gastroscopy and who

were positive for H. pylori with a positive RUT were enrolled. Patients with gastric

surgery or history of malignancy, severe reflux oesophagitis, and PPI or antimicrobial

therapy during the preceding 4 weeks were excluded. The age range was 24 to 79

years, median 53, and 46 patients were females. Patients were considered H. pylori

infected if they had, in addition to the RUT, positive histology or culture. All patients

received H. pylori eradication therapy and the treatment combination was chosen at

the discretion of the endoscopist.

Study II

A total of 1574 volunteers without significant abdominal complaints were collected

following a newspaper announcement and screened for H. pylori infection with a

rapid whole-blood antibody test (Pyloriset Screen II, Orion Diagnostica, Espoo

Finland, or Biocard Helicobacter pylori IgG, AniTech Ltd, Vantaa, Finland) and the

positive screening results were confirmed by EIA-based serology and 13C UBT.

Subjects with positive results in both of the confirmatory tests were enrolled.

Exclusion criteria included antimicrobial treatment during the previous 2 months, use

of H2-blockers, PPIs or bismuth during the previous 2 weeks, H. pylori eradication

therapy during the previous 5 years, gastric surgery, chronic gastrointestinal diseases,

pregnancy and lactation, and contraindications to drugs comprising the H. pylori

eradication regimen (allergy to penicillin and prolonged QT-interval and anti-fungal

therapy for dermatophyte infection inhibiting the use of clarithromycin) or allergy to

fruit juices containing the probiotics used in the subsidiary study (Myllyluoma et al.,

2005). Of the 1574 subjects who were screened initially, 300 were positive for H.

pylori using the screening tests, 196 were positive in both of the confirmatory tests,

but 11 of them fulfilled one of the above exclusion criteria and thus, the H. pylori

positive study population consisted of 185 subjects. From among the screened

population who were negative in the H. pylori tests, 114 subjects were invited as a

control group and of the 97 who accepted the invitation all were negative in both of

the confirmatory tests. The final study population consisted of 185 H. pylori positive

subjects, age range 25 to 71 years, median 55, and 97 H. pylori negative subjects, age

47

range 23 to 64 years, median 43 years. In total, there were 186 females and 96 males.

All H. pylori positive subjects received the same eradication regimen viz: amoxicillin

1g bid, clarithromycin 500mg bid, and lanzoprazole 30mg bid for one week.

Study III

This study was retrospective and included 1241 mostly dyspeptic patients assessed by

gastroscopy and an additional 167 patients who had previously undergone HSV.

Patients with gastric malignancies were excluded. The HSV patients had been

gastroscopied previously median 14 years (range 3 to 18 years) after vagotomy to

assess the long-term efficacy of the surgical procedure. The endoscopy and

histological records were reviewed. A total of 198 patients were excluded; duodenal

biopsies were unavailable for 185 patients and 13 of the patients with vagotomy had

undergone gastric resection. Thus, the final study population consisted of 1056

dyspeptics (age range 16 to 88 years, median 52) and 154 patients with vagotomy

(age range 32 to 76 years, median 52 years).

Two duodenal bulb biopsies from two gastroscopies were available for 419 of

the 1056 dyspeptic patients (median interval of gastroscopies five months) and for

102 of the 154 HSV patients (median interval 12.6 months). The patients were

classified as H. pylori positive if they had positive H. pylori culture or serology and as

H. pylori negative if all tests (culture, histology and serology) were negative.

Study IV

Of 345 consecutive patients with a clinical indication for gastroscopy, a subpopulation

of 108 patients, who had at least one endoscopy performed earlier and successful H.

pylori eradication therapy on a known date, comprised the study population. The age

range was 28 to 87 years, median 65 years, and 83 were females. Seventy-seven of the

patients gave a blood sample.

The success of H. pylori eradication therapy that had been verified as

clinically appropriate was reviewed. In 54 patients verification had been made by 13C

UBT, in 26 by a new gastroscopy including histology, in 11 with a significantly

declining IgG antibody titre in paired sera, in three by the stool antigen test, and in six

by culture and histology, and in eight by only histology following the gastroscopy

performed in the present study.

48

ETHICAL CONSIDERATIONS

The Ethics Committee of the Hospital District of Pirkanmaa approved the study I, the

Ethics Committee of the Hospital District of Helsinki and Uusimaa approved the

studies II and IV, and the register-based study III was approved by the Hospital

District of Helsinki and Uusimaa. All patients in studies I, II, and IV gave their

written informed consent before the study and all studies were conducted in

accordance with the Declaration of Helsinki.

METHODS

Gastroscopy and collection of biopsies

During the gastroscopies (I and IV) the evident macroscopical pathology was

recorded and two biopsies were taken from the antrum and the corpus for histology.

In addition, in the study I, one biopsy was taken from the antrum and the corpus for

RUT, and the results were read within three hours according to the manufacturer’s

instructions (Pyloriset Urease, Orion Diagnostica, Espoo, Finland) and the positive

specimens were sent in the original test tubes for culture. In the study III, during the

original gastroscopies most patients had had two biopsies taken from the duodenal

bulb, antrum and corpus.

H. pylori culture

Gastric biopsies (I) were cultured on Brucella agar plates supplemented with 7%

horse blood and also on selective Brucella agar plates containing 1% Iso-Vitalex, 6

mg/L vancomycin, 2 mg/L amphotericin B, and 20 mg/L nalidixic acid. Plates were

incubated at 35 to 37 °C in a microaerobic atmosphere for a maximum of 12 days.

Identification of H. pylori was based on colony appearance, Gram staining, and

positive reactions for oxidase, urease, and catalase. The culture result was available

for most patients in study III.

Histology

All specimens in studies I and IV were stained with haematoxylin-eosin, Alcian blue

(pH 2.5)-periodic acid-Schiff and modified Giemsa stains and the samples were

examined by one experienced pathologist unaware of the identity of the samples. The

49

assessment of gastric histology had been performed according to the updated Sydney

System (Dixon et al., 1996). The retrospective specimens in study III had been stained

with haematoxylin-eosin and Alcian blue (pH 2.5)-periodic acid-Schiff and examined

by several pathologists in the Department of Pathology of the University Hospital

(tertiary referral clinic). The assessment of DGM was graded as 0 (no gastric

metaplasia was evident), 1 (less than 20%), 2 (20 to 50%), and 3 (more than half of

the surface epithelium was covered with gastric epithelium). The duodenal biopsies

with an indistinct grade of metaplasia were re-reviewed by the same pathologist. Of

the biopsies available, the one with the most widespread DGM was included.

Serum tests

Serum samples at baseline (I, II, IV) and after therapy (I, II) were stored at -20 °C

until analysed. To detect H. pylori antibodies the paired sera were analyzed in

parallel. The measurements for both H. pylori antibodies of the immunoglobulin G

(IgG) and IgA classes were performed with an in-house EIA as described earlier and

demonstrated a sensitivity of 99% and specificity of 93% in detecting H. pylori

infection as compared to histology (Oksanen et al., 1998). H. pylori eradication

therapy was considered successful when antibody titres of at least the IgG class had

fallen 40% or more from the pre-treatment level 4 to 6 months after therapy.

Pepsinogen I and II, and fasting gastrin-17 levels (Gastropanel, Biohit PLC.

Diagnostics, Helsinki, Finland), and PCA (Varelisa, Pharmacia Diagnostics, Freiburg,

Germany) were investigated according to the manufacturers instructions. The normal

ranges used were those recommended by the manufacturers: for PG I 30-120 g/l; for

PG II 3-10 g/l; for PG I/PG II 3-20; for gastrin-17 2-10 pmol/l, and for PCA, below

10 U/ml.

UBT

UBTs (I, II) were performed after an overnight fast either with Diabact tablets

(Diabact AB, Uppsala, Sweden) (collection of breath sample 10 min after ingestion)

or liquid 75 mg urea-citric acid solution (collection of breath sample 30 min after

ingestion). Both were validated 13C-labelled UBT methods and the results were

measured either with isotope ratio mass spectrometry or with non-dispersive isotope-

selective infrared spectroscopy. The cut-off values were determined as delta over

50

baseline (DOB) >2.2‰ in the tablet method and >4‰ positive and <2.5‰ negative in

the liquid method.

Stool antigen tests

The stool specimens in studies I and II were collected at baseline and at least 4 weeks

after eradication therapy. They were stored at -20 °C before analysis and were thawed

twice at most. Stool samples before and after eradication therapy were run in parallel

in all three antigen tests. All examinations were performed according to the

manufacturers’ instructions.

In the polyclonal antibody-based Premier Platinum HpSA test (Meridian Inc,

Cincinnati, Ohio, USA), the recommended sample size was diluted with buffer and

added to the micro-titre wells with a peroxidase-conjugated polyclonal antibody and

incubated for 1 hour at room temperature. After washing, substrate was added and

after further incubation of 10 min the stop solution was added. Results were read at

450 nm by spectrophotometry (Titertek Multiscan analyser, Eflab Oy, Helsinki,

Finland) and interpreted according to the manufacturer’s recommendations. An

optical density (OD) value < 0.140 was regarded as negative, 0.140-0.159 as

equivocal (grey zone), and > 0.160 as positive. The grey zone results were retested

using the same stool samples stored at -20 °C. In addition, in the study I, all samples

were also examined after centrifuging the diluted stool specimens and using the

supernatant in the further analysis.

For the Amplified IDEIA HpStAR monoclonal antibody-based test

alternatively known as FemtoLab Cnx (Dako, Glostrup, Denmark) stool samples

suspended in sample diluent were centrifuged for 5 min and the supernatant plus the

monoclonal antibody solution were pipetted into the wells and incubated for 1 hour.

After four washes followed by 10 min incubation with the substrate, the reaction was

stopped. The OD values at 450 nm > 0.190 were interpreted as positive and < 0.190 as

negative.

For the ImmunoCard STAT!HpSA test (Meridian Bioscience Europe, Milan,

Italy), which is based on monoclonal antibodies and a lateral flow chromatography,

stool specimen with diluent was emulsified with an applicator stick and vortexed for

15 s and dispensed to the sample port of the test cassette. After incubation for 5 min at

room temperature, the appearance of a pink-red line in the reading window was

interpreted as a positive test result.

51

Statistical analysis

The comparisons between the different patient groups in study III were analysed using

a multinomial logistic regression model. The results were expressed as odds ratios

(OR), confidence intervals (CI) and P values. The Fisher’s Exact Test was used for

categorical variables (III, IV). Two-sided P values less than 0.05 were considered

statistically significant. Data were analysed using SPSS for Windows software

version 12.0 (SPSS Inc, Chicago, IL). In study IV the decline in chronic gastritis and

elevated antibodies were assessed by the Linear-By-Linear Association Test, the

strata being the time. The comparisons between the patient groups in study IV were

analysed with the Pearson Chi-Square test. These data were analysed using StatXact 7

(Cytel Inc, MA, USA, 2006).

RESULTS

The performance of stool antigen tests (Studies I and II)

Of the stool antigen tests studied, the monoclonal antibody-based EIA test HpStAR

had the highest sensitivity, specificity, accuracy, and likelihood ratio both in primary

diagnosis and in a post-treatment setting. The results of the different stool antigen

tests at baseline and after eradication therapy investigated in studies I and II are

combined and presented in Tables 2 and 3. The statistics of the performance of these

tests before and after eradication therapy are presented in Tables 4 and 5.

In the study I, 18 patients did not submit a follow-up serum sample (five patients had

a therapy failure and received new eradication regimen before the 4 months follow-up

in the study protocol) and one patient was seronegative at baseline; the assessment of

the eradication result in these patients was based on UBT only. In study II, the results

of 182 patients were available after eradication therapy. Of the original 185 patients

who received eradication therapy, one died before the follow-up serum sample was

drawn, one interrupted the eradication therapy for severe headache suspected to be a

side-effect of the medication (this turned out to be viral meningitis), and one had

discrepant results in the confirmatory tests after eradication therapy and was thus

excluded from the analysis.

In the HpSA test, the grey zone values were re-examined using the same stool

specimens according to the manufacturer’s recommendations. Of the 628 test results,

52

19 (3.0%) were within the grey zone in the initial test, but after retesting only three

remained in this category. When the HpSA test was repeated using centrifuged

supernatant for 82 patients in study I the result changed for nine (11%) patients (six at

baseline and three after eradication therapy). However, the results (at baseline a

sensitivity of 96% and after eradication therapy a sensitivity of 75% and specificity of

92%) were only marginally affected. When the data was statistically analyzed the

grey zone results, even after re-examination were considered neither true positive nor

true negative.

Table 2. The results of the three stool antigen tests in 364 patients (82 in Study I and

282 in Study II) at baseline as compared to reference methods. H. pylori positivity

was defined by RUT and culture or histology in patients studied with invasive

methods and by both serology and 13C UBT in patients not endoscopied.

H. pylori HpSA HpStAR ImmunoCard

Positive 267 Positive 2441 258 251

Negative 212 9 16

Grey zone 23

Negative 97 Negative 93 93 86

Positive 44 4 11

1One patient originally had a grey zone result2Three patients originally had a grey zone result3Grey zone results even in re-examination4Three subjects originally had grey zone results

53

Table 3. The results of the three stool antigen tests in 264 patients (82 in Study I and

182 in Study II) after H. pylori eradication therapy as compared with the reference

methods. The patients’ eradication success was classified according to the agreement

of both UBT and serology1.

H. pylori HpSA HpStAR ImmunoCard

Negative 232 Negative 2242 227 224

Positive 7 5 8

Grey zone 13

Positive 32 Positive 25 31 29

Negative 74 1 3

1Serology unavailable for 18 patients and one was originally seronegative at baseline2Seven patients originally had grey zone results3Grey zone even in re-examination4Two patients originally had grey zone results

Table 4. The results of the three stool antigen tests in the diagnosis of H. pylori

infection of 364 patients (82 in Study I and 282 in Study II) at baseline.

Test Sensitivity Specificity PPV NPV Accuracy Likelihood

ratio

% % Pos Neg

HpSA 91.4 95.9 0.98 0.82 0.93 22.3 0.09

HpStAR 96.9 95.9 0.99 0.91 0.96 23.4 0.03

ImmunoCard 94.0 88.7 0.96 0.84 0.93 8.3 0.07

PPV = positive predictive value

NPV = negative predictive value

54

Table 5. The results of the three stool antigen tests in the diagnosis of H. pylori

infection of 264 patients (82 in Study I and 182 in Study II) after eradication therapy.

Test Sensitivity Specificity PPV NPV Accuracy Likelihood

ratio

% % Pos Neg

HpSA 78.1 96.6 0.78 0.97 0.94 23.0 0.22

HpStAR 96.9 97.8 0.86 1.00 0.98 44.0 0.03

ImmunoCard 90.6 96.6 0.78 0.99 0.96 26.6 0.10

PPV = positive predictive value

NPV = negative predictive value

The prevalence of DGM (Study III)

Among the 26 patients who had an ulcer but were H. pylori negative, widespread

DGM was detected in four (15%) patients and there was no association with the

location of the ulcer. The results of the remaining 1030 dyspeptic patients and 154

HSV patients, grouped according to the macroscopic findings in gastroscopy, H.

pylori infection status and the graded extent of DGM are presented in Table 6. The

statistical associations between grade 1 DGM and macroscopic findings were not

significant (OR values between 1.44 and 3.57). The association between widespread

(grade 2 and 3) DGM and endoscopy findings is presented in Table 7. Of the 256 H.

pylori positive duodenal ulcer patients, with or without gastric ulcer, 55% had

widespread (more than 20%) DGM (Table 6). In patients with duodenal ulcer only,

the association with widespread DGM was very high, OR 42 (Table 7). In addition to

duodenal ulcer, widespread DGM was also significantly associated with H. pylori

positive distal gastric ulcer (Table 7). In vagotomy patients, widespread DGM

occurred only in 8.4% (Table 6), although 32 (21%) of them had had recurrent ulcers

after the primary operation; most recurrences had been gastric ulcers (19, of which 16

were prepyloric, and one further patient had both gastric and duodenal ulcer

recurrences). The widespread DGM was an uncommon finding in patients with H.

pylori gastritis and proximal gastric ulcer.

55

Table 6. Prevalence and extent of DGM in the duodenal bulb in 1030 patients grouped

according to the gastroscopy findings (including a history of endoscopy-verified

ulcer) and 154 patients who had had highly selective vagotomy (HSV).

DGM

Patient Group Grade 0

(%)

Grade 1

(%)

Grade 2 and 3

(%)

Duodenal ulcer, n=211 65 (30.8) 20 (9.5) 126 (59.7)

Duodenal & gastric ulcer1, n=45 22 (48.9) 8 (17.8) 15 (33.3)

Distal gastric ulcer2, n=83 51 (61.4) 12 (14.5) 20 (24.1)

Proximal gastric ulcer2, n=81 63 (77.8) 16 (19.8) 2 (2.5)3

Distal & proximal gastric ulcer2, n=17 13 (76.5) 2 (11.8) 2 (11.8)

Gastric ulcer, unknown location, n=4 2 (50.0) 2 (50.0) 0

H. pylori positive gastritis, n=248 216 (87.1) 22 (8.9) 10 (4.0)

H. pylori negative, no ulcer, n=341 260 (76.2) 56 (16.4) 25 (7.3)

After HSV, n=154(H. pylori positive

142, 92%)

123 (79.9) 18 (11.7) 13 (8.4)

All the ulcer patients were H. pylori positive. Of the biopsies available, the one with

the most widespread DGM was included1The location of gastric ulcer was distal in 25, proximal in 15 and both proximal and

distal in 4 patients.2Distal gastric ulcer included ulcers at pylorus, prepylorus (less than 3 cm from

pylorus), and antrum. Proximal gastric ulcer included ulcers in angulus, corpus, and

cardia.3One patient had ulcerative lymphoma

56

Table 7. Odds Ratios (OR), Confidence Intervals (CI) and P values in H. pylori

positive ulcer patients, H. pylori negative non-ulcer patients and HSV patients

according to the prevalence of widespread (grade 2 and 3) DGM. The group of

patients with H. pylori gastritis and no ulcer disease was used as the reference

category.

Patient group No. patients OR (CI) P

Duodenal ulcer 126 41.87 (20.77 to 84.41) 0.001

Duodenal and gastric ulcer 15 14.73 (5.91 to 36.68) 0.001

Distal gastric ulcer1 20 8.47 (3.74 to 19.20) 0.001

Proximal gastric ulcer1 2 0.69 (0.15 to 3.21) 0.63

Distal & proximal gastric ulcer 2 3.32 (0.66 to 16.76) 0.15

H. pylori negative, no ulcer 25 2.08 (0.98-4.42) 0.06

After HSV 13 2.28 (0.97 to 5.36) 0.06

1Distal gastric ulcer included ulcers at pylorus, prepylorus (less than 3 cm from

pylorus), and antrum. Proximal gastric ulcer included ulcers in angulus, corpus, and

cardia.

Persistence of chronic gastric inflammation and elevated H. pylori antibodies

after successful eradication (Study IV)

None of the patients had acute inflammation in the gastric mucosa in gastroscopy

performed in the present study, median 6.4 years, range 0.1-15.4 years after

successful H. pylori eradication therapy. However, the persistence of the chronic

gastric inflammation and elevated H. pylori antibodies of the IgG class were quite

common findings although the prevalence of both declined with time, as shown in

Table 8. The decline in the prevalence of chronic gastritis (P<0.0001) and of the

elevated H. pylori antibodies (P<0.001) was highly significant (Linear-By-Linear

Association Test, strata is time).

57

Table 8. Proportion of patients (n=108) with chronic inflammation in gastric mucosa

and proportion of patients (sera available n=77) with elevated H. pylori antibodies in

different time cohorts after successful H. pylori eradication therapy.

Number of patients with (%)

Time after H. pylori

eradication (years)

Chronic gastric

inflammation

Elevated H. pylori

antibodies

< 1 12/13 (92) 9/11 (82)

1- 2.9 5/10 (50) 3/3 (100)

3- 4.9 6/17 (35) 2/12 (17)

5- 9.9 12/51 (24) 14/37 (38)

>10 2/17 (12) 3/14 (21)

To assess the long term sequelae, the patients were divided into two groups

according to the time (< 5years, >5years) elapsed from successful eradication therapy.

Persistent chronic gastritis, even in those patients who had less than five years from

therapy, was mostly mild (Table 9). Five or more years after successful eradication

therapy, there was no association between chronic inflammation and elevated H.

pylori antibodies (Tables 10 and 11). The average age of patients at the time of

eradication therapy was 56 years for those with chronic gastric inflammation and 57

years for patients not showing chronic inflammation at least five years after successful

H. pylori eradication therapy.

The prevalence of atrophic corpus gastritis was rare throughout follow-up

(Table 9). One of the study patients developed corpus atrophy suggestive for the

autoimmune type. She had had no gastroscopy performed at the time of eradication

therapy and thus the time point for the atrophy development is not known. In the

assessment of serum markers of corpus atrophy, all patients with low PG I also had a

low ratio of PG I/II and they all had elevated levels of gastrin-17. Although rare,

atrophic corpus gastritis, whether assessed from histological specimens or by serum

markers was not associated with chronic inflammation or elevated H. pylori

antibodies after five years of eradication therapy, as shown in Tables 10 and 11. The

prevalence of intestinal metaplasia in antrum clearly declined with time, however, the

58

proportion of patients with moderate to severe metaplasia remained the same (Table

9). Low gastrin-17 was found in four of the 32 patients with intestinal metaplasia in

antrum. After five years of eradication therapy, intestinal metaplasia in the antrum

was more common in patients with persistent chronic inflammation Table 10) and in

patients with no elevated H. pylori antibodies (Table 11), but the associations were

not statistically significant.

Table 9. The findings in gastric histology of 108 patients presented according to the

time elapsed from successful H. pylori eradication therapy.

Time elapsed from eradication therapy

Finding < 5 y, n=40 (%) > 5 y, n=68 (%)

Chronic gastritis1 23 /40 (58) 14/68 (21)

antrum & corpus 14/23 (61) 8/14 (57)

antrum 8/23 (35) 5/14 (36)

corpus 1/23 (4) 1/14 (7)

grade 2 to 3 gastritis 4/23 (17) 1/14 (7)

Antral intestinal metaplasia2 19/40 (48) 13/68 (19)

grade 2 to 3 7/19 (37) 5/13 (38)

Atrophic corpus gastritis 2/403 (5) 1/684 (1)

1P<0.0001, Pearson Chi-Square2P<0.0024, Pearson Chi-Square3In both, atrophy was of grade 14Atrophy was of grade 3

59

Table 10. Histological findings in 68 patients with and without chronic inflammation

in the gastric mucosa at least five years after successful H. pylori eradication therapy;

serum markers were available for 51 patients.

Marker Chronic inflammation,

n=14 (%)

No chronic inflammation,

n=54 (%)

Chronic atrophic corpus

gastritis

1 (7) 0

Intestinal metaplasia in

antrum1

42 (29) 93 (17)

Serum markers:

Elevated IgG antibodies 3/10 (30) 14/41 (34)

Elevated PCA 1/10 (10) 6 /41 (15)

Low PG I 1/10 (10) 01The difference is not statistically significant, Fisher’s Exact Test2 In two patients intestinal metaplasia was of grade 13In six patients intestinal metaplasia was of grade 1

Table 11. Serum and histological findings in 51 patients with and without

persistently elevated IgG antibodies five years or more after successful H. pylori

eradication therapy.

Marker Elevated IgG antibodies,

n=17 (%)

Not elevated IgG

antibodies, n=34 (%)

Elevated PCA 2 (12) 5 (15)

Low PG I 0 1 (3)

Chronic inflammation

in gastric mucosa

3 (18) 7 (21)

Atrophic corpus gastritis 0 11 (3)

Intestinal metaplasia in

antrum2

2 (12) 9 (26)

1Atrophy was of grade 3.2The difference is not statistically significant, Fisher’s Exact Test.

60

DISCUSSION

Evaluation of stool antigen tests in the diagnosis of H. pylori infection (Studies I

and II)

Stool antigen tests have been under intensive investigation and development during

the last few years. In our study, where three stool antigen tests were compared, the

monoclonal antibody-based stool antigen EIA test (HpStAR) showed the best overall

performance. Our results are in accordance with a recent meta-analysis, in which the

same test showed a pooled sensitivity of 94% and a specificity of 97% (our results,

97% and 96%, respectively) in pre-treatment setting and a sensitivity of 93% and a

specificity of 96% (our results, 97% and 98%, respectively) in post-treatment setting

(Gisbert et al., 2006b). In the same meta-analysis, the polyclonal antibody-based EIA

test (HpSA) showed almost equal specificity but clearly lower sensitivity both in the

pre-treatment and post-treatment settings. Especially in the post-treatment setting, the

differences between the sensitivities of monoclonal and polyclonal antibody-based

EIA tests were substantial (sensitivities 91% and 76%, respectively). This was also

shown in our study, as the specificities of the tests were almost equal (96% for both

tests in pre-treatment, and 98% and 97% in post-treatment setting, respectively),

whereas the sensitivities were for the monoclonal and polyclonal antibody-based tests

at baseline 97% and 92%, respectively, and in the post-treatment setting 97% and

78%, respectively (Table 4 and 5). Furthermore, the differences between positive and

negative test results obtained with the monoclonal antibody-based test are generally

greater as compared with the polyclonal antibody-based test, allowing better

distinction, reproducibility and no need for a grey zone.

In study I, at baseline, the reference methods used were invasive methods

i.e.RUT, histology and culture; and in study I, after eradication therapy and in study II

in the pre- and post-eradication settings, serology and UBT were used. This is in line

with the recommendation that two reference methods should be used when new

methods are being evaluated (Mégraud and Lehours, 2007) and also according to the

fact that biopsy-based methods show low sensitivity in the post-treatment setting

(Laine et al., 1998b, Laine et al., 2000). The different screening and confirmation

methods used in Studies I (positive RUT in screening and confirmation with histology

and serology) and II (positive whole blood serology in screening and confirmation by

61

EIA serology and UBT) had insignificant impact on the performance of the stool tests

at baseline. Of the 300 H. pylori positive subjects screened in Study II, only 196 had

positive results in the both confirmatory tests.The excluded patients were mostly

positive in screening only, but 28 of them were also positive in EIA serology. These

28 UBT negative patients were not studied further with either the stool antigen tests or

by any invasive tests.

Although the experience with rapid “in-office” stool antigen tests is still

limited, including the ImmunoCard studied by us, they were assessed in a meta-

analysis which showed that the results were inferior when compared to the EIA tests

(Gisbert and Pajares, 2004b). As the results of the in-office tests are available within

10 minutes they have been suggested to be suitable for the use at the doctor’s office.

No studies, however, exist of the patients’ ability to give a stool specimen at the office

visit. In some studies, patients have shown an overall reluctance to collect stool

samples (O’Connor et al., 2002). Subsequent to the latest meta-analysis, several other

studies on the performance of the ImmunoCard test have been published with widely

varying sensitivity and specificity values (Lu et al., 2006) as well as inferior results

compared to EIA tests (Chisholm et al., 2004). Also in our study, the ImmunoCard

had only a moderate specificity (89%) when used as a primary diagnostic test,

although after eradication therapy the test had a better overall performance than the

polyclonal antibody-based test.

Although not explored in our study, in some special clinical situations stool

antigen tests have shown the same restrictions described for many other diagnostic

methods. The accuracy of stool antigen tests is hampered by PPI therapy to the same

magnitude as described for UBT (Gisbert and Pajares, 2004b). Despite the promising

reports of the stool antigen tests among patients with a bleeding peptic ulcer (Peitz et

al., 2003) (Gisbert et al., 2004d), none of the stool antigen tests can be recommended

to diagnose H. pylori infection in patients with upper gastrointestinal bleeding;

performance of the polyclonal antibody-based test has been controversial. Moreover,

the information on the monoclonal antibody-based test is limited (Gisbert and

Abraira, 2006a). Partial gastrectomy (Sheu et al., 2002), cirrhosis (Gisbert et al.,

2004b) and atrophic gastritis (Lahner et al., 2004) are other clinical settings when the

performance of stool antigen tests has been suboptimal. The sensitivity of the HpSA

test was only 40% in the study by Scheu and colleagues (Scheu et al., 2002) in

gastrectomy patients after eradication therapy and in the study by Lahner and

62

colleagues (Lahner et al., 2004) in patients with atrophic corpus gastritis having H.

pylori positive histology and serology.

The prevalence of H. pylori infection is rapidly declining as shown in young

age groups and even in the older age groups. Due to the active treatment the number

of H. pylori positive subjects is lower than earlier (Malaty, 2007). In Finland and

other developed countries, H. pylori diagnostic tests are at present used in an

environment with a low prevalence of the infection. This is also highlighted after

eradication therapy, as the eradication rates in Finnish patients may be exceptionally

high, 88% in our studies (I and II combined) and in a recent multi-centre study, 91%

(Koivisto et al., 2005). In practice this means that among 100 patients with H. pylori

prevalence of 10% (for instance H. pylori prevalence among volunteers in our study II

was 12%), a test with sensitivity and specificity of 98% will produce 10 true positive

and 2 false positive test results and thus one of six patients is treated in vain.

Therefore, even if highly sensitive and specific diagnostic methods are used, the low

prevalence of H. pylori infection in the population increases the relative proportion of

false positive test results as compared to true positive ones and warrants the use of

several tests (Laheij et al., 2000).

Among the non-invasive tests, the monoclonal antibody-based stool antigen

EIA may be according to our study somewhat inferior to UBT and serology, but even

so it is a good and practical alternative in the diagnosis of H. pylori infection. In the

pre-treatment setting if endoscopy is performed, the stool antigen test is hardly

needed. However, among younger patients with no need for gastroscopy, the stool

antigen test could be used in combination with, for instance, serology to overcome the

problems caused by the low prevalence. In addition, in post-treatment settings, the

stool antigen test is an alternative to UBT.

The use of DGM in selecting patients for H. pylori eradication therapy (Study

III)

Widespread DGM was a common finding in H. pylori positive patients with duodenal

(60%) or distal gastric ulcer (24%), and it was rare in patients with proximal gastric

ulcer (2.5%) and in patients with H. pylori gastritis but no ulcer (4%).The prevalence

of H. pylori infection varies widely and in many developing countries 80% of adult

population is infected (Malaty, 2007). As in such countries eradication of the infection

63

in the whole population is not possible, the Maastricht III Consensus Report

(Malfertheiner et al., 2007) introduces guidelines for patient selection. Duodenal ulcer

is a generally accepted indication for eradication therapy. However, at the time of

gastroscopy, an ulcer may have healed with only minimal scar formation and the

treatment of patients with bulbar deformity does not prevent new ulcers. The risk for a

new peptic ulcer was 10% during a 10 year follow-up in patients with H. pylori

gastritis and the highest risk was in patients with antral predominant gastritis

(Sipponen et al., 1990). However, in that particular study, only 20% of patients with a

new ulcer that had developed during the follow-up would have received eradication

therapy on the basis of the scar formation seen in the initial gastroscopy. On the other

hand, antral predominant gastritis, suggested as an indication for H. pylori

eradication, occurs in half of the patients with chronic dyspepsia (Koskenpato et al.,

2002) and during a 6-7 year follow-up, patients with antral predominant gastritis did

not use more gastrointestinal medications or health services as compared to

individuals with other topographic types of gastritis (Heikkinen et al., 2004). Thus,

better diagnostic tools are needed to allocate the H. pylori eradication therapy

properly.

Although the association between DGM and duodenal ulcer is recognised, the

existence of DGM also in H. pylori negative subjects and the controversial results of

its prevalence in different entities have led to an underestimation of the significance

of DGM. As early as the late 1980’s, the presence of H. pylori in histological

specimens of the duodenal bulb was found to be a very strong risk factor for a

duodenal ulcer (RR 51), stronger than gastric metaplasia alone (RR 7.6). (Carrick et

al., 1989) Since the colonization of the duodenal bulb is more probable in association

with widespread DGM, these results are in line with our own study showing an OR of

42 for widespread DGM and duodenal ulcer. Recently, in a prospective study,

positive H. pylori culture from the duodenal bulb was shown to predict the duodenal

ulcer risk in a one-year follow-up (OR 6.29). (Pietroiusti et al., 2005) The sensitivity

of culture is, however, prone to errors, and in order to find H. pylori in the duodenal

bulb, histological specimens with special stains are needed. Instead, DGM and its

extent are easily detected in histological specimens. However, despite the high OR

values in the present study, widespread DGM only detected 55% of patients with

duodenal ulcer but even so widespread DGM could be added to to the list of

indications to eradicate H. pylori. On the other hand, the proportion of gastritis

64

patients with widespread DGM but no ulcer was 4% in our study. This is an

acceptable number of patients receiving eradication therapy with no shown benefit (at

least as duodenal ulcer is concerned) if widespread DGM is used to detect patients for

treatment. Widespread DGM was rare or non-existent in patients with proximal

gastric ulcers or with atrophic gastritis, both recommended indications for eradication

therapy to prevent gastric cancer (Malfertheiner et al., 2007).

Based on previous studies (Jönsson et al., 1988) the low prevalence of

widespread DGM (8.4%) in HSV patients was an unexpected finding in our study.

Duodenal ulcer disease has been for all patients the indication for HSV and thus, it is

conceivable that at the time of the operation the patients had had a high prevalence of

widespread DGM which then vanished after operation probably in parallel with the

decline in duodenal acidity. As 92% of the HSV patients were H. pylori infected

many years after the operation, this opposes the hypothesis that DGM and duodenal

H. pylori infection with resulting inflammation (and extension of DGM in a vicious

circle) are crucial in the pathogenesis of duodenal ulceration. If anything, this finding

supports the observation of the significance of acid output in determining the location

of the ulcer since most of the ulcer recurrences of these HSV patients had occurred in

the prepyloric area. This is also in accordance with the finding of the high acid

exposure impairing the function of Cdx2, the master regulator of intestinal

differentiation, the suggested mechanism behind the DGM (Faller et al., 2004).

The association between persistent chronic inflammation and elevated H. pylori

antibodies after successful eradication therapy (Study IV)

The persistence of chronic inflammation in the gastric mucosa (21%) and elevated H.

pylori antibodies (33%) were common findings after five years or more of successful

H. pylori eradication, both are new findings as most of former follow-up studies have

only lasted for a few years (Hojo et al. 2002). There was no association between these

two phenomena and neither of them was associated with atrophic gastritis. Five years

after eradication therapy, the persistent chronic inflammation was mild in all patients

except one. For intestinal metaplasia in the antrum, although the prevalence decreased

with time, the grade was stable. One patient developed an autoimmune type of total

gastric atrophy between two consecutive gastroscopies, however, at the time of the

65

eradication therapy the gastroscopy was not performed and thus, the precise time for

the development of the atrophy can only be guessed.

Most of the studies assessing the reversibility of chronic gastritis after

eradication therapy have shown significant improvement (Hojo et al. 2002). However,

the follow-up time in these studies has been mostly one year or less. In one of these

studies with multiple biopsies taken both from the antrum and the corpus, chronic

inflammation was present in all patients after one year of successful eradication

therapy (Franceschi et al., 2002). However, studies with a longer follow-up (up to

four years) revealed conflicting results; one study showed normal antral mucosa in

only half of the patients (Zerbib et al. 2000) whereas in the other, normal antrum and

corpus mucosa was detected in the majority of the patients (Tepeš et al., 1999b).

Some of the discrepancies in the studies arise from the difficulties in the assessment

of mild chronic gastritis (Tepeš et al., 1999b). The advantage of our study is that one

experienced pathologist assessed all patient specimens unaware of the study protocol.

The possible role of the intensity of the inflammation before eradication therapy on

the persistence of the chronic inflammation after eradication has not been studied so

far.

Animal studies of experimental gastric cancer (testing the inflammatory

hypothesis) have shown that a longer duration of the infection increases the risk of

cancer, and there may be a point of no return when eradication therapy cannot

anymore prevent the cancer development (Cai et al., 2005). In our study, the average

age of the patients at the time of eradication therapy did not differ among those with

and those without chronic inflammation and therefore it is improbable that the

duration of H. pylori infection (presumably obtained in the childhood) would have

contributed to the persistence of chronic inflammation. Although we were not able to

show any association between the persistence of chronic inflammation and atrophic

gastritis, the clinical significance of the sustained inflammation in gastric

carcinogenesis remains to be shown.

Although a significant decline in H. pylori antibody titres 4 to 6 months after

eradication therapy indicates successful therapy (Kosunen et al., 1992, Lerang et al.,

1998), it may take several years before the antibodies decline below the cut-off level

(Cutler et al., 1998). The reason for the persistent positive serology is not known and

in our study it was not shown to be associated with persistent chronic inflammation

either. However, the persistent seropositivity emphasizes the importance of early

66

verification of the eradication success. If later on diagnostic methods of H. pylori

infection are needed, tests capable of detecting a current infection should be used.

In our study, there was no association between persistent mostly mild chronic

gastric inflammation and elevated H. pylori antibodies and the results do not support

the hypothesis of H. pylori triggering an autoimmune process which could continue

after eradication therapy. Neither was there any association between persistently

elevated H. pylori antibodies or chronic inflammation and precancerous conditions

e.g. atrophic gastritis and intestinal metaplasia. Considering the disappointing results

of the human gastric cancer prevention studies (Wong et al., 2004), the question as to

whether sustained chronic inflammation has any role to play remains to be studied.

67

CONCLUSIONS

1. Of the stool antigen tests evaluated, the monoclonal antibody-based EIA

demonstrated the best performance both in pretreatment and posttreatment settings.

This is an accurate test for the diagnosis of H. pylori infection and the performance is

comparable to other diagnostic tools. The polyclonal antibody-based EIA and the

rapid test were less accurate.

2. The presence of widespread DGM was strongly associated with a H. pylori positive

duodenal ulcer and could possibly be used to detect patients at a high risk for

duodenal ulceration for H. pylori eradication therapy.

3. Chronic gastric inflammation and elevated H. pylori antibodies persist in some

individuals, due to unknown reasons, for years after successful eradication therapy.

These phenomena are not associated with each other or with atrophic changes in the

gastric mucosa.

68

ACKNOWLEDGEMENTS

This study was carried out at the Department of Bacteriology and Immunology,

Haartman Institute, University of Helsinki during the years 2002 to 2007. The study

was conducted in close co-operation with Malmi and Herttoniemi Hospitals, Health

Centre of Helsinki, and Peijas Hospital, Vantaa. I wish to express my gratitude to a

friendly and supportive atmosphere.

I am greatly indebted to my supervisor Docent Hilpi Rautelin, MD, she is

“The Super Coach”. I have enjoyd the ease of the dealings with her and her

constructive criticism. She has succeeded to transmit some of her endless energy and

optimism to her trainee. I admire her thorough devotion to teaching and supervising.

She was always available as a friend and supervisor even after office hours. This

thesis is to her credit. I am far behind her social skils and despite her intense

persuasion I regret not to learn to share her delight to give lectures and to champagne.

Knowing there is still more to learn I look forward the future challenges and

collaboration as a part of her “extended family”.

I am grateful to Professor Seppo Meri, MD, the Head of Haartman Institute for

the possibility to work in his department with excellent facilities for scientific work.

I owe my sincerest thanks to the reviewers of this thesis Docent Tuomo Karttunen,

MD, and Docent Olli Meurman, MD, for their expert comments and gentle criticism.

I wish to express my warmest thanks to Aino Oksanen, MD, PhD, at present

also my unpretentious taskmaster, for her scientific enthusiasm, friendship and

sharing the love for horses.

I owe my gratitude to Professor Kari Seppälä, MD, for teaching me

gastroenterology and not forgetting the long philosophical discussions of the medicine

and hospital insitutions and elaborate memories when reviewing the patient records.

I am grateful for Professor Pentti Sipponen, MD, for his high expertise in

histological analyses and his cheerful support and admire for “female energy”.

I owe my respectful gratitude to Professor Seppo Sarna for his skilful help in

statistical analyses.

I owe my sincerest thanks to collaborators and co-authors: Tuuli Löfgren, MD,

and Docent Anna-Liisa Karvonen, MD, for the collaboration in Study I, Eveliina

Myllyluoma, PhD, Docent Riitta Korpela, PhD, and Leena Seppo, MSc, for fruitful

69

collaboration in Study II at Valio Ltd. Research Centre, Anna Sankila, MD, PhD,

Professor Timo Kosunen, MD, Professor Reijo Tilvis, MD, Hannu Hyvärinen, MD,

PhD, and Perttu Arkkila, MD, PhD, in Study III and Auli Linnala, PhD, at Biohit PLC

Diagnostics in Study IV. I also owe my gratitude to the patients who volunteered to

donate their specimens.

My warmest gratitude I like to express to late Pirjo Kosonen for her skilful and

assiduous laboratory assistance in many of my works.

My appreciation and thanks are also due to Docent Malcolm Richardson, PhD,

for revising the language of this thesis and my sincere apology for torturing his

mother tongue.

I am deeply grateful to Ilkka Vanhatalo for expert and expeditious assistance

with my obstinate computer and to Marja-Liisa Merivirta for versatile and efficient

library assistance.

I owe also my gratitude to the fellow workers in Malmi, Herttoniemi and

Peijas hospitals for their respecting attitude to research and extra workload without

grumble.

I love my friends and relatives for making my life sweet and worth living and

for accepting my cocooned hermit nature.

This study was supported by a grant from City of Helsinki. I thank the

manufacturers of the stool antigen tests for providing the test kits for Study I.

Helsinki, September 2007

70

REFERENCES

Anonymous 1994. International Agency for Research on Cancer. Schistosomes, liver flukesand Helicobacter pylori. IARC monographs on the evaluation of carcinigenic risks tohumans. Lyon, France. IARC 61:177-220.

Anonymous 1997. Technical annex: tests used to assess Helicobacter pylori infection.Working Party of the European Helicobacter pylori Study Group. Gut 41:S10-18.

Adachi, K., A. Kawamura, M. Ono, K. Masuzaki, T. Takashima, M. Yuki, H. Fujishiro, S.Ishihara, Y. Kinoshita. 2002. Comparative evaluation of urine-based and other minimallyinvasive methods for the diagnosis of Helicobacter pylori infection. J. Gastroenterol. 37:703-708.

Agha-Amiri, K., U. Peitz, D. Mainz, S. Kahl, A. Leodolter, P. Malfertheiner. 2001. A novelimmunoassay based on monoclonal antibodies for the detection of Helicobacter pyloriantigens in human stool. Z. Gastroenterol. 39:555-560.

Ahmed, F., U. K. Murthy, W. D. Chey, P. P. Toskes, D. A. Wagner. 2005. Evaluation of theEz-HBT Helicobacter blood test to establish Helicobacter pylori eradication. Aliment.Pharmacol. Ther. 22:875-880.

Albillos, A., M. Alvarez-Mon, I. Rossi, M. A. Gonzalo, M. C. Marín, L. Abreu. 1990.Different HCl and pepsinogen I secretion patterns in anatomically defined gastric ulcersubsets. Am. J. Gastroenterol. 85:535-538.

Alm, R. A., L. S. Ling, D. T. Moir, B. L. King, E. D. Brown, P. C. Doig, D. R. Smith, B.Noonan, B. C. Guild, B. L. deJonge, G. Carmel, P. J. Tummino, A. Caruso, M. Uria-Nickelsen, D. M. Mills, C. Ives, R. Gibson, D. Merberg, S. D. Mills, Q. Jiang, D. E. Taylor,G. F. Vovis, T. J. Trust. 1999. Genomic-sequence comparison of two unrelated isolates of thehuman gastric pathogen Helicobacter pylori. Nature 397:176-180.

Amarapurkar, D. N., S. S. Parikh, S. R. Prabhu, R. H. Kalro, H. G. Desai. 1993. Is gastricmetaplasia essential for duodenal ulcer? J. Clin. Gastroenterol. 17:204-206.

Andersen, L. P. and F. Espersen. 1992. Immunoglobulin G antibodies to Helicobacter pyloriin patients with dyspeptic symptoms investigated by the western immunoblot technique. J.Clin. Microbiol. 30:1743-1751.

Andersen, L. P., S. Holck, L. Elsborg, T. Justesen. 1991. The Helicobacter (Campylobacter)pylori-colonized duodenal mucosa and gastric metaplasia. APMIS 99:244-248.

Andersen, L. P., S. Kiilerick, G. Pedersen, A. C. Thoreson, F. Jørgensen, J. Rath, N. E.Larsen, O. Børup, K. Krogfelt, J. Scheibel, S. Rune. 1998. An analysis of seven differentmethods to diagnose Helicobacter pylori infections. Scand. J. Gastroenterol. 33:24-30.

Annibale, B., E. Di Giulio, P. Caruana, E. Lahner, G. Capurso, C. Bordi, G. Delle Fave. 2002.The long-term effects of cure of Helicobacter pylori infection on patients with atrophic bodygastritis. Aliment. Pharmacol. Ther. 16:1723-1731.

Annibale, B., E. Lahner, A. Santucci, D. Vaira, A. Pasquali, C. Severi, R. Mini, N. Figura, G.Delle Fave. 2007. CagA and VacA are immunoblot markers of past Helicobacter pyloriinfection in atrophic body gastritis. Helicobacter 12:23-30.

71

Annibale, B., R. Negrini, P. Caruana, E. Lahner, C. Grossi, C. Bordi, G. Delle Fave. 2001Sep. Two-thirds of atrophic body gastritis patients have evidence of Helicobacter pyloriinfection. Helicobacter 6:225-233.

Archimandritis, A., P. Apostolopoulos, S. Sougioultzis, I. Delladetsima, P. Davaris, M.Tzivras. 2000a. The CLO test is unreliable in diagnosing H. pylori infection in post-surgicalstomach; is there any role of H. pylori in peptic ulcer recurrence? Eur. J. Gastroenterol.Hepatol. 12:93-96.

Archimandritis, A., M. Tzivras, S. Sougioultzis, I. Papaparaskevas, P. Apostolopoulos, A.Avlami, P. S. Davaris. 2000b. Rapid urease test is less sensitive than histology in diagnosingHelicobacter pylori infection in patients with non-variceal upper gastrointestinal bleeding. J.Gastroenterol. Hepatol. 15:369-373.

Ashton-Key, M., T. C. Diss, P. G. Isaacson. 1996. Detection of Helicobacter pylori in gastricbiopsy and resection specimens. J. Clin. Pathol. 49:107-111.

Bago, J., D. Strini , Z. B. Halle, D. Jandri , M. Tomi , A. Bili , P. Bago. 2002. Effect ofHelicobacter pylori eradication on extent of duodenal gastric metaplasia and grade ofgastritis. Coll. Antropol. 26:557-563.

Baxter, A. G., M. A. Jordan, P. A. Silveira, W. E. Wilson, I. R. Van Driel. 2005. Geneticcontrol of susceptibility to autoimmune gastritis. Int. Rev. Immunol. 24:55-62.

Bergman, M. P., C. M. Vandenbroucke-Grauls, B. J. Appelmelk, M. M. D'Elios, A. Amedei,A. Azzurri, M. Benagiano, G. Del Prete. 2005. The story so far: Helicobacter pylori andgastric autoimmunity. Int. Rev. Immunol. 24:63-91.

Bermejo, F., D. Boixeda, J. P. Gisbert, V. Defarges, J. M. Sanz, C. Redondo, C. Martín deArgila, A. García Plaza. 2002. Rapid urease test utility for Helicobacter pylori infectiondiagnosis in gastric ulcer disease. Hepatogastroenterology 49:572-575.

Bizzozero, G. 1893. Ueber die schlauchförmigen Drüsen des Magendarmkanals und dieBeziehungen ihres Epithels zu dem Oberflächenepithel der Schleimhaut. Arch Mikr Anat42:82-152.

Blanchard, T. G., M. L. Drakes, S. J. Czinn. 2004. Helicobacter infection: pathogenesis. Curr.Opin. Gastroenterol. 20:10-15.

Blaser, M. J. 1998. Helicobacters are indigenous to the human stomach: duodenal ulcerationis due to changes in gastric microecology in the modern era. Gut 43:721-727.

Blaser, M. J. 1999a. Hypothesis: the changing relationships of Helicobacter pylori andhumans: implications for health and disease. J. Infect. Dis. 179:1523-1530.

Blaser, M. J. 1999b. In a world of black and white, Helicobacter pylori is gray. Ann. Intern.Med. 130:695-697.

Brandi, G., B. Biavati, C. Calabrese, M. Granata, A. Nannetti, P. Mattarelli, G. Di Febo, G.Saccoccio, G. Biasco. 2006. Urease-positive bacteria other than Helicobacter pylori in humangastric juice and mucosa. Am. J. Gastroenterol. 101:1756-1761.

Brooks, H. J., D. Ahmed, M. A. McConnell, G. O. Barbezat. 2004. Diagnosis of Helicobacterpylori infection by polymerase chain reaction: is it worth it? Diagn. Microbiol. Infect. Dis.50:1-5.

72

Bynum, T. E. 1991. Non-acid mechanisms of gastric and duodenal ulcer formation. J. Clin.Gastroenterol. 13:S56-64.

Cai, X., J. Carlson, C. Stoicov, H. Li, T. C. Wang, J. Houghton. 2005. Helicobacter feliseradication restores normal architecture and inhibits gastric cancer progression in C57BL/6mice. Gastroenterology 128:1937-1952.

Campbell, M., D. Machin. 1993. Medical statistics: a commonsense approach, p. 36-38. JohnWiley & Sons Ltd, West Sussex, England.

Cao, Q., Z. H. Ran, S. D. Xiao. 2007. Screening of atrophic gastritis and gastric cancer byserum pepsinogen, gastrin-17 and Helicobacter pylori immunoglobulin G antibodies. J. Dig.Dis. 8:15-22.

Capurso, G., A. Carnuccio, E. Lahner, F. Panzuto, F. Baccini, G. D. Fave, B. Annibale. 2006.Corpus-predominant gastritis as a risk factor for false-negative 13C-urea breath test results.Aliment. Pharmacol. Ther. 24:1453-1460.

Cardenas, V. M., Z. D. Mulla, M. Ortiz, D. Y. Graham. 2006. Iron deficiency andHelicobacter pylori infection in the United States. Am. J. Epidemiol. 163:127-134.

Carrick, J., A. Lee, S. Hazell, M. Ralston, G. Daskalopoulos. 1989. Campylobacter pylori,duodenal ulcer, and gastric metaplasia: possible role of functional heterotopic tissue inulcerogenesis. Gut 30:790-797.

Caselli, M., M. R. Bovolenta, A. Aleotti, L. Trevisani, G. Stabellini, N. Ricci. 1988. Epithelialmorphology of duodenal bulb and Campylobacter-like organisms. J. Submicrosc. Cytol.Pathol. 20:237-242.

Casswall, T. H., H. O. Nilsson, M. Bergström, P. Aleljung, T. Wadström, A. K. Dahlström,M. J. Albert, S. A. Sarker. 1999. Evaluation of serology, 13C-urea breath test, and polymerasechain reaction of stool samples to detect Helicobacter pylori in Bangladeshi children. J.Pediatr. Gastroenterol. Nutr. 28:31-36.

Castillo-Rojas, G., M. A. Ballesteros, S. Ponce de León, R. Morales-Espinosa, A. Cravioto,Y. L pez-Vidal. 2002. Bleeding peptic ulcers and presence of Helicobacter pylori by varioustests: a case-control study. Eur. J. Gastroenterol. Hepatol. 14:1113-1118.

Castro-Fernández, M., D. Sánchez-Muñoz, E. García-Díaz, J. Miralles-Sanchiz, J. Vargas-Romero. 2004. Diagnosis of Helicobacter pylori infection in patients with bleeding ulcerdisease: rapid urease test and histology. Rev. Esp. Enferm. Dig. 96:395-398.

Cavallaro, L. G., B. Egan, C. O'Morain, F. Di Mario. 2006. Treatment of Helicobacter pyloriinfection. Helicobacter 11:36-39.

Chakravorty, M., A. Ghosh, A. Choudhury, A. Santra, J. Hembrum, S. Roychoudhury. 2006.Interaction between IL1B gene promoter polymorphism in determining the susceptibility toHelicobacter pylori associated duodenal ulcer. Human mutation 27:411-419.

Chang, C. C., S. Pan, G. S. Lien, S. H. Chen, C. J. Cheng, J. D. Liu, Y. S. Cheng, F. M. Suk.2001. Investigation of the extent of gastric metaplasia in the duodenal bulb by usingmethylene blue staining. J. Gastroenterol. Hepatol. 16:729-733.

73

Chey, W. D., U. Murthy, P. Toskes, S. Carpenter, L. Laine. 1999. The 13C-urea blood testaccurately detects active Helicobacter pylori infection: a United States, multicenter trial. Am.J. Gastroenterol. 94:1522-1524.

Chisholm, S. A., R. J. Owen, E. L. Teare, S. Saverymuttu. 2001. PCR-based diagnosis ofHelicobacter pylori infection and real-time determination of clarithromycin resistance directlyfrom human gastric biopsy samples. J. Clin. Microbiol. 39:1217-1220.

Chisholm, S. A., C. L. Watson, E. L. Teare, S. Saverymuttu, R. J. Owen. 2004. Non-invasivediagnosis of Helicobacter pylori infection in adult dyspeptic patients by stool antigendetection: does the rapid immunochromatography test provide a reliable alternative toconventional ELISA kits? J. Med. Microbiol. 53:623-627.

Chung, I. K., S. J. Hong, E. J. Kim, J. Y. Cho, H. S. Kim, S. H. Park, M. H. Lee, S. J. Kim, C.S. Shim. 2001. What is the best method to diagnose Helicobacter infection in bleeding pepticulcers?: a prospective trial. Korean J. Intern. Med. 16:147-152.

Claeys, D., G. Faller, B. J. Appelmelk, R. Negrini, T. Kirchner. 1998. The gastric H+,K+-ATPase is a major autoantigen in chronic Helicobacter pylori gastritis with body mucosaatrophy. Gastroenterology 115:340-347.

Correa, P. 1988. A human model of gastric carcinogenesis. Cancer Res. 48:3554-3560.

Cutler, A. F., S. Havstad, C. K. Ma, M. J. Blaser, G. I. Pérez-Pérez, T. T. Schubert. 1995.Accuracy of invasive and noninvasive tests to diagnose Helicobacter pylori infection.Gastroenterology 109:136-141.

Cutler, A. F., V. M. Prasad, P. Santogade. 1998. Four-year trends in Helicobacter pylori IgGserology following successful eradication. Am. J. Med. 105:18-20.

Cutler, A. F. and P. Toskes. 1999. Comparison of 13C-urea blood test to 13C-urea breath testfor the diagnosis of Helicobacter pylori. Am. J. Gastroenterol. 94:959-961.

de Block, C. E., I. H. de Leeuw, J. J. Bogers, P. A. Pelckmans, M. M. Ieven, E. A. van Marck,V. van Hoof, E. Máday, K. L. van Acker, L. F. van Gaal. 2002. Helicobacter pylori, parietalcell antibodies and autoimmune gastropathy in type 1 diabetes mellitus. Aliment. Pharmacol.Ther. 16:281-289.

de Boer, W. A. 1997. Diagnosis of Helicobacter pylori infection. Review of diagnostictechniques and recommendations for their use in different clinical settings. Scand. J.Gastroenterol. 32:35-42.

Delaney, B., A. C. Ford, D. Forman, P. Moayyedi, M. Qume. 2007. Initial managementstrategies for dyspepsia. Cochrane Database of Syst Rev.

D'Elios, M. M., A. Amedei, A. Azzurri, M. Benagiano, G. Del Prete, M. P. Bergman, C. M.Vandenbroucke-Grauls, B. J. Appelmelk. 2005. Molecular specificity and functionalproperties of autoreactive T-cell response in human gastric autoimmunity. Int. Rev. Immunol.24:111-122.

D'Elios, M. M., B. J. Appelmelk, A. Amedei, M. P. Bergman, G. Del Prete. 2004. Gastricautoimmunity: the role of Helicobacter pylori and molecular mimicry. Trends Mol. Med.10:316-323.

74

Delport, W. and S. W. van der Merwe. 2007. The transmission of Helicobacter pylori: Theeffects of analysis method and study population on inference. Baillieres Best Pract. Res. Clin.Gastroenterol. 21:215-236.

Dickey, W., B. D. Kenny, J. B. McConnell. 1996. Effect of proton pump inhibitors on thedetection of Helicobacter pylori in gastric biopsies. Aliment. Pharmacol. Ther. 10:289-293.

Dixon, M. F. 2000. Patterns of inflammation linked to ulcer disease. Baillieres Best Pract.Res. Clin. Gastroenterol. 14:27-40.

Dixon, M. F., R. M. Genta, J. H. Yardley, P. Correa. 1996. Classification and grading ofgastritis. The updated Sydney System. International Workshop on the Histopathology ofGastritis, Houston 1994. Am. J. Surg. Pathol. 20:1161-1181.

Donahue, P. E., C. Griffith, H. M. Richter. 1996. A 50-year perspective upon selective gastricvagotomy. Am. J. Surg. 172:9-12.

Egan, B. J. and C. A. O'Morain. 2007. A historical perspective of Helicobactergastroduodenitis and its complications. Baillieres Best Pract. Res. Clin. Gastroenterol.21:335-346.

Ekesbo, R., E. Toth, F. T. Fork, M. Held, I. Nilsson, T. Wadström, K. Sjölund. 2006. ChronicHelicobacter pylori infection in a population in southern Sweden analysed by histopathology,immunoblot and ELISA serology. Eur. J. Gastroenterol. Hepatol. 18:589-593.

Ekström, A. M., M. Held, L. E. Hansson, L. Engstrand, O. Nyrén. 2001. Helicobacter pyloriin gastric cancer established by CagA immunoblot as a marker of past infection.Gastroenterology 121:784-791.

El-Omar, E. M. 2006. Role of host genes in sporadic gastric cancer. Baillieres Best Pract.Res. Clin. Gastroenterol. 20:675-686.

El-Omar, E. M., M. Carrington, W. H. Chow, K. E. McColl, J. H. Bream, H. A. Young, J.Herrera, J. Lissowska, C. C. Yuan, N. Rothman, G. Lanyon, M. Martin, J. F. Fraumeni Jr, C.S. Rabkin. 2000. Interleukin-1 polymorphisms associated with increased risk of gastriccancer. Nature 404:398-402.

Epple, H. J., F. W. Kirstein, C. Bojarski, J. Frege, M. Fromm, E. O. Riecken, J. D. Schulzke.1997. 13C-urea breath test in Helicobacter pylori diagnosis and eradication. Correlation tohistology, origin of 'false' results, and influence of food intake. Scand. J. Gastroenterol.32:308-314.

Eriksson, N. K., M. A. Färkkilä, M. E. Voutilainen, P. E. Arkkila. 2005. The clinical value oftaking routine biopsies from the incisura angularis during gastroscopy. Endoscopy 37:532-536.

Fabre, R., I. Sobhani, P. Laurent-Puig, N. Hedef, N. Yazigi, C. Vissuzaine, I. Rodde, F. Potet,M. Mignon, J. P. Etienne. 1994. Polymerase chain reaction assay for the detection ofHelicobacter pylori in gastric biopsy specimens: comparison with culture, rapid urease test,and histopathological tests. Gut 35:905-908.

Faller, G., A. Dimmler, T. Rau, S. Spaderna, F. Hlubek, A. Jung, T. Kirchner, T. Brabletz.2004. Evidence for acid-induced loss of Cdx2 expression in duodenal gastric metaplasia. J.Pathol. 203:904-908.

75

Faller, G., H. Steininger, J. Kränzlein, H. Maul, T. Kerkau, J. Hensen, E. G. Hahn, T.Kirchner. 1997. Antigastric autoantibodies in Helicobacter pylori infection: implications ofhistological and clinical parameters of gastritis. Gut 41:619-623.

Fallone, C. A., V. G. Loo, A. N. Barkun. 1998. Utility of serology in determiningHelicobacter pylori eradication after therapy. Can. J. Gastroenterol. 12:117-124.

Feldman, R. A., J. J. Deeks, S. J. W. Evans, and the Helicobacter pylori Serology Group.1995. Multi-laboratory comparison of eight commercailly available Helicobacter pyloriseroplogy kits. Eur. J. Clin. Microbiol. Infect. Dis. 14:428-433.

Ferrucci, P. F. and E. Zucca. 2007. Primary gastric lymphoma pathogenesis and treatment:what has changed over the past 10 years? Br. J. Haematol. 136:521-538.

Fitzgibbons, P. L., C. P. Dooley, H. Cohen, M. D. Appleman. 1988. Prevalence of gastricmetaplasia, inflammation, and Campylobacter pylori in the duodenum of members of anormal population. Am. J. Clin. Pathol. 90:711-714.

Forbes, G. M., J. R. Warren, M. E. Glaser, D. J. Cullen, B. J. Marshall, B. J. Collins. 1996.Long-term follow-up of gastric histology after Helicobacter pylori eradication. J.Gastroenterol. Hepatol. 11:670-673.

Ford, A. C., B. C. Delaney, D. Forman, P. Moayyedi. 2006. Eradication therapy for pepticulcer disease in Helicobacter pylori positive patients.Update of Cochrane Database Syst Rev.

Franceschi, F. and A. Gasbarrini. 2007. Helicobacter pylori and extragastric diseases.Baillieres Best Pract. Res. Clin. Gastroenterol. 21:325-334.

Franceschi, F., R. M. Genta, A. R. Sepulveda. 2002. Gastric mucosa: long-term outcome aftercure of Helicobacter pylori infection. J. Gastroenterol. 37:17-23.

Futami, H., M. Takashima, T. Furuta, H. Hanai, E. Kaneko. 1999. Relationship betweenHelicobacter pylori infection and gastric metaplasia in the duodenal bulb in the pathogenesisof duodenal ulcer. J. Gastroenterol. Hepatol. 14:114-119.

Gatta, L., N. Vakil, C. Ricci, J. F. Osborn, A. Tampieri, F. Perna, M. Miglioli, D. Vaira. 2003.A rapid, low-dose, 13C-urea tablet for the detection of Helicobacter pylori infection beforeand after treatment. Aliment. Pharmacol. Ther. 17:793-798.

Gatta, L., N. Vakil, C. Ricci, J. F. Osborn, A. Tampieri, F. Perna, M. Miglioli, D. Vaira. 2004.Effect of proton pump inhibitors and antacid therapy on 13C urea breath tests and stool test forHelicobacter pylori infection. Am. J. Gastroenterol. 99:823-829.

Gisbert, J. P. and V. Abraira. 2006a. Accuracy of Helicobacter pylori diagnostic tests inpatients with bleeding peptic ulcer: a systematic review and meta-analysis. Am. J.Gastroenterol. 101:848-863.

Gisbert, J. P., F. de la Morena, V. Abraira. 2006b. Accuracy of monoclonal stool antigen testfor the diagnosis of H. pylori infection: a systematic review and meta-analysis. Am. J.Gastroenterol. 101:1921-1930.

Gisbert, J. P., S. Khorrami, F. Carballo, X. Calvet, E. Gené, J. E. Dominguez-Muñoz. 2004a.H. pylori eradication therapy vs. antisecretory non-eradication therapy (with or without long-term maintenance antisecretory therapy) for the prevention of recurrent bleeding from pepticulcer.Update of Cochrane Database Syst Rev.

76

Gisbert, J. P. and J. M. Pajares. 2001. Diagnosis of Helicobacter pylori infection by stoolantigen determination: a systematic review. Am. J. Gastroenterol. 96:2829-2838.

Gisbert, J. P. and J. M. Pajares. 2004b. Stool antigen test for the diagnosis of Helicobacterpylori infection: a systematic review. Helicobacter 9:347-368.

Gisbert, J. P. and J. M. Pajares. 2004c. Review article: C-urea breath test in the diagnosis ofHelicobacter pylori infection -- a critical review. Aliment. Pharmacol. Ther. 20:1001-1017.

Gisbert, J. P. and J. M. Pajares. 2005. 13C-urea breath test in the management of Helicobacterpylori infection. Dig. Liver Dis. 37:899-906.

Gisbert, J. P., M. Trapero, X. Calvet, J. Mendoza, M. Quesada, M. Güell, J. M. Pajares.2004d. Evaluation of three different tests for the detection of stool antigens to diagnoseHelicobacter pylori infection in patients with upper gastrointestinal bleeding. Aliment.Pharmacol. Ther. 19:923-929.

Gomollón, F., J. A. Ducons, S. Santolaria, I. Lera Omiste, R. Guirao, M. Ferrero, M.Montoro. 2003. Breath test is very reliable for diagnosis of Helicobacter pylori infection inreal clinical practice. Dig. Liver Dis. 35:612-618.

González, C. A., S. Peña, G. Capellá. 2003. Clinical usefulness of virulence factors ofHelicobacter pylori as predictors of the outcomes of infection. What is the evidence? Scand.J. Gastroenterol. 38:905-915.

Goodman, K. J., S. L. Joyce, K. P. Ismond. 2006. Extragastric diseases associated withHelicobacter pylori infection. Curr. Gastroenterol. Rep. 8:458-464.

Graham, D. Y., A. R. Opekun, M. Jogi, Y. Yamaoka, H. Lu, R. Reddy, H. M. El-Zimaity.2004a. False negative urea breath tests with H2-receptor antagonists: interactions betweenHelicobacter pylori density and pH. Helicobacter 9:17-27.

Graham, D. Y., A. R. Opekun, M.S. Osato, H. M. El-Zimaity, C. K. Lee, Y. Yamaoka, W. A.Qureshi, M. Cadoz, T. P. Monath. 2004b. Challenge model for Helicobacter pylori infectionin human volunteers. Gut 53:1235-1243.

Graham, D. Y. and A. Shiotani. 2005. The time to eradicate gastric cancer is now. Gut54:735-738.

Griñó, P., S. Pascual, J. Such, J. A. Casellas, M. Niveiro, M. Andreu, J. Sáez, J. R. Aparicio,E. Griñó, L. Compañy, R. Laveda, M. Pérez-Mateo. 2003. Comparison of stool immunoassaywith standard methods for detection of Helicobacter pylori infection in patients with upper-gastrointestinal bleeding of peptic origin. Eur. J. Gastroenterol. Hepatol. 15:525-529.

Griñó, P., S. Pascual, J. Such, J. A. Casellas, M. Niveiro, M. Andreu, J. Sáez, E. Griñó, J. M.Palazón, F. Carnicer, M. Pérez-Mateo. 2001. Comparison of diagnostic methods forHelicobacter pylori infection in patients with upper gastrointestinal bleeding. Scand. J.Gastroenterol. 36:1254-1258.

Grove, D. I., G. Koutsouridis, A. G. Cummins. 1998. Comparison of culture, histopathologyand urease testing for the diagnosis of Helicobacter pylori gastritis and susceptibility toamoxycillin, clarithromycin, metronidazole and tetracycline. Pathology 30:183-187.

77

Güell, M., E. Artigau, V. Esteve, J. Sánchez-Delgado, F. Junquera, X. Calvet. 2006.Usefulness of a delayed test for the diagnosis of Helicobacter pylori infection in bleedingpeptic ulcer. Aliment. Pharmacol. Ther. 23:53-59.

Hahn, M., M. B. Fennerty, C. L. Corless, N. Magaret, D. A. Lieberman, D. O. Faigel. 2000.Noninvasive tests as a substitute for histology in the diagnosis of Helicobacter pyloriinfection. Gastrointest. Endosc. 52:20-26.

Hamlet, A., L. Stage, H. Lönroth, C. Cahlin, C. Nyström, A. Pettersson. 1999. A novel tablet-based 13C urea breath test for Helicobacter pylori with enhanced performance during acidsuppression therapy. Scand. J. Gastroenterol. 34:367-374.

Harris, A. W., P. A. Gummett, M. M. Walker, J. J. Misiewicz, J. H. Baron. 1996. Relationbetween gastric acid output, Helicobacter pylori, and gastric metaplasia in the duodenal bulb.Gut 39:513-520.

Heikkinen, M., P. Pikkarainen, M. Vornanen, S. Hollmén, R. Julkunen. 2001. Prevalence ofgastric metaplasia in the duodenal bulb is low in Helicobacter pylori positive non-ulcerdyspepsia patients. Dig. Liver Dis. 33:459-463.

Heikkinen, M., M. Vornanen, S. Hollmén, M. Färkkila. 2004. Prognostic significance ofantrum-predominant gastritis in functional dyspepsia. Scand. J. Gastroenterol. 39:227-231.

Held, M., L. Engstrand, L. E. Hansson, R. Bergström, T. Wadström, O. Nyrén. 2004. Is theassociation between Helicobacter pylori and gastric cancer confined to CagA-positivestrains? Helicobacter 9:271-277.

Herbrink, P. and L. J. van Doorn. 2000. Serological methods for diagnosis of Helicobacterpylori infection and monitoring of eradication therapy. Eur. J. Clin. Microbiol. Infect. Dis.19:164-173.

Hershko, C., A. Lahad, D. Kereth. 2005. Gastropathic sideropenia. Best Pract. Res. Clin.Haematol. 18:363-380.

Hirschl, A. M. and M. L. Rotter. 1996. Serological tests for monitoring Helicobacter pylorieradication treatment. J. Gastroenterol. 31:33-36.

Hoang, T. T., A. S. Rehnberg, T. U. Wheeldon, C. Bengtsson, D. C. Phung, R. Befrits, M.Sörberg, M. Granström. 2006. Comparison of the performance of serological kits forHelicobacter pylori infection with European and Asian study populations. Clin. Microbiol.Infect. 12:1112-1117.

Hojo, M., H. Miwa, T. Ohkusa, R. Ohkura, A. Kurosawa, N. Sato. 2002. Alteration ofhistological gastritis after cure of Helicobacter pylori infection. Aliment. Pharmacol. Ther.16:1923-1932.

Hopkins, R. J., L. S. Girardi, E. A. Turney. 1996. Relationship between Helicobacter pylorieradication and reduced duodenal and gastric ulcer recurrence: a review. Gastroenterology110:1244-1252.

Horvitz, G. and B. D. Gold. 2006. Gastroduodenal diseases of childhood. Curr. Opin.Gastroenterol. 22:632-640.

Houghton, J. and T. C. Wang. 2005. Helicobacter pylori and gastric cancer: a new paradigmfor inflammation-associated epithelial cancers. Gastroenterology 128:1567-1578.

78

Jass, J. R. and M. D. Walsh. 2001. Altered mucin expression in the gastrointestinal tract: areview. J. Cell. Mol. Med. 5:327-351.

Jeen, Y. T., S. W. Lee, S. I. Kwon, H. J. Chun, H. S. Lee, C. W. Song, S. H. Um, J. H. Choi,C. D. Kim, H. S. Ryu, J. H. Hyun. 2001. Differentiation between reinfection andrecrudescence of Helicobacter pylori strains using PCR-based restriction fragment lengthpolymorphism analysis. Yonsei Med. J. 42:41-45.

Johnson, H. D. 1957. The classification and principles of treatment of gastric ulcers. Lancet.14:518-520.

Jolley, C. D. and D. A. Wagner. 2007. Comparison of the 13C-urea blood test to histology andrapid urease testing in the diagnosis of Helicobacter pylori infection in children. J. Pediatr.Gastroenterol. Nutr. 44:68-70.

Jones N. L., P. Sherman, C. Fallone, N. Flook, F. Smaill, S. van Zanten, R. Hunt, A.Thomson. Canadian Helicobacter Study. 2005. Canadian Helicobacter Study GroupConsensus Conference: Update on the approach to Helicobacter pylori infection in childrenand adolescents--an evidence-based evaluation. Can. J. Gastroenterol. 19:399-408.

Jonkers, D., G. Houben, A. de Bruine, J. W. Arends, E. Stobberingh, R. Stockbrügger. 1998.Prevalence of gastric metaplasia in the duodenal bulb and distribution of Helicobacter pyloriin the gastric mucosa. A clinical and histopathological study in 96 consecutive patients. Ital. J.Gastroenterol. 30:481-483.

Jonkers, D., P. Houben, E. Stobberingh, R. Stockbrügger. 1997a. Ongoing gastric acidinhibition is a confounding factor in Helicobacter pylori diagnosis. Eur. J. Gastroenterol.Hepatol. 9:49-53.

Jonkers, D., E. Stobberingh, A. de Bruine, J. W. Arends, R. Stockbrügger. 1997b. Evaluationof immunohistochemistry for the detection of Helicobacter pylori in gastric mucosal biopsies.J. Infect. 35:149-154.

Josenhans, C., D. Beier, B. Linz, T. F. Meyer, S. Suerbaum. 2007. Pathogenomics ofHelicobacter. Int. J. Med. Microbiol. E-pub ahead of print.

Jönsson, K. A., M. Ström, G. Bodemar, K. Norrby. 1988. Histologic changes in thegastroduodenal mucosa after long-term medical treatment with cimetidine or parietal cellvagotomy in patients with juxtapyloric ulcer disease. Scand. J. Gastroenterol. 23:433-441.

Jüttner, S., M. Vieth, S. Miehlke, W. Schneider-Brachert, C. Kirsch, T. Pfeuffer, N. Lehn, M.Stolte. 2004. Reliable detection of macrolide-resistant Helicobacter pylori via fluorescence insitu hybridization in formalin-fixed tissue. Mod. Pathol. 17:684-689.

Kabir, S. 2003. Review article: clinic-based testing for Helicobacter pylori infection byenzyme immunoassay of faeces, urine and saliva. Aliment. Pharmacol. Ther. 17:1345-1354.

Kabir, S. 2004. Detection of Helicobacter pylori DNA in feces and saliva by polymerasechain reaction: a review. Helicobacter 9:115-123.

Kaneko, H., T. Konagaya, K. Kusugami. 2002. Helicobacter pylori and gut hormones. J.Gastroenterol. 37:77-86.

79

Karttunen, T., S. Niemelä, J. Lehtola. 1991. Helicobacter pylori in dyspeptic patients:quantitative association with severity of gastritis, intragastric pH, and serum gastrinconcentration. Scand. J. Gastroenterol. Suppl. 186:124-134.

Kassa, E., E. Tsega, W. Gebre. 1996. Comparison of diagnostic methods for detection ofHelicobacter pylori. East Afr. Med. J. 73:239-241.

Kato, S., N. Furuyama, K. Ozawa, K. Ohnuma, K. Iinuma. 1999. Long-term follow-up studyof serum immunoglobulin G and immunoglobulin A antibodies after Helicobacter pylorieradication. Pediatrics 104:e22.

Khulusi, S., S. Badve, P. Patel, R. Lloyd, J. M. Marrero, C. Finlayson, M. A. Mendall, T. C.Northfield. 1996. Pathogenesis of gastric metaplasia of the human duodenum: role ofHelicobacter pylori, gastric acid, and ulceration. Gastroenterology 110:452-458.

Khulusi, S., M. A. Mendall, S. Badve, P. Patel, C. Finlayson, T. C. Northfield. 1995. Effect ofHelicobacter pylori eradication on gastric metaplasia of the duodenum. Gut 36:193-197.

Kim, N., S. H. Lim, K. H. Lee, S. E. Choi. 1998. Long-term effect of Helicobacter pylorieradication on gastric metaplasia in patients with duodenal ulcer. J. Clin. Gastroenterol.27:246-252.

Kindermann, A., N. Konstantopoulos, N. Lehn, H. Demmelmair, S. Koletzko. 2001.Evaluation of two commercial enzyme immunoassays, testing immunoglobulin G (IgG) andIgA responses, for diagnosis of Helicobacter pylori infection in children. J. Clin. Microbiol.39:3591-3596.

Klaukka, T., A. Helin-Salmivaara, R. Huupponen, J. E. Idänpään-Heikkilä. 2005.Tulehduskipulääkkeet tarkastelussa. Suomen lääkärilehti - Finlands läkartidning 60:3266.

Kobayashi, T., S. Kikuchi, Y. Lin, K. Yagyu, Y. Obata, A. Ogihara, A. Hasegawa, K. Miki,E. Kaneko, H. Mizukoshi, T. Sakiyama, H. Tenjin. 2004. Trends in the incidence of gastriccancer in Japan and their associations with Helicobacter pylori infection and gastric mucosalatrophy. Gastric Cancer. 7:233-239.

Koch, H. K., B. Baumert, U. Koch, M. Oehlert, W. Oehlert. 1990. Prevalence ofCampylobacter pylori as demonstrated by histology or CLO-test in different types of gastritis.A study in 5 clinically predefined groups of patients. Pathol. Res. Pract. 186:154-158.

Kodama, M., Y. Kitadai, M. Ito, H. Kai, H. Masuda, S. Tanaka, M. Yoshihara, K. Fujimura,K. Chayama. 2007. Immune response to CagA protein is associated with improved plateletcount after Helicobacter pylori eradication in patients with idiopathic thrombocytopenicpurpura. Helicobacter 12:36-42.

Koivisto, T. T., H. I. Rautelin, M. E. Voutilainen, M. T. Heikkinen, J. P. Koskenpato, M. A.Färkkilä. 2005. First-line eradication therapy for Helicobacter pylori in primary health carebased on antibiotic resistance: results of three eradication regimens. Aliment. Pharmacol.Ther. 21:773-782.

Koivisto, T. T., H. I. Rautelin, M. E. Voutilainen, S. E. Niemelä, M. Heikkinen, P. I.Sipponen, M. A. Färkkilä. 2004. Primary Helicobacter pylori resistance to metronidazole andclarithromycin in the Finnish population. Aliment. Pharmacol. Ther. 19:1009-1017.

80

Kokkola, A., T. U. Kosunen, P. Puolakkainen, P. Sipponen, M. Härkönen, F. Laxén, J.Virtamo, R. Haapiainen, H. Rautelin. 2003. Spontaneous disappearance of Helicobacterpylori antibodies in patients with advanced atrophic corpus gastritis. APMIS 111:619-624.

Kokkola, A., H. Rautelin, P. Puolakkainen, P. Sipponen, M. Färkkilä, R. Haapiainen, T. U.Kosunen. 1998. Positive result by serology indicates active Helicobacter pylori infection inpatients with atrophic gastritis. J. Clin. Microbiol. 36:1808-1810.

Kokkola, A., H. Rautelin, P. Puolakkainen, P. Sipponen, M. Färkkilä, R. Haapiainen, T. U.Kosunen. 2000. Diagnosis of Helicobacter pylori infection in patients with atrophic gastritis:comparison of histology, 13C-urea breath test, and serology. Scand. J. Gastroenterol. 35:138-141.

Kokkola, A., P. Sipponen, H. Rautelin, M. Härkönen, T. U. Kosunen, R. Haapiainen, P.Puolakkainen. 2002 Mar. The effect of Helicobacter pylori eradication on the natural courseof atrophic gastritis with dysplasia. Aliment. Pharmacol. Ther. 16:515-520.

Koletzko, S. and A. Feydt-Schmidt. 2001. Infants differ from teenagers: use of non-invasivetests for detection of Helicobacter pylori infection in children. Eur. J. Gastroenterol. Hepatol.13:1047-1052.

Koletzko, S., N. Konstantopoulos, D. Bosman, A. Feydt-Schmidt, A. van der Ende, N.Kalach, J. Raymond, H. Rüssmann. 2003. Evaluation of a novel monoclonal enzymeimmunoassay for detection of Helicobacter pylori antigen in stool from children. Gut 52:804-806.

Kolho, K. L., T. Klemola, A. Koivusalo, H. Rautelin. 2006. Stool antigen tests for thedetection of Helicobacter pylori in children. Diagn. Microbiol. Infect. Dis. 55:269-273.

Kolho, K. L., J. Korhonen, M. Verkasalo, H. Lindahl, E. Savilahti, H. Rautelin. 2002.Helicobacter pylori serology at diagnosis and follow-up of biopsy-verified infection inchildren. Scand. J. Infect. Dis. 34:177-182.

Konno, M., N. Fujii, S. Yokota, K. Sato, M. Takahashi, K. Sato, E. Mino, T. Sugiyama. 2005.Five-year follow-up study of mother-to-child transmission of Helicobacter pylori infectiondetected by a random amplified polymorphic DNA fingerprinting method. J. Clin. Microbiol.43:2246-2250.

Koskenpato, J., M. Färkkilä, P. Sipponen. 2002. Helicobacter pylori and different topographictypes of gastritis: treatment response after successful eradication therapy in functionaldyspepsia. Scand. J. Gastroenterol. 37:778-784.

Kosunen, T. U., A. Aromaa, P. Knekt, A. Salomaa, H. Rautelin, P. Lohi, O. P. Heinonen.1997. Helicobacter antibodies in 1973 and 1994 in the adult population of Vammala, Finland.Epidemiol. Infect. 119:29-34.

Kosunen, T. U., J. Hook-Nikanne, A. Salomaa, S. Sarna, A. Aromaa, T. Haahtela. 2002.Increase of allergen-specific immunoglobulin E antibodies from 1973 to 1994 in a Finnishpopulation and a possible relationship to Helicobacter pylori infections. Clin. Exp. Allergy32:373-378.

Kosunen, T. U., K. Seppälä, S. Sarna, P. Sipponen. 1992. Diagnostic value of decreasing IgG,IgA, and IgM antibody titres after eradication of Helicobacter pylori. Lancet 339:893-895.

81

Kreuning, J., A. M. v.d. Wal, G. Kuiper, J. Lindeman. 1989. Chronic nonspecific duodenitis.A multiple biopsy study of the duodenal bulb in health and disease. Scand. J. Gastroenterol.Suppl. 167:16-20.

Krienitz, W. 1906. Ueber das Auftreten von Spirochäten verschiedener Form im Mageninhaltbei Carcinoma ventriculi. Dtsch. Med. Wochenschr. 28:872-889.

Kuipers, E. J. and P. Sipponen. 2006. Helicobacter pylori eradication for the prevention ofgastric cancer. Helicobacter 11:52-57.

Kuo, C. H., D. C. Wu, C. Y. Lu, Y. C. Su, F. J. Yu, Y. C. Lee, I. C. Wu, S. R. Lin, C. S. Liu,C. M. Jan, W. M. Wang. 2002. The media of rapid urease test influence the diagnosis ofHelicobacter pylori. Hepatogastroenterology 49:1191-1194.

Kusters, J. G., A. H. van Vliet, E. J. Kuipers. 2006. Pathogenesis of Helicobacter pyloriinfection. Clin. Microbiol. Rev. 19:449-490.

Lage, A. P., E. Godfroid, A. Fauconnier, A. Burette, J. P. Butzler, A. Bollen, Y. Glupczynski.1995. Diagnosis of Helicobacter pylori infection by PCR: comparison with other invasivetechniques and detection of cagA gene in gastric biopsy specimens. J. Clin. Microbiol.33:2752-2756.

Laheij, R. J., W. A. de Boer, J. B. Jansen, H. J. van Lier, P. M. Sneeberger, A. L. Verbeek.2000. Diagnostic performance of biopsy-based methods for determination of Helicobacterpylori infection without a reference standard. J. Clin. Epidemiol. 53:742-746.

Lahner, E., D. Vaira, N. Figura, E. Pilozzi, A. Pasquali, C. Severi, F. Perna, G. Delle Fave, B.Annibale. 2004. Role of noninvasive tests (C-urea breath test and stool antigen test) asadditional tools in diagnosis of Helicobacter pylori infection in patients with atrophic bodygastritis. Helicobacter 9:436-442.

Lai, L. H. and J. J. Sung. 2007. Helicobacter pylori and benign upper digestive disease.Baillieres Best Pract. Res. Clin. Gastroenterol. 21:261-279.

Laine, L., R. Estrada, M. Trujillo, K. Knigge, M. B. Fennerty. 1998a. Effect of proton-pumpinhibitor therapy on diagnostic testing for Helicobacter pylori. Ann. Intern. Med. 129:547-550.

Laine, L., D. N. Lewin, W. Naritoku, H. Cohen. 1997. Prospective comparison of H&E,Giemsa, and Genta stains for the diagnosis of Helicobacter pylori. Gastrointest. Endosc.45:463-467.

Laine, L., D. Lewin, W. Naritoku, R. Estrada, H. Cohen. 1996. Prospective comparison ofcommercially available rapid urease tests for the diagnosis of Helicobacter pylori.Gastrointest. Endosc. 44:523-526.

Laine, L. A., R. A. Nathwani, W. Naritoku. 2005. The effect of GI bleeding on Helicobacterpylori diagnostic testing: a prospective study at the time of bleeding and 1 month later.Gastrointest. Endosc. 62:853-859.

Laine, L., L. Suchower, E. Johnson, P. Ronca, G. Neil. 1998b. Accuracy of CLO test afterHelicobacter pylori therapy. Gastrointest. Endosc. 47:250-253.

Laine, L., J. Sugg, L. Suchower, G. Neil. 2000. Endoscopic biopsy requirements for post-treatment diagnosis of Helicobacter pylori. Gastrointest. Endosc. 51:664-669.

82

Laurén, P. 1965. The two histological main types of gastric carcinoma: diffuse and so-calledintestinal-type carcinoma. An attempt at a histo-clinical classification. Acta Pathol. Microbiol.Scand. 64:31-49.

Lauwers, G. Y. 2003. Defining the pathologic diagnosis of metaplasia, atrophy, dysplasia, andgastric adenocarcinoma. J. Clin. Gastroenterol. 36:S37-43.

Lee, J. M., N. P. Breslin, C. Fallon, C. A. O'Morain. 2000. Rapid urease tests lack sensitivityin Helicobacter pylori diagnosis when peptic ulcer disease presents with bleeding. Am. J.Gastroenterol. 95:1166-1170.

Leodolter, A., U. Peitz, M. P. Ebert, K. Agha-Amiri, P. Malfertheiner. 2002. Comparison oftwo enzyme immunoassays for the assessment of Helicobacter pylori status in stoolspecimens after eradication therapy. Am. J. Gastroenterol. 97:1682-1686.

Leodolter, A., D. Vaira, F. Bazzoli, K. Schütze, A. Hirschl, F. Mégraud, P. Malfertheiner.2003. European multicentre validation trial of two new non-invasive tests for the detection ofHelicobacter pylori antibodies: urine-based ELISA and rapid urine test. Aliment. Pharmacol.Ther. 18:927-931.

Lerang, F., J. B. Haug, B. Moum, P. Mowinckel, T. Berge, E. Ragnhildstveit, A. Bjørneklett.1998. Accuracy of IgG serology and other tests in confirming Helicobacter pylori eradication.Scand. J. Gastroenterol. 33:710-715.

Lerang, F., B. Moum, P. Mowinckel, J. B. Haug, E. Ragnhildstveit, T. Berge, A. Bjørneklett.1998. Accuracy of seven different tests for the diagnosis of Helicobacter pylori infection andthe impact of H2-receptor antagonists on test results. Scand. J. Gastroenterol. 33:364-369.

Leung, W. K., S. R. Lin, J. Y. Ching, K. F. To, E. K. Ng, F. K. Chan, J. Y. Lau, J. J. Sung.2004. Factors predicting progression of gastric intestinal metaplasia: results of a randomisedtrial on Helicobacter pylori eradication. Gut 53:1244-1249.

Li, H. C., C. Stoicov, A. B. Rogers, J. Houghton. 2006. Stem cells and cancer: evidence forbone marrow stem cells in epithelial cancers. World J. Gastroenterol. 12:363-371.

Lin, S. K., J. R. Lambert, M. Schembri, L. Nicholson, M. Finlay, C. Wong, A. Coulepis.1992. A comparison of diagnostic tests to determine Helicobacter pylori infection. J.Gastroenterol. Hepatol. 7:203-209.

Linz, B., F. Balloux, Y. Moodley, A. Manica, H. Liu, P. Roumagnac, D. Falush, C. Stamer, F.Prugnolle, S. W. van der Merwe, Y. Yamaoka, D. Y. Graham, E. Pérez-Trallero, T.Wadström, S. Suerbaum, M. Achtman. 2007. An African origin for the intimate associationbetween humans and Helicobacter pylori. Nature 445:915-918.

Lochhead, P. and E. M. El-Omar. 2007. Helicobacter pylori infection and gastric cancer.Baillieres Best Pract. Res. Clin. Gastroenterol. 21:281-297.

Lu, C. Y., F. C. Kuo, S. W. Wang, Y. C. Lo, I. C. Wu, L. L. Chang, F. J. Yu, Y. C. Su, W. M.Wang, C. M. Jan, D. C. Wu. 2006. The clinical applications and accuracy of 2 rapid near-patient tests in detecting Helicobacter pylori infection. Diagn. Microbiol. Infect. Dis. 56:241-246.

Macintyre, I. M. and A. Millar. 1991. Highly selective vagotomy--a safe operation forduodenal ulcer. Immediate and long-term complications and sequelae in 500 patients. Eur. J.Surg. 157:261-265.

83

Maconi, G., L. Vago, G. Galletta, V. Imbesi, O. Sangaletti, F. Parente, C. Cucino, S. Bonetto,G. Bianchi Porro. 1999. Is routine histological evaluation an accurate test for Helicobacterpylori infection? Aliment. Pharmacol. Ther. 13:327-331.

Makristathis, A., W. Barousch, E. Pasching, C. Binder, C. Kuderna, P. Apfalter, M. L. Rotter,A. M. Hirschl. 2000. Two enzyme immunoassays and PCR for detection of Helicobacterpylori in stool specimens from pediatric patients before and after eradication therapy. J. Clin.Microbiol. 38:3710-3714.

Malaty, H. M. 2007. Epidemiology of Helicobacter pylori infection. Baillieres Best Pract.Res. Clin. Gastroenterol. 21:205-214.

Malfertheiner, P., J. Enrique Domínguez-Muñoz, H. Heckenmüller, M. Neubrand, H. P.Fischer, T. Sauerbruch. 1996. Modified rapid urease test for detection of Helicobacter pyloriinfection. Eur. J. Gastroenterol. Hepatol. 8:53-56.

Malfertheiner, P., L. C. Fry, K. Mönkemüller. 2006. Can gastric cancer be prevented byHelicobacter pylori eradication? Baillieres Best Pract. Res. Clin. Gastroenterol. 20:709-719.

Malfertheiner, P., F. Mégraud, C. O'Morain, F. Bazzoli, E. El-Omar, D. Graham, R. Hunt, T.Rokkas, N. Vakil, E. J. Kuipers. 2007. Current concepts in the management of Helicobacterpylori infection - The Maastricht III Consensus Report. Gut 56: 772-781.

Malfertheiner P., F. Mégraud, C. O'Morain, A. Hungin, R. Jones, A. Axon, D.Y. Graham, G.Tytgat. European Helicobacter Pylori Study Group (EHPSG). 2002. Current concepts in themanagement of Helicobacter pylori infection--the Maastricht 2-2000 Consensus Report.Aliment. Pharmacol. Ther. 16:167-180.

Malfertheiner, P., P. Sipponen, M. Naumann, P. Moayyedi, F. Mégraud, S. D. Xiao, K.Sugano, O. Nyrén. Lejondal H. pylori-Gastric Cancer Task Force. 2005 Sep. Helicobacterpylori eradication has the potential to prevent gastric cancer: a state-of-the-art critique. Am. J.Gastroenterol. 100:2100-2115.

Mana, F., W. Van Laere, A. Bossuyt, D. Urbain. 2005. The early effect of proton pumpinhibitor therapy on the accuracy of the 13C-urea breath test. Dig. Liver Dis. 37:28-32.

Manes, G., A. Balzano, G. Iaquinto, C. Ricci, M. M. Piccirillo, N. Giardullo, A. Todisco, M.Lioniello, D. Vaira. 2001. Accuracy of the stool antigen test in the diagnosis of Helicobacterpylori infection before treatment and in patients on omeprazole therapy. Aliment. Pharmacol.Ther. 15:73-79.

Marais, A., L. Monteiro, A. Occhialini, M. Pina, H. Lamouliatte, F. Mégraud. 1999. Directdetection of Helicobacter pylori resistance to macrolides by a polymerase chainreaction/DNA enzyme immunoassay in gastric biopsy specimens. Gut 44:463-467.

Marshall, B. J. and J. R. Warren. 1984. Unidentified curved bacilli in the stomach of patientswith gastritis and peptic ulceration. Lancet 1:1311-1315.

Marzio, L., D. Angelucci, L. Grossi, M. G. Diodoro, E. Di Campli, L. Cellini. 1998. Anti-Helicobacter pylori specific antibody immunohistochemistry improves the diagnosticaccuracy of Helicobacter pylori in biopsy specimen from patients treated with triple therapy.Am. J. Gastroenterol. 93:223-226.

84

Matsukura, N., T. Tajiri, S. Kato, A. Togashi, G. Masuda, A. Tokunaga, N. Yamada. 2004.Diagnostic value of culture, histology and PCR for Helicobacter pylori in the remnantstomach after surgery. Aliment. Pharmacol. Ther. 20:33-38.

McColl, K. E. 1996. Helicobacter pylori, gastric acid, and duodenal gastric metaplasia. Gut39:615-616.

McColl, K. E. 2005. Screening for early gastric cancer. Gut 54:740-742.

McColl, K. E. 2006. Cancer of the gastric cardia. Baillieres Best Pract. Res. Clin.Gastroenterol. 20:687-696.

Mégraud, F. and European Paediatric Task Force on Helicobacter pylori. 2005. Comparisonof non-invasive tests to detect Helicobacter pylori infection in children and adolescents:results of a multicenter European study. J. Pediatr. 146:198-203.

Mégraud, F. and P. Lehours. 2007. Helicobacter pylori Detection and AntimicrobialSusceptibility Testing. Clin. Microbiol. Rev. 20:280-322.

Meijer, B. C., J. C. Thijs, J. H. Kleibeuker, A. A. van Zwet, R. J. Berrelkamp. 1997.Evaluation of eight enzyme immunoassays for detection of immunoglobulin G againstHelicobacter pylori. J. Clin. Microbiol. 35:292-294.

Meining, A., A. Morgner, S. Miehlke, E. Bayerdörffer, M. Stolte. 2001. Atrophy-metaplasia-dysplasia-carcinoma sequence in the stomach: a reality or merely an hypothesis? BaillieresBest Pract. Res. Clin. Gastroenterol. 15:983-998.

Mera, R., E. T. Fontham, L. E. Bravo, J. C. Bravo, M. B. Piazuelo, M. C. Camargo, P. Correa.2005. Long term follow up of patients treated for Helicobacter pylori infection. Gut 54:1536-1540.

Metz, D. C., E. E. Furth, D. O. Faigel, J. A. Kroser, A. Alavi, D. M. Barrett, K. Montone.1998. Realities of diagnosing Helicobacter pylori infection in clinical practice: a case for non-invasive indirect methodologies. Yale J. Biol. Med. 71:81-90.

Moayyedi, P., S. Soo, J. Deeks, B. Delaney, A. Harris, M. Innes, R. Oakes, S. Wilson, A.Roalfe, C. Bennett, D. Forman. 2006. Eradication of Helicobacter pylori for non-ulcerdyspepsia.Update of Cochrane Database Syst Rev.

Monteiro, L., A. de Mascarel, A. M. Sarrasqueta, B. Bergey, C. Barberis, P. Talby, D. Roux,L. Shouler, D. Goldfain, H. Lamouliatte, F. Mégraud. 2001. Diagnosis of Helicobacter pyloriinfection: noninvasive methods compared to invasive methods and evaluation of two newtests. Am. J. Gastroenterol. 96:353-358.

Morris, J. M., A. L. Reasonover, M. G. Bruce, D. L. Bruden, B. J. McMahon, F. D. Sacco, D.E. Berg, A. J. Parkinson. 2005. Evaluation of seaFAST, a rapid fluorescent in situhybridization test, for detection of Helicobacter pylori and resistance to clarithromycin inparaffin-embedded biopsy sections. J. Clin. Microbiol. 43:3494-3496.

Murakami, K., R. Sato, T. Okimoto, K. Watanabe, M. Nasu, T. Fujioka, M. Kodama, J.Kagawa. 2003. Influence of anti-ulcer drugs used in Japan on the result of 13C-urea breath testfor the diagnosis of Helicobacter pylori infection. J. Gastroenterol. 38:937-941.

Myllyluoma, E., L. Veijola, T. Ahlroos, S. Tynkkynen, E. Kankuri, H. Vapaatalo, H.Rautelin, R. Korpela. 2005. Probiotic supplementation improves tolerance to Helicobacter

85

pylori eradication therapy--a placebo-controlled, double-blind randomized pilot study.Aliment. Pharmacol. Ther. 21:1263-1272.

Mäkinen, J., Niemelä, S., Kerola,T., Lehtola, J., Karttunen, T. 2003. Epithelial cellproliferation and glandular atrophy in lymphocytic gastritis: effect of H. pylori treatment.World J. Gastroenterol. 9:2706-2710.

Nakagawa, S., M. Asaka, M. Kato, T. Nakamura, C. Kato, T. Fujioka, M. Tatsuta, K. Keida,S. Terao, S. Takahashi, N. Uemura, T. Kato, N. Aoyama, D. Saito, M. Suzuki, A. Imamura,K. Sato, H. Miwa, H. Nomura, M. Kaise, S. Oohara, T. Kawai, K. Urabe, N. Sakaki, S. Ito, Y.Noda, A. Yanaka, K. Kusugami, H. Goto, T. Furuta, M. Fujino, F. Kinjyou, T. Ookusa. 2006.Helicobacter pylori eradication and metachronous gastric cancer after endoscopic mucosalresection of early gastric cancer. Aliment. Pharmacol. Ther. 24:214-218.

Ndip, R. N., W. G. MacKay, M. J. Farthing, L. T. Weaver. 2003. Culturing Helicobacterpylori from clinical specimens: review of microbiologic methods. J. Pediatr. Gastroenterol.Nutr. 36:616-622.

Nilsson, I., A. Ljungh, P. Aleljung, T. Wadström. 1997. Immunoblot assay for serodiagnosisof Helicobacter pylori infections. J. Clin. Microbiol. 35:427-432.

Noach, L. A., T. M. Rolf, N. B. Bosma, M. P. Schwartz, J. Oosting, E. A. Rauws, G. N.Tytgat. 1993. Gastric metaplasia and Helicobacter pylori infection. Gut 34:1510-1514.

O'Connor H. J., B. M. Broderick, K. P. Cullen, B. Flynn, J. Jayaram. 2002. Evaluation of theHelicobacter pylori stool antigen (HpSA) test in routine clinical practice--is it patient-friendly? Ir. Med. J. 95:305-306.

Odenbreit, S., J. Püls, B. Sedlmaier, E. Gerland, W. Fischer, R. Haas. 2000. Translocation ofHelicobacter pylori CagA into gastric epithelial cells by type IV secretion. Science 287:1497-1500.

Oh, J. D., H. Kling-Bäckhed, M. Giannakis, L. G. Engstrand, J. I. Gordon. 2006a. Interactionsbetween gastric epithelial stem cells and Helicobacter pylori in the setting of chronic atrophicgastritis. Curr. Opin. Microbiol. 9:21-27.

Oh, J. D., H. Kling-Bäckhed, M. Giannakis, J. Xu, R. S. Fulton, L. A. Fulton, H. S. Cordum,C. Wang, G. Elliott, J. Edwards, E. R. Mardis, L. G. Engstrand, J. I. Gordon. 2006b. Thecomplete genome sequence of a chronic atrophic gastritis Helicobacter pylori strain:evolution during disease progression. Proc. Natl. Acad. Sci. U. S. A. 103:9999-10004.

Ohara, S., M. Kato, M. Saito, S. Fukuda, C. Kato, S. Hamada, R. Nagashima, K. Obara, M.Suzuki, H. Honda, M. Asaka, T. Toyota. 2004. Comparison between a new 13C-urea breathtest, using a film-coated tablet, and the conventional 13C-urea breath test for the detection ofHelicobacter pylori infection. J. Gastroenterol. 39:621-628.

Oksanen, A., P. Sipponen, R. Karttunen, A. Miettinen, L. Veijola, S. Sarna, H. Rautelin.2000. Atrophic gastritis and Helicobacter pylori infection in outpatients referred forgastroscopy. Gut 46:460-463.

Oksanen, A., L. Veijola, P. Sipponen, K. O. Schauman, H. Rautelin. 1998 Apr. Evaluation ofPyloriset Screen, a rapid whole-blood diagnostic test for Helicobacter pylori infection. J.Clin. Microbiol. 36:955-957.

86

Oleastro, M., A. Ménard, A. Santos, H. Lamouliatte, L. Monteiro, P. Barthélémy, F.Mégraud. 2003. Real-time PCR assay for rapid and accurate detection of point mutationsconferring resistance to clarithromycin in Helicobacter pylori. J. Clin. Microbiol. 41:397-402.

Pacheco, N., V. Mago, I. Gómez, P. Gueneau, M. Guelrud, N. Reyes, L. R. Pericchi, M. G.Domínguez-Bello. 2001. Comparison of PCR and common clinical tests for the diagnosis ofH. pylori in dyspeptic patients. Diagn. Microbiol. Infect. Dis. 39:207-210.

Paimela, H., L. Paimela, R. Myllykangas-Luosujärvi, E. Kivilaakso. 2002. Current features ofpeptic ulcer disease in Finland: incidence of surgery, hospital admissions and mortality for thedisease during the past twenty-five years. Scand. J. Gastroenterol. 37:399-403.

Palli, D., G. Masala, G. Del Giudice, M. Plebani, D. Basso, D. Berti, M. E Numans, M.Ceroti, P. H. Peeters, H. B. de Mesquita, F. L. Buchner, F. Clavel-Chapelon, M. C. Boutron-Ruault, V. Krogh, C. Saieva, P. Vineis, S. Panico, R. Tumino, O. Nyrén, H. Siman, G.Berglund, G. Hallmans, M. J. Sanchez, N. Larranaga, A. Barricarte, C. Navarro, J. R. Quiros,T. Key, N. Allen, S. Bingham, K. T. Khaw, H. Boeing, C. Weikert, J. Linseisen, G. Nagel, K.Overvad, R. W. Thomsen, A. Tjonneland, A. Olsen, A. Trichoupoulou, D. Trichopoulos, A.Arvaniti, G. Pera, R. Kaaks, M. Jenab, P. Ferrari, G. Nesi, F. Carneiro, E. Riboli, C. A.Gonzalez. 2007. CagA+ Helicobacter pylori infection and gastric cancer risk in the EPIC-EURGAST study. Int. J. Cancer 120:859-867.

Pan, S., G. S. Lien, C. H. Liao, S. H. Chen. 1996. Gastric metaplasia of regenerating duodenalmucosa and deformity of duodenal bulb: a correlative study. J. Gastroenterol. Hepatol.11:108-112.

Parente, F., G. Maconi, G. Bianchi Porro, M. Caselli. 2002. Stool test with polyclonalantibodies for monitoring Helicobacter pylori eradication in adults: a critical reappraisal.Scand. J. Gastroenterol. 37:747-749.

Parsonnet, J., H. Shmuely, T. Haggerty. 1999. Fecal and oral shedding of Helicobacter pylorifrom healthy infected adults. JAMA 282:2240-2245.

Patrick, W. J., D. Denham, A. P. Forrest. 1974. Mucous change in the human duodenum: alight and electron microscopic study and correlation with disease and gastric acid secretion.Gut 15:767-776.

Peek, R. M.,Jr, G. G. Miller, K. T. Tham, G. I. Pérez-Pérez, T. L. Cover, J. C. Atherton, G. D.Dunn, M. J. Blaser. 1995. Detection of Helicobacter pylori gene expression in human gastricmucosa. J. Clin. Microbiol. 33:28-32.

Peitz, U., A. Leodolter, S. Kahl, K. Agha-Amiri, T. Wex, K. Wolle, T. Günther, B.Steinbrink, P. Malfertheiner. 2003. Antigen stool test for assessment of Helicobacter pyloriinfection in patients with upper gastrointestinal bleeding. Aliment. Pharmacol. Ther. 17:1075-1084.

Peitz, U., A. Leodolter, T. Wex, D. Schütze, K. Wolle, T. Welte, T. Günther, U. Schmidt, P.Malfertheiner. 2004. Diagnostics of Helicobacter pylori infection in patients with peptic ulcerbleeding. Z. Gastroenterol. 42:141-146.

Pérez-Pérez, G. I. 1997. Role of Helicobacter pylori infection in the development ofpernicious anemia. Clin. Infect. Dis. 25:1020-1022.

Pérez-Pérez, G. I. 2000. Accurate diagnosis of Helicobacter pylori. Culture, includingtransport. Gastroenterol. Clin. North Am. 29:879-884.

87

Pietroiusti, A., I. Luzzi, M. J. Gomez, A. Magrini, A. Bergamaschi, A. Forlini, A. Galante.2005. Helicobacter pylori duodenal colonization is a strong risk factor for the development ofduodenal ulcer. Aliment. Pharmacol. Ther. 21:909-915.

Pikkarainen, P., A. L. Karvonen, I. Kunnamo. 2002. Endoskopistin käsikirja, p. 42-43,Kustannus Oy Duodecim ja Suomen Gastroenterologiayhdistys ry., Helsinki.

Pilotto, A. and N. Salles. 2002. Helicobacter pylori infection in geriatrics. Helicobacter 7:56-62.

Rautelin, H., K. Seppälä, H. Nuutinen, P. Kärkkäinen, P. Sipponen, T. U. Kosunen. 1997.Culture of Helicobacter pylori from gastric biopsies transported in biopsy urease test tubes.Eur. J. Clin. Microbiol. Infect. Dis. 16:380-383.

Rehnberg-Laiho, L., J. Louhija, H. Rautelin, J. Jusufovic, R. Tilvis, A. Miettinen, T. U.Kosunen. 1999. Helicobacter antibodies in Finnish centenarians. J. Gerontol. A Biol. Sci.Med. Sci. 54:M400-3.

Rehnberg-Laiho, L., H. Rautelin, P. Koskela, S. Sarna, E. Pukkala, A. Aromaa, P. Knekt, T.U. Kosunen. 2001. Decreasing prevalence of helicobacter antibodies in Finland, withreference to the decreasing incidence of gastric cancer. Epidemiol. Infect. 126:37-42.

Rehnberg-Laiho, L., H. Rautelin, M. Valle, T. U. Kosunen. 1998. Persisting Helicobacterantibodies in Finnish children and adolescents between two and twenty years of age. Pediatr.Infect. Dis. J. 17:796-799.

Ricci, C., J. Holton, D. Vaira. 2007. Diagnosis of Helicobacter pylori: Invasive and non-invasive tests. Baillieres Best Pract. Res. Clin. Gastroenterol. 21:299-313.

Rimbara, E., N. Noguchi, T. Yamaguchi, K. Narui, T. Kawai, M. Sasatsu. 2005. Developmentof a highly sensitive method for detection of clarithromycin-resistant Helicobacter pylorifrom human feces. Curr. Microbiol. 51:1-5.

Robinson, K., R. H. Argent, J. C. Atherton. 2007. The inflammatory and immune response toHelicobacter pylori infection. Baillieres Best Pract. Res. Clin. Gastroenterol. 21:237-259.

Rollán, A., R. Giancaspero, M. Arrese, C. Figueroa, V. Vollrath, M. Schultz, I. Duarte, P.Vial. 1997. Accuracy of invasive and noninvasive tests to diagnose Helicobacter pyloriinfection after antibiotic treatment. Am. J. Gastroenterol. 92:1268-1274.

Rowland, M., L. Daly, M. Vaughan, A. Higgins, B. Bourke, B. Drumm. 2006. Age-specificincidence of Helicobacter pylori. Gastroenterology 130:65-72.

Rudi, J., C. Kolb, M. Maiwald, I. Zuna, A. von Herbay, P. R. Galle, W. Stremmel. 1997.Serum antibodies against Helicobacter pylori proteins VacA and CagA are associated withincreased risk for gastric adenocarcinoma. Dig. Dis. Sci. 42:1652-1659.

Rugge, M., P. Correa, M. F. Dixon, R. Fiocca, T. Hattori, J. Lechago, G. Leandro, A. B.Price, P. Sipponen, E. Solcia, H. Watanabe, R. M. Genta. 2002. Gastric mucosal atrophy:interobserver consistency using new criteria for classification and grading. Aliment.Pharmacol. Ther. 16:1249-1259.

Rüssmann, H., K. Adler, R. Haas, B. Gebert, S. Koletzko, J. Heesemann. 2001a. Rapid andaccurate determination of genotypic clarithromycin resistance in cultured Helicobacter pyloriby fluorescent in situ hybridization. J. Clin. Microbiol. 39:4142-4144.

88

Rüssmann, H., A. Feydt-Schmidt, K. Adler, D. Aust, A. Fischer, S. Koletzko. 2003. Detectionof Helicobacter pylori in paraffin-embedded and in shock-frozen gastric biopsy samples byfluorescent in situ hybridization. J. Clin. Microbiol. 41:813-815.

Rüssmann, H., V. A. Kempf, S. Koletzko, J. Heesemann, I. B. Autenrieth. 2001b.Comparison of fluorescent in situ hybridization and conventional culturing for detection ofHelicobacter pylori in gastric biopsy specimens. J. Clin. Microbiol. 39:304-308.

Sadowski, D., H. Cohen, L. Laine, P. Greenberg, J. Goldstein, M. Mihalov, A. F. Cutler.1998. Evaluation of the FlexSure HP whole blood antibody test for diagnosis of Helicobacterpylori infection. Am. J. Gastroenterol. 93:2119-2123.

Saito, K., K. Arai, M. Mori, R. Kobayashi, I. Ohki. 2000. Effect of Helicobacter pylorieradication on malignant transformation of gastric adenoma. Gastrointest. Endosc. 52:27-32.

Salles-Montaudon, N., S. Dertheil, N. Broutet, N. Gras, L. Monteiro, A. De Mascarel, F.Mégraud, J. P. Emeriau. 2002. Detecting Helicobacter pylori infection in hospitalized frailolder patients: the challenge. J. Am. Geriatr. Soc. 50:1674-1680.

Salomaa-Räsänen, A., T. U. Kosunen, J. Mattila, S. Sarna, H. Rautelin. 2004. Age-dependentaccuracy of Helicobacter pylori antibody assays for adults, with special emphasis on atrophicgastritis. Clin. Diagn. Lab. Immunol. 11:1185-1188.

Samloff, I. M. and R. T. Taggart. 1987. Pepsinogens, pepsins, and peptic ulcer. Clin. Invest.Med. 10:215-221.

Sande, N., M. Nikulin, I. Nilsson, T. Wadström, F. Laxén, M. Härkönen, O. Suovaniemi, P.Sipponen. 2001. Increased risk of developing atrophic gastritis in patients infected withCagA+ Helicobacter pylori. Scand. J. Gastroenterol. 36:928-933.

Savarino, V., G. S. Mela, P. Zentilin, G. Lapertosa, P. Ceppa, S. Vigneri, M. R. Mele, C.Mansi, D. Tracci, G. Bisso, G. Celle. 1997. 24-hour gastric pH and extent of duodenal gastricmetaplasia in Helicobacter pylori-positive patients. Gastroenterology 113:741-745.

Savarino, V., G. S. Mela, P. Zentilin, C. Mansi, S. Vigneri, F. Di Mario, A. Malesci, A.Calabrò, P. Sossai, G. Celle. 1995. Circadian acidity pattern in gastric ulcers at different sites.Am. J. Gastroenterol. 90:254-258.

Savarino, V., G. S. Mela, P. Zentilin, M. R. Mele, G. Bisso, M. Pivari, C. Mansi, L. Tessieri,G. Lapertosa, P. Ceppa, S. Vigneri. 2000. Effect of Helicobacter pylori eradication on 24-hour gastric pH and duodenal gastric metaplasia. Dig. Dis. Sci. 45:1315-1321.

Schilling, D., M. G. Ott, M. Nilius, G. Köhler, A. Zober, P. Messerer, J. F. Riemann. 2000.Specific Helicobacter pylori antigens unable to distinguish nonucler dyspepsia or peptic ulcercases from asymptomatic seropositive controls: a nested case-control study in employees of alarge company. Dig. Dis. Sci. 45:2444-2450.

Schwarz, E., G. Plum, G. Mauff, H. Hasbach, S. Eidt, M. Schrappe, W. Kruis. 1997.Molecular biology in diagnosis and epidemiology of Helicobacter pylori: PCR for thedetection and AP-PCR for characterization of patient isolates. Zentralbl. Bakteriol. 285:368-378.

Segni, M., O. Borrelli, I. Pucarelli, G. Delle Fave, A. M. Pasquino, B. Annibale. 2004. Earlymanifestations of gastric autoimmunity in patients with juvenile autoimmune thyroid diseases.J. Clin. Endocrinol. Metab. 89:4944-4948.

89

Sen, N., O. Yilmaz, I. im ek, A. A. Küpelio lu, H. Ellidokuz. 2005. Detection ofHelicobacter pylori DNA by a simple stool PCR method in adult dyspeptic patients.Helicobacter 10:353-359.

Sgouros, S. N. and C. Bergele. 2006. Clinical outcome of patients with Helicobacter pyloriinfection: the bug, the host, or the environment? Postgrad. Med. J. 82:338-342.

Sheu, B. S., H. B. Yang, Y. L. Wang, A. W. Kao, C. H. Chuang, P. W. Lin, Y. C. Chang.2002. Stool antigen assay to screen H. pylori infection and to assess the success of 3-Day and7-Day eradication therapy in the patients with partial gastrectomy. Helicobacter 7:199-204.

Sipponen, P. and D. Y. Graham. 2007. Importance of atrophic gastritis in diagnostics andprevention of gastric cancer: application of plasma biomarkers. Scand. J. Gastroenterol. 42:2-10.

Sipponen, P., T. U. Kosunen, J. Valle, M. Riihelä, K. Seppälä. 1992. Helicobacter pyloriinfection and chronic gastritis in gastric cancer. J. Clin. Pathol. 45:319-323.

Sipponen, P., P. Ranta, T. Helske, I. Kääriäinen, T. Mäki, A. Linnala, O. Suovaniemi, A.Alanko, M. Härkönen. 2002. Serum levels of amidated gastrin-17 and pepsinogen I inatrophic gastritis: an observational case-control study. Scand. J. Gastroenterol. 37:785-791.

Sipponen, P., M. Riihelä, H. Hyvärinen, K. Seppälä. 1994. Chronic nonatropic ('superficial')gastritis increases the risk of gastric carcinoma. A case-control study. Scand. J. Gastroenterol.29:336-340.

Sipponen, P., K. Varis, O. Fräki, U. M. Korri, K. Seppälä, M. Siurala. 1990. Cumulative 10-year risk of symptomatic duodenal and gastric ulcer in patients with or without chronicgastritis. A clinical follow-up study of 454 outpatients. Scand. J. Gastroenterol. 25:966-973.

Smith, V. C. and R. M. Genta. 2000. Role of Helicobacter pylori gastritis in gastric atrophy,intestinal metaplasia, and gastric neoplasia. Microsc. Res. Tech. 48:313-320.

Sobala, G. M., J. E. Crabtree, D. F. Dixon, C, J. Schorah, J. D. Taylor, B. J. Rathbone, R. V.Heatley, A. T. R. Axon. 1991. Acute Helicobacter pylori infection: clinical features, local andsystematic immune response, gastric mucosal histology, and gastric juice ascorbic acidconcentrations. Gut 32:1415-1418.

Sokic-Milutinovic, A., T. Wex, V. Todorovic, T. Milosavljevic, P. Malfertheiner. 2004. Anti-CagA and anti-VacA antibodies in Helicobacter pylori-infected patients with and withoutpeptic ulcer disease in Serbia and Montenegro. Scand. J. Gastroenterol. 39:222-226.

Steer, H. W. 1984. Surface morphology of the gastroduodenal mucosa in duodenal ulceration.Gut 25:1203-1210.

Storskrubb, T., P. Aro, J. Ronkainen, M. Vieth, M. Stolte, K. Wreiber, L. Engstrand, H.Nyhlin, E. Bolling-Sternevald, N. J. Talley, L. Agréus. 2005. A negative Helicobacter pyloriserology test is more reliable for exclusion of premalignant gastric conditions than a negativetest for current H. pylori infection: a report on histology and H. pylori detection in the generaladult population. Scand. J. Gastroenterol. 40:302-311.

Suerbaum, S. and P. Michetti. 2002. Helicobacter pylori infection. N. Engl. J. Med.347:1175-1186.

90

Take, S., M. Mizuno, K. Ishiki, Y. Nagahara, T. Yoshida, K. Yokota, K. Oguma, H. Okada,Y. Shiratori. 2005. The effect of eradicating Helicobacter pylori on the development ofgastric cancer in patients with peptic ulcer disease. Am. J. Gastroenterol. 100:1037-1042.

Talley, N. J., J. R. Lambert, S. Howell, H. H. Xia, S. K. Lin, L. Agréus. 1998. An evaluationof whole blood testing for Helicobacter pylori in general practice. Aliment. Pharmacol. Ther.12:641-645.

Talley, N. J., V. Stanghellini, R. C. Heading, K. L. Koch, J. R. Malagelada, G. N. Tytgat.1999. Functional gastroduodenal disorders. Gut 45:37-42.

Tatematsu, M., T. Tsukamoto, K. Inada. 2003. Stem cells and gastric cancer: role of gastricand intestinal mixed intestinal metaplasia. Cancer. Sci. 94:135-141.

Tepeš, B., V. Ferlan-Marolt, A. Juteršek, B. Kav , L. Zaletel-Kragelj. 1999a. Interobserveragreement in the assessment of gastritis reversibility after Helicobacter pylori eradication.Histopathology 34:124-133.

Tepeš, B., B. Kav , L. K. Zaletel, M. Gubina, A. Ihan, M. Poljak, I. Križman. 1999b. Two-to four-year histological follow-up of gastric mucosa after Helicobacter pylori eradication. J.Pathol. 188:24-29.

Testoni, P. A., U. Bonassi, F. Bagnolo, E. Colombo, R. Scelsi. 2002. In diffuse atrophicgastritis, routine histology underestimates Helicobacter pylori infection. J. Clin.Gastroenterol. 35:234-239.

Thijs, J. C., A. A. van Zwet, W. J. Thijs, H. B. Oey, A. Karrenbeld, F. Stellaard, D. S. Luijt,B. C. Meyer, J. H. Kleibeuker. 1996. Diagnostic tests for Helicobacter pylori: a prospectiveevaluation of their accuracy, without selecting a single test as the gold standard. Am. J.Gastroenterol. 91:2125-2129.

Thomas, J. E., G. R. Gibson, M. K. Darboe, A. Dale, L. T. Weaver. 1992. Isolation ofHelicobacter pylori from human faeces. Lancet 340:1194-1195.

Tiwari, S. K., A. A. Khan, K. S. Ahmed, I. Ahmed, F. Kauser, M. A. Hussain, S. M. Ali, A.Alvi, A. Habeeb, Z. Abid, N. Ahmed, C. M. Habibullah. 2005. Rapid diagnosis ofHelicobacter pylori infection in dyspeptic patients using salivary secretion: a non-invasiveapproach. Singapore Med. J. 46:224-228.

Tomb, J. F., O. White, A. R. Kerlavage, R. A. Clayton, G. G. Sutton, R. D. Fleischmann, K.A. Ketchum, H. P. Klenk, S. Gill, B. A. Dougherty, K. Nelson, J. Quackenbush, L. Zhou, E.F. Kirkness, S. Peterson, B. Loftus, D. Richardson, R. Dodson, H. G. Khalak, A. Glodek, K.McKenney, L. M. Fitzegerald, N. Lee, M. D. Adams, E. K. Hickey, D. E. Berg, J. D.Gocayne, T. R. Utterback, J. D. Peterson, J. M. Kelley, M. D. Cotton, J. M. Weidman, C.Fujii, C. Bowman, L. Watthey, E. Wallin, W. S. Hayes, M. Borodovsky, P. D. Karp, H. O.Smith, C. M. Fraser, J. C. Venter. 1997. The complete genome sequence of the gastricpathogen Helicobacter pylori. Nature 388:539-547.

Trebesius, K., K. Panthel, S. Strobel, K. Vogt, G. Faller, T. Kirchner, M. Kist, J. Heesemann,R. Haas. 2000. Rapid and specific detection of Helicobacter pylori macrolide resistance ingastric tissue by fluorescent in situ hybridization. Gut 46:608-614.

Tseng, C. A., W. M. Wang, D. C. Wu. 2005. Comparison of the clinical feasibility of threerapid urease tests in the diagnosis of Helicobacter pylori infection. Dig. Dis. Sci. 50:449-452.

91

Tu, T. C., C. L. Lee, C. H. Wu, T. K. Chen, C. C. Chan, S. H. Huang, S. C. Lee MS. 1999.Comparison of invasive and noninvasive tests for detecting Helicobacter pylori infection inbleeding peptic ulcers. Gastrointest. Endosc. 49:302-306.

Udd, M., P. Miettinen, A. Palmu, R. Julkunen. 2003. Effect of short-term treatment withregular or high doses of omeprazole on the detection of Helicobacter pylori in bleeding pepticulcer patients. Scand. J. Gastroenterol. 38:588-593.

Uemura, N., S. Okamoto, S. Yamamoto, N. Matsumura, S. Yamaguchi, M. Yamakido, K.Taniyama, N. Sasaki, R. J. Schlemper. 2001. Helicobacter pylori infection and thedevelopment of gastric cancer. N. Engl. J. Med. 345:784-789.

Vaira, D., J. Holton, M. Menegatti, F. Landi, C. Ricci, A. Ali, L. Gatta, S. Farinelli, C.Acciardi, B. Massardi, M. Miglioli. 1998. Blood tests in the management of Helicobacterpylori infection. Italian Helicobacter pylori Study Group. Gut 43:S39-46.

Vaira, D., C. Ricci, M. Menegatti, L. Gatta, S. Berardi, A. Tampieri, M. Miglioli. 2001a.Stool test for Helicobacter pylori. Am. J. Gastroenterol. 96:1935-1938.

Vaira, D. and N. Vakil. 2001b. Blood, urine, stool, breath, money, and Helicobacter pylori.Gut 48:287-289.

Vakil, N., D. Rhew, A. Soll, J. J. Ofman. 2000. The cost-effectiveness of diagnostic testingstrategies for Helicobacter pylori. Am. J. Gastroenterol. 95:1691-1698.

Valle, J., M. Kekki, P. Sipponen, T. Ihamäki, M. Siurala. 1996. Long-term course andconsequences of Helicobacter pylori gastritis. Results of a 32-year follow-up study. Scand. J.Gastroenterol. 31:546-550.

van Doorn, L. J., Y. Glupczynski, J. G. Kusters, F. Mégraud, P. Midolo, N. Maggi-Solcà, D.M. Queiroz, N. Nouhan, E. Stet, W. G. Quint. 2001. Accurate prediction of macrolideresistance in Helicobacter pylori by a PCR line probe assay for detection of mutations in the23S rRNA gene: multicenter validation study. Antimicrob. Agents Chemother. 45:1500-1504.

van IJzendoorn, M. C., R. J. Laheij, W. A. de Boer, J. B. Jansen. 2005. The importance ofcorpus biopsies for the determination of Helicobacter pylori infection. Neth. J. Med. 63:141-145.

van Keeken, N., E. van Hattum, W. A. de Boer. 2006. Validation of a new, commerciallyavailable dry rapid urease test for the diagnosis of Helicobacter pylori infection in gastricbiopsies. Neth. J. Med. 64:329-333.

van Leerdam, M. E., A. van der Ende, F. J. ten Kate, E. A. Rauws, G. N. Tytgat. 2003. Lackof accuracy of the noninvasive Helicobacter pylori stool antigen test in patients withgastroduodenal ulcer bleeding. Am. J. Gastroenterol. 98:798-801.

Varis, K., P. Sipponen, F. Laxén, I. M. Samloff, J. K. Huttunen, P. R. Taylor, O. P. Heinonen,D. Albanes, N. Sande, J. Virtamo, M. Härkönen. 2000. Implications of serum pepsinogen I inearly endoscopic diagnosis of gastric cancer and dysplasia. Helsinki Gastritis Study Group.Scand. J. Gastroenterol. 35:950-956.

Vilaichone, R. K., V. Mahachai, C. Kositchaiwat, D. Y. Graham, Y. Yamaoka. 2003. Relationbetween seroreactivity to low-molecular-weight Helicobacter pylori-specific antigens anddisease presentation. Clin. Diagn. Lab. Immunol. 10:1025-1028.

92

Warren, J. R. and B. J. Marshall. 1983. Unidentified curved bacilli on gastric epithelium inactive chronic gastritis. Lancet i:1273-1275.

Watanabe, T., S. Tomita, M. Kudo, M. Kurokawa, A. Orino, A. Todo, T. Chiba. 1998.Detection of Helicobacter pylori gene by means of immunomagnetic separation-basedpolymerase chain reaction in feces. Scand. J. Gastroenterol. 33:1140-1143.

Weiss, J., J. Mecca, E. da Silva, D. Gassner. 1994. Comparison of PCR and other diagnostictechniques for detection of Helicobacter pylori infection in dyspeptic patients. J. Clin.Microbiol. 32:1663-1668.

Whittingham, S. and I. R. Mackay. 2005. Autoimmune gastritis: historical antecedents,outstanding discoveries, and unresolved problems. Int. Rev. Immunol. 24:1-29.

Wilcox, M. H., T. H. Dent, J. O. Hunter, J. J. Gray, D. F. Brown, D. G. Wight, E. P. Wraight.1996. Accuracy of serology for the diagnosis of Helicobacter pylori infection--a comparisonof eight kits. J. Clin. Pathol. 49:373-376.

Winiarski, M., W. Bielanski, M. Plonka, M. Dobrzanska, A. Kaminska, A. Bobrzynski, P. C.Ronturek, S. J. Konturek. 2003. The usefulness of capsulated 13C-urea breath test in diagnosisof Helicobacter pylori infection in patients with upper gastrointestinal bleeding. J. Clin.Gastroenterol. 37:34-38.

Wi niewska, M., H. O. Nilsson, L. Bak-Romaniszyn, T. Rechci ski, W. Biela ski, I. P aneta-Ma ecka, M. P onka, S. Konturek, T. Wadstrom, W. Rudnicka, M. Chmiela. 2002. Detectionof specific Helicobacter pylori DNA and antigens in stool samples in dyspeptic patients andhealthy subjects. Microbiol. Immunol. 46:657-665.

Wong, B. C., S. K. Lam, W. M. Wong, J. S. Chen, T. T. Zheng, R. E. Feng, K. C. Lai, W. H.Hu, S. T. Yuen, S. Y. Leung, D. Y. Fong, J. Ho, C. K. Ching, J. S. Chen, China GastricCancer Study,Group. 2004. Helicobacter pylori eradication to prevent gastric cancer in ahigh-risk region of China: a randomized controlled trial. JAMA 291:187-194.

Wong, W. M., S. K. Lam, K. C. Lai, K. M. Chu, H. H. Xia, K. W. Wong, K. L. Cheung, S. K.Lin, B. C. Wong. 2003. A rapid-release 50-mg tablet-based 13C-urea breath test for thediagnosis of Helicobacter pylori infection. Aliment. Pharmacol. Ther. 17:253-257.

Wyatt, J. I. and B. J. Rathbone. 1989. Gastric metaplasia in the duodenum and Campylobacterpylori. Gastroenterol. Clin. Biol. 13:78B-82B.

Wyatt, J. I., B. J. Rathbone, M. F. Dixon, R. V. Heatley. 1987. Campylobacter pyloridis andacid induced gastric metaplasia in the pathogenesis of duodenitis. J. Clin. Pathol. 40:841-848.

Wyatt, J. I., B. J. Rathbone, G. M. Sobala, T. Shallcross, R. V. Heatley, A. T. Axon, M. F.Dixon. 1990. Gastric epithelium in the duodenum: its association with Helicobacter pyloriand inflammation. J. Clin. Pathol. 43:981-986.

Yakoob, J., W. Jafri, S. Abid, N. Jafri, Z. Abbas, S. Hamid, M. Islam, K. Anis, H. A. Shah, H.Shaikh. 2005. Role of rapid urease test and histopathology in the diagnosis of Helicobacterpylori infection in a developing country. BMC Gastroenterol. 5:38.

Yamaoka, Y., T. Kodama, D. Y. Graham, K. Kashima. 1998. Comparison of four serologicaltests to determine the CagA or VacA status of Helicobacter pylori strains. J. Clin. Microbiol.36:3433-3434.

93

Yang, H., M. F. Dixon, J. Zuo, F. Fong, D. Zhou, I. Corthesy, A. Blum. 1995. Helicobacterpylori infection and gastric metaplasia in the duodenum in China. J. Clin. Gastroenterol.20:110-112.

Yilmaz, O. and E. Demiray. 2007. Clinical role and importance of fluorescence in situhybridization method in diagnosis of H pylori infection and determination of clarithromycinresistance in H pylori eradication therapy. World J. Gastroenterol. 13:671-675.

Yilmaz, O., E. Demiray, S. Tümer, O. Altungöz, K. Yörüko lu, M. Soytürk, I. im ek. 2007.Detection of Helicobacter pylori and determination of clarithromycin susceptibility usingformalin-fixed, paraffin-embedded gastric biopsy specimens by fluorescence in situhybridization. Helicobacter 12:136-141.

Yoshida, S., T. Kozu, T. Gotoda, D. Saito. 2006. Detection and treatment of early cancer inhigh-risk populations. Baillieres Best Pract. Res. Clin. Gastroenterol. 20:745-765.

Zerbib, F., C. Lenk, B. Sawan, R. Cayla, N. Broutet, B. Carles, A. de Mascarel, F. Mégraud,H. Lamouliatte. 2000. Long-term effects of Helicobacter pylori eradication on gastric antralmucosa in duodenal ulcer patients. Eur. J. Gastroenterol. Hepatol. 12:719-725.

Zhou, L. Y., S. R. Lin, S. G. Ding, X. B. Huang, L. Zhang, L. M. Meng, R. L. Cui, J. Zhu.2005. The changing trends of the incidence of gastric cancer after Helicobacter pylorieradication in Shandong area. Chin. J. Dig. Dis. 6:114-115.

Zullo, A., C. Hassan, R. Lorenzetti, S. Winn, S. Morini. 2003. A clinical practice viewpoint:to culture or not to culture Helicobacter pylori? Dig. Liver Dis. 35:357-361.

94

95

ORIGINAL PUBLICATIONS