Aliment Pharmacol Ther 2002; 16: 281–289.

Helicobacter pylori, parietal cell antibodies and autoimmune

gastropathy in type 1 diabetes mellitus
C. E. M . D E BLOCK*, I. H. DE LEEUW*, J. J. P. M. BOG ERS , P. A. PELCKM ANSà, M. M. IEVEN§,
E. A. E. VAN MARCK , V. VA N H OOF–, E. MÁDAY *, K. L. VA N A CKER* & L. F. VAN GAAL*
From the Departments of *Endocrinology-Diabetology, Pathology, àGastroenterology and Hepatology,
§Microbiology, –Biochemistry, University of Antwerp, University Hospital Antwerp, Edegem, Belgium
Accepted for publication 15 October 2001

associated with HLA-DQA1*0103-B1*0603 (OR ¼ 0.12,

SUMMARY
P ¼ 0.015) and positively with DQA1*0501-B1*0201
Background: Fifteen to 20% of type 1 diabetic patients (OR ¼ 1.9, P ¼ 0.032). PCA-positivity was linked to
exhibit parietal cell antibodies (PCA), which are asso- HLA-DQA1*0501-B1*0301 (OR ¼ 3.9, P ¼ 0.017). A
ciated with autoimmune gastritis, hypochlorhydria, link between H. pylori and PCA was observed when
iron deficiency and pernicious anaemia. PCA-positivity was defined as a titre ‡ 1/20 (OR ¼ 2.0,
Aim: To examine whether Helicobacter pylori infection P ¼ 0.03), but not if ‡ 1/40 was the cut-off point. PCA-
could explain the high prevalence of PCA and auto- positivity, but not H. pylori infection, was associated
immune gastropathy in diabetes. If so, H. pylori with iron deficiency anaemia (OR ¼ 2.7, P ¼ 0.008),
eradication could prevent autoimmune gastritis. pernicious anaemia (OR ¼ 33.5, P < 0.0001), hypo-
Methods: In 229 type 1 diabetics (M/F: 135/94; age: chlorhydria (OR ¼ 12.1, P ¼ 0.0008) and autoimmune
41 ± 12 years) PCA were measured. H. pylori infection gastritis (OR ¼ 12.5, P < 0.0001).
was assessed by serology, urea breath test in all and by Conclusions: The HLA-bound susceptibility of H. pylori
histology (updated Sydney system) in 88 subjects. and PCA differed. PCA-positivity but not ongoing
Pentagastrin tests were performed in 42 patients. H. pylori infection is associated with autoimmune
Results: Sixty-nine patients were PCA-positive. H. pylori gastritis. Low titres of PCA might reflect H. pylori
infection was present in 72 patients and was negatively infection rather than autoimmune gastropathy.

pernicious anaemia,8, 9 and may predispose to gastric

INTRODUCTION
cancer10, 11 and carcinoid tumours.11–13
Type 1 diabetes mellitus results from autoimmune Helicobacter pylori infection has been recognized as a
destruction of pancreatic b-cells. In addition, 15–20% risk factor for development of atrophic antral gastri-
of these patients exhibit parietal cell antibodies (PCA) tis,14, 15 intestinal metaplasia16 and gastric cancer.16–19
compared to 2–10% of nondiabetic subjects.1, 2 PCA Recently, gastric autoantibodies have been demonstra-
target the gastric proton pump (H+/K+ATPase) and are ted in subjects with H. pylori-positive gastritis20–25 and
a marker of autoimmune atrophic gastritis,3–5 which thus the question arises whether H. pylori may be
is associated with iron deficiency anaemia6, 7 and implicated in the pathogenesis of autoimmune atrophic
corpus gastritis.
Genetic (e.g. HLA type), immune and environmental
Correspondence to: Dr C. E. M. De Block, Department of Endocrinology- factors, in addition to the differential pathogenicity of
Diabetology, Faculty of Medicine, University of Antwerp (UA), University
Hospital Antwerp, Wilrijkstraat 10, B-2650 Edegem, Belgium. H. pylori strains may be important in modifying the risk
E-mail: [email protected] associated with H. pylori infection.26 Diabetic patients

Ó 2002 Blackwell Science Ltd 281

 13652036, 2002, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1046/j.1365-2036.2002.01186.x by Nat Prov Indonesia, Wiley Online Library on [28/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
282 C. E. M. DE BLOCK et al.

may be at high risk for H. pylori infection because they Upjohn, Germany, N < 10 U/mL) (n ¼ 175; r ¼ 0.85;
may be more frequently exposed to it, owing to regular P < 0.0001). Antibodies to intrinsic factor (AIF) were
hospital visits,27 and because their immune system is measured by radiobinding assay (Diagnostic Products
affected.28 However, there are only a few studies on the Corporation, Los Angeles, USA, Nl < 1.1). Serum
relationship between H. pylori and PCA in type 1 gastrin was measured by RIA technique (Euro-
diabetes with incomplete investigation of H. pylori Diagnostics, Malmö, Sweden, Nl < 110 ng/L) and sol-
infection and suboptimal or no documentation of uble transferrin receptor (sTf R) levels by nephelometry
histological and gastric acid secretion data.29–31 (BN II; Dade Behring, Germany, Nl: 0.83–1.76 mg/L).
We investigated whether H. pylori infection could be a Iron deficiency anaemia was defined as described
reason for the increased prevalence of PCA, autoim- before.6 Pernicious anaemia was defined as a macrocy-
mune gastritis and associated pathology (anaemia, tic anaemia with low vitamin B12 levels, hypergastri-
hypochlorhydria) in type 1 diabetic patients. Should naemia, positive AIF and/or PCA, and responsive to
any of these be the case, H. pylori eradication could vitamin B12 therapy.
prevent autoimmune gastric disease in these patients. HLA-DQ (Human Leucocyte Antigen) typing was
We also studied the relationship between HLA-DQ performed as described previously.32
polymorphism and susceptibility to H. pylori and to PCA. Gastric acid secretion was studied in 42 patients in
euglycaemic conditions after a 12-h overnight fast
using pentagastrin-stimulated acid output (6 lg/kg
PATIENTS AND METHODS body weight s.c., Pentagastrin Injection BP; Cambridge
Laboratories Ltd, Newcastle upon Tyne, UK). Gastric
Patients
juice was collected for 1 h before and 1 h after
Between October 1998 and August 2000, 229 con- pentagastrin administration in 15-min aliquots. Hypo-
secutive type 1 diabetic patients (male/female: 135/94; chlorhydria was defined as maximal acid output
PCA +/): 69/160; mean age: 41 ± 12 years, diabetes (MAO) < 15 mmol H+/h.33
duration: 18 ± 10 years) attending the out-patient Serum was assayed for H. pylori IgG antibodies by ELISA
Antwerp University diabetes clinic, were studied. (Roche Diagnostics, Brussels, Belgium). Urea breath test
Eighty-eight patients (M/F: 51/37), comprising 47 (UBT) were performed in all subjects after a 12-h
PCA-positive and 41 PCA-negative subjects matched overnight fast using 75 g 13C urea, and breath samples
for age, gender, duration of diabetes and metabolic taken at 15, 30, 45 and 60 min were analysed by
control, underwent gastroscopy, and 42 patients mass-spectrophotometry.34 Antrum, corpus and fundus
(PCA +/): 15/27) agreed to participate in gastric acid biopsies of 88 patients were examined for H. pylori
secretion studies. They were consecutively recruited colonization using a modified Giemsa stain and/or
according to PCA status and irrespective of symptoms. immunostaining (cfr infra). H. pylori infection was
One-hundred sex- and age-matched controls (M/F: 46/ diagnosed if any test was positive. The period between
54; age: 39 ± 14 years; PCA +/): 7/93) were tested for urea breath test and gastroscopy was maximally 14 days.
H. pylori serology. None of the diabetic and control
subjects used antibiotics, anti-inflammatory drugs, Gastroscopy and histology. At upper gastrointestinal
prokinetics or proton pump inhibitors. The study was endoscopy (Olympus Videoscope GIF/Q140, Melville,
approved by the local Ethics Committee. Each subject NY, USA) at least two biopsies from fundus, corpus,
gave informed consent, in accordance with the Helsinki antrum and descending duodenum were taken. Five
Declaration. micrometer sections were evaluated by two investiga-
tors, unaware of the patient’s clinical and laboratory
findings. The visual analogue scale of the updated
Methods
Sydney system was used to evaluate inflammation,
PCA were assayed by indirect immunofluorescence activity, atrophy, intestinal metaplasia and H. pylori
(Medical Diagnostics California, CA, USA, Nl < 1/20 colonization, which were graded as follows: 0 ¼
dilution).2 This indirect immunofluorescence assay absent, 1 ¼ mild, 2 ¼ moderate and 3 ¼ severe.35
correlated well with the enzyme immunoassay for Autoimmune gastritis, formerly known as type A
H+/K+ATPase antibodies (Varelisa; Pharmacia & atrophic corpus gastritis, spares the antrum.3 Atrophy

Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 281–289

 13652036, 2002, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1046/j.1365-2036.2002.01186.x by Nat Prov Indonesia, Wiley Online Library on [28/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
H. PYLORI AND AUTOIMMUNE GASTROPATHY IN TYPE 1 DIABETES 283

was defined as a genuine loss of glands, according to ANOVA was used to determine differences between
Genta.36 Furthermore, histological criteria for active groups. Bonferroni adjustments for multiple compar-
autoimmune gastritis without total atrophy (diffuse isons were made when appropriate. Differences in
lymphocytic infiltration of the lamina propria, focal distributions of categorical data were investigated by
lymphocyte-mediated destruction of oxyntic glands and v2 or Fisher’s Exact test. Stepwise forward logistic
parietal cell pseudohypertrophy) were evaluated.37 regression analysis was used to assess the strength and
H. pylori immunostaining was performed, particularly independency of associations. A two-tailed P < 0.05
if lymphoid aggregates, atrophy or intestinal metaplasia was considered significant.
were seen. A polyclonal rabbit H. pylori IgG antibody
(DAKO B0471; Glostrup, Denmark, dilution: 1/200)
was used as primary antibody. Sites of immunoreacti- RESULTS
vity were revealed using polyclonal goat-antirabbit H. pylori findings
antibody (DAKO E0432, 1/500) and streptavidine-
biotinylated horseradish peroxidase (DAKO K0377; H. pylori infection was present in 72 patients (31%)
100 lL/slide; 20¢), followed by incubation in a 3,3¢- with a similar gender distribution. The positive concor-
diamino-benzidine tetrahydrochloride solution (HK dance between H. pylori IgG and UBT was 86%, and
153–5K; BioGenex, San Ramon, CA, USA). between H. pylori IgG and histology was 91%. The
respective negative concordance rates were 96% and
100%. In comparison, H. pylori IgG was positive in 42 of
Statistical analysis
100 healthy sex- and age-matched controls.
Results were analysed using SPSS (SPSS Inc., Chicago, H. pylori-positive subjects were older than H. pylori-
IL, USA). The unpaired t-test, Mann–Whitney U-test or negative patients (P ¼ 0.046) (Table 1). H. pylori
Table 1. Clinical, histological and acid
H. pylori +ve H. pylori )ve P-value
secretion parameters in H. pylori-positive
vs. H. pylori-negative type 1 diabetic n (M/F) 72 (46/26) 157 (89/68) N.S.
patients Age (y) 43 ± 13 40 ± 12 0.046
Duration of diabetes (years) 18 ± 12 18 ± 10 N.S.
HbA1c (%) 7.8 ± 1.0 7.9 ± 1.1 N.S.
aTPO > 100 lU/mL (n) 23 (32%) 36 (23%) N.S.
PCA ‡ 1/20 (n) 29 (40%) 40 (25%) 0.03
PCA ‡ 1/40 (n) 24 (33%) 34 (22%) N.S. (0.07)
HLA-DQA1*0501-B1*0201 46 (64%) 75 (48%) 0.032
HLA-DQA1*0103-B1*0603 1 (1%) 16 (10%) 0.015
sTfRec (mg/L) 1.22 ± 0.38 1.16 ± 0.28 N.S.
Gastrin (ng/L) 154 ± 147 156 ± 213 N.S.
Iron deficiency anaemia (n) 14 (19%) 27 (17%) N.S.
Pernicious anaemia (n) 2 (3%) 11 (7%) N.S.
Histological data (n = 88) 47 41
Autoimmune gastritis (n) 16 (34%) 15 (37%) N.S.
Corpus inflammation 1 [0–2] 1 [0–2] N.S.
Corpus atrophy 1 [0–3] 1 [0–3] N.S.
Corpus metaplasia 0 [0–3] 0 [0–2] N.S.
Corpus H. pylori 1 [0–3] 0 [0–0] N.S.
ECL hyperplasia (n) 3 (6%) 4 (10%) N.S.
Acid secretion data (n = 42) 12 30
Hypochlorhydria (n) 5/12 11/30 N.S.
BAO (mmol H+/h) 1.61 ± 2.13 1.32 ± 1.34 N.S.
MAO (mmol H+/h) 11.91 ± 9.53 12.74 ± 9.41 N.S.
PAO (mmol H+/h) 17.42 ± 13.92 18.51 ± 12.14 N.S.
Fasting pH 4.14 ± 2.51 3.34 ± 2.20 N.S.

Data are expressed as means ± standard deviation, median [range] or numbers (percentages).
BAO, MAO, PAO: basal, maximal, peak acid output.

Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 281–289

 13652036, 2002, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1046/j.1365-2036.2002.01186.x by Nat Prov Indonesia, Wiley Online Library on [28/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
284 C. E. M. DE BLOCK et al.

infection showed a negative association with HLA- autoimmune gastritis were also similar between
DQA1*0103-B1*0603 (OR ¼ 0.12 [0.02–0.95]; P ¼ H. pylori-positive and H. pylori-negative subjects. Antral
0.015) and a positive one with HLA-DQA1*0501- inflammation was more pronounced in H. pylori
B1*0201 (OR ¼ 1.9 [1.1–3.4]; P ¼ 0.032). H+/K+ infected patients (P ¼ 0.005).
ATPase antibody levels, gastrin levels and acid secretion
parameters did not differ between H. pylori-positive
Parietal cell antibody findings
and H. pylori-negative subjects. Of interest, we noted an
association between H. pylori and PCA when PCA were equally common in males and females. Age
PCA-positivity was defined as a titre ‡ 1/20 (OR ¼ 2.0 and diabetes duration were similar in PCA-positive and
[1.1–3.6]; P ¼ 0.03), but no longer if the cut-off point PCA-negative subjects (Table 2). The presence of PCA
was ‡ 1/40. Prevalences of iron deficiency anaemia, was associated with HLA-DQA1*0501-B1*0301
pernicious anaemia and autoimmune atrophic gastritis (OR ¼ 3.9 [1.3–11.3]; P ¼ 0.017). A link between
were similar in H. pylori-positive and H. pylori-negative thyroid peroxidase (aTPO) and PCA was found
individuals (Table 1). (OR ¼ 2.1 [1.1–3.9]; P ¼ 0.021). Soluble transferrin
Gastroscopy and histology were performed in 18 PCA- receptor (sTf R) levels (P ¼ 0.039), a marker of iron
negative H. pylori-negative; 23 PCA-positive H. pylori- deficiency anaemia, and gastrin levels (P ¼ 0.012) were
negative; 23 PCA-negative H. pylori-positive; and 24 higher in PCA-positive subjects. In 42 patients tested,
PCA-positive H. pylori-positive patients. Two-way ANOVA basal (P ¼ 0.012), maximal (P ¼ 0.004) and peak acid
with PCA and H. pylori status as fixed factors output (P ¼ 0.011) were lower, fasting pH was higher
showed that H. pylori status did not influence the (P ¼ 0.049), and hypochlorhydria was more prevalent
degree of corpus or antrum atrophy, intestinal meta- (OR ¼ 12.1 [2.7–54.3]; P ¼ 0.0008) in PCA-postive
plasia, or the prevalence of enterochromaffin-like (ECL) subjects. PCA status was the only risk factor (b ¼ 0.08;
cell hyperplasia. Histological parameters of pre-atrophic P ¼ 0.028) for hypochlorhydria in a logistic regression

Table 2. Clinical, histological and acid

PCA ‡ 1/20 PCA < 1/20 P-value
secretion parameters in PCA-positive vs.
n (M/F) 69 (39/30) 160 (96/64) N.S. PCA-negative type 1 diabetic patients
Age (years) 41 ± 13 41 ± 12 N.S.
Duration of diabetes (years) 19 ± 11 18 ± 10 N.S.
aTPO > 100 lU/mL (n) 25 (36%) 34 (21%) 0.021
H. pylori-positive (n) 29 (42%) 43 (27%) 0.03
HLA DQA1*0501-B1*0301 9 (13%) 6 (4%) 0.017
sTfRec (mg/L) 1.25 ± 0.34 1.13 ± 0.30 0.039
Gastrin (ng/L) 197 ± 246 102 ± 46 0.012
Iron deficiency anaemia (n) 20 (29%) 21 (13%) 0.008
Pernicious anaemia (n) 12 (17%) 1 (0.6%) < 0.0001
Histological data (n = 88) 47 41
Autoimmune gastritis (n) 27 (57%) 4 (10%) < 0.0001
Corpus inflammation 1 [0–2] 1 [0–2] N.S.
Corpus atrophy 1 [0–3] 0 [0–2] 0.02
Corpus metaplasia 0 [0–3] 0 [0–2] N.S.
Corpus H. pylori 1 [0–3] 0 [0–2] N.S.
ECL hyperplasia (n) 7 (15%) 0 (0%) 0.013
Acid secretion data (n = 42) 15 27
Hypochlorhydria (n) 11 (73%) 5 (18%) 0.0008
BAO (mmol H+/h) 0.65 ± 0.77 1.89 ± 1.73 0.012
MAO (mmol H+/h) 7.76 ± 6.99 16.90 ± 10.38 0.004
PAO (mmol H+/h) 12.62 ± 12.03 23.12 ± 13.44 0.011
Fasting pH 4.63 ± 2.56 2.97 ± 1.96 0.049

Data are expressed as means ± standard deviation, median [range] or numbers (percentages).
BAO, MAO, PAO: basal, maximal, peak acid output.

Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 281–289

 13652036, 2002, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1046/j.1365-2036.2002.01186.x by Nat Prov Indonesia, Wiley Online Library on [28/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
H. PYLORI AND AUTOIMMUNE GASTROPATHY IN TYPE 1 DIABETES 285

model testing PCA, H. pylori infection, age, gender and involvement of corpus and antrum, and study
diabetes duration. More PCA-positive than PCA-negative design.33, 42 Acute H. pylori gastritis is associated with
subjects had iron deficiency anaemia (OR ¼ 2.7 acidic gastric pH and hypergastrinaemia in the first
[1.4–5.4]; P ¼ 0.008), pernicious anaemia (OR ¼ 33.5 week, followed by achlorhydria, which lasts for months.
[4.3–263.4]; P < 0.0001) and autoimmune gastritis Thereafter, gastric acid and gastrin levels usually return
(OR ¼ 12.5 [3.8–40.7]; P < 0.0001). to normal,42 although not in subjects who develop
Signs of pre-atrophic autoimmune gastritis were more mucosal atrophy.33, 43 H. pylori could influence acid
prevalent in PCA-positive subjects, as documented by a secretion by changing the number of parietal cells,
more pronounced lymphocytic infiltration in corpus which we did not observe, or by modifying their
mucosa (P ¼ 0.038) and partial atrophy of oxyntic function via H. pylori toxins or inflammatory medi-
glands (P ¼ 0.014). Logistic regression showed that ators.44, 45 Achlorhydria may cause hypergastrinaemia
PCA status (b ¼ 1.44; P ¼ 0.04) and gastrin levels by interrupting the negative feedback of somatostatin
(b ¼ 0.01; P ¼ 0.026), but not H. pylori infection, age, on gastrin-cells.46 We confirmed that gastrin levels
duration of diabetes or gender were risk factors for correlated negatively with peak acid output and posi-
autoimmune atrophic gastritis. PCA-positive subjects tively with the degree of corpus atrophy.47
had a more severe corpus atrophy (P ¼ 0.02) and a H. pylori infection might also cause iron deficiency
higher prevalence of ECL cell hyperplasia (7/47 vs. anaemia, probably through atrophy-induced hypo/
0/41; P ¼ 0.013) and corpus intestinal metaplasia achlorhydria, or through competition with the host
(11/47 vs. 1/41; P ¼ 0.004) than PCA-negative dia- for the absorption of iron.48 We could not support this
betic patients. Moreover, one PCA-positive patient with observation. PCA are a marker of pernicious anaemia,
autoimmune gastritis had a gastric carcinoid tumour14 which is 10 times more frequent in type 1 diabetic than
and another one was diagnosed with gastric cancer. nondiabetic subjects.9 In contrast, we and others have
However, this last case also showed H. pylori coloniza- found no or a negative association between H. pylori
tion at the stage of gastritis, but no longer at the stage of and pernicious anaemia.7, 49, 50 PCA-positive diabetic
adenocarcinoma. patients may be at increased risk to develop neuropathy
due to both diabetes and cobalamin deficiency, which is
amenable to vitamin B12 therapy.
DISCUSSION

Approximately 15–20% of type 1 diabetic patients

Histological data
exhibit parietal cell antibodies,1, 2 a marker of perni-
cious anaemia and autoimmune gastritis.3, 8 We inves- As supported by our data, PCA are a good marker of
tigated whether H. pylori infection, present in 30–40% autoimmune atrophic corpus gastritis, which occurs in
of 35–40-year-old Belgian subjects,38 is associated with 0.1–10% of the general population, whereas in type 1
an increased prevalence of PCA and autoimmune diabetes its prevalence is 5–10%, amounting to 20–40%
gastropathy in type 1 diabetes. of PCA-positive patients.2, 8, 9 However, the terms
‘atrophic gastritis’ and ‘gastric atrophy’ were poorly
defined in the past36 and data on antral or corpus
Gastric acid production data and haematological findings
involvement were usually not mentioned, making
PCA target gastric H+/K+ATPase and may compromise studies using a good definition of autoimmune atrophic
gastric acid production,4, 5 which is important for iron gastritis3, 35 needed. Recently, it has been suggested
absorption, and thus leads to iron deficiency anae- that H. pylori infection induces autoimmune gastri-
mia.39, 40 We and others have shown that acid tis14, 20–25, 51 by stimulating granulocytes to produce
secretion is impaired, gastrin levels are higher and iron autodestructive oxygen radicals, which are mutagenic
deficiency anaemia is more frequent in PCA-positive and may cause corpus atrophy and H. pylori self-
than PCA-negative subjects,6, 31, 40 whereas H. pylori destruction due to intestinal metaplasia or achlorhy-
infection did not influence these parameters.41 How- dria.19 This hypothesis is supported by Karnes et al. who
ever, variable effects of H. pylori infection on acid noticed a high prevalence of H. pylori-seropositivity and
secretion, iron and gastrin levels have been reported, a low prevalence of positive H. pylori staining in subjects
depending on the duration of infection, the relative with atrophic corpus gastritis.20 One of our patients also

Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 281–289

 13652036, 2002, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1046/j.1365-2036.2002.01186.x by Nat Prov Indonesia, Wiley Online Library on [28/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
286 C. E. M. DE BLOCK et al.

showed H. pylori infection at the stage of gastritis, but

Pathogenetic considerations
no longer at the stage of adenocarcinoma. Cohort
follow-up studies have suggested a 0.3–2% annual Possible mechanisms leading to the formation of gastric
increase of atrophic corpus gastritis in subjects with antibodies and autoimmune gastritis in H. pylori
H. pylori-positive gastritis.14, 52, 53 However, others infection are still under investigation. First, molecular
have found no or a negative link between H. pylori mimicry may occur between H. pylori lipopolysaccha-
and atrophic corpus gastritis,15, 44, 54 and earlier stud- ride Lewis epitopes and Lewis blood group antigens
ies, for historical reasons, did not report the H. pylori expressed by gastric epithelial cells,23 or between the
status.55, 56 We observed no association between b-subunits of H. pylori urease and gastric H+/K+
H. pylori status and pre-atrophic autoimmune gastritis, ATPase,22 or between the NH2-terminal sequence of a
when the mucosa is still hospitable for H. pylori, and we vacuolating toxin that is secreted by H. pylori and the
identified PCA and gastrin as risk markers for autoim- H+/K+ATPase a-subunit in the region of residues 243–
mune gastritis. A correlation between H. pylori and 257 (30% homology).22, 61 However, others have
gastric antibodies has been reported in some21–25 shown that H. pylori antibodies did not cross-react with
but not all studies.15, 54 We observed an association H+/K+ATPase and could not demonstrate a common
between H. pylori and PCA when the cut-off point of epitope.21 In addition, a recent study showed that Lewis
PCA-positivity was ‡ 1/20, but not anymore when it blood group antigens were not involved in the process of
was ‡ 1/40, suggesting that low PCA titres might reflect molecular mimicry and that the anti-Lewis immune
H. pylori infection. response is probably H. pylori-independent.25 H. pylori-
The HLA-bound susceptibility to H. pylori infection and infected patients seem to develop antibodies to a peptide
to PCA differed in this study. Others have noted of H+/K+ATPase but not through molecular mimicry.25
that HLA-DQB1*0301 was more common in patients Second, H. pylori infection results in mucosal influx of
with gastric adenocarcinoma than in controls and T-helper 1 (Th1) cells,62 producing IL-12 which acti-
that H. pylori-seropositivity was less frequent in vates interferon-c (IFN-c) production. IFN-c may stimu-
DQB1*0301 positive patients with gastric cancer com- late aberrant HLA-DR expression on gastric epithelial
pared with DQB1*0301 negative patients.57 We report cells, enabling them to present autoantigens such as
an assocation between PCA and DQB1*0301.58 This gastric H+/K+ATPase, which induces the activation of
suggests that a genetic influence may be involved in the autoreactive T-cells.63 These T-cells help to induce
pathogenesis of gastric cancer in PCA-positive patients autoantibodies and destruct the glands, resulting in
with autoimmune gastritis. mucosal atrophy. In a last step T-cells are thought to
In the 88 patients subjected to gastroscopy, nearly become H. pylori-independent and an autoimmune
40% of PCA-positive subjects with autoimmune gastritis response perpetuates.
showed (pre)malignant gastric lesions. Gastric carcinoid H. pylori eradication may improve inflammation, acid
tumours, developing on a background of ECL cell production and reduce gastrin levels, but the effects on
hyperplasia induced by hypergastrinaemia, occur in gastric atrophy and intestinal metaplasia are contro-
4–7% of patients with autoimmune gastritis and/or versial.33, 49
pernicious anaemia, which is 13 times more frequent In conclusion, the presence of PCA, but not H. pylori
than in controls.12 Autoimmune gastritis with hypo- infection was associated with autoimmune gastritis,
chlorhydria may also result in overgrowth of bacteria iron deficiency anaemia, pernicious anaemia, hypergas-
producing mutagenic N-nitroso compounds, which can trinaemia and hypochlorhydria. However, our data do
lead to intestinal meta-/dysplasia-precancerous condi- not preclude an early influence of H. pylori in the later
tions.19 Patients with atrophic gastritis/pernicious development of autoimmune gastritis. No clear rela-
anaemia have a 3–5-fold increased gastric cancer risk tionship between PCA and H. pylori infection was
(ranging between 1 and 9%).59, 60 H. pylori infection observed. Low titres of PCA might suggest H. pylori
has also been associated with a 2–4-fold increased risk infection rather than autoimmune gastropathy, and the
of gastric adenocarcinoma, with a life-time risk amount- HLA-bound susceptibility of H. pylori infection and PCA
ing to 1–2%.60 This risk seems not to be affected by the differed. Longitudinal and post-H. pylori-eradication
aetiology of the atrophic condition, particularly whether studies are needed to clarify whether H. pylori is involved
it is related to autoimmune gastritis or to H. pylori.60 in the induction of PCA and in the pathogenesis of

Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 281–289

 13652036, 2002, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1046/j.1365-2036.2002.01186.x by Nat Prov Indonesia, Wiley Online Library on [28/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
H. PYLORI AND AUTOIMMUNE GASTROPATHY IN TYPE 1 DIABETES 287

well-defined autoimmune gastritis. Nevertheless, pernicious anaemia in diabetes mellitus. Lancet 1968; 2:
because of the beneficial effects on inflammation, acid 415–7.
10 Armbrecht U, Stockbrügger RW, Rode J, Menon GG, Cotton
production and gastrin levels, we suggest eradication of
PB. Development of gastric dysplasia in pernicious anaemia: a
H. pylori in PCA-positive subjects in order to prevent clinical and endoscopic follow-up study of 80 patients. Gut
possible autoimmune gastropathy. 1990; 31: 1105–9.
11 Kokkola A, Sjoblom SM, Haapiainen R, Sipponen P, Puolak-
kainen P, Järvinen H. The risk of gastric carcinoma and car-
ACKNOWLEDGEMENTS cinoid tumours in patients with pernicious anaemia: a
prospective follow-up study. Scand J Gastroenterol 1998; 33:
Supported by a grant from the European Foundation for 88–92.
the Study of Diabetes (EFSD). 12 Solcia E, Fiocca T, Villani L, et al. Morphology and patho-
We gratefully acknowledge Dr C. Van Campenhout genesis of endocrine hyperplasias, precarcinoid lesions, and
and K. Van Cotthem (Laboratory of Immunology & carcinoids arising in chronic atrophic gastritis. Scand J Gas-
Protein Chemistry), B. Raeymaeckers and R. Borrie troenterol 1991; 26(Suppl. 180): 146–59.
13 De Block CEM, Leeuw IH, Pelckmans PA, et al. Autoimmune
(Department of Gastroenterology), and Apr. M. Martin
hepatitis, autoimmune gastritis, and gastric carcinoid in a
(Laboratory of Hormonology). We thank Dr F. Peiffer type 1 diabetic patient—a case report. J Diabetes Complica-
and Dr P. Abrams (Department of Endocrinology- tions 2000; 14: 116–20.
Diabetology) who contributed to the recruitment of 14 Kuipers EJ, Uyterlinde AM, Pena AS, et al. Long-term sequelae
patients, and Dr B. Manuel y Keenoy, J. Vertommen, of Helicobacter pylori gastritis. Lancet 1995; 345: 1525–8.
P. Aerts, S. Schrans and M. Vinckx (Laboratory of 15 Oksanen A, Sipponen P, Karttunen R, et al. Atrophic gastritis
and Helicobacter pylori infection in outpatients referred for
Endocrinology-Diabetology), University of Antwerp. gastroscopy. Gut 2000; 46: 460–3.
16 Sipponen P, Hyvärinen H, Seppälä K, Blaser MJ. Review
article: pathogenesis of the transformation from gastritis to
REFERENCES
malignancy. Aliment Pharmacol Ther 1998; 12(Suppl. 1):
1 Riley WJ, Toskes PP, Maclaren NK, Silverstein J. Predictive 61–71.
value of gastric parietal cell autoantibodies as a marker for 17 Parsonnet J, Friedman G, Vandersteen D, et al. Helicobacter
gastric and hematologic abnormalities associated with insulin pylori infection and the risk of gastric cancer. N Engl J Med
dependent diabetes. Diabetes 1982; 31: 1051–5. 1991; 325: 1127–31.
2 De Block CEM, Van Gaal LF, Leeuw IH, Belgian Diabetes 18 Uemura N, Okamoto S, Yamamoto S, et al. Helicobacter pylori
Registry. High prevalence of manifestations of gastric auto- infection and the development of gastric cancer. N Engl J Med
immunity in parietal cell-antibody positive type 1 (insulin- 2001; 345: 784–9.
dependent) diabetic patients. J Clin Endocrinol Metab 1999; 19 Correa P. Human gastric carcinogenesis: a multistep and
84: 4062–7. multifactorial process. Cancer Res 1992; 52: 6735–40.
3 Strickland RG, Mackay IR. A reappraisal and significance of 20 Karnes WE Jr, Samloff IM, Siurala M, et al. Positive serum
chronic atrophic gastritis. Dig Dis 1973; 18: 426–40. antibody and negative tissue staining for Helicobacter pylori in
4 Burman P, Mardh S, Norberg L, Karlsson FA. Parietal cell subjects with atrophic body gastritis. Gastroenterology 1991;
autoantibodies in pernicious anemia inhibit H,K-adenosine 101: 167–74.
triphosphatase, the proton pump of the stomach. Gastroen- 21 Ma JY, Borch K, Sjöstrand SE, Janzon L, Mardh S. Positive
terology 1989; 96: 1434–8. correlation between H,K-Adenosine Triphosphatase autoan-
5 Toh BH, Sentry JW, Alderuccio F. The causative H+/K+ tibodies and Helicobacter pylori antibodies in patients with
ATPase antigen in the pathogenesis of autoimmune gastritis. pernicious anemia. Scand J Gastroenterol 1994; 29: 961–5.
Immunol Today 2000; 21: 348–54. 22 Uibo R, Vorobjova T, Metsküla K, Kisand K, Wadström T,
6 De Block CEM, Van Campenhout CM, Leeuw IH, et al. Soluble Kivik T. Association of Helicobacter pylori and gastric auto-
transferrin receptor level: a new marker of iron deficiency immunity: a population-based study. FEMS Immunol Med
anaemia, a common manifestation of gastric autoimmunity in Microbiol 1995; 11: 65–8.
type 1 diabetes. Diabetes Care 2000; 23: 1384–8. 23 Negrini R, Savio A, Poiesi C, et al. Antigenic mimicry between
7 Marignani M, Delle Fave G, Mecarocci S, et al. High preval- Helicobacter pylori and gastric mucosa in the pathogenesis of
ence of atrophic body gastritis in patients with unexplained body atrophic gastritis. Gastroenterology 1996; 111: 655–65.
microcytic and macrocytic anemia. Am J Gastroenterol 1999; 24 Faller G, Steiniger H, Kränzlein J, et al. Antigastric autoanti-
94: 766–72. bodies in Helicobacter pylori infection: implications of histologi-
8 Toh BH, Van Driel IR, Gleeson PA. Pernicious anemia. N Engl cal and clinical parameters of gastritis. Gut 1997; 41: 619–23.
J Med 1997; 337: 1441–8. 25 Appelmelk BJ, Faller G, Claeys D, Kirchner T, Vandenbroucke-
9 Ungar B, Stocks AE, Whittingham S, Martin FIR, Mackay IR. Grauls CM. Bugs on trial: the case of Helicobacter pylori and
Intrinsic-factor antibody, parietal-cell antibody, and latent autoimmunity. Immunol Today 1998; 19: 296–9.

Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 281–289

 13652036, 2002, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1046/j.1365-2036.2002.01186.x by Nat Prov Indonesia, Wiley Online Library on [28/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
288 C. E. M. DE BLOCK et al.

26 Forman D. Review is there significant variation in the risk of 43 Katelaris PH, Seow F, Lin BPC, Napoli J, Ngu MC, Jones DB.
gastric cancer associated with Helicobacter pylori infection? Effect of age, Helicobacter pylori infection, and gastritis with
Aliment Pharmacol Ther 1998; 12(Suppl. 1): 3–7. atrophy on serum gastrin and gastric acid secretion in healthy
27 Oldenburg B, Diepersloot RJA, Hoekstra JBL. High seropreva- men. Gut 1993; 34: 1032–7.
lence of Helicobacter pylori in diabetes mellitus patients. Dig Dis 44 Wallace JL, Cucala M, Mugridge K, Parente L. Secretagogue-
Sci 1996; 41: 458–61. specific effects of interleukin-1 on gastric acid secretion. Am
28 Nolan CM, Beaty HN, Bagdade JD. Further characterization J Physiol 1991; 261: G559–G564.
of the impaired bactericidal function of granulocytes in 45 Beil W, Birkholz C, Wagner S, Sewing KF. Interaction of
patients with poorly controlled diabetes. Diabetes 1978; 27: Helicobacter pylori and its fatty acids with parietal cells and
889–94. gastric H+/K (+) -ATPase. Gut 1994; 35: 1176–80.
29 de Luis DA, de la Calle H, Roy G, et al. Helicobacter pylori 46 Trudeau WL, McGuigan JE. Relations between serum gastrin
infection and insulin-dependent diabetes mellitus. Diabetes levels and rates of gastric hydrochloric acid secretion. N Engl
Res Clin Pract 1998; 39: 143–6. J Med 1971; 284: 408–12.
30 Barrio R, Roldan MB, Alonso M, Canton R, Camarero C. 47 Sipponen P, Valle J, Varis K, Kekki M, Ihamäki T, Siurala M.
Helicobacter pylori infection with parietal cell antibodies in Fasting levels of serum gastrin in different functional and
children and adolescents with insulin dependent diabetes morphologic states of the antrofundal mucosa. An analysis of
mellitus. J Pediatr Endocrinol Metab 1997; 10: 511–5. 860 subjects. Scand J Gastroenterol 1990; 25: 513–9.
31 De’Angelis GL, Apollionio G, Boselli E, et al. Anticorps 48 Marignani M, Angeletti S, Bordi C, et al. Reversal of
antiestomac et gastrite au cours du diabète insulino-depend- long-standing iron deficiency anaemia after eradication of
ant de l’enfant. Arch Fr Pediatr 1993; 50: 475–8. Helicobacter pylori infection. Scand J Gastroenterol 1997; 32:
32 Heimberg H, Nagy ZP, De Somers G, Leeuw I, Schuit FC. 617–22.
Complementation of HLA DQA and -DQB genes confers sus- 49 Annibale B, Aprile MR, D’ambra G, Caruana P, Bordi C,
ceptibility and protection to insulin-dependent diabetes mell- Delle Fave G. Cure of Helicobacter pylori infection in atrophic
itus. Human Immunol 1992; 33: 10–7. body gastritis patients does not improve mucosal atrophy
33 El-Omar EM, Oien K, El-Nujumi A, et al. Helicobacter pylori but reduces hypergastrinemia and its related effects on body
infection and chronic gastric acid hyposecretion. Gastroen- ECL-cell hyperplasia. Aliment Pharmacol Ther 2000; 14:
terology 1997; 113: 15–24. 625–34.
34 Peeters M, Ghoos Y, Geypens B, Rutgeerts P. Breath tests in 50 Fong TL, Dooley CP, Dehesa M, et al. Helicobacter pylori
Helicobacter pylori infection: methodological aspects. J Physiol infection in pernicious anemia: a prospective controlled study.
Pharmacol 1997; 48(Suppl. 4): 67–73. Gastroenterology 1991; 100: 328–32.
35 Dixon MF, Genta RM, Yardley JH, Correa P and the Partici- 51 Varis O, Valle J, Siurala M. Is Helicobacter involved in the
pants in the International Workshop on the Histopathology of pathogenesis of the gastritis characteristic of pernicious
Gastritis. Houston 1994. Classification and grading of gastri- anaemia? Comparison between pernicious anaemia relatives
tis. The updated Sydney system. Am J Surg Pathol 1996; 20: and duodenal ulcer relatives. Scand J Gastroenterol 1993; 28:
1161–81. 705–8.
36 Genta RM. Review article: gastric atrophy and atrophic gas- 52 Valle J, Kekki M, Sipponen P, Ihamaki T, Siurala M. Long-term
tritis–nebulous concepts in search of a definition. Aliment course and consequences of Helicobacter pylori gastritis.
Pharmacol Ther 1998; 12(Suppl. 1): 17–23. Results of a 32-year follow-up study. Scand J Gastroenterol
37 Stolte M, Baumann K, Bethke B, Ritter M, Lauer E, Eidt H. 1996; 31: 546–50.
Active autoimmune gastritis without total atrophy of the 53 Villako K, Kekki M, Maaroos HI, et al. Chronic gastritis: pro-
glands. Z Gastroenterol 1992; 30: 729–35. gression of inflammation and atrophy in a six-year endoscopic
38 Blecker U, Lanciers S, HauSeries B, Vandenplas Y. Helicobacter follow-up of a random sample of 142 Estonian urban subjects.
pylori positivity in Belgium. Acta Gastroenterol Belg 1995; 58: Scand J Gastroenterol 1991; 186(Suppl.): 135–41.
31–4. 54 Kohlstadt IC, Antunez De Mayolo EA. Parietal cell antibodies
39 Schade SS, Cohen RJ, Conrad ME. The effect of hydrochloric among Peruvians with gastric pathology changes. Scand
acid on iron absorption. N Engl J Med 1968; 279: 672–4. J Gastroenterol 1993; 28: 973–7.
40 Kokkonen J. Parietal cell antibodies and gastric secretion in 55 Ihamaki T, Kekki M, Sipponen P, Siurala M. The sequelae and
children with diabetes mellitus. Acta Paediatr Scand 1980; course of chronic gastritis during a 30- to 34-year bioptic
69: 485–9. follow-up study. Scand J Gastroenterol 1985; 20: 485–91.
41 Hurlimann S, Dür S, Schwab P, et al. Effects of Helicobacter 56 Correa P, Haenzel W, Cuello C, et al. Gastric precancerous
pylori on gastritis, pentagastrin-stimulated gastric acid secre- process in a high risk population: cohort follow-up. Cancer
tion, and meal-stimulated plasma gastrin release in the Res 1990; 50: 4737–40.
absence of peptic ulcer disease. Am J Gastroenterol 1998; 93: 57 Lee JE, Lowy AM, Thompson WA, et al. Association of gastric
1277–85. adenocarcinoma with the HLA class II gene DQB1*0301.
42 McGowan CC, Cover TL, Blaser MJ. Helicobacter pylori and Gastroenterology 1996; 111: 426–32.
gastric acid: biological and therapeutic implications. Gastro- 58 De Block CEM, Leeuw IH, Rooman RPA, et al. Gastric parietal
enterology 1996; 110: 926–38. cell antibodies are associated with glutamic acid decarboxy-

Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 281–289

 13652036, 2002, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1046/j.1365-2036.2002.01186.x by Nat Prov Indonesia, Wiley Online Library on [28/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
H. PYLORI AND AUTOIMMUNE GASTROPATHY IN TYPE 1 DIABETES 289

lase-65 antibodies and the HLA DQA1*0501-DQB1*0301 62 d’Elios MM, Manghetti M, de Carli M, et al. T helper 1 effector
haplotype in type 1 diabetes. Diabet Med 2000; 17: 618–22. cells specific for Helicobacter pylori in the gastric antrum of
59 Sipponen P, Kekki M, Haapakoski J, Ihamäki T, Siurala M. patients with peptic ulcer disease. J Immunol 1997; 158:
Gastric cancer risk in chronic atrophic gastritis: statistical cal- 962–7.
culations of cross-sectional data. Int J Cancer 1985; 35: 173–7. 63 Engstrand L, Scheynius A, Pahlson C, Grimelius L, Schwan
60 Kuipers EJ. Review article: relationship between Helicobacter A, Gustavsson S. Association of Campylobacter pylori with
pylori, atrophic gastritis and gastric cancer. Aliment Phar- induced expression of class II transplantation antigens on
macol Ther 1998; 12(Suppl. 1): 25–36. gastric epithelial cells. Infect Immunol 1989; 57: 827–32.
61 Cover TL, Blaser MJ. Purification and characterization of the
vacuolating toxin from Helicobacter pylori. J Biol Chem 1992;
267: 10570–5.

Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 281–289