Effects of Helicobacter pylori on Gastritis, Pentagastrin-

Stimulated Gastric Acid Secretion, and Meal-Stimulated Plasma

Gastrin Release in the Absence of Peptic Ulcer Disease
S. Hurlimann, M.D., S. Du r, M.D., P. Schwab, M.D., L. Varga, M.D., L. Mazzucchelli, M.D., R. Brand, Ph.D., and
F. Halter, M.D.
Gastrointestinal Unit, Inselspital, Institute of Pathology and Department of Mathematical Statistics, University of Berne,
Berne, Switzerland
Objective: There is strong evidence accumulating that
chronic infection with Helicobacter pylori (H. pylori) in-
terferes with inhibitory pathways of the regulation of
acid secretion. The increase in maximum acid output
(MAO), and the increase in the sensitivity of the parietal
cell to gastrin commonly observed in patients suffering
from duodenal ulcer disease (DU), however, remains
largely unexplained. Insufficient evidence is available
concerning how these parameters are influenced by H.
pylori infection in patients not suffering from peptic
ulcer disease (PUD) and how they are related to H.
pylori-induced gastritis. The aim of this study was to
compare basal gastric acid secretion (BAO), MAO, and
the sensitivity of the parietal cell to gastrin in H. pylori-
positive and H. pylori-negative patients not suffering
from PUD, and to study the relationship with their in-
dividual postprandial gastrin release and the degree of
gastric antral and corpus gastritis. Methods: H. pylori
status was assessed by CLO test and histology (two
biopsies each from the antrum and the corpus) in 14 H.
pylori-positive and 16 H. pylori-negative nonulcer pa-
tients of comparable age, weight and gender. Gastritis
score was assessed by a pathologist, who was unaware of
the acid secretory data. Following determination of
BAO, the relation of pentagastrin and gastric acid se-
cretion was established with a cumulative pentagastrin
dose response curve for the dose range 0.036.0 g/kg
1
h
1
and MAO (V
max
) and pentagastrin sensitivity (ED
50
)
were determined. Basal and postprandial gastrin release
was measured by radioimmunoassay. Results: There was
a significant higher gastritis score in the H. pylori-posi-
tive compared with the H. pylori-negative subjects. The
dose response curves of the pentagastrin stimulated gas-
tric acid secretion were not different between H. pylori-
positive and H. pylori-negative groups. No correlation
was seen between the gastritis score, basal acid output
(BAO) peak acid output (PAO), maximum acid output
(MAO), ED
50
values and the plasma gastrin values.
There was, however, a considerable larger variation of
the PAO and MAO data of the H. pylori-infected sub-
jects and >50% of the respective data was above or
below the relatively low range of the respective values of
the noninfected subjects. Conclusions: H. pylori-induced
gastritis does not regularly enhance maximum acid out-
put in nonulcer patients, nor does it modify the sensitiv-
ity of the parietal cell to gastrin. H. pylori infection is
thus unlikely to be directly responsible for an increase of
these parameters in DU disease. Our data support, how-
ever, the concept that chronic H. pylori infection can
either enhance or attenuate maximum acid secretory
capacity in certain subgroups of patients. (Am J Gas-
troenterol 1998;93:12771285. 1998 by Am. Coll. of
Gastroenterology)
INTRODUCTION
Gastric hyperacidity is common in peptic ulcer disease
and particularly in duodenal ulcer disease (DU). This con-
cerns basal acid output (BAO), maximum acid output
(MAO), or peak acid output (PAO), meal-stimulated secre-
tion, and the sensitivity of the parietal cell to gastrin (GS)
(14). A considerable overlap of the data on all these
secretory parameters of DU patients with those of control
subjects has, however, long been recognized (5). The in-
creased MAO may reflect, in part an increase in the parietal
cell mass, commonly found in DU patients (6, 7). Addition-
ally, the ECL cell, whose secretion of histamine is predom-
inantly influenced by gastrin from the antral G-cell, is in-
volved (8). It is now proved beyond doubt that the infection
with the bacterium Helicobacter pylori (H. pylori) plays the
dominant role in the pathogenesis of duodenal ulcer disease
(DU) (9). Yet, only 1015% of the general population
develop DU in their lifetime (10, 11), even though 3366%
of the Western population have this infection (12, 13). There
is abundant evidence that in DU disease, both the antral
predominant gastritis regularly accompanying this disease
and the exaggerated gastrin release are generally caused by
H. pylori infection (1422). Attempts have been made to Received Nov. 21, 1997; accepted Apr. 1, 1998.
THE AMERICAN JOURNAL OF GASTROENTEROLOGY Vol. 93, No. 8, 1998
Copyright 1998 by Am. Coll. of Gastroenterology ISSN 0002-9270/98/$19.00
Published by Elsevier Science Inc. PII S0002-9270(98)00290-1
1277
link all abnormalities of DU disease to gastrin and its
established trophic effects (23). It is now well established
that the paucity of somatostatin, the principal physiologic
inhibitor of acid secretion, is one of the prime reasons for the
dysregulation of acid secretion in H. pylori infected DU
disease patients (24). This is best documented by the potent
acid stimulatory effect of the so-called gastrin releasing
peptide GRP in H. pylori-infected subjects, through its com-
bined effect on gastrin and somatostatin release. GRP-stim-
ulated acid secretion has been shown to be increased 3-fold
in H. pylori-positive subjects and 6-fold in DU patients and
has been shown to slowly normalize following H. pylori
eradication (22). Similarly, the exaggerated gastrin release
is reversible once H. pylori infection has ceased, but this
occurs long before normalization of acid secretion. There is
no indication that H. pylori eradication leads to a resolution
of the increased parietal cell mass (22), as is observed in
patients suffering from a Zollinger-Ellison syndrome within
3 to 6 months after curative resection of the gastrinoma
tumor (25). It is thus doubtful whether the increase in
parietal cell mass often observed in DU patients is caused by
H. pylori infection, and this similarly relates to the increase
in GS (26). In this context, it is of particular interest how H.
pylori infection modifies acid secretion in otherwise healthy
subjects. It was the aim of this study to compare the relation
between gastritis, gastrin release, and gastric acid parame-
ters related to the amount of functioning parietal cells
(MAO) and the sensitivity of the parietal cells to gastrin in
H. pylori-positive and H. pylori-negative subjects not af-
fected by ulcer disease.
MATERIALS AND METHODS
All patients attending our endoscopy unit and presenting
with an upper-GI endoscopy without a history of peptic
ulcer, and who had normal endoscopy findings in the esoph-
agus, stomach, and duodenum, were tested for H. pylori
infection. Three biopsies were sampled from the antrum and
corpus and one from either localization was used for the
measurement of urease activity (CLO-test), the others for
histological assessment of the degree of gastritis and micro-
scopic analysis of the H. pylori status. If they had neither
taken any acid inhibitory agents in the last 1 wk nor anti-
biotics in the last 2 months before the endoscopy, they were
invited to participate in the study and to undergo analysis of
the postprandial gastrin profile and acid secretory studies.
They were only enlisted in the study when congruent results
were obtained for absence or presence of H. pylori infection
with both the CLO-test and histology. A total of 16 H.
pylori-negative and 14 H. pylori-positive, healthy subjects
were admitted to the study.
For acid secretory studies, all subjects reported at 8 AM
after a 12-h fast. They were fitted with an indwelling venous
cannula in the forearm to facilitate intravenous infusion. A
French 16 rubber tube was positioned with the tip in the
most dependent part of the stomach. The aspiration tech-
nique and titration of the gastric juice obtained were as
described in previous studies by our group (4). The volume
of each sample was measured and an aliquot was reserved
for analysis. Acid was measured by electrometric titration to
pH 7.4, using 50 mmol/L NaOH. At the beginning, residual
gastric juice was collected and discarded. Basal acid secre-
tion was then sampled for 40-min periods (four 10-min
samples). To establish gastrin dose-response relationship,
pentagastrin (Peptavlon, ICI-Pharma) was administered by
continuous intravenous infusion over the dose range of 0.03,
0.1, 0.3, 1.0, 3.0, and 6.0 g kg
1
h
1
in consecutive
40-min periods (four 10-min samples). To allow for equi-
libration of the response, the first 10-min sample following
commencement of each dose level was discarded. Through-
out the study, extrapolation was made to acid output per
hour for basal acid (four pooled 10-min samples) and for
each infusion period of the stepwise dose-response study
(three pooled 10-min samples each).
Basal and postprandial plasma gastrin was measured by
radioimmunoassay, by an antibody that can simultaneously
detect gastrin17 and gastrin 34 (27) before and 1, 2, and 3 h
after a standard meal (two rolls of bread with butter and jam
and a cup of coffee with milk).
Mucosal biopsies were embedded in paraffin and 4 m
sections were obtained. Sections were stained with hema-
toxylin and eosin and graded for the degree of gastritis
according to the following methodology: Five morpholog-
ical parameters were assessed using the Sydney system (28,
29) in accordance with a recent study of El Omar et al. (30).
The biopsies were graded from 0 to 3 (equating to none,
mild, moderate, and severe). The five parameters were as
follows: 1) inflammation, which scored the chronic inflam-
matory infiltrate in the lamina propria; 2) activity, which
scored the acute inflammatory infiltrate. Chronic and acute
inflammatory score (1 and 2) were added together, giving a
total inflammatory score with a possible maximum of 6
(range 06); 3) atrophy, scored form 0 to 3 on the basis of
proportion of glandular loss; 4) intestinal metaplasia, scored
from 0 to 3 on whether absent or occupying less than one
third, more than one third, or more than two thirds of the
mucosa present; and 5) H. pylori colonization, scored only
when definitely recognized by morphology and graded from
0 to 3. When felt necessary special stains (modified Giemsa
and immunohistochemistry) were used in selected cases to
obtain a reliable information on H. pylori colonization. This
grading was applicable to both antral and body changes
(15).
Statistics
Demographic characteristics such as age and body
weight, values of fasting and meal-stimulated postprandial
plasma gastrin, basal acid output (BAO), and peak acid
output (PAO) in the H. pylori-negative and H. pylori-posi-
tive healthy subjects were compared using the Wilcoxon
rank-sum test. A p value of 0.05 was taken as significant.
The results of the stepwise pentagastrin dose-response curve
1278 HURLIMANN et al. AJG Vol. 93, No. 8, 1998
were analyzed by multiple linear regression analysis by
comparing the following parameters: BAO, calculated
MAO, calculated integrated response of the total amount of
acid produced during the entire pentagastrin dose response
curve (acid-AUC), and the pentagastrin ED
50
values with
the variables age, sex, weight, integrated gastrin response
(gastrin-AUC), gastric antral, and gastric corpus gastritis
score for H. pylori-positive and H. pylori-negative patients.
Additionally, the degree of variance of data collected from
H. pylori-positive and H. pylori-negative subjects was com-
pared for all of the above mentioned parameters with the
Siegel-Tukey test (31) between H. pylori-negative and H.
pylori-positive subjects. The Kendals-Tau Somers D-test
(31) was used for analysis of the correlation between gas-
tritis scores and acid secretory data (BAO and calculated
MAO values, acid-AUC), sex, age, body weight, and the
integrated postprandial gastrin response gastrin-AUC data.
The study was approved by the Ethics Committee of the
Medical Faculty of the University of Bern.
RESULTS
The demographic parameters of the 14 H. pylori-positive
and 16 H. pylori-negative healthy subjects were well
matched. There were no differences in the median age of the
H. pylori-positive group: 34 yr (range 2461) and of the H.
pylori-negative group: 27 yr (range 2651), p 0.16, and
the body weight (66.0 kg [range 4993]) vs 68.5 kg [range
5388], respectively). The female-to-male ratio was 6:8 in
the H. pylori-positive and 7:9 in the H. pylori-negative
group, respectively.
Acid secretion
The BAO values and the pentagastrin dose-response
curves of both groups are shown in Figure 1. There was no
difference in any acid secretory parameters between the H.
pylori-positive group and the H. pylori-negative group: me-
dian basal acid output was 1.94 mmol H

h
1
(range 00.05
10.9) vs 1.64 mmol H

h
1
(range 0.078.22). The median
observed PAO value in the H. pylori-positive group was
26.75 mmol H

h
1
(range 0.17556.03) vs 26.11 mmol
H

h
1
(range 20.2938.91) in the H. pylori-negative group.
The respective calculated MAO data derived from the fit of
the dose response curve were 24.11 mmol H

h
1
(range
0.2168.26) vs 27.87 mmol H

h
1
(range 22.6141.03).
There was a positive correlation between BAO and MAO
values of both groups (H. pylori-positive, p 0.0088, H.
pylori-negative, p 0.0151). The median value of the
integrated acid response during the pentagastrin dose-re-
sponse curve (acid-AUC) was 146.4 mmol H

/ 5h (range
0.5307.9 in the H. pylori-positive population vs 140.5
mmol H

/5h (range 99.5216.9) in the H. pylori-negative

population. The difference between the two groups was not
significant, but as outlined from the differences in the stan-
dard errors of each data point of the pentagastrin dose-
response curve (Fig. 1), the overall variation of the MAO
data was considerably smaller in the H. pylori-negative
subjects as compared with the H. pylori-positive subjects
(p 0.0029). There was no significant difference in pen-
tagastrin ED
50
values (H. pylori-positive subjects: median
0.289 g kg
1
h
1
, (range 0.0495.710); H. pylori-negative
subjects: median 0.553 g kg
1
h
1
(range 0.0034.840)].
Plasma gastrin
The basal and meal stimulated plasma gastrin values are
shown in Figure 2. The median value for the integrated
response (gastrin-AUC values) was 87.4 pmol/l 3h (range
9.22539.9) in the H. pylori positive subjects and 49.8
pmol/l 3h (range 23.2145.2) in the H. pylori negative
subjects (Fig. 2). The difference failed to reach significance
(see also Fig. 4).
Gastritis score
The biopsy material sampled allowed reliable determina-
tion of the total inflammatory score in the biopsies sampled
from the antral area in 14 of 14 subjects who had a positive
CLO test and in 13 of the 16 subjects who had a negative
test. This was 13 of 14 for the biopsies sampled from the
corpus area in H. pylori-positive subjects and in 16 of 16 of
the H. pylori-negative subjects. In two of the 14 patients
who had a positive CLO test, H. pylori colonization could
only be identified with certainty in biopsies sampled from
the gastric corpus. All H. pylori-positive subjects had a
FIG. 1. Cumulative pentagastrin/acid dose response curve (dose range
0.036.0 g kg
1
h
1
). There was no difference between the H. pylori-
positive and H. pylori-negative group for BAO, PAO, acid-AUC, and the
ED
50
values of the pentagastrin dose response curve.
AJG August 1998 EFFECTS OF H. PYLORI ON GASTRITIS 1279
varying degree of superficial gastritis, which was antral-
predominant in all except one 53-yr-old male patient, in
whom a mild-to-moderate pangastritis was associated with a
marked gastric corpus atrophy accompanied by virtual ana-
cidity and the highest observed gastrin value (AUC-gastrin
2539.9 pmol/L 3h). As shown in Table 1, the H. pylori-
positive population had a significantly higher combined
inflammatory and H. pylori colonization density score both
in the antrum and corpus as compared with H. pylori-
negative subjects.
No correlation was found between the parameters of the
Sydney system gastritis scores in the gastric antrum or
gastric corpus and any of the acid secretory parameters such
as BAO (data not shown) and calculated MAO (Fig. 3) and
the ED
50
values of the pentagastrin dose response curve
(data not shown).
Figure 3a clearly outlines the considerably wider scatter
of the MAO data in H. pylori-positive subjects; the values of
three subjects were above, and five were below, the range of
the H. pylori-negative subjects. The difference in calculated
MAO values between the two groups were not statistically
different, but there was a highly significant difference in the
degree of variation within each group (p 0.0029). Allo-
cation of the individual gastritis scores to each individual
calculated MAO value highlights the lack of correlation
between MAO and gastritis scores.
There was a larger, but nonsignificant difference in the
variation of the integrated plasma gastrin data (gastrin-AUC
values) in the H. pylori-positive patients, as compared with
the H. pylori-negative group (Fig. 4). Although numerically
higher values were observed in the H. pylori-positive group,
the differences failed to reach significance. There was no
significant correlation between integrated gastrin values and
the combined inflammatory score and score for H. pylori
colonization density.
DISCUSSION
This study performed on 30 predominantly young to
middle-aged nonulcer patients, who were well matched for
age and sex, confirmed that gastritis regularly accompanies
chronic H. pylori infection. However, all acid secretory
parameters such as BAO, MAO, PAO, or GS showed, in
agreement with other studies, no significant differences
between H. pylori-positive and H. pylori-negative subjects
not suffering from DU disease (3235). Similarly, the ab-
sence of a correlation between the degree of gastritis and the
acid secretory parameters is in line with several previous
observations (14, 36, 37). Of particular interest, however,
was the considerably larger variation of MAO data of H.
pylori-infected subjects as observed in this study, whereby
50% had values above and below the range of noninfected
subjects. Some previous studies supply putative explana-
tions for this phenomenon. The notoriously large variation
characterizing gastric acid secretory data (1, 2, 5) points
toward the possibility that a sampling error could be a
confounding factor. A large between and within patient
variation is a particularly well recognized problem for BAO
data (38) and the overlap between health and disease for this
parameter has long been established before the role of H.
pylori infection in gastritis and PUD had been noticed (1, 2).
There is convincing evidence that this can be reduced in DU
patients through H. pylori eradication (22). In a well de-
signed longitudinal study, BAO was modestly higher in H.
pylori-infected patients not suffering from DU disease (22).
In that study, the difference was significant both in com-
FIG. 2. The basal and postprandial gastrin response, as measured during
3 h. At each data point the values were numerically higher in H. pylori-
infected subjects, but the difference failed to reach significance.
TABLE 1
Gastritis Score
1280 HURLIMANN et al. AJG Vol. 93, No. 8, 1998
parison with non-H. pylori-infected healthy control subjects
and DU patients, in whom the increase was considerably
more pronounced; however, some 60% of DU disease pa-
tients showed an overlap of their BAO data with those of
both H. pylori-infected and non-infected healthy control
subjects. H. pylori eradication failed, however, in agreement
with a previous observation (39), to decrease BAO in oth-
erwise healthy, previously H. pylori-infected patients (22).
By contrast, one study reported significantly lower BAO
values in H. pylori-infected volunteers compared with un-
FIG. 3. Gastritis scores and calculated maximum acid output (MAO): In each H. pylori-positive or H. pylori-negative subject, the individual combined
inflammatory score (A) as well as the total score for H. pylori colonization density (B) is allocated to the MAO value as calculated from the pentagastrin
dose-response curve. There was no significant difference between the MAO data: median 24.11 mmol H

h
1
(range 0.2168.26) vs 27.87 mmol H

h
1
(range 22.6141.03), but the difference in the degree of variation within each group was significantly different (p 0.0029).
FIG. 4. Gastritis scores and integrated postprandial gastrin release: In each H. pylori-positive or H. pylori-negative subject the individual combined
inflammatory score (A) as well as the total score for H. pylori colonization density (B) is allocated to the integrated postprandial gastrin value. There was
no significant difference between the gastrin data of the two groups (median 87.4 pmol/L 3h [range 9.22539.9] in the H. pylori-positive subjects and 49.8
pmol/L 3h [range 23.2145.2]) in the H. pylori-negative subjects.
AJG August 1998 EFFECTS OF H. PYLORI ON GASTRITIS 1281
infected control subjects (40). The large variation may well
mask a tendency of H. pylori infection to modify basal acid
secretion. It is, however, evident from the result of the
above-quoted study (22) that a large proportion of otherwise
healthy H. pylori-infected subjects and even DU patients do
not have an elevated basal acid secretion and, in our study,
the degree of variation of BAO-data was similarly large
independent on whether or not the subjects were infected by
H. pylori.
Several studies found no change in MAO or PAO data
after H. pylori eradication (19, 22, 32, 39). Some notewor-
thy observations nevertheless suggest that chronic H. pylori
infection may not be fully devoid of an influence on MAO
or PAO. A considerable proportion of H. pylori-infected
patients not suffering from DU disease have been found to
have high acid secretory values that overlap with those of
DU patients (41). Harris et al. (42) demonstrated a consid-
erable fall in pentagastrin-stimulated PAO in DU patients 6
months after H. pylori eradication. Their observation was,
however, based on seven patients only and the reduction
was substantial in only five subjects; but this was in line
with earlier observations of Moss et al. (43), who found a
decrease in MAO values 1 yr after H. pylori eradication. By
contrast, El-Omar et al. (30) recently reported reversal of
total or near total anacidity through H. pylori eradication in
a subgroup of nonulcer patients with corpus predominant
gastritis. The latter finding helps to appreciate better the
observations made by Chandrakumaran et al. (45) that
MAO was lower in both H. pylori-infected DU and non-DU
patients in comparison with the data of respective control
groups. This is in line with earlier observations made con-
cerning male patients by Katelaris et al. (46). These appar-
ently contradictory findings make it likely that H. pylori
infection can lead to either an increase or decrease of MAO
values in subgroups of patients and thus may lead to dis-
cordant results in studies with small patient samples. On the
other hand, the disparate effects that H. pylori eradication
can exert on MAO may compensate each other in studies
with larger samples and thus obscure the inconsistent
changes that chronic H. pylori infection may exert on max-
imum acid secretory capacity of the stomach. Circumstantial
evidence for such a hypothesis is supplied by our observa-
tion that MAO data showed a significantly larger variation
in the H. pylori-infected subjects, whereby more than 50%
of the values of the H. pylori-positive subjects were either
below or above the range of the noninfected subjects. In line
with all other studies our data show, however, unequivo-
cally that many H. pylori-infected subjects have normal
MAO data and that this even relates in our population to the
basal and postprandial gastrin values.
There are at least two possible mechanisms through
which H. pylori infection could increase or decrease maxi-
mum acid output. This is either through a change in the
number of parietal cells or ECL cells or through a modifi-
cation of the function of these highly specialized cells, by H.
pylori toxins or by inflammatory mediators. Even though
there is no unequivocal evidence available that H. pylori
infection can directly enhance the parietal cell or ECL mass,
it is, however, well established that a subset of H. pylori-
infected, usually asymptomatic, patients develop a special
form of chronic multifocal atrophic gastritis (MAG type,
according to the classification of Correa [47]) involving
both the mucosa of the gastric antrum and corpus, leading to
a gradual reduction of epithelial glandular structures, with
permanent loss of the parietal cells, associated with hypo- or
even total achlorhydria (16, 17, 30, 44, 48). The proportion
of such patients appears to be higher in older subjects and
relatively large in the presence of high H. pylori prevalence
(37), and it is less frequently observed in populations with
a lower H. pylori prevalence (14). In our study performed in
a country where H. pylori prevalence is 30% (49), only
one of 14 H. pylori-positive subjects fell into this category
and, in contrast to previous observations (30, 50), the ma-
jority of our nonulcer patients had antral predominant gas-
tritis, the type of gastritis often encountered in DU disease
(40, 41). It is thus more likely that secretory products of H.
pylori (51) or the toxin-induced inflammatory response (52)
did lead to either an enhancement or attenuation in the
maximum acid secretory capacity of the parietal cell in some
of our patients. There is convincing evidence available that
subtotal to total loss of acid secretion can occur after acute
H. pylori infection (5355). Despite persistence of H. pylori
infection and histological gastritis, acute hypochlorhydria
generally resolves spontaneously (16). The mechanisms
through which H. pylori infection modifies the acid output
is not yet fully elucidated, but the rapid partial reversibility
of anacidity observed in patients with advanced gastritis
(30) clearly underlines the existence of the inhibition of the
parietal cell through H. pylori toxins or its inflammatory
mediators, even in patients with long-standing chronic in-
fection.
There is no indication from this and the majority of
previous studies that H. pylori enhances the sensitivity of
the parietal cell to gastrin in non-DU patients (35, 39, 56)
and it is noteworthy that several previous studies using
histamine and pentagastrin could not confirm a difference in
sensitivity to stimulation between DUs and control subjects
(5658). In this context, it is of particular interest that in a
recent study of the McColl group of Glasgow (44), volun-
teers not infected with H. pylori had an equal sensitivity to
gastrin as DU patients, whereas this was 2-fold lower in H.
pylori infected healthy volunteers, thus indicating that the
difference in gastrin sensitivity between DU and non-DU
patients may not be due to an increase in the sensitivity of
ulcer patients but rather to a reduction of this parameter in
H. pylori infected, nonulcer patients. Their technology used
to establish the gastrin sensitivity differed, however, from
that used in this and many previous studies and it was more
sophisticated. They measured plasma gastrin before and
repeatedly during the gastrin infusion in each individual
with an antibody that detects not only endogenous gastrin 17
but also gastrin 34. Plasma gastrin was plotted in each
1282 HURLIMANN et al. AJG Vol. 93, No. 8, 1998
individual against acid output at each measuring point of the
dose-response curve. By contrast the pentagastrin concen-
tration of the infusate was the standard in our study. The
advantage of their method is that it corrects for the overall
increase of basal gastrin in H. pylori-infected subjects. It can
thus not be fully excluded that our technique overestimated
gastrin sensitvity in the H. pylori-infected subjects who in
many previous studies have been shown to have increased
basal gastrin levels (18, 19, 21, 23, 60). Because, in the
present study only few H. pylori-positive individuals had
basal gastrin levels above the range observed in H. pylori-
negative subjects it is unlikely that this difference in meth-
odology fully explains the conflict between the results of the
two studies. It is more likely that differences in the preva-
lence of subforms of gastritis were responsible for the dis-
cordant results, and the Glasgow group has putatively linked
(44) the impairment of gastrin sensitivity in their H. pylori-
positive nonulcer patients to atrophic gastritis within the
gastric corpus, which previously had been observed in non-
ulcer patients (30, 50). The absence of a decreased gastrin
sensitivity in our H. pylori-infected nonulcer patients, who
all except one only had a mild gastric corpus gastritis,
supplies some indirect support for this hypothesis. Simi-
larly, the only mild and insignificant elevation of gastrin as
observed in our H. pylori-infected subjects may also be
related to the low grade of gastric corpus gastritis, because
in nonulcer patients excessive gastrin release is especially
favored by atrophy of the corpus mucosa (8).
Difference in virulence of H. pylori strains has been
considered as a putative explanation as to why only a
minority of DU patients develop peptic ulcers, inasmuch as
it has been shown that patients who develop DU disease are
more likely to be infected with the more aggressive, CagA-
positive strains of H. pylori, which are associated with
enhanced inflammatory responses (61). These strains, how-
ever, have been shown not to disturb gastric function to a
greater degree than do CagA-negative strains (61), but they
are likely to cause greater mucosal damage and hence
weaken the ability of the mucosa to withstand the increased
levels of acid induced by H. pylori infection. In this context
it is noteworthy that the risk for developing gastric cancer is
approximately 4-fold less in DU patients than in the general
population (62), whereas it is considerably higher in patients
with MAG-type gastritis (14, 37).
It is thus more likely that host-related mechanisms, pos-
sibly of genetic origin, are causally related to the fact that
only a small proportion of H. pylori-positive subjects de-
velop significant pathology such as PUD or gastric carci-
noma. Theoretically, the tendency toward a lower or higher
MAO could help to differentiate the two groups early, as
DU disease is expected to occur in subjects with a high
MAO, whereas a low acid output would favor the later
occurrence of a gastric malignancy.
Our data supply some indirect evidence that the notori-
ously large overlap between MAO or PAO data of DU
patients and healthy subjects, which intrigued many ob-
servers before the pathogenic role of H. pylori had been
established (1, 2, 5), may be related, at least in part, to the
fact that many of the healthy subjects were infected by H.
pylori. The significantly greater range of the MAO data
among H. pylori-infected nonulcer patients points toward
the necessity to establish new control data in large groups of
subjects who, in the absence of H. pylori infection, have a
really healthy mucosa both in the gastric antrum and corpus.
It would be of great value to extend such studies to all other
acid secretory parameters and also to review the time-
honored concept of the direct correlation between MAO
data and the parietal cell mass (6).
In conclusion, the results of this study suggest that a large
proportion of H. pylori-infected nonulcer patients have no
abnormality in their gastrin release and in conventional acid
secretory parameters such as BAO, MAO, and sensitivity of
the parietal cell to gastrin, despite various degrees of antral
predominant gastritis. However, maximum acid output of H.
pylori-infected nonulcer subjects showed a considerably
larger variation than in noninfected subjects. Furthermore,
the relatively equal deviation of the data of infected subjects
above and below the range of the healthy control subjects
can be reconciled with previous discordant observations that
H. pylori infection can induce either an increase or a de-
crease of maximum acid secretory capacity of the parietal
cells in subgroups of patients. It appears unlikely that a
difference in the parietal cell mass is the exclusive reason
for the larger variation of MAO data of H. pylori-infected
subjects.
ACKNOWLEDGMENT
We would like to thank Prof. I.N. Marks, Cape Town, and
Prof. K.E.L. McColl for valuable suggestions with the
manuscript.
Reprint requests and correspondence: F. Halter, M.D., F.R.C.P., Edin,
Visiting Professor, Division of Gastroenterology, UCI Medical Center, 101
The City Drive, Bl. Dg 53, Route 81, Orange, CA 92868.
REFERENCES
1. Wormsley KG, Grossman MI. Maximal histalog test in control sub-
jects and patients with peptic ulcer. Gut 1965;6:42735.
2. Baron JH. Pathophysiology of gastric acid secretion. In: Domschke W,
Wormsley KG, eds. Magen-und Magenkrankheiten. Stuttgart: Georg
Thieme Verlag, 1981:13149.
3. Isenberg JI, Grossman MI, Maxwell V, et al. Increased sensitivity to
stimulation of acid secretion by pentagastrin in duodenal ulcer. J Clin
Invest 1975;55:3307.
4. Halter F, Bangerter U, Haecki WH, et al. Sensitivity of the parietal cell
to pentagastrin in health and duodenal ulcer disease: A reappraisal.
Scand J Gastroenterol 1982;17:53944.
5. Lam SK. Pathogenesis and pathophysiology of duodenal ulcer. In:
Isenberg JI, Johansson C, eds. Clinics in gastroenterology, Vol. 13.
London: W.B. Saunders, 1984:44772.
6. Card WI, Marks IN. The relationship between the acid output of the
stomach following maximal histamine stimulation and parietal cell
mass. Clin Sci 1960;19:14763.
7. Cox AJ. Stomach size and its relation to chronic peptic ulcer. AMA
Arch Pathol 1952;54:407.
8. Modlin IM, Tang LH. The gastric enterochromaffin-like cell: An
enigmatic cellular link. Gastroenterology 1996;111:783810.
AJG August 1998 EFFECTS OF H. PYLORI ON GASTRITIS 1283
9. Peura DA, Graham DY. Helicobacter pylori: Consensus reached:
Peptic ulcer is on the way of becoming and historic disease [editorial].
Am J Gastroenterol 1989;89:11379.
10. Watkinson G. The incidence of chronic peptic ulcer found at necropsy.
Gut 1960;1:1431.
11. Levi IS, De la Fuente AA. A post mortem study of gastric and
duodenal peptic lesions. Gut 1963;4:34959.
12. Graham D, Klein PD, Opekun AR, et al. Effect of age on frequency of
active Campylobacter pylori infection diagnosed by the
13
C-urea
breath test in normal subjects and patients with peptic ulcer. J Infect
Dis 1988;157:77780.
13. McDonagh TA, Woodward M, Morrison C, et al. Lack of independent
association of H. pylori and coronary heart disease. Gut 1996;38(suppl
1):A1.
14. Sipponen P, Kekki M, Seppala K, et al. The relationship between
chronic gastritis and gastric acid secretion. Aliment Pharmacol Ther
1996;10 (suppl 1):10318.
15. Maaroos HI, Kekki M, Villako K, et al. The occurrence and extent of
Helicobacter pylori colonization and antral and body gastritis profiles
in an Estonian population sample. Scand J Gastroenterol 1990;25:
10107.
16. McGowan CC, Cover TL, Blaser MJ. Helicobacter pylori and gastric
acid: Biological and therapeutic implications. Gastroenterology 1996;
10:92638.
17. Rubin CE. Are there three types of Helicobacter pylori gastritis?
Gastroenterology 1997;112:210810.
18. Oderda G, Varia D, Holton J, et al. Amoxycillin plus tinidazle for
Campylolbacter pylori gastritis in duodenal ulcer subjects: Assessment
by serum IgG antibody, pepsionogen I, and gastrin levels. Lancet
1989;1:6902.
19. Levi S, Beardshall K, Swift I, et al. Antral Helicobacter pylori,
hypergastrinaemia, and duodenal ulcers: Effect of eradicating the or-
ganism. Br Med J 1989;299:15045.
20. Prewett EJ, Smith JTL, Nwokolo CU. Eradication of Helicobacter
pylori abolishes 24-hour hypergastrinaemia: A prospective study in
healthy subjects. Aliment Pharmacol Ther 1991;5:28390.
21. Graham DY, Opekum A, Lew GM, et al. Ablation of exaggerated
meal-stimulated gastrin release in duodenal ulcer patients after clear-
ance of Helicobacter (Campylobacter) pylori infection. Am J Gastro-
enterol 1990;85:3948.
22. El-Omar EM, Penman, ID, Ardill JE, et al. Helicobacter pylori infec-
tion and abnormalities of acid secretion in patients with duodenal ulcer.
Gastroenterology 1995;109:68191.
23. Levi S, Beardshall K, Playford R, et al. Campylobacter pylori and
duodenal ulcers: The gastrin link. Lancet 1989;1:11678.
24. Calam J. Helicobacter pylori, acid and gastrin. Eur J Gastroenterol
Hepataol 1995;7:3107.
25. Pisegna JR, Norton JA, Slimak GG, et al. Effects of curative gastri-
noma resection on gastric secretory function and antisecretory drug
requirement in the Zollinger-Ellison syndrome. Gastroenterology
1992;102:766778.
26. McColl KEL, El-Omar E. Helicobacter pylori and disturbance of
gastric function associated with duodenal ulcer disease and gastric
cancer. Scand J Gastroenterol 1996;31(suppl 215):327.
27. Witzel L, Heitz P, Halter F, et al. Effects of prolonged administration
of metiamide on serum gastrin, gastrin content of the antrum and
gastric corpus and G-cell population in the rat. Gastroenterology 1979;
76:9459.
28. Price AB. The Sydney system: Histological division. J Gastroenterol
Hepatol 1991;6:20922.
29. Dixon MF, Genta RM, Yardley JH, et al. Classification and grading of
gastritis: The updated Sydney system. Am J Surg Pathol 1996;20:
116181.
30. El-Omar E, Oien K, El-Nuhjumi, et al. Helicobacter pylori infection
and chronic gastric acid hyposecretion. Gastroenterology 1997;113:
1524.
31. Lehmann EL. Nonparametrics: Statistical methods based on ranks.
New York: McGraw-Hill, 1977.
32. Monbriand JR, Appleman HD, Cotner EK, et al. Treatment of Campy-
lobacter pylori does not alter gastric acid secretion. Am J Gastroenterol
1989;84:15136.
33. Goldschmiedt M, Karnes WE, Feldman M. Relationship between
Helicobacter pylori (H pylori) and gastric acid secretion/serum gastrin
concentrations in healthy human. Gastroenterology 1990;98:A50 (Ab-
stract).
34. Haruma K, Kawaguchi H, Yoshihara M, et al. Relationship between
Helicobacter pylori infection and gastric acid secretion in young
healthy subjects. J Clin Gastroenterol 1994;19:202.
35. Kovacs TOG, Sytnik B, Walsh JH. Helicobacter pylori infection does
not produce increased gastrin-stimulated acid secretion. Gastroenter-
ology 1994;106:4:A112.
36. Testino G, Coraggia M, Cheli R. Stimulated gastrinemia, parietal cell
mass and stimulated acid secretion in autonomous chronic gastritis:
Helicobacter pylori influence. Hepato-Gastroenterology 1995;42:650
4.
37. Ruiz B, Correa P, Fontham ETH, et al. Antral atrophy, Helicobacter
pylori colonization, and gastric pH. Am J Clin Pathol 1996;105:96
101.
38. Makhlouf GM, McManus JPA, Card WI. A comparative study of the
effects of gastrin, histamine, histalog, and mechothane on the secretory
capacity of the human stomach in two normal subjects over 20 months.
Gut 1965;6:52534.
39. Moss SF, Calam J: Acid secretion and sensitivity to gastrin in patients
with duodenal ulcer: effect of eradication of Helicobacter pylori. Gut
1993;34:88892.
40. Peterson WL, Barnett CC. Evans DJ, et al. Acid secretion and serum
gastrin in normal subjects and patients with duodenal ulcer. Am J
Gastroenterol 1993;88:203843.
41. El-Omar E, Penman J, Ardill JES, et al. A substantial proportion of
non-ulcer dyspepsia patients have the same abnormality of acid secre-
tion as duodenal ulcer patients. Gut 1995;36:5348.
42. Harris AW, Gummett PA, Misiewicz, et al. Eradication of Helicobac-
ter pylori in patients with duodenal ulcer lowers basal and peak acid
outputs to gastrin releasing peptide and pentagastrin. Gut 1996;38:
6637.
43. Moss SF, Ayesu K, Calam J. Gastrin and gastric acid output 1 year
after eradication of Helicobacter pylori in duodenal ulcer patients.
Regul Pept 1991;35:251.
44. Gillen D, El-Omar EM, Wirz AA, et al. The acid response to gastrin
distinguishes duodenal ulcer patients from helicobacter pylori-infected
healthy subjects. Gastroenterology 1998;114:507.
45. Chandrakumaran K, Varia D, Hobsley M. Duodenal ulcer, Helicobac-
ter pylori and gastric secretion. Gut 1994;35:10336.
46. Katelaris PH, Seow F, Lin BPC, et al. Effect of age, Helicobacter
pylori infection, and gastritis with atrophy on serum gastrin and gastric
acid secretion in healthy men. Gut 1993;34:10327.
47. Correa P. Chronic gastritis. In: Whitehead R, ed. Gastrointestinal
and oesophageal pathology. London: Churchill Livingstone, 1989:
40220.
48. Correa P. Human gastric carcinogenesis: A multistep and multifacto-
rial process: First American cancer society award lecture on cancer
epidemiology and prevention. Cancer Res 1992;52;673540.
49. Brignoli R, Merki H, Miazza B. Endoskopische Befunde bei freiwil-
ligen und dyspeptischen Patienten. Schweiz Med Wochenschr 1994;
124:12408.
50. Eidt S, Stolte M. Differences between Helicobacter pylori associated
gastritis in patients with duodenal ulcer, pyloric ulcer, other gastric
ulcer, and gastritis without ulcer. In: Malfertheiner P, Ditschuneit H,
ed., Helicobacter pylori, gastritis and peptic ulcer, 22836. Berlin,
Heidelberg: Springer-Verlag.
51. Cave DR. Helicobacter pylori and its interrelation with chief and
parietal cells. Yale J Biol Med 1996;96:918.
52. Rademaker W, Hunt RH. Helicobacter pylori and gastric acid
secretion. The ulcer link? Scand J Gastroenterol 1991:26(suppl):
187:717.
53. Marshall BJ, Armstrong JA, McGechie DB, et al. Attempt to fulfill
Kochs postulates for pyloric campylobacter. Med J Aust 1985;142:
4369.
54. Morris A, Nicholson G: Ingestion of Campylobacter pyloridis causes
gastritis and raised fasting gastric pH. Am J Gastroenterol 1987;82:
1929.
55. Graham DY, Alpert LC, Smith JL, et al. Latrogenic Campylobacter
pylori infection is a cause of epidemic achlorhydria. Am J Gastroen-
terol 1988;83:97480.
56. Chittajallu RS, Howie CA, McColl KEL. Effect of Helicobacter pylori
on parietal cell sensitivity to pentagastrin in duodenal ulcer subjects.
Scand J Gastroenterol 1992;27:85762.
1284 HURLIMANN et al. AJG Vol. 93, No. 8, 1998
57. Roxburgh JC, Whitfield P, Hobsley M. Parietal cell sensitivity in
humans: Control and duodenal ulcer subjects, smokers and non-smok-
ers. Eur J Gastroenterol Hepatol 1994;6:23540.
58. Hunt JN, Kay AW. The nature of gastric hypersecretion of acid in
patients with duodenal ulcer. Br Med J 1954;2:14446.
59. Wormsley KG, Mahoney MP Parietal cell responsiveness in duodenal
ulcer. Br Med J 1967;1:278.
60. McColl KEL, Fullarton GM, El-Nujumi AM, et al. Lowered gastrin
and gastric acidity after eradication of Campylobacter pylori in duo-
denal ulcer. Lancet 1989;ii:499500.
61. McColl KEL, El-Omar E, El-Nujumi A, et al. H pylori-induced dis-
turbance of acid secretion is unrelated to bacterial CagA status. Gut
1995;36(suppl)1:A11.
62. Hansson LE, Nyren O, Hsing AW, et al. The risk of stomach cancer in
patients with gastric or duodenal ulcer disease. N Engl J Med 1996;
335:2429.
AJG August 1998 EFFECTS OF H. PYLORI ON GASTRITIS 1285