Why guidelines are required for the treatment of Helicobacter
pylori infection in children
Philip M. Sherman, MD
Richard H. Hunt, MD
Clin Invest Med 1996; 19 (5) : 362-7.
[résumé]
Dr. Sherman is in the Division of Gastroenterology and Nutrition,
Hospital for Sick Children, and the departments of Paediatrics
and Microbiology, University of Toronto, Toronto, Ont., and Dr.
Hunt is in the Division of Gastroenterology, McMaster University
Medical Centre, Hamilton, Ont.
Correspondence to: Dr. Philip M. Sherman, Rm. 8411, Hospital for
Sick Children, 555 University Ave., Toronto ON M5G 1X8; fax:
416 813-6531
Helicobacter pylori is firmly established as a human pathogen;
it fulfils all of Koch's postulates as the infectious agent causing
chronic, active (type B) gastritis. Infection is strongly associated
with duodenal and gastric ulcer. Recently, gastric mucosal-associated
lymphoid tissue lymphoma has been successfully treated by curing
H. pylori infection. Because of the evidence that the organism
causes chronic gastritis and an increased risk of gastric cancer,
it has been classified as a category I carcinogen by the World
Health Organization. However, the overwhelming majority of people
infected have no symptoms. Current eradication therapy is not
ideal; there are treatment failures and substantial side effects.
As a result, therapy should be reserved for people with clinical
symptoms and complications. The infection, if present, should
be treated in patients who have endoscopic evidence of mucosal
ulcers in the stomach or duodenum. Current evidence does not support
treating the infection to prevent gastric carcinogenesis or to
alleviate symptoms of abdominal discomfort in the absence of peptic
ulcers.
Il est bien établi que l'Hélicobacter pylori
est un pathogène humain, qui répond à tous
les postulats de Koch en tant qu'agent microbien causant la gastrite
chronique active (de type B). Cette infection est fortement associée
à l'ulcère duodénal et gastrique. Un lymphome
du tissu lymphoïde associé à la muqueuse gastrique
a été récemment traité avec succès
en éliminant l'infection à H. pylori. L'Organisation
mondiale de la santé a classifié H. pylori
comme un carcinogène de catégorie 1, puisque ce
micro-organisme cause une gastrite chronique et est associé
à un risque accru de cancer gastrique. Par contre, la grande
majorité des sujets infectés sont asymptomatiques.
Le traitement actuel d'éradication n'est pas idéal,
puisqu'il y a des échecs thérapeutiques ainsi que
des effets secondaires substantiels. En conséquence, le
traitement devrait être réservé aux sujets
symptomatiques et avec des complications. Si l'infection est présente,
elle doit être traitée chez les sujets chez lesquels
l'endoscopie démontre des ulcères muqueux à
l'estomac ou le duodénum. À l'heure actuelle, les
données ne justifient pas le traitement de l'infection
à H. pylori dans le but de prévenir le cancer
gastrique ou de supprimer les symptômes abdominaux en l'absence
d'ulcère peptique.
[Table of contents]
Helicobacter pylori is an established human pathogen that
fulfils each of Koch's postulates as a cause of chronic, active
(type B) gastritis.[1] Whether H. pylori is a cause of
peptic ulcers is not as well established. Nevertheless, the evidence
of causation is compelling.[2] There is a strong epidemiologic
association between H. pylori infection and both duodenal
ulcers (90% to 95%) and gastric ulcers (60% to 80%) in the absence
of concurrent ingestion of nonsteroidal anti-inflammatory drugs
(NSAIDs). The most convincing evidence of a role for H. pylori
in peptic ulcers is provided by studies that show that ulcer recurrence
is reduced from around 70% or 80% of patients not treated to 2.6%
of those treated in the year after eradication of the infection
from the gastric antrum.[3]
[Table of contents]
The precise mechanism by which H. pylori infection causes
mucosal ulcers in the gastroduodenal region remains unknown. It
is clear that bacterial heterogeneity and host factors are both
important in determining the clinical outcome of initial infection.
Gastric metaplasia in the duodenum and resulting H. pylori
colonization of this site may predispose to duodenal ulcers. In
fact, gastric metaplasia is more commonly found in patients with
a fasting gastric juice of a pH of less than 2.5.[4] The organism
may cause not only gastritis but also mucosal inflammation in
the duodenum. Altered levels of gastrin and somatostatin in the
gastric mucosa of subjects with H. pylori infection may
result from a repertoire of mediators released in response to
the infection. There is an increase in both basal and meal-stimulated
gastrin, and the area under the curve for gastrin is increased
throughout the entire day. Since gastrin is a secretagogue for
acid and is also trophic to the parietal cell, it may lead to
an increase in the parietal cell mass and an increased production
of acid by parietal cells. Moreover, H. pylori infection
induces the release of both chemokines (such as interleukin-8)
and cytokines (interleukin-6) from infected gastric epithelial
cells, setting the stage for inflammatory injury to gastroduodenal
mucosa. Previous consideration of a genetic predisposition to
peptic ulcers (thought to be caused by autosomal dominant hyperpepsinogenemia
I and defects in bicarbonate production in the proximal duodenum)
need to be reviewed and revised in the context of this previously
unrecognized gastric bacterial pathogen.
Recent findings raise the possibility that only certain H.
pylori strains are ulcerogenic. It is suggested that the toxin
production and a cytotoxin-associated outer membrane protein found
in certain strains are indirect markers of the H. pylori
strains that induce peptic ulcers. Most studies conducted in Western
countries suggest that the strains of H. pylori that express
the Cag A protein are more virulent and are associated with more
severe gastritis and with peptic ulcer. Recent data suggest that
the strains that produce both the Cag A and the Vac A proteins
are associated with peptic ulcer. Vac A is directly ulcerogenic
in an animal model.[5] DNA hybridization studies also suggest
that ulcer-associated H. pylori isolates are related.
As I mentioned earlier, accumulating evidence points to H.
pylori infection of the stomach as the pathogenic cause of
most cases of peptic ulcers.[6,7] This conclusion has resulted
in a major reassessment of previously held notions about the cause
of mucosal ulcers in the stomach and duodenum. It also brings
forward the promise of a major advance in therapy, because treatment
of H. pylori infection cures most cases of peptic ulcers.[2,8]
Epidemiologic evidence indicates that H. pylori infection
is a risk factor for a variety of types of gastric cancer, including
carcinoma and mucosal-associated lymphoid tissue (MALT) lymphoma.
Acquisition of infection in childhood appears to be a critical
risk factor for neoplastic complications. A recent report from
China indicates that H. pylori infection occurs earlier
in regions with a high prevalence of gastric cancer, and that
a greater than usual rate of gastric atrophy is seen in children
in these regions.[9] Type III intestinal metaplasia in the gastric
mucosa is a known pathological marker for the risk for gastric
cancer, although type I intestinal metaplasia does not carry this
risk and is more commonly seen in association with H. pylori
infection.[10] Eradication of H. pylori infection has been
reported to cause regression of intestinal metaplasia and to cure
MALT lymphoma. In an animal model, infection with a related Helicobacter
species, in conjunction with exposure to a chemical carcinogen,
results in an increased frequency of tumours in the infected and
inflamed stomach.[11] However, the reason why gastric cancer does
not develop in most infected humans is unclear, although the preponderance
of type I intestinal metaplasia seen in most cases of H. pylori
infection may be one explanation. Similarly, it has not been established
whether H. pylori eradication prevents gastric lymphoma
and adenocarcinoma, but it is unlikely that a study would be carried
out to determine this because of the very long time periods involved.
A provocative and very critical editorial[12] recently questioned
the validity of treating H. pylori, given that not all
of Koch's postulates have been fulfilled to establish the organism
as a human pathogen, at least not as a pathogen causing peptic
ulcers and gastric cancer. Moreover, current treatment options
are subject to several limitations: failure to eradicate H.
pylori because of bacterial resistance to the drugs employed,
poor patient compliance with the combination therapy, and an unacceptable
rate of side effects.[13] Clearly, novel forms of safe and effective
prevention (e.g., a vaccine) or alternative therapy (e.g., nutritional
supplements)[14] need to be developed.
Critical reviews do not advocate therapy for all patients with
H. pylori infection.[2,13,15] The National Institutes of
Health consensus conference held in 1994 advocated treatment for
all patients who have an H. pylori infection and (1) a
peptic ulcer or (2) a known history of peptic ulcer for which
they are taking long-term maintenance therapy.[2] Recently, at
least one editorial has advocated an aggressive approach to H.
pylori infection in children.[16] A trial of treatment was
considered reasonable even in patients with only recurrent abdominal
pain. However, a consideration of the spectrum of disease is required
in defining indications for therapy.
I now proceed to an assessment of the strength of the data supporting
a cause-and-effect relation between H. pylori infection
and chronic, active gastritis, duodenal and gastric ulcers and
gastric cancer, because these data are important in judging the
merits of initiating antibacterial therapy. Any benefits must
be balanced, of course, against the costs, including side effects,
associated with currently available treatment options.
[Table of contents]
H. pylori infection is now the accepted cause of chronic,
active (type B) gastritis, since it fulfils all of Koch's postulates
as the causal agent in this condition.[1] The bacteria cause mucosal
inflammation in the previously unaffected antrum in animal models;
they have also caused gastritis in several well-documented accidental
infections and in two human volunteers.[17] Therefore, a recent
suggestion that H. pylori is simply an opportunistic infection
is not valid.[12] The question remains, however, whether this
human infection requires treatment in every affected patient.
Cross-section evaluations of the frequency of infection in given
populations have shown that most people with an H. pylori
infection are asymptomatic.[1,2] Current evidence suggests that
most people remain free of sequelae of the infection although,
without treatment, the infection appears to be lifelong. People
with H. pylori infection who later have peptic ulcers and
gastric cancer should be considered as special targets for intervention;
for them, eradication of the organism could prevent or cure disease.
However, current knowledge does not permit an accurate identification
of the subset of asymptomatic people with H. pylori infection
who will go on to have complications of the resulting gastritis.
Some authorities recommend the initiation of therapy in all children
with an infection, but current treatment options limit the appeal
of this proposal.
[Table of contents]
Not all of Koch's postulates have been fulfilled to conclusively
establish H. pylori infection and the resulting gastritis
as a cause of mucosal ulcers in the duodenum. The absence of documented
duodenal ulcers in human volunteers with the infection and in
animals infected as part of challenge studies is a significant
deficiency in our ability to establish causation. Nonetheless,
there is convincing evidence that H. pylori infection is
indeed related to duodenal ulcers. The most compelling evidence
is the change in the natural history of duodenal ulcers with eradication
of the organism. The rate of ulcer recurrence is reduced from
77% after 1 year of follow-up when ulcers are healed with acid-suppressing
agents (which do not affect H. pylori status) to less than
3% when H. pylori infection is eradicated (p <
0.0001).[3] This change in the recurrence rate continues for at
least 7 years after therapy (the longest reported follow-up period
to date) and is evident in both adults and children with duodenal
ulcers.[6] The change in the natural history of duodenal ulcers
is also clinically relevant because it dramatically alters the
rate of recurrence of intestinal hemorrhage and is markedly cost-effective
when compared with ulcer-healing regimens alone.[8] Therefore,
treatment of patients with an H. pylori infection and endoscopically
proven duodenal ulcers is currently recommended.[2,13] Moreover,
this approach has strong implications for a marked reduction in
the health care costs of duodenal ulcers.[18]
[Table of contents]
Gastric ulcers
In adults, in the absence of concurrent use of NSAIDs, rates of
gastric ulcer recurrence are reduced from 50% at 1-year follow-up
to less than 3% (p < 0.001) when H. pylori is
eradicated. A recent report of gastric ulcers developing after
H. pylori infection of gnotobiotic piglets also provides
support for the role of this infection in ulcers of the gastric
mucosa.[19] A consensus conference sponsored by the National Institutes
of Health recommended treatment of H. pylori infection
in cases of gastric ulcers even if there is a history of recent
NSAID ingestion.[2] The issue is contentious, however, because
most ulcers of the gastric mucosa involve NSAID or corticosteroid
use, without H. pylori infection.[6,7]
[Table of contents]
Whether H. pylori infection and chronic, active gastritis
alone cause clinical symptoms remains contentious. The data currently
available do not provide compelling support for this view.[1,15,20]
Simply identifying H. pylori in a patient with symptoms
does not provide evidence, of course, that the infection is the
cause of the symptoms, since there is a comparable rate of infection
in asymptomatic, age- and community-matched controls. Similarly,
resolution of symptoms with therapy is of little significance
because the placebo response rate provides a similar frequency
in the improvement of symptoms. Current recommendations, therefore,
do not support therapy to eradicate H. pylori in the absence
of endoscopic evidence of peptic ulcers.[2,13] Nonetheless, many
gastroenterologists have seen an excellent response to eradication
of H. pylori infection in patients with nonulcer dyspepsia.[16]
It must be emphasized that much of the literature upon which these
observations are based is fraught with methodologic limitations.
For example, the definition of nonulcer dyspepsia may be imprecise,
resulting in a heterogeneous group of patients being lumped together.
This imprecision could mask the real benefit of therapy in symptomatic
patients with an H. pylori infection. In most published
studies, the duration of follow-up after intervention is also
quite short. This could be an important drawback because at least
one report, albeit uncontrolled, indicates that symptomatic improvement
is sustained for a long period as a result of therapy.[21] The
symptoms of dyspepsia experienced by patients who have acquired
H. pylori infection through self-inoculation or inadvertent
inoculation via a contaminated endoscope also suggest that infection
may play a role in dyspepsia. More carefully defined and controlled
studies could broaden current guidelines for treatment.
The results of trials involving children with recurrent abdominal
pain are equally inconclusive. Meta-analysis of these studies
suggests that there is little or no benefit of therapy, but at
least two studies[22,23] contradict this conclusion. A trial of
classic triple therapy for this indication has been advocated
by at least one pediatric gastroenterologist.[16]
[Table of contents]
Casecontrol studies provide epidemiologic evidence that
long-standing H. pylori infection and chronic gastritis
play a role in the development of a variety of types of gastric
cancer, including adenocarcinoma, lymphoma and MALT lymphoma.[1]
Nested casecontrol studies indicate a relative risk of gastric
cancer of between 3 and 6 among those with H. pylori infection,
with the relative risk rising to about 10 with increased duration
of infection.
Cure of the infection results in the regression of MALT lymphoma
and of intestinal metaplasia in the stomach. The latter is important
if such metaplasia is type III, which is a precursor of adenocarcinoma.
It is unclear from these studies whether there is regression of
all types of intestinal metaplasia. In patients with MALT lymphoma,
H. pylori infection should be eradicated and the patient
followed closely in a defined protocol and in collaboration with
hematologic oncologists, as needed.
It is not clear whether it is cost-effective or therapeutically
justifiable to initiate H. pylori treatment in the hope
of preventing cancer that will develop in only a small proportion
of all people with an infection. It certainly makes little sense
to begin therapy if the patient's only symptom is nonulcer dyspepsia,
with the justification that treatment will prevent cancer in the
long term. There is no evidence that patients with nonulcer dyspepsia
are at a higher risk of cancer than asymptomatic patients with
an H. pylori infection. Acquisition of the infection in
childhood appears to be a risk factor for gastric cancer. Therefore,
some advocate treatment of all children with an infection. However,
the costbenefit ratio of such an aggressive approach has
not been established.
[Table of contents]
H. pylori infection is associated with a wide variety of
other medical conditions, including coronary artery disease, short
stature, protein-losing gastropathy and recurring enteric infections.[24]
Additional data are required to confirm these observations before
considering the merits of instituting currently available treatments
for these conditions.
The National Institutes of Health consensus conference wisely
chose not to advocate treatment of H. pylori in the absence
of documented infection.[2] In these guidelines, treatment was
advocated only for patients with endoscopic evidence of mucosal
ulcers in the stomach or duodenum. Whether infected patients with
nonulcer dyspepsia or symptomatic infection should be treated
in an attempt to relieve symptoms and to prevent cancer is less
certain. Available data do not provide support for an aggressive
approach to the initiation of treatment. Further careful clinical
research is obviously required. Current guidelines for initiating
therapy to eradicate H. pylori infection (Table 1)
may well have to be modified and revised according to expected
advances in the field.[25]
[Table of contents]
Dr. Sherman is the recipient of a career scientist award from
the Ontario Ministry of Health. The publication of this article
was supported in part by a grant from the Fonds de la recherche
en santé du Québec. We would like to thank Astra
Pharma Canada for its sponsorship of the invited speakers' program.
[Table of contents]
- Veldhuyzen Van Zanten SJO, Sherman PM. A systematic overview
of Helicobacter pylori infection as the cause of gastritis, duodenal
ulcer, gastric cancer, and non-ulcer dyspepsia: a systematic overview.
Can Med Assoc J 1994; 150: 177-85.
- Helicobacter pylori in peptic ulcer disease. NIH Consensus
Statement. JAMA 1994; 272: 65-71.
- Mohamed AH, Wilkinson J, Hunt RH. The predictive value of
Helicobacter pylori (Hp) negativity for duodenal ulcer (DU) recurrence
[abstract]. Gut 1994; 35(5): W66.
- Rademaker JW, Hunt RH. H. pylori and gastric acid secretion:
the ulcer link? Scand Gastroenterol 1991; 26(suppl 187): 71-7.
- Telford JL, Ghiara P, Dell'Orco M, Comanducci M, Burroni D,
Bugnoli M, et al. Gene structure of the Helicobacter pylori cytotoxin
and evidence of its key role in gastric disease. J Exp Med 1994;
179: 1653-8.
- Sherman PM. Peptic ulcer disease in children. Diagnosis, treatment,
and the implication of Helicobacter pylori. Gastroenterol Clin
North Am 1994; 23(4): 707-25.
- Bourke B, Sherman PM, Drumm B. Peptic ulcer disease: What
is the role for Helicobacter pylori? Semin Gastrointest Dis 1994;
5: 24-31.
- Sonnenberg A, Townsend WF. Cost of duodenal ulcer therapy
with antibiotics. Arch Intern Med 1995; 155: 922-8.
- Li YY, Hu PJ, Liu DQ, Lin HL, Chen MH, Hazell SL, et al. Helicobacter
pylori and gastric cancer: an endoscopic survey of healthy children
in high and low risk areas for gastric cancer in China [abstract].
Gut 1995; 37(suppl 1): A38.
- Lechago J, Correa P. Prolonged achlorhydria and gastric neoplasia:
Is there a causal relationship? Gastroenterology 1993; 104(5):
1554-7.
- Fox JG, Wishnok JS, Murphy JC, Tannenbaum SR, Correa P. MNNG-induced
gastric carcinoma in ferrets infected with Helicobacter mustelae.
Carcinogenesis 1993; 14(9): 1957-61.
- Graham JR. Helicobacter pylori: human pathogen or simply an
opportunist? Lancet 1995; 345: 1095-7.
- Veldhuyzen Van Zanten SJO, Sherman PM. Indications for treatment
of Helicobacter pylori infection: a systematic overview. Can Med
Assoc J 1994; 150: 189-98.
- Al Somal N, Coley KE, Molan PC, Hancock BM. Susceptibility
of Helicobacter pylori to the antibacterial activity of manuka
honey. J R Soc Med 1994; 87: 9-12.
- Macarthur C, Saunders N, Feldman W. Helicobacter pylori, gastroduodenal
disease, and recurrent abdominal pain in children. JAMA 1995;
273: 729-34.
- Tolia V. Helicobacter pylori in pediatric nonulcer dyspepsia:
Pathogen or commensal? [editorial; comment]. Am J Gastroenterol
1995; 90(6): 865-7; Comment on: Am J Gastroenterol 1995; 90(6):
906-9.
- Hunt RH. Epidemic hypochlorhydria. In: Heatly RV, editor.
Helicobacter pylori and gastroduodenal disease. 2nd ed. Oxford:
Blackwell Scientific Publications, 1992: 187-97.
- O'Brien B, Goeree R, Mohamed AH, Hunt RH. Cost-effectiveness
of Helicobacter pylori eradication for the long-term management
of duodenal ulcer in Canada. Arch Intern Med 1995; 155: 1958-64.
- Krakowka S, Eaton KA, Rings DM. Occurrence of gastric ulcers
in gnotobiotic piglets colonized by Helicobacter pylori. Infect
Immun 1995; 63: 2352-5.
- Veldhuyzen Van Zanten SJO, Malatjalian D, Tanton R, Leddin
D, Hunt RH, Blanchard W. The effect of eradication of Helicobacter
pylori (HP) on symptoms of non-ulcer dyspepsia (NUD): a randomized
double-blind placebo controlled trial [abstract]. Gastroenterology
1995; 108(4): A250.
- McCarthy C, Patchett S, Collins RM, Beattie S, et al. Long-term
prospective study of Helicobacter pylori in non-ulcer dyspepsia.
Dig Dis Sci 1995; 40(1): 114-9.
- Heldenberg D, Wagner Y, Heldenberg E, Keren S, Auslaender
L, Kaufshtein M, et al. The role of Helicobacter pylori in children
with recurrent abdominal pain. Am J Gastroenterol 1995; 90(6):
906-9.
- Oderda G, Vaira D, Ainley C, et al. Eighteen month follow-up
of Helicobacter pylori positive children treated with amoxycillin
and tinidazole. Gut 1992; 33: 1328-30.
- Mendall MA, Goggin PM, Molineaux N, Levy J, et al. Relation
of Helicobacter pylori infection and coronary heart disease. Br
Heart J 1994; 71: 437-9.
- Hunt RH, Tytgat GNJ. Helicobacter pylori: basic mechanisms
to clinical cure. Dordecht, Netherlands: Kuwer Academic Publishers,
1994.
[Table of contents]
| CIM: October 1996 / MCE: octobre 1996
|
CMA Webspinners / >