Sherman, Philip*; Czinn, Steven*; Drumm, Brendan†; Gottrand, Frederic†; Kawakami, Elisabete‡; Madrazo, Armando‡; Oderda, Giuseppina†; Seo, Jeong-Kee¶; Sullivan, Peter#; Toyoda, Shigeru¶; Weaver, Lawrence#; Wu, Tzee-Chung¶
I. SUMMARY OF THE PROBLEM
Helicobacter pylori–induced gastritis causing mucosal ulceration either in the stomach or the proximal duodenum is a relatively uncommon event in children compared with adults. When H. pylori–associated ulcers occur in children, duodenal ulceration is much more frequently identified than gastric ulcers. In children younger than 6 to 10 years of age, peptic ulcers are usually due to noxious agents (such as corticosteroids and nonsteroidal antiinflammatory agents) or after major stresses (for example, burns, head injury, systemic illness). In these settings, upper gastrointestinal tract hemorrhage, vomiting, or perforation are frequent presenting features. The ulcers tend not to recur after healing if either the offending agent or underlying disease predisposing to mucosal ulceration can be removed.
In older children and adolescents, the clinical presentation and natural history of peptic ulcers are more comparable to those observed in adults. The ulcers present as epigastric and nocturnal abdominal pain in teenagers with a positive family history of peptic ulceration. In this setting, despite healing of the acute ulceration, the natural history is for the ulcer to recur. It is now clear that such ulcers are not related to a genetic predisposition to hyperpepsinogenemia but to infection of the stomach with the pathogen H. pylori.
Recurrences of H. pylori–associated peptic ulceration are markedly reduced by treatment directed at eradicating the gastric infection. For example, duodenal ulcer recurrence rates in H. pylori–infected children have been reported to be reduced from 65% recurrence at 1 year of follow-up if ulcers are healed with acid suppressive agents alone compared with 1-year ulcer recurrence rates below 5% if the gastric pathogen is eradicated. This intervention therefore results in reduced global health care costs.
Peptic ulcer disease has been considered to arise as a result of an imbalance between mucosal defenses and aggressive factors, including acid and pepsin. Mucosal defenses against acid and pepsin include mucus and bicarbonate, mucosal blood flow, prostaglandins, trefoil peptides, and epidermal growth factor. Primary peptic ulcers are usually duodenal, chronic, and related to H. pylori, whereas secondary ulcers are gastric, acute, and unrelated to infection status. It is clear that marked overproduction of gastric acid, such as occurs in the Zollinger-Ellison syndrome and, more commonly in children, antral G cell hyperplasia (also referred to as pseudo-Zollinger Ellison syndrome) can cause duodenal ulceration. However, excess acid production does not account for the overwhelming majority of peptic ulcers in children.
It has been known for many years that recurrent peptic ulcer disease in humans is accompanied by a chronic-active gastritis. In 1983, Drs. Barry Marshall and Robin Warren in Perth, Australia provided the explanation for the cause of the gastritis and accompanying peptic ulcer disease when they successfully cultured H. pylori from the gastric antrum of humans.
Subsequent studies in children confirmed the findings first reported in adults. There is specificity to the infection, the infection is associated with previously unexplained recurrent peptic ulcers, and both gastritis and ulcers resolve when the infection is successfully eradicated (1). Although there is general consensus worldwide to treat H. pylori infection when there is endoscopic evidence of peptic ulceration, whether to intervene in other more frequently encountered clinical settings remains uncertain. Increasing knowledge of the natural history of the infection, routes of transmission in human populations, and potential environmental reservoirs will help to clearly define why, when, and in whom to treat H. pylori.
II. MAJOR ISSUES IN NEED OF INVESTIGATION OR IMPLEMENTATION
A. Epidemiology of H. pylori Infection in Children
No environmental reservoir for H. pylori has been identified. Although nonhuman primates and domestic cats have been reported to harbor the infection, it appears that this is due to transmission of the infection from people to nonhumans.
It is now clear that H. pylori infection is usually acquired in childhood. In developing countries, H. pylori is acquired early in infancy, with up to two thirds of children infected by 2 years of age. The prevalence of infection appears to be lower amongst breast-fed infants whose mothers are infected with the organism. A reduced rate of infection could relate to the presence of specific H. pylori immunoglobulin A antibodies present in the milk. In contrast, in developed nations only 10% of all children will be infected with H. pylori at 10 years of age. There are higher prevalence rates observed in certain childhood populations, such as the clustering of infection among family members and in children living in institutions. Such epidemiologic features suggest that the infection is spread amongst humans.
Person-to-person transmission could occur via oral transmission because the organism has been detected in the oral cavity and in vomitus (2). Alternatively the route of transmission of infection could be fecal to oral because the organism has, on a few occasions, been cultured from feces, and bacterial antigens are detected in human feces of those harboring the organism in the stomach. Moreover, fecal-to-oral transmission has been demonstrated convincingly in animal models of Helicobacter infection (such as H. mustelae infection of ferrets).
The prevalence of H. pylori infection increases with advancing chronologic age due to a cohort effect. Prevalence rates are higher in developing nations and amongst children living in lower socioeconomic groups in developed countries. Previous reports of ethnic and genetic susceptibilities to H. pylori likely reflect insufficient consideration of the confounding effects of socioeconomic deprivation as an independent risk factor.
There is controversy regarding the natural history of H. pylori infection in children. Recent reports highlight the changes in histology in the stomach over time. In one study, children followed over a period of 2 years demonstrated increasing mucosal inflammation in the stomach without evidence of changes in bacterial density. These interesting findings need to be confirmed by researchers evaluating children longitudinally in other clinical and geographic settings. Several recent reports also indicate that the acute inflammatory response in the gastric mucosa is less pronounced in children compared with adults. These findings could indicate that the host humoral and cellular responses differ depending on the age at which the gastric infection is first acquired and might serve to explain varying rates of disease outcome (ulcer, cancers) evident in different parts of the world.
At least four studies, using indirect markers of gastric infection, report spontaneous clearing and re-acquisition of the gastric infection in preschoolers. These interesting findings need to be confirmed using highly accurate methods to definitively establish the presence or absence of infection. If confirmed, the findings would indicate marked differences in the natural history of infection in young children compared with the rest of the human population because spontaneous eradication of infection does not appear to occur in adults. Re-acquisition of H. pylori after successful therapy in adolescents and adults is an exceedingly uncommon event; at least amongst those individuals living in developed nations. It has been considered that re-infections are, in reality, recrudescences after treatment failure.
B. Consequences of H. pylori Infection
1. Gastritis:H. pylori infection fulfills each of Koch's postulates and all of Hill's criteria for causation as a cause of chronic-active pangastritis in humans. Therefore, there is no doubt that the organism is a human pathogen. Questions remaining at this time relate to the clinical consequences of the infection and the accompanying mucosal inflammation in the stomach.
2. Duodenal ulcer: The evidence in support of H. pylori infection as a cause of duodenal ulceration is convincing. Ulcers heal faster when anti-helicobacter therapy is used compared with acid suppression therapy alone. More importantly, duodenal ulcers do not recur when H. pylori infection is successfully eradicated.
3. Gastric ulcer: The natural history of non-NSAID–associated gastric ulceration in adults is changed by eradication of H. pylori infection. However, the causative role of H. pylori in causing gastric ulcers in children and adolescents is less certain. In some, but not all, reports in children with gastric ulceration the presence of H. pylori infection is uncommon and no more frequent than in age-matched reference populations of asymptomatic individuals. This finding may simply reflect the fact that a large proportion of gastric ulcers is secondary in nature in children.
4. Gastric adenocarcinoma: There is accumulating evidence supporting long-standing H. pylori infection, gastric atrophy, and intestinal metaplasia with the development of intestinal-type and undifferentiated adenocarcinomas in the antrum and body of the stomach in adults. To date, H. pylori-induced gastric cancer has not been reported in children. Epidemiologic studies associating H. pylori infection with gastric cancers were judged to be sufficiently compelling by the World Health Organization for the organism to be the first bacterium classified as a human carcinogen. Adenocarcinomas developing in the stomach of animals infected with helicobacter species (H. pylori infection of Mongolian gerbils, and H. mustelae infection of ferrets in conjunction with oral delivery of a chemical carcinogen) provide strong support for the role of the organism as a carcinogen. Whether the possible progression of childhood H. pylori infection to gastric carcinoma in adult life is sufficient reason to advocate eradication of the organism as soon as it is encountered is currently a highly contentious issue. Additional research is required to clarify current controversies.
5. Gastric lymphoma and MALToma: Seroepidemiologic studies support an association between long-standing H. pylori infection and both lymphoma and B cell monoclonal proliferative responses—also referred to as MALTomas. H. pylori-associated MALTomas have been reported only very occasionally in children. Studies in adults suggest that at least a proportion of MALTomas appear to demonstrate regression after eradication of the gastric infection.
C. Helicobacter pylori Infection and Abdominal Pain
The link between H. pylori infection and the development of recurrent abdominal pain is controversial. A recent meta-analysis of a large number of controlled studies in adults with epigastric abdominal pain in the absence of endoscopic evidence of peptic ulceration provides support for the contention that H. pylori-induced gastritis may well be a risk factor in non-ulcer dypepsia (3). The meta-analysis indicated that the presence of H. pylori was associated with symptoms of non-ulcer dyspepsia (odds ratio 1.6, 95% confidence intervals: 1.4–1.8), and symptoms improve when the organism is successfully eradicated (1.9, 1.3–2.6).
A comparable evaluation of published studies available in the pediatric literature does not provide support for H. pylori gastritis as an etiologic factor in recurrent abdominal pain of childhood; at least, as defined by Apley (4). The role of H. pylori as a cause of uninvestigated dyspepsia (that is, epigastric pain before diagnostic upper endoscopy has been performed) and non-ulcer dyspepsia (epigastric symptoms in the absence of mucosal ulceration at esophagogastroduodenseopy) in children and adolescents is an important unanswered question. There are no definitive recommendations for universal testing or treatment of H. pylori in children with either recurrent pain referable to the epigastrium or recurrent periumbilical abdominal pain. It remains to be determined whether H. pylori–induced gastritis is responsible for clinical symptoms in at least a subgroup of infected children and adolescents.
D. Extraintestinal Features of H. pylori Infection
H. pylori has been associated with a wide variety of neurologic, dermatologic, and cardiac manifestations. Current critical reviews of the literature comment on the poor quality of the available evidence and the need for additional carefully conducted, high-quality studies with appropriate control groups (5). There are also reports from some centers associating H. pylori infection in children with short stature and diarrheal disease, but this finding has not been confirmed in studies by other investigators. Additional studies in this important but contentious area are awaited with interest.
A few case reports describe sideropenic refractory anemia, occurring both in adults and in children, that respond to the eradication of H. pylori with an increase in hemoglobin and ferritin levels. These findings are of particular interest because case-control studies describe lower body iron stores in H. pylori–infected children compared to age-matched controls without evidence of gastric colonization.
E. Methods to Diagnose Peptic Ulcers and H. pylori
1. Invasive Methods
Endoscopy is the only method to accurately diagnose peptic ulceration in children. At the time of diagnostic upper endoscopy, nodularity in the antrum of the stomach is a specific (but not sensitive) feature indicative of active H. pylori infection in children. Biopsy specimens taken from the stomach can be tested for the presence of H. pylori by a variety of complementary techniques. These approaches include staining of biopsy sections (silver, Giemsa, Genta, acridine orange) and microscopic evaluation, culture, and testing for the presence of urease activity.
Staining with silver and microscopic evaluation by an expert pathologist is the most accurate invasive method of diagnosis of H. pylori infection in pediatric populations. Giemsa staining is a reasonable alternative because it is easy to perform, less susceptible to technical artifacts, and is relatively inexpensive. Much less commonly than H. pylori, a morphologically distinct helicobacter also has been associated with gastritis in humans, including children (6). Currently referred to as “Helicobacter heilrnaneii” (previously called “Gastrospirillum horn mis”), the nonculturable organism has not been associated with other complication such as peptic ulcer disease or gastric cancers.
Culture of gastric biopsy specimens is less sensitive and more expensive, but it offers the opportunity to test for antimicrobial resistance that may be relevant for those children failing to respond to traditional therapeutic regimens. Polymerase chain reaction using appropriate specific primers can be employed to successfully detect bacterial genomic DNA in gastric biopsy specimens, as well as in feces and in saliva. The technology can be employed to differentiate between recrudescence and re-infection and to detect point mutations in genes accounting for bacterial resistance to antibiotics, including nitroimidazoles and macrolides. In children and teenagers receiving concurrent therapy with a proton pump inhibitor, biopsies should be performed on the body and cardia (and, possibly, transition zones) of the stomach as well as from the antrum to reduce the chances of false-negative results.
Follow-up endoscopy is rarely necessary in pediatric populations, except in the setting of peptic ulceration associated with complications (for example, massive hemorrhage or perforation). If repeat diagnostic upper endoscopy is undertaken, biopsies from antrum, body, cardia, and transition zones are recommended in adults. However, published data supporting similar recommendations in children and adolescents are currently lacking.
2. Noninvasive Methods
Infection with H. pylori induces a vigorous immune response resulting in the presence of local and systemic antibodies. H. pylori–specific immunoglobulin G antibodies present in serum, plasma, whole blood, saliva, gastric juice, and urine have each been used to successfully detect the presence of infection in adults. The humoral immune response is less vigorous in children. Therefore, the cutoff values established for use in adults to determine the presence or absence of H. pylori infection are not appropriate for use in young children. Increasing concerns have been raised about the utility of testing for antibodies in low-prevalence settings such as is the case for children and adolescents living in developed nations (7). Although immunoassays are relatively inexpensive and technically easy to perform, those employed in children must first be validated in each country or region, using serum samples obtained from relevant populations.
Office-based tests using whole blood or saliva have been advocated for use in the diagnosis of H. pylori infection in adults. The accuracy of these methodologies in varying pediatric populations should be the subject of carefully controlled comparative studies before they are employed in the clinical setting.
Breath testing using the stable isotope 13C- or radiolabeled 14C- urea as substrate shows great promise as an alternative approach to noninvasively diagnose H. pylori infection (8). Although the amount of radioisotope is very small, 14C-labeled substrates are forbidden for use in children and women during their reproductive years in some nations. Urea breath testing appears to be highly accurate in children older than 5 years of age, although the specific parameters of the testing protocols need to be standardized between laboratories. Whether the urea breath test will prove to have sufficiently high positive and negative predictive values in preschoolers is a matter of current investigation (9). At least in subjects older than 5 years of age, the presence of other urease-producing organisms in the oral cavity and in the gut do not provide confounding results.
A commercial stool antigen test has recently been reported to be highly accurate as a diagnostic test for use in adults. In contrast to serology, in which antibody titers may remain elevated for months after successful eradication of the organism, the stool antigen test is reported to turn negative within a week after elimination of H. pylori. The accuracy of the stool antigen test when used in pediatric populations in a variety of clinical settings requires further testing (10).
F. Treatment of Ulcers and H. pylori Infection
The discovery of H. pylori has radically changed the management of patients with primary peptic ulcers. Eradication of the organism prevents recurrence of ulcer disease. Therefore, treatment strategies should now focus on the eradication of H. pylori. Traditional therapeutic agents that heal peptic ulcers including antacids and histamine-2 receptor antagonists have no effect on H. pylori and the accompanying chronic-active antral gastritis. Although proton pump inhibitors have anti-helicobacter activity in vitro, the agents are not effective in eradicating the organism in vivo when employed alone as monotherapy. In vitro sensitivities to antimicrobial agents are not always an effective means of predicting response to therapy in the clinical setting (11).
Current treatment protocols advocated in a variety of consensus conferences (12–14) as first-line therapies include a proton pump inhibitor plus two antibiotics (choosing two of the following: amoxycillin, clarithromycin, or metronidazole). Meta-analyses indicate successful eradication in more than 80% of subjects when evaluated on an intent-to-treat protocol.
Although controlled studies in children are not yet available in the peer-reviewed biomedical literature, open case series suggest comparable rates of efficacy when using the same treatment regimens in children (15). Optimal compliance may be obtained when treatment regimens provide twice-daily dosing for a period of 7 days. Compliance is an important consideration because it is a major determinant of the success of treatment. H. pylori strains resistant to either nitroimidizoles or clarithromycin prove more difficult to eradicate. In this setting, the use of quadruple therapy, which can include a treatment regimen containing bismuth, may prove efficacious. It should be noted that the use of bismuth-containing compounds is not allowed in some countries (for example, France). Similarly, in some nations, approval for the use of clarithromycin in combination regimens to eradicate H. pylori infection has not been approved by the appropriate regulatory agency.
G. Prevention of H. pylori Infection
H. pylori meets two important criteria that give it high priority for vaccine development. Firstly, there is widespread infection in both developing world and in adults living in more developed nations. Secondly, the infection imposes a considerable burden of illness in developed nations so that sufficient resources should be able to be attracted to support the necessary research required for vaccine development. Moreover, increasing rates of resistance to antibiotics and suboptimal compliance with multiple drug regimens provide additional justification for actively promoting vaccine development.
Studies in animals indicate that there is potential for the development of a vaccine to both prevent H. pylori infection and to eradicate existing bacterial colonization in the stomach. In mouse models of Helicobacter infection, a variety of immunogens (bacterial sonicates, urease, heat stress proteins, outer membrane proteins, and possibly lipoproteins) are effective in eliciting a change the cellular and humoral immune responses that result in successful therapeutic vaccinations. To date, all of the vaccine constructs require the use of small amounts of either cholera toxin or Eschericha coil heat-labile toxin as mucosal adjuvants. The development of effective immunogens that can be employed successfully in animals and humans without the need for the potentially toxic enterotoxins is a challenge for the future.
III. PROPOSED PLAN TO ACHIEVE GOALS
Research of the highest quality, conducted both in the laboratory and in the clinical area, is required to clearly define the mode of transmission of the organism. Such studies should resolve the conflicting reports regarding potential environmental reservoir for H. pylori. The viability of coccoid forms of the microbe and potential role as a vehicle for transmission requires clarification.
Clinical research and appropriate animal models of infection are required to clearly delineate the natural history of infection. Studies in young children, using validated questionnaires and highly accurate diagnostic techniques, are needed to clarify the frequency of spontaneous eradication of infection and factors modulating host responses to gastric colonization by the bacterium. The role, if any, of spiral bacteria other than H. pylori, including, for example, Helicobacter heilmanii, in gastrointestinal and hepatobiliary diseases in humans requires definition. The evolution of gastritis induced by helicobacter infection to mucosal atrophy and intestinal metaplasia in childhood populations in regions where the frequency of gastric cancers later in life varies widely should be monitored over time. Such studies will require careful attention to potential confounding variables including other environmental factors that impact on carcinogenesis.
The role of novel preventative and treatment strategies for H. pylori infection requires clarification. For instance, in children the potential role of nutriceuticals as anti-helicobacter therapy requires careful investigation in the clinical research setting or with the use of appropriate animal models, and not simply reliance on data generated in vitro in the laboratory setting. The development of an effective oral vaccine for use in humans, without the need for potentially toxic mucosal adjuvants, is required. Prevention studies and attempts at eradication of established infection (therapeutic vaccination) in children likely should await pilot studies (so-called phase 1 and phase 2 investigations) in nonhuman primates and adult humans to establish safety and provide initial evidence of potential efficacy.
The potential for H. pylori to cause clinical symptoms in the absence of mucosal ulceration requires additional study. These clinical investigations require the use of strict study group definitions to definitively establish whether there are subsets of infected children and adolescents who have clinical symptoms arising from gastric infection and the resulting chronic-active pangastritis. Functional dyspepsia, non-ulcer dyspepsia, and recurrent abdominal pain must be evaluated as separate entities. All such studies require the use of appropriate age-matched comparison groups. Such research will help to better define current controversies regarding appropriate indications for the diagnosis and treatment of H. pylori infection in children. Case reports and reports of aggregate case series will add little to current uncertainties in the field.
Additional important issues warrant clinical research in children. These research areas include the validation of accurate indirect diagnostic tests for use in children under 5 years of age, a clear definition of the frequency of reinfection in young children after eradication therapy, the choice and timing of diagnostic tests for use in follow-up of children receiving eradication therapy, and the optimal management of initial treatment failures, including the role of antibiotic sensitivity testing in this setting.
Support for the research in the priority areas identified above should be broad based and subject to the rigors of the peer review process. Soliciting of funds to support the necessary research should include traditional governmental granting agencies, the World Health Organization, appropriate national agencies, and partnerships with industry. These organizations should be encouraged to more actively support research activities in this field. Similarly, charitable organizations and societies concerned with issues of the health children should be engaged in this research program.
Many of these proposals could be enhanced by the development of multicenter and cross-national evaluations. Priority should be given to the highest quality research programs that form new collaborative research partnerships. In this way, research can evaluate critical issues in the primary care setting as well as in children living in a variety of socioeconomic circumstances.
1. Sherman P, Hassall E, Hunt RH, et al. Canadian Helicobacter study group consensus conference on the approach to Helicobacter pilori infection in children and adolescents. Can J Gastroenterol 1999; 13:553–9.
2. Parsonct J, Shmuely H, Haggerty T. Fecal and oral shedding of Helicobacter pylori from healthy infected adults. J Am Med Assoc 1999; 282:2240–5.
3. Jaakkimainen RL, Boyle E, Tudiver F. Is Helicobacterpylorii associated with non-ulcer dyspepsia and will eradication improve symptoms? A meta-analysis. Br Med J 1999; 319:1040–4.
4. Macarthur C. Helicobacter pylori infection and childhood recurrent abdominal pain: lack of evidence for a cause and effect relationship. Can J Gastroeneterol 1999; 13:607–10.
5. Leontiadis GI, Sharma VK, Howden CW. Non-gastrointestinal tract associations of Helicobacter pilori infection. What is the evidence? Arch Intern Med 1999; 159:925–40.
6. Mention K, Michaud L, Guimber D, et al. Characteristics and prevalence of Helicobacter heibnanii infection in children undergoing upper gastrointestinal endoseopy. J Pediatr Gastroenterol Nutr 1999; 29:533–9.
7. Drumm B, Koletzko S, Oderda G. Helicobacterpylori infection in children: a consensus statement. J Pediatr Gastroenterol Nutr 2000; 30:207–13.
8. Vincent P, Michaud L, de LaSalle EM, et al. 13C-urea breath test and gastric mucosal colonziation by Helicobacter pylori in children: quantitative relation and usefulness for diagnosis of infection. Helicobacter 1999; 4:233–7.
9. Thomas JE, Dale A, Harding M, et al. Interpreting the 13C-urea breath test in a large population of young children from a developing country. Pediatr Res 1999; 46:147–51.
10. Oderda C, Rapa A, Ronchi B, et al. Detection of Helicobacter pylori in stool specimens by non-invasive antigen enzyme immunoassay in children: multicentre Italian study. Br Med J 2000; 320:347–8.
11. Graham DY. Antibiotic resistance in Helicobacter pylori: implications for therapy. Gastroenterology 1998; I 15:1272–7.
12. Hunt RH, Thomson ABR. Canadian Helicobacterpylorii consensus conference. Can J Gastroenterol 1998; 12:31–41.
13. Lam SK, Talley NJ. Report of the 1997 Asia Pacific consensus conference on the management of Helicobacterpylorii infection. J Gastroenterol Hepatol 1998; 13:1–12.
14. European Helicobacter Study Group. Current European concepts in the management of Helicobacter pylori infection. The Maastricht consensus report. Gut 1997;41:8–13.
15. Rowland M, Mine C, Bourke B, et al. How should Helicobacter pylori infected children be managed? Gut 1999; 45(Suppl 1):36–9.
© 2002 Lippincott Williams & Wilkins, Inc.