Diagnosi e trattamento delle infezioni da H. Pylori in pediatria

Medical Position Statement: The North American Society for PediatricGastroenterology and Nutrition

Helicobacter pylori Infection in Children: Recommendations forDiagnosis and Treatment

*Benjamin D. Gold, †Richard B. Colletti, ‡Myles Abbott, §Steven J. Czinn, �Yoram Elitsur,¶Eric Hassall, #Colin Macarthur, **John Snyder, and ††Philip M. Sherman

Division of *Pediatric Gastroenterology and Nutrition, Department of Pediatrics, Emory University School of Medicine,Atlanta, Georgia; †Pediatric Gastroenterology, University of Vermont, Fletcher Allen Health Center, Burlington, Vermont;‡Berkeley California; §Pediatric Gastroenterology, Rainbow Babies and Children’s Hospital, Cleveland, Ohio; �Division ofGastroenterology, Marshall University School of Medicine, Huntington, West Virginia; ¶Division of Gastroenterology, BCChildren’s Hospital, Vancouver, British Columbia, Canada; Divisions of #General Pediatrics and ††Gastroenterology and

Nutrition, The Hospital for Sick Children, Toronto, Ontario, Canada; and **Pediatric Gastroenterology, University of CaliforniaSan Francisco, San Francisco, California, U.S.A.

Helicobacter pylori infects at least 50% of the world’shuman population (1). However, most individuals in-fected with H. pylori do not experience symptoms orhave signs of recognizable disease. In most children, thepresence of H. pylori infection does not lead to clinicallyapparent disease, even when the organism colonizing thegastric mucosa causes chronic active gastritis (2).Knowledge about H. pylori infection is evolving, par-ticularly in the pediatric age group for which there arestill large gaps in knowledge.

Additional multicenter, randomized, placebo-controlled treatment trials in children infected by H. py-lori are critically needed to definitively characterize theeffect of H. pylori eradication treatment during child-hood on symptoms and gastroduodenal mucosal disease.

There is compelling evidence that this organism isassociated with a significant proportion of duodenal ul-cers and, to a lesser extent, with gastric ulcers in children(3). There are epidemiologic data linking chronic H. py-lori infection, probably beginning in childhood, with thedevelopment of gastric adenocarcinoma and gastric lym-phoma (4). Findings in recently reported animal modelssupport the role of H. pylori in the pathogenesis of gas-tric cancers (5).

There are many studies describing the prevalence ofH. pylori infection. Most epidemiologic studies of H.pylori infection have been performed in adults who had

been infected for many years before clinical symptomsappeared (6). The incidence of H. pylori infection inindustrialized countries is estimated to be approximately0.5% of the susceptible population per year. In contrast,there is a significantly higher estimated incidence of H.pylori infection in developing countries of approximately3% to 10% per year (7). The limited data on the inci-dence of H. pylori infection in children consist largely ofretrospective seroprevalence studies.

Humans appear to be the primary natural reservoir ofH. pylori infection. Other reservoirs that have been pro-posed include water, domestic cats, and houseflies (8–10). The risk factors described for acquiring infectioninclude residence in a developing country, poor socio-economic conditions, family overcrowding, and possiblyan ethnic or genetic predisposition. In North America,the prevalence rates of H. pylori among Asian-Ameri-cans, African-Americans and Hispanics are similar tothose of residents of developing countries (11). The routeof transmission of H. pylori in humans is not known butis postulated to be fecal-oral, gastric-oral (in vomitus), ororal-oral (12).

Although H. pylori infection may be acquired duringchildhood, there are limited guidelines regarding its di-agnosis and treatment in children and adolescents. Suchevidence-based consensus guidelines are needed for bothprimary care and specialty medical providers to ensurejudicious use of diagnostic testing and appropriate thera-peutic regimens for the management of children with H.pylori infection. Therefore, the North American Societyfor Pediatric Gastroenterology and Nutrition (NASPGN)appointed the Helicobacter pylori Infection GuidelineCommittee to develop a clinical practice guideline forthe child with H. pylori infection.

Received and accepted September 8, 2000.Address correspondence and reprint requests to Dr. Benjamin Gold,

Emory University School of Medicine, 2040 Ridgewood Drive, NE,Atlanta, GA 30322, U.S.A.; or to Dr. Richard B. Colletti, University ofVermont, Department of Pediatrics, A-121 Given Medical Building,Burlington, VT 05405-5557, U.S.A.

Journal of Pediatric Gastroenterology and Nutrition31:490–497 © November 2000 Lippincott Williams & Wilkins, Inc., Philadelphia

490

These clinical practice guidelines are designed to as-sist primary care physicians, nurse practitioners, physi-cian assistants, and pediatric gastroenterologists in theevaluation and treatment of suspected or diagnosed H.pylori–associated disease. The desired outcomes of theserecommendations are the detection of children and ado-lescents with H. pylori who need treatment. These rec-ommendations are applicable to children in developedcountries where the prevalence of infection is low butmay not be directly relevant to children living in com-munities where there is a higher frequency of gastriccolonization by H. pylori. These recommendations havebeen endorsed by the Executive Council of NASPGNand by the American Academy of Pediatrics. They aregeneral guidelines to assist medical care providers in thediagnosis and treatment of H. pylori infection in chil-dren. They are not intended as a substitute for clinicaljudgment or as a protocol for the management of allpatients.

In its deliberations, the committee addressed four is-sues about H. pylori infection in children: How reliableare tests to detect H. pylori? When is testing for H. pyloriindicated? When is treatment of H. pylori infection in-dicated? What is the preferred treatment of H. pylori? Asummary of the recommendations of the H. pylori Infec-tion Guideline Committee is presented in Table 1.

METHODS

The H. pylori Infection Guideline Committee consisted of aprimary care pediatrician, a clinical epidemiologist, and sevenpediatric gastroenterologists. To develop evidence-basedguidelines, articles published in English from January 1966through May 1999 on H. pylori in children were searched.Articles on diagnosis and treatment were sought separately.Letters, editorials, case reports, abstracts, and reviews wereexcluded. Evidence tables were prepared based on 16 articleson clinical presentation, 9 articles on diagnostic studies, and 30articles on therapy. Subsequently, additional articles were iden-tified and reviewed. When the pediatric literature was insuffi-cient, the adult literature was also considered. Articles wereevaluated using published criteria (13). The Committee basedits recommendations on an integration of a review of the medi-cal literature and expert opinion. Consensus was achieved byusing the nominal group technique, a structured quantitativemethod, as described previously (14,15). By using the methodsof the Canadian Preventive Services Task Force (16), the qual-ity of evidence of each of the recommendations made by thecommittee was determined and is summarized (Table 1).

HOW RELIABLE ARE TESTS FORH. PYLORI INFECTION?

Several invasive and noninvasive tests are available todetect H. pylori infection (Table 2). An ideal test for H.pylori is noninvasive or minimally invasive, highly ac-curate, inexpensive, and readily available and enablesdifferentiation between active or past infection with the

TABLE 1. Summary of recommendations and the quality ofthe supporting evidence

RecommendationsQuality ofevidencea

How reliable are tests for H. pylori infection?Currently, the diagnosis of H. pylori-mediated disease

can be made reliably only through the use ofendoscopy with biopsy. II, III

Presently available commercial serologic tests arefrequently unreliable for screening children for thepresence of H. pylori infection. II

Urea breath testing, although promising, has not beenstudied sufficiently in children. II

When is testing indicated?It is recommended that testing be performed in

children with endoscopically diagnosed, orradiographically definitive, duodenal or gastric ulcers. I

It is recommended that children with recurrentabdominal pain, in the absence of documentedulcer disease, not be tested for H. pylori infection. II

Testing for H. pylori infection is not recommended inasymptomatic children. II

Routine screening of children with a family history ofgastric cancer or recurrent peptic ulcer disease isnot recommended. II

Testing following treatment of documented H. pyloriis recommended, especially with complicatedpeptic ulcer disease (i.e., bleeding, perforation, orobstruction). For patients who remain symptomaticafter treatment, it is recommended that endoscopyand biopsy be performed to evaluate for thepersistence of H. pylori-associated peptic ulcerdisease. I, II

If pathological evidence of MALT lymphoma isdocumented, then testing for H. pylori isrecommended. II

When is treatment of H. pylori infection indicated?Eradication treatment is recommended for children who

have a duodenal ulcer or gastric ulcer identified atendoscopy and H. pylori detected on histology. I

A prior history of documented duodenal or gastriculcer disease is an indication for treatment if activeH. pylori infection is documented. I

There is no compelling evidence for treatingchildren with H. pylori infection and non-ulcerdyspepsia or functional recurrent abdominal pain. III

Treatment is not recommended for H. pylori-infectedchildren residing in chronic care facilities; childrenwith unexplained short stature; or children atincreased risk for acquisition of infection, includingasymptomatic children who have a family memberwith either peptic ulcer disease or gastric cancer. III

What is the preferred treatment of H. pylori infections in children?It is recommended that treatment consist of three or

four medications, given once or twice daily, forone to two weeks. I

a Categories of the quallity of evidence: I Evidence obtained from atleast one properly designed randomized controlled study. II-1 Evidenceobtained from well-designed cohort or case-controlled trials withoutrandomization. II-2 Evidence obtained from well-designed cohort orcase-control analytic studies, preferably from more than one center orresearch group. II-3 Evidence obtained from multiple time series withor without the intervention. Dramatic results in uncontrolled experi-ments (such as the results of the introduction of penicillin treatment inthe 1940’s) could also be regarded as this type of evidence. III Opinionsof respected authorities, based on clinical experience, descriptive stud-ies, or reports of expert committees.

TREATMENT RECOMMENDATIONS FOR HELICOBACTER INFECTION 491

J Pediatr Gastroenterol Nutr, Vol. 31, No. 5, November 2000

organism. In addition, such a test enables discriminationbetween the presence of H. pylori infection and H. py-lori–associated disease. Because no such ideal test cur-rently exists, the advantages and drawbacks of tests thatare available require critical appraisal and must be as-sessed for their suitability for use in children.

Failure to reach an accurate diagnosis carries consid-erable financial and social costs including the expense ofmore tests, repeated visits to health care providers, inap-propriate treatment, and missed school or work. A de-finitive test, even if it is expensive, may result in overallcost savings (17).

It is important to emphasize that the accuracy of adiagnostic test is greatly impacted by the prevalence ofH. pylori in the population tested. There is a need forstudies to assess the accuracy and potential utility ofvarious noninvasive diagnostic tests in populations inNorth America that differ in demographic factors thatmay influence the prevalence and natural history of H.pylori infection (18).

Invasive Testing Through Endoscopy

Biopsies and Histopathology

The definitive diagnosis of H. pylori and the evidenceof the consequences of infection can be made reliablyonly by endoscopy with multiple biopsy specimens ob-tained in one or more regions of the stomach includingantrum, body, and transition zones (i.e., cardia and inci-sura). Histology provides information regarding the pres-ence of H. pylori and the severity and topographic dis-tribution of gastritis including the presence of atrophicgastritis, intestinal metaplasia, and mucosa-associatedlymphoid tissue (MALT) lymphoma (3). As in adults,biopsy specimens obtained in the prepyloric antrum havethe highest yield in H. pylori infection. Tissue specimensoften are also obtained from the body and the transitionzones of the stomach, particularly if the patient has re-cently taken acid-suppressing medication (19). It is rec-ommended that multiple biopsies be performed in chil-dren with endoscopically documented peptic ulcer dis-ease or peptic ulcer suspected as a result of radiographic

study. The optimal staining of biopsy sections is bestdetermined by local expert pathologists. Endoscopic ex-amination of and specimens obtained in the esophagus,stomach, and duodenum also provide information aboutother upper gastrointestinal disorders that may be thecause of clinical symptoms including, for example,esophagitis and peptic ulcer disease that is not due to H.pylori.

There are drawbacks to diagnostic gastrointestinal en-doscopy. It is a relatively invasive procedure requiringsedation or anesthesia. Furthermore, the test remainsrelatively expensive in many centers, and access to anendoscopist with specific pediatric expertise is limited inmany geographic areas.

Rapid Urease Testing of Biopsy Tissues

Urease testing (CLO, TriMed, Kansas City, MO; Hp-Fast, GI Supply, Division of ChekMed Systems Inc.,Camphill, PA; PyloriTek, Horizons International, Agua-dilla, Puerto Rico) provides indirect identification of H.pylori infection within a few hours of endoscopy (20).However, these tests have a poor positive predictivevalue (as low as 50%) in children, even though the nega-tive predictive value is high (97–98%) (20,21). The ac-curacy of the test is dependent on the number of tissuespecimens tested, the location of biopsy sites, bacterialload, and previous usage of antibiotics and proton pumpinhibitors, as well as the prevalence of H. pylori in thepopulation tested.

Bacterial Culture

Culture of H. pylori from the gastric mucosa providesan opportunity to obtain a profile of antibiotic sensitivitythat could identify potential treatment failure due to an-tibiotic resistance (22). Culture also provides a bacterialstrain for use in epidemiologic studies to examine asso-ciations of virulence characteristics with disease out-come. However, bacterial culture for H. pylori is rela-tively expensive and success rates for recovery of theorganism in many clinical laboratories are low (23). Cur-rently, standardization of culture procedures has not beenestablished, and bacterial cultures are only obtained rou-tinely in research settings.

Polymerase Chain Reaction

Polymerase chain reaction (PCR) is a highly sensitivetechnique that can be used to detect the presence of H.pylori in body fluids (e.g., gastric juice and stool), tissues(e.g., gastric mucosa), and water (24). Testing of H. py-lori genomic DNA by PCR can be used to advanceknowledge at the molecular level—for example, by pro-viding information about point mutations conferringresistance to antibiotics and about putative bacterialvirulence factors. However, PCR is expensive, the assay

TABLE 2. Tests for Helicobacter pylori andHelicobacter-related disorders

Invasive tests requiring endoscopyBiopsies and histologyRapid urease testingBacterial culturePolymerase chain reaction of bacterial DNA

Non-invasive testsSerum and whole blood antibodySaliva antibodyUrine antibodyStool antigenUrea breath testing

GOLD ET AL.492


is difficult to set up, specificity may be compromised byinadvertent contamination, and it is not widely availableoutside the research laboratory.

Noninvasive Testing

Immunoassay Tests to Detect H. pylori Antibodies

Enzyme-linked immunosorbent assays (ELISAs) todetect H. pylori antibodies are relatively inexpensive andeasy to implement in the clinical setting. Many tests areavailable for use to test whole blood, plasma, or serum.However, compared with histology, the sensitivity andspecificity of serologic assays are poor in both adults andchildren unless used in the populations in which theywere initially developed (25). In general, the accuracy ofserum-based immunoassays and whole-blood tests foruse in the physician’s office in symptomatic children indeveloped countries is poor, with a range of sensitivity ofonly 60% to 70% (26–28). Furthermore, age-related cut-off values for commercial immunologic tests have notbeen established for children. One immunoassay devel-oped in a research center to detect H. pylori–specificimmunoglobulin (Ig)G in children was 91% sensitivecompared with sensitivity of less than 70% in three com-mercially available assays (28). In areas with low preva-lence of H. pylori infection, such as in developed coun-tries, testing of serum and whole blood is not sufficientlyaccurate to diagnose H. pylori infection in children. Ac-cordingly, treatment regimens based on the results ofthese tests cannot be recommended. Serologic tests maynot be used reliably to verify eradication of H. pylori,because antibody titers can remain positive for months,despite resolution of infection.

Saliva and Urine Tests for H. pylori Antibodies

Similar to serologic tests, saliva-based tests also detectthe presence of H. pylori–specific IgG antibodies. Thetests are easy to perform, painless, and inexpensive. Sa-liva tests are less sensitive than assays of serum or wholeblood (29). The protein concentration of saliva appears toaffect the accuracy of test results. Urine-based assays areeasy to perform, require minimal labor for collection,and are painless (30). However, these assays are highlyvariable and are not yet commercially available. There-fore, saliva and urine assays for the detection of H. pyloriantibodies cannot be recommended.

Stool Test for H. pylori Antigens

Testing of H. pylori antigens in stools has shownpromising results in adults for the noninvasive diagnosisof gastric infection using a commercially available kit(31). Testing for H. pylori antigens in feces also appearsto be accurate for use in monitoring the success of eradi-

cation therapy. However, patients may be reluctant tocollect stool specimens. In addition, refrigerated stoolsare more difficult to test. Additional pediatric studiesevaluating the accuracy of stool antigen testing for bothinitial diagnosis and posttreatment follow-up are re-quired before specific recommendations can be consid-ered (32).

Urea Breath Testing

Urea breath tests are noninvasive and have high sen-sitivity and specificity (>95%) both in adults (33) andchildren (34,35). The test requires the ingestion of eitherradiolabeled 14C-urea or urea tagged with the stable iso-tope 13C. Test results may be influenced by concurrentuse of antibiotics and acid-suppressing medications andby the presence of other urease-producing organismspresent in the oral cavity. Test parameters are currentlylaboratory-specific (e.g., dosages for differing ages ofchildren, cutoff values, duration of fasting, use of a testmeal, times of sampling, and timing of posttherapy test-ing) and have not been well standardized for children(36). In addition, urea breath testing is technically moredifficult to perform in small children and infants, withfailure rates in collection up to 10%, especially outsidethe clinical research setting (34).

In summary, the diagnosis of H. pylori–associated dis-eases currently can be made reliably only by endoscopywith biopsies. The most commonly used noninvasive testto screen adults for H. pylori infection is serology. Un-fortunately, currently available commercial serologictests are frequently unreliable for screening children forthe presence of H. pylori infection. Current whole-blood,saliva, and urinary immunoassays are insufficiently sen-sitive or specific to be effective as diagnostic tools. In-sufficient data are available in children to confirm theaccuracy of the recently approved H. pylori stool antigentest. The urea breath test has the promise to providenoninvasive and accurate diagnosis of H. pylori infec-tion; but currently, there is insufficient evidence that itcan be used to reliably diagnose or exclude H. pylori–associated diseases.

WHEN IS TESTING INDICATED?

The primary goal of testing is to diagnose the cause ofclinical symptoms and not simply to detect the presenceof H. pylori infection. Testing is not helpful unless it willalter the management of the disease.

A variety of invasive and noninvasive tests exist forthe detection of H. pylori infection, but their degree ofsensitivity and specificity vary, as do their suitability forclinical use in children. Thus, there is potential for inap-propriate testing or misuse of tests in children.



Endoscopically Diagnosed or RadiographicallyDefinitive Peptic Ulcer

The causal relationship between H. pylori infectionand primary duodenal ulcers is compelling (37). There-fore, it is recommended that testing for the presence of H.pylori infection be performed in children with endo-scopically diagnosed or radiographically definitive duo-denal ulcer. Although the data in children are less com-plete, evidence from studies in adults (38) supports therecommendation that testing for H. pylori also be per-formed in subjects with a documented gastric ulcer.

Abdominal Pain Unrelated to Peptic Ulcers

Several lines of evidence, including serologic surveys,endoscopic evaluations, and treatment trials indicate thatH. pylori is not a frequent cause of recurrent abdominalpain in children. There have been six studies performedin North America, Europe, and Australia, with 2715 chil-dren evaluated by esophagogastroduodenoscopy and bi-opsy, serology, or urea breath test (39–44). Although 5%to 17% of children with abdominal pain had evidence ofinfection with H. pylori, 5% to 29% of children withoutabdominal pain were also infected with H. pylori. Thereare no convincing data to support routine testing of chil-dren with recurrent abdominal pain (39–45). Investiga-tors have also looked for specific symptom patterns in H.pylori–infected children, but none so far has been de-tected (46–50). Future studies are needed to determinewhether subsets of children with abdominal pain can beidentified in whom signs and symptoms are caused by H.pylori infection. It is recommended that children withrecurrent abdominal pain, in the absence of documentedulcer disease, not be tested for H. pylori infection.

Asymptomatic Children, Including Those atIncreased Risk of Acquiring H. pylori Infection

There are no compelling data to support routine testingin asymptomatic children. Testing for H. pylori infectionis not recommended in children without clinical symp-toms, including those residing in long-term care facili-ties, children with short stature, and those at increasedrisk of acquiring H. pylori infection. In addition, pur-ported extraintestinal manifestations of H. pylori infec-tion have not been demonstrated in a convincing fashion(51). Accordingly, a test-and-treat approach is not rec-ommended in these circumstances.

Family History of Gastric Cancer or RecurrentPeptic Ulcer Disease

No currently available data support routine testing inchildren with a positive family history of diseases relatedto H. pylori infection (52). Epidemiologic evidence in-

dicates that there is a link between gastric cancers (bothadenocarcinoma and lymphoma) and H. pylori infection.However, no studies have shown that H. pylori eradica-tion during childhood prevents subsequent developmentof gastric malignancies. Until evidence is available tobetter define the role of H. pylori in a variety of gastriccancers and the role of H. pylori eradication in diseaseprevention, routine screening of children with a familyhistory of gastric cancer or recurrent peptic ulcer diseaseis not recommended.

Histologic Evidence of Lymphoma

In the rare circumstance in which histopathologic evi-dence of MALT lymphoma is documented in a child,testing for H. pylori is recommended.

Follow-up of Therapy for H. pylori Infection

Testing to confirm eradication of infection and theresolution of associated symptoms and disease sequelaeis advisable in selected children. Guidelines in adultsrecommend testing after treatment of complicated pepticulcer (52), but studies in children are limited. As such,few data are available on the effectiveness of therapy inchildren, testing after treatment is recommended in thosewith complicated peptic ulcer disease (i.e., bleeding, per-foration, or obstruction) or lymphoma. For patients whoremain symptomatic, it is recommended that endoscopyand biopsy be performed to evaluate for the persistenceof H. pylori-associated peptic ulcer disease. For patientswith an uncomplicated ulcer who are asymptomatic aftercompletion of eradication therapy, testing for persistenceof infection is not necessary. However, some physiciansadvocate the use of urea breath testing in this clinicalsetting.

WHEN IS TREATMENT OF H. PYLORIINFECTION INDICATED?

Eradication therapy is recommended for children whohave both known active H. pylori infection and symp-tomatic gastrointestinal disease. Known active H. pyloriinfection is defined as identification of the organisms byhistopathologic examination or as a positive culture fromendoscopic gastric biopsy. Serology is not a reliable testfor active disease, because it may indicate past but notcurrent infection with H. pylori.

There are no randomized controlled trials in childrenthat determine the precise clinical settings in whicheradication therapy is indicated. Although additionalstudies in children are needed (53), the available evi-dence supports the following recommendations.

GOLD ET AL.494


Duodenal and Gastric Ulcers

Eradication treatment is recommended for childrenwho have a duodenal ulcer or gastric ulcer identified atendoscopy and H. pylori documented by histopathology.A prior history of duodenal or gastric ulcer disease is alsoan indication for treatment if active H. pylori infection isdocumented. If a definitive ulcer is present on contrastradiography (e.g., an ulcer crater is present), eradicationtherapy is indicated if either a noninvasive or invasivetest result is positive for H. pylori.

Lymphoma

The rare child with pathologic evidence of MALTlymphoma and H. pylori infection should be treated witheradication therapy. Further studies of pediatric patientswith lymphoma should be performed to monitor the re-currence, progression, or remission of the tumor aftertherapy.

Atrophic Gastritis With Intestinal Metaplasia

Eradication treatment is recommended for the rarechild who has pathologically proven atrophic gastritiswith intestinal metaplasia, according to the updated Syd-ney classification of gastritis (54), plus coexisting H.pylori infection. Because of the preneoplastic nature ofthese pathologic changes, follow-up endoscopy is rec-ommended to confirm that the H. pylori infection hasbeen eradicated and to ensure that there is no subsequentprogression of gastric mucosal disease.

Gastritis Without Peptic Ulcer Disease

The finding of H. pylori–associated gastritis in theabsence of peptic ulcer disease during diagnostic endos-copy poses a dilemma for the endoscopist. The decisionto treat H. pylori–associated gastritis without duodenal orgastric ulcer in this situation is subject to the judgment ofthe clinician and deliberations with the patient and fam-ily. Studies in adults on the effect of eradication treat-ment on abdominal symptoms have produced conflictingresults (55–58). There are no randomized controlled tri-als in children. The long-term impact of the eradicationof H. pylori and the healing of gastritis on the subsequentdevelopment of peptic ulcer disease, adenocarcinoma, orlymphoma is uncertain. Although there is a small life-time risk of development of peptic ulcer disease associ-ated with H. pylori gastritis, there are no randomizedcontrolled trials demonstrating that eradication of H. py-lori results in prevention of peptic ulcer disease. In ad-dition, there are no data showing that eradication therapyinfluences the long-term risk for development of gastriccancers. Antibiotic treatment can result in adverse drugreactions, promote antibiotic resistance, and increase the

cost of care. Therefore, the H. pylori Infection GuidelineCommittee concludes that there is insufficient evidenceto support either initiating or withholding eradicationtreatment in this situation.

Recurrent Abdominal Pain andAsymptomatic Children

There is no compelling evidence, at the present time,for treating children with H. pylori infection and eithernonulcer dyspepsia or functional recurrent abdominalpain. There is also no convincing evidence currentlyavailable that asymptomatic children who have a familymember with H. pylori infection, peptic ulcer, or gastriccancer need treatment.

WHAT IS THE PREFERRED TREATMENT OFH. PYLORI INFECTION IN CHILDREN?

The optimum treatment regimen for eradicating H. py-lori in children has not been determined (59). Effectivetherapy in adults is defined as successful eradication ofH. pylori infection in a minimum of 80% of treated sub-jects (60). Although it appears that treatment options thathave been effective in adults will also be efficacious inchildren, controlled studies in pediatric populations areneeded to confirm or refute this supposition. Unfortu-nately, the limited data currently available in children areopen-label, case series and uncontrolled, anecdotal ob-servations that do not meet the minimum criteria fordetermining efficacy. In vitro sensitivity of H. pylori to aspecific drug does not guarantee that the bacterium willbe effectively eradicated from the human stomach.Therefore, current treatment strategies to eradicate H.pylori have been developed primarily by trial-and-errormethodology (61).

The single most important determinant of successfuleradication therapy is compliance with the prescribedcombination treatment regimen (62). There are well-described treatment failures due to suboptimal compli-ance. To enhance adherence to the treatment regimen,the number of medications prescribed, the frequency ofadministration, and the duration of therapy are best keptto the minimum required for successful treatment.

It is recommended that initial treatment consist ofthree medications, administered twice daily, for 1 to 2weeks (63). Specifically, as shown in Table 3, three first-line therapy options are recommended for use in childrenand adolescents. For patients in whom initial treatmenthas failed, two other options are recommended, includ-ing one option with four medications. It is recommendedthat monotherapy and two-drug regimens be avoided,because they are ineffective and increase the likelihood



of acquired antibiotic resistance (64). Primary antimicro-bial resistance also can result in treatment failure evenwhen a three- or four-drug regimen is used. Resistance ofH. pylori to nitroimidazoles causes an increase in the rateof treatment failures in regimens using metronidazole.An increasing prevalence of resistance to clarithromycin,documented in the past few years, particularly in Europe,could eventually impair the therapeutic effectiveness ofthis antibiotic in H. pylori treatment regimens. Results insome studies suggest that prior therapy with a proton

pump inhibitor also reduces the effectiveness of eradica-tion treatment protocols. Studies are needed to determinethe relative importance of these risk factors in pediatricpopulations.

REFERENCES

1. Ernst PB, Gold BD. Helicobacter pylori in childhood: new insightsinto the immunopathogenesis of gastric disease and implicationsfor managing infection in children. J Pediatr Gastroenterol Nutr.1999;28:462–73.

2. Drumm B. Helicobacter pylori in the pediatric patient. Gastroen-terol Clin North Am. 1993;22:169–82.

3. Dohil R, Hassall E, Jevon G, et al. Gastritis and gastropathy ofchildhood. J Pediatr Gastroenterol Nutr. 1999;29:378–94.

4. Huang J-Q, Sridhars CY, Hunt RH. Meta-analysis of the relation-ship between Helicobacter pylori seropositivity and gastric cancer.Gastroenterology. 1998;114:1169–79.

5. Watanabe T, Tada M, Nagai H, et al. Helicobacter pylori infectioninduces gastric cancer in Mongolian gerbils. Gastroenterology.1998;115:642–8.

6. Webb PM, Knight T, Greaves S, et al. Relation between infectionwith Helicobacter pylori and living conditions in childhood: evi-dence for person to person transmission in early life. BMJ. 1994;308:750–3.

7. Parsonnet J. The incidence of Helicobacter pylori infection. Ali-ment Pharmacol Ther. 1995;9:45–51.

8. Handt LK, Fox JG, Stalis IH, et al. Characterization of felineHelicobacter pylori strains and associated gastritis in a colony ofdomestic cats. J Clin Microbiol. 1995;33:2280–9.

9. Grubel P, Hoffman JS, Chong FK, et al. Vector potential of house-flies (Musca domestica) for Helicobacter pylori. J Clin Microbiol.1997;35:1300–3.

10. Klein PD, Graham DY, Gaillour A, et al. Water source as a riskfactor for Helicobacter pylori infection in Peruvian children. Lan-cet. 1991;337:1503–6.

11. Staat MA, Kruszon-Moran D, McQuillan GM, et al. A population-based serologic survey of Helicobacter pylori infection in childrenand adolescents in the United States. J Infect Dis. 1996;174:1120–3.

12. Goodman KJ, Correa P. The transmission of Helicobacter pylori:a critical review of the evidence. Int J Epidemiol. 1995;24:875–87.

13. Sackett DL, Richardson WS, Rosenberg W, et al. Evidence-basedMedicine: How to Practice and Teach EBM. Edinburgh: ChurchillLivingston; 1998.

14. McMurray AR. Three decision-making aids: brainstorming, nomi-nal group, and Delphi technique. J Nurs Staff Dev. 1994;10:62–5.

15. Cockeram AW. Clinical practice guidelines: help or hindrance? JPediatr Gastroenterol Nutr. 1999;28:362–3.

16. Canadian Task Force on the Periodic Health Examination: theperiodic health examination. Can Med Assoc J. 1979;121:1193–254.

17. Olson AD, Fendrick AM, Deutsch D, et al. Evaluation of initialnoninvasive therapy in pediatric patients presenting with suspectedulcer disease. Gastrointest Endosc. 1996;44:554–1.

18. Rothman KJ, Greenland S. Modern Epidemiology. 2nd ed. Phila-delphia: Lippincott–Raven; 1998.

19. Genta RM, Graham DY. Comparison of biopsy sites for the his-topathologic diagnosis of Helicobacter pylori: a topographic studyof H. pylori density and distribution. Gastroenterology. 1997;112:2108–10.

20. Elitsur Y, Neace C. Detection of Helicobacter pylori organisms byHp-fast in children. Dig Dis Sci. 1999;44:1169–72.

21. Elitsur Y, Hill I, Lichtman SN, et al. Prospective comparison ofrapid urease tests (Pyloritek, CLO test) for the diagnosis of Heli-cobacter pylori infection in symptomatic children: a pediatric mul-ticenter study. Am J Gastroenterol. 1998;93:217–9.

TABLE 3. Recommended eradication therapies for H. pyloridisease in children

First-lineoptions Medications Dosage

1 amoxicillin 50 mg/kg/day up to1 g bid

clarithromycin 15 mg/kg/day up to500 mg bid

proton pump inhibitor:omeprazole (or comparableacid inhibitory doses ofanother PPI)

1 mg/kg/day up to 20mg bid

2 amoxicillin 50 mg/kg/day up to1 g bid

metronidazole 20 mg/kg/day–500mg bid



3 clarithromycin 15 mg/kg/day up to500 mg bid

metronidazole 20 mg/kg/day up to500 mg bid



Second-line options4 bismuth subsalicylate 1 tablet (262 mg) qid

or 15 ml (17.6mg/mL qid)

metronidazole 20 mg/kg/day–500mg bid

proteon pump inhibitor:omeprazole (or comparableacid inhibitory doses ofanother PPI)


pus, an additional antibiotic:amoxicillin

50 mg/kg/day up to1 g bid

or tetracyclinea 50 mg/kg/day up to1 g bid

or clarithromycin 15 mg/kg/day–500mg bid

5 ranitidine bismuth-citrate 1 tablet qidclarithromycin 15 mg/kg/day–500

mg bidmetronidazole 20 mg/kg/day–500

mg bid

Initial treatment should be provided in a twice daily regimen (toenhance compliance) for 7 to 14 days.

a Only for children 12 years of age or older.bid, twice daily; qid, four times daily.

GOLD ET AL.496


22. van der Hulst RW, van der Ende A, Homan A, et al. Influence ofmetronidazole resistance on efficacy of quadruple therapy for Heli-cobacter pylori eradication. Gut. 1998;42:166–9.

23. Holton J. Clinical relevance of culture: why, how, and when. Heli-cobacter. 1997;2(suppl 1):S25–33.

24. Westblom TU. Molecular diagnosis of Helicobacter pylori. Immu-nol Invest. 1997;26:163–74.

25. Breslin NP, O’Morain CA. Noninvasive diagnosis of Helicobacterpylori: a review. Helicobacter. 1997;2:111–7.

26. De Oliveira AMR, Rocha GA, Queiroz DMM, et al. Evaluation ofenzyme-linked immunosorbent assay for the diagnosis of Helico-bacter pylori infection in 157 children from different age groupswith and without duodenal ulcer. J Pediatr Gastroenterol Nutr.1999;28:157–61.

27. Czinn SJ. Serodiagnosis of Helicobacter pylori in pediatric pa-tients. J Pediatr Gastroenterol Nutr. 1999;28:132–4.

28. Khanna B, Cutler A, Israel NR, et al. Use caution with serologictesting for Helicobacter pylori infection in children. J Infect Dis.1998;178:460–5.

29. Fallone CA, Elizov M, Cleland P, et al. Detection of Helicobacterpylori infection by saliva IgG testing. Am J Gastroenterol. 1996;91:1145–9.

30. Alemohammad MM, Foley TJ, Cohen H. Detection of immuno-globulin G antibodies to Helicobacter pylori in urine by an enzymeimmunoassay method. J Clin Microbiol. 1993;31:2174–7.

31. Vaira D, Malfertheiner P, Megraud F, et al. Diagnosis of Helico-bacter pylori with a new non-invasive antigen-based assay. Lancet.1999;354:30–3.

32. Oderda G, Rapa A, Ronchi B, et al. Detection of Helicobacterpylori in stool specimens by non-invasive antigen enzyme immu-noassay in children: multicentre Italian study. BMJ. 2000;320:347–8.

33. Cutler AF, Havstad S, Ma CK, et al. Accuracy of invasive andnoninvasive tests to diagnose Helicobacter pylori infection. Gas-troenterology. 1995;109:136–41.

34. Rowland M, Lambert I, Gormally S, et al. Carbon 13-labeled ureabreath test for the diagnosis of Helicobacter pylori infection inchildren. J Pediatr. 1997;131:815–20.

35. Bode G, Rothenbacher D, Brenner H, et al. Variation in the 13C-urea breath test value by nationality in Helicobacter pylori-infectedchildren. Scand J Gastroenterol. 1998;33:468–72.

36. Jones NL, Bourke B, Sherman PM. Breath testing for Helicobacterpylori infection in children: a breath of fresh air? J Pediatr. 1997;131:791–3.

37. Sherman PM, Hassall E, Hunt RH, et al. Canadian Helicobacterstudy group consensus conference on the approach to Helicobacterpylori infection in children and adolescents. Can J Gastroenterol.1999;13:553–9.

38. Hunt R, Thomson AB. Canadian Helicobacter pylori consensusconference. Can J Gastroenterol. 1998;12:31–41.

39. Van der Meer SB, Forget PP, et al. The prevalence of Helicobacterpylori serum antibodies in children with recurrent abdominal pain.Eur J Pediatr. 1992;151:799–801.

40. McCallion WA, Bailie AG, Ardill JE, et al. Helicobacter pylori,hypergastrinemia, and recurrent abdominal pain in children. J Pe-diatr Surg. 1995;30:427–9.

41. Chong SK, Lou Q, Asnicar MA, et al. Helicobacter pylori infec-tion in recurrent abdominal pain in childhood: comparison of di-agnostic tests and therapy. Pediatrics. 1995;96:211–5.

42. Bode G, Rothenbacher D, Brenner H, et al. Helicobacter pylori andabdominal symptoms: a population based study among preschoolchildren in southern Germany. Pediatrics. 1998;101:634–7.

43. O’Donahoe JM, Sullivan PB, Scott R, et al. Recurrent abdominal

pain and Helicobacter pylori in a community-based sample ofLondon children. Acta Paediatr. 1996;85:961–4.

44. Hardikar W, Feekery C, Smith A, et al. Helicobacter pylori andrecurrent abdominal pain in children. J Pediatr Gastroenterol Nutr.1996;22:148–52.

45. Macarthur C, Saunder N, Feldman W. Helicobacter pylori, gas-troduodenal disease, and recurrent abdominal pain in children.JAMA. 1995;273:729–34.

46. Gormally SM, Prakash N, Durnin MT, et al. Association of symp-toms with Helicobacter pylori infection in children. J Pediatr.1995;126:753–6.

47. Hardikar W, Davidson PM, Cameron DJ, et al. Helicobacter pyloriinfection in children. J Gastroenterol Hepatol. 1991;6:450–4.

48. Glassman MS, Schwarz Sm, Medow MS, et al. Campylobacterpylori-related gastrointestinal disease in children. Incidence andclinical findings. Dig Dis Sci. 1989;34:1501–4.

49. Reifen R, Rasooly I, Drumm B, et al. Helicobacter pylori infectionin children: is there specific symptomatology. Dig Dis Sci. 1994;39:1488–92.

50. Snyder JD, Hardy SC, Thorne GM, et al. Primary antral gastritis inyoung American children: low prevalence of Helicobacter pylori.Dig Dis Sci. 1994;39:1859–63.

51. Leontiadis GI, Sharma VK, Howden CW. Non-gastrointestinaltract associations of Helicobacter pylori infection. What is theevidence? Arch Intern Med. 1999;159:925–40.

52. Howden CW. For what conditions is there evidence-based justifi-cation for treatment of Helicobacter pylori infection? Gastroen-terology. 1997;113:S107–112.

53. Sherman PM, Hunt RH. Why guidelines are required for treatmentof Helicobacter pylori infection in children. Clin Invest Med. 1996;19:362–7.

54. Dixon MF, Genta RM, Yardley JH, et al. Classification and grad-ing of gastritis: the updated Sydney system. Am J Surg Pathol.1994;20:1161–81.

55. Blum AL, Talley NJ, O’Morain C, et al. Lack of effect of treatingHelicobacter pylori infection in patients with nonulcer dyspepsia:omeprazole plus clarithromycin and amoxicillin effect one yearafter treatment (OCAY) study group. N Engl J Med. 1998;339:1875–81.

56. McColl K, Murray L, El-Omar E, et al. Symptomatic benefit fromeradicating Helicobacter pylori infection in patients with nonulcerdyspepsia. N Engl J Med. 1998;339:1869–74.

57. Talley NJ, Janssens J, Lauritsen K, et al. Eradication of Helico-bacter pylori in functional dyspepsia: randomized double blindplacebo controlled trial with 12 months follow up. The OptimalRegimen Cures Helicobacter Induced Dyspepsia (ORCHID) studygroup. BMJ. 1999;318:833–7.

58. Talley NJ, Vakil N, Ballard D, et al. Absence of benefit fromeradicating Helicobacter pylori in patients with nonulcer dyspep-sia. N Engl J Med. 1999;3412:15:1106–11.

59. Blecker U, Gold BD. Treatment of Helicobacter pylori infection:a review. Pediatr Infect Dis J. 1997;16:391–9.

60. Harris A. Current regimens for treatment of Helicobacter pyloriinfection. Br Med Bull. 1998;54:195–205.

61. Peura D. Helicobacter pylori: rational management options. Am JMed. 1998;105:424–30.

62. Huang H-Q, Hunt RH. Treatment after failure: the problem ofnon-responders. Gut. 1999;45:140–5.

63. Rowland M, Imrie C, Bourke B, et al. How should Helicobacterpylori infected children be managed? Gut. 1999;45:1336–9.

64. Behrens R, Lang T, Keller KM, et al. Dual versus triple therapy ofHelicobacter pylori infection: results of a multicentre trial. ArchDis Child. 1999;81:68–70.



Health & Medicine

Diagnosi e trattamento delle infezioni da H. Pylori in pediatria