Although Helicobacter pylori infection is now recognised to be typically acquired during childhood,[1–6] the natural history of the infection remains poorly understood. H. pylori is present in most cases of children with primary gastritis, implicating H. pylori as the aetiological agent in chronic antral gastritis in children.[7,8] Although the prevalence of peptic ulcer disease in children is very low, the association between H. pylori infection and peptic ulcer disease has been established in children as in adults.[8] Because H. pylori infection is most prevalent during childhood, there is a need for all physicians to become knowledgeable regarding the infection. The focus of the present article is on the epidemiology of H. pylori infection, symptoms and clinical manifestations associated with H. pylori, and diagnosis and management of the infection in children.

1. Epidemiology

H. pylori infection is a chronic infection. Early data suggests that, once acquired, it is often lifelong. The prevalence of H. pylori in a community depends on the rate of acquisition (i.e. incidence of the infection) and the rate of loss of the infection. Among children, the rate of acquisition of H. pylori varies between, and within, populations.[9–14] It has been established that the prevalence of H. pylori is inversely related to socioeconomic status,[9,12,15–17] with the major variable being socioeconomic status during childhood, the period of highest risk. Attempts to understand the different rates of infection in defined groups have focused on differences in socioeconomic status defined by occupation, family income level and living conditions. Each of these variables measures a different component of the socioeconomic complex.

Within the US, studies in a cohort of Blacks and White Hispanics examined the relationship between current and childhood socioeconomic status and the prevalence of H. pylori infection.[1] There was an inverse correlation between socioeconomic status during childhood and the prevalence of H. pylori infection, irrespective of the present social class. For populations in which the social class is more or less homogeneous, such as in China and Russia, the density of living conditions has been shown to be the most significant risk factor.[3,10] A study of the effect of childhood conditions in a sample of monozygotic and dizygotic twins, reared together or apart in different socioeconomic status, revealed that the strongest effects on the acquisition of H. pylori were the density of living and low household income during childhood.[18]

Cross-sectional studies have consistently indicated a gradual increase in H. pylori seroprevalence with age,[9–16] which has been interpreted as a reflection of the fall in the rate of acquisition in successive generations of children as sanitation improved and standards of living increased.[19,20] It is now apparent that the decrease in the prevalence of H. pylori infection in industrialised countries is related to improved living conditions during childhood.[5] Although H. pylori infection is a chronic, and possibly a lifelong, infection,[20] spontaneous elimination of the infection has been reported.[2] A recent longitudinal study that examined Black and White children living in the same community and attending the same schools over a 12-year period found that the rates of acquisition and disappearance of H. pylori were identical for the total population.[2] However, there were differences between the 2 races as more Black children remained infected while more White children lost the infection during the observation period. The study suggested that the higher rate of acquisition and the lower rate of loss of infection among Black children might be because of differences in access to healthcare facilities or a more intense exposure. Another 9-year follow-up study conducted on children and adults from a typical mountain village in Japan also found that the rate of disappearance was greater than the rate of acquisition.[5] The study reported that there was a fall in the prevalence during the observation period which did not reflect changes in the rate of acquisition but rather the higher rate of loss of infection. These changes may be related to changes in medical practices leading to the more frequent use of antimicrobials for other common infections.

The human is the only known host reservoir for H. pylori infection.[20] The mode of transmission is probably person-to-person, but whether the route is oral-oral or faecal-oral transmission is still unclear.[21–28] H. pylori can reach the oral cavity via reflux of gastric juice and can be found in saliva and dental plaque;[29,30] however, dental workers are not at high risk of acquiring the infection.[31] The organism may also survive in faeces[32,33] which could be the vehicle of transmission through anything contaminated by them (e.g. food or water). Studies have reported that unclean water supplies in Peru[34] and consumption of fresh vegetables grown using human wastes for fertiliser in Chile[35] have both been associated with acquisition of H. pylori infection. Although water-borne sources of H. pylori infection may be important in some developing countries,[34] it does not play a significant role in transmission of the infection in developed countries.[16]

Genetic differences in susceptibility to acquire H. pylori infection and outcome of the infection have also been reported.[36–38] A study on monozygotic and dizygotic twins investigated the importance of genetic and environmental factors on the acquisition of H. pylori.[36] The results of the twin study confirmed that there was a genetic influence for susceptibility to acquire the infection, but that environmental factors are more important.

There are conflicting results regarding the relationship between H. pylori infection, nutritional factors and growth in children. A study from Colombia indicated that children infected with H. pylori were significantly shorter than their uninfected peers.[39] However, 2 well designed studies, 1 from Nicaragua[40] and the other from South Africa,[41] failed to confirm such associations. Literature on H. pylori and breast-feeding practices is limited. Few studies have hinted at the protective effect of breast milk on acquiring H. pylori infection.[42,43] A prospective study of infants from a Gambian village found that a high level of immunoglobulin A in mothers’ breast milk may protect against infection during the first year of life.[42] A very recent study from the US has provided additional evidence in support of an association between breast-feeding practices and the prevention of acquisition of H. pylori infection (Malaty HM, Graham DY, Logan ND, unpublished data). Although in that study breast-feeding practice was significantly associated with the mother’s education, children who were breast-fed consistently had a lower rate of H. pylori infection than children who were not breast-fed, regardless of the mother’s education (fig. 1).

Fig. 1
figure 1

Effects of maternal factors on Helicobacter pylori infection in children. Each pair of bars represents the prevalence of H. pylori infection in children in relation to breast-feeding practices and maternal education (Malaty HM, Graham DY, Logan ND, unpublished data).

Full size image

2. Clinical Manifestations

Primary gastritis is uniformly associated with H. pylori infection, which implicates H. pylori as an aetiological agent in chronic gastritis in adults and children.[7,8] As in adults, most infected children who have gastritis are asymptomatic. A study from Arkansas reported a 30% infection rate among healthy children with no upper gastrointestinal symptoms.[12] The endoscopic appearance of gastric mucosa in children with H. pylori colonisation can vary from completely normal to erosions, ulceration and nodularity.[44]

H. pylori is also an established organism for both duodenal ulcer and gastric ulcer disease in children and adults.[8,9,20] However, among infected children with H. pylori, peptic ulcer disease is uncommon.[45] There is a distinction between primary and secondary ulcer in children. Secondary ulcers are usually caused from severe stress or medication and patients usually present with acute haematemesis. Primary ulcer is very rare, particularly before adolescence, and mostly occurs in the duodenum.[45] H. pylori infection accounts for 90 to 100% of paediatric duodenal ulcers.[8] Kilbridge et al.[46] reported an 89% infection rate among 9 paediatric patients with duodenal ulcers. If H. pylori is not effectively treated, duodenal ulcers in infected children can persist or relapse with long term morbidity.[47,48] There are no significant studies that have reported an association between H. pylori and gastric ulcer disease in children because of the very low rate of gastric ulcers in this population. A study conducted by Prieto et al.[49] investigated 270 children who underwent endoscopy; 12 had gastric ulcers, of whom 75% were infected with H. pylori.

Worldwide, H. pylori infection may affect up to 50% of children with recurrent abdominal pain (RAP).[50,51] Apley’s criteria for RAP are still considered the gold standard, defining it as ‘the occurrence of 3 separate pain episodes over a 3-month period which are severe enough to interfere with normal activity’.[52] RAP affects approximately 10 to 15% of children and adolescents.[53] However, there is weak and inconsistent evidence of an association between RAP and H. pylori infection. Although a study from Germany[54] found no difference regarding abdominal pain between infected and noninfected children, a study from Peru[55] found a significant association between H. pylori infection and RAP among young children enrolled in the study. A study that evaluated 80 children with RAP reported a 54% infection rate among these children.[51] At 2 months after anti-H. pylori therapy, they observed that all of the children cured of H. pylori became asymptomatic compared with those with persistent infection who still had symptoms. Interestingly, another study reported that a course of triple therapy improved RAP symptoms significantly in children, irrespective of the elimination of the infection.[56]

Because of the conflicting results of the association between RAP and H. pylori in childhood, there is considerable debate regarding the treatment of infected children with RAP, and as yet there are no uniform guidelines regarding this issue. Therefore, clinicians and primary care physicians should make all possible efforts to convey to parents that RAP in their child may have a multifactorial origin and that H. pylori infection may be only one of the possible contributing elements to their symptoms. Testing for H. pylori infection in children with RAP is appropriate only when treatment is planned if the test is positive. Well controlled multicentre prospective studies on eradication of H. pylori from infected children with RAP, with long term follow-up, are required to answer these questions.

3. Diagnostic Methods

The link between H. pylori and a number of important gastrointestinal diseases prompted development of a variety of methods to detect the presence of the infection. The same tests to diagnose H. pylori infection in adults are applied to children. Diagnostic tests for H. pylori can be categorised into 2 groups; noninvasive tests and invasive tests that require endoscopy. As in adults, the gold standard procedure for the diagnosis of H. pylori infection is the histological examination of mucosal biopsies using special stains (Diff-Quick, EL-Zimaity or Genta stains).[57,58] An additional specimen could be taken, during the endoscopy procedure, from the greater curvature of the prepyloric antrum for rapid urease testing (‘CLO Test’). Obtaining gastric biopsies requires the child to have an endoscopy procedure under deep sedation or general anaesthesia. Although a study from Ireland reported concerns regarding the small biopsy specimen obtained from paediatric endoscopy,[7] others have reported that biopsy size was not as critical for the diagnosis of H. pylori infection.[59]

Noninvasive tests for the diagnosis of H. pylori infection are desirable in the paediatric setting. However, these tests cannot distinguish between H. pylori gastritis and peptic ulcer disease. Serological tests and the urease breath test fall into this category.

Laboratory-based immunoglobulin G (IgG) tests are typically done using multiwell enzyme-linked immunosorbent assays (ELISA) [e.g. ‘HM-CAP’], which have proven to be accurate in symptomatic children.[60] ELISA is a technique that measures IgG antibodies to H. pylori and is an indicator of current or past infection. However, there is still an argument for its use as a screening test to preselect patients for endoscopy. There are also a number of rapid office-based IgG kits using serum, including ‘QuickVue’ and ‘FlexSure HP’. Rapid whole blood−based IgG tests have also become available but their sensitivity and specificity may be lower than those of serum-based methods.[61] Although the rapid tests are reliable in adults, their use in children remains controversial because of their low sensitivity.[61,62] Children who have recently acquired H. pylori will test negative in serological tests because of the delay in developing a sufficient humoral immune response.[61] Several studies are still investigating the validation of tests based on measuring antibodies in children.

Attempts have been made to provide diagnostic testing that does not require blood sampling. The [13C]urea breath test is the noninvasive method of choice to determine H. pylori status in both children and adults.[63,64] This test is based upon the urease activity of the organism, which splits CO2 from ingested urea. Ingestion of labelled urea allows for the labelled CO2 produced in this reaction to be detected in the breath. The [13C]urea breath test is nonradioactive and it is an easy and safe test to perform in children. Klein et al.[65] recently determined a single cutoff value for children of all ages by calculating the urea hydrolysis rate (UHR) independently of the differences in anthropometric measurements.

4. Management

Several different regimens have been evaluated for the treatment of H. pylori infection in adults, but to date there is no standard treatment or guidelines for the management of H. pylori infection in children. It is still debatable whether treating infected asymptomatic children is beneficial, since the reinfection rate and the exact age of acquisition of H. pylori infection have not been identified. However, therapy to cure the infection should be started in all children with peptic (duodenal or gastric) ulcer who are still infected. Successful eradication of H. pylori with antibacterial therapy has resulted in the cure of duodenal ulcers and lowering of the ulcer relapse rate in children.[66,67] The ideal anti-H. pylori regimen should be safe, cheap, easy to comply with, well tolerated by children and able to achieve a high cure rate.

Studies involving children have investigated different treatment regimens (monotherapy and dual or triple therapies) for durations ranging from 1 to 6 weeks. There was a wide variation in eradication rates in these paediatric studies, ranging from 27% to 94% when the treatment regimen lasted 4 to 6 weeks.[67–69] Triple therapy yielded a higher cure rate than dual therapy or monotherapy. 1-week and 2-week regimens of triple therapy that aim to optimise compliance and cure rates are summarised in table I.

Table I
figure Tab1

Clinical trials of Helicobacter pylori eradication regimens in paediatric patients

Full size image

Walsh et al.[73] in Ireland evaluated a 1-week treatment regimen of bismuth subcitrate, metronidazole and clarithromycin (at dosages adapted for a 1-week course), and reported a 95% cure rate among 22 children infected with H. pylori. Concern has been expressed about the use of bismuth in the treatment of H. pylori gastritis in children because of reports of bismuth toxicity in adults.[72,74] However, De Giacomo et al.[69] could not confirm these reports; no children who received colloidal bismuth subcitrate with amoxicillin for treatment of H. pylori gastritis displayed toxicity attributable to bismuth administration. Another study used a 2-week regimen of metronidazole and clarithromycin with omeprazole to achieve a 93% cure rate of infection in 15 Canadian children.[75]

Dual therapy yields a relatively better cure rate of H. pylori infection in children than in adults, perhaps because of the lower rate of antimicrobial resistance in children. In 3 earlier studies conducted in Italy, Oderda et al.[67,71,76] investigated the combination of amoxicillin and tinidazole for 6 weeks in the treatment of H. pylori−associated gastritis. Among the total of 67 children who completed the 6-month follow-up in the 3 studies combined, 54 (81%) were cured of the infection. For long term cure of infection, the combination of amoxicillin and tinidazole for 6 weeks appears to be more effective than using amoxicillin alone. Combination of amoxicillin with H2-antagonists or omeprazole is not recommended to treat children with H. pylori infection because of the low cure rate.[77,78]

There are several important factors influencing the efficacy of therapy in paediatric patients, including poor compliance, adverse effects and development of antimicrobial resistance. Although prospective studies have examined the effect of such factors on the efficacy of treatment regimens in the adult population,[79] there is a lack of such studies in the paediatric setting. There is a great need for well designed and controlled clinical trials on large numbers of H. pylori−infected children to evaluate different regimens for treatment. Follow-up studies to measure reinfection rates are also needed.

5. Vaccine Development

Extensive research has been directed towards the development of vaccination against H. pylori for prevention and possibly treatment of an existing infection. Czinn and Nedrud[80,81] reported the possibility of oral immunisation in animal models. Several animal models have been used over the past few years and many H. pylori antigens have been tested, but no single antigen has been fully successful. Examination of mixtures of antigens is under way, but the final formulation of a vaccine is not yet fully defined. The target population for prophylactic vaccination is children residing in areas with a high incidence of gastric carcinoma. In developing countries and areas with a high rate of infection among children, vaccination may be a valuable treatment regimen. Establishing the value of therapeutic immunisation alone or combined with anti-H. pylori treatment remains an important objective for research.

6. Conclusions

H. pylori infection is common in both adults and children, and H. pylori−associated gastritis and peptic ulcer disease have been reported in both age groups. H. pylori infection causes 90 to 100% of paediatric duodenal ulcers. If H. pylori is not effectively treated, duodenal ulcers in infected children can persist or relapse with long term morbidity. There is still only weak and inconsistent evidence for an association between RAP in children and H. pylori infection. Evidence to date indicates that routine investigation for H. pylori infection in children with RAP is not warranted. Therefore, testing for H. pylori infection in children with RAP is appropriate only when treatment is planned if the test is positive. Current H. pylori eradication treatment for children is based on regimens for adults, using triple therapy or dual therapy with a proton pump inhibitor. An ideal treatment for H. pylori infection in children has yet to be determined.

Acknowledgements

This study was supported by the Department of Veterans Affairs and the generous support of Hilda Schwartz.