The changing prevalence of Helicobacter pylori infection in Canadian children: Should screening be performed in high-risk children?

Department of Pediatrics, Division of Gastroenterology, British Columbia Institute for Children’s and Women’s Health, University of British Columbia, Vancouver, British Columbia Correspondence: Dr Kevan Jacobson, Division of Gastroenterology, British Columbia Children’s Hospital, 4480 Oak Street, Room K4-181, Vancouver, British Columbia V6H 3V4. Telephone 604-875-2332, fax 604-875-3244, e-mail kjacobson@cw.bc.ca K Jacobson. The changing prevalence of Helicobacter pylori infection in Canadian children: Should screening be performed in high-risk children? Can J Gastroenterol 2005;19(7):412-414.

The changing prevalence of Helicobacter pylori infection in Canadian children: Should screening be performed in high-risk children?

PREVALENCE OF HELICOBACTER PYLORI
Helicobacter pylori infection, frequently acquired during childhood and adolescence, is linked to socioeconomic status and living conditions (1-3).While the prevalence of H pylori is decreasing in developed countries (4), it remains high in developing countries, in populations of low socioeconomic status and in immigrant populations within developed countries (5)(6)(7).
Seroepidemiological studies of H pylori-infected children in western Europe (8)(9)(10), the United States (11)(12)(13) and, to a limited degree, in Canada, have demonstrated variability in H pylori seropositivity (5,6).While these data are useful and convenient for the evaluation of large populations, they must also be considered in the context of geographical location, population studied, lack of appropriate symptom assessment instruments, lack of clearly defined controls, accuracy of the serological assay (13) and lack of confirmation of H pylori infection with the urease breath test (UBT) or endoscopy with biopsy.The latter is important because a positive serology indicates either a past or present H pylori infection.
In a recent, multisite, cross-sectional prospective study (13) involving 992 children from 12 children's hospitals or medical centres predominantly in the eastern and southeastern regions of the United States, the overall prevalence of H pylori antibodies was 17.2%.A wide geographical variation in seropositive rates was observed, with rates ranging from 8.3% (one of 12) to 38.6% (39 of 101) at different centres.Among symptomatic children referred to pediatric gastrointestinal (GI) clinics with abdominal pain (n=182), abdominal pain and vomiting (n=168), or vomiting alone (n=12), the seropositivity rate was 22.5%.In contrast, among children (n=619) seen in non-GI outpatient clinics (who had no GI complaints and who required blood drawing as part of standard clinical management) during the same period, the seropositivity rate was 14.1%.Although the seropositive rates were significantly different between the two groups, the GI referral patients were older than the non-GI patients (mean age 10.1 years [range one to 18 years] versus mean age 7.7 years [range two months to 18 years], respectively).In keeping with previous studies (9,11,14), H pylori seropositivity was strongly associated with increasing age, being nonwhite, living in poverty, crowded household conditions, an annual household income less than US$36,000 and the mother's education being grade 12 or less (Table 1).
Similarly, in Canada, H pylori infection rates vary considerably.Low prevalence rates of 5.28% (13 of 246) have been observed in children aged five to 18 years in an ongoing study (prevalence, symptomatology and investigation of pediatric H pylori infection [PSI-PHI] study [unpublished data]) involving four Canadian academic centres where upper GI endoscopy is performed for upper GI symptoms.In contrast, high seroprevalence rates have been observed in indigenous people, with rates ranging from 51% in children from the Arctic communities of Chesterfield Inlet and Repulse Bay, Nunavut (5) to 95% in First Nation's children of the subarctic community Wasagamack, Manitoba (6).In keeping with American studies, in communities with widespread H pylori infection, overcrowding and poor living conditions, including primitive toileting (Arctic community households had flush toilets versus only 0.3% in the Wasagamack community), there was increased likelihood of H pylori infection.In the Arctic communities, antibodies to H pylori were identified in 32% (nine of 28) of children by 15 years of age and 58.9% (63 of 107) of individuals by 29 years of age.However, in the children in Wasagamack, using an enzyme immunoassay for detection of stool antigen, approximately 70% (16 of 23) of the seven-to nine-year-old children were positive.Moreover, approximately 20% of children (five of 23) younger than one year of age were stool antigen-positive.The number of children studied was small and the stool antigen test was not validated with UBT or endoscopy with biopsy.However, these results provide valuable data regarding acquisition of H pylori and experience with stool testing in a population where UBT might be challenging, particularly in younger children, and where endoscopy is not always possible.
Immigrants to Canada provide a further burden of H pylori infection.The immigrant population comprises approximately 200,000 individuals per year with approximately 80% to 90% of immigrants settling in Montreal, Quebec; Toronto, Ontario; or Vancouver, British Columbia.Approximately 70% of immigrants belong to minority groups that contribute to higher prevalence rates and variability of H pylori infection across the country.In a recent study in the United States (7), 226 children adopted from 18 countries (mean age 24±2.05months, range four months to 16 years) demonstrated a seroprevalence rate of 31%.H pylori-positive children were older at arrival in the United States than their H pylori-negative counterparts (40±11 versus 15±1 months for Chinese children; 61±28 versus 24±9 months for Romanian children; and 48±5 versus 22±3 months for Russian children) and were more likely to have resided in orphanages rather than foster care.Sixty-six per cent of children who resided entirely in orphanages, or other institutional settings such as hospitals, were H pylori antibody-positive compared with 22% who lived in institutions and family settings, and 12% who lived only in family settings.That study (7) also demonstrated variability in seropositivity, with higher rates of H pylori antibodies observed in Russian children (49%) compared with Romanian (20%) and Chinese children (16%) (Figure 1).

CHANGING PREVALENCE OF H PYLORI
With improved socioeconomic status and H pylori eradication therapy (whether intentional or the result of another treatment, ie, when proton pump inhibitors are used to treat gastroesophageal reflux disease or antibiotics to treat infectious illness), studies have begun to demonstrate a change in H pylori prevalence.Recently, in a study from Greece (4), H pylori seroprevalance was compared in banked serum from 200 consecutive adult outpatients collected in 1987, from serum samples from 201 similarly selected outpatients from the same unit, and from 120 consecutive blood donors from the same hospital collected in 1997.An overall 10.75% decrease in seroprevalence was demonstrated (from 59.5% to 48.75%).In particular, the decrease in seroprevalence was most marked in the 15 to 24 year old age group.Presently, no published data are available in Canada and the United States; however, if one extrapolates from available data, the decline appears evident.
A 1991 study from Arkansas ( 14) demonstrated a seroprevalence of 45% in children aged three to 20 years.While this rate is higher than the seropositivity rate of 17.2% recently reported by Chong et al (13) (suggesting a decline in seroprevalence), this may simply reflect geographical variability and a different population studied.Moreover, older patients were included in the study by Fiedorek et al (14), adding to the higher prevalence rate.
Children referred to the gastroenterology department of the Hospital for Sick Children, Toronto, Ontario, for upper GI symptoms in the early 1990s were found to have H pylori infection rates of 26% to 43% (15,16)   studies (15,16) were not specifically designed for comparison of trends, and the populations are from different centres, the data nevertheless agree with current trends.The variability in infection rates are likely also influenced by the proportions of subjects who belong to immigrant and First Nations populations.

COMPLICATIONS ASSOCIATED WITH H PYLORI INFECTION
Because H pylori infection contributes to the development of peptic ulcer disease and gastric cancer, the burden of these diseases in the Western world may fall primarily on the high-risk minority groups.).However, it is possible that fewer patients received medical attention and that different diagnostic and therapeutic medical practices were followed, leading to a potential underestimation of the severity of the disease in the Wasagamack community.In contrast, the mean annual ageadjusted gastric cancer incidence rates were statistically similar for treaty-status Aboriginals (11.2 of 100,000) and non-Aboriginals (11.6 of 100,000).Low rates of gastric cancer have also been observed in specific Chinese and African populations associated with a high prevalence of childhood H pylori infection (18,19).While these data require further study, it is possible that the Wasagamack community is infected with strains of H pylori, and has a host genetic profile that is associated with a diminished susceptibility to gastric cancer.
Clearly, studies should be performed on populations at risk.However, the declining prevalence of H pylori infection in the Western world suggests changing patterns of transmission.Therefore, studies should be performed on children in both low-and high-risk geographical areas, and with diverse socioeconomic and ethnic backgrounds.In Canada, studies need to be undertaken to identify populations at risk and to determine whether the high prevalence rates of H pylori infection are associated with high complication rates.Moreover, improving the socioeconomic status and living conditions of populations at risk will reduce the burden of infection and, thus, the burden of disease.
To ensure the validity of such population-based studies, one needs to ensure that the screening tools are both validated and safe.Moreover, we need to determine whether there are indeed consequences to H pylori infection in childhood, whether eradication is associated with reduced H pylori-related consequences, and whether the eradication regimens are safe and if eradication of H pylori infection in childhood produces a significant benefit in adulthood.Finally, testing also requires that we consider an intent-to-treat approach for H pylori-infected individuals.
(Table2); whereas elsewhere, infection rates appear to be decreasing, with lower infection rates of approximately 5% recently observed in Canadian children referred to tertiary care institutions in the ongoing PSI-PHI study (unpublished data).Although these

TABLE 1
While the complications associated with H pylori infection have not been systematically studied in Canada,

TABLE 2
Prevalence of Helicobacter pylori (Hp) infection in Canadian children undergoing upper gastrointestinal (GI) endoscopy for upper GI symptoms from 1990 to 1994