Connect
MJA
MJA

The epidemiology of Helicobacter pylori infection in African refugee children resettled in Australia

Sarah Cherian, David Forbes, Frank Sanfilippo, Angus Cook and David Burgner
Med J Aust 2008; 189 (8): 438-441.
Published online: 20 October 2008

Helicobacter pylori infection is usually acquired in childhood.1,2 Acute infection is often silent, with symptoms and disease manifesting later in life, as does an increased risk of H. pylori-related malignancy.

The prevalence of H. pylori infection is markedly increased in developing countries,3,4 and risk factors include increasing age, large family size and socioeconomic deprivation.5-7 Refugees resettled in developed countries generally come from regions of high H. pylori prevalence, whereas those who have grown up in industrialised nations have a lower prevalence of H. pylori infection.3,8 Increased prevalence is also found in migrants and indigenous populations; adverse socioeconomic conditions in these groups account for some of the excess risk.7,9,10

Western Australia resettles about 1300 humanitarian refugees annually, representing over 10% of Australia’s refugee intake.11 Currently, many families are African, and about half the refugees are children.12

In this study, we investigated the prevalence and epidemiological associations of H. pylori infection in a high-risk paediatric population. The main outcome measure was H. pylori infection diagnosed by monoclonal faecal antigen enzyme immunoassay testing (MFAT). The effects of age, sex, transit through refugee camps, comorbidities and treatment interventions were investigated.

Methods

We conducted a cross-sectional study at the Migrant Health Unit (MHU) in Perth, WA, the sole screening unit for humanitarian refugees resettled in WA. About 80% of targeted refugees in WA receive an initial health assessment at the MHU.12 African children (aged less than 16 years) who presented for initial health assessment between 1 February and 30 November 2006 were included. Blood samples were obtained for routine screening investigations at the first clinic visit as part of standard clinical care and stool samples were collected 1 week later. Children were excluded from our study if they had received antibiotics or proton-pump inhibitors in the preceding month, if they had an immunodeficiency or active tuberculosis. Ethical approval was obtained from the Women and Children’s Ethics Committee, Princess Margaret Hospital for Children. Informed consent was obtained in the presence of trained interpreters, as appropriate.

Data on age, sex, ethnicity, country of last transit, transit period, country of birth, type of dwelling in country of transit, and recent drug administration were obtained at the first visit by means of a structured questionnaire. Breastfeeding history was recorded for children under 2 years of age. Details of premigration antihelminthic and antimalarial treatment were obtained from accompanying International Office of Migration documentation.

Helicobacter pylori diagnosis

Fresh faecal samples were obtained from each child and frozen at 20°C for batch analyses. We assessed H. pylori status using Amplified IDEIA HpStAR kits (Dako, Glostrup, Denmark) following the manufacturer’s instructions and as previously reported.13

Identification of other infections

Details of helminth infection, tinea capitis, tuberculosis and malaria were obtained for each child. Helminth infection was defined as the presence of any of the following results: positive serological test for schistosomiasis and/or strongyloidiasis, positive stool microscopy for ova, cysts or parasites of pathogenic helminths, peripheral eosinophilia (≥ 0.7 × 109/L) or elevated immunoglobulin E (IgE) levels (> 280 kU/L). Pre-migration administration of albendazole was documented in 80% of children, with the remainder receiving empiric albendazole at the first health assessment visit.

A clinical diagnosis of tinea capitis was made at the initial visit based on skin examination. Latent or active tuberculosis infection was diagnosed by QuantiFERON-TB Gold testing (Cellestis International, Melbourne, Vic) in children over 2 years of age, with chest radiographs as indicated. All children had a single blood film and smears and rapid immunochromatographic testing (BinaxNOW, Portland, Ore, USA), irrespective of symptoms or premigration antimalarial treatment.

Statistical analyses

All data were analysed with SPSS, version 14.0 for Windows (SPSS Inc, Chicago, Ill, USA). Continuous variables were compared by independent t tests or Mann–Whitney tests, as appropriate. Associations between categorical variables and H. pylori infection were initially analysed by Pearson χ2 or Fisher’s exact tests. Logistic regression was used to determine the effect of independent variables separately on H. pylori infection, adjusting only for age and sex. Multivariate logistic regression was used to evaluate the effect of covariates on H. pylori infection. Statistical significance was set at the 5% level and two-sided P values were calculated.

Results

Two-hundred and one African refugee children presenting for screening at the MHU were recruited consecutively, with a 100% response rate. Eight children were excluded (five received antibiotics before screening and three were of non-African ethnicity). Of the 193 eligible children, 100 (52%) were male. The mean age was 7.9 years (SD, 4.4 years). Box 1 shows the main demographic features of the cohort.

Eighteen children (9%) were breastfeeding at the time of enrolment (mean age of breastfeeding children, 11.3 months; SD, 5.2 months). There were 116 children (60%) who had lived in refugee camps, with the remainder living in urban dwellings (apartments or houses). Almost all children who transited through Tanzania or Kenya (96 of 97; 99%) had lived in refugee camps, while all 28 who transited through Egypt had lived in apartments. The overall median transit time before resettlement in WA was 5.5 years (interquartile range [IQR], 3.0–8.4). Protracted refugee stays (more than 5 years in transit) were common (98 of 193; 51%); the median transit time for these 98 children was 8.0 years (IQR, 6.5–10.0).

Helicobacter pylori infection

MFAT was performed in 182 children, and H. pylori infection was diagnosed in 149 (82%). MFAT results clearly discriminated between populations that were infected (median positive optical density [OD], 2.85; IQR, 1.18–3.61) and uninfected (median negative OD, 0.10; IQR, 0.08–0.18), with no equivocal results (P < 0.001).

Children with H. pylori infection were significantly older (mean age, 8.5 years [SD, 4.2] v mean age, 5.8 years [SD, 4.5]; P < 0.001) with no sex differences. The prevalence of H. pylori infection was 63% for children under 2 years of age, rising to 95% for those older than 14 years. When analysed by age strata, the odds of infection were more than fourfold higher for children aged over 10 years compared with those aged less than 5 years (odds ratio [OR], 4.42; 95% CI, 1.58–12.35). Where two or more children in a family were enrolled (51 families), 45 of the 51 oldest siblings (88%) had H. pylori infection compared with 35 of the youngest siblings (69%; OR 3.43; 95% CI, 1.21–9.67).

Logistic regression was used to assess the effect of various factors on the odds of H. pylori infection (Box 2). Age was a significant predictor of H. pylori infection, with the odds of infection increasing by 17% for each year of age (OR, 1.17; 95% CI, 1.07–1.28). While the odds of infection were numerically largest for children transiting through Kenya, the overall relationship between country of transit and infection was not significant after adjusting for age and sex. Premigration antimalarial treatment (79 of 182 children; 43%) significantly reduced the odds of H. pylori infection after adjusting for age and sex (OR, 0.31; 95% CI, 0.14–0.72). Multivariate regression showed that only age and premigration antimalarial treatment were significantly associated with H. pylori infection (Box 3).

Effect of other infectious diseases on Helicobacter pylori infection

In total, 76 of the 182 children (42%) had evidence of helminth infection, 15 (8%) had tinea capitis, 16 (9%) had Plasmodium falciparum infection, and 11 of the 153 children (7%) tested had positive QuantiFERON-TB Gold results (with normal chest radiographs), indicative of latent tuberculosis infection. After adjusting for age and sex, the prevalence of H. pylori infection was not affected by the presence any of these infections (Box 2). No difference in IgE levels or peripheral eosinophilia counts were found in children with or without H. pylori infection (data not shown).

Discussion

Our study shows a high prevalence of H. pylori infection in African refugee children, confirming that children from developing countries are at greater risk of infection.14 Our results support the observation that early childhood is the main period of acquisition of H. pylori infection in high-prevalence populations.1,2,4 H. pylori infection was present in 82% of this cohort, and the odds of infection increased significantly with age. In comparison, the prevalence of H. pylori infection in Australian children is low,8 although the prevalence in Australian Aboriginal children is significantly higher, especially in those from remote areas,9 reflecting differences in socioeconomic status.

The protective effect of antimalarial treatment on H. pylori infection is a potentially important and unexpected finding. In children who received empirical premigration antimalarial treatment, this was given about 6 weeks before study enrolment. This correlated with the median time between arrival in Australia and the MHU health screening (Dr A Thambiran, Medical Director, MHU, Perth, WA, personal communication). Empirical premigration antimalarial treatment was ceased in mid 2006 because of concerns about efficacy and coverage.15

The antimalarial therapy may have eradicated existing H. pylori infection. The period between administration of antimalarial drugs and collection of faecal samples was short, and so reacquisition of H. pylori infection during this intervening period is unlikely. The elimination half-lives of pyrimethamine and sulfadoxine are relatively long (3–4 and 6–9 days, respectively) and that of dihydroartemisinin, the active metabolite of artesunate, is less than 1 hour.16 Antimalarial therapy is unlikely to have affected MFAT performance. The effect of antimalarial therapy remained significant in our final regression analyses, and was independent of albendazole therapy. To our knowledge, this in-vivo association has not been previously reported, although artemisinins are known to have antibacterial properties.17 It has been postulated that artemisinin derivatives may interact with iron-dependent bacteria (such as H. pylori) and potentially provide a mechanism for targeted bacterial death.18 The possible therapeutic role of artemisinins, which are cheap and well tolerated, in H. pylori eradication warrants further investigation.

A limitation of our study is the lack of a traditional “gold standard” for the diagnosis of H. pylori infection. Methods based on endoscopy and biopsy, or urea breath testing, are neither practical nor ethical for population-based screening of children, particularly in non-English speaking and often traumatised families. Recent international guidelines now recommend MFAT as an alternative in both adult and paediatric populations.19,20

In this study, we investigated potential epidemiological risk factors that may predispose refugee children to H. pylori infection. Surprisingly, transit through refugee camps did not place children at increased risk of infection, despite harsh environmental and nutritional conditions. The ubiquitous deprivation and overcrowding that characterise urban refugee conditions (eg, Egyptian apartments) may instead contribute to the non-significant association between dwelling type and H. pylori infection.

Intrafamilial spread of H. pylori, particularly mother-to-child transmission2,21,22 or from infected older siblings,2,23 is a potentially important mechanism for acquisition. In our cohort, older siblings had odds of H. pylori infection three times higher than those of their youngest siblings, supporting this premise. Parental H. pylori infection was not assessed in this study, but a high prevalence would be expected, in keeping with analogous published results.22-24 Reliable data on family size could not be obtained, as many siblings and/or parents were displaced in transit.

The relationship between breastfeeding and infection was not found to be statistically significant; however, the number of children being breastfed at the time of enrolment was small. Our study did not address the phenomenon of transient H. pylori infections.25 Over 60% of children aged under 2 years had H. pylori infection, which is similar to other reports of African infants.4 Given the high levels of overall infection in this cohort, reinfection is likely even if there were cases of transient infection in infancy.

Clinicians should be aware of the high prevalence of H. pylori infection in resettled refugee children, including the potential development of chronic complications. Longitudinal studies of this population are warranted.

1 Ethnicity, country of birth and transit profiles of the 193 African refugee children in the study

Ethnicity


Country of birth


Country of last transit


Group

No.

Country

No.

Country

No.

Median transit time (IQR)


Sudanese

66 (34%)

Sudan

54 (28%)

Tanzania

64 (33%)

6.1 years (3.0–9.0)

Burundian

55 (29%)

Tanzania

40 (21%)

Kenya

33 (17%)

5.5 years (4.0–9.4)

Liberian

23 (12%)

Kenya

21 (11%)

Egypt

28 (15%)

2.0 years (1.2–2.9)

Congolese

22 (11%)

Burundi

14 (7%)

Guinea

26 (13%)

5.8 years (4.0–7.3)

Eritrean

20 (10%)

Democratic Republic of Congo

13 (7%)

Sudan

19 (10%)

6.4 years (3.3–9.4)

Other*

7 (4%)

Other

51 (26%)

Other

23 (12%)

7.0 years (3.3–10.0)


IQR = interquartile range. * Other ethnicity: Sierra Leonian (7). Other countries of birth: Liberia (11); Egypt (9); Guinea (9); Zambia (8); Sierra Leone (6); Ethiopia (2); Ghana (2); Zimbabwe (2); Ivory Coast (1); Nigeria (1). Other transit countries: Zambia (11); Ghana (3); Ethiopia (3); Uganda (3); Zimbabwe (2); Nigeria (1).

2 Adjusted odds ratios for factors tested for independent associations with Helicobacter pylori infection

Variable

No. of children

No. infected with H. pylori

Odds ratio* (95% CI)

P


Age (years)

182

149 (82%)

1.17 (1.07–1.28)

0.002

Age strata (sex-adjusted)

0.013

< 5 years

51

35 (69%)

1.0

5–10 years

67

56 (84%)

2.29 (0.95–5.53)

0.064

> 10 years

64

58 (91%)

4.35 (1.55–12.19)

0.005

Sex (age-adjusted)

Female

89

70 (79%)

1.0

Male

93

79 (85%)

1.51 (0.69–3.31)

0.3

Protracted refugee stay (> 5 years)

Yes

98

85 (87%)

1.0

No

84

64 (76%)

1.03 (0.38–2.80)

0.95

Type of dwelling

0.42

Refugee camp

123

99 (81%)

1.0

Apartment

28

22 (79%)

0.96 (0.33–2.74)

0.93

House

31

28 (90%)

2.34 (0.65–8.63)

0.2

Ethnicity

0.35

Sudanese

61

54 (89%)

1.0

Liberian

23

19 (89%)

0.48 (0.12–1.94)

0.3

Congolese

19

16 (84%)

0.61 (0.13–2.28)

0.53

Burundian

52

37 (71%)

0.30 (0.11–0.85)

0.023

Eritrean

20

17 (85%)

0.73 (0.16–3.22)

0.68

Sierra Leonian

7

6 (86%)

0.61 (0.06–6.29)

0.68

Last transit country

0.23

Tanzania

61

45 (74%)

1.0

Kenya

32

31 (97%)

11.3 (1.40–91.28)

0.023

Guinea

26

23 (88%)

2.34 (0.59–9.30)

0.23

Egypt

24

19 (79%)

1.61 (0.49–5.27)

0.44

Sudan

19

16 (84%)

2.02 (0.49–8.20)

0.33

Other

20

15 (75%)

0.96 (0.28–3.24)

0.95

Breastfeeding

Yes

16

10 (63%)

1.0

No

166

139 (84%)

1.12 (0.30–4.21)

0.87

Premigration antihelminthic treatment

Yes

144

119 (83%)

1.0

No

38

30 (79%)

1.05 (0.41–2.68)

0.92

Premigration antimalarial treatment

No

103

91 (88%)

1.0

Yes

79

58 (73%)

0.31 (0.14–0.72)

0.006

Helminth infection

Yes

76

66 (87%)

1.0

No

106

83 (78%)

1.01 (0.40–2.55)

0.98

QuantiFERON-TB Gold test result

Negative

120

99 (83%)

1.0

Positive

11

11 (100%)

Indeterminate

22

17 (77%)

0.85 (0.27–2.71)

0.79

Plasmodium falciparum infection

Yes

16

12 (75%)

1.0

No

166

137 (83%)

2.06 (0.57–7.23)

0.26

Tinea capitis

Yes

15

11 (73%)

1.0

No

167

138 (83%)

1.84 (0.52–6.46)

0.34


* Odds ratios are age- and sex-adjusted by logistic regression. Other transit countries: Zambia (11); Ghana (3); Ethiopia (3); Uganda (3); Zimbabwe (2); Nigeria (1). No odds ratio reported for positive QuantiFERON-TB Gold test results because of small numbers.

3 Multivariate logistic regression model of variables associated with Helicobacter pylori infection

Variable

No. of children

No. infected with H. pylori

Odds ratio (95% CI)

P


Age (years)

182

149 (82%)

1.18 (1.07–1.31)

< 0.05

Premigration antimalarial treatment

No

103

91 (88%)

1.0

Yes

79

58 (73%)

0.33 (0.15–0.75)

< 0.05

  • Sarah Cherian1,2
  • David Forbes1,2
  • Frank Sanfilippo1
  • Angus Cook1
  • David Burgner1,2

  • 1 University of Western Australia, Perth, WA.
  • 2 Princess Margaret Hospital for Children, Perth, WA.


Acknowledgements: 

We thank Dr Aesen Thambiran and the staff of the Migrant Health Unit for their assistance.

Competing interests:

Sarah Cherian received a 2007 Royal Australasian College of Physicians Research Grant and a 2007 University of Western Australia Research Grant for consumables used in this study. Sarah Cherian also received a 2007 Paediatric Research Society of Australia and New Zealand (PRSANZ) travel grant and a 2007 UWA Graduate Travel Award to present related data at the 2007 RACP Congress. Faecal antigen kits were supplied by Dako, Denmark and Oxoid, Australia without charge. Neither company had any influence on the design, analyses, interpretation or content of this manuscript.

  • 1. Malaty HM, El-Kasabany A, Graham DY, et al. Age at acquisition of Helicobacter pylori infection: a follow-up study from infancy to adulthood. Lancet 2002; 359: 931-935.
  • 2. Rowland M, Daly L, Vaughan M, et al. Age-specific incidence of Helicobacter pylori. Gastroenterology 2006; 130: 65-72.
  • 3. Pounder RE, Ng D. The prevalence of Helicobacter pylori infection in different countries. Aliment Pharmacol Ther 1995; 9 Suppl 2: 33-39.
  • 4. Langat AC, Ogutu E, Kamenwa R, et al. Prevalence of Helicobacter pylori in children less than three years of age in health facilities in Nairobi Province. East Afr Med J 2006; 83: 471-477.
  • 5. Go MF. Review article: natural history and epidemiology of Helicobacter pylori infection. Aliment Pharmacol Ther 2002; 16 Suppl 1: 3-15.
  • 6. Lindkvist P, Enquselassie F, Asrat D, et al. Risk factors for infection with Helicobacter pylori — a study of children in rural Ethiopia. Scand J Infect Dis 1998; 30: 371-376.
  • 7. Malaty HM, Graham DY. Importance of childhood socioeconomic status on the current prevalence of Helicobacter pylori infection. Gut 1994; 35: 742-745.
  • 8. Hardikar W, Grimwood K. Prevalence of Helicobacter pylori infection in asymptomatic children. J Paediatr Child Health 1995; 31: 537-541.
  • 9. Windsor HM, Abioye-Kuteyi EA, Leber JM, et al. Prevalence of Helicobacter pylori in Indigenous Western Australians: comparison between urban and remote rural populations. Med J Aust 2005; 182: 210-213. <MJA full text>
  • 10. Everhart JE, Kruszon-Moran D, Perez-Perez GI, et al. Seroprevalence and ethnic differences in Helicobacter pylori infection among adults in the United States. J Infect Dis 2000; 181: 1359-1363.
  • 11. Australian Government Department of Immigration and Multicultural and Indigenous Affairs. Demographic trends in humanitarian arrivals: Western Australia. Canberra: Commonwealth of Australia, 2006.
  • 12. Martin JA, Mak DB. Changing faces: a review of infectious disease screening of refugees by the Migrant Health Unit, Western Australia in 2003 and 2004. Med J Aust 2006; 185: 607-610. <MJA full text>
  • 13. Cherian S, Burgner DP, Carson CF, et al. Diagnosis of Helicobacter pylori infection in a high-prevalence pediatric population: a comparison of 2 fecal antigen testing methods and serology. J Pediatr Gastroenterol Nutr 2008; 47: 130-135.
  • 14. Jacobson K. The changing prevalence of Helicobacter pylori infection in Canadian children: should screening be performed in high-risk children? Can J Gastroenterol 2005; 19: 412-414.
  • 15. Cherian S, Fagan JM, Thambiran A, et al. Severe Plasmodium falciparum malaria in refugee children despite reported predeparture antimalarial treatment [letter]. Med J Aust 2006; 185: 611. <MJA full text>
  • 16. Davis TME, Karunajeewa HA, Ilett KF. Artemisinin-based combination therapies for uncomplicated malaria. Med J Aust 2005; 182: 181-185. <MJA full text>
  • 17. Wang J, Zhou H, Zheng J, et al. The antimalarial artemisinin synergizes with antibiotics to protect against lethal live Escherichia coli challenge by decreasing proinflammatory cytokine release. Antimicrob Agents Chemother 2006; 50: 2420-2427.
  • 18. Velayudhan J, Hughes NJ, McColm AA, et al. Iron acquisition and virulence in Helicobacter pylori: a major role for FeoB, a high-affinity ferrous iron transporter. Mol Microbiol 2000; 37: 274-286.
  • 19. Gisbert JP, de la Morena F, Abraira V. Accuracy of monoclonal stool antigen test for the diagnosis of H. pylori infection: a systematic review and meta-analysis. Am J Gastroenterol 2006; 101: 1921-1930.
  • 20. Malfertheiner P, Megraud F, O’Morain C, et al. Current concepts in the management of Helicobacter pylori infection: the Maastricht III Consensus Report. Gut 2007; 56: 772-781.
  • 21. Weyermann M, Adler G, Brenner H, et al. The mother as source of Helicobacter pylori infection. Epidemiology 2006; 17: 332-334.
  • 22. Escobar ML, Kawakami E. Evidence of mother–child transmission of Helicobacter pylori infection. Arq Gastroenterol 2004; 41: 239-244.
  • 23. Mbulaiteye SM, Gold BD, Pfeiffer RM, et al. H. pylori-infection and antibody immune response in a rural Tanzanian population. Infect Agent Cancer 2006; 1: 3.
  • 24. Fernando N, Holton J, Zulu I, et al. Helicobacter pylori infection in an urban African population. J Clin Microbiol 2001; 39: 1323-1327.
  • 25. Goodman KJ, O’Rourke K, Day RS, et al. Dynamics of Helicobacter pylori infection in a US–Mexico cohort during the first two years of life. Int J Epidemiol 2005; 34: 1348-1355.

Author

remove_circle_outline Delete Author
add_circle_outline Add Author

Comment
Do you have any competing interests to declare? *

I/we agree to assign copyright to the Medical Journal of Australia and agree to the Conditions of publication *
I/we agree to the Terms of use of the Medical Journal of Australia *
Email me when people comment on this article

Responses are now closed for this article.