30th April 2010, Volume 123 Number 1313

Alan G Fraser, Robert Scragg, David Schaaf, Patricia Metcalf, Cameron C Grant

Iron deficiency is an important problem in New Zealand children and young adults. A survey of girls from 8 Auckland high schools found that 18.3% of girls had iron deficiency and 11.5% had anaemia.1 Iron deficiency was more common in Māori and Pacific Island ethnic groups. A longitudinal study of Dunedin children (predominantly European) found lower but still significant rates of anaemia and iron deficiency by the age of 21 years (5.8% and 6.7% respectively).2 The reasons for these ethnic differences remain unclear.

Helicobacter pylori (H. pylori) infection is found more frequently in Māori and Pacific Island ethnic groups and this is a potential explanation for the difference in prevalence of iron deficiency.3,4 There are a number of possible mechanisms by which H. pylori may affect iron metabolism. There may be occult bleeding because of gastritis or ulceration, impaired absorption of non-haem iron because of decreased acid secretion or the bacteria itself may be a scavenger for iron thereby competing with the host for dietary iron or ferritin.5,6

Results from cross-sectional studies suggest that the impact of H. pyloriinfection may vary between different countries and also vary according to socioeconomic status within a country.7–12 One previous New Zealand study has shown a significantly lower serum iron in H. pylori-infected adults selected from the electoral roll but no difference in serum ferritin or the prevalence of anaemia.8

This study sought to determine if H. pylori was an explanation for the high prevalence of iron deficiency in female teenage children in Auckland schools that had a high proportion of Māori and Pacific Island students.


The participants in this report come from a cross-sectional survey, carried out during May 1997 to September 1998, of 2549 Year 11-13 students attending schools in South, Central and West Auckland. The main aim of the survey was to compare cardiovascular risk factors in Pacific Island students with those of other ethnicities. We randomly selected 10 (out of 32) schools with a proportion (>15%) of Pacific Island students and invited all Year 11–13 students at each school to take part. Information on iron and anaemia status was collected from students at 8 of these schools. The results and methods have been reported previously.1. The current report on H. pylori and iron status is limited to female students (at 7 schools) because of their increased prevalence of iron deficiency and anaemia compared with males.
After an overnight fast, students were interviewed in the mornings in groups of up to 10. At the time of this interview a fasting venous blood sample was collected from each student to measure glucose, lipids, iron indices and haemoglobin. Each student then completed a 15-minute self-administered questionnaire which documented age, gender, ethnicity (self-defined), past medical history, and general lifestyle patterns. The Ministry of Education classification of schools by socioeconomic decile, from 1 (low) to 10 (high), was used to code socioeconomic status (SES).13 The number of surveyed students at each school, in this report, ranged from 34 to 373.
All blood samples were separated within 2 hours of collection at a community laboratory, and tested for cardiovascular risk factors and iron status. The following methods were used to determine iron status: serum ferritin was measured by microparticle enzyme immunoassay (Abbott Laboratories); iron (transferrin) saturation was derived from serum iron and unsaturated iron binding capacity by a colourimetric method (Roche); C-reactive protein (CRP) was measured by nephelometry (Behring Diagnostics); haemoglobin (using a cyanmethaemoglobin method) and mean cell volume (MCV) were measured on a Technicon H*3. Any remaining serum was stored at -80ºC.
Definition of anaemia and low ferritin—Cut-points from the United States NHANES III survey are used in the current report.7 Anaemia was defined as haemoglobin <120 g/L for females. Iron deficiency was defined as any two (or more) of the following three: serum ferritin <12 ug/L; iron (transferrin) saturation <14%; MCV <81 fl which was the 5th percentile for the reference group of students in the sample who met all the following criteria: CRP <4 mg/L, haemoglobin g/L >118 (10thpercentile), iron saturation >8%, and ferritin ≥10 ug/L. This definition of iron deficiency is based on that used in the NHANES surveys.7
The only difference is that we have used MCV as a third test of iron deficiency in place of measuring erythrocyte protoporphyrin, since the latter test was not readily available in Auckland at the time of the survey.14 H. pylori antibody was determined by enzyme immunoassay using commercially available kits which was been validated by comparison with urea breath test.15
Data analyses in this report are restricted to 792 (out of 922 surveyed) female students with measured H. pylori antibodies, after excluding 61 students with missing information on H. pylori status, 44 students with CRP >5 mg/L (to exclude any students with possible acute phase reaction (which affects serum ferritin, iron saturation and and haemoglobin concentration) and 25 students with missing CRP values. Statistical analyses were made using SAS SURVEYFREQ and SURVEYMEANs procedures (Release 9.2, Research Triangle Park, NC, 2005) which corrects standard errors and confidence intervals for any design effect from clustering of students by school. The Rao-Scott modified Chi-Squared test was used. The STATA command binreg was used to calculate unadjusted and adjusted risk ratios. The natural logarithm of ferritin was used in analyses to normalise its distribution.


278 students tested positive for H. pylori antibodies, giving a prevalence of 35.1% (95%CI 23.5–46.7). Table 1 shows the prevalence of H. pylori for demographic variables. The prevalence of positive serology was highest for Pacific Island students (49.0%), intermediate for Māori (26.7%) and Asian (24.7%) and lowest for European (13.7%) (p<0.0001). The prevalence of positive H. pylori serology varied across schools from 24.6% up to 60.9%. The prevalence varied with school SES decile, being nearly twice as high for students in SES decile 1 schools (53.1%) compared with decile 2 and 3 schools (28.6%) (p=0.036), but did not vary with age (p=0.19).

Table 1. Prevalence of positive H. pylori serology by level of demographic variable
H. pylori positive % (95% CI)
P value
Age (years)


35.5 (21.7–46.6)
31.5 (16.8–46.1)
36.0 (23.0–49.0)
42.2 (33.1–51.4)



49.0 (38.0–60.0)
26.7 (16.9–36.4)
24.7 (12.6–36.7)
13.7 (6.0–21.4)

School SES decile
1 (low)
2 and 3


53.1 (44.9–61.2)
28.0 (27.0–30.2)

35.1 (23.5–46.7)

Comparisons of adjusted mean values for serum iron measures and haemoglobin, between students with positive and negative H. pylori serology are shown in Table 2. Students with positive serology had significantly lower mean iron saturation (p=0.013), compared with students with negative serology. Mean ferritin (p=0.068), haemoglobin (p=0.08) and MCV (p=0.16) did not vary with H. pylori serology.

Table 2. Mean (95% confidence interval) of blood iron measures and haemoglobin, by category of H. pylori serology, adjusted for age, ethnicity and school SES decile
Blood variable
H. pylori serology
P value
Ferritin (ug/L)
Iron saturation (%)
Mean cell volume (fl)
Haemoglobin (g/L)
21.7 (18.5–25.4)
20.1 (18.7–21.4)
85.9 (85.4–86.3)
130.3 (129.4–131.2)
23.8 (21.4–26.6)
22.0 (20.7–23.3)
86.7 (86.0–87.4)
131.5 (130.9–132.0)

Table 3 shows risk ratios for iron deficiency and anaemia associated with positive H. pylori serology, adjusting for age, ethnicity and school SES decile. There was a significantly higher risk of iron deficiency in students with positive H. pylori serology (p=0.001), but not anaemia (p=0.88).

The prevalence of iron deficiency ranged from 24.2% (se=5.9) in Māori students, 18.9% (3.1) in Pacific, 12.3% (2.1) in Asian to 7.2% (1.9) in European students (p=0.0049). Mean ferritin levels were not significantly different for Māori, Pacific and Asian students, when compared with European, adjusting for age and school SES (p>0.05).

In contrast, after adjusting for age and school SES, mean iron saturation was significantly lower in Māori (by 6.0%), and in Pacific (by 6.5%) students (p<0.05), but not in Asian students, when compared with European; while mean MCV also was significantly lower in all three non-European groups compared with European students (Māori by 4.4 fl, Pacific by 3.2 fl, Asian by 3.1 fl).

Table 3. Risk ratios for iron deficiency and anaemia, associated with positive H. pylori serology († row percents)
H. pyloriserology
Iron deficiency
Relative risk (95% CI)

Adjusted for age, ethnicity and school SES decile
56 (20.1%)
74 (14.4%)
1.28 (1.19–1.39
1.20 (1.08–1.34)



33 (11.9%)
56 (10.9%)
1.06 (0.90–1.26)
1.01 (0.87–1.18)


This study has shown a significant association between H. pylori and iron deficiency, including iron saturation. The magnitude of this effect is small but may be of some clinical importance. There was no association between H. pylori and anaemia.

Data from some other population studies has suggested an association of H. pylori with iron deficiency but the magnitude of the effect may be small and may depend on the definition of iron deficiency.

A US study using the National Health and Nutrition Examination survey 1999–2000 found a significantly lower serum ferritin in H. pylori infected adults (56 compared with 65 ug/L); there was a significant adjusted odds ratio for iron deficiency anaemia but not for iron deficiency (similar definition to this study).7 A further report from the 2003 NHANES survey has shown similar results.16

Several studies have shown a lower serum ferritin with H. pylori infection without showing any effect on iron deficiency anaemia. In a large population study from Germany of subjects aged 18–89 years serum ferritin in H. pylori-infected individuals was 54.5 ug/dL compared with 63.8 ug/dL in the uninfected subjects. The association did not vary by age, gender or iron intake.9

In another population study of 2794 adults in Denmark serum ferritin was lower in men (114 vs 120ug/L, p=0.01) and in post-menopausal women (63 vs 77 ug/L, p=0.02) who were H. pylori positive.10 An small Australian study found significantly lower ferritin levels among women with H. pylori infection compared with non-infected controls (59 vs 88 ug/L)17 In a study of 2080 native Alaskans (78% positive for H. pylori) there was a significant association between low serum ferritin and positive H. pylori serology, particularly in subjects less than 20 years of age.11

Most of these reports have included mainly adult subjects. There is limited data on the possible association in children.12,18,19 In a study of 688 Alaskan children 38% were found to have iron deficiency and 7.8% iron deficiency anaemia. There was an association with H. pylori infection but it is perhaps difficult to make strong conclusions as 86% of children had H. pyloriinfection.19

The other form of evidence for a relationship between H. pylori and iron deficiency comes from H. pylori eradication studies. Most studies have involved small numbers of subjects and have been non-randomised.20–22 In a small randomised study from Korea, 22 H. pylori-positive subjects (mean age 15 years) were randomised to H. pylori eradication, ferrous sulphate alone or both H. pylori eradication and iron supplementation. Eradication treatment, with or without iron resulted in a significant increase in haemoglobin.23

A randomised study of children in rural Alaska showed no difference between treatment with iron supplementation alone and iron treatment combined withH. pylori eradication treatment at 14 months follow-up although there was a difference in iron-deficiency anaemia at 40 months follow-up despite a 50% re-infection rate with H. pylori.24,25

A randomised study of 200 Bangladeshi children aged form 2–5 years showed no additional benefit of H. pylori eradication on iron deficiency compared to iron treatment alone at 90 days. The strength of this study was a randomization to iron treatment alone, H. pylori eradication alone or the combination. In addition, H. pylori status was assessed again at 90 days by urea breath testing and the analysis of children with successful eradication versus persisting infection showed no difference in iron deficiency.26

The reasons for these variable results are unclear. It is likely that H. pylori has a small effect on iron balance by decreasing absorption or increasing iron loss leading to iron deficiency but that this effect is not large enough to lead to iron deficiency anaemia in most populations. There may be regional differences or unknown factors that determine whether there is a pangastritis (causing a decrease in acid secretion) or perhaps a hemorrhagic gastritis (as has been commonly observed in Alaskans).

An older population may have more advanced H. pylori-related changes in the gastric mucosa (although limited negative data in elderly subjects does not support this concept).28 Several studies have shown that acid secretion does improve after H. pylori eradication particularly if there is corpus-predominant gastritis or pangastritis.28,29 However the impact of optimal gastric acid secretion on iron absorption may have been overstated.

A study of iron absorption in children aged 2–5 years in Bangladesh showed that gastric acid secretion improved after H. pylori eradication but iron absorption did not improve after eradication treatment.30

H. pylori infection is claimed to be associated with a number of extra-gastric manifestations. Most of these associations have not been confirmed by larger well conducted studies that control for confounding issues such as socioeconomic status. One major example is the debate over the possible association of H. pylori with coronary heart disease.31

This study shows that H. pylori is associated with iron deficiency but this weak effect is unlikely to explain the major ethnic differences in iron deficiency anaemia. Dietary intake of iron appears to be similar in different ethnic groups in New Zealand.32

The prevalence rates of H. pylori infection in European, Māori and Pacific island children in this study (collected 1997–98) are similar to data reported from samples taken from 11 year old South Auckland school children in 1988.3

This study confirms the significant public health issue of H. pylori infection particularly for Pacific Islanders who will carry a burden of upper gastrointestinal disease into coming decades, causing both peptic ulceration and gastric cancer, because of high rates of H. pylori infection.4

Further studies on the reasons for high rates of acquisition of H. pyloriinfection in some ethnic groups are required. Targeted testing and eradication depending on ethnicity may be sensible to prevent later consequences of H. pylori infection.4


Iron deficiency is an important problem in New Zealand children and young adults. A previous study of 8 Auckland high schools showed that 18.3% of girls had iron deficiency. This study assessed the potential role of Helicobacter pylori (H. pylori), a bacterium that can infect the stomach, in causing iron deficiency. 792 female students (median age 16 years) from 7 Auckland high schools had H. pylori serology and tests for iron deficiency. The prevalence of positive H. pylori serology was highest for Pacific Island students (49.0%), intermediate for Maori (26.7%) and Asian (24.7%) and lowest for European (13.7%). Positive H. pylori serology was associated with increased risk of iron deficiency (RR 1.20) (after adjusting for age, ethnicity and school SES decile). This study indicates that H. pylori infection is associated with iron deficiency and should be considered as a possible cause when iron deficiency is diagnosed. This study also confirms the significant public health issue of H. pylori infection particularly for Pacific Islanders who will carry a burden of upper gastrointestinal disease into coming decades, causing both peptic ulceration and gastric cancer, because of high rates of H. pylori infection.


Iron deficiency is an important problem in New Zealand children and young adults. Iron deficiency and Helicobacter pylori (H. pylori) infection are each more common in Māori and Pacific Island ethnic groups.


This study seeks to determine if H. pylori infection is associated with iron deficiency.


792 female students from 7 Auckland high schools (median age 16 years) had H. pylori serology and tests for iron deficiency assessed by a combination of serum ferritin, iron saturation and mean cell volume.


The prevalence of positive H. pylori serology was highest for Pacific Island students (49.0%; CI 38.0–60.0), intermediate for Māori (26.7%; CI 16.9–36.4) and Asian (24.7%; CI 12.6–36.7) and lowest for European (13.7%; 6.0–21.4) p<0.0001. Students with positive H. pylori serology had lower mean levels of iron saturation (p=0.013), but not of ferritin (p=0.068), haemoglobin (p=0.08) or mean cell volume (p=0.16), compared to those with negative serology. Positive H. pylori serology was associated with increased risk of iron deficiency (RR 1.20; CI 1.08–1.34), but not anaemia (RR 1.01; CI 0.87–1.18), after adjusting for age, ethnicity and school SES decile.


This study indicates that H. pylori infection is associated with iron deficiency in adolescent females. There are significant differences in H. pylori serology amongst different ethnic groups in New Zealand.

Author Information

Alan G Fraser, Associate Professor, Department of Medicine, University of Auckland; Robert Scragg, Associate Professor in Epidemiology, Epidemiology & Biostatistics, School of Population Health, University of Auckland; David Schaaf, Research Fellow, Pacific Health, School of Population Health, University of Auckland; Patricia Metcalf, Senior Lecturer, Department of Statistics, University of Auckland; Cameron C Grant, Associate Professor, Department of Paediatrics, University of Auckland


Assoc Prof Alan Fraser, Department of Medicine, University of Auckland, Private Bag 92019, Auckland, New Zealand. Fax: +64 (0)9 3737677

Correspondence Email


Competing Interests

None known.


  1. Schaaf D, Scragg R, Metcalf P, Grant C, Buchanan J. Prevalence of iron deficiency in Auckland high school students. NZ Med J. 2000;113:347-50.
  2. Fawcett JP, Brooke M, Beresford CH, et al. Iron deficiency and anaemia in a longitudinal study in New Zealand at ages 11 and 21 years. NZ Med J. 1998;111:400-2.
  3. Fraser AG, Scragg R, Metcalf P, McCullough S, Yeates NJ. Prevalence of Helicobacter pylori in different ethnic groups in New Zealand children and adults. Aust NZ J Med. 1996;26:646-51.
  4. Fraser AG. Helicobacter pylori: a historical perspective 1983-2003. NZ Med J. 2004;117:1194.
  5. DuBois S, Kearney DJ. Iron-deficiency anaemia and Helicobacter pylori infection: a review of literature. Am J Gastroenterol. 2005;100:453-459.
  6. Barabino A. Helicobacter pylori-related iron deficiency anemia: a review. Helicobacter. 2002;7:71-5.
  7. Cardenas VM, Mulla ZD, Ortiz M, Graham DY. Iron deficiency and Helicobacter pylori infection in the Unites States. Am J Epidemiol. 2006;163:127-34.
  8. Collett JA, Burt MJ, Frampton CM, et al. Seroprevalence of Helicobacter pylori in the adult population of Christchurch: risk factors and relationship to dyspeptic symptoms and iron studies. NZ Med J. 1999;112:292-5.
  9. Berg G, Bode G, Blettner M, Boeing H, Brenner H. H.pylori infection and serum ferritin: a population-based study among 1806 adults in Germany. Am J Gastroenterol. 2001;96:1014-8.
  10. Milman N, Rosenstock S, Anderson L, Jorgenson T, Bonnevie O. Serum ferritin, hemoglobin, and Helicobacter pylori infection: a seroepidemiologic survey comprising 2794 Danish adults. Gastroenterology. 1998;115:268-274.
  11. Parkinson AJ, Gold BD, Bulkow L, et al. High prevalence of Helicobacter pylori in the Alaskan native population and association with low serum ferritin levels. Clin Diag Lab Immunol. 2000;7:885-8.
  12. Choe YH. Kim SK. Hong YC. The relationship between Helicobacter pylori infection and iron deficiency: seroprevalence study in 937 pubescent children. Arch Dis Childhood. 2003;88:178.
  13. Data Management and Analysis Section. Ministry of Education socioeconomic indicator for schools. Wellington: Ministry of Education.
  14. Dallman PR, Looker AC, Johnson CL, Carroll M. Influence of age on laboratory criteria for the diagnosis of iron deficiency anaemia and iron deficiency in infants and children. In: Hallberg L, Asp N-G, ed. Iron nutrition in health and disease. John Libby Co, London, 1996:65-74.
  15. Fraser AG, Williamson S, Lane MR, Hollis B. Nurse-led dyspepsia clinic using the urea breath test for H.pylori. NZ Med J. 2003;116:479-83.
  16. Kearney DJ, Ioannou GN. The association between Helicobacter pylori seropositivity and serum markers of iron stores in the United States. Gastroenterology 2005;128: Suppl 2: W952 (abstract).
  17. Peach HG, Bath NE, Farish SJ. Helicobacter pylori infection: an added stressor on iron status of women in the community. Med J Aust. 1998;169:188-90.
  18. Seo JK. Ko JS. Choi KD. Serum ferritin and Helicobacter pylori infection in children: a sero-epidemiologic study in Korea. J Gastroenterol Hepatol. 2002;17:754-7.
  19. Baggett HC, Parkinson AJ, Muth PT, Gold BD, Gessner BD. Endemic iron deficiency associated with Helicobacter pylori infection among school-aged children in Alaska. Pediatrics. 2006;117:E396-404.
  20. Ashorn M, Ruuska T, Makipernaa A. Helicobacter pylori and iron deficiency anaemia in children. Scand J Gastroenterol. 2001;36:701-5.
  21. Konno M, Muraoka S, Takahashi M, Imai T. Iron deficiency anaemia associated with Helicobacter pylori gastritis. J Pediat Gastroenterol Nutrition. 2000;31:52-56.
  22. Annibale B, Marignani M, Monarca B, et al. Reversal of iron deficiency anaemia after H.pylori eradication in patients with asymptomatic gastritis. Annals Int Med. 1999;131:668-72.
  23. Choe YH, Kim SK, Son BK, Lee DH, Hong YC, Pai SH. Randomised placebo-controlled trial of Helicobacter pylori eradication for iron deficiency anaemia in pre-adolescent children and adolescents. Helicobacter. 1999;4:135-9
  24. Gessner BD, Baggett HC, Muth PT, et al. A controlled, household-randomized, open label trial of the effect that treatment of Helicobacter pylori has on iron deficiency in rural Alaska. J Infect Dis. 2006;193:537-46.
  25. Fagan RP. Dunaway CE. Bruden DL. Parkinson AJ. Gessner BD. Controlled, household-randomized, open-label trial of the effect of treatment of Helicobacter pylori infection on iron deficiency among children in rural Alaska: results at 40 months. J Infect Dis 2009;199:652-60.
  26. Sarker SA, Mahmud H, Davidsson L, et al. Causal relationship of Helicobacter pylori With iron-deficiency anemia or failure of iron supplementation in children. Gastroenterology 2008;135:1534-42.
  27. Pilott A, Fabrello R, Franceschi M, et al. Helicobacter pylori infection in asymptomatic elderly subjects living at home or in a nursing home: effects on gastric function and nutritional status. Age Aging. 1996;25:245-9.
  28. Guitierrez O, Melo M, Segura AM, Angel A, Genta RM, Graham DY. Cure of Helicobacter pylori infection improves gastric acid secretion in patients with corpus gastritis. Scand J Gastroenterol. 1997;32:664-8.
  29. Haruma K, Mihara M, Okamoto E, Kusinoki H, Hananoki M, Tanaka S. Eradication of Helicobacter pylori increases gastric acidity in patients with atrophic gastritis of the corpus – evaluation of 24-hr pH monitoring. Aliment Pharmacol Therap. 1999;13:155-62.
  30. Sarker SA, Davidsson L, Mahmud H, et al. Helicobacter pylori infection, iron absorption and gastric acid secretion in Bangladeshi children. Am J Clin Nutrition. 2004;80:149-53.
  31. Fraser AG, Scragg RK, Cox B, Jackson RT. Helicobacter pylori, Chlamydia pneumoniae and myocardial infarction. Int Med J. 2003;33:267-272.
  32. Ferguson EL, Morison IM, Faed JM, et al. Dietary iron intakes and biochemical iron status of 15-49 year old women in New Zealand: is there cause for concern? NZ Med J. 2001;114:134-8.