Incidence of upper gastrointestinal haemorrhage in Maori and New Zealand European ethnic groups, 2001-2010

Incidence of upper gastrointestinal haemorrhage in Maori andNew Zealand European ethnic groups, 2001–2010J. Irwin1, R. Ferguson2, F. Weilert1 and A. Smith1

1Department of Gastroenterology, Waikato Hospital, Hamilton and 2School of Medicine, The University of Auckland, Auckland, New Zealand

Key wordsupper gastrointestinal bleeding, peptic ulcer

disease, epidemiology, upper gastrointestinal

endoscopy.

CorrespondenceJames Irwin, Department of Gastroenterology,

Royal Brisbane and Women’s Hospital,

Butterfield Street, Herston, Qld 4006, Australia.

Email: [email protected]

Received 5 November 2013; accepted 25 May

2014.

doi:10.1111/imj.12489

Abstract

Background: To date no incidence figures for upper gastrointestinal haemorrhage

(UGIH) in New Zealand have been published.

Aims: To determine the incidence of UGIH in a demographically defined population,

and to assess variation in incidence associated with demographic parameters.

Methods: Between 17 March 2001 and 12 October 2010, all patients presenting to

Waikato Hospital with UGIH were prospectively ascertained, and clinical, epidemiologi-

cal and laboratory data were collected. Incidence rates were calculated, and were age

adjusted to the World Health Organization world standard population. Parameters

associated with trends in incidence were examined.

Results: There were 1360 UGIH events, yielding a crude incidence of 59.2 per 100 000

adults (age ≥ 15 years) per year (all quoted incidence figures per 100 000 adults per

year), and an age-adjusted incidence (AAI) of 46.4. AAI was higher for Maori compared

with New Zealand Europeans (91.3 vs 37.0, rate ratio (RR) = 2.47, P < 0.001). Maori

were more likely to have a gastric ulcer at endoscopy (odds ratio (OR) = 2.21, P < 0.001).

For those tested for Helicobacter pylori (n = 702), Maori were more likely to be infected

(OR = 2.12, P < 0.001). AAI was higher for males (61.1 vs 33.6, RR = 1.82, P < 0.001).

Males were more likely to have a duodenal ulcer at endoscopy (OR = 1.79, P < 0.001).

AAI incidence decreased from the first to the second half of the study period (53.6 vs

45.8, RR = 1.17, P < 0.001).

Conclusion: AAI of UGIH in the Waikato region was 46.4. This was significantly higher

in Maori and in males, and decreased over the study period. These data will provide a

comparison for future assessment of trends in UGIH.

Introduction

Upper gastrointestinal haemorrhage (UGIH) is defined asbleeding into the gastrointestinal tract, distal to the upperoesophageal sphincter and proximal to the ligament ofTreitz. Patients with UGIH present with one or a combi-nation of haematemesis, melaena, hypotension and col-lapse. Mortality in published series lies between 7% and14%.1–4

Over the past 20 years, gastroscopy has become routinein the assessment of UGIH. This has led to the moreaccurate description of underlying pathologies causingUGIH.1–4 However, given that a significant proportionof UGIH cases have no cause identified at endoscopy,estimation of incidence remains reliant on clinical param-eters.3 Published estimates vary significantly, between 48

and 172 per 100 000 adults per year (all incidence figuresquoted per 100 000 adults per year).3–8 To our knowledge,no description of incidence of UGIH in Maori has beenpublished.

This study reports the incidence of UGIH over a10-year period in a demographically defined populationin the Waikato region of New Zealand (NZ). The influ-ence of ethnicity, sex and time period on incidence isexamined. The association between incidence, clinicalparameters and demographic parameters is analysed,illuminating factors influencing differences in incidencebetween demographic groups.

Methods

Patient identification

A prospective database, dedicated to recording gastroin-testinal haemorrhage, captured all patients undergo-ing endoscopy with haematemesis or melaena as theindication from March 2001 to October 2010, in a single

Funding: We would like to acknowledge the support of theWaikato District Health Board for contributing to funding for thisresearch.Conflict of interest: None.

bs_bs_banner

© 2014 The AuthorsInternal Medicine Journal © 2014 Royal Australasian College of Physicians 735

mailto:[email protected]

tertiary referral hospital (Waikato Hospital, Hamilton,NZ). This is the only hospital in the region performingendoscopy for acute gastrointestinal bleeding.

Presenting symptoms were retrospectively confirmedand those without evidence of haematemesis or melaenawere excluded. Bleeding events which occurred within30 days of a sentinel bleeding event were classified asrebleeding events and were excluded. Patients whounderwent more than one endoscopy for UGIH duringthe study period had their first event only included.

The study region was defined as the seven core districtsof the Waikato region as geographically described by Sta-tistics New Zealand: Hamilton City, Hamilton district,Waipa, South Waikato, Matamata-Piako, Otorohangaand Waitomo.9 Patients whose recorded domicile was notin the study area were excluded.

Data collection

Patient demographics, presenting symptoms and findingsat endoscopy were prospectively recorded on the gastro-intestinal haemorrhage database. Medication history wascollected retrospectively from the paper clinical record andincluded use on admission of a proton pump inhibitor oranticoagulant (warfarin, aspirin, clopidogrel or dipyri-damole), and non-steroidal anti-inflammatory drug(NSAID) use within 30 days of admission. Helicobacterpylori status was also assessed retrospectively, and wasinterpreted as positive or negative if performed within 30days of the bleeding event. Testing was performed withserology (Genesis Diagnostics Hp-G ELISA Screen,Product code GD01, Genesis Diagnostics Ltd, Cambridge-shire, United Kingdom), histological examination ofbiopsy material using a Warthin–Starry stain, a urease testof stomach biopsy tissue using standardised urea broth(Fort Richard Laboratories Ltd, Auckland, NZ) or H. pylorifaecal antigen testing (Hp StAR amplified IDEIA, Code:K663011, Oxoid Limited, Hampshire, United Kingdom).

Age, sex and ethnicity data for each patient wereobtained from the hospital’s electronic patient informa-tion system, and matched to our cohort by National Hos-pital Index number, date of birth and surname.

Baseline demographic data for the study region wereobtained from the statistics NZ website.9 Data taken fromthe 2006 census were used as the denominator to assessthe influence of sex and ethnicity on incidence, and datafrom the 2001 and 2006 censuses were used to assesstemporal trends in incidence. Demographic data werestratified by sex, ethnicity and age.

Statistics

Statistical analysis was performed using the R statisticalcomputing environment.10 Crude incidence was calcu-

lated per 100 000 adults aged 15 and over. For assess-ment of temporal trends, the first half of the study period(17 March 2001–29 December 2005 using 2001 censusdata to define the at risk population) was compared withthe second half of the study period (29 December2005–12 October 2010 using 2006 census data to definethe at risk population). To allow comparison betweendemographic groups with different age structures, inci-dence was adjusted to the World Health Organization(WHO) world standard population.11 Comparison ofcrude and age-adjusted incidence (AAI) rates betweendifferent demographic groups was made using Rate.Ratiotest in R. Proportions of patients with risk factors forUGIH were compared between demographic groupsusing Prop.test in R, as were proportions of patients witheach endoscopic diagnosis. This allowed associationanalysis between demographic parameters and riskfactors for UGIH.

Results

Two thousand and eighty-nine bleeding events were rec-orded in the gastrointestinal haemorrhage database.Seven hundred and twenty-nine were excluded, and1360 included in the final dataset (Fig. 1). Nine hund-red and ninety nine patients (73%) identified as NZ

Figure 1 Exclusions.

Irwin et al.

© 2014 The AuthorsInternal Medicine Journal © 2014 Royal Australasian College of Physicians736

European, 268 (20%) Maori and 93 (7%) as otherethnicity. Eight hundred and four (59%) were male.

The adult population (those aged 15 or older) of thestudy region in 2006 was 239 690. Crude incidence overthe 10-year study period was 59.2 per 100 000 adults peryear. Incidence adjusted to the WHO standard population

was 46.4. Incidence increased markedly with increasingage (Fig. 2).

Crude incidence stratified by ethnicity was not signifi-cantly different, while AAI was 2.5 times higher in Maori(Table 1, Fig. 2). Crude and AAI was 1.5 and 1.7 timeshigher respectively in males compared with females

020

040

060

080

0

Age (years)

Inci

denc

e pe

r 10

0 00

0 pe

r ye

ar

15−40 40−50 50−60 60−70 70−80 >80

(a)

15−40 40−50 50−60 60−70 70−80 >80

020

040

060

080

0

Age (years)

Ethnicity

(b)0

200

400

600

800

Age (years)

Inci

denc

e pe

r 10

0 00

0 pe

r ye

ar

15−40 40−50 50−60 60−70 70−80 >80

Sex

(c)0

200

400

600

800

Age (years)

15−40 40−50 50−60 60−70 70−80 >80

Year

(d)

Figure 2 Incidence of upper gastrointestinal haemorrhage (UGIH) stratified by (a) age, (b) by age and ethnicity ( , Maori (n = 268); , NZ European

(n = 999)), (c) by age and sex ( , male (n = 804); , female (n = 556)), and (d) by age and study period ( , 2001–2005 (n = 694); , 2006–2010

(n = 666)) (Error bars show 95% confidence interval).

Incidence of upper GI haemorrhage


(Table 1, Fig. 2). Crude incidence was lower in the secondhalf of the study period, and AAI significantly so(Table 1). This difference was most apparent in patients60 years or older (Fig. 2).

The proportion of patients receiving each diagnosis,comparing Maori to NZ European, male to female, andthe first and second part of the study period, is shown inTable 2. Of note, the proportion of Maori with gastriculceration was significantly higher compared with NZEuropeans, and the proportion of males with duodenalulceration was significantly higher compared withfemales. There was an increase in diagnosis of oesopha-geal or gastric varices over the study period.

One hundred and forty-five out of 268 (54%) of Maoriand 500/999 (50%) of NZ Europeans were tested forH. pylori infection. For those tested, Maori had higherrates of H. pylori infection in comparison to NZ Europeans(59/145, 40.7% vs 122/500 24.4%, P < 0.001). Maori hadlower rates of aspirin use (93/268, 34.7% vs 488/999,48.8% P < 0.001). Males had lower rates of proton pumpinhibitor (PPI) use (174/804, 22% vs 163/556, 29%, P =

0.014). The proportion of patients taking warfarin (81/694, 12% vs 120/666, 18%, P = 0.001) and clopidogrel(4/694, 0.6% vs 19/666, 2.8%, P = 0.011) increasedsignificantly over the course of the study period, whilethere was a trend towards increased aspirin use (297/694, 43% vs 318/666, 48%, P = 0.07).

Discussion

Incidence

The incidence of UGIH in the population observed in thisstudy is at the lower end of the spectrum of publishedfigures.3–6 The wide range of published incidence figuresfor UGIH is due in part to differing study methodology.Our methodology, relying on identification of bleedingevents through referral for endoscopy, may have led toan underestimation of true incidence. This method ofpatient selection will not record episodes of UGIH occur-ring in patients who do not present to hospital or inpatients who present to hospital but are not referred forendoscopy. However, this method of case ascertainmenthas advantages over collection of UGIH events throughdischarge coding. Patients with a dubious UGIH event(e.g. suspected haematemesis with coffee groundvomitus, or positive faecal occult blood testing) may berecorded as having had UGIH on discharge coding. This isespecially true in institutions where there is financial orother incentive to code multiple diagnoses for eachadmitted patient.

We performed an analysis of recorded UGIH by dis-charge coding in our hospital over the same time periodas our study. This recorded a crude incidence of UGIH of89.4, 50% higher than that recorded in our study. Webelieve that the difference between the two figures islikely to be the result of two factors: the presence ofpatients with UGIH who were not referred for endoscopy,

Table 1 Incidence of UGIH per 100 000 adults per year, stratified by eth-

nicity, sex and study period

Rate ratio P-value

NZ European Maori

CI 54.7 63.0 1.15 0.047

AAI 37.0 91.3 2.49 <0.001

Female Male

CI 47.0 72.3 1.54 <0.001

AAI 33.6 61.1 1.74 <0.001

2001–2005 2006–2010

CI 65.8 58.3 0.89 0.027

AAI 53.6 45.8 0.83 <0.001

AAI, age-adjusted incidence; CI, crude incidence; NZ, New Zealand; UGIH,

upper gastrointestinal haemorrhage.

Table 2 Percentage of patients with UGIH with each endoscopic diagnosis

NZ European (n = 1129) Maori (n = 297) Female (n = 556) Male (n = 804) 2001–2005 (n = 759) 2006–2010 (n = 763)

Gastric ulcer 14 27** 18 17 16 19

Duodenal ulcer 19 23 15 24** 20 21

Gastric cancer 2.1 3 1.8 2.4 1.7 2.5

Oesophageal cancer 1 0.4 0.7 0.9 0.7 0.9

Gastric erosions 12 12 14 11 11 14

Duodenal erosions 7 7.5 6.1 7.7 5.8 8.4

Vascular lesion 2.8 1.5 2.9 2.1 1.7 3.1

Oesophageal varices 3.2 2.6 2 4.5* 1.9 5.1*

Oesophagitis 18 12* 15 17 17 15

Mallory–Weiss tear 5.4 2.6 4.7 4.9 5.5 4.1

Normal endoscopy 18 16 20 15* 19 15

P-value of significance for comparison of two rates: *P < 0.05; **P < 0.001. NZ, New Zealand; UGIH, upper gastrointestinal haemorrhage.

Irwin et al.


and the presence of patients without clear evidence ofclinically significant UGIH who received this dischargecode.

Patients who suffered UGIH and attended a distanthospital because they were away from home will nothave been recorded in our study, and will have contrib-uted to an underestimation of true population incidence.Eleven per cent (166 of 1526) of UGIH events in ourdataset were in patients who lived outside our hospital’scatchment area, giving an approximate estimate of thiserror.

Reported incidence figures vary depending on the defi-nition of the population at risk. Studies using the entirepopulation tend to report lower incidence figures,3 whilethose using an ‘adult’ population over a cut-off age ofbetween 15 and 20 tended to report higher incidencefigures.4,6,12,13 Incidence of UGIH in children is signifi-cantly lower than in adults,3 and children make up asignificant proportion of the population. In our studyregion in 2006, 24.7% of the population was less than 15years of age,9 and inclusion of children in the at-riskpopulation would have reduced estimation of incidenceby approximately 25%. Despite the exclusion of children,the incidence we have observed is still at the lower end ofthe published range.

Incidence in Maori

AAI of UGIH was significantly higher in Maori, andincreased rates of H. pylori infection and gastric ulcerationwere associated with this difference. H. pylori testing wascompromised in our study – only 50% of patients hadtesting performed, and four different methods of testingwere used. Additionally, urease tests taken during epi-sodes of UGIH have previously been described as havinga reduced sensitivity to H. Pylori infection.14 The inci-dence of H. pylori in our cohort is lower than publishedrates in NZ ethnic groups of 35.8% (NZ European adults)and 57.4% (Maori adults).15 Despite these limitations, wefeel the H. pylori results give useful insight into the aeti-ology of UGIH in Maori.

The difference in incidence could be attributed togenetic or environmental factors. Maori underperform inmeasures of education, income and life expectancy incomparison to the NZ European population. The higherrate of H. pylori infection in Maori15 is consistent withhigher observed rates in other indigenous popula-tions.16,17 UGIH has previously been negatively correlatedwith socio-economic status,6 an association that may bedue in part to higher H. pylori infection rates in deprivedpopulations.

There was no difference in NSAID medication usebetween Maori and NZ Europeans, and Maori had a

lower rate of aspirin use. Maori had a higher incidence ofwarfarin use that could plausibly be considered to becausally associated with an increased incidence of UGIH,but not with gastric ulceration.

The genotype of H. pylori prevalent in Maori (hpMaori)differs from that in Europe (hpEurope).18 Previous effortsto identify differing prevalence of virulence factors inH. pylori isolated from NZ Europeans and Polynesians(including Maori) demonstrated similar CagA levels byimmunoblot, and higher cytotoxic VacA levels in H. pyloristrains isolated from Polynesians.19 VacA-producingstrains of H. pylori are associated with duodenal ulcera-tion,20 and a higher prevalence of VacA-producing H.pylori strains among Maori is unlikely to explain theobserved increased rate of gastric ulceration.

H. pylori infection has been more strongly associatedwith duodenal ulceration than with gastric ulceration.20

Our Maori cohort had an increased incidence of gastricand not duodenal ulceration, despite a higher proport-ion of H. pylori infection. The typical distribution ofH. pylori infection associated with duodenal and gastriculceration differs. Duodenal ulceration is associated withpredominantly gastric antral infection and consequenthypersecretion of acid from the gastric body.21 Gastriculceration is associated with infection of the entire bodyof the stomach, and low or normal acid secretion. Geneticor environmental factors may predispose to differing pat-terns of infection in Maori, and may explain the unusualpredominance of gastric ulceration despite high H. pyloriinfection rates. Further research in this area may identifyunrecognised contributors to the pathogenesis of gastriculceration.

Maori have been shown to have high rates of intestinalmetaplasia22 and gastric cancer, both associated withH. pylori infection.22,23 A significant proportion of gastriccancer in Maori is histologically diffuse.24 This studyshows that in addition to increased rates of gastric cancer,there is an increased rate of gastric ulceration in Maori.Although unlikely to be part of the sequence of diffusegastric cancer development, a higher rate of gastriculceration in Maori is likely to be a related phenomenon,similarly consequent to chronic inflammation of thegastric mucosa.

Incidence in males

UGIH incidence was higher in males, as observed in pre-vious studies.2,6 Male UGIH was associated with a signifi-cantly higher rate of duodenal ulceration, and a lowerrate of PPI use.

The most likely explanation for a difference in UGIH andduodenal ulcer incidence between sexes is the modifyingeffect of sex hormones on ulcer formation and healing.



In animal studies, exogenous administration of oestrogenand progestogen has been shown to protect againstulcer formation, while testosterone administrationaggravates it.25,26

Temporal trends in incidence

The incidence of UGIH has decreased over the past 20years.3,13 Accepted risk factors include aspirin use,clopidogrel use, NSAID use,27 anticoagulation with war-farin28 and H. pylori infection,20 while PPI use is protec-tive.29 Temporal alteration in prevalence of these factorsmay have contributed to the observed decrease. PPI usehas increased over this time period30; however, aspirinuse30 and warfarin use31 have also increased. Addition-ally, the incidence of UGIH is known to increase sharplywith age, and our ageing population should contribute toan increasing crude (but not age adjusted) rate.

In our study, despite ageing of our population, crudeincidence of UGIH reduced over the study period by11%. This reduction was associated with an increase inwarfarin and clopidogrel use, and a trend towardsincreasing aspirin use. There was no significant change inH. pylori infection rate, NSAID use or PPI use. It is likelythat as the population ages further, there will be furtherincreases in antiplatelet and anticoagulant use. The inter-play of these factors is complex and it is not immediatelyevident whether incidence rates are likely to continue todecrease, or to increase with time.

Study limitations

This study has several limitations. First, incidence in ourstudy may have been underestimated due to case ascer-tainment bias, and non-capture of bleeding events thatoccurred in our population while outside of our hospital’scatchment area. Second, endoscopic diagnosis was a sub-jective interpretation made by the 20 individualendoscopists performing endoscopy over the 10 years ofthe study. Last, retrospective collection of medicationhistory from the clinical record is known to be inaccurate,particularly regarding NSAID use.

Conclusion

AAI of UGIH in the Waikato region was 46.4. This wassignificantly higher in Maori, in males, and decreasedover the study period. Maori were noted to have higherrates of H. pylori infection and gastric ulceration. Thesedata will provide a comparison for future assessment oftrends in UGIH.

Acknowledgements

We would like to thank Steve Holmes (Statistical andDatabase Support, Waikato Hospital) for providing statis-tical assistance, and Carolyn Gibbs (Department of Gas-troenterology, Waikato Hospital) for maintaining ourdatabase.

References

1 Marmo R, Koch M, Cipolletta L,

Capurso L, Pera A, Bianco MA et al.

Predictive factors of mortality from

nonvariceal upper gastrointestinal

hemorrhage: a multicenter study. Am J

Gastroenterol 2008; 103: 1639–47, quiz

1648.

2 Yavorski RT, Wong RK, Maydonovitch

C, Battin LS, Furnia A, Amundson DE.

Analysis of 3294 cases of upper

gastrointestinal bleeding in military

medical facilities. Am J Gastroenterol

1995; 90: 568–73.

3 Van Leerdam ME, Vreeburg EM, Rauws

EAJ, Geraedts AAM, Tijssen JGP,

Reitsma JB et al. Acute upper GI

bleeding: did anything change? Time

trend analysis of incidence and outcome

of acute upper GI bleeding between

1993/1994 and 2000. Am J Gastroenterol

2003; 98: 1494–9.

4 Rockall TA, Logan RF, Devlin HB,

Northfield TC. Incidence of and

mortality from acute upper

gastrointestinal haemorrhage in the

United Kingdom. Steering Committee

and members of the National Audit of

Acute Upper Gastrointestinal

Haemorrhage. BMJ 1995; 311:

222–6.

5 Lewis JD, Bilker WB, Brensinger C,

Farrar JT, Strom BL. Hospitalization and

mortality rates from peptic ulcer disease

and GI bleeding in the 1990s:

relationship to sales of nonsteroidal

anti-inflammatory drugs and acid

suppression medications. Am J

Gastroenterol 2002; 97: 2540–9.

6 Blatchford O, Davidson LA, Murray WR,

Blatchford M, Pell J. Acute upper

gastrointestinal haemorrhage in west of

Scotland: case ascertainment study. BMJ

1997; 315: 510–14.

7 Paspatis GA, Konstantinidis K,

Chalkiadakis I, Tribonias G,

Chlouverakis G, Roussomoustakaki M.

Changing trends in acute upper

gastrointestinal bleeding in Crete,

Greece: a population-based study.

Eur J Gastroenterol Hepatol 2012; 24:

102–3.

8 Hreinsson JP, Kalaitzakis E,

Gudmundsson S, Björnsson ES. Upper

gastrointestinal bleeding: incidence,

etiology and outcomes in a

population-based setting. Scand J


9 Statistics New Zealand. Home – statistics

New Zealand: 2001 and 2006 census

data [Internet]. [cited 2014 Jan 12].

Available from URL:

http://www.stats.govt.nz/

10 R Development Core Team. R: a

Language and Environment For Statistical

Computing. [Internet]. Vienna, Austria: R

Foundation for Statistical Computing;

2011. [cited 2014 Jun 14]. Available

from URL: http://www.R-project.org/

11 Omar B, Ahmad CB-P. Age

standardization of rates: a new WHO

standard. GPE Discuss Pap Ser

EIPGPEEBD World Health Organ, No.

31, 2001.

Irwin et al.


http://www.stats.govt.nz/

http://www.R-project.org/

12 Longstreth GF. Epidemiology of

hospitalization for acute upper

gastrointestinal hemorrhage: a

population-based study. Am J


13 Loperfido S, Baldo V, Piovesana E,

Bellina L, Rossi K, Groppo M et al.

Changing trends in acute upper-GI

bleeding: a population-based study.

Gastrointest Endosc 2009; 70: 212–24.

14 Gisbert JP, Abraira V. Accuracy of

Helicobacter pylori diagnostic tests in

patients with bleeding peptic ulcer: a

systematic review and meta-analysis. Am

J Gastroenterol 2006; 101: 848–63.

15 Fraser AG, Scragg R, Metcalf P,

McCullough S, Yeates NJ. Prevalence of

Helicobacter pylori infection in different

ethnic groups in New Zealand children

and adults. Aust N Z J Med 1996; 26:

646–51.

16 Goodman KJ, Jacobson K, van Zanten

SV. Helicobacter pylori infection in

Canadian and related Arctic Aboriginal

populations. Can J Gastroenterol 2008; 22:

289.

17 Windsor HM, Abioye-Kuteyi EA, Leber

JM, Morrow SD, Bulsara MK, Marshall

BJ. Prevalence of Helicobacter pylori in

Indigenous Western Australians:

comparison between urban and remote

rural populations. Med J Aust 2005; 182:

210–13.

18 Falush D, Wirth T, Linz B, Pritchard JK,

Stephens M, Kidd M et al. Traces of

human migrations in Helicobacter

pylori populations. Science 2003; 299:

1582–5.

19 Campbell S, Fraser A, Holliss B, Schmid

J, O’Toole PW. Evidence for ethnic

tropism of Helicobacter pylori. Infect Immun

1997; 65: 3708–12.

20 Schöttker B, Adamu MA, Weck MN,

Brenner H. Helicobacter pylori infection is

strongly associated with gastric and

duodenal ulcers in a large prospective

study. Clin Gastroenterol Hepatol 2012; 10:

487–93, e1.

21 Schultze V, Hackelsberger A, Günther T,

Miehlke S, Roessner A, Malfertheiner P.

Differing patterns of Helicobacter pylori

gastritis in patients with duodenal,

prepyloric, and gastric ulcer disease.

Scand J Gastroenterol 1998; 33:

137–42.

22 Fraser AG, Peng SL, Jass JR. Intestinal

metaplasia subtypes and Helicobacter

pylori infection: a comparison of ethnic

groups in New Zealand. J Gastroenterol

Hepatol 1998; 13: 560–5.

23 Nevalainen TJ, Laurén P, Gavin JB. The

intestinal and diffuse types of gastric

carcinoma in Maori and non-Maori

patients in Auckland. Scand J


24 Biggar M, Srinivasa S, Wickramarachchi

B, Babor R, Poole GH, Hill AG. Gastric

cancer location and histological subtype

in Pacific people and Maori defies

international trends. N Z Med J 2011;

124: 1331. [cited 2013 Jul 2]. Available

from URL: http://journal.nzma.org.nz/

journal/124-1331/4584/

25 Aguwa CN. Effects of exogenous

administration of female sex hormones

on gastric secretion and ulcer formation

in the rat. Eur J Pharmacol 1984; 104:

79–84.

26 László F, Varga C, Montoneri C, Drago

F. Damaging actions of testosterone on

cysteamine-induced gastroduodenal

ulceration and vascular leakage in

the rat. Eur J Pharmacol 1997; 337:

275–8.

27 García Rodríguez LA, Cattaruzzi C,

Troncon MG, Agostinis L. Risk of

hospitalization for upper gastrointestinal

tract bleeding associated with ketorolac,

other nonsteroidal anti-inflammatory

drugs, calcium antagonists, and other

antihypertensive drugs. Arch Intern Med

1998; 158: 33–9.

28 Hallas J, Dall M, Andries A, Andersen

BS, Aalykke C, Hansen JM et al. Use of

single and combined antithrombotic

therapy and risk of serious upper

gastrointestinal bleeding: population

based case-control study. BMJ 2006;

333: 726.

29 Lin KJ, Hernández–Díaz S,

García Rodríguez LA. Acid suppressants

reduce risk of gastrointestinal bleeding

in patients on antithrombotic or

anti-inflammatory therapy.

Gastroenterology 2011; 141: 71–9.

30 Cai S, García Rodríguez LA,

Massó-González EL, Hernández-Díaz S.

Uncomplicated peptic ulcer in the UK:

trends from 1997 to 2005. Aliment

Pharmacol Ther 2009; 30: 1039–48.

31 Smith NLPB. Temporal trends in the use

of anticoagulants among older adults

with atrial fibrillation. Arch Intern Med

1999; 159: 1574–8.



http://journal.nzma.org.nz/journal/124-1331/4584/

http://journal.nzma.org.nz/journal/124-1331/4584/

Documents

Incidence of upper gastrointestinal haemorrhage in Maori and New Zealand European ethnic groups, 2001-2010