Q J Med 2002; 95: 519-525
© 2002 Association of Physicians
Relationship between the birth cohort pattern of Helicobacter pylori infection and the epidemiology of duodenal ulcer
From the Department of Medicine, Frenchay Hospital, Bristol, and the Department of Social Medicine, University of Bristol, Bristol, UK
Received 7 January 2002 and in revised form 7 April 2002
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: Helicobacter-pylori-related duodenal ulcer (DU) is an important cause of dyspepsia.
Aim: To determine the relationship between the pattern of H. pylori infection and the epidemiology of duodenal ulcer in a single population.
Design: Prospective two-part study of (i) patients with DU referred for endoscopy because of dyspepsia, and (ii) the incidence of H. pylori infection in the general population of the same area.
Methods: Details of 533 DU patients were recorded, and related to the pattern of H. pylori infection among 10 537 adults in the same community, determined by the 13C-urea breath test.
Results: In patients with DU, birth year was more important than age in determining the rate of presentation for endoscopy (the ‘birth cohort’ effect). H. pylori infection showed a similar birth cohort effect, and the prevalence decreased steadily in those born in successive years, from 28.8% in the 1930s to 3.5% in the 1970s. The proportion of dyspeptic patients who had duodenal ulcers also fell progressively, from 22.2% in 1979 to 5.7% in 1998. H. pylori prevalence and duodenal ulcer incidence were closely correlated at all ages.
Discussion: Duodenal ulcer prevalence (as judged by the rate of referral of duodenal ulcer patients for endoscopy) is determined principally by the distribution of H. pylori infection in the local population. The birth cohort effect seen in adult duodenal ulcer patients reflects the acquisition of H. pylori in childhood. In Bristol, H. pylori prevalence and duodenal ulcer incidence are both declining to very low levels.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
At least 25% of the adult population in the UK have dyspepsia, and about 2% of the population per year consult their general practitioner for this reason.1 Duodenal ulcer has been recognized as a common cause of dyspepsia, which is a major and costly public health problem. Large amounts of money are spent on medications and investigations for dyspepsia, with substantial further costs incurred as a result of hospitalization, time lost from work and the payment of sickness benefit.2,3
The management of patients presenting with dyspepsia due to a duodenal ulcer has been revolutionized by the discovery that nearly all duodenal ulcers occurring in patients who are not taking non-steroidal anti-inflammatory drugs (NSAIDs) are associated with chronic Helicobacter pylori infection,4–8 which is usually acquired in childhood.9,10 Eradication of the H. pylori infection cures the ulcer,5–8,11 usually permanently, since reacquisition of H. pylori infection is rare.12 Various strategies have been developed that aim to identify patients whose dyspepsia is due to a duodenal ulcer resulting from chronic H. pylori infection. These include performing early endoscopy in all dyspeptic patients,13,14 testing for H. pylori infection and treating any infected patients to eradicate the organism,15 testing for H. pylori infection and referring for endoscopy only those who are infected,16 or treating dyspeptic patients symptomatically and investigating only if the symptoms are severe or persistent.17 It is not easy to compare the cost-effectiveness of such strategies, since this depends not only on the costs of the investigations and treatment, but also on the prevalence of H. pylori infection in the local community and on the incidence of H. pylori-related duodenal ulcer among patients from that community presenting with dyspepsia. This information is rarely available, and the exact relationship between H. pylori infection and duodenal ulcer incidence in a single population has not been described previously.
We have studied the incidence of Helicobacter-related duodenal ulcer as a cause of dyspepsia over a period of 20 years in a single health district, and compared this with the H. pylori prevalence in different age groups in the same population, obtained as part of the community-based Bristol Helicobacter Project.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The study consisted of two separate parts, one based in the district general hospital responsible for providing a diagnostic service for all dyspeptic patients living in the defined catchment area of Frenchay Health District, and one based in the community served by that hospital.
In part 1, patients with undiagnosed dyspepsia (defined as intermittent epigastric pain or discomfort) who lived in Frenchay Health District and were referred to the medical gastroenterology service at Frenchay Hospital, Bristol, for diagnostic endoscopy during the 20-year period 1979–1998 were studied. An open-access endoscopy service was available throughout this period, with a waiting time usually of about 4–6 weeks. Patients with chest pain, heartburn or reflux, but without epigastric pain, and any patients with a previously diagnosed ulcer or who were taking NSAIDs, were excluded. Age, sex, indications for endoscopy and findings were recorded prospectively for all patients. Duodenal ulcer was defined conventionally, as an ulcer >5 mm in diameter, situated in the duodenum and/or pyloric channel. Acid-suppressant medication was discontinued in all patients for a period of at least 2 weeks before their endoscopy.
In part 2, the Bristol Helicobacter Project, all people aged 20–59 years who were registered with seven general practices within the Frenchay Hospital catchment area in North East Bristol (total 26 203) were invited to participate in a community-based prospective randomized controlled trial of the effects of H. pylori eradication on dyspepsia, quality of life, resource utilization and various other outcome measures. These practices serve about 20% of the population of the Frenchay Hospital catchment area. Overall, 10 537 (40.2%) gave informed consent to take part in the study, and had a 13C-urea breath test for active H. pylori infection, using a standard orange juice and citric acid test meal, with a cut-off of 
3.5 per ml.18
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Part 1: Study population and changes over the 20-year period
533 patients (354 men and 179 women) presenting with dyspepsia had active non-NSAID-related duodenal ulcers (Table 1
). The age distribution was essentially the same for men and women, except for a small but significant increase in the proportion of women in the over-70 year age group (
2=11.2, p<0.01). The distribution of birth years showed a peak in the 1930s, with nearly half (49.7%) of all the patients having been born in the 1920s and 1930s.
|
There was relatively little difference in the median year of birth between those patients diagnosed between 1979 and 1983 and those diagnosed between 1994 and 1998 (1933 and 1936, respectively) (Table 2
), despite the 15-year difference in mean date of endoscopy (Figure 1). Consequently, patients in whom the diagnosis of duodenal ulcer was made more recently were an average of 8.4 years older (Table 2). The date of birth was thus more important than the age of the patients in determining their duodenal ulcer risk, indicating the existence of a birth cohort effect for duodenal ulcer in this population.
|
|
The proportion of patients whose dyspepsia was due to a duodenal ulcer declined steadily from 22.2% in 1979 to 5.7% in 1998 (Figure 2
). If this decline in the proportion of dyspeptic patients with duodenal ulcer continues at the same rate, their contribution to the burden of dyspepsia in this community should be negligible by the year 2005.
|
Part 2: H. pylori infection rates in the local community, and their relationship with duodenal ulcer incidence
Of the 10 537 subjects who had a 13C-urea breath test done, 1634 (15.5%) were positive for H. pylori infection. Infection rates correlated strongly with age and inversely with year of birth (r=0.99, p<0.001 for both correlations) (Figure 3). The population of the Frenchay Hospital catchment area (taken from Area Health Authority statistics for 1995) was subdivided by years of birth into 5-year cohorts. The numbers of dyspeptic patients with duodenal ulcers seen in 1989–1998 who were born in those same 5-year periods were recorded, and the duodenal ulcer incidence for each cohort was calculated (Table 3). The duodenal ulcer incidence (as reflected in referrals for diagnostic endoscopy) for each cohort was then compared with the corresponding H. pylori infection rate in the catchment population. An average of 0.78±0.23 (SD) dyspeptic patients per 1000 helicobacter-positive subjects per year were diagnosed by the endoscopy service as having duodenal ulcers, a rate of incidence which was similar for all but the youngest age groups. A plot of the rate of presentation of duodenal ulcer patients born in each of these 5-year cohorts vs. the corresponding H. pylori prevalence found in the same cohorts (Figure 4), indicates that when H. pylori prevalence reaches zero, the duodenal ulcer incidence will be negligible (0.005 per 1000 per year).
|
|
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
In most of the UK, as in many other parts of the developed world, the incidence of duodenal ulcer has been falling steadily since the 1960s,19–22 probably as a result of a decline in H. pylori infection as living conditions in childhood have improved. However, in areas where the H. pylori prevalence remains high, such as Glasgow and Northern Ireland,23,24 there does not as yet appear to be much of a decline in duodenal ulcer incidence.
Susser was the first to note that peptic ulcer death rates correlated closely with years of birth,25 giving rise to a so-called ‘cohort effect’.26,28 The cause of such cohort effects was unknown at that time, but now can be explained by differences in the rates of acquisition of H. pylori infection in childhood, which show a similar cohort effect.29
The figures obtained in this study need careful interpretation, because of several potential sources of bias. Firstly, there would have been a change with time in the pattern of referral of dyspeptic patients for diagnosis. Differences in the overall referral rate over the years, and the steadily declining use of other methods of diagnosis such as barium meals, would have had an effect. Secondly, in recent years, the introduction of new policies for dyspepsia management in general practice, such as those involving testing dyspeptic patients for H. pylori infection and then either giving H. pylori eradication therapy (‘test and treat’) or else referring H. pylori positive patients for endoscopy (‘test and refer’), might be expected to have affected the rate of decline of duodenal ulcer as a cause of dyspepsia. In the population that we studied, these policies seem in to have had little effect up to 1998, since the decline in duodenal ulcer among dyspeptic patients was exactly the same in the second 10 years of the study (when such policies were being introduced) as it was in the first 10 years (before they had been developed) (Figure 2
). This may have resulted from a slow take-up of such policies in this area, although there are no objective figures on this. Thirdly, we have not included patients whose ulcers presented with complications such as haemorrhage, because it was not possible to record their details prospectively, and the quality of the hospital records is inadequate for any retrospective study. These sources of bias will have resulted in a significant underestimate of the total incidence of duodenal ulcers, but should have had only a minor effect on the proportional figures that we calculated. Any study of this type can never give a true total incidence for duodenal ulcer, since some are asymptomatic, some do not present to their general practitioners, some are never referred for accurate diagnosis, and some are referred elsewhere.
Our finding of a relatively constant rate of presentation of duodenal ulcers with dyspepsia at different ages (except in the youngest age group, in whom the rate was lower), when corrected for H. pylori prevalence, is consistent with previous observations on duodenal ulcer.30–32 Doll, Avery Jones and Buckatzsch in 1951,31 at a time when H. pylori prevalence was probably high in all age groups of their study population, calculated a total annual incidence of duodenal ulcer in men in the UK of 3.2 per 1000, which was almost constant between the ages of 35 and 64 years. The reason for the lower rate of duodenal ulcer presentation in the youngest cohort that we studied, those born in 1970–1974, may be that many of those in this cohort who were destined to develop ulcers were still too young to have presented. They would have been aged 15–19 at the beginning of the 10-year period of observation and 25–29 at the end: according to previous studies,31,32 fewer than half of all duodenal ulcer patients would have presented by those ages.
We have assumed for the purposes of this study that the Helicobacter Project participants and the duodenal ulcer patients were representative of the population of Frenchay Health District (the catchment area of Frenchay Hospital). The General Practices participating in the Bristol Helicobacter Project are all within the catchment area of Frenchay Hospital, and serve about 20% of its population. The population characteristics of the different practices differ slightly from each other and from those of the whole of the district, mainly in the relative proportion of elderly people in each. This should not have had any effect on our results, since we confined our analysis to the age range 20–59. Markers of social deprivation in childhood or in adulthood showed non-significant differences between the individual general practices, and were similar to the figures for the district as a whole. The H. pylori infection rates and the changes in these between successive birth cohorts did not differ significantly between individual general practices, and are therefore likely to be representative of those throughout the whole of the district.
The duodenal ulcer population was derived solely from the Frenchay Hospital catchment area, since we only included patients living there. The most significant source of error in our figures is probably related to our confining our analysis to the well-documented patients seen by the medical endoscopy service, and excluding those diagnosed elsewhere (e.g. by radiology). This would have reduced the total number of duodenal ulcer patients available for analysis, and thus would have underestimated the figures for total incidence, but should not materially have affected the relative changes in the proportion of dyspeptic patients with duodenal ulcers that we demonstrated, nor the relationship with H. pylori prevalence in individual birth cohorts.
Our findings are all consistent with the suggestion that the community prevalence of H. pylori infection is the most important determinant of non-NSAID-related duodenal ulcer incidence. The decline in duodenal ulcer incidence in Bristol in recent years is largely accounted for by the progressive decline in the number of younger people who are infected by H. pylori. This is happening at such a rate that, in this area, duodenal ulcer will probably become of minor importance as a cause of dyspepsia in the near future. In this part of North-East Bristol, the cost-effectiveness of any policies for dyspepsia management that include testing for H. pylori infection is already borderline for younger patients who were born in this area.
Accurate knowledge of the age-related H. pylori prevalence in any area increases the accuracy of estimates of the cost-effectiveness of different strategies (such as ‘test and treat’ or ‘test and refer’) for the management of dyspeptic patients in that area. The costs of such strategies will vary greatly between different parts of the UK, depending on the local pattern of H. pylori infection, which is still known accurately for only a few areas. For example, the H. pylori infection rate in adults aged 40–49 years found in community-based studies in the UK varies from 15% in Bristol33 to 28% in Leeds34 and 60% in Glasgow23. The contribution of duodenal ulcers to the burden of dyspepsia in Bristol is already very small, whereas in Glasgow it remains large, and in Leeds it is intermediate. A public health policy for the management of dyspepsia needs to be tailored to suit the location, and to be based on accurate knowledge of the local prevalence of H. pylori infection.
|
Acknowledgments |
|---|
We thank the participants in the Bristol Helicobacter Project and the general practitioners and Health Centre staff; the nursing team of Lynne Bradshaw, Julie Watson, Tina Critchley, Jo Lee, Carol Everson-Coombe, Penny Nettlefield and Joanne Smith; Judy Millward, Helen Davies, Amy Hawkins and Sarah Pike for secretarial support and Erwin Brown, Paul Thomas, Nick Pope and Phil Hedges of the Microbiology Department and Peter Spurr, Martin Bullock and Fiona Greenwood of the Pharmacy Department, Frenchay Hospital, for help with the 10 537 breath tests. This study was funded jointly by the NHS South and West Regional Research and Development Directorate and GlaxoSmithKline UK. The Department of Social Medicine is the lead centre for the MRC Health Services Research Collaboration.
|
Notes |
|---|
Address correspondence to Dr R.F. Harvey, Department of Medicine, Frenchay Hospital, Bristol BS16 1LE. e-mail: richard.harvey1{at}virgin.net
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Jones R, Lydeard S. Prevalence of symptoms of dyspepsia in the community. Br Med J1989; 298:30–2.
2. Nyren O, Adami HO, Gustavsson S, Loof L. Social and economic effects of non-ulcer dyspepsia. Scand J Gastroenterol2001; 209:41–7.
3. Levin TR, Schmittdiel JA, Kunz K, Henning JM, Henke CJ, Colby CJ, et al. Costs of acid-related disorders to a health maintenance organization. Am J Med1997; 103:520–8.[Web of Science][Medline]
4. Marshall BJ, Warren JR. Unidentified curved bacilli in the stomach of patients with gastritis and peptic ulceration. Lancet1984; 1:1311–15.[Web of Science][Medline]
5. O'Connor HJ, Dixon MF, Wyatt JI, Axon AT, Dewar EP, Johnston D. Campylobacter pylori and peptic ulcer disease. Lancet1987; 2:633–4.[Medline]
6. Coghlan JG, Gilligan D, Humphries H, McKenna D, Dooley C, Sweeney E, et al. Campylobacter pylori and recurrence of duodenal ulcers: a 12-month follow-up study. Lancet1987; 2:1109–11.[Web of Science][Medline]
7. Marshall BJ, Goodwin CS, Warren JR, Murray R, Blincow ED, Blackbourn SJ, et al. Prospective double-blind trial of duodenal ulcer relapse after eradication of Campylobacter pylori. Lancet1988; 2:1437–42.[Web of Science][Medline]
8. Rauws EA, Tytgat GN. Cure of duodenal ulcer associated with eradication of Helicobacter pylori. Lancet1990; 335:1233–5.[Web of Science][Medline]
9. Thomas JE, Dale A, Harding M, Coward WA, Cole TJ, Weaver LT. Helicobacter pylori colonization in early life. Pediat Res1999; 45:218–23.[Web of Science][Medline]
10. Rothenbacher D, Inceoglu J, Bode G, Brenner H. Acquisition of Helicobacter pylori infection in a high-risk population occurs within the first 2 years of life. J Pediatrics2000; 136:744–8.[Web of Science][Medline]
11. Hentschel E, Brandstatter G, Dragosics B, Hirschl AM, Nemec H, Schutze K, et al. Effect of ranitidine and amoxicillin plus metronidazole on the eradication of Helicobacter pylori and the recurrence of duodenal ulcer. N Engl J Med1993; 328:308–12.
12. Parsonnet J, Blaser MJ, Perez-Perez GI, Hargrett-Bean N, Tauxe RV. Symptoms and risk factors of Helicobacter pylori infection in a cohort of epidemiologists. Gastroenterology1992; 102:41–6.[Web of Science][Medline]
13. Fjosne U, Kleveland P, Waldum H, Halvorsen T, Petersen H. The clinical benefit of routine upper gastrointestinal endoscopy. Scand J Gastroenterol1986; 21:433–40.[Web of Science][Medline]
14. Lassen AT, Pedersen FM, Bytzer P, Schaffalitzky de Muckadell OB. Helicobacter pylori test-and-eradicate versus prompt endoscopy for management of dyspeptic patients: a randomised trial. Lancet2000; 356:455–60.[Web of Science][Medline]
15. Heaney A, Collins JS, Watson RG, McFarland RJ, Bamford KB, Tham TC. A prospective randomised trial of a ‘test and treat’ policy versus endoscopy based management in young Helicobacter pylori positive patients with ulcer-like dyspepsia, referred to a hospital clinic. Gut1999; 45:186–90.
16. Patel P, Khulusi S, Mendall MA, Lloyd R, Jazrawi R, Maxwell JD, et al. Prospective screening of dyspeptic patients by Helicobacter pylori serology. Lancet1995; 346:1315–18.[Web of Science][Medline]
17. Bytzer P, Hansen JM, Schaffalitzky de Muckadell OB. Empirical H2-blocker therapy or prompt endoscopy in management of dyspepsia. Lancet1994; 343:811–16.[Web of Science][Medline]
18. Dominguez-Munoz JE, Leodolter A, Sauerbruch T, Malfertheiner P. A citric acid solution is an optimal test drink in the 13C-urea breath test for the diagnosis of Helicobacter pylori infection. Gut1997; 40:459–62.
19. Bonnevie O. The incidence of duodenal ulcer in Copenhagen county. Scand J Gastroenterol1975; 10:385–93.[Web of Science][Medline]
20. Coggon D, Lambert P, Langman MJ. 20 years of hospital admissions for peptic ulcer in England and Wales. Lancet1981; 1:1302–4.[Web of Science][Medline]
21. Quartini A, Negri E, La Veccia C. Trends in peptic ulcer mortality in Italy 1955–1985. J Epidemiol Comm Health1992; 46:494–7.
22. Munnangi S, Sonnenberg A. Time trends of physician visits and treatment patterns of peptic ulcer disease in the United States. Arch Intern Med1997; 157:1489–94.
23. Woodward M, Morrison C, McColl K. An investigation into factors associated with Helicobacter pylori infection.J Clin Epidemiol2000; 53:175–81.[Web of Science][Medline]
24. Murray LJ, McCrum EE, Evans AE, Bamford KB. Epidemiology of Helicobacter pylori infection among 4742 randomly selected subjects from Northern Ireland. Int J Epidemiol1997; 26:880–7.
25. Susser M. Environmental factors and peptic ulcer. The Practitioner1961; 186:302–18.[Web of Science]
26. Sonnenberg A. Occurrence of a cohort phenomenon in peptic ulcer mortality from Switzerland. Gastroenterology1984; 86:398–401.[Web of Science][Medline]
27. Katschinski BD. Logan RF. Changes in birth-cohort pattern of peptic ulcer mortality in England and Wales. Postgrad Med J1991; 67:825–8.
28. Westbrook JI, Rushworth RL. The epidemiology of peptic ulcer mortality 1953–1989: a birth cohort analysis. Int J Epidemiol1993; 22:1085–92.
29. Banatvala N, Mayo K, Megraud F, Jennings R, Deeks JJ, Feldman RA. The cohort effect and Helicobacter pylori. J Infect Dis1993; 168:219–21.[Web of Science][Medline]
30. Jamieson RA, Smith WE, Scott LDW. Peptic ulcer in Glasgow. Br Med J1949; 1:298–301.
31. Doll R, Avery Jones F, Buckatzsch M. Occupational factors in the aetiology of gastric and duodenal ulcers. Special Report Series No. 276. Medical Research Council, London1951.
32. Fry J. Peptic ulcer: a profile. Br Med J1964; 1:809–12.
33. Lane AJ, Egger M, Murray LJ, Donovan JL, Harvey IM, Nair P, et al. Socioeconomic status in childhood but not adulthood predicts H. pylori infection: The Bristol Helicobacter Project. Gastroenterology2000; 118:a2557.
34. Moayyedi P, Feltbower R, Brown J, Mason S, Mason J, Nathan J, et al. Effect of population screening and treatment for Helicobacter pylori on dyspepsia and quality of life in the community: a randomised controlled trial. Lancet2000; 355:1665–9.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. C Ford and N. J Talley Does Helicobacter pylori really cause duodenal ulcers? Yes BMJ, August 14, 2009; 339(aug14_1): b2784 - b2784. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||