Correspondence
Helicobacter pylori, Gastric Ulcer, and Duodenal Ulcer
N Engl J Med 1996; 335:1841-1843December 12, 1996
- Article
To the Editor:
In her editorial on the article by Hansson and associates (July 25 issue),1,2 Dr. Parsonnet suggests that the age at which Helicobacter pylori infection is acquired is the predictive factor for the different diseases induced by this bacterium (i.e., duodenal ulcer disease or gastric cancer). Yet our studies in China and studies in several developed countries all indicate that childhood is usually the period when H. pylori infection is acquired.3,4 Indeed, Dr. Parsonnet has stated elsewhere, “Adults who currently harbor the organism are more likely to have been infected in childhood than adulthood.”5
To examine the importance of the age at which H. pylori infection is acquired in relation to disease development, we compared the prevalence of H. pylori infection in an area where the incidence of gastric cancer is high (Lanzhou, northern China) with the prevalence of infection in an area where the incidence of gastric cancer is low (Guangzhou, southern China) and found that the age–prevalence patterns of H. pylori infection and the age-standardized prevalence were similar in the two cities.6 In addition, among patients undergoing endoscopic examination in these two cities, the prevalence of gastric cancer was high, with a lower prevalence of duodenal ulcer disease in Lanzhou, whereas in Guangzhou the prevalence of duodenal ulcer disease was high and that of gastric cancer was low.6 Indeed, our data indicated that the age at which H. pylori infection was acquired was not a specific marker of an increased risk of gastric cancer. Instead, such factors as the prevalence of atrophy appeared to be markers of an increased risk of gastric cancer.
Far from unmasking a paradox, this editorial has returned us to a hypothesis that is not supported by Dr. Parsonnet's own elegant studies, or those of other researchers, which have demonstrated that a cohort effect is responsible for the declining incidence of both gastric cancer and peptic ulcer disease in developed countries.7
We would argue against Dr. Parsonnet's view that the age at which H. pylori infection is acquired predicts the development of specific disease. We must look at the host, the environment, and perhaps the strain of H. pylori to understand the paradox that an increase in the incidence of duodenal ulcer disease reduces the risk of gastric cancer.
7 ReferencesHazel M. Mitchell, Ph.D.
Stuart L. Hazell, Ph.D.
University of New South Wales, Sydney, NSW 2052, Australia1
Parsonnet J . Helicobacter pylori in the stomach -- a paradox unmasked. N Engl J Med 1996;335:278-280
Full Text | Web of Science | Medline2
Hansson L-E ,Nyren O ,Hsing AW , et al. The risk of stomach cancer in patients with gastric or duodenal ulcer disease. N Engl J Med 1996;335:242-249
Full Text | Web of Science | Medline3
Mitchell HM ,Li YY ,Hu PJ , et al. Epidemiology of Helicobacter pylori in southern China: identification of early childhood as the critical period for acquisition. J Infect Dis 1992;166:149-153
CrossRef | Web of Science | Medline4
Banatvala N ,Mayo K ,Megraud F ,Jennings R ,Deeks JJ ,Feldman RA . The cohort effect and Helicobacter pylori. J Infect Dis 1993;168:219-221
CrossRef | Web of Science | Medline5
Parsonnet J . The incidence of Helicobacter pylori infection. Aliment Pharmacol Ther 1995;9:Suppl 2:45-51
Web of Science | Medline6
Hu PJ ,Li YY ,Lin HL , et al. Gastric atrophy and regional variation in upper gastrointestinal disease. Am J Gastroenterol 1995;90:1102-1106
Web of Science | Medline7
Westbrook JI ,Rushworth RL . The epidemiology of peptic ulcer mortality 1953-1989: a birth cohort analysis. Int J Epidemiol 1993;22:1085-1092
CrossRef | Web of Science | Medline
To the Editor:
Dr. Parsonnet finds it paradoxical that H. pylori infection causes different ulcers with different risks of cancer. She hypothesizes that “childhood infection predisposes patients to gastric ulcers and stomach cancer, but it inhibits the development of duodenal ulcers.” Most researchers believe that the cause of H. pylori–associated ulcers and cancer is multifactorial. Diet, rather than early infection, may be one of several environmental determinants. For example, in two ethnically similar Chinese populations, H. pylori infection is acquired at an early age (less than five years in 34 percent of cases), but the patterns of ulcer and cancer development are markedly different: Guangzhou, with more duodenal ulcers and fewer gastric cancers, has fresh vegetables available year-round; Lanzhou, with more gastric ulcers and cancers, has only seasonal fresh vegetables, with year-round salted and spiced foods.1 It is not paradoxical for the same organism to cause different diseases if the cause is multifactorial.
Kuipers et al. (April 18 issue)2 state that long-term treatment with omeprazole causes atrophic fundal gastritis, and Dr. Parsonnet concludes, “Current therapies might be advancing the cancer clock by converting relatively benign gastric inflammation into a more destructive, premalignant process.” Kuipers et al. do, however, report the virtual absence of intestinal metaplasia, which is especially pertinent, because in patients with atrophic gastritis, intestinal metaplasia is the only generally accepted precursor of gastric cancer.
What do Kuipers et al. mean by atrophic gastritis? I have found two illustrations elsewhere that indicate their concept of atrophic gastritis: one, purporting to show a “loss of glands” (Figure 3 of the cited article),3 shows only mild, superficial gastritis; the other, in Kuipers's doctoral thesis, misinterprets a lymphoid aggregate in a biopsy specimen obtained after eight weeks of omeprazole therapy as “invasion of inflammatory cells throughout the complete glandular layer with loss of glands.” However, lymphoid aggregates are seen in many untreated and treated patients with H. pylori infection.4
Physicians now worry whether long-term treatment with omeprazole predisposes patients to cancer. The evidence is clearly against such a predisposition, but the worry can be eliminated by eradicating preexisting H. pylori infection before initiating long-term treatment with proton-pump inhibitors.
4 ReferencesCyrus E. Rubin, M.D.
University of Washington, Seattle, WA 981951
Hu PJ ,Li YY ,Lin HL , et al. Gastric atrophy and regional variation in upper gastrointestinal disease. Am J Gastroenterol 1995;90:1102-1106
Web of Science | Medline2
Kuipers EJ ,Lundell L ,Klinkenberg-Knol EC , et al. Atrophic gastri-tis and Helicobacter pylori infection in patients with reflux esophagitis treated with omeprazole or fundoplication. N Engl J Med 1996;334:1018-1022
Full Text | Web of Science | Medline3
Kuipers EJ ,Uyterlinde AM ,Pena AS , et al. Increase of Helicobacter pylori-associated corpus gastritis during acid suppressive therapy: implications for long-term safety. Am J Gastroenterol 1995;90:1401-1406
Web of Science | Medline4
Genta RM ,Hamner HW ,Graham DY . Gastric lymphoid follicles in Helicobacter pylori infection: frequency, distribution, and response to triple therapy. Hum Pathol 1993;24:577-583
CrossRef | Web of Science | Medline
To the Editor:
Hansson et al. report an inverse relation between duodenal ulcer disease and gastric cancer. Both conditions have frequently been associated with the presence of H. pylori. Parsonnet sees a paradox if H. pylori is an etiologic factor in each condition and explains it by proposing that the nature of the illness caused by H. pylori depends on the duration of the presence of the organism in the host.
For Parsonnet's proposition to be credible, duodenal ulcer would have to be found commonly in young adults in underprivileged communities, and gastric cancer would have to be found commonly in the elderly in industrialized countries. The facts do not support the proposition. A more plausible explanation is that H. pylori is simply an opportunistic organism in any mucosal lesion in the upper gastrointestinal tract, as suggested elsewhere.1,2
Despite the association between H. pylori infection and peptic ulceration, gastric cancer, and gastric lymphoma proposed by many researchers and even the National Institutes of Health,3 Marshall, the original proponent of the association, admits that Koch's postulates have never been fulfilled for any such association, despite a decade of concerted attempts.4
Surely it is time that gastroenterologic researchers started looking at the facts, instead of trying to explain the myriad paradoxes about H. pylori with what is undoubtedly guesswork.
4 ReferencesJohn R. Graham, M.B.
Sydney Hospital, Sydney, NSW 2000, Australia1
Graham JR . Helicobacter pylori: human pathogen or simply an opportunist? Lancet 1995;345:1095-1097
CrossRef | Web of Science | Medline2
Fallingborg J ,Poulsen LO ,Grove A ,Teglbjaerg PS . Frequency of Helicobacter pylori and gastritis in healthy subjects without gastrointestinal symptoms. Scand J Gastroenterol 1992;27:388-390
CrossRef | Web of Science | Medline3
NIH Consensus Development Panel on Helicobacter pylori in Peptic Ulcer Disease
CrossRef | Web of Science4
Marshall BJ . Helicobacter pylori in peptic ulcer: have Koch's postulates been fulfilled? Ann Med 1995;27:565-568
CrossRef | Web of Science | Medline
Author/Editor Response
The author replies:
To the Editor: Mitchell and Hazell, as well as Rubin, submit that other factors, besides the age at which H. pylori infection is acquired, determine the dichotomous outcome of H. pylori infection. The proposed factors include the strain of the infecting organism, the host's genetic characteristics, and environmental exposure. Undoubtedly, these factors play a part in H. pylori–related diseases. Yet the paradox is not that, given different predisposing influences, H. pylori infection causes different diseases — many infectious agents do that. The paradox rests in a peculiar segregation of outcomes — namely, that persons with duodenal ulcer disease actually have a diminished likelihood of contracting gastric cancer. There is no evidence that the strain of the infecting organism dichotomizes the outcome in this manner. Virulent strains of H. pylori are similarly linked to both ulcers and cancer.1
Mitchell and Hazell's own work controverts a strong genetic effect on the segregation of duodenal ulcer and cancer. In ethnically matched populations from different regions of China, they found markedly different rates of the two diseases.2 It is possible that an unrecognized environmental factor increases the risk of cancer while decreasing the risk of ulcer disease. This factor would have to exert its influence in childhood, because the risk of gastric cancer appears to be determined at a young age. Since such a factor has not been identified, the reductionist view is that the factor is H. pylori infection itself. This hypothesis at least provides a starting point for explaining the paradoxical findings of Hansson et al.
Mitchell and Hazell argue that the age hypothesis cannot hold true because H. pylori infection is almost uniformly a childhood infection. This is not the case. Adult infections do occur, and as a proportion of prevalent infections, they increase as the overall incidence declines. In parts of China where 6 percent of children may acquire H. pylori infection each year, approximately 85 percent of infections in adults are acquired before the age of 10 years. In the United States, where the incidence in childhood is much lower (perhaps 1 percent per year), a much smaller proportion of the prevalence of H. pylori infection is attributable to childhood acquisition (the minority, according to my calculations). These differences among various populations in the age at which the infection is acquired help explain nonparallel declines in cases of cancer and ulcer disease, despite a similar prevalence of H. pylori infection among adults.
I agree with Dr. Rubin that the relation between acid inhibition and the acceleration of precursors of gastric cancer remains hypothetical — hence, the somewhat tentative nature of my comments on this subject. Dr. Graham reiterates the skepticism about H. pylori that he expressed in detail one year ago.3 At that time, other writers eloquently refuted his arguments.4,5 I will not revisit the debate here, except to note that requiring the fulfillment of Koch's postulates to establish causality is not always tenable. Should we systematically infect humans with the human immunodeficiency virus to see whether it causes AIDS? It is to be hoped that in the future, model systems will yield the evidence that Dr. Graham requires. Until that time, however, it is imprudent to discount a preponderance of clinical, physiologic, and epidemiologic evidence linking H. pylori to disease.
5 ReferencesJulie Parsonnet, M.D.
Stanford University, Stanford, CA 94305-50921
Blaser MJ . Intrastrain differences in Helicobacter pylori: a key question in mucosal damage? Ann Med 1995;27:559-563
CrossRef | Web of Science | Medline2
Hu PJ ,Li YY ,Lin HL , et al. Gastric atrophy and regional variation in upper gastrointestinal disease. Am J Gastroenterol 1995;90:1102-1106
Web of Science | Medline3
Graham JR . Helicobacter pylori: human pathogen or simply an opportunist? Lancet 1995;345:1095-1097
CrossRef | Web of Science | Medline4
Harris A ,Misiewicz JJ . Helicobacter pylori: a human pathogen. Lancet 1995;345:1579-1579
CrossRef | Web of Science | Medline5
Murray L ,Harvey I . Helicobacter pylori: a human pathogen. Lancet 1995;345:1579-1580
CrossRef | Web of Science | Medline
Author/Editor Response
Dr. Rubin believes that we state that omeprazole causes atrophic gastritis. However, we showed exactly the opposite: omeprazole does not cause atrophic gastritis. The cause of atrophy is H. pylori, not omeprazole. We reported that patients can be treated for many years with omeprazole, without atrophy developing. We do state that profound acid suppression increases the severity of H. pylori–induced inflammation in the gastric corpus; this inflammation is associated with an increased incidence of atrophic gastritis in H. pylori–infected persons.
Dr. Rubin also questions our ability to diagnose atrophy, referring to two illustrations. These were not presented for the purpose of showing atrophy. Another picture, intended to show atrophic gastritis,1 was included in the same thesis and is therefore known to Dr. Rubin. It is very similar to one that he himself used.2 Our standards for identifying atrophy are in agreement with the revised Sydney classification.3
To ensure reliable and reproducible results in our study, biopsy samples were obtained according to a strict protocol, encoded, and then analyzed by one of us, an experienced pathologist who has been involved in some 40 studies of gastritis. Our findings are supported by those of other studies4,5 and are biologically plausible, given the effect of pH on H. pylori and H. pylori gastritis. We do, of course, accept the possibility that other factors such as diet may influence the development of atrophy, but their relevance is not clear with respect to our study.
The clinical implications of our findings are still unknown. Atrophic gastritis is considered to represent an initial step in a process that in some people can lead to dysplasia and cancer.6 In the intestinal type of carcinoma, this process often proceeds through intestinal metaplasia types II and III. The absence of these types of metaplasia in our patients is noteworthy. We do not know what drives the different steps of the process leading to gastric cancer, and it is too early to know whether metaplasia and dysplasia will occur in patients with gastric atrophy that developed during profound acid suppression. Neither are there any data on the effects of eradicating H. pylori in patients undergoing long-term acid suppression. Both our study and others demonstrate the relation among acid, H. pylori, and atrophic gastritis. We hope that these findings will stimulate further research and discussion, including the consideration of whether to eradicate H. pylori in patients requiring long-term, profound acid suppression.
6 ReferencesErnst J. Kuipers, M.D., Ph.D.
Stephan G.M. Meuwissen, M.D., Ph.D.
Free University Hospital, 1007 MB Amsterdam, the NetherlandsNiilo Havu, M.D., Ph.D.
Astra, S-151 85 Södertälje, Sweden1
Kuipers EJ ,Uyterlinde AM ,Pena AS , et al. Long-term sequelae of Helicobacter pylori gastritis. Lancet 1995;345:1525-1528
CrossRef | Web of Science | Medline2
MacDonald WC ,Rubin CE . Gastric biopsy -- a critical evaluation. Gastroenterology 1967;53:143-170
Web of Science | Medline3
Dixon MF, Genta RM, Yardley JH, Correa P, Participants in the International Workshop on the Histopathology of Gastritis, Houston 1994
CrossRef | Web of Science | Medline4
Eissele R ,Brunner G ,Solcia E ,Arnold R . Gastric mucosa during treatment with lansoprazole: Helicobacter pylori is a major risk factor for argyrophil-cell hyperplasia. Gastroenterology 1996;110:Suppl:A101-A101 abstract.
Web of Science5
Lamberts R ,Creutzfeldt W ,Struber HG ,Brunner G ,Solcia E . Long-term omeprazole therapy in peptic ulcer disease: gastrin, endocrine cell growth, and gastritis. Gastroenterology 1993;104:1356-1370
Web of Science | Medline6
Correa P . Human gastric carcinogenesis: a multistep and multifactorial process. Cancer Res 1992;52:6735-6740
Web of Science | Medline
- Citing Articles (6)
Citing Articles
1
Igor Charvet, Gaëlle Ory, Philippe Thueler, Marie-Anne Brundler, Michel Saint-Ghislain, Nadereh Azarpey, Antoine Hadengue, Christian Depeursinge, Bernard Vermeulen, Paolo Meda. (2004) Diagnosis and grading of gastritis by non-invasive optical analysis. European Journal of Gastroenterology & Hepatology 16:11, 1189-1198
CrossRef2
Jonathan D. Kaunitz, Yasutada Akiba. (2004) Gastroduodenal mucosal defense: role of endogenous mediators. Current Opinion in Gastroenterology 20:6, 526-532
CrossRef3
Luke C. Bi, Jonathan D. Kaunitz. (2003) Gastroduodenal mucosal defense: an integrated protective response. Current Opinion in Gastroenterology 19:6, 526-532
CrossRef4
J Torres. (2000) A Comprehensive Review of the Natural History of Helicobacter pylori Infection in Children. Archives of Medical Research 31:5, 431-469
CrossRef5
F Pakodi. (2000) Helicobacter pylori. One bacterium and a broad spectrum of human disease! An overview. Journal of Physiology-Paris 94:2, 139-152
CrossRef6
Ranjan Dohil, Eric Hassall, Gareth Jevon, James Dimmick. (1999) Gastritis and Gastropathy of Childhood. Journal of Pediatric Gastroenterology & Nutrition 29:4, 378-394
CrossRef
