Helicobacter pylori is one of the few remaining major pathogens that accompanied humans on their travels from Africa. A recently published study reports the unexpected finding of a low H. pylori prevalence among pregnant women in Zanzibar (Farag, T.H., Stolzfus, R.J., Khalfan, S.S., Tielsch, J.M., 2007. Unexpectedly low prevalence of Helicobacter pylori infection among pregnant women on Pemba Island, Zanzibar. Trans. R. Soc. Trop. Med. Hyg. 101). The apparent epidemiology of higher prevalence with higher socioeconomic status and decrease with age are unprecedented. As with many ‘unexpected’ events, a search of the literature reveals evidence of low prevalence populations in Java and Malaysia, with clues dating back to the mid-twentieth century. Why some populations apparently lost H. pylori infection remains an open question. However, the tools needed to resolve the dilemma are readily available and we hope investigators will soon rise to the challenge.
The introductory paragraphs of research publications about Helicobacter pylori typically include statements about H. pylori being one of the most common pathogens affecting humankind and listing the various diseases it causes. The many studies of H. pylori epidemiology have repeatedly confirmed that (a) the infection is typically acquired in childhood; (b) there are inverse correlations between prevalence and socioeconomic status, household hygiene and sanitation; and (c) birth in a developing country correlates strongly with high prevalence. Genetic comparisons of H. pylori that have studied the patterns of human migration have shown that one can typically identify the region of origin of an individual by examination of their H. pylori strain (Falush et al., 2003; Linz et al., 2007; Vilaichone et al., 2004; Yamaoka et al., 2000a, 2000b; Yamaoka et al., 2002). For example, Amerindians living in remote areas tend to have ‘Asian-type’ H. pylori and those living in Central or South America who have had significant contact with immigrants of European descent typically have H. pylori originating from the Iberian peninsula (Falush et al., 2003; van Doorn et al., 1999; Yamaoka et al., 2002). Now, in this issue of Transactions, Farag et al. (2007) present a carefully conducted and analyzed prospective cohort study on pregnant women from Zanzibar, an island off the east coast of Africa. They used state-of-the-art noninvasive diagnostic methods to examine 857 pregnant women between the ages of 20 and 40 years who were attending antenatal care clinics in Pemba Island, Zanzibar, Tanzania, a location with limited sanitation and a generally poor standard of living (Farag et al., 2007).
The prevalence of H. pylori was expected to be high and similar to neighboring mainland Africa where the prevalence of infection is typically 80% or greater. They report that the overall prevalence of H. pylori infection was suprisingly low (17.5%). They also show that there was a correlation between the prevalence of the infection and residence near a major truck route (Farag et al., 2007). The epidemiologic correlates were almost exactly the opposite of what one has come to expect for H. pylori infections. The available data from other populations suggest that results from pregnant women are representative of the population as a whole. For example, Goodman et al. (2003) confirmed the expected high prevalence of H. pylori among pregnant women residing in the US-Mexico border region and the correlation with poor socioeconomic conditions. Similarly, in France H. pylori seroprevalence was significantly higher in non-French pregnant women than in French pregnant women, consistent with the effect of socioeconomic status on the acquisition of the infection (Kalach et al., 2002). In the study from Zanzibar, instead of H. pylori being predominately a disease of the disadvantaged, there was a direct correlation between H. pylori prevalence and socioeconomic status measured by education, occupation, possessions, housing, and even weight, height and hemoglobin levels. The typical age-related prevalence pattern of H. pylori infection in developing countries is for the prevalence to increase rapidly in childhood and then to plateau at a high proportion of the population (e.g. 80%) by age 20. By contrast, the prevalence among women on Pemba Island decreased with age (e.g. from 23% among women aged 20 or less decreasing to 13% by age 40). In developed countries, H. pylori prevalence increases with age reflecting a series of birth cohorts in which the risk of acquisition of the infection has continued to decrease over time. The data from Pemba Island imply the reverse (i.e. the presence of birth cohorts in which the risk has increased over time) suggesting that the infection is in the process of being introduced into the population.
The authors propose explanations for the low prevalence of H. pylori and the concentration of the infection along established truck routes and among the relatively well-off inhabitants. They find no obvious evidence of a practice or food that would have prevented infection in childhood, resulted in its loss in young adulthood or led to false-negative urea breath tests. They point out other populations in developing countries where the prevalence of infection is also markedly lower than expected. The current paradigm is that H. pylori has been a co-traveler with humans and that only recently has it begun to disappear (Falush et al., 2003; Linz et al., 2007; Yamaoka et al., 2002). Its disappearance has been correlated with improved standards of living and hygiene. Helicobacter pylori in humans can be traced back to Africa and the pattern is most consistent with humans having carried H. pylori as part of their parasitic burden (Linz et al., 2007). As noted before, peoples living remotely in the South American jungles are known to be infected with H. pylori most closely related to strains currently circulating in Asia, consistent with their origin from Asia. In contrast, Amerindians living in villages but with limited contact to those of European descent typically have European-type strains prompting investigators to suggest that H. pylori infection in the New World was a post-Columbian event (Kersulyte et al., 2000).
If absence of H. pylori implies loss of the infection in a population, and improved sanitation and standards of living lead to a break in the chain of transmission and eventual loss of the infection, why do some populations in developing countries have a low prevalence of the infection? There are no data to suggest that other low prevalence populations in developing countries (e.g. the Javanese) are descendents of a civilization with modern standards of living and hygiene. It appears more likely that these populations passed through bottlenecks in which the founding populations were uninfected. For example, founding populations may have been by chance made up of individuals without H. pylori. There are other data that suggest that populations existed where H. pylori infection was either absent or low. For example, investigators from the early and mid-twentieth century reported that in Malaysia and Java, gastric cancer and peptic ulcer, diseases typically associated with H. pylori, were rare in the indigenous populations but were common among the more recently arrived Chinese and Indian immigrants (Doll, 1952, 1958). After the discovery of H. pylori, these indigenous populations were shown to have lower than expected prevalence of H. pylori, and even those with peptic ulcer typically had a non-H. pylori associated ulcer (Goh, 1997; Goh and Parasakthi, 2001; Raj et al., 2001; Tokudome et al., 2005).
Clearly, the road map for unraveling this mystery would involve a more comprehensive epidemiologic study in Zanzibar that would cover a wider sample of the population, including men, women, and children, as well as additional tests to confirm the presence of the infection (e.g. stool antigen testing and serology). It is possible that increasing wealth is associated with abandoning a practice or food that inhibits H. pylori, so dietary practices should be investigated. It would be particularly important to examine and compare the actual H. pylori strains. Strain comparison is most likely to be definitive in Indonesia and Malaysia as strains from India and China are distinctive allowing one to state definitely whether infections among the original inhabitants were acquired from later immigrants, as was the case in village populations in Central and South America (Falush et al., 2003; Mukhopadhyay et al., 2000; Vilaichone et al., 2004; Yamaoka et al., 2000a, 2000b; Yamaoka et al., 2002). For example, one would expect that if the Chinese and Indian immigrants reintroduced H. pylori into the population, many infected Malays would have Chinese or Indian strains. Minimally invasive techniques such as the oral brush or string methods allow gastric sampling to be rapidly and easily accomplished (Graham et al., 2005; Windsor et al., 2005). The use of transport media and dry ice allows for samples to be collected and transported from even remote locations such that the experiments can be done quickly and easily.
In summary, as with many ‘unexpected’ events a search of the literature, in this instance the reviews by Richard Doll, gave clues to the presence of populations with low H. pylori prevalence. The questions about the reason for the apparent low prevalence of H. pylori remain open, but the problems in answering them are not technical as the tools needed to resolve the ‘whats’ and the ‘whys’ are readily available.
This material is based upon work supported in part by the Office of Research and Development Medical Research Service Department of Veterans Affairs and by Public Health Service grant DK56338, which funds the Texas Gulf Coast Digestive Diseases Center.
Conflicts of interest
Dr Graham is a consultant for Novartis in relation to vaccine development for treatment or prevention of H. pylori infection, and is a paid consultant for Otsuka Pharmaceuticals and a member of the Board of Directors of Meretek, Diagnostics, the manufacturer of the 13C-urea breath test. Dr Graham also receives royalties on the Baylor College of Medicine patent covering the serologic test, HM-CAP. Dr Malaty has received research support from AstraZeneca and Jannsen/Eisai. Dr Yamaoka is supported in part by NIH grant DK 62813.