RESUMEN
BACKGROUND: Studies conducted in Europe as well as in North and South America have tried to link Helicobacter pylori colonization with the drinking water supply, especially since H. pylori is known to survive quite well in water. METHODS: In 2000, a cohort of 1884 grade-two children from two rural counties surrounding the city of Leipzig, Germany (77.4% of the 1991/1992 birth cohort) were tested for H. pylori colonization using the [13C]urea breath test. A parent-completed questionnaire elicited details on living conditions and lifestyle habits including questions on the children's drinking water from sources other than public water supplies, swimming in natural waters, etc. In a second independent study, samples of well water, taken from 157 private wells still used in the two counties, were being tested for the presence of H. pylori, using polymerase chain reaction (PCR) method to determine relevant target DNA fragments of H. pylori. RESULTS: In county I, 5.7% of the children and in county II 6.6% tested H. pylori-positive. Cluster analyses of the questionnaire data in both counties pointed to 'drinking water from other than municipal sources', as the closest H. pylori-associated cluster variable. The cluster estimations were supported by odds ratio (OR) calculations with an OR=16.4 (95% confidence interval (CI) 3.1,...,88.5) for county I and OR=4.0 (95% CI 1.3,...,12.4) for county II. The PCR analyses showed H. pylori DNA fragments in 10.8% of the wells in county I and 9.2% in county II. The detection limit was set at 10 DNA copies corresponding to 125 bacteria/L, the average infestation of these wells was 931 bacteria/L. CONCLUSION: Despite the fact that the microbiological and epidemiological data do not correspond except that both studies were conducted in the same geographical areas, the independent findings of H. pylori in well water in the same general areas where children do seem to drink water other than from the public water supply suggests that water may be an important source of H. pylori infection.