Results

A total of 448 (44% (95% confidence interval 41–47%)) of the 1025 men were seropositive to CagA antibodies, representing about 61% (57–65%) of the men seropositive to mixed antigenH pylori (table 1). Seropositivity to CagA antibodies was more common in men resident in northern England and Scotland than in southern England (49%v 28%, p<0.00001), with a similar geographical pattern for seropositivity to mixed antigenH pylori (79% v53%; p<0.00001). Reported bedroom sharing in childhood was more frequent in CagA seropositive men compared with H pylori infected CagA seronegative men (76%v 66%; p<0.01). The strong correlation between H pylori seropositivity andChlamydia pneumoniae IgG titres (p<0.0001) did not change much after adjustment for age, cigarette smoking, and markers of socioeconomic status (suggesting that these two infective agents may share as yet unrecognised risk factors).10Associations between H pylori seropositivity and low FEV₁ and short stature, however, weakened substantially after adjustments. The data were too sparse to subdivide reliably by town of residence and racial origin (the large majority of men were Caucasian).

Discussion

Previous studies have reported great variation in the prevalence of potentially virulent H pylori strains with much higher prevalences generally reported in developing countries than in North America and western Europe.4 ,6 ,11 ,12Most, however, have been small studies conducted in selected populations or in selected regions within certain countries. By contrast, we conducted a nationwide community based study of cagA⁺ H pylori strains, including detailed information on a number of possible risk factors and other characteristics in more than 1000 individuals. We found that about 45% of middle aged British men resident in 18 different towns in 1980 were infected with potentially more virulent strains of H pylori. Moreover, we found a greater prevalence of these organisms in the less affluent regions of Scotland and northern England than in southern England. It may be that some of the individuals seropositive for CagA antibodies in this study were also coinfected with other H pylori strains, but this would not alter the interpretation of the present findings.

As H pylori is believed to be transmitted mainly by person-to-person contact, our observation that men infected with CagA positive strains of H pylorireported bedroom sharing during childhood more frequently than men infected with other strains is consistent with suggestions that potentially virulent strains may be acquired earlier in life than other strains.13 More direct evidence is needed, however, to determine reliably both the timing and mode of transmission of various strains of the infection. We observed no strong associations between CagA seropositivity and blood markers of systemic inflammation, despite previous suggestions of increased intragastric inflammation in CagA positive individuals.3 We also observed no important differences between men infected and those not infected with CagA positive strains for a number of other characteristics listed in table1, suggesting that these characteristics cannot explain differences in the strengths of association reported between differentH pylori subtypes and gastric diseases.