D. M. Weir

in 47 cases by the finding of L.E. cells (with necropsy confirmation in seven), in one by typical changes at necropsy, and in one by skin biopsy.

ABO blood group and secretor state was determined in 319 women with recurrent urinary tract infection and compared with those of a control group of 334 women of similar age ranges. Women of blood groups B and AB who are non-secretors of blood group substances showed a significant relative risk of recurrent urinary tract infection of 3.12 (95% confidence limits, 1.49 and 6.52) in comparison with other types. This appears to be a genuine example of synergy in which absence of anti-B isohaemagglutinin and secretor substances combines to give an increased risk of recurrent urinary tract infection. Determination of blood group and secretor state may provide additional information in identifying those at risk.

Smoking is associated with an increased risk of respiratory tract infection in adults. In children, exposure to cigarette smoke is a risk factor for respiratory tract infection and bacterial meningitis: Active smoking and passive exposure to cigarette smoke is also associated with carriage of some potentially pathogenic species of bacteria in both adults and children. The aims of the study were to determine the effect of active smoking on: (1) bacterial binding to epithelial cells; (2) expression of host cell antigens that act as receptors for some species; and (3) the effects of passive exposure to water-soluble components of cigarette smoke on bacterial binding. Flow cytometry was used to assess binding to buccal epithelial cells of the following species labelled with fluorescein isothiocyanate: Neisseria meningitidis, Neisseria lactamica, Streptococcus pneumoniae, Bordetella pertussis, Haemophilus influenzae, Moraxella catarrhalis, Staphylococcus aureus. Flow cytometry was also used to assess expression of host cell antigens which have been identified as bacterial receptors. For each species, binding to cells of smokers was significantly higher than to cells of non-smokers; however, expression of host cell antigens was similar on epithelial cells of both groups. Non-dilute cigarette smoke extract reduced binding of bacteria to epithelial cells, but dilutions between 1 in 10 and 1 in 320 enhanced binding. We conclude that smokers might be more densely colonised by a variety of potentially pathogenic bacteria. The enhanced bacterial binding to epithelial cells of smokers is not related to enhanced expression of host cell antigens that can act as receptors for some species, but possibly to components in the smoke that alter charge or other properties of the epithelial cell surface. Passive coating of mucosal surfaces with components of cigarette smoke might enhance binding of potentially pathogenic bacteria.