Protection from natural infections with enterotoxigenic Escherichia coli: longitudinal study

Abstract

Enterotoxigenic Escherichia coli (ETEC) are an important cause of diarrhoea and diarrhoeal deaths in children living in developing countries and of travellers' diarrhoea. During the past 25 years, vaccine development efforts have been focused on induction of protective immunity against surface colonisation factors (CFs) and the heat-labile enterotoxin. Although vaccines that induce immunity to heat-labile toxin offer protection against diarrhoea from ETEC that produce this toxin, the benefit of including CF antigens remains uncertain. We aimed to estimate the protection that natural ETEC infections induce against new infections. In Guinea-Bissau, we followed up 200 neonates until up to age 2 years, most of whom were breastfed throughout the study. We collected stool specimens from the children every week irrespective of whether they had diarrhoea. As a measure of protection, we used Cox regression models to estimate the change in infection rates after a primary ETEC infection. We thus estimated the protection attributable to CFs, toxins, and to any other factors that could be shared by ETEC with the same toxin-CF profile. ETEC infections induced a 47% (95% CI 12 to 69) protection against new infections with ETEC that had the same toxin-CF profile; the corresponding estimate attributable to CFs was -1% (-40 to 27). Infection with heat-labile toxin-positive ETEC conferred a 45% (-1 to 70) protection against symptomatic infections with ETEC positive for this toxin. For breastfed children living in endemic areas, other antigens are substantially more important than CFs for induction of protective immunity against ETEC infection.