Escherichia coli gastro-enteritis.

Mild gastro-intestinal upset is common in babies, but severe disease is not now the scourge it formerly was, and part of the great improvement in infantile mortality is due to its decline in the twentieth century. The reasons for this are not clear, and many theories have been advanced. Formerly, a considerable number of cases of gastro-enteritis were thought to be due to dietetic, that is to say chemical upsets, but it is now believed that infection is the responsible factor for almost all cases. It has been known for a long time that dysentery bacilli and salmonellae cause only a small proportion of the cases, and many other bacteria have been suspected from time to time. One theory dating back to the early years of this century is that the disease is caused by the invasion of the small gut by the flora of the large gut (Moro, 1916). Adam (1922, 1923, 1927) advanced the view that there were certain varieties of Escherichia coli associated with the disease and based his classification upon biochemical tests. Goldschmidt (1933) showed that the strains of E. coli described by Adam fell into a few serological groups. The significance of this earlier work was not at first fully appreciated when certain 0 groups of E. coli, classified according to the Kauffmann scheme, were shown to have an association with gastro-enteritis in babies. The first of these, E. coli 0111, was described by Bray in 1945, and soon afterwards 055, 026 and 01 19 were described by Giles and Sangster (1948), by Smith (1949, 1953), and by Taylor and Charter (1952). The evidence incriminating these 0 groups was until recently epidemiological, but Thomson (1955a) showed that the whole length of the gut is colonized in the disease. The syndrome of gastro-enteritis has thus striking similarities to the syndrome of cholera: the whole length of the gut and often the gall bladder swarm with the pathogen; there is no acute inflammation of the gut in the pathological sense; there is vomiting as well as diarrhoea, and collapse occurs because of fluid loss and electrolyte imbalance. The old name 'cholera infantum would appear to be a very fitting one. With few exceptions, most of the literature on gastro-enteritis associated with E. coli deals with hospital outbreaks, many of w.iich have been very troublesome because infection, once introduced, may linger for a long time (Rogers, 1951; Rogers and Koegler, 1951: Wright and Roden. 1953; Jameson, Mann and Rothfield, 1954). In most areas where a study has been made of the disease the majority of cases yielding the suspected varieties of E. coli have arisen from cross-infection in hospitals. For example, in the Cardiff area in 1953, the year before the surveys described in this publication, 80O of the recognized cases had such an association as the following table shows.