Intended for healthcare professionals

Editorials

Haemolytic uraemic syndrome and E coli 0157

BMJ 1999; 318 doi: https://doi.org/10.1136/bmj.318.7185.684 (Published 13 March 1999) Cite this as: BMJ 1999;318:684

Prevention rests with sound public health measures

  1. Margaret Fitzpatrick, Consultant paediatric nephrologist
  1. St James's University Hospital, Leeds LS9 7TF

    Deaths from renal failure after outbreaks of acute gastroenteritis in Scotland in 1996, which were traced back to poor hygiene in butchers‘ shops, focused attention on Escherichia coli 0157 and its links with the haemolytic uraemic syndrome. Although in the Scottish outbreak deaths occurred mainly in elderly people, haemolytic uraemic syndrome associated with diarrhoea generally affects children and is the commonest cause of acute renal failure in children in Europe and North America.

    Haemolytic uraemic syndrome is the most important complication of infection by verocytotoxin producing E coli—usually of the serotype 0157:H7, though other serotypes are also implicated. 1 2 It is characterised by the sudden onset of haemolytic anaemia with fragmentation of red blood cells, thrombocytopenia, and acute renal failure after a prodromal illness of acute gastroenteritis often with bloody diarrhoea. The gastrointestinal disease may be severe, with haemorrhagic colitis, and the central nervous system, pancreas, lungs and heart may also be affected. Fortunately more than 95% of people affected with this condition do recover from the acute illness, and recurrence is unusual. The longer term prognosis remains unclear; follow up studies have shown 15–40% of survivors with renal sequelae. 3 4

    The annual incidence of infection with verotoxin E coli ranges from 1–30 cases per 100 000 in industrialised countries and is highest in young children under 5 years and during the warmer months. During outbreaks of infection about 8% of patients develop haemolytic uraemic syndrome.5 The infective dose of E coli 0157:H7 is 50–100 organisms and the incubation period to onset of diarrhoea 1–8 days. Younger children may continue to excrete the bacteria for more than three weeks after the infection, but asymptomatic prolonged carriage of E coli 0157 is unusual.

    The natural reservoir of verotoxin producing E coli is the intestinal tracts of domestic animals, especially cattle, and bovine foods such as ground beefburgers and unpasteurised milk are major sources for human infection. Other foods, including cider and vegetables such as lettuce and radish sprouts, are becoming increasingly recognised as sources. E coli 0157:H7 has been recovered from many retail foods including fresh seafood, lamb, chicken, pork, venison, and veal. Person to person transmission of E coli 0157, facilitated by the low infectious dose, is a source of human infection.

    The initial cases of an outbreak and many sporadic cases of E coli 0157:H7 disease may go unrecognised because routine stool cultures do not differentiate these E coli from normal enteric flora. Not all laboratories in the United Kingdom screen for the pathogen. Also, although the disease eventually manifests as bloody diarrhoea, it usually begins with non-bloody diarrhoea and this may be difficult to differentiate from the acute viral gastroenteritis commonly seen in children.

    Children at particular risk of developing haemolytic uraemic syndrome after infection with verotoxin producing E coli are younger, have a high presenting white blood cell count, and have bloody rather than non-bloody diarrhoea.5 The use of antimotility or antidiarrhoeal agents is an additional risk factor, and the toxin type of the infecting strain may also be important.6 The prevention of secondary and epidemic spread would be greatly facilitated by looking for E coli 0157 in faecal samples. Clearly not all children with diarrhoea should have stool cultures, but when stools are sent to the laboratory from children, particularly those with risk factors, they should be tested for E coli 0157:H7.

    Once haemolytic uraemic syndrome caused by verotoxin producing E coli has developed there is no specific treatment yet available. Prompt diagnosis and good supportive care to maintain hydration, electrolyte balance, and nutrition are crucial, as is the management of the critical complications of cerebral oedema, severe ischaemic colitis, myocardial dysfunction, and diabetes mellitus. Studies are in progress to identify novel treatments that may halt the development of haemolytic uraemic syndrome in E coli infection, such as the use of a synthetic oligosaccharide receptor for the verotoxin.7 Understanding of the pathogenetic mechanisms and of the pivotal role of verocytotoxin at a cellular level is increasing,8 and this will clearly be valuable in developing treatments. Of greatest benefit is preventing primary and secondary infection with verotoxin producing E coli, and this should be the focus of public health policy.

    Preventing E coli 0157:H7 disease is difficult as the organisms colonise the intestines of healthy cattle and other food animals. They are also resistant to acidic conditions, dehydration, and high salt concentrations. Public health control thus requires vigilance at all stages of the food chain. Safe food handling and proper cooking will destroy the organism, and the risk of foodborne disease can be reduced by limiting possible cross contamination. Irradiation dramatically reduces the risk of E coli 0157:H7 infection by killing these organisms, and this approach has been recommended as safe and effective by consumer groups and an independent scientific panel.9 In the United States the Food and Drug Administration approved irradiation of red meat in December 1997. In future the use of a toxoid vaccine,10 which has potential given the established central role of verocytotoxin in the pathogenesis of haemolytic uraemic syndrome, may help to eradicate the disease altogether. For now, sound public health measures remain paramount.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    View Abstract