microbe of the month: Citrobacter freundii

A vegetable grown in `organic' conditions has resulted in death and critical illness in a German kindergarten. Bernard Dixon explains why

Common sense tells us that organic foods, because they are prepared "naturally", are safer than those made using synthetic chemicals, whether as fertilisers, preservatives or insecticides. But common sense is not always reliable, as a report this month from the Robert Koch Institute in Wernigerode, Germany, makes horrifyingly clear.

The report describes an outbreak of severe gastro-enteritis and kidney failure which affected nine children, one of whom died, in a nursery school and adjacent kindergarten in a small town in north-west Germany. It was caused by the bacterium Citrobacter freundii, which was traced to parsley grown in an organic garden, contaminated when pig slurry was used in place of artificial fertiliser.

The incident occurred a year ago, when the early summer temperature seems to have facilitated the multiplication of C. freundii - a close relative of the better-known food poisoner salmonella. Such microbes are always more likely to cause trouble when the warmth of summer allows them to proliferate very rapidly. In this case, however, the initial trigger was the contamination of the supposedly pristine organic garden, which public health officials did not begin to suspect until they started to investigate what was happening at the school and kindergarten.

The outbreak first came to notice when 14 of the children, aged between one and six, developed severe gastro-enteritis. Then three of them - two boys aged two and a three-year-old girl - became seriously unwell with clear signs of acute kidney failure, anaemia and other ill-effects. Together, these conditions comprise the life-threatening "haemolytic uraemic syndrome" (HUS), which was clearly recognised for the first time only in the early Eighties.

All of the victims had to be admitted urgently to hospital, where they not only received powerful antibiotics but also had to be treated on a kidney machine. Over the next five days, six more children, all aged three, were hospitalised for the same reasons. One of the patients died and the others recovered only after several weeks on the kidney machine.

Dr Heinz Tschape and his fellow investigators found that faeces from the eight survivors all contained a particular strain of C. freundii. It was one that produced a poison known as verotoxin, which has previously been associated with gastro-enteritis and HUS caused by other bacteria. This was the first clue to the original source of the organism, since most outbreaks of both gastro-enteritis and HUS attributable to other verotoxin-producing bacteria have been associated with animal husbandry. Usually, milk or meat has proved to be contaminated - the chances of this happening being greatly heightened by mediocre hygiene. In this case, tests on the stools of the children, as well as those of staff and members of the children's families, revealed a further 28 individuals who were infected with C. freundii, though without becoming ill. Using questionnaires, the investigators next set out to pinpoint any particular food(s) which those infected had consumed over the past few days.

It soon became clear that a batch of green butter sandwiches, prepared in the nursery school kitchen, was the most probable source of infection. The sandwiches had been eaten four days before the outbreak began. Since bread is very rarely a carrier of disease-causing bacteria, suspicions focused on the green butter, which contained parsley.

Unfortunately, none of the butter was left to be tested in the laboratory, so Dr Tschape and his colleagues examined some fresh parsley leaves from the organic garden where the original parsley had been grown. These proved to be highly contaminated, not only with various strains of C. freundii but also with many other species of bacteria that had in all likelihood originated in human or other animal intestines.

As reported in this month's Epidemiology and Infection, the use of probes to identify the microbial DNA, together with other sophisticated tests, soon established that one of the C. freundii strains on the parsley was indistinguishable from that isolated from the patients. The conclusion seemed inescapable. Yet kitchen staff said that they had washed the parsley vigorously before making the sandwiches. It was at this point that investigators realised that the remaining bacteria had multiplied subsequently in the early summer warmth, thereby reaching a concentration sufficient to cause illness.

The combination of exceptionally high levels of C. freundii and Escherichia coli and other intestinal bacteria convinced Dr Tschape and his collaborators that the organic garden had been contaminated with faecal material. When they discovered that it had indeed been treated not with synthetic fertiliser but with pig manure, their chain of reasoning was complete. So much for the excellence of foods prepared as nature intended.