SEATTLE--Children with gastrointestinal (GI) infections caused by Escherichia coli 0157:H7 should not receive antibiotic treatment because it markedly raises their risk of hemolytic-uremic syndrome (HUS). That warning is from Seattle researchers who recently completed a study of 71 children treated for E coli 0157:H7 GI infections between April 1997 and August 1999.[1]

"The risk of antibiotic-related HUS in children with bloody diarrhea, especially those infected with E coli 0157:H7, has been controversial," said Phillip I. Tarr, MD, the senior author, in an interview with PULMONARY REVIEWS. "However, in our study, antibiotic administration remained a strong and independent risk factor for the development of HUS even when we controlled for objective indices of severity."

Thus, the relationship between antibiotic use and HUS in children with E coli 0157:H7 cannot be dismissed simply as a function of disease severity. In other words, the relationship appears not to arise because antibiotics were given only to the most severely ill children, who are most likely to develop HUS anyway, as some researchers have argued.

In North America, HUS develops in about 15% of children with E coli 0157:H7 GI infections. It typically arises soon after the onset of diarrhea and is marked by thrombocytopenia, hemolytic anemia, and nephropathy. Antibiotics may increase the HUS risk in these patients by triggering the release of E coli Shiga toxins.

STUDY DESIGN

The study subjects, all children age 10 years or younger, were prospectively identified through a network of 47 laboratories in Washington, Oregon, Idaho, and Wyoming. To facilitate rapid enrollment, the laboratories reported the children's E coli 0157:H7 infections to the authors immediately upon discovery.

To be diagnosed with HUS, patients had to have hemolytic anemia characterized by a hematocrit below 30% and erythrocyte destruction on a peripheral-blood smear. They also had to have thrombocytopenia (a platelet count below 150,000/mm3) and renal insufficiency, defined as a serum creatinine level above the upper limit of the normal range for age. Clinical observation took place during the 14 days after diarrhea onset to coincide with the HUS risk period. However, the analyses only took into account medications that the patients received on or before the seventh day of illness; those administered after HUS diagnosis were excluded.

THE ANTIBIOTIC-HUS LINK

HUS developed in 10 (14%) of the children--five of the nine (56%) who received antibiotics versus only five of the 62 (8%) not given antibiotics. The two sets of children with HUS had similar clinical and laboratory characteristics, noted Dr. Tarr, a gastroenterologist at Children's Hospital and Regional Medical Center in Seattle.

The children who developed HUS following antibiotic administration had been given either trimethoprim-sulfamethoxazole or a ß-lactam. Both types of antibiotics appeared to markedly increase the relative risk of HUS in this study, Dr. Tarr told PULMONARY REVIEWS.

As expected, the initial white-cell count and the day on which the initial stool culture was ordered--both considered measurements of disease severity in this study--correlated with the risk of HUS. For example, the incidence of HUS was highest in those who had an initial white-cell count above 14,300/mm3 and in those from whom a stool culture was obtained within two days of illness onset. However, the risk associated with antibiotic use remained even after the authors controlled for these factors.

In the multivariate analysis, antibiotic treatment during the first seven days after disease onset was found to have a relative risk for HUS of 17.3, said Dr. Tarr, who is also a Professor of Pediatrics at the University of Washington School of Medicine in Seattle. The relative risk rose to 32.3 when antibiotic therapy fell within the first three days of illness onset.

Four children who developed HUS became oligoanuric and required renal dialysis. Seven needed erythrocyte and/or platelet transfusions. None died during hospitalization.

In light of their findings, the authors recommend against antibiotic therapy in children with suspected E coli 0157:H7 infections unless (and until) stool cultures demonstrate a pathogen that can be appropriately treated by an antibiotic. The same strategy may also be prudent in adult patients, suggested Dr. Tarr, even though no studies have assessed the HUS risk from antibiotic therapy in adults with E coli 0157:H7 infections. "My concern is that the same risks apply in these cases," Dr. Tarr concluded.

--Timothy Begany

Reference
1. Wong CS, Jelacic S, Habeeb RL, et al. The risk of the hemolytic-uremic syndrome after antibiotic treatment of Escherichia coli 0157:H7 infections. N Engl J Med. 2000;342:1930-1936.

Return to table of contents