July 2000
SEATTLE - Hemolytic-uremic syndrome (HUS) developed in more than half of the children treated with antibiotics vs. less than 10% of those not treated with antibiotics for Escherichia coli O157:H7 infection, according to a recent study. Because of the known biologic plausibility, the rigorous efforts to reduce potential study bias and the strength of the association, the researchers concluded that antibiotic treatment in children with E. coli O157:H7 infection increases the risk of HUS.
"Antibiotics are certainly not helpful in preventing HUS when given to children infected with E. coli O157:H7, and they are probably associated with a worse outcome, i.e. the development of HUS," said Phillip Tarr, MD, astroenterologist at Children's Hospital and Regional Medical Center and professor of pediatrics at the University of Washington School of Medicine, Seattle.
The study, presented at the 35th United States-Japan Cholera and Related Diarrheal Diseases Conference and published in the New England Journal of Medicine, was conducted with 46 cooperating laboratories in four states under the sponsorship of the National Institutes of Health. It looked at 71 children younger than 10 years who had diarrhea caused by E. coli O157:H7 and whose stool cultures were obtained within seven days of the onset of illness.
![[bar]](/art/gradred.gif){kind=small}
--- A
photomicrograph of a kidney biopsy from a child with hemolytic-uremic
syndrome. Note the thrombus in the afferent arteriole.
PHOTO COURTESY OF PHILLIP TARR, MD
The researchers
defined the syndrome as hemolytic anemia, thrombocytopenia and renal
insufficiency. A 14-day risk period, beginning with onset of diarrhea in
children with a positive culture for E. coli O157:H7, was used for
clinical observation, with the duration of the risk period based on that seen
in previous studies.
Nine of 71 study participants (13%) received antibiotics. HUS occurred in 10 study participants, including five of the nine patients (56%) that received antibiotics.
"Perhaps there is a group where the pathophysiologic cascade toward HUS is already underway by the time of presentation. There are multiple overlapping risks for the development of HUS, some perhaps are bacteria-associated, some are host-associated and some are exogenous, i.e., medication administration," said Tarr.
According to the study, those factors significantly associated with HUS include higher white-cell count, earlier time to medical evaluation with stool culture after diarrhea onset and treatment with antibiotics. A multivariate analysis, adjusted for initial white-cell count and the day stool culture was obtained, showed antibiotic treatment remained a risk for developing the syndrome.
Of the 10 total children who developed HUS, four had oligoanuria and required dialysis. Seven required erythrocyte transfusions, platelet transfusions or both. None of the children died during hospitalization.
Frequency of HUS was related to initial white-cell count. No children with an initial count of 3,200/mm3 to 8,700/mm3 developed HUS; 6% with an initial count of 8,800/mm3 to 11,800/mm3 developed HUS; 17% with an initial count of 11,900/mm3 to 14,200/mm3 developed HUS; and 35% with an initial count of 14,300/mm3 to 24,600/mm3 developed HUS.
Incidence rates of HUS were also higher among study patients who had laboratory studies soon after the onset of illness and those treated with antibiotics.
In logistic-regression analyses, initial white-cell count was significantly associated with HUS, with relative risk proportional to white-cell count, according to the study. The time from onset of diarrhea until the day the initial stool culture was obtained was also strongly associated with a risk of developing the syndrome.
Multivariate analysis, adjusted for both initial white-cell count and the day stool culture was obtained, showed patients treated with antibiotics had a higher risk of HUS than patients who were not treated with antibiotics.
Two children with HUS were treated with trimethoprim-sulfamethoxazole (TMP-SMX) and three were treated with cephalosporins. The four study patients who were treated with antibiotics but did not develop HUS received TMP-SMX, a cephalosporin (each one patient) and amoxicillin (two patients).
The relative risk of HUS used as a function of antibiotic class, after adjustment for initial white-cell count, was 17.7 for TMP-SMX and 13.4 for b-lactam antibiotics.
"The time when most children are given antibiotics is at presentation in advance of microbiologic diagnosis. While there might be some benefit to treating a child with Campylobacter, Yersinia or Shigella on presentation, these infections can be difficult to differentiate from E. coli O157:H7 infections on initial evaluation," said Tarr. "We think now the harm of treating E. coli O157:H7 infection with antibiotics exceeds the potential benefit of treating the others with such agents. I encourage withholding antibiotics unless and until a pathogen is recovered that can be appropriately treated by such drugs, and E. coli O157:H7 is confirmed to be absent."
Analysis of an outbreak in Japan showed treatment with fosfomycin (Monurol Sachet, Forest) was associated with a decreased risk of HUS - but only for children who received fosfomycin on the second day of their illness. The study compared fosfomycin with other antibiotics, but not with treatment without antibiotics.
"The E. coli that cause HUS have a common denominator: the ability to produce Shiga-toxin 1, Shiga-toxin 2 or both," said Tarr. "E. coli O157:H7 is the best-known and most frequently isolated member of this group, although there are others that can cause HUS. It is believed this is a nonbacteremic infection. The organism does not get out of the gastrointestinal (GI) tract, but the working hypothesis is that the toxin leaves the GI tract, circulates and causes vascular and organ damage."
Japanese studies have also demonstrated the extra intestinal deposition of toxin, mainly in the kidneys and in the lungs.
For more information:
- Wong C, Jelacic S, Habeeb R, et al. The risk of hemolytic-uremic syndrome after antibiotic treatment of Escherichia coli O157:H7 infections. N Engl J Med Final version appears in the June 29 issue.
- Uchida H, Kiyokawa N, Taguchi T, et al. Shiga toxins induce apoptosis in pulmonary epithelium-derived cells. J Infect Dis. 1999;180(6):1902-1911.
- Uchida H, Kiyokawa N, Horie H, et al. The detection of Shiga toxins in the kidney of a patient with hemolytic uremic syndrome. Pediatr Res. 1999;45(1):133-137.
You can
express your views on this article, or other relevant themes, in the
Infectious Diseases in Children
Specialty Forums.