CHICAGO — A child's death from unsuspected meningococcal disease can keenly heighten an emergency physician's awareness that there are few clues about which children with fever and petechiae are safe to send home, Dr. Jane Knapp said at a meeting sponsored by the American College of Emergency Physicians.
“You can't pick them out,” cautioned Dr. Knapp, professor of pediatrics at the University of Missouri-Kansas City and a pediatric emergency physician at Children's Mercy Hospital in Kansas City, Mo. “But we can't admit every child with fever and petechiae.”
Neither clinical nor hematologic features are reliable predictors of meningococcal infection, she added.
Dr. Knapp presented a case from early in her career of a 7-year-old boy who was afebrile on admission. Mental status changes followed by the development of a petechial rash in the emergency department prompted treatment for meningococcemia, but he died shortly afterward.
The case highlights the fact that lack of fever is not always a reassuring sign and does not exclude meningococcal infection, she said.
One study of 24 children with meningococcal disease found that 5 had axillary temperatures of less than 37.5° C (Arch. Dis. Child. 2001;85:218).
Another study of 381 febrile children with meningococcal infection found that 10% did not have a petechial/purpuric rash, although they did appear unwell (Pediatrics 1999;103:E20). An additional 45 (12%) of the children had what the authors called “unsuspected meningococcal disease” (UMD), meaning they were seen in the hospital and discharged with a later positive culture. Of those 45 children, 24 were recalled when their blood culture results came in positive, 14 returned because they had worsened or developed a rash, 5 returned for a scheduled follow-up, and 2 returned because of persistent fever. Two children in the UMD group died after returning to the hospital—one 6 hours and the other 12 hours later.
Comparing the children with UMD to a control group of culture-negative febrile patients, the authors found that the UMD group was on average significantly younger (9 months vs. 14 months), with 82% of them aged between 3 and 36 months. The UMD group also had significantly higher band counts on average (14 vs. 7), compared with the culture-negative patients. However, the authors concluded that the predictive value of the band count is low in this group, because UMD is uncommon in young febrile pediatric patients.
“That study suggests that neither the clinical examination nor the CBC [complete blood count] reliably distinguishes young children with UMD from those with viral illnesses,” Dr. Knapp said.
Because this is an area of legal risk for physicians, she said, they are left with a perplexing challenge. One guideline to follow for managing children with a nonblanching rash is whether they appear unwell. If they appear unwell, Dr. Knapp suggested simply admitting them and treating for invasive meningococcal infection. “Could you compromise and send someone home with antibiotics?” she asked. “I would be pretty liberal.”