Approaches to evaluating the febrile child have changed over time

VANCOUVER, British Columbia — The way physicians evaluate febrile children has evolved, especially with the advent of conjugate vaccines, Eugene D. Shapiro, MD, an Infectious Diseases in Children Editorial Board member said here during a presentation at the 2010 Pediatric Academic Societies Annual Meeting.

In the past, children aged younger than three months with fever underwent extensive laboratory testing, hospitalization and prophylactic antibiotic treatment, Shapiro said, as physicians feared that clinical and lab features may be insufficient for diagnosing severe illnesses.

Criteria based on clinical presentation and test results were eventually developed in the 1980s and 1990s to characterize children at low risk for serious bacterial infection. These guidelines were designed to preclude unnecessary presumptive antimicrobial treatment, Shapiro noted. However, they were not 100% effective.

The Pediatric Research in Office Settings (PROS) Study from 1995 indicated that the use of clinical judgment as opposed to following set algorithms might also be beneficial, according to Shapiro. Practitioners in the study identified and provided antibiotic treatment to 97% of children with bacteremia using clinical judgment, he explained, without following guidelines or algorithms. The use of these tools would not have improved care and may have resulted in more testing and hospitalizations, Shapiro said.

He pointed out, however, that lab testing is still important, especially for febrile infants aged younger than 1 to 2 months.

Tests used to identify febrile children aged 3 to 36 months at risk for occult bacteremia can be problematic. The evidence supporting these methods is flawed, Shapiro noted, because studies paid inadequate attention to the predictive value of diagnostic tests. Research also showed the predictive positive value of elevated white blood cell (WBC) counts was poor.

Unofficial guidelines developed in the 1990s, however, recommended that physicians presumptively treat a febrile child with an elevated WBC count with an antimicrobial drug like ceftriaxone. Clinical trials evaluating this process yielded positive results, but the study design, specifically the definitions of the outcomes, was biased in favor of ceftriaxone. Statistical analyses were also inappropriate, according to Shapiro.

“Sadly enough, these data drove clinical practices for many years, and even up to today, this is what’s cited as evidence that presumptive treatment, like ceftriaxone, is justified,” Shapiro said.

He noted that there are risks to presumptive treatments, including cost, adverse side effects, selective pressure on bacteria and follow-up complications.

Shapiro also addressed the effect that conjugate vaccines have had on how pediatricians approach diagnosing febrile children. The frequency of occult bacteremia related to Haemophilus influenzae type B and Streptococcus pneumoniae declined dramatically since the availability of the Hib conjugate vaccine and the seven-valent pneumococcal conjugate vaccine (PCV7, Prevnar, Wyeth). This decrease in bacteremia has led to a further drop in positive predictive value for diagnostic tests for febrile children.

Shapiro also stressed the fact that urinary tract infections are much more common than meninigitis, and collecting uncontaminated urine specimens is difficult. Furthermore, testing urine from bag specimens is unreliable because of frequent contamination.

“It is unlikely, in my opinion, that the benefits of diagnostic testing and presumptive treatment are worth the risks/costs for febrile children aged 3 to 36 months,” said Shapiro. “My recommendation is for careful clinical assessment of the patient over time." – by Melissa Foster

For more information:

Shapiro ED. #3075. Presented at: 2010 Pediatric Academic Societies Annual Meeting; May 1-4, 2010; Vancouver, British Columbia.

Follow the PediatricSuperSite.com on Twitter.