Both the CRP and WBC values were generally substantially higher in confirmed bacterial than in viral infections. The trouble was that there were important exceptions to that broad trend that seriously compromised the lab tests' predictive value.
For example, the WBC in confirmed infections with Staphylococcus aureus—one of the two most common pathogens causing bacteremia—was only one-third to one-fourth the level seen in infections caused by Streptococcus pneumoniae or Escherichia coli. In fact, the WBC in S. aureus infections was similar to levels associated with most of the key viral infections. The viral outlier was adenovirus, which produced WBCs and CRPs in the bacterial infection range (J. Pediatr. 2006;149:721-4).
“This is a very well-done study—and these are results that can be used in real life,” Dr. Radetsky commented.
He added that the best role for CRP in his view is in following the progress of antibiotic treatment and deciding when to discontinue it.
“In my part of the world now, virtually all of the go home on oral antibiotics types of diseases [are] treated exactly the same way: You treat until the patient is better and the CRP is normal,” the pediatrician continued.
▸ Standard- vs. double-dose amoxicillin for nonsevere pneumonia in young children. Investigators at the Pakistan Institute of Medical Sciences, Islamabad, randomized 876 children aged 2–59 months who had nonsevere pneumonia to 3 days of amoxicillin at either the standard dose of 45 mg/kg per day or to 90 mg/kg per day. Nonsevere pneumonia was defined using World Health Organization criteria: an elevated respiratory rate with no retractions.
On day 5, clinical resolution of the respiratory illness was observed in 98.4% of patients, with no difference between the two study arms. On follow-up on day 14, there was a 2% relapse rate, again with no difference between the two treatment groups (Arch. Dis. Child. 2007;92:291-7).
The Pakistani investigators treated for just 3 days with antibiotics on the basis of two earlier randomized trials conducted in more than 4,100 young children in India and Pakistan, which demonstrated 3 days of treatment was as good as 5.
▸ The questionable value of chest x-rays in acute bronchiolitis. Dr. Suzanne S. Schuh and colleagues at the Hospital for Sick Children, Toronto, prospectively studied 265 children aged 2–23 months who had a clinical diagnosis of acute bronchiolitis. All were given a chest x-ray. Comparison of the pre- and postradiography patient management plans revealed that the imaging study led to an alteration in treatment plan in a mere 2.6% of cases (J. Pediatr. 2007;150:429-33).
“Chest x-rays may not be necessary at all, because they rarely give rise to information [that] would require a change in your management,” Dr. Radetsky observed.
▸ Negative blood cultures: How reliable are they? They are not very reliable, according to pediatricians at Royal Children's Hospital, Melbourne. They conducted a prospective study of all 1,358 blood culture bottles obtained during a 6-month period. The culture-positive rate was 5.2% in bottles deemed of adequate volume, significantly better than the 2.2% positive rate in bottles of insufficient volume.
The Australians defined adequate blood sample volume as at least 4 mL in children aged 36 months and up, at least 0.5 mL in those 1 month of age or younger, and at least 1 mL in all those in-between. Fifty-four percent of bottles were judged to be of inadequate volume, a rate that declined to 36% following a staff education campaign (Pediatrics 2007;119:891-6).
▸ Accuracy of rapid influenza tests. A multicenter study involving 270 children under age 5 years hospitalized for acute respiratory symptoms or fever who underwent influenza culture and/or reverse transcriptase-polymerase chain reaction testing as well as one of four rapid influenza tests concluded that the rapid tests were moderately accurate, with an overall sensitivity of 65% and a specificity of 97%.
But the investigators from the New Vaccine Surveillance Network and the Centers for Disease Control and Prevention noted that the accuracy of the rapid tests varied considerably week by week over the course of flu season, as a function of the local background influenza prevalence at the time of testing.
They concluded that when the local prevalence was less than 10%, the false-positive test rate was so high that rapid tests were of little value (Pediatrics 2007;119:e6-11).
“Many people may argue that in areas of high prevalence, the rapid tests are not necessary, and in areas of low prevalence, they are not accurate,” according to Dr. Radetsky.
▸ Scalp ringworm disease patterns in preschoolers. Susan M. Abdel-Rahman, Pharm.D., and colleagues at Children's Mercy Hospital, Kansas City, Mo., prospectively followed 446 children at a single urban preschool for 2 years. On a monthly basis, they performed scalp exams and gathered scalp samples for genotyping of Trichophyton tonsurans isolates. Their goal was to learn the natural course of preschool infection with T. tonsurans, by far the No. 1 cause of tinea capitis in the United States.