The procalcitonin assay is superior to C-reactive protein, neutrophil, and white blood cell measurements at identifying invasive bacterial infection in very young febrile infants, according to a study published Nov. 23 in JAMA Pediatrics.
Procalcitonin assays allow earlier detection of certain infections compared with other biomarker assays in older children, and a few small studies have hinted at their usefulness in infants, but to date no large prospective studies have assessed procalcitonin assays in the youngest infants. So investigators evaluated the diagnostic accuracy of procalcitonin and other biomarkers in a prospective study involving 2,047 febrile infants aged 7-91 days who presented to 15 pediatric emergency departments in France during a 30-month period.
“We did not include infants 6 days or younger because they are likely to have early-onset sepsis related to perinatal factors and because physiologic procalcitonin concentrations during the first [few] days of life are higher than thereafter,” said Dr. Karen Milcent of Hôpital Antoine Béclère, Clamart (France), and her associates.
Serum samples were collected from the infants at the initial clinical examination, but procalcitonin assays were not performed at that time. Attending physicians diagnosed the infants as having either bacterial or nonbacterial infections without knowing the procalcitonin results. Then procalcitonin tests were done retrospectively on frozen serum samples by lab personnel who were blinded to the infants’ clinical features. Thirteen (1.0%) infants had bacteremia and 8 (0.6%) had bacterial meningitis.
The procalcitonin assay was significantly more accurate at identifying invasive bacterial infections than was C-reactive protein level, absolute neutrophil count, or white blood cell count. At a threshold of 0.3 ng/mL or more, procalcitonin level had a sensitivity of 90%, a specificity of 78%, and a negative predictive value of 0.1. In addition, the procalcitonin assay was the most accurate in a subgroup analysis restricted to patients whose fever duration was less than 6 hours and another subgroup analysis restricted to patients younger than 1 month of age, the investigators said (JAMA Ped. 2015 Nov 23. doi:10.1001/jamapediatrics.2015.3210).
“Our results suggest that it may be possible to improve clinical practice for the treatment of young febrile infants” by combining procalcitonin assay results with a careful case history, a thorough physical examination, and other appropriate testing. Among other benefits, this approach offers the potential of avoiding lumbar punctures, they added.
These study findings “should encourage the development of decision-making rules that incorporate procalcitonin,” Dr. Milcent and her associates said.