Laurie Barclay, MD
Medscape Medical News 2009. © 2009 Medscape
Procalcitonin (PCT) and C-reactive protein (CRP), along with currently used laboratory tests, may facilitate the treatment of infants younger than 3 months seen in emergency departments for fever of unknown origin, according to the results of a retrospective study reported in the January 21 Online First issue of Archives of Disease in Childhood.
“Although the majority of children present [with] minor infections, it is important to identify those with serious bacterial infection (SBI) in order to start antibiotic treatment early,” write Izaskun Olaciregui, from Donostia Hospital in San Sebastian, Spain, and colleagues. “The Rochester scale includes a good general appearance in a previously healthy child, the absence of focal infection, and certain laboratory values (leucocyte count of 5000 – 15 000, ≤ 1500 band neutrophils, urinalysis with ≤ 10 leucocytes per high power field, and ≤ 5000 leucocytes per high power field in faeces in patients with diarrhoea) as indicators of good prognosis; the negative predictive value of this scale was 98.9% for SBI and 99.5% for bacteraemia….The objective of this study was to determine the ability of CRP and PCT to predict SBI in febrile infants under 3 months of age and to compare them with the Rochester criteria.”
The study sample consisted of all infants younger than 3 months seen in the emergency department between January 2004 and December 2006 for a febrile syndrome with no evident cause. Clinical features, PCT, CRP, and leukocyte count were compared for their ability to differentiate SBI from non-SBI. The laboratory markers were evaluated with receiver operating characteristic curves and analyzed with multivariate logistic regression.
Among 347 patients studied, 23.63% had SBI. Patients with SBI had significantly higher mean PCT, CRP, and leukocyte and neutrophil counts vs patients with non-SBI. The other criteria evaluated did not differentiate SBI from non-SBI.
PCT and CRP were each better predictors of SBI vs leukocyte count, with an area under the receiver operating characteristic curve of 0.77 for PCT (95% confidence interval [CI], 0.72 – 0.81), 0.79 for CRP (95% CI, 0.75 – 0.84), and 0.67 for leukocyte count (95% CI, 0.63 – 0.73).
Among the 15 infants who had sepsis, bacteremia, bacterial meningitis, or other more invasive bacterial infections, the diagnostic value of PCT was higher than CRP (area under the receiver operating characteristic curve, 0.84; 95% CI, 0.79 – 0.88 vs 0.68; 95% CI, 0.63 – 0.73, respectively). Among infants in whom the duration of fever was less than 12 hours, the differences between PCT, CRP, and leukocyte count were statistically significant in both SBI and non-SBI groups, with increasing predictive value of PCT and decreasing value of CRP.
Limitations of this study include retrospective design, data from history and physical examination obtained from the clinical records, cultures not performed in a small percentage of the infants, and measurement of PCT with use of a semiquantitative method.
“PCT, CRP, and leucocyte count have intrinsic predictive value for SBI in febrile infants under 3 months of age,” the study authors write. “The diagnostic value of PCT is greater than CRP for more invasive bacterial infections and for fever of short duration.”
The study authors have disclosed no relevant financial relationships.
Arch Dis Child. Published online January 21, 2009.