Risk Stratification in Febrile Infants Using Complete Blood Cell Counts

Complete blood cell count (CBC) parameters did not accurately identify febrile infants with invasive bacterial infections (IBIs), according to the results of new research published in JAMA Pediatrics.

In this planned secondary analysis of a prospective observational study, 4313 febrile infants (temperature of ≥38°C, <60 days old; median age 38 days) from 26 emergency departments were enrolled.

The researchers analyzed blood cultures and determined the presence of an IBI with a cerebrospinal fluid culture or a telephone follow-up within 7 days of the emergency department visit. Premature infants, critically ill infants, and infants receiving antibiotics within 4 days of the visit, with congenital malformations, or with a comorbid medical diagnosis were excluded from the analysis.

In the study, a total of 97 infants (2.2%; 95% CI, 1.8% to 2.7%) were found to have IBIs (n=73; 95% CI, 1.4% to 2.1% isolated bacteremia; n=24; 95% CI, 0.4% to 0.8% bacterial meningitis). Of the 24 infants with bacterial meningitis 11 also had bacteremia.

CBC parameters were unable to distinguish between infants with and without IBIs. Sensitivities were low for CBC thresholds; white blood cell count (WBC) <5000/µL (10% sensitivity; 95%, CI 4% to16%), WBC >15,000/µL (27%; 95%, CI,18% to 36%), absolute neutrophil count (ANC) >10,000/µL (18%; 95%, CI10% to 25%), and platelets <100×10³/µL (7%; 95%, CI,  2% to 12%).

ANC had the highest predictive value, but the optimal calculated threshold of 4100 cell/µL would have missed approximately 33% of the infants with IBIs. Moreover, the researchers noted that using WBC cutoffs of <5000 or >14,900 cells/µL would have missed 63% (n=61) of IBIs, and an ANC count cutoff of 10×10³ cells/µL would have missed 82% (n=80) of IBIs.

The predictive value of CBC parameters did not improve when infants were categorized according to age (0 to 28 days and 29 to 60 days).

The study investigators concluded that “complete blood cell count parameters had poor accuracy in distinguishing febrile infants age 60 days and younger with and without invasive bacterial infections in the postpneumococcal conjugate vaccine era, although the ANC had the highest sensitivity. Physicians who use CBC thresholds in an attempt to risk stratify febrile young infants may be falsely reassured by normal CBC parameters.”

Reference

Cruz AT, Mahajan P, Bonsu BK, et al. Accuracy of complete blood cell counts to identify febrile infants 60 days or younger with invasive bacterial infections [published online September 11, 2017]. JAMA Pediatr. doi:10.1001/jamapediatrics.2017.2927