The 130th Annual Meeting of APHA

5164.0: Wednesday, November 13, 2002 - 2:45 PM

Abstract #42050

Utility of the serum c-reactive protein (CRP) for detection of occult bacterial infection in children

Daniel J. Isaacman, MD, Division of Pediatric Emergency Medicine, Children's Hospital of The King's Daughters, 601 Children's Lane, Norfolk, VA 23507 and Bonnie L. Burke, MS, Clinical Outcomes, Research and Epidemiology, Children's Hospital of The King's Daughters, 601 Children's Lane, Norfolk, VA 23507, (757) 668-8781, Disaacma@chkd.com.

Background: Prior studies suggest serum C-reactive protein (CRP) is a useful screening test for neonatal bacteremia. This study assessed the utility of CRP as a screening test for occult bacterial infection (OBI) in comparison to current screens. Methods: Febrile children ages 3-36 months receiving a complete blood count and culture were enrolled between 2/2000 and 6/2001. OBI was defined as occult pneumonia, UTI or bacteremia. Logistic regression and ROC curves were modeled for each screen to identify optimal test values. Results: Analysis of 256 children found 50 (19.5%) patients with identified focal infection (otitis (24), UTI (9), pneumonia (17)) and 206 without. OBI was identified in 29 (11.3%): three bacteremia, 17 pneumonia and 9 UTI. Median WBC count was 12.9 [range: 3.6,39.1] cells/mm3, median ANC level was 7.12 [range: 0.56,28.16] cells/mm3 and median CRP level was 1.7 [range: 0.2,43.3] mg/dl. The optimal CRP cut-point (4.4mg/dl) achieved a sensitivity of 63% and a specificity of 81%. Among individual laboratory predictors (ANC, WBC or CRP), ANC with a cut-point of 10.6 cells/mm3 best detected OBI, yielding a sensitivity of 69% and a specificity of 80%. Adding CRP to ANC decreased sensitivity (59%), increased specificity (92%) while roughly doubling screening costs. Active monitoring in a similar population showed an overall bacteremia rate of 1.1%. Conclusion: An ANC cut-point of 10.6 cells/mm3 is best for detection of occult bacterial infection using a single test. Adding CRP to ANC adds little diagnostic utility. Lowered incidence of occult bacteremia supports less use of diagnostic screening.

Learning Objectives:

Keywords: Emergency Department/Room, Screening

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

Outbreaks, Surveillance and Patterns

The 130th Annual Meeting of APHA