TY - JOUR
T1 - C-reactive protein diagnostic value for bacterial infections in the paediatric emergency department setting
AU - Adam, Iben Hamad
AU - Kestenbom, Inbal
AU - Shmueli, Moshe
AU - Hassan, Lior
AU - Lendner, Idan
AU - Ben-Shimol, Shalom
N1 - Publisher Copyright:
© 2024 Paediatrics and Child Health Division (The Royal Australasian College of Physicians).
PY - 2024/1/1
Y1 - 2024/1/1
N2 - Aim: C-reactive protein (CRP) is commonly used to assess the probability of bacterial infection and the need for empiric antibiotic treatment. We assessed the relationship between CRP levels and bacterial infection in the paediatric emergency room (PER) setting. Methods: A retrospective cohort study. Relative risks (RRs) were calculated for the relation between CRP levels and various demographic, clinical and diagnosis parameters. Additionally, the sensitivity and specificity of different CRP levels (2, 5, 7 and 10 mg/dL) for bacterial infection diagnosis was calculated. Results: Overall, 13 092 cases were recorded. Fever, leucocytosis, hospitalisation and bacterial infection diagnosis were associated with elevated CRP levels, while lower CRP levels were associated with young age. Gender, ethnicity, hypoxemia and neutrophilia were not associated with CRP levels. The sensitivity of CRP ≥2 mg/dl for bacterial disease was 65%–70%, declining to <50% in CRP ≥5 mg/dL. The specificity of CRP ≥2 mg/dL for bacterial disease was ~60%, increasing to >80% in CRP ≥5 mg/dL. Conclusions: CRP levels of ≥5 mg/dl high specificity for bacterial disease, allow for its use as an indicator to start empiric antibiotic treatment, while the low sensitivity of various CRP levels probably excludes the possibility of withholding empiric antibiotic treatment, based only on CRP level.
AB - Aim: C-reactive protein (CRP) is commonly used to assess the probability of bacterial infection and the need for empiric antibiotic treatment. We assessed the relationship between CRP levels and bacterial infection in the paediatric emergency room (PER) setting. Methods: A retrospective cohort study. Relative risks (RRs) were calculated for the relation between CRP levels and various demographic, clinical and diagnosis parameters. Additionally, the sensitivity and specificity of different CRP levels (2, 5, 7 and 10 mg/dL) for bacterial infection diagnosis was calculated. Results: Overall, 13 092 cases were recorded. Fever, leucocytosis, hospitalisation and bacterial infection diagnosis were associated with elevated CRP levels, while lower CRP levels were associated with young age. Gender, ethnicity, hypoxemia and neutrophilia were not associated with CRP levels. The sensitivity of CRP ≥2 mg/dl for bacterial disease was 65%–70%, declining to <50% in CRP ≥5 mg/dL. The specificity of CRP ≥2 mg/dL for bacterial disease was ~60%, increasing to >80% in CRP ≥5 mg/dL. Conclusions: CRP levels of ≥5 mg/dl high specificity for bacterial disease, allow for its use as an indicator to start empiric antibiotic treatment, while the low sensitivity of various CRP levels probably excludes the possibility of withholding empiric antibiotic treatment, based only on CRP level.
KW - CRP
KW - bacterial infection
KW - children
UR - http://www.scopus.com/inward/record.url?scp=85212861146&partnerID=8YFLogxK
U2 - 10.1111/jpc.16752
DO - 10.1111/jpc.16752
M3 - Article
C2 - 39716368
AN - SCOPUS:85212861146
SN - 1034-4810
JO - Journal of Paediatrics and Child Health
JF - Journal of Paediatrics and Child Health
ER -