Emergency Department Markers of Severe Pediatric Pneumonia

Pneumonia is a common infection in childhood. We tested the utility of initial clinical assessment tools and of early laboratory markers for predicting severe disease among children admitted with pneumonia.

This was a retrospective cohort study of previously healthy children admitted to a children's hospital with pneumonia from June 2017 to June 2021. We tested the predictive utility of the Canadian Triage and Acuity Scale (CTAS) score and of the pediatric assessment triangle (PAT) upon first patient contact. We also evaluated the predictive value of elevated WBC, ANC, CRP, lower serum albumin, and phosphorus drawn in the ED, with severe pneumonia defined as concurrent sepsis, need for pleural drainage, or intensive care during index admission. Analyses were bivariate.

During the study period, 554 children admitted with pneumonia were included in the analysis [median age, 25.5 mo (IQR: 14.3 to 63.0); 58.3% males]. Seventy-four (13.3%) had severe pneumonia.Severe pneumonia was diagnosed in 17% of children classified as CTAS 2 (emergent), 13% with CTAS 3 (urgent), 13% with CTAS 4 (less urgent), and 0% with CTAS 5 (nonurgent). Among children with an abnormal PAT, 20% had severe pneumonia versus 5% among those with a normal PAT. The receiver operating characteristic area under the curve (95% CIs) was 0.73 (0.66-0.80) for serum albumin, 0.66 (0.61-0.72) for CRP, 0.61 (0.58-0.66) for phosphorus, 0.58 (0.51-0.64) for ANC, and 0.55 (0.49-0.62) for WBC. The AUC for PAT was 0.71 (95% CI: 0.65-0.77) and for CTAS was 0.56 (95% CI: 0.49-0.62).

For the prediction of severe pediatric pneumonia, no single marker offered robust utility. CTAS showed no clear utility, while PAT was comparable to albumin, the best-performing laboratory marker. Hypoalbuminemia was marginally superior to elevated CRP. The remaining laboratory markers had at best moderate accuracy in isolation.