Objective To determine the agreement between peripheral and central capillary refill time (pCRT/cCRT) and their diagnostic value for the detection of serious bacterial infection (SBI) in febrile children presenting to the paediatric emergency department (ED).

Methodology A prospective observational study at the Paediatric ED, Erasmus MC-Sophia Children’s hospital, the Netherlands. We included 1193 previously healthy febrile children (1 month^-15 years) with data on both pCRT- and cCRT-measurements available, as recorded in categories (normal <2 s., prolonged 2–4 s. and severely prolonged >4 s.) by triage nursing staff.

Main outcome measures were agreement between pCRT and cCRT (determined by weighted kappa), diagnostic odds ratio (DOR) and area under the receiver-operating characteristic curve (AUC) for the detection of SBI.

Results Abnormal pCRT was observed in 153 (12,8%) and abnormal cCRT in 55 (4,6%) children.

Overall agreement was 0,343 (considered as ‘fair’). Agreement stratified for age was lower (0,143) in the group of 1–5 years of age. Stratified for body-temperature, agreement showed a declining trend ranging from 0,509 (temperature <37,5°C) to 0,201 (temperature >39,5°C).

The DOR of abnormal pCRT (>2 s.) for SBI was 1.10 (95% CI:0,65–1,84), with an AUC of 0,505 (95% CI:0,454–0,557). For abnormal cCRT (>2 s), the DOR was 0,43 (95% CI:0,13–1,39) with an AUC of 0,514 (95% CI:0,464–0,564). Presence of both abnormal pCRT and cCRT did not improve diagnostic performance.

Conclusions pCRT and cCRT showed fair agreement in febrile children at the ED which was reduced particularly in children with high temperature and age 1–5 years. Both abnormal pCRT and cCRT showed low diagnostic value for the detection of SBI.