Objective: The purpose of this study was to establish the usefulness of capillary refill time when measured during the initial assessment of children.
Methods: All children with spontaneous illness attending a paediatric accident and emergency department over a 7-month period were eligible for entry into the study. Capillary refill time was measured at the fingertip, using a standard technique, as part of the initial assessment. Each child was then followed up to ascertain clinical progress, including the need for admission, intravenous fluids, length of stay and diagnosis, as well as the white cell count when this was available. The value of capillary refill time as a predictor of the markers of illness severity was then assessed.
Results: Capillary refill time measurements were recorded on 4878 children. There was no significant association of capillary refill time with meningococcal disease, other significant bacterial illness or the white cell count. A prolonged capillary refill time was associated with a more urgent triage category, the administration of a fluid bolus and the length of hospital stay (P<0.0001). The best performance was obtained when a capillary refill time of 3 s or more is taken to be 'prolonged'. However, this gave positive predictive values of only 9% for a triage category of 1 or 2 (negative predictive value 97%), 11% for requiring a fluid bolus (negative predictive value 99%), 55% for hospital admission (negative predictive value 65%) and 22% for stay over 2 days/death (negative predictive value 91%).
Conclusion: The prolongation of capillary refill time is a poor predictor of the need for intravenous fluid bolus or hospital admission.