Respiratory rate is an early predictor of clinical deterioration in children.

OBJECTIVE To assess the value of respiratory rate (RR) as a predictor of clinical deterioration in children, compared with other vital sign measurements. DESIGN A retrospective case-control study, comparing children who deteriorated, requiring admission to critical care with children who did not deteriorate. METHODS RR, heart rate (HR), and blood pressure (BP) measurements were collected from each patient for a 48-hour duration. The 95th centile was identified for each and 5% to 30% thresholds above the 95th centile were calculated. For each threshold the sensitivity, specificity, odds ratio, positive, and negative predictive value for deterioration was calculated. RESULTS Forty cases (age range 7 weeks-15 years) and 40 control patients matched for age, gender, and hospital location were recruited. In 30/40 patients who deteriorated at least one RR ≥ 30% above the 95th centile for their age was recorded in the 48 hours before deterioration, compared with 10/40 controls, regardless of clinical diagnosis. Only 3/40 children that deteriorated had a HR > 30% greater than the 95th centile, compared with 2/40 controls. An elevated RR was the only vital sign whose odds ratios were significant at each threshold level above the 95th centile. Maximum RR occurred 16.8 hours before deterioration. CONCLUSION RR is a more accurate predictor of clinical deterioration in children than other vital signs. Greater weighting and importance should be placed on RR, which is often omitted in children due to difficulties with its measurement.