Heart rate cut-offs to identify non-febrile children with dehydration and acute kidney injury

Abstract

We hypothesized that the heart rate (HR) variation in an acute setting compared with HR in wellbeing status could be a good marker of both dehydration and acute kidney injury (AKI). Since HR in wellbeing status is unknown in most cases, we assumed as reliable surrogate the 50th percentile of HR according to age and gender. We evaluated if the estimated percentage of heart rate variation in acute setting compared with 50th percentile of HR (EHRV) could be marker of dehydration and AKI in children. Two independent cohorts, one prospective comprehending 185 children at type 1 diabetes mellitus onset (derivation) and one retrospective comprehending 151 children with acute gastroenteritis and pneumonia (validation), were used to develop and externally validate EHRV as predictor of the ≥ 5% dehydration and/or AKI composite outcome. Febrile patients were excluded. EHRV was calculated as ((HR at admission–50th percentile of HR)/HR at admission) × 100. The prevalences of ≥ 5% dehydration and AKI were 61.1% and 43.8% in the derivation and 34.4% and 24.5% in the validation cohort. For the ≥ 5% dehydration and/or AKI composite outcome, the area under receiver-operating characteristic curve of the EHRV in the derivation cohort was 0.69 (95%CI, 0.62–0.77; p < 0.001) and the best EHRV cut-off was > 24.5%. In the validation cohort, EHRV > 24.5% showed specificity = 100% (95%CI, 96.2–100.0), positive predictive value = 100%, and negative predictive value = 67.1% (95%CI, 64.7–69.5). The positive likelihood ratio was infinity, and odds ratio was not calculable because all the patients with EHRV > 24.5% showed ≥ 5% dehydration and/or AKI. Conclusions: EHRV appears a rather reliable marker of dehydration and AKI. Further validations could allow implementing EHRV in the clinical practice.What is Known:• Increased heart rate (HR) is an easily and quickly detectable sign of dehydration in childhood, but its cut-off to suspect dehydration or acute kidney injury (AKI) is not defined.What is New:• We found that a percentage of estimated HR variation in acute setting in comparison with 50th percentile of HR (EHRV)>24.5% predicted ≥5% dehydration and/or AKI in non-febrile patients.• We provide a one-page tool to suspect ≥5% dehydration and/or AKI on the basis of the HR. If furtherly validated, this tool could be implemented in the daily clinical practice.