Abstract

Defining Distributions of physiological variables on admission to a Pediatric Intensive Care Unit and their course over time

Author: Danny Eytan

Background: Variables such as heart rate, respiratory rate and blood pressure form the basis for evaluating and monitoring the physiological state of patients. Acutely-ill children in the critical care settings are usually heavily monitored, these variables often are continuously recorded and serve for modifying interventions and assessing response to treatments. Although ranges for these values in healthy children have appeared in textbooks, they have only recently been evaluated in an evidence based manner in hospitalized children and to the extent of our knowledge, there are no studies reporting the distributions of such basic physiological variables in children managed in the intensive care unit during critical illness and over recovery. Knowledge of the expected distribution of values on admission and their trajectory can serve to better define target ranges for alarms for example and to identify patients at risk as those who deviate from the population norms and expected trajectory. This study was aimed at filling the above mentioned knowledge gap by providing estimates for the vital signs distributions in critically-ill children admitted to a high-flux pediatric intensive care unit. Methods: We used an electronic signal repository containing all recorded physiological variables from the clinical bedside monitoring units for all patients admitted to The Hospital for Sick Children’s critical care unit since April 2013. ~130 million values collected from 1289 children served to estimate distributions on admission and over the first 3 days post admission for heart rate and arterial blood pressure. These data allowed us to generate detailed distributions of physiological variables for 19 distinct surgical procedures and admission diagnoses. Results: Heart rate and blood pressure on admission are significantly dependent on patient age and admission diagnosis by multiple regression analysis. We show that heart rate tends to decrease over time post admission while blood pressure is relatively stable with significant differences in the directions and magnitude of these trends when analyzed by admission diagnosis. Moreover, for any given time interval the variance of either heart rate or arterial blood pressure for the individual patient is several folds smaller than the variance of the group of patients with the same diagnosis. Thus, for any given individual patient and admission diagnosis it is possible to define the expected trajectory over time in the state space defined by the recorded physiological variables. Conclusion: The distributions of basic physiological variables in children in a pediatric intensive care setting are described and defined by admission diagnosis and time from admission. These distributions may serve as useful references for bedside clinicians to help better define diagnostic and alarm thresholds to and identify early patients that deviate from the expected trajectory. Moreover, the marked difference in variance between the individual patient and the population of patients with the same diagnosis (or age) hints that alarm thresholds may be better set by using each patient as their own control rather than by using the range defined by the population of similar patients.

Co Author/Co-Investigator Names/Professional Title: Danny Eytan, MD PhD, Department of Critical Care Medicine, Neurosciences and Mental Health Program, Research Institute, The Hospital for Sick Children, University of Toronto, Toronto, Canada. Andrew Goodwin, B.Sc, Department of Critical Care, The Hospital for Sick Children, University of Toronto, Toronto, Canada. Robert Greer, B.Eng, Department of Critical Care, The Hospital for Sick Children, University of Toronto, Toronto, Canada. Anirudh Thommandran, B.Sc, Department of Critical Care, The Hospital for Sick Children, University of Toronto, Toronto, Canada. Anne-Marie Guerguerian, MD PhD, Department of Critical Care, Neurosciences and Mental Health Program, Research Institute, The Hospital for Sick Children, University of Toronto, Toronto, Canada. Peter Laussen, MBBS, FCICM. Department of Critical Care, David and Stacey Cynamon Chair in Critical Care Medicine, The Hospital for Sick Children; Department of Anaesthesia, University of Toronto, Toronto, Canada.