SCCM Resource Library-54-Performance of Various Predictive Equations Compared to Indirect Calorimetry in Ventilated Children

Video Transcription

Hello, all. I am Prithvi Sandhy. I am one of the attending physicians in the Pediatric Intensive Care Unit at Nicholas Children's Hospital in Miami. And today I will be talking  to you about performance of various predictive equations compared to indirect calorimetry in ventilated children. The authors have no disclosures. Energy expenditure is variable  in critically ill children. Indirect calorimetry can accurately measure energy expenditure.  However, the use of indirect calorimetry can be expensive. It can utilize a lot of resources.  It may not be readily available in many ICUs, and it may not be practical for day-to-day use. Because of these reasons, there are few equations which could be used to calculate  energy expenditure. These equations consider age, sex, weight, and height of the patient.  One of the advantages of using equations are they are easy and inexpensive. However, variable  accuracy of data has been reported while using equations, and making nutritional decisions based on these equations could lead to under or overfeeding patients.  The objective of our study was to evaluate the bias and precision of various predictive equations in estimating energy expenditure compared to indirect calorimetry in critically  ill ventilated patients. Moving on to methods, our study was a single institution retrospective  chart review study. It consisted of 40 patients. We compared four predictive energy expenditure  equations, that is the WHO, Schofield, resting energy expenditure by Institute of Medicine,  and energy expenditure calculation using VCO2, and compared them to energy expenditure calculated  using indirect calorimetry. We used SPSS 25.0 for analysis. The various statistical methods used were blind Altman test, one sample t-test, and linear regression analysis.  Moving on to the results section, among 40 patients in our study, 32 of them were endotracheally  intubated, and eight of them had tracheostomies. 30 were males, and 11 of them were on vasoactive  medication. Among 40 patients, 11 were underweight, 19 had normal weight, and 10 of them were overweight. The bottom table shows the mean age, weight, height, BMI, CO2 production,  and energy expenditure measured by indirect calorimetry. This bar graph demonstrates mean calories as calculated by indirect calorimetry, which is the bar in red, and mean calories  calculated by various other energy expenditure equations, which are shown by the blue bars.  You can note the energy expenditure as calculated by indirect calorimetry was 1248 kilocalories  per day, and energy expenditure calculated by the other equations were slightly higher or below in estimating energy expenditure.  Over the next few slides, I will be showing the comparison of energy expenditure as calculated  by indirect calorimetry to various predictive equations. In this Bland-Altman plot, we are  comparing energy expenditure as calculated by indirect calorimetry to the Scofield calculator.  The mean value is depicted on the x-axis, and the difference between the tests are depicted on the y-axis. The solid horizontal line you see is the bias, which here when you compare  the mean difference between indirect calorimetry to the Scofield calculator was minus 27 kilocalories.  The dotted line above and below the solid red line are the upper and lower line of agreement,  which depicts precision, and here it was 243 kilocalories. The p-value was 0.4. Henceforth, there was no statistically significant difference amongst the tests.  In this Bland-Altman plot, we are comparing energy expenditure as calculated by indirect calorimetry to the WHO calculator. The bias as you can see depicted by the solid red line  was minus 88 kilocalories. That is the mean difference between these tests. The precision was 269 kilocalories, and the p-value was 0.049, denoting a statistically significant  difference between these tests. In this plot, we are comparing energy expenditure as calculated  by indirect calorimetry to EEVCO2 calculator. The bias was 151 kilocalories, precision was  246 kilocalories, and the p-value was less than 0.001. Here we are comparing energy expenditure  as calculated by indirect calorimetry to EER calculator. The bias was 306 kilocalories, precision was 265 kilocalories, and the p-value was less than 0.001.  In this table, we show a comparison between the results from indirect calorimetry to various predictive equations. You can note energy expenditure values calculated by Scofield  equation was closest to the values obtained by indirect calorimetry method compared to various other equations. From this, we conclude the Scofield equation compared most favorable  across all BMI groups in predicting energy expenditure in critically ill ventilated children  compared to indirect calorimetry. Scofield equation has the smallest bias at 2%. However, the limits of agreement are wide for all equations, including the Scofield equation.  I would like to thank the audience for your participation, and now I am open for questions.

Video Summary

Dr. Prithvi Sandhy discusses the performance of various predictive equations compared to indirect calorimetry in estimating energy expenditure in ventilated children. Indirect calorimetry is an accurate but expensive method for measuring energy expenditure. Due to cost and availability constraints, equations based on age, sex, weight, and height are used as alternatives. However, these equations may not provide accurate results, leading to under or overfeeding of patients. The study compared four predictive equations to indirect calorimetry and found that the Scofield equation performed the best, although all equations had wide limits of agreement.

Asset Subtitle

Pediatrics, GI and Nutrition, 2021

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2021 Critical Care Congress held virtually from January 31-February 12, 2021.

Prithvi Raj Sendi Keshavamurthy

Introduction/Hypothesis: Energy expenditure (EE) is variable in critically ill children. This may lead to underfeeding or overfeeding with serious consequences. Predictive equations are commonly used to assess the nutritional needs of hospitalized children. There is a paucity of data regarding the performance of various predictive equations in critically ill children. Our objective was to evaluate the bias and precision of various predictive equations in estimating EE compared to indirect calorimetry (IC).

Methods: We retrospectively analyzed the performance of four predictive equations to estimate EE; WHO, Schofield, Resting Energy Expenditure by the Institute of Medicine (EER), EE calculation using VCO2 (EEVCO2) and compared them to EE calculated using IC (Ultima CCM indirect calorimeter, MGC Diagnostics) in 40 ventilated children in the intensive care unit. Agreement between two methods was tested by calculating the bias (mean difference) and precision (standard deviation of the bias) for predicted EE from each predictive equation compared to IC using Bland-Altman methodology. We used one-sample-t-testing to determine if bias was significantly different from zero. Linear regression analysis (difference vs mean of the two methods) was performed to evaluate for the presence of a proportional error (evaluates the performance across the range of the values). We used SPSS version 25 for analysis.

Results: Of the 40 ventilated children, 32 were endotracheally intubated, 8 had tracheostomies, 30 were male and 11 were on vasoactive medication. Based on BMI for age, 11 children were underweight, 19 normal weight, and 10 overweight or obese. The mean (+/- SD) age (years), weight (kg), height (m), BMI (kg/m2), VCO2 (ml/min), and IC (Kcal/day) were, 10.8Â±4.3, 34.6Â±15.7, 1.34Â±0.23, 18.6Â±4.8, 169Â±56, and 1248Â±355, respectively. The bias [%, p] (precision) was, Schofield -27 [2.2, p=0.4] (243), WHO -88 [7.3, p=0.049] (269), EEVCO2 151 [11.4, p<0.001] (246), and EER 306 [21.9, p<0.001] (265) Kcal compared to IC. Proportional errors were not present with Schofield and EER.

Conclusions: The Schofield equation compared most favorably across all BMI groups in predicting EE in critically ill ventilated children compared to indirect calorimetry. Studies with separate analyses in different BMI and age categories are needed.

Meta Tag

Content Type Presentation

Knowledge Area Pediatrics

Knowledge Area GI and Nutrition

Knowledge Level Advanced

Membership Level Select

Tag Calorimetry

Year 2021

Keywords

predictive equations

indirect calorimetry

energy expenditure

ventilated children

Scofield equation

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.