Allometric Scaling in the ICU: a proof of concept

Indirect calorimetry provides an accurate measurement of resting energy expenditure (REE) in critically ill patients, but the cumbersome nature of the technique and need for special training, specific clinical circumstances and equipment prevents broad application. As a result, surrogate methods of mathematically estimating REE are more commonly used to guide management of fluids and nutrition (Harris Benedict, WHO, Schofield and Holliday-Segar). There is no consensus on the optimal mathematical method in critically ill patients. Although some patient characteristics vary linearly with mass, such as blood volume, more complicated processes like REE, free water requirement; cardiac output (CO), carbon dioxide elimination (VCO2) and oxygen consumption (VO2) have been shown to scale according to surface area. This concept is termed allometric scaling and is encompassed in the mathematical relationship known as the surface law. Surface-area-based relationships for estimating energy variables determined by Brody’s number (weight (kg) to the ¾ power) have been validated in clinical pediatric medicine in the practice of closed circuit anesthesia in well, healthy children. We hypothesize that applying these surface-area-based relationships to critically ill children will allow us to create a predictive framework that more accurately measures resting energy expenditure (REE) and its dependent variables than the current standard of care. As a proof of concept, we obtained a retrospective cohort of 104 children who underwent serial measures of indirect calorimetry with mass spectroscopy following cardiopulmonary bypass. We have performed calculations using the standard estimation equations that drive clinical management (Harris-Benedict, WHO, and Schofield) as well as our novel allometric method (REE = k*x3/4 ) and compared these estimated values to measured values of REE to determine relative accuracy of each equation. We also compared the level of accuracy amongst equations in an interrupted time-series analysis to determine at which point following a cardiopulmonary bypass, these equations are most likely to be accurate. Preliminary analysis demonstrates non-inferiority of allometric estimation of REE relative to current clinical estimation equations; however, specific models and analyses are forthcoming.