Accuracy of resting energy expenditure predictive equations in coronavirus disease 2019 (COVID-19) survivors

Summary: Background &amp; Aims: Coronavirus disease 2019 (COVID-19) may be associated with abnormal energy metabolism and lead to inaccurate resting energy expenditure (REE) estimations by predictive equations. Here, we report measured REE (mREE) of a group of COVID-19 survivors and compared its accuracy against predicted REE (pREE). Methods: This was a cross-sectional analysis of patients who survived COVID-19 prior to July 2021. An indirect calorimeter was used for mREE and compared against 21 pREE equations, 10 of which used a measure of body composition. Paired t-tests and Bland-Altman analysis were used to evaluate agreement and relative accuracy or bias for percentage error between pREE and mREE; measurements within ±10% were considered accurate. Results: We assessed 38 COVID-19 survivors; age: 48.5y (interquartile range: 40.2, 60.0), body mass index: 29.3±5.6 kg/m2, mREE: 1520± 275 kcal/d, time since COVID-19: 183.2 ±34.4 days. Ten (47.6%) pREE equations were significantly different from mREE (P <0.05). Harris-Benedict equation had the smallest limits of agreement, ranging from -14.3% to 25.8% (or -249 to 393 kcal/d). Mifflin St-Jeor was the most accurate equation (within 10% of mREE). The best performing equation (Mifflin St-Jeor) still over or under-estimated pREE in ∼37% of the patients. Conclusion: A large variability in mREE versus pREE was observed in COVID-19 survivors. Even the most accurate equation (Mifflin St-Jeor) exhibited higher inaccuracies compared to mREE. We need to explore better methods to estimate energy requirements during the COVID-19 recovery period, until more accurate predictive equations are developed this population.