Dear Editor, Advances in paediatric critical care have resulted in the increased survival of chronically ill patients. In many units, such patients represent [50 % of the workload and are often malnourished [1]. Prediction of energy expenditure (PEE) equations are simple alternatives to the gold standard of assessing resting energy expenditure (REE) by indirect calorimetry (IC). However, recent work has suggested that PEE equations fail to adequately predict REE. New paediatric intensive care unit (PICU)-oriented equations have been found to be no better than well-established tools, such as the Schofield-HW equation [2]. We have compared PICU-specialized and commonly used PEE equations with REE using a new modular metabolic monitor (E-COVX) in well-nourished and malnourished critically ill children. The E-COVX compact metabolic module has the advantage of not being influenced by uneventful open endotracheal suctioning [3]. It is suitable for repeated 30-min measurements in well-sedated mechanically ventilated children with stable respiratory patterns in a variety of ventilation modes [4]. We hypothesized that IC using the E-COVX would show that energy expenditure is unpredictable in malnourished children and that replacing REE with any PEE equation, including the new PICU-oriented equations, might not be applicable for estimating energy expenditure in patients suffering from malnutrition during a critical illness. The Ethics Committee of the Institutional Review Board approved this study, and parents or guardians of the children gave informed, written consent.Mechanically ventilated critically ill children, consecutively admitted to the PICU of the University Hospital, Heraklion, Crete, were enrolled in the study. Thirty consecutive 1-min gas exchange measurements (VO 2 and VCO 2 ) were taken and the respiratory quotient (RQ) and REE were calculated for each patient (results were blinded from the attending physician). The nutritional status was evaluated for the presence of proteinenergy malnutrition as defined by Waterlow [5]. The PEE was estimated and basal metabolic rate (PBMR) was predicted using the common Harris-Benedict, Schofield-HW, Seashore, Fleisch, CaldwellKennedy and Henrys formulas, as well as equations specifically developed for the PICU by White and by Meyer [2]. Patients were classified as hypermetabolic, normometabolic and hypometabolic when the REE was [110, 90-110 and \90 % of the Schofield-PBMR, respectively, and as overfed, adequately fed and underfed when the caloric intake was [110, 80-110 and \80 % of the REE, respectively.Forty-four patients (28 boys, 16 girls) were studied (total measurements 1,320). Four patients were admitted after an elective procedure; the remaining patients were admitted as a result of an acute illness or injury. Twelve (27.3 %) patients presented severe (4) or moderate (8) malnutrition. All patients survived.The REE did not differ between patients according to a body temperature of \37.1°C (1,068 ± 460 kcal/ day) or C37.1°C (824 ± 217 kcal/ day) or with a Ramsey s...