Proefschrift Kerklaan

Chapter 1

child’s REE can accurately be reflected by a measurement of at least 5 minutes 104 . However, most measurements will take at least 30 minutes, taking into account the time to connect the metabolic monitor and the time to reach steady state. Within-day and between-day variations in REE from the acute phase to the stable phase are small in the majority of critically ill children 1,94,105-107 , so a single measurement early during admission may serve to guide nutritional therapy. Since REE remains stable, but requirements are likely to change during the different phases of critical illness, the optimal caloric intake in relation to REE is likely to vary as well. Despite its superiority in predicting REE, only a minority of PICUs uses IC to determine energy requirements 9 , because measurements are time consuming and limited to stabilised mechanically ventilated children with mild ventilator settings or spontaneously breathing children without need for oxygen. High purchase and maintenance expenses of metabolic monitors further limit availability of IC. Alternatively, a simplified metabolic equation using ventilator-derived VCO 2 measurements, could allow measurement of energy expenditure in absence of a metabolic monitor 108 . However, this approach needs to be validated for use in critically ill children. Due to the limited availability and practice of IC, REE is predominantly predicted by age- dependent equations based on weight and/or height. These equations, derived from measurements in healthy children, do not predict energy requirements accurately in critically ill children, resulting in an increased risk of malnutrition during PICU stay 105,109,110 . Several factors, commonly present in the PICU, affect measured REE; fever is found to increase REE, while sedatives and muscle relaxants have shown to decrease it 111 . An increase of REE is also seen in children with burns 112 , septic neonates 30,113 and in children after major surgeries, but only temporarily 31 . However, despite these established effects, the application of uniform correction factors to REE for the whole PICU population is simplistic and likely to be inaccurate 4 . Therefore, when IC is not possible, it is preferred to derive REE from Schofield’s formula for weight, without the addition of stress or activity factors 4 . 2 ), known as the respiratory quotient (RQ), reflects the utilisation of different substrates. A value >1.0 indicates lipogenesis and can be used to identify carbohydrate overfeeding 114-116 . A high amount of carbohydrates will not always result in an RQ >1.0 because ongoing utilisation of fat for energy, as seen in critical illness, will lower the measured RQ 117 . RQ is also affected by hyperventilation and metabolic acidosis. Therefore, a cautious interpretation of this variable is necessary before adjusting nutritional practices. The measured RQ value can also function as an indicator of caloric overfeeding when it is compared to the predicted RQ based on the macronutrients provided (RQ macr ) 118,119 . Its adequacy to detect overfeeding is affected by the presence of endogenous energy production, as seen in children with caloric intake below Respiratory Quotient The VCO 2 and VO 2 values obtained by IC are not solely used to calculate REE. Their ratio (VCO 2 / VO

18