73 / 208
Use of indirect calorimetry to detect overfeeding
71
4
INTRODUCTION
Nutritional support affects outcome in critically ill children
1-3
. Undernutrition has long been the
primary focus for nutritional research, but overfeeding is also prevalent in pediatric intensive
care units (PICUs)
1,4-6
. Caloric overfeeding is associated with increased mortality in critical ill
adults
7
. It may lead to liver dysfunction by increasing the risk for hepatobiliary complications,
such as steatosis and cholestasis, and might increase the risk of infection secondary to
hyperglycemia
8
. Overfeeding of glucose leads to lipogenesis with an increase in carbon
dioxide
9
, resulting in a difficulty to wean from the ventilator
10,11
. Furthermore, overfeeding
during critical illness might evoke a phenotype of autophagy deficiency as a potentially
important contributor to mitochondrial, organ and skeletal muscle damage, particularly when
amino acid enriched parenteral nutrition (PN) is provided
12,13
. Also in critically ill children,
unintended consequences of overfeeding are likely to occur
14
.
To prevent these detrimental effects, nutritional therapies are ideally guided by resting energy
expenditure (REE) throughout the course of illness
15
. REE can be measured (mREE) by indirect
calorimetry or predicted by use of equations, and might be affected by the type, severity and
stage of disease
16-18
. Because there is a lack of studies using clinical endpoints to determine
the optimal caloric intake in critically ill children, recommendations on minimum caloric
intake are often based on equilibrating energy or protein balances
19,20
. So far, however, no
clinical endpoint or (surrogate) marker has been studied to determine the optimal maximum
caloric intake in this population. Overfeeding is arbitrarily defined as a ratio caloric intake/
REE >110%
7,21-23
or >120%
14,24-28
(see related studies in Table 1). As an alternative method
the comparison of measured respiratory quotient (RQ) to the predicted RQ based on the
macronutrient intake (RQ
macr
) is suggested
29,30
. The measured RQ is derived from the ratio of
CO
2
production over O
2
consumption and reflects the use of different substrates. An RQ value
>1.0 indicates lipogenesis, and is frequently used to identify carbohydrate overfeeding
29
.
RQ
macr
is the weighted average of the RQs of the different macronutrients administered, which
can be obtained from the modified Lusk table. A difference >0.05 between RQ and RQ
macr
has
been proposed to define overfeeding
29,30
.
The aim of the present study was to compare different definitions of overfeeding in critically
ill mechanically ventilated children based on measurements of mREE, RQ and caloric intake
and to find an appropriate definition to study the effect of overfeeding on clinical endpoints
in future trials.