Continuous versus Intermittent Nutrition in Paediatric Intensive Care: a Proof-of-concept

We have recently shown in a large randomised study in paediatric ICU patients
that withholding Parenteral Nutrition during the first week of critical
illness, as compared with an early start (<48 hours), reduced length of
intensive care dependency and the number of nosocomial infections. The benefit
of this counterintuitive nutritional strategy, through which low macronutrient
intakes were accepted, is presumed to be caused by activation of autophagy due
to a fasting response. Autophagy is an intracellular recycling process crucial
for maintaining cellular integrity and function. Its protective role against
various forms of critical illness induced organ failure, including ICU acquired
muscle weakness, is strongly activated during periods of fasting. Currently,
artificial feeding is usually administered through continuous infusion,
although solid evidence supporting this practise is lacking. Intermittent, as
compared with continuous, (par)enteral nutrition may provide a more
physiological feeding/fasting pattern which activates autophagy, while
providing sufficient nutrition during critical illness. A physiological
feeding/fasting pattern could also sustain circadian rhythm. Thus, such
strategy could impact essential mechanisms such as immunology, metabolism,
neuropsychology, and pharmacodynamics, to improve recovery from critical
illness.

The main research objectives for the proof-of-concept study are to show the
feasibility and safety of a daily cycle of feeding and fasting in critically
ill children of different age-groups, that will trigger an adequate fasting
response while providing equal amounts of daily nutrients as with standard
continuous feeding. We hypothesise that in critically ill children intermittent
versus continuous feeding is feasible and safe, and will lead to a fasting
response, which could potentially activate autophagy while providing sufficient
nutrition. Ultimately we hypothesize that such strategy will lead to
accelerated recovery and reduced ICU dependency, which is to be tested in a
large multicentre RCT with clinically relevant outcome parameters in a
multicentre setting within the Rotterdam-Leuven consortium.

A randomized non-blinded proof-of-concept study will be performed to explore
the feasibility and safety of intermittent feeding in critically ill neonates
(<28d, n=30), infants (<1yr, n=30) and children (*1yr, n=30) while overall
providing equal amounts of daily nutrients as with standard continuous feeding.
Based upon adult pilot data (unpublished data) and glycaemic control data from
the PEPaNIC RCT (n=1440, CCMO projectnr: NL38772.000.12 / MEC-2012-412 /
EudraCT 2012-000811-10 / S54127 / NCT01536275), a 12hrs fasting period is
expected to be optimal. The study intervention will last a maximum of 14 days
or until PICU discharge, or the ability for *oral nutrition*, whichever comes
first. Methods for evaluating the fasting response, autophagy and the circadian
rhythm in the planned subsequent RCT will be tested during the proof-of-concept
study.

The *Continuous Nutrition* strategy will be the nutritional management
currently recommended and acting as *control*. Both enteral nutrition (EN) and
(after day 7) parenteral nutrition (PN) will be provided continuously 24
hrs/day. The *Intermittent Nutrition* strategy, acting as the *intervention*,
comprises a physiological *overnight fast during which no artificial nutrition
(EN and/or PN) will be provided for a duration of 12 hours.
For both study arms, daily caloric targets are similar in both randomization
groups and increase during the first week in the PICU. Isocaloric nutrition
between study arms will be achieved with higher intakes during the day and/or
with energy-dense formulas in the intermittent group. Enteral and/or parenteral
nutrition (timing and requirements) will be initiated and administered
according to the most recent nutritional guidelines for all patients; EN will
be started as early as possible (< 24 hrs) unless strictly contraindicated, and
increased in a step-wise manner. PN (glucose/amino acids/lipids) will be
started after day 7 if EN is still insufficient (<80% target intake). To
prevent hypoglycaemia during the overnight fasting period, 1) insulin (if
administered) will be stopped one hour prior the fasting interval and 2)
neonates (±1 mg/kg/min glucose) and infants (±0.5 mg/kg/min glucose) will be
provided with a guaranteed glucose intake, unless they develop hyperglycaemia
with such regimen (at what time the glucose intake will be tapered down and
eventually stopped).

The burden is expected to be minimal. Based on studies in the literature, there
is equipoise over the feeding patterns. The risk in participating to the study
and being randomised to one of the nutritional strategies are minimal (based on
PEPaNIC data), and specifically compass an increased risk of developing
hypoglycaemia in the fasting periods of the *Intermittent* nutrition. Another
potential burden or risk in the intervention group (*Intermittent*) might be
gastrointestinal intolerance to the administration of more nutritional intake
in a shorter time frame. However, safety measures (standardized and regular
checks; hypoglycaemia / lactate / gastro-intestinal tolerance) will be taken to
further decrease these risks. Further risks are negligible, and will only
entail 1) additional blood draws, which will be taken from clinical lines or in
addition to pricks for clinical purpose and 2) neuro-monitoring / sleep
measurements, which will be performed non-invasively.
Clinical equipoise between Continuous and Intermittent nutrition is a specific
problem for critically ill patients in an ICU environment. A similar study has
also been performed in adult ICU patients in Leuven (unpublished data).
However, although the results of this study have provided essential information
for the design of our study, this pilot study can NOT be translated one-on-one
to the paediatric patient, as critically ill children of different age groups
have different metabolic and nutritional issues. The patient group of
critically ill children (0 * 18 yrs), represents a significant proportion of
the worldwide ICU population. Therefore, the feasibility and safety needs to be
determined in our population of critically ill children.