Malnutrition and outcome in hospitalized children in Europe
(ESPEN Network Grand Project)

Disease-related malnutrition affects 20-50 % of adult hospitalized patients and
was shown to seriously affect relevant outcomes, such as recovery from disease
or surgery, length of hospital stay and cost of care (Norman et al 2008).
In paediatric patients, disease associated malnutrition has even more severe
consequences on a short- and long-term basis, for example with respect to
disease course and mortality, complication rates, growth, development,
well-being and long-term health outcomes (Goulet and Koletzko 2004; Koletzko et
al 2008). According to prior studies, malnutrition affects about 15-30 % of
hospitalized children in Europe (ESPGHAN 2005, Pawellek et al 2008, Joosten and
Hulst 2008). However, available criteria for defining malnutrition in
paediatric patients are inconsistent, not based on firm evidence, and not
generally agreed upon. Current guidelines do not address assessment of and
screening for childhood malnutrition. Therefore, a large number of affected
children are not adequately diagnosed.
One aim of this study is to assess the prevalence of malnutrition and patients
at risk for malnutrition among hospitalized children across Europe. In addition
criteria to link anthropometric measurements and the prediction of outcome,
i.e. length of hospital stay, shall be established. A further goal then is to
establish agreed, evidence-based criteria for malnutrition in children with the
purpose of leading to an agreed, evidence-based screening tool for paediatric
malnutrition in developed countries. This tool shall include a set of simple
questions, based on previously suggested tools. Thereby this study will provide
a strong basis for implementing evidence-based nutritional interventions in
paediatric patients by harmonisation of diagnostic criteria for childhood
malnutrition in developed countries.

These two hypotheses will be tested:
• Global nutritional status of hospitalized children in Europe can be assessed
in a meaningful way by standardized anthropometric measures and simple
questions related to patient*s history, based on previously suggested
paediatric screening
• Nutritional status of hospitalized children at admission in Europe is
associated with a relevant clinical outcome.

The purpose of this study is to characterize in hospitalized children across
Europe:
• the prevalence of malnutrition on admission and at discharge,
• the effects of malnutrition on outcomes, such as length of hospital stay and
occurrence of infectious complications,
• the predictive values of different descriptors of malnutrition for relevant
outcomes,
• the prevalence of paediatric hospital patients at risk for malnutrition, as
well as predictors of increased risk,
• the relative value of proposed paediatric screening tools, for identifying a
subpopulation with increased risk for malnutrition, for relevant outcomes

This is a prospective European multi-centre cohort study.
The following data will be collected by the named assessors using specifically
designed case report forms. Data will be obtained at admission and at discharge
of all patients which have been rated eligible by the assessor. All collected
data will be treated, analysed and archived confidentially.

1. Interview at admission
A. Demographic and medical data: gender, date of birth, date of admission,
ethnic back ground, chronic disease, ICD 10 classification for main diagnosis,
ICD 10 classification for reason of admission, quality of life assessment

B. Questionnaire data for nutritional status: high risk disease, acute
admission/condition effect on nutrition, subjective clinical assessment,
dietary intake and nutritional losses, frequency and duration of
gastrointestinal symptoms, weight loss or poor weight gain, previous dietary
support.
This questionnaire integrates the questionnaires of three previous nutrition
screening tools, sorted by items: The Paediatric Yorkhill Malnutrition Score
(PYMS), the nutritional risk screening tool, called STRONGKids (Hulst et al
2009) and the screening tool for the Assessment of Malnutrition in Paediatrics
(STAMP) (McCarthy et al 2008).

2. Anthropometric and BIA measures at admission: weight (kg), height (cm),
mid-upper arm circumference (cm) , Triceps skin fold thickness (mm), Handgrip
strength (kg), Bioelectrical impedance analysis (BIA), in those centres where
feasible, due to available BIA equipment
The admission data has to be collected within 24 hours after admission time.
BIA measurement will be performed with all children aged three years or older.
Handgrip strength will be performed with all children aged six years or older.
Hydration status, handedness and time of last eating/drinking are to be
recorded.

3. Anthropometric and BIA measures at discharge: weight (kg), mid-upper arm
circumference (cm) , Triceps skin folder thickness (mm), Handgrip strength
(kg), Bioelectrical impedance analysis (BIA), in those centres where feasible
Anthropometry is going to be repeated at discharge for all patients with LOS >=
96 hours, when possible. Otherwise the last measured weight is to be recorded.

4. Collection on outcome data: date of discharge (days of hospital stay),
number of days with temperature >38.5oC, number of days with diarrhoea, number
of days with vomiting, number of days with antibiotic use, number of days with
need for nutritional support during stay in any form, in addition to any prior
to admission, further complications (Infections, other)

Data will be derived from the hospital record.

The study takes place while the child is hospitalized. Data are collected at
two time points: admission and discharge. There is no need for additional
hospital visits, blood drawings or invasive measurements. The study comprises
of:
- Short interview with questions related to previous health, weight loss,
gastrointestinal symptoms and nutritional intake.
- Measurements of weight and length (standard care)
- Measurements of mid-upper arm circumference and triceps skin fold thickness
- Measurements of bioelectrical impedance analysis (BIA) (children >= 3 years)
- Measurement of handgrip strength (children >= 6 years)

The burden for the child is minimal, because some of the measurements and
questions are part of the standard hospital care and admission procedure. The
additional measurements are the measurements of mid-upper arm circumference,
triceps skin fold thickness, BIA and measurement of handgrip strength. The
skinfold measurement can be experienced as unpleasant. However, this
measurement takes only a few seconds and can be compared in terms of burden
with the measurement of blood pressure. The BIA measurement has no burden: two
electrodes are placed at a hand and a foot of a child, which are comparable tot
electrodes placed when taking an ECG. The child does not notice anything from
the measurement itself. In order to measure the hand grip strength the child
must squeeze in a tool, this has no burden.

The research question can only be answered by including children as subjects,
because the aims of the study are specifically focused on children. Measurement
of nutritional status and screening tools for malnutrition that have been
developed for adults, can not be applied to children because age specific
characteristics such as growth, have to be taken into account.