Metabolic rate analysis of healthy preterm and full-term infants
during the first weeks of life1,2
Jacqueline Bauer, Claudius Werner, and Joachim Gerss
ABSTRACT
Background: Longitudinal data on resting energy expenditure
(REE) in extremely immature infants and full-term neonates are
scarce but are necessary to understand the energy requirements in
neonatal nutrition during the first weeks of life.
Objective: The aim of the present study was to measure REE and
its main components longitudinally during the first weeks of life to
quantify their significant determinants.
Design: REE was investigated longitudinally over a period of 6 wk
in healthy, stable, and growing preterm infants and over 5 wk in
full-term neonates by means of indirect calorimetry.
Results: A total of 197 infants, including 183 premature infants and
14 full-term neonates, were recruited for the study. REE values
increased in all gestational age groups from the first week to
5–6 wk of postnatal age, with the most pronounced increase in
the smallest infants (+140%) and the smallest increase in the full-
term neonates (+47%). Univariate calculations showed that for each
postnatal week, REE increased by 6.93–9.64 kcal
kg21
d21 with
each additional kcal administered, for an average increase of
0.701 kcal, and increased by 1.78 kcal for each 1 g gain in weight.
Postnatal age was the strongest predictor to influence REE (r2 =
0.727, P , 0.0001).
Conclusions: This study provides comprehensive data on longitu-
dinally determined REE values of healthy premature and full-
term infants. Results may serve as a basis for comparative studies
that address various disease states as well as different nutritional
protocols. Am J Clin Nutr 2009;90:1517–24.
INTRODUCTION
One of the goals of neonatal nutritional practices is to provide
adequate nutrition to achieve anthropometric indexes and body
composition similar to those of the normal fetus in utero of the
same gestational age. However, providing an adequate energy
supply to extremely immature infants exposed to many different
conditions that can lead to considerable variability in energy
expenditure (EE) is difficult. The key issue in nutritional man-
agement is the understanding of energy requirements, because
premature infants are not comparable with fetuses, because they
are in a completely different environment. Energy expended by
term and preterm infants is influenced by infant maturity, growth,
calorie intake, physical activity, and several diseases (1–7). If
energy intake is insufficient for EE and energy losses, growth may
be restricted. It has been shown that insufficient energy intakes by
preterm infants during early postnatal life results in an energy
deficit as high as 800 kcal/kg after a period of 5 wk (2). Moreover, it
Am J Clin Nutr 2009;90:1517–24. Printed in USA. Ó 2009 American Society for Nutrition
Downloaded from https://academic.oup.com/ajcn/article-abstract/90/6/1517/4598078/ by guest on 06 October 2019
is well known that periods of nutritional deprivation in early in-
fancy, especially of protein, lead to permanent defects that cannot be
subsequently repaired (1, 2). Recent investigations of the effect of
early postnatal nutrition indicated that the rate of weight gain in
premature infants is influenced by the amount of calories given,
whereas gains in length and head circumferences are influenced by
the amount of protein in the diet (1).
During the first weeks after birth, EE increases in preterm and
full-term neonates. This has been explained by increasing energy
intake and physical activity (1). However, most of the previously
reported data in preterm infants are cross-sectional and were
obtained by means of a variety of measurement techniques and in
different circumstances (8–13). Three longitudinal studies of EE
after birth were published in the 1980s (11–13), which included
3–15 infants at different gestational and postnatal ages. The
results of these studies varied because of different contributing
factors that influence EE. Recently, 2 small longitudinal studies
of resting EE (REE) in preterm infants with a birth weight
,1500 g (14) or ,1000 g (15) during the first 3 and 5 wk,
respectively, were published. However, longitudinal studies of
“normal” values of REE in a large cohort of healthy, growing,
and stable preterm infants and full-term neonates have not been
found in the literature. In addition, using tables and charts of
nutrient utilization of preterm infants that have been estimated
rather than measured is of limited use for designing balanced
nutrition protocols. Moreover, estimated data are only realistic
under the assumption that normal in utero fetal growth repre-
sents ideal growth of the preterm infant. The present in-
vestigation was designed to investigate and to provide reference
values of REE in preterm and in full-term infants longitudinally
during the first 5–6 wk after birth.
SUBJECTS AND METHODS
The eligible study population consisted of 197 stable, healthy,
and growing neonates. Infants were divided into 4 groups
according to their gestational ages: 1) 75 preterm infants with
1
From the Departments of Pediatrics (JB and CW) and Medical Informat-
ics and Biomathematics (JG), University Children’s Hospital of Muenster,
University of Muenster, Muenster, Germany.
2
Address correspondence to J Bauer, University Children’s Hospital of
Muenster, Department of Pediatrics, Albert-Schweitzer Strasse 33, 48149
Muenster, Germany. E-mail: jacqueline.bauer@ukmuenster.de.
Received June 29, 2009. Accepted for publication September 12, 2009.
First published online October 7, 2009; doi: 10.3945/ajcn.2009.28304.
1517
1518 BAUER ET AL
a gestational age of 26 to 28 wk, 2) 49 infants with a gestational
age of 29 to 32 wk, 3) 59 infants with a gestational age of 33 to
35 wk, and 4) 14 full-term infants with a gestational age of 38 to
41 wk. The last group of infants had been admitted to the hos-
pital because of maternal indications (eg, adoption or social
problems) after birth. Studies after discharge were done as part
of weekly outpatient clinic visits. The Ethical Committees of the
Medical Faculties of Muenster and Heidelberg have approved
the investigation and parental consent was obtained before
studying each individual infant.
All infants were spontaneously breathing and free of major
congenital malformations, septicemia, and pulmonary or circu-
latory compromises and did not receive supplemental oxygen
during the entire study period. Because infants with intrauterine
growth retardation may have increased EE (16), only infants
appropriate for gestational age (10th to 90th birth weight per-
centile) were included (17). Infants with hyperirritability or
lethargy and infants of diabetic mothers were excluded, because
EE increases with increasing physical activity (9) and may be less
in infants with reduced muscle tone (7). All infants with a ges-
tational age ,28 wk were treated with caffeine for apnea until
they reached 34 wk of gestational plus postnatal age by using
a loading dose of 10 mg
kg21
d21 and a single maintenance
dose of 5 mg
kg21
d21 during the entire study period. The
dosage was adjusted according to weight gain to maintain
a dosage of 5 mg
kg21
d21 during the study period. Preterm
infants born after 28 wk and treated with caffeine were excluded
from the study. To prevent interindividual variability of caf-
feine’s influence on EE, all infants received the medication
between 0900 and 1000. No other stimulating drugs, cortico-
steroids, or antibiotics that could alter EE were given during the
entire study period.
Monitoring and behavioral state
All infants were monitored continuously with a cardiorespi-
ratory monitor (model Vicom-smu; Marquette-Hellige GmbH,
Freiburg, Germany). Heart rate, respiratory rate, and oxygen
saturation were monitored continuously, and event data (apneas,
tachycardias, bradycardias, periodic breathing, and oxygen sat-
uration) were recorded for 24 h on a computer. Skin (lower leg)
and rectal temperatures were measured continuously for 2 h
before, during, and after indirect calorimetry. Room humidity
(ranging from 35% to 45%) and temperature (ranging from 24 to
26°C) were kept constant and recorded during the observation
periods. Indirect calorimetry was performed in neonates with
a body weight ,1500 g in an air temperature–controlled in-
cubator (Hill Room Air Shields; Heinen & Löwenstein, Bad
Ems, Germany). Incubator temperature was adjusted to maintain
the rectal temperature between 36.6 and 37.5°C and the pe-
ripheral skin temperature (at the sole of the right foot) between
34.5 and 35.8°C (18, 19). Depending on the postnatal age of the
infant, incubator humidity ranged from 80% in the first week to
65% thereafter. In more mature infants with a body weight
.1500 g, who could maintain body temperature in a conven-
tional warming bed, REE was recorded in an open warming bed
providing conductive heat by using a heated gel mattress
(Babytherm 8000; Draeger Medical AG, Luebeck, Germany).
The warming bed temperature was adjusted to achieve the
abovementioned rectal and skin temperatures. All infants were
treated in the same type of incubator or bed. After the second
week of life, term neonates were studied in their regular beds,
which included skin and rectal temperature monitoring.
The behavioral states of all infants were recorded throughout
the measurement period, based on the modified Freymond Be-
havioral State Scale (20). Four different behavioral states were
distinguished: 0 (eyes open or closed, regular respiration, no
movements), 1 (small movements), 2 (vigorous movements), and
3 (crying). All the infants were studied only during the sleep
(state 0–1). In infants who showed periods of activity states .1
during the measurements, calorimetry was stopped and the in-
vestigation was repeated 4 h later or on the next day.
Downloaded from https://academic.oup.com/ajcn/article-abstract/90/6/1517/4598078/ by guest on 06 October 2019
Nutrition and weight
During the entire study period, all full-term neonates were fed
every 4 h with the same term formula milk. The formula (Milupa
GmbH, Friedrichsdorf, Germany) contained 1.8 g protein, 8.1 g
carbohydrate, 3.7 g fat, and 72 kcal/100 mL. To rule out in-
terindividual differences in energy intake in infants fed mother’s
milk, breastfed infants were not included in this study. Weight
gain was calculated as the mean value from the consecutive days
around the calorimetric measurements.
In the preterm infants, parenteral nutrition was started im-
mediately after birth with glucose, amino acids, lipids, and
electrolytes. Gastric nutrition was initiated with small amounts of
preterm formula on the first day of birth. In the first study week
(3–5 d after birth), infants received 20–50% of the total calorie
intake parenterally and 50–80% enterally via a gastric tube. From
the second week on, all infants were exclusively fed preterm
formula. All preterm infants received the same formula, which
was given every 2 or 4 h via gastric tubes as bolus or orally. The
formula (Milupa GmbH, Friedrichsdorf, Germany) contained
2.4 g protein, 7.8 g carbohydrate, 4.4 g fat, and 80 kcal/100 mL.
When the preterm infants reached a corrected gestational age of
36 wk, preterm formula was replaced by the formula for full-term
infants within 4 d.
Body weight was measured daily at 0800 with a calibrated
electronic scale with a resolution of 1 g. To estimate the energy
content of the formula, the Atwater system was used (protein =
4 kcal/g, carbohydrate = 4 kcal/g, fat = 9 kcal/g).
Indirect calorimetry
Measurements of maximal oxygen uptake (VO _ 2) and maximal
carbon dioxide uptake (VCO_ 2 ) were performed with a portable
open-circuit continuous indirect calorimetry device designed for
neonates (Deltatrac II Metabolic Monitor; Datex-Ohmeda, In-
strumentarium Corp, Helsinki, Finland), which calculated REE.
Details of the technique, the precision of the device, and its
validation for indirect calorimetry in neonates and preterm in-
fants were as previously described by our group and others (3–5,
21, 22). The accuracy of the device was tested as a mean ex-
_ 2 measurements of 2 6 2% (22). This
perimental error for VO
transportable device consists of a fast differential paramagnetic
oxygen sensor and an infrared carbon dioxide sensor attached to
a silicon face mask or to a transparent hood that is continuously
ventilated by a constant-flow generator. The face mask, with
a capacity of 40 mL, covered the infant’s mouth and nose
(Laerdal Medical, Graefelfing, Germany), and VO _ 2 and VCO
_ 2
 ENERGY EXPENDITURE IN PRETERM INFANTS
measurements were accurate at a flow of 3 L/min (error: 21 6
0.8%), and a tidal volume under 15 mL/breath as recently de-
scribed and tested by Bauer et al (22). The device does not
measure inspiratory and expiratory oxygen concentrations sep-
arately, but measures 0.01% (vol) O2 differences as a 1-min
average very accurately. The device was calibrated before each
measurement with a standard calibration gas (5% O2 and 95%
CO2). The analyzer was set at zero according to room air. Cali-
bration gases were prepared to an accuracy of 60.03% and
certified gravimetrically. A silicon face mask was used for the
preterm infants, and a transparent hood was used for the term
neonates. Marks et al (23) and Bauer et al (22) showed that a face
mask gives more reliable results in very-low-birth weight infants
(,1500 g) than does a hood and offers the advantage of ready
access to the infant. The Deltatrac II monitor stores each minute-
to-minute VO_ 2 and VCO
_ 2 values and REE electronically. At the end
of the measurements, the values were transmitted to a personal
computer and processed by using SAS for Windows (SAS In-
stitute, Cary, NC). The nonprotein EE was calculated from VO _ 2
_
and VCO 2 as EE = 5.50 _
VO 2 +1.76 _
VCO 2 (in kcal
kg21

d21
) (24).
Study protocol
According to the recommendations for neonatal calorimetry
(1, 25), REE was defined as EE at rest or during sleep in ther-
moneutrality, 1 h after the previous feeding. It incorporates the
basal metabolic rate, which is difficult to measure in preterm
infants and neonates because they cannot be fasted long enough
and because of additional EE arising from postprandial metab-
olism that still lingers after the last feed.
If inclusion criteria were fulfilled, measurements of REE
began in preterm and in full-term neonates on days 3–5 of
postnatal life. Indirect calorimetry was started at 1500–1600 and
lasted for ’6 h continuously per day in each infant, except for
the period of breastfeeding and feeding. Because EE is influ-
enced strongly by feeding (25), each period of indirect calo-
rimetry began 1 h after feeding, including an equilibration time
of 15 min, as recommended by the manufacturer. Therefore,
each period of calorimetry began 60 min after feeding and lasted
for 60, 120, or 180 min depending on the feeding intervals to
minimize the effects of postprandial thermogenesis on the re-
sults. Several periods of measurements were conducted sub-
sequently in each infant to obtain a total of 6 h of calorimetry
data. The measurement was only interrupted for breastfeeding
and feeding procedures. Indirect calorimetry was repeated in the
same infants under identical conditions every week, over a 5-wk
period in the full-term neonates and over a 6-wk period in the
preterm infants.
Statistical analysis
Data were analyzed by using SPSS statistical software (version
17.0 for Windows; SPSS Inc, Chicago, IL) and SAS version
9.2 for Windows (SAS Institute Inc). The results are given as
means 6 SDs. Linear regression and analysis of covariance on
the whole set of data were used to determine the significant
predictors of REE. Mixed linear models were established, ie,
random subjects effects were included in all models to account
for intrasubject correlation of repeated measurements over time.
P values ,0.05 were considered statistically significant.
1519
RESULTS
A total of 197 consecutive infants, including 183 preterm
infants and 14 full-term neonates were eligible for the study.
Because only healthy preterm and full-term neonates were se-
lected, the infants were recruited over a period of 7 y by the
principal investigator (JB). Seven healthy full-term neonates and
8 preterm infants from the total infants with a gestational age of
26 to 28 wk were included in our previous publications (5, 15). A
total of 28 infants were excluded during the recruitment period
because they needed antibiotics for sepsis treatment (n = 12),
they needed other drugs that could alter EE measurements (n = 6),
they required surgery (n = 4), they were transferred to other
Downloaded from https://academic.oup.com/ajcn/article-abstract/90/6/1517/4598078/ by guest on 06 October 2019
hospitals (n = 4), and because their parents did not consent to their
participation in the study (n = 2). The excluded preterm infants
and full-term neonates did not differ from the study population
with regard to gestational age and birth weight (data not shown).
All 197 infants studied spent the entire time of calorimetry in
states 0–1, which means that 80–85% of the measuring time
was recorded in sleep periods with low activity, and REE was
90–95% of EE.
In the premature infants with a gestational age of 26 to 28 wk,
REE values increased 2.4-fold during the 5-wk observation
period (Table 1). Energy intake was always higher than REE
and reached a maximum of 123 kcal
kg21
d21 at 5 wk, and no
energy deficits were detected at any study time. In the more
mature infants with a gestational age of 29 to 32 wk, REE in-
creased by 77% from weeks 1 to 5 and by 60% from weeks 1 to
4 in the infants with a gestational age of 33 to 35 wk. Full-term
neonates showed a moderate increase in REE of 47% from
weeks 1 to 4 (Tables 2–4). The increases in REE from one study
week to the next were significant for all study periods and
gestational age groups (P , 0.0001 for all comparisons). The
relation between REE and energy intake in preterm infants of all
gestational ages is shown in Figures 1–3.
We conducted linear regression and analysis of covariance
analyses to study the influence of energy intake, postnatal age,
and body weight gain as independent variables on REE. In
a univariate analysis the regression of REE by energy intake
showed a close relation (r2 = 0.606, P , 0.0001). We calculated
that, for each additional kcal given, REE increased by 0.701 kcal
(95% CI: 0.679, 0.724), ie, 70% of the energy intake was ex-
pended and 30% was available for growth. The figures were
qualitatively identical at all gestational ages. REE was also
strongly related to postnatal age (r2 = 0.727, P , 0.0001). By
analyzing the regression of REE by postnatal age and adjusted
by gestational age, we found that, for each postnatal week, REE
increased by 9.638 kcal (95% CI: 9.369, 9.907) in the infants
with a gestational age of 26 to 28 wk, by 6.934 kcal (95%
CI: 6.601, 7.267) in the infants with a gestational age of 29 to
32 wk, and by 7.556 kcal (95% CI: 7.154, 7.958) in the more
mature infants with a gestational age of 33–35 wk.
In univariate analyses, a significant correlation between REE
and weight gain was found (r2 = 0.338, P , 0.0001). Regression
analyses of REE by weight gain showed that a gain of 1 g in body
weight resulted in a rise in REE of 1.78 kcal (95% CI: 1.658, 1.901).
DISCUSSION
One of the most remarkable findings in our investigation was
the lower REE values (range: 23–56 kcal
kg21
d21) obtained
1520 BAUER ET AL
TABLE 1
Resting energy expenditure (REE), energy intake, and weight gain in 75 preterm infants with a gestational age of 26 to 28
wk (n = 75)1
Postnatal age (wk)

1 2 3 4 5 6

REE (kcal
kg21
d21)

Mean 6 SD 33 6 6 39 6 8 49 6 8 57 6 10 70 6 8 80 6 8
Minimum 23 26 31 34 56 63
Maximum 56 60 67 74 89 98
Energy intake (kcal
kg21
d21)
Mean 6 SD 66 6 8 73 6 8 86 6 11 104 6 15 112 6 13 123 6 5
Minimum 51 59 72 80 89 110

Downloaded from https://academic.oup.com/ajcn/article-abstract/90/6/1517/4598078/ by guest on 06 October 2019

Maximum 89 95 136 145 139 144
Weight gain (g
kg21
d21)
Mean 6 SD 26.1 6 3.9 8.9 6 2.7 10.5 6 2.6 12.2 6 2.8 14.0 6 2.1 14.7 6 2.8
Minimum 215 2 5 6 9 6
Maximum 9 19 19 19 18 20
1
Conversion factor: 1 kcal = 4.184 kJ. The increases in REE from one study week to the next were significant for all
study periods and gestational age groups (P , 0.0001 for all comparisons).

in the micropremies (gestational age: 26–28 wk) during the first
week after birth. These values are similar to the data provided by
Bauer et al (14) of from 27 to 35 kcal
kg21
d21 in the first
week of life in stable preterm infants with a similar gestational
age. We confirmed our own previously published data in a small
number of premature infants (birth weight ,1000 g), which
showed lower REE values in the first week after birth (15). Other
comparable REE data of stable and healthy micropremies
shortly after birth have not been found in the literature. These
values may be explained by the inclusion of selected subjects in
both our studies without respiratory complications and major
postnatal morbidities. Other investigators studied premature in-
fants with severe respiratory disease or mechanical ventilation
and reported higher REE values (10–15%), which are attrib-
uted to the increased oxygen consumption of injured lungs and
consequently showed elevated REE values compared with our
group. Our immature infants were nursed in double-walled in-
cubators to minimize heat loss, whereas some of the previous
investigators used open incubators, thereby increasing the heat
loss significantly. Few longitudinal studies of REE have been
conducted in extremely premature infants (,30 wk of gesta-
tion); most investigations have been conducted in considerably
more mature infants who received more calories and presented
greater physical activities, which have shown that REE increases
with advancing postnatal age. Additionally, early provision of
a higher energy supply in other studies may have increased REE
in their infants (13). We found a close correlation between REE
and energy intake in accordance with previous trials (11, 13–15).
In the smallest infants (26–28 wk) in our present study, REE
values increased by 140% during the observation time, by 77%
in the more mature infants (29–32 wk), and by 60% in the ne-
onates of advanced gestational age (33–35 wk). These results are
similar to our own previous report and to 2 other studies (14, 15)
that included a very small number of subjects but that studied
healthy and growing immature neonates during the first weeks
after birth. However, the present data show a major difference

TABLE 2
Resting energy expenditure (REE), energy intake, and body weight gain in 49 preterm infants with a gestational age of 29
to 32 wk (n = 49)1
Postnatal age (wk)

1 2 3 4 5 6
21 21
REE (kcal
kg
d )
Mean 6 SD 46 6 8 53 6 10 59 6 11 65 6 11 72 6 9 81 6 11
Minimum 31 32 40 47 59 64
Maximum 67 78 88 90 96 115
Energy intake (kcal
kg21
d21)
Mean 6 SD 78 6 9 91 6 10 98 6 9 112 6 10 117 6 6 122 6 5
Minimum 64 78 80 96 102 110
Maximum 96 108 115 125 126 136
Weight gain (g
kg21
d21)
Mean 6 SD 7.5 6 1.1 10.5 6 1.1 12.2 6 1.6 13.9 6 1.5 14.7 6 1.3 17 6 1.7
Minimum 4 9 9 10 11 14
Maximum 10 13 15 18 17 22
1
Conversion factor: 1 kcal = 4.184 kJ. The increases in REE from one study week to the next were significant for all
study periods and gestational age groups (P , 0.0001 for all comparisons).
 ENERGY EXPENDITURE IN PRETERM INFANTS 1521
TABLE 3
Resting energy expenditure (REE), energy intake, and weight gain in 59 preterm infants with a gestational age of 33 to 35
wk (n = 59)1
Postnatal age (wk)

1 2 3 4 5

REE (kcal
kg21
d21)

Mean 6 SD 48 6 6 52 6 8 60 6 9 69 6 9 77 6 8
Minimum 39 40 50 54 60
Maximum 65 78 89 98 102
Energy intake (kcal
kg21
d21)
Mean 6 SD 81 6 8 92 6 9 102 6 9 113 6 7 122 6 3
Minimum 66 78 89 99 115

Downloaded from https://academic.oup.com/ajcn/article-abstract/90/6/1517/4598078/ by guest on 06 October 2019

Maximum 99 118 120 126 129
Weight gain (g
kg21
d21)
Mean 6 SD 8.2 6 2.5 11.3 6 2.7 12.5 6 2.1 13.8 6 2.2 15.5 6 2.8
Minimum 1 6 6 9 10
Maximum 15 17 17 19 21
1
Conversion factor: 1 kcal = 4.184 kJ. The increases in REE from one study week to the next were significant for all
study periods and gestational age groups (P , 0.0001 for all comparisons).

between the minimum and maximum REE values. We thought
that the current values expose or reflect probably more realistic
levels of energy expended in this population without any alter-
ations related to illness, thermal environment, or medications.
Caffeine was the only medication with a potential effect on
metabolic rate that was given to all infants during the study
period. Previously, we showed that long-term administration of
caffeine was associated with an increase in REE of 21 kcal

kg21
d21 after 48 h of treatment compared with pretreatment
levels (4). Other researchers found a 19% higher REE in infants
who received long-term caffeine than in nontreated infants (14).
In this study there was no information about the time span be-
tween the first administration of caffeine and the first calori-
metric measurement. The REE values in this study in the first
week after birth were similarly low compared with our values.
We hypothesize that the lower REE values acquired in our
smallest neonates shortly after birth were not altered by caffeine
therapy, because calorimetry was conducted in almost all infants
5 h, on average, after the first caffeine dose. Nevertheless, long-
term metabolic stimulation associated with caffeine therapy
should be considered in evaluations of REE in preterm infants
treated with this medication.
We found that the difference between energy intake and REE
was positive during the entire observation time and reached
40 kcal
kg21
d21 in the first week of postnatal age in all
groups. These values were more rapidly achieved in the full-
term infants (second week) than in the premature infants (third
week). Assuming that ’10% of the energy intake is lost via
feces and urine (1, 26) and that physical activity accounts for
losses of 3% in premature infants (9) and of 5% in term infants
(27, 28), ’35 and 34 kcal
kg21
d21 of the energy intake was
available for growth. This agrees with the findings of a multiple
regression analysis performed in preterm infants weighing
,1000 g studied during the first 6 wk of postnatal life (15).
Daily weight gains in the fourth week of postnatal life were 12,
14, 14, and 12 g
kg21
d21 in the infants with gestational ages

TABLE 4
Resting energy expenditure (REE), energy intake, and weight gain in 14 full-term neonates with a gestational age of 38 to
41 wk (n = 14)1
Postnatal age (wk)

1 2 3 4 5
21 21
REE (kcal
kg
d )
Mean 6 SD 42 6 4 51 6 5 59 6 6 60 6 8 62 6 9
Minimum 36 43 50 47 48
Maximum 48 59 67 70 73
Energy intake (kcal
kg21
d21)
Mean 6 SD 57 6 5 87 6 12 99 6 12 104 6 11 112 6 8
Minimum 50 59 74 87 96
Maximum 68 104 117 122 125
Weight gain (g
kg21
d21)
Mean 6 SD 212 6 10 11 6 12 11 6 11 12 6 16 11 6 12
Minimum 214 8 10 9 12
Maximum 10 13 15 17 23
1
Conversion factor: 1 kcal = 4.184 kJ. The increases in REE from one study week to the next were significant for all
study periods and gestational age groups (P , 0.0001 for all comparisons).
1522 BAUER ET AL

Downloaded from https://academic.oup.com/ajcn/article-abstract/90/6/1517/4598078/ by guest on 06 October 2019

FIGURE 1. Relation between resting energy expenditure and energy intake in 75 preterm infants with a gestational age of 26 to 28 wk. The lines represent
linear regression at different postnatal ages. The total regression line corresponds to the entire data across all postnatal ages.

of 26 to 28 wk, 29 to 32 wk, 33 to 35 wk, and 38 to 41 wk,
respectively (Tables 1–4). Assuming a daily energy storage of
’35 kcal
kg21
d21 in premature infants with gestational ages
of 26 to 32 wk and 34 kcal
kg21
d21d in more mature infants
with gestational ages of 33 to 41 wk, energy storages of 2.9, 2.5,
3.5, and 2.8 kcal, respectively, per 1 g of newly formed tissue
can be calculated for the 4 groups. Because fat accretion in
newly formed tissue contains more energy than protein, high-
energy storage per 1 g of weight gain indicates that weight gain
is associated with high fat accretion. In the fetus, energy storage
increases from 0.6 kcal/g at 24 to 26 wk of gestational age to
2.6 kcal/g at 39 to 40 wk of gestational age (29). Preterm infants
frequently have higher levels of energy storage of up to 4 kcal/g.
Reichman et al (30) calculated that the fat content of new tissue
in the fetus is 11%, whereas energy storage of 4 kcal/g indicates
that new tissue contains 32% fat, which is far above the normal
level of fetal fat accretion (29). In full-term infants, newly
formed tissue contains 41% fat (31). This corresponds to an
energy storage of 2.8 kcal/g in full-term neonates per 1 g of
weight gain in our study. The consequences of the higher fat
accretion in the preterm infant compared with adipose tissue
deposition during the fetal period are unknown with respect to
subsequent growth, development, and effect on lipid metabolism
and development of metabolic syndrome as obesity and diabetes
in later life. A recent study showed that preterm infants have
a highly significant decrease in subcutaneous adipose tissue and
a significant increase in intraabdominal adipose tissue (32). In
adults, intraabdominal or visceral adiposity is associated with

FIGURE 2. Relation between resting energy expenditure and energy intake in 49 preterm infants with a gestational age of 29 to 32 wk. The lines represent
linear regression at different postnatal ages. The total regression line corresponds to the entire data across all postnatal ages.
 ENERGY EXPENDITURE IN PRETERM INFANTS
FIGURE 3. Relation between resting energy expenditure and energy intake in 59 preterm infants with a gestational age of 33 to 35 wk. The lines represent
linear regression at different postnatal ages. The total regression line corresponds to the entire data across all postnatal ages.
insulin resistance, even in lean individuals and in those with
dyslipidemia (33). Inadequate nutritional intakes can lead to
excessively rapid growth of body fat stores during early life,
which can result in obesity, insulin resistance, and the de-
velopment of the metabolic syndrome in later life.
A potential limitation of our study was the long period of
subject recruitment due to the careful selection of a large group of
healthy and growing premature infants and to the longitudinal
design of the investigation. To minimize possible confounding
variables associated with a long recruitment period, all partic-
ipants received the same formula and almost all infants were
measured by one principal investigator (JB) throughout the ob-
servation period. The nutritional regimen and study protocol were
strictly followed over the years.
We concluded that, in preterm and full-term infants, REE
increases during the first weeks of life with rising energy intakes,
whereas a positive energy balance could be achieved from the first
week of life with our nutritional regimen. The difference between
energy intake and REE reaches ’40 kcal
kg21
d21 at 3 to
6 wk of postnatal age, independent of gestational age, without
the inclusion of energy lost via feces and urine. Consequently,
energy disposition was much lower. Our longitudinal study
provides REE reference values from a relatively large population
of healthy growing preterm infants and full-term neonates
measured under standardized conditions and extending over the
first weeks after birth. This approach helps to predict EE from
energy intake with acceptable accuracy and is very helpful in
determining not only the energy needs of the infants but also
their response to consistent nutritional protocols. Nevertheless,
future investigations of nutrient requirements, energy balance,
and body composition should focus on infants with more vari-
able energy metabolism, such as in infants with intrauterine
growth restriction and or with severe postnatal diseases.
We gratefully acknowledge the contributions of our infants and their fam-
ilies and those of our nurses and colleagues.
The authors’ responsibilities were as follows—JB: responsible for the over-
all study design, performed the measurements, wrote the manuscript, and was
involved in patient recruitment and data collection and analysis; JG: performed
1523
Downloaded from https://academic.oup.com/ajcn/article-abstract/90/6/1517/4598078/ by guest on 06 October 2019
the statistical analysis of the data and participated in the discussion of the
results and in the correction of the manuscript; CW and JB: participated in
the correction of the manuscript; and CW: provided clinical support and assis-
ted in the clinical experiments. None of the authors had a conflict of interest.
REFERENCES
1. Sauer PJ. Neonatal energy metabolism. In: Richard CM, ed. Principles
of perinatal-neonatal metabolism. Heidelberg, Germany: Springer-Verlag,
1997:583–1043.
2. Cooke RJ, Embleton ND. Feeding issues in preterm infants. Arch Dis
Child Fetal Neonatal Ed 2000;83:F215–8.
3. Bauer J, Maier K, Muehlbauer B, Poeschl J, Linderkamp O. Energy
expenditure and plasma catecholamines in preterm infants with mild
chronic lung disease. Early Hum Dev 2003;72:147–57.
4. Bauer J, Maier K, Linderkamp O, Hentschel R. Effect of caffeine on
oxygen consumption and metabolic rate in very low birth weight infants
with idiopathic apnea. Pediatrics 2001;107:660–3.
5. Bauer J, Hentschel R, Linderkamp O. Effect of sepsis syndrome on
neonatal oxygen consumption and energy expenditure. Pediatrics 2002;
110:e69.
6. Leitch CA, Karn CA, Ensing GJ, Denne SC. Energy expenditure after
surgical repair in children with cyanotic congenital heart disease.
J Pediatr 2000;137:381–5.
7. Bauer J, Teufel U, Doege C, Gausepohl HJ, Beedgen B, Linderkamp O.
Energy expenditure in neonates with Down syndrome. J Pediatr 2003;
143:264–6.
8. DeMarie MP, Hoffenberg A, Biggerstaff SL, Jeffers BW, Hay WW Jr,
Thureen PJ. Determinants of energy expenditure in ventilated preterm
infants. J Perinat Med 1999;27:465–72.
9. Thureen PJ, Phillips RE, Baron KA, DeMarie MP, Hay WW Jr. Direct
measurement of the energy expenditure of physical activity in preterm
infants. J Appl Physiol 1998;85:223–30.
10. Leitch CA, Denne SC. Energy expenditure in extremely low-birth
weight infants. Clin Perinatol 2000;27:181–95.
11. Chessex P, Reichman BL, Verellen GJ, et al. Influence of postnatal age,
energy intake and weight gain on energy metabolism in the very-low-birth
weight infant. J Pediatr 1981;99:761–6.
12. Sauer PJ, Dane HJ, Visser HK. Longitudinal studies on metabolic rate,
heat loss, and energy cost of growth in low birth weight infants. Pediatr
Res 1984;18:254–9.
13. Gudinchet F, Schutz Y, Micheli JL, Stettler E, Jequier E. Metabolic cost
of growth in very low-birth-weight infants. Pediatr Res 1982;16:1025–30.
14. Bauer K, Laurenz M, Ketteler M, Versmold H. Longitudinal study of
energy expenditure in preterm neonates ,30 weeks of gestation during
the first three postnatal weeks. J Pediatr 2003;142:390–6.
1524 BAUER ET AL
15. Bauer J, Maier K, Hellstern G, Linderkamp O. Longitudinal evaluation
of energy expenditure in preterm infants with birth weight below 1000
grams. Br J Nutr 2003;89:533–7.
16. Bohler T, Kramer T, Janecke AR, Hoffmann GF, Linderkamp O. In-
creased energy expenditure and faecal fat excretion do not impair weight
gain in small-for-gestational-age preterm infants. Early Hum Dev 1999;
54:223–34.
17. Monset-Couchard M, de Bethmann O, Relier JP. Long term outcome of
small versus appropriate size for gestational age co-twins/triplets. Arch
Dis Child Fetal Neonatal Ed 2004;89:F310–4.
18. Bell EF, Gray JC, Weinstein MR, Oh W. The effects of thermal envi-
ronment on heat balance and insensible water loss in low-birth-weight
infants. J Pediatr 1980;96:452–9.
19. Malin SW, Baumgart S. Optimal thermal management for low birth
weight infants nursed under high-powered radiant warmers. Pediatrics
1987;79:47–54.
20. Freymond D, Schutz Y, Decombaz J, Micheli JL, Jequier E. Energy
balance, physical activity, and thermogenic effect of feeding in pre-
mature infants. Pediatr Res 1986;20:638–45.
21. Bauer J, Sontheimer D, Fischer CH, Linderkamp O. Metabolic rate and
energy balance in very low birth weight infants during kangaroo holding
by their mothers and fathers. J Pediatr 1996;129:608–11.
22. Bauer K, Pasel K, Uhrug C, Sperling P, Versmold H. Comparison of face
mask, head hood, and canopy for breath sampling in flow-through in-
direct calorimetry to measure oxygen consumption and carbon dioxide
production of preterm infants ,1500 grams. Pediatr Res 1997;41:139–
44.
23. Marks KH, Coen P, Kerrigan JR, Francalancia NA, Nardis EE, Snider
MT. The accuracy and precision of an open-circuit system to measure
oxygen consumption and carbon dioxide production in neonates. Pediatr
Res 1987;21:58–65.
24. Weir JB. New methods for calculating metabolic rate with special ref-
erence to protein metabolism. J Physiol 1949;109:1–9.
25. Stothers JK, Warner RM. Effect of feeding on neonatal oxygen con-
sumption. Arch Dis Child 1979;54:415–20.
26. Heim T. Energy and lipid requirements of the fetus and the preterm
infant. J Pediatr Gastroenterol Nutr 1983;2(suppl.1):S16–41.
27. Davies PSW, Clough H, Bishop NJ, Lucas A, Cole JJ, Cole TJ. Total
energy expenditure in small for gestational age infants. Arch Dis Child
Fetal Neonatal Ed 1996;75:F46–8.
28. Wells JCK, Davies PSW. Estimation of the energy cost of physical ac-
tivity in infancy. Arch Dis Child 1998;78:131–6.
Downloaded from https://academic.oup.com/ajcn/article-abstract/90/6/1517/4598078/ by guest on 06 October 2019
29. Ziegler EE, O’Donnell AM, Nelson SE, Fomon SJ. Body composition of
the reference fetus. Growth 1976;40:329–41.
30. Reichman B, Chessex P, Putet G, et al. Diet, fat accretion and growth in
premature infants. N Engl J Med 1981;305:1495–500.
31. Fomon SJ. Body composition of the male reference infant during the
first year of life. Pediatrics 1967;40:863–70.
32. Uthaya S, Thomas EL, Hamilton G, Doré CJ, Bell J, Modi N. Altered
adiposity after extremely preterm birth. Pediatr Res 2005;57:211–5.
33. Wajchenberg BL, Giannella-Neto D, da Silva ME, Santos RF. Depot-
specific hormonal characteristics of subcutaneous and visceral adipose
tissue and their relation to the metabolic syndrome. Horm Metab Res
2002;34:616–21.