Acta Anaesthesiol Scand 1993: 37: 170-1 75

Metabolic consequences of different perioperative

fluid therapies in the neonatal period
K. NILSSON,
K. SANDSTROM, S. ANDR~ASSON, and L. E. LARSSON
A. NIKLASSON’
Department of Paediatric Anaesthesia and Intensive Care and ‘Department of Paediatrics, &tra Hospital, Gothenburg, Sweden

Carbohydrate and fat metabolism during and after anaesthesia and surgery was studied in 14 neonates with
major congenital non-cardiac anomalies. They were either given a glucose solution until surgery or starved
for at least 4 h before surgery. Ringer-acetate alone or Ringer-acetate plus 10% glucose was used for the
intraoperative fluid therapy. After anaesthesia all neonates were given a 10% glucose solution. Concentrations
of glucose, free fatty acids, triglycerides, lactate, pyruvate, alanine, glycerol and 3-hydroxybutyrate were
measured at predetermined intervals pre-, intra- and postoperatively. Blood glucose concentrations rose
during surgery both in neonates given glucose before and during surgery ( n = 6 ) and in neonates not
given glucose before and during surgery ( n = 6 ) . Increased intraoperative levels of free fatty acids and 3-
hydroxybutyrate were found in neonates not given glucose before and during surgery. The triglyceride levels
were equal in both groups. In two neonates given glucose before surgery and Ringer-acetate during surgery
increased levels of 3-hydroxybutyrate were found, particularly in one patient who became hypoglycaemic.
In conclusion, starved neonates without intraoperative glucose supply mobilized fat and maintained blood
glucose concentrations.
Received 26 March 1991, accebted f o r publication in a revised version 4 July 1992
Key words: Anesthesia; fluid therapy; free fatty acids; 3-hydroxybutyrate; metabolism; neonates.

Glucose-containing solutions are usually recommended
during anaesthesia in the neonatal period (1). The rea-
sons for this are the low stores of glycogen and the im-
paired gluconeogenesisin the neonate (2). I n a previous
study blood glucose concentrations in neonates during
and after surgery were found to be related to both the
pre- and intraoperative glucose infusion (3). Blood glu-
cose concentrations were mostly stable when glucose
was given before and during surgery. Preoperative glu-
cose followed by intraoperative Ringer-acetate resulted
in an increased risk for hypoglycaemia. O n the other
hand, hypoglycaemia was not found when glucose was
not given before and during surgery.
In normal healthy newborns an increasing pro-
portion of the energy requirement is derived from glu-
coneogenesis and fat mobilization (4).Our previous
study on blood glucose concentrations during neonatal
surgery showed maintained blood glucose levels in
starved neonates without intraoperative glucose
supply. The aim of this report is to describe the contri-
bu tion of fat mobilization and gluconeogenesis to the
energy homeostasis before, during and after surgery in
the neonatal period with and without peroperative
glucose infusions.
PATIENTS AND METHODS
The study was approved by the Ethics Committee of the Medical
Faculty, University of Goteborg. Informed consent was obtained
from the parents of all infants studied.
Fourteen neonates with major congenital non-cardiac defects were
studied in connection with operation within their first 144 h of life.
The basic layout and the blood glucose values of 1 1 of the 14 patients
have been presented earlier (3). These 11 patients constitute a sub-
group of this earlier randomized study (30 patients) on the influence
of fluid regimens on perioperative blood glucose concentrations.
Three new patients (nos. 6, 11 and 12, Table 1) were recruited, as
only blood glucose and blood gases were available for a majority of
the original patients. The mean weight was 3.3 kg (range 2.4-4.0
kg), mean gestational age was 38 weeks (range 36-42 weeks) and
mean age at operation was 39 h (range 6-144 h). The preoperative
nutrition was not manipulated for the study and followed the routines
of the referring hospital or physician. Five neonates from the ran-
domized blood glucose study plus one added neonate had had a
running infusion of 2 4 ml kg-’ h-’ 10% glucose (Glucos, ACO,
Sweden) for at least 3 h when presented for anaesthesia (Group 1).
These neonates received an intraoperative fluid therapy of 3 ml kg-’
h-’ of 10% glucose (0.3 g kg-l h-I) in addition to Ringer-acetate
(Ringerdex, Pharmacia, Sweden). Four neonates from the blood
glucose study plus two recruited neonates had been starved for at
least 4 h (Group 2). Before that, one had been given breast milk and
two had had a glucose infusion. This “starved” group received only
Ringer-acetate intraoperatively. Group 3 consisted of two neonates
who also had a glucose infusion running preoperatively, but were
randomized to receive a glucose-free intraoperative fluid supply. A
total infusion rate of 20 ml kg-’ was aimed at in all infants during
the first hour of surgery and after that the rate was reduced to 10 ml
’
kg- h-’. Colloids were given in addition, if indicated. Postopera-
tively, all infants had a 10% glucose solution with electrolytes at a
rate of 5&1OO ml kg-’ 24 h-’ (5-10 g kg-l 24 h-I) according to the
age at time of operation.
Anaesthesia was induced with thiopentone 5 mg kg-’ and main-
tained with fentanyl (incremental doses of 2 pg kg-’, maximum 10
pg kg-’) and nitrous oxide-oxygen as needed, judged by Sao, or
Spo,. Pancuronium or vecuronium 0.1 mg kg-’ was used for muscle
relaxation. All children were ventilated using a Mapleson-D system.
 PERIOPERATIVE FLUID THERAPY
After induction of anaesthesia, blood samples were taken from an
indwelling arterial cannula. Samples were taken immediately before
the start of the operation and repeated every 30 min until the oper-
ation was completed. Immediately upon arrival in the ICU, another
sample was taken (End) followed by samples taken at 1, 4 and 8 h
postoperatively. Samples were analysed for glucose and lactate in
blood and triglycerides (TG), free fatty acids (FFA), glycerol, lactate,
pyruvate, alanine and 3-hydroxybutyrate (3-OHB) in plasma. Blood-
gases and acid-base status were determined at every sampling oc-
casion.
Blood glucose was measured using a glucose-6-phosphate dehydro-
genase method (5). Plasma for TG, FFA, 3-OHB, alanine and glycer-
ol was separated by centrifugation and plasma fractions were por-
tioned and frozen separately. Blood for lactate analysis was frozen
after precipitation in 0.6 mol I-' perchloric acid. TG and FFA were
determined by gas-liquid chromatography (6). 3-OHB (7), pyruvate,
alanine (E), glycerol (9) and lactate (10) were analysed using enzy-
matic tests. Blood gases and acid-base status were analysed with a
Radiometer ABL 2 (Copenhagen, Denmark).
Two-way analysis of variance of the Wilcoxon signed rank test
was used for statistical differences in substrate concentrations before,
during and after surgery within each group. Student's btest was used
for differences between intraoperative and postoperative levels within
the two study groups where all intraoperative concentrations (includ-
ing End-concentrations) were pooled and compared to all pooled
postoperative concentrations. Bonferroni correction was used to ad-
just the P-value for multiple comparisons. A P-value<0.05% was
considered significant.
RESULTS
The mean values presented in the text, tables and
figures originate from the two main study groups
(Group 1, glucose and Group 2, starved Ringer-ace- + higher (P<0.05, t-test) compared to neonates given
tate). Some data of interest from Group 3, the two
neonates with glucose preoperatively and only Ringer-
acetate intraoperatively, are commented on in the text
but not used in the statistical comparisons of the study.
Weight, gestational age and age at operation did not
differ between the groups (Table 1). Except for the
171
preoperative fluid supply there were no statistically
significant differences in total fluid infusion rates be-
tween Groups 1 and 2. The total intraoperative in-
fusion rate (glucose+Ringer) was 17.7 k 3.5 ml kg-'
h-I in Group 1 and 15.3 f 5.8 m l . kg-]. h-' in Group
2. In Group 1 the mean preoperative glucose infusion
rate was 3.4k0.9 ml kg-I h-' and the mean intra-
operative glucose infusion rate was 2.9 _+ 0.2 ml kg-'
'.
h - The mean Ringer-acetate infusion rate during
anaesthesia in Group 1 was 14.8 & 3.6 ml kg-' h-I.
After surgery the mean glucose infusion rate was 3.6 f
1.O ml kg-' h-' in all neonates (identical mean infusion
rates in Groups 1 and 2 ) .
Blood glucose levels did not differ between the neo-
nates given glucose (Group 1) and the starved neonates
(Group 2). In both groups, blood glucose concen-
trations rose during surgery (P<O.Ol, Table 2 and
Table 3). Lactate and pyruvate tended to be higher,
although not statistically significantly different in neo-
nates given glucose compared to starved neonates. Ala-
nine and glycerol tended to be lower, although not
statistically significantly different, in neonates given
glucose compared to starved neonates. TG values
showed no difference between the neonates with differ-
ent fluid therapies during and after surgery. FFA did
not change during and after operation among the neo-
nates given glucose. In the starved neonates mean in-
traoperative concentrations of FFA were significantly
glucose. Furthermore, FFA increased from the pre-
operative concentrations to the last measured intra-
operative concentrations in the starved group
( P < 0.05, Wilcoxon's signed rank test). After surgery
the starved neonates showed a falling mean FFA con-
centration compared to the mean concentration during

Table 1
Patient data and fluid therapy.
Gestational Age at Preop. Intraop. Intraop.
Weight age surgery glucose glucose Ringer
Group Patient Diagnosis (kg) (weeks) (h) (ml kg-' h-')
1 1 Oesophageal atresia 2.8 37 24 3.6 2.5 19
2 Oesophageal atresia 2.7 36 13 2.6 3.0 19
3 Duodenal atresia 2.4 36 19 2.9 2.9 14
4 Pulmonary cyst 3.7 38 7 2.8 2.8 15
5 Hydrocephalus 3.3 38 102 5.0 3.0 11
6 Myelomeningocele 4.0 39 14 3.3 3.0 11
2 7 Myelomeningocele 3.3 38 72 0 0 24
8 Abdominal tumour 4.0 42 114 0 0 21
9 Myelomeningocele 3.5 37 9 0 0 13
10 Myelomeningocele 3.8 36 6 0 0 12
11 Myelomeningocele 3.7 40 22 0 0 9
12 Duodenal atresia 2.4 36 71 0 0 13
3 13 Oesophageal atresia 3.3 42 11 2.4 0 23
14 Duodenal atresia 2.6 37 25 2.6 0 25
172 K. SANDSTROM ET AL.
Table 2
Metabolic measures (mmol.l-') preoperatively, after 30 and 60 min of surgery, at the end of anaesthesia and 4 and 8 h postoperatively in
Group 1, neonates given glucose preoperatively and during surgery (n=6). Mean values f 1 s.d. are given. Abbrevations: TG - triglycerides,
FFA - free fatty acids, 3-OHB-3-hydroxybutyrate.
Preop. 30' 60' End 4h 8h
Glucose 3.8f 1.5 4.3 f 1.5 4.4f 1.1 5.45 1.0 5.7 f 3.1 5.2 f 1.8
TG 0.56 f 0.21 0.58 f 0.20 0.47 f 0.09 0.67 f 0.25 0.69 f 0.38 0.60 f 0.26
FFA 0.57k0.19 0.6 1 f 0.20 0.63 f 0.21 0.47 f 0.23 0.50 f 0.26 0.36 f 0.19
Lactate 2.49 f 0.99 2.24 f 0.80 2.23 5 1.24 2.46 f 1.27 1.58 f 0.76 2.03 f 0.81
Pyruvate 0.1 1 f 0.06 0.1 1 f 0.06 0.1 I f 0.06 0.11 f 0.07 0.11 f0.04 0.10 f 0.04
Alanine 0.32 f 0.10 0.25 f 0.10 0.26 f 0.06 0.25f0.11 0.30 f 0.10 0.27 f 0.08
3-OHB 0.32 k 0.33 0.29 f 0.26 0.28 f 0.18 0.40 f 0.36 0.13f0.14 0.09 f 0.07
Glycerol 0.51 f 0.53 0.54 f 0.59 0.79 f 0.97 0.61 f 0.66 0.41 f 0.35 0.32 f 0.14

surgery (P<0.05, t-test). Mean postoperative concen-
trations of FFA did not differ between the two study
groups. Mean 3-OHB was elevated both before and
during surgery among the starved neonates compared
to the neonates given glucose ( P < O . O O l , t-test). After
surgery the starved neonates had the same concen-
trations of 3-OHB as the neonates given glucose. Mean
3-OHB was higher during surgery compared to mean
postoperative concentration in the starved neonates
(P<O.OOl, t-test). The mean concentrations of blood
glucose, FFA and 3-OHB in the two groups at each
sampling occasion are plotted in Fig. 1 and 2. The
preoperative concentrations and the last concentrations
during surgery of FFA and 3-OHB are plotted in Fig.
3.
In one of the two infants in Group 3, in which
the glucose infusion was stopped and replaced with
Ringer-acetate, hypoglycaemia followed (Fig. 4).
Blood glucose concentrations in this neonate were re-
peatedly below 2 mmol 1-' during the first hour of
surgery. The lowest concentration was 1.2 mmol 1-I.
This patient also had the highest levels of 3-OHB.
Standard bicarbonate in this patient decreased from
21.6 to 19.6 mmol 1-'. Simultaneously FFA was the
second highest noted in this study. The mean blood
glucose concentration during surgery in the other in-
fant in Group 3 was 3.5 mmol 1-'. Compared to the
hypoglycaemic neonate, this neonate had less pro-
nounced increases in FFA and 3-OHB concentrations.
All neonates had blood gases and acid-base status
within the normal ranges during and after anaesthesia
without any significant differences between the groups.
For all the patients, peroperative p H (mean s.d.) was
7.45 k 0.1 1 Po, was 12.0 5 6.7 kPa and Pco, was 4.0 f
1.2 kPa. Standard bicarbonate concentrations in the
neonates were in the normal ranges at all sampling
occasions. The only exception was the hypoglycaemic
neonate mentioned above, who coincidentally with
the high 3-hydroxybutyrate concentration developed a
slight acidosis.
DISCUSSION
Blood glucose levels have previously been reported in
11 of 14 of these infants (3). The neonates not given
glucose before and during surgery maintained ade-
quate blood glucose concentrations at the same level
as the neonates given glucose before and during
surgery, as did the three additional patients recruited
to this study. The increased blood glucose concen-
trations are probably caused by the surgical trauma,
as is seen in adults ( 1 1) and older children ( 12). The

Table 3
Metabolic measures (mmol ' I - ' ) preoperatively, after 30 and 60 min of surgery, at the end of anaesthesia and 4 and 8 h postoperatively in
Group 2, starved neonates given no fluid before and Ringer during surgery (n =6). Means f 1 s.d. are given. Abbrevations: TG - triglycerides,
FFA - free fatty acids, 3-OHB-3-hydroxybutyrate.
Preop. 30' 60' End 4h 8h
Glucose 3.4f 1.1 4.4f 1.3 5.5 f 2.2 5.8 k 2.0 5.5 f 0.7 5.9 f 1.2
TG 0.57 f 0.24 0.60 f 0.25 0.62 f 0.25 0.61 f 0.25 0.65 k 0.23 0.60 f 0.26
FFA 0.76 f 0.27 0.76 f 0.35 1.13 f 0.57 0.75 f 0.26 0.40 f 0.23 0.35 f 0.25
Lactate 2.08 i 0.73 2.25f 1.14 2.25 f 0.71 1.61 f 0.67 1.58 f 0.76 1.89 _+ 0.80
Pyruvate 0.08 f 0.04 0.08 f 0.05 0.08 f 0.02 0.07 f 0.03 0.08 f 0.05 0.07 f 0.03
Alanine 0.34 f 0.10 0.31 f 0.09 0.35 f 0.08 0.35 k 0.09 0.38 f 0.07 0.37 f 0.08
3-OHB 0.74 f 0.35 0.95 f 0.46 1 .OO f 0.60 0.78 f 0.48 0.08 f 0.05 0.10 f 0.06
Glycerol 0.63 f 0.47 0.68 f 0.52 0.86 f 0.78 0.68 f 0.63 0.46 k 0.32 0.31 f0.16
 PERIOPERATIVE FLUID THERAPY 173
Glucose FFA Mean &EM
mmol.L' 3-OHB rnrnol.l-'

1.5 1 FFA 3-OHB

1.0 4

0.5 - I

-
Preop 30 rnin 60 min End 4h 8h
Fig. I . Mean concentrations of glucose, FFA and 3-OHB in neonates
given glucose pre- and intraoperatively (n = 6 ) ;preop, at 30, 60 min,
at the end of anaesthesia and 4 and 8 h postoperatively. Glucose
(mmol.1-I) is plotted on the left axis, FFA 0 (rnmol.l-') and 3-
OHB A (mmol . I - ' ) are plotted on the right axis.
ability of the neonate to respond to trauma and the
endocrine and metabolic reactions to surgical trauma
have recently been extensively studied (13, 14). Con-
siderably elevated levels of noradrenaline, adrenaline
and also of glucose, lactate, pyruvate, glycerol and
ketone bodies have been documented immediately
after surgery in infants and neonates (13). All infants
in these studies received a glucose infusion during
surgery. They could therefore be compared to the
group in our study receiving glucose pre- and intra-
Glucose FFA
rnrno1.i' 3-OHB
6- mmot.C'
5- 2.0
4-
9
-
01
Preop End of Preop End of
surgery surgery
Fig. 3. Mean preoperative (Preop) and last mean intraoperative
concentrations (End of surgery) of FFA in the starved neonates 0
and in the neonates given glucose 0 and of 3-OHB in the starved
neonates A and in the neonates given glucose A . * denotes a statisti-
cally significant difference in concentrations between Preop and End
of surgery (Wilcoxon's signed rank test).
operatively. These neonates showed only minor
changes in all measured parameters, except glucose.
Different fluid regimes may have contributed. In
earlier reports only maintenance volumes of electro-
lyte-free glucose were used during surgery, with insuf-
ficient peripheral circulation as a possible consequence.
Different anaesthesic practice in providing analgesia
during operation may also be an important factor. All
our patients were given up to 10 pg kg-' of fentanyl.
I n the first of the cited studies the analgesic agents were
limited to nitrous oxide in oxygen and low inspired
fractions of halothane (13). In the second study it was
shown that fentanyl almost eliminated the metabolic
reactions seen in the control group, which was only
given nitrous oxide as analgesia (14).
FFA
Glucose 3-OHB
mrno1.i' mmol.I -'

3-
1.6

1.2
1- 2.0
1.5

2- 0.8 - 1.0
1- 0.4
- 0.5
0
L 0
Preop 30 min 60 min End 4h 8h 0
Preop 30 rnin 60 min 90 min End 4 h
Fig. 2. Mean concentrations of glucose, FFA and 3-OHB in starved
neonates without fluid preoperatively and Ringer acetate intraopera- Fig. 4. One patient with glucose preoperatively and Ringer-acetate
tively ( n = 6 ) ; preop, at 30, 60 min, at the end of anaesthesia and 4 intraoperatively. Concentrations preop, at 30, 60, 90 min, at the end
and 8 h postoperatively. Glucose H (mmol . I-') is plotted on the left of anaesthesia and 4 h postoperatively. Glucose H (mrnol ' 1 - I ) is
axis, FFA 0 (mmol.l-') and 3-OHB A (mmol.1-I) are plotted on plotted on the left axis, FFA 0 (mmol.l-'), 3-OHB A (mmol.1-I)
the right axis. are plotted on the right axis.
174 K. SANDSTROM ET AL.
During surgery, slightly higher concentrations of lac-
tate and pyruvate were seen in the group given glucose
intraoperatively compared to the group given only
Ringer-acetate. The difference did not reach signifi-
cant levels. Similarly alanine and glycerol concen-
trations tended to be higher in the surgical period
in the starved neonates compared to the non-starved
neonates. Glycogen mainly from the liver might pro-
vide the maintained blood glucose concentrations dur-
ing surgery in the starved neonates. Gluconeogenesis
from alanine and glycerol might also contribute to the
maintained blood glucose concentrations.
High levels of 3-hydroxybutyrate before and during
surgery were found in neonates without glucose supply.
Free fatty acids were also higher before surgery in the
starved neonates and during surgery concentrations
rose. This indicates a higher fat utilization in these
neonates. A rapid transition of metabolism from glu-
cose to fat as the main source of energy occurs in
normal neonates after birth (15). Rising concentrations
of free fatty acids, glycerol and ketone bodies have
been found during the first days of life (4,16). As a
result of the increased fat metabolism, the respiratory
quotient is decreased ( 1 7 ) .
In our earlier study we found a substantial risk for
intraoperative hypoglycaemia in those infants whose
preoperative glucose infusion was interrupted. This
hypoglycaemia was not counteracted by the simul-
taneous start of surgery. All infants who became hypo-
glycaemic were younger than 48 h (3). Because of this
finding, only two infants with this fluid regimen were
available for the extended metabolic analysis. High
levels of 3-hydroxybutyrate and free fatty acids were
found in one of these neonates with interrupted glucose
infusion. Ketone bodies can be used by the brain as
energy substrate (18). In adults, after prolonged star-
vation, ketone bodies replace glucose as the predomi-
nant energy substrate of the brain ( 19). Cerebral arte-
rio-venous differences of ketone bodies have been
studied in children after induction of elective cardiac
anaesthesia (20). Depending on the arterial concen-
tration, the brain has a net uptake of ketone bodies in
this situation (20).
The postoperative metabolism showed similar pat-
terns in all neonates. Glucose concentrations were of
the same magnitude. Free fatty acids and S-hydroxy-
butyrate decreased from the elevated intraoperative
levels in the previously starved neonates as a conse-
quence of the postoperative glucose infusion.
This study of carbohydrate and fat metabolism dur-
ing and after neonatal surgery has revealed two differ-
ent metabolic reactions, depending on the fluid ther-
apy. In neonates given glucose before and during
surgery, rising blood glucose concentrations were seen
intraoperatively. As a consequence of this, the fat util-
ization and gluconeogenesis were both low. In starved
neonates given only Ringer-acetate during surgery,
equally rising blood glucose concentrations were also
found. The most probable explanation is an adequate
gluconeogenesis. Simultaneously, these neonates
showed a substantial release and utilization of fat. In
conclusion, it appears that the starved neonates
adapted to and could cope with a glucose-free peri-
operative fluid therapy and that the neonates given
peroperative glucose could handle the amount of glu-
cose given in this study. Interruption of a running
glucose infusion is unfavourable.
ACKNOWLEDGEMENTS
This study was supported by grants from The Goteborg Medical
Society.
REFERENCES
1. Hatch D J, Sumner E. Current topics in anaesthesia 5. Neonatal
anaesthesia, 1st edn. London: Edward Arnold, 1981: 57-58.
2. Swyer P R. Nutrition, growth and metabolism in the newborn.
In: Prakash 0, ed. Critical care of the child. Boston: Martin
Nijhoff Publishers, 1984: 1-29.
3. Larsson L E, Nilsson K, Niklasson A, Andrtasson S, Ekstrom-
Jodal B. Influence of fluid regimens on perioperative blood-
glucose concentrations in neonates. Br 3 Anaesth 1990: 64:
4 19-424.
4. Elphick M C, Wilkinson A W. The effects of starvation and
surgical injury on the plasma levels of glucose, free fatty acids
and neutral lipids in newborn infants suffering from various
congenital anomalies. Pedialr Res 1981: 15: 313-318.
5. Carrol J J, Smith N, Babson A L. A colorimetric serum glucose
determination using hexokinase and glucose-6-phosphate de-
hydrogenase. Biochem Med 1970: 4: 171-180.
6. Olegird R. Essential fatty acids in the postnatal period. In:
Metabolism of blood lipids in newborn infants. Thesis Goteborg:
Gotab, 1974.
7. Persson B. Determination of plasma acetoacetate and D-betahy-
droxybutyrate in newborn infants by an enzymatic micro-
method. Scand J Clin Lab Inuesl 1970: 25: 9-18.
8. Karl I E, Pagliara A S, Kipnis D M. A microfluorimetric enzy-
matic assay for the determination of alanine and pyruvate in
plasma and tissues. 3 Lab Clan Med 1972: 80: 434-441.
9. Laurel1 S, Tibbling G. An enzymatic micromethod for the deter-
mination of glycerol. Clin Chim Acta 1966: 13: 317-322.
10. Lowry 0 H, Passoneau J V. Collections of metabolic assays. In:
Lowry 0 H, Passoneau J V, eds. A flexible system of enzymatic
analysis. New York: Academic Press, 1972: 199-201.
11. Wiklund L, Thortn L. Intraoperative blood compartment and
fluid therapy. ,Ada Anaesthesiol Scand 1985: 29 (Suppl. 82): 1-8.
12. Ryhanen P, Knip M, Puukka R, Huttunen N-P, Lahti Y, Perkila
L. Regulatory response to intravenous glucose infusion in
children undergoing surgery. Anesthesiology 1988: 68: 147-1 52.
13. Anand K J S, Brown M J, Causon R C, Christofides N N D,
Bloom S D, Aynsley-Green A. Can the human neonate mount
an endocrine and metabolic response to surgery? J Pediatr Surg
1985: 20: 41-48.
14. Anand K J S, Sippell W G, Aynsley-Green A. Randomised trial
 PERIOPERATIVE FLUID THERAPY 175
of fentanyl anaesthesia in preterm babies undergoing surgery: 19. Owen 0 E, Morgan A P, Kemp H G , Sullivan J M, Herrera
effects on the stress response. Lancet 1987: i: 62-65. M G, Cahill G F. Brain metabolism during fasting. 3 Clin lnuest
15. Persson B, Tunell R. Influence of environmental temperature 1967: 46: 1589.
and acidosis on lipid mobilization in the human infant during the 20. Settergren G , Lindblad B S, Persson B. Cerebral blood flow and
first two hours after birth. Acta Pediatr Scand 1971: 61: 385-389. exchange of oxygen, glucose, ketone bodies, lactate, pyruvate
16. Persson B, Gentz J. The pattern of blood lipids, glycerol and and amino acids in infants. Acta Pediatr Scand 1976: 65: 343-353.
ketone bodies during the neonatal period, infancy and childhood.
Acta Paediah Scand 1966: 55: 353.
17. Senterre J, Karlberg P. Respiratory quotient and metabolic rate Address:
in normal full term and small-for-date newborn infants. Ada Kcrstin Sandstrom, M D
Paediatr Scand 1970: 59: 653. Department of Paediatric Anaesthesia
18. Robinson A M, Williamson D H. Physiological roles of ketone &tra Sjukhuset
bodies as substrates and signals in mammalian tissues. Physiol S-41685 Goteborg
Rev 1980: 60: 143-187. Sweden