British Journal of Anaesthesia 112 (3): 540–5 (2014)
Advance Access publication 4 November 2013 . doi:10.1093/bja/aet374
PAEDIATRICS
Postoperative decrease in plasma sodium concentration after
infusion of hypotonic intravenous solutions in neonatal surgery†
G. Edjo Nkilly 1, D. Michelet 1, J. Hilly1, T. Diallo 1, B. Greff 1, N. Mangalsuren 1, E. Lira 1, I. Bounadja 1, C. Brasher 1,
A. Bonnard2, S. Malbezin 1, Y. Nivoche 1 and S. Dahmani 1,3*
1
Department of Anaesthesia, Intensive Care and Pain Management,2 Department of General and Urological Surgery, AP-HP, and 3 UMR INSERM
U 676, Robert Debré University Hospital, Paris Diderot University, Paris Sorbonne Cité, Paris, France
* Corresponding author. E-mail: [email protected]
Editor’s key points
† The role of hypotonic
solution infusion in
causing hyponatraemia in
neonatal surgery is
unclear.
† In a prospective study,
reductions in
postoperative plasma
sodium concentration
correlated with
intraoperative, but not
preoperative, free water
intake.
† Plasma sodium
concentration should be
closely monitored.
† Routine use of hypotonic
i.v. solutions is
questionable in neonatal
surgery.
Background. Hypotonic i.v. solutions can cause hyponatraemia in the context of paediatric
surgery. However, this has not been demonstrated in neonatal surgery. The goal of this
study was to define the relationship between infused perioperative free water and plasma
sodium in neonates.
Methods. Newborns up to 7 days old undergoing abdominal or thoracic surgery were included
in this prospective, observational study. Collected data included type and duration of surgery,
calculated i.v. free water intake, and pre- and postoperative plasma sodium. Statistical
analyses were performed using the Pearson correlation, Mann– Whitney test, and receiver
operating characteristic analysis with a 1000 time bootstrap procedure.
Results. Thirty-four subjects were included. Postoperative hyponatraemia occurred in four
subjects (11.9%). The difference between preoperative and postoperative plasma sodium
measurements (DNaP) correlated with calculated free water intake during surgery (r¼0.37,
P¼0.03), but not with preoperative free water intake. Calculated operative free water intake
exceeding 6.5 ml kg21 h21 was associated with DNaP≥4 mM with a sensitivity and
specificity [median (95% confidence interval)] of 0.7 (0.9 –1) and 0.5 (0.3– 0.7), respectively.
Conclusions. Hypotonic solutions and i.v. free water intake of more than 6.5 ml kg21 h21 are
associated with reductions in postoperative plasma sodium measurements ≥4 mM. In the
context of neonatal surgery, close monitoring of plasma sodium is mandatory. Routine use
of hypotonic i.v. solutions during neonatal surgery should be questioned as they are likely to
reduce plasma sodium.
Keywords: fluids; hypotonic; i.v.; neonatal surgery; neonates; postoperative hyponatraemia
Accepted for publication: 23 August 2013
Recent studies highlight the potential for long-term neurocognitive developmental delay in neonates suffering from hyponatraemia.1 – 3 This is of significant concern as hyponatraemia
rates of up to 24% have been reported in preterm infants
receiving i.v. fluids.4
Recommendations regarding perioperative i.v. infusions in
children have changed, particularly with respect to tonicity.5 6
Many authors now recommend the use of isotonic or nearisotonic infusions perioperatively to prevent hyponatraemia.5 7
Syndrome of inappropriate antidiuretic hormone (ADH)
occurs commonly in children after spinal surgery.6 8 Precipitating factors include pain, mechanical ventilation, and opioid
administration. The risk of profound hyponatraemia in
paediatric surgical patients is higher when receiving hypotonic
†
i.v. solutions and there have been numerous reports of
cerebral oedema, coning, and patient death or permanent
handicap.9 10 However, the perioperative risks of hyponatraemia have been poorly explored in neonates, and there are no
such recommendations.11 Perioperative hypotonic i.v. fluids
are still used due to concerns over the expansion of their
larger extracellular space with isotonic solutions and renal
immaturity. Many anaesthesiologists still follow neonatal
intensive care guidelines recommending hypotonic i.v. infusions containing 20–40 mM of sodium and daily sodium
intakes of 2–3 mmol kg21 day21.11 These protocols do not
take into account the metabolic and volume homeostasis
requirements of surgery, and there is some discussion as to
their appropriateness given hyponatraemia rates in neonates.4
This article is accompanied by Editorial III.
& The Author [2013]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
For Permissions, please email: [email protected]
BJA
Postoperative hyponatraemia in neonates
The goal of this study was to assess the change in plasma
sodium concentration during neonatal surgery in the first
week of life and its relationship with calculated i.v. free water
administration.
Methods
This is a prospective single-centre observational study undertaken with ethical committee approval (Comité d’Evaluation
de l’Ethique des Projets de Recherche Biomédicale—CEERB—
Paris Nord). Written consent was obtained from all parents.
RL with 1% dextrose. The first fluid challenge was systematically performed after induction. Fluid challenges were repeated
according to changes in haemodynamic parameters (.20%
increase in heart rate or decrease in MAP compared with baseline), decreased urine output when measured, intraoperative
surgical circumstances such as bleeding or prolonged
abdominal surgery, and at the anaesthesiologist’s discretion.
Hypotension refractory to 20 ml kg21 of i.v. fluid and bradycardia (heart rate ,100 beats min21) were treated with i.v. atropine 20 mg kg21.
Inclusion criteria
Data collected
Patients were included if they fulfilled the following criteria:
thoracic or abdominal surgery in neonates up to 7 days old requiring preoperative i.v. fluid therapy, absence of preoperative
mechanical ventilation and haemodynamic support, and
absence of cardiac or vascular malformations.
The following variables were recorded: age, weight, gestational
age, surgery type and duration, pre- and postoperative plasma
osmolarity [OSMp: (plasma Na+ +plasma K+)×2+plasma
urea+plasma glucose], computed i.v. solution theoretical
osmolarity (OSMIV: manufacturer specifications minus
glucose), i.v. solution volumes, and calculated free water
administration {total i.v. fluid volume×[1–(OSMIV /OSMp)]}.
I.V. solution volumes was used to calculate free water administration {total i.v. fluid volume×[1–(OSMIV ×0.923/OSMp)]}.
The corrected factor for OSMIV, called the osmotic coefficient,
was used in order to take in account the decrease in the
theoretical osmolarity of perfused solutions in plasma
(Table 1). All data were collected for preoperative and operative
time periods. Hyponatraemia was defined as plasma sodium
concentration (NaP),135 mmol litre21.
Anaesthetic management
Subjects were not premedicated. Standardized laboratory
examinations included: prothrombin time, activated partial
thromboplastin time, haemoglobin and platelet counts,
plasma fibrinogen, protein, sodium (NaP), potassium, bicarbonate, urea, creatinine, and blood sugar levels. The timing
of preoperative laboratory testing was not standardized. Postoperative testing was performed immediately on arrival in the
post-anaesthetic care unit. Anaesthesia was performed
according to local protocols with standard monitoring. Induction of anaesthesia used propofol (3–5 mg kg21) or thiopental
(3–5 mg kg21) and succinylcholine (2 mg kg21). After tracheal
intubation, sufentanil (0.2 mg kg21) was administered. Hypnosis was maintained using sevoflurane (0.8 –1 age-adjusted
minimum alveolar concentration) in 50% O2 –air. Ventilation
was pressure-controlled without end-expiratory pressure. Endtidal CO2 measurements were maintained between 4 and
4.7 HPa (30–35 mm Hg). Analgesia was administered as
sufentanil boluses of 0.1 mg kg21 when heart rate or mean
arterial pressure (MAP) increased by 20% of baseline. Subjects
were actively warmed from entry into theatre until exit, and
room temperature was kept at 21–238C. Thirty minutes
before the end of surgery, patients received 10 mg kg21 i.v.
paracetamol. At the end of surgery, regional analgesia was
performed, where indicated, consisting of transversus abdominis plane block. After operation, subjects received paracetamol (10 mg kg21 6 hourly), and nalbuphine or morphine
nurse-controlled analgesia protocols.
I.V. fluid management
Fluid management was physician and speciality (anaesthesiologist or neonatologist) dependent. Preoperative fluid intake
ranged from 100 to 120 ml kg21 day21 of various solutions containing 5–10% dextrose. Intraoperative fluid intake consisted
of a maintenance rate of 4 ml kg21 h21 (Holliday and Segar
formula) plus compensation for surgical fluid losses (4–20 ml
kg21 h21). Fluid challenges were performed using 10 ml kg21
over 15 min of either normal saline, Ringer’s lactate (RL), or
Statistical analysis
Plasma sodium changes between pre- and postoperative measurements (DNaP) were analysed along with their relationship
to pre- and intraoperative i.v. free water intakes.
Continuous variables were compared using analysis of
variables (ANOVA), and correlations described using the
Pearson correlation coefficient. In sample sizes ,30, the
Spearman correlation and Mann–Whitney test were used.
Discrete variables were compared using the x 2 test or Fisher’s
exact test. A P-value of ,0.05 was considered the threshold
to reject the null hypothesis. Thresholds were defined using
the receiver operating characteristic (ROC) curve analysis and
determination of the Younden index (the maximal value of
sensitivity+specificity). The latter was performed using a
1000 time bootstrap method that allows for re-sampling of
original data in order to increase statistical precision.
Regression to the mean was assessed by linear correlation
and cofactor analysis (ANCOVA),12 13 after describing distribution
normality using the Shapiro–Wilk test. Regression to the
mean occurs when repeated measurements of a factor are performed. It consists of spontaneous or bias-influenced evolution
in extreme values measurements towards the mean or in contrast, those close to the mean towards extreme values.13 14
Descriptive statistics were displayed as median (minimum –
maximum). The area under the curve (AUC), sensitivity and
specificity were expressed as median (95% confidence interval). Statistical analyses were performed using SPSS 20.0 software (IBM, Chicago, IL, USA) and the pROC module (for ROC
541
BJA
Edjo Nkilly et al.
Table 1 I.V. solutions: composition and osmolarity. G 10%w, G 5%w, B45w, B46w, B66w, Pediavenw: Pharmacie Centrale des Hôpitaux de Paris, Paris,
France. B27w: Baxter France, Maurepas, France. B26w, normal saline, RL: Fresenius France, Sèvres, France
Solution
Glucose
(mM)
Lactate
(mM)
Na
(mM)
Normal
saline
0
0
154
Ringer’s
lactate
0
28.4
131
K
(mM)
5.4
Cl
(mM)
Ca/P/Mg
(mM)
Theoretical osmolarity without
glucose (mOsmol litre21)
Theoretical osmolarity with
glucose (mOsmol litre21)
154
0
308
308
112
1.8
278
278
B26
277
0
68
27
95
0
190
467
B27
277
0
34
20
54
2.3
110
387
B66
51
20.1
120
4
108
2.2
260
311
B45
277
0
34
20
59
4.7
117
394
B46
550
0
51
20
71
4.5
146
696
Pediaven
550
0
20
17
20
18.3
75
625
G5%
277
0
0
0
0
0
0
277
G 10%
550
0
0
0
0
0
0
550
Preoperative
I.V. free
water
(ml kg21)
Intraoperative
I.V. free
I.V. free
water
water
(ml kg21 h21) (ml kg21)
I.V. free
water
(ml kg21 h21)
30
Intraoperative free water
intake (ml kg–1 h–1)
Table 2 Linear correlations between the difference of preand postoperative plasma sodium measurements (DNaP) and
postoperative plasma sodium measurements (NaP) vs
perioperative free water intake and overall i.v. solution tonicity. r,
partial Pearson correlation
20
10
DNaP
r
0.07
0.24
0.37
0.36
P-value
0.7
0.18
0.03
0.03
20.03
20.17
0.17
0.16
0.8
0.9
0.3
0.3
Preoperative NaP
r
P-value
Postoperative NaP
r
P-value
20.12
0.5
20.31
0.07
20.3
0.08
–5
5
0
DNaP (mmol litre–1)
10
Fig 1 Linear correlation between intraoperative i.v. free water
intake and the difference between pre- and postoperative plasma
sodium concentration (DNaP). r¼0.37, P¼0.03.
20.31
0.08
analysis with resampling) for R software 3.0.1 (The R Foundation for Statistical Computing, Vienna, Austria).
Pooled data from the literature in neonatal intensive care,
preterm infants, and other paediatric populations were used
to predict the occurrence of hyponatraemia, as no data are
available for surgical neonates (n¼370).3 15 16 There was a
mean incidence of hyponatraemia of 40% over all studies
where patients received hypotonic solutions (range: 14–100%).
The mean incidence for premature neonates was 26% (n¼58).3
Therefore, a sample size of 30 should result in four to 30 cases
of postoperative hyponatraemia.
Results
Thirty-four patients were eligible for the study and all were enrolled. Surgery was abdominal (n¼26: colostomy, colectomy,
small bowel atresia, omphalocoele, gastroschisis, ovarian
542
0
–10
hernia, and oesophageal atresia) or thoracic (n¼6: diaphragmatic hernia).
Three subjects were found to be hyponatraemic before
surgery (NaP¼133 mM in all three), and all normalized
plasma sodium during surgery. Four others were found to be
hyponatraemic after operation (133 mM: n¼1, 134 mM: n¼3)
with a DNaP ranging from 1 to 9 mmol litre21.
A statistical correlation was found between DNaP and free
water volume administered intraoperatively (Table 2 and
Fig. 1). However, there was no significant correlation with free
water administered before operation (Table 2). Nor was there
a significant correlation between pre- and postoperative
plasma sodium values and administered free water volume,
whether before operation or intraoperatively (Table 2).
Performing an ROC curve analysis using DNaP values and
postoperative hyponatraemia resulted in an AUC of 0.88
(95% confidence interval 0.7 –1). The best threshold for DNaP
was 4 mM with a sensitivity of 0.75 (95% confidence interval
BJA
Postoperative hyponatraemia in neonates
Table 3 Comparison of subjects with a postoperative variation of plasma sodium concentration (DNaP) of ≥4 mM with those with DNaP,4 mM.
Data expressed as median (range)
Overall (n534)
Age (days)
Gestational age (weeks)
Weight (kg)
1.5 (0 –7)
DNaP<4 mM (n525)
1.9 (0–7)
DNaP≥4 mM (n59)
P-value
1 (0 –2)
0.16
38.6 (32.5 – 42)
38 (33– 42)
39 (32 – 39)
0.4
2.8 (2.1 –3.8)
2.8 (2.1– 3.9)
2.9 (2.2 –3.7)
0.9
139 (133 –144)
138 (133 –144)
142 (138 –144)
Preoperative
NaP (mM)
Plasma K (mM)
Plasma urea (mM)
4.3 (3 –6)
0.9
4.1 (1.7 –7.4)
0.8
61 (18 – 108)
62 (18– 108)
61 (53 – 90)
0.3
Plasma protein (g litre21)
57 (31 – 72)
55 (31– 72)
60 (50 – 70)
0.06
288 (276 –301)
286 (276 –296)
290 (285 –301)
0.015
17 (10 – 20)
16 (9.8– 20)
17.6 (13 – 20)
0.4
Hb (g dl21)
Urine output (ml kg21 litre21)
4 (1.4– 13.7)
0.004
4.4 (3 –5.4)
Plasma creatinine (mM)
Plasma osmolarity (mOsm litre21)
4 (1.4 –13.7)
4.3 (3.2– 6)
2 (1 –2.7)
1.8 (0.4– 5)
Infusion duration (h)
24 (2 –24)
24 (3–24)
24 (2 –24)
0.4
I.V. free water intake (ml kg21)
52 (0 –136)
52 (21 to 136)
53 (8 –96)
0.9
I.V. free water intake (ml kg21 h21)
Infused solution tonicity (mOsm litre21)
2 (1.1 –2.7)
3 (0 –5.6)
3 (0–5.6)
3 (2 –5.6)
0.7
75 (0 –308)
75 (0–308)
75 (0 –179)
0.3
139 (133 –149)
140 (134 –149)
136 (133 –140)
0 (29 to 9)
22 (29 to 3)
Postoperative
NaP (mM)
DNaP (mM)
Plasma K (mM)
5 (4 –9)
0.002
,0.0001
4 (2.6 –6.4)
4.2 (3.1– 6.4)
4 (2.6 –6)
Plasma urea (mM)
3.6 (1.1 –8.7)
3.5 (1.1– 8.7)
3.7 (1.3 –7.8)
Plasma creatinine (mM)
61 (4 –101)
53 (4–101)
624 (45 – 78)
0.24
47.5 (40 – 67)
48 (40– 67)
47 (40 – 59)
0.13
Plasma protein (g litre21)
D Plasma protein (g litre21)
Plasma osmolarity (mOsm litre21)
6.5 (211 to 28)
289 (275 –304)
4 (211 to 20)
289 (277 –304)
15 (2 –28)
281 (275 – 298)
0.3
1
0.004
0.017
Hb (g dl21)
14 (8 –20)
15 (9.8 –20)
0.7
D Hb (g dl21)
1.6 (23.9 to 10.8)
1.5 (27.7 to 8.6)
1.7 (22.2 to 10.8)
0.7
Surgery duration
90 (30 – 240)
90 (60– 180)
90 (30 – 240)
0.8
I.V. free water intake (ml kg21)
10 (0 –37)
9 (0–34)
13 (5 –37)
0.6
8 (0 –23)
7 (0–23)
10 (3 –19)
243 (0 –308)
273 (0–308)
I.V. free water intake (ml kg21 h21)
Infused solution tonicity (mOsm litre21)
0.25 –1) and a specificity of 0.93 (95% confidence interval
0.83 –1).
Using this DNaP threshold and comparing subjects with
DNaP.4 mM (n¼9, 26.5%) and those with lesser DNaP
values (n¼25, 73.5%), a significant difference was again
demonstrated for increased intraoperative free water intake
(Table 3). There was, however, no significant difference in preoperative i.v. free water intake. Plasma protein levels also
decreased significantly (DPlasma protein, Table 3) in subjects
with DNaP.4 mM. Preoperative plasma sodium was significantly higher in subjects exhibiting a DNaP.4 mM.
Preoperative NaP was significantly higher in subjects exhibiting a DNaP,4 mM (Table 3). As a consequence, regression
to the mean might have contributed to the higher value
of DNaP in this group. Data concerning DNaP were normally distributed within all subgroups (subjects with DNaP , or .4 mM).
A significant correlation was found between preoperative NaP
and DNaP (r¼0.6, P,0.0001). ANCOVA using preoperative NaP as
a covariate still found a significant difference in DNaP between
14.8 (8–20)
231 (168 –276)
0.05
0.1
the two groups (P,0.0001, F¼22.8), but a significant effect of
preoperative hyponatraemia on this difference was also
observed (P¼0.013, F¼7).
ROC curve analysis of the association of intraoperative free
water intake and a DNaP≥4 mM found an AUC of 0.73 (95%
confidence interval 0.53–0.93, Fig. 2). The Younden index
was 6.5 ml kg21 h21 of intraoperative free water, and the sensitivity and specificity [expressed as mean (95% confidence
interval)] at this threshold were 0.7 (0.9– 1) and 0.5 (0.3 –0.7),
respectively. Twenty-one subjects received .6.5 ml kg21 h21
free water intraoperatively. As expected, the percentage of
subjects exhibiting DNaP≥4 mM was higher in those receiving
.6.5 ml kg21 h21 of free water intraoperatively (n¼8, 38%)
than in those receiving ,6.5 ml kg21 h21 (n¼1, 7.7%, P¼0.05).
Discussion
The use of hypotonic solutions during neonatal surgery was
associated with reduced postoperative plasma sodium levels.
543
BJA
Edjo Nkilly et al.
1.0
Sensitivity
0.8
0.6
0.4
0.2
0.0
0.0
0.2
0.4
0.6
1 – Specificity
0.8
1.0
Fig 2 ROC curve analysis of intraoperative free water intake according to measured postoperative variation of plasma sodium concentration of ≥4 or ,4 mM.
Hyponatraemia occurred in 12% of cases and DNaP correlated
with the amount of i.v. free water administered. DNaP values
.4 mM were statistically associated with increased intraoperative i.v. free water volumes.
Our population was selected to assess the effect of i.v. fluids
on perioperative plasma sodium levels. Patients receiving preoperative ventilatory or haemodynamic support, factors
known to favour hyponatraemia,1 – 3 10 17 were excluded. Surgical illness and operative fluid therapy were therefore the major
determinants of water homeostasis. The DNaP threshold value
of 4 mM used was defined by the occurrence of postoperative
hyponatraemia in the sample, independent of preoperative
serum sodium values. A significant correlation was found
between preoperative NaP and DNaP. Given the coefficient of
correlation ,1, a regression to the mean effect in the observed
decrease in DNaP cannot be excluded.14 However, when preoperative NaP was introduced as a covariate, a statistically significant difference in DNaP was still found between those
subjects with a DNaP≥4 mM and those with ,4 mM. As such,
even if present, regression to the mean does not explain all
variation of DNaP. Preoperative plasma sodium concentration
did not correlate with free water intake (Table 2) and as such,
it appears unlikely that preoperative sodium levels influenced
perioperative fluid therapy.
Intraoperative free water intake correlated significantly
with DNaP and was significantly higher in subjects with
DNaP.4 mM. In addition, protein levels decreased significantly in subjects where DNaP was .4 mM (Table 3). These results
support the role of free water intake and increasing intravascular water as causal factors for hyponatraemia in neonates
undergoing surgery. Neonates respond to ADH, from
26 weeks gestation (although they are less sensitive to its
544
effects than older children and adults).18 – 23 ADH secretion is
likely to occur perioperatively as a result of hypovolaemia and
metabolic stimuli, including pain, inflammation, and opioid administration.6 11 24 – 26 Accordingly, the combined effects of
perioperative hypotonic fluids and ADH on water reabsorption
in the kidney significantly favour a decrease in plasma sodium
in neonates as in older children.6 Future studies examining
non-osmotic ADH secretion via urine biochemistry and serum
ADH assays would be useful.21 23
No statistical association was found between preoperative
free water intake and DNaP despite larger free water intakes
during this period (Tables 2 and 3). This could be a result of insensible hypotonic water losses in the absence of ADH secretion. One previous study of non-surgical neonates found they
can cope with water and sodium intakes of 200 and 3 mg
kg21 day21, respectively.27
Theoretical average flow rates for hypotonic solutions
to cause plasma sodium reductions of .4 mM (free water
.6.5 ml kg21 h21) ranged from 6.5 to 18 ml kg21 h21 (6.5 ml
kg21 h21 for G10% and G5%; 8 ml kg21 h21 for Pediavenw;
10 ml kg21 h21 for B27w and B45w; 12 ml kg21 h21 for B46w;
and 18 ml kg21 h21 for B26w). As expected, near-isotonic
solutions require much higher flow rates (no limit, 87 and
44 ml kg21 h21 for normal saline, RLw and B66w, respectively).
Given the relatively high i.v. infusion rates used during neonatal
surgery, anaesthesiologists should be aware of the possibility
of a reduction in postoperative plasma sodium with any
hypotonic solution.
Our sample did not include any cases of profound hyponatraemia. Furthermore, the overall incidence of hyponatraemia
was relatively low, despite high free water intakes. This might
be explained by neonatal renal insensitivity to ADH, lack of
renal concentrating power, or both.1 22 23
Hyponatraemia with NaP of 130 mM in neonates is detrimental to long-term neurocognitive development, especially
in preterm children.4 28 However, the consequences of perioperative hyponatraemia in this age group are unknown. One
subject in the study did experience a decrease in postoperative
NaP of 9 mM which would have caused profound hyponatraemia in a patient with preoperative NaP of 139 mM or less.
This study emphasizes the potential for hypotonic solutions
to reduce postoperative plasma sodium concentration and
cause postoperative hyponatraemia. Close monitoring of
plasma sodium concentration is mandatory for surgery in
the first week of life. The use of isotonic and near-isotonic i.v.
solutions during the operative period is recommended.
Authors’ contributions
G.E.N.: participated in study design, patient recruitment, data
analysis and interpretation, drafting and revising the manuscript, and approved the final version; D.M., J.H., and S.M.:
participated in study design, patient recruitment data interpretation, and approved the final version; T.D., B.G., N.M., E.L.,
I.B., and A.B.: participated in patient recruitment and approved
the final version; C.B.: participated in data collection analysis
and interpretation, drafting and revising the manuscript, and
Postoperative hyponatraemia in neonates
approved the final version. Y.N.: participated in data collection
analysis and interpretation, revised the manuscript, and
approved the final version. S.D.: participated in study design,
patient recruitment, data collection analysis and interpretation, drafting and revising the manuscript, and approved the
final version
Acknowledgements
This study is dedicated to Professor Isabelle Murat and her
long and fruitful clinical and academic career as a paediatric
anaesthetist in France. The study was greatly influenced by
her previous work on this subject, which led to the manufacture
of a specific perioperative paediatric i.v. solution (B66—1%
glucose in Ringer’s lactate) which is available in the Paris
public hospital system.
Declaration of interest
None declared.
Funding
Support was provided solely from institutional and departmental sources.
References
1 Al-Dahhan J, Haycock GB, Chantler C, Stimmler L. Sodium
homeostasis in term and preterm neonates. II. Gastrointestinal
aspects. Arch Dis Child 1983; 58: 343– 5
2 Al-Dahhan J, Haycock GB, Chantler C, Stimmler L. Sodium
homeostasis in term and preterm neonates. I. Renal aspects.
Arch Dis Child 1983; 58: 335– 42
3 Al-Dahhan J, Haycock GB, Nichol B, Chantler C, Stimmler L. Sodium
homeostasis in term and preterm neonates. III. Effect of salt supplementation. Arch Dis Child 1984; 59: 945–50
4 Moritz ML, Ayus JC. Hyponatremia in preterm neonates: not a
benign condition. Pediatrics 2009; 124: e1014– 6
5 Sumpelmann R, Becke K, Crean P, et al. European consensus
statement for intraoperative fluid therapy in children. Eur J
Anaesthesiol 2011; 28: 637–9
6 Bailey AG, McNaull PP, Jooste E, Tuchman JB. Perioperative crystalloid and colloid fluid management in children: where are we and
how did we get here? Anesth Analg 2010; 110: 375– 90
7 Murat I, Dubois MC. Perioperative fluid therapy in pediatrics.
Paediatr Anaesth 2008; 18: 363– 70
8 Lieh-Lai MW, Stanitski DF, Sarnaik AP, et al. Syndrome of inappropriate antidiuretic hormone secretion in children following spinal
fusion. Crit Care Med 1999; 27: 622– 7
9 Brazel PW, McPhee IB. Inappropriate secretion of antidiuretic
hormone in postoperative scoliosis patients: the role of fluid
management. Spine (Phila Pa 1976) 1996; 21: 724–7
BJA
10 Burrows FA, Shutack JG, Crone RK. Inappropriate secretion of
antidiuretic hormone in a postsurgical pediatric population. Crit
Care Med 1983; 11: 527– 31
11 Modi N. Management of fluid balance in the very immature
neonate. Arch Dis Child Fetal Neonatal Ed 2004; 89: F108– 11
12 Barnett AG, van der Pols JC, Dobson AJ. Regression to the
mean: what it is and how to deal with it. Int J Epidemiol 2005; 34:
215– 20
13 Bland JM, Altman DG. Comparisons within randomised groups can
be very misleading. Br Med J 2011; 342: d561
14 Bland JM, Altman DG. Some examples of regression towards the
mean. Br Med J 1994; 309: 780
15 Kannan L, Lodha R, Vivekanandhan S, Bagga A, Kabra SK
Kabra M. Intravenous fluid regimen and hyponatraemia among
children: a randomized controlled trial. Pediatr Nephrol 2010; 25:
2303– 9
16 Choong K, Kho ME, Menon K, Bohn D. Hypotonic versus isotonic
saline in hospitalised children: a systematic review. Arch Dis Child
2006; 91: 828–35
17 Coulthard MG, Cheater LS, Long DA. Perioperative fluid therapy in
children. Br J Anaesth 2007; 98: 146– 7
18 Boone M, Deen PM. Physiology and pathophysiology of the
vasopressin-regulated renal water reabsorption. Pflugers Arch
2008; 456: 1005– 24
19 Rees L, Brook CG, Shaw JC, Forsling ML. Hyponatraemia in the first
week of life in preterm infants. Part I. Arginine vasopressin
secretion. Arch Dis Child 1984; 59: 414– 22
20 Rees L, Shaw JC, Brook CG, Forsling ML. Hyponatraemia in the first
week of life in preterm infants. Part II. Sodium and water balance.
Arch Dis Child 1984; 59: 423– 9
21 David R, Ellis D, Gartner JC. Water intoxication in normal infants: role
of antidiuretic hormone in pathogenesis. Pediatrics 1981; 68:
349– 53
22 Holtback U, Aperia AC. Molecular determinants of sodium and
water balance during early human development. Semin Neonatol
2003; 8: 291–9
23 Kovacs L, Sulyok E, Lichardus B, Mihajlovskij N, Bircak J. Renal
response to arginine vasopressin in premature infants with late
hyponatraemia. Arch Dis Child 1986; 61: 1030– 2
24 Modi N. Hyponatraemia in the newborn. Arch Dis Child Fetal
Neonatal Ed 1998; 78: F81 –4
25 Modi N. Fluid balance and renal function. Semin Neonatol 2003; 8:
267
26 Modi N. Clinical implications of postnatal alterations in body water
distribution. Semin Neonatol 2003; 8: 301–6
27 Coulthard MG, Hey EN. Effect of varying water intake on renal
function in healthy preterm babies. Arch Dis Child 1985; 60:
614– 20
28 Ertl T, Hadzsiev K, Vincze O, Pytel J, Szabo I, Sulyok E. Hyponatremia
and sensorineural hearing loss in preterm infants. Biol Neonate
2001; 79: 109–12
Handling editor: H. C. Hemmings
545