PEDIATRIC CARDIAC
Tight Glycemic Control Protects the Myocardium
and Reduces Inflammation in Neonatal Heart
Surgery
Dirk Vlasselaers, MD, PhD, Dieter Mesotten, MD, PhD, Lies Langouche, PhD,
Ilse Vanhorebeek, PhD, Ingeborg van den Heuvel, MD, Ilse Milants, RN,
Pieter Wouters, MS, Patrick Wouters, MD, PhD, Bart Meyns, MD, PhD,
Mette Bjerre, PhD, Troels Krarup Hansen, MD, PhD, and
Greet Van den Berghe, MD, PhD
Departments of Intensive Care Medicine and Cardiac Surgery, Katholieke Universiteit Leuven, Belgium; Department of
Anesthesiology, Universiteit Gent, Belgium; and Immunoendocrine Research Unit, Aarhus University Hospital, Denmark
Background. Neonatal cardiac surgery evokes hyper-
glycemia and a systemic inflammatory response. Hyper-
glycemia is associated with intensified inflammation and
adverse outcome in critically ill children and in pediatric
cardiac surgery. Recently we demonstrated that tight
glycemic control (TGC) reduced morbidity and mortality
of critically ill children. Experimental data suggest that
insulin protects the myocardium in the setting of isch-
emia-reperfusion injury, but this benefit could be blunted
by coinciding hyperglycemia. We hypothesized that insu-
lin-titrated TGC, initiated prior to myocardial ischemia
and reperfusion, protects the myocardium and attenuates
the inflammatory response after neonatal cardiac surgery.
Methods. This is a prospective randomized study at a
university hospital. Fourteen neonates were randomized
to intraoperative and postoperative conventional insulin
therapy or TGC. Study endpoints were effects on myo-
cardial damage and function; inflammation, endothelial
activation, and clinical outcome parameters.
T he perioperative period of congenital heart surgery
(CHS) can be challenging as these procedures generate a
systemic inflammatory reaction and endocrine-metabolic
stress responses. Consequently, hyperglycemia is common
during and after CHS [1, 2], which is associated with increased
morbidity and mortality [3–5]. A recent study showed that
tight glycemic control (TGC) significantly reduced morbidity
and mortality in critically ill children [6].
Impaired myocardial function is often present after
neonatal CHS [7] complicating the postoperative course.
It is triggered by multiple mechanisms like surgical
trauma, ischemia and reperfusion (I-R) injury, local and
systemic inflammation, and oxidative stress [8].
Hyperglycemia is associated with poor outcome in
patients with myocardial ischemia, yet a direct causative
Accepted for publication March 30, 2010.
Address correspondence to Dr Vlasselaers, University Hospitals Leuven,
Department of Intensive Care Medicine, Herestraat 49, B-3000 Leuven,
Belgium; e-mail: dirk.vlasselaers@uzleuven.be.
© 2010 by The Society of Thoracic Surgeons
Published by Elsevier Inc
Results. Tight glycemic control significantly reduced
circulating levels of cardiac troponin-I (p ⴝ 0.009), heart
fatty acid-binding protein (p ⴝ 0.01), B-type natriuretic
peptide (p ⴝ 0.002), and the need for vasoactive support
(p ⴝ 0.008). The TGC suppressed the rise of the proin-
flammatory cytokines interleukin-6 (p ⴝ 0.02) and inter-
leukin-8 (p ⴝ 0.05), and reduced the postoperative in-
crease in C-reactive protein (p ⴝ 0.04). Myocardial
concentrations of Akt, endothelial nitric-oxide synthase,
and their phosphorylated forms were not different be-
tween groups.
Conclusions. In neonates undergoing cardiac surgery,
intraoperative and postoperative TGC protects the myo-
cardium and reduces the inflammatory response. This
appears not to be mediated by an early, direct insulin
signaling effect, but may rather be due to independent
effects of preventing hyperglycemia during reperfusion.
(Ann Thorac Surg 2010;90:22–30)
© 2010 by The Society of Thoracic Surgeons
role in aggravating I-R injury is speculative [9]. Experi-
mental data suggest that hyperglycemia may induce
oxidative stress, generate proinflammatory cytokines,
and increase myocardial apoptosis. Insulin, given at the
time of reperfusion, reduces myocardial I-R injury in
animal models, partially by attenuation of apoptosis [10].
This is mediated by phosphatidylinositol-3-kinase (PI3K)
and endothelial nitric oxide synthase (eNOS), and the
concurrent local increase of nitric oxide (NO) production
[11]. However, myocardial protection by insulin may be
abolished by hyperglycemia during reperfusion [12].
A cocktail of glucose, insulin, and potassium (GIK) can
protect the ischemic myocardium in patients with myo-
cardial ischemia and during cardiac surgery [13, 14].
Other trials reported no favorable effect of GIK, possibly
explained by concomitant hyperglycemia [15].
Congenital heart surgery with cardiopulmonary by-
pass (CPB) may impair endothelium-dependent vasodi-
latation [16]. We previously showed that TGC protects
0003-4975/$36.00
doi:10.1016/j.athoracsur.2010.03.093
Ann Thorac Surg VLASSELAERS ET AL 23

PEDIATRIC CARDIAC
2010;90:22–30 GLUCOSE CONTROL PROTECTS THE MYOCARDIUM

volume of the CPB circuit was 220 mL, consisting of a

Abbreviations and Acronyms
BG ⫽ blood glucose
BNP ⫽ brain-natriuretic peptide
CHS ⫽ congenital heart surgery
CIT ⫽ conventional insulin therapy
CO ⫽ cardiac output
CPB ⫽ cardiopulmonary bypass
CRP ⫽ C-reactive protein
cTnI ⫽ cardiac troponin-I
DSC ⫽ delayed sternal closure
d-TGA ⫽ transposition great arteries
eNOS ⫽ endothelial nitric oxide synthase
GIK ⫽ glucose-insulin-potassium
HFABP ⫽ heart fatty acid binding-protein
IL ⫽ interleukin
I-R ⫽ ischemia and reperfusion
NO ⫽ nitric oxide
PI3K ⫽ phosphatidylinositol-3-kinase
(P)ICU ⫽ (pediatric) intensive care unit
PM ⫽ temporary pacemaker
TA ⫽ truncus arteriosus
TGC ⫽ tight glycemic control
the endothelium of adult intensive care unit (ICU) pa-
tients [17]. In an animal model we demonstrated that
hyperglycemia inhibits normal endothelium-dependent
vasorelaxation, which can be prevented by maintaining
normoglycemia [18].
We hypothesized that in neonatal CHS, TGC to age-
adjusted normal fasting levels using insulin infusion,
initiated prior to surgery and continued postoperatively,
protects the myocardium and attenuates the inflamma-
tory and endothelial responses. We investigated a poten-
tial direct insulin-mediated mechanism through the
PI3K/Akt pathway and eNOS activation.
Material and Methods
Patients
This study comprised a predefined subgroup analysis as
part of a prospective randomized controlled trial [6]. A
subgroup of neonates with transposition of the great
arteries (d-TGA) or truncus arteriosus, scheduled for
surgical repair, was prospectively and randomly assigned
to conventional insulin therapy (CIT) or TGC, initiated at
induction of anesthesia and continued throughout sur-
gery and stay in the pediatric (P) ICU. Allocation to
treatment groups was done by sealed envelopes. The
protocol was approved by the hospital ethical committee
and informed consent was obtained from the parents.
The study was registered at ClinicalTrials.gov by the
identifier number NCT00214916.
Perioperative Management
Anesthesia was induced and maintained with sevoflu-
rane, sufentanil, pancuronium, and midazolam. Sedation
in PICU comprised piritramide in continuous infusion
and intermittent bolus injections of midazolam. Priming
mixture of packed red blood cells (hematocrit 30%),
albumin 20% (10 mL/kg), crystalloids, and 30 mg/kg
methylprednisolone. Cardioprotection was delivered by
anterograde crystalloid cardioplegia (Plegivex 40 mL/kg;
Larne, Northern Ireland) and topical cooling. Patients
were cooled to 32°C and received ultrafiltration during
CPB. For weaning of CPB, all neonates received dobut-
amine and milrinone. Hemodynamic support was indi-
vidually adapted by the attending physician based on
clinical evolution.
Experimental Protocol
During surgery all neonates received a baseline infusion
of glucose 20% with 20 units of insulin (Actrapid; Novo
Nordisk A/S, Bagsvaerd, Denmark) and 20 milliequiva-
lent potassium chloride (500 mL, running at 3 mL/kg/
hour). In the TGC group, the principle of a hyperinsu-
linemic-normoglycemic clamp was applied. We used a
modified version of a previously described protocol [19].
In brief, the target blood glucose (BG) range during and
after surgery was set at 50 to 80 mg/dL. This target level
was chosen, incorporating a safety margin, based on
normal fasting BG levels in healthy neonates (31to 60
mg/dL) [20, 21]. A continuous insulin infusion (Actrapid)
was started and continued throughout the surgical pro-
cedure at 0.3 international units kg⫺1 · hour⫺1. If neces-
sary, the speed of the glucose infusion was adjusted to
keep the BG in target. The BG was analyzed every 15
minutes in arterial blood. Postoperatively, BG policy was
continued by insulin-titration to the BG target range in
the presence of a standard intravenous and enteral
feeding protocol. Arterial BG was checked at least hourly
until the BG was in the target range and stable. The BG
control was left to the discretion of the bedside nurse and
checked at least every four hours. Insulin infusions were
continued until discharge from PICU or stopped earlier
when more than two-thirds of caloric intake was admin-
istered as intermittent bolus feeding.
In the CIT group hyperglycemia was only treated with
insulin infusion when BG exceeded 215 mg/dL twice, and
insulin was stopped when BG was below 180 mg/dL.
Neonatal hypoglycemia was defined as BG less than 30
mg/dL [22]. Insulin infusions were stopped when BG was
less than 50 mg/dL and 1mL/kg of a 50% dextrose
solution was given when BG was less than 30 mg/dL.
Myocardial biopsies of the right atrium were taken at
time of cannulation and decannulation to study the
effects of I-R injury on the PI3K/Akt signaling pathway
and eNOS. Small skeletal muscle biopsies from the
musculus abdominis rectus were taken at the start and
end of surgery. All biopsies were snap frozen in liquid
nitrogen and stored at minus 80°C.
Biochemical Analyses
Blood glucose was exclusively determined with the ABL
715 blood gas analyzer (Radiometer, Bronshoj, Denmark)
[23]. Serum insulin, C-peptide, N-terminal pro brain
natriuretic peptide (NT-proBNP), cardiac troponin-I
(cTnI), heart fatty acid binding-protein (HFABP), C-reac-
 24 VLASSELAERS ET AL Ann Thorac Surg
PEDIATRIC CARDIAC

GLUCOSE CONTROL PROTECTS THE MYOCARDIUM 2010;90:22–30

Table 1. Inotrope Score Clinical Endpoints

0 points No inotropes or vasodilators
1 point Dobutamine ⱕ 3 ␮g/kg/min and(or) milrinone
ⱕ 0.5 ␮g/kg/minute
2 points Dobutamine ⬎ 3 and ⱕ 6 ␮g/kg/minute and(or)
milrinone ⬎ 0.5 ␮g/kg/minute
3 points Dobutamine ⬎6 and ⱕ10 ␮g/kg/minute and(or)
(nor-)adrenaline ⱕ 0.15 ␮g/kg/minute and(or)
milrinone: ⬎ 0.5 ␮g/kg/minute
4 points Dobutamine: ⬎ 10 ␮g/kg/min and(or)
(nor-)adrenaline ⬎ 0.15 ␮g/kg/minute and(or)
milrinone ⬎ 0.5 ␮g/kg/minute
5 points LVAD, RVAD, BiVAD, or ECMO
ECMO ⫽ extracorporeal membrane oxygenation; (L) (R) (Bi)VAD ⫽
left, right, biventricular assist device.
tive protein, intercellular adhesion molecule-1, E-
selectin, and serum concentrations of cytokines (interleu-
kin [IL]-1beta, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, tumor
necrosis factor-alpha) were all analyzed according the
manufacturers’ directions. Serum mannose binding lec-
tin and membrane attack complex were measured using
in-house developed assays [24, 25].
Protein Levels of Signal Transduction Pathways in
Tissue Biopsies
Homogenates and immunoprecipitates of tissue samples
were immunoblotted with specific Ab against eNOS,
phospho-eNOS, Akt, and phosphor-Akt and analyzed
with Image Master Software (Amersham Biosciences,
Piscataway, NJ).
Hemodynamic performance was assessed by the need for
temporary pacing (PM), incidence of delayed sternal
closure (DSC), and a predefined inotrope score (Table 1).
We also evaluated time to extubation, days in PICU and
hospital as well as prolonged PICU and hospital stay.
Statistical Analysis
Baseline and outcome variables were compared using
repeated measures analysis of variance (nonnormally
distributed data are log-transformed prior to analysis of
variance testing), the Student’s t test, or the Mann-
Whitney U test. Data are presented as means (⫾ SD) or
medians (25th to 75th percentiles). Two-sided p values
less than 0.05 were considered statistically significant.
Statistical tests were performed with Statview version
5.0.1 (SAS Institute, Inc, Cary, NC).
Results
Demography and Clinical Outcome
Fourteen neonates were included, 7 in each group. There
were no differences between groups in baseline charac-
teristics and CPB variables. Median time to extubation
tended to be shorter in the TGC group. There were no
significant differences in renal function and duration of
stay in PICU and the hospital. There were no early or late
deaths. The lower incidence of DSC (29% vs 0%) and
need for PM (57% vs 14%) in the first 48 hours in the TGC
group was not significant (Table 2).

Table 2. Demography and Clinical Outcome

Variables CIT TGC p

Weight (kg) 3.00 (2.98–3.48) 3.36 (3.02–3.82) 0.62

Diagnosis (n):
TGA/IVS 4 5
TGA/VSD 1 1
TA 2 1
Aorta clamp time (minutes) 77 (64–106) 80 (62–83) 0.62
CPB duration (minutes) 113 (100–166) 112 (108–123) 0.80
Reperfusion (minutes) 30 (27–55) 31 (17–41) 0.45
Ultrafiltration (mL) 250 (200–287) 250 (165–286) 0.71
PICU stay (day) 9 (5–12) 7 (5–11) 0.62
Prolonged PICU stay (⬎9 days) 71% 28% 0.28
Hospital stay (day) 13 (8–23) 10 (7–16) 0.45
Prolonged hosp stay (⬎13 days) 71% 28% 0.57
Time to extubation (hours) 141 (81–181) 74 (35–95) 0.06
DSC 2 (29%) 0 (0%) 0.22
PM 4 (57%) 1 (14%) 0.13

Data are expressed as median (P25–P75) or as numbers (proportions) (diagnosis DSC, PM).
Prolonged stay: longer than median stay for the CIT group.
CIT ⫽ conventional insulin therapy; CPB ⫽ cardiopulmonary bypass; DSC ⫽ delayed sternal closure; IVS ⫽ intact ventricular septum;
PICU ⫽ pediatric intensive care unit; PM ⫽ temporary pacemaker; TA ⫽ truncus arteriosus; TGA ⫽ transposition of the great arteries;
TGC ⫽ tight glycemic control group; VSD ⫽ ventricular septal defect.
Ann Thorac Surg
2010;90:22–30
Glucose and Insulin
In the TGC group BG was within target range and
significantly different from the CIT group. During sur-
gery there were no hypoglycemic events in both groups.
In the postoperative period, 4 hypoglycemic events
(range, 23 to 30 mg/dL) occurred in 2 patients (28%) in the
TGC group, representing 0.7% of all BG measurements
No hypoglycemic events occurred in the CIT group. Glu-
cose administration was similar in both groups, while
insulin infusion significantly differed. Infusion of exoge-
nous insulin in the TGC group suppressed the endogenous
production of insulin and C-peptide. The endogenous pro-
duction of insulin with CIT, reflected by increased levels of
C-peptide, was insufficient to counter hyperglycemia, re-
flecting insulin resistance (Table 3).
Myocardial Function and Damage
Tight glycemic control reduced blood lactate and the
need for hemodynamic support, reflected by the inotrope
score. The cTnI, HFABP, and NT-proBNP release were
also significantly lowered by TGC (Table 3). On postop-
erative day 1 a significant correlation existed between
cTnI and HFABP, and NT-proBNP (Fig 1).
Inflammation
The rise in C-reactive protein in the postoperative period
was significantly lower with TGC. A similar dampening
of a rise in IL-6 and IL-8 was observed (Table 3). The time
course of tumor necrosis factor-alpha, other measured
cytokines, serum mannose binding lectin, and serum
membrane attack complex was not significantly different
between groups.
Endothelial Activation
The measured changes of E-selectin and intercellular
adhesion molecule-1 levels in serum did not differ be-
tween the two groups (data not shown).
Protein Levels of Akt and eNOS
In atrial biopsies, myocardial tissue exposed to I-R, TGC
did not change concentrations of Akt, eNOS, and their
phosphorylated forms as compared with CIT. In skeletal
muscle biopsies, tissue not exposed to I-R during the
surgical procedure, levels of Akt, pAkt, eNOS, and
p-eNOS, were likewise similar for both groups (Fig 2).
Comment
This study demonstrates that intraoperative and postop-
erative TGC in neonatal CHS protected the myocardium
and attenuated the inflammatory response evoked by the
surgical procedure, which may favorably affect clinical
outcome. This did not appear to be mediated by direct
insulin signaling effects, but to effects of preventing
hyperglycemia during reperfusion.
Implementing TGC carries the risk of evoking bio-
chemical hypoglycemia. We could not identify any dele-
terious symptoms of these hypoglycemic events through-
out stay in PICU. In order to exclude possible longer term
VLASSELAERS ET AL 25
PEDIATRIC CARDIAC
GLUCOSE CONTROL PROTECTS THE MYOCARDIUM
effects, a cardiac, neurocognitive, and developmental
follow-up study of these patients is currently ongoing.
High exogenous insulin dose lowered BG to normal
fasting levels and suppressed endogenous production of
insulin, demonstrated by reduced C-peptide. From day 2,
circulating levels of insulin were comparable in both groups
despite a significant difference in BG, suggesting that insu-
lin resistance remained more pronounced during CIT and
indicating that early prevention of hyperglycemia by exog-
enous insulin results in maintained insulin sensitivity.
Our data strongly suggest that myocardial injury
evoked by I-R was significantly reduced and myocardial
function was better preserved with TGC. Release of cTnI
was lower with TGC. Because surgical trauma and renal
function between both groups were comparable, allow-
ing to infer a similar cTnI clearance, reduced cTnI levels
likely reflect less myocardial damage. This finding may
be relevant for longer term outcome as cTnI levels early
after CHS are a powerful independent predictor of out-
come [26, 27]. A similar myocardial protective effect of
TGC was described in adult cardiac surgery patients [28].
Circulating levels of HFABP were also reduced by TGC.
In CHS, HFABP after aortic declamping reflects myocar-
dial damage and is associated with postoperative clinical
outcome [29]. Brain-natriuretic peptide is released in
response to ventricular dysfunction and increased wall
stress. Elevated postoperative BNP levels are also asso-
ciated with postoperative cardiac dysfunction [30], pro-
longed hospital stay, and one-year mortality in adult
cardiac surgery patients [31]. After CHS, BNP levels
correlate closely with ventricular function [32], duration
of mechanical ventilation, and postoperative low cardiac
output [33]. Moreover, neonates after arterial switch with
higher postoperative BNP values appear to have a more
complicated evolution, as reflected by prolonged me-
chanical ventilation, inotropic support, and PICU stay
[34]. Tight glycemic control reduced the rise in BNP,
possibly related to the observed trends for improved
clinical outcome like inotrope score and earlier extuba-
tion. However, in view of the small sample size, the
clinical implications remain to be confirmed.
Postoperative PM is applied for treating arrhythmias in
anticipation of recovery of normal rhythm or definitive
pacemaker implantation, or to improve CO by increasing
heart rate. Neonates have little inotropic reserve due to
the decreased density of contractile elements and re-
spond less to preload because of decreased ventricular
compliance [35]. Therefore, they are particularly depen-
dent on heart rate to increase CO. The increased
incidence of early PM with CIT reflects decreased
myocardial function and subsequent need for aug-
menting CO. In addition, TGC possibly decreases local
inflammation and myocardial edema causing less con-
duction disturbances.
Delayed sternal closure is a common strategy in CHS
to avoid further compromising myocardial function. The
higher incidence of DSC in the CIT group, albeit not
significant, possibly reflects a higher incidence of myo-
cardial dysfunction.
Blood lactate and its time course in the immediate
 26 VLASSELAERS ET AL Ann Thorac Surg
PEDIATRIC CARDIAC

GLUCOSE CONTROL PROTECTS THE MYOCARDIUM 2010;90:22–30

Table 3. Data of Metabolic, Cardiac, and Inflammatory Parameters at Different Times

Within Between Residual
Time CIT IIT Groups Groups Value

Blood glucose (mg/dL) Perop 215 (208–244) 102 (93–105)

Day 1 137 (111–149) 72 (70–80)
Day 2 99 (94–116) 76 (70–87)
LD ICU 99 (91–106) 87 (82–90) ⬍0.001 ⬍0.001 ⬍0.001
IV glucose (g/kg BW) Perop 2.5 (1.9–2.9) 1.3 (1.2–1.6)
Day 1 8.2 (6.7–8.4) 7.6 (7.2–8.8)
Day 2 8.2 (6.9–8.4) 10.2 (9.2–10.9)
LD ICU 6.4 (5.6–10.8) 11.2 (10.3–13.4) ⬍0.001 0.12 0.015
Insulin serum (mU/L) Day 1 60 (19–83) 110 (41–337)
Day 2 35 (19–64) 38 (19–65)
Day 3 16 (11–22) 36 (27–55)
LD ICU 17 (10–22) 32 (26–38) 0.003 0.12 0.07
C-peptide (pmol/L) Day 1 1.38 (0.76–1.79) 0.30 (0.20–0.66)
Day 2 1.4 (1.05–1.53) 0.13 (0.10–0.32)
Day 3 1.0 (0.8–1.3) 0.08 (0.06–0.10)
LD ICU 0.45 (0.40–0.59) 0.26 (0.09–0.37) 0.035 ⬍0.0001 0.008
Insulin infusion (U/day) perop 0 (0–0) 7.7 (7.0–9.2)
Day 1 0 (0–0.2) 4.2 (2.9–5.1)
Day 2 0 (0–0) 3.7 (2.6–5.1)
LD ICU 0 (0–0) 2.0 (1.0–2.6) ⬍0.001 ⬍0.001 ⬍0.001
Inotrope score ICU 2 (2–3.8) 2 (2–3)
⫹12h 2 (2–3) 1 (1–1.8)
⫹24h 2 (1.3–2.8) 1 (1–1)
⫹36h 2 (1–2) 1 (1–1)
⫹48h 2 (1.3–2) 1 (1–1) ⬍0.001 0.01 0.008
Lactate (mmol/L) ICU 3 (1.8–4.3) 1.6 (1–2.5)
⫹12h 1.9 (1.7–2.2) 1.5 (1.3–1.9)
⫹24h 1.5 (1.3–2.2) 1.4 (1.3–1.6)
⫹36h 1.2 (0.9–1.3) 1.1 (1–1.2)
⫹48h 0.8 (0.7–1.2) 0.9 (0.7–1) ⬍0.0001 0.03 0.05
Troponin I (␮g/L) ICU 31.80 (19.15–67.13) 17.55 (10.30–21.30)
⫹12h 25.00 (22.30–32.05) 10.60 (9.43–16.15)
⫹24h 28.50 (20.35–31.88) 9.55 (7.70–14.10)
⫹36h 22.90 (16.15–29.95) 6.80 (6.13–9.73)
⫹48h 16.70 (10.85–24.82 5.80 (5.28–8.90) ⬍0.0001 0.03 0.009
HFABP (ng/mL) day 1 90872 (62867–220071) 27964 (19051–43436)
day 2 33496 (27996–58162) 12840 (7971–21109)
day 3 15858 (14120–18259) 4801 (2991–5521)
LD ICU 2855 (1628–8650) 1688 (1082–4118) ⬍0.0001 0.006 0.01
NT-proBNP (ng/L) day 1 62002 (22404–90848) 16971 (14277–21758)
day 2 31064 (19355–38399) 11750 (6117–17765)
day 3 20606 (12185–24635) 10376 (9282–14802)
LD ICU 12570 (7816–16683) 8803 (5358–20303) 0.0002 0.04 0.002
C-reactive protein (mg/L) day 1 13 (9–31) 11 (5–16)
day 2 44 (31–81) 28 (10–52)
day 3 67 (40–116) 30 (6–48)
LD ICU 16 (14–22) 14 (10–16) ⬍0.0001 0.04 0.03
IL-6 (% change) day 1 2721 (1446–4835) 487 (51–1359)
day 2 5070 (1139–5516) 358 (93–905)
day 3 1682 (582–1962) 294 (27–431)
LD ICU 418 (257–711) 50 (–74–396) ⬍0.0001 0.01 0.02

Continued
Ann Thorac Surg VLASSELAERS ET AL 27

PEDIATRIC CARDIAC
2010;90:22–30 GLUCOSE CONTROL PROTECTS THE MYOCARDIUM

Table 3. Continued
Within Between Residual
Time CIT IIT Groups Groups Value

IL-8 (% change) day 1 111 (60–185) –26 (–33–105)

day 2 96 (76–178) 50 (1–130)
day 3 91 (46–150) 42 (–33–99)
LD ICU –13 (–15–129) 43 (–26–52) ⬍0.001 0.06 0.05

All analyses were done by repeated measures analysis of variance.

p values for differences within and between groups and residual value.
BW ⫽ body weight; CIT ⫽ conventional insulin therapy; h ⫽ hours; HFABP ⫽ heart fatty acid binding-protein; IIT ⫽ intensive insulin
therapy; IL ⫽ interleukin; IV ⫽ intravenous; LD ICU ⫽ last day intensive care unit; NT-proBNP ⫽ N-terminal prohormone brain
natriuretic peptide.

postoperative period correlate with outcome parameters
such as inotrope score, length of stay, and mortality [36].
The observed lower lactate levels with TGC could reflect
a better hemodynamic profile.
The myocardial protective effect of TGC in the context
Fig 1. (A) (B). Correlation between markers of myocardial damage
and function. (D1 ⫽ first postoperative day regression line with
95% confidence interval; cTnI ⫽ cardiac troponin-I; HFABP ⫽ heart
fatty acid binding–protein; NT–proBNP ⫽ N–terminal pro brain
natriuretic peptide.) ( ⫽ tight glycemic control [TGC]; e ⫽ con-
ventional insulin therapy [CIT].)
of I-R may at least partially be explained by an attenuated
inflammatory response. Tight glycemic control signifi-
cantly reduced inflammation, as indicated by lower C-re-
active protein. The CHS, CPB, and the associated I-R
injury cause an inflammatory response. Newborns are at
increased risk, and higher levels of IL-6 and IL-8 after
arterial switch correlate with myocardial dysfunction and
damage, reflected by higher serum levels of troponin
[37]. Furthermore, concentrations of IL-6 and IL-8 after
CHS are associated with the degree of postoperative
organ dysfunction and the need for medical intervention,
including inotropic support [38]. Tight glycemic control
suppressed the early release of IL-6 and IL-8 in this
study. As TGC also decreased cTnI, HFABP, and NT-
proBNP, this could reflect reduced myocardial damage
and better preservation of myocardial function provoked
by an attenuated inflammatory response. Indeed, proin-
flammatory cytokines are implicated in decreasing con-
tractility and uncoupling beta-adrenergic receptors [39].
A similar effect of intraoperative TGC on the inflamma-
tory response, reflected by decreased levels of IL-6 and
IL-8, was described in adults [40].
As a potential mechanism for cardioprotection with
insulin-titrated TGC, we postulated a direct action of
insulin, through the insulin receptor-mediated activation
of PI3K/Akt, a pathway known to evoke protection
against myocardial I-R-injury. This pathway increases
local nitrous oxide production resulting from phosphor-
ylation of eNOS. Experimental animal models suggested
insulin as the active component of GIK against myocar-
dial I-R-injury [11]. However, hyperglycemia not only
exacerbates myocardial I-R-injury but may also counter-
act any cardioprotective effect of GIK, due to myocardial
PI3K/Akt inactivation [12]. In the cardiac biopsies we did
not observe an effect on PI3K/Akt or eNOS. Possibly this
is explained by the timing of the biopsy; 30 minutes after
commencing myocardial reperfusion may have been too
early to detect an effect on the studied pathway in
contrast with the animal experiments where myocardial
tissue was reperfused for 4 hours before being analyzed.
Whether eNOS or PI3K/Akt played a role beyond the
studied 30 minutes reperfusion remains unknown. Alter-
natively, the myocardial protection observed in this study
was not mediated through this insulin-receptor mediated
 28 VLASSELAERS ET AL Ann Thorac Surg
PEDIATRIC CARDIAC

GLUCOSE CONTROL PROTECTS THE MYOCARDIUM 2010;90:22–30

Fig 2. The Akt and endothelial

nitric oxide synthase (eNOS) pro-
tein (P) levels in skeletal muscle
(A) and myocardium (B). Data are
presented as mean (⫾ standard
error). For each signaling mole-
cule, a representative blot is added
with pre- and post-ischemia-
reperfusion (I-R) samples. ( ⫽
tight glycemic control [TGC]; e ⫽
conventional insulin therapy
[CIT].)

pathway and may point to prevention of distinct glucose
toxicity.
In conclusion, this exploratory study provided arguments
for a protective effect on the myocardium of insulin-titrated
TGC during and after CHS. An attenuation of the early
inflammatory response to CHS may have contributed to
this myocardial protection. Further study is needed to
unravel the molecular mechanisms involved.
Study Limitations
First, the study was not performed in a blinded fashion,
as TGC requires careful BG monitoring. Second, the
small sample size and the selection of CHS make that the
results remain preliminary and needs confirmation in a
larger, more heterogeneous population. To avoid poten-
tial interference of large differences in baseline charac-
teristics and surgical procedures, we chose to perform
this study in a population with comparable baseline and
CPB characteristics and surgical trauma. Third, absence
of validated serial cardiac output measurements or echo-
cardiographic exams focusing on myocardial function
limits the conclusions regarding myocardial perfor-
mance. Finally, delaying the timing of myocardial tissue
harvesting during reperfusion would have substantially
increased the duration of the surgical procedure, which is
ethically unacceptable.
Work was supported by the Fund for Scientific Research (FWO),
Belgium (G.0533.06) and the Research Council of Katholieke
Universiteit Leuven (GOA2007/14). Dirk Vlasselaers is a clinical
PhD Fellow (FWO), Dieter Mesotten a Postdoctoral Fellow of the
Clinical Research Fund University Hospital Leuven, Lies
Langouche a Postdoctoral Fellow (FWO), and Ilse Vanhorebeek
a Postdoctoral Fellow of Research Fund KUL. Greet Van den
Berghe, MD, PhD receives research financing through the
Methusalem program funded by the Flemish Government.
References
1. Anand K, Hansen D, Hickey P. Hormonal-metabolic stress
responses in neonates undergoing cardiac surgery. Anesthe-
siology 1990;73:661– 670.
Ann Thorac Surg
2010;90:22–30
2. Ballweg J, Wernovsky G, Ittenbach R, et al. Hyperglycemia
after infant cardiac surgery does not adversely impact neu-
rodevelopmental outcome. Ann Thorac Surg 2007;84:2052– 8.
3. Yates A, Dyke P 2nd, Taeed R, et al. Hyperglycemia is a
marker for poor outcome in the postoperative pediatric
cardiac patient. Pediatr Crit Care Med 2006;7:351–5.
4. Falcao G, Ulate K, Kouzekanani K, Bielefeld M, Morales J,
Rotta A. Impact of postoperative hyperglycemia following
surgical repair of congenital cardiac defects. Pediatr Cardiol
2008;29:628 –36.
5. Polito A, Thiagarajan R, Laussen P, et al. Association be-
tween intraoperative and early postoperative glucose levels
and adverse outcomes after complex congenital heart sur-
gery. Circulation 2008;118:2235– 42.
6. Vlasselaers D, Milants I, Desmet L, et al. Intensive insulin
therapy for patients in paediatric intensive care: a prospec-
tive, randomised controlled study. Lancet 2009;373:547–56.
7. Wernovsky G, Wypij D, Jonas R, et al. Postoperative course
and hemodynamic profile after the arterial switch operation
in neonates and infants. A comparison of low-flow cardio-
pulmonary bypass and circulatory arrest. Circulation 1995;
92:2226 –35.
8. Anselmi A, Abbate A, Girola F, et al. Myocardial ischemia,
stunning, inflammation, and apoptosis during cardiac sur-
gery: a review of evidence. Eur J Cardiothorac Surg 2004;25:
304 –11.
9. Deedwania P, Kosiborod M, Barrett E, et al. Hyperglycemia
and acute coronary syndrome. A scientific statement from
the American heart association diabetes committee of the
council on nutrition, physical activity, and metabolism. Cir-
culation 2008;117:1610 –9.
10. Jonassen AK, Brar BK, Mjøs OD, Sack MN, Latchman DS,
Yellon DM. Insulin administered at reoxygenation exerts a
cardioprotective effect in myocytes by a possible anti-
apoptotic mechanism. J Mol Cell Cardiol 2000;32:757– 64.
11. Gao F, Gao E, Yue T-L, et al. Nitric oxide mediates the
antiapoptotic effect of insulin in myocardial ischemia-
reperfusion: the roles of PI3-kinase, Akt, and endothelial
nitric oxide synthase phosphorylation. Circulation 2002;105:
1497–502.
12. Su H, Sun X, Ma H, et al. Acute hyperglycemia exacerbates
myocardial ischemia/reperfusion injury and blunts cardio-
protective effect of GIK. Am J Physiol Endocrinol Metab
2007;293:E629 –35.
13. Bothe W, Olschewski M, Beyersdorf F, Doenst T. Glucose-
insulin potassium in cardiac surgery: a meta-analysis. Ann
Thorac Surg 2004;78:1650 –7.
14. Kloner R, Nesto R. Glucose-insulin-potassium for acute
myocardial infarction: continuing controversy over cardio-
protection. Circulation 2008;117:2523–33.
15. Van den Berghe G. Insulin vs strict blood glucose control to
achieve a survival benefit after AMI? Eur Heart J 2005;26:
639 – 41.
16. Wessel D, Adatia I, Giglia T, Thompson J, Kulik T. Use of
inhaled nitric oxide and acetylcholine in the evaluation of
pulmonary hypertension and endothelial function after cardio-
pulmonary bypass. Circulation 1993;88(5 Pt 1):2128 –38.
17. Langouche L, Vanhorebeek I, Vlasselaers D, et al. Intensive
insulin therapy protects the endothelium of critically ill
patients. J Clin Invest 2005;115:2277– 86.
18. Ellger B, Debaveye Y, Vanhorebeek I, et al. Survival benefits
of intensive insulin therapy in critical illness: impact of
maintaining normoglycemia versus glycemia-independent
actions of insulin. Diabetes 2006;55:1096 –105.
19. Carvalho G, Moore A, Qizilbash B, Lachapelle K, Schricker
T. Maintenance of normoglycemia during cardiac surgery.
Anesth Analg 2004;99:319 –24.
20. Mosby’s Diagnostic and Laboratory Test Reference, 6th
edition. New York: Mosby, Elsevier Science; 2003:457.
21. Brown T, Connelly J, Dunlop M, McDougall P, Tibballs J.
Fasting plasma glucose in children. Aust Paediatr J 1980;16:
28 –9.
VLASSELAERS ET AL 29
PEDIATRIC CARDIAC
GLUCOSE CONTROL PROTECTS THE MYOCARDIUM
22. Cornblath M, Hawdon JM, Williams AF, et al. Controversies
regarding definition of neonatal hypoglycemia: suggested
operational thresholds. Pediatrics 2000;105:1141–5.
23. Scott M, Bruns D, Boyd J, Sacks D. Tight glucose control in
the intensive care unit: are glucose meters up to the task?
Clinical Chemistry 2009;55:18 –20.
24. Hansen TK, Thiel S, Wouters PJ, Christiansen JS, Van den
Berghe G. Intensive insulin therapy exerts antiinflammatory
effects in critically ill patients and counteracts the adverse
effect of low mannose-binding lectin levels. J Clin Endocri-
nol Metab 2003;88:1082– 8.
25. Haahr-Pedersen S, Bjerre M, Flyvbjerg A, et al. Level of
complement activity predicts cardiac dysfunction after acute
myocardial infarction treated with primary percutaneous
intervention. J Invasive Cardiol 2009;21:13–9.
26. Mildh L, Pettilä V, Sairanen H, Rautiainen. Cardiac troponin
T levels for risk stratification in pediatric open heart surgery.
Ann Thorac Surg 2006;82:1643–9.
27. Imura H, Modi P, Pawade A, et al. Cardiac troponin I in
neonates undergoing the arterial switch operation. Ann
Thorac Surg 2002;74:1998 –2002.
28. Albacker T, Carvalho G, Schricker T, Lachapelle K. Myocar-
dial protection during elective coronary artery bypass graft-
ing using high-dose insulin therapy. Ann Thorac Surg 2007;
84:1920 –7.
29. Hasegawa T, Yoshimura N, Oka S, Ootaki Y, Toyoda Y,
Yamaguchi M. Evaluation of heart fatty acid-binding protein as
a rapid indicator for assessment of myocardial damage in
pediatric cardiac surgery. J Thorac Cardiovasc Surg 2004;127:
1697–702.
30. Provenchère S, Berroeta C, Reynaud C, et al. Plasma brain
natriuretic peptide and cardiac troponin I concentrations
after adult cardiac surgery: association with postoperative
cardiac dysfunction and 1-year mortality. Crit Care Med
2006;34:995–1000.
31. Hutfless R, Kazanegra R, Madani M, et al. Utility of B-type
natriuretic peptide in predicting postoperative complica-
tions and outcomes in patients undergoing heart surgery.
J Am Coll Cardiol 2004;43:1873–9.
32. Koch A, Zink S, Singer H. B-type natriuretic peptide in
paediatric patients with congenital heart disease. Eur Heart J
2006;27:861– 6.
33. Shih C-Y, Sapru A, Oishi P, et al. Alterations in plasma
B-type natriuretic peptide levels after repair of congenital
heart defects: a potential perioperative marker. J Thorac
Cardiovasc Surg 2006;131:632– 8.
34. Cannesson M, Biona C, Gostoli B, et al. Time course and
prognostic value of plasma B-type natriuretic peptide con-
centrations in neonates undergoing the arterial switch oper-
ation. Anesth Analg 2007;104:1059 – 65.
35. Epstein D, Wetzel R. Cardiovascular physiology and shock.
In: Critical heart disease in infants and children, 2nd edition.
New York: Mosby, Elsevier; 2006:17–20.
36. Basaran M, Sever K, Kafali E, et al. Serum lactate level has
prognostic significance after pediatric cardiac surgery. J Car-
diothorac Vasc Anesth 2006;20:43–7.
37. Hövels-Gürich H, Vazquez-Jimenez J, Silvestri A, et al.
Production of proinflammatory cytokines and myocardial
dysfunction after arterial switch operation in neonates with
transposition of the great arteries. J Thorac Cardiovasc Surg
2002;124:811–20.
38. Madhok A, Ojamaa K, Haridas V, Parnell V, Pahwa S,
Chowdhury D. Cytokine response in children undergoing
surgery for congenital heart disease. Pediatr Cardiol 2006;27:
408 –13.
39. Blum A, Miller H. Role of cytokines in heart failure. Am
Heart J 1998;135(2 Pt 1):181– 6.
40. Albacker T, Carvalho G, Schricker T, Lachapelle K. High-
dose insulin therapy attenuates systemic inflammatory re-
sponse in coronary artery bypass grafting patients. Ann
Thorac Surg 2008;86:20 – 8.