Pediatric Diabetes 2014: 15: 75–83 © 2014 John Wiley & Sons A/S.

doi: 10.1111/pedi.12134 Published by John Wiley & Sons Ltd.

All rights reserved
Pediatric Diabetes

Review Article

Tight glucose control in critically ill

children – a systematic review and
meta-analysis

Srinivasan V, Agus MSD. Tight glucose control in critically ill children – a Vijay Srinivasana and
systematic review and meta-analysis. Michael S.D. Agusb
Pediatric Diabetes 2014: 15: 75–83. a Department of Anesthesiology and

Background: It is unclear if tight glucose control (TGC) with intensive insulin Critical Care Medicine, Children’s
therapy (IIT) can improve outcomes in critically ill children admitted to the Hospital of Philadelphia and Perelman
School of Medicine, University of
intensive care unit (ICU). The objective of this systematic review and
Pennsylvania, Philadelphia, PA, USA;
meta-analysis is to describe the benefits and risks of TGC with IIT in critically and b Division of Medicine Critical Care,
ill children and explore differences between published studies. Department of Medicine, Boston
Methods: Prospective randomized controlled trials (RCTs) of TGC with IIT Children’s Hospital and Harvard
in critically ill children admitted to the ICU were identified through a search Medical School, Boston, MA, USA
of MEDLINE, PubMed, EMBASE, Scopus, ISI Web of Science and
Cochrane Database of Systematic Reviews as well as detailed citation review
of relevant primary and review articles. RCTs of TGC with IIT in critically ill Key words: children – critical illness –
adults and preterm neonates were excluded. Data on study design and setting, glucose control – intensive insulin
sample size, incidence of hypoglycemia, incidence of acquired infection, and therapy – stress hyperglycemia
30-day mortality were abstracted. Meta-analytic techniques were used for
analysis of outcomes including 30-day mortality, acquired infection, and Corresponding author: Michael S.D.
incidence of hypoglycemia. Agus, MD,
Division of Medicine Critical Care,
Results: We identified four RCTs of TGC with IIT in critically ill children
Department of Medicine, Boston
that included 3288 subjects. Overall, TGC with IIT did not result in a decrease Children’s Hospital,
in 30-day mortality [odds ratio (OR): 0.79; 95% confidence interval (CI): Harvard Medical School,
0.55–1.15, p = 0.22]. TGC with IIT was associated with decrease in acquired Boston,
infection (OR: 0.76; 95% CI: 0.59–0.99, p = 0.04). TGC with IIT was also MA,
associated with significant increase in hypoglycemia (OR: 6.14; 95% CI: USA.
2.74–13.78, p < 0.001). Tel: (617) 355-5849;
Conclusions: TGC with IIT does not result in decrease in 30-day mortality, fax: (888) 883-9238;
but appears to reduce acquired infection in critically ill children. However, e-mail:
TGC with IIT is associated with higher incidence of hypoglycemia. Large michael.agus@childrens.harvard.edu
multi-center studies of TGC with IIT using continuous glucose monitoring in Submitted 17 January 2014.
critically ill children are needed to determine if this strategy can definitively Accepted for publication 23 January
improve clinical outcomes in this population without increasing hypoglycemia. 2014

Critically ill children often manifest stress hyper-
glycemia (SH), including those with previously normal
glucose metabolism (1–7). An estimated 49–72% of
critically ill children experience blood glucose (BG)
concentrations >150 mg/dL (>8.3 mmol/L) whereas
BG concentrations >200 mg/dL (>11 mmol/L) occur
in as many as 20–35% of critically ill children
(1–7). SH typically develops via a combination of
increased gluconeogenesis (relative to glucose uptake
and clearance) and development of insulin resistance
(8). Both these mechanisms are mediated by an increase
in counter-regulatory hormones (catecholamines, cor-
tisol, glucagon, and growth hormone) and proinflam-
matory cytokines (9, 10). SH may also develop because
of impaired pancreatic beta-cell function with decrease
in insulin secretion (11). SH during critical illness there-
fore results in conditions of readily available glucose
as a fuel for vital organs in the body. Whereas SH
75
Srinivasan et al.
may be adaptive and favor survival in the acute phase
of critical illness when metabolic demands are high,
persistence of SH during the chronic phase of critical
illness may be harmful because of increased oxida-
tive damage with potentiation of the proinflammatory
response and impairment of coagulation pathways,
cellular repair and tissue healing (12). Historically, SH
was ignored by practitioners and usually justified as an
adaptive response to stress (13, 14). However, several
studies in critically ill children have challenged this
belief by showing the association of SH with increase
in morbidity and mortality across a variety of disease
states (1–7, 15–21). Consequently, the practice of tight
glucose control (TGC) with intensive insulin therapy
(IIT) has emerged as a plausible strategy to improve
outcomes in critically ill children.
Studies of TGC with IIT in critically ill adults were
initially very promising with remarkable improvements
in survival as well as reduction in various morbidities
(22, 23). However, subsequent studies of TGC with
IIT observed outcomes that were no better or worse
than conventional care (24–27). Notably, all studies
of TGC with IIT in critically ill adults observed
marked increases in hypoglycemia prompting concerns
about the impact of such a strategy on long-term and
neurologic outcomes (22, 24–27). As a result, the initial
enthusiasm to embrace TGC with IIT as ‘a magic bullet
therapy’ has given way to a more careful and nuanced
approach in the adult critical care community (28). The
pediatric critical community has embarked cautiously
on this approach because of diagnostic considerations,
practice variations, and developmental issues specific
to children that are significantly different from adults.
In 2005, Pham and colleagues retrospectively
reviewed their practice of TGC with IIT in 33 children
with severe burns in comparison to 31 historical
controls treated conventionally and showed a decrease
in infection rates and improved survival with TGC
using IIT (29). In 2009, Vlasselaers and colleagues
published their single center experience of TGC
with IIT in 700 infants and children predominantly
recovering from cardiac surgery (30). They showed
reductions in inflammation and length of ICU stay
as well as ICU mortality, but at the cost of a
significant increase in hypoglycemia. In 2010, Jeschke
and colleagues studied TGC with IIT in 239 severely
burned children and observed that TGC with IIT
improved postburns morbidity with a non-significant
decrease in mortality (31). In 2012, Agus and colleagues
studied TGC with IIT in 980 children undergoing
cardiac surgery and did not observe any reduction in
morbidities or mortality (32). However, they showed
that TGC with IIT could be safely achieved with a
very low incidence of hypoglycemia using continuous
glucose monitoring technologies together with an
explicit insulin dosing algorithm. In 2013, Macrae
76
and colleagues observed that TGC with IIT in 1369
critically ill children from 13 centers did not lead to
improvement in their primary endpoint but did shorten
hospital stay in one subgroup, and resulted in an
increased incidence of hypoglycemia (33).
The objective of this systematic review and meta-
analysis is to describe the impact of TGC with IIT
compared to conventional control of BG on key
outcomes in critically ill children and explore important
differences between published studies.
Methods
Identification of trials
All relevant randomized controlled trials (RCTs)
in critically ill children that compared the impact
of TGC with IIT to conventional control of
BG on mortality outcomes were identified. The
primary outcome measure was 30-day mortality
with summary effects estimate expressed as odds
ratio (OR). Secondary outcome measures included
the incidence of hospital acquired infections and
incidence of hypoglycemia defined as BG < 40 mg/dL
(<2.2 mmol/L). We systematically searched the
National Library of Medicine’s MEDLINE database
for relevant studies published between 1946 and
2014 using the following medical subject headings
and keywords: insulin, glucose, hyperglycemia,
hypoglycemia, intensive care, critical illness, strict
glucose control, intensive insulin therapy, children,
child, and RCT. The search strategy is depicted
in Fig. 1. Additionally, we also searched PubMed,
EMBASE, Scopus, ISI Web of Science and Cochrane
Database of Systematic Reviews for relevant studies.
We performed a detailed review of citations of all
relevant primary and review articles. Studies performed
in adults and preterm neonates as well as studies that
were not RCTs were excluded from this meta-analysis.
This meta-analysis was performed according to the
guidelines proposed by the Preferred Reporting Items
for Systematic Reviews and Meta-Analyses (PRISMA)
group (34).
Data extraction and assessment of quality
Data from all eligible studies were extracted in
standardized fashion. Data were abstracted on study
design and setting, sample size, patient characteristics,
incidence of hypoglycemia, incidence of acquired
infection, and 30-day mortality. Reporting of study
results were categorized as intention to treat, or not.
Allocation of concealment and likelihood of bias were
assessed to determine the methodological quality of the
included RCTs. Allocation concealment was ranked as
adequate, uncertain, or inadequate, and the five-point
Jadad scale was used to score likelihood of bias (35).
Pediatric Diabetes 2014: 15: 75–83
 Tight glucose control in critically ill children: potential benefits and risks
Fig. 1. Search strategy used in OVID MEDLINE (run on 22 January 2014).
Statistical analysis
Statistical analysis was performed using compre-
hensive meta-analysis 3.0 (Biostat; Englewood, NJ,
USA). Heterogeneity among studies was assessed using
the Cochran Q statistic with p ≤ 0.10 indicating signifi-
cant heterogeneity (36, 37). True vs. observed variance
was assessed by I2 with suggested thresholds for low
(25–49%), moderate (50–74%), and high (≥75%) val-
ues (38). We used a random effects model if the eligible
studies differed widely in terms of population studied;
otherwise, we used a fixed effects model. Summary
effects estimates were presented as ORs with 95%
confidence interval (CI). Significance was reported
as two-sided p < 0.05. Publication bias was assessed
visually using a funnel plot.
Results
Quality of trials included
Details of study identification with inclusion and
exclusion criteria are described in Fig. 2. The search
strategy identified 32 citations. Of these, two unique
studies met our inclusion criteria (30, 32). An additional
unique study was identified from the bibliography of a
review article identified in the 32 citations (31). Another
unique study was identified as a new publication (33).
In all, four RCTs met inclusion criteria with a total
of 3288 subjects. Sample size ranged from 239 to
1369 subjects. All the studies were of adequate quality
(Jadad score ≥3), used random allocation, and clearly
reported concealment of allocation. Treating clinicians
were not blinded to treatment allocation in any of
the studies. Study investigators were, however, blinded
to treatment and outcomes (in at least three studies).
Two of four studies reported results with intention to
treat analysis with complete follow-up of all reported
subjects (30, 32). The qualitative characteristics of the
studies are described in Table 1. Details of the studies
including design and setting as well as sample size and
select variables are described in Table 2.
Pediatric Diabetes 2014: 15: 75–83
Results of meta-analysis
We used a random effects model given the significant
differences in population and BG target ranges in
both intervention and control groups in the four
RCTs. Overall, TGC with IIT did not result in a
significant decrease in 30-day mortality (OR: 0.79;
95% CI: 0.55–1.15, p = 0.22, Q statistic = 2.99 with
p = 0.39, I2 = 0) (Fig. 3). TGC with IIT was associated
with decrease in acquired infection (OR: 0.76; 95% CI:
0.59–0.99, p = 0.04, Q statistic = 3.37 with p = 0.34,
I2 = 11) (Fig. 4). TGC with IIT was also associated
with significant increase in hypoglycemia (OR: 6.14;
95% CI: 2.74–13.78, p < 0.001, Q statistic = 11.28 with
p = 0.01, I2 = 73) (Fig. 5). There was no evidence of
publication bias on visual inspection of the funnel plot.
Discussion
There are few published pediatric RCTs of TGC with
IIT and most are limited to single or two-center studies
in specific populations of critically ill children. To our
knowledge, this is the first systematic review and meta-
analysis examining the benefits and risks of TGC with
IIT in critically ill children. Our data suggest that TGC
with IIT does not appear to confer any survival benefits,
but may reduce the incidence of acquired infections
in critically ill children. Importantly, in the absence
of continuous glucose monitoring technologies, TGC
with IIT can significantly increase the incidence of
hypoglycemia. While the study by Vlasselaers et al.
showed benefits of TGC with IIT with reductions in
inflammation, length of stay, and mortality in a single
center highly experienced in this approach and caring
predominantly for children recovering from cardiac
surgery (30), the subsequent two-center study by Agus
et al. did not show any benefit of TGC with IIT on
hospital acquired infections or mortality in critically
ill children recovering from cardiac surgery (32). The
study by Jeschke and colleagues was carried out in
a highly specific population of critically ill children
with severe burns (>30% total body surface area)
77
Srinivasan et al.

Fig. 2. Flow diagram of study identification with inclusions and exclusions.

Table 1. Quality assessment of included RCTs of TGC with IIT in critically ill children
Study, reference, year Clinician blinding Intention to treat Allocation concealment Jadad score (0–5)
Vlasselaers (30), 2009 No Yes Yes 3
Jeschke (31), 2010 No No Yes 3
Agus (32), 2012 No Yes Yes 3
Macrae (33), 2014 No No Yes 3

IIT, intensive insulin therapy; RCT, randomized controlled trial; TGC, tight glucose control.

and showed benefits of TGC with IIT in reducing
burn-specific morbidities, but was underpowered to
detect mortality benefits (31). The recently published
Control of Hyperglycaemia in Paediatric Intensive
Care (CHiP) study by Macrae et al. attempted to
overcome weaknesses of the earlier studies by enrolling
a mixed population of critically ill children across
multiple centers in the United Kingdom, but was
unable to show any convincing clinical benefits of TGC
with IIT (33). However, in the subgroup of critically
ill children that did not undergo cardiac surgery, TGC
with IIT (in comparison to conventional management)
78
significantly reduced hospital length of stay within
12 months as well as per patient health care costs at
12 months (33).
There are several important differences between
the four RCTs that could explain the variability
in observed outcomes in the individual studies
(30–33). In the study by Vlasselaers et al., BG
concentrations were controlled to age-specific fasting
ranges [50–80 mg/dL (2.8–4.4 mmol/L) for infants and
70–100 mg/dL (3.9–5.5 mmol/L) in older children]
in the intervention group that were much lower
than the BG ranges in the intervention groups
Pediatric Diabetes 2014: 15: 75–83
 Tight glucose control in critically ill children: potential benefits and risks

Table 2. Study details of included RCTs of TGC with IIT in critically ill children
Acquired Acquired Hypoglycemia* – Hypoglycemia* – Mortality†– Mortality†–
Size sepsis – sepsis – TGC control TGC control
Study, year, reference, Setting (n) TGC (%) control (%) (%) (%) (%) (%)
Vlasselaers (30), 2009 Mixed 700 29.2 36.8 24.9 1.4 2.3 5.1
Jeschke (31), 2010 Burns 239 8.2 22.6 26 9 8.3 12.3
Agus (32), 2012 Cardiac 980 5 5 3 1 1 1
Macrae (33), 2014 Mixed 1369 5.5 6.4 7.3 1.5 5 5

IIT, intensive insulin therapy; RCT, randomized controlled trial; TGC, tight glucose control.
*Blood glucose concentration ≤40 mg/dL (≤2.2 mmol/L).
†30-day mortality.

Fig. 3. Effect of tight glucose control (TGC) with intensive insulin therapy (IIT) on 30-day mortality in critically ill children.

in the other three RCTs [Agus and Jeschke:
80–110 mg/dL (4.4–6.1 mmol/L); Macrae: 72–126
(4.0–7.0 mmol/L)]. The control group BG range also
differed between the four RCTs [Vlasselaers and
Macrae: 180–215 mg/dL (10–11.9 mmol/L); Jeschke:
140–180 mg/dL (7.8–10 mmol/L); Agus: standard care
without a specified target BG range]. Additionally,
there was greater separation of BG concentrations
in the study by Vlasselaers et al. [infants: 86 mg/dL
(4.8 mmol/L) in the TGC group vs. 115 mg/dL
(6.4 mmol/L) in the control group; older children:
95 mg/dL (5.3 mmol/L) in the TGC group vs.
148 mg/dL (8.2 mmol/L) in the control group] as
well as in the study by Jeschke et al. compared
with separation achieved by Agus et al. [112 mg/dL
(6.2 mmol/L) in the TGC group vs. 121 mg/dL
(6.7 mmol/L) in the control group] and Macrae
et al. [107 mg/dL (5.9 mmol/L) in the TGC group vs.
114 mg/dL (6.3 mmol/L) in the control group]. Another
important difference between the studies pertained to
the variation in overall incidence of acquired infection
in study subjects (Vlasselaers: 33%, Jeschke: 19%,
Agus: 5%, and Macrae: 6%). Whereas the incidence
of acquired infection in the Jeschke study is not
unexpected given the population of children with
severe burns, Vlasselaers et al. observed a much higher
Pediatric Diabetes 2014: 15: 75–83
incidence of acquired infection than Agus et al. or
Macrae et al. suggesting important definitional and
possibly epidemiological differences between these
RCTs. Consequently, the trial by Vlasselaers was more
probably than the trials by Agus et al. or Macrae et al.
to have identified a positive benefit from TGC with
IIT. If the Vlasselaers trial is excluded from the meta-
analysis, TGC with IIT is no longer associated with
reduction in acquired infection in critically ill children
(OR: 0.79, 95% CI: 0.51–1.24, p = 0.31).
Hypoglycemia because of TGC with IIT is of
major concern to pediatric critical care practitioners
resulting in their hesitancy to embrace this strategy
more widely (39). The studies by Vlasselaers et al.,
Jeschke et al., and Macrae et al. observed significantly
greater rates of severe hypoglycemia [BG ≤ 40 mg/dL
(≤2.2 mmol/L)] in the intervention group using TGC
with IIT highlighting these concerns (30, 31, 33)
(Table 2). Macrae et al. observed the association
of hypoglycemia with mortality especially in the
cardiac surgery subgroup (33). However, the trial
by Agus et al. showed that TGC with IIT could
be safely carried out with minimum increase in
hypoglycemia when continuous glucose monitoring
technologies together with an explicit insulin dosing
algorithm were employed (32). The long-term impact of
79
Srinivasan et al.
Fig. 4. Effect of tight glucose control (TGC) with intensive insulin therapy (IIT) on acquired sepsis in critically ill children.
Fig. 5. Effect of tight glucose control (TGC) with intensive insulin therapy (IIT) on hypoglycemia [defined as blood glucose (BG) concentration
≤40 mg/dL (≤2.2 mmol/L)] in critically ill children.
hypoglycemia on the developing pediatric brain during
critical illness is largely unknown. However, when
the cohort enrolled in the study by Vlasselaers et al.
was followed for 4 years, there were no differences in
measures of intelligence [full scale intelligence quotient,
(IQ)]) or other neurocognitive measures between the
intervention group receiving TGC with IIT (median
IQ: 88) and the control group receiving usual care
(median IQ: 88.5) (40). The cohort enrolled by Agus
et al. is currently being followed in similar fashion for
long-term neurologic outcomes.
TGC with IIT in critically ill adults has not
showed unequivocal benefits when generalized to large
populations of critically ill adults. A recent meta-
analysis of seven RCTs of TGC with IIT in critically
ill adults concluded that there was no evidence to
support this practice in patients admitted to general
medical-surgical ICUs (41). Specifically, the analysis
concluded that TGC with IIT did not reduce 28-
day mortality (OR: 0.95; 95% CI: 0.87–1.05) or
80
incidence of bloodstream infections (OR: 1.04; 95% CI:
0.93–1.17), but was associated with significantly higher
incidence of hypoglycemia (OR: 7.7; 95% CI: 6.0–9.9,
p < 0.001). Using techniques of meta-regression to
control for percentage of calories given intravenously,
this meta-analysis showed reduced mortality from
TGC with IIT when parenteral nutrition was utilized
as the predominant method of nutrition. After
excluding the two Leuven trials that predominantly
employed parenteral nutrition, the authors observed
that mortality was lower with control patients receiving
usual care (OR: 0.90, 95% CI: 0.81–0.99, p = 0.04,
I2 = 0) suggesting that TGC with IIT was harmful
in patients fed enterally (41). In comparison, all
four RCTs in our meta-analysis enrolled considerable
proportions of subjects that were enterally fed. The
American College of Physicians now recommends
against TGC with IIT in critically ill adults as a routine
practice and suggests a target range of 140–200 mg/dL
(7.8–11.1 mmol/L) if glucose control is warranted
Pediatric Diabetes 2014: 15: 75–83
 Tight glucose control in critically ill children: potential benefits and risks
(42). Similarly, in preterm neonates, TGC with IIT
has not showed any benefits, but results in greater
hypoglycemia. In the only large multi-center trial to
date, treatment with insulin infusion in very low birth
weight infants, according to a defined protocol, led
to greater mortality at 28 postnatal days compared to
standard care (11.9 vs. 5.7%, p = 0.04) resulting in early
termination of the trial for potential harm (43). The
incidence of hypoglycemia [defined as BG < 47 mg/dL
(<2.6 mmol/L) for greater than 1 h] was also higher in
the intervention group compared to the control group
(29 vs. 17%, p = 0.005) even though continuous glucose
monitoring technologies were employed.
The limitations of our meta-analysis should be
considered when interpreting our results. Pediatric
RCTs of TGC with IIT in critically ill children are few
and include widely disparate populations [Vlasselaers
(30): mixed with 75% cardiac surgery, Jeschke (31):
severe burns, Agus (32): cardiac surgery, Macrae (33):
mixed with 60% cardiac surgery] of varying age groups
and differing ICU practices. Of the four RCTs included
in this review, only one (Macrae et al.) was a multi-
center study. The RCTs employed different target
ranges for both TGC with IIT and control, and used
different protocols to guide IIT with different BG
monitoring methods. We attempted to address this
limitation by employing a random effects model for
analysis. Additionally, we were unable to assess the
effect of important variables such as route of nutrition,
percentage of calories given intravenously, and BG
monitoring because of lack of adequate studies. We
were also unable to formally assess differences in
outcomes between cardiac and non-cardiac groups
because of lack of availability of data broken down by
group, but the results from published studies seem to
suggest that there are important differences between
these two groups. Nevertheless, this meta-analysis
represents an important attempt to systematically
review TGC with IIT in critically ill children in studies
performed till date.
What should the pediatric critical care practitioner
do when faced with hyperglycemia in a critically ill child
in the ICU? The evidence for (or against) using TGC
with IIT to improve survival outcomes in critically ill
children remains inconclusive with little knowledge
about long-term outcomes. It is possible that any
clinical benefit from TGC with IIT is overshadowed
by harm from hypoglycemia because of TGC with IIT.
Strategies that combine continuous glucose monitoring
and TGC with IIT may yet afford the best chance to
examine whether TGC with IIT in critically ill children
is associated with improvements in clinical outcomes
while limiting hypoglycemia. The Heart and Lung
Failure – Pediatric Insulin Titration (HALF-PINT)
trial is a large multi-center trial in North America
currently underway that employs such an approach
Pediatric Diabetes 2014: 15: 75–83
and is expected to conclusively answer the question
of whether TGC with IIT improves clinical outcomes
in critically ill children (excluding cardiac surgery).
This study aims to enroll 1880 critically ill children
across 30 centers to study the impact of TGC with
IIT on 28-day survival adjusted for ICU length of
stay (registered at ClinicalTrials.gov NCT01565941).
Subjects are being randomized to one of two BG
ranges – either 80–110 mg/dL (4.4–6.1 mmol/L) or
150–180 mg/dL (8.3–10 mmol/L). Continuous glucose
monitoring is performed in both groups to minimize
the incidence of hypoglycemia along with the use
of standardized BG testing, and an explicit insulin
dosing algorithm across all sites to minimize variation
in observed results. Subjects in both groups are also
being randomly selected for long-term follow-up of
neurological function and cognitive behavior.
Conclusions
TGC with IIT is not associated with decrease in
30-day mortality, but appears to be associated with
reduction in acquired infection when evaluated in
mixed populations of critically ill children. However,
TGC with IIT is also associated with higher incidence
of hypoglycemia. Large multi-center studies of TGC
with IIT using continuous glucose monitoring in
critically ill children are needed to determine if this
strategy can definitively improve clinical outcomes in
this population without increasing hypoglycemia.
Conflict of interest
The authors declared no conflicting interests.
Financial support
HALF-PINT study funded by NIH/NHLBI (NIH/
NHLBI: 1U01HL107681).
References
1. Srinivasan V, Spinella PC, Drott HR, Roth CL,
Helfaer MA, Nadkarni V. Association of timing,
duration, and intensity of hyperglycemia with intensive
care unit mortality in critically ill children. Pediatr Crit
Care Med 2004: 5: 329–336.
2. Faustino EV, Apkon M. Persistent hyperglycemia in
critically ill children. J Pediatr 2005: 146: 30–34.
3. Wintergerst KA, Buckingham B, Gandrud L,
Wong BJ, Kache S, Wilson DM. Association of
hypoglycemia, hyperglycemia, and glucose variability
with morbidity and death in the pediatric intensive care
unit. Pediatrics 2006: 118: 173–179.
4. Yung M, Wilkins B, Norton L, Slater A. Paediatric
Study Group; Australian and New Zealand Intensive
Care Society. Glucose control, organ failure, and
81
Srinivasan et al.
mortality in pediatric intensive care. Pediatr Crit Care
Med 2008: 9: 147–152.
5. Hirshberg E, Larsen G, Van Duker H. Alterations in
glucose homeostasis in the pediatric intensive care unit:
hyperglycemia and glucose variability are associated
with increased mortality and morbidity. Pediatr Crit
Care Med 2008: 9: 361–366.
6. Preissig CM, Rigby MR. Pediatric critical illness
hyperglycemia: risk factors associated with development
and severity of hyperglycemia in critically ill children. J
Pediatr 2009: 155: 734–739.
7. Ognibene KL, Vawdrey DK, Biagas KV. The
association of age, illness severity, and glycemic status
in a pediatric intensive care unit. Pediatr Crit Care Med
2011: 12: e386–e390.
8. Mechanick JI. Metabolic mechanisms of stress
hyperglycemia. JPEN J Parenter Enteral Nutr 2006:
30: 157–163.
9. Dufour S, Lebon V, Shulman GI, Petersen
KF. Regulation of net hepatic glycogenolysis and
gluconeogenesis by epinephrine in humans. Am J
Physiol Endocrinol Metab 2009: 297: E231–E235.
10. Kulp GA, Herndon DN, Lee JO, Suman OE, Jeschke
MG. Extent and magnitude of catecholamine surge in
pediatric burned patients. Shock 2010: 33: 369–374.
11. Mizock BA. Alterations in fuel metabolism in critical
illness: hyperglycaemia. Best Pract Res Clin Endocrinol
Metab 2001: 15: 533–551.
12. Srinivasan V. Stress hyperglycemia in pediatric critical
illness: the intensive care unit adds to the stress!. J
Diabetes Sci Technol 2012: 6: 37–47.
13. Weise K, Zaritsky A. Endocrine manifestations of
critical illness in the child. Pediatr Clin North Am 1987:
34: 119–130.
14. Gupta P, Natarajan G, Agarwal KN. Transient
hyperglycemia in acute childhood illnesses: to attend or
ignore? Indian J Pediatr 1997: 64: 205–210.
15. Gore DC, Chinkes D, Heggers J, Herndon DN,
Wolf SE, Desai M. Association of hyperglycemia with
increased mortality after severe burn injury. J Trauma
2001: 51: 540–544.
16. Cochran A, Scaife ER, Hansen KW, Downey EC.
Hyperglycemia and outcomes from pediatric traumatic
brain injury. J Trauma 2003: 55: 1035–1038.
17. Michaud LJ, Rivara FP, Longstreth WT Jr, Grady
MS. Elevated initial blood glucose levels and poor
outcome following severe brain injuries in children.
J Trauma 1991: 31: 1356–1362.
18. Branco RG, Garcia PC, Piva JP, Casartelli CH,
Seibel V, Tasker RC. Glucose level and risk of mortality
in pediatric septic shock. Pediatr Crit Care Med 2005:
6: 470–472.
19. Yates AR, Dyke PC 2nd, Taeed R et al. Hyperglycemia
is a marker for poor outcome in the postoperative
pediatric cardiac patient. Pediatr Crit Care Med 2006:
7: 351–355.
20. Day KM, Haub N, Betts H, Inwald DP.
Hyperglycemia is associated with morbidity in critically
ill children with meningococcal sepsis. Pediatr Crit Care
Med 2008: 9: 636–640.
21. Tuggle DW, Kuhn MA, Jones SK, Garza JJ, Skinner
S. Hyperglycemia and infections in pediatric trauma
patients. Am Surg 2008: 74: 195–198.
82
22. van den Berghe G, Wouters P, Weekers F et al.
Intensive insulin therapy in the critically ill patients. N
Engl J Med 2001: 345: 1359–1367.
23. Krinsley JS. Effect of an intensive glucose management
protocol on the mortality of critically ill adult patients.
Mayo Clin Proc 2004: 79: 992–1000.
24. Van den Berghe G, Wilmer A, Hermans G et al.
Intensive insulin therapy in the medical ICU. N Engl J
Med 2006: 354: 449–461.
25. Brunkhorst FM, Engel C, Bloos F et al. Intensive
insulin therapy and pentastarch resuscitation in severe
sepsis. N Engl J Med 2008: 358: 125–139.
26. Preiser JC, Devos P, Ruiz-Santana S et al. A
prospective randomised multi-centre controlled trial on
tight glucose control by intensive insulin therapy in adult
intensive care units: the Glucontrol study. Intensive Care
Med 2009: 35: 1738–1748.
27. NICE-SUGAR Study Investigators, Finfer S, Chit-
tock DR et al. Intensive versus conventional glucose
control in critically ill patients. N Engl J Med 2009: 360:
1283–1297.
28. Preiser JC. NICE-SUGAR: the end of a sweet dream?
Crit Care 2009: 13: 143.
29. Pham TN, Warren AJ, Phan HH, Molitor F,
Greenhalgh DG, Palmieri TL. Impact of tight
glycemic control in severely burned children. J Trauma
2005: 59: 1148–1154.
30. Vlasselaers D, Milants I, Desmet L et al. Intensive
insulin therapy for patients in paediatric intensive care: a
prospective, randomised controlled study. Lancet 2009:
373: 547–556.
31. Jeschke MG, Kulp GA, Kraft R et al. Intensive
insulin therapy in severely burned pediatric patients:
a prospective randomized trial. Am J Respir Crit Care
Med 2010: 182: 351–359.
32. Agus MS, Steil GM, Wypij D et al. Tight glycemic
control versus standard care after pediatric cardiac
surgery. N Engl J Med 2012: 367: 1208–1219.
33. Macrae D, Grieve R, Allen E et al. A randomized trial
of hyperglycemic control in pediatric intensive care. N
Engl J Med 2014: 370: 107–118.
34. Liberati A, Altman DG, Tetzlaff J et al. The
PRISMA statement for reporting systematic reviews
and meta-analyses of studies that evaluate healthcare
interventions: explanation and elaboration. BMJ 2009:
339: b2700.
35. Jadad AR, Moore RA, Carroll D et al. Assessing
the quality of reports of randomized clinical trials: is
blinding necessary? Control Clin Trials 1996: 17: 1–12.
36. Cochran W. The combination of estimates from
different experiments. Biometrics 1954: 10: 101–129.
37. Berlin JA, Laird NM, Sacks HS, Chalmers TC. A
comparison of statistical methods for combining event
rates from clinical trials. Stat Med 1989: 8: 141–151.
38. Higgins JP, Thompson SG, Decks JJ, Altman DG.
Measuring inconsistency in meta-analyses. BMJ 2003:
327: 557–560.
39. Hirshberg E, Lacroix J, Sward K, Willson D,
Morris AH. Blood glucose control in critically ill adults
and children: a survey on stated practice. Chest 2008:
133: 1328–1335.
40. Mesotten D, Gielen M, Sterken C et al. Neurocogni-
tive development of children 4 years after critical illness
Pediatric Diabetes 2014: 15: 75–83
 Tight glucose control in critically ill children: potential benefits and risks
and treatment with tight glucose control: a randomized
controlled trial. JAMA 2012: 308: 1641–1650.
41. Marik PE, Preiser JC. Toward understanding tight
glycemic control in the ICU: a systematic review and
metaanalysis. Chest 2010: 137: 544–551.
42. Qaseem A, Humphrey LL, Chou R, Snow V, Shekelle
P. Clinical Guidelines Committee of the American
College of Physicians. Use of intensive insulin therapy
Pediatric Diabetes 2014: 15: 75–83
for the management of glycemic control in hospitalized
patients: a clinical practice guideline from the American
College of Physicians. Ann Intern Med 2011: 154:
260–267.
43. Beardsall K, Vanhaesebrouck S, Ogilvy-Stuart
AL et al. Early insulin therapy in very-low-birth-weight
infants. N Engl J Med 2008: 359: 1873–1884.
83