Content uploaded by Nancy Elizabeth Wight
Author content
All content in this area was uploaded by Nancy Elizabeth Wight on Nov 10, 2014
Content may be subject to copyright.
BREASTFEEDING MEDICINE
Volume 1, Number 4, 2006
© Mary Ann Liebert, Inc.
Hypoglycemia in Breastfed Neonates
NANCY E. WIGHT
ABSTRACT
Healthy, full-term infants are programmed to make the transition from their intrauterine con-
stant flow of nutrients to their extrauterine intermittent nutrient intake without the need for
metabolic monitoring or interference with the natural breastfeeding process. Homeostatic
mechanisms ensure adequate energy substrate is provided to the brain and other organs, even
when feedings are delayed. The normal pattern of early, frequent, and exclusive breastfeed-
ing meets the needs of healthy full-term infants. Routine screening or supplementation are
not necessary and may harm the normal establishment of breastfeeding. Screening should be
restricted to at-risk and symptomatic infants. Symptomatic infants need immediate assess-
ment and intravenous glucose therapy, not forced feedings.
253
INTRODUCTION
I
N THE LAST ISSUE OF
B
REASTFEEDING
M
EDICINE
,
the Academy of Breastfeeding Medicine pub-
lished its updated Clinical Protocol #1: Guide-
lines for Glucose Monitoring and Treatment of Hy-
poglycemia in Breastfed Neonates.
1
This revised
protocol was based on an extensive review of
the literature, but because of the nature of a
protocol, not all the background information
and supporting documentation could be in-
cluded. This article provides much of the in-
formation in the protocol, but also additional
discussion and references that may be of value
to clinicians seeking to establish effective and
breastfeeding-supportive care of neonates who
have, or are at risk for, hypoglycemia.
GLUCOSE HOMEOSTASIS
DURING TRANSITION
Throughout gestation the fetus receives its
entire supply of glucose (70% of its energy
needs) from the maternal circulation by facili-
tated diffusion via the placenta, with fetal
plasma glucose levels 70% to 80% of maternal
venous plasma levels.
2
Glucose utilization by
the fetus is approximately 5 mg/kg per minute
with amino acids and lactate as additional en-
ergy sources.
2
Although present, the enzyme
activity necessary for gluconeogenesis is mini-
mal in the fetus as there is no need for glucose
production. However, an important sequence
of integrated metabolic adaptations occurs at
birth, allowing the newborn to produce glucose
Sharp Healthcare Lactation Service, San Diego, California; Department of Pediatrics, University of California—San
Diego School of Medicine, San Diego, California; Division of Neonatology, Children’s Specialists of San Diego, San
Diego, California.
Review
and regulate its own metabolic homeostatic
processes.
3
At birth the infant must supply its
glucose needs of approximately 5 to 8 mg/kg
per minute (70% used by the brain) through a
balance of exogenous sources (breast milk) and
endogenous glucose production through glu-
coneogenesis, glycogenolysis, and ketogenesis,
provided adequate substrates are available.
3,4
Within minutes of cutting the umbilical cord,
there is a three- to fivefold surge in glucagon
and catecholamines, which initiate glycogen
breakdown. High endogenous growth hormone
and cortisol facilitate the onset of gluco-
neogenesis within several hours, and insulin se-
cretion is blunted so that serum concentrations
of insulin fall.
3
The processes that ensure the
availability of glucose and other fuels are col-
lectively described as counter-regulation, and
are activated primarily by glucagon and adren-
alin.
3
The term hypoglycemia refers to a low blood
glucose concentration. Neonatal hypoglycemia
is not a medical condition in itself, but a fea-
ture of illness or a failure to adapt from the fe-
tal state of continuous transplacental glucose
consumption to the extrauterine pattern of in-
termittent nutrient supply.
2
Transient hypo-
glycemia in the immediate newborn period is
common and occurs in almost all mammalian
species. In healthy term human infants, even if
early enteral feeding is withheld, this phe-
nomenon is self-limited as glucose levels spon-
taneously rise within 2 to 3 hours.
5–7
This early
self-limited period of hypoglycemia cannot be
considered pathologic, and there is little prac-
tical value in measuring the blood glucose
concentrations of asymptomatic babies in the
first 2 hours after birth.
2,4,8
Furthermore, even
in those situations in which low blood glu-
cose concentrations develop secondary to pro-
longed intervals (8 hours) between breast
feedings,
6
there is a marked ketogenic response
that provides glucose-sparing fuel to the
brain.
6,9–11
The neonatal brain has an enhanced
capability to use ketone bodies relative to in-
fants (fourfold) and adults (40-fold).
12
Hawdon
6
and others have studied this pat-
tern of metabolic adaptation.
13,14
Breastfed
term infants have lower blood glucose
6,13–15
and higher ketone bodies
6
than formula-fed in-
fants. Breastfed infants up to 1 week old had a
significantly lower mean blood glucose con-
centration (range 1.5 to 5.3 mmol/L [27 to 95
mg/dL] and mean 3.6 mmol/L [58 mg/dL])
than formula-fed infants of the same age (range
2.5 to 6.2 mmol/L [45 to 111 mg/dL] and mean
4.0 mmol/L [72 mg/dL]).
6
Those infants who
lost the most weight postnatally had the high-
est ketone body concentrations,
6
which sug-
gests that the provision of alternate fuels
constitutes a normal adaptive response to tran-
siently low nutrient intake during the estab-
lishment of breastfeeding.
6,16
Because the opti-
mally breastfed infant’s lower blood glucose is
the physiologic norm, it has been suggested
that breastfed infants may well tolerate lower
plasma glucose levels without any significant
clinical manifestations or sequelae, assuming
adaptive metabolic response systems are func-
tioning normally.
16
One case series of hypo-
glycemia in “healthy, breastfed term new-
borns” (all of whom were feeding poorly at
discharge) revealed no urinary ketones in any
of the three symptomatic infants, suggesting a
defective ketogenic response to fasting.
17
As yet there has been no systematic study of
urinary ketones in symptomatic or asympto-
matic hypoglycemic newborns. The presence
of urinary ketones may prove to be a reas-
suring finding in asymptomatic, mildly to
moderately hypoglycemic infants. Multiple
regression analysis from Hawdon’s study,
6
correcting for postnatal age and feeding
method, revealed that only the interval be-
tween feeds was inversely correlated with glu-
cose concentration, supporting the need for
continuous mother–infant contact and fre-
quent on-demand nursing.
DEFINITIONS OF HYPOGLYCEMIA
The definition of hypoglycemia in the new-
born infant has remained controversial because
of a lack of significant correlation among
plasma glucose concentration, clinical symp-
toms, and long-term sequelae.
16,18–20
There
have been four main approaches to the defini-
tion of hypoglycemia: (a) epidemiologic/sta-
tistical approach based on measured range of
glucose values; (b) clinical manifestations; (c)
acute changes in metabolic/endocrine re-
WIGHT254
sponses and neurologic function; and (d) long-
term neurologic outcome.
2,16
Epidemiologic approach
“Normal” blood glucose results vary enor-
mously with the source of the blood sample,
assay method, and whether blood or plasma
glucose concentration is determined. Plasma
or serum glucose concentrations are 10% to
15% higher than in whole blood.
21
In addi-
tion, early feeding schedules have a prominent
effect on blood glucose concentrations but
have changed a great deal since early studies.
2
Even now they vary from hospital to hospital.
The healthy, term exclusively breastfed infant
represents the biological norm, yet until re-
cently, very few breastfed infants had been
studied.
Breastfed, formula-fed, and mixed-fed in-
fants follow the same pattern of glucose values
with an initial fall in glucose over the first 2
hours, followed by a gradual rise in glucose
over the next 96 hours, whether fed or not
(Table 1).
5,14,22
As noted, when compared,
breastfed infants tend to have slightly lower
glucose and higher ketone bodies than artifi-
cially fed infants.
6,13–15
Although helpful, the
findings of the epidemiologic approach have
been erroneously used to define the cutoff be-
tween normoglycemia and hypoglycemia
rather than recognizing that hypoglycemia re-
flects a continuum of biologic abnormalities,
ranging from mild to severe.
16
The incidence of “hypoglycemia” varies with
the definition.
25
Many authors have suggested
numeric definitions of hypoglycemia, usually
between 30 and 50 mg/dL (1.7 to– 2.8 mmol/L)
HYPOGLYCEMIA IN BREASTFED NEONATES 255
T
ABLE
1. N
EONATAL
B
LOOD
G
LUCOSE
C
ONCENTRATION
(
MG
/
D
L-
TOP
,
AND MMOL
/L-
BOTTOM
)
IN
T
ERM
I
NFANTS BY
P
OSTNATAL
A
GE FROM
S
EVEN
S
URVEY
S
TUDIES
Swenne Hoseth 2000
22
Adejuyigbe Diwaker
Srinivasan 1986
5
Heck 1987
14
Hawdon 1992
6
1994
13
Whole blood: 2001
23
2002
24
Plasma: Serum: Whole blood: Serum: breastfed Plasma: Plasma:
Age mixed fed mixed fed mixed fed breastfed median (10– breastfed breastfed
(h) (mean SD) (mean SD) (mean SD) (mean SD) 90 percentile) (mean SD) (mean SD)
0 107 35.. 97 29065 2 — 88 (50–128) 64 320—
156 19 60 180— — 50 (36–70)0——
260 11 61 150— — 56 (36–81)0——
370 13 — — 38 9.5 — — 54 19
468 14 — — (3–15 h) 54 (47–70)0——
665 13 56 110— — 50 (40–61)0—53 13.5
12 67 14 56 12058 5.4 — 50 (41–63)0——
(12–24 h)
24 71 10 61 10063 3.6 53 9.7 50 (41–63)057 15052 14
(25–48 h)
48 73 13 64 10063 3.6 — 63 (58–79)059 170—
(49–72 h)
72 83 12 — 61 2 — 61 (52–83)0—54 14
(73–96 h)
96 80 12 — 74 2 — 68 (58–81)0—
(97–168 h)
mmol/L
0 5.94 1.94 5.4 1.603.6 0.1 — 4.9 (2.8–7.1).3.55 1.760—
1 3.1 1.06 3.3 100. — — 2.8 (2–3.9)00 ——
2 3.3 0.61 3.4 0.83 — — 3.1 (2–4.5)00 ——
3 3.9 0.72 — — 2.1 0.53 — — 3 1.05
4 3.8 0.78 — — (3–15 h) 3 (2.6–3.9) — —
6 3.6 0.72 3.1 0.61 — — 2.8 (2.2–3.4) — 2.95 0.75
12 3.7 0.78 3.1 0.67 3.2 0.3 — 2.8 (2.3–3.5) — —
24 3.9 0.56 3.4 0.56 3.5 0.2 2.93 0.54 2.8 (2.3–3.5) 3.18 0.8502.89 0.79
48 4.1 0.72 3.6 0.56 3.5 0.2 — 3.5 (3.2–4.4) 3.29 0.967 —
72 4.6 0.67 — 3.4 0.1 — 3.4 (2.9–4.6) — 3 0.79
96 4.4 0.67 — 4.1 0.1 — 3.8 (3.2–4.5) — —
and varying by postnatal age.
2,5,14,18,25–29
Corn-
blath et al.
16
summarized the problem well:
“Significant hypoglycemia is not and can never
be defined as a single number that can be ap-
plied universally to every individual patient.
Rather, it is characterized by a value(s) that is
unique to each individual and varies with both
their state of physiologic maturity and the in-
fluence of pathology.” Instead of specifying a
number to represent hypoglycemia, Cornblath
et al.
16,30
suggest “operational thresholds” as
defined by Table 2. These “operational thresh-
olds” represent values to which one should re-
act, either by further testing and/or treatment.
They do not represent either normal or abnor-
mal values.
A recent metaanalysis (studies published 1986
to 1994) of low plasma glucose thresholds in full-
term normal newborns who were mostly mixed
fed (formula and breastfeeding) or formula fed,
presented recommended low thresholds for
plasma glucose based on hours after birth (Table
3). They specifically noted that given the lower
plasma glucose levels in normal breastfed in-
fants, the low thresholds for exclusively breast-
fed infants might even be lower.
31
Clinical manifestations of hypoglycemia
The clinical manifestations of hypoglycemia
are nonspecific, occurring with a variety of other
neonatal problems. Even in the presence of an
arbitrary low glucose level, the physician must
assess the general status of the infant by ob-
servation and physical examination to rule out
other disease entities and processes that may
need additional laboratory evaluation and
treatment. Some common clinical signs are
listed in Table 4. A diagnosis of hypoglycemia
also requires that symptoms abate after nor-
moglycemia is restored.
Acute physiologic changes
Neurophysiologic monitoring, including
electroencephalography (EEG), visual evoked
potentials (VEP), and brainstem auditory
evoked responses (BAER) have failed to define
a safe blood glucose concentration or a thresh-
old for neurologic damage.
4
Koh et al.
32
re-
ported BAER abnormalities only when blood
glucose fell below 2.6 mmol/L (47 mg/dL). Un-
fortunately, only 5 of their 17 patients were less
than 4 days old, only one was symptomatic,
and the symptoms did not correlate with the
lowest blood glucose level. Kinnala et al.
33
found four times the number of head MRI and
ultrasound abnormalities in 18 symptomatic
full-term infants than 19 normoglycemic con-
trols. Most lesions were absent by 2 months of
age, and only one infant appeared neurologi-
cally affected.
Evidence from tissue culture and animal
models indicate that the neural damage attrib-
WIGHT256
T
ABLE
2. O
PERATIONAL
T
HRESHOLDS
Operational threshold Plasma glucose
Intervention 36 mg/dL (2.0 mmol/L)
Intravenous glucose 20–25 mg/dL (1.1–1.4 mmol/L)
Preterm infants Same as term
Infants on TPN 45 mg/dL (2.5 mmol/L)
Therapeutic objective
Transient hypoglycemia 45 mg/dL (2.5 mmol/L)
Profound/persistent hypoglycemia 60 mg/dL (3.3 mmol/L)
Operational threshold by age after birth
1st 24 h
Healthy term or preterm 34–37 wk, formula-fed 30–35 mg/dL (1.7–2.0 mmol/L)
Sick, LBW, preterm 34 wk 45–50 mg/dL (2.5–2.8 mmol/L)
24 h 40–50 mg/dL (2.2–2.8 mmol/L)
Any age 20–25 mg/dL (1.1–1.4 mmol/L)
From: Cornblath M, Hawdon JM, Williams AF, et al. Controversies regarding definition of neona-
tal hypoglycemia: Suggested operational thresholds. Pediatrics 2000;105:1141–1145; Cornblath M,
Ichord R. Hypoglycemia in the neonate. Semin Peinatol 2000;24:136–149.
uted to hypoglycemia is not simply a matter of
inadequate energy stores, but rather a result of
accumulation of toxic substances, such as as-
partic acid
4
and glutamate.
34
Because this pro-
cess requires time (hours to days), clinicians
can be reassured that transient, single, brief pe-
riods of hypoglycemia are unlikely to cause
permanent neurologic damage.
4,34,35
Long-term neurologic outcome
The data correlating neonatal hypoglycemia
with long-term neurologic outcome are limited
because of lack of suitable nonhypoglycemic
controls, a failure to consider other pathology,
and the small number of asymptomatic infants
followed.
16,19,20
Animal studies suggest that
the immature brain is incredibly resistant (via
many different mechanisms) to damage from
even profound hypoglycemia.
36
Koivisto et al.
37
found no difference in neurologic outcome be-
tween asymptomatic hypoglycemic infants and
euglycemic control infants, with 94% and 95%
of each group normal on 1- to 4-year follow-up.
There was a significant increase in neurologic
abnormalities (12%) in symptomatic (tremor,
cyanosis, paleness, limpness, irritability, apathy,
or tachypnea that disappeared during treatment
with glucose) hypoglycemic infants, and 50%
incidence of neurologic abnormalities when
seizures were present. The estimated duration
of hypoglycemia was 37 hours in the asympto-
matic hypoglycemic infants, 49 hours in the
symptomatic/nonseizure group, and 105 hours
in the seizure group. The follow-up evaluation
included a physical examination, anthropome-
try, and neurologic and developmental assess-
ments, as well as an ophthalmologic examina-
tion. Interestingly, this study was done in
Finland from 1967 to 1970, at a time when in-
fants were fed 5% “saccharose” for 24 hours
prior to initiating breastfeeding.
More recently, Brand et al.
38
studied hypo-
glycemia (defined as 2.2 mmol/L [40 mg/
dL] 1 hour after birth or 2.5 mmol/L [45
mg/dL] thereafter) in term LGA infants born to
nondiabetic mothers. Screening was done at 1, 3,
and 5 hours after birth and continued if the glu-
cose was low. Intravenous glucose was started if
the glucose was less than 1.5 to 2 mmol/L (27 to
36 mg/dL) or symptoms were present. There
were no significant differences between the hy-
poglycemic and control groups at 4-year follow-
up, which consisted of the Dutch version of
the Denver Developmental Scale, the Snijders-
Oomen nonverbal intelligence test, and the
Dutch version of the Child Behavior Check List.
Unfortunately, the type of feeding was not dis-
closed and only 64% of the original population
completed the assessment at 4 years.
A recent systematic review of cohort studies
on subsequent neurologic development after
episodes of hypoglycemia in the first week of
life found that major clinical and methodologic
heterogeneity of available studies precluded any
true metaanalysis.
20
Of the 18 eligible studies,
the overall methodologic quality was consid-
ered poor in 16 studies and high in two studies.
Pooling of the results of the two high-quality
studies was deemed inappropriate because of
this heterogeneity. None of the studies provided
a valid estimate of the effect of neonatal hypo-
glycemia on neurodevelopment. Building on the
strengths and weaknesses of existing studies,
they proposed an “optimal” future study design
and invited content experts and clinicians
worldwide to comment, refine the design, and
HYPOGLYCEMIA IN BREASTFED NEONATES 257
T
ABLE
3. R
ECOMMENDED
L
OW
T
HRESHOLDS
:
P
LASMA
G
LUCOSE
L
EVEL
Hour after birth 5th Percentile PGL-mg/dL (mmol/L)
1–2 (nadir) 28 (1.6)
3–47 40 (2.2)
48–72 48 (2.7)
From: Alkalay AL, Sarnat HB, Flores-Sarnat L, et al.
Population meta-analysis of low plasma glucose thresh-
olds in full-term normal newborns. Am J Perinatol
2006;23:115–119.
T
ABLE
4. C
LINICAL
M
ANIFESTATIONS
OF
P
OSSIBLE
H
YPOGLYCEMIA
Irritability, tremors, jitteriness
Exaggerated Moro reflex
High-pitched cry
Seizures or myoclonic jerks
Lethargy, listlessness, limpness, hypotonia
Coma
Cyanosis
Apnea or irregular breathing
Tachypnea
Hypothermia
Vasomotor instability
Poor suck or refusal to feed
participate in a prospective collaborative study.
Hopefully, such a study will be done.
In the absence of definitive data regarding
a “safe” blood glucose concentration in any
given population, the “operational threshold”
approach suggested by Cornblath
16,30
and Al-
kalay
31
seems most appropriate. Hawdon
39
summarized it well: “Evidence from studies of
humans and other animals suggests that corti-
cal damage and long-term sequelae occur after
prolonged hypoglycemia sufficiently severe to
cause neurological signs.”
ASSESSMENT OF GLUCOSE LEVELS
Bedside glucose testing strips are inexpen-
sive and practical, but are not reliable with sig-
nificant variance from true blood glucose lev-
els.
29,40,41
Studies comparing different reagent
strips have estimated that as many as 20% of
truly normoglycemic infants are falsely labeled
as hypoglycemic, leading to unnecessary labo-
ratory tests and treatment.
4
A recent study
evaluated readily available “point-of-care” glu-
cose measuring devices and concluded that
none of the five glucometers was satisfactory
as the sole measuring device.
34
Newer bedside
glucose systems simplify procedures, but ap-
propriate test strip storage, handling, and ad-
herence to expiration dates are still essential to
prevent error.
42
Even newer point-of-care glucose electrode
(YSI) and cuvette-based glucose oxidation opti-
cal methods (HemoCue), although possible im-
provements over reagent strips, do not have the
reliability of laboratory measurement.
2,43–45
Bed-
side glucose tests may be used for screening, but
laboratory levels must confirm results before a
diagnosis of hypoglycemia can be made, espe-
cially in asymptomatic infants.
2,21,35,40
RISK FACTORS FOR HYPOGLYCEMIA
Neonates at increased risk for developing
neonatal hypoglycemia should be routinely
monitored for blood glucose levels irrespective
of the mode of feeding. These at-risk infants
should be screened before any symptoms man-
ifest. Neonates at risk fall into two main cate-
gories: (a) Excess use of glucose, which in-
cludes the hyperinsulinemic states; and (b) in-
adequate production or substrate delivery.
30
Table 5 shows infant categories at increased
risk for hypoglycemia.
4,16,30,46–48
MANAGEMENT RECOMMENDATIONS
General
Glucose screening should be performed only
on at-risk infants and those with clinical symp-
toms compatible with hypoglycemia. Routine
monitoring of blood glucose in asymptomatic,
term newborns is potentially harmful to the es-
tablishment of a healthy mother–infant rela-
tionship and successful breastfeeding pat-
terns.
2,4,8,35,40,49–54
This recommendation has
been supported by the World Health Organi-
zation,
2
the AAP,
51
and the National Childbirth
Trust of the United Kingdom.
52
At-risk infants
should be screened for hypoglycemia with a
frequency and duration related to the specific
risk factors of the individual infant.
4
It is sug-
gested that monitoring begin within 30 to 60
minutes of age for infants with suspected hy-
perinsulinemia, and no later than 2 hours of age
for infants in other risk categories. Monitoring
should continue, until normal preprandial lev-
els are consistently obtained. Bedside glucose
screening, tests must be confirmed by formal
laboratory testing.
Early and exclusive breastfeeding meets the
nutritional needs of healthy, term, newborn in-
fants. Healthy term infants do not develop
symptomatic hypoglycemia simply as a result
of underfeeding.
2,4,51
Therefore, routine sup-
plementation of healthy, term infants with wa-
ter, glucose water, or formula is unnecessary
and may interfere with establishment of nor-
mal breastfeeding and normal metabolic com-
pensatory mechanisms.
6,13,51,52
As with gen-
eral breastfeeding recommendations, healthy
term infants should initiate breastfeeding
within 30 to 60 minutes of life and continue on
demand, recognizing that crying is a very late
sign of hunger.
51,55
Feedings should be fre-
quent, 10 to 12 times per 24 hours in the first
few days after birth.
51
Early breastfeeding is
WIGHT258
not precluded just because the infant meets the
criteria for glucose monitoring. Initiation and
establishment of breastfeeding is also facili-
tated by skin-to-skin contact of mother and in-
fant. Such practices will maintain normal in-
fant body temperature and reduce energy
expenditure while stimulating suckling and
milk production.
15,51
Documented hypoglycemia in an
asymptomatic infant
As noted, the asymptomatic “hypoglycemic”
infant is at extremely low risk of long-term neu-
rologic sequelae. Such an infant should con-
tinue breastfeeding (approximately every 1 to
2 hours) or feed 3 to 5 mL/kg of expressed
breast milk or substitute nutrition (pasteurized
donor human milk, elemental formulas, par-
tially hydrolyzed formulas, or routine formu-
las). This volume is based on normal volumes
of colostrum
56
and the average size of the in-
fant’s stomach in the first week postpartum.
57
There is no research available delineating what
amount of glucose or what volume of glucose-
containing oral fluids is needed to raise serum
glucose a certain amount in any population.
The blood glucose concentration should be
rechecked before subsequent feedings until the
value is acceptable and stable. If the neonate is
unable to suck or feedings are not tolerated, avoid
forced feedings (e.g., nasogastric tube) and begin
intravenous therapy (see the following). Such an
infant is not normal and requires a careful ex-
amination and evaluation in addition to more in-
tensive therapy. If glucose remains low despite
feedings, intravenous glucose therapy should be
initiated and intravenous rate adjusted by blood
glucose concentration. Of course breastfeeding
may continue during intravenous (IV) glucose
therapy if the infant is interested and will suckle.
As with any medical therapy, clinical signs, phys-
ical examination, screening values, laboratory
confirmation, treatment, and changes in clinical
condition (i.e., response to treatment) should be
carefully documented.
Symptomatic hypoglycemic infants
Infants with symptoms consistent with hy-
poglycemia (see Table 4) or infants with plasma
glucose levels less than 20 to 25 mg/dL (1.1
to 1.4 mmol/L) should have more aggressive
therapy. Current neonatal texts suggest initiat-
ing intravenous glucose using a 2 mL/kg bolus
of 10% glucose solution, followed by a contin-
uous infusion of 6 to 8 mg/kg per minute (ap-
proximately 80 to 100 mL/kg per 24 hours).
58
Attempts to rely on oral or intragastric feeding
to correct extreme (20 to 25 mg/dL) or symp-
tomatic hypoglycemia are inappropriate and
may be dangerous if the infant aspirates the
oral supplement. Such an infant is not normal
and requires an immediate and careful exami-
nation and evaluation. To allow for minute-to-
minute variations in blood glucose, the glucose
HYPOGLYCEMIA IN BREASTFED NEONATES 259
T
ABLE
5. A
T
-R
ISK
I
NFANTS FOR
W
HOM
R
OUTINE
M
ONITORING
OF
B
LOOD
G
LUCOSE
I
S
I
NDICATED
Small for gestational age (SGA); 10th percentile for weight
Large for gestational age (LGA); 90th percentile for weight*
Discordant twin; weight 10% larger twin
Infant of diabetic mother, especially if poorly controlled
Low birth weight (2500 g)
After perinatal stress; severe acidosis or hypoxia-ischemia
Cold stress
Polycythemia (venous Hct 70%)/hyperviscosity
Erythroblastosis fetalis
Beckwith-Wiedemann syndrome
Microphallus or midline defect
Suspected infection
Respiratory distress
Known or suspected inborn errors of metabolism or endocrine disorders
Maternal drug treatment (e.g., terbutaline, propranolol, oral hypoglycemics)
Infants displaying symptoms associated with hypoglycemia (see Table 4)
*In unscreened populations in which LGA may represent undiagnosed and untreated
maternal diabetes.
concentration in symptomatic infants should
be maintained at greater than 45 mg/dL (2.5
mmol/L).
The intravenous rate should be adjusted by
blood glucose concentration and frequent
breastfeeding should be encouraged after the
relief of symptoms. As feedings are initiated,
glucose concentrations should be monitored
before feedings as the IV is weaned, until val-
ues are stabilized off intravenous fluids. Again,
clinical signs, physical examination, screening
values, laboratory confirmation, treatment, and
changes in clinical condition (i.e., response to
treatment) should be carefully documented.
SUPPORTING THE MOTHER
Having an infant thought to be normal and
healthy develop hypoglycemia is both con-
cerning to the mother and family, and may
jeopardize breastfeeding. Mothers should be
reassured that there is nothing wrong with
their milk, and that supplementation is usually
temporary. Having the mother hand express or
pump milk that is then fed to her infant can
overcome feelings of maternal inadequacy as
well as help establish a full milk supply. In or-
der to protect the mother’s milk supply, it is
important to provide stimulation to the breasts
by manual or mechanical expression with ap-
propriate frequency (8 times in 24 hours) until
her baby is latching and suckling well. Keep-
ing the infant at the breast or returning the in-
fant to the breast as soon as possible is impor-
tant. Skin-to-skin care is easily done with an IV
and may soften the trauma of intervention,
while providing physiologic thermoregulation,
contributing to metabolic homeostasis.
FUTURE RESEARCH
Boluyt et al.
20
proposed a study design to an-
swer two questions: (a) What is the effect of
various blood glucose concentrations in the
first three postnatal days on long-term neu-
rodevelopment? and (b) What is the effect of
treatment with additional carbohydrates in
neonates with moderate hypoglycemia on
long-term neurodevelopment compared with
expectant observation? In addition, the devel-
opment of simple, rapid, and reliable bedside
testing would improve the efficiency of diag-
nosis and treatment. A clearer understanding
of the role of other metabolic fuels and meth-
ods to measure them in a clinically useful way
may improve the ability to predict which in-
fants are truly at risk for neurologic sequelae,
and facilitate more rapid and appropriate treat-
ment. Finally, it is unclear how much enteral
glucose in what form is necessary to raise blood
glucose to acceptable levels, once those levels
are determined. At present, many breastfed in-
fants are being grossly overfed with the good
intention of avoiding intravenous glucose. It
may be that more long-term problems are be-
ing caused by aggressively feeding artificial
milks.
CONCLUSION
Healthy, full-term infants are programmed
to make the transition from their intrauterine
constant flow of nutrients to their extrauterine
intermittent nutrient intake without the need
for metabolic monitoring or interference with
the natural breastfeeding process. Homeostatic
mechanisms ensure adequate energy substrate
is provided to the brain and other organs, even
when feedings are delayed. The normal pattern
of early, frequent, and exclusive breastfeeding
meets the needs of healthy full-term infants.
Routine screening or supplementation are not
necessary and may harm the normal establish-
ment of breastfeeding. Screening should be re-
stricted to at-risk and symptomatic infants.
Symptomatic infants need immediate assess-
ment and intravenous glucose therapy, not
forced feedings.
REFERENCES
1. Academy of Breastfeeding Medicine. Clinical Proto-
col #1 (Revised): Guidelines for Glucose Monitoring
and Treatment of Hypoglycemia in Breastfed
Neonates. Breastfeeding Med 2006;1:178–184.
2. Williams, AF. Hypoglycaemia of the Newborn: Review of
the Literature. World Health Organization, Geneva,
1997. Accessed August 7, 2006: www.who.int/child-
adolescent-health/publications/NUTRITION/
WHO_CHD_97.1.htm.
WIGHT260
3. Sperling MA, Menon RK. Differential diagnosis and
management of neonatal hypoglycemia. Pediatr Clin
North Am 2004;51:703–723.
4. Eidelman AI. Hypoglycemia and the breastfed
neonate. Pediatr Clin North Am 2001;48:377–387.
5. Srinvasan G, Phildes RS, Cattamanchi G, et al. Plasma
glucose values in normal neonates: A new look. J Pe-
diatr 1986;109:114–117.
6. Hawdon JM, Ward Platt MP, Aynsley-Green A. Pat-
terns of metabolic adaptation for preterm and term
neonates in the first postnatal week. Arch Dis Child
1992;67:357–365.
7. Cornblath M, Reisner SH. Blood glucose in the neonate
and its clinical significance. NEJM 1965;273:378–381.
8. Hawdon JM, Ward Platt MP, Aynsley-Green A. Pre-
vention and management of neonatal hypoglycemia.
Arch Dis Child Fetal Neonatal Ed 1994;70:F60–F65.
9. Lucas A, Bayes S, Bloom SR, et al. Metabolic and en-
docrine responses to a milk feed in 6 day old term in-
fants: Differences between breast and cow’s milk for-
mula feeding. Acta Paediatr Scand 1981;70:195–200.
10. Edmond J, Auestad N, Robbins RA, et al. Ketone body
metabolism in the neonate: development and the ef-
fect of diet. Fed Proc 1985;44:2359–2364.
11. Yager JY, Heitjan DF, Towfighi J, et al. Effect of in-
sulin-induced and fasting hypoglycemia on perinatal
hypoxic-ischemic brain damage. Pediatr Res 1992;31:
138–142.
12. Mehta A. Prevention and management of neonatal
hypoglycaemia. Arch Dis Child 1994;70:F54–F65.
13. Swenne I, Ewald U, Gustafsson J, et al. Inter-rela-
tionship between serum concentrations of glucose,
glucagon and insulin during the first two days of life
in healthy newborns. Acta Paediatr 1994;83:915–
919.
14. Heck LJ, Erenberg A. Serum glucose levels in term
neonates during the first 48 hrs of life. J Pediatr 1987;
110:119–122.
15. Durand R, Hodges S, LaRock S, et al. The effect of
skin-to-skin breast-feeding in the immediate recovery
period on newborn thermoregulation and blood glu-
cose values. Neonatal Intensive Care 1997;23–29.
16. Cornblath M, Hawdon JM, Williams AF, et al. Con-
troversies regarding definition of neonatal hypo-
glycemia: Suggested operational thresholds. Pediatrics
2000;105:1141–1145.
17. Moore AM, Perlman M. Symptomatic hypoglycemia
in otherwise healthy, breastfed term newborns. Pedi-
atrics 1999;103:837–839.
18. Kalhan S, Peter-Wohl S. Hypoglycemia: What is it for
the neonate? Am J Perinatol 2000;17:11–18.
19. Sinclair JC. Approaches to the definition of neonatal
hypoglycemia. Acta Paediatr Jpn 1997;39:S17–S20.
20. Boluyt N, van Kempen A, Offringa M. Neurodevel-
opment after neonatal hypoglycemia: A systematic re-
view and design of an optimal future study. Pediatrics
2006;117:2231–2243.
21. Cornblath M, Schwartz R. Disorders of Carbohydrate
Metabolism in Infancy, 3rd ed. Blackwell Scientific,
Boston, 1991.
22. Hoseth E, Joergensen A, Ebbesen F, et al. Blood glu-
cose levels in a population of healthy, breast fed, term
infants of appropriate size for gestational age. Arch
Dis Child Fetal Neonatal Ed 2000;83:F117–119.
23. Adejuyigbe EA, Fasubaa OB, Ajose OA, et al. Plasma
glucose levels in exclusively breastfed newborns in
the first 48 hours of life in Ile-Ife, Nigeria. Nutrition
Health 2001;15:121–126.
24. Diwakar KK, Sasidhar MV. Plasma glucose levels in
term infants who are appropriate size for gestation
and exclusively breast fed. Arch Dis Child Fetal Neona-
tal Ed 2002; 87:F46–F48.
25. Sexson WR. Incidence of neonatal hypoglycemia: A
matter of definition. Editorial. J Pediatr 1984;105:149–150.
26. Cole MD, Peevy K. Hypoglycemia in normal neonates
appropriate for gestational age. J Perinatol 1994;14:
118–120.
27. Stanley CA, Baker L. The causes of neonatal hypo-
glycemia, Editorial. NEJM 1999;3040:1200–1201.
28. Schwartz RP. Neonatal hypoglycemia: How low is too
low? Editorial. J Pediatr 1997;131:171–173.
29. Alkalay AL, Klein AH, Nagel RA, et al. Neonatal non-
persistent hypoglycemia. Neonatal Intensive Care
2001;14:25–34.
30. Cornblath M, Ichord R. Hypoglycemia in the neonate.
Semin Peinatol 2000;24:136–149.
31. Alkalay AL, Sarnat HB, Flores-Sarnat L, et al. Popu-
lation meta-analysis of low plasma glucose thresholds
in full-term normal newborns. Am J Perinatol 2006;23:
115–119.
32. Koh THHG, Aynsley-Green A, Tarbit M, et al. Neural
dysfunction during hypoglycemia. Arch Dis Child
1988;63:1353–1358.
33. Kinnala A, Rikalainen H, Lapinleimu H, et al. Cere-
bral magnetic resonance imaging and ultrasonogra-
phy findings after neonatal hypoglycemia. Pediatrics
1999;103:724–729.
34. McGowan JE. Neonatal hypoglycemia. NeoReviews
July 1999; e6-15. Available at: www.neoreviews.org
or from the AAP Reference Library.
35. AAP Committee on Fetus and Newborn, American
Academy of Pediatrics. Routine evaluation of blood
pressure, hematocrit, and glucose in newborns. Pedi-
atrics 1993;92:474–476.
36. Vannucci RC, Vannucci SJ. Hypoglycemic brain in-
jury. Semin Neonatol 2001;6:147–155.
37. Koivisto M, Blanco-Sequeiros M, Krause U. Neonatal
symptomatic and asymptomatic hypoglycemia: A fol-
low-up study of 151 children. Dev Med Child Neurol
1972;14:603–614.
38. Brand PLP, Molenaar NLD, Kaaijk C, et al. Neurode-
velopmental outcome of hypoglycemia in healthy,
large for gestational age, term newborns. Arch Dis
Child 2005;90:78–81.
39. Hawdon JM. Hypoglycemia and the neonatal brain.
Eur J Pediatr 1999;158:S9–S12.
40. Hawdon JM, Ward Platt MP, Aynsley-Green A.
Neonatal hypoglycemia: Blood glucose monitoring
and infant feeding. Midwifery 1993;9:3–6.
41. Ho HT, Yeung WKY, Young BWY. Evaluation of
HYPOGLYCEMIA IN BREASTFED NEONATES 261
“point-of-care” devices in the measurement of low
blood glucose in neonatal practice. Arch Dis Child Fe-
tal Neonatal Ed 2004;89:F356–F359.
42. Sirkin A, Jalloh T, Lee L. Selecting an accurate point-
of-care testing system: Clinical and technical issues
and implications in neonatal blood glucose monitor-
ing. JSPN 2002;7:104–112.
43. Ellis M, Manandhar DS, Manandhar N, et al. Com-
parison of two cotside methods for the detection of
hypoglycemia among neonates in Nepal. Arch Dis
Child Fetal Neonatal Ed 1996;75:F122–125.
44. Dahlberg M, Whitelaw A. Evaluation of HemoCue
blood glucose analyzer for the instant diagnosis of hy-
poglycaemia in newborns. Scand J Clin Lab Invest
1997;57:719–724.
45. Sharief N, Hussein K. Comparison of two methods of
assessment of whole blood glucose in the neonatal pe-
riod. Acta Paediatr 1997;86:1246–1252.
46. Cowett RM, Loughead JL. Neonatal glucose metabo-
lism: Differential diagnosis, evaluation, and treatment
of hypoglycemia. Neonat Network 2002;21:9–19.
47. de Lonlay P, Giurgea I, Touati G, et al. Neonatal
hypoglycaemia: Aetiologies. Semin Neonatol 2004;9:
49–58.
48. Sunehag AL, Haymond MW. Glucose extremes in
newborn infants. Clin Perinatol 2002;29:245–260.
49. Hawdon JM. Neonatal hypoglycemia: The conse-
quences of admission to the special care nursery. Child
Health 1993;Feb:48–51.
50. Haninger NC, Farley CL. Screening for hypoglycemia
in healthy term neonates: Effects on breastfeeding. J
Midwifery Women’s Health 2001;46:292–301.
51. American Academy of Pediatrics, Section on Breast-
feeding, Policy Statement: Breastfeeding and the use
of human milk. Pediatrics 2005;115:496–506.
52. National Childbirth Trust, United Kingdom. Hypo-
glycemia of the Newborn: Guidelines for appropriate
blood glucose screening and treatment of breast-fed
and bottle-fed babies in the UK. Midwives 1997;110:
248–249.
53. Nicholl R. What is the normal range of blood glucose
concentrations in healthy term newborns? Arch Dis
Child 2003;88:238–239.
54. AAP & ACOG. Guidelines for Perinatal Care, 5th ed.
American Academy of Pediatrics, Elk Grove Village,
IL, 2002.
55. WHO/UNICEF. Protecting, Promoting and Supporting
Breast-Feeding: The Special Role of Maternity Services. A
Joint WHO/UNICEF Statement. World Health Orga-
nization, Geneva, 19890.
56. Neville MC, Morton J, Umemura S. Lactogenesis: The
transition from pregnancy to lactation. Pediatr Clin
North Am 2001;48:35–52.
57. Scammon RE, Doyle LO: Observations on the capac-
ity of the stomach in the first ten days of postnatal
life. Am J Dis Child 1920;20:516–538.
58. Lilien LD, Phildes RS, Srinivasan G, et al. Treatment
of neonatal hypoglycemia with minibolus and intra-
venous glucose infusion. J Pediatr 1980;97:295–298.
Address reprint requests to:
Nancy E. Wight, M.D., FABM
Division of Neonatology
Children’s Specialists of San Diego
3020 Children’s Way, MC 5102
San Diego, CA 92123-4282
E-mail: wightsd@aol.com
APPENDIX A
To convert mmol/L of glucose to mg/dL,
multiply by 18.
To convert mg/dL of glucose to mmol/L,
divide by 18 or multiply by 0.055.
WIGHT262
