Eur J Pediatr (1998) 157 [Suppl 1] : S11–S15 © Springer-Verlag 1998

O. D. Saugstad

Practical aspects
of resuscitating asphyxiated newborn infants

Abstract Of all newborn infants, 5% require some degree of basic life support at birth.
Newborn resuscitation therefore is one of the most frequent procedures carried out in med-
icine. It is therefore important that the routines in use are evidence based. Newborn resusci-
tation can be divided into ten steps: (1) initial stabilisation; (2) evaluation; (3) ventilation;
(4) oxygen supplementation; (5) external heart massage; (6) medication; (7) response as-
sessment; (8) withdrawal; (9) post resuscitation care; and (10) documentation. The proce-
dures used in these steps are rarely based on scientific investigation and there is a need for
more research in this field. Training programmes, identification of risk cases and prepara-
tion for resuscitation should be part of the routine in all delivery units. It is underlined that
the need for oxygen, external heart massage or medication is rare. Most depressed newborn
infants manage well with suctioning, gentle tactile stimulation or a few ventilations with a
bag and mask.

Key words Newborn resuscitation · Room air ventilation

Abbreviations AHA American Heart Association · BD base deficit

It is obvious that a large number of newborn infants re-

Introduction
Approximately 4–7 million (3%–5%) of the 140 million
newborn infants born worldwide every year need some
kind of resuscitation at birth. Of these 0.8–1 million will
die and an equal number develop sequelae due to birth as-
phyxia. In developing countries it has been estimated that
30% of perinatal deaths are due to perinatal asphyxia [16].
In the USA, 6% of all newborn infants require basic life
support in the delivery room, or nursery, constituting
21,000 newborn infants only in the USA every year [2]. In
Europe 2–6/1000 live births develop acute clinical seque-
lae of intra-uterine asphyxia, socalled hypoxic-ischaemic
encephalopathy; of these 25% of the moderate and 75%–
100% of the severe cases seem to later develop major neu-
rodevelopmental sequelae [4].
O. D. Saugstad
Department of Paediatric Research, Rikshospitalet,
N-0027 Oslo, Norway
Tel.: +47 22 86 91 10, Fax: +47 22 86 91 17,
e-mail: o.d.saugstad@rh.uio.no
quire resuscitation at birth both in developed and develop-
ing countries and it is therefore highly important that ex-
isting resuscitation routines are based on scientific evi-
dence. Further, trained personnel and adequate equipment
should be available at every delivery. Since most of the
deliveries in the world still occur at home far away from a
hospital and without health personnel present, it is also
important to develop simple resuscitation routines which
can be handled by traditional birth attendants.
Preparation for birth
It is important to anticipate which infants who are at risk
for needing resuscitation. The antepartum and intrapartum
history can often be a help in predicting which infants will
need resuscitation. Most deliveries, however, are unevent-
ful and the adaptation to extra-uterine life occurs
smoothly. These infants require only to be delivered into a
warm room where they can be dried immediately. Their
cord should be cut with sterile equipment and their breath-
ing pattern should be observed. The mother herself is
S12
most often the best caregiver and provider of warmth,
food and protection from infections. Occasionally the
heart rate of the infant should be monitored preferably by
auscultation or alternatively by palpating the pulse. Only
in some infants should more vigorous resuscitation proce-
dures be carried out [17]. Most of these can be performed
with bag and a face mask but in a few newborn infants en-
dotracheal intubation and more sophisticated treatment
and observation are needed.
Initial stabilisation
As soon as the infant is delivered a clock is started and
during its transport to the resuscitation table the person-
nel should make a preliminary opinion about the serious-
ness of the situation. A completely lifeless and pale in-
fant probably needs a more aggressive approach than a
blue infant with some movements and perhaps also a
feeble cry. At the resuscitation table the heart rate is
quickly recorded by auscultation. An experienced person
needs to listen only a few seconds to get an impression of
the degree of bradycardia. The nose and mouth should
then be suctioned thoroughly but quickly and for not
more than 5–10 s. A bulb syringe or a mechanical suction
aparatus attached to an 8F or 10F suction catheter should
be used and the negative pressure should not exceed 136
cm H2O (100 mm Hg). Suctioning can induce bradycar-
dia and even heart arrest via vagal reflexes especially
when deep suctioning in the oropharynx is performed.
Laryngospasm and pulmonary artery vasospasm have
also been reported after suctioning. Suction bulbs are
less likely than suction catheters to induce cardiac arry-
thmias but they are probably not as efficient. There is no
justification for the practice of routine gastric suctioning
which even may be harmful.
Simultanously the other member(s) of the resuscitation
team should continue drying the infant with warm and
clean towels and changing the initial linens which usually
are wet. This also provides tactile stimulation which may
help initiate breathing in mildly depressed infants. This
initial stabilisation should require only seconds.
The most important indication of the need for resusci-
tation is failure to breathe after birth. If the newborn does
not cry or breathe or is only gasping after the initial
quick suctioning, drying, and tactile stimulation, resuscita-
tion should be started. An apneoic infant with heart rate
> 100/min probably represents primary apnoea and the in-
fant requires often only tactile stimulation and a few in-
flations with a bag and mask. Tactile stimulation of the
spine in the direction from the caudal to the cervical re-
gion causes extension of the body and subsequent expan-
sion of the chest in the newborn rat [3].The mother rat
stimulates its newborn offspring by licking the spine from
caudal to cervical regions [3].
Apnoea and heart rate < 100/min may represent sec-
ondary apnoea and ventilation should be initiated. In most
of these cases the heart rate improves and spontaneous
ventilation develops quickly. Ventilation when needed
should start as soon as possible not many seconds after the
initial stabilisation.
Ventilation
After initial correct stabilisation, positioning, suctioning
and drying, the infants who fullfil the criteria for further
resuscitation should be given ventilatory support. In the
majority of these a bag and face mask is adequate. Differ-
ent bags and masks are available but a bag with a reser-
voir and a “pop off” valve which automatically opens if
the inflation pressure exceeds a certain figure is recom-
mended especially in preterm infants. A self inflating bag
is often preferable since it is independent of gas supply
and ambient air can be used. Only a soft mask provides a
tight seal where the mask is applied to the newborn’s face
to achieve adequate pressures. A soft rimmed mask with a
small dead space gives the best face seal.
The first breath of the newborn often generates a nega-
tive pressure of 70 cm H2O and even more. The first arti-
ficial breath given to the infant therefore requires a high
pressure usually in the range of 30–40 cm H2O or even
higher in the term infant, however, lower in preterm in-
fants. The subsequent inflations usually require less pres-
sure, around 20 cm H2O. It is preferable to have a
manometer coupled to the bag so peak inflation pressures
can be recorded. The recommended rate of ventilation ac-
cording to the American Heart Association (AHA) is
40–60 breaths/min [2].
The best way to monitor a successful ventilation is to
observe movements of the chest wall and to follow the
heart rate. The pulse very quickly picks up during a suc-
cessful resuscitation procedure. At least three out of four
of asphyxiated infants have spontaneous breathing after
establishing these initial steps.
The effect of bag and mask ventilation has been inves-
tigated in some studies [5–8, 14, 15]. In unexpanded lungs
a satisfactory tidal volume is not achieved and bag and
mask ventilation on apnoeic newborn infants is probably
futile. In aerated lungs correct bag and mask ventilation
provides tidal volumes of 5–15 ml.
After adequate ventilation has been established for
15–30 s the heart rate should be recorded for not more
than 5–10 s or preferably an assistant records the heart
rate continuously. If the heart rate is at least 100/min and
spontaneous respirations are established positive pressure
ventilation may be discontinued. If the heart rate is be-
tween 60–80/min and improving, ventilation should con-
tinue. If the heart rate is less than 60–80/min and not im-
proving, assisted ventilation is continued and external
heart massage is started [2].
Ventilating the preterm infant
Preterm infants are more often in need for resuscitation
than term ones, because of hypotonia, their Apgar scores
are also lower. When resuscitating premature infants spe-

cial precautions should be taken. Rapid changes in blood
pressure and cerebral blood flow probably increase the
risk for intracranial haemorrhage in preterm infants. Care-
ful stabilisation and positioning are therefore very impor-
tant in these infants. Abrupt changes in positioning should
be avoided as well as rapid boluses of volume expanders.
Since hypoxia seems to increase the likelihood for in-
tracranial bleeding, adequate resuscitation should be initi-
ated without any delay.
The preterm infant has a high surface area compared
with body mass and a thin skin accelerating heat loss and
increasing the risk for hypothermia. The lung compliance
is lower, and since the respiratory drive and muscles are
weaker than in term infants they have an increased need
for assisted ventilation. Tidal volumes and peak ventila-
tory pressures should be lower than in term infants. The
more preterm, the more vulnerable the infant.
Oxygen supplementation
High concentrations of supplemental oxygen (80%–100%)
are usually recommended in most textbooks and guide-
lines dealing with newborn resuscitation [2, 10, 13].
There is, however, no scientific basis for such recommen-
dations. A pilot study did not show any short-term differ-
ences between room air and 100% oxygen resuscitated in-
fants [9]. This was further tested in the socalled “Resair 2
study” [12]. In this internationally controlled multicentre
study, 288 newborn infants were resuscitated with room
air and 321 with 100% oxygen.This study has shown that
most, probably all, newborn infants can be resuscitated as
efficiently with room air as with 100% oxygen. Even the
most depressed newborn infants seemed to be adequately
resuscitated with room air when assessed by response in
heart rate, acid base status, or neurological status in the
neonatal period. In fact, times from birth to first cry and to
first breath were significantly later in oxygen resuscitated
compared with room air resuscitated newborn infants.
There was even a tendency to a lowered neonatal survival
in 100% oxygen compared with room air resuscitated
newborn infants indicating some unknown adverse effects
of using oxygen even for the short duration of time resus-
citation is performed. These results are a strong reminder
that establishing ventilation in the asphyxiated newborn
infant is the most important goal in resuscitation.
If room air resuscitation is not successful in the course
of 90 s I recommend to supplement with oxygen. In such
cases it is probably optimal to monitor the arterial oxygen
saturation by pulse oximetry. The oxygen supply in all de-
livery units should have a blender so the oxygen concen-
tration can be adjusted to the need. If justified to give sup-
plemental oxygen, for instance if cyanosis or bradycardia
persists more than 90 s, I recommend to start out with
40%–60% of oxygen and adjust according to the clinical
response and if possible oxygen saturations measured by
pulse oximetry. The role of oxygen supplementation in
newborn resuscitation has been discussed [1, 11].
S13
External heart massage
Chest compressions are rarely needed and are not recom-
mended for basic newborn resuscitation. If bradycardia
persists with a heart rate less than 60–80/min and no signs
of improvement, chest compressions should be carried out
according to the guidelines of the AHA [2]. In the Resair
2 study only 1.7% had a heart rate of 60 or less after 3 min
and 1.2% after 5 min. Still almost 20% recieved chest
compression.
The AHA recommends that chest compression should
be performed by placing the two thumbs on the middle
third of the sternum with the fingers encircling the chest
and supporting the back [2]. The thumbs should be placed
side by side on the sternum just below an imaginary line
drawn between the nipples. In small, premature infants
the two thumbs may have to be superimposed instead of
being placed side by side. The tips of the middle finger
and either the index or ring finger of one hand positioned
directly above the chest can also be used for chest com-
pressions. The sternum is compressed 1–2 cm with an
equal time for the relaxation phase. When the spontanous
pulse rate has reached 80/min chest compressions should
be discontinued.
Chest compressions should always be carried out si-
multanously with adequate ventilation. A synchronised
chest compression and ventilatory rate of 3:1 giving 90
compressions and 30 breaths/min is recommended. An
adequately performed resuscitation with chest compres-
sions obviously requires at least two trained persons.
Medications
When oxygenation is established via adequate ventilation
of the asphyxiated infant it is extremely rare that drugs are
needed.
Epinephrine is indicated in asystole or in sustained
bradycardia in spite of adequate ventilation and oxygena-
tion. Epinephrine (1:10, 000 solution) is given intra-
venously or intratracheally in a dose of 0.01–0.03 mg/kg
(0.1–0.3 ml/kg) which may be repeated every 3–5 min if
needed. Bradycardia due to insufficient ventilation tech-
niques should be treated by correction of the technique in-
stead of adminstration of drugs such as epinephrine.
Naloxone hydrochloride is a narcotic antagonist, how-
ever, few studies have documented any benefit of this
drug given during resuscitation. The potential negative ef-
fects are concerning as it increases metabolic require-
ments and redistributes blood flow. Naloxone administra-
tion should therefore be restricted only to infants who are
considered depressed from narcotics administrated to the
mother during labour within 4 h of the delivery and who
require active resuscitation in the immediate post natal pe-
riod. Adequate ventilation should always be established
before administration of naloxone. The appropriate dose
is debated but the AHA recommends an initial dose given
intravenously or intratracheally of 0.1 mg/kg of a 1 mg/ml
S14
or 0.4 mg/ml solution, which can be repeated every 2–3
min as needed [2].
Blood volume expanders during acute resuscitation are
only indicated when there are unmistakable signs of shock
with evidence of acute blood loss, including feto-maternal
haemorrhage. The optimal treatment seems to be unclear,
shock may, however, be treated with repeated transfusions
of volume expanders, usually 10 ml/kg. As volume ex-
panders, 5% albumin or other plasma substitutes, he-
parinized placental blood (administrated through a filter
from a blood transfusion set), 0-negative blood cross
matched with the mothers blood, Ringer lactate, Ringer
acetate, or normal saline can be used. The volume ex-
pander may be given over 5–10 min.
Sodium bicarbonate or Tham has no routine place in
newborn resuscitation. No randomized studies have
shown beneficial effects of these drugs. In fact, sodium
bicarbonate may be counterproductive by decreasing in-
tracellular pH. Only in prolonged cardiac arrest not re-
sponding to other therapy, could administration of sodium
bicarbonate in a dose of 2 meq/kg infused slowly be used.
There is no evidence that atropine or calcium have any
beneficial effect during the acute phase of resuscitation of
the newborn infant.
In the Resair 2 study, 15% of infants were given one or
more drugs, i.e. 7% were given epinephrine, 6% sodium
bicarbonate, 1% received naloxone, and only 0.6% were
infused with albumin or blood transfusion.
Response to resuscitation
In the 609 infants included in the Resair 2 study heart rate
rose from a mean of 92/min at 1 min of age to 112/min at
90 s, and 128 at 3 min, only 8% and 5% of the infants had
a heart rate < 100/min at 3 and 5 min of age respectively.
Apgar scores were a median of 4 at 1 min and increased to
7 at 5 min and 8 at 10 min of age. The median duration of
ventilation was 2 min, 3/4 required less than 4 min, 1/10
10 min or more, with approximately 1/20 being ventilated
more than 30 min. The mean base deficit (BD) and pH in
umbilical cord were 14.3 mmol/l and 7.11 respectively.
After 10 min arterial values did not show much differ-
ence. This means that heart rate is a much more sensitive
indicator of the success of resuscitation than acid-base
status which very slowly normalises.
Post resuscitation care
After a successful resuscitation heat loss should be pre-
vented by drying the child with dry towels. The child
should be labelled and frequently checked with regard to
breathing efforts, respiratory rate, colour, heart rate, signs
of birth injury or malformations. If the resuscitation was
brief and the situation was not too dramatic, the newborn
could be monitored in the nursery provided an adequate
respiration is established. The child could be placed so
skin to skin contact with the mother is obtained. However,
Table 1 The 10 steps of new-
1. Initial stabilisation
born resuscitation
2. Evaluation
3. Ventilation
4. Oxygen supplementation
5. External heart massage
6. Medication
7. Response assessment
Training programmes, identifi- 8. Withdrawal
cation of risk cases, and prepa- 9. Post resuscitation care
ration for resuscitation should 10. Documentation
be part of the routine
this depends on the local conditions and the possibilities
for adequate observation by a trained observer.
Even if the infant is not in need of artificial ventilation
following resuscitation often the child is brought to the in-
tensive care unit for further close follow up which in-
cludes monitoring of the heart rate, ventilation, determi-
nation of arterial pH and blood gases, treatment of any hy-
potension with volume expanders or pressors, appropriate
fluid therapy, and treatment of any seizures. The first
hours following the resuscitation the newborn infant
should be regularly screened for any hypoglycaemia or
hypocalcaemia. Breast-feeding should be encouraged if
possible already 1 h after birth. If the resuscitation was
unsucessful and the child died the parents need a close
follow up.
The ten steps of newborn resuscitation
Resuscitation of the newborn infant is one of the most fre-
quent procedures carried out in medicine. Simple and
clear guidelines for this procedure should therefore ex-
ist.Training programmes, identification of risk cases, and
preparation for the resuscitation procedures should be part
of the routine. The resuscitation itself can be divided into
ten steps (Table 1). Not all infants requiring resuscitation
go through all the ten steps listed but the different steps of
resuscitation should be clearly identified and be based on
scientific evidence. Today this is not the case and there is
obviously a need for more research in this field.
Acknowledgement This study was supported by the Laerdal
Foundation for Acute Medicine.
References
1. Carcillo JA (1997) 100% oxygen in the delivery room is just
fine, for now. Crit Care Med 25:1269
2. Emergency Cardiac Care Committee and Subcommittees,
American Heart Association (1992) Guidelines for cardiopul-
monary resuscitation and emergency cardiac care.VII: Neona-
tal resuscitation. JAMA 268:2276–2281
3. Faridy EE (1983) Instinctive resuscitation of the newborn rat.
Respir Physiol 51:1–19
4. Levene MI, Kornberg J, Williams THC (1985) The incidence
of and severity of postasphyxial encephalopathy in full-term
infants. Early Hum Dev 11:21–28

5. Milner AD, Vyas H, Hopkin IE (1984) Efficacy of facemask
resuscitation at birth. BMJ 289:1563–1565
6. Palme C, Nystrom B, Thunell R (1985) An evaluation of the ef-
ficiency of face masks in the resuscitation of newborn infants.
Lancet II:207–210
7. Palme-Kilander C (1992) Methods of resuscitation in low-Ap-
gar score newborn infants – a national survey. Acta Paediatr
81:739–744
8. Palme-Kilander C, Tunell R (1992) Pulmonary gas exchange
during face mask ventilation immediately after birth. Arch Dis
Child 68:11–16
9. Ramji S, Ahuja S, Thirupuram S, Rootwelt T, Rooth G,
Saugstad OD (1993) Resuscitation of asphyxic newborn infants
with room air or 100% oxygen. Pediatr Res 34: 809–812
10. Roberton NRC (1992) Resuscitation of the newborn. In:
Roberton NRC (ed) Textbook of neonatology. Churchill Liv-
ingstone, Edinburgh, pp 173–195
11. Saugstad OD (1996) Resuscitation of newborn infants; do we
need new guidelines? Prenat Neonat Med 1:26–28
S15
12. Saugstad OD, Rootwelt T, Aalen OO (1997) Resuscitation of
asphyxiated newborn infants with room air or oxygen: an inter-
national randomised trial, the Resair 2 study (submitted)
13. Tyson JE (1992) Immediate care of the newborn infant. In: Sin-
clair JC, Bracken MB (eds) Effective care of the newborn in-
fant. Oxford University Press, Oxford, pp 21–39
14. Vyas H, Milner AD, Hopkin IE, Boon AW (1981) Physiologic
responses to prolonged and slow-rise inflation in the resuscita-
tion of the asphyxiated newborn infant. J Pediatr 99:635–639
15. Vyas H, Milner AD, Hopkin IE (1981) Intrathoracic pressure
and volume changes during the spontaneous onset of respira-
tion in babies born by cesarean section and by vaginal delivery.
J Pediatr 99:787–791
16. World Health Organisation (1991) Child health and develop-
ment: health of the newborn. World Health Organisation,
Geneva
17. Zupan J (1997) Basic newborn resuscitation. World Health Or-
ganisation, Geneva