CE: Swati; ACO/300303; Total nos of Pages: 5;

ACO 300303

REVIEW

CURRENT
OPINION A different perspective: anesthesia for extreme
premature infants: is there an age limitation or
how low should we go?
Per-Arne Lönnqvist a,b

Purpose of review
To put in perspective, the various challenges that faces pediatric anesthesiologists because of the recently
lowered limits with regards to the viability of a fetus. Both medical and ethical considerations will be
highlighted.
Recent findings
Issues related to: who should anesthetize these tiny babies; can we provide adequate and legal monitoring
during the anesthetic; does these immature babies need hypnosis and amnesia and the moral/ethical
implications associated with being involved with care of doubtful long-term outcome are reviewed.
Summary
There does currently not exist sufficient research data to provide any evidence-based guidelines for the
anesthetic handling of extreme premature infants. Current practice relies on extrapolations from other
patient groups and from attempting to preserve normal physiology. Thus, focused research initiatives within
this specific field of anesthesia should be a priority. Furthermore, in-depth multiprofessional ethical
discussions regarding long-term outcome of aggressive care of extremely premature babies are urgently
needed, including the new concepts of disability-free survival and number-need-to-suffer.
Keywords
anesthesia, ethics, extreme premature, infant, medical

INTRODUCTION Several issues related to this context will be

The evolution of modern medicine is very impressive
but is also associated with new and often unforeseen
difficulties, some involving important ethical issues.
One very topical illustration of this is the fact
that many first world countries have decided to
lower the limit for viability of a fetus from previous
24 gestational weeks to only 22 gestational weeks.
Thus, a fetus born at 22 gestational weeks should
now be subjected to full resuscitation and neonatal
intensive care if there are not specific circumstances
against this. Not only does this result in concerns
regarding the right to late abortion because of med-
ical reasons but it does also provide new ‘challenges’
for pediatric anesthesiologists.
The number of surgical procedures performed in
these ultra-premature babies are limited but some
represents quite significant surgeries, for example,
surgical ductus closure and laparotomy for perfo-
rated and septic necrotizing enterocolitis, which are
associated with substantial pain and neuroendo-
crine stress responses.
highlighted in the following sections.
WHO IS BEST SUITED TO TAKE CARE OF
THESE ULTRA-PREMATURE BABIES
DURING MAJOR SURGERY?
The ultra-premature babies are exclusively handled
by neonatologists, who readily takes care of tasks
frequently performed by anesthesiologists, for
example, endotracheal intubation, venous and
a
Department of Physiology & Pharmacology, Karolinska Institutet and
b
Paediatric Anaesthesia, PICU & ECMO Services, Karolinska University
Hospital, Solna, Stockholm, Sweden
Correspondence to Per-Arne Lönnqvist, MD, DEAA, FRCA, PhD, Pae-
diatric Anaesthesia, PICU & ECMO Services, Karolinska University
Hospital, Solna, SE 171 76 Stockholm, Sweden.
Tel: +46 70 721 06 50; fax: +46 8 5177 7260;
e-mail: per-arne.lonnqvist@ki.se
Curr Opin Anesthesiol 2018, 31:000–000
DOI:10.1097/ACO.0000000000000581

0952-7907 Copyright ß 2018 Wolters Kluwer Health, Inc. All rights reserved. www.co-anesthesiology.com

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: Swati; ACO/300303; Total nos of Pages: 5;
ACO 300303
Pediatric anesthesia
KEY POINTS
 Data is clearly insufficient to allow for any evidence-
based guidelines regarding anesthesia of the extreme
premature infant.
 Further research is urgently needed in this field
of anesthesiology.
 Adequate analgesia is of paramount importance for
reduction of the neuroendocrine stress response and for
patient wellbeing.
 Performing anesthesia in extreme premature infants is
associated with very important ethical issues; to just
provide service may not always be the right thing
to do.
arterial line placements, transfusion of blood prod-
ucts and titration of vasoactive medications. Thus,
the exposure to these important tasks in this specific
group of patients is very limited even for the dedi-
cated pediatric anesthesiologist.
As our patients deserve the best competency
(and the parents demand it!), it does seem logical
that the neonatologists perhaps should replace the
pediatric anesthesiologists in this specific setting. By
doing so, we would also resolve the ‘blame game’
that on occasion takes place. Many of us have expe-
rienced our neonatology colleagues complaining of
us giving too much fluids and opioids, criticizing
our chosen ventilator strategy and using excessive
inspired oxygen fractions, and even blaming the
occurrence of an intracranial hemorrhage on the
anesthetic team. Depending on the judicial frame-
work, this could potentially have legal benefits for
pediatric anesthesiologists to hand over this task to
the neonatologists.
A further benefit for the patient is that it will
avoid a hazardous transport from the NICU to the
operating room as neonatologists most certainly
will prefer to take care of the babies on their own
home turf.
Contrary to neonatologists, we anesthesiolo-
gists have specific knowledge regarding the use of
volatile agents, muscle relaxants and the use of
ketamine. However, with all due respect for our
specialty, I do believe that neonatologists are fully
capable of mastering the use of muscle relaxants and
ketamine after a limited educational effort. Thus, to
perform high-dose opioid anesthesia in reasonably
stable ultra-premature babies [e.g. surgical patent
ductus arteriosus (PDA) repair] or ketamine anesthe-
sia in hemodynamically unstable babies [e.g. perfo-
rated septic necrotizing enterocolitis (NEC)] does
in my opinion appear to be well within reach for
2 www.co-anesthesiology.com
Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
adequately upgraded neonatologists. Remains, the
issue of hypnosis and volatile anesthetics.
IS THERE AN ABSOLUTE NEED TO
PROVIDE HYPNOSIS AND AMNESIA FOR
THE ULTRA-PREMATURE BABY?
Premature babies 22–24 gestational weeks have not
yet developed the corticothalamic connections that
are required to truly cortically process various
important input from the periphery [1]. This can
potentially be interpreted as if the fetus within this
age bracket ‘lacks the prerequisites for having a soul’
[1]. Surgery for PDA or NEC rarely take place before
25–27 gestational weeks but it is still questionable to
what extent the corticothalamic connectivity has
matured.
Thus, although adequate analgesia and hemo-
dynamic control are essential, it is not currently
clear if the ultra-premature baby truly benefit from
hypnosis and amnesia. Furthermore, premature
babies are hemodynamically very sensitive to both
propofol and sevoflurane, and thus, easily react with
hypotension if these agents are administered [2–5].
The issue whether premature babies and more
term neonates benefit from proper intraoperative
hypnosis and amnesia has duly been questioned by
Davidson [6]. Apart from the risk of hypotension
caused by volatile agents or propofol, these agents
may also add further burden with regards to the
neurodevelopmental outcome of these tiny patients
because of the potential neurotoxic properties of
these agents, especially in these very immature
&&
babies [7 ].
If accepting these argument, a high-dose opioid
or adequate ketamine anesthetic, perhaps laced with
a small dose of alpha-2 agonists (that appear to
ameliorate the neurotoxicity of other anesthetics)
& &
[8 ,9 ,10] appear an interesting alternative.
Furthermore, whenever excluding volatile
agents, there is no need to change the baby’s NICU
ventilator to an anesthetic machine [11]. Thereby,
a potentially hazardous maneuver that often is
accompanied by a period of unstable oxygenation
is avoided.
THE ISSUE OF INTRAOPERATIVE
MONITORING
To properly monitor a variety of different physio-
logic parameters is of paramount interest whenever
performing anesthesia. In fact, both in the United
States and in Europe, there are imperative guidelines
regarding what monitoring must be performed dur-
ing an anaesthetic (https://www.asahq.org//media/
Sites/ASAHQ/Files/Public/Resources/standards-
Volume 31  Number 00  Month 2018
CE: Swati; ACO/300303; Total nos of Pages: 5;
ACO 300303
guidelines/standards-for-basic-anesthetic-monitoring.
pdf; http://www.eba-uems.eu/resources/PDFS/safety-
guidelines/EBA-Minimal-monitor.pdf). The require-
ments are not optional, and there are no general
exemptions made for premature babies. Thus, from
a strict legal point of view, it is highly questionable if
we can (or should?) perform anesthesia in this patient
category where it, almost without exception, is virtu-
ally impossible to maintain proper monitoring
throughout the case.
Those of us who are involved with providing
anesthesia for ultra-premature babies all know how
hard or even impossible it is to have even the most
basic degree of monitoring present during a case of
PDA or even more so in a NEC laparotomy proce-
dure. Despite your best efforts, you are frequently
trapped in a situation where you will be flying by the
seats of your pants, mainly having to rely on the
ECG to guide your anesthetic (bradycardia is bad. . .).
I have personally, in one occasion, felt forced to ask
the surgeon to take a blood gas from the aorta just to
get some help in my anesthetic management.
The introduction of near infra-red spectroscopy
(NIRS) is by many seen as the solution in this specific
circumstance but it is not properly validated in
babies less than 2.5 kg and is associated with several
other limitations in this setting [12]. To completely
rely on the NIRS monitor in this situation may, to a
certain extent, be considered wishful thinking and
can provide a false sense of security.
Interestingly, there are no imperative monitor-
ing guidelines whenever performing procedures in
neonatology, which can be taken as a further argu-
ment to hand over this task to our neonatologist
colleagues.
ETHICAL CONSIDERATIONS
Despite not being primarily responsible for the care
of these very small babies, we are medical profes-
sionals who need to perform our services on a proper
ethics foundation and not just see ourselves as pas-
sive tools for the surgeons and neonatologists. Thus,
we should not take part in care that is ethically
questionable.
In the context of anesthesia of the ultra-prema-
ture, the ethics issue can be divided into at least two
major issues.
First, to subject individuals, in this case our most
fragile patients, to the pain and suffering that is
inevitable associated with major surgery, can only
be justified if the chance of long-term outcome is
good or at least acceptable for the group at hand.
However, the neurodevelopmental outcome of
22–26 gestational weeks old premature babies is
clearly questionable [13]. An example of this is that
0952-7907 Copyright ß 2018 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Anesthesia for extreme premature infants Lönnqvist
approximately 30% of the surviving babies will be
diagnosed with autism spectrum disorder already at
3 years of age [14] and that very few will be able to
start school without substantial external support.
Furthermore, many surviving individuals also suffer
from substantial somatic comorbidity. Thus, just to
focus on the number of babies that are discharged
from the NICU alive is a highly questionable mea-
sure of successful NICU care, something that is of a
serious ethical concern.
Second, it is not only the individual that might
be saved to a reasonable life that is of interested in
this context, it is also how many other babies who
must suffer greatly to have this one ‘successful’ case.
I have recently proposed that the outcome of neo-
natal intensive care should stop focusing on mortal-
ity and instead start to adopt the principles of
‘Disability-free survival’ and introduced a new
dimension whenever evaluating various resource-
intensive therapies. In line with the established
entities ‘Number needed to treat’ (NNT) and ‘Num-
ber needed to harm’ (NNH), I suggest that also
‘Number needed to suffer’ (NNS) should be applied
to take into consideration the very real and substan-
tial suffering that other ultra-premature babies must
endure for each patient that is pulled through to a
&& &&
marginal life at best [15 ,16 ,17]. One should also,
in this process, consider the anguish of parents and
siblings as well as the extreme costs and burdens in
the healthcare system. This new way of trying to
more critically evaluate the outcome of aggressive
neonatal care of extremely premature babies
has recently also received support from certain US
centers [18,19].
In summary, it is our duty to take care of patients
who by various reasons have been struck by diseases
such as stroke, dementia and autism. However, it is
not within the ethical scope of healthcare to know-
ingly and willingly create such individuals by over-
aggressive neonatal care that is associated with very
significant pain and suffering. We as pediatric anes-
thesiologists must decide ourselves whether we will-
ingly will take part in such activities or if we just
should say no. It is just not right that we should
be taken hostages by neonatologists and pediatric
surgeon and thereby be coerced into performing
unmonitored anesthesia with very doubtful long-
term outcome. We must be strong enough to take
our own moral stance in the context of caring for the
ultra-premature baby and voice our opinion strongly.
ANESTHESIA OF THE ULTRA-PREMATURE
BABY
Unfortunately, the literature is very sparse and,
thus, no proper evidence-based guidelines exists
www.co-anesthesiology.com 3
CE: Swati; ACO/300303; Total nos of Pages: 5;
ACO 300303

Pediatric anesthesia

on how best to anesthetize this specific group of
patients. However, all the fundamentals of neonatal
anesthesia will of cause apply [20–22]. In general,
the scenario can be broken down into two typical
scenarios that are discussed as follows.
Reasonably stable baby often with
acceptable vascular access
The typical patient in this category is a baby coming
for surgical closure of a PDA, and can be viewed as a
semi-elective case. In this situation, a high-dose
fentanyl anesthetic supplemented by intravenous
atropine and muscle relaxation is an adequate
choice. Remifentanil may be an attractive alterna-
tive to fentanyl because of the faster clearance
[11,23] but the issues of acute tolerance and rebound
&&
pain associated with the use of remifentanil [24 ]
have not yet been adequately assessed in the ultra-
premature baby. If the blood pressure situation
allows, small and incremental doses of clonidine
or dexmedetomidine can be added. If inotropic
support is required, either dopamine or adrenaline
can be used. The use of wound catheters to provide
adequate postoperative analgesia, resulting in opi-
oid sparing, has recently been described [25].
Hemodynamically unstable baby with septic
perforated necrotizing enterocolitis
These patients are often desperately ill at the time of
first surgery. Not infrequently they are already on
massive volume replacement with blood products
and high-dose infusions of vasoactive drugs. Venous
vascular access is frequently borderline or
insufficient and attempts at placing an arterial line
has failed. These patients are often coagulopathic
from their septic condition. This is the scenario
where it is also very difficult or virtually impossible
to get adequate monitoring in place. The baby may
also be on high-frequency oscillatory ventilation,
which makes it impossible to even get the slightest
end-tidal carbon dioxide trace.
It is not possible to give any firm advice on how
to handle this very demanding and unsatisfactory
situation. Personally, I prefer a ketamine anesthetic
supplemented by atropine and muscle relaxation.
This is the situation where it may be adequate to
request a blood gas drawn from the surgical field to
at least get any information to guide your anesthetic
by. It also represents the situation where a quick but
adequate preanesthetic discussion should be taken
between the anesthesiologist, neonatologist and
pediatric surgeon to assess whether it is the interest
of the baby and its parents to progress with active
treatment, including a surgical intervention. If con-
sensus cannot be reached, my personal view, con-
sidering the long-term outcome of these babies,
is that we as anesthesiologists should decline to
be involved.
Table 1 represents some of the important issues
that needs to be kept in mind whenever anesthetiz-
ing this very specific patient category.
CONCLUSION
Anesthesia for extreme premature infants still rep-
resents largely uncharted territory where the pedi-
atric anesthesiologists must stick to common sense
based on preserving normal physiology, providing

Table 1. Some issues related to anesthesia of the ultra-premature baby

Hypotension How to define hypotension in need of treatment is not possible, although the common rule of thumb to aim for a
(see [26]) mean arterial pressure equal to the gestational week may provide some type of guide. However, this is very
problematic as most of these babies will have a PDA with left-to-right shunting, which seriously will affect the level
of the diastolic pressure. If the blood gas does not show signs of metabolic acidosis then the blood pressure at
hand might be judged as acceptable.
Hypocapnia Trustworthy end-tidal carbon dioxide traces to estimate arterial pCO2 cannot even at the best of circumstances be
(see [26]) recorded because of the high respiratory rate. It is, therefore, of great value to get repeated either arterial or
venous blood gas analyses during the case. This is sometimes not possible, which results in a highly unsatisfactory
situation. As hypocapnia can produce serious cerebral vasoconstriction, unintentional hyperventilation must be
avoided if possible.
Calcium substitution Both the premature myocardium and the peripheral resistance vessels are highly dependent on plasma calcium
concentration. Both sepsis and the need for substantial transfusion of blood product may reduce plasma calcium.
Thus, it may be considered to administer small incremental doses of intravenous calcium to support cardiac output
and blood pressure, preferably guided by plasma analyses (included in the blood gas analysis).
Choice of In the situation with a severely ill baby with NEC, the massive neuroendocrine stress response makes it illogical to
intraoperative continue with total parental nutrition during the operation as this will produce a clearly problematic substrate
fluid situation with hyperlipidemia and substantial hyperglycemia. A balanced intravenous fluid that contain close-to-
normal sodium concentrations, with adequate glucose content to avoid hypoglycemia, appears a better and more
logic choice. During a major laparotomy, a basic infusion rate of 10 ml/kg/h may suffice, further need for volume
replacement is often better accomplished with plasma or albumin 5%.

4 www.co-anesthesiology.com Volume 31  Number 00  Month 2018

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: Swati; ACO/300303; Total nos of Pages: 5;
ACO 300303
adequate analgesia, and extrapolations from other
areas of anesthesiology. Further research in this
specific field of anesthesia is urgently needed in
order to provide better knowledge and guidelines
on how best to perform anesthesia in these tiny
individuals. An integral part of this should also
be in-depth multiprofessional ethical discussions
regarding long-term outcome of aggressive care
of extremely premature babies, including the new
concepts of disability-free survival and number-
need-to-suffer.
Acknowledgements
None.
Financial support and sponsorship
This work has only received funding from departmental
resources.
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED
READING
Papers of particular interest, published within the annual period of review, have
been highlighted as:
& of special interest
&& of outstanding interest
1. Lagercrantz H. The emergence of consciousness: science and ethics. Semin
Fetal Neonatal Med 2014; 19:300–305.
2. Welzing L, Kribs A, Eifinger F, et al. Propofol as an induction agent for
endotracheal intubation can cause significant arterialhypotension in preterm
neonates. Paediatr Anaesth 2010; 20:605–611.
3. Vanderhaegen J, Naulaers G, Van Huffel S, et al. Cerebral and systemic
hemodynamic effects of intravenous bolus administration of propofol in
neonates. Neonatology 2010; 98:57–63.
4. Smits A, Thewissen L, Caicedo A, et al. Propofol dose-finding to reach optimal
effect for (semi-)elective intubation in neonates. J Pediatr 2016; 179:
54.e9–60.e9.
5. McCann ME, Withington DE, Arnup SJ, et al., GAS Consortium. Differences in
blood pressure in infants after general anesthesia compared to awake
regional anesthesia (GAS Study-a prospective randomized trial). Anesth
Analg 2017; 125:837–845.
6. Davidson AJ. Neurotoxicity and the need for anesthesia in the newborn: does
the emperor have no clothes? Anesthesiology 2012; 116:507–509.
0952-7907 Copyright ß 2018 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Anesthesia for extreme premature infants Lönnqvist
7. Graham MR. Clinical update regarding general anesthesia-associated neu-
&& rotoxicity in infants and children. Curr Opin Anaesthesiol 2017; 30:682–687.
Nice and timely overview of this complex topic.
8. Li J, Xiong M, Nadavaluru PR, et al. Dexmedetomidine attenuates neurotoxicity
& induced by prenatal propofol exposure. J Neurosurg Anesthesiol 2016;
28:51–64.
Highlights the potential neuroprotective properties of dexmedetomidine.
9. Perez-Zoghbi JF, Zhu W, Grafe MR, Brambrink AM. Dexmedetomidine-
& mediated neuroprotection against sevoflurane-induced neurotoxicity extends
to several brain regions in neonatal rats. Br J Anaesth 2017; 119:506–516.
Highlights the potential neuroprotective properties of dexmedetomidine.
10. Laudenbach V, Mantz J, Lagercrantz H, et al. Effects of alpha(2)-adrenoceptor
agonists on perinatal excitotoxic brain injury: comparison of clonidine and
dexmedetomidine. Anesthesiology 2002; 96:134–141.
11. Sammartino M, Garra R, Sbaraglia F, et al. Experience of remifentanil in
extremely low-birth-weight babies undergoing laparotomy. Pediatr Neonatol
2011; 52:176–179.
12. Lönnqvist PA, Wallin M. A healthy measure of monitoring fundamentals.
Pediatr Anesth 2017. (submitted for publication).
13. Serenius F, Kallen K, Blennow M, et al. Neurodevelopmental outcome in
extremely preterm infants at 2.5 years after active perinatal care in Sweden.
JAMA 2013; 309:1810–1820.
14. Johnson S, Hollis C, Kochhar P, et al. Autism spectrum disorders in extremely
preterm children. J Pediatr 2010; 156:525.e2–531.e2.
15. Lönnqvist PA. Medical research and the ethics of medical treatments: dis-
&& ability-free survival. Br J Anaesth 2017; 118:286–288.
First description of the concept number-needed-to-suffer.
16. Lönnqvist PA. The potential implications of using disability-free survival and
&& number needed to suffer as outcome measures for neonatal intensive care.
Acta Paediatr 2017; 107:200–202.
Important discussion regarding outcome of neonatal care of extreme premature
infants.
17. Lönnqvist PA. Number needed to suffer: replying to comments on my paper.
Acta Paediatr 2017; 107:204–205.
18. Kaempf JW, Kockler N, Tomlinson MW. Shared decision-making, value
pluralism and the zone of parental discretion. Acta Paediatr 2017;
107:206–208.
19. Kaempf JW, Tomlinson MW, Tuohey J. Extremely premature birth and the
choice of neonatal intensive care versus palliative comfort care: an 18-year
single-center experience. J Perinatol 2016; 36:190–195.
20. Kinouchi K. Anaesthetic considerations for the management of very low and
extremely low birth weight infants. Best Pract Res Clin Anaesthesiol 2004;
18:273–290.
21. Lönnqvist PA. Major abdominal surgery of the neonate: anaesthetic consid-
erations. Best Pract Res Clin Anaesthesiol 2004; 18:321–342.
22. McCann ME, Soriano SG. Progress in anesthesia and management of the
newborn surgical patient. Semin Pediatr Surg 2014; 23:244–248.
23. Penido MG, Garra R, Sammartino M, Pereira e Silva Y. Remifentanil in
neonatal intensive care and anaesthesia practice. Acta Paediatr 2010;
99:1454–1463.
24. Yu EH, Tran DH, Lam SW, Irwin MG. Remifentanil tolerance and hyperalgesia:
&& short-term gain, long-term pain? Anaesthesia 2016; 71:1347–1362.
Nice overview of this timely topic that may be of well relevance in premature babies.
25. Anell-Olofsson M, Lönnqvist PA, Bitkover C, et al. Plasma concentrations of
levobupivacaine associated with two different intermittent wound infusion
regimens following surgical ductus ligation in preterm infants. Paediatr
Anaesth 2015; 25:711–718.
26. McCann ME, Schouten AN, Dobija N, et al. Infantile postoperative encepha-
lopathy: perioperative factors as a cause for concern. Pediatrics 2014;
133:e751–e757.
www.co-anesthesiology.com 5