R E S U S C I T A T I O N P L U S 13 (2023) 100360

Available online at www.sciencedirect.com

Resuscitation Plus
journal homepage: www.elsevier.com/locate/resuscitation-plus

Review
The use of induced hypothermia in extracorporeal
membrane oxygenation: A narrative review

Anthony Moreau a,b, Bruno Levy c,d, Filippo Annoni a,b, Roberto Lorusso e, Fuhong Su a,b,
Mirko Belliato f, Fabio Silvio Taccone a,b,*

Abstract
Despite venovenous or venoarterial extracorporeal membrane oxygenation (ECMO) being increasingly used in patients with severe acute respiratory
disease syndrome, severe cardiogenic shock, and refractory cardiac arrest, mortality rates still remain high mainly because of the severity of the
underlying disease and the numerous complications associated with initiation of ECMO. Induced hypothermia might minimize several pathological
pathways present in patients requiring ECMO; even though numerous studies conducted in the experimental setting have reported promising results,
there are currently no recommendations suggesting the routine use of this therapy in patients requiring ECMO. In this review, we summarized the
existing evidence on the use of induced hypothermia in patients requiring ECMO. Induced hypothermia was a feasible and relatively safe intervention
in this setting; however, the effects on clinical outcomes remain uncertain. Whether controlled normothermia has an impact on these patients com-
pared with no temperature control remains unknown. Further randomized controlled trials are required to better understand the role and impact of
such therapy in patients requiring ECMO according to the underlying disease.
Keywords: Brain, Monitoring, ECMO, Ischemia, Perfusion, Induced hypothermia, Targeted temperature management, Cardiogenic shock,
Acute distress respiratory syndrome, Cardiac arrest

Targeted temperature management (TTM) encompasses all the

Introduction interventions required to reach and maintain a specific level of body
temperature in critically ill patients. In particular, TTM is a common
Extracorporeal membrane oxygenation (ECMO) therapy has been intervention in the management of brain-injured patients, following
increasingly implemented in the management of critically ill patients an anoxic or a traumatic injury.6,7 Within different TTM strategies,
worldwide. Despite the lack of strong evidence, venovenous (V-V the interest in induced hypothermia (IH), i.e., cooling the patient
ECMO) configuration has become a valuable therapy for severe below normal body temperature (<37 °C), has always been very high
acute respiratory distress syndrome (ARDS), which remains refrac- because of its potential protective effects not only on the brain but
tory to all recommended interventions, including protective ventila- also on other organs,8 at least as suggested in experimental studies.
tion, titration of positive end-expiratory pressure (PEEP), and Indeed, minutes or hours after an acute injury (i.e., ischemia, inflam-
prone position,1 while venoarterial (V-A ECMO) mode has been mation, trauma), destructive processes begin at the cellular level,
implemented for cardiogenic shock (CS) and refractory cardiac including cellular apoptosis, mitochondrial dysfunction, excessive
arrest [i.e., extracorporeal cardiopulmonary resuscitation (ECPR)].2 free radical production, reperfusion injury, increased permeability of
Nevertheless, mortality rates in those populations remain high3–5 the blood–brain barrier and of the cellular membranes, accumulation
and there is an urgent need to better understand how to improve of excitatory neurotransmitters, production of pro-inflammatory
the performance of ECMO devices to minimize complications and cytokines, and microthrombi formation.8–10 These processes may
potentially provide organ protection. continue hours or days after the initial trigger, and most of those

* Corresponding author at: Department of Intensive Care, Hopital Universitaire de Bruxelles (HUB), Université Libre de Bruxelles (ULB), Route de
Lennik, 808, 1070 Brussels, Belgium.
E-mail address: [email protected] (F.S. Taccone).
https://doi.org/10.1016/j.resplu.2023.100360
Received 28 November 2022; Received in revised form 28 December 2022; Accepted 9 January 2023

2666-5204/Ó 2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/
licenses/by/4.0/).
2 R E S U S C I T A T I O N P L U S 13 (2023) 100360

are temperature-dependent, i.e., they are enhanced by increasing beta-blocker agents or increasing sedatives. Although experimental,
body temperature.8 In this setting, IH can diminish the systemic IH is a practical option to reduce QCO and therefore increase QECMO/
and cerebral metabolism, resulting in a reduction in glucose and oxy- QCO ratio.27 Indeed, basal metabolic rate decreases by 6–10% for
gen consumption, which would minimize the cellular metabolic dis- each degree of reduction in body temperature below 37 °C8,28; as
tress.8,11 IH can also decrease the carbon dioxide production, such, oxygen consumption and carbon dioxide production will also
which would decrease the risk of intracranial and pulmonary hyper- decrease by a similar proportion, resulting in higher arterial oxygen
tension.8 By inhibiting the caspase activation,12 IH would also pre- pressure (PaO2) and decreased arterial carbon dioxide pressure
vent mitochondrial dysfunction and interrupt the apoptotic pathway, (PaCO2), without changes in the ventilator or ECMO settings. In par-
thus preventing cellular death.13 Finally, IH can attenuate the neu- ticular, tissues and organs with a high baseline oxygen consumption,
roexcitatory cascade, which occurs after cerebral ischaemia–reper- similar to the brain or the heart, have a proportionally greater reduc-
fusion,14 reduce the local neutrophil and macrophage activation, tion in oxygen consumption. Interestingly, reduction in carbon diox-
which would sustain neuroinflammation,8,15 and limit the production ide production may favor an even further reduction in minute
of free radicals.16 In addition to these cellular effects, IH can also ventilation (i.e., tidal volume and respiratory rate), reducing the res-
affect the cardiovascular system [i.e., elevation of mean arterial pres- piratory mechanical power29 and then preventing the risk of
sure and stroke volume (SV)].17 ventilator-induced lung injury. Moreover, IH can enhance fat metabo-
In this review, we summarized the existing evidence on the use of lism, leading to increased levels of glycerol, free fatty acids, ketonic
TTM [i.e., including IH or controlled normothermia (NT)] in adult acids, and lactate, resulting in mild metabolic acidosis8; through the
patients requiring ECMO. The PubMed database was searched from Bohr effect, acidemia then induces a reduced hemoglobin affinity for
1990 to 10 December 2022 using the keywords “((ECMO) OR (extra- oxygen and high oxygen tissue availability.25,30
corporeal membrane oxygenation) OR (ECPR) OR (extracorporeal CS is the most severe form of acute heart failure and is frequently
cardiopulmonary resuscitation) OR (ECLS) OR (extracorporeal life associated with an increased risk of death.31 There are numerous
support)) AND ((hypothermia) OR (target temperature management) causes of CS, but the pathophysiology comprises several unique
OR (TTM)). Moreover, the following MESH terms were also used – overlapping components31: an initial cardiac insult decreasing car-
(ECMO) AND (hypothermia, induced)” – to identify the most relevant diac output (CO), central hemodynamic alterations, microcirculatory
publication in this field. Only peer-reviewed published studies (case dysfunction, systemic inflammatory response syndrome, and multior-
reports, retrospective/prospective studies, randomized controlled gan dysfunction. Central hemodynamic alterations include a reduc-
studies, and meta-analyses) written in English were considered to tion of stroke volume (SV), reduction in systemic arterial blood
prepare this article. pressure (ABP), and elevation of left ventricular end-diastolic pres-
sure.31 In this setting, IH may result in increased systemic vascular
resistances19 and could potentially stabilize the patients with CS,
The rationale for induced hypothermia during although the reduction in heart rate would result in a slight reduction
ECMO in CO. The use of IH in these patients may also help limit the severity
of the ischemia–reperfusion injury.31 Importantly, V-A ECMO could
The application of TTM in patients requiring V-V or V-A ECMO is be associated with some complications, such as an increased after-
supported by different pathophysiological considerations. In these load (i.e., retrograde arterial flow), an increased myocardial work
two situations, initiation of ECMO is associated with a complex (i.e., more extensive ischemic injury), and an important inflammatory
inflammatory reaction characterized by the activation of coagulative response (i.e., because of the blood contact with a large surface of
and inflammatory cascades, the production of proinflammatory the extracorporeal circuits).32 The use of IH might mitigate all these
cytokines, the activation of the complement, and the innate immune phenomena related to ECMO implementation.
system.18,19 If severe and persistent, this inflammatory response Finally, IH may be interesting to improve neurological outcome in
may lead to endothelial injury, disrupted microcirculation, and organ patients with refractory CA requiring ECPR. Indeed, these patients
dysfunction. In some clinical settings (i.e., ARDS, CS), an over- are exposed to a prolonged cardiopulmonary resuscitation and there-
whelming inflammation is also frequently reported. Because of its fore are more susceptible to develop hypoxic ischemic brain injuries
anti-inflammatory properties, TTM may be then beneficial in these (HIBI). Pathophysiology of HIBI is complex and could be described
conditions.20–23 as a two-hit model; the primary lesion is secondary to the cessation
In patients with refractory and persistent hypoxemia despite V-V of oxygen transport during CA and the secondary lesion because of
ECMO, some strategies are currently available. The main determi- apoptosis, activation of a neuroexcitatory cascade with an excessive
nant of oxygen saturation (SaO2) in patients undergoing V-V ECMO production of glutamate, and an inflammatory response. By limiting
is the ratio of ECMO blood flow to cardiac output (QECMO/QCO). In or minimizing all these mechanisms, IH could therefore be neuropro-
particular, a QECMO/QCO ratio of at least 0.6 was associated with a tective in these patients (Fig. 1).
SaO2 value greater than 90%.24 Indeed, arterial oxygen content
can be improved either by increasing ECMO flow or increasing blood
oxygen delivery (i.e., red blood cell transfusion).25,26 Recirculation, a Induced hypothermia in animal models of
phenomenon that occurs when the arterial cannula and the veinous ECMO
cannula are too close or in case of very high ECMO blood flow, must
also be minimized. Optimalization of mechanical ventilation (i.e., Few studies evaluating the use of TTM in animal models with
PEEP titration, alveolar recruitment maneuvers) or reduction in the ECMO have been published so far; all of them have evaluated
intrapulmonary shunt (i.e., prone position, inhaled nitric oxide) may the use of IH. Moreover, with the exception of one experiment
also be considered. If the patient remains hypoxemic despite all realized in the setting of CS,40 all other studies were conducted
these interventions, a reduction in QCO could be considered using in CA models.33–39
 R E S U S C I T A T I O N P L U S 13 (2023) 100360 3

Fig. 1 – Potential benefits from induced hypothermia (IH) in patients with venoarterial (V-A) or venovenous (V-V)
extracorporeal membrane oxygenation (ECMO) therapy.

Table 1 summarizes the studies conducted on animal models. Induced hypothermia in patients undergoing V-
These studies showed several benefits from IH compared with the V ECMO
NT group. Regardless of the animal species, six studies demon-
strated positive effects of IH on brain physiology by either resulting Implementation of IH in patients requiring V-V ECMO has been
in better neurological outcomes (evaluated by neurological scores poorly reported. The majority of publications on the use of IH in
such as Neurologic Deficit Score or Overall Performance Cate- refractory hypoxemia are case reports describing patients with respi-
gory),33,35,39 reducing neuronal deaths (assessed by histology and ratory failure without ECMO. Three prospective studies described
molecular biology),33,36,39 or improving cerebral oxygenation.36 IH the effects of IH in patients with respiratory failure without V-V ECMO
was also associated with reduced myocardial necrosis on histologi- implementation. Despite some limitations (i.e., unknown tidal vol-
cal and blood samples33,35 and improved haemodynamics.35 Sur- ume, no standardization of IH protocol), Villar et al.41 reported a
vival time was also longer in the IH group compared with lower mortality in the IH group (32–35 °C) compared with the stan-
normothermia.34,38 dard therapy in a small cohort (n = 19) of septic patients with ARDS.
In a swine model of CS requiring V-A ECMO support, Vanhuyse Matsuno et al.42 demonstrated a lower overall mortality in kidney-
et al.40 evaluated the effects of mild IH on the cardiovascular system; transplanted patients admitted to the intensive care unit for acute
lower fluid balance and norepinephrine requirements were required respiratory failure when treated by IH (i.e., 35 °C) rather than no tem-
in the IH group compared with normothermia. Interestingly, higher perature control. Schorgten et al.43 also observed a lower 14-day
vascular reactivity and dP/dtmax, a marker of left ventricular function, mortality in septic patients treated with controlled normothermia
were also observed in the IH group. All these studies have several (36.5–37 °C) compared with no fever control; however, only 50%
limitations. First, the number of animals used for these experiments of these patients had ARDS and no subgroup analysis was realized
was very limited. Second, protocols applied in these studies were not to assess the effects of such intervention on respiratory function and
comparable to the clinical scenario [i.e., healthy animals; short time gas exchanges. Hayek et al.44 reviewed all case reports evaluating
between the initiation of cardiac arrest and return of spontaneous cir- the benefits of IH in patients with acute respiratory failure, reporting
culation (ROSC); short follow-up period]. Third, study protocols were various effects on the avoidance of ECMO and/or improvement in
all different (i.e., species, depth and duration of hypothermia, speed PaO2 values.
of rewarming), which limited all further comparisons. Fourth, different Table 2 summarizes the studies conducted in patients needing V-
temperature ranges were used in the NT groups, according to the V ECMO support for severe ARDS. The first two treated cases of
studies species, which would limit the translation of these findings patients with severe ARDS (PaO2/FiO2 < 100) due to H1N1 infection
to the clinical setting. remained hypoxemic despite V-V ECMO implementation44; IH (32–
 4
Table 1 – Summary of experimental data evaluating the use of induced hypothermia (IH) during the use of extracorporeal membrane oxygenation (ECMO).

Studies Species N Study protocol TTM protocol applied Main outcomes

33
Ao et al., 2001 Dog 17 Induction of VF by electric n = 8: 33 °C for 20 h, then 37.5 °C until the end of NDS significantly lower in IH group
shock the study Less degenerating pyramidal hippocampal neurons at brain biopsy in IH group
No Flow: 15 min n = 9: 37.5 °C throughout the study Myocardial necrosis significantly lower in IH group
ECPR during 24 h Core temperature reached
ICU stay during 72 h 34 °C within the 30 min of the start of the cooling
Han et al., Rodent 35 CA by asphyxia (8 min) n = 10: 37 °C 72 h survival significantly higher in the 2 groups with IH than in NT group (80%
201034 ECPR during 1 h n = 10: 34 °C in mild IH group vs 60% in moderate IH group vs 0% in NT group)
ICU support during 1 h n = 10: 30 °C NDS better at 24 h post resuscitation in the moderate IH group compared to
TTM during additional 6 h n = 5: Sham the other groups
Observation during OPC better in moderate IH group at 72 h post resuscitation than the two other
additional 3 days No information about time to target temperature groups

R E S U S C I T A T I O N P L U S
Ostadal et al., Pig 8 Induction of VF by electric n = 4: 33 °C during ECPR ABP and cerebral oxygenation values significantly higher in IH group
201335 shock n = 4: 36.8 °C during ECPR Levels of troponin I, myoglobin, CPK, and ALT significantly lower in the IH
No Flow: 20 min group
ECPR during 90 min 33 °C within the 5 min of the start of the cooling Levels of NSE and CysC comparable
ICU stay during 72 h
Janata et al., Rodent 70 Induction of VF by electric n = 13: ECPR + NT (37 °C for 12 h after ROSC) NDS, OPC and surviving neurons in hippocampus were not statically different
201336 shock n = 13: ECPR + IH (33 °C for 12 h after ROSC) at day 14 between the 4 groups
No flow: 6 min n = 13: CPR + NT (37 °C for 12 h after ROSC) Significant reduction of neuronal death in subiculum and thalamus in IH
ECPR groups: ECPR n = 21: CPR + IH (33 °C for 12 h after ROSC) groups
weaned from ECMO 2 min n = 10: sham No synergistic benefit from the combination of ECPR and IH
after ROSC
33 °C within the 10 min of the start of the cooling

13 (2023) 100360
Bergan et al., Pig 20 Induction of VF by electric n = 10: 38 °C during 1500 No effect of two-hour IH on resuscitation success
201637 shock n = 10: 32–33 °C during 1200 followed by 300 of No effect of two-hour IH on post-arrest left ventricular function (evaluated by
No Flow: 15 min rewarming LV pressure measurements and MRI)
ECPR during 150 min No effect of two-hour IH on magnitude of myocardial injury (estimated by
33 °C within the 20 min of the start of the cooling serum concentrations of Troponin T and AST)
Foerster et al., Pig 21 Induction of VF by electric n = 11: 37 °C during 600 Seven-day mortality, NDS at 24 h and at 120 h after CA significantly lower in
201838 shock n = 10: 32 °C during 300 followed by 300 of the IH group
No Flow: 20 min rewarming to a target of 36 °C Significantly less pyrexia or seizures at 24 h in IH group
ECPR during 60 min NDS on day 7 lower (NS) in the IH group
32 °C within the 15 min of the start of the cooling
Zhang et al., Pig 18 Induction of VF by electric n = 6: 37 °C during 32 h IH significantly increased Bcl-2 and decreased Bax protein and mRNA expres-
202139 shock n = 6: 34 °C during 24 h followed by 8 h of sion levels, reflecting a decrease in apoptotic cell death in the frontal cortex
No Flow: 6 min rewarming (rewarming rate  0.5 °C/h) Anti-apoptotic effects of IH may be in part mediated by the upregulation of
ECPR till 32 h post ROSC n = 6: 33 °C for 1 hour, alternate 35 °C for another GRP78 and downregulation of CHOP, implicating that ERS may play an
hour during 24 h followed by 8 h of rewarming important role in this process
(rewarming rate  0.5 °C/h) Temperature variability did not affect the anti-apoptotic effects of IH

IH within 2 h post ROSC

Vanhuyse et al.,Pig 12 CS induced by coronary n = 6: 38 °C Lower fluid balance in IH group
201740 ligation n = 6: 34 °C Lower norepinephrine doses required in IH group
Better vascular reactivity in IH group
Better dP/dtmax in IH group
N = number; TTM = targeted temperature management; VF = ventricular fibrillation; ECPR = extracorporeal cardiopulmonary resuscitation; ICU = intensive care unit; NDS = Neurological Deficit Score (NDS 0% = normal, NDS
100% = brain death); IH = induced hypothermia; CA = cardiac arrest; NT = normothermia; ABP = arterial blood pressure; CPK = creatine-phosphokinase; ALT = alanine-aminotransferase; NSE = neuron-specific enolase;
OPC = overall performance category; CysC = cystatin-C; ECMO = extracorporeal membrane oxygenation; ROSC = return of spontaneous circulation; CPR = cardiopulmonary resuscitation; LV = left ventricle; MRI = magnetic
resonance imaging; AST = aspartate-aminotransferase; CS = cardiogenic shock.
Table 2 – Summary of clinical data evaluating the use of induced hypothermia (IH) during the use of extracorporeal membrane oxygenation (ECMO).

Studies Design ECMO Indication N Study protocol Main outcomes

configuration
Hayek et al., 201744 Case report V-V ECMO ARDS 2 IH (32–34 °C) for 24 h Correlation between each degree drop in body
temperature and PaO2 raising and PCO2 drop
No information about time to target temperature

Device used for IH induction: surface cooling

Luc et al., 201945 Case report V-V ECMO ARDS 1 IH (34 °C) for 10 days Correlation between each degree drop in body
temperature and PaO2 raising
No information about time to target temperature

Device used for IH induction: not mentioned

R E S U S C I T A T I O N P L U S
Kimmoun et al., Case report V-V ECMO ARDS 1 IH (34 °C) Correlation between each degree drop in body
201327 temperature and SaO2 raising
No information about the period in IH and the time to
target temperature

Device used for IH induction: heat exchanger

connected to the ECMO circuit
Levy et al., 202250 RCT V-A ECMO CS 374 n = 168: 33–34 °C for 24 h Non-significant reduction in 30-day mortality rate
Multicentric n = 166: 36–37 °C for 24 h in IH group (42% vs 51%; p = 0.15)
Beneficial effects of IH in a composite secondary
outcome (i.e., death, heart transplant, escalation
IH obtained within 2 h post randomization to long-term mechanical support or stroke)

13 (2023) 100360
Similar rate of complications
Device used for IH induction: heat exchanger
connected to the ECMO circuit
Beppu et al., 201356 Case report V-A ECMO CA - ECPR 1 IH (34 °C) for 24 h Good neurological outcome

IH obtained 3 h after collapse

Device used for IH induction: not mentioned

Nusbaum et al., Case report V-A ECMO CA - ECPR 1 IH (32–34 °C) Good neurological outcome
201457
No information about the period in IH and the time to
target temperature

Device used for IH induction: not mentioned

Moreno et al., Case report V-A ECMO CA - ECPR 1 IH (33 °C) for 24 h Good neurological outcome
201458
No information about time to target temperature

Device used for IH induction: heat exchanger

connected to the ECMO circuit
(continued on next page)

5
 6
Table 2 (continued)

Studies Design ECMO Indication N Study protocol Main outcomes

configuration
Thooft et al., 201459 Case report V-A ECMO CA - ECPR 1 IH (33 °C) for 24 h Good neurological outcome

No information about time to target temperature

Device used for IH induction: heat exchanger

connected to the ECMO circuit
Ikejiri et al., 202160 Case report V-A ECMO CA - ECPR 1 IH (34 °C) for 24 h Good neurological outcome

No information about time to target temperature

Device used for IH induction: heat exchanger

R E S U S C I T A T I O N P L U S
connected to the ECMO circuit
Mita et al., 201761 Case report V-A ECMO CA - ECPR 1 IH (34 °C) for 24 h Good neurological outcome

No information about time to target temperature

Device used for IH induction: not mentioned

Kim et al., 201762 Case report V-A ECMO CA - ECPR 1 IH (34.5 °C) for 96 h Good neurological outcome

No information about time to target temperature

Device used for IH induction: surface cooling

Kim et al., 201463 Retrospective V-A ECMO CA - ECPR 55 No information about the period in IH and the time to Relation between the application of IH and an

13 (2023) 100360
Monocentric target temperature increased probability of good neurological out-
come in patients with ECPR
Device used for IH induction: not mentioned
Kagawa et al., Retrospective V-A ECMO CA - ECPR 87 n = 48: <34 °C In ECPR subgroup, target temperatures < 34 °C
201564 Multicentric n = 39  34 °C associated with better neurological outcome at
3 months (82 vs 49%, p = 0.012)
Target temperature and durations of cooling/
rewarming assigned by the medical physicians
regardless of the patients’ conditions
Device used for IH induction: cold saline
(66%), surface cooling with ethanol evaporation (2%)
or with pad (16%), gastric lavage with cold saline
(8%), water circulating mattress (60%) patients,
cooling device attached to a hemodiafiltration system
(9%) patients, heat exchanger with ECMO (28%)
Pang et al., 201765 Retrospective V-A ECMO CA - ECPR 79 n = 14: 34 °C for 24 h Better neurological outcome at hospital discharge
Monocentric n = 65: normothermia for 24 h in patients treated by IH (42.9% vs 15.4%,
p = 0.02)
No information about time to target temperature

Device used for IH induction: heat exchanger

connected to the ECMO circuit
Table 2 (continued)

Studies Design ECMO Indication N Study protocol Main outcomes

configuration
Kim et al., 201966 Retrospective V-A ECMO CA - ECPR 101 n = 25: 33–34 °C for 24 h Good neurological outcome in IH group (32%)
Monocentric n = 76: no control of temperature for 24 h compared with 34% in the other group (p = 0.84)
No difference observed in the rate of hospital sur-
No information about time to target temperature vival (48 vs 46%; p = 0.91)
Limitations:
Device used for IH induction: surface cooling o Level of target temperature in the “control”
group = 35.6 °C
o Indications for IH subjectively determined by
physician and patients’ families
Sakurai et al., Retrospective V-A ECMO CA - ECPR 977 n = 471: TTM Better neurological outcome (CPC 1–2 at 1 month
202267 Multicentric 32 °C (1%) after collapse) in ECPR + TTM group (16 vs 10%,
33 °C (4%) OR (95% CI) 1.546 (1.046 – 2.286), p 0.029)

R E S U S C I T A T I O N P L U S
34 °C (65%) In subgroup analysis according to interval from
35 °C (11%) collapse to pump start (>300 , >450 , >600 ), better
36 °C (19%) neurological outcome in ECPR + TTM group,
n = 506: any control of body temperature except for interval from collapse to pump
start > 600
Median interval from collapse to reach target
temperature: 2490
Median interval of temperature management: 43
hours

Device used for IH induction: surface cooling or heat

exchanger connected to the ECMO circuit

13 (2023) 100360
Nakashima et al., Retrospective V-A ECMO CA - ECPR 609 2 sub-groups analysis: Significantly lower risk for in-hospital mortality in
202268 Multicentric  36 °C for < 12 h (n = 136) the 34–36 °C group when compared with >36 °
34–36 °C for  12 h (n = 250) C group (HR 0.73 [0.55–0.96]; p = 0.025)
 34 °C for  12 h (n = 223)IH duration Significantly lower risk for in-hospital mortality in
(36 °C) the 36 °C for 12–48 h group when compared
< 12 h (n = 136) with 36 °C for <12 h group (HR 0.69 [0.53–
12–48 h (n = 394) 0.90]; p = 0.005)
 48 h (n = 79)

No information about time to target temperature

Device used for IH induction: heat exchanger
connected to the ECMO circuit
(86%), others (14%)
Watanabe et al., Retrospective V-A ECMO CA - ECPR 890 n = 249: 35–36 °C 30-day survival with favorable neurological out-
202269 Multicentric n = 641: 32–34 °C come (CPC 1–2) was similar between the 2
groups (16.5% in NT group, 15.9% in IH group)
No information about period with TTM 30-day TTM complications similar between the 2
Median interval from collapse to reach target groups
temperature: 600 (NT group) and 1100
(IH group)

Device used for IH induction: not mentioned

7
(continued on next page)
 8
Table 2 (continued)

Studies Design ECMO Indication N Study protocol Main outcomes

configuration
Pang et al., 201671 RCT V-A ECMO CA - ECPR 21 Survival rate at discharge with sufficiently good
n = 9: 34 °C for 24 h neurological function observed in 2/9 patients in
n = 12: 37 °C for 24 h the IH group (34 °C for 24 h) and in 1/12 patients
in the NT group (37 °C for 24 h; p = 0.37)
No information about time to target temperature 6-month mortality and the length of hospital stay
similar between groups
Device used for IH induction: heat exchanger Limitations:
connected to the ECMO circuit o Number of patients included very small
o Patients included in IH group younger
o Left ventricular ejection fraction before ECPR
lower in the IH group

R E S U S C I T A T I O N P L U S
Chen et al., 202072 Meta-analysis V-A ECMO CA - ECPR 806 Heterogeneity in TTM protocol applied Association statistically significative between IH
9 studies and neurologic outcome / survival in overall
Huang et al., 202273 Meta-analysis V-A ECMO CA - ECPR 2643 For patients who underwent TTM, goal target No difference in neurologic outcome/survival
35 studies temperature ranged from 33 to 36 °C with a median between TTM and non-TTM group
cooling duration of 24 hours across the studies

No information about the time to target temperature

Duan et al., 202174 Meta-analysis V-A ECMO CA - ECPR 2035 For patients who underwent TTM, goal target ECPR + IH significantly improved the short-term
23 studies temperature ranged from 33 to 36 °C survival (OR = 2.27, 95% CI (1.60–3.23),
p < 0.00001) and neurologic outcomes
No information about the period in IH and the time to (OR = 2.60, 95% CI (1.92–3.52), p < 0.00001)
target temperature At 3 months of follow-up, ECPR + IH significantly

13 (2023) 100360
improved survival (OR = 3.36, 95% CI (1.65–
6.85), p < 0.0008) and favorable neurologic out-
comes (OR = 3.02, 95% CI (1.38–6.63),
p < 0.006)
Bertic et al., 202275 Meta-analysis V-A ECMO CA - ECPR 6793 TTM applied in 37% of patients TTM associated with favorable neurologic out-
92 studies come assessed by CPC score
No information about the period in IH and the time to
target temperature
N = number; ARDS = acute respiratory distress syndrome; TTM = targeted temperature management; ECPR = extracorporeal cardiopulmonary resuscitation; IH = induced hypothermia; RCT = randomized controlled trial;
CA = cardiac arrest; NT = normothermia; V-A ECMO = venoarterial extracorporeal membrane oxygenation; V-V ECMO = venovenous extracorporeal membrane oxygenation; CPR = cardiopulmonary resuscitation;
CS = cardiogenic shock; HR = hazard ratio; CPC = cerebral performance category (CPC 1 = good cerebral performance, CPC 2 = moderate cerebral disability, CPC 3 = severe cerebral disability, CPC 4 = coma/vegetative
state, CPC 5 = brain death); OR = odd ratio; CI = confidence interval.
 R E S U S C I T A T I O N P L U S 13 (2023) 100360 9

34 °C) was therefore maintained for 24 hours using an external are based on the results of several prospective randomized stud-
device. Interestingly, for each degree drop in body temperature, ies.51–55 Currently, there is no recommendation for the use of lower
mean PaO2 concomitantly rose by 7.4 (case 1) and 2.6 (case 2) temperature targets in patients with refractory CA needing an ECMO
mmHg, with an average drop in PaCO2 of 0.8 (case 1) and 1.5 (case support.
2) mmHg, respectively. Luc et al.45 also reported one patient with sev- Table 2 summarizes the studies conducted in patients needing V-
ere ARDS secondary to a Legionella pneumophila community- A ECMO support for refractory cardiac arrest. Most of the published
acquired pneumonia. Despite the implementation of V-V ECMO, data in this field are case reports or retrospective case series. With
the patient remained severely hypoxemic; IH at 34 °C was then the exception of four of them, the time from CA to IH initiation was
induced for 10 days; and an elevation in PaO2 (i.e., from 45 to not reported, while this parameter is the main determinant of cooling
75 mmHg) and SaO2 (i.e., from 80% to 90%) during the cooling phase effectiveness in this setting.
was observed and remained consistent over time. Interestingly, Kim- Few case reports have assessed the effects of IH (33–34 °C) on
moun et al.27 also reported one patient with a refractory hypoxemia neurological outcomes in this setting; IH was applied for 24 hours,
secondary to a severe H1N1 infection; by decreasing body tempera- and the causes of CA were ventricular fibrillation secondary to acute
ture from 37 to 34 °C through the heat exchanger of the ECMO circuit, coronary disease or Brugada syndrome,56–58 Taxus or tricyclic intox-
an improvement in SaO2, from 82% to 94%, was observed. ication,59,60 pheochromocytoma crisis, and amniotic fluid embolism61
To the best of our knowledge, there is no prospective trial evalu- or massive pulmonary embolism.62 All these cases consistently
ating the effects of IH in ARDS patients on V-V ECMO support. Cur- reported a favorable neurological outcome with cardiac recovery,
rently, one study is scheduled in France, which will compare the suggesting the possibility of publication biases.
effects of two levels of body temperature (i.e., 33–34 °C vs 36 °C Kim et al.63 retrospectively analyzed a cohort, including 599
for 48 h) on SaO2 in ARDS patients under V-V ECMO (HypoLun- patients who experienced an out-of-hospital cardiac arrest (OHCA)
gECMO, NCT05306392). Whether aiming at NT in these patients in one single center between 2006 and 2013. Using propensity-
would be beneficial in comparison with untreated fever remains score matching, the authors observed that the application of IH
unknown. was associated with an increased probability of good neurological
outcome in patients with ECPR. However, this study had some lim-
itations; first, no information was given in the manuscript about the IH
Induced hypothermia in patients with V-A protocol that was applied. Second, the criteria for ECPR indication
ECMO for cardiogenic shock and the upper limit of CPR duration for ECMO implantation were
not clearly established. Third, because of a modification in the
Many studies assessing IH in CS animal models without V-A ECMO CPR guidelines during the period of the study, IH was not used
support demonstrated improved physiological parameters, including before 2010 in patients requiring ECPR.
increased contractility and SV, reduced heart rate and left ventricle Kagawa et al.64 investigated whether lower target temperatures
end-diastolic pressure, reduced systemic and myocardial oxygen and/or prolonged cooling could provide improved neurological out-
consumption, increased mean arterial pressure, and higher mixed comes in 237 comatose cardiac arrest survivors between 2003 and
venous oxygen saturation.46,47 Systemic biomarkers of ischemia– 2014. Target temperature and durations of cooling/rewarming were
reperfusion were also reduced with IH in two models of CS.48,49 assigned by the medical physicians regardless of the patients’ condi-
In one randomized trial, Levy et al.50 recently studied the effects tions, which would create a selection bias in the analysis of the over-
of IH in patients with CS needing V-A ECMO support. The authors all results. However, in the ECPR subgroup, target
included 374 patients, 168 in the IH group (33–34 °C for 24 h) and temperatures < 34 °C was associated with better neurological out-
166 in the normothermia group (36–37 °C for 24 h). A trend toward come at 3 months (82% vs 49%, p = 0.012) compared with other
a nonsignificant reduction in 30-day mortality rate was observed in temperature subgroups. The authors hypothesized that the benefits
the IH group (42% vs 51%; p = 0.15). Interestingly, beneficial effects of ECMO were due to faster cooling and better hemodynamic stabil-
of IH were also observed in a composite secondary outcome (i.e., ity in these patients.
death, heart transplant, escalation to long-term mechanical support, Pang et al.65 analyzed retrospectively a cohort of 79 patients
or stroke). However, this study had limitations; only patients requiring requiring ECPR treated between 2003 and 2016; 14 of them
mechanical ventilation and sedation were included. Moreover, there received IH (34 °C for 24 h) with a rewarming rate not exceeding
was consistent heterogeneity in the causes of CS, including a large 0.5 °C/h after 24 h. Compared with patients with ECPR and normoth-
proportion of patients receiving ECMO after cardiac arrest. Finally, ermia, patients treated by IH had better neurological outcome at hos-
the trial was likely underpowered to statistically detect a survival ben- pital discharge (42.9% vs 15.4%, p = 0.02). Kim et al.66
efit of 10% with IH. Nevertheless, the use of IH was not associated retrospectively analyzed a cohort including 101 patients who were
with more complications (i.e., similar rate of infections and bleeding treated with ECPR, 25 of those receiving IH (33–34 °C for 24 h);
between groups). The post hoc Bayesian analysis also suggested at discharge, 32% in the IH group had a good neurological outcome
a potential benefit for V-A ECMO in these patients. compared with 34% in the other group (p = 0.84); moreover, no dif-
ference was observed in the rate of hospital survival (48 vs 46%;
p = 0.91). On the multivariate analysis, the use of IH was not inde-
Induced hypothermia in patients undergoing pendently associated with neurological outcomes and hospital sur-
ECPR vival. Importantly, the level of target temperature in the “control”
group was 35.6 °C. Indications for IH were also subjectively deter-
European guidelines6 currently recommend maintaining a target mined by physician and patients’ families as the use of the active
temperature < 37.7 °C for at least 72 h after ROSC in patients who cooling system was not covered by the national health insurance
remain unconscious after cardiac arrest; these recommendations system.
10 R E S U S C I T A T I O N P L U S 13 (2023) 100360

Sakurai et al.67 recently performed a retrospective analysis of 977 retrospective (i.e., very low levels of evidence and high risk of bias)
patients who underwent ECPR for OHCA between 2014 and 2019; and most of them were unbalanced for baseline characteristics, CA
the use of TTM (defined in the manuscript as using cooling devices characteristics, and initial neurological status between groups. Third,
and have any active target temperature strategy) was applied in 471 most of these studies were heterogeneous regarding the inclusion/
of them. In ECPR patients undergoing TTM, target temperature was exclusion criteria for ECPR. Fourth, most of the studies were realized
32 °C (1%), 33 °C (4%), 34 °C (65%), 35 °C (11%), and 36 °C (19%); in Asia and could not be generalizable to other settings with other
the median interval from collapse to reach target temperature and medical standards.
interval of temperature management were 249 minutes and 43
hours, respectively. A higher proportion of patients with favorable
neurological outcome at 1 month was observed in the ECPR group Complications
(16% vs 10%; odds ratio 1.55 [95% CI 1.05–2.29]; p = 0.03). These
effects were not observed in the subgroup of patients with a time Induction of IH is associated with several physiological changes,
from arrest to ECPR exceeding 60 minutes. Importantly, ECPR which could also be harmful in patients requiring ECMO support.8,76
patients receiving TTM were younger and had more frequently a car- Shivering and cutaneous vasoconstriction, ions disturbances (i.e.,
diac etiology of arrest than others. hypomagnesemia, hypokalemia, hypophosphatemia), cardiovascu-
Nakashima et al.68 recently performed a retrospective analysis of lar and hemodynamic effects (i.e., bradycardia, arrhythmias), coro-
the Extracorporeal Life Support Organization Registry, including nary vasoconstriction in patients with severely atherosclerotic
1511 adult patients who underwent ECPR from 2010 to 2019. Of coronary arteries, mild coagulopathy, unpredictable drugs clearance,
those, 849 patients received IH; the mean duration of risks for infections, and impaired bowel function have been largely
temperature  36 °C was 24 hours, and the mean maximum temper- described.
ature within 72 hours of ECPR initiation was 36.9 °C. No significant In one retrospective cohort, Kagawa et al.66 observed a higher
difference in in-hospital mortality at 90 days was observed between rate of arrhythmias and pneumonia in patients with cooling exceed-
the two groups (hazard ratio 1.06 [95% CI 0.93–1.21]; p = 0.39). ing 28 hours during ECMO than in the others. The rate of bleeding
After excluding patients without data on body temperature or who was also significantly higher in patients treated at target
died before IH was completed ended (n = 240), IH was associated temperatures < 34 °C than in others. However, Mecklenburg et al.77
with a lower probability of in-hospital mortality. recently reported the lack of increased risk for major bleeding in
Watanabe et al.69 recently analyzed a retrospective cohort of 890 patients needing V-A ECMO for ECPR and treated with IH compared
OHCA patients undergoing ECPR between 2014 and 2019; 249 with normothermic ECPR. Similarly, two other retrospective stud-
patients were treated with NT (35–36 °C) and 641 with IH (32–34 ° ies65,66 also reported IH to be relatively safe during ECPR. Regard-
C). Whether by using multivariable logistic regression or inverse ing ions disturbances, a retrospective study78 including 116 ECPR
probability weighting, 30-day survival with favorable neurological out- patients recently yielded that IH was associated with hypokalemia
come (CPC 1–2) was similar between the two groups (16.5% in the and hypophosphatemia. Shiba et al.79 also described an increased
NT group and 15.9% in the IH group). risk of pneumonia in ECRP patients treated with IH compared with
In a prospective observational study, OHCA patients undergoing others. In patients with CS needing V-A ECMO support, IH was also
ECPR had a better neurological outcome than those receiving con- associated with a higher requirement of red blood cell transfusion
ventional resuscitation; IH was applied for 24 hours in most of sur- compared with others, although the percentage of bleeding was
vivors in the ECPR-treated group compared with the control not statistically different between groups.50
(91.5% vs 54.1%; p < 0.0001).70 A higher probability of favorable
neurological outcome was also observed for a duration of cooling
of 24–48 hours and a target temperature of 33 °C compared with Conclusions
other TTM strategies.
One randomized controlled trial71 has prospectively evaluated Severe ARDS, refractory CA, and severe CS are relatively frequent
the effects of IH in patients undergoing ECPR. Survival rate at dis- diseases with poor prognosis, which might require the use of V-V or
charge with favorable neurological function was observed in 2/9 V-A ECMO. Implementation of IH in these patients appears to be a
patients in the IH group (34 °C for 24 h) and in 1/12 patients in feasible and valuable intervention with some robust rationale and
the normothermia group (37 °C for 24 h; p = 0.37). Six-month physiological effects. However, the effects on clinical outcome
mortality and the length of hospital stay were also similar between remain uncertain as most of the studies realized in these settings
the groups. However, this study had some limitations. First, the are animal reports, case reports, and retrospective studies. Whether
number of patients included was very small. Moreover, patients controlled normothermia has an impact on these patients compared
included in the IH group were younger, with possibly greater with temperature control remains unknown. In this context, further
potential for neurological recovery compared with older patients. randomized controlled trials are required to better understand the
Left ventricular ejection fraction before ECPR was also lower in role and impact of TTM/IH in patients requiring ECMO, according
the IH group, which might increase the risk of multiple organ fail- to the underlying disease.
ure in these patients.
Four meta-analyses were published evaluating the potential ben-
eficial effects of TH on neurological outcomes after ECPR.72–75
Conflicts of interest
Except for one of them,73 all analyses reported an improvement in
neurological outcomes for ECPR patients treated with IH. However, BL received fees for consulting and research grant from Baxter and
the eligible studies and sample size of the included cohorts were rel- Amomed. RL and FST received consulting fees from Eurosets. FST
atively limited. Second, most of the studies were observational and received lecture fees from BD and ZOLL.
 R E S U S C I T A T I O N P L U S 13 (2023) 100360 11

Conflicts of interest of other authors: none. International Severe Traumatic Brain Injury Consensus Conference
(SIBICC). Intensive Care Med 2019;45:1783–94.
8. Polderman KH. Mechanisms of action, physiological effects, and
CREDIT authors statement complications of hypothermia. Crit Care Med 2009;37:S186–202.
9. Polderman KH. Induced hypothermia and fever control for prevention
and treatment of neurological injuries. Lancet 2008;371:1955–69.
AM and FST contributed to the study conception and design. AM and 10. Sekhon MS, Ainslie PN, Griesdale DE. Clinical pathophysiology of
FST completed data extraction and analysis. AM and FST drafted hypoxic ischemic brain injury after cardiac arrest: a “two-hit” model.
the manuscript. BL, FA, RL, FS, MB, and FST revised the manu- Crit Care 2017;21:90.
script. All authors approved the submitted version of the manu- 11. Erecinska M, Thoresen M, Silver IA. Effects of hypothermia on
script. All authors agreed both to be personally accountable for energy metabolism in Mammalian central nervous system. J Cereb
Blood Flow Metab 2003;23:513–30.
the author’s own contributions and to ensure that questions related
12. Xu L, Yenari MA, Steinberg GK, Giffard RG. Mild hypothermia
to the accuracy or integrity of any part of the work are appropriately
reduces apoptosis of mouse neurons in vitro early in the cascade. J
investigated, resolved, and the resolution documented in the Cereb Blood Flow Metab 2002;22:21–8.
literature. 13. Ning XH, Chen SH, Xu CS, et al. Hypothermic protection of the
ischemic heart via alterations in apoptotic pathways as assessed by
gene array analysis. J Appl Physiol (1985) 2002;92:2200–7.
Acknowledgments
14. Auer RN. Non-pharmacologic (physiologic) neuroprotection in the
treatment of brain ischemia. Ann N Y Acad Sci 2001;939:271–82.
AM and FST were supported by the Fonds Erasme pour la 15. Kimura A, Sakurada S, Ohkuni H, Todome Y, Kurata K. Moderate
Recherche Médicale. hypothermia delays proinflammatory cytokine production of human
peripheral blood mononuclear cells. Crit Care Med
Author details 2002;30:1499–502.
16. Globus MY, Busto R, Lin B, Schnippering H, Ginsberg MD. Detection
a
of free radical activity during transient global ischemia and
Department of Intensive Care, Hôpital Universitaire de Bruxelles recirculation: effects of intraischemic brain temperature modulation. J
b
(HUB), Brussels, Belgium Laboratoire Expérimental des Soins Neurochem 1995;65:1250–6.
Intensifs, Université Libre de Bruxelles (ULB), Brussels, Bel- 17. Götberg M, van der Pals J, Olivecrona GK, Götberg M, Koul S,
giumcService de Médecine Intensive et Réanimation Brabois, CHRU Erlinge D. Mild hypothermia reduces acute mortality and improves
Nancy, Pôle Cardio-Médico-Chirurgical, Vandoeuvre-les-Nancy, hemodynamic outcome in a cardiogenic shock pig model.
d Resuscitation 2010;81:1190–6.
France INSERM U1116, Faculte´ de Médecine, Universite´ de
e 18. Hirthler M, Simoni J, Dickson M. Elevated levels of endotoxin,
Lorraine, 54000 Nancy, France Department of Cardio-Thoracic
oxygen-derived free radicals, and cytokines during extracorporeal
Surgery, Heart & Vascular Centre, Maastricht University Medical membrane oxygenation. J Pediatr Surg 1992;27:1199–202.
Centre, Cardiovascular Research Institute Maastricht (CARIM), 19. Millar JE, Fanning JP, McDonald CI, McAuley DF, Fraser JF. The
Maastricht, The Netherlands fUOC AR 2-Anestesia e Rianimazione inflammatory response to extracorporeal membrane oxygenation
Cardiotoracica Foundation IRCCS Policlinico San Matteo, Pavia, (ECMO): a review of the pathophysiology. Crit Care 2016;20:387.
Italy 20. Matthay MA, Zemans RL, Zimmerman GA, et al. Acute respiratory
distress syndrome. Nat Rev Dis Primers 2019;5:18.
21. Hong SB, Koh Y, Lee IC, et al. Induced hypothermia as a new
approach to lung rest for the acutely injured lung. Crit Care Med
R E F E R E N C E S 2005;33:2049–55.
22. Kohsaka S, Menon V, Lowe AM, et al. Systemic inflammatory
response syndrome after acute myocardial infarction complicated by
1. Papazian L, Aubron C, Brochard L, et al. Formal guidelines: cardiogenic shock. Arch Intern Med 2005;165:1643–50.
management of acute respiratory distress syndrome. Ann Intensive 23. Hochman JS. Cardiogenic shock complicating acute myocardial
infarction: expanding the paradigm. Circulation
Care 2019;9:69.
2. McDonagh TA, Metra M, Adamo M, et al. 2021 ESC Guidelines for 2003;107:2998–3002.
the diagnosis and treatment of acute and chronic heart failure. Eur 24. Schmidt M, Tachon G, Devilliers C, et al. Blood oxygenation and
Heart J 2021;42:3599–726. decarboxylation determinants during venovenous ECMO for
3. Combes A, Hajage D, Capellier G, et al. Extracorporeal Membrane respiratory failure in adults. Intensive Care Med 2013;39:838–46.
Oxygenation for Severe Acute Respiratory Distress Syndrome. N 25. Montisci A, Maj G, Zangrillo A, Winterton D, Pappalardo F.
Engl J Med 2018;378:1965–75. Management of refractory hypoxemia during venovenous
4. Schmidt M, Zogheib E, Rozé H, et al. The PRESERVE mortality risk extracorporeal membrane oxygenation for ARDS. ASAIO J
score and analysis of long-term outcomes after extracorporeal 2015;61:227–36.
membrane oxygenation for severe acute respiratory distress 26. Levy B, Taccone FS, Guarracino F. Recent developments in the
syndrome. Intensive Care Med 2013;39:1704–13. management of persistent hypoxemia under veno-venous ECMO.
Intensive Care Med 2015;41:508–10.
5. Alba AC, Foroutan F, Buchan TA, et al. Mortality in patients with
cardiogenic shock supported with VA ECMO: A systematic review 27. Kimmoun A, Vanhuyse F, Levy B. Improving blood oxygenation
and meta-analysis evaluating the impact of etiology on 29,289 during venovenous ECMO for ARDS. Intensive Care Med
patients. J Heart Lung Transplant 2021;40:260–8. 2013;39:1161–2.
28. Sessler DI. Temperature monitoring. In: Miller RD, editor. Miller’s
6. Nolan JP, Sandroni C, Böttiger BW, et al. European Resuscitation
Council and European Society of Intensive Care Medicine guidelines Anesthesia. Philadelphia: Elsevier, Churchill Livingstone; 2005. p.
2021: post-resuscitation care. Intensive Care Med 2021;47:369–421. 1571–97.
7. Hawryluk GWJ, Aguilera S, Buki A, et al. A management algorithm 29. Cressoni M, Gotti M, Chiurazzi C, et al. Mechanical Power and
for patients with intracranial pressure monitoring: the Seattle Development of Ventilator-induced Lung Injury. Anesthesiology
2016;124:1100–8.
12 R E S U S C I T A T I O N P L U S 13 (2023) 100360

30. Kaufman DP, Kandle PF, Murray I, Dhamoon AS. Physiology, 49. Hamamoto H, Sakamoto H, Leshnower BG, et al. Very mild
Oxyhemoglobin Dissociation Curve. StatPearls. Treasure Island hypothermia during ischemia and reperfusion improves
(FL): StatPearls Publishing; 2021. postinfarction ventricular remodeling. Ann Thorac Surg
31. Chioncel O, Parissis J, Mebazaa A, et al. Epidemiology, 2009;87:172–7.
pathophysiology and contemporary management of cardiogenic 50. Levy B, Girerd N, Amour J, et al. Effect of Moderate Hypothermia vs
shock - a position statement from the Heart Failure Association of the Normothermia on 30-Day Mortality in Patients With Cardiogenic
European Society of Cardiology. Eur J Heart Fail 2020;22:1315–41. Shock Receiving Venoarterial Extracorporeal Membrane
32. Swain L, Reyelt L, Bhave S, et al. Transvalvular Ventricular Oxygenation: A Randomized Clinical Trial. JAMA 2022;327:442–53.
Unloading Before Reperfusion in Acute Myocardial Infarction. J Am 51. Hypothermia after Cardiac Arrest Study Group. Mild therapeutic
Coll Cardiol 2020;76:684–99. hypothermia to improve the neurologic outcome after cardiac arrest.
33. Ao H, Tanimoto H, Yoshitake A, Moon JK, Terasaki H. Long-term N Engl J Med 2002;346:549–56.
mild hypothermia with extracorporeal lung and heart assist improves 52. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose
survival from prolonged cardiac arrest in dogs. Resuscitation survivors of out-of-hospital cardiac arrest with induced hypothermia.
2001;48:163–74. N Engl J Med 2002;346:557–63.
34. Han F, Boller M, Guo W, et al. A rodent model of emergency 53. Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature
cardiopulmonary bypass resuscitation with different temperatures management at 33°C versus 36°C after cardiac arrest. N Engl J Med
after asphyxial cardiac arrest. Resuscitation 2010;81:93–9. 2013;369:2197–206.
35. Ostadal P, Mlcek M, Kruger A, et al. Mild therapeutic hypothermia is 54. Lascarrou JB, Merdji H, Le Gouge A, et al. Targeted Temperature
superior to controlled normothermia for the maintenance of blood Management for Cardiac Arrest with Nonshockable Rhythm. N Engl
pressure and cerebral oxygenation, prevention of organ damage and J Med 2019;381:2327–37.
suppression of oxidative stress after cardiac arrest in a porcine 55. Dankiewicz J, Cronberg T, Lilja G, et al. Hypothermia versus
model. J Transl Med 2013;11:124. Normothermia after Out-of-Hospital Cardiac Arrest. N Engl J Med
36. Janata A, Drabek T, Magnet IA, et al. Extracorporeal versus 2021;384:2283–94.
conventional cardiopulmonary resuscitation after ventricular 56. Beppu K, Doi T, Hosokawa A, et al. Full neurological recovery,
fibrillation cardiac arrest in rats: a feasibility trial. Crit Care Med Cerebral Performance Category (CPC) 1, after 65-minute cardiac
2013;41:e211–22. arrest using percutaneous cardiopulmonary system and therapeutic
37. Bergan HA, Halvorsen PS, Skulstad H, Fosse E, Bugge JF. Does hypothermia. Int J Cardiol 2013;168:4893–5.
therapeutic hypothermia during extracorporeal cardiopulmonary 57. Nusbaum DM, Bassett ST, Gregoric ID, Kar B. A case of survival
resuscitation preserve cardiac function? J Transl Med 2016;14:345. after cardiac arrest and 3½ hours of resuscitation. Tex Heart Inst J
38. Foerster K, Benk C, Beyersdorf F, et al. Twenty minutes of 2014;41:222–6.
normothermic cardiac arrest in a pig model: the role of short-term 58. Moreno J, Magaldi M, Fontanals J, et al. Use of therapeutic
hypothermia for neurological outcome. Perfusion 2018;33:270–7. hypothermia and extracorporeal life support after an unusual
39. Zhang B, Gu Q, Chen X, et al. Temperature Variability Does Not response to the ajmaline challenge in a patient with Brugada
Attenuate the Beneficial Effects of Therapeutic Hypothermia on syndrome. J Cardiol Cases 2014;10:34–8.
Cellular Apoptosis and Endoplasmic Reticulum Stress in the 59. Thooft A, Goubella A, Fagnoul D, et al. Combination of veno-arterial
Cerebral Cortex of a Swine Cardiac Arrest Model. Neurocrit Care extracorporeal membrane oxygenation and hypothermia for out-of-
2021;34:769–80. hospital cardiac arrest due to Taxus intoxication. CJEM
40. Vanhuyse F, Ducrocq N, Louis H, et al. Moderate Hypothermia 2014;16:504–7.
Improves Cardiac and Vascular Function in a Pig Model of Ischemic 60. Ikejiri K, Akama Y, Ieki Y, et al. Veno-arterial extracorporeal
Cardiogenic Shock Treated with Veno-Arterial ECMO. Shock membrane oxygenation and targeted temperature management in
2017;47:236–41. tricyclic antidepressant-induced cardiac arrest: A case report and
41. Villar J, Slutsky AS. Effects of induced hypothermia in patients with literature review. Medicine (Baltimore) 2021;100:e24980.
septic adult respiratory distress syndrome. Resuscitation 61. Mita K, Tsugita K, Yasuda Y, et al. A successfully treated case of
1993;26:183–92. cardiac arrest after Caesarean section complicated by
42. Matsuno N, Nagao T, Uchiyama M, et al. Beneficial effect of induced pheochromocytoma crisis and amniotic fluid embolism. J Anesth
hypothermia on mortality and graft survival in renal transplant 2017;31:140–3.
patients with severe respiratory failure. Transplant Proc 62. Kim YS, Choi W, Hwang J. Resuscitation of prolonged cardiac arrest
1997;29:229–30. from massive pulmonary embolism by extracorporeal membrane
43. Schortgen F, Clabault K, Katsahian S, et al. Fever control using oxygenation. Eur J Cardiothorac Surg 2017;51:1206–7.
external cooling in septic shock: a randomized controlled trial. Am J 63. Kim SJ, Jung JS, Park JH, Park JS, Hong YS, Lee SW. An optimal
Respir Crit Care Med 2012;185:1088–95. transition time to extracorporeal cardiopulmonary resuscitation for
44. Hayek AJ, White HD, Ghamande S, Spradley C, Arroliga AC. Is predicting good neurological outcome in patients with out-of-hospital
Therapeutic Hypothermia for Acute Respiratory Distress Syndrome cardiac arrest: a propensity-matched study. Crit Care 2014;18:535.
the Future? J Intensive Care Med 2017;32:460–4. 64. Kagawa E, Dote K, Kato M, et al. Do Lower Target Temperatures or
45. Luc JGY, Meyer SR, Murtha WJ, Singh G. Therapeutic hypothermia Prolonged Cooling Provide Improved Outcomes for Comatose
as an adjunct to extracorporeal membrane oxygenation for acute Survivors of Cardiac Arrest Treated With Hypothermia? J Am Heart
respiratory distress and refractory hypoxemia. Perfusion Assoc 2015;4:e002123.
2019;34:422–4. 65. Pang PYK, Wee GHL, Huang MJ, et al. Therapeutic Hypothermia
46. Nishimura Y, Naito Y, Nishioka T, Okamura Y. The effects of cardiac May Improve Neurological Outcomes in Extracorporeal Life Support
cooling under surface-induced hypothermia on the cardiac function in for Adult Cardiac Arrest. Heart Lung Circ 2017;26:817–24.
the in situ heart. Interact Cardiovasc Thorac Surg 2005;4:101–5. 66. Kim YS, Cho YH, Sung K, et al. Target Temperature Management
47. Boyer NH, Gerstein MM. Induced hypothermia in dogs with acute May Not Improve Clinical Outcomes of Extracorporeal
myocardial infarction and shock. J Thorac Cardiovasc Surg Cardiopulmonary Resuscitation. J Intensive Care Med
1977;74:286–94. 2019;34:790–6.
48. Haendchen RV, Corday E, Meerbaum S, Povzhitkov M, Rit J, 67. Sakurai T, Kaneko T, Yamada S, Takahashi T. Extracorporeal
Fishbein MC. Prevention of ischemic injury and early reperfusion cardiopulmonary resuscitation with temperature management could
derangements by hypothermic retroperfusion. J Am Coll Cardiol improve the neurological outcomes of out-of-hospital cardiac arrest:
1983;1:1067–80.
 R E S U S C I T A T I O N P L U S 13 (2023) 100360 13

a retrospective analysis of a nationwide multicenter observational Cardiopulmonary Resuscitation (ECPR)? J Intensive Care Med
study in Japan. J Intensive Care 2022 Jun 17;10:30. 2022;37:157–67.
68. Nakashima T, Ogata S, Noguchi T, et al. Association of intentional 74. Duan J, Ma Q, Zhu C, Shi Y, Duan B. eCPR Combined With
cooling, achieved temperature and hypothermia duration with in- Therapeutic Hypothermia Could Improve Survival and Neurologic
hospital mortality in patients treated with extracorporeal Outcomes for Patients With Cardiac Arrest: A Meta-Analysis. Front
cardiopulmonary resuscitation: An analysis of the ELSO registry. Cardiovasc Med 2021;13 703567.
Resuscitation 2022;177:43–51. 75. Bertic M, Worme M, Foroutan F, et al. Predictors of Survival and
69. Watanabe M, Matsuyama T, Miyamoto Y, Kitamura T, Komukai S, Favorable Neurologic Outcome in Patients Treated with eCPR: a
Ohta B. The impact of different targeted temperatures on out-of- Systematic Review and Meta-analysis. J Cardiovasc Transl Res
hospital cardiac arrest outcomes in patients receiving extracorporeal 2022;15:279–90.
membrane oxygenation: a nationwide cohort study. Crit Care 76. Stegman BM, Newby LK, Hochman JS, Ohman EM. Post-myocardial
2022;26:380. infarction cardiogenic shock is a systemic illness in need of systemic
70. Sakamoto T, Morimura N, Nagao K, et al. Extracorporeal treatment: is therapeutic hypothermia one possibility? J Am Coll
cardiopulmonary resuscitation versus conventional cardiopulmonary Cardiol 2012;59:644–7.
resuscitation in adults with out-of-hospital cardiac arrest: a 77. Mecklenburg A, Stamm J, Angriman F, et al. Impact of therapeutic
prospective observational study. Resuscitation 2014;85:762–8. hypothermia on bleeding events in adult patients treated with
71. Pang PY, Wee GH, Hoo AE, et al. Therapeutic hypothermia in adult extracorporeal life support peri-cardiac arrest. J Crit Care
patients receiving extracorporeal life support: early results of a 2021;62:12–8.
randomized controlled study. J Cardiothorac Surg 2016;11:43. 78. Ravipati P, Murray S, Yannopoulos D, Drawz PE, Bartos JA. Impact
72. Chen X, Zhen Z, Na J, Wang Q, Gao L, Yuan Y. Associations of of AKI in Patients with Out-of-Hospital Cardiac Arrest Managed with
therapeutic hypothermia with clinical outcomes in patients receiving VA ECMO. Kidney360 2021;2:1827–30.
ECPR after cardiac arrest: systematic review with meta-analysis. 79. Shiba D, Hifumi T, Tsuchiya M, et al. Pneumonia and Extracorporeal
Scand J Trauma Resusc Emerg Med 2020;28:3. Cardiopulmonary Resuscitation Followed by Targeted Temperature
73. Huang M, Shoskes A, Migdady I, et al. Does Targeted Temperature Management in Patients With Out-of-Hospital Cardiac Arrest –
Management Improve Neurological Outcome in Extracorporeal Retrospective Cohort Study. Circ Rep 2019;1:575–81.