Prevention of hypothermia in patients undergoing orthotopic liver transplantation using the humigard® open surgery humidification system: a prospective randomized pilot and feasibility clinical trial
Weinberg et al. BMC Surgery
Prevention of hypothermia in patients undergoing orthotopic liver transplantation using the humigard® open surgery humidification system: a prospective randomized pilot and feasibility clinical trial
Laurence Weinberg 0 1
Andrew Huang 0
Daniel Alban 0
Robert Jones 2
David Story 3
Larry McNicol 0 1
Brett Pearce 0
0 Department of Anaesthesia, Austin Hospital , Heidelberg , Australia
1 Department of Surgery, and Anaesthesia Perioperative and Pain Medicine Unit, The University of Melbourne , Melbourne , Australia
2 Liver and Intestinal Transplant Unit, Austin Hospital and The University of Melbourne , Heidelberg , Australia
3 Perioperative and Pain Medicine Unit; The University of Melbourne , Victoria , Australia
Background: Perioperative thermal disturbances during orthotopic liver transplantation (OLT) are common. We hypothesized that in patients undergoing OLT the use of a humidified high flow CO2 warming system maintains higher intraoperative temperatures when compared to standardized multimodal strategies to maintain thermoregulatory homeostasis. Methods: We performed a randomized pilot study in adult patients undergoing primary OLT. Participants were randomized to receive either open wound humidification with a high flow CO2 warming system in addition to standard care (Humidification group) or to standard care alone (Control group). The primary end point was nasopharyngeal core temperature measured 5 min immediately prior to reperfusion of the donor liver (Stage 3 − 5 min). Secondary endpoints included intraoperative PaCO2, minute ventilation and the use of vasoconstrictors. Results: Eleven patients were randomized to each group. Both groups were similar for age, body mass index, MELD, SOFA and APACHE II scores, baseline temperature, and duration of surgery. Immediately prior to reperfusion (Stage 3 − 5 min) the mean (SD) core temperature was higher in the Humidification Group compared to the Control Group: 36.0 °C (0.13) vs. 35.4 °C (0.22), p = 0.028. Repeated measured ANOVA showed that core temperatures over time during the stages of the transplant were higher in the Humidification Group compared to the Control Group (p < 0.0001). There were no significant differences in the ETCO2, PaCO2, minute ventilation, or inotropic support. Conclusion: The humidified high flow CO2 warming system was superior to standardized multimodal strategies in maintaining normothermia in patients undergoing OLT. Use of the device was feasible and did not interfere with any aspects of surgery. A larger study is needed to investigate if the improved thermoregulation observed is associated with improved patient outcomes.
Temperature; Anaesthesia; Nasopharyngeal; Thermoregulation; Liver; Transplant
The Victorian Liver Transplantation Unit at Austin Health
provides liver transplantation services to people in
Victoria, Tasmania and parts of Southern New South
Wales, Australia. Over 1000 liver transplants have been
performed since 1988. Local data from our service shows
that during orthotopic liver transplantation (OLT) over
70% of recipients are hypothermic (core temperature less
than 36 °C prior to reperfusion of the donor liver), despite
standardized measures employed to maintain temperature
homeostasis. Hypotherrmia during OLT can result in
cardiac arrhythmias and ischaemia, coagulopathy,
increased allogeneic transfusion, wound infection delayed
post-anaesthetic recovery, shivering and patient discomfort,
and prolonged hospitalization. During OLT, hypothermia
results from the combination of a large and open surgical
wound, prolonged exposure of abdominal organs to room
air, compounded by blood loss, massive transfusion, and
the need for large volumes of intravenous replacement.
Further, being a highly metabolically active organ, removal
of the native liver further diminishes overall heat
production. Replacement with a cold donor liver preserved in ice
introduces an additional hypothermic insult upon
revascularization, and the use of extracorporeal circuits (for
veno-venous bypass or renal replacement therapy) can
further compound thermal stress.
The Humigard® system (Fisher and Paykel Healthcare,
Auckland, New Zealand) is a heat delivery system
allowing insufflation of carbon dioxide (CO2) into the surgical
wound. A schematic overview of the humidification
system is presented in Fig. 1. Using this system wound
ventilation with warmed, humidified CO2 has been
shown to reduce hypothermia in laparoscopic [1–3],
colonic [4, 5] and cardiac surgery [6, 7], however there are
no studies evaluating its use in OLT. Therefore, we
hypothesized that in patients undergoing OLT the use of the
Fisher & Paykel Humigard® system maintains higher
intraoperative temperatures when compared to standardized
multimodal strategies to maintain thermoregulatory
The study was approved by the Austin Health Research
Ethics Committee (HREC no: 2012/04674), and was
conducted between July 2013 to July 2014 at a
university teaching hospital with expertise in liver
transplantation. We retrospectively registered the trial with the
Australian New Zealand Clinical Trials Registry: ACTRN
12616001631493 (registered 25 November 2016).
Inclusion criteria included adult recipients (age >18 years)
undergoing primary OLT. Exclusion criteria included
pregnancy, fulminant hepatic failure, redo-OLT,
requirement for continuous veno-venous bypass,
haemofiltration, and multi-visceral transplantation. All patients
were evaluated preoperatively at a dedicated
anaesthesia pre-admission clinic and provided written
Fig. 1 Schematic representation of the Humigard® surgical humidification system
Primary end point
The primary end point was core temperature measured
5 min prior to reperfusion of the donor liver (Stage 3 −
5 min). Measurement of core temperature was measured
using a nasopharyngeal temperature probe (CareFusion
Incorporation, Australia) inserted in the upper third of the
nasopharynx . Nasopharyngeal temperature has been
reported to be an accurate and precise measurement of
core body temperature [9–11]. As the insertion of a
continuous Cardiac Output Swan-Ganz pulmonary artery
catheter (Edward Lifesciences CCO-Combo, IL, USA) is
part of our standard anaesthesia technique for all patients
undergoing OLT, we also used temperature measurements
from the PAC. The PAC is considered as the gold
standard measurement of core body temperature [12–14].
Secondary end points
Secondary end points included temperature
measurements at the following time points:
Stage 1 + 60 min: (recorded 60 min after start of the
Stage 2 + 30 min (recorded 30 min after start of the
Stage 3 + 5 min: (recorded 5 min post reperfusion)
Stage 3 + 60 min: (recorded 60 min post reperfusion)
Closure of the surgical wound
Other data collected included baseline patients
characteristics, baseline temperature (recorded at surgical
incision), indication for transplant, MELD, SOFA and
APACHE II scores, use of vasoactive drugs and
inotropes, fluid intervention, transfusion requirements,
estimated blood loss, minute ventilation and PaCO2
(measured at the same temperature time points) and
duration of surgery.
Standardization of perioperative temperature
Perioperative temperature homeostasis was standardized
for all participants. One hour prior to surgery,
participants were pre-warmed with a full body warming
blanket (Bair Hugger 3 M™, Model 315) set at 43 °C. On
arrival to the operating room, the ambient operating
room temperature was set at 21 °C, and participants
were placed on sterile full access underbody warming
blanket (Bair Hugger 3 M™, Model 637) set at 43 °C,
placed over a standard operating table. These warming
blanket devices were continued during induction of
anaesthesia and during insertion of all invasive monitoring
lines, after which the full body warming blanket was
replaced with an intraoperative upper body warming
device that covered both upper limbs and face (Bair
Hugger 3 M™, Model 523XL) for the remainder of the
case. During the anhepatic phase, ambient operating
temperature was increased to 23 °C, with no further
adjustments for the remainder of the case.
Intraoperatively, crystalloid and colloid fluid intervention,
including the use of packed red blood cells were delivered via
a Belmont® Rapid Infuser RI-2 system that delivered
preheated fluids at 42 °C. If clinically indicated, fresh
frozen plasma, platelet and cryoprecipitate transfusions
were delivered via a separate HOTLINE® (Smiths
Medical, Kent, UK) intravenous warming fluid device,
set at 42 °C.
Standardization of anaesthesia
Anaesthesia management followed a protocol designed
to standardize patient care. This included insertion of
two arterial lines, central venous and Continuous
Cardiac Output Swan-Ganz pulmonary artery catheters
(Edwards Lifesciences, Vigilance II, CCO- Combo, IL, USA)
used in conjunction with the Vigilance II monitor
(Edwards Lifesciences, IL, USA). Three liver transplant
anesthesiologists and three dedicated transplant surgeons
provided anaesthesia and surgical care to all participants.
For all participants, general anaesthesia was induced
with fentanyl (3–5 ug/kg), propofol (1–2 mg/kg), and
cisatracurium (0.2 mg/kg). Maintenance of anaesthesia
included inhalation of isoflurane in a 50% oxygen:air
mixture. Mechanical ventilation was standardized and a
circle circuit with absorption of CO2 at 1 L/minute fresh
gas flow was used. End-tidal CO2 was maintained at 35–
45 mmHg. Noradrenaline was administered at the
discretion of the anaesthetists maintaining the mean
arterial pressure within 20% of the patients’ preoperative
values. Fluid intervention included the acetate buffered
crystalloid solution, Plasmalyte-148TM
(BaxterHealthcare, Toongabie, NSW, Australia), and Albumex 20%
(CSL, Biotherapy, Victoria, Australia). Blood and other
blood products were administered at the discretion of
the anaesthetists and based on conventional
laboratory coagulation tests, in addition to point of care
As part of our intuition’s OLT surgical protocol, all
patients had a Makuuchi’s incision (reverse L- Incision) for
surgical exposure to allow dissection and hepatectomy
of the native liver. Immediately after surgical excision
and opening of the abdomen, all patients had the
Humigard system deployed into the right cranial quadrant of
the abdomen by the surgical team. The diffuser system
was attached as per device instructions. Participants
randomized to the Humigard system had CO2 delivered
into the abdominal cavity at 37 °C and 100% relative
humidity. Participants randomized to the control group did
not receive delivery of CO2. Removal of the native liver
began with dissection of the hilum carried down to the
hepatic artery. Anhepatic phase commenced when the
portal vein was transected. After the removal of the
native liver, the allograft implantation began with the
suturing of the donor upper vena cava to the hepatic veins
preserving flow through the inferior vena cava using a
partial side clamp (the ‘piggy-back’ technique). After the
portal and arterial anastomoses were completed, the
biliary tract reconstruction was completed using an
endto-end choledocostomy without a T-tube stent. In select
cases, due to inadequacy of the biliary duct size, or
depending on the underlying disease, a Roux-en-Y
hepatojejunostomy was performed. Prior to surgical closure the
Humigard system was removed from the abdomen in all
Samples size calculations were performed using
inferences for means comparing two independent samples
. This was established with an internal audit of 60
anaesthesia charts of patients undergoing OLT in our
institution over the previous 2 years, which
demonstrated a mean (standard deviation) temperature prior
to reperfusion of 35.4 °C (0.8). Nominating a clinically
important difference of 1 °C in the intervention group,
using a two-sided test, with an alpha value of 0.05 and
a desired power of 0.80, the sample size required for
each group was 11 participants. A computer generated
randomization program was used to ensure all that
participants received individual randomization codes.
Random permutations of treatments for each
participant were created using the randomization program
first generator application entering “Humigard Group”
and “Control Group” as the treatment labels.
Participant randomization was sealed in an opaque envelope.
Study investigators, anaesthesiologists and surgeons
were blinded to the intervention. Randomization was
performed by a dedicated liver theatre technician at the
commencement of surgery who either turned the
Humigard device on, or left the device off. Treatment
allocation was revealed after data analysis was
performed. All clinicians involved in postoperative patient
care were also blinded to the intervention.
Continuous data was tested for normality using the
D’Agnostino-Pearson omnibus test. For the primary end
point between groups, comparisons for continuous data
were performed with the Students’s t-test. All test were
considered two-tailed and a p-value <0.05 indicated
statistical significance. Values were reported as mean and
standard deviation (SD) or medians and interquartile
range (IQR). Changes in NPP and PAC temperatures
over the given timepoints were measured with repeated
measures ANOVA. Given the exploratory nature of this
study, no formal adjustment for multiplicity of testing
was undertaken, and p = 0.05 was regarded as significant
for every outcome. This may yield a potential increase in
Type 1 error rate, which is acceptable given the pilot
and feasibility nature of the study. Statistical analyses
were performed using GraphPad Prism version 6.0
(GraphPad Software, La Jolla California). The study is
reported according to the updated CONSORT guidelines
for reporting parallel group randomized trials .
Twenty-six participants consented for this study. Four
participants did not proceed to transplantation, as the
donor organ was considered unsuitable. In total 22
participants were randomized, 11 to the Humigard Group
and 11 to the Control Group. All participants had the
Humigard system positioned without complication.
There were no violations in the study protocol, and all
participants received the interventions as per allocation.
No patients were excluded, and all results have been
included in the final analysis. Both groups were evenly
matched for baseline characteristics, indications for OLT
and intraoperative factors (Table 1). Immediately prior
to reperfusion (Stage 3 − 5 min) (primary end point),
mean (SD) core temperature was higher in the
Humigard Group compared to the Control Group: NPP: 36.0 °
C (0.13) vs. 35.4 °C (0.22), p = 0.028; PAC: 35.9 °C (0.16)
vs. 35.5 °C (0.24), p = 0.14). Temperature at each stage of
the surgical procedure are summarized in Table 2. There
were no significant differences in the ETCO2, PaCO2,
minute ventilation, or inotropic support (Table 2). The
median noradrenaline infusion rate prior to reperfusion
(Stage 3 − 5 min) was 5 ug/min (IQR 0:9) mcg/min in
the Control group and 4 mcg/min (IQR 1:10) in the
Humigard group (p = 0.96) After reperfusion of the
donor liver (Stage 3 + 5 min), mean (SD) NPP
temperature decreased to 34.7 °C (0.23) in the Humigard
Group and to 35.9 °C (0.22) in the Control Group, p =
0.41. Similarly at this time point PAC temperature
decreased to 34.9 °C (0.21) in the Humigard Group and to
35.1 °C (0.28) in the Control group, p = 0.46. Core
temperatures continued to increase in both groups post
reperfusion and after 60 min (Stage 3 + 60 min),
temperature was higher in the Humigard Group compared
to the Control group: Humigard Group NPP: 35.8 °C
(0.16) vs. Control Group NPP 35.7 °C (0.15), p = 0.77;
Humigard Group PAC: 35.8 °C (0.18) vs. Control Group
35.7 °C (0.16), p = 0.59.
At closure of the abdomen, mean (SD) temperatures
were higher in the Humigard group compared to the
Control group NPP: 36.7 °C (0.21) vs. 36.1 °C (0.13), p =
0.0.045; PAC: 36.8 °C (0.23) vs. 36.3 °C (0.09), p = 0.091).
At closure, the median noradrenaline requirements were
4 mcg/min (IQR 3:6) in the Humigard Group and 4
mcg/min (IQR 1:7) in the Control Group (p = 0.8). There
were no differences observed in time on the ventilator,
vasoactive support in the ICU, duration of stay in ICU,
Table 1 Baseline and intraoperative patient characteristics
Body mass index (kg/m2)
Indications for transplantation
Primary sclerosing cholangitis
Primary biliary cirrhosis
Acetate-buffered crystalloid (ml)
Albumex (20%) (ml)
Platelets (pooled adult units)
Red Blood cells (units)
Donor blood administered (ml)
Cell saved blood returned (ml)
Duration of surgery (hours)
Data presented as mean and standard deviation or median and interquartile range
or length of hospital stay (Table 3). Repeated measured
ANOVA showed that both NPP and PAC temperatures
over time were higher in the Humigard Group compared
to the Control Group (p < 0.0001) (Figs. 2 and 3).
We conducted a randomized controlled pilot and
feasibility study in patients undergoing liver transplantation
to evaluate whether the use of the Fisher & Paykel
Humigard® system was superior to standardized
multimodal strategies in maintaining normothermia in
patients undergoing OLT. Patients in the Humigard Group
had higher NPP and PAC temperatures immediately
prior to reperfusion (Stage 3 − 5 min) and at skin
closure. Importantly, use of the device did not interfere with
any aspects of the surgery with 100% adherence and
compliance to the study protocol.
Our study demonstrates that core temperature can be
increased in patients undergoing OLT by insufflating the
wound cavity with humidified CO2 via a gas diffuser.
There may be several mechanisms underlying this.
Firstly, the near 100% relative humidification of the
wound cavity substantially reduces heat loss by
evaporation. Secondly, as this CO2 is warmed to 37 °C, not
only are convective losses greatly reduced, but there is a
net gain of heat by convection because the humidified
CO2 is warmer than the wound edges. Moreover, the
greenhouse properties of CO2, that is, its ability to
absorb and reemit certain wavelengths, reduce heat loss by
radiation effectively insulating the patient by providing a
thermal blanket . The Humigard system has also
Table 2 Temperatures, vasoactive support, End-tidal CO2 and minute ventilation
Stage Control Group Humigard Group
(n = 11) (n = 11)
Baseline Ambient Temp (°C) 21.1 ± 0.1 21.1 ± 0.1
PAC Temp (°C) 35.9 ± 0.1 35.8 ± 0.2
NPP Temp (°C) 35.9 ± 0.1 36.0 ± 0.1
Norad (ug/min) 0 (0,0) 0 (0,0)
PaCO2 (mmHg) 37.6 ± 03.59 38.4 ± 4.50
ETCO2 (mmHg) 32.0 ± 2.4 30.9 ± 1.8
Min Vent (L/min) 5.5 ± 0.2 5.7 ± 0.4
1 + 60 min Ambient Temp (°C) 21.2 ± 0.1 21.3 ± 0.1
PAC Temp(°C) 35.8 ± 0.2 35.7 ± 0.2
NPP Temp (°C) 35.8 ± 0.2 35.8 ± 0.2
Norad (ug/min) 1 (0,2) 0.5 (0,2)
PaCO2 (mmHg) 37.9 ± 4.2 37.4 ± 6.8
ETCO2 (mmHg) 32.1 ± 1.8 31.0 ± 1.3
Min Vent (L/min) 5.9 ± 0.3 5.7 ± 0.5
2 + 30 min Ambient Temp (°C) 22.6 ± 0.4 22.5 ± 0.3
PAC Temp (°C) 35.6 ± 0.3 36.2 ± 0.2
NPP Temp (°C) 35.6 ± 0.23 36.4 ± 0.2
Norad (ug/min) 5 (0,9) 1 (0,7)
PaCO2 (mmHg) 33.0 ± 3.5 36.1 ± 5.8
ETCO2 (mmHg) 27.8 ± 1.2 30.1 ± 1.3
Min Vent (L/min) 6.3 ± 0.3 6.3 ± 0.5
3 − 5 min Ambient Temp (°C) 22.5 ± 0.53 22.6 ± 0.39
PAC Temp (°C) 35.5 ± 0.2 35.9 ± 0.16
NPP Temp (°C) 35.4 ± 0.2 36.0 ± 0.13
Norad (ug/min) 5 (0,9) 4 (1,10)
PaCO2 (mmHg) 33.5 ± 2.9 36.4 ± 5.5
ETCO2 (mmHg) 28.0 ± 1.2 28.9 ± 1.08
Min Vent (L/min) 6.3 ± 0.3 6.4 ± 0.48
3 + 5 min Ambient Temp (°C) 22.5 ± 0.5 22.6 ± 0.5
PAC Temp (°C) 35.1 ± 0.28 34.9 ± 0.21
NPP Temp (°C) 35.0 ± 0.22 34.7 ± 0.23
Norad (ug/min) 8 (5,15) 10 (6,15)
PaCO2 (mmHg) 35.9 ± 3.9 38.3 ± 4.5
ETCO2 (mmHg) 30.1 ± 1.10 29.5 ± 1.00
Min Vent (L/min) 6.3 ± 0.34 7.3 ± 0.42
3 + 60 min Ambient Temp (°C 23.0 ± 0.6 22.9 ± 0.5
PAC Temp (°C) 35.7 ± 0.2 35.8 ± 0.2
NPP Temp (°C) 35.7 ± 0.2 35.8 ± 0.2
Norad (ug/min) 5 (3,10) 6 (5,9)
PaCO2 (mmHg) 35.2 ± 4.2 36.4 ± 4.0
ETCO2 (mmHg) 31.0 ± 1.9 29.9 ± 1.3
Min Vent (L/min) 6.7 ± 0.4 7.14 ± 0.4
Table 2 Temperatures, vasoactive support, End-tidal CO2 and minute ventilation (Continued)
Ambient Temp (°C
Min Vent (L/min)
−0.4 ± 0.7
−2.3 ± 1.6
−1.8 to 1.0
−0.1 to 0.1
−1.2 to 3.1
−5.6 to 1.1
−1.2 to 1.2
Data presented as mean (standard deviation) or median (interquartile range)
Temp temperature, PAC pulmonary artery catheter, NPP nasopharyngeal probe, Norad noradrenalin, ET CO2 end tidal carbon dioxide, Min Vent minute ventilation
been developed to fill an open surgical wound cavity
with a laminar flow of CO2 for the prevention of arterial
air embolism in open-heart surgery [18–21]. However,
the localized warming effect of CO2 also prevents
hypothermia-induced vasoconstriction, which in turn
increases tissue oxygen tension of the wound . As CO2
has a greater density than air when insufflated with a
continuous laminar flow into the wound cavity it
gravitates into the wound flooding the cavity, exerting a
thermal insulating effect, minimizing the redistribution of
heat from the body core to the wound surface, and
shielding the patient from diffusion and convective air
currents caused by the operating room ventilation [6,
17]. This provides a stable environment of warm humid
CO2, which increases the efficiency of these
mechanisms. Given that the surgical incision during OLT is
large and the operation is prolonged, we would also
expect these changes to have a more pronounced effect in
this group of patients.
Results of our study are consistent with the
publication by Frey and colleagues , who conducted a
randomized controlled trial in patients undergoing open
colonic surgery. Frey et al. showed a significant
improvement of approximately 0.6 °C in mean core temperature
when the Humigard was used compared to controls .
Our paper demonstrated temperatures changes of
similar magnitude. Similarly our findings are consistent with
other studies where the creation of a humified CO2
atmosphere in the surgical wound cavity increased the
total wound temperature by 0.5 °C . Whilst such
changes may appear insignificant, mild hypothermia has
been associated with surgical-wound infection [22, 23],
adverse effects on the coagulation system , platelet
dysfunction , increased increases blood loss and
allogeneic transfusion requirements , and increase
adverse cardiac events , all of which can be fatal in the
context of OLT. Given the long duration and complexity
of liver transplantation surgery as well as a high rate of
post-operative complications in this patient group, small
differences in temperature may be clinically important.
Due to the relatively small sample size, differences in
clinical outcomes were not observed, however the
magnitude of temperature changes observed are consistent
with clinical benefits seen in other studies .
“Normal” core temperature in healthy adults range
between 36.5 °C and 37.5 °C” according to the National
Institute for Health and Clinical Excellence guidelines .
It is interesting to note that despite intense
preoperative warming strategies employed in both groups in
our study, baseline core temperatures at the start
fulfilled the definition of “mild” hypothermia in both
groups of patients. Reasons for this include exposure of
the patient during the insertion of invasive lines, the use
of cold antiseptic solution for the line insertions and
impaired thermoregulatory function associated with both
hepatic failure and general anaesthesia. As expected, we
did not observe any significant differences in core
temperature at Baseline and at Stage 1 + 60 min, due to
the lag time required for the Humigard system to
increase core temperature. Similarly, immediately post
reperfusion, there were no significant differences in core
temperatures measured by the NPP or the PAC.
Changes in core temperature immediately post
reperfusion depend on the size and temperature of the donor
Table 3 Intensive care and hospital ventilator times, inotropic support and length of stay
ICU Ventilation time (hours)
ICU Noradrenalin support (hours)
ICU length of stay (hours)
Hospital length of stay (days)
−27.6 to 63.4
−48.4 to 105.9
Data presented as mean and standard deviation, or median and interquartile range
Fig. 2 Changes in nasopharyngeal temperature during orthotopic liver transplantation. Data is presented as mean and standard deviation
liver, donor liver warm and cold ischaemic times, and
temperature and volume of crystalloid solution used to
flush the donor liver prior to reperfusion. The difference
in temperature change between the two groups at
reperfusion was striking. The change of temperature from
prior to reperfusion (Stage 3 − 5 min) to post-perfusion
(Stage 3 + 5 min) differed by approximately 0.7 °C
between the two groups completely reversing the
improvement in core temperature seen in the Humigard group
up to this point. This anomaly not only explains why we
failed to show any significant difference in temperature
at Stage 3 + 5 min but also explains the lack of
statistically significant difference between the groups at Stage
3 + 60 min, given that 1 h is insufficient to warm a
patient by 1 °C. That said, at the Stage 3 + 60 min
timepoint, temperatures were noticeably higher in the
Humigard group compared to the Control group, and
this accelerated warming continued to skin closure,
Fig. 3 Changes in pulmonary artery pressure temperature during orthotopic liver transplantation. Data is presented as mean and
reinforcing the efficacy of the Humigard device in
maintaining thermoregulatory homeostasis during OLT.
This study has several methodological strengths.
Measurements of core temperatures were all objective
variables not amenable to ascertainment bias or
manipulation, and our findings were further strengthened
by measurements of temperature from two validated
physiological temperature-monitoring devices.
Selection bias was minimized by randomization and
blinding. Comprehensive electronic data collection using
electronic temperature monitoring validated the
accuracy of our results further increasing the validity of the
findings. There are several limitations to our study.
First, the sample size was small and the study was
powered to demonstrate changes in core temperature, and
not changes in adverse outcomes associated with
hypothermia. The Humigard system has many other
potential benefits, which were not evaluated in the
present study. The humidification of the wound
prevents dessication of the exposed tissues , which in
turn may prevent ileus and assist tissue recovery. In
addition, given that CO2 is bacteriostatic, a reduced
rate of surgical site infections might be expected .
Pure CO2 significantly decreased the growth rate of
Staphylococcus aureus at body temperature . Its
bacteriostatic effect may explain the low infection rates
in patients who undergo laparoscopic procedures. Frey
and colleagues also showed that the exposed wound
edges were significantly warmer with the Humigard
system . This may be associated with improved
blood flow and improved healing. These benefits
require further clinical research and larger clinical trials
before such findings can be validated in the setting of
OLT. Finally, the use of any surgical suction apparatus
within the abdominal cavity may theoretically deplete
the humidified CO2 layer. We were not able to measure
the impact that the surgical suction had on temperature
homeostasis. However, with a humidified CO2 flow rate
of 10 L/min, any CO2 losses through surgical suction
would be replaced within seconds. As the surgical
suction is not used continuously, we think that it had a
negligible effect on the warming capacity of Humigard
In summary, the use of the Fisher & Paykel Humigard®
system with standard warming strategies was superior to
standardized multimodal strategies in maintaining
normothermia in patients undergoing OLT. Use of the
device was feasible and did not interfere with any aspects
of the surgery. A larger study is needed to investigate if
the improved thermoregulation observed is associated
with improved patient outcomes.
°C: Degrees celsius; APACHE: Acute physiology and chronic health
evaluation; CO2: Carbon dioxide; HREC: Health research ethics committee;
IQR: Interquartile range; MELD: Model for end-stage liver disease;
NPP: Nasopharyngeal temperature probe; OLT: Orthotopic liver
transplantation; PAC: Pulmonary artery catheter; PaCO2: Partial pressure of
carbon dioxide; SOFA: Sepsis-related organ failure assessment
Availability of data and materials
All data generated or analyzed during this study titled “Prevention of
Hypothermia in Patients Undergoing Orthotopic Liver Transplantation using
the Humigard® Open Surgery Humidification System: A Prospective
Randomized Pilot and Feasibility Clinical Trial” are included in this published
LW performed the literature review, designed this study, analyzed the data
and assisted in the writing of the manuscript. He was one of the principle
anaesthetists involved in the management of these patients. BP analyzed the
data and assisted in the writing of the manuscript. He was one of the
principle anaesthetists involved in the management of these patients. DS
analyzed the data and assisted in the writing of the manuscript. He was one
of the principle anaesthetists involved in the management of these patients.
LM assisted in the writing of the manuscript. He was one of the principle
anaesthetists involved in the management of these patients. AH performed
the literature review, collected and assisted in data analysis, and assisted in
the writing of the manuscript. DA performed the literature review, collected
and assisted in data analysis, and assisted in the writing of the manuscript.
RJ assisted in the writing of the manuscript. He was the principle transplant
surgeon involved in the management of these patients. All authors read and
approved the final manuscript.
LW, BP, DS and LM are specialist anaesthetists involved in the management
of liver transplant patients at Austin Hospital. RJ is the Director of Liver and
Intestinal Transplant Unit at Austin Hospital. AH and DA are anaesthesia
residents at Austin Hospital.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
This study has received Ethics Committee approval and all participants
provided written consent.
Ethics Committee: Austin Health Research Ethics Committee
Ethics Approval number: HREC no: 2012/04674
1. Birch DW , Manouchehri N , Shi X , Hadi G , Karmali S. Heated CO(2) with or without humidification for minimally invasive abdominal surgery . Cochrane Database Syst Rev . 2011 ; 19 ( 1 ): CD007821 .
2. Hamza MA , Schneider BE , White PF , Recart A , Villegas L , Ogunnaike B , Provost D , Jones D. Heated and humidified insufflation during laparoscopic gastric bypass surgery: effect on temperature, postoperative pain, and recovery outcomes . J Laparoendosc Adv Surg Tech A . 2005 ; 15 : 6 - 12 .
3. Sammour T , Kahokehr A , Hill AG . Meta-analysis of the effect of warm humidified insufflation on pain after laparoscopy . Br J Surg . 2008 ; 95 : 950 - 6 .
4. Frey JM , Janson M , Svanfeldt M , Svenarud PK , van der Linden JA . Local insufflation of warm humidified CO2 increases open wound and core temperature during open colon surgery: a randomized clinical trial . Anesth Analg . 2012 ; 115 : 1204 - 11 .
5. Frey JM , Janson M , Svanfeldt M , Svenarud PK , van der Linden JA . Intraoperative local insufflation of warmed humidified CO2 increases open wound and core temperatures: a randomized clinical trial . World J Surg . 2012 ; 36 : 2567 - 75 .
6. Frey JM , Svegby HK , Svenarud PK , van der Linden JA . CO2 insufflation influences the temperature of the open surgical wound . Wound Repair Regen . 2010 ; 18 : 378 - 82 .
7. Persson M , van der Linden J. Wound ventilation with carbon dioxide: a simple method to prevent direct airborne contamination during cardiac surgery ? J Hosp Infect . 2004 ; 56 : 131 - 6 .
8. Lee J , Lim H , Son KG , Ko S. Optimal nasopharyngeal temperature probe placement . Anesth Analg . 2014 ; 119 : 875 - 9 .
9. Akata T , Setoguchi H , Shirozu K. Yoshino reliability of temperatures measured at standard monitoring sites as an index of brain temperature during deep hypothermic cardiopulmonary bypass conducted for thoracic aortic reconstruction . J Thorac Cardiovasc Surg . 2007 ; 133 : 1559 - 65 .
10. Cork RC , Vaughan RW , Humphrey LS . Precision and accuracy of intraoperative temperature monitoring . Anesth Analg . 1983 ; 62 : 211 - 4 .
11. Togawa T. Body temperature measurement . Clin Phys Physiol Meas . 1985 ; 6 : 83 - 108 .
12. Hooper VD , Andrews JO . Accuracy of noninvasive core temperature measurement in acutely ill adults: the state of the science . Biol Res Nurs . 2006 ; 8 : 24 - 34 .
13. Giuliano KK , Scott SS , Elliot S , Giuliano AJ . Temperature measurement in critically ill orally intubated adults: a comparison of pulmonary artery core, tympanic, and oral methods . Crit Care Med . 1999 ; 27 : 2188 - 93 .
14. Davie A , Amoore J. Best practice in the measurement of body temperature . Nurs Stand . 2010 ; 24 : 42 - 9 .
15. http://www.stat.ubc.ca/~rollin/stats/ssize/n2.html. Accessed 22 Oct 2015 .
16. Schulz KF , Altman DG , Moher D , for the CONSORT Group. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials . Ann Intern Med . 2010 ; 152 : 726 - 32 .
17. Persson M , Elmqvist H , van der Linden J. Topical humidified carbon dioxide to keep the open surgical wound warm: the greenhouse effect revisited . Anesthesiology . 2004 ; 101 : 945 - 9 .
18. Persson M , Svenarud P , van der Linden J. What is the optimal device for carbon dioxide deairing of the cardiothoracic wound and how should it be positioned ? J Cardiothorac Vasc Anesth . 2004 ; 18 : 180 - 4 .
19. Persson M , van der Linden J. De-airing of a cardiothoracic wound cavity model with carbon dioxide: theory and comparison of a gas diffuser with conventional tubes . J Cardiothorac Vasc Anesth . 2003 ; 17 : 329 - 35 .
20. Svenarud P , Persson M , van der Linden J. Intermittent or continuous carbon dioxide insufflation for de-airing of the cardiothoracic wound cavity? An experimental study with a new gas-diffuser . Anesth Analg . 2003 ; 96 : 321 - 7 .
21. Svenarud P , Persson M , van der Linden J. Effect of CO2 insufflation on the number and behavior of air microemboli in open-heart surgery: a randomized clinical trial . Circulation . 2004 ; 109 : 1127 - 32 .
22. Kurz A , Sessler DI , Lenhardt R. Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization . N Engl J Med . 1996 ; 334 : 1209 - 15 .
23. Melling AC , Ali B , Scott EM , Leaper DJ . Effects of preoperative warming on the incidence of wound infection after clean surgery: a randomised controlled trial . Lancet . 2001 ; 358 : 876 - 80 .
24. Rohrer M , Natale A. Effect of hypothermia on the coagulation cascade . Crit Care Med . 1992 ; 20 : 1402 - 5 .
25. Valeri RC , Cassidy G , Khuri S , Feingold H , Ragno G , Altschule MD . Hypothermiainduced reversible platelet dysfunction . Ann Surg . 1987 ; 205 : 175 - 81 .
26. Schmied H , Kurz A , Sessler DI , Kozek S , Reiter A. Mild intraoperative hypothermia increases blood loss and allogeneic transfusion requirements during total hip arthroplasty . Lancet . 1996 ; 347 : 289 - 92 .
27. Frank SM , Fleisher LA , Breslow MJ , Higgins MS , Olson KF , Kelly S , Beattie C. Perioperative maintenance of normothermia reduces the incidence of morbid cardiac events. A randomized clinical trial . JAMA . 1997 ; 277 : 1127 - 34 .
28. National Institute for Health and Clinical Excellence . Perioperative hypothermia (inadvertent): the management of inadvertent peri-operative hypothermia in adults NICE Clinical Guideline 65 London: National Institute for Health and Clinical Excellence . London: National Institute for Health and Clinical Excellence ; 2008 .
29. Persson M , van der Linden J. Can wound desiccation be averted during cardiac surgery? An experimental study . Anesth Analg . 2005 ; 100 : 315 - 20 .
30. Persson M , van der Linden J. Intraoperative CO2 insufflation can decrease the risk of surgical site infection . Med Hypotheses . 2008 ; 71 : 8 - 13 .
31. Persson M , Svenarud P , Flock JI , van der Linden J. Carbon dioxide inhibits the growth rate of Staphylococcus aureus at body temperature . Surg Endosc . 2005 ; 19 : 91 - 4 .