Content uploaded by Guillaume Besch
Author content
All content in this area was uploaded by Guillaume Besch on Apr 19, 2023
Content may be subject to copyright.
CORRESPONDENCE
Comparison of the efficacy of high-flow nasal oxygenation and
spontaneous breathing with face mask ventilation during
panendoscopy
Cl
ement Conti
1
, Olivier Mauvais
2
, Emmanuel Samain
1
,
3
, Laurent Tavernier
2
,
S
ebastien Pili Floury
1
,
3
, Guillaume Besch
1
,
3
and David Ferreira
1
,
4
,
*
1
D
epartement d’Anesth
esie R
eanimation Chirurgicale, Centre Hospitalier Universitaire de Besanc¸on, Besanc¸on,
France,
2
D
epartement d’ORL, Centre Hospitalier Universitaire de Besanc¸on, Besanc¸on, France,
3
EA3920, University of
Franche Comt
e, Besanc¸on, France and
4
Laboratoire de Recherches Int
egratives en Neurosciences et Psychologie Cognitive
(LINC), Universit
e de Franche-Comt
e, Besanc¸on, France
*Corresponding author. E-mail: dferreira@chu-besancon.fr
Keywords: endoscopy; face mask ventilation; high-flow nasal oxygenation; hypoxaemia; panendoscopy; spontaneous
breathing
EditordPanendoscopy is a high-risk procedure where
oxygenation and patient safety in the absence of tracheal
intubation is compromised by the depth of anaesthesia
required for the procedure. Currently, there are several
methods to ensure oxygenation during this procedure but
none is universally accepted. Whatever the method chosen,
the risks to the patient remain significant, particularly
during hypoxaemia during episodes of haemoglobin
desaturation. Several studies have shown an increase in safe
apnoea time with high-flow nasal oxygenation (HFNO) in
patients at risk of difficult tracheal intubation.
1e4
The
feasibility of this oxygenation method has been described in
ear, nose, and throat surgery but in small samples. There
was significantly increased apnoea time without
desaturation, enabling surgery to be carried out in good
conditions and without interference by a tracheal tube.
5e10
Since 2017, most panendoscopies in our centre have been
performed with HFNO. Before 2017, panendoscopies were
mostly performed with intermittent face mask ventilation
(FMV). The objective of this study was to show that HFNO
used in 2018 and 2019 during panendoscopies reduced the
rate of intraprocedural hypoxaemia (SpO
2
<90%) compared
with spontaneous breathing with FMV as used in 2015 and
2016.
An original anonymised database was retrospectively
created from our anaesthesia informatics software (AMI
(Assistance M
edicale par Informatique), Aegle Informatique
m
edicale, Cergy-pontoise). No discrepancies between the
extracted and analysed data and the file data were found in 50
randomly selected files. The study was registered in the clin-
ical trials registry (NCT05593718) and approved by the ethics
committee of the hospital of Besanc¸ on (02-130422). All patients
who received panendoscopy with FMV (during the years
2015e16) or HFNO (during the years 2018e19) were included.
The year 2017 was considered to be a necessary washout
period to avoid the learning effect of HFNO. Patients who un-
derwent panendoscopy in combination with another invasive
procedure were excluded. Only the first panendoscopy was
selected for each patient. Total intravenous anaesthesia
(TIVA) was titrated in both groups. Spontaneous ventilation
was preserved in the FMV group when possible,
11
whereas for
HFNO the aim was to maintain immobility whether the patient
was apnoeic or not. For the FMV group, spontaneously
breathing patients were preoxygenated via a face mask with
an inspired fraction of oxygen (FiO
2
) of 100%. After induction of
anaesthesia, the patient was intermittently ventilated with a
face mask in ventilation mode according to the anaesthetist’s
choice. For the HFNO group, patients were preoxygenated via
HFNO with a flow rate of 30e50 L min
1
with an FiO
2
of 100% or
spontaneously breathing FMV with an FiO
2
of 100%. During the
procedure, the flow rate of HFNO was increased to 70 L min
1
.
Patients did not experience jaw thrust or chin lift to maintain a
©2023 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.
For Permissions, please email: permissions@elsevier.com
e1
British Journal of Anaesthesia, xxx (xxx): xxx (xxxx)
patent airway before scope insertion. Panendoscopy started
immediately after induction of anaesthesia. In both groups,
frequency and indication of FMV was left at the discretion of
the clinician.
The primary endpoint was the rate of severe hypoxaemia
during a panendoscopic procedure under general anaesthesia,
defined as SpO
2
<90% for more than 60 s. Secondary endpoints
were the incidence of intraoperative tracheal intubation, dura-
tion of surgery (time in minutes from induction of anaesthesia
to discharge from the operating room including performance of
surgery and the time for reoxygenation with mask ventilation)
and the incidence of complications. We classified complications
into: 1) minor complications (i.e. bradycardia between 45 and 30
beats min
1
, haemodynamic instability with MAP between 65
and 40 mm Hg, hypercapnia with EtCO
2
between 8 and 10.7kPa
[60 and 80 mm Hg] [highest EtCO
2
measured during face mask
recoveries or intubation]) and 2) major complications (i.e.
bradycardia <30 beats min
1
, haemodynamic instability with
MAP<40 mm Hg, hypercapnia with EtCO
2
>10.7 kPa [80 mm Hg],
rescue tracheal intubation, and rescue tracheotomy [performed
to overcome desaturation]).
Patient characteristics are described for each group. Cate-
gorical variables were compared using the
c
2
test or Fisher’s
exact test. Student’s t-test was used to compare normally
distributed continuous variables, and the Kruskal Wallis test
for variables that were not normally distributed. Statistical
significance was set at P<0.05. To control the alpha risk
inflation related to multiple analyses on the same variable,
Bonferroni correction was applied (for SpO
2
, minor and major
complications). Statistical analysis was performed using R
4.0.3 software (R Foundation for Statistical Computing,
Vienna, Austria).
12
There were 251 patients included in the FMV group
(including 211 [84%] with oesophagoscopy), and 215 patients in
the HFNO group (including 154 [72%] with oesophagoscopy).
Each group was clinically comparable according to sex,
American Society of Anesthesiologists (ASA) physical status,
age, and BMI (Table 1). The number of patients with at least
one hypoxaemia episode (SpO
2
<90% for at least 1 min) per
procedure did not differ between the FMV (20%) and HFNO
groups (25%), P¼0.24 (not significant). Other thresholds
(SpO
2
<88%, SpO
2
<80%, and SpO
2
<70%) were analysed, and no
significant difference was observed (Table 1). The duration of
surgery was shorter in the HFNO group (40 [24] min) than in the
FMV group (50 [23] min), P<0.001. Recourse to oxygenation
with face mask (>2 interventions per procedure) was higher in
the FMV group (23%) than in the HFNO group (4%), P<0.001. The
rate of orotracheal intubation was higher in the FMV group
(211 [84%]) than in the HFNO group (75 [35%]), P<0.001. Minor or
major complications (low arterial pressure, bradycardia) were
similar in both groups, as was the rate of rescue tracheal
intubation or rescue tracheotomy.
No difference between the two oxygenation methods was
found in terms of haemoglobin desaturation (whatever the
Table 1 Patient characteristics and main and secondary outcomes. ASA, American Society of Anesthesiologists; BMI, body mass index;
EtCO
2
, end-tidal CO
2
concentration; FMV, face mask ventilation; HFNO, high-flow nasal oxygenation; MAP, mean arterial pressure; NS,
not significant; SpO
2
, pulse oxygen saturation.
FMV group, n(%) HFNO group, n(%) P-value
Panendoscopy, n¼466 (100%) 251 (54) 215 (46)
Oesophagoscopy, n(%) 211 (84) 154 (72) 0.002
Patients characteristics
Sex (male), n(%) 183 (73) 156 (73) NS
Age, yr, mean (
SD
) 63.1 (11.4) 64.8 (11.3) <0.001
BMI, kg m
2
, mean (
SD
) 23.9 (5.4) 24.9 (5.2) <0.001
ASA physical status 1 37 (15) 41 (19) NS
ASA physical status 2 135 (54) 110 (51)
ASA physical status 3 71 (28) 59 (28)
ASA physical status 4 8 (3) 5 (2)
Main outcome
SpO
2
<90%, n(%) 50 (20) 53 (25) NS
Secondary outcomes
Lowest SpO
2
, mean (
SD
) 93.0 (5.5) 91.3 (8.5) <0.001
SpO
2
<88%, n(%) 33 (13) 43 (20) NS
SpO
2
<80%, n(%) 6 (2) 16 (7) NS
SpO
2
<70%, n(%) 2 (1) 6 (3) NS
Recourse to oxygenation with FMV
1 Time per procedure, n(%) 86 (34) 44 (20) <0.001
2 Times per procedure, n(%) 58 (23) 9 (4) <0.001
3 Times per procedure, n(%) 36 (14) 4 (2) <0.001
Orotracheal intubation per procedure, n(%) 211 (84) 75 (35) <0.001
Duration of surgery (min), mean (days) 50.1 (23.1) 40.0 (23.5) <0.001
Minor intraprocedural complications
Hypercapnia (EtCO
2
/8-10.7 kPa), n(%) 80 (32) 39 (25) NS
Low arterial pressure (MAP 65e40 mm Hg/), n(%) 144 (57) 109 (43) NS
Bradycardia (45e30 beats min
1
), n(%) 15 (6) 9 (4) NS
Major intraprocedural complications
Hypercapnia (EtCO
2
>10.7kPa), n(%) 7 (3) 7 (5) NS
Low arterial pressure (MAP<40 mm Hg/), n(%) 0 (0) 1 (0.5) NS
Bradycardia (<30 beats min
1
), n(%) 1 (0.4) 0 (0) NS
Rescue orotracheal intubation, n(%) 3 (1.2) 3 (1.4) NS
Rescue tracheostomy,n(%) 2 (0.8) 1 (0.5) NS
e2
-
Correspondence
chosen threshold) and complications. This study has the
largest sample size to our knowledge. The 90% SpO
2
threshold
was used in most studies assessing haemoglobin desaturation
in ear, nose, and throat surgery.
9,13e15
This value was chosen
as both a warning signal of hypoxaemia requiring action by
the anaesthetic team to reoxygenate the patient, and as the
best marker of failure to use HFNO under general anaesthesia.
The high rate of hypoxaemia in both groups might be attrib-
utable to the tolerance of moderate desaturation in order to
prevent interruption to the surgical procedure each time an
airway intervention was performed by the anaesthetist. There
was no reported case of airway obstruction with instrumen-
tation. Neuromuscular blocking drugs had no impact on
desaturation, as they were only used during tracheal intuba-
tion procedures, and desaturation only occurred when pa-
tients were not intubated. Intubation was systematic in the
FMV group during oesophagoscopy, whereas only half of the
patients were intubated during oesophagoscopy in the HFNO
group. Rescue intubation did not differ between the two
groups, nor did moderate and severe hypercapnia. As the FMV
group was mask ventilated and intubated much more often
than the HFNO group, it seems difficult to draw conclusions on
the equivalence of hypercapnia between the two groups. The
need to deepen depth of anaesthesia was frequent, resulting in
minimal or moderate arterial hypotension in a large propor-
tion of patients. The fact that HFNO yields shorter surgical
times, as found in a study by Nekhendzy and colleagues
6
with
a small sample size, is probably related to fewer facemask
recoveries and fewer surgeon repositionings, which might
lead to improved surgical comfort and diagnostic perfor-
mance, warranting prospective evaluation. An assessment of
the optimal SpO
2
threshold requiring initiation of reoxygena-
tion procedures when using HFNO during panendoscopy is
essential to limit desaturation while optimising procedural
comfort.
Declaration of interest
The authors declare that they have no conflicts of interest.
Acknowledgements
Pascale Franc¸ois and M
elanie Claveau collected data. The au-
thors would like to thank Fiona Ecarnot (Department of Car-
diology, University Hospital of Besancon, and EA 3920,
University of Franche-Comte, Besancon, France) for her
assistance in preparing the manuscript.
References
1. Patel A, Nouraei SAR. Transnasal Humidified Rapid-
Insufflation Ventilatory Exchange (THRIVE): a physiolog-
ical method of increasing apnoea time in patients with
difficult airways. Anaesthesia 2015; 70: 323e9
2. Booth AWG, Vidhani K. Spontaneous ventilation using
target-controlled propofol infusion for microlaryngoscopy
in adults: a retrospective audit. Anaesth Intensive Care 2016;
44: 285e93
3. Booth AWG, Vidhani K, Lee PK, Thomsett C-M. SponTa-
neous Respiration using IntraVEnous anaesthesia and Hi-
flow nasal oxygen (STRIVE Hi) maintains oxygenation and
airway patency during management of the obstructed
airway: an observational study. Br J Anaesth 2017; 118:
444e51
4. Guy L, Christensen R, Dodd B, et al. The effect of trans-
nasal humidified rapid-insufflation ventilator exchange
(THRIVE) versus nasal prongs on safe apnoea time in
paralysed obese patients: a randomised controlled trial. Br
J Anaesth 2022; 128: 375e81
5. Lyons C, Callaghan M. Apnoeic oxygenation with high-
flow nasal oxygen for laryngeal surgery: a case series.
Anaesthesia 2017; 72: 1379e87
6. Nekhendzy V, Saxena A, Mittal B, et al. The safety and
efficacy of transnasal humidified rapid-insufflation
ventilatory exchange for laryngologic Surgery. Laryngo-
scope 2020; 130: E874e81
7. Waters E, Kellner M, Milligan P, Adamson RM, Nixon IJ,
McNarry AF. The use of Transnasal Humidified Rapid-
Insufflation Ventilatory Exchange (THRIVE) in one hun-
dred and five upper airway endoscopies. A case series. Clin
Otolaryngol 2019; 44: 1115e9
8. Maupeu L, Raguin T, Hengen M, Diemunsch P, Schultz P.
Indications of transnasal humidified rapid-insufflation
ventilatory exchange (THRIVE) in laryngoscopy, a pro-
spective study of 19 cases. Clin Otolaryngol 2019; 44: 182e6
9. Yang S-H, Wu C-Y, Tseng W-H, et al. Nonintubated lar-
yngomicrosurgery with transnasal humidified rapid-
insufflation ventilatory exchange: a case series. J Formos
Med Assoc 2019; 118: 1138e43
10. Huang L, Dharmawardana N, Badenoch A, Ooi EH.
A review of the use of transnasal humidified rapid insuf-
flation ventilatory exchange for patients undergoing sur-
gery in the shared airway setting. J Anesth 2020; 34: 134e43
11. Besch G, Chopard-Guillemin A, Monnet E, et al. Propofol-
remifentanil anesthesia for upper airway endoscopy in
spontaneous breathing patients: the ENDOTANIL ran-
domized trial. Minerva Anestesiol 2016; 82:11
12. R Core Team. R: a language and environment for statistical
computing. Vienna, Austria: R Foundation for Statistical
Computing; 2014. https://www.r-project.org/. [Accessed
20 October 2022] (20/10/2022)
13. Gustafsson I-M, A
˚Lodenius, Tunelli J, Ullman J, Jonsson
Fagerlund M. Apnoeic oxygenation in adults under gen-
eral anaesthesia using Transnasal Humidified Rapid-
Insufflation Ventilatory Exchange (THRIVE) ea physio-
logical study. Br J Anaesth 2017; 118: 610e7
14. Lau J, Loizou P, Riffat F, Stokan M, Palme CE. The use of
THRIVE in otolaryngology: our experiences in two
Australian tertiary facilities. Aust J Otolaryngol 2019; 2: 22.
22
15. Rajan S, Joseph N, Tosh P, Kadapamannil D, Paul J,
Kumar L. Effectiveness of transnasal humidified rapid-
insufflation ventilatory exchange versus traditional pre-
oxygenation followed by apnoeic oxygenation in delaying
desaturation during apnoea: a preliminary study. Indian J
Anaesth 2018; 62: 202
doi: 10.1016/j.bja.2023.03.015
Correspondence
-
e3
