Full text of DOI:10.1007/BF03020531

CARDIOTHORACIC ANESTHESIA, RESPIRATION AND AIRWAY
297
The OxyArm™ – a new minimal contact oxygen
delivery system for mouth or nose breathing
[Le nouveau système de distribution d’oxygène à contact minimal OxyArm™ pour
la respiration buccale ou nasale]
Elizabeth Ling BSc MD MSc FRCPC,* Lee McDonald RN,† Tim R.J. Dinesen PhD MBA,‡ Donald DuVall MD FRCPC§
Purpose: To describe the development and performance of a new
minimal contact oxygen (O2) delivery system for both nasal and oral
breathing, with capnographic capabilities.
Méthode : Les défis du développement et de la conception du proto-
type de l’OxyArm™ (OA) sont présentés. De conception innovatrice, il
utilise un casque de monture semi-rigide et un diffuseur d’O2. L’OA a
été comparé au masque Venturi chez huit adultes en bonne santé
ayant consenti au test. Les fractions d’O2 inspiré ont été mesurées
dans l’hypopharynx grâce à un échantillonnage gazeux continu d’O2
de bas débit à haut débit. Les données moyennes ont été comparées
selon des tests t bilatéraux appariés comportant des niveaux signifi -
catifs à 0,05.
Methods: The development and design challenges of the
OxyArm™ (OA) prototype are described. The innovative design
utilizes a headset with a semi-rigid boom and an O2 diffuser. The
OA was compared to the Venturi mask in eight healthy adults after
informed consent. Inspired O fractions were measured in the
2
hypopharynx using continuous gas sampling at low to high O2 flow
rates. Mean data were compared using two-tailed paired t tests
with significance set at 0.05.
Résultats: La concentration d’O2 inspiré a été plus élevée avec l’OA
pour des débits de 2 (26,3± 2,5 vs 23,3 ± 0,5,P < 0,01) et de 6
L·min–1 (33,5± 3,3 vs 28,8 ± 1,2, P <0,01). Pour un débit de 12
L·min–1, la concentration d’O2 a été plus faible avec l’OA (39,2 ±
6,3vs 46,0 ± 2,7, P < 0,02). Tous les sujets ont jugé les appareils
confortables pendant l’étude de courte durée.
Results: The measured inspired O2 concentration was higher in
the OA at 2 (26.3 ± 2.5 vs 23.3 ± 0.5, P <0.01) and 6 L·min–1
(33.5 ± 3.3 vs 28.8 ± 1.2, P <0.01) flow rates. At 12 L·min–1, the
O2 concentration was less in the OA (39.2 ± 6.3 vs 46.0 ± 2.7, P
<0.02). All subjects found both systems comfortable for the short
duration of the study.
Conclusion : L’OA a distribué des concentrations prévisibles d’O2 de
débits bas à moyens. Ce système est confortable, facile à utiliser, non
obstructif, inodore et sans latex. Les possibilités capnographiques de
cet appareil font qu’il est idéal pour l’anesthésie avec monitorage ou
d’autres installations (soins intensifs) où le monitorage de la respiration
est justifié. L’appareil n’entre pas en contact avec le visage, le rendant
idéal pour la pédiatrie et l’oxygénothérapie de longue durée à domi -
cile. Il faut d’ailleurs poursuivre les essais cliniques dans ce sens.
Conclusions: The OA delivered predictable concentrations of O2
over low to medium flow rates. This system is comfortable, easy to
use, non-obtrusive, odorless, and latex-free. The ability to monitor
capnography makes this device ideal for monitored anesthesia care
or in other settings (intensive care) where monitoring of respiration
is warranted. This device does not contact the face and thus may
be ideal for pediatric patients and those on long-term home O
2
therapy. Further clinical trials in these areas are warranted.
REATMENT of respiratory insufficiency
and postoperative hypoxemia most often
necessitates supplemental oxygen (O ).1,2
O2 is most commonly administered to
patients using a variety of different face masks and
Objectif : Décrire la mise au point et la performance d’un nouveau
système de distribution d’oxygène à contact minimal pour la respira-
tion buccale et nasale, doté de capacités capnographiques.
2
T
From the Departments of Anesthesia, McMaster University,* Hamilton, Southmedic Inc.,† Barrie, Dinesen Research Group,‡ Toronto,
and the Royal Victoria Hospital,§ Barrie, Ontario, Canada.
Address correspondence to: Dr. Elizabeth Ling, Assistant Clinical Professor, McMaster University, Department of Anesthesia, Hamilton
Health Sciences, Hamilton General Site, 237 Barton Street East, Hamilton, Ontario L8L 2X2, Canada. Phone: 905-527-0271, ext.
46698; Fax: 905-577-8023; E-mail: linge@mcmaster.ca
Developmental work was carried out at Southmedic Inc., and clinical work at Dr. S. McDonald’s office in Barrie, Ontario. Supported by a
grant from Southmedic, Inc. of Barrie, Ontario, Canada.
Accepted for publication September 5, 2001.
Revision accepted November 16, 2001.
CAN J ANESTH 2002 / 49: 3 / pp 297–301

298
CANADIAN JOURNAL OF ANESTHESIA
nasal cannulae. The performance of such devices is
somewhat variable, particularly in terms of the
inspired O concentration delivered to the patient.
2
The standard O mask ensures supplementation of
inspired air with2 O for mouth and nose breathing.
2
However, some patients find these masks uncomfort-
able, claustrophobic and their speech is hindered. In
those with borderline hypoventilation, dead space is
increased.3 Routine mouth care, postoperative nausea
and vomiting, and restlessness all lead to mask removal
or displacement, resulting in no O delivery at all.4
Nasal cannula delivery systems li2e in close proximity
with mucous membranes. With dry O flow, this may
lead to local irritation, infection and 2bleeding.3,5 Air
embolism is a rare but described complication.6 Mouth
breathing presents another underestimated problem
with these devices, since O is not inhaled.7,8 This
occurs when talking and whi2le snoring during sleep.9
Nasal cannulae are less likely to be removed than face
masks, although they are frequently dislodged.1 0
The OxyArm™ (OA) (Southmedic Inc., Barrie,
Ontario, Canada), a new minimal contact open O
2
delivery system, has been designed to overcome the
problems of face masks and nasal cannulae.
Capnography monitoring has also been incorporated
into this novel device. We describe the development of
this system and results of preliminary clinical studies.
FIGURE 1 Cross-sectional drawing illustrating shape of the cup
that houses the pin diffuser unique to the OxyArm™.
supply tubing. The design of the boom had to be flex-
ible enough to allow for adjustments, but sufficiently
rigid to stay in place. This was achieved by inserting a
length of stainless steel wire through the lumen of the
boom. This semi-rigid boom was designed to be
either left-or-right-sided and also made adjustable so it
could be shortened or lengthened to fit. The shape of
the boom follows the contour of the cheek and rests
out of the line of sight.
Materials and methods
The concept of an open O delivery system came from
2
anecdotal evidence of open tubes fixed in proximity to
the face.1 1 The headset design originated from adapt-
ing the headsets used in hands-free telecommunica-
tions device. Subsequent modification of the
commercial headset allowed delivery of O through an
open-ended tube. The first prototype 2of the OA
included a tension band that traversed the top of the
The requirements that impacted the design of the
head (headset), O supply and carbon dioxide (CO )
diffuser were the need to incorporate O delivery and
2
2
sampling lines attached to an adjustable boom, and2a
CO2 sampling, without the patient feeling the O sup-
diffuser consisting of both a “cup” and “pin” to deliv-
ply stream. Initial testing revealed that an open O2 hose
2
er a premixed O plume and sample end-tidal CO .
continually blowing in the face was not well tolerated
2
The patented cup and pin consists of a plastic molde2d
‘pin’ centrally located in the diffuser cup (Figure 1).
Several design challenges led to the further devel-
opment of the prototype. The headset went through
various transformations after studying available head-
sets in the marketplace. Modifications were incorpo-
rated so the headset could rest either behind or on top
of the head to allow for various patient positions. The
final headset prototype fits around the back or the top
of the patient’s head, with arms resting on top of the
ears (Figure 2). Positional stability of the headset and
and also not effective at delivering O . O2 needed to be
2
focused and directed at the mouth and nose area.
Various shapes of diffuser prototypes were initially con-
structed using dental molds, and their subsequent air-
flow properties studied. A variety of cup and pin
configurations were tested for both O delivery efficacy
2
and reproducibility of end-tidal CO waveforms.
Utilizing computer simulated wind tunne2l modeling, it
was possible to understand and illustrate the open O
2
flow dynamics from the diffuser. Thus computerized
fluid dynamics were studied on the O plume at the
2
boom became an issue due to the weight of the O
Boundary Layer Wind Tunnel Laboratory at the
2

Ling et al.: A NEW OXYGENDELIVERY SYSTEM
299
FIGURE 4 Oxygen (O ) mass fraction gradient of the
2
OxyArm™ at 4 L·min– 1 with inspiration from computer simulated
wind tunnel testing. The shape of the diffuser (recuperator region)
causes velocity vortexes to develop which concentrate the flow of
O2 from the pin into a flame-like shape towards the face.
FIGURE 2 Final prototype of the OxyArm™ in situ.
of velocities, pressures and O concentrations at every
2
point in space inside and around the device. A stan-
dard -epsilon turbulence model was employed in the
context of steady state simulations. Figure 3 shows the
O2 concentration profile at passive conditions, as cal-
culated through the computer simulation. O diffuses
out from the inlet and pin in the shape of2a mush-
room. During inspiration, the shape of the diffuser
causes complex velocity vortexes to form, which forces
O2 into a flame-like shape towards the face (Figure 4).
This dynamic property of the diffuser is unique to the
OA, concentrating O delivery during inspiration, and
allowing sampling of2CO during expiration.
Initial clinical testing c2ompared inspired O concen-
2
FIGURE 3 Oxygen (O ) mass fraction gradient of the
trations in the hypopharynx with varying O flows
2
2
OxyArm™ at 4 L·min– 1 from computer simulated wind tunnel
between the OA prototype and the Venturi mask
(Airlife™, Percento2 Mask, Allegiance) in healthy adults.
testing. O diffuses from the inlet region out from the pin like a
2
mushroom cloud towards the face in passive conditions (no inspi-
ration). The recuperator region is the concave area of the diffuser
cup.
Oxygen concentrations in the hypopharynx
Eight healthy adults (ages 26–59 yr, ASA I) were stud-
ied after giving informed consent. Subjects were seat-
ed at rest under normal respiratory conditions. No
instructions were given on breathing pattern. Subjects
were shown how to use the OA once and no further
feedback was provided. A sampling catheter was
attached to a pediatric suction catheter (Baxter T64C,
Chicago, Illinois, USA) and inserted nasally by an
anesthesiologist to lie between 8–10 cm in the
hypopharynx. The sampling catheter was connected
to a multigas analyzer and recorder (Datex-Ohmeda
AS/3) calibrated according to manufacturer’s guide-
University of Western Ontario (Dr. H. Hangan,
London, Ontario, Canada).
The three-dimensional geometry of the OA was
reproduced in a computer simulation, based on an
AutoCAD (computer-aided design) generated draw-
ing. The geometry was meshed with a finite volume
grid consisting of many independent cells. The
momentum (Navier-Stokes), continuity and a passive
scalar equation were solved for every cell in the
domain. The result of this process was the calculation
lines. Medical grade pure O was supplied to each sub-
2

300
CANADIAN JOURNAL OF ANESTHESIA
TABLE Oxygen concentrations measured in the hypopharynx using the OxyArm™ and the Venturi mask at incremental flow rates
O2 (L·min– 1)
Device
Mean [O] %
SD
95% CI
P*
2
2
OA
Mask (24%)
OA
Mask (28%)
OA
Mask (31%)
OA
Mask (35%)
OA
Mask (40%)
OA
26.3
23.3
29.4
26.8
33.5
28.8
34.0
32.8
35.9
37.0
39.2
46.0
2.5
0.5
3.2
0.3
3.3
1.2
4.2
1.4
2.7
1.2
6.3
2.7
24.3, 28.3
22.9, 23.7
26.8, 32.0
26.6, 27.0
30.9, 36.1
27.8, 29.8
30.6, 37.4
31.7, 33.9
33.7, 38.1
36.0, 38.0
34.2, 44.2
43.8, 48.2
0.01
0.06
0.01
0.47
0.39
0.02
4
6
8
10
12
Mask (50%)
OA=OxyArm; O =oxygen; Mask=Venturi mask; SD=standard deviation; CI=confidence interval. *two-tailed paired t test.
2
ject at flow rates of 2, 4, 6, 8, 10, and 12 L·min– 1
using the OA and the Venturi mask, in no predeter-
mined order. In addition, the corresponding diluter
jets were attached to the Venturi mask at the appro-
centrations of inspired O to prevent postoperative
hypoxemia. Evaluation of2O delivery systems should
2
be based on adequacy of O2 delivery, simplicity and
ease of use, patient acceptability and compliance, and
cost effectiveness. Traditional O masks and nasal can-
nulae are most commonly used,2but both have disad-
vantages.3–10 The OA was designed primarily as a
priate flow rate to provide O concentrations of 24,
2
28, 31, 35, 40, and 50%. Flow rates were determined
by an O2 regulator needle valve (Western Medica XA-
means of offering some advantages over the O mask
2831). At each stage, the concentration of inspired O
2
2
and nasal cannula, while also incorporating end- tidal
CO2 monitoring capabilities.
was measured in the hypopharynx at least ten times
over 90-sec intervals, after steady state inspired O and
end-tidal CO concentrations had been reached. 2
Mean data2and standard deviations were compared
for the two delivery systems using two-tailed paired t
tests with significance set at 0.05.
Compared to the Venturi mask, the OA delivered
the same or a greater fraction of inspired O concen-
trations at flow rates from 2–10 L·min–1. At2the high-
est flow rate of 12 L·min– 1, the fraction of measured
inspired O in the hypopharynx was less using the OA.
However, 2delivery of O at this flow rate lies outside
Results
2
that of the conventional mask and one would likely
consider using a rebreathing mask. It should be stated
that this study was of a very preliminary nature with a
small number of subjects. Nevertheless, the OA was
well tolerated without any side effects, easy to adjust
and simple to use.
Oxygen concentration in the hypopharynx
The mean O concentrations and 95% confidence inter-
2
val measured in the hypopharynx for each flow rate are
presented in the Table. Both the OA and Venturi mask
delivered acceptable O concentrations for clinical
administration at low to2medium flow rates. The mea-
sured O concentration in the hypopharynx using the
OA was 2higher than with the Venturi mask at flow rates
of two (26.3 2.5 vs 23.3 0.5, P=0.01) and 6 L·min–1
(33.5 3.3 vs 28.8 1.2, P=0.01). At a flow rate of 12
The OA is a novel, minimal contact, open O deliv-
2
ery system and is well suited for either oral or nasal O
2
delivery. The unique headset design with the baffled
cup diffuser offers controlled delivery of variable con-
centrations of O in an unencumbered and comfortable
L·min– 1, the OA delivered lower O concentrations as
2
compared to the Venturi mask (39.2 6.3 vs 46 2.7,
P=0.02). At all other flow rates, there was no difference
way. It is simple2to modulate the O concentration with
2
the OA by adjusting the supply flow rate. To achieve
this with the Venturi mask, matching combinations of
diluter jets and flow rates must be changed.
The minimal contact design allows patients to talk
comfortably, and routine nursing tasks may be accom-
between the two O delivery systems. All subjects found
2
both systems comfortable for the short duration of the
study period. Noise levels from the OA were perceived
to be low and consistent at all flow rates.
plished without disturbing O delivery. Patient anxiety
2
Discussion
and claustrophobia may be reduced due to the lack of
facial contact and unhindered line of sight. The entire
system is odorless and latex-free. The OA received
Postanesthetic care units handle large volumes of
patients each year, the majority requiring higher con-

Ling et al.: A NEW OXYGENDELIVERY SYSTEM
301
tion to postoperative patients. Br J Anaesth 1993; 70:
Food and Drug Administration approval in Canada in
May 2000 and is currently pending a worldwide
patent. It is currently available in 57 countries.
440–2.
11 Kumar RM, Kabra SK, Singh M. Efficacy and accept-
ability of different modes of oxygen administration in
children: implications for a community hospital. J Trop
Pediatr 1997; 43: 47–9.
In further studies, use of the OA for long-term O
2
therapy may prove beneficial. Compliance in these
patients may be improved due to improved comfort,
minimal contact, and the global acceptability of head-
sets giving it a non-medical appearance. Talking, eat-
ing and drinking may be permitted with
uninterrupted O delivery. It may also prove to be an
important altern2ative in the postoperative pediatric
population, as some are intolerant of traditional O
2
delivery systems. The ability of the OA to incorporate
capnography may prove useful in the operating room
during monitored anesthesia care. Future clinical trials
are ongoing and will further define the clinical utility
of the OA in these various settings.
Acknowledgements
The authors wish to thank Dr. James Paul for statisti-
cal assistance and Dr. Sandy McDonald for his support
in this project.
References
1 Stewart BN, Hood CI, Block AJ. Long-term results of
continuous oxygen therapy at sea level. Chest 1975;
68: 486–92.
2 Craig DB. Postoperative recovery of pulmonary func-
tion. Anesth Analg 1981; 60: 46–52.
3 Miller WF. Oxygen therapy, catheter, mask, hood and
tent. Anesthesiology 1962; 23: 445–51.
4 Nolan KM, Baxter MK, Winyard JA, Roulson CJ,
Goldhill DR. Video surveillance of oxygen administra-
tion by mask in postoperative patients. Br J Anaesth
1992; 69: 194–6.
5 Hoffman LA, Dauber JH, Ferson PF, Openbrier DR,
Zullo TG. Patient response to transtracheal oxygen
delivery. Am Rev Respir Dis 1987; 135: 153–6.
6 Merino-Angulo J, Perez de Diego I, Casas JM.
Subcutaneous emphysema as a complication of oxygen
therapy using nasal cannulas (Letter). N Engl J Med
1987; 316: 756.
7 Camner P, Bakke B. Nose or mouth breathing?
Environ Res 1980; 21: 394–8.
8 Canet J, Sanchis J. Performance of a low flow O
2
Venturi mask: diluting effects of the breathing pattern.
Eur J Respir Dis 1984; 65: 68–73.
9 Kauffmann F, Annesi I, Neukirch F, Oryszczyn MP,
Alpérovitch A. The relation between snoring and smok-
ing, body mass index, age, alcohol consumption and res-
piratory symptoms. Euro Respir J 1989; 2: 599–603.
10 Nolan KM, Winyard JA, Goldhill DR. Comparison of
nasal cannulae with face mask for oxygen administra-