British Journal of Anaesthesia 1993; 7 1 : 453-457

APPARATUS

ACCURACY OF 94 ANAESTHETIC AGENT VAPORIZERS IN

CLINICAL USEf

J. NIELSEN, F. M. PEDERSEN, F. KNUDSEN, M. B. JENSEN AND M. IBSEN

circumstances resembling clinical use. In addition,

SUMMARY
Using the Bruel & Kjaer Anaesthetic Gas Monitor
type 1304, we have monitored the output of 94
anaesthetic agent vaporizers (Fluotec 3: 58, En-
fluratec 3: 24, Isotec 3: 12), in seven departments
of anaesthesia, at different dial settings and flow
rates. The range of output, for one type of vaporizer
and dial setting (flow: 6 litre m/n~') was largest
with the Fluotec 3 (0.85-1.55% when dial set to
1%) and smallest with the Isotec 3 (0.85-1.15%
when dial set to 1 %). In determining the number of
vaporizers with unacceptable inaccuracy, we ap-
plied acceptance limits of ±15% relative on each
vaporizer and each dial setting. Using a flow of
oxygen 6 litre min~' 17 %of Fluotec 3,8% ofIsotec 3
and 71% of Enfluratec 3 vaporizers had outputs
outside those limits. Even when some specific
conditions (vaporizers giving output beyond the
limits at any two or more dial settings; output
beyond the limits in the clinically relevant range
(0.5-2%)) were added, a substantial number of
vaporizers did not perform within the limits. We
found a significantly greater accuracy of the vapor-
izers after 3-monthly calibration checks (? < 0.05)
compared with vaporizers undergoing service and
calibration only annually. Using a questionnaire, we
found that fewer than 30% of the anaesthetists
using the vaporizers would accept aberrance be-
yond ±10% relative of the dial setting. (Br. J.
Anaesth. 1993; 7 1 : 453-457)
KEY WORDS
Equipment: vaporizers, safety.
As the concern for patient safety during anaesthesia
is increasing, so is interest in the performance of
equipment used during anaesthesia. Performance of
anaesthetic vaporizers has been examined for many
years [1-17], but the majority of these investigations
are case reports or are based upon a small number of
vaporizers. To our knowledge, only one report [17]
described the performance of a large number of
vaporizers in clinical use; this was based on main-
tenance records obtained during a 4-year period.
The aim of this study was to investigate the
accuracy of a large number of vaporizers in clinical
use, under controlled conditions, at several flow rates
and dial settings throughout the range, under
using a questionnaire, we sought the views of
anaesthetists on vaporizer performance.
MATERIALS AND METHODS
We approached the departments of anaesthesia at
seven hospitals in the Copenhagen area and asked
them to participate in this study. All agreed, thus
providing three university clinics, one county hos-
pital and three smaller urban hospitals. Each de-
partment stated the type, age and number of
vaporizers in daily use and the maintenance policy.
The Bruel & Kjaer Anaesthetic Gas Monitor Type
1304 (Bruel & Kjaer, DK-2850 Naerum, Denmark)
was used to measure vaporizer output. This monitor
has been investigated in our own laboratory and
results showed excellent linearity and accuracy [18].
The resolution of the monitor is +0.05 vol%, and
the accuracy in determining a given concentration
was found to be ^ ± 5 % relative (rel.).
This monitor was warmed up for 30 min before
calibration, which was performed before each test
sequence at the respective department, using a
calibration gas mixture (Alphagaz, Cambridge,
U.K.) with certified purity (anaesthetic agent
0.4 + 0.008 % rel.). The same calibration gas canister
was used throughout the investigation. In addition,
despite the fact that all vaporizers had keyed filling
systems, the type of anaesthetic vapour delivered
from each vaporizer was determined using the agent-
specific anaesthetic gas monitor Nellcor N-2500
(Nellcor, U.S.A.).
Testing was carried out in an operating room of
each department, the ambient air temperature being
approximately 22 °C. All vaporizers were stored at
the same temperature. If any vaporizer needed
refilling before the test, the liquid anaesthetic agent
used had the same temperature as the ambient air.
The level of liquid in the vaporizers was between the
JACOB NIELSEN*, M.D.; FINN MOLLER PEDERSEN, M.D.; FREDDY
KNUDSEN, M.D.; MAIKEN BRITT JENSEN, M.D.; MERETE IBSEN,
M.D. ; Department of Anaesthesia and Intensive Care Unit, Herlev
Hospital, University of Copenhagen, Medical School, DK-2730
Herlev, Denmark. Accepted for Publication: March 11, 1993.
•Present address, for correspondence: University of Copen-
hagen, Department of Anaesthesia and Intensive Care Medicine,
KAS Gentofte, DK-2900 Hellerup, Denmark.
fPart of this study was presented orally at the 1991 ASA
meeting, San Francisco, U.S.A.
454
TABLE I. Part of questionnaire distributed to each anaesthetist employed at the departments
1. What magnitude of abberance from the vaporizer
dial setting do you find acceptable?
2. Are you using Isotec or Fluotec vaporizers (Tec 3)
with the dial set between the fixed marks in the
dial setting range 0-1 % ?
3. Are you using Enfluratec vaporizers with (Tec 3)
the dial set at 0.2/0.4/0.6/0.8%?
4. If you have answered seldom/never to question
2, is the reason that you
5. If you have answered seldom/never to question 3,
is the reason that you
"half" and "full" marks. All were transported
upright and with the dial in the "off" position. Each
vaporizer was mounted on an anaesthetic machine. If
this was done using the Selectatec system, we
ensured that the manifold port valve " O " rings were
intact, and the lever-lock was in the "lock" position.
Similarly, for every vaporizer we checked that the
drain port was fully closed.
At the vaporizer outlet, an ISO-cone with a 50-cm
p.v.c. hose was mounted. The sampling tube of the
monitor was connected via a Y-piece at the end of the
hose, thus providing a homogenous gas mixture and
a free outlet, avoiding pressure variations. Before
each test, the vaporizer was flushed with the dial set
at 4 % and an oxygen flow of 8 litre min"1 for a few
minutes, in order to remove any nitrous oxide
dissolved in the anaesthetic liquid, as this may have
a major influence on output [19]. In addition, we
monitored the eventual output of nitrous oxide
during the test sequence. The vaporizers were tested
at three different flow rates of oxygen: 0.5, 2 and
6 litre min"1. Fluotec and Isotec vaporizers (Tec 3,
Ohmeda Ltd, U.K.) had the dials set at 0.5, 1, 2, 3.5
and 5%. As these vaporizers are used commonly
with the dial set between the fixed marks, they were
tested also at settings of "0.25%" and "0.75%".
Transparent plastic scales (log scale) with marks
representing 0, "0.25" and "0.75" (log centre)
settings for each type of vaporizer were produced.
The 0 marks of the plastic scale and the dial were
aligned and the dial was then set at "0.25" or
"0.75", thus testing each vaporizer under uniform
conditions. Enfluratec vaporizers (Tec 3, Ohmeda
Ltd, U.K.) had the dials set at 0.2, 0.4, 0.6, 0.8, 1, 2,
3.5 and 5%.
The output was noted at each dial setting and flow
rate, and readings were made continuously until
stability was reached, usually after 30 s. The date of
the last service check was also noted. After the test
sequence at each department, the Briiel & Kjasr
Anaesthetic Gas Monitor Type 1304 was checked
against the calibration gas in order to evaluate any
drift. Before our investigation of the vaporizers, we
distributed a questionnaire to each anaesthetist
employed at the departments (table I).
Maintenance policy
According to guidelines issued by the Danish
National Board of Health, all vaporizers should be
BRITISH JOURNAL OF ANAESTHESIA
0%rel. 0-5%rel. 5-10% rel. 10-20% rel. >20%rel.
• • D • •
• Dial setting irrelevant, dosage according
to clinical parameters.
D Often • Seldom/never
D Often • Seldom/never
D do not trust the vaporizer at these dial settings
• do not find these dial settings clinically relevant
D do not use this type of vaporizer
• do not trust the vaporizer at these dial settings
• do not find these dial settings clinically relevant
• do not use this type of vaporizer
sent annually to the manufacturer for a complete
service check; this includes disassembling, cleaning
and calibration. All departments stated that these
guidelines had been followed. Apart from this, one
department had implemented 3-monthly calibration
checks using an anaesthetic agent monitor (Datex
Capnomac, Datex, Finland). If the vaporizer output
exceeded some specific limits (+ 20 % rel. in the dial
setting range 0 - 1 % , +15% rel. in the dial setting
range 2—4%), the vaporizer was sent for recali-
bration. Oxygen was used as carrier gas at a flow rate
of 5 litre min"1.
Criteria for acceptance
Because of lack of explicit data (in the Operators'
Manuals) regarding the actual calibration limits, we
decided that the output was acceptable if it was
within + 10% rel. of the dial setting. Added to this
was the error in measurement, caused by the gas
monitor and calibration gas ( ± 5 % rel.). As a
consequence, our acceptance limits were +15% rel.
(or +0.1 vol%, greatest value chosen). As the
Fluotec 3 and Isotec 3 vaporizers do not have fixed
marks at dial settings "0.25" and "0.75", the
outputs at these dial settings are not included in the
part of the investigation dealing with acceptability.
Consequently, the Fluotec and Isotec vaporizers
were tested for acceptability at five dial settings and
the Enfluratecs at eight dial settings; all vaporizers at
three flow rates.
Statistical analysis
The accuracy of the Tec vaporizers undergoing
quarterly calibration checks and of those submitted
only to annual service and calibration was compared
using chi-square analysis, as was the discrepancy
between answers in the questionnaire. P < 0.05 was
regarded as statistically significant.
RESULTS
A total of 114 vaporizers was reported to be in
clinical use at the seven departments. One hundred
and two vaporizers (90 %) were tested; the rest were
in use for anaesthetic procedures at the time of our
visit. We decided to omit the data for two Penlon
vaporizers and seven Drager Vapor 19.n vaporizers
because of small numbers. Data for the equipment
tested are given in table II. The date for next
ACCURACY OF ANAESTHETIC AGENT VAPORIZERS 455

TABLE II. Type of vaporizers and time from last calibration (number
100 -
or mean (range))

Number Time from last

Vaporizer studied calibration (months)
Fluotec 3 (5%) 58 7.0(2-16)
Isotec3(5%) 12 9.0(1-12)
Enfluratec 3 (5 %) 24 8.6(1-12)

Fluotec 3 (n = 58)

1
60-;
40-: [71 Fluotec 3 Isotec 3 Enfluratec 3

—
c 20^ (n=58) (n=12) (n = 24)
o
o • v, If FIG. 2. Vaporizers giving outputs beyond the acceptance limits
u
•
(±15% rel.) at one or more combinations of the five or eight dial
-20-: settings (0.2-5%) and the three flow rates of oxygen:
0.5 litre min"1 ( • ) ; 2 litre mhr 1 ( • ) ; 6 litre min"1 (0).
-40 ~ •
-60^ 1 1 1 1 1 1 1
"d.25" 0.5 "0.75" 1 2 3.5 100 - ,
i '• Dial setting (vol %)
80 -
Isotec 3 (n E

c 60 -
T3
o
>
ID

c
o
o
s Q.

Fluotec 3 Isotec 3 Enfluratec 3

"0.25 0.5 "0.75" 1 2 (n = 58) (n=12) (n = 24)
Dial setting (vol %)
FIG. 3. Vaporizers giving outputs beyond the acceptance limits
Enfluratec 3 (n = 24) (±15% rel.) under various conditions, but all at 6 litre min"1
flow. • = Beyond limits at any one or more of the five or eight
dial settings; • = beyond limits at any two or more of the five or
eight dial settings; 0 = beyond the limits at the dial settings
0.5 %, 1 % and 2 % (Fluotec 3 and Isotec 3) or 0.6 %, 1 % and 2 %
c (Enfluratec 3).
o

Isotec 3 (dial set to 1 %: output 0.85-1.15 %). Using

0.4 0.6 0.8 1 3.5
Dial setting (vol %)
FIG. 1. Relative variation from dial setting in Tec 3 vaporizers.
Each column shows range of relative variation ( # = mean) at
fixed vaporizer settings. Carrier gas: oxygen 6 litre min"1.
calibration was overdue in three vaporizers (by 1, 2
and 4 months). The mean age of the Tec vaporizers
(time in use) was approximately 10 yr.
The department using 3-monthly calibration
checks—performed by their own medical physics
staff—stated that two vaporizers (of a total of 33 Tec
3) had been sent for recalibration during the past
year. Figure 1 illustrates the relative range and mean
deviations of the vaporizer output from the dial
settings at a flow rate of 6 litre min"1. The range of
outputs, for one type of vaporizer and one dial
setting, was largest with the Fluotec 3 (e.g. dial set to
1%: output 0.85-1.55%) and smallest with the
the Fluotec 3 and Isotec 3 vaporizers with the dial set
at "0.25"%, the mean output was substantially less
than the dial setting, and the relative range of
outputs was wider, as in the fixed settings. Only one
of the Fluotec 3 vaporizers gave no output at the dial
setting "0.25"%; thus a "cut off" phenomenon
was not demonstrated generally. When the dial was
set at "0.75"%, the vaporizer performance was of
the same relative proportion as in the fixed settings.
A small flow dependency was seen with the Tec 3
vaporizers, giving excessive output at the greatest
dial setting and slowest flow rate; this diminished as
the flow increased. The output was not influenced by
changes in flow rate (0.5-6.0 litre min"1) at dial
settings most relevant clinically (0.5-2%).
Figure 2 illustrates the number of vaporizers with
an unacceptable inaccuracy at any one of the 15 or 24
combinations of flow and vaporizer setting (Fluotec
and Isotec, three flow rates and five dial settings;
Enfluratec, three flow rates and eight dial settings).
In order to derive the percentage of vaporizers with
an unacceptable output, especially at clinically
456 BRITISH JOURNAL OF ANAESTHESIA
TABLE III. Number of vaporizers (Tec 3, n = 94) with outputs After the test sequence at each department, the
beyond the acceptance limits { + 20% rel. at dial settings 0-1 %; monitor was again supplied with the calibration gas;
+ 15% rel. at dial settings 2-4%) at a department using 3-monthly
calibration checks, compared with vaporizers undergoing only annual
drift was =$2.5%.
calibration. Carrier gas: oxygen 6 litre min~'. P < 0.05 (x2 test) Each of the 307 anaesthetists at the seven depart-
between groups ments received a questionnaire (140 anaesthetists,
167 certified registered nurse anaesthetists (CRNA))
Unacceptable Acceptable
output output
(table I). One hundred and ninety-four (63%) were
returned (anaesthetists 55 %, CRNA 70 %). Figure 4
Vaporizers submitted to shows the degree of vaporizer inaccuracy that was
3-monthly calibration 0 30 considered acceptable by the 194 anaesthetists.
checks (n = 30) Fewer than 30% would accept variation beyond
Vaporizers undergoing
service and calibration 10 54 ± 10 % rel. of the dial setting.
once a year (n = 64) In the dial setting range 0-1 %, the anaesthetists
used those vaporizers with several, fixed settings
(Enfluratec) significantly (P < 0.01) more often
45 -, (58%) than the vaporizers (Fluotec & Isotec) with
40 -
only one fixed setting (39%). Among those anaes-
thetists who seldom/never used the vaporizers in the
35 - low range of dial settings, significantly more (P <
o 0.01) stated the reason to be lack of confidence in the
TO 30 - vaporizers without fixed set points compared with
!t
to 25 - those vaporizers possessing them.
20 -
DISCUSSION
15 -
In general, we have found that the vaporizers had an
CD
Q. 10 - unacceptable inaccuracy and the outputs varied
greatly from one vaporizer to another. The accuracy
5 -
of vaporizers in clinical use at a department using 3-
EL monthly calibration checks was significantly greater
0% 0-5% 5-10% 10-20% >20% Clinical compared with vaporizers of the same type in use at
Acceptable variation (% rel.) departments using only annual service and cali-
FIG. 4. Variation in vaporizer output from the dial setting (% rel.)
bration.
considered acceptable by 78 anaesthetists ( 0 ) and 116 Certified The variation in output within one type of
Registered Nurse Anaesthetists (D)- Clinical = using clinical vaporizer was substantial (fig. 1), with relative
criteria. variations in the range — 50 % to + 70 %, at the fixed
settings. This was most pronounced using the
relevant settings, we then applied the acceptance Fluotec 3 compared with the Enfluratec 3 and Isotec
limits (+ 15 % rel.) to two specific conditions (fig. 3): 3, but this difference was caused by one Fluotec 3
vaporizers giving output beyond the limits at any with large errors. This lack of predictability of
two or more of the five or eight dial settings, and output is not clinically acceptable, as it may mask
output beyond the limits in the clinically relevant patient responses.
range of 0.5-2 % (0.5 or 0.6 %, 1 % 2 %; three data Another consequence of variability in output is
points for each type of vaporizer). The flow rate used that it places some vaporizers beyond the limits of
was 6 litre min"1. For comparison, the percentage of acceptance of accuracy. When choosing these ac-
vaporizers failing the test at any one or more of the ceptance limits, we have used primarily criteria of
five or eight dial settings throughout the range at the ± 10 % rel., adding to this the variability of the
flow rate of 6.0 litre min"1 is shown. measurement technique setup. These limits are the
There was no difference between the percentage of same as those used by other investigators [17]. The
vaporizers with unacceptable or acceptable output resulting limits are in overall accordance with the
regarding the mean time from last service (Tec 3 specifications stated by the manufacturer of Tec 3
vaporizers (« = 94): mean time from last service 7.5 vaporizers [20]. Applying the acceptance limits of
months for both unacceptable and acceptable per- + 15% rel. on measured output, a large proportion
formers). of the vaporizers performed beyond these limits.
Applying less restrictive acceptance limits This was most pronounced at low flows, but even at
( ± 2 0% rel. in the dial setting range 0 - 1 % , 6 litre min"1, 17 % of Fluotec 3, 8 % of Isotec 3 and
+ 15 % rel. in the dial setting range 2-4%) because 71 % of Enfluratec 3 gave outputs outside these
of the lesser resolution of the monitor used by the limits. As this also includes vaporizer settings
department performing its own calibration checks unlikely to be used in ordinary clinical procedures,
every 3 months, among all the Tec 3 vaporizers we added two specific conditions to the acceptance
tested (6 litre min"1), we found a significantly greater limits and applied these to the vaporizer output at
accuracy in the vaporizers submitted to the quarterly 6 litre min"1—a flow close to that at which the
calibration checks (/> < 0.05), than in vaporizers vaporizers were calibrated. The first condition was
undergoing only annual service and recalibration excessive output at any two or more of the dial
(table III). settings, thus including the vaporizers with the more
ACCURACY OF ANAESTHETIC AGENT VAPORIZERS
substantial variation: 10% of the Fluotec 3 were in
this group, compared with none of the Isotec 3 and
15% of the Enfluratec 3. The second condition was
output beyond the acceptance limits in the clinically
relevant dial setting range (0.5 or 0.6%, 1%, 2 % ;
three data points for each vaporizer); on this basis,
15 % of the Fluotec 3, 10% of the Isotec 3 and 35 %
of the Enfluratec 3 delivered excessive output.
One reason for the greater accuracy of the Isotec 3
vaporizers could be that, because of the cost of
isoflurane, these have been used less frequently.
Only a relatively small number (12) of Isotec 3
vaporizers was investigated. As the precise measure-
ment of anaesthetic agent concentration in the gas
mixture delivered to the patient during anaesthesia is
still not routine, the accuracy of anaesthetic agent
vaporizers is important. While cardiovascular de-
pression caused by excessive vaporizer output may
be recognized and corrected, it is better avoided
altogether.
Carter and colleagues [17] examined 631 main-
tenance records of vaporizers in the West of
Scotland; among these were 236 Fluotec 3 and 214
Enfluratec 3. All vaporizers were given 6-monthly
calibration checks; apart from this there was no
uniform policy for maintenance. Oxygen was used as
carrier gas at 5-7 litre min"1 and the monitors were
Riken interferometers. The acceptance limits were
+ 10%rel., and to this was added the error of
measurement. They demonstrated calibration errors
in 9 % of the Fluotec 3 and in 4 % of the Enfluratec
3. Isotec 3 vaporizers were not examined. More
vaporizers failed the acceptance limits in our study,
although the acceptance limits were identical and the
flow rates similar. The apparently smaller variation
in the output found by Carter's group could be a
result of the procedure of 6-monthly calibration
checks and recalibration of the vaporizers failing the
acceptance limits, compared with the general policy
of maintenance revealed in our study, namely annual
service and recalibration. In addition, the vaporizers
we investigated had been in use for 10 years, as
opposed to 4 years in use for the vaporizers
investigated by Carter and colleagues.
At the one department in our study which had
implemented 3-monthly calibration checks, the ac-
curacy of the vaporizers was significantly better.
Because of the smaller resolution of the monitor used
for measurement at this department, compared with
that used in the present study, the limits were
somewhat less restricted in the dial setting range
0-1%, and measurements at 5% were not per-
formed.
The mean output of the Fluotec 3 and the Isotec
3 vaporizers with the dial setting between the fixed
marks of 0%, 0.5% and 1%, was somewhat less
than that expected, especially at "0.25%". This
could be attributable to the fact that the vaporizers
were not calibrated at these points. We could not
demonstrate the "cut-off" phenomenon found by
Latto [5] using Fluotec 3 vaporizers with the dial
setting positioned midway between "OFF" and 0.5 %,
although there might be a difference in the exact
position of the dial setting. As the survey among the
457
anaesthetists at the seven departments showed that
the Fluotec 3 and Isotec 3 vaporizers were used quite
often (39 %) with the dial setting positioned between
the fixed marks in the low concentration range, we
recommend calibration and marking of the "0.25"%
and "0.75"% set points.
From our results we believe it is reasonable to
presume that all vaporizers could achieve acceptable
accuracy by implementing regular field calibration
checks, and sending for recalibration only those
vaporizers with output beyond certain limits. How-
ever, the regular service check of the vaporizers with
acceptable output should not be omitted, but the
interval extended.
ACKNOWLEDGEMENTS
We are grateful to the Briiel & Kjsr Company (DK-2850 Naerum,
Denmark) and the Nellcor Company (California, U.S.A.) for
providing the gas monitors for the study.
REFERENCES
1. Morgan M, Lumley J. Reliability of halothane vaporizers.
Anaesthesia 1968; 23: 440-445.
2. Paterson GM, Hulands GH, Nunn JF. Evaluation of a new
halothane vaporizer: The Cyprane Fluotec mark 3. British
Journal of Anaesthesia 1969; 41: 109-119.
3. Noble WH. Accuracy of halothane vaporizers in clinical use.
Canadian Anaesthetists Society Journal 1970; 17: 135-144.
4. Murray WJ, Fleming P. Fluotec mark 2 halothane output:
Nonlinearity from "off" to 0.5 per cent dial settings.
Anesthesiology 1972; 36: 180-181.
5. Latto IP. Administration of halothane in the 0-0.5%
concentration range with the Fluotec mark 2 and mark 3
vaporizers. British Journal of Anaesthesia 1973; 45: 563-569.
6. Dorsch JA. Understanding Anesthetic Equipment. Baltimore:
Williams & Wilkins Co., 1975; 92-122.
7. Inadvertent contamination of anaesthetic circuits with halo-
thane. British Journal of Anaesthesia 1977; 49: 745-754.
8. Cook TL, Eger El II, Behl RS. Is your vaporizer off?
Anesthesia and Analgesia 1977; 56: 793-800.
9. Anonymous. Health Devices 1980; 10: 31-51.
10. Lin CY. Assessment of vaporizer performance in low-flow
and closed-circuit anesthesia. Anesthesia and Analgesia 1980;
59: 359-366.
11. Miller JM, Cascorbi HF. Yet another vaporizer hazard.
Anesthesia and Analgesia 1980; 59: 805.
12. Novae GD, Ursillo RC. Malfunctioning halothane vaporizer.
Anesthesia and Analgesia 1981; 60: 121.
13. Steffey EP, Woliner M, Howland D. Evaluation of an
isoflurane vaporizer. The Cyprane Fortec. Anesthesia and
Analgesia 1982; 61: 457^64.
14. Carter JA, McAtteer P. A serious hazard associated with the
Fluotec mark 4 vaporizer. Anaesthesia 1984; 39: 1257-1258.
15. Reike H, Raffouf EM, Zuchner K, Hauschke D, Sonntag H.
Genauigkeit von Halothanverdampfern in Abhangigkeit von
Temperatur, Tragergaszusammensetzung und Gasflussrate.
(Abstract in English.) Anaesthesist 1985; 34: 522-527.
16. Duncan JA. Equipment malfunction: possible hazard.
Anaesthesia 1986; 41: 1271.
17. Carter KB, Gray WM, Railton R, Richardson W. Long-term
performance of Tec vaporizers. Anaesthesia 1988; 43:
1042-1046.
18. Nielsen J, Kann T, Moller JT. Evaluation of three multigas
anesthetic monitors: The Briiel & Kjaer Anesthetic Gas
Monitor 1304, the Datex Capnomac Ultima and the Nellcor
N-2500. Journal of Clinical Monitoring 1993; 9: 91-98.
19. Scheller MS, Drummond JC. Solubility of N2O in volatile
anesthetics contributes to vaporizer aberrancy when changing
carrier gases. Anesthesia and Analgesia 1986; 65: 88—90.
20. Operators Manual. Tec 3. Continuous flow vaporizer.
Ohmeda, England.