Evaluation of Two Methods of Endotracheal

Tube Selection in Dogs

This study evaluated two methods of endotracheal tube selection using 28 fresh canine carcasses of various ages, weights, and genders. The two selection methods were 1) nasal septal width pairing with outer diameter of an endotracheal tube, and 2) digital palpation of the
tracheal outer diameter to determine the endotracheal tube size. All dogs were dolichocephalic breeds. Results of this study showed that the canine nasal septal width method of endotracheal tube selection was correlated with the size of the tracheal internal (r=0.72) and outer
(r=0.73) diameters. However, evidence shows that the digital palpation method is slightly
more effective than the nasal width method in selecting the best-fitting endotracheal tube.
The percentage of the best-fit tube selection for the nasal septal width method was 21%,
while the digital palpation method was 46%. With these two methods, selecting an endotracheal tube that is too small is possible, especially when the tube internal diameter is 7 mm.
J Am Anim Hosp Assoc 2008;44:236-242.

Jim Lish, PhD

Jeff C. H. Ko, DVM, MS,
Diplomate ACVA
Mark E. Payton, PhD

From the Center for Veterinary

Health Sciences (Lish) and the
Department of Statistics (Payton),
College of Arts and Sciences,
Oklahoma State University,
Stillwater, Oklahoma 74078
and the Department of
Veterinary Clinical Sciences (Ko),
School of Veterinary Medicine,
Purdue University,
West Lafayette, Indiana 47908.
Address all correspondence to Dr. Ko.
236

Selection of a correctly sized endotracheal tube is important in practicing

anesthesia. Too large of an oral-tracheal tube may cause laryngotracheal
trauma during the intubation attempt or a complete failure to intubate.1
An inappropriately small tube may result in gas leakage and subsequent
pollution of the operation room. If a smaller cuffed endotracheal tube is
used, the cuff may require overinflation to maintain a seal, exerting high
pressure on the tracheal wall and thereby increasing the risk of tracheal
necrosis. Inappropriately small endotracheal tubes may also increase the
work of breathing because of increased airway resistance. For each 1-mm
decrease in the internal diameter (ID) of the endotracheal tube, the work
of breathing has been demonstrated to increase by 34% to 154%, and the
airway resistance increases by 25% to 100%.1,2 Therefore, when selecting an endotracheal tube for a dog, one must select as large of an ID as
possible in order to minimize the work of breathing and reduce airway
resistance during inhalant anesthesia.
Size selection of endotracheal tubes for oral-tracheal intubation in
dogs has not been standardized due to variations in age, breeds, and body
weights of animals. In addition to selecting the estimated optimum-sized
endotracheal tube, most practitioners also select two additional endotracheal tubes. One is a size larger and one is a size smaller than the estimated appropriate size. These additional tubes are kept at hand to ensure
a close fit.
In practice, several methods have been used to select endotracheal
tubes. The first commonly used method is based on the dogs body
weight, with the assumption that the dogs tracheal diameter is directly
proportional to its body mass. The accuracy of this method is highly
dependent on the experience of the anesthetist. The second method used
is based on direct digital palpation of the dogs tracheal diameter just
above the thoracic inlet, with the assumption that the tracheas outer
diameter (OD) will approximate the appropriate size of endotracheal tube
to be used. The third method of selecting endotracheal tubes is to hold an

Introduction

JOURNAL of the American Animal Hospital Association

September/October 2008, Vol. 44

endotracheal tube up to the nasal septum (at the narrowest

point between two nostrils) of the dog and match the width
of the nasal septum with the endotracheal tube OD. A
review of current literature reveals no published information supporting the assumption of a direct anatomical correlation between the nasal septal width and the tracheal ID
and OD. Furthermore, no information has been published
comparing the accuracies of nasal septal width versus direct
palpation methods for endotracheal tube size selection.
In this study, the authors hypothesized that 1) the correlation is low between nasal septal width of the dog and its
tracheal ID or OD, and 2) endotracheal tube size selection is
more accurate when using digital palpation of the dogs tracheal OD than when using the dogs nasal septal width. The
objectives of this study, therefore, were to 1) evaluate the
correlations between the nasal septal width, tracheal ID and
OD, age, and body weight of the dog; and 2) compare two
methods of selecting endotracheal tube size in dogs using
digital palpation of the tracheal OD and nasal septal width
of the dog.
Twenty-eight fresh dog cadavers were used in this study.
Eight dogs were females, and the rest were males. The ages
of the mostly mixed-breed dogs were between 6 months and
5 years, with body weights between 5.4 kg and 29.3 kg
[Table 1]. All dogs were dolichocephalic. Two methods
were used to select the endotracheal tubes for each dog. The
first method used the nasal septal width as a guide, and the
second method used digital palpation of the tracheal OD as
a guide. For the nasal septal width method, the midportion
of the endotracheal tube was held against the nasal septum
of the dog, and the tube size was estimated at the septums
narrowest point. An endotracheal tube with an OD matching
this septal width was then selected. The same person also
selected an endotracheal tube using digital palpation by
selecting the OD that most closely approximated the OD of
the dogs trachea. The digital palpation of the tracheal OD
was performed immediately cranial to the thoracic inlet,
over hair and other tissues situated around the trachea at that
location, just as would be done in a clinical setting. The
endotracheal tube size selected for each dog with each
method was then recorded.
Following the selection of endotracheal tubes, the nasal
septal widths of these dogs were measured in millimeters
with calipers between the two nostrils at the narrowest
point. For actual measurement of the tracheal ID and OD,
the whole airway was harvested, and the tracheal ID and
OD were measured in millimeters with calipers at the
approximate site immediately above the thoracic inlet
where digital palpation occurred. Different sizes of endotracheal tubes were used to actually perform tracheal intubation on the harvested cadaver trachea, and the size of the
endotracheal tube that best fit the dogs actual trachea was
selected and recorded.
Endotracheal tubes are clinically indexed with their ID.
Because both methods of selection were based on the use of

Materials and Methods

Endotracheal Tube Selection

237

the OD, in order to simulate the clinical situation, each

endotracheal tube selected by the OD had to be converted to
the ID and recorded for comparison between the two methods. The same person recorded all measurements to ensure
consistency. All of the endotracheal tubes used were the
polyvinyl chloride type.a

A correlation analysis of the variables using PC SAS

Version 9b was conducted. The variables used in this analysis were: nasal septal width; tracheal ID; tracheal OD; ID of
endotracheal tube used for actual intubation; OD of endotracheal tube used for actual intubation; ID and OD of endotracheal tube selected by using digital palpation; ID and OD
of endotracheal tube selected by using nasal septal width
method; age of the dog; body weight of the dog; and cube
root of body weight.
The authors calculated the Pearsons linear correlation
coefficient and reported a P value (P<0.01) to assess the
coefficients statistical significance. In addition, endotracheal selection methods were compared with two statistical
methods using the best-fit endotracheal tube to the cadaver trachea as a standard for comparison.
The first statistical method calculated an estimate of the
percentage of correct endotracheal tube selection based
on the ID of the best-fit endotracheal tube. This measure
served as a calculation of the accuracy of the two endotracheal tube selection methods by dividing the number of estimates that matched the correct endotracheal tube by the
total number of dogs. The second statistical method for calculation of the accuracy was expressed as a mean absolute
difference of the tube diameter determined by either method
(i.e., palpation-selected ID or nasal septal width-selected
ID) and the ID of the best-fit endotracheal tube directly
inserted into the trachea. For the digital palpation selection
method, the formula would be di = average(|pi bi|), where
di is the mean absolute difference, pi is the diameter of the
tube determined by palpation, and bi is the diameter of the
tube determined to be the best fit for the ith dog. The same
calculation was applied to the nasal septal width selection
method. This method evaluates how much the proposed
methods differ from what was considered best. Smaller
values of this statistic would reflect a higher level of precision. In addition, McNemars test was used to compare the
accurate selection rates between two methods based on the
previous statistical result. The null hypothesis tested with
McNemars was that both methods are equally accurate in
selecting the optimal size of tracheal tube.

Statistical Analysis

The results for major variables are presented in Table 1, and

the linear correlations and P values for major variables are
presented in Table 2. As expected, the endotracheal tube
selected and inserted directly into the cadavers trachea was
highly correlated with the tracheas ID and OD. The age of
the dog was poorly correlated with the tracheas ID or OD,
whereas the body weight of the dog was better correlated

Results

238

JOURNAL of the American Animal Hospital Association

September/October 2008, Vol. 44

Table 1
Tracheal Variables* of the Dogs
Variables

Mean SD

Minimum

Maximum

Age (mo)

28.720.7

6.0

60.0

Weight (kg)

15.47.7

4.5

29.0

Nasal septal width (mm)

7.61.4

5.0

11.0

Tracheal OD (mm)

20.04.8

12.0

28.0

Tracheal ID (mm)

16.04.9

8.0

23.5

Endotracheal tube OD (mm)

11.83.1

6.0

15.7

Endotracheal tube ID (mm)

8.52.1

5.0

11.0

Palpation of neck selection of OD

of endotracheal tube (mm)

10.51.9

6.8

13.5

Palpation of neck selection of ID

of endotracheal tube (mm)

7.81.5

5.0

10.0

Nasal septal width selection of

endotracheal tubeOD (mm)

9.31.6

6.8

12.1

Nasal septal width selection of

endotracheal tubeID (mm)

6.81.2

5.0

9.0

* Data are presented as mean standard deviation (SD) with maximum and minimum values.
OD=outer diameter
ID=internal diameter

with the tracheal ID (r=0.85) and OD (r=0.84). The nasal

width of the dog was poorly correlated with the age of the
dog but was significantly correlated with body weight. The
nasal septal width was significantly correlated with tracheal
ID (r=0.73; P<0.0001) and OD (r=0.72; P<0.0001). The
digital palpation of the tracheal OD was also significantly
(P<0.0001) correlated with the dogs tracheal OD. The
percentages of correct estimates are presented in Tables 3
and 4.
The digital palpation method of selecting endotracheal
tubes provided a higher percentage (46%) of correct estimates than the nasal septal width method (21%). For both
methods, the percentage of correct estimates was calculated
by dividing the number of correct estimates by the total
number of dogs [Tables 3, 4]. The absolute mean difference
also showed that palpation of the tracheal OD (mean difference = 0.857) is a more precise method to select the best-fit
tube than the nasal septal width (mean difference = 1.75)
method. McNemars test was performed on the data in

Tables 3 and 4, analyzing the null hypothesis that both

methods are equally accurate. The P value was 0.07. This is
not statistically significant at the 0.05 level, but it is marginally significant and is indicative of evidence that the
digital palpation method is more effective than the nasal
septal width method.
This study hypothesized that 1) the correlation was low
between the nasal septal width of the dog and the tracheal
ID or OD; and 2) that endotracheal tube size selection is
more accurate using digital palpation of the dogs tracheal
OD than using the dogs nasal septal width. The results of
this study did not support the first hypothesis; nasal septal
width correlated well with tracheal ID and OD. However,
this study confirmed the hypothesis that digital palpation of
the OD of the trachea is a more accurate method for selecting a best-fit endotracheal tube than using the width of the
nasal septum.

Discussion

r=0.82
P<0.0001

r=0.72
P<0.0001

r=0.73
P<0.0001

r=0.79
P<0.0001

r=0.79
P<0.0001

r=0.82
P<0.0001

Body weight

Tracheal OD (mm)

Tracheal ID (mm)

Optimal ET tube
OD (mm)

Optimal ET tube
ID (mm)

Body volume
(cube root of
body weight)

* OD=outer diameter
ET=endotracheal
ID=internal diameter

r=0.34
P=0.08

Age

Nasal
Septal
Width

r=0.90
P<0.0001

r=0.90
P<0.0001

r=0.88
P<0.0001

r=0.90
P<0.0001

r=0.88
P<0.0001

r=0.87
P<0.0001

r=0.46
P=0.01

Palpation of
Neck Selection
of OD* of ET
Tube (mm)

r=0.90
P<0.0001

r=0.90
P<0.0001

r=0.88
P<0.0001

r=0.90
P<0.001

r=0.88
P<0.0001

r=0.87
P<0.0001

r=0.46
P=0.01

Palpation of
Neck Selection
of ID of ET
Tube (mm)

r=0.88
P<0.0001

r=0.82
P<0.0001

r=0.80
P<0.0001

r=0.92
P<0.0001

r=0.76
P<0.0001

r=0.86
P<0.0001

r=0.34
P=0.08

r=0.87
P<0.0001

r=0.81
P<0.0001

r=0.79
P<0.0001

r=0.93
P<0.0001

r=0.75
P<0.0001

r=0.86
P<0.0001

r=0.34
P=0.08

Nasal Septal
Nasal Septal
Width Selection of Width Selection of
ET TubeOD (mm) ET TubeID (mm)

r=0.99
P<0.0001

r=0.88
P<0.0001

r=0.87
P<0.0001

r=0.85
P<0.0001

r=0.84
P<0.0001

1.00

r=0.51
P=0.01

Body
Weight
(kg)

The Linear Correlations (r) and P Values for Major Tracheal Variables in the Dog

Table 2

1.00

r=0.90
P<0.0001

r=0.89
P<0.0001

r=0.87
P<0.0001

r=0.87
P<0.0001

r=0.99
P<0.0001

r=0.52
P=0.0044

Body Volume
(Cube Root of
Body Weight)

September/October 2008, Vol. 44

Endotracheal Tube Selection
239

240

JOURNAL of the American Animal Hospital Association

September/October 2008, Vol. 44

Table 3
Contingency Table Representing Accuracy of the Nasal Septal Width Method of Endotracheal
Tube Selection
Optimal
Endotracheal
Tube Internal
Diameter Size
(mm)

Total No.
of Dogs

15

2*

4*

16

Endotracheal Tube Internal Diameter Size (mm)

Selected by Nasal Septal Width

17

0*

18

0*

19

0*

10

11

Total no. of dogs

28

* Note: The numbers marked with an asterisk represent a correct match of the internal diameter between the optimal endotracheal tube and the endotracheal tube selected by the nasal septal width method.

Table 4
Contingency Table Representing Accuracy of the Digital Palpation Method of Endotracheal
Tube Selection
Optimal
Endotracheal
Tube Internal
Diameter Size
(mm)

10

Total No.
of Dogs

15

1*

5*

2*

3*

16
17
18

Endotracheal Tube Internal Diameter Size (mm)

Selected by Digital Palpation

0
0

19

1*

10

1*

11

Total no. of dogs

28

* Note: The numbers marked with an asterisk represent a correct match of the internal diameter between the optimal endotracheal tube and the digitally selected endotracheal tube.

September/October 2008, Vol. 44

Based on the data from Tables 3 and 4, McNemars test

was able (with marginally statistical significance, P=0.07)
to identify that the percentage of correct tube selection for
the digital palpation method was greater than the percentage
of correct tube selection for the nasal septal width method.
Clinically, the difference in percentage of correct tube selection (46% for digital palpation versus 21% for nasal septal
width) is significant. However, McNemars test has limited
power when used with small sample sizes, such as those in
this study.
Although body weight was not used as part of the endotracheal tube selection method, the results of this study
revealed that body weight correlated better with tracheal
size than did the age of the dog; nasal septal width of the
dog correlated poorly with the age of the dog but did correlate well with the dogs body weight. A strong correlation
has been shown between tracheal diameter and body mass
in Dalmatian puppies (<5 months of age)3 and in small,
mixed-breed dogs.4 The current results are similar to these
studies. However, because of breed differences, it is difficult to use body weight as a generalized method for selecting a proper endotracheal tube in a clinical setting.5
A high value of the correlation coefficient means that
the two variables change in a linear fashion. However, just
because the correlation is high does not imply that the prediction is accurate. The current results showed that digital
palpation of the tracheal OD was a more precise method for
selecting a properly fitting tube than the nasal septal width
measurement method. The percentages of obtaining a correct estimate with the digital palpation and nasal septal
width methods were 46.4% and 21.4%, respectively.
Body masses of adult dogs vary dramatically between
breeds, and, therefore, the tracheal size varies accordingly.6
Despite this range of variation, a correlation was found
between body size and nasal septal width for the dogs in
this study. Body mass variation among breeds poses a challenge for unifying a standard method of selecting endotracheal tubes for dogs. Review of literature revealed minimal
information about the correlation among age, body weight,
and size of the trachea in mixed-breed dogs. It has been
suggested that mammalian linear dimension, such as tracheal diameter or nasal septal width, is not expected to vary
with body weight, but with cube root of body weight.7,8
The positive correlation seen with this study does not disprove this; it is most likely related to the relatively limited
range of body weights in the study sample. Clinical studies
have suggested that endotracheal tube size can be estimated by palpating the dogs trachea.6 However, others consider this to be a subjective method, especially for
brachycephalic dogs.9 In this study, only dolichocephalic
dogs were used.
In the current study, the tracheas of the cadaver dogs
were measured with calipers after they were harvested, and
the optimal endotracheal tube was selected by direct insertion of the best-fit tube into the trachea. Some factors one
would encounter in a clinical setting were not considered
during the selection of the endotracheal tube, such as size

Endotracheal Tube Selection

241

limitation of the laryngeal openings or obesity of the dog,

where the fat surrounding the trachea might also affect the
digital palpation of the trachea.
Besides the anatomical considerations, the endotracheal
tube itself should also be considered. For example, wall
thickness is not the same in all endotracheal tubes; therefore, ID will vary. Also, as an endotracheal tube ages, it may
lose wall rigidity, which may increase its tendency to collapse under ordinary cuff pressure. This would cause turbulent air flow and increase the work of breathing.
Furthermore, cleaning and sterilizing an endotracheal tube
for reuse can cause loss of pliable polymers, resulting in
stiffening or hardening of the tube wall. Thus, a slight
increase in cuff pressure to achieve tracheal wall cuff sealing can transfer all the pressure to the tracheal wall membrane, suppressing tissue blood flow.
Some challenges were encountered in this study. In a
clinical setting, endotracheal tubes are indexed by their
ID, yet clinical selections of endotracheal tubes are based
on the tubes OD. In other words, clinicians use either the
OD of the endotracheal tube for measuring against the
nasal septal width of the dog, or they use digital palpation
of the tracheal OD for estimation against the OD of the
endotracheal tube. Because of the clinical nature of these
processes, both endotracheal tube selection methods used
in this study exhibit some error for estimating the optimal
best-fit size of the endotracheal tube, resulting in a tendency for selecting smaller sizes of endotracheal tubes
rather than optimal best-fit tubes as the size of the trachea
increases. For example, the data in Tables 3 and 4 show
that a too-large tube was selected only once, whereas all
other incorrectly matched endotracheal tubes were ones
that were too small. This inaccuracy rate increased as the
optimal endotracheal tube ID sizes (that are approximate
to the tracheal ID sizes) increased from 7 mm to 11 mm in
both selection methods. Minimal impact was observed
when the endotracheal tube size was <7 mm ID. However,
when the correct tube size exceeded 7 mm, both methods
exhibited a tendency to select tubes that were too small by
1 to 2 mm.
These results show that the digital palpation method is more
efficient in selecting a best-fit endotracheal tube than the
nasal septal width method, and both methods are relatively
imprecise as tracheal size increases. Future studies are needed in this area, including evaluation of dogs in different
breeds and sizes.

Conclusion

a Sheridan/CF, Cuffed, Murphy Eye, oral/nasal tracheal tube; Hudson

Respiratory Care, Inc., Temecula, CA 92589-593
b SAS Institute, Cary, NC 27511-513

Footnotes

The authors thank Dr. Jennifer Hubbard and Meaghan

McMonagle for their technical assistance.

Acknowledgments

242

JOURNAL of the American Animal Hospital Association

11. Florete OG, Jr. Airway devices and their application. In: Kirby RR,
Gravenstein N, eds. Clinical Anesthesia Practice. Philadelphia: WB
Saunders, 1994:298-319.
12. Bolder PM, Healye TEJ, Bolder AR. The extra work of breathing
through adult endotracheal tubes. Anest Analg 1989;65(8):853-859.
13. Avki S, Yigitarslan K, Ozgel O. Comparison of airway size with
some phenotypic parameters in Dalmatian puppies: a practical
method to estimate endotracheal tube size. Vet Anaesth Analg
2006;33:24-27.
14. Dallman MJ, Brown EM. Structural considerations in tracheal disease. Am J Vet Res 1979;40:555-558.
15. Hartsfield SM. Airway management and ventilation. In: Thurmon
JC, Tranquilli WJ, Benson GJ, eds. Lumb and Jones Veterinary
Anesthesia. 3rd ed. Baltimore: Williams & Wilkins, 1996:515-556.

References

September/October 2008, Vol. 44

16. Brock N. Questions about anaesthetic equipment. Can Vet J
1998;39:311-312.
17. Woldehiwot Z, Horsfield K. Diameter, length and branching angles
of the upper airways in the dog lung. Respir Physiol 1978;33:
213-218.
18. Mortola JP, Fisher JT. Comparative morphology of the trachea in
newborn mammals. Respir Physiol 1980;39:297-302.
19. Harvey CE, Fink EA. Tracheal diameter: analysis of radiographic
measurements in brachycephalic and nonbrachycephalic dogs. J Am
Anim Hosp Assoc 1982;18:570-576.