. 5.
s...
I.
x

 

Hfiummwnwwwfiéfluufiéx 5.4...

1.x IN? [74.15%

152:“).- ‘
a; « l; 3

‘11. «I . .

:32; 1.53%..
uiwg S

gran;
6.

s. u ( flu .
.1 Kiwi... in f
.K‘I~ isz,

3.9. .5}!

I. 7:}
01.! a. .

.2. 1v 13

Illllllllllll\lllllllllll ll ill lllll llllllll

LIBRARY
Mlchlgan State
Unlverslty

 

 

 

 

This is to certify that the
thesis entitled

USE OF FLOW-VOLUME LOOPS T0 EVALUATE
EQUINE LEFT LARYNGEAL HEMIPLEGIA AND ITS
TREATMENT BY PARTIAL ARYTENOIDECTOMY

presented by

Jonathan Mark Lumsden

has been accepted towards fulfillment
of the requirements for

Masters Science

 

 

degree in

Department of Large Animal Clinical Sciences
’5 fWS‘
/ Majo or professor
BIRM—

O-7639 MS U is an Affirmative Action/Equal Opp oooooo ' 1y Institution

 

 

PLACE IN RETURN BOX to remove thie checkout from your record.
TO AVOID FINES return on or before ode due.

  

DATE DUE DATE DUE DATE DUE

E-Cj
TLJ-LJ

m

| l——|

MSU to An Affirmative Action/Equal Opportunlty Institution
omens-9.1

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

USE OF FLOW-VOLUME LOOPS To EVALUATE
EQUINE LEFT LARYNGEAL HEMIPLEGIA AND ITS
TREATMENT ev PARTIAL ARYTENOIDECTOMY

By

Jonathan Mark Lumsden

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Large Animal Clinical Sciences

1993

ABSTRACT
use OF FLOW-VOLUME LOOPS To EVALUATE
EQUINE LEFT LARYNGEAL HEMlPLEGIA AND ITS
TREATMENT BY PARTIAL ARYTENOIDECTOMY
By

Jonathan Mark Lumsden

Upper airway function was evaluated in 6 horses at rest and during
exercise. Expiratory and inspiratory impedance (2E and 2.) were calculated and
tidal breathing flow-volume loops (TBFVL) were acquired. TBFVL indices
included inspiratory and expiratory flow at 50% of tidal volume (IF50 and EFso,
respectively). After left recurrent laryngeal neurectomy (LRLN), ZI and EFSOIIF50
significantly increased and IF60 significantly decreased from baseline (before
LRLN) values during exercise. After partial arytenoidectomy, Z. returned to
baseline values though IF,so and EFm/IF,50 remained significantly different from
baseline values. After LRLN, TBFVL showed marked inspiratory airflow
limitation. Following partial arytenoidectomy, TBFVL shape approximated that
seen at the baseline evaluation, although partial inspiratory flow limitation
persisted. When obtained during exercise, TBFVL allowed non-invasive,
objective, specific, and repeatable detection of upper airway obstruction.
Partial arytenoidectomy improved upper airway function during exercise in

horses with left laryngeal hemiplegia.

To the Lumsden family; Janice-Mary, Ian-Henry, Andrew and Simon, for sharing
my enthusiasm throughout my residency through your exciting and positive

approach to life.

ACKNOWLEDGEMENTS

To Dr. Fred Derksen for your guidance, your drive for quality research and your

attention to goals.

To Dr. Ed Robinson for your never ending enthusiastic and logical contribution

to my research training.

To Dr. John Stick for your good faith, inspiration and tireless positive approach

to all aspects of my residency training.

To Dr. John Caron for going beyond the expected to cultivate my clinical desire

to question and appreciate the questioned.

To Dr. Frank Nickels and Dr. Chris Brown for your instruction and friendship

throughout my residency.

To the prettiest girl in Holt, for sharing your understanding of people and of

”what it's all about”.

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

CHAPTER

INTRODUCTION

ASSESSMENT OF UPPER AIRWAY FUNCTION

A.
B
C.
D

E.

Principles of Flow Mechanics

. Flow Mechanics in the Upper Respiratory Tract

Upper Airway Resistance Measurement Techniques in Humans

. Measurement of Upper Airway Obstruction in Horses

Development of Flow-Volume Loop Analysis

EQUINE LEFT LARYNGEAL HEMIPLEGIA - A REVIEW

A.
B.
C.

D.

Anatomy and Physiology of the Larynx
Incidence
Pathogenesis

Diagnosis

vii

viii

15

18

22

38

42

50

TABLE OF CONTENTS IcontJ

IV. TREATMENT OF LEFT LARYNGEAL HEMIPLEGIA - A REVIEW

A. Ventriculectomy 59
B. Prosthetic Laryngoplasty 61
C. Arytenoidectomy 66
D. Laryngeal Reinnervation 72

V. USE OF FLOW-VOLUME LOOPS TO EVALUATE UPPER AIRWAY
OBSTRUCTION IN EXERCISING STANDARDBRED HORSES

A. Summary 78
B. Introduction 79
C. Materials and Methods 82
D. Results 86
E. Discussion 99

VI. EVALUATION OF PARTIAL ARYTENOIDECTOMY AS A TREATMENT FOR
LEFT LARYNGEAL HEMIPLEGIA

A. Summary 109
B. Introduction 110
C. Materials and Methods 112
D. Results 115
E. Discussion 123
Vll. SUMMARY AND CONCLUSIONS 128

LIST OF REFERENCES 130

vi

LIST OF TABLES

lame Base

1. Effect of exercise on the measured and calculated
variables in the 6 horses before (baseline) and after left .
recurrent laryngeal neurectomy (LRLN). 94

2. Effect of exercise on selected tidal breathing flow-
volume loop indices from the 6 horses before (baseline)
and after left recurrent laryngeal neurectomy (LRLN). 95

3. Coefficients of variation for various tidal breathing flow-
volume loop indices from the 6 horses before (baseline)
and after left recurrent laryngeal neurectomy (LRLN). 96

4. Effect of surgery on measured and calculated variables
from 6 horses with surgically induced left laryngeal
hemiplegia before (pre-operatively) and after (post-
operatively) left partial arytenoidectomy and bilateral
ventriculectomy. 1 17

vii

LIST OF FIGURES

Einute

1. Diagrammatic representation of Bernoulli's effect creating
collapse of a narrowed airway. For total enerQY (Emmi, +
Emmi conservation within the system, the increase in
airflow velocity (V)(EM) at the narrowed portion of the
airway from V1 to V2, there must be a reciprocal decrease
in intraluminal pressure IPIIEmme from P1 to P2 at the
site of the airway narrowing.

2. a, Relationship of transpulmanory pressure and flow for a
normal subject at different degrees of lung inflation
(curves 1, 2, 3, and 4). b, A plot of maximum achievable
flow against degree of lung inflation. Flow and volume
co-ordinates of points A, B, C, and D from fig 1a are
plotted as closed circles on this curve. 0 corresponds to
maximum expiration point and y to maximum inspiration
point (From Hyatt et al,1958l.

3. Representative flow-volume loop shape, showing the
effect of fixed, variable extrathoracic, and intrathoracic
airway obstructions.

4. Mean (:1: SEM) heart rate (beats/min;bpm) in 6 horses
performing an incremental treadmill exercise test prior to
left recurrent laryngeal neurectomy (LRLN). The treadmill
had a 3-degree slope.

viii

13

25

29

87

LIST OF FIGURES (cont)

Eigum

5. Representative tidal breathing flow-volume loops (TBFVL)
from 4 horses at rest. Variation in TBFVL shape was
seen within and between horses. Tidal volume at peak
inspiratory flow was variable. The TBFVL indicate peak
inspiratory flow near the end of inspiration (A) and early
in inspiration (B). Variations in the inspiratory limb of
TBFVL included monophasic (A and B), biphasic (C), and
triphasic (D) patterns.

6. Representative TBFVL from a horse at rest (period A),
exercising at 75% of maximal heart rate (period B), and at
maximal heart rate (period C). The effect of increasing
exercise on shape, airflow rate, and tidal volume is
shown. The representative TBFVL (C) at period C shows
the flow measurements used to calculate TBFVL indices.
Curved arrows indicate the direction of the respiratory
cycle.

7. Airflow curves from different horses before LRLN
exercising at maximal heart rate (period C). Horses had a
constant biphasic breathing pattern (A), a combination of
biphasic and monophasic patterns (B), or a constant
monophasic pattern (C). The TBFVL derived from flow
and volume tracings are at the right of the tracing. A
typical biphasic TBFVL shape is shown in A, composed of
biphasic expiratory and inspiratory airflow curves.
Monophasic expiratory airflow curves were seen with a
biphasic (B) or monophasic (C) inspiratory airflow curve.
V = airflow (50 L/s/division), VT = tidal volume (10
L/division) and, t = time (0.5 sec/division).

8. Typical TBFVL from a horse exercising at maximal heart
rate (period C) before LRLN (a) and after LRLN (b).
Preservation of the expiratory flow curve and inspiratory
flow limitation are seen after LRLN.

89

9O

91

92

LIST OF FIGURES (contJ

Elmira

10.

11.

12.

Peak expiratory flow-to-peak inspiratory flow ratio
(PEF/PIF) and expiratory-to-inspiratory flow ratio at mid-
tidal volume (EFso/IFSO) before and after LRLN at rest and
exercise. Baseline = prior to LRLN; LRLN = 14 days
after LRLN; * = data significantly (P < 0.05) different
from baseline measurement at same speed; * * = data
significantly (P < 0.01) different from baseline
measurement at same speed.

Peak inspiratory flow (PIF), inspiratory flow at mid-tidal
volume (lFsoI. and inspiratory flow at 25% tidal volume
(le5) before and after LRLN at rest and exercise. Baseline
= prior to LRLN; LRLN = 14 days after LRLN; * = data
significantly (P < 0.05) different from baseline
measurement at same speed; ** = data significantly (P
< 0.01) different from baseline measurement at same
speed.

TBFVL's from a horse at baseline (3), after LRLN (b), and
following partial arytenoidectomy and bilateral
ventriculectomy (c). TBFVL were generated at a speed
corresponding to maximal heart rate.

Peak inspiratory pressure (Pm), peak inspiratory flow (PIF),
and inspiratory impedance (2.) before and after LRLN, and
after partial arytenoidectomy at rest (A) and during
moderate (B) and strenuous (C) exercise. Baseline =
prior to LRLN; LRLN = 2 weeks after LRLN; ARYT = 16
weeks after partial arytenoidectomy * = data
significantly different from baseline measurement at same
speed (p < 0.05). f = data significantly different from
LRLN measurement at same speed (p < 0.05).

97

98

116

118

LIST OF FIGURES lcont.)

Elmira Eage

13. Peak expiratory pressure (Pug), peak expiratory flow (PEF),
and expiratory impedance (2;) before and after LRLN, and
after partial arytenoidectomy at rest (A) and during
moderate (B) and strenuous (C) exercise. Baseline =
prior to LRLN; LRLN = 2 weeks after LRLN; ARYT = 16
weeks after partial arytenoidectomy. 119

14. Peak inspiratory flow (PIF), inspiratory flow at mid-tidal
volume (lel, and inspiratory flow at 25% tidal volume
(le5) before and after LRLN, and after partial
arytenoidectomy at rest (A) and moderate (BI and
strenuous (C) exercise. Baseline = prior to LRLN; LRLN
= 2 weeks after LRLN; ARYT = 16 weeks after partial
arytenoidectomy. * = data significantly different from
baseline measurement at same speed (p < 0.05). 120

15. Peak expiratory flow/peak inspiratory flow ratio (PEF/PIF)
and expiratory/inspiratory flow ratio at mid-tidal volume
(EFso/IFBOI before and after LRLN, and after partial
arytenoidectomy at rest (A) and moderate (B) and
strenuous (C) exercise. Baseline = prior to LRLN; LRLN
= 2 weeks after LRLN; ARYT = 16 weeks after partial
arytenoidectomy * = data significantly different from
baseline measurement at same speed (p < 0.05).f =
data significantly different from LRLN measurement at 121
same speed (p < 0.05).

16. Peak expiratory flow (PEF), expiratory flow at mid-tidal
volume (EFso), and expiratory flow at 25% tidal volume
(E1325) before and after LRLN, and after partial
arytenoidectomy at rest (A) and moderate (B) and
strenuous (C) exercise. Baseline = prior to LRLN; LRLN
= 2 weeks after LRLN; ARYT = 16 weeks after partial
arytenoidectomy. 1 22

xi

I. INTRODUCTION

Reduced athletic performance as a result of left laryngeal hemiplegia
(LLH) is a common cause of wastage in the racing industry, and may also result
in early retirement of non-racing equine athletes (Rossdale et al,1985;
Cook,1974). The observed exercise intolerance in horses with LLH is a direct
result of paralysis of the left cricoarytenoideus dorsalis (CAD) muscle (Dupey,
1815 in MacDueen,1896; Duncan et al,1974). Paralysis of the CAD muscle,
in the majority of cases, has been attributed to a distal axonopathy of the left
recurrent laryngeal nerve (RLN) (Cole,1946; Duncan et al,1974). The etiology
of this neuropathy remains undetermined.

Use of the flexible fiberoptic endoscope for examination of the upper
respiratory tract of horses revolutionized the accuracy with which LLH could be
diagnosed (Cook,1974). More recently, videoendoscopic examination of
laryngeal function during treadmill exercise has provided more sensitive
detection of this upper airway obstruction (Derksen,1988; Morris and
Seeherman,1988). Further information, regarding the pathophysiology of LLH
at rest and during exercise, has been provided through quantitative evaluation
of upper airway flow mechanics and rima glottidis cross-sectional area (Derksen

et al,1986; Martin et al,1986). Unfortunately, these objective measurement

2

techniques have afforded limited clinical use because of their invasive and/or
time consuming nature.

In human medicine, objective and sensitive evaluation of upper and lower
airway function is routinely achieved in a clinical setting using analysis of
maximal effort flow-volume loops (FVL). A maximal effort flow-volume loop
represents an x-y plot of airflow rate versus volume, when a patient inspires
and then expires with maximal effort (Hyatt et al,1958l. This commonly used
test of airway function in humans has had limited application in veterinary
medicine because of the uncooperative nature of animals. Therefore, the first
objective of my study, was to determine if flow-volume loop analysis could be
used to quantitate airway obstruction in horses with surgically-induced LLH at
rest and during exercise.

Objective evaluation of upper airway function in horses has been
reported for treatments of LLH including ventriculectomy, prosthetic
laryngoplasty, subtotal arytenoidectomy and laryngeal reinnervation (Shappell
et al,1988; Belknap et al,1990; Fulton et al,1991). At present, based on
subjective and objective evaluations, prosthetic laryngoplasty, is the preferred
method of surgical treatment. Despite this, prosthetic laryngoplasty may be
associated with numerous post operative complications, therefore the search
for the ideal treatment of LLH continues.

Following the recognition of arytenoid chondropathy and because of the
need to treat unsuccessful prosthetic laryngoplasty procedures,

arytenoidectomy was advocated in the early 1980's (Haynes et al 1980; White

3

and Blackwell et al,1980). In addition, surgeons have suggested the use of
arytenoidectomy as a primary treatment for LLH (Haynes et al,1984).
Presently, recommendations regarding the arytenoidectomy technique ofchoice
remain controversial. Subtotal arytenoidectomy, which leaves the corniculate
cartilage in situ, is preferred by some authors because of the high incidence of
coughing and dysphagia reported with partial arytenoidectomy, which involves
removal of the corniculate cartilage (Haynes,1981;Speirs,1986).

Recent objective information indicates that subtotal arytenoidectomy fails
to improve upper airway function and that the upper airway obstruction results
from axial displacement of the corniculate cartilage during inspiration (Belknap
et al.1990; Stick and Derksen,1989). In addition, recent studies report a lower
rate of post-operative complications after partial arytenoidectomy than
previously described (Tulleners et al,1988zl). Therefore, the second objective
of this study was to evaluate the efficacy of partial arytenoidectomy, in

restoring upper airway function in horses with LLH.

ll. ASSESSMENT OF UPPER AIRWAY FUNCTION

A. Principles of Flow Mechanics

In its simplest form the upper airway acts as a conduit for airflow from
the atmosphere to the lung. The flow of fluid in a rigid cylindrical tube is
described and predicted by the laws of fluid mechanics. In contrast, airflow
through the mammalian respiratory system is much less well defined by
equations used in aerodynamics. Factors limiting application of the principles
of hydrodynamics and aerodynamics to airways arise from the complex
geometry and dynamic nature of the respiratory tract (West,1990). Despite
such limitations, use of laws of fluid mechanics in studies of flow-volume
relationships in airways has increased the understanding of mammalian airflow
mechanics and has also allowed quantitation of the mechanical properties of
the respiratory system.
Laminar flow

Fluid flow, whether it is air or liquid, requires a force. The force for flow
in a tube is provided by a pressure difference between the two ends of the
tube, referred to as the driving pressure. The flow of air through a rigid smooth
cylinder at low flow rates is characterized by stream lines or flow laminae,
which are parallel to the sides of the tube. This kind of fluid flow is defined as
laminar flow.

As airflow enters a tube, the leading edge of flow is initially flat, because
all molecules enter at the same velocity. Over a finite period of time the flow

profile becomes parabolic. During the development of this parabolic profile,

4

5

high shear forces exist between fluid molecules and the cylinder wall. Shear
forces gradually decrease from the periphery, where there is zero velocity of
fluid molecules adjacent to the cylinder wall, to the centrally located flow
laminae. This allows acceleration of velocities at the center of the cylinder,
increasing shear at the center of the cylinder and thus equalizing forces across
the profile (Plint and Bosworth,1978). Thus in a finite distance (entrance
length) and time, an even distribution of shear between laminae results in a
parabolic flow profile where flow near the center has a velocity almost twice
the average velocity. Mathematically the relationship between pressure and

flow in a laminar flow regime may be described by Poiseuille’s Law.

 

Where V = airflow, P = driving pressure, r = tube radius, 11 = gas
viscosity, and l = tube length. Thus, when tube dimensions are constant,
pressure is directly proportional to flow, or P = KN , where K1 represents the
mechanical energy loss within the system from laminar flow.

Laminar flow resistance

In air- or liquid-filled systems resistance to flow is defined as the
difference between inflow and outflow pressures divided by the instantaneous
rate of flow of fluid (Pride, 1 970). The term, resistance, is analogous to Ohm's

law in electricity, where the resistance of a conducting wire is the potential

6

difference across its length divided by the instantaneous current. Thus
resistance may be represented by:

R a P‘ ' P2
\‘I
where R = resistance, P1 - P2 = pressure change along the tube, and V =

flow rate.
By combining the resistance formula and Poiseuilles equation, resistance

to flow may be expressed solely by the geometry of the tube and the viscosity

of the gas.
n = §Jfl
x r‘
Where u = gas viscosity, l = length of the tube, and r= radius of the tube.

Clearly from this equation it can be seen that relatively small changes in tube
diameter will result in large changes in flow resistance. if the radius is halved
the resistance increases by 16-fold. However, doubling the length only doubles
the resistance.
Turbulent flow

At higher flow rates complete disorganization of stream lines occurs.
Particle velocity throughout the flow varies in magnitude and direction resulting
in turbulent flow. The properties of turbulent flow differ from those of laminar

flow. The flow profile is flat, flow at the center of the cylinder is only 1.2

7

times the average velocity and driving pressure is not proportional to flow rate,
but approximately to its square. When flow is turbulent, P is equal to the
product of K2 (anatomically dependent constant) and V2 (Plint and
Bosworth,1978). in addition, the pressure-flow relationship in turbulent flow
is dependent on gas density.
Turbulent flow resistance

Turbulent flow results in an increase in shear stresses between flow
laminae. This increase in shear stresses is associated with a greater pressure
drop for a given flow, and therefore the driving pressure for a given flow is
greater when the flow regime is turbulent rather than laminar. When flow
through a passage is both turbulent and laminar in places, it may be described
by Rohrer’s equation, where flow has both laminar and turbulent properties

(Hamilton, 1 979):

P = K1uV + szV2

where P = driving pressure, K1 = mechanical energy loss due to laminar flow,
[I = the viscosity of the gas, K2 = mechanical energy loss due to turbulent
flow, p = the density of the gas, and V = airflow. As flow is not constant
when the flow regime is turbulent, resistance is defined as the ratio of the

change in pressure over the change in flow:

Resistance = 5—1:

v1 -v2

8

Resistance under turbulent conditions is thus dependent upon the flow rate at
which it is measured.
Reynoid's Number

Reynolds (1883) was the first to calculate the constant value at which
the flow regime in a tube changes from laminar to turbulent. This dimension
less value is dependent on the characteristics of the tube, the density, and
viscosity of the fluid and flow velocity, and is represented by:

Basie—v
cur

wherep = density, V = airflow velocity, r = radius, and u = viscosity.

The validity of this equation has been confirmed through experiments
using a wide range of tube diameters, flow rates, and fluid properties. Laminar
flow occurs when Re is less than 2300 and turbulent flow occurs when Re is
greater than 3000 (Patel and Head,1968). Transitional flow, occurring at
moderate flow rates, has characteristics of both laminar and turbulent flow and
is associated with a Re between 2300 and 3000.

Entrance conditions of a branching system of tubes have been shown to
influence the critical Re at which flow regime is altered (Pedley et al,1970).
Typically, if incoming fluid has disturbances, then the Re at which turbulence
occurs will be reduced, the driving pressure needed for flow will be greater. The
larger airways of the respiratory system represent the greatest resistive

pressure drop of the respiratory system. This is because airflow velocities are

9

greater in larger central airways compared to smaller peripheral airways and the
total cross-sectional area is much smaller in the central airways than the
peripheral airways. Therefore, the driving pressure needed for flow from
atmosphere to alveoli is significantly affected by entrance effects in the upper
airway and larger diameter parts of the lower respiratory tract (Pedley et

aL1970L

8. Flow Mechanics in the Upper Respiratory Tract

The upper airway begins at the nostrils and includes the nasal cavity, the
pharyngeal cavity, and finally the larynx. The flow of air from atmosphere
through the upper airway, to the alveoli, is the result a force provided by the
pressure difference between the pleural cavity and the atmosphere (driving
pressure). inspiratory driving pressure is produced by diaphragmatic and
intercostal muscle contraction, resulting in a decrease in intrapleural and
subsequently alveolar pressure.

The net driving pressure required for gas flow through the upper airway
is dependent on the flow rate, whether flow is laminar or turbulent, the density
(when airflow is turbulent) and the viscosity of the gas, and several important
impedance factors. These impedance factors include airway resistance,
resulting from narrow and irregular geometry of the upper airway and
reactance, consisting of airway narrowing associated with airway wall
compliance and inertial forces created by acceleration and deceleration of

airflow when airflow reverses.

10

Turbulent flow in the upper airway

it has been shown in the upper respiratory tract of humans,that laminar
flow only occurs at very low flow rates (< 1.0 L/s) (Olson et al,1970). In the
horse, the Reynolds number calculated for airflow at 1.3 Us in the trachea
suggests that the flow regime is turbulent (Attenburrow et al,1983). Since
airflow in the upper respiratory tract of the horse at rest is approximate 5 Us
and may increase 20—fold during strenuous exercise (Hornicke et al,1987), the
laws of fluid dynamics defining turbulent flow are appropriate.

With the reversal of flow direction during the respiratory cycle, inertial
forces are created. This inertance increases the driving pressure required to
produce a given flow. At rest, inertance appears to be of minor significance,
but when flow rates and respiratory frequencies increase, as during exercise,
inertial forces are likely to increase greatly. Techniques used to evaluate upper
airway function in man, minimize the need to account for inertial forces
associated with measurement of airflow and its’ required driving pressure. This
is because airflow measurement is performed either during quiet breathing or
during forceful, unidirectional, controlled respiration. in contrast, airflow
measurement in exercising horses is associated with high airflow rates and
respiratory rates and inertance forces are likely to have a significant effect on

driving pressure.

11

Upper airway compliance

An additional complicating factor, making direct application of Rohrers
equation inadequate, is that the respiratory system is not a rigid tube, but
instead represents a series of airways whose dimensions change throughout the
respiratory cycle. Therefore, the constant values used in his equations,
representing mechanical losses due to geometry, are unlikely to remain truly
constant. The amount of airway diameter change in response to transmural
pressure during inspiration and expiration is determined by the compliance of
the airway wall. During inspiration, intraluminal upper airway pressure is sub-
atmospheric, whereas pressures are positive with respect to atmosphere during
expiration. Collapsible structures in the upper airway rely on adequate dilatator
muscle function to resist negative intraluminal pressures during inspiration.
Dilator muscle contraction maintains airway caliber and thus airflow during
inspiration (Robinson and Sorenson,1978). Collapsible structures of the upper
airway include the external nares, pharynx, and the arytenoid cartilages of the
larynx, which are supported principally by muscle rather than bone or cartilage.
The ability of the upper airway to resist dynamic collapse on inhalation is
apparently incomplete because, inspiratory resistance has been shown to be
50% and 200% greater than expiratory resistance during quiet breathing at rest
and during strenuous exercise respectively (Derksen et al,1986). Other
methods that may reduce flow impedance in the upper airway are straightening
of the airways and sympathetic vasoconstriction of vascular sinuses in the

nasal mucosa (Derksen,1988).

12

In humans, the effect of compliance on measurement of airway function
is reduced by measuring airflows and pressure differences during expiration.
Additionally, the minimal contribution of the soft palate to the nasopharynx in
man (Procter,1977), results in a low compliance upper airway. In contrast to
man, the upper airway of the horse may represent a more compliant conduit,
as the horse has a long soft palate that forms an intimate seal with the ventral
aspect of the epiglottis (Cook, 1 981 :I). The potential of the equine nasopharynx
to narrow during inspiration and the substantial inertia associated with the
reversal of flow direction at high flow rates warrant the description of total
upper airway resistance to flow as impedance rather than resistance.

Upper airway resistance

The resistance to airflow is a combination of turbulence and drag created
by respiratory tract geometry, friction between air and airway walls, and the
shear forces present within a flowing gas (Olson et al,1970). Clearly, the upper
respiratory tract is anything but a "smooth, rigid, cylinder," but rather a conduit
of varying diameter with extensive surface irregularities. Considering the
effects of varying airway diameter, we can predict that if flow rate is to remain
constant and airway diameter decreases, then flow velocity must increase.
This is referred to as "convective acceleration” (Fry and Hyatt,1960).
Acceleration requires a force. This force is provided by an increase in driving
pressure, that is, a greater pressure drop along the airway. Associated with
convective acceleration is a further drop in pressure along the stream in an

effort to conserve energy within the system (Bernoulli’s principle) (Figure 1).

13

This additional pressure drop further increases the tendency for collapsible

segments of the airway to narrow, thus perpetuating a vicious cycle of

 

 

 

’__‘ _

 

 

Figure 1. Diagrammatic representation of Bernoulli's effect creating collapse
of a narrowed airway. For total energy (EM 4» Emmi conservation within
the system, the increase in airflow velocity WHEN) at the narrowed portion
of the airway from V1 to V2, requires a reciprocal decrease in intraluminal
pressure (P)(E,,,,,,,,,,,) from P1 to P2 at the site of the airway narrowing.

This cycle is epitomized in the equine athlete suffering from left laryngeal
hemiplegia. Paralysis of the abductor muscle of the arytenoid cartilage prevents
the normal dilatator function needed to maintain airflow during exercise. The
cross-sectional area of the rima glottidis (opening to the trachea) is smaller than
normal as the corniculate cartilage cannot be fully abducted. Air is subject to
increased convective acceleration through the narrowed rima glottidis, which

is associated with a further pressure drop (Bernoulli effect). The greater

14

negative intraluminal pressure results in further narrowing, as the corniculate
cartilage is unable to abduct, and becomes displaced axially. When subject to
high airflow rates, the progressive airway narrowing and negative intraluminal
pressure may result in complete obstruction of the rima glottidis by the
corniculate cartilage (dynamic collapse) during inspiration (Derksen,1988).
Upper airway impedance

The relationship between resistance and airway radius (Poiseuilles law)
provides important insight into the relatively mild changes in airway diameter
that can cause significant increases in resistance to flow. in an attempt to
better describe pressure-flow relationships in the airway, parallels drawn from
the field of electrical engineering accommodate the role of inertance and
compliance. The sum of these two components (reactance) is combined with
resistance in one term called the impedance (Young and Hall,1989). Impedance
is defined as the total opposition to flow in the part of the respiratory tract

examined.

where Z = impedance, AP = driving pressure, and AV = change in flow rate.

15

C. Upper Airway Resistance Measurement Techniques in Humans
Upper airway resistance in Humans

Early studies showed a nonlinear relationship between pressure and flow
in the upper airway, indicative of turbulent flow (Hyatt and Wilcox, 1 961 ; Ferris
et al,1964). For this reason, resistance was consistently calculated at a low
flow rate of 0.5 US, at which flow was considered to be laminar. Increase in
lung volume decreases upper airway resistance (Hyatt and Wilcox, 1 961; Ferris
et al,1964) because, during maximal inspiration head and neck position
straightens the upper airway and widens the rima glottidis.
Partitioning of upper airway resistance

Hyatt and Wilcox (1961) were the first to partition respiratory resistance
into lower and upper airway components by measuring intra-tracheal pressure
with a blunt 19-gauge needle introduced perpendicularly into the trachea, 2-3
cm below the cricoid cartilage. The intra-tracheal-oral pressure gradient was
recorded during various respiratory maneuvers. Oral pressure was recorded just
inside to the lips through a 19-gauge needle and subjects mouth breathed with
their nose clamped. Airflow was measured using a concentric cylinder
flowmeter, and respiratory volumes were recorded using a spirometer. Upper
airway resistance was estimated to contribute approximately 45% of total
airway resistance in normal subjects. This value is similar to that predicted by
Rohrer in 1915 based on computations from anatomical models.

The upper airway not only provides a large portion of resistance to

airflow, this resistance is variable within and between subjects (Hyatt and

16

Wilcox, 1 961 ). The observed variability was considered to result from changes
in the laryngeal airway size, which comprised the most significant site of airway
resistance during mouth breathing. The importance of the larynx as a source
of airway obstruction during mouth breathing, was supported by the
observation that during hyperventilation the vocal cords remained abducted,
and airway resistance decreased by approximately 40%.

In a later study, Ferris et al (1964) suggested that the mouth and
pharynx were also significant and variable sources of airway resistance. Spann
and Hyatt (1971), after measuring mouth, pharynx, and larynx resistances,
considered pharyngeal resistance negligible, and mouth resistance accounted
for 45% and 70% of upper airway resistance (Ruaw) during quiet mouth
breathing at low and high lung volumes, respectively.

Posterior rhinomanometry

As man is primarily a nasal breather, Ferris et al (1964) attempted to
quantify the contribution of nasal passages to upper airway and total
respiratory resistance. Upper airway resistance was partitioned by employing
a new and less invasive method of obtaining airway pressure. Nasal resistance
was measured by obtaining the pressure difference between the inside of a
mask placed over the face of the subject and a tube placed via the mouth into
the oropharynx . The pressure difference between mask and oropharynx was
divided by the measured airflow from a mask flowmeter while subjects
breathed through their nostrils. This technique is referred to as posterior

rhinomanometry. A similar technique was described by Speizer and Frank

17

(1964), except that the orally placed catheter measured oropharyngeal
pressures through pressure changes in a latex balloon attached to its tip. This
technique uses mouth pressure as an index of nasopharyngeal pressure,
because upper airway anatomy in humans allows direct communication
between the ore-and naso-pharynx (Proctor,1977). Disadvantages of posterior
rhinomanometry include, 1) the need to diligently hold the mouth tube to
prevent air from escaping around the tube, and 2) avoiding sealing off from the
pharynx that part of the oral cavity containing the catheter (Ferris et al,1964;
Nolte and Liider-Liiher,1972). Reports using posterior rhinomanometry
concluded that nasal resistance accounts for nearly half the total and two-thirds
of upper airway resistance during nose breathing (Ferris et al,1964; Speizer et
al,1964). It has been stated that nasal breathing nearly doubles the total
respiratory resistance compared to oral breathing (Procter,1977).
Anterior rhinomanometry

During anterior rhinomanometry one nasal passage contains a catheter
for measurement of pharyngeal pressure while the patient breaths through the
other nasal passage (Nolte and Luder-Lfiher,1972). This method has clinical
advantages over posterior rhinomanometry in that it eliminates the
inconvenience of a pharyngeal tube and the inability of some patients to hold
the tube correctly.

Anterior rhinomanometry assumes that left and right nasal resistance are
equal. The limitations of this measurement technique were documented by

Nolte and Liider-Liihr (1972) in a study comparing measurement of nasal

18

resistance by posterior rhinomanometry, anterior rhinomanometry, and whole-
body plethysmography. They concluded that anterior rhinomanometry over-
estimates the measured resistance when laminar flow is assumed and under-
estimates the value when turbulent flow is assumed.
Body plethysmography

Early on in the search for a clinical, quantitative, and practical method of
measuring nasal resistance, body plethysmography was suggested as a
potential diagnostic tool (Butler,1960). Dubois et al (1955) performed
pulmonary function tests using a non-invasive technique that employs a
constant-volume, variable-pressure, whole-body plethysmograph.

Measurement of nasal resistance (Rn), using body plethysmography is
based on the fact that total resistance between alveoli and nose openings
consists of two partial resistances: the resistance while mouth breathing (Raw)
and Rn. Thus the difference in resistance measurements during mouth
breathing and nose breathing will represent nasal resistance. The advantages
of body plethysmography as a clinical diagnostic tool are that it is efficient and
reliable and requires minimal demand for skill and cooperation by the patient.
The obvious limitation of this technique is the cost of equipment and technical

skill necessary.

D. Measurement of Upper Airway Obstruction in Horses
Measurements of inspiratory and expiratory airflow in horses walking on

a treadmill were performed as early as 1898 (Zuntz and Hagemann,1898). In

19

1966, airflow measurements were obtained from standing horses, using a
pneumotachograph attached to a close-fitting face mask covering the horse’s
nose and mouth (Gillespie et al,1966). The pneumotachograph functions as a
flow laminator and provides a constant resistance. Differential transducers
connected to either side of the pneumotachograph, provide a measurement of
the change in pressure across the pneumotachograph, which is proportional to
flow rate.

Calculation of upper airway impedance in the resting horse was initially
described by Robinson et al, (1975). Airflow was measured through a
pneumotachograph attached to a face mask, and trans-upper airway pressure
was determined as the difference in pressure between a catheter at the level
of the nares and an 18-gauge needle inserted into the trachea just below the
larynx. The contribution of upper airway resistance to total airflow resistance
was further investigated by measuring pressure in the trachea and in the pleural
space. Using a dead horse head, upper airway resistance was partitioned by
a catheter pull through technique at constant airflows of 4 and 7 Us. In the
upper respiratory tract 80% of resistance to airflow was at the nostrils, with
the larynx contributing the other 20%. The upper respiratory tract was
determined to contribute only 30% of total airway resistance. In contrast, in
a later study by Art et al (1988) upper airway resistance, principally that of the
external nares, was considered to account for 82% of total respiratory
resistance. The differences in these two studies may be due in part to

differences in measuring techniques of pleural pressure. The

20

intercostally-placed mushroom catheter used by Robinson et al,(1975) to
measure pleural pressure may result in pleural deformation and altered pressure
values, whereas the indirect pleural pressure measurement via a esophageal
balloon used in the study by Art at al,(1988) does not cause such errors. In
addition, Art at al, positioned their intra-tracheal catheter closer to the thoracic
inlet than that of Robinson et al.; consequently, the measured pressure gradient
between trachea and pleural cavity may have been smaller.

Efforts to quantify upper airway function in exercising horses, began with
the measurement of airflow rates while riding a horse wearing a face mask
attached to a respiratory tube (Hornicke et al,1983). Flow was sensed by a
tube-mounted flag whose deflection was measured electronically. The
electrical signals were transmitted telemetrically and stored on magnetic tape.
The ability to quantify upper airway flow mechanics during exercise, was
further advanced with the development of a technique for measuring intra-
tracheal pressure in the horses during exercise (Mangseth,1984). The
technique involved the percutaneous placement of an intra-tracheal catheter,
which was connected to a differential pressure transducer. Upper airway
pressure changes were defined as the pressure difference between atmospheric
and intra-tracheal pressure. In later studies, this percutaneous intra-tracheal
catheter technique was compared to a nasotracheal catheter in maximally
exercising horses with and without surgically-induced LLH (Williams et
al,1990:l). Pressure measurements obtained using these two catheter systems

were found not to differ significantly. Subsequently, intra-tracheal pressure

21

measurements were obtained before and after surgical treatment of
experimentally induced and naturally occurring upper airway obstructions
(Williams et al,1990:l,ll).

Because variations in flow rates rather than changes in airway geometry
may account for differences in intra-tracheal pressures within and between
horses, studies in which only intra-tracheal pressure is measured should be
interpreted with caution. Therefore, in 1986, Derksen et al. quantified upper
airway impedance in the horse at rest and during treadmill exercise. Airflows
were measured via a mask-pneumotachograph unit. Trans-upper airway
pressure, was the difference between mask pressure and mid-cervical intra-
tracheal pressure. Impedance was determined as the ratio of peak trans-upper
airway pressure and peak flow for a given breath. Subsequently, impedance
values were calculated in horses to evaluate numerous surgical treatments for
induced left laryngeal hemiplegia (Derksen et al,1986; Shappell et al,1988;
Belknap et al,1990; Fulton et al,1991).

Recently, measurement of airflow rates using ultrasound has been
adapted to the exercising horse (Woakes et al,1987). As the
pneumotachograph used by Derksen et al, was of considerable weight, airflow
measurement by ultrasound greatly reduces the weight of the mask worn by
the horse. Variation in airflow rates change the transit time of ultrasound
beams projected diagonally across flow tubes mounted on a mask. The
resulting phase shift of the received signal is detected and produces a voltage

output proportional to flow.

22

lntra-tracheal pressure measurement using a lateral tracheal catheter has
clinical limitations because of the invasive nature of the procedure and
complications such as cellulitis, granulomas and chondritis (Nielan et al,1992).
Although nasotracheal catheter systems may be less invasive, ventilatory
impairment (although'small), is likely to be greater than that seen with a lateral
transtracheal catheter (Nielan et al,1992). in addition to the invasive nature of
tracheal pressure measurement, the time-consuming calculation of impedance
values from pressure and flow tracings is clinically undesirable. For these
reasons, quantitation of upper airway obstructions is rarely used in a clinical
setting.

In human medicine, flow-volume loop analysis allows clinically useful,
quantitative evaluation of upper airway obstructive diseases, and it is possible
that the technique may be clinically useful for quantifying upper airway
obstruction in the horse. Therefore, I chose to document this technique in

horses with surgically-induced LLH at rest and during exercise.

E. Development of Flow-Volume Loop Analysis
1 Currently, the most commonly used human pulmonary function test is
the forced vital capacity (FVC) maneuver. To complete this procedure the
patient inhales to maximal lung volume then immediately exhales as rapidly,
forcefully and completely as possible. This maneuver requires patient
cooperation, and coaxing is often necessary to generate maximal effort. This

test is performed at rest, breathing into a heated pneumotachograph, which is

23

considered the most accurate flow monitoring system (Yeh et al,1987). During
a PVC maneuver, the x-y plot of airflow rate and its' integral with time
(volume), produces a maximum expiratory- and-inspiratory flow-volume curve
(MEFVC and MlFVC, respectively). These 2 curves comprise a maximal effort
flow-volume loop (FVL). Flow-volume loop analysis involves qualitative and
quantitative evaluation of this continuous x-y plot of airflow versus volume
(Hyatt and Black,1973). As a clinical diagnostic test, FVL analysis
approximates the criteria stated by Cochrane and Holland (1971) for the goals
of any diagnostic test: simplicity, patient acceptability, accuracy, repeatability,
sensitivity, specificity, and cost. For this reason it is well established as a
reproducible and sensitive indicator of respiratory disease in man (Becklake and
Permutt,1977).

The use of the FVL, as a pulmonary function test occurred following the
recognition of the functional relationship existing between transpulmonary
pressure, airflow rate and degree of lung inflation (Fry et al,1954; Hyatt et
al,1958l. The relationship between pressure and flow was expressed as a
family of isovolume pressure-flow (PF) curves differing in shape at varied lung
volumes. Figure 2 shows 4 PF curves obtained from initial experiments
performed on healthy people by Hyatt et al (1958). The expiratory flows of PF
curves 1, 2, and 3, measured low in the vital capacity, increase to a maxima,
A, B, and C respectively, and then diminish with increasing transpulmonary
pressure. Therefore, despite increasing effort (increasing transpulmonary

pressure), once a flow maximum for that volume is reached, increasing

24

transpulmonary pressure results in dynamic compression of airways and flow
rate can not be increased (flow limitation).

At lung volumes less than approximately 70% of vital capacity, flow is
effort independent, as increasing transpulmonary pressures do not result in an
increase in airflow. Therefore, the maximal achievable expiratory flow over the
lower half of the vital capacity theoretically bears a constant relationship to
lung inflation, that is uniquely determined by the physical properties of the
lower respiratory tract and does not depend on maximal effort (Fry et al,1958l.
Using the plot of flow maxima at various lung volumes, the information
obtained through measurement of transpulmonary pressure and flow can be
represented as a flow-volume (FV) curve (Figure 25). Representation of a PVC
maneuver as a FV curve, rather than a PV curve, is clinically desirable as the
measurement of pleural pressure required to derive transpulmonary pressure,
involves the invasive placement of an esophageal balloon. In a series of studies
evaluating the factors influencing FV curves at lung volumes less than 70% of
vital capacity, Hyatt et al (1958) and Fry et al (1960), repeatedly showed that
increased extrathoracic resistance had minimal effect on this effort independent
portion of the FV curve. Therefore, this characteristic portion of the MEFVC,
termed the alpha segment, is relatively invariant within an individual and is
therefore a potential indicator of pulmonary disease states.

Over the upper 30% of vital capacity, the relationship between maximal
expiratory flow and degree of lung inflation is influenced by the generated

transpulmonary pressure and therefore the shape of the FV curve is effort

25

 

 

 

 

 

 

 

a b
EXPlR - L/Sec MAX
EplrUSec
.. m D ’6
.. 5) [1
I
- 40 I
c I
PRESSURE " 3° "
cmHzOfl 10 - 20 5
“1.0 " Q A, I
. . . . ll
"'2” o 10 20 so 40
-a.o VOLUME (Uters. from max. expir. point)
INSPlR - USec

Figure 2. a, Relationship of transpulmonary pressure and flow for a normal
subject at differing degrees of lung inflation (curves 1, 2, 3, and 4). b, A plot
of maximum achievable flow against degree of lung inflation. Flow and volume
co-ordinates of points A, B, C, and D from Figure 1a are plotted as closed
circles on this curve. 0 corresponds to maximum expiration point and y to
maximum inspiration point (From Hyatt et al,1958l.

dependent. As seen in Figure12a, the PF curve high in the vital capacity (4)
does not achieve expiratory flow maxima, presumably because the patient
cannot generate sufficient transpulmonary pressure to cause dynamic
compression of larger more rigid airways (Hyatt et al,1958l. The effect of
added extrathoracic airway resistance on the effort dependent portion of the
FV curve has been shown to significantly decrease peak expiratory flow and the
volume of air expired in the first second (FEVm). Therefore, in contrast to the

alpha segment of the MEFVC, this effort-dependent, beta-gamma segment is

26

related to the dimensions and physical properties of the entire airway in a
complicated manner and therefore its' contour is not consistent.

In addition to reflecting the properties of the respiratory tract, the effort
dependent nature of this part of the FV curve is influenced by variation in flow
rates with effort within individuals. Therefore, in order to standardize the
contour and indices describing the FV curve, peak expiratory flow and flow
generated early in the upper half of the MEFVC at high lung volume requires a
maximal expiratory effort to attain a reproducible beta—gamma FV curve (Fry
and Hyatt,1960). In practice, the effort dependent segment acquired from a
FVC maneuver has been shown to be remarkably repeatable and therefore
clinically useful (Fry et al,1960; Hyatt and Black,1973).

Effect of lung disease on MEFVC shape

in man certain pulmonary diseases produce characteristic MEFVC shapes.
The interpretation of the contour of the MEFVC generated, in conjunction with
evaluation of indices quantifying the MEFVC such as the (FEVm), peak
expiratory flow (PEF) and expiratory flow at 50% of lung volume (EFso), have
become the backbone of clinically used screening programs for a plethora of
pulmonary conditions (asthma, emphysema, cystic fibrosis etc.) as well as
providing prognostic measurement of airway function (Hyatt and Black,1973).
Effect of airway obstruction on MlFVC

In contrast to expiratory PF curves, inspiratory PF curves at all lung
volumes increase monotonically with decreasing transpulmonary pressure

(Hyatt et al,1958l. Thus, flow throughout the inspiratory FV curve is effort

27

dependent. The effort dependent portions of the FVL, the inspiratory curve and
the upper one third of the MEFVC, are affected by alveolar pressure (elastic
recoil and pleural pressure) and resistance of the respiratory tract. Therefore,
maximal flow is directly proportional to the driving pressure and inversely
proportional to the resistance (Miller and Hyatt,1969). As inspiratory flow is
effort dependent at all lung volumes the muscle force dissipated to overcome
added airway resistance, renders less force available to overcome normal
airway resistance, and as a result maximal flow decreases (Miller and
Hyatt,1969).

In 1968 Jordanoglou and Pride reported the clinical use of flow-volume
loop analysis to detect an upper airway obstruction. This was the first report
of the use of this non-invasive pulmonary function test in the detection of
changes in upper airway resistance. Subsequent to this, Miller and Hyatt
(1969) studied the pathophysiology of upper airway obstructions and described
how these abnormalities could be reflected by spirometry and FVL. When
pressure and flow rates are high, typically at high lung volumes, a given upper
airway resistance should reduce flow more than at flow rates observed at lower
lung volumes. At lower volumes a greater resistance should be required to
permit detection of reduced flow. These properties were demonstrated by
evaluating normal patients breathing through different sized orifices and in
clinical patients with upper airway lesions. lndices describing the FVL
generated in these patients, afforded detection of upper airway obstructions

(UAO) at high flow rates. In addition, subsequent clinical studies have

28

documented that only severe upper airway obstruction can be detected if low
flow rates are used (Abramson,1982; Amis and Kupershoek,1986).

From these initial studies Miller and Hyatt proposed a classification of
upper airway obstructions (UAO) into two functional groups based on the
pressure-flow relationships around the site of obstruction.

Fixed obstruction - The airway is unable to appreciably change cross-
sectional area in response to transmural pressure differences during either
inspiration or expiration. Oualitatively this lesion would be represented by equal
reduction of both inspiratory and expiratory flow and hence appear as a
flattening of inspiratory and expiratory curves of the FVL (Figure 3).
Quantitative description of this loop shape includes reduced inspiratory and
expiratory flow rates and near normal ratios of expiratory-to-inspiratory flow at
mid-tidal volume. This characteristic pattern is seen in lesions such as ”fixed
high tracheal stenosis" in man.

Variable extra-thoracic obstruction - The airway responds to increased
transmural pressure on inspiration. When intraluminal pressure is negative with
respect to atmospheric pressure, cross-sectional area is reduced resulting in
obstruction to airflow. During expiration, intraluminal pressures are positive
with respect to extraluminal pressures which tends to dilate the airway and
reduce the degree of the airway obstruction. The characteristic appearance of
the FVL is a flattened inspiratory curve, representing a low, fairly constant flow
and a near normal shaped expiratory curve (Figure 3). Ouantitatively, FVL

analysis of this type of lesion is typified by an increase in the ratio of peak

29

expiratory-to~inspiratory flow and ratio of flows at mid-tidal volume, as well as
a reduction in inspiratory flow rates throughout inhalation. This characteriStic
FVL pattern is typically associated with patients with laryngeal paralysis and

laryngeal space occupying lesions (Miller and Hyatt,1973; Kashima,1984).

 

 

Explr Fixed Variable Variable Normal
‘ lesion extrathoracic intrathoracic
p
4 ..
a 2 -
2 Ole
3
& \a-I—J
a ..
4 p
.L
l____: l_._._I L_____I I———I
'"sPlr loo 0

Vital capacity (96)

Figure 3. Representative flow-volume loop shape, showing the effect of fixed,
variable extrathoracic, and intrathoracic airway obstructions, (Hyatt and

Black,1973).

In addition to the 2 defined UAO types, a characteristic FVL pattern
expected to occur with variable intro-thoracic obstruction was proposed. The
expiratory FV curve tends to have a fixed flow over a large portion of the vital
capacity, with the mid-tidal flow being extremely low. The constant expiratory
flow suggests a lesion such as intra-thoracic tracheal tumor which develops an

orifice like configuration shortly after onset of expiration.

30
Quantitative FVL analysis

Since this initial report, FVL analysis has been used extensively in the
clinical diagnosis and evaluation of upper airway obstructions (Miller and
Hyatt,1973; Kashima,1984; Polverino et al,1990l. Peak, mid-tidal-volume
flows and their ratios were initially proposal by Miller and Hyatt (1969) as
useful indicators of airway resistance. Subsequently, several other FVL indices
and their “normal” values have been suggested as being beneficial in the
diagnosis of UAO. Evaluation of FVL from healthy patients and from those
with confirmed fixed UAO, has shown that reduction in flow occurs mainly at
high lung volumes, while the FEVm could be almost normal. The FEV”, is
relatively unaffected by upper airway obstruction as the volume exhaled in 1
second incorporates a whole range of lung volumes many of which are on the
effort-independent part of the FVL. It was therefore suggested that an
increased in FEVm/PEF would be a useful indicator of UAO (Empey,1972).
Topham (1974) also recommended use of this index, particularly in preference
to PEF/PIF. He considered FEV1 _o/PEF advantageous, because of the difficulty
experienced by many patients when performing the forced inspiratory
maneuver, resulting in inadequate inspiratory efforts.

Evaluation of the extrathoracic respiratory tract in conjunction with the
lower airway, provides a simple and rapid assessment of respiratory function
(Carilli et al,1974; Bass,1973). FVL indices permit differentiation of patients
with UAO from patients with chronic obstructive pulmonary disease. Rotman

et al (1975) suggested that when FEV,.o/PEF> 10 and/or EFso/lFso> 1.0, then

31

the possibility of UAO should be further investigated. Numerous other
investigators have published “normal" values, obtained from a relatively small
number of subjects, for various FVL indices useful in diagnosing UAO
(Shim,1972; Bass,1976; Hoffstein,1986). The mid-tidal expiratory-to-
inspiratory flow ratio is considered a sensitive indicator of UAO by many
investigators (Miller et al,1973; Rotman et al,1975; Shim et al,1972;
Kashima,1984). Unfortunately, reported normal values vary from as high as
1.5 (Shim,1972; Haponic et al,1987) to as low as 0.8 (Vincken et al,1987).
Recently, values were reported for FVL indices obtained from a study of a large
number of healthy subjects. By documenting mean values and the distribution
of various FVL indices from a large population, these values may represent a
more accurate account of the degree of variation that exists (Polverino et
al,1990). In that study, the normal EFm/IF50 was reported as 1.3+l— 0.288,
and interestingly, when 'normal" values for various indices were compared to
values obtained from 17 patients with UAO, IFso was found to be the most
consistent indicator of UAO.
Tidal breathing flow-volume loops

The FVC maneuver (recorded as a FVL) as described in man obviously
necessitates patient cooperation for a maximal (voluntary) effort. Children
younger than 6 years of age are usually unable to perform a full FVC maneuver
(Wall et al,1984). Evaluation of pulmonary function in such patients has been
performed using partial expiratory flow-volume curves (PEFVC), generated as

a result of an expiratory effort which is greater than obtained during quiet

32

breathing at rest ,though not originating at total lung capacity (Taussig,1977).
When used in conjunction with measurement of functional residual capacity,
to standardize lung volume, the technique has been reported as a useful clinical
tool in the assessment of lung function in infants and neonates (Buist,1980;
Wall et al,1984). Although PEFVC are considered useful in the detection of
more severe lower airway obstructive diseases such as cystic fibrosis, mild
airway dysfunction may not be accurately detected because of variable flow
rates within healthy children (Wall et al,1984). In addition to evaluation of
lower airway function, the shape and indices of flow-volume leaps generated
from less than maximal respiratory effort have been reported as a useful
indicator of upper airway obstructions in young children (Smith and
Cooper,1981 ).

Use of PEFVC and flow-volume loops in infants involves some degree of
voluntary respiratory effort, and for that reason it is not a test that can be
executed in human neonates and animals because of lack of subject
compliance. For this reason, the technique would seem limited as a diagnostic
tool in veterinary medicine. Despite this fact, flow-volume loops may be
recorded from such non cooperative subjects during quiet breathing at rest with
resultant graphs representing tidal breathing flow-volume loops (T BFVL). Tidal
breathing flow-volume loops have been used to evaluate sleeping neonates
suffering from laryngotracheal disease (Abramson et al,1982). The TBFVL
obtained were useful in detecting upper airway obstruction based on

characteristic shapes, though no correlation between TBFVL indices and clinical

33

degree of airway obstruction was demonstrable. This study suggested that the
low airflow rates associated with tidal breathing limit the diagnostic sensitivity
of the TBFVL. At low flow rates, relatively severe airway obstruction is
necessary to detect flow changes, and the associated low transmural pressures
may be insufficient to cause dynamic collapse of variable extrathoracic lesions.
The significance of the severity and nature of the airway obstruction in
permitting quantitative detection by TBFVL is seen in patients suffering from
sleep apnea, where this severe airway obstruction is associated with
characteristic TBFVL shape and indice changes (Haponik et al,1981; Tammelln
et al,1983). Although TBFVL analysis has some benefit in aiding detection of
airway obstruction in non cooperative patients, the ability of these subjects to
voluntarily increase driving pressure, and thereby increase airflow, may obscure
flow limitation. For these reasons, the lack of standardization of TBFVL makes
strict numerical interpretation difficult (Abramson et al,1982).

In 1986, Amis and Kurpershoek first introduced the TBFVL analysis
technique to clinical veterinary practice. In this initial report TBFVL from normal
dogs (33) were categorized into 4 predominant TBFVL shapes. In the
predominant TBFVL shape (20 dogs), PEF occurred early in expiration , and PIF
occurred late in inspiration. In their evaluation of dogs with fixed upper airway
obstructions and chronic bronchitis, the TBFVL shapes generated by these
conscious dogs were useful in their assessment of airway obstruction.
Significant differences in TBFVL indices from dogs with airway obstruction,

compared to indices obtained in normal dogs were scarce and significant

34

changes tended to be associated with the most severe clinical airway
obstructions. Repeatable quantitative assessment of airway function using
TBFVL analysis was limited by body size, amount of respiratory effort, variation
in breathing pattern and respiratory rate. This variability of TBFVL observed in
normal subjects was documented by the high coefficients of variation for
TBFVL indices (up to 46%).

In a subsequent study, evaluating upper airway function in dogs with
bilateral laryngeal paralysis, 2 TBFVL shape types were reported in the 27 of
30 TBFVL evaluations considered abnormal (Amis et al,1986). All dogs in this
study had stridor or respiratory distress, suggesting moderate to severe airway
obstruction. Although inspiratory airflows were reportedly reduced in many of
the dogs, TBFVL indices consistent with a variable extrathoracic lesion
(decreased inspiratory flows and increased in expiratory-to-inspiratory flow
ratios) were not significantly different from their normal values. This lack of
significance would suggest either large variability and/or insensitivity in
detecting less severe airway obstruction.

Despite the limitations in strict quantitative assessment of airway
function using TBFVL, the technique has been reported as being useful in
evaluating surgical treatment of various upper airway obstructions in dogs
(Smith et al,1986; 1990). In a recent study, TBFVL were used to evaluate
airway function in healthy cats, and cats with bronchial disease. Using the
technique, investigators were able to detect lower airway obstruction

(McKiernan et al,1993). In this study, as with that reported by Amis et al

35

(1986) significant changes in TBFVL indices were associated with moderate to
severe lower airway obstructions. In the past, clinical application of TBFVL
analysis involved laborious, time consuming measurement of TBFVL indices
from either photographs, storage oscilloscopes or x-y plotter graphs. Recently,
TBFVL analysis has become more clinically feasible with the availability of
computer assisted flow-volume analysis programs‘. This program has been
used to evaluate TBFVL from normal dogs with similar results to those
previously reported (Amis and Kurpershoek,1986), but was much easier and
rapid to use (McKiernan et al,1987).
Stimulation of respiratory effort in noncooperative subjects

Although TBFVL are considered of value in non cooperative patients,
methods of improving diagnostic sensitivity by stimulating increased airflow
rates and lung volumes have been investigated. Motoyama (1977) produced
MEFVC in children by applying negative pressure to the airway via an
endotracheal tube, though obviously this technique has clinical limitations, as
subjects need to be either deeply sedated or anesthetized. Other investigators
have measured MEFVC in infants by application of positive pressure to the
external chest wall by use of a chest chamber (Adler and Wohl,1978). In
further attempts to obtain maximal flow curves in newborns and infants, vital
capacity of the crying infant has been measured (Wise et al,1980). In crying
infants, it is difficult to distinguish voluntary vocal cord apposition from airway

obstruction.

 

‘Respiratory Loop Analysis Software, BUXCO Electronics, Inc., Sharon, CT

36

In veterinary medicine the use of respiratory stimulants such as increased
concentrations of inspired carbon dioxide in cats (McKiernan et al,1989) and
horses (Tesarowaski et al,1989) have been investigated. Whiting (1988), used
10% CO, inspired air to stimulate tidal volumes which approached predicted
vital capacities in horses (Davidson et al,1986). however, maximum flows
predicted for adult horses of 65-90 L/sec were not approached. in Whitings'
study, although patient acceptance of the technique was good, coefficients of
variation of FVL indices were approximately twice those reported in human
PEFV studies (Barnes et al,1981), suggesting poor repeatability. Although,
TBFVL indices were significantly increased in all of these studies, their
usefulness and suitability as a routine diagnostic procedure in clinical veterinary
practice remains to be determined.

The effect of exercise-associated ventilatory mechanics and increased
airflow on the TBFVL has been evaluated in normal humans during exhausting
exercise (Olafsson and Hyatt,1969; Babb et al,1991). The TBFVL, from
healthy non-athletic people, obtained during strenuous exercise impinges on
their FVL only near and expiration. in contrast, patients with obstructive lung
disease breathe along or outside their MEFVC during exercise (Grimby and
Stiksa,1970). In well trained athletic young men, TBFVL obtained during sub-
maximal exercise generally do not approach their FVL (Grimby et al,1971). In
contrast, during maximal exercise in the same subjects airflows are near-
maximal, and flow approaches much of their MEFVC, and tidal volumes are

approximately 50% of vital capacity.

37

In exercising horses, airflow rate is greatly increased whereas variability
in breathing strategy is reduced (Hornicke et al,1983; 1987). In 1988, Art et
al, evaluated the sub-maximal exercise-induced changes in breathing pattern in
ponies by analyzing TBFVL. This study reported that TBFVL were variable
between ponies though relatively constant within individuals. The TBFVL
indices which significantly changed in response to exercise were poorly
correlated with pulmonary resistance. Based on these findings the authors
questioned the diagnostic usefulness of exercise associated TBFVL in
pulmonary function testing. However, the relatively low airflow rates reported
(23.56 +/- 1.26 Us) in Art's study during submaximal exercise, are likely to
limit the sensitivity of the test in assessing airway function under those
exercise conditions.

The sensitivity of TBFVL in detecting airway obstruction may improve as
airflow rates approach maximum (Abramson et al,1982). and the usefulness of
TBFVL in detecting airway obstruction in non—cooperative patients, like horses,
may be enhanced during more strenuous treadmill exercise. Furthermore, it is
possible, that if a ventilatory response to exercise in performance horses is
similar to that reported in well trained young men, then TBFVL obtained during
maximal exercise may approach their FVL. The increased sensitivity of TBFVL
obtained during exercise may therefore allow detection of less than severe

upper airway obstruction.

lli. EQUINE LEFT LARYNGEAL HEMIPLEGIA - A REVIEW

A. Anatomy and Physiology of the Larynx

The equine larynx consists of a framework of cartilages and associated
tissues which provide a conduit for the flow of air from the pharynx to the
trachea. The larynx contains 3 paired cartilages (cuneiform, corniculate and
arytenoid) and 3 unpaired cartilages (cricoid, thyroid and epiglottis).

The pyramidally shaped arytenoid cartilages are located rostral to the
cricoid cartilage and medial to the laminar portion of the thyroid cartilage. Each
arytenoid cartilage consists of a caudo-dorsal base and a rostrally located apex.
Caudally, each arytenoid cartilage has a synovial articulation with the cricoid
cartilage, and rostrally the apex fuses with the corniculate cartilage. Lateral
and rostral to the base is a large muscular process, while the vocal process is
situated ventrally (Getty,1975). The arytenoid cartilages, and the epiglottis,
have the greatest range of motion of the laryngeal cartilages. The cuneiform
cartilages represent two caudal extensions from the epiglottis which are
situated just rostral to the thyroid cartilage.

Broad fibre-elastic ligaments exist between the cricoid cartilage and the
first tracheal ring, and between the cricoid and the thyroid cartilage. The
transverse arytenoid ligament is a narrow band that connects the dorsomedial
angles of the opposing arytenoid cartilages. The vestibular ligament extends
from the cuneiform cartilages and lateral aspect of the epiglottis, to the ventral
aspect of the arytenoid cartilages. The vocalis ligament spans the distance

between the vocal process of the arytenoid cartilage to the caudal border of the

38

39

body of the thyroid cartilage. The mucous membrane lining the vocalis
ligament and muscle (vocal folds), and medial surface of the arytenoid cartilage,
outlines the narrowest cross sectional area of the laryngeal airway; the rima
glottidis. The lateral ventricles constitute an infolding of mucous membrane
between the vocal fold and the vestibular fold (covering the vestibularis muscle
and ligament). The aryepiglottic folds consist of mucous membrane gathered
into discrete folds extending from the caudo-lateral aspect of the epiglottis to
the corniculate and arytenoid cartilages. The folds are continuous on either
side of the epiglottis ventral to the epiglottis and are contiguous with the
glossoepiglottic fold rostrally.

The extrinsic muscles of the larynx; the thyrohyoideus, hyoepiglottis,
sternothyroideus and omohyoideus muscles, move the whole larynx as a unit.
The hypoglossal nerve (cranial nerve XII) is the efferent nerve to the
thyrohyoideus and hyoepiglottis muscles, whereas the ventral branches of the
first and second cervical nerve innervate the sternothyrohyoideus and
omohyoideus.

The intrinsic laryngeal muscles, which originate from the muscular
process (except the vocalis, cricothyroideus and tensor ventriculi lateralis mm.),
move individual cartilages in relation to each other. The cricothyroideus muscle
tenses the vocal ligaments by moving the thyroid cartilage caudally, causing
adduction of the vocal folds and increased dorsoventral diameter of the rima
glottidis. Adduction of the vocal process of the arytenoid cartilage and vocal

fold results from contraction of thyroarytenoideus (vocalis and vestibularis

40

components), arytenoideus transversus and cricoartenoideus lateralis muscles.
The cricoarytenoideus dorsalis (CAD) muscles cause abduction and dorsal
displacement of the arytenoid cartilage and vocal fold. The tensor ventriculi
lateralis also causes abduction of the vocal fold.

The innervation of the intrinsic muscles, except for the cricothyroideus
muscle, is via the recurrent laryngeal nerves (RLN). Unlike the dog and man,
there is no passage of motor nerve fibers from one side of the larynx to the
other (Quinlan et al,1982). The cell bodies of the RLN are located in the
nucleus ambiguus of the medulla oblongata. Nerve fibers of the RLN travel
within the vagus nerve then branch from the right vagus nerve at the level of
the first or second rib, and from the left vagus nerve at the aortic arch. After
passing over the ligamentum arteriosum, the left RLN then ascends to the
larynx along the ventral surface of the common carotid artery within the carotid
sheath, as does the right RLN. Only the cricothyroideus muscle is innervated
by the cranial laryngeal nerve, which also is sensory to the mucous membrane
of the larynx.

In addition to providing a patent airway for respiration, the complex
structure of the larynx is involved in swallowing, phonation, coughing,
explosive efforts (straining) and airway protection from aspiration
(Bartlett,1989). Laryngeal movements are coordinated in the central nervous
system during respiration. Laryngeal response to respiratory demands is
regulated through afferent nervous input and numerous reflexes involving the

entire respiratory tract. Central stimulus of the RLN, supplying intrinsic

41

laryngeal muscles, results in abduction of the arytenoid cartilage and vocal fold
during inspiration and relative adduction during expiration. Abduction of the
vocal fold occurs prior to contraction of the diaphragm, thereby ensuring
minimal resistance to airflow at the larynx during inspiration. In addition to the
CAD muscle, contraction of the cricothyroideus muscle may also aid abduction
(Suzuki and Kirchner,1969). Decreased motor tone is observed in the RLN
during expiration and therefore, the decreased CAD muscle activity is thought
responsible for the relative adduction of the vocal folds. This adduction of the
vocal folds may play a role in the control of respiratory frequency, by slowing
expiratory airflow (England at al,1982).

The controlled movement of the vocal folds by the respiratory center
may be influenced by reflexes arising from receptors sensitive to changes in pH
of cerebrospinal fluid, stretch receptors in the lung and chemoreceptors in the
carotid sinus. Physiological states, such as hypercapnia which may cause
prolonged vocal fold abduction, alter laryngeal function (Bartlett,1989).
Increased RLN activity associated with abduction of vocal folds and arytenoid
cartilage in response to increased resistance to airflow has also been reported
(Glogowska et al,1974).

Receptors in the airway epithelium, including the laryngeal epithelium
may alter the cross sectional area of the rima glottidis. Stimulation of irritant
receptors in the laryngeal mucosa causes adduction of the vocal folds and
therefore increased laryngeal resistance to airflow (Stransky et al,1973).

Laryngeal stretch receptors have also been described (Bartlett,1989). The

42

reflex activity of the larynx is also associated with changes in respiratory
pattern and depth of breathing. Thus, afferent laryngeal nerves are involved

with the control of ventilation (Bartlett,1989l.

8. Incidence of LLH

As a cause of wastage in athletic horses, respiratory disorders are
second only to lameness (Rossdale et al,1985). Although the exact prevalence
of upper airway obstructions, as performance limiting diseases, is difficult to
determine due to their varied clinical manifestations, LLH has been reported as
the next most common disease of the respiratory system after respiratory
infections (Cook,1970). The disease, LLH, has been considered to range from
a subclinical hemiparesis to hemiplegia associated with overt clinical signs.
Therefore, the reported incidence of LLH will depend on the manifestation of
the disease and the diagnostic criteria used. A subclinical form of LLH has
been reported to occur in as many as 77% of clinically normal horses greater
than 14.2 hands, and recently a 95% incidence was reported from evaluation
of a large number of Thoroughbreds, based on palpation of the atrophied CAD
muscles (Duncan et al,1974: Cook,1988).

Based on endoscopic evaluation in the resting horse, the prevalence of
the clinical form of laryngeal hemiplegia has been reported to range from 1.3%
to 8% (Pascoe et al,1981; Raphel,1982; Baker,1983:ll; Hillidge,1985;
Sweeney et al,1991). Breeds recognized as being commonly affected with LLH

are the Thoroughbred, Standardbred, Quarterhorse and draught type horses,

43

though the observed breed distribution may be skewed by the athletic use of
such breeds. The incidence of LLH in draught horses has been reported from
9.0% and 22% (Goulden et al,1985; Archer et al,1989).

Many of the reported surveys were conducted on Thoroughbreds in race
training, except for Hillidge's study who reported an 8% incidence. Therefore,
the incidence of LLH reported by others (1 .3-4.7%l. may be artificially low, as
horses are likely to be culled at the earliest signs of a performance limiting
disease. in a study of horses presented for poor performance, endoscopic
examination of the larynx during high speed treadmill exercise, identified LLH
in 36% of horses that had an upper airway abnormality detected (Morris and
Seeherman,1991).

In a number of surveys, entire and castrated males are found to be more
commonly effected by LLH than females (Cook,1970; Goulden and Anderson,
1981 :I). Goulden and Anderson (1981zl), reported 6 times as many
Thoroughbred males to be suffering from LLH than females. In addition, all
horses diagnosed with LLH were shown to be significantly (P<0.05) heavier
and taller at the withers than unaffected animals. Although foals and geriatric
horses have been observed with LLH (Hillidge,1985), the condition is more
commonly diagnosed in Thoroughbred horses between 1 and 5 years of age,
and in draught horses at 8 years (Archer et al,1989). In contrast to the
findings cited above, using percutaneous palpation of laryngeal musculature as
the criteria for a diagnosis of LLH, a poor correlation was found between LLH

and large horses and males (Cook,1988). These findings, although

44

controversial, primarily serve to emphasize the lack of agreement between
investigators, regarding the disease syndrome being described and the method

of diagnosis.

C. Pathogenesis
Nerves

The major neural lesion in LLH is suggestive of a progressive axonal
degeneration in the face of attempted repair. The disease selectively effects
large myelinated nerve fibers in the distal regions of the left, and to a lesser
extent the right recurrent laryngeal nerves (Duncan et al,1974: Cahill and
Goulden,1986:l,ll). There is loss of large diameter myelinated fibers distally,
in conjunction with regeneration of axonal clusters, thinly myelinated axons and
I'onion bulbs”, particularly in the middle and proximal portions of the nerve.
These characteristic neural changes, referred to as a distal axonopathy, were
originally described by Cole (1946), though it wasn't until later, after Duncan
et al,(1974) confirmed these findings that this terminology became recognized.
Muscles

The intrinsic laryngeal muscle lesions, in clinical and sub-clinical LLH, are
typical of chronic neurogenic atrophy (Cole,1946; Duncan et al,1974). Fiber
type grouping, evidence of denervation and reinnervation by separate nerve
fibers, has been reported in all muscles supplied by the RLN (Cahill and
Goulden, 1986:lll). Fiber type grouping occurs when denervated muscle fibers

in the vicinity of a surviving nerve fiber are reinnervated by collateral sprouts

45

from this fiber, and thus assume the same metabolic type, as determined by the
innervating nerve fiber. Other changes consistent with neurogenic atrophy
include the presence of atrophic and hypertrophic muscle fibers (Duncan et
al,1974). variation in muscle fibre size, and degenerate muscle fascicles being
replaced by connective tissue (Cole,1946).

Of the intrinsic laryngeal muscles, the earliest and most pronounced
atrophy occurs in the cricoarytenoideus lateralis (CAL), the main adductor
(Duncan and Griffiths 1973; Duncan et al,1974: Cahill and Goulden,1986:lll).
The preferential denervation of the adductor muscles, CAL and transverse
arytenoid muscles, has recently been quantified (Duncan et al,1991:l).
Presently, the reason for the observed preferential adductor muscle denervation
is unknown. Previously, it has been suggested that the adductor muscles may
be innervated by a greater percentage of large diameter nerve fibers, which are
more severely affected in LLH (Duncan and Griffiths,1973). Recent
morphometric studies of the adductor and abductor branch of the RLN, showed
no significant difference between the population of large diameter fibers in
these nerves (Duncan et al,1991:ll). It has also been proposed that preferential
adductor muscle denervation is due to the fibers innervating the CAL muscles
being more distal than that of the CAD muscle.

Neurogenic atrophy, although less severe, is also commonly observed in
the right intrinsic laryngeal muscles (Cole,1946), as well as the extensor
digitorum Iongus muscle (Cahill and Goulden 1986:lll) in horses with LLH,

supporting the view that the disease is a generalized peripheral neuropathy.

46

Although mild degenerative changes in the more distal parts of longer limb
nerves have been reported, their clinical importance is questionable, as they
may represent age related changes rather than that of a generalized disease
state (Wheeler and Plummer,1989). Therefore, whether the pathology
associated with LLH is a generalized process or not is unclear.

Despite the characteristic neuromuscular findings in clinical cases of LLH,
it is now well established that many horses may have histologically apparent
neuromuscular changes in the absence of clinical LLH. In one study, where
laryngeal asynchrony and trembling, but not paralysis was observed
endoscopically in 6 horses, neuromuscular changes were found, similar to those
reported in clinical cases of LLH (Duncan et al,1977). This study suggested
these endoscopic findings represent the early signs of a progressive lesion
which may develop into left sided laryngeal hemiplegia. This subclinical or early
stage of LLH has been termed “recurrent laryngeal neuropathy”, as opposed to
laryngeal hemiplegia, which is considered by Cook to more accurately describe
a total inability to abduct the arytenoid cartilage (Cook,1988). in contrast to
the conclusions drawn from Duncan et al'sl1977) study, results from a large
number of endoscopic examinations in horses over a 5 year period suggested
that the observed variations in symmetry and synchrony were a normal finding
(Baker, 1 983zl). Baker (1983zl) reported that none of the 168 horses exhibiting
asynchronous laryngeal activity, progressed to clinical disease, and questioned

whether these endoscopic findings were a prelude to the clinical state of LLH.

47
Cartilages

Little attention has been paid to the effect of LLH on the cartilages of the
larynx. In a study were the RLN was sectioned in 30 horses, bony ankylosis
of the crico-arytenoid joint was reported to develop in 10 horses within 1 year
(Denecke in Elies et al,1983). These changes may be of significance when
considering the prognosis following prosthetic laryngoplasty in long standing
cases of LLH.

ETIOLOGY

The etiology of the neuromuscular pathological findings in LLH is most
commonly idiopathic, hence the common use of the term “idiopathic laryngeal
hemiplegia" in place of LLH. The involvement of both right and left RLN make
focal injury or compression unlikely as major factors. If LLH is part of a
generalized neuropathy, possibly involvement of the left RLN may be
exacerbated by, as yet undefined, “local factors“ (Griffiths,1991).

Injury

In a small number of cases, isolated incidents of known injury have been
reported to result in laryngeal hemiplegia, including cervical lacerations and
blunt trauma (Gilbert,1972 and Goulden and Anderson, 1981:"). Known
causes of recurrent laryngeal nerve injury also include, injection of perivascular
irritants and a retropharyngeal abscess caused by W (Barber, 1 981 ).

Mechanical causes of left RLN injury have been proposed based on the
unique anatomy of the RLNs. A flattening of the left RLN, reported in some

horses, as it passes around the aorta was thought associated with the site of

48
nerve damage (Argyle in Mason,1973). The lack of pathology at this site and

the similar flattening of the right RLN observed by Cahill and Goulden (1986zl),
do not support this theory.

In 1970, Rooney suggested the anatomical path of the left RLN,
compared to the right RLN results in greater tensile forces experienced by the
left RLN when the neck is stretched. In a recent report, head position during
general anesthesia was postulated to result in bilateral laryngeal paralysis after
general anesthesia (Abrahamson et al,1990). Despite the theory that this
periodic stretching would result in ischemia and nerve damage, peripheral
nerves can be stretched up to 8% of their length before there is vascular
compromise (Lundborg,1988). Further evidence questioning Rooney's theory
is that microscopic lesions in the left RLN of horses with LLH are not typical of
ischemic nerve injury (Duncan and Hammang,1987).

Central nervous system lesions

Focal CNS lesions have been reported in horses with LLH. Significantly
higher numbers of axonal spheroids (areas of degenerating and regenerating
axons), have been described in the lateral cuneate nuclei of horses with LLH
compared to horses unaffected by the disease. Despite the difference observed
between these 2 groups, the increased frequency of axonal spheroids may be
an age related process, and therefore their significance remains uncertain (Cahill

and Goulden,1986:lV; Cahill and Goulden,1989).

49

Causes of peripheral neuropathy

Polyneuropathies with a distal distribution can have a variety of causes
including inherited, toxic, nutritional deficiencies, and metabolic disorders.
Despite the fact that in the majority of cases the etiology of LLH is unknown,
isolated reports of laryngeal hemiplegia resulting from poisonings such as
organophosphates (Rose et al,1981; Duncan and Brook, 1985), lead
(Burrows,1982l and plant toxicities from Lathyrus, alfalfa, and false dandelion
(Cahill and Goulden,1987; Huntington et al,1989) exist. Thiamine (vitamin B1)
deficiency was proposed as a possible cause of LLH, based on a similar
association known to occur in people (Loew,1973). Despite the fact that low
thiamine levels have been recorded in horses with LLH, the disease has not
been induced by a thiamine deficient diet (Cymbaluk et al,1977). Other
suggested causes have included various viral and bacterial respiratory tract
infections (Goulden and Anderson,1981zlll.
Genetics

The possibility of an inherited basis for LLH has substantial evidence.
Clinical LLH (Cook,1988l and fiber type grouping (Gunn,1973) in laryngeal
muscles of foals has been reported. Clearly, the demonstration of the disease
in very young foals or even near-term fetuses (Gunn,1973), would be indicative
of a congenital or hereditary association. Cook postulated that a single
recessive gene was responsible for LLH (Cook,1981:ll). To date, no recorded
matings between horses with clinical LLH have produced 100% affected foals

(Hillidge, 1985). The inheritance of phenotype was thought to be the only

5O

genetic component to LLH since affected horses tended to be taller and heavier
(Marks et al,1970:l).

In a recent report, 47 offspring of an affected stallion were compared
with a similar number of controls. A significantly greater number (11) of the
stallions offspring had LLH compared to the controls (Poncet et al,1989). In
this study a dominant gene was hypothesized as explaining the high occurrence
rate, rather than the high incidence having a congenital basis. The findings
from such organized breeding studies should be viewed with caution, as the
widespread nature of LLH, make selection of an adequate control population

difficult.

D. Diagnosis
Clinical Signs

Decreased ability to abduct the arytenoid and corniculate cartilages
during exercise causes inspiratory dyspnea, reduced tolerance for sustained
high intensity exercise, and a characteristic stridor or "roaring” sound during
exercise. Loss of adductor tone may result in a voice of lower than normal
pitch. Typically these clinical signs gradually increase in severity over a period
of weeks to months. A diagnosis of LLH requires consideration of the animals
signalment (breed and height), history (previous performance, race times),
careful physical examination and an endoscopic examination of the larynx

(Baker,1983:ll).

51
Physical Examination

A general physical examination is particularly important to identify other
potential performance limiting conditions such as lameness and lower airway
disease. The presence of fibrous scars in the skin from previous laryngeal
surgery should be ascertained, either ventral to the larynx or immediately
ventral to the linguofacial vein. Percutaneous laryngeal palpation may
demonstrate evidence of CAD muscle atrophy, by a more prominent muscular
process of the left arytenoid cartilage compared to the right. Although such
atrophy is pathognomonic for LLH, experience is necessary for the examiner to
be comfortable with such findings. An “arytenoid depression test" may also
be performed. This test involves manual depression of the muscular process
in a medial, rostral, and ventral direction which may reveal reduced abductor
tone and inspiratory stridor in horses with LLH (Marks et al,1970:l).

Tests of adductor muscle function

Knowledge that adductor muscle dysfunction precedes abductor muscle
dysfunction (Duncan et al,1974). potentially provides an avenue for tests that
may allow early detection of LLH. In the “grunt test“, the horse attempts to
close its larynx in response to having its abdomen or thorax threatened. In
affected horses this is associated with a ”grunt” as air is able to escape
through a partially occluded rima glottidis (Cook,1965). The ”slap test",
involves 3 or 4 sharp slaps on either side of the withers to evoke an adductory
movement from the contralateral arytenoid (Greet et al,1980). A normal

response requires that the reflex arc terminating in the recurrent laryngeal nerve

52

be intact. Although it may be useful to verify laryngeal hemiplegia, it does not
allow differentiation of degrees of paresis, it becomes less responsive with
repeated stimulation, and is absent in tense or frightened horses and in horses
with cervical spinal cord lesions. Generally adductor function tests are
considered less reliable than abductor function evaluation via endoscopic
examination of the larynx.

Recently, objective evaluation of the integrity of the reflex are associated
with the slap test has been attempted by the use of measured latency times of
the left and right reflex arcs (Cook and Thalhammer,1991). The validity of this
measurement in detecting and grading recurrent laryngeal neuropathy remains
unproven. Previously, electromyographic studies have been shown to detect
CAD muscle paralysis (Moore et al,1988). However, this diagnostic technique
has limited ability to discern grades of LLH and appears impractical because of
its invasive nature, cost and operator skill required.

Endoscopy

Although history and laryngeal palpation are useful in attaining a
diagnosis of LLH, definitive diagnosis and identification of other potential upper
airway causes of stridor is accomplished by use of endoscopy (Cook,1965l.
Endoscopic evaluation of the larynx is presently regarded as the most accurate
and specific diagnostic test for LLH. Careful endoscopic evaluation of laryngeal
form and function facilitates differentiation of LLH from other upper airway
conditions associated with exercise intolerance and respiratory noise during

exercise. These include arytenoid chondropathy (Haynes et al,1980).

53

ossification of the arytenoid cartilages (Shapiro et al,1979), epiglottic
entrapment (Boles et al,1978), dorsal displacement of the soft palate (Cook
1974) and subepiglottic cysts (Raker,1976). Further diagnostic assistance with
differentiation between these upper airway obstructions may be obtained from
pharyngeal radiography (Linford et al,1983; Orsini et al,1989).

The endoscopic appearance of the larynx may vary with the degree of
abductor muscle atrophy and loss of function (Duncan et al,1977). With
hemiplegia, there is obvious asymmetry of the rima glottidis, failure of the
arytenoid and corniculate cartilage to abduct normally, a kinking of the
aryepiglottic fold and dilation of the laryngeal ventricle on the left side
(Cook,1965). Evaluation of abductor muscle function may be enhanced by
stimulation of swallowing, nasal occlusion, exercise and systemic
administration of doxapram hydrochloride (Marks et al,1970:l, Baker 1983:",
Lane,1987). The repeatability of some of these techniques has recently been
evaluated, using measurement of rima glottidis cross sectional area and multiple
independent observers. Nasal occlusion was considered the most simple and
reliable method of inducing arytenoid abduction (Archer et al,1991). Although
in the clinical setting, induction of swallowing is currently considered the 'acid
test“ in post-sale examinations of laryngeal function.

The significance of tremor, asynchronous movement, and reduced
abductor activity of the arytenoid cartilages as a cause of upper airway
obstruction remains a controversial issue. The prevalence of asynchronous

movement of the arytenoids has lead to it's incorporation into a grading scheme

54

for laryngoscopic examinations. Recently a grading scheme has been proposed

for clinical use (Rakestraw et al,1991), based on endoscopic classifications

previously reported (Goulden et al,1985).

Grade I.

Grade ll.

Grade III.

Grade IV.

Synchronous full abduction and adduction of the left and
right arytenoid cartilages

Asynchronous movement of the left arytenoid cartilage
during any phase of respiration. Full abduction of the left
arytenoid cartilage (as compared to the right) inducible by
nasal occlusion or swallowing.

Asynchronous movement of the left arytenoid cartilage
during any phase of respiration. Full abduction of the left
arytenoid cartilage cannot be induced and maintained by
nasal occlusion or swallowing.

Marked asymmetry of the larynx at rest and no substantial
movement of the left arytenoid cartilage during any phase

of respiration.

Several commonly used practices may influence the appearance of the

larynx. These include sedation, use of a twitch and endoscopy through the

right or left nostril. In 1975, Robinson and Sorenson reported altered upper

airway function following xylazine administration in the resting horse, and

recommended that upper airway endoscopy should be performed in the

unsedated horse. Despite their findings, Cook (1988) has recommended the

use of tranquilizers to facilitate examination of the larynx. Recently, some of

55

the potential factors influencing laryngoscopic appearance were evaluated by
multiple experienced observers, using the above cited laryngeal function grades,
in conjunction with measurement of the cross sectional area of the rima
glottidis. Sedation with xylazine, use of the alternate nostril and day of
examination had a statistically significant influence on the laryngeal score and
rima glottidis measurement (Archer et al,1991; Ducharme et al,1991; Hackett
et al,1991). These studies suggest that the most reliable evaluation of
laryngeal function at rest is obtained in the twitched, unsedated horse, viewing
the larynx via the left nasal passage and stimulating arytenoid cartilage
abduction by nasal occlusion.
Endoscopy During Exercise

In some horses the larynx may appear normal at rest but during and
immediately after exercise, signs of LLH may be present. Strenuous exercise,
sufficient to reveal exercise intolerance and respiratory noise as described by
the owner, immediately followed by endoscopic examination may demonstrate
incomplete arytenoid abduction (Lane,1987). This technique may be useful in
detecting LLH in horses which may not have typical signs of LLH at rest
(recurrent laryngeal neuropathy). Despite this, examination in the resting
horses immediately after exercise is of limited value in detecting the dynamic
nature of LLH, as respiratory airflow rates rapidly return toward normal resting
values. The sensitivity of laryngoscopy has recently been enhanced through
videoendoscopic examination of larynx during high speed treadmill exercise

(Derksen,1988; Morris and Seeherman,1988).

56

During exercise, high airflow rates and negative upper airway pressures
compared to atmosphere, may result in axial displacement or "dynamic
collapse“ of the left corniculate cartilage across the airway in horses with LLH
(Robinson and Sorenson,1978; Derksen,1988l. Recently, the graded
endoscopic appearance of horses larynges at rest where compared to
videoendoscopic evaluation during strenuous treadmill exercise (Rakestraw et
al,1991). All grade I and II horses were able to fully abduct the left arytenoid
cartilage during exercise, and all grade IV horses showed evidence of airway
closure, ”dynamic collapse", during inspiration. However, in horses with grade
III resting laryngeal function, videoendoscopic findings during exercise could not
be predicted. This suggests that endoscopic evaluation of resting horses with
grade III LLH is an unreliable method for assessing laryngeal function during
exercise.

The controversy surrounding the interpretation and clinical significance
of endoscopic findings at rest, and to a lesser extent during exercise, is
confounded by the fact that evaluation is based entirely on subjective
impressions. In the last few years there has been an increasing effort to adapt
quantitative measurements of upper airway function, used at an experimental
level, to enable their use in a clinical setting. Such objective evaluations of the
upper airway include measurements of upper airway impedance (Derksen et
al,1986) and measurement of rima glottidis cross sectional area (Martin et

aL1983L

57

Recently, tracheal pressure measurements from exercising horses with
LLH have been used as an indicator of upper airway function in the diagnosis
and surgical treatment in a clinical setting (Williams et al,1990:l). Similarly,
quantitative evaluation of the rima glottidis has been determined in a limited
number of clinical cases, and may have potential as a indicator of airway
obstruction during exercise (Rakeshaw et al,1991). Upper airway impedance
measurements document airway function more thoroughly than these
techniques, as discussed in Chapter I. Although it has been used clinically on
a single horse (Stick and Derksen,1989l. the technique has clinical limitations,
which were discussed in the Chapter II (E).

In chapter V, I will show for the first time that TBFVL analysis in
exercising horses may prove to be a clinically useful, sensitive, and quantitative
test of upper airway function. In the present chapter, I reviewed the incidence,
pathogenesis, etiology and diagnosis of LLH. In the next chapter I will review

the surgical treatments of this condition.

IV. TREATMENT OF LEFT LARYNGEAL HEMIPLEGIA - A REVIEW

As early as 1664 horsemen reported horses making an abnormal
“roaring” sound during exercise, which was associated with a diminished
tolerance for work (in MacQueen,1896l. The relationship between ”roaring"
and left laryngeal paralysis was reported by Dupuy in 1807 and subsequently
confirmed in 1815 after he induced 'roaring" by transecting the recurrent
laryngeal nerve (in MacQueen,1896l.

Soon after the recognition of the association between the degeneration
of the abductor muscle of the larynx and the commonly observed clinical signs
of LLH, the search for a treatment was undertaken. Fleming stated that
”roaring” and reduced exercise tolerance resulted from the ventricle filling with
air and billowing into the airway and upon severe exercise the arytenoid
cartilage also contributing to airway narrowing (Fleming,1882). The basic
principle of all the techniques attempted for treatment of LLH is to relieve the
airway obstruction, by either by—passing the larynx or removal or stabilization
of the lateral ventricle, vocal fold, and corniculate and arytenoid cartilages.

For horses that are only required to perform minimal exercise or race for
short distances, surgical intervention with its associated cost, potential
complications and predicted success rate, may be unnecessary. However,

effective surgical intervention is necessary for horses to exercise maximally.

58

59

A. Ventriculectomy

Ventriculectomy or sacculectomy involves the removal of the mucosal
lining of the lateral ventricle. A form of ventriculectomy was first described by
Gunther Junior in 1866 (in Williams,1911). The goal of this procedure is to
induce a fibrous scar between the vocal fold, thyroid and arytenoid cartilages
and thereby abduct and stabilize the vocal fold and arytenoid cartilage. The
modern version of ventriculectomy was initially used by Williams in North
America in 1906 (Williams,1911), and subsequently popularized by Hobday in
Great Britain. As the lateral ventricle is considered the primary site of
respiratory noise during exercise in horses with LLH (Attenburrow et al,1983).
it has considerable value in reducing and modifying exercise related stridor.
Although the procedure can be performed in the standing horse, it is usually
performed with the horse in dorsal recumbency (Haynes,1984). The lateral
ventricle is exposed through a Iaryngotomy created by a midline incision
through the skin and subcutaneous tissues. Subsequently, the paired
sternothyrohyoideus muscles are separated, and the entire length of the
cricothyroid ligament is incised. Eversion of the ventricle may be accomplished
using either a toothed burr or traction with forceps. After excision of the
everted membrane the edges of the ventricle may be sutured (Pouret,1966) and
the vocal cord removed (Baker,1983:ll), though no extra benefit is considered
to be gained from these additional procedures (Haynes,1984). Recently
ablation of the lateral ventricle by use of transendoscopic laser produced

satisfactory results, although thermal damage to structures adjacent the treated

6O

ventricle were reported (Shires et al,1990). The advantages of the laser
technique are that it obviates the need for general anesthesia and Iaryngotomy.

The Iaryngotomy wound is usually left to heal by secondary intention.
Primary closure is possible, and it may be associated with more rapid healing,
and decreased discharge from the wound. lf primary closure is attempted, a
meticulous airtight seal of the cricothyroid membrane and thorough wound
lavage is paramount in preventing the development of subcutaneous edema and
wound abscessation. Complications following ventriculectomy are rare, though
may include, inflammatory polyps, laryngeal edema, chondritis and wound
infections. Post-operatively, horses are stall rested and observed closely for 48
hours for signs of respiratory distress associated with laryngeal edema.
Healing is generally complete within 21-30 days and training generally should
resume after 45 days of rest.

Early reports of the use of ventriculectomy for treatment of LLH,
document restoration of athletic endeavors and reduction in respiratory noise
in 85-95% and 20-71% of horses, respectively (Williams, 1 91 1; Hobday,1935l.
Subjective evaluations of the ability of ventriculectomy to successfully reduce
stridor during exercise, and/or improve tolerance to exercise range from 28%
to 99% (Baker,1983:ll; Pouret,1966l. This large range reported reflects
variation between authors as to the criteria determining success or failure.
Recently, a successful outcome, defined as a reduction in inspiratory noise and
improved exercise tolerance, was reported in 82% of draft horses with LLH

treated by ventriculectomy (Bohannon et al,1990). From this study it may be

61

concluded, that ventriculectomy may be sufficient to allow sustained low level
exercise.

To date only 1 study has objectively evaluated the efficacy of
ventriculectomy. Shappell et al (1988) evaluated the effect of ventriculectomy
in horses with surgical-induced LLH using measurements of airflow and airway
pressure to evaluate airflow mechanics. This study found that unilateral
ventriculectomy did not reduce resistance to airflow in these horses at either
sub-maximal or near maximal exercise levels. Others have also deemed
ventriculectomy an unjustifiable procedure on its own in horses with LLH
(Speirs,1987; Cook,1988l. Despite the findings of Shappell et al (1988), many
surgeons treat equine athletes suffering from LLH by ventriculectomy in

conjunction with prosthetic laryngoplasty.

B. Prosthetic Laryngoplasty

Based on published literature to date, their appears to be a consensus
that, when performed by experienced surgeons, prosthetic laryngoplasty (PL)
is currently the best available method for treating horses suffering from LLH.
The objective of PL, or abductor muscle prosthesis, is to abduct and stabilize
the affected vocal fold, and corniculate and arytenoid cartilage. The first
detailed report of the use of this procedure, whereby a prosthetic suture is
placed to mimic the abductor function of the CAD muscle, appeared in 1970

(Marks et al,1970:ll).

62

For prosthesis placement, the horse is positioned in lateral recumbency,
under general anesthesia, with the affected side uppermost. An 8 cm curvi-
linear incision is made from the level of the cricotracheal space extending
rostrally along the ventral border of the linguofacial vein. The plane of
dissection is continued between the omohyoideus muscle and the linguofacial
vein to the lateral wall of the larynx. The crico- and thyropharyngeus muscles
are then separated along their septum. After identification of the cricoid
cartilage, a cutting curved or half circle needle is passed from the caudal border
of the cricoid rostrally about 1.0-1.5 cm. The needle should be directed
submucosally and penetrate the dorsal lamina as close to the median ridge of
the cricoid cartilage as possible. Generally the prosthesis consists of No.2
nonabsorbable material, either single or double strand. The suture ends are
then passed under the cricopharyngeus muscle emerging at the septum
between the crico-and-thyropharyngeus muscles. Using a 16 gauge needle 3
hole is drilled through the muscular process of the arytenoid cartilage and the
suture and arising from the most cranial aspect of the cricoid cartilage is pulled
through with a #10 crochet hook from medial to lateral. The site of suture
placement should be through the cranial half of the base of the muscular
process, rather than at the apex. Accurate placement through the dorsal lamina
of the cricoid and muscular process of the arytenoid cartilage is essential. The
prosthesis must be positioned to approximate the normal line of action of the
CAD muscle, and also to minimize migration of the prothesis through the

cartilage when tension is applied. The suture is then tied to create permanent

63

abduction of the arytenoid cartilage. The pharyngeus muscles, subcutaneous
tissue and skin are then apposed. A unilateral or bilateral ventriculectomy may
then be performed via Iaryngotomy.

Since the original description of the technique, various modifications
have been described including an approach dorsal to the linguofacial vein
(Merriam,1973). The elastic material described with the original technique
(Marks et al,1970:ll), is now rarely used because of the difficulty of achieving
sterility, inferior long term strength, and increased tissue reactivity
(Merriam, 1 973). Use of a tracer surgical needle to pass the prosthesis through
the muscular process is now commonly used in place of a crochet hook
(Shappell at al,1988). Speirs (1983) considers placement of a second
prosthesis beneficial, although placement of the second suture in the cricoid
cartilage too far lateral to the first may result in adduction, rather than
abduction of the vocal fold and process (Speirs,1987).

The degree of abduction of the arytenoid cartilage by the prosthesis
remains a controversial issue. Some surgeons suggest that maximum arytenoid
abduction is necessary while others propose that maximum abduction may
result in chronic coughing, dysphagia and interference with the lateral food
channel (Greet et al,1979). At present it is generally recommended that the
left corniculate cartilage is positioned midway between full abduction and the
resting position.

There are many potential complications associated with laryngoplasty

including coughing, chondritis, fistulation, laryngeal granuloma due to

64

penetration of mucosa, dysphagia, laryngospasm, tracheitis, pneumonia,
seroma formation, wound sepsis and dehiscence, choke, and osseous
metaplasia of cartilage (Marks et al,1970:ll, Mackay-Smith et al,1973;
Merriam, 1 973, Haynes et al,1980; Goulden and Anderson, 1 982; Speirs,1987).
Fortunately, the incidence of complication is low, with seroma, coughing and
nasal discharge of food and water being the most common complications
reported (Goulden and Anderson,1982). Coughing is reported to occur in up
to 57% of horses immediately post operatively (Merriam,1973) and to remain
chronic in 4% to 6% (Speirs,1987). Aspiration of food particles during eating
may be confirmed by endoscopic examination of the proximal trachea, and
would appear the most common reported cause of coughing. In some horses,
chronic coughing will resolve if the prosthesis is cut or removed (Greet et
al,1979). The incidence of nasal reflux has been reported from 1% to 43%
(Speirs,1983; Goulden and Anderson,1982). Although regurgitation may be
associated with excessive arytenoid cartilage abduction, coughing and
dysphagia have been observed after a sham operation in which no suture was
placed (Greet et al,1979). This suggests that interference with neuromuscular
control of deglutition resulting from the surgical approach for laryngoplasty may
be involved in the pathogenesis of this complication.

Based on relatively large numbers of horses treated by PL, a ”successful"
outcome, determined from subjective evaluation of performance and stridor,
has been reported in 44% to 90% of cases (Goulden and Anderson,1982;

Baker,1983:ll; Marks et al,1970:ll; Haynes,1984l. These reports were

65

supported by the finding that post-operative racing performance between
treated horses and a control group of horses of similar ability was not
significantly different (Speirs,1983).

Objective assessment of this surgical procedure using evaluation of
airway flow mechanics was initially provided by Derksen et al (1986).
Although horses were evaluated at speeds substantially less than experienced
during racing, PL significantly reduced the resistance to inspiratory airflow in
horses with surgically induced LLH. This finding was later substantiated using
treadmill speeds, which resulted in airflow rates in horses approaching that
experienced during racing (Shappell et al,1988). Recently, tracheal airway
pressure was measured in induced and naturally occurring forms of LLH, before
and after PL (Williams,1990:l,li). These measurements were performed under
simulated race conditions, and showed that pressure measurements were
substantially reduced following prosthetic laryngoplasty, further supporting the
initial objective evaluations of this surgical procedure (Derksen et al,1986).
Laryngoplasty has also been shown to reduce the level of hypoxemia recorded
in exercising horses with LLH (Bayley et al,1984; Tate et al,1993).

The long term effects of laryngoplasty on performance are less well
documented. Failure of the prosthesis to maintain arytenoid cartilage
abduction, is reported to occur in some horses (Marks et al,1970:ll;
Baker,1983; Goulden and Anderson, 1 982). Loss of abduction may be sudden,
as a result of suture breakage or loosening, or gradual over a period of years,

due to migration of the prosthesis through laryngeal cartilage. Variations of the

66

original technique described have been aimed at decreasing the incidence of the
prosthetic material tearing from the muscular process. Goulden and Anderson
(1982) reported on placement of a second suture, anchored to the prosthesis,
through the arytenoid and thyroid cartilages, as well as a teflon implant
between the caudal border of the cricoid cartilage and the prosthesis.
Placement of the prosthesis using a horizontal mattress pattern in the cricoid
cartilage and use of an additional prosthesis have also been advocated
(Speirs,1983). In vitro studies suggest that the age of the horse and type of
prosthetic material, are unlikely to influence the rate at which abduction is lost,
and the most common place for the prosthesis to pull through the cartilage is
at the muscular process of the arytenoid cartilage (Dean et al,1990).

Recent reports suggest that transection of the left RLN, preventing any
residual contraction of the CAD and CAL muscles, following prosthesis
placement may reduce the rate at which the prothesis cuts through laryngeal
cartilage (Ducharme and Hackett,1991). Convincing justification for these

additional procedures is lacking at the present time.

C. Arytenoidectomy

In the late 19th century, Gi‘lnther (Hanover,1845l and Mc'iller, in their
attempts to find a surgical treatment for LLH, performed partial and total
arytenoidectomies with the aim of removing the obstructing arytenoid cartilage
(in Liautard,1892). Subsequently, Giinther Junior reported that total

arytenoidectomy often resulted in severe post-operative hemorrhage and

67

dysphagia with death from aspiration pneumonia (in Williams,1911). He also
found that although these life threatening complications could be avoided by
removing less cartilage, particularly the corniculate cartilage, his results were
still unsatisfactory because the remaining cartilage continued to interfere with
air flow. The introduction of the ventriculectomy in 1906 (Williams,1911)
provided a less traumatic method of improving horses that had LLH, and
subsequently arytenoidectomy fell into disuse.

The terminology classifying the 3 types of arytenoidectomy, although
adequate, remains somewhat confusing. Total arytenoidectomy involves the
complete removal of both the corniculate and arytenoid cartilages. Partial
arytenoidectomy involves resection of both cartilages with retention of the
muscular process, while in the subtotal method, only the arytenoid cartilage is
removed with both the corniculate cartilage and usually the muscular process
being retained (Haynes,1984). More recently an alternative classification
scheme has been proposed based on incorporating the arytenoid and
corniculate cartilage into the terminology, and is suggested to be more
anatomically correct (Speirs,1987).

Use of the modern day form of arytenoidectomy was stimulated in the
late 1970's by isolated reports describing the successful treatment of horses
with arytenoid chondropathies, and of horses in which a PL had failed, using
either subtotal or partial arytenoidectomy (Wheat,1978; White and
Blackwell,1980; Haynes et al,1980). These reports suggested that the life

threatening post-operative complications experienced in the 19th century were

68

not observed, presumably the result of improved anesthetic and surgical
technique, and the availability of anti-inflammatory drugs.

Following recognition of the performance limiting effects of arytenoid
chondropathy (Haynes et al,1980). and it's apparent increasing incidence,
arytenoidectomy has become a common place procedure (Tulleners et
al,1988:l; Speirs,1987). Although the other main indication for
arytenoidectomy is management of unsuccessful prosthetic Iaryngoplasties,
some authors consider arytenoidectomy a primary treatment for LLH, as the
potential coughing, aspiration and short lived abduction associated with the
prosthesis may be avoided (Haynes et al,1984; Mcllwraith and Turner,1987).

The objective of arytenoidectomy is to remove sufficient cartilage to
produce an airway adequate for the projected activity of the horse. Anesthesia
is maintained via an endotracheal tube placed through a mid cervical
tracheotomy which can be performed either before or after induction of general
anesthesia (Haynes,1984). With the horse in dorsal recumbency, a
Iaryngotomy is performed and the ventricle on the same side as the affected
arytenoid cartilage is usually removed first. Subsequently, mucosal incision and
dissection allows complete or partial removal of the arytenoid and corniculate
cartilages. Mucosal edges are then apposed in a continuous or interrupted
pattern using absorbable suture material.

A tracheostomy tube is often required for up to 5 days post-operatively.
The Iaryngotomy is left to heal by second intention and anti-inflammatory drugs

and broad spectrum antibiotics are commonly employed for 2-5 days. Stall rest

69

is recommended for 30 days and followed by an additional 30-60 days of
paddock turnout. A more detailed surgical description of partial
arytenoidectomy is found in Chapter VI.

Numerous variations regarding surgical technique have been described
since the resurgence of this surgical procedure including, not transecting the
cricoid or thyroid cartilages (Speirs,1986), and the submucosal injection of
1:10,000 epinephrine prior to mucosal incision (White and Blackwell,1980).
Others have suggested that the inter-arytenoid ligament should be preserved
to minimize post operative dysphagia (Speirs,1986). Excision of the corniculate
and arytenoid cartilages using CO, and Nd:YAG laser has been performed in the
standing and anesthetized horse (Montgomery,1985; Tate et al,1986; 1989),
and recently partial arytenoidectomy has been reported using an extra-laryngeal
approach (Hay and Tulleners,1993).

Some authors advocate healing of mucosal defects, particularly in the
treatment of arytenoid chondropathies, by second intention rather than
attempting primary closure. Second intention healing following partial
arytenoidectomy is associated with shorter surgery times, reduced hematoma
formation, and comparable healing quality when compared to primary closure
techniques, though healing time appears longer compared to primary closure
(Tulleners et al.1988zll). Recently, the use of systemic corticosteroids has
been recommended to control post-operative swelling (Dean,1990).

lntra-operative complications include difficult access and visualization,

hemorrhage, and defects in the mucous membranes. Post-operatively

7O

complications may include wound swelling and dehiscence (Tulleners et
al,1988:l), coughing, dysphagia, nasal discharge of food and water, and
cartilage mineralization (Haynes,1984; Speirs,1986). Complications referable
to dysphagia and inadequate laryngeal protection of the lower airway during
swallowing are reportedly more common after partial arytenoidectomy
(Speirs,1986l and have been suggested to result from the absence of the
corniculate cartilage (Haynes et al,1984; Speirs,1986). The pathogenesis of
the observed dysphagia following partial arytenoidectomy may result from the
inability of the arytenoids to contribute to glottal closure during swallowing
(Speirs,1986; 1987). However, the real need to retain any portion of the
corniculate cartilage is still uncertain, and if epiglottic function is normal it is
questionable whether the rima glottidis needs to be completely sealed on
adduction (Mcllwraith and Turner,1987).

Subjective evaluation, based on performance of intended activity and
complications following subtotal and partial arytenoidectomy have been
reported. Reports of successful outcome following arytenoidectomy vary
greatly depending on the criteria used by the author to define "success“. In
addition, accurate interpretation of studies evaluating arytenoidectomies is
difficult as the underlying disease process may be either arytenoid
chondropathy (AC), LLH, or a combination of both. Improved athletic
performance and reduction in respiratory noise following subtotal

arytenoidectomy has been reported in 50% of horses with AC and 83% with

71
LLH (Haynes et al,1984). Coughing and evidence of dysphagia was reported in

25% of horses in this study, but were not considered to effect performance.
Improved athletic ability following unilateral partial arytenoidectomy is
reported in 56% of horses treated for either LLH or failed PL (Speirs,1986l and
63% to 80% of horses with AC (Speirs,1986, and Tulleners et al,1988:ll.
Complications such as dysphagia and coughing following this procedure have
been reported as high as 45% (Speirs,1986l and more recently as low as 10%
(Tulleners et al,1988:ll. The reason for the disparity between the complication
rates reported by Speirs and Tulleners may reflect minor differences in surgical
procedure or may be influenced by the nature of the condition being treated.
In 1990, quantitative evaluation of subtotal arytenoidectomy for the
treatment of surgically-induced LLH, unequivocally demonstrated the failure of
this technique to improve upper airway function in exercising horses (Belknap
et al,1990). Similarly, evaluation of tracheal pressures (as an indicator of upper
airway resistance) revealed that, in horses with surgically-induced LLH, and in
horses with arytenoid chondropathy, subtotal arytenoidectomy did not
significantly decrease tracheal pressures (Williams et al,1990:l,ll). These
findings are in direct contrast to that of Haynes et al,(1984), who reported
improved exercise tolerance following subtotal arytenoidectomy, based on
subjective assessment. Differences between these authors may be explained
by the subjective method of evaluation used by Haynes (1984) compared to
quantitation of airway function (Belknap et al,1990; Williams et al,1990:ll).

Differences in the outcome reported following subtotal arytenoidectomy may

72

also reflect the varied athletic requirements of the horses evaluated and

differences in the amount of corniculate cartilage removed.

D. Laryngeal Reinnervatlon

From the above review of treatments for LLH, their varied degrees of
"success“, result from either a less than adequate airway cross sectional area
or post-operative complications. As stated by Goulden and Anderson in 1982,
optimal airway cross sectional area and minimal post-operative complications,
may be achieved by using more physiologically compatible techniques, such as
reinnervation of the abductor muscle of the larynx.

Treatment of laryngeal paralysis in people by laryngeal reinnervation
using a nerve muscle pedicle (NMP) graft has been described and is considered
a valuable procedure (Lyons and Tucker, 1 974; Tucker, 1 976). The mechanism
by which denervated muscle is reinnervated using either direct nerve
implantation or NMP transposition is not completely understood. It has been
proposed that the NMP graft technique transplants motor end plates (the
synapse of a nerve and a muscle fiber) in the NMP to the recipient paralysed
muscle. Subsequently, contraction of the recipient muscle is caused by
propagation of depolarization from the NMP graft (Johnson and Tucker,1987).
An alternative theory is that axons transplanted either in the NMP or the
transected nerve, sprout and form junctions with the recipient muscle fibers

(Hall et al,1988l.

73

For NMP transplantation to be successful the recipient muscle must be
able to be reinnervated despite the underlying disease process and the extent
of neurogenic atrophy of the muscle. As the pathogenesis of equine LLH is
considered a distal axonopathy and not a primary myopathy, laryngeal
reinnervation may have potential value in the treatment of LLH. Clinical signs
of LLH typically occur in horses from 2-5 years (Goulden and Anderson,
1981:“). Therefore, the duration of neurogenic atrophy of the CAD muscle is
unlikely to preclude the muscles ability reinnervate, as reinnervation is
commonly successful after 3 years and up to 22 years of denervation atrophy
in humans (Tucker,1976l.

Following successful reinnervation, the muscle must assume the
functional and biological characteristics of the muscle originally innervated by
the transferred nerve (Tucker,1976). As the pathophysiology of equine LLH
involves the inability of the CAD muscle to abduct the arytenoid cartilage during
inspiration, donor nerves supplying accessary muscles of respiration which
contract during inspiration have been used.

In 1989, Ducharme and co-workers reported results of the efficacy of
CAD muscle reinnervation in ponies with induced laryngeal hemiplegia. The
second cervical nerve (with and without NMP grafts from the omohyoideus
muscle) was used to reinnervate the equine larynx (Ducharme et al,1989zl-lll).

The surgical technique described here is as reported by Fulton ( 1990)
using a NMP graft. Other techniques of reinnervation are reported elsewhere

(Ducharme et al,1989:l-Ill; Harrison et al,1992). Using similar patient

74

positioning and surgical approach to that for PL, a branch of the first cervical
nerve is identified as it travels caudal to the cricopharyngeus muscle to insert
in the omohyoideus muscle. The cricopharyngeus muscle is retracted cranially
to expose the CAD muscle in which a 1-cm opening is created parallel to its
muscle fibers. An omohyoideus muscle pedicle (5 x 5 mm) is created
containing the insertion of the first cervical nerve and is transposed into the
previously created opening in the CAD muscle. The NMP graft is anchored by
2 simple interrupted 4—0 absorbable sutures, and the wound is then closed
routinely.

Since the initial description of this procedure, modifications to surgical
technique have included, electrical stimulation for identification of nerves
(Harrison et al,1992), local anesthetic induced desensitization of the first
cervical nerve prior to excision of the muscle pedicle, and implantation of
multiple NMP grafts at various locations in the CAD muscle (Fulton,1990).

Success of laryngeal reinnervation techniques has been evaluated using
endoscopic appearance of the larynx at rest (Ducharme et al,1989:l; Fulton et
al,1992; Harrison et al,1992) and during treadmill exercise (Fulton et al,1991).
The ability to reinnervate the CAD muscle has also been evaluated using
histologic evidence of reinnervation (Ducharme et al,1989; Fulton et al,1992;
Harrison et al,1992) and objective measurement of upper airway function
during exercise (Fulton et al,1991).

Using a NMP graft harvested from the second cervical nerve (C2) and the

omohyoideus muscle, Ducharme et al.(1989:l) found that 30 weeks after

75

surgery, the endoscopic appearance of the larynx at rest, showed only slight
arytenoid abduction in 1 of 4 ponies. Despite these findings, histologic
evidence of reinnervation after NMP graft technique, seen as muscle fiber
grouping (Eisele et al,1988), was observed in 3 of 4 ponies. Subsequent to
this report, the same researchers, using a direct nerve (CZ) implantation
technique, observed partial arytenoid abduction in 4 of 6 ponies, though
convincing histological evidence of reinnervation in the CAD muscles was not
found (Ducharme et al,1989;"). In a third study reinnervation was attempted
by anastomosing a sectioned branch of C2 with the abductor branch of a
distally sectioned left RLN (Ducharme et al,1989:lll). Endoscopic examination,
while horses were breathing 10% 00,, revealed clonic abductor movement of
the arytenoid cartilage which was synchronous with inspiration in 5 of 6
horses. Histologic evidence for reinnervation supported the observed abductor
muscle function in these horses. Despite the fact that physiologically, the
reinnervation was successful, the partial restoration of function did not appear
to be sufficient for racing. In addition, the technical difficulty of the nerve
anastomosis technique, and the questionable viability of the distal RLN in cases
of LLH, render this procedure of questionable value (Fulton,1990).

Following the initial findings of Ducharme et al.(1989), Fulton et
al.(1991) reported the successful use of a NMP graft in restoring upper airway
function in horses with induced LLH. Measurements of upper airway function
in 5 horses receiving a NMP graft, and 2 horses undergoing a sham procedure,

provided objective data that laryngeal reinnervation significantly improves upper

76

airway function in horses with LLH during exercise. Substantial abduction of
the arytenoid cartilage was observed endoscopically following electrical
stimulation of a proximal segment of C1, during nasal occlusion (Fulton et
al,1992), and strenuous exercise (Fulton pers comm.) though not during quiet
breathing at rest (Fulton et al,1991). Although reinnervation studies by
Ducharme and Fulton report histological evidence of CAD muscle reinnervation,
the lack of observed abduction of the arytenoid cartilage in the studies of
Ducharme et al,(1989zl-ll), may be explained by insufficient stimulation of the
accessary muscles of respiration during the resting endoscopic examination.
Despite the encouraging findings of Fulton's study, of improved upper airway
function and lack of significant post-operative complications, the duration for
reinnervation to allow significant improvement in upper airway function took up
to 52 weeks. Thus, the clinical application of this technique may be limited by
the economic feasibility of the time taken to return athletes to competitive
levels of performance. Following these findings, it has been proposed that
multiple NMP grafts may potentially reduced the time for reinnervation, though
to date preliminary findings have not supported this hypothesis (Derksen et
al,unpublished data).

Recently, the ability of a right CAD muscle pedicle graft to innervate the
contralateral denervated CAD muscle was evaluated. Results from this study
suggested that muscle-to-muscle neurotization of the paralyzed muscle did not

occur, and that this technique appeared to be of limited clinical value in the

77

treatment of LLH in horses (Harrison et al,1992). Continued investigation and
refinement of laryngeal reinnervation techniques is clearly warranted.
Conclusion

The development of videoendoscopic and upper airway function
measurement techniques to document the effect of LLH during exercise,
represents a major advancement in our understanding of the pathophysiology
of LLH in the horse. As discussed previously, the objective techniques
currently available have limited clinical application. Therefore, the first goal in
my study is to evaluate the usefulness of flow-volume loop analysis in providing
measurement of airway function that may be a useful clinical diagnostic test.
Evaluation of this technique in documenting the effect of LLH in resting and
exercising horses is presented in chapter V.

As revealed in the preceding review of the various treatments for LLH,
as yet none is considered ideal. The treatment of choice when performing
arytenoidectomy remains controversial between advocates of subtotal and
partial arytenoidectomy techniques. To date the efficacy of partial
arytenoidectomy in treating LLH has not been objectively evaluated. Therefore,
the second objective of this study, presented in chapter VI, is to determine the
efficacy of partial arytenoidectomy for treatment of LLH using previously
documented techniques (calculation of impedance) and flow-volume loop

analysis.

V. USE OF FLOW-VOLUME LOOPS TO EVALUATE UPPER AIRWAY

OBSTRUCTION IN EXERCISING STANDARDBRED HORSES

A. Summary

Flow-volume loops generated from 6 Standardbred horses at rest and
during treadmill exercise were evaluated for their use in detecting upper airway
obstruction. Tidal breathing flow-volume loops (TBFVL) were obtained from
horses at rest and exercising at speeds corresponding to 75% of maximal heart
rate and at maximal heart rate. The TBFVL were evaluated, using a pulmonary
function computer; calculated indices describing airflow rate and expiratory-to-
inspiratory airflow ratio for individual loops were determined. In addition to
TBFVL indices, standard variables of upper airway function also were
measured: peak airflow, peak pressure, and calculated inspiratory and
expiratory impedances. Measurements were recorded before left recurrent
laryngeal neurectomy (LRLN; baseline) and 14 days after surgically induced left
laryngeal hemiplegia.

When horses were at rest, TBFVL shape and the indices describing the
loop were highly variable. In contrast, in exercising horses, TBFVL shape was
consistent and coefficients of variation of loop indices were less during exercise
than at rest. After LRLN, TBFVL from exercising horses indicated marked
inspiratory airflow limitation, while the expiratory airflow curve was preserved.
Peak inspiratory flow rate and inspiratory flow at 50 and 25% of tidal volume

decreased, and the ratio of peak expiratory to inspiratory airflow (PEF/PIF) and

78

79

that of mid-tidal volume expiratory and inspiratory airflow rates (EFso/IFSOI
increased significantly (P = 0.05). lnspiratory impedance also increased after
LRLN.

Although in resting horses TBFVL were not a useful indicator of upper
airway obstruction, examination of TBFVL from exercising horses allowed
objective, specific, and repeatable detection of upper airway obstruction. The
technique was non-invasive, rapid, and well tolerated by horses; thus, it is a

potentially valuable clinical diagnostic test.

B. Introduction

Upper airway obstruction is a common cause of reduced exercise
tolerance in performance horses (Raphel,1982; Sweeney et al,1991 ; Morris and
Seeherman,1991) and is frequently associated with inspiratory noise
(Fleming,1882; Haynes, 1 984). Lesions of the upper airway resulting in airflow
obstruction include laryngeal hemiplegia, subepiglottic cysts, arytenoid
chondropathy, and aryepiglottic entrapment. Of these, left laryngeal hemiplegia
(LLH) is most commonly diagnosed (Morris and Seeherman,1990; 1991 ). The
inspiratory flow limitation caused by axial displacement of the left arytenoid
cartilage is the direct result of paralysis of its abductor muscle (Robinson and
Sorenson,1978; Derksen et al,1986). Endoscopic examination of the larynx
and pharynx in resting horses permits descriptive evaluation of upper airway
lesions (Cook,1974). However, some upper airway lesions, such as dorsal

displacement of the soft palate, may be inapparent in the resting horses and are

80

best identified by videoendoscopy during exercise (Derksen,1988; Morris and
Seeherman,1988). The limitation of this technique is that it requires subjective
interpretation of airway function. Upper airway impedance has been measured
experimentally in horses to objectively evaluate the efficacy of prosthetic
laryngoplasty (Derksen et al,1986), ventriculectomy (Shappell et al,1988),
subtotal arytenoidectomy (Belknap et al, 1990), and laryngeal reinnervation
(Fulton et al,1991) as treatments for LLH. The clinical use of objective
measurement techniques of upper airway function has been limited (Stick and
Derksen,1989; Williams et al,1990:ll). primarily because of the inconvenient
and invasive nature of the technique (Derksen et al,1986).

Flow-volume loop analysis is a common test of respiratory tract function
in human medicine, because it is non-invasive, convenient, and sensitive.
Airflow rate is continuously plotted against volume during a single maximal
inspiratory and expiratory effort (FVL). Initially used in the clinical evaluation
of lower airway disease (Hyatt et al,1958l. FVL analysis remains one of the
most widely used pulmonary function tests in human medicine. Determination
of FVL has been used for the clinical diagnosis of upper airway obstruction in
human beings since 1968 (Jordanoglou and Pride,1968. The sensitivity,
specificity, and repeatability of this test rely on patient co—operation to perform
maximal inhalation and exhalation. In doing so, flow rates approach maximal
and are associated with significant intraluminal pressure changes allowing

detection of subtle airway obstruction (Fry and Hyatt,1960). In veterinary

81

medicine, the non-cooperative nature of our patients has prevented clinical use
of maximal FVL.

Clinical evaluation of upper airway function, using a tidal breathing flow-
volume loop (TBFVL) has been attempted in human neonates and infants who
are incapable of performing a maximal voluntary respiratory effort on demand
(Abramson et al,1982). When breathing with less than maximal effort, such
non-cooperative subjects have low airflow rates that are associated with lack
of sensitivity of the TBFVL to detect airway obstruction (Abramson et al,1982;
Wise et al,1980). The great flow variability associated with TBFVL prevents
strict numerical interpretation of loop indices, therefore limiting its usefulness
in evaluating airway obstruction (Abramson et al,1982). Tidal breathing flow-
volume loop analysis performed in dogs with upper airway obstruction also
lacks sensitivity and cannot reliably quantitate less than severe obstruction
(Amis and Kurpershoek,1986; Amis et al,1986).

The sensitivity of the TBFVL in detecting airway obstruction may improve
as flow rates approach maximal (Fry and Hyatt,1960). In horses, high-speed
treadmill exercise results in near-maximal airflow rates (Belknap et al,1990;
Fulton et al,1991). Therefore, the objective of the study reported here was to
evaluate the ability of TBFVL, generated during exercise, to detect upper airway

obstruction induced by left recurrent laryngeal neurectomy (LRLN).

82
C. Materials and Methods

Horses—Six adult Standardbreds (3 male and 3 female) 5.2 :I: 2.0 (mean
:I: sd; range 3 to 8) years old and weighing 437.8 :l: 40.2 kg (range, 390.0 to
490) were studied. All horses were maintained on pasture for at least 30 days
before experiments, treated with anthelmintics, and vaccinated against tetanus,
equine influenza, and rhinopneumonitis. Prior to measurements or surgical
procedures, clinical examination and endoscopic evaluation of the upper airway
at rest indicated no abnormalities. Horses were kept at pasture between
surgical procedures and measurement protocols.

W—Techniques for measurement of airflow, tidal
volume, and transupper airway pressure have been described (Derksen et
al,1986). Briefly, horses were fitted with a 15.2-cm diameter
pneumotachograph (Laminar flow straightener element, Merriam instruments,
Grand Rapids, Mich.) mounted on a face mask. The mask covered mouth and
nostrils and allowed complete nostril dilatation. The combined dead space of
the mask and pneumotachograph was approximately 3.5 L. A wire mesh
(Mesh SS Screen, McMaster Carr, Chicago, Ill.) located between the muzzle
and pneumotachograph, prevented contamination of the pneumotachograph
with secretions and functioned as a flow straightener element. The resistance
of the pneumotachograph was 0.04 cm of H,O/L/s, up to an airflow of 90 Us.
The combined resistance of the facemask, wire mesh, and pneumotachograph
was 0.05 cm of H,O/L/s at the peak airflows recorded in this study. Pressure

changes across the pneumotachograph were measured by use of a differential

83
pressure transducer (Model DP45-22, Validyne Sales, Northbridge, Calif.) which

produced a signal proportional to flow. The pneumotachograph was calibrated
using a rotameter flow meter (Model FP-2-37-P-10/77, Fisher 81 Porter Co,
Warminster, Pa.) capable of measuring airflow rate up to 90 L/s.

lnspiratory and expiratory transupper airway pressure (PU. and Fuel was
measured as the pressure difference between tracheal and mask pressure.
Tracheal pressure was measured via a lateral tracheal catheter placed
percutaneously at the mid-cervical level. A second catheter was connected to
measure pressure within the face mask. Catheters were connected to each
side of a second differential pressure transducer, which was calibrated by use
of a water manometer. Flow and pressure signals were phase-matched up to
10 Hz, as described (Derksen and Robinson,1980).

Airflow and pressure signals were passed through 4-and 8-Hz low-pass
filters, respectively, and were fed into a respiratory function computer (Buxco
LS-14, Buxco Electronics, Inc., Sharon, Conn.). Selection of filters was derived
from preliminary studies. To obtain tidal volume (VT) the flow signal was
digitally integrated with respect to time. lnspiratory and expiratory impedances
(Z, and 25) were calculated as the ratio of peak transupper airway pressure and
peak airflow over 10 consecutive breaths. Minute ventilation (V E) was
calculated as the product of respiratory rate If) and VT. Heart rate (HR) was
recorded, using a telemetry system (Digital UHF Telemetry System, M1403A,

Hewlett Packard, Palo Alto, Calif.).

84

Ten representative TBFVL were selected from each exercise period and
were analyzed, using computer software (Buxco LS-14, Buxco Electronics Inc,
Sharon, Conn.) specifically designed to analyze FVL. Criteria for selection of
FVL included adequate loop closure (< 5% difference between inspiratory and
expiratory volumes) and lack of artifact (snorting and swallowing), as outlined
by Amis et al,(1986).

The shape of the TBFVL was qualitatively evaluated. For quantitative
analysis of TBFVL, VT, inspiratory and expiratory times (T. and TE), total breath
time (Trorl. f, peak inspiratory and expiratory airflow rates (PIF and PEF), flow
rates at 50%, 25%, and 12.5% of tidal volume (le, EFso, IF,5, EF,5, IF1 ,5, and
EF,,.5), ratios of these flow rates, UTE, VT/T., T./TOT, and volume at PIF and PEF
were calculated. The mean :I: SEM and coefficient of variation of measured
values and derived indices were obtained.

W— Over a 3 day period, prior to any measurement
procedures, all horses were acclimatized to work on the treadmill while wearing
the breathing apparatus. Maximal HR (HRm) during exercise was determined
prior to upper airway function tests. All horses underwent a single rapid
incremental exercise test (RIET) to estimate the relation between HR and
treadmill speed for each individual horse. The RIET was performed as reported
by Rose et al,(1990), except that horses were the mask previously described
and a 3-degree treadmill incline was used. Horses were first exercised at a
treadmill speed of 4 m/s for 3 minutes. Subsequently, treadmill speed was

increased to 6 m/s for 90 seconds. Treadmill speed was then increased at

85

intervals of 60 seconds to 8, 10, 11, and 12 m/s, respectively. The RIET was
terminated when horses could no longer maintain position on the treadmill.
Heart rate was recorded in the last 15 seconds of each exercise period. From
these measurements, the treadmill speed at which further increase failed to
substantially increase HR, speed at HR,” was determined. The treadmill speed
at which 75% of maximal HR (HRMw) was observed was also calculated.

WWW-Measurements were initially made with horses
at rest, standing adjacent to the treadmill (period A) throughout the last minute
of a 2-minute interval. Subsequently, horses were exercised on the treadmill
with a 3-degree incline at 4 m/s for a 2-minute period, then at the treadmill
speed corresponding to their HROJW for 2 minutes (period 8). All horses
rested for 1 minute, then exercised at a treadmill speed corresponding to HR“,
(period C) for a 2-minute period. Data were collected in the last minute of each
2-minute exercise period. On the day after each TBFVL measurement, the
larynx was examined by use of videoendoscopy when the horse was at rest
and exercising at HRM.

Measurements were obtained before LRLN (baseline) and 14 days after
LRLN. Left laryngeal neurectomy was performed through a 10-cm skin incision
over the left jugular vein in the mid-cervical region (Derksen et al. 1986). The
recurrent laryngeal nerve was isolated by blunt dissection and transected.
Subcutaneous tissues and skin were then apposed in routine manner.

Wham-Data obtained from individual RIET were analyzed

using linear regression analysis to determine predicted treadmill velocities

86

corresponding to HROJM. Significant differences between mean HR were
determined by use of two-way ANOVA. When F values were significant at p
< 0.05, means were compared, using Fischer's least significant difference test.

A three-way ANOVA was used to analyze the effect of exercise and
surgical treatment according to the model Yijk = Al + B] + ABij + Ck + ACik
+ BCjk + error (Gill 1978), where Al was the fixed effect of exercise (3
levels), Bj was the fixed effect of surgical procedure (2 levels), and Ck was the
random effect of 6 horses. When F values were significant at p < 0.05,

treatment means were compared by use of the Tukey test.

D. Results
The HR“, determined from the RIET for the 6 horses was 225.5 :I: 4.92
beats/min. This rate was achieved at a treadmill speed of 11.0 :I: 0.2 m/s and
was significantly different from mean HR at 10.0 m/s but not from that at 12.0
m/s. The HRMM was 166.8 :I: 3.9 beats/min and corresponded to a treadmill
speed of 6.2 :I: 0.7 m/s (Figure 4).
At period 8, trotting and pacing gaits were observed, whereas at period
C all horses galloped prior to LRLN, except for 1 that paced. After LRLN at
period 8, 3 horses galloped and 3 trotted or paced. After LRLN at period C, all
horses galloped. Because of the severity of upper airway obstruction after
LRLN, 1 horse was unable to work at the required treadmill speed at period C.
W— A variety of different TBFVL shapes was recorded

at rest (period A). On the basis of variation of the inspiratory flow curve and

87

250

 

200 *

Heart Rate (b.p.m.)

iso~ /

100 l I I L
4 6 8 10 11 12

Speed (m/s)

 

 

l—
r.—

 

Figure 4. Mean (:t SEM) heart rate (beats/min;bpm) in 6 horses performing an
incremental treadmill exercise test prior to left recurrent laryngeal neurectomy
(LRLN). The treadmill had a 3-degree slope.

PIF location, 4 basic shapes predominated within and between horses. The
inspiratory curve was monophasic, biphasic, or less frequently, triphasic. The
PIF was either early or late in inspiration. The expiratory flow curve was
typically biphasic, with peak flow observed early in expiration (Figure 5).
When airflow rates and tidal volume increased in response to exercise,

TBFVL shape was markedly modified (Figure 6). At period C, before LRLN, 3

88
of the 6 horses (2 galloping and 1 pacing) had a regular biphasic breathing

pattern (Figure 7A), 2 had a combination of biphasic and monophasic breathing
patterns (Figure 7B), and 1 had a monophasic breathing pattern throughout the
gallop (Figure 7C). Of these 3 TBFVL shapes observed during exercise prior to
LRLN, predominantly biphasic, and less frequently, monophasic expiratory limb
shapes were observed during periods 8 and C. A biphasic expiratory limb
shape was always associated with biphasic inspiratory limb shape (Figure 7A),
whereas monophasic expiratory limb shape was associated with monophasic
and biphasic inspiratory limb shapes (Figures 78 and C). In the biphasic
inspiratory flow curves, peak flow was observed at approximately 30% of tidal
volume. The smaller second inspiratory flow peak was observed late in
inspiration and was preceded by a more pronounced reduction in flow rate than
that seen in the biphasic expiratory flow curve. Although a clear and
consistent relation between gait and loop shape was not evident, the mono-
phasic loop shape illustrated in Figure 7C was only seen in galloping horses
during period C. This loop shape was associated with smaller tidal volume and
higher respiratory rate than those associated with the other loop shapes.
After LRLN, videoendoscopy confirmed the presence of LLH at rest and
during exercise in all 6 horses. No changes in TBFVL shape were seen at
period A. In contrast, marked changes were observed in the shape of the
inspiratory limb of the TBFVL at periods 8 and C; similar shapes were seen in
all 6 horses. Although the shape of the expiratory limb of the TBFVL was

approximately the same as that seen prior to LRLN, the inspiratory flow curve

89

_ A ,3
C I:
O h- o l—
-.—_: '5
.g .g
Q. "’ Q .—
X X
LU U1

 

-(h-
-l-

Flow (4 00 Us/dIv)
)-
Flow (4 00 Usldlv)
fl I

I

lnspiratlon
l

Inspiration
I

 

 

— Volume (2.00 Udiv) - Volume (2.00 Udiv)

p-
h—
l J I l I l J

Expiration
I
Expiration

qr-
c-Il-

Flow (4 00 Us/dlv)
I 1

Flow (4.00 Us/div)

j.-

_

Inspiration

Inspiration

pr—

 

I

 

Volume (2.00 Udiv) L Volume (2.00 Udiv)

Figure 5. Representative tidal breathing flow-volume loops (TBFVL) from 4
horses at rest. Variation in TBFVL shape was seen within and between horses.
Tidal volume at peak inspiratory flow was variable. The TBFVL indicate peak
inspiratory flow near the end of inspiration (A) and early in inspiration (8).
Variations in the inspiratory limb of TBFVL included monophasic (A and B),
biphasic (C), and triphasic (D) patterns.

9O

Expiration
I
/
0|

 

 

 

3..

._ .

D

E; i i (: I! 1 ‘xfflfi‘77"" I
a k———’ 7
5..

SE

3 l.

r‘.’

a N

E_ J

 

 

 

L. Volume (4.00 L/dtvI

 

Figure 6. Representative TBFVL from a horse at rest (period A), exercising at
75% of maximal heart rate (period 8), and at maximal heart rate (period C).
The effect of increasing exercise on shape, airflow rate, and tidal volume is
shown. The representative TBFVL (C) at period C shows the flow
measurements used to calculate TBFVL indices. Curved arrows indicate the
direction of the respiratory cycle.

VT

=O-5s/div

91

 

Flow (20.00 Lia/(IV)

  

Flow (2000 wow;

 

 

 

~ Volume (4.00 Udiv)

- Volume (4.00 Udv)

 

I
b
.—
..
I-
>-
F

 

4, I f G -<I

Volume (4.00 Udiv)

Figure 7. Airflow curves from different horses before LRLN exercising at

maximal heart rate (period C).

Horses had a constant biphasic breathing

pattern (A), a combination of biphasic and monophasic patterns (8), or a
constant monophasic pattern (C). The TBFVL derived from flow and volume
tracings are at the right of the tracing. A typical biphasic TBFVL shape is
shown in A, composed of biphasic expiratory and inspiratory airflow curves.
Monophasic expiratory airflow curves were seen with a biphasic (B) or
monophasic (C) inspiratory airflow curve. V -= airflow (50 L/s/division), VT =
tidal volume (10 L/division) and, t = time (0.5 sec/division).

92

was typified by a flow peak that occurred early in inspiration and was followed
by a pronounced reduction in airflow (Fig 8). This reduction in inspiratory flow
rate was accentuated with increasing treadmill speed. After LRLN, all horses
adopted either a biphasic or triphasic inspiratory breathing pattern at exercise

period C; monophasic inspiratory shapes were not observed.

l—

_

Expiration

(20.00 LIs/oivl

 

Flow

Inspiration

h—

 

L- Volume (4.00 L/div)
Figure 8. Typical TBFVL from a horse exercising at maximal heart rate (period

C) before LRLN (a) and after LRLN (b). Preservation of the expiratory flow
curve and inspiratory flow limitation are seen after LRLN.

Winn—Prior to LRLN, increasing intensity of exercise

resulted in a progressive increase in HR, Vt, V5, Fur. PIF, and PEF. Although f

and P0: increased with treadmill speed, significant differences between the

93

exercise periods 8 and C were not documented. Significant differences were
not observed in either ZI and 2E between rest and exercise (Table 1).

With regard to the TBFVL indices determined before LRLN (Table 2),
increasing treadmill speed progressively decreased T. and TE, and increased VT,
VT/T., lung volume at PIF and PEF, and all airflow rates during inspiration and
expiration. Other indices were not affected by exercise.

After LRLN, PU. and 2. increased significantly at exercise periods 8 and
C (Table 1), whereas PIF was decreased, compared to baseline (pre-LRLN)
measurements. The LRLN procedure did not have significant effect on f, VT,
2,, Pug, and PEF at any exercise level. After LRLN at period 8, HR was
significantly greater than the baseline value, and at period C, V E significantly
decreased. After LRLN, PEF/PIF and EFw/IFSO increased significantly, compared
with baseline values at exercise periods 8 and C (Figure 9), inspiratory flow rate
at 50 and 25% of tidal volume and lung volume at PIF decreased at period C,
and PIF decreased at periods 8 and C (Table 2;Figure 10). The LRLN procedure
had no effect on indices describing the expiratory limb of the TBFVL (Table 2).

After LRLN at period C, TE/T. was significantly decreased, whereas at
period 8 and C, T./'l'mT was significantly increased. The VT/T. value was also
significantly decreased at periods 8 and C after LRLN (Table 2). Some
coefficients of variation of TBFVL indices which changed significantly after

LRLN are shown in Table 3.

T

able 1 -

94

Effect of exercise on the measured and calculated variables in the

(baseline) and after left recurrent laryngeal neurectomy (LRLN).

6 horses before

 

Exercise level

 

 

Surgical
VITIODIO Status A 1 B t C *
HR (beets/min) baseline 40.17 1 4.18 173.67 1 2.79' 217.17 1 2.37'1
LRLN 40.00 1 1.98 189.00 1 2.53! 222.80 1 3.72
f (breaths/min) baseline 27.83 1 2.07 77.33 1 6.21" 92.67 1 15.16"
LRLN 23.66 1 1.43 68.50 1 3.96 70.80 1 3.00
VT (Ll baseline 5.39 1 0.39 12.50 1 0.78’ 14.80 1 1.18'1‘
LRLN 5.98 1 0.29 11.78 1 0.68 14.51 1 0.08
v, (lein) baseline 151.41 1 15.01 950.23 1 59.02:: 1295.97 1 127.52%
LRLN 140.75 1 7.69 808.40 1 68.29 1027.60 1 71.175
Pu. (cm of H,Ol baseline 1.94 1 0.22 22.29 1 1.15‘ 38.57 1 3.93‘t
LRLN 2.16 1 0.32 49.40 1 4.08§ 62.73 1 6.705
PUE (cm of H,Ol baseline 1.10 1 0.40 14.16 1 3.15' 11.86 1 3.410
LRLN 2.08 1 0.91 9.65 1 0.73 11.61 1 2.49
PIF (Llsecl baseline 5.09 1 0.34 51.19 1 4.05’ 75.52 1 9.35'1
LRLN 5.84 1 0.39 36.58 1 2.745 40.87 1 3.235
PEF (Llsecl baseline 7.71 1 0.36 56.94 1 4.39 66.05 1 5.58'1
LRLN 7.16 1 0.64 56.89 1 4.98 65.76 1 4.59
2. (cm H,O/L/sec) baseline 0.39 1 0.03 0.46 1 0.06 0.53 1 0.04
LRLN 0.38 1 0.06 1.41 1 0.205 1.63 1 0.285
ZE (cm H,O/L/sec) baseline 0.14 1 0.03 0.25 1 0.06 0.19 1 0.06
LRLN 0.26 1 0.09 0.18 1 0.02 0.19 1 0.06

 

HR = heart rate, I

Hut-Ir 0

respiratory frequency, VT = tidal volume, VE = minute ventilation, Pu. =
inspiratory pressure, P0: = expiratory pressure, PIF = peak inspiratory flow, PEF = peak expiratory
flow, 2, = inspiratory impedance, ZE = expiratory impedance.

Data significantly different (P < 0.05) from period A.
Data significantly different (P < 0.05) from period 8.
Data significantly different (P < 0.05) from baseline measurement at same speed.
Values represent means and SEM.

95

Table 2 - Effect of exercise on selected tidal breathing flow-volume loop values from the 6
horses before (baseline) and after left recurrent laryngeal neurectomy (LRLN).

 

Exercise level

 

 

 

Sur ical
Variable staotus At 81 Ct
T. (msl baseline 1234.12 1 105.28 420.67 1 33.52' 387.50 1 65.70'
LRLN 1426.80 1 115.28 540.83 1 26.00 536.51 1 17.82
TE (ms) baseline 1196.87 1 86.67 396.00 1 37.14' 366.17 1 57.62'
LRLN 1398.61 1 122.44 373.50 1 9.8 351.19 1 8.4
TEI'I‘, baseline 0.99 1 0.05 0.94 1 0.03 0.98 1 0.05
LRLN 0.97 1 0.05 0.70 1 0.02 0.65 1 0.015
Tflm, baseline 0.51 1 0.01 0.51 1 0.01 0.51 1 0.01
LRLN 0.52 1 0.01 0.59 1 0.015 0.61 1 0.015
VTIL) baseline 6.02 1 0.92 13.11 1 0.80’ 15.73 1 1.27‘
LRLN 6.39 1 0.35 12.47 1 0.44 14.07 1 0.53
VT/T, baseline 4.84 1 0.34 31.77 1 2.08' 44.78 1 5.03't
LRLN 4.40 1 0.14 23.87 1 1.705 25.70 1 1.025
PIF (Us) baseline 7.02 1 0.40 50.42 1 3.9 ' 77.02 1 8.43't
LRLN 6.03 1 0.26 36.21 1 2.095 38.22 1 1.365
IF“, (Us) baseline 5.24 1 0.31 49.23 1 4.08‘ 70.77 1 8.47%
LRLN 4.84 1 0.18 23.77 1 1.61 24.09 1 1.035
IF,,5 (Us) baseline 6.17 1 0.56 43.57 1 2.70' 59 48 1 5.04% -
LRLN 5.17 1 0.38 34.34 1 2.14 35 53 1 1.645
PEF (L/sl baseline 9.25 1 0.94 56.68 1 5.24‘ 71.71 1 5.86%
LRLN 8.06 1 0.53 56.54 1 3.25 65.22 1 2.19
EFlio (Llsl baseline 6.15 1 0.35 47.02 1 7.38‘ 65.76 1 8.09’t
LRLN 5.27 1 0.65 50.35 1 3.67 61.40 1 1.81
EF,. (L/sl baseline 5.25 1 0.54 36.61 1 3.38‘ 53.15 1 5.50'1
LRLN 4.88 1 0.49 34.75 1 2.58 45.02 1 1.14
PEF/PIF baseline 1.33 1 0.07 1.13 1 0.06 0.97 1 0.07
LRLN 1.35 1 0.10 1.58 1 0.055 1.70 1 0.035
Elele baseline 1.25 1 0.11 1.19 1 0.11 1.08 1 0.17
LRLN 1.11 1 0.15 2.15 1 0.125 2.56 1 0.045
Vol PIF (Ll baseline 1.87 1 0.41 3.56 1 0.32‘ 5.79 1 0.33't
LRLN 2.74 1 0.42 2.52 1 0.17 2.74 1 0.175
Vol PEF (Ll baseline 5.07 1 0.90 8.94 1 0.80' 10.15 1 1.31’
Li" 5.61 1 0.34 8.15 1 0.26 8.59 £0.40

 

 

 

 

T. = inspiratory time, TE = expiratory time, V, = tidal volume, PIF = peak inspiratory flow, IFS,
= inspiratory flow at 50% of VT, IF” = inspiratory flow at 25% of VT, PEF = peak expiratory
flow, EFI50 = expiratory flow at 50% of V,, EF,,5 = expiratory flow at 25% of VT, Vol PIF =
volume at peak inspiratory flow, Vol PEF = volume at peak expiratory flow.

' Data significantly different (P < 0.05) from period A.

t Data significantly different (P < 0.05) from period 8.

5 Data significantly different (P < 0.05) from baseline measurement at same speed.
t Values represent means and SEM.

96

Table 3 - Coefficients of variation for various tidal breathing flow-volume loop
indices from the 6 horses before (baseline) and after left recurrent
laryngeal neurectomy (LRLN).

 

Exercise level

 

 

. Surgical
Variable Status A B C
2mm 2mm 2mm
TEIT. baseline 19.8 (0.5) 4.78 (0.5) 4.35 (0.4l
LRLN 12.8 (0.5) 4.2 (0.3) 5.0 (0.7)
Tflm baseline 9.82 (0.5) 2.35 (0.5) 2.115 (0.4)
LRLN 6.0 (0.3) 1.3 (0.4) 1.8 (0.7)
V,/T| baseline 16.32 (0.3) 3.96 (0.6) 3.255 (0.5)
LRLN 12.7 (0.4) 2.7 (0.8) 3.2 (1.2)
PIF (L/sl baseline 19.93 (0.3) 6.57 (0.4) 3.16 (0.8)
LRLN 12.3 (0.4) 3.5 (0.7) 3.6 (0.7)
IF“ (L/s) baseline 24.14 (0.2) 9.04 (0.5) 5.83 (0.7)
LRLN 13.5 (0.3) 5.7 (0.3) 6.2 (0.8)
IF,,5 (L/sl baseline 21.9 (0.5) 6.69 (0.4) 5 28 (0 7)
LRLN 15.0 (0.4) 5.0 (0.6) 6 6 (1.2)
PEF/PIF baseline 20.80 (0.5) 7.38 (0.4) 5.1 (0.6)
LRLN 12.2 (0.3) 5.5 (0.6) 3.8 (0.8)
EFw/IF50 baseline 31.86 (0.5) 11.81 (0.3) 7.44 (0.5)
LRLN 16.0 (0.4) 8.0 (0.3) 7.2 (0.7)
Vol PIF (Ll baseline 66.1 (0.5) 10.37 (0.4) 16.42 (1.4)
LRLN 68.0 (0.6) 5.2 (0.4) 7.2 (0.4)

 

iCV = Mean coefficients of variation calculated from the coefficient of variation
for the 10 flow-volume loops analyzed for each horse at each exercise
period.

(CV) = Coefficient of variation of the calculated mean coefficient of variation of
the 6 horses.

See Table 2 for key.

97

21

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[:IBaseline :
ILRLN * T
“-1.5- E T \ §
E . N 1% .§
0.5— \ \ \
31 CIBaseline :
2.5— ILRLN : T\
T
2_ s
as, \

 

 

 

 

 

 

 

 

 

 

 

Figure 9. Peak expiratory flow-to-peak inspiratory flow ratio (PEF/PIF) and expiratory-
to—inspiratory flow ratio at mid-tidal volume (EFw/lFsol before and after LRLN at rest
and exercise. Baseline = prior to LRLN; LRLN = 14 days after LRLN; * = data
significantly (P < 0.05) different from baseline measurement at same speed;
" = data significantly (P < 0.01) different from baseline measurement at same
speed.

98

UBasellne

EILRLN
°°‘ 1

PIF (Llsec)

 

 

 

 

 

1001 Daeeellm
Guam

“-1
a
a
m ,
A B

 

 

 

 

 

 

 

 

 

C
‘°°' Deeeellm
BLRLN
80* i
§ 001
a .
o _, *
I; ‘0
ma
0 r‘m ,
A B C

Figure 10. Peak inspiratory flow (PIF), inspiratory flow at mid-tidal volume (lFso), and
inspiratory flow at and tidal volume plus 25% tidal volume (IF,5) before and after LRLN
at rest and exercise. Baseline = prior to LRLN; LRLN = 14 days after LRLN; * = data
significantly (P < 0.05) different from baseline measurement at same speed; “ =
data significantly (P < 0.01) different from baseline measurement at same speed.

99

E. Discussion

Quantitative evaluation of upper airway function during exercise has
been repeatedly performed by measurement of peak airflow rates, peak
pressure changes, and the calculation of impedance in clinically normal horses
and horses with upper airway obstruction induced by LRLN (Derksen et
al,1986; Shappell et al,1988; Belknap et al,1990; Fulton et al,1991;
Mangseth,1984; Funkquist at al,1988). This standard technique of measuring
upper airway flow mechanics necessitates the invasive and time-consuming
placement of a transtracheal catheter. Furthermore, the time required for
calculation of inspiratory and expiratory impedances makes the technique
inconvenient. Measurement of tracheal pressures via a nasotracheal catheter
provides a less invasive alternative to percutaneous placement (Williams et
al,1990:ll. Such measurement of tracheal pressures has been advocated as a
clinically useful test of upper airway function (Williams et al,1990:l; ll).
However, tracheal pressure is not only influenced by upper airway geometry,
but also by the pattern of breathing and airflow rates (Fry and Hyatt,1960).
Therefore, changes in tracheal pressure in exercising horses are difficult to
interpret.

In contrast, the measurement of TBFVL in horses at rest and during
exercise was non-invasive and easily performed. As noted, horses adapted
easily to the treadmill and tolerated the face mask well (Morris and
Seeherman,1991; Shappell et al,1988; Belknap et al,1990; Fulton et al,1991).

Furthermore, with the aid of recently developed computer software, analysis

100

of TBFVL was convenient and rapid, and was completed immediately after the
measurement period.

Wigwam—Tidal breathing flow-volume loops
recorded from horses at rest indicated inter- and intrahorse variability in shape,
tidal volume, and airflow rates. Interestingly, the inspiratory limb of the TBFVL
was uniphasic in some horses and polyphasic in others (Figure 5). Similar
variations in TBFVL shape were reported in resting ponies by Art and Lekeux
(1988). In contrast, TBFVL in resting dogs have uniphasic inspiratory and
expiratory limbs (Amis and Kurpershoek,1986). In dogs, Amis and Kurpershoek
described 4 distinct TBFVL shapes, which differed principally on the basis of
location of the peak inspiratory flow. Similar variations in the location of peak
inspiratory flow were found in our horses. Differences in TBFVL shapes
between these 2 species are probably a reflection of differing breathing
strategies known to exist between dogs and horses (Kosch,1985l.

Coefficients of variation of TBFVL indices ranged from 9.82 to 66.1%
in horses at rest and were much higher than those seen at exercise. This
reflects the greater variability in breathing strategy adopted by resting horses,
compared with exercising horses. This large coefficient of variation limits the
clinical usefulness of TBFVL obtained in standing horses. Similar large
coefficients of variation (1 to 46%) have been reported in dogs (Amis and
Kupershoek,1986l.

WWW—After LRLN, horses at rest did not have

significant changes in TBFVL indices or in Z, and 25. At rest, during quiet

101

breathing, airway pressures are insufficient to cause dynamic collapse of the
arytenoid cartilage in horses with LLH (Robinson and Sorenson,1978; Derksen
et al,1986). Likewise, no differences in loop shapes were appreciable before
and after LRLN in horses at rest. The TBFVL are also insensitive in detection
of upper airway obstruction in conscious dogs; thus, the clinical usefulness of
the technique is limited to patients with advanced upper airway obstruction
(Amis and Kurpershoek 1986; Amis et al,1986). From the aforementioned
discussion, it is clear that determination of TBFVL does not enable detection of
LLH in standing horses.

WWW-In human medicine.
determination of maximal FVL is considered to be a non-invasive, rapid,
sensitive, and specific clinical diagnostic tool, aiding in the detection and
evaluation of upper airway obstruction (Miller and Hyatt,1969; Miller and
Hyatt,1973; Kashima,1984). However, to achieve a maximal FVL, the
technique requires patient cooperation and maximal respiratory effort, resulting
in high airflow rates (Miller and Hyatt,1969l.

In exercising horses, airflow rate is greatly increased whereas variability
in breathing strategy is reduced (H6rnicke et al,1983; Hornicke et al,1987).
Therefore, we hypothesized that the sensitivity of TBFVL in detecting upper
airway obstruction in non-cooperative patients, like our horses, may be
markedly enhanced during strenuous treadmill exercise. The TBFVL obtained
in healthy, non-athletic people during exercise only impinge on their maximal

FVL near and expiration (Olafsson and Hyatt,1969; Babb et al,1991).

102

However, in well-trained young men, TBFVL obtained during maximal exercise
approach much of the maximal expiratory flow curve of their maximal FVL
(Grimby et al, 1971 ). These exercising TBFVL in people are associated with
near-maximal airflow rate and tidal volume that approximates 50% of vital
capacity. Similarly, in our study, airflow rate approached maximal flow rate
(Leith,1976l and tidal volume was approximately 50% of predicted vital
capacity (Davidson et al,1986). We reasoned that if a similar ventilatory
response to exercise is also true in performance horses, TBFVL obtained during
maximal exercise may be as sensitive as maximal FVL for detection of airway
obstruction. In this study, the ability of TBFVL determination to detect upper
airway obstruction in horses at rest and during exercise, using TBFVL analysis,
was compared with the standard technique that involves measurement of peak
driving pressure, airflow change, and calculation of inspiratory and expiratory
impedances.

To standardize the exercise protocol, maximal heart rate was first
determined in each individual horse. In subsequent measurement protocols,
horses were exercised at treadmill speeds correlating to HROJM and HR“.
Exercise was standardized in this way in an attempt to minimize variability in
airflow rates between horses at each measurement period. Indeed, coefficients
of variation for TBFVL indices obtained during exercise in our horses were
smaller than those obtained in resting horses. The progressive decrease in
coefficients of variation for TBFVL indices, with increasing exercise level (Table

3), indicates that respiratory patterns became less variable during exercise. The

103

greater consistency of TBFVL seen during exercise, compared with that of
TBFVL at rest is likely attributable to the high ventilatory requirements imposed.
Furthermore, breathing pattern is imposed by a coupling of respiratory rate and
stride frequency (Hornicke et al,1983); thus the observed decrease in
coefficients of variation for TBFVL indices may, in part, reflect the decreased
variation in gait between horses at exercise period C. The shape of the TBFVL
during exercise is likely influenced by gait, stride frequency, and the mechanical
events related to locomotion, such as acceleration, deceleration, and weight
bearing (Attenburrow,1983). Because we did not measure aspects of
locomotion such as stride frequency and length and because we did not
standardize gait, the effects of locomotion on shape of the TBFVL cannot be
determined.

Different gaits observed between horses within exercise periods may
potentially influence values for various respiratory variables; including f, VT,
airflow, and airway pressures. Although, differences in VT and f in apparently
normal horses, using trotting/pacing and galloping gaits, has been reported
(Hornicke et al,1983) in our horses the different gaits observed at period B after
LRLN did not result in large interhorse variability of values for the respiratory
variables measured. This is most likely attributable to the uncoupling of the 1 :1
ratio between respiratory rate and stride frequency, allowing galloping horses
to use a higher VT after LRLN.

The observed ventilatory patterns and TBFVL shapes seen at rest and

during exercise may reflect influences of rebreathing of gases, the added

104

impedance imposed by the mask-pneumotachograph breathing system, or a
combination of these 2 factors. Previous studies indicated that various types
of breathing apparatuses worn during exercise reduces respiratory frequency,
may change stride frequency-to-ventilation ratios, and alter arterial blood gas
tensions (Bisgard et al,1978; Bayley et al,1987). As neither arterial blood gas
nor end-tidal CO2 evaluation was measured in this study the extent of
rebreathing C02 and its effects on TBFVL shapes observed is unknown. The
pressure change across the mask-pneumotachograph at peak inspiratory airflow
rates in this study was approximately 10% of peak inspiratory transupper
airway pressures and, therefore, is likely to minimally effect ventilatory
response to exercise.

WWW—The increases in peak flow
rates, peak transupper airway pressures, and the minimal change in peak
impedance associated with increased exercise prior to LRLN have been reported
(Table 1) (Shappell et al,1988; Belknap et al,1990; Fulton et al,1991). This
study extended those observations and documented that a progressive increase
in exercise intensity was associated with an increase in airflow rates
throughout the respiratory cycle. Indeed, because flow ratios were not
significantly affected by increasing exercise, these data indicate that inspiratory
and expiratory flows may be increased equally.

Effeet ef LRLN in exercising nereee—After LRLN, Pm, 2., and T./TTOT
increased and PIF, TE/T" and VT/T. decreased in exercising horses, whereas Pug,

PEF, and 2E remained unaffected (Tables 1 and 2). These data indicate that

105

LRLN causes upper airway obstruction on inhalation as a result of dynamic
collapse of the left arytenoid cartilage. The validity of LRLN as a model of
naturally acquired LLH was confirmed by a recent report (Williams et
al,1990:ll). The tracheal pressure changes measured after experimentally
induced LLH were comparable to those measured in horses with naturally
acquired LLH.

Flow-volume loop shapes obtained from exercising horses after LRLN
were easily distinguished from those recorded prior to LRLN at the same level
of exercise. The TBFVL shape was typified by a pronounced flattening of the
inspiratory flow curve. After the initial PIF, inspiratory flow decreased markedly
and remained constant until immediately before the onset of exhalation. This
flow profile indicates flow limitation during the last two-thirds of the inspiratory
portion of the tidal volume. Videoendoscopy during exercise in horses with
induced LLH indicates that, on exhalation, the affected arytenoid cartilage is
moved abaxially, probably as a result of intraluminal pressures that are positive
with respect to atmospheric pressure. Immediately after the onset of
inhalation, the affected arytenoid cartilage moves axially to almost completely
obstruct the rima glottidis. This cartilage remains in this axial position
throughout inhalation. The initial PIF observed in TBFVL obtained from
exercising horses after LRLN probably develops before the axial movement of
the affected cartilage has been completed. The period of flow limitation
apparently corresponds to the period during which the affected arytenoid

cartilage is in the maximally displaced position.

106

The expiratory limb of the TBFVL in exercising horses was unaffected
by LRLN. The shape of the inspiratory portion of the TBFVL, coupled with the
preservation of the shape of the pre-LRLN expiratory flow curve, is
characteristic of dynamic upper airway lesions (Miller and Hyatt,1969). The
effects of upper airway obstruction on FVL depends on size and location of the
lesion and the fixed or variable nature of the lesion (Miller and Hyatt,1969).
Miller and Hyatt,(1969l proposed a classification of upper airway obstruction
and described how these abnormalities affect the FVL. They classified
obstructions into 3 functional groups on the basis of pressure-flow relations
around the lesion site: fixed obstruction; variable extrathoracic obstruction; and
variable intrathoracic obstruction. Variable obstructions are substantially
affected by changes in transmural pressure and subsequently may result in
dynamic collapse. Thus, intrathoracic lesions result in predominantly expiratory
flow limitation, whereas extrathoracic lesions cause a reduction in inspiratory
gas flow. Fixed obstructions exert flow-limiting effects on both phases of
respiration. The changes in the shape of TBFVL induced by LRLN in exercising
horses were characteristic of variable extrathoracic obstruction.

The observed changes in TBFVL shape are supported by the numerical
data (Table 2). The TBFVL indices describing exhalation were unaffected by
LRLN. In contrast, in exercising horses, LRLN resulted in significant reduction
of PIF, IF“, and IF”, and an increase in PEF/PIF and EFso/IFSO. The largest
percentage change was observed in the IFso, which decreased by 40.4 and

63.5% at periods B and C, respectively. The change in the EFm/IFI50 after LRLN

107

was highly significant (P < 0.01) even at measurement period 8. This
indicates that these 2 loop indices may be most useful in the detection of
dynamic upper airway obstruction in exercising horses. Originally identified as
a potential sensitive indicator of variable extrathoracic obstruction in human
beings (Jordanoglou and Pride,1968), the IF,50 and EFm/IF$0 may reflect near-
maximal inspiratory airflow limitation associated with the point of complete
dynamic collapse of the left arytenoid cartilage (Derksen,1988). This earlier
prediction was recently confirmed by a large study in people that concluded
that the IFI50 is the most reliable indicator of dynamic upper airway obstruction
(Polverino et al,1990). The mean value of 2.56 for EFw/IF60 in our study was
similar to values reported in a series of clinical cases involving human patients
with dynamic extrathoracic lesions (Kashima,1984; Miller and Hyatt,1969).
However, in such adult human studies, maximal FVL are achieved, whereas in
our study, TBFVL were obtained through exercise. Comparison of these data
therefore indicates that TBFVL obtained during high-speed exercise may be as
sensitive for detection of upper airway obstruction as is maximal FVL.

In conclusion, results of this study indicate that analysis of TBFVL in
exercising horses allows detection of upper airway obstruction induced by
LRLN. After LRLN in exercising horses, the shape of the TBFVL was
characteristic for a variable extrathoracic obstruction. lndices of mid-tidal
inspiratory flow (IF,so and EFso/IFSO) were most sensitive for detection of upper
airway obstruction. Because the technique was quantitative, rapid, non-

invasive, and well tolerated by horses, TBFVL analysis may become a useful

108

diagnostic tool in conjunction with upper airway endoscopy during exercise for

evaluation of naturally acquired upper airway obstruction in horses.

VI. EVALUATION OF PARTIAL ARYTENOIDECTOMY AS A TREATMENT

FOR EQUINE LARYNGEAL HEMIPLEGIA

A. Summary

The efficacy of partial arytenoidectomy was assessed in 6 Standardbred
horses, with surgically-induced laryngeal hemiplegia, at rest (period A) and
during exercise at speeds corresponding to maximum heart rate (period C) and
75% of maximum heart rate (period 8). Peak expiratory and inspiratory airflow
rate (PEF and PIF), and expiratory and inspiratory transupper airway pressure
IPUE and Pu.) were measured and, expiratory and inspiratory impedance (2.; and
2.) were calculated. Simultaneously, tidal breathing flow-volume loops (TBFVL)
were acquired using a respiratory function computer. lndices derived from
TBFVL included, airflow rates at 50 and 25% of tidal volume (EFBO, le, EF25,
and |F25I and the ratios of expiratory to inspiratory flows. Measurements were
made before left recurrent laryngeal neurectomy (baseline), 2 weeks after left
recurrent laryngeal neurectomy (LRLN), and 16 weeks after left partial
arytenoidectomy coupled with bilateral ventriculectomy (ARYT).

After LRLN, during exercise periods B and C, Zl and the ratio of EFm/IFso
significantly increased and PIF, IF.so and IF25 significantly decreased from
baseline values. Sixteen weeks after ARYT, ZI returned to baseline values at
exercise periods B and C. Although PIF, IF”, ":25: PEF/PIF, and EFm/IF50
returned to baseline values at exercise period 8, these indices remained

significantly less than baseline measurements during exercise period C. After

109

110

ARYT, TBFVL shapes from horses at exercise period C approached that seen
at the baseline evaluation.

Partial arytenoidectomy improves upper airway function in exercising
horses with surgically-induced left laryngeal hemiplegia, though qualitative and
quantitative evaluation of TBFVL's suggest that some flow limitation remains
at near maximal airflow rates. These results suggest that, although the
procedure does not completely restore the upper airway to normal, partial
arytenoidectomy is a viable treatment option for failed laryngoplasty and

arytenoid chondropathy in the horse.

8. Introduction

Partial arytenoidectomy, the removal of the arytenoid and corniculate
cartilages while leaving the muscular process in_s_i_te, was first introduced as a
treatment for left laryngeal hemiplegia (LLH) in 1845 (Liautard,1892). This
procedure was considered unsatisfactory because it was associated with a high
rate of post-operative complications. With the development of the
ventriculectomy (Williams,1911) and later prosthetic laryngoplasty (Marks et
al,1970:ll), the low morbidity and mortality of these procedures obviated the
use of arytenoidectomy as a treatment for LLH. Subtotal arytenoidectomy, the
removal of the arytenoid body leaving the muscular process and corniculate
cartilage intact, and partial arytenoidectomy have been recommended as
treatments for arytenoid chondropathy (Haynes et al,1980; Tulleners et

al,1988:l), arytenoid malformations and failed prosthetic laryngoplasty (White

111

and Blackwell, 1 980; Haynes et al,1984). The merits of these two techniques
have been evaluated subjectively based on assessment of performance and
post-operative complications (Haynes et al,1984; Speirs,1986; Tulleners et
aL1988flL

In 1990, Belknap et al, quantitatively evaluated subtotal arytenoidectomy
and demonstrated that subtotal arytenoidectomy failed to improve upper airway
function in horses with surgically-induced LLH exercising submaximally. To
date, there has been no objective evaluation of the efficacy of partial
arytenoidectomy.

Upper airway function in exercising horses has been quantitated by
measurement of transupper airway pressure and airflow, and calculation of
inspiratory and expiratory impedance (Derksen et al, 1986). As reported in
chapter V, analysis of tidal breathing flow-volume loops (TBFVL), is a useful,
sensitive, convenient and non-invasive technique, for quantitation of upper
airway obstruction in exercising horses. In the present study, the effect of
partial arytenoidectomy on upper airway function was evaluated by TBFVL
analysis, measurement of upper airway impedance and videoendoscopic
examination of the larynx in strenuously exercising horses with surgically-

induced LLH.

1 12
C. Materials and Methods

fierees—As outlined in chapter V, the 6 adult Standardbred horses used
in this project were maintained on pasture between surgical procedures and
measurement protocols.

Wages—Airflow rates, tidal volume and transupper
airway pressure were recorded as described in chapter V. Simultaneous with
measurements of airflow and airway pressure, TBFVL were obtained for horses
at rest and during exercise as outlined in chapter V. The shape of TBFVL
generated and the mean :1; SEM of measured values and derived indices were
also obtained.

ExeeflmegteLdeejgn—As described in the preliminary study (chapter V),
treadmill speeds at which maximum heart rate (Hi-1m) and 75% of maximum
heart rate (HRmM) occurred were estimated.

All data were recorded during the last minute of each 2-min
measurement period. Measurements were made while horses were at rest,
standing adjacent the treadmill (period A). Subsequently, horses were
exercised on the treadmill with a 3—degree incline at 4 m/sec for 2 min, then at
the treadmill speed corresponding to HRMW for 2 min (period B). Horses
rested for 1 min and then exercised at a treadmill speed corresponding to HRM
(period C) for 2 min.

Measurements obtained before left recurrent laryngeal neurectomy
(baseline) and 2 weeks after LRLN (LRLN) reported in Chapter V were designed

to evaluate the usefulness of TBFVL indices for detection of upper airway

113

obstruction. Sixteen weeks after left partial arytenoidectomy and bilateral
ventriculectomy (ARYT), horses underwent the same measurement protocol as
for baseline and LRLN measurements. On the day following each measurement
of upper airway function, the larynx of each horse was examined using
videoendoscopy at rest and during exercise at HRmu.
Surgical Procedures

Induction of laryngeal hemiplegia was performed by transection of the
left recurrent laryngeal nerve (Derksen et al,1986). Left partial
arytenoidectomy was performed as follows. Anaesthesia was induced with
guiafenesin (Aceto Chemical Co, Flushing, NY.) and thiamylal (Biotal, Bioceutic
Division, Boehringer lngelheim Animal Health Inc, St Joseph, Mo.). The animals
were intubated via a mid-cervical tracheotomy and anaesthesia was maintained
with halothane (Fluothane, Ayerst Laboratories Inc, New York, NY.) in oxygen.
Partial arytenoidectomy was performed as previously described by Mcllwraith
and Turner (1987), except that the corniculate cartilage and its mucosa were
removed en bloc, and a bilateral ventriculectomy was performed. Via a ventral
Iaryngotomy a fiberoptic retractor (Shea fiberoptic retractor, Stryker Corp,
Kalamazoo, Mich.) was used to illuminate the lumen of the larynx. The mucosa
was incised beginning at the ventral aspect of the junction of the corniculate
and arytenoid cartilages and extending caudad along the ventral border of the
arytenoid cartilage. This incision was continued along two-thirds of the caudal
border of the arytenoid cartilage. The cranial aspect of the incision was

extended along the entire junction of the corniculate and arytenoid cartilages.

114

The laryngeal mucosa was elevated from the luminal side of the arytenoid
cartilage using a blunt Asheim knife. A separate incision was then made
extending from the left laryngeal ventricle dorsally along the rostral border of
the corniculate cartilage. The arytenoid and corniculate cartilages were then
separated. The corniculate cartilage and attached luminal mucosa was removed
en bloc, after dissecting the laryngeal musculature from its lateral aspect. The
body of the arytenoid cartilage was similarly dissected from its lateral
musculature and transected from its muscular process and removed, leaving the
muscular process and intact luminal mucosa. Bilateral ventriculectomy was
then performed, and all free mucosal edges were apposed with 2-0
polydioxanone (PDS, Ethicon Inc, Somerville, NJ.) in a simple continuous
pattern. Horses were treated preoperatively and postoperatively with procaine
penicillin (Procaine penicillin G, Pfizer Inc, New York, NY) (10,000 lU/kg lM BID
for 5 days), and phenylbutazone paste (Butazolidin paste, Coopers Animal
Health Inc, Kansas City, Mo.) (3 mg/kg P0 BID for 5 days). Food was withheld
for 48 hours after surgery and tracheotomy tubes (Dyson self-retaining tracheal
tubes, Coopers Animal Health Inc, Mishawaka, lnd.) were removed following
endoscopic confirmation of an adequate airway at 24 hours. The larynx of
each horse was endoscopically examined at rest 1, 4, 8 and 16 weeks
postoperatively.

Statistical analysis of data obtained from this study were analyzed in the

same manner as described in the preliminary study in chapter V.

1 15
D. Results

The HR,” and l-IR0_75,,_,, averaged 225.5 :I: 4.92 bpm (mean +/- SEM)
and 166.8 :I: 3.9 bpm at treadmill speeds of 11.0 :1: 0.2 m/s and 6.2 :I: 0.7
m/s, respectively. The effect of exercise on all physiological parameters is
shown in Table 1 and discussed in chapter V.

The effect of LRLN and partial arytenoidectomy on TBFVL shape is
shown in Figure 11. The flattened shape of the inspiratory flow curve after
LRLN (b) was altered after partial arytenoidectomy (c), to a shape that
approached the inspiratory flow curve (a) generated at baseline evaluation. The
effects of LRLN on the indices of upper airway function are shown in Table 4
and Figures 12-16.

Sixteen weeks after partial arytenoidectomy and bilateral
ventriculectomy, PU. and Z| were not significantly different from baseline values
during exercise periods 8 and C (Table 4;Figure 12). Furthermore, all indices
describing TBFVL returned to baseline (ore-LRLN) values at period 8 (Table
4;Figures 14-16). However, PIF, lFso, IF25, PEF/PIF, EFso/IFSO, TE/T" T./TTOT, and
V7/Tl remained significantly different from the baseline value at exercise period
C (Table 4;Figures 12,14,15). Expiratory flow rates were unaffected by either
LRLN or ARYT at all exercise levels (Figures 13,16).

Left laryngeal hemiplegia was confirmed using videoendoscopic
evaluation of the larynx at rest and during exercise in all horses. Following
partial arytenoidectomy and bilateral ventriculectomy, no post-operative

complications were encountered other than minor dehiscence of the sutured

116

mucosa in two horses. Evidence of dysphagia or coughing was not noted
either at rest or during exercise. Endoscopic examination in the resting horse
at 16 weeks showed complete mucosal healing. The right corniculate cartilage

appeared rotated toward the left side in all horses at rest and during exercise.

 

 

 

C

0

up. I—-

;J

IO

L

or!

a F—

x A

I“ >
0H
U
E _.
\
H
o

I 1 g

o. l l 1
O
L“.
I l—‘
O
H

8 LL

H

u _

I‘U

L

H

(1

m —

C

H

L Volume (4.00 1/div)

Figure 11. TBFVL's from a horse at baseline (a), after LRLN (b), and following
partial arytenoidectomy and bilateral ventriculectomy (c). TBFVL were
generated at a speed corresponding to maximal heart rate.

117

 

 

Table 4 - Effect of surgery on measured and calculated variables from 6 horses with surgically
induced left laryngeal hemiplegia before (pm-operative) and after (post-operative) left
partial arytenoidectomy and bilateral ventriculectomy.

Exercise
Variable Period Baseline: Pre-operativet Post-operativet
\7E A 151.41 1 15.01 140.75 1 7.69 138.40 1 12.12
(litre/min) B 950.23 1 59.02 808.40 1 68.29 963.80 1 39.60
C 1,295.97 1 127.52 1,027.60 1 71.17' 1178.50 1 38.58
Pu. A 1.94 1 0.22 2.16 1 0.32 1.87 1 0.23
Icm of H20) 8 22.29 1 1.15 49.40 1 4.08. 26.12 1 3.20f
C 38.57 1 3.93 62.73 1 6.70' 39.99 1 4.39f
PIF A 5.09 1 0.34 5.84 1 0.39 6.49 1 0.55
(litre/sec) B 51.19 1 4.05 36.58 1 2.74’ 43.28 1 1.82
C 75.52 1 9.35 40.87 1 3.23“ 50.09 1 2.160
2. A 0.39 1 0.03 0.38 1 0.06 0.30 1 0.04
Icm of H,O/litre/sec) B 0.46 1 0.06 1.41 1 0.20" 0.62 1 0.101
C 0.53 1 0.04 1.63 1 0.28‘ 0.81 1 0.101
IFso A 5.24 1 0.31 4.84 1 0.18 5.82 1 0.31
(litre/sec) B 39.86 1 4.08 23.77 1 1.61' 37.87 1 1.95
C 65.98 1 8.47 24.09 1 1.03’ 36.24 1 1.30“
IF,s A 6.17 1 0.56 5.17 1 0.38 5.90 1 0.31
(litre/sec) B 49.77 1 2.70 34.34 1 2.14“ 42.03 1 1.92
C 70.78 1 5.04 35.53 1 1.64' 47.89 1 2.17'
PEF/PIF A 1.33 1 0.07 1.35 1 0.10 1.25 1 0.07
B 1.13 1 0.06 1.58 1 0.05“ 1.19 1 0.021
C 0.97 1 0.07 1.70 1 0.03“ 1.36 1 0.02’1
EFm/IF50 A 1.25 1 0.11 1.11 1 0.15 1.13 1 0.12
B 1.19 1 0.11 2.15 1 0.12‘ 1.27 1 0.02f
C 1.08 1 1.07 2.56 1 0.04' 1.72 1 0.08‘1
TJI'. A 0.99 1 0.05 0.97 1 0.05 0.95 1 0.06
B 0.94 1 0.03 0.70 1 0.02' 0.83 1 0.01
C 0.98 1 0.05 0.65 1 0.01 0.73 1 0.01“
T,/T,o, A 0.51 1 0.01 0.52 1 0.01 0.52 1 0.02
B 0.51 1 0.01 0.59 1 0.01' 0.55 1 0.004
C 0.51 1 0.01 0.61 1 0.01' 0.58 1 0.004‘
V,/T. A 4.84 1 0.34 4.40 1 0.14 5.19 1 0.27
B 31.77 1 2.08 23.87 1 1.70“ 29.98 1 1.03
C 44.78 1 5.03 25.70 1 1.02. 34.47 1 1.38'

 

' Data significantly (P < 0.05) different from baseline (before left recurrent laryngeal neurectomy)
measurement at same speed. 1 Data significantly different from pro-operative measurement at same
speed. 1 Values represent mean and sem. V, = tidal volume, VE = minute ventilation, Pu. = peak
inspiratory pressure, PlF = peakinspiratory flow, 2. = peak inspiratory impedance, IFso = inspiratory
flow at 50% of V,, IF25 = inspiratory flow at 25% of V,, TE = expiratory time, T. = inspiratory time,
T101 3 tOtBI brath time

118

‘9‘?

Pu(cmofH,0)
B 8 8 S 8

.5
O
1

 

'8‘

PIF (Llsec)
«3 as g

9}

 

O
l

ZI (cm H,O/Iitre/sec)
-; a; .3

P
on

 

 

O
l

 

Figure 12. Peak inspiratory pressure (Pm), peak inspiratory flow (PIF), and
inspiratory impedance (2.) before and after LRLN, and after partial
arytenoidectomy at rest (A) and during moderate (B) and strenuous (C)
exercise. Baseline = prior to LRLN; LRLN = 2 weeks after LRLN; ARYT = 16
weeks after partial arytenoidectomy * = data significantly different from
baseline measurement at same speed (p < 0.05). f = data significantly
different from LRLN measurement at same speed (p < 0.05).

“fig

 

 

q

1

6f .o .5. 31

q . .
5 o 5 o
1

904‘
MV©0©O©

 

 

.......

EEEEEEE

     

  

sausage

«EV

 

OA-

_ q _
2 1 o

a Q Q
1356“: a so. an

and after partial

before and after LRLN,

Figure 13. Peak expiratory pressure (PUE), peak expiratory flow (PEF), and
impedance (2;)

expiratory
exercise. Baseline = prior to LRLN; LRLN = 2 weeks after LRLN; ARYT = 16

arytenoidectomy at rest (A) and during moderate (B) and strenuous (C)
weeks after partial arytenoidectomy.

120

10°“ UBaaeline

 

 

IF” (litre/sec)

 

 

1°01 Emeline
.LRLN

 

 

Figure 14. Peak inspiratory flow (PIF), inspiratory flow at mid-tidal volume
(lFsol. and inspiratory flow at 25% tidal volume (le5) before and after LRLN,

and after partial arytenoidectomy at rest (A) and moderate (B) and strenuous
(C) exercise. Baseline = prior to LRLN; LRLN = 2 weeks after LRLN; ARYT =

16 weeks after partial arytenoidectomy. * = data significantly different from
baseline measurement at same speed (p < 0.05).

121

2 - Baseline
LRLN *

  

A
l
o'e‘e'e‘e‘e'e'ii'
02030292020324)

01

I
.'
O
20.

:9

0".
202020

 

7'
r’
I

_ IBaseline *
2.51 ILRLN
4 @ARYT

   

vb.

 
   

   

0'
O O
0.4

.6

v

    
   
 

'0'.
O
.9

?

    
  

     

  

'9
9:9:

 
 
   
 
    
   

'e'e'e
0 O
. 96?.

  

 

V/l/l/A

we.
.9
'efi? . .

Figure 15. Peak expiratory flow I peak inspiratory flow ratio (PEF/PIF) and
expiratory/ inspiratory flow ratio at mid-tidal volume (EFso/IFSO) before and after
LRLN, and after partial arytenoidectomy at rest (A) and moderate (B) and
strenuous (C) exercise. Baseline = prior to LRLN; LRLN = 2 weeks after LRLN;
ARYT = 16 weeks after partial arytenoidectomy * = data significantly
different from baseline measurement at same speed (p < 0.05M = data
significantly different from LRLN measurement at same speed (p < 0.05).

122

9
wonononononon nounononon.

 

 

 

 

w
as; “an.

204
O

 

(C) exercise. Baseline = prior to LRLN; LRLN = 2 weeks after LRLN; ARYT =

and after partial arytenoidectomy at rest (A) and moderate (B) and strenuous
16 weeks after partial arytenoidectomy.

Figure 16. Peak expiratory flow (PEF), expiratory flow at mid-tidal volume
(EFso), and expiratory flow at 25% tidal volume (EFZS) before and after LRLN,

123

Sixteen weeks post-operatively, videoendoscopy during exercise at HRM
revealed an apparently adequate airway in all horses. Slow motion playback of
the video recording made during exercise revealed minor centripetal movement
of the left side of the larynx on inspiration in 4 horses with more pronounced
inward movement observed in 2 horses. The latter two horses had the highest

Z, values during exercise period C.

E. Discussion

The aim of arytenoidectomy is to provide an airway of maximal cross-
sectional area during exercise while preserving normal swallowing
(Speirs,1986). Although Haynes et al,1984 suggested that these objectives
can best be achieved using subtotal arytenoidectomy, dynamic inspiratory
collapse of the remaining corniculate cartilage, may result in continued airway
obstruction during exercise (Stick and Derksen,1989; Belknap et al,1990).
Partialarytenoidectomyreportedlyeliminates the collapsing corniculatecartilage
(Stick and Derksen,1989) and may allow return to racing in approximately 50-
7596 of cases (Speirs,1986; Tulleners et al,1988:l). The technique has been
associated with an unacceptably high incidence (36%) of dysphagia and/or
coughing (Speirs,1986). but a large recent study found a much lower incidence
of these complications (Tulleners et al,1988:l). Therefore, partial
arytenoidectomy may be the preferred treatment for arytenoid chondritis and
failed prosthetic laryngoplasty in performance horses, and was chosen for

evaluation in the present study. We evaluated our horses 16 weeks after

124

surgery, because preliminary studies, using endoscopic examination of the
larynx with horses standing, confirmed that the diameter of the rima glottidis
appeared to continued to enlarge for up to 4 months after partial
arytenoidectomy (Speirs,1986).

During moderate and strenuous exercise (Periods B and C, respectively),
after partial arytenoidectomy, the airflow-generating transupper airway
inspiratory pressures and inspiratory impedance did not differ from baseline
values (Table 4; Figure 12), suggesting that partial arytenoidectomy restored
upper airway function in horses with induced LLH. Additionally, during Period
C, partial arytenoidectomy reversed the decrease in V 5 observed after LRLN
(Table 4). However, although ZI and PU. returned to baseline values after partial
arytenoidectomy, PIF remained significantly reduced suggesting that this
operation did not completely restore upper airway geometry. This became
apparent only at the higher exercise level (Period C). Lack of tissue support
may have allowed partial dynamic collapse of the left side of the larynx after
partial arytenoidectomy.

Further information regarding the effects of partial arytenoidectomy on
upper airway function was obtained using TBFVL analysis. After partial
arytenoidectomy, measurements of inspiratory airflow during Period 8 did not
differ significantly from baseline measurements (Table 4; Figures 14,15). In
contrast, during more strenuous exercise (Period C), PIF and the TBFVL indices
describing airflow throughout inspiration lle and IF”) remained significantly

less than baseline values. Although PEF/PIF and EFm/IFI50 (Table 4; Figure 15),

125

were less than values obtained after LRLN, they remained significantly greater
than baseline values.

The TBFVL in Figure 11 illustrates the reduced airflow throughout
inspiration and preservation of airflow during expiration after LRLN. After
partial arytenoidectomy, the TBFVL shape at Period C was characterized by an
inspiratory curve intermediate between those observed post-LRLN and at
baseline. Therefore, TBFVL evaluation suggests that partial arytenoidectomy
eliminated the LLH-induced dynamic upper airway obstruction at submaximal
exercise (Period 8), but that some inspiratory flow limitation persisted during
more strenuous exercise. When extending these findings to racing conditions
it may be concluded that partial arytenoidectomy does not completely restore
upper airway function in horses with LLH.

The leftward displacement of the right corniculate cartilage observed at
rest and during exercise may have resulted from caudal retraction of rostral
mucosa of the left side of the larynx, while leaving the transverse arytenoid
ligament intact (Speirs,1986). The varied degrees of centripetal movement of
the left side of the larynx during inspiration was probably the result of the semi-
rigid nature of the tissue of the left side of the larynx after partial
arytenoidectomy and the lack of left side abductor function. This centripetal
movement, which appeared more pronounced in horses with more abundant
mucosa, in conjunction with an airway of less than normal cross-sectional area,
may explain the partial inspiratory flow limitation. The results of the present

study support previous observations by others (Stick and Derksen,1989;

126

Belknap et al,1990) that unsupported structures, such as the corniculate
cartilage, are a significant cause of upper airway obstruction after subtotal
arytenoidectomy. Therefore, we recommend, removal of as much laryngeal
mucosal as practical during surgery while permitting primary closure under
minimal tension.

Although the decrease in 2. that followed partial arytenoidectomy
parallelled the decrease in 2. after prosthetic laryngoplasty (Shappell et
al,1988), accurate comparison of the efficacy of these two techniques in
improving upper airway function in horses with LLH is impossible due to the
higher treadmill speeds, higher peak airflows and lower treadmill incline
reported in the present study. Despite these differences in exercise conditions,
comparisons of upper airway impedance between these two studies would
suggest that the two different surgical techniques have similar efficacy. In the
present study, there were obvious discrepancies in the results from TBFVL and
the calculation of 2' during period C. Detection of persistent partial inspiratory
obstruction after partial arytenoidectomy using TBFVL analysis, but not by
evaluation of 2., indicates that the former technique is a more sensitive measure
of upper airway function. Re-evaluation of prosthetic laryngoplasty for the
treatment of LLH using TBFVL analysis may provide more detailed assessment
of the ability of this surgical technique to restore upper airway function.

The horses in our study showed no evidence of coughing or dysphagia.
However, in clinical cases, pre-existing laryngeal and peri-Iaryngeal pathology

associated with arytenoid chondropathies and failed prosthetic laryngoplasty

127

may influence the incidence of serious post-operative complications and the
efficacy of partial arytenoidectomy in restoring upper airway function
(Speirs,1986; Tulleners et al,1988:l). Based on studies of upper airway flow
mechanics, videoendoscopy in exercising horses and minimal post-operative
complications, we recommend partial arytenoidectomy as the treatment of

choice for arytenoid chondropathy and failed prosthetic laryngoplasty.

VII. CONCLUSIONS AND RECOMMENDATIONS

The results of this project provide important preliminary evidence for the
usefulness of flow-volume loop analysis in the evaluation of upper airway
obstructions in horses during exercise. The instantaneous generation of TBFVL
and their flow indices by specifically programmed computer software facilitates
the clinical use of this objective diagnostic tool. The significant change in
TBFVL indices, seen with airway obstruction, in conjunction with the non-
invasive nature of this test are a testimony for the use of this respiratory
function test in a clinical setting.

The TBFVL acquired during exercise, following LRLN, are in agreement
with laryngeal function observed videoendoscopically after induced and
naturally occurring cases of LLH. Our understanding of the pathophysiology of
various other upper airway obstructions, such as dorsal displacement of the
soft palate, arytenoid chondropathy and epiglottic entrapment, may allow
prediction of the shape of TBFVL acquired from such horses during exercise.
Thus TBFVL analysis of these other causes of reduced exercise tolerance and
poor performance may provide objective information as to nature and severity
of the airway obstruction.

Although induced LLH, used in this project, represents a severe form of
airway obstruction, it appears that TBFVL analysis may be a relatively sensitive
indicator of airway obstruction. Following partial arytenoidectomy, Zl returned

to baseline values, but TBFVL indices improved only partially. This suggests

128

129

that TBFVL analysis is more sensitive than the measurement of Z| in detection
of upper airway obstruction. The clinical use of this technique may be of value
in the detecting the effect of upper airway lesions such as recurrent laryngeal
neuropathy, arytenoid chondropathy, dorsal displacement of the soft palate,
epiglottic entrapment and pharyngeal lymphoid hyperplasia. The use of TBFVL
analysis to evaluate these clinical entities may provide insight as to there
significance as causes of airway obstruction.

Further studies evaluating the variability of TBFVL indices in a larger
population of horses would provide an important step toward establishing a
"normal" range of values for the upper airway. Investigation into the effect of
the number of representative TBFVL evaluated and the effect of using varied
levels of signal filtering may also provide important information to allow valid

evaluations of clinical cases of airway obstruction.

LIST OF REFERENCES

LIST OF REFERENCES

Abrahamsen EJ, Bohanon TC, Bednarski RM, Hubbell JAE, Muir WW. Bilateral
arytenoid cartilage paralysis after inhalation anesthesia in a horse. JA VMA
197: 1363-1365, 1990.

Abramson AL, Goldstein MN, Stenzler A, Steel A. The use of the tidal
breathing flow volume loop in laryngotracheal disease of neonates and infants.
Laryngoscope 92: 922-926, 1982.

Adler S, Wohl ME. Flow-volume relationship at low lung volumes in healthy
term newborn infants. Pediatrics 61: 636-640, 1978.

Amis TC, Kurpershoek C. Tidal breathing flow-volume loop analysis for clinical
assessment of airway obstruction in conscious dogs. Am J Vet Res 47: 1002-
1006. 1986.

Amis TC, Smith MM, Gaber CE, Kurpershoek C. Upper airway obstruction in
canine laryngeal paralysis. Am J Vet Res 47: 1007-1010, 1986.

Archer RM, Lindsay WA, Duncan ID. A comparison of techniques to enhance
the evaluation of equine laryngeal function. Equine Vet J 23: 104-107, 1991.

Archer RM, Lindsay WA, Duncan ID. Equine laryngeal hemiplegia—Endoscopic
survey of 400 draft horses. Vet Surg abstract 18: 62-63, 1989.

Art T, Lekeux P. Respiratory airflow patterns in ponies at rest and during
exercise. Can J Vet Res 52: 299-303, 1988.

Art T, Serteyn D, Lekeux P. Effect of exercise on the partitioning of equine
respiratory resistance. Equine VetJ 20: 268-273, 1988.

Attenburrow DP. Respiration and locomotion. In: Snow DH, Persson SGB,
Rose RJ, eds. Equine exercise physiology. Cambridge, England: Granta
Editions, 17-22, 1983.

130

131

Attenburrow DP, Flack FC, Hornicke H, Meixner R, Pollman U. Respiratory
airflow and sound intensity. Equine Exercise Physiology (Snow DH, Persson
868, Rose RJ, eds). Cambridge, England: Granta Editions, 23-26, 1983.

Babb TG, Viggiano R, Hurley B, Staats B, Rodarte JR. Effect of mild-to-
moderate airflow limitation on exercise capacity. JAppl Physiol 70: 223-230,
1991.

Baker GJ. Laryngeal asynchrony in the horse: definition and significance. In:
Snow DH, Persson 568, Rose RJ, eds. Equine exercise physiology.
Cambridge, England: Granta Editions, 1983:l,11: 46-50.

Baker GJ. Laryngeal hemiplegia in the horse. Comp Cont Ed Pract Vet 5: 561 -
567, 1983:".

Barber SM. Paralyngeal abscess with laryngeal hemiplegia and fistulation in a
horse. Can VetJ 22: 389-392, 1981.

Barlett D. Respiratory functions of the larynx. Physiol Rev 69: 33-57, 1989.

Barnes PJ, Gribbin RH, Osmanliev D, Pride NB. Partial flow-volume curves to
measure bronchodilator dose-response curves in normal humans. J Appl
Physiol: Respirat Environ Exercise Physiol 50: 1193-1197, 1981.

Bass H. The flow volume loop: normal standards and abnormalities in chronic
obstructive pulmonary disease. Chest 63: 171-176, 1973.

Bayly WM, Grant BD, Modransky PD. Arterial blood gas tensions during
exercise in a horse with laryngeal hemiplegia, before and after corrective
surgery. Res Vet Sci 36: 256-258, 1984.

Bayley WM, Schulz DA, Hodgson DR, Gollnick PD. Ventilatory responses of
the horse to exercise: effect of gas collection systems. J Appl Physiol 63:
1210-1217, 1987.

Becklake MR, Permutt S. Evaluation of tests of lung function for "screening”
for early detection of chronic obstructive lung disease. In: The lung in
transition between health and disease. Macklem PT, Permutt S (eds). Marcel
Dekker, lnc.: New York, 12: 345-387, 1979.

Belknap JK, Derksen FJ, Nickels FA, Stick JA, Robinson NE. Failure of subtotal
arytenoidectomy to improve upper airway flow mechanics in exercising
Standardbreds with induced laryngeal hemiplegia. Am J Vet Res 51: 1481-
1487. 1990.

 

132

Bisgard GE, Forster BB, Byrnes B, Stanek K, Klein J, Manohar M. Cerebrospinal
fluid acid-base balance during muscular exercise. J Appl Physiol 45: 94-101,
1978.

Bohanon TC, Beard WL, Robertson JT. Laryngeal hemiplegia in draft horses:
a review of 27 cases. Vet Surg 19: 456-459, 1990.

Boles CL, Raker CW, Wheat JD. Epiglottic entraphment by arytenoepiglottic
folds in the horse. J Am Vet Med Assoc 172: 338-342, 1978.

Buist AS, Adams BE, Sexton GJ. Azzam AH. Reference values for functional
residual capacity and maximal expiratory flow in young children. Am Rev Resp
Dis 122: 983-988. 1980.

Burrows GE. Lead poisoning in the horse. Eq Pract 4: 30-36, 1982.

Butler J. The work of breathing through the nose. Clin Sci 19: 55-62, 1960.

Cahill Jl, Goulden BE. Equine laryngeal hemiplegia: Part I. A light microscopic
study of peripheral nerves. N Zea/and Vet J 34: 161-169, 1986:l.

Cahill Jl, Goulden BE. Equine laryngeal hemiplegia: Part II. An electron
microscope study of peripheral nerve. N Zea/and VetJ 34: 170-175, 1986:".

Cahill Jl, Goulden BE. Equine laryngeal hemiplegia: Part IV. Muscle pathology.
N Zea/and Vet J 34: 186-190, 1986:lll.

Cahill Jl, Goulden BE. Equine laryngeal hemiplegia: Part V. Central nervous
system pathology. N Zea/and Vet J 34: 191-193, 1986:IV.

Cahill JI, Goulden BE. The pathogenesis of equine laryngeal hemiplegia—a
review. N Z VetJ 35: 82-90, 1987.

Cahill Jl, Goulden BE. Nervous system component in equine laryngeal
hemiplegia. NZealand VetJ 37: 89-90, 1989.

Carilli AD, Denson LJ, Rock F, Malabanan S. The flow-volume loop in normal
subjects and in diffuse lung disease. Chest 66: 472-477, 1974.

Cochrane AL, Holland WW. Validation of screening procedures. Br Med Bull
27: 3’8: 1971.

Cole CR. Changes in the equine larynx associated with laryngeal hemiplegia.
Am J Vet Res 7: 69-77, 1946.

133

Cook WR. The diagnosis of respiratory unsoundness in the horse. Vet Rec 77:
516-528. 1965.

Cook WR. A comparison of idiopathic laryngeal paralysis in man and horse.
J Laryngo and Otol 84: 819-835, 1970.

Cook WR. Some observations on diseases of the ear, nose and throat in the
horse, and endoscopy using a flexible fibreoptic endoscope. Vet Rec 94: 533-
541. 1974.

Cook WR. Some observations on form and function of the equine upper airway
in health and disease: I. The pharynx. Proc 27th AAEP 40: 355-391, 1981:l.

Cook WR. Some observations on form and function of the equine upper airway
in health and disease. II. The larynx. Proc 27th AAEP 393-451, 1981:".

Cook WR. Diagnosis and grading of hereditary recurrent laryngeal neuropathy
in the horse. J Eq Vet Sci 8: 432-455, 1988.

Cook WR, Thalhammer JG. Electrodiagnostic test for the objective grading of
recurrent laryngeal neuropathy in the horse. 37th Ann Proc Am Assoc Eq Pract
275-296, 1991.

Cymbaluk RL, Fretz PB, Loew FM. Thiamine measurements in horses with
laryngeal hemiplegia. Vet Rec 96: 611-616, 1977

Davidson IWF, Parker JC, Beliles RP. Biological basis for extrapolation across
mammalian species. Reg Toxicol Pharmacol 6: 211-237, 1986.

Dean PW. Diagnosis, treatment, and prognosis of arytenoid chondropathy.
36th Meeting Am Assoc Eq Pract 415-422, 1990.

Dean PW, Nelson JK, Schumacher J. Effects of age and prosthesis material on
in vitro cartilage retention of laryngoplasty prosthesis in horses. Am J Vet Res
51:114-117,1990.

Derksen FJ. Physiology of airflow in the athletic horse. Proc 34th Meeting Am
Assoc Equine Pract 149-158, 1988.

Derksen FJ, Robinson NE. Esophageal and intrapleural pressures in the healthy
conscious pony. Am J Vet Res 41: 1756-1761, 1980.

Derksen FJ, Stick JA, Scott EA, Robinson NE, Slocombe RF. Effect of
laryngeal hemiplegia and laryngoplasty on airway flow mechanics in exercising
horses. Am J Vet Res 47: 16-20, 1986.

 

134

Dubois AB, Botelho SY, Comroe JH. A new method for measuring airway
resistance in man using a bodyplethysmograph: values in normal subjects and
in patients with respiratory disease. J Clin Invest 35: 327-335, 1956.

Ducharme NG, Hackett RP. The value of surgical treatment of laryngeal
hemiplegia in horses. Compendium of Cont Educ 13: 472-475, 1991.

Ducharme NG, Hackett RP, Fubini SL, Erb HN. The reliability of endoscopic
examination in assessment of arytenoid cartilage movement in horses: Part II:

Influence of side of examination, reexamination, and sedation. Vet Surg 20:
180-184. 1991.

Ducharme NG, Horney FD, Partlow GD, Hulland TJ. Attempts to restore
abduction of the paralysed equine cartilage cartilage I. Nerve-muscle pedicle
transplants. Can J Vet Res 53: 202-209, 1989.

Ducharme NG, Horney FD, Hulland TJ, Partlow GD, Schnurr D, Zutrauen K.
Attempts to restore abduction of the paralysed equine arytenoid cartilage ll.
Nerve implantation (pilot study). Can J Vet Res 53: 210-215, 1989.

Ducharme NG, Viel L. Partlow GD, Hulland TJ, Horney FD. Attempts to restore

abduction of the paralyzed equine arytenoid cartilage III. Nerve anastomosis.
Can J Vet Res 53: 216-223, 1989.

Duncan ID, Amundson J, Cuddon PA, Sufit R, Jackson KF, Lindsay WA.
Preferential denervation of the adductor muscles of the equine larynx I: muscle
pathology. Equine VetJ 23: 94-98, 1991:I.

Duncan ID, Baker GJ, Heffron CJ, Griffiths IR. A correlation of endoscopic and
pathological changes in subclinical pathology of the horses’ larynx. Eq Vet J
9: 220-225, 1977.

Duncan ID, Brook D. Bilateral laryngeal paralysis in the horse. Eq Vet J 17:
228-233. 1985.

Duncan ID, Griffiths IR. Pathological changes in equine laryngeal muscles and
nerves. Proc 19th Ann Conv Am Assoc Eq Pract 97-113, 1973.

Duncan ID, Griffiths IR, McQueen A, Baker GJ. The pathology of equine
laryngeal hemiplegia. Acta Neuropath (Ber!) 27: 337-348, 1974.

Duncan ID and Hammang JP. Ultrastructural observations of organelle
accumulation in the equine recurrent laryngeal nerve. J Neurocytol 16: 269-
280. 1987.

 

135

Duncan ID, Reifenrath P, Jackson KF, Clayton M. Preferential denervation of
the adductor muscles of the equine larynx II: nerve pathology. Equine Vet J
23: 99-103, 1991:".

Eisele DW, Hillel AD, Trachy RE, Little JW. Reinnervation of the trapezius
muscle. Otolaryngol-Head Neck Surg 98: 34-43, 1988.

Elies W, Pusalkar A. Histological findings in arytenoid cartilages after long-
lasting recurrent nerve paralysis. J Laryngol and Otol 97: 949-957, 1983.

Empey DW. Assessment of upper airways obstruction. Brit Med J 3: 503-
505, 1972.

England SJ, Bartlett D, Daubenspeck JA. Influence of human vocal cord
movement on airflow and resistance during eupnea. J Appl Physiol: Respirat
Environ Exercise Physiol 52: 773-779, 1982.

Ferris BG, Mead J, 0pie LH. Partitioning of respiratory flow resistance in man.
J Appl Physiol 19: 653-658. 1964.

Fleming G. Laryngismus paralytica (“roaring"). VetJ of Camp Path 14: 1-12,
1882.

Fry DL, Ebert RV, Stead W, Brown CC. The mechanics of pulmonary

ventilation in normal subjects and in patients with emphysema. Am J Med 16:
80—97, 1954.

Fry DL, Hyatt RE. Pulmonary mechanics: a unified analysis of the relationship
between pressure, volume and gasflow in the lungs of normal and diseased
human subjects. Am J Med 29: 672-689, 1960.

Fulton IC. The effectiveness of a nerve muscle pedicle graft for the treatment
of equine left laryngeal hemiplegia. Michigan State University: Thesis. 56-58,
1 990.

Fulton IC, Derksen FJ, Stick JA, Robinson NE, Walshaw R. Treatment of left
laryngeal hemiplegia in Standardbreds using a nerve muscle pedicle graft. Am
J Vet Res 52: 1461-1467, 1991.

Fulton lC, Derksen FJ, Stick JA, Robinson NE, Duncan ID. Histologic
evaluation of nerve muscle pedicle graft used as a treatment for left laryngeal
hemiplegia in Standardbreds. Am J Vet Res 53: 592-596, 1992.

136

Funkquist B, Holm K, Karlsson A, Kvart C, Molander C, Obel N. Studies on
intratracheal pressure in the exercising horse. J Vet Med A 35: 424-441,
1988.

Getty R, ed . Sisson and Grossman ’s 7719 Anatomy of the domestic animals.
5th ed. WB Saunders Co., Philadelphia, 376-454, 498-524, 650-670, 1975.

Gilbert GH. Laryngeal hemiplegia following jugular injury. J Am Vet Med
Assoc 171: 1686-1687. 1972.

Gill JL. Design and analysis of experiments in the animal and medical sciences.
Vol. 1. Ames, IA: Iowa State University Press, 1978.

Gillespie JR, Tyler WS, Eberly VE. Pulmonary ventilation and resistance in
emphysematous and control horses. JAppl Physiol 21: 416-422, 1966.

Glogowska M, Stransky A, Widdicombe JG. Reflex control of discharge in
motor fibres to the larynx. J Physiol 239: 365-379, 1974.

Goulden BE, Anderson LJ. Equine laryngeal hemiplegia. Part I: Physical
characteristics of affected animals. N Zealand Vet J 29: 150-154, 1981:l.

Goulden BE, Anderson LJ. Equine laryngeal hemiplegia. Part II: Some clinical
observations. N Zealand Vet J 29: 194-198, 1981:".

Goulden BE, Anderson LJ. Equine laryngeal hemiplegia. Part III: Treatment by
laryngeal prosthesis. N Zealand Vet J 30: 1-5, 1982.

Goulden BE, Anderson LJ, Cahill Jl. Roaring in Clydesdales. N Zealand Vet J
33: 73-76. 1985.

Greet TRC. The slap test for laryngeal adductory function in horses with
suspected cervical spinal cord damage. Eq Vet J 12: 127-131, 1980.

Greet TRC. Baker GJ, Lee R. The effect of laryngoplasty on pharyngeal
function in the horse. Eq VetJ 11: 153-158, 1979.

Griffiths IR. The pathogenesis of equine laryngeal hemiplegia. Equine Vet J
23: 75-76, 1991.

Grimby G, Saltin B, Wilhelmsen L. Pulmonary flow-volume and pressure-
volume relationship during submaximal and maximal exdrcise in young well-
trained men. Bull Physio-path Respir 7: 157-168, 1971.

137

Grimby G, Stiksa J. Flow-volume curves and breathing patterns during exercise
in patients with obstructive lung disease. J Clin Lab Invest 25: 303-313,
1970.

Gunn HM. Further observations on laryngeal skeletal muscle in the horse. Eq
Vet J 5: 77-80, 1973.

Hackett RP, Ducharme NG, Fubini SL, Erb HN. The reliability of endoscopic
examination in assessment of arytenoid cartilage movement in horses: Part I:
Subjective and objective laryngeal evaluation. Vet Surg 20: 174-179, 1991.

Hall SJ, Trachy RE, Cummings CW. Facial muscle reinnervation: a comparison
of neuromuscular pedicle with direct nerve implant. Ann Otol Rhinol Laryngol
97: 229-233, 1988.

Hamilton LH. Nasal airway resistance: its measurement and regulation.
Physiologist 22: 43-49, 1978.

Haponik EF, Bleecker ER, Allen RP, Smith PL, Kaplan J. Abnormal inspiratory
flow-volume curves in patients with sleep-disordered breathing. Am Rev Respir
Dis 124: 571-574, 1981.

Haponik EF, Meyers DA, Munster AM, Smith PL, Britte EJ, Wise RA, Bleeker
ER. Acute upper airway injury in burn patients. Am Rev Respir Dis 135: 360-
366. 1987.

Harrison IW, Speirs VC, Braund KG, Steiss JE. Attempted reinnervation of the
equine larynx using a muscle pedicle graft. Cornell Vet 82: 59-68, 1992.

Hay WP, Tulleners EP, Ducharme NG. Arytenoidectomy in the horse using an
extralaryngeal approach. Vet Surg 22: 50-56, 1993.

Haynes PF. Arytenoid chondritis in the horse. Proc 27th Am Assoc Equine
Pract pp 63-66, 1981.

Haynes PF. Surgery of the equine respiratory tract. In: Jennings PB, ed. The
practice of large animal surgery, vol. 1. Philadelphia, PA: WB Saunders Co:
388-487. 1984.

Haynes PF, McClure JR, Watters, JW. Subtotal arytenoidectomy in the horse:
an update. Proc Am Assoc Equine Pract: 21-33, 1984.

Haynes PF, Snider TG, McClure JR, McClure JJ. Chronic chondritis of the
equine arytenoid cartilage. J Am Vet Med Assoc 177: 1135-1142, 1980.

138

Hillidge CJ. Prevalance of laryngeal hemiplegia in a Thoroughbred horse farm.
Eq Vet Sci 5: 252-254. 1985.

Hobday F. The surgical treatment of roaring in horses. Vet Rec 15: 1535-
1539. 1935.

Hoffstein V, Brown I, Taylor R, McLean P, Zamel N. Maximum flow ratios at
mid-vital capacity in young healthy adults. Chest 90: 857-860, 1986.

H6rnicke H, Meixner R, Pollmann U. Respiration in exercising horses. In: Snow
DH, Persson 868, Rose RJ, eds. Equine exercise physiology. Cambridge,
England: Granta Editions, 7-16, 1983.

H6rnicke H, Weber M, Schweiker W. Pulmonary ventilation in Thoroughbred
horses at maximum performance. In: Gillespie JR, Robinson NE, eds. Equine
exercise physiology 2. Davis, CA: ICEEP Publications, 216-224, 1987.

Huntington PJ, Jeffcott LB, Friend SCE, Luff AR, Finkelstein DI, Flynn RJ.
Australian stringhalt: epidemiological, clinical and neurological investigations.
Eq VetJ 21: 266-273, 1989.

Hyatt EH, Black LP. The flow-volume curve: a current perspective. Am Rev
Respir Dis 107: 191-199, 1973.

Hyatt RE, Schilder DP, Fry DL. Relationship between maximum expiratory flow
and degree of lung inflation. J Appl Physiol 13: 331-336, 1958.

Hyatt RE, Wilcox RE. Extrathoracic airway resistance in man. J Appl Physiol
16: 326-330, 1961.

Johnson J, Tucker HM. Selective experimental reinnervation of paralyzed facial
muscles. Arch Otolaryngol 103: 22-25, 1987.

Jordanoglou J, Pride NB. A comparison of maximum inspiratory and expiratory
flow in health and in lung disease. Thorax 23: 38-45, 1968.

Kashima HK. Documentation of upper airway obstruction in unilateral vocal
cord paralysis: flow-volume loop studies in 43 subjects. Laryngoscope 94:
923-937, 1984.

Kosch PC. Respiratory muscle function and breathing strategy: comparative
and developmental differences. Proc 4th Vet Resp Symposium, May 9-11,
Anaheim; 26-38, 1985.

139

Lane JG. Fibreoptic endoscopy of the equine upper respiratory tract: a
commentary on progress. Eq Vet J 19: 495-499, 1987.

Leith DE. Comparative mammalian respiratory mechanics. The Physiologist 19:
485-510. 1976.

Liautard, A. Manualof operative veterinary surgery. Sabiston and Murray, New
York, 1892.

Linford RL, O'Brien TR, Wheat JD, Meagher DM. Radiographic assessment of
epiglottic length and pharyngeal and laryngeal diameters in the Thoroughbred.
Am J Vet Res 44: 1660-1666, 1983.

Loew FM. Thiamin and equine laryngeal hemiplegia. Vet Rec 97: 372-373, 1973.

Lundborg G. lntraneural microcirculation. Orthopedic Clinics of Nth Am 19: 1-
1 2, 1988.

Lyons RM, Tucker HM. Delayed restoration of abduction in the paralyzed
canine larynx. Arch 0tolaryngol100: 176-179, 1974.

Mackay-Smith MP, Johnson JH, Baker RH. Laryngoplasty: a progress report.
19th AAEP 133-146, 1973.

MacQueen J. Roaring, whistling and grunting. J Comp Path and Therapy 9:
112-119, 1896.

Mangseth G. Evaluation of tracheal pressures in the running horse. Proc Assoc
Equine Sports Med 74-80, 1984.

Marks D, Mackay-Smith MP, Cushing LS, Leslie JA. Use of a prosthetic device
for surgical correction of left laryngeal hemiplegia in horses. J Am Vet Med
Assoc 157: 157-163, 1970:l.

Marks D, Mackay-Smith MP, Cushing LS, Leslie JA. Etiology and diagnosis of
laryngeal hemiplegia in horses. JAm Vet Med Assoc 157: 429-436, 1970:".

Martin GS, Beadle RE, Haynes PF, Watters JM. Cross-sectional area of the
aditus laryngus and rima glottidis before and after transection of the left
recurrent laryngeal nerve in the horse. Am J Vet Res 47: 422-425, 1986.

Mason JE. Laryngeal hemiplegia: a further look at Haslam’s anomaly of the left
recurrent laryngeal nerve. Eq Vet J 5: 150-155, 1973.

140

Mcllwraith CW, Turner AS. Surgery of the upper respiratory tract. Equine
Surgery Advanced Techniques. Lea & Febiger, Philadelphia, pp 210-215,
1987.

McKiernan BC, Dye JA, Rozanski EA. Tidal breathing flow-volume loops in
healthy and bronchitic cats. J Vet Int Med 1993 (accepted).

McKiernan BC, Jones SD. Computerized tidal breathing flow-volume loop
acquisition in the dog. Proc 6th Vet Res Sym 28, 1987.

McKiernan BC, Rozanski EA, Jones SD, Hayes 8. The effect of CO, on tidal
breathing flow-volume loops in conscious cats. Proc 8th Vet Res Sym 23,
1989.

Merriam JG. Laryngoplasty: an evaluation of three abductor muscle
prosthetics. Proc 19th AAEP123-131, 1973.

Miller RD, Hyatt RE. Obstructing lesions of the larynx and trachea: clinical and
physiologic characteristics. Mayo Clin Proc 44: 145-161 , 1969.

Miller RD, Hyatt RE. Evaluation of obstructing lesions of the trachea and larynx
by flow-volume loops. Am Rev Respir Dis 108: 475-481, 1973.

Montgomery TC. C02 Laser and Microsurgery in the Horse. Proc 31$t Meeting
Am Assoc Equine Pract 399-407, 1985.

Moore MP, Andrews F, Reed SM, Grant 80. Electromyographic evaluation of
horses with laryngeal hemiplegia. J Eq Vet Sci 8: 424-427, 1988.

Morris EA, Seeherman HJ. The dynamic evaluation of upper respiratory
function in the exercising horse. Proc 34th Meeting Am Assoc Equine Pract
159-165, 1988.

Morris EA, Seeherman HJ. Evaluation of upper respiratory function during
strenuous exercise in racehorses. J Am Vet Med Assoc 196: 431-438, 1990.

Morris EA, Seeherman HJ. Clinical evaluation of poor performance in the
racehorse: the results of 275 evaluations. Equine Vet J 23: 169-174, 1991.

Motoyama EK. Pulmonary mechanics during early postnatal years. Pediatr Res
11: 220-223, 1977.

Neilan GJ, Rehder RS, Ducharme NG, Hackett RP. Measurement of tracheal
static pressure in exercising horses. Vet Surg 21: 423-428, 1992.

 

141

Nolte D, Liider-Liihr I. Comparing measurings of nasal resistance by body
plethysmography and by rhinomanometry. Respiration 30: 31-38, 1973.

Olafsson S, Hyatt RE. Ventilatory mechanics and expiratory flow limitation
during exercise in normal subjects. J Clin Invest 48: 564-573, 1969.

Olson DE, Dart GA, Filley GF. Pressure drop and fluid flow regime of air
inspired into the human lung. J Appl Physiol 28: 482-494, 1970.

Orsini PG, Raker CW, Reid CF, Mann P. Xeroradiographic evaluation of the
equine larynx. Am J Vet Res 50: 845-849, 1989.

Pascoe JR, Ferraro GL, Cannon JH. Exercise-induced pulmonary hemorrhage
in racing Thoroughbreds: a preliminary study. Am J Vet Res 42: 703-707,
1981.

Patel V, Head MR. Reversion of turbulent to laminar flow. J Fluid Mech 34:
371-392. 1968.

Pedley TJ, Schroder RC, Sudlow MF. The prediction of pressure drop and
variation of resistance with the human bronchial airways. Respir Physiol 9:
387-405, 1970.

Flint MA, Boswirth L. Fluid Mechanics: a Laboratory Course. High Wycombe,
Bucks, England: Charles Griffin & Co, 1978.

Polverino M, De Lucas P, Marin JM. et al. Normal standards for diagnostic
tests of upper airway obstruction. Am Rev Res Dis. 161: 63, 1990.

Poncet PA, Montavon S, Gaillard C, Barrelet F, Straub R, Gerber H. A
preliminary report on the possible genetic basis of laryngeal hemiplegia. Eq Vet
J21: 137-138, 1989.

Pouret E. Laryngeal ventriculectomy with stitching of the laryngeal saccules.
Proc 12th AAEP 207-213, 1966.

Pride NB. The assessment of airflow obstructions: role of measurements of
airways resistance and of tests of forced expiration. Brit J Dis Chest 65: 135,
1 971 .

Proctor OF. The upper airways: I. Nasal physiology and defense of the lungs.
Am Rev Respir Dis 115: 97-120, 1977.

Ouinlan TJ, Goulden BE, Barnes GRG, Anderson LJ, Cahill JI. Innervation of
the equine laryngeal muscles. N Zealand Vet J 30: 43-45, 1982.

142

Raker CW. Endoscopy of the upper respiratory tract of the horse. Proc let
Ann Meet Am Assoc Eq Pract 23-28, 1976.

Rakestraw PC, Hackett RP, Ducharme NG, Nielan GJ, Erb HN. Arytenoid
cartilage movement in resting and exercising horses. Vet Surg 20: 122-127,
1991.

Raphel CF. Endoscopic findings in the upper respiratory tract of 479 horses.
JAm Vet Med Assoc 181: 470-473, 1982.

Robinson NE, Sorenson PR. Pathophysiology of airway obstruction in horses:
a review. J Am Vet Med Assoc 172: 299-303, 1978.

Robinson NE, Sorenson PR, Goble DO. Patterns of airflow in normal horses and
horses with respiratory disease. Proc let AAEP 11-22, 1975.

Rooney JR, Delaney PM. A hypothesis on the causation of laryngeal hemiplegia
in the horse. Eq Vet J 2: 35-37, 1970.

Rose RJ, Hartley WJ, Baker W. Laryngeal paralysis in Arabian foals associated
with oral haloxon administration. Equine Vet J 13: 171-176, 1981.

Rose RJ, Hendrickson DK, Knight PK. Clinical exercise testing in the normal
Thoroughbred racehorse. Aust Vet J 67: 345-348, 1990.

Rossdale PD, Hopes R, Digby W, Offord K. Epidemiological study of wastage
among racehorses 1982 and 1983. Vet Rec 116: 66—69, 1985.

Rotman HH, Liss HP, Weg JG. Diagnosis of upper airway obstrucion by
pulmonary function testing. Chest 68: 796-799, 1975.

Shapiro J, White NA, Schlafer DH, Rowland GN. Hypertrophic ossification of
the laryngeal cartilages of a horse. J Eq Med and Surg 3: 370-374, 1979.

Shappell KK, Derksen FJ, Stick JA, Robinson NE. Effects of ventriculectomy,
prosthetic laryngoplasty, and exercise on upper airway function in horses with
induced left laryngeal hemiplegia. Am J Vet Res 49: 1760-1765, 1988.

Shim C, Corro P, Park SS, Williams MH. Pulmonary function studies in patients
with upper airway obstruction. Am Rev Respir Dis 106: 233-238, 1972.

Shires GMH, Adair HS, Patton CS. Preliminary study of laryngeal sacculectomy
in horses, using a neodymiumzyttrium aluminum garnet laser technique. Am J
Vet Res 51 : 1247-1249, 1990.

 

143

Smith GJ, Cooper DM. Laryngomalacia and inspiratory obstruction in later
childhood. Arch Dis Childhood 56: 345-349. 1981.

Smith MM, Gourley IM, Kurpershoek CJ, Amis TC. Evaluation of a modified
castellated laryngofissure for alleviation of upper airway obstruction in dogs
with laryngeal paralysis. JAm Vet Med Assoc 188: 1279-1283, 1986.

Smith MM, Gourley IA, Amis TC, Kurpershoek C. Management of tracheal
stenosis in a dog. JAm Vet Med Assoc 196: 931-934, 1990.

Spann RW, Hyatt RE. Factors affecting upper airway resistance in conscious
man. J Appl Physiol 31: 708-712, 1971.

Speirs VC. Partial arytenoidectomy in horses. Vet Surg 15: 316-320, 1986.
Speirs VC. Laryngeal surgery—150 years on. Eq Vet J 19: 377-383, 1987.

Speirs VC, Bourke JM. Anderson GA. Assessment of the efficacy of an
abductor muscle prosthesis for treatment of laryngeal hemiplegia in horses.
Aust VetJ 60: 294-299. 1983.

Speizer FE, Frank NR. A technique for measuring nasal and pulmonary flow
resistance simultaneously. J Appl Physiol 19: 176-178, 1965.

Stick JA, Derksen FJ. Use of videoendoscopy during exercise for determination
of appropriate surgical treatment of laryngeal hemiplegia in a colt. J Am Vet
Med Assoc 195: 619-622, 1989.

Stransky A, Szereda M, Widdicombe JG. The effects of lung reflexes on
laryngeal resistance and motor neuron discharge. J Physiol 231: 417-438,
1973.

Suzuki M, Kirchner JA. The posterior cricoarytenoid as an inspiratory muscle.
Laryngoscope 79: 849-864, 1969.

Sweeney CR, Maxson AD, Soma LR. Endoscopic findings in the upper
respiratory tract of 678 Thoroughbred racehorses. JAm Vet Med Assoc 198:
1037-1038. 1991.

Tammelln BR, Wilson AF, Borowiecki BDB, Sassin JF. Flow-volume curves
reflect pharyngeal airway abnormalities in sleep apnea syndrome. Am Rev
Respir Dis 128: 712-715. 1983.

144

Tate LP, Corbett WP, Bishop B, Foreman J. Blood gas tensions, acid-base
status, heart rates, and venous profiles in exercising horses with laryngeal
hemiplegia before and after corrective surgery. Vet Surg 22: 177-183, 1993.

Tate LP, Newman HC, Cullen JM, Sweeney C. Neodymium (Nd):YAG laser
surgery in the equine larynx: a pilot study. Lasers in Surgery and Medicine 6:
473-476, 1986.

Tate LP, Sweeney CL, Cullen JM, Corbett WT, Newman HC, Brown TC, Ketner
MT. Transendoscopic neodymiumzyttrium aluminum garnet laser irradiation in
horses. Am J Vet Res 50: 786-791, 1989.

Taussig LM. Maximal expiratory flows at functional residual capacity: a test of
lung function for young children. Am Rev Respir Dis 116: 1031-1038, 1977.

Tesarowski DB, Viel L, McDonelI WN. Analysis of flow-volume loops and
arterial blood gas during breathing of 8, 10, and 12% C02/air mixtures in
horses. Proc 8th Vet Res Sym 18, 1989.

Topham JH, Empey DW. Practical assessment of obstruction in the larynx and
trachea. J Laryng Otol 88: 1185-1193, 1974.

Tucker HM. Human laryngeal reinnervation. Laryngoscope 86: 769-779,
1976.

Tulleners EP, Harrison IW, Raker CW. Management of arytenoid chondropathy
and failed laryngOplasty in horses: 75 cases (1979-1985). J Am Vet Med
Assoc 192: 670-675, 1988:l.

Tulleners EP, Harrison IW, Mann P, Raker CW. Partial arytenoidectomy in the
horse with and without mucosal closure. Vet Surg 17: 252-257, 1988:".

Vincken W, Ghezzo H, Cosio MG. Maximal static respiratory pressures in
adults: normal values and their relationship to determinants of respiratory
function. Bull Eur Physiopathol Resp 23: 435-439, 1987.

Wall MA, Misley MC, Dickerson D. Partial expiratory flow-volume curves in
young children. Am Rev Respir Dis 129: 557-562, 1984.

West JB. Respiratory Physiology- The Essentials. 4th ed. Baltimore, MD: The
Williams and Wilkins Co, 86-114, 1979.

Wheat JD. Pathology and related surgery of the larynx. 6th Annu Forum Am
Coll Vet Surg 1978.

145

Wheeler and Plummer. Age-related changes in the fibre composition of equine
peripheral nerve. J Neurol Sci 90: 53-66, 1989.

White, N. A. and Blackwell, R. 8. Partial arytenoidectomy in the horse. Vet.
Surg. 9, 5-12, 1980.

Whiting T. Evaluation of the partially forced expired flow-volume loop derived
by C02 inhalation challenge in the horse. Thesis: University of Guelph, 1988.

Williams WL. The surgical relief of roaring. 48th Ann Conf A VMV 299-317,
191 1 .

Williams JW, Pascoe JR, Meagher DM, Hornof WJ. Effects of left recurrent
laryngeal neurectomy, prosthetic laryngoplasty, and subtotal arytenoidectomy
on upper airway pressure during maximal exertion. Vet Surg 19: 136-141,
1990a.

Williams JW, Meagher DM, Pascoe JR, Hornof WJ. Upper airwy function
during maximal exercise in horses with obstructive upper airway lesions: effect
of surgical treatment. Vet Surg 19: 142-147, 1990:".

Wise PH, Krauss AN, Waldman S, et al. Flow-volume loops in newborn infants.
Crit Care Med 8: 61-63, 1980.

Woakes AJ, Butler PJ, Snow DH. The measurement of respiratory airflow in
exercising horses. Equine Exercise Physiology 2 (Gillespie JR, Robinson NE,
eds). Davis, CA: ICEEP Publications, 194-205, 1987.

Yeh MP, Adams TD, Gardner RM, Yanowitz FG. Turbine flowmeter vs. Fleisch

pneumotachometer: a comparative study for exercise testing. J Appl Physiol
63: 1289-1295. 1987.

Young SS, Hall LW. A rapid, non-invasive method for measuring total
respiratory impedance in the horse. Equine VetJ 21: 99-105, 1989.

Zuntz N, Hagemann O. Untersuchungen fiber den Stoffwechsel des Pferdes bei
Ruhe und Arbeit. Landw Jb 27: 1-438, 1898.