Veterinary Research Communications, 17 (1993) 227-239 Copyright © Kluwer Academic Publishers bv - Printed in the Netherlands

MEASUREMENT OF TOTAL RESPIRATORY IMPEDANCE IN DOGS BY THE FORCED OSCILLATION TECHNIQUE

C. CLERCX 1, P. GUSTIN 2, F.J. LANDSER 3 AND K.P. VAN DE WOESTIJNE 1Clinique M6dicale des Petits Animaux, 2Chaire de Pharmacologie- Pharmacoth6rapie-Toxicologie, Facult6 de M6decine V6t6rinaire, Universit6 de Li6ge, Sart Tilman, 4000 Li6ge, Belgium; 3Laboratorium voor Pneumologie, U.Z. Gasthuisberg, 3000 Leuven, Belgium

ABSTRACT

Clercx, C., Gustin, P., Landser F.J. and Van de Woestijne, K.P., 1993. Measurement of total respiratory impedance in dogs by the forced oscillation technique. Veterinary Research Communications, 17 (3), 227-239

The resistance ( R ) and reactance ( X ) of the total respiratory system were determined at various frequencies in 14 [~ealthy conscious beagle dogs• A pseudorandom noise pressure wave was produced at the nostrils of the animals by means of a loudspeaker adapted to the nose by a tightly fitting mask. A Fourier analysis of the pressure and flow signals yielded mean R and X , over 16 s, at frequencies from 2 to 26 Hz. The influence of the posture of the dog, the posi~on of i~ head, the linearity of the respiratory system, the reproducibility of the method and the effects of upper and lower airway obstructions were studied. In sitting and standing healthy dogs with the head in the extended position,

• . . ÷ - 1 R values mcreased progresswely with frequency from 5.4 _ 0.4 (SEM) cmH O L s at 6 Hz up to 8.8 r s - 1 . 2 . . .

-+ 0.7 cmHzO L s at 26 Hz, the mean resonant frequency being 6.1 _+ 0.5 Hz. No significant differences were observed between measurements performed with the head in the normal or the extended position. In a recumbent posture, all R values were increased but R was still dependent on the frequency in the same way (7.1 _+ 0.7 cmH'SO L-Is at 6Hz up to 10.0 -+ ~.5 cmH O L-Is at 26 Hz). Tracheal compression also induced highZer R values without changes in t~e frequency dependence or in the resonant frequency, rs

In anaesthetized dogs, airway obstruction was induced by inhalation of histamine (4 mg/ml for 5 rain; the R values tended to decrease with increasing frequency, and the resonant frequency was

, r s

markedly mcreased.

Keywords: airways, dog, forced oscillation, respiratory impedance, obstruction to airways

Abbreviations: FOT, forced oscillation technique; IV, intravenous(ly)

INTRODUCTION

The mechanical characteristics of the canine respiratory system have been studied extensively in anaesthetized and tracheotomized animals or in excised lungs by the forced oscillation technique (Hull and Long, 1961; Tsai et al., 1977; Kappos et aL, 1981; Fullton et aL, 1982; Van Brabandt et aL, 1983; Jackson et aL, 1984; Fredberg et al., 1984, 1985; Harf et al., 1985). These preparations have been very useful in elucidating the mechanical behaviour of each segment of the respiratory system in physiological and pathological conditions and in validating theoretical models predicting the mechanical properties and the gas distribution within the lungs.

The FOT should also be very useful for investigating respiratory diseases in the dog. The main advantage of this non-invasive technique is that the same animal can

228

be investigated on different occasions in order to demonstrate the evolution of a respiratory disease or the effects of a drug. The aim of this work was to adapt the FOT to conscious dogs and to describe the frequency characteristics of their respiratory system under physiological conditions. In order to test the ability of the FOT to characterize airway changes, the method was used to assess changes induced in experimental upper and lower airway obstructions.

MATERIALS AND METHODS

The experimental setup and data processing were similar to those described by Landser et al. (1976) for humans and adapted to calves by Gustin et al. (1988a). Fourteen adult healthy beagle dogs aged 9 months to 8 years and weighing 9-15 kg were used for this study. The conscious animals breathed through 30-era long, 2.5-cm ID tubing into a pneumotachograph (Screen-type semidisposable flowhead type F300L, max flow 300 L/min, Mercury Electronics LDT, Newton Mearns, Glasgow, Scotland). The tubing was adapted to the nose of the animals by means of a tight-fitting mask (Figure 1). Care was taken to keep the mouth closed during measurements. The lateral pressure was sampled from a tap situated in the tubing close to the mask. Pressure and flow were measured by two identical differential pressure transducers (Sensym SCX 0 -+ 1 psid, Sensortechnics GmbH, Puchheim, Germany). A loudspeaker, which produced a pseudorandom pressure wave containing all harmonics of 2-26 Hz was connected to the tubing. Volume displacement of the loudspeaker system was about 50 ml peak to peak. Pressure signal amplitude was 1.5 cmH20. A 1.5-cm ID and 0.6-m long side tubing allowed the animal to breathe through a low-resistance, high-inertiance pathway. The dead space of this setup was reduced by a constant flow in the mask from a compressed air bottle. Under these conditions, the dog was breathing quietly. The pressure and flow signals were filtered (3.5 Hz) by means of a fourth-order high-pass filter.

Figure 1. Tight-fitting nasal mask applied upon the nose of a beagle dog

2 2 9

A spectral analysis was then performed on-line on the pressure and flow signals, by means of a Fourier analyser system, yielding a resistance (Rrs) and a reactance ( X ) value for each of the harmonics (Michaelson et al., 1975; Landser et al., 1976). The measurements were repeated three times and the mean values of Rrs and X at each frequency were computed. Before measurement, correction factors for the frequency characteristics of the setup were calculated for each of the frequencies investigated.

A coherence function (Michaelson et aL, 1975; Landser et al., 1976) was calculated for all harmonics from 2 to 26 Hz. Only values with a coherence function greater than 0.95 were considered.

TABLE I Reproducibility of measurements successive days in 13 healthy dogs a

of resistances and reactances performed on

6 Hz

R bat X bat r s r s

12 Hz 24 Hz 6 Hz 12 Hz 24 Hz

Mean 5.42 6.05 6.96 0.03 1.01 2.17

D c 0.21 0.32 0.28 0.08 0.01 0.06 SD c 1.23 1.08 1.58 0.46 0.37 0.85

aValues are in cmH20 L -1 b • • • R resplratoryres~stance;X resplratoryreactance C r s ~s . .

D and SD: mean value (with standard devlanon) of the differences (in absolute value) between the lowest and the highest values of Rrs or X obtained in each animal on different days (see text for further explanation)

The reproducibility and linearity of the respiratory system were checked to validate the technique. The reproducibility was assessed by calculating the differences between the lowest and the highest values of Rrs and X determined at 6, 12 and 24 Hz in 13 dogs, in which measurements were taken ori 3 different days (Table I). This was repeated for at least two positions. The larger difference was selected. For each of the above frequencies, the differences determined in each of the 13 animals were averaged and a mean difference was obtained (D _+ SD). The linearity of the respiratory system was tested in 7 dogs by comparing the values of Rrs and X for two different amplitudes of forced pressure signal (1.50 and 0.75 cmH20 ).

230

Effect of the position of the animal

Fourteen beagle dogs aged 1-8 years were used for this study. In order to determine the most suitable position for the dog during the measurements, the method was applied to dogs that were standing (ST), sitting (SI), in sternal recumbency (SR) and in right lateral recumbency (LR) with their heads in the normal position (the longitudinal axis of the head making an angle of 90 ° with the axis of the neck).

Effect of the position of the head

The effect of the position of the head on the mechanical properties of the respiratory system was investigated in standing and sitting dogs (n = 14). Three positions of the head were compared: normal, extension (angle of 180 ° between head and neck) and flexion (angle of 45°). Measurements with the head in flexion were performed in only 6 dogs, as the position appeared to cause the dogs some discomfort and the results obtained were relatively variable.

Effect of upper airway obstruction

Measurements were performed before and after the application of external manual compression of the lowest part of the extrathoracic trachea close to the thoracic inlet in 7 standing dogs with their heads in normal position.

Effects of lower airway obstruction

Three dogs were anaesthetized with thiopental (up to 20 mg/kg IV) (Penthotal; Abbot, Belgium) and placed in the right lateral recumbent position with their heads in the extended position. After measuring the reference values, an endotracheal tube was introduced into the trachea and the animal was challenged for 5 min with an aerosol of isoosmotic phosphate buffered saline (pH 7.4, 0.05 mol/L phosphate), either alone or containing histamine (4 mg/ml) (Sigma Chemical Co., St Louis, MO, USA), nebulized with an Accorn Vix Nebulizer powered by air at a flow rate of 2 L/min. The endotracheal tube was then withdrawn and measurements of Rrs and X were repeated until the maximum effects had been recorded.

Statistical analysis

The data are presented as the mean + SEM. A Student's t-test for paired data was used to compare the results obtained with dogs in different postures and with different positions of the head andp = 0.05 was chosen as a level of significance.

231

RESULTS

When two different pressure amplitudes were used to drive the respiratory system, identical values of R andX were obtained (the mean differences in Rrs a n d X in 5 animals were <0.5 c~nH20 l~-ls).

The reproducibility of measurements performed on successive days is illustrated in Table I.

In standing animals with the normal head position (n = 14), Rrs increased strongly with increasing frequency. The mean data are illustrated in Figure 2. X was negative at low frequencies and became positive at high frequencies; mean resonant frequency, defined as the frequency at which the reactance is zero, was 7.9 -+ 0.8 Hz (mean + SEM). In three of the dogs, X decreased at the highest frequencies.

G" 4~

.-I

-1-

E

rr

10

g

8

7

6

5 4

i i i i i i i i i t i

6 8 10 12 14 16 18 2 0 22 2 4 2 6 Hertz

3

2 .,d

¢q

- r 1 E

P~ o X

-1 i i i i | i i I i i i

6 8 1 0 12 14 16 18 2 0 2 2 2 4 26 Hertz

Figure 2. Mean resistance (Rrs) and reactance (37) values (_+ SEM) obtained in 14 healthy standing dogs with the head in normal position, as functions of oscillatory frequency. Lines join values at 2-Hz intervals

The mean Rrs and X values were not significantly influenced by the posture of the dog or the positioning of the head (Figure 3) except for when the head was in flexion and when the dog was in a recumbent posture. Positioning the head in flexion always induced a marked increase in the resonant frequency and in the Rrs values and a large decrease in X , although the severity of the changes differed from one dog to

0

C

i • i I I I • i

(~os/-ItO;~HW~) sJU

O1 O

¢-

E

, i

e E

"O "1~

¢- i,-

: \ \ = . . *+

I I I i

0

E

"0 "O

. .~ r"

I I I I I

(~ )S l l lO~ HI,U:)) sJU

N

0 @l

~0

'd"

0

cO

N

-r" ~ . 0

o

cO

O

E

( ~ e S t l l O Z H U . l ~ ) s i x

0 m

t - i n

o_~

CP

232

I ! I I

(~eStltOEH~U~) s ix

N

"2

-r-

- 0 . 1

,d"

- c ' d

¢N

0

cO

'7,

N

-I-

¢M

_ o

. 00

- co

O~

0

- ¢ 0

- co

o~

+1

=~..=

. 0 ~.~

=~..=

1 1

~ 10

~ 9 - - /

I 7 E

v 6 P ~ 5

- ~ h e a d n o r m a l ÷ h e a d in extensioq - ~ head in f lexion

i i i i i ~ i i i i i i i

2 4 6 8 10 12 14 16 18 2 0 22 24 2 6 Hertz

233

3 ¢,,>

2 . J

"-r-

E o

P -1 N

- 2

- ~ h e a d n o r m a l --,- head in extens~ort -e- head in f lexion

i i i i i i i i i ~ I

6 8 10 12 14 1 6 18 2 0 22 2 4 2 6 Hertz

Figure 4. Resistance (Rrs) and reactance ( X ) as functions of oscillatory frequency in a healthy standing dog (12 kg) with 3 different positions of the head: normal position (~); extension (*); flexion (=)

another. Typical data for one dog weighing 12 kg are shown in Figure 4. Although the Rrs and X values were not significantly modified by the extension of the head, a considerable decrease in the Rrs values at all frequencies was observed in three dogs.

In recumbent postures, all the Rrs values were increased; however, at all frequencies this was only significant when the standing or sitting postures were compared with sternal recumbency (Figure 5). The frequency dependence of Rrs and the resonant frequency were not significantly modified.

Reduction of the tracheal diameter induced a significant increase in Rrs at all frequencies, without modification in the frequency dependence of Rrs (Figure 6). X values were not modified.

In dogs with histamine-induced bronchospasm, maximum bronchoconstriction occurred within 5 min. The Rrs values were mostly increased. However, they tended to decrease with increasing frequencies, at least at the lower frequencies (Figure 7a,b,c). All X values were decreased, especially at lower frequencies, and the resonant frequency was markedly increased. Phosphate buffer alone had no effect.

234

11

• -J 9

-r" 8

E ~ 7 p~ m- 6

r e c u m b e n c y J. J- ~ --~ right lateral r e c u m b e n c y

"~ standing dogs i i i i i i i i t

6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4 2 6 Hertz

4

4> 3

"~ 2

C~J t

E ( J ~-~ 0 P X . 1 ¸

-2

B

~ c u m b e n c y ~ ' ~ ~ right lateral r e c u m b e n c y ~" ~ standing dogs

i i i i i i i i i I

6 8 tO 12 14 16 18 20 22 24 26 Hertz

Figure 5. Mean resistance (Rrs) and reactance ( X ) as functions of oscillatory frequency in animals (n = 14) in sternal recumbent ([]), right lateral recumbent ( ,) and standing postures (=). The head was in extension during the measurements. Lines join mean values (_+ SEM) at 2-Hz intervals. Asterisks indicate significant differences from values measured in the standing position (p ~ 0.05)

DISCUSSION

The technique described was developed in humans (Michaelson et al., 1975; Landser et al., 1976), and has been successfully applied to the measurement of pulmonary mechanisms in conscious calves (Gustin et al., 1988a) and ponies (Art et al., 1989). It now appears that it can be successfully adapted to conscious dogs.

One of the requirements of this technique is that the system investigated is linear within the range of the forced disturbances. The magnitude of the pressure impulses used in this study was well tolerated yb the animals. When smaller impulses were used, identical values of Rrs and X were obtained, indicating that the requirement for linearity was satisfied.

In calves, the reproducibility of FOT measurements performed in quick succession has been shown to be better than the reproducibility of measurements performed on successive days (Gustin et aL, 1988a). In this study in dogs, the reproducibility of measurements performed on successive days gave smaller mean differences between the lowest and the highest values of Rrs and X than was the case using the same

235

,.J

C~J I

E

¢-e

1 4

1 2

1 0

8 erence - + tracheal compression

t i i i i i i i i 1 i

6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4 2 6 H e r t z

" - " 4 - ¢,a

3 - . i I

"I"

E v

I,. X 0

1

I I i

-e- Reference

+ Tracheal compression

' ' o 2 ' ' ' 8 ' ' '4 '6 Hertz 6 8 I I 14 16 I 20 22 2 2

Figure 6. Resistance (Rrs) and reactance ( X ) as functions of oscillatory frequency before (6) and after ( ,) extrathoracic compression in 7 healthy dogs. Lines join values (+ SEM) at 2-Hz intervals. Asterisks indicate significant differences from values measured before compression (p ~< 0.05)

technique in calves (Gustin et al., 1988a). The set-up was applied to the respiratory system of the dog by means of a

tight-fitting nasal mask. In humans a mouthpiece is used, and in conscious calves or ponies a tight-fitting mask was adapted to the shape of the animals' head (Gustin et al., 1988a; Art et al., 1989). It is important to avoid any leakage of air from the mask and also to minimize the dead space in the mask. Moreover, as this technique is intended to be of practical use in conscious diseased dogs, the mask must be comfortable. A mouthpiece clearly cannot be used in conscious dogs. It is difficult to adapt a head mask because of the large pendulous lips of the beagle dogs, and also because of the extreme variety of head morphology among the different breeds of dogs. Therefore, the option of a nasal mask was chosen. The nasal mask was designed to rest on the nasal bone and just above the superior lip, in order to get a perfect fit for the mask without deforming the dog's soft nose (Figure 1). The nasal mask chosen gave satisfactory results, but artefacts and non-repeatable measurements had been obtained with various types of face mask previously investigated. The mask was well tolerated during the brief measurements. Another advantage of the nasal mask is minimization of the shunt impedance constituted by the cheeks (Michaelson et al.,

2 3 6

. . -=. " e l

o

e~ ~ o m m r, . ~ m

m

F-

E

'° 1 • m

o m o m o

(~s/-I~OZH tu~) S~l

~ r

o

Z

o

r0

E

E o

( ~ H ~ )

.

(~SI-IIOZHtU:>) sJX

0 "

U .

m

\ m o m o

"~ "7 (:~st-I/O~H tu~) sJX

o

- m

- ~ r

I I I

I I .

. o

- ~ r

O " CP

U .

Z

©

(:>a £ Iq,'OZ H tu.~) sIX

U

<~

& ©

N

m

~ s

~ m

~.~ 0

m ~

~ ° ~

~ E ~.~

g e ° ~

~N e g

~.~

237

1975). This was confirmed by the fact that no difference in X values was obtained whether the mouth was kept open or closed.

We are not aware of other available data on the mechanical behaviour of canine lungs at high frequencies (2-26 Hz) in conscious intact dogs. However, good correspondance between the measurements obtained using the FOT and those obtained by other techniques has been established in dogs (Harf et al., 1985), as well as in human subjects (Landser et al., 1976; Frank et al., 1971; Nagels et al., 1980; Pimmel et al., 1981) and in calves (Gustin et al., 1988a).

The R values in the 14 healthy conscious beagle dogs obtained by FOT ranged rs 1

from 4.3 cmH20 L- s at 4 Hz to 10.9 cmH20 L-is at 26 Hz. Reported values of total pulmonary resistance measured in conscious dogs at spontaneous breathing frequency by the isovolume technique using an oesophageal balloon are variable (Crosfill and Widdicombe, 1961; Pickrell et al., 1971; Gillespie and Hyatt, 1974; Muggenburg and Mauderly, 1974; Dain and Gold, 1975). Muggenburg and Mauderly (1974) found a value of 20 cmH20 L-is in standing, awake dogs and Pickrell et aL (1971) found values ranging from 5.64 to 18.48 (mean 10.44) cmH20 L-is in standing, conscious beagles. Dain and Gold (1975) and Gillespie and Hyatt (1974), both workinglwith conscious tracheotomized dogs, reported values of about 1 and 1.30 cmH20 L- s, in standing and prone postures, respectively. Rrs measured in tracheotomized animals are probably relatively underestimated because of the bypass of the high resistance in the upper airways. In the FOT, the chest wall resistance is included in the measurement of Rrs. Therefore, one would expect resistance values higher than those measured by the oesophageal method. Discrepancies may also be attributed to the fact that the frequency of spontaneous breathing is approximately 0.2 Hz and Rrs by FOT should be extrapolated to this frequency.

It is difficult to compare our results with measurement of impedance at high frequency using a forced oscillation technique reported in anaesthetized dogs (Hull and Long, 1961; Fullton et al., 1982; Jackson et aL, 1984; Fredberg et aL, 1984, 1985; Harf et aL, 1985) or excised dog lungs (Van Brabandt et al., 1983; Kappos et al., 1981) because the models are far too different.

The posture of the dog, and the position of the head, have been shown to have little influence upon the results in conscious dogs, except with the head in flexion, when the frequency dependence of Rrs decreased, Rrs increased and X became negative up to 26 Hz. The same observations were made in calves (Gustin et al., 1988a) and were attributed to an enhanced shunt at the level of the upper airway, due to narrowing of the upper airways related to the head position. Observations in humans also led to the conclusion that the position of the head is not critical for routine measurements, if flexion of the neck is avoided (Michels et aL, 1991).

The FOT appears to be well suited for detection of an airway narrowing and for demonstrating its localization: Reduction in the tracheal diameter induced an increase of all Rrs values, without change in frequency dependence of Rrs, and did not modify the X values, except at higher frequencies where it was significantly decreased. These results are not surprising since only the resistive properties have been modified; no effect upon elastance or inertance was expected. After tracheal banding in dogs, Harf et al. (1985) noticed that the FOT yielded slightly underestimated values of resistance, which they attributed to the presence of a marked alinearity in the respiratory system. Although the manual compression was consistently applied by the same experimenters, wide error bars may be attributed to differences in the reduction

238

of the tracheal diameter among the animals. In histamine-induced bronchoconstriction, our results are in agreement with those

reported in human patients with obstructive lung disease; features considered as typical of obstructive lung disease include increased Rrs values at low frequencies, associated with a marked negative frequency dependence of Rrs at low frequencies, and an overall decrease in X , resulting in an increase in resonant frequency (Grimby et al., 1968; Michaelson et al., 1975; C16ment et aL, 1983; B6gin et aL, 1988). These modifications can be simulated by Mead 's model of the lung (Mead, 1969) and explained by non-homogeneous behaviour of the respiratory system. The results are also in agreement with experimental results reported following bronchoconstrict ion induced by methacholine (Duiverman et aL, 1986) and carbachol (Chinet et al., 1987), in excised dog lung after administration of histamine (Kappos et al., 1981; Fredberg et al., 1985) and also after histamine-induced bronchoconstrict ion in calves (Gustin et

al., 1988b).

C O N C L U S I O N

The F O T is less invasive than common mechanical measurements and is practical for use in conscious dogs. Our results demonstrate the ability of the technique to detect and localize airway obstruction in dogs. Reference measurements need to be obtained in different breeds of dogs, and the effects of age and weight need to be investigated before the technique can be widely applied in the measurement of respiratory mechanics in dogs with clinical respiratory disease.

REFERENCES

Art, T., Lekeux, P., Gustin, P., Desmecht D., Amory, H. and Paiva, M., 1989. Inertance of the respiratory system in ponies. Journal of Applied Physiology, 67, 534-540

B6gin, R., Cantin, A. and Landser, F.J., 1988. R6sistance respiratoire par oscillation forc6e et VEMS pour 6valuer la r6activit6 bronchique non sp6cifique. Revue des Maladies Respiratoires, 5, 389-392

Chinet, T., Harf, A. and Loring, H., 1987. Forced oscillatory conductance of the respiratory system allows assessment of bronchopulmonary reactivity. Bulletin Europ(en Physiopathologie Respiratoire, 23, 322S

Cl6ment, J., Landser F.J. and Van de Woestijne, K.P., 1983. Total resistance and reactance in patients with respiratory complaints with and without airways obstruction. Chest, 83, 215-220

Crosfill, M.L. and Widdicombe, J.G., 1961. Physical characteristics of the chest and lungs and the work of breathing in different mammalian species. Journal of Physiology, 158, 1-14

Dain, D. and Gold, W.M., 1975. Mechanical properties of the lungs and experimental asthma in conscious allergic dogs. Journal of Applied Physiology, 38, 96-100

Duivermann, E.J., Neijens, H.J., Vander Snee-Van Smaalen M. and Kerrebijn, K.J., 1986. Comparison of forced oscillometry and forced expirations for measuring dose-related responses to inhaled metacholine in asthmatic children. Bulletin Europ(en Physiopathologie Respiratoire, 22, 433-436

Frank, N.R., Mead, J. and Whinenberger, J.L., 1971. Comparative sensitivity of four techniques for measuring changes in respiratory flow resistance in man. Journal of Applied Physiology, 31, 934-938

Fredberg, J.F., Keefe, D.H., Glass, G.M., Castile, R.G. and Frantz, I.D., 1984. Alveolar pressure non- homogeneity during small-amplitude high-frequency oscillation. Journal of Applied Physiology, 57, 788-800

Fredberg, J.F., Ingram, R.H., Castile, R.G., Glass, G.M. and Drazen, J.M., 1985. Nonhomogeneity of lung response to inhaled histamine assessed with alveolar capsules. Journal of Applied Physiology, 58, 1914-1922

Fullton, J.M., Hayes, D.M. and Pimmel, R.L., 1982. Pulmonary impedance in dogs measured by forced random noise with a retrograde catheter. Journal of Applied Physiology, 52, 725-733

Gillespie, D.J. and Hyatt, R.E., 1974. Respiratory mechanics in the unanesthetized dog. Journal of Applied Physiology, 36, 98-102

239

Grimby, G., Takishima, T., Graham, W., Macklem, P. and Mead, J., 1968. Frequency dependence of flow resistance in patients with obstructive lung disease. The Journal of Clinical Investigation, 47, 1455-1465

Gustin, P., Dhem, A.R., Lomba, F., Lekeux, P., Van de Woestijne, K.P. and Landser, F.J., 1988a. Measurement of total respiratory impedance in calves by forced oscillation technique. Journal of Applied Physiology, 64, 1786-1791

Gustin, P., Dhem, A.R., Lekeux, P., Lomba, F., Landser, F.J. and Van de Woestijne, K.P., 1988b. Investigation of the effects of histamine inhalation on the tracheobronchial tree of calves by the forced oscillation technique. Journal of Veterinary Pharmacology and Therapeutics, 11, 374-380

Harf, A., Decramer, M., Zin, A., Milic-Emili, J. and Chang, K., 1985. Respiratory resistance in dogs by the single-breath and forced oscillation methods. Journal of Applied Physiology, 59, 262-265

Hull, W.E. and Long, E.C., 1961. Respiratory impedance and volume flow at high frequency in dogs. Journal of Applied Physiology, 16, 439-443

Jackson, A.C., Watson, J.W. and Kotlikoff, M.I., 1984. Respiratory system, lung and chest wall impedances in anesthetized dogs. Journal of Applied Physiology, 57, 34-39 '

Kappos, A.D., Rodarte, J.R. and Lai-Fook, S.J., 1981. Frequency dependence and partitioning of respiratory impedance in dogs. Journal of Applied Physiology, 51, 621-628

Landser, F.J., Nagels, J., Demedts, M., Billiet, L. and Van de Woestijne, ICP., 1976. A new method to determine frequency characteristics of the respiratory system. Journal of Applied Physiology, 41, 101-106

Mead, J., 1969. Contribution of compliance of airways to frequency dependent behaviour of lungs. Journal of Applied Physiology, 26, 670-673

Michels, A., Loveniers, M. and Van de Woestijne, K.P., 1991. Influence of head position on the impedance of the respiratory system. European Respiratory Reviews, 1, 229-231

Michaelson, E.D., Grassman, E.D. and Peters, W.R., 1975. Pulmonary mechanics by spectral analysis of forced random noise. Journal of Clinical Investigation, 56, 1210-1230

Muggenburg, B.A. and Maudedy, J.L., 1974. Cardiopulmonary function of awake, sedated, and anesthetized beagle dogs. Journal of Applied Physiology, 37, 152-157

Nagels, J., Landser, F.J., Van der Linden, L., Cl6ment, J. and Van de Woestijne, K.P., 1980. Mechanical properties of lung and chest wall during spontaneous breathing. Journal of Applied Physiology, 49, 408-415

Pickrell, J.A., Dubin, S.E. and Elliott, J.C., 1971. Normal respiratory parameters of unanesthetized beagle dogs. Laboratory Animal Science, 21, 677-679

Pimmel, R.L., Fullton, J.M., Ginsberg, J.F., Hazucha, M.J., Haak, E.D., McDonnell, W.F. and Bromberg, P.A., 1981. Correlation of airway resistance with forced random noise resistance parameters. Journal of Applied Physiology, 51, 33-39

Tsai, M.J., Pimmel, R.L., Stiff, E.J., Bromberg, P.A. and Hamlin, R.L., 1977. Respiratory parameter estimation using forced oscillatory data. Journal of Applied Physiology, 43, 322-330

Van Brabandt, H., Cauberghs, M., Verbeken, E., Moerman, P., Lauweryns, J.M. and Van de Woestijne, K.P., 1983. Partitioning of pulmonary impedance in excised human and canine lungs. Journal of Applied Physiology, 55, 1733-1742

(Accepted: 7 May 1993)