Upsala J Med Sci 97: 1 15-126 

Ariway Pressures During Positive Pressure Ventilation 
with Superimposed Oscillations Before and After Lung 

Injury in the Cat 
By A. Jonzon, T. Norsted, and G. Sedin 

The Department of Pediatrics, University Hospital and the Department of Physiology and Medical 
Biophysics, University of Uppsala, Uppsala, Sweden 

ABSTRACT 

This study was made to determine how oscillations superimposed 

on intermittent positive pressure ventilation (IPPV) influence 

the arterial blood gases, pH and the airway pressures during 

adequate alveolar ventilation i.e. at inhibition of 

inspiratory activity, before and after experimentally induced 

lung injury in the anaesthetized cat. Two IPPV frequencies 

were studied. The lung was injured by instillation of xanthine 

oxidase into the upper airways during IPPV. The peak, mean and 

end-expiratory intrapleuraland airway (intratracheal) pressu- 

res at two levels were measured and the arterial blood gases 

and pH were determined at inhibition of inspiratory activity 

with and without superimposition of oscillations on the 

ventilatory pattern. Before lung injury, superimposed oscilla- 

tions lowered the airway pressures only at an IPPV rate of 15 

breaths per minute (b.p.m.). After lung injury, such oscilla- 

tions increased the airway pressures only at 15 b.p.m. The 

airway pressures were always lower at 60 than at 15 b.p.m. 

115 



INTRODUCTION 

We have previously shown that superimposition of oscillations 

on intermittent positive pressure ventilation (IPPV) lowers 

the mean intratracheal airway pressure at IPPV frequencies of 

15 breaths per minute (b.p.m.) but not at 100 b.p.m. (3). The 

pressure reduction with superimposed oscillations may be of 

value during ventilation of newborn infants, since high airway 

pressures during IPPV together with oxygen administration are 

considered to play an important pathogenetic role in broncho- 

pulmonary dysplasia (5). In infants with interstitial emphyse- 

ma, superimposed oscillations are of advantage in most cases 

( 4 , 8 ) .  

Experimental studies of the effects of ventilatory patterns 

on the airway pressures and arterial blood gases in the 

presence of a lung disease require a good and stable experi- 

mental model which allows repeated measurements under stan- 

dardized conditions. Recently Saugstad et a1 (7) demonstrated 

that injection of xanthine oxidase into the trachea results 

in lung damage with formation of perivascular oedema, 

dilatation of lymph vessels and infiltration of neutrophils 

with reduced lung compliance. The lung damage in this experi- 

mental model has some histopathological characteristics in 

common with that which may be seen in prolonged neonatal 

ventilation ( >  5 days). The model could therefore be appli- 

cable in studies of the way in which IPPV with and without 

superimposed oscillations influences the airway pressures and 

arterial blood gases after lung injury. 

The present study was undertaken to determine whether pressure 

116 



oscillations superimposed on IPPV influence the arterial blood 

gases and pH and the airway pressures differently at inhibi- 

tion of inspiratory activity in cats with normal and injured 

lungs. 

METHODS 

Subjects and preparation 

Seven cats, weighing between 2.5 and 4.7 kg, were studied. 

Anaesthesia was induced with chloroform and maintained with 

intermittent injections of chloralose (Merck AG, G F R ) .  

Catheters were introduced through the femoral vein and artery 

into the inferior vena cava and aorta. The left phrenic nerve 

was exposed through a frontal, medial incision in the neck. 

An endotracheal tube was inserted just below the larynx and 

a ligature was placed so that no air could leak between the 

endotracheal tube and the tracheal wall. Two equally long 

catheters had previously been attached to the endotracheal 

tube so that one had its tip 1 cm above the carina and the 

other had its tip 5 cm further down. To measure the pleural 

pressure, another catheter was inserted through the rib cage, 

without letting any air into the pleural space. 

Measurements 

Airway pressures were measured at the tip of the endotracheal 

tube ("tip pressuret1) and also about 1 cm above the carina and 

5 cm below the tip of the endotracheal tube ("distal pres- 

surev1). Pleural pressure was measured through a catheter 

117 



inserted into the pleural space. Arterial blood pressure was ' 

measured through a catheter placed in the abdominal aorta. All 

catheters were connected to identical transducers (Druck AG, 

GFR) and amplifiers (Hellige AG, GFR) . All signals were ampli- 
fied with an 8-channelled medical amplifier system (Hellige 

AG, GFR) and fed to a recorder (Recorder 330-P, Hellige AG, 

GFR). Measurements of arterial blood gases and pH were made 

with an automatic acid-base analyser (Radiometer, Denmark). 

The phrenic nerve activity was recorded by placing the left 

phrenic nerve on bipolar hook electrodes. The nerve and 

electrodes were immersed in mineral oil. For amplification a 

Neurolog system (Digitimer, U.K.) was used. 

Ventilators 

A Siemens Elema Servo Ventilator (SV 9OOC) was used in the 

experiments. A positive end-expiratory pressure (PEEP) of 0.5 

kPa was applied. The set PEEP was not altered when oscillatory 

ventilation was superimposed. Oscillations of the ventilation 

gas were accomplished by attaching metal bellows to the tubing 

between the ventilator and the endotracheal tube. The bellows 

were controlled by a motor on which the stroke volume and the 

number of strokes per minute could be set independently. The 

stroke volume of the bellows in these experiments was 19 ml. 

The number of strokes was 570-600/minute. 

118 



Experimental procedure 

Before any measurements were made a check was made to see that 

all airway pressures were at zero during expiratory rest with 

the cat breathing spontaneously and also that the acid-base 

status was normal and that the base excess (BE) was above -5 

mEg/l. The experiments were performed during ventilation with 

15 or 60 breaths per minute with the ventilator in volume- 

controlled mode. The inspiratory time was always 3 3 %  of the 

cycle and a plateau of 10 % was used. 

First the cat was ventilated at a frequency of 15 or 60 per 

minute, and the minute ventilation was slowly increased until 

the phrenic nerve activity was inhibited. About 20 seconds 

after inhibition of inspiratory activity, measurements of the 

peak, mean and end-expiratory airway and pleural pressures 

were made. A blood sample was drawn for determination of 

arterial blood gases and pH. Oscillations were then superim- 

posed and the tidal volume of the ventilator was reduced until 

the phrenic nerve activity reappeared; it was then again 

slowly increased until the phrenic nerve activity was inhib- 

ited and the same measurements were repeated. Subsequently 10 

U/kg b.w. of xanthine oxidase was injected into the airways 

and the fraction of inspired oxygen ( F I 0 2 )  was increased to 

0 . 4 .  After 30 minutes the same procedure was repeated, with 

ventilation to inhibition without and with superimposed 

oscillation. 

119 



Histopathology of the lungs 

After the experiment the lungs were perfused with a mixture 

of formaldehyde (10%) and glutaraldehyde ( 4 % )  through a 

catheter inserted into the pulmonary artery. The lungs were 

then removed and placed in a 4 %  solution of formaldehyde. 

Histological examination revealed a non-homogeneous distribu- 

tion of atelectatic areas with capillary dilatation, oedema, 

and fluid accumulation in the alveolar spaces. In some animals 

there was also local infiltration of leucocytes in the 

atelectatic areas. 

RESULTS 

Before lung injury, without oscillations, at inhibition 

At inhibition of inspiratory activity, without oscillations 

the arterial POzl P C 0 2 ,  pH and BE were the same at 15 and 60 

b.p.m. The peak and mean airway pressures were lower at 60 

than at 15 b.p.m. (p<O.Ol). 

Before lung injury, with oscillations, at inhibition 

When oscillations were superimposed the arterial blood gases 

and pH were the same as without superimposed oscillations at 

inhibition of inspiratory activity. With oscillations, as 

without the peak and mean airway pressures were lower at 60 

than at 15 b.p.m. (p<O.Ol). At 15 b.p.m. the peak and mean 

airway pressures were lower with than without superimposed 

oscillations (p<O.O1). 

120 



N O R M A L  L U N G S  

a O " i  18 

J 
o f  - 

15 60 

I N J U R E D  LUNGS 

15 60 

15 60 

V e n t i l a t o r y  frequency 

Fiq. 1 

NORMAL LUNGS I N J U R E D  LUNGS 

N 
0 
0 a 
m 
a, 

- 
.- 
L 

c L 

Q 

8 -  

6 -  
L------ 

L -  

15 60 15 60 

Ventilatory frequency 

Arterial P O 2  and P C 0 2 ,  pH and base excess (BE) during ventila- 
tion at 15 and 60 b.p.m. at inhibition of inspiratory activity 
without superimposed oscillations (whole lines) and with 
superimposed oscillations (dotted lines). Triangles represent 
experiments made after lung injury. 



E
N

D
 O

F
 

E
T

-T
U

B
E

 

N
O

R
M

A
L 

LU
N

G
S

 
IN

JU
R

E
D

 
LU

N
G

S
 

k
P

a
* 

0
-

 

z
 

I
 5 
-

0
2

-
 

-0
 L

 

k7
;l 

1
5

 

A 

"-# 
- 
-
-
Y

 
- 

F
IV

E
 C

M
 

B
E

LO
W

 
TI

P
 O

F
 E

T
-T

U
B

E
 

N
O

R
M

A
L 

LU
N

G
S

 
lN

JU
R

E
D

 
LU

N
G

S
 

u
 

15
 

6
0

 

k
P

a
* 

A
 

0
6

-
 
I

-
=

 
1
 

a
 

0
4

-
 

a
 

0
2

-
 

w
 

W
 

IP
P

V
 

IP
P

V
+

o
sc

 

-
 

IP
P

V
 

IP
P

V
 +

 0
5

c
 

u
 

0
-
 -
 - u
 

60
 

6
0

 
15

 
6

0
 

V
e

n
ti

la
to

ry
 f

re
q

u
en

cy
 

V
e

n
ti

la
to

ry
 

fr
e

q
u

e
n

c
y

 

P
LE

U
R

A
L 

N
O

R
M

A
L 

LU
N

G
S

 
IN

JU
R

E
D

 L
U

N
G

S
 

I -0 L 
IP

P
V

'O
S

C
 

I 
-
 lPP

V
 

60
 

15
 

6
0

 

V
e

n
ti

la
to

ry
 f

re
q

u
en

cy
 

Fi
g.
 
2
 

P
e
a
k
,
 m
e
a
n
 a
nd
 e
n
d
-
e
x
p
i
r
a
t
o
r
y
 a
i
r
w
a
y
 a
nd
 p
l
e
u
r
a
l
 p
r
e
s
s
u
r
e
s
 d
u
r
i
n
g
 v
e
n
t
i
l
a
t
i
o
n
 a
t 
1
5
 a
nd
 

6
0

 
b.
p.
m.
, 

a
t
 
i
n
h
i
b
i
t
i
o
n
 
of
 
i
n
s
p
i
r
a
t
o
r
y
 
a
c
t
i
v
i
t
y
 
w
i
t
h
o
u
t
 
s
u
p
e
r
i
m
p
o
s
e
d
 
o
s
c
i
l
l
a
t
i
o
n
s
 

(
w
h
o
l
e
 l
in
es
) 
an
d 
w
i
t
h
 
s
u
p
e
r
i
m
p
o
s
e
d
 
o
s
c
i
l
l
a
t
i
o
n
s
 
(
d
o
t
t
e
d
 l
in
es
).
 
T
h
e
 a
i
r
w
a
y
 p
re
ss
ur
es
 

a
r
e
 m
e
a
s
u
r
e
d
 
a
t
 t
h
e
 d
i
s
t
a
l
 e
nd
 
of
 t
h
e
 e
n
d
o
t
r
a
c
h
e
a
l
 
(E
T)
 t
u
b
e
 a
nd
 
5
 c
m
 b
el
ow
 
t
h
e
 e
nd
 
of
 

t
h
e
 E
T 
tu
be
. 
T
r
i
a
n
g
l
e
s
 r
e
p
r
e
s
e
n
t
 
e
x
p
e
r
i
m
e
n
t
s
 m
a
d
e
 
a
f
t
e
r
 
lu
ng
 
in
ju
ry
. 



After lung injury, without oscillations, at inhibition 

The arterial PC02 was lower after than before lung injury 

(p<O.Ol - 0.05). P O 2  was always higher after than before lung 
injury due to the higher F I 0 2 .  Arterial pH and BE were the 

same as before lung injury. The peak and mean airway pressures 

were lower at 60 than at 15 b.p.m. (p<O.Ol). Pleural pressure 

was lower at 60 than at 15 b.p.m. (~~0.05). 

After lung injury, with oscillations, at inhibition 

The arterial PC02 was lower after than before lung injury 

(p<O.Ol - 0.05). PO2 was always higher after than before lung 
injury, because of the higher F I 0 2 .  Arterial pH and BE were 

the same as before lung injury. The peak and mean airway 

pressures were lower at 60 than at 15 b.p.m. (p<O.Ol). Pleural 

pressure was lower at 60 than at 15 b.p.m. (pCO.05). The peak 

and mean airway pressures were higher with than without 

superimposed oscillations at 15 (p<O.Ol) but not at 60 b.p.m.. 

DISCUSSION 

This study shows that airway pressures at inhibition of 

inspiratory activity differ before and after lung injury. 

Thus, the airway pressures were higher after than before lung 

injury at a ventilatory frequency of 60 b.p.m., but the same 

under both conditions at a ventilatory frequency of 15 b.p.m. 

Superimposition of oscillation lowered the airway pressures 

before but not after lung injury. 

9-928572 123 



Attempts have been made to standardize animal models of 

bronchopulmonary dysplasia to be used in experimental studies 

( 6 ) .  Of all species studied, only the newborn baboon has been 

kept on a ventilator long enough to mimic stages 2 and 4 of 

this condition ( 6 ) .  Even if damage caused by long-term 

ventilation may differ from that induced by chemical agents, 

the latter is the only alternative in most experimental 

situations. We therefore considered several alternatives such 

as oxygen injury, oleic acid injury and injury caused by 

xanthine oxidase. 

Instillation of oleic acid into the lungs results in a very 

unstable condition, usually with progressive deterioration of 

lung function, and does not permit valid comparisons of 

different respirator settings. Administration of 100% oxygen 

needs 4 to 5 days to cause injury with pulmonary oedema and 

alveolar damage. We have therefore chosen the technique used 

by Saugstad et a1 (7) with instillation of xanthine oxidase 

into the airways. The adverse effect of xanthine oxidase may 

at least partly be explained by formation of superoxide 

radicals (2), and the amount of oxygen radicals formed is 

dependent on the hypoxanthine concentration, oxygen concentra- 

tion and the presence of xanthine oxidase (1). We have made 

no attempt to characterize in detail the lung injury caused 

by instillation of xanthine oxidase into the airways, but have 

verified that the histopathological lung changes in our cats 

resemble those described by Saugstad et a1 (7). We have also 

considered it important to document the fact that during the 

period of the study there were no undue changes in acid-base 

status which could have influenced the inspiratory activity 

124 



and its inhibition. 

We chose the model described by Saugstad et a1 (7) in our 

attempt to determine whether superimposed oscillations could 

be of advantage after a lung injury that has some features in 

common with a neonatal lung disease seen after some days of 

treatment. During this phase the lung disease may necessitate 

the use of high ventilatory pressures to achieve adequate 

alveolar ventilation. X-ray of the lung may then show variable 

aeration and sometimes localized densities. 

Our results do not support the hypothesis that superimposed 

oscillations could be beneficial in reducing airway pressures 

after lung injury induced with xanthine oxidase. This injury 

is non homogenous with capillary dilatation, oedema and fluid 

accumulation in alveolar spaces and would probably cause 

localized densities on chest x-ray examination. Clinical 

results indicate, however, that superimposed oscillations are 

beneficial during ventilation of lungs (8) with interstitial 

emphysema without localized densities. 

CONCLUSIONS 

2. 

In cats with injured lungs: 

1. Oscillations superimposed on IPPV do not lower the 

airway pressure at inhibition of inspiratory activ- 

ity. 

A low arterial PC02 is needed to inhibit inspiratory 

activity. 

125 



ACKNOWLEDGEMENTS 

This study was supported by grants from the Swedish Medical 

Research Council (19M-7544), Stiftelsen Allmanna BB:s Minnes- 

fond, the Samaritan Foundation and the Sigurd and Elsa Golje 

Memorial Fund. The authors thank Siemens AB, Stockholm for 

kindly providing the Servoventilator 9OOC. They are also 

indebted to Barbro Kjallstrom, Harry Johansson, Gunno Nilsson 

and Erik Ekstrom for technical assistance. We thank Susanne 

Thorell and Sabina Albinsson for typing assistance. 

REFERENCES 

1. 

2. 

3. 

4. 

5. 

6. 

7. 

8. 

Fridovich I. Quantitative aspects of the production 
of superoxide anion radical by mild xanthine 
oxidase. J Biol Chem, 245:4053, 1970. 

Johnson KJ, Fanton JC, Kaplan J, Ward P. In vivo 
damage of rat lungs by oxygen metabolites. J Clin 
Invest, 67:983, 1981. 

Jonzon A, Tantalean J, Norsted T, Sedin G. Airway 
pressures during positive pressure ventilation with 
superimposed oscillations. Upsala J Med Sci, 95:53- 
61, 1990. 

Ng KPK, Easa D Management of interstitial emphysema 
by high frequency low positive pressure hand venti- 
lation in the neonate. J Pediatr, 95:117-118, 1979. 

Phillip AGS, Oxygen plus pressure plus time: the 
etiology of bronchopulmonary dysplasia. Pediatrics, 
55 : 44-50, 1975. 

Recommendations of the Workshop on Bronchopulmonary 
Dysplasia. J Pediatr, 95:920, 1979. 

Saugstad OD, Hallman M, Becher G, Oddoy A, Lium B, 
Lachmann B. Respiratory failure caused by intratrac- 
heal saline. Additive effect of xanthine oxidase. 
Biol Neonate, 54: 61-67, 1988. 

Sedin,G., Jonzon,A. Kaniklides,C. Superimposed high- 
frequency oscillatory ventilation in the treatment 
of neonatal interstitial pulmonary emphysema. In 
manuscript. 

126