Eur J Pediatr (1992) 151 : 846-850
European Journal of
Pediatrics
9 Springer-Verlag1992
Dynamic lung inflation during high frequency oscillation in neonates E. W. Hoskyns, A. D. Milner, and I. E. Hopkin Department of Neonatal Medicine and Surgery, City Hospital, Nottingham, United Kingdom Received September 25, 1990 / Accepted after revision March 25, 1992
Abstract. The effects of high frequency oscillation (HFO) on dynamic lung inflation were examined in 22 neonates ventilated for respiratory disease. H F O was combined with conventional ventilation and a series of frequencies from 2-25 Hz was tested. Dynamic lung inflation was measured using a jacket plethysmograph which was converted to a measure of alveolar pressure using the complicance of the respiratory system obtained during conventional ventilation. The results showed an increase in dynamic lung inflation with frequency such that volume increased by 0.4 ml for each increase of 10 Hz. Alveolar pressure increased by 1 . 2 c m H a O for each increase of 10Hz. Dynamic lung inflation also increased with increased volumes of oscillation.
Key words: High frequency oscillation - Dynamic lung inflation - Positive end-expiratory pressure - Neonates
Introduction Artificial ventilation at high frequencies has provoked much interest in the last few years. This is particularly evident in neonatal ventilation. At high frequencies with conventional ventilators [16], jet ventilators [3] or oscillators [1] air trapping may occur. This may account for some of the observed improvement seen when high frequency ventilation of any sort is used. It is important to know the degree of air trapping as the increase in alveolar pressure that this reflects may have negative effects in terms of pressure damage to the lungs (eg. air leaks) and effects on cardiac output. The explanation for this phenomenon is that the airway walls are not completely rigid, and they tend to narrow on expiration when the pressure in the lumen is less Correspondence to: A.D.Milner, Division of Paediatrics, St. Thomas's Hospital, London SEI 7EH, UK Abbreviations: CPAP = continuous positive airways pressure;
ET = endotracheal; HFO = high frequency oscillation; IPPV = intermittent positive airways pressure; PEEP = positive end expiratory pressure
than in the surrounding tissue. This results in expiratory impedance being higher than inspiratory impedance and may in extreme circumstances lead to airway collapse. There is also an effect from the branching nature of the airways [12] which also tends to increase expiratory impedance. If the inspiratory time: expiratory time ratio is 1 : 1 then at the end of one cycle there will be a net increase of gas in the lungs. Over multiple cycles the gas in the lungs will continue to build up until the excess pressure is enough to overcome the difference in impedance. The magnitude of this effect depends on the size of the difference in impedance, the frequency of ventilation and also on the expiratory time constant of the lung. As the effect depends on a change in impedance within the lung it is not measureable at the airway opening. Thus mean alveolar pressure may be considerably higher than mean airways pressure measured either in the trachea or at the top of the endotracheal tube. The effect can be assessed by a volume change in the lungs or by some measure of true alveolar pressure. These two are related by the static compliance of the respiratory system [15]. Because it can be measured in a variety of ways it is known by various synonyms: dynamic lung inflation, dynamic hyperinflation, air trapping, mean alveolar pressure, alveolar positive end expiratory pressure (PEEP), auto-PEEP or inevitable PEEP. Different forms of high frequency ventilation are not equivalent in the degree of dynamic lung inflation produced. Froese and Bryan [8] have proposed a classification based on whether the expiratory phase is active (providing additional suction) or passive (relying on the elastic recoil of the respiratory system). This focuses on a significant difference between high frequency oscillation (HFO) and the other systems. In expiration the H F O generator is actively aiding expiration by sucking gas back into the delivery chamber whereas with conventional ventilators or jet ventilators the motive force is all directed towards inspiratory flow. This is particularly important when dynamic lung inflation is considered as this effect should be less marked during HFO. We have investigated the effect of H F O on dynamic lung inflation in neonates ventilated for respiratory disease. H F O was combined with conventional ventilation and the effect of frequency and volume of oscillation were examined.
847 motor across a moveable fulcrum. This arrangement allows the volume delivered to be altered while the oscillator is running and makes the volume delivered into the circuit independent of frequency. The oscillator was connected to the suction port of the patient manifold by non-compliant plastic tubing 28 cm length by 0.7 cm diameter. The bias flow gas was humidified by the Drfiger circuit. The fulcrum was adjusted to displace 8 mI into the circuit and oscillated for each baby at a series of frequencies from 2 to 25 Hz. In a few babies this was repeated at 4 ml or 12 ml displacement. Airway pressure was measured with a pressure transducer (Validyne MP45) attached to a 21 gauge butterfly needle inserted into the ET tube at the level of the babies mouth. Changes in lung volume during HFO were measured using a jacket plethysmograph. One of two sizes was used depending on the size of the baby. The use of similar respiratory jackets has been described in older children for measuring tidal volumes [9] and can be used to assess changes in lung volume although it cannot be used to measure absolute lung volumes. The jackets are essentially a cylinder made of natural rubber which the baby lies inside. Volume changes in the chest are measured by putting the baby in the jacket and inflating it so that it apposes the baby's chest and abdomen. Any respiratory movement results in a change in pressure in the jacket. Changes of pressure in the jacket were measured with a Pye Ether UP 1 differential pressure transducer attached to one of the jacket ports by a length of non-compliant tubing. The other port was connected to a pressure dial for an approximate measure of jacket pressure and to a syringe for inflation and calibration. Before each study the jacket was inflated to a pressure of 4-6 cmH20 and calibrated by injection of 2-5 ml of air. At this pressure the response of both jackets was linear and this was not affected by the weight of the baby inside the jacket. Compliance of the jackets was approximately 55 ml c m H 2 0 - l a n d 90 ml c m H 2 0 -1 for the small and large jackets respectively. The jackets have been validated for use at high frequencies [11] but for the purposes of this study were used to measure the constant distending pressure theoretically associated with high frequency ventilation. Any damping of very high frequency components should not affect measurement of this constant distending pressure. At each frequency the oscillator was switched on for 15 s (or longer if this was needed to allow the baby to settle and a stable baseline end-expiratory volume to be obtained). The oscillator was then switched off for a further 15 s. The mean of the end expiratory volume of the 10 breaths prior to switching off the oscillator was compared with the mean of the subsequent 10 breaths and the difference taken as the degree of dynamic lung inflation (see Fig. 2). Volume changes caused by the oscillator as well as those caused by the concentional ventilator were transmitted to the jacket. The end expiratory volume was therefore calculated as the mid-point of the peak to peak variation of the oscillations. This calculation is valid provided that the waveform of the oscillation is symmetrical (i.e. sinusoidal).
Subjects A t o t a l o f 19 b a b i e s v e n t i l a t e d f o r r e s p i r a t o r y p r o b l e m s w e r e s t u d i e d o n 22 o c c a s i o n s . G e s t a t i o n a l a g e v a r i e d from 26-40 weeks (mean 29.2 weeks), post-natal age f r o m 12 h - 1 7 d a y s ( m e a n 3.6 d a y s ) a n d w e i g h t f r o m 8 2 0 2200g ( m e a n 1330g). O f t h e s e studies, 15 w e r e p e r f o r m e d on babies who were ventilated with intermittent positive airways pressures (IPPV). The ventilator settings for these babies varied with peak pressures 14-40 cmH20, end-expiratory pressures 2-4cmH20, respiratory rates 10-120/min and inspired oxygen concentrations 25%100%. Of the babies on IPPV, 5 were paralysed with p a n c u r o n i u m . T h e r e m a i n i n g 7 b a b i e s h a d less s e v e r e lung disease requiring continuous positive airways press u r e ( C P A P ) o n l y , at o x y g e n c o n c e n t r a t i o n s o f 2 1 % 35%.
Methods Informed consent was obtained from the parents prior to all studies. In the first part of the study the compliance of the respiratory system was determined by the method of Field et al. [6]. A Fleish type 0 pneumotachograph with a pressure independent bias flow to eliminate the added dead space was connected between the ventilator and the endotracheal (ET) tube to measure tidal volume and a no. 21 butterfly needle was inserted into the ET tube and attached to a pressure transducer to measure pressure changes at the airway opening. For this part of the study babies on CPAP were transfered to IPPV. Measurements of compliance were made on a minimum of 20 breaths by dividing the tidal volume by the ventilator pressure swing with an inspiratory time of at least 0.3 s. Traces of tidal volume were examined to ensure that a plateau was reached with each breath. Breaths were excluded if there was evidence of breathing movements by the baby, either by observation or by examination of the pressure and volume record but oesophageal balloons were not used in this study to validate these observations. All babies were ventilated with Dr~iger babylog ventilators (models 1 or 1R). In the second part of the study each baby was continued on the ventilation prescribed by the clinical team. HFO was given by connecting an oscillator to the patient manifold of the conventional ventilator as shown in Fig. 1. The oscillator was designed and built in the Nottingham University Medical School workshop and consists of a stainless steel syringe with a polytetrafluoroethane piston. All parts in direct contact with the ventilator circuit were removed and sterilised at the end of each study. The piston is driven by a
H.F.O. syringe
Fulcrum
•
Pressure / monrtor f
Y
J
Motor
Fig. 1. Diagram of the combined ventilation system. The conventional ventilator provides the bias flow and humidification. The oscillator is driven by an electric motor
I !
Fig. 2. Tidal volume trace from a paralysed baby. In the first part of the trace volume oscillations at 12 Hz are seen superimposed on the tidal volume trace. In the second part of the trace the oscillator is turned off. The end-expiratory volumes during and after oscillation are shown and the difference between the two (0.8 ml in this case) is the dynamic lung inflation
848 In some studies the oscillations caused a rise in pressure within the ventilator circuit. This was seen when the end-expiratory pressure during normal ventilation was lower than the average end-expiratory pressure during the preceeding period of oscillation. Every episode of oscillation was examined for the magnitude of this effect. The compliance of the respiratory system from the first part of the study was used to calculate the excess pressure (alveolarP E E P ) within the lung causing the dynamic lung inflation. The values obtained for alveolar P E E P and dynamic lung inflation were then adjusted to take account of the effect of P E E P within the ventilator circuit using the formula;
6 5 4
E
3
o
e-
2 Ial l.g e~
1
[PPV CPAP
0
AP,~tv = 1/C x AV - APpEEp.
"1
0
The results were analysed by linear regression of the volume and pressure changes against frequency.
I
I
I
10
20
30
n = 15 n =7
Frequency in HZ
Fig. 4. Mean values for alveolar PEEP are shown against frequency for the IPPV and CPAP babies. The difference between the two groups appears less than in Fig. 3
Results
Values for compliance of the respiratory system varied from 0 . 1 1 - 0 . 6 2 m l / c m H 2 0 (mean 0.40) with generally lower values for those babies on I P P V (mean 0.35) compared to those on C P A P (mean 0.51). H F O was well tolerated by the babies studied and no untoward side-effects were noticed. In the unparalysed babies starting and stopping H F O was sometimes associated with body movements which interfered with measurement of end expiratory volume. The waveform of the oscillations was examined in three of the studies chosen at r a n d o m using traces at 5, 10, 15, 20 and 25Hz. When compared to a pure sine wave these varied by less than 5 ~ in each baby. The variation in dynamic lung inflation with frequency of oscillation is shown in Fig. 3 for I P P V and C P A P babies. Despite the large variation in the results, particularly for the C P A P babies, lung volumes tend to increase with increasing frequency. This effect was statistically significant on linear regression analysis (P < 0.0001, r = 0.23, slope = 0.0423 ml/Hz) although the correlation coefficient was low. In general the volume increase was larger for the CPAP babies although the differences were not statistically significant. When the results were converted to measures of alveolar P E E P (Fig. 4) the difference between the two groups of babies was reduced.
1.2 -
1.0"
0.8&
0.6 i
G
0.4O
0.2 :
-6 >
0.0 -0.2
I
i
20
30
Frequency in HZ 7 6 O r
5
-r"
E o
.C r, 1.1.I ILl
,',
4
3 2 1 0 -1
t
I
o
3
i
10
10
20 Frequency in HZ
I
30
Fig.5. Means (+ SD) of Iung inflation and alveolar PEEP are shown for the group of paralysed babies studied. Variation in results at adjacent frequencies is reduced due to less movement artefact
.E t~ o O
>
0
~ *
-1
I
0
10
I
20 Frequency in HZ
IPPV
n = 15
CPAP
n = 7
I
30
Fig. 3. Mean values for dynamic lung inflation are shown against frequency for the IPPV and CPAP babies. Both show an increase with increased frequency but the CPAP results are more variable
Linear regression analysis of these data showed; P < 0.0001, r = 0.268, slope = 0 . 1 1 6 c m H 2 0 / H z ) . There was no relationship between the P E E P induced in the circuit and the level of alveolar P E E P measured by the jacket. The variation due to m o v e m e n t of the baby can be eliminated by using the data from babies who were paralysed during the study (Fig. 5). When this was done the variation from frequency to frequency was much reduced. Eight babies had H F O at 4 ml and 8 ml volume displacement and four had H F O at 8 ml and 12 ml (Fig. 6).
849 3-
g .I2 O
"5 >
0
---B---
8ml 1 2 mt
9
-I
I
0
10
I
n=4
I
20 Frequency in HZ
30
2*= p
European Journal of
Pediatrics
9 Springer-Verlag1992
Dynamic lung inflation during high frequency oscillation in neonates E. W. Hoskyns, A. D. Milner, and I. E. Hopkin Department of Neonatal Medicine and Surgery, City Hospital, Nottingham, United Kingdom Received September 25, 1990 / Accepted after revision March 25, 1992
Abstract. The effects of high frequency oscillation (HFO) on dynamic lung inflation were examined in 22 neonates ventilated for respiratory disease. H F O was combined with conventional ventilation and a series of frequencies from 2-25 Hz was tested. Dynamic lung inflation was measured using a jacket plethysmograph which was converted to a measure of alveolar pressure using the complicance of the respiratory system obtained during conventional ventilation. The results showed an increase in dynamic lung inflation with frequency such that volume increased by 0.4 ml for each increase of 10 Hz. Alveolar pressure increased by 1 . 2 c m H a O for each increase of 10Hz. Dynamic lung inflation also increased with increased volumes of oscillation.
Key words: High frequency oscillation - Dynamic lung inflation - Positive end-expiratory pressure - Neonates
Introduction Artificial ventilation at high frequencies has provoked much interest in the last few years. This is particularly evident in neonatal ventilation. At high frequencies with conventional ventilators [16], jet ventilators [3] or oscillators [1] air trapping may occur. This may account for some of the observed improvement seen when high frequency ventilation of any sort is used. It is important to know the degree of air trapping as the increase in alveolar pressure that this reflects may have negative effects in terms of pressure damage to the lungs (eg. air leaks) and effects on cardiac output. The explanation for this phenomenon is that the airway walls are not completely rigid, and they tend to narrow on expiration when the pressure in the lumen is less Correspondence to: A.D.Milner, Division of Paediatrics, St. Thomas's Hospital, London SEI 7EH, UK Abbreviations: CPAP = continuous positive airways pressure;
ET = endotracheal; HFO = high frequency oscillation; IPPV = intermittent positive airways pressure; PEEP = positive end expiratory pressure
than in the surrounding tissue. This results in expiratory impedance being higher than inspiratory impedance and may in extreme circumstances lead to airway collapse. There is also an effect from the branching nature of the airways [12] which also tends to increase expiratory impedance. If the inspiratory time: expiratory time ratio is 1 : 1 then at the end of one cycle there will be a net increase of gas in the lungs. Over multiple cycles the gas in the lungs will continue to build up until the excess pressure is enough to overcome the difference in impedance. The magnitude of this effect depends on the size of the difference in impedance, the frequency of ventilation and also on the expiratory time constant of the lung. As the effect depends on a change in impedance within the lung it is not measureable at the airway opening. Thus mean alveolar pressure may be considerably higher than mean airways pressure measured either in the trachea or at the top of the endotracheal tube. The effect can be assessed by a volume change in the lungs or by some measure of true alveolar pressure. These two are related by the static compliance of the respiratory system [15]. Because it can be measured in a variety of ways it is known by various synonyms: dynamic lung inflation, dynamic hyperinflation, air trapping, mean alveolar pressure, alveolar positive end expiratory pressure (PEEP), auto-PEEP or inevitable PEEP. Different forms of high frequency ventilation are not equivalent in the degree of dynamic lung inflation produced. Froese and Bryan [8] have proposed a classification based on whether the expiratory phase is active (providing additional suction) or passive (relying on the elastic recoil of the respiratory system). This focuses on a significant difference between high frequency oscillation (HFO) and the other systems. In expiration the H F O generator is actively aiding expiration by sucking gas back into the delivery chamber whereas with conventional ventilators or jet ventilators the motive force is all directed towards inspiratory flow. This is particularly important when dynamic lung inflation is considered as this effect should be less marked during HFO. We have investigated the effect of H F O on dynamic lung inflation in neonates ventilated for respiratory disease. H F O was combined with conventional ventilation and the effect of frequency and volume of oscillation were examined.
847 motor across a moveable fulcrum. This arrangement allows the volume delivered to be altered while the oscillator is running and makes the volume delivered into the circuit independent of frequency. The oscillator was connected to the suction port of the patient manifold by non-compliant plastic tubing 28 cm length by 0.7 cm diameter. The bias flow gas was humidified by the Drfiger circuit. The fulcrum was adjusted to displace 8 mI into the circuit and oscillated for each baby at a series of frequencies from 2 to 25 Hz. In a few babies this was repeated at 4 ml or 12 ml displacement. Airway pressure was measured with a pressure transducer (Validyne MP45) attached to a 21 gauge butterfly needle inserted into the ET tube at the level of the babies mouth. Changes in lung volume during HFO were measured using a jacket plethysmograph. One of two sizes was used depending on the size of the baby. The use of similar respiratory jackets has been described in older children for measuring tidal volumes [9] and can be used to assess changes in lung volume although it cannot be used to measure absolute lung volumes. The jackets are essentially a cylinder made of natural rubber which the baby lies inside. Volume changes in the chest are measured by putting the baby in the jacket and inflating it so that it apposes the baby's chest and abdomen. Any respiratory movement results in a change in pressure in the jacket. Changes of pressure in the jacket were measured with a Pye Ether UP 1 differential pressure transducer attached to one of the jacket ports by a length of non-compliant tubing. The other port was connected to a pressure dial for an approximate measure of jacket pressure and to a syringe for inflation and calibration. Before each study the jacket was inflated to a pressure of 4-6 cmH20 and calibrated by injection of 2-5 ml of air. At this pressure the response of both jackets was linear and this was not affected by the weight of the baby inside the jacket. Compliance of the jackets was approximately 55 ml c m H 2 0 - l a n d 90 ml c m H 2 0 -1 for the small and large jackets respectively. The jackets have been validated for use at high frequencies [11] but for the purposes of this study were used to measure the constant distending pressure theoretically associated with high frequency ventilation. Any damping of very high frequency components should not affect measurement of this constant distending pressure. At each frequency the oscillator was switched on for 15 s (or longer if this was needed to allow the baby to settle and a stable baseline end-expiratory volume to be obtained). The oscillator was then switched off for a further 15 s. The mean of the end expiratory volume of the 10 breaths prior to switching off the oscillator was compared with the mean of the subsequent 10 breaths and the difference taken as the degree of dynamic lung inflation (see Fig. 2). Volume changes caused by the oscillator as well as those caused by the concentional ventilator were transmitted to the jacket. The end expiratory volume was therefore calculated as the mid-point of the peak to peak variation of the oscillations. This calculation is valid provided that the waveform of the oscillation is symmetrical (i.e. sinusoidal).
Subjects A t o t a l o f 19 b a b i e s v e n t i l a t e d f o r r e s p i r a t o r y p r o b l e m s w e r e s t u d i e d o n 22 o c c a s i o n s . G e s t a t i o n a l a g e v a r i e d from 26-40 weeks (mean 29.2 weeks), post-natal age f r o m 12 h - 1 7 d a y s ( m e a n 3.6 d a y s ) a n d w e i g h t f r o m 8 2 0 2200g ( m e a n 1330g). O f t h e s e studies, 15 w e r e p e r f o r m e d on babies who were ventilated with intermittent positive airways pressures (IPPV). The ventilator settings for these babies varied with peak pressures 14-40 cmH20, end-expiratory pressures 2-4cmH20, respiratory rates 10-120/min and inspired oxygen concentrations 25%100%. Of the babies on IPPV, 5 were paralysed with p a n c u r o n i u m . T h e r e m a i n i n g 7 b a b i e s h a d less s e v e r e lung disease requiring continuous positive airways press u r e ( C P A P ) o n l y , at o x y g e n c o n c e n t r a t i o n s o f 2 1 % 35%.
Methods Informed consent was obtained from the parents prior to all studies. In the first part of the study the compliance of the respiratory system was determined by the method of Field et al. [6]. A Fleish type 0 pneumotachograph with a pressure independent bias flow to eliminate the added dead space was connected between the ventilator and the endotracheal (ET) tube to measure tidal volume and a no. 21 butterfly needle was inserted into the ET tube and attached to a pressure transducer to measure pressure changes at the airway opening. For this part of the study babies on CPAP were transfered to IPPV. Measurements of compliance were made on a minimum of 20 breaths by dividing the tidal volume by the ventilator pressure swing with an inspiratory time of at least 0.3 s. Traces of tidal volume were examined to ensure that a plateau was reached with each breath. Breaths were excluded if there was evidence of breathing movements by the baby, either by observation or by examination of the pressure and volume record but oesophageal balloons were not used in this study to validate these observations. All babies were ventilated with Dr~iger babylog ventilators (models 1 or 1R). In the second part of the study each baby was continued on the ventilation prescribed by the clinical team. HFO was given by connecting an oscillator to the patient manifold of the conventional ventilator as shown in Fig. 1. The oscillator was designed and built in the Nottingham University Medical School workshop and consists of a stainless steel syringe with a polytetrafluoroethane piston. All parts in direct contact with the ventilator circuit were removed and sterilised at the end of each study. The piston is driven by a
H.F.O. syringe
Fulcrum
•
Pressure / monrtor f
Y
J
Motor
Fig. 1. Diagram of the combined ventilation system. The conventional ventilator provides the bias flow and humidification. The oscillator is driven by an electric motor
I !
Fig. 2. Tidal volume trace from a paralysed baby. In the first part of the trace volume oscillations at 12 Hz are seen superimposed on the tidal volume trace. In the second part of the trace the oscillator is turned off. The end-expiratory volumes during and after oscillation are shown and the difference between the two (0.8 ml in this case) is the dynamic lung inflation
848 In some studies the oscillations caused a rise in pressure within the ventilator circuit. This was seen when the end-expiratory pressure during normal ventilation was lower than the average end-expiratory pressure during the preceeding period of oscillation. Every episode of oscillation was examined for the magnitude of this effect. The compliance of the respiratory system from the first part of the study was used to calculate the excess pressure (alveolarP E E P ) within the lung causing the dynamic lung inflation. The values obtained for alveolar P E E P and dynamic lung inflation were then adjusted to take account of the effect of P E E P within the ventilator circuit using the formula;
6 5 4
E
3
o
e-
2 Ial l.g e~
1
[PPV CPAP
0
AP,~tv = 1/C x AV - APpEEp.
"1
0
The results were analysed by linear regression of the volume and pressure changes against frequency.
I
I
I
10
20
30
n = 15 n =7
Frequency in HZ
Fig. 4. Mean values for alveolar PEEP are shown against frequency for the IPPV and CPAP babies. The difference between the two groups appears less than in Fig. 3
Results
Values for compliance of the respiratory system varied from 0 . 1 1 - 0 . 6 2 m l / c m H 2 0 (mean 0.40) with generally lower values for those babies on I P P V (mean 0.35) compared to those on C P A P (mean 0.51). H F O was well tolerated by the babies studied and no untoward side-effects were noticed. In the unparalysed babies starting and stopping H F O was sometimes associated with body movements which interfered with measurement of end expiratory volume. The waveform of the oscillations was examined in three of the studies chosen at r a n d o m using traces at 5, 10, 15, 20 and 25Hz. When compared to a pure sine wave these varied by less than 5 ~ in each baby. The variation in dynamic lung inflation with frequency of oscillation is shown in Fig. 3 for I P P V and C P A P babies. Despite the large variation in the results, particularly for the C P A P babies, lung volumes tend to increase with increasing frequency. This effect was statistically significant on linear regression analysis (P < 0.0001, r = 0.23, slope = 0.0423 ml/Hz) although the correlation coefficient was low. In general the volume increase was larger for the CPAP babies although the differences were not statistically significant. When the results were converted to measures of alveolar P E E P (Fig. 4) the difference between the two groups of babies was reduced.
1.2 -
1.0"
0.8&
0.6 i
G
0.4O
0.2 :
-6 >
0.0 -0.2
I
i
20
30
Frequency in HZ 7 6 O r
5
-r"
E o
.C r, 1.1.I ILl
,',
4
3 2 1 0 -1
t
I
o
3
i
10
10
20 Frequency in HZ
I
30
Fig.5. Means (+ SD) of Iung inflation and alveolar PEEP are shown for the group of paralysed babies studied. Variation in results at adjacent frequencies is reduced due to less movement artefact
.E t~ o O
>
0
~ *
-1
I
0
10
I
20 Frequency in HZ
IPPV
n = 15
CPAP
n = 7
I
30
Fig. 3. Mean values for dynamic lung inflation are shown against frequency for the IPPV and CPAP babies. Both show an increase with increased frequency but the CPAP results are more variable
Linear regression analysis of these data showed; P < 0.0001, r = 0.268, slope = 0 . 1 1 6 c m H 2 0 / H z ) . There was no relationship between the P E E P induced in the circuit and the level of alveolar P E E P measured by the jacket. The variation due to m o v e m e n t of the baby can be eliminated by using the data from babies who were paralysed during the study (Fig. 5). When this was done the variation from frequency to frequency was much reduced. Eight babies had H F O at 4 ml and 8 ml volume displacement and four had H F O at 8 ml and 12 ml (Fig. 6).
849 3-
g .I2 O
"5 >
0
---B---
8ml 1 2 mt
9
-I
I
0
10
I
n=4
I
20 Frequency in HZ
30
2*= p
