Anaesthesia, 1995, Volume 50, pages 601-606

APPARATUS

The Hayek Oscillator

Nomograms for tidal volume and minute ventilation using external high frequency oscillation

A. J . PETROS, S. S. D. F E R N A N D O , V. S. SHENOY AND N . M . AL-SAADY

Summary
The Hayek Oscillator is a recent development in external high ,frequency centilution and is a useful mode of ventilatory support
during anaesthesia, in chronic respiratory failure and weaning in intensi1.e care. The Hayek Oscillator is new and its application
is growing, as are the number of patients who have benefited,from its use. Horcerer, there are no clear guidelines on how best to
adjust the oscillator to achieve optimum ventilation. A simple method qf predicting changes in tidal volume and minute ventilation
following adjustment wouldmake the oscillator more useful. W epresent nomograms,fi)r tidal colume, minute oentilation andeffective
alveolar ventilation when the three variables, oscillator,frequency, mean chamber pressure and peak-to-trough pressure span were
adjusted. Thefrequency-tidal volume relationship was unaflected by a mean chamber pressure o f 0 , - 5, - 10 crnH20,but altered
with changes in peak-to-trough pressure span. W e have also determined the eflkct of’ increasing negutiiie extrathoracic pressure
on functional residual capacity. The relationship between tidal volume and ,frequency ua s non-lineur and related to the
peak-to-trough pressure span. Mean functional residual capacity signlficantlj, increased,from 2.25 1 ( S E M 0.10) without the cuirass
at rest to 2.61 1 ( S E M 0.14) at - 10 cmHzO dp < 0.05;n = 5) and 2.47 ( S E M 0.12) at - 20 c.mH20 of mean chamber pressure.
Vital capacity was unchanged by increasing extrathoracic pressure as M’US total lung cupacity .

Key words
Equipment; Hayek Oscillator, nomograms.

With the introduction of the Hayek OscillatorTM(Flexco
Medical Instruments AG, Zurich, Switzerland) (Fig. 1)
external high frequency oscillation (EHFO) is a practical
option in ventilatory support. It has been used to improve
gas exchange in adults with acute respiratory failure
on the intensive therapy unit [ l , 21, has been shown
to increase pulmonary blood flow in children following
cardiac surgery [3] and has even been used to ventilate
anaesthetised and paralysed patients undergoing micro-
laryngeal surgery, as tracheal intubation was not necessary
[4, 51.
The Hayek Oscillator can operate around a negative
pressure baseline. Previous attempts at EHFO with
compression alone have resulted in a gradual reduction in
functional residual capacity [6] and lung collapse [7];
they relied upon the elastic recoil of the chest wall to
restore normal lung volumes. By operating around a negative
baseline there is more control over end expiratory lung
volume and the reduction in functional residual capacity can
be minimised.
The mean pressure within the cuirass is termed mean
chamber pressure. A positive end expiratory pressure of
+ 5 cmHIO will compress the chest and a negative
inspiratory pressure of - 5 cmHzO will expand it, resulting
in an overall mean chamber pressure of OcmH20. The
pressure difference between end expiratory (+ 5 cmH,O) and
end inspiratory chamber pressure ( - 5 cmHzO) is termed,
the span. Changes in these two variables together with
variations in frequency have been shown to alter minute and
tidal volume in an animal model [8]. However, the effect of
these changes has not been characterised in humans. We
report the frequencyltidal volume and frequency/minute

A.J. Petros*, FFARCSI, S.S.D. Fernandof, MD, MRCP, PhD, V.S. Shenoyt, MSc, PhD, N.M. Al-Saady, MB, ChB, PhD,
Department of Medicine, Division of Physiological Medicine, St George’s Hospital Medical School, Cranmer Terrace, London
SW17 ORE. *Intensive Care Unit, and ?Respiratory Function Laboratory, St George’s Hospital Medical School, Blackshaw
Road, London SW17 OQT, fDepartment of Medicine, Faculty of Medicine, University of Ruhuwa, Galle, Sri Lanka.

0003-2409/95/070601 + 06 $08.00/0 @ 1995 The Association of Anaesthetists of G t Britain and Ireland 601
602 A . J . Petvos et al.

volume relationship at varying mean chamber pressures and
spans in healthy volunteers and construction nomograms,
which we believe should simplify the use of the Hayek
Oscillator.
Methods
Local ethics committee approval was obtained to study
10 healthy male subjects aged 2 5 4 7 years. Before the
study the cuirass was placed over the subjects' chest wall
and oscillated for brief periods at different frequencies
to allow them to become familiar with the sensation of
chest oscillation and also to ensure that no air leaks were
present between the chest and the cuirass. All studies were
performed in the upright sitting position. The subjects
breathed through a pneumotachograph (frequency response
3 Hz) wearing nose-clips attached to a Bicore Pulmonary
Monitor (BICORE Monitoring System, USA). This
device gives a breath-by-breath or time-averaged integrated
output for VT and VE and also allows the inspiratory and
expiratory waveform to be monitored in real-time on an
oscilloscope.
The study consisted of three protocols: (1) oscillation at
0.25,0.5, 1, 1.5,2, 3 Hz at different mean chamber pressures
(0, - 5 , -10cmH20) with the span held constant at
30 cmH20;(2) oscillation at 0.5, 1,2, 3 Hz at different spans
(20, 30 and 40 cmH20) with mean chamber pressure held
constant at - 5 cmH,O; (3) on five subjects, functional
residual capacity was measured using the helium dilution
technique [9] before and after fitting the cuirass and at a
continuous negative chamber pressure of - 10 and
- 20 cmH1O. Functional residual capacity was measured,
with subjects seated, using a closed circuit helium
equilibration method, with P. K. Morgan Transfer Test
model C apparatus (P. K. Morgan, Gillingham, Kent). Vital
capacity was recorded immediately after measuring
functional residual capacity. Residual volume was derived by
subtracting expiratory reserve volume from functional
residual capacity. Total lung capacity was obtained from the
sum of vital capacity and residual volume. The measured
values were corrected to BTPS. Duplicate measurements of
functional residual capacity were made and the average value
is reported. If the two readings varied by more than lo%,
they were discarded.
Three readings for VT and VE were taken at each stage
when it was judged by examining the flow waveform on
the monitor that breathing was in harmony with the
oscillation and no spontaneous breaths were added by the
subject. The results are presented as group means, standard
deviations (SD) and standard errors for functional residual
capacity. Statistical analysis of data was performed using the
paired t-test and analysis of variance at a significance of
p < 0.05.
there was no change in the VT. The lower panel of Figure 2
shows that, by increasing the span from 20 to 40 cmH20,a
significant increase in VT could be achieved, particularly at
0.5 and 1 Hz (p < 0.005, analysis of variance). With a span
of 20 cmHIO at 0.5 Hz, V r was 0.58 1 (SD 0.14 1) which
increased to 0.84 I (SD 0.13 1) when the span was increased
to 40 cniH20.
Similarly, minute volume increased from 10 to 30 l.min-'
as the frequency increased from 0.25-1.5 Hz (Fig. 3).
Minute volume was calculated by the BICORE as the
product of tidal volume and frequency, hence standard
deviations are not shown in Figure 3. At 2 Hz, at all mean
chamber pressures VE decreased, but above 2 Hz it increased
in a linear fashion. Changes in mean chamber pressure did
not affect VE.
As VT decreases below the deadspace volume of 150 ml,
effective alveolar ventilation is reduced. By subtracting an
arbitrary deadspace of 150 ml from all volumes measured the
effective alveolar ventilation can be appreciated (Fig. 4).
These figures illustrate that VE above 2 Hz becomes a
mathematical function of frequency and when expressed in
this manner VE above 2 Hz falls below zero. A nomogram
with differing mean chamber pressure and a fixed span of
30 cmH20is shown in the upper panel of Figure 5. A second
nomogram for the frequency/ VT relationship with a fixed
mean chamber pressure of - 5 c m H 2 0 and an increasing
span is shown in the lower panel of Figure 5 . As different
spans did not significantly alter VT at - 5 cmHzO, all three
values for VT were averaged and are presented within an area
which encompasses 1 SD.
Vital capacity and functional residual capacity at rest
without the cuirass or other chest wall restriction are shown

Results
The upper panel of Figure 2 shows the mean VT(SD)
achieved at 0.25, 0.5, I , 1.5, 2 and 3 Hz at different mean
chamber pressures. At higher frequencies, the subjects made
no independent effort to breathe as confirmed by observation
of the VT waveform. The highest tidal volume was achieved Fig. 1. The Hayek Oscillator. The external perspex cuirass attached
via wide bore tubing to the pressure generator fits over the chest and
at 0.5 Hz. At 2 Hz and a span of 20 c m H 2 0 VT approached abdomen. The second fine bore tube detects extrathoracic chamber
0.15 1, equivalent to deadspace volume (VD). Despite a pressure. (Reproduced with permission from Medicom UK, for
decreasing mean chamber pressure from 0 to - 10 cmH20 Flexco Medical Instruments AG.)

Anaesthesia, Volume 50, July 1995

 The Hayek Oscillator 603
1.o

t I
0.8

0.6

0.4

0.2

I , I , I 1 I
0 1 2 3 4
2
v
Frequency (Hz)

33
2
3

I I I I , I I I
0 1 2 3 4
Frequency (Hz)
Fig. 2. Each data point represents a group mean (SD). The upper panel shows the frequency-tidal volume relationship at decreasing mean
chamber pressure of O(n), - 5(.), - lo@) c m H 2 0 with a fixed span of 30 cmH20. The lower panel shows a fixed chamber pressure of
- 5 cmHzOwith spans of 20(0), 30(.) and 40(A)cmH20.Mean chamber pressure does not affect VT at any frequency, whereas decreasing
the span decreases VT over the same frequency range.

in Table 1. When the cuirass was applied, there was a Discussion

non-significant decrease in functional residual capacity.
As the extrathoracic pressure or mean chamber pressure
decreased from -10 to -20cmHz0, functional residual
capacity increased significantly from 2.25 (SD 0.1) 1 at rest
without the cuirass to 2.47 (SD 0.12) 1 at a mean chamber
pressure of - 20 cmH20.
We have characterised the changes in tidal volume
and minute volume produced by the Hayek Oscillator at
varying frequencies, mean chamber pressures and pressure
spans. The Hayek Oscillator can also be used in addition
to conventional ventilation by providing a continuous

Anaesthesia, Volume 50, July 1995

604 A . J . Petros el al.

30 ~

20 -

10 -

d 1 I 1 -

1 2 3 4

, 1 I I 1 I 1
101
0 1 2 3 4
Frequency (Hz)
Fig. 3. Upper panel shows the frequency-minute volume relationship at decreasing mean chamber pressures ofO(n), - 5(.), - 1O(A)cmH2O
with a fixed span of 30 cmH20.The lower panel shows a fixed chamber pressure of 5 cmH20 with spans of 2 0 ( 0 ) , 30(.) and 40(A)cmH20.
~

Changes in span increase minute volumes, changing mean chamber pressures do not.

negative extrathoracic pressure. In an attempt to increase
functional residual capacity we also determined how
functional residual capacity and vital capacity altered with
increasing negative extrathoracic pressures.
Above 2 Hz the tidal volumes generated were less than
anatomical deadspace [ 101. If effective minute ventilation
is calculated by subtracting anatomical deadspace from each
tidal volume and multiplying by the respiratory frequency,
again above 120 breath.min-’ there is no effective alveolar
ventilation with any chamber pressure of a span of
30 cmH20.The Hayek Oscillator behaves as two ventilators
in one offering two principles of ventilation. Below
2 Hz the Hayek Oscillator acts as a conventional ventilator
giving maximal ventilation at 0.25-0.5 Hz and a span of
30-40 c m H 2 0 . However, as the frequency approaches
2 Hz the Hayek Oscillator behaves more like a high
frequency oscillator, presumably achieving oxygenation by
diffusion [ I 1-1 31.
The Hayek Oscillator also significantly increases func-
tional residual capacity at increasing negative extrathoracic
pressures. When used in synchrony with conventional
positive pressure ventilation it may be possible to reduce
the peak inspiratory pressure and positive end expiratory
pressure (PEEP), both of which have been implicated

Ancrrsthszu, Volume 50, July 1995

 The Huyek Oscillator 605

i-
I
-10
0
‘ 1
I 1
2 3
I ’ 4

v
4
E

20

10 L

-10 I I , I , I I I
0 1 2 3 4
Frequency (Hz)
Fig. 4. Arbitary estimated (VE = F x (VT - 0.15 I)) effective alveolar ventilation-frequency relationship with a fixed span or mean chamber
pressure. Above 2 Hz there is no apparent alveolar ventilation at a span of 30 cmHIO.

in causing lung parenchymal damage. A lower mortality need to be established. From this study the Hayek Oscillator
rate occurs in association with low volume pressure-limited seems to have the attractive properties of being a
ventilation [14], but the advantages of the Hayek Oscillator conventional ventilator at low frequencies ( < 2 Hz) and

Table 1. Changes in functional residual capacity, vital capacity and total lung capacity at increasing continuous negative
mean chamber pressures. Values are expressed as mean (SEM).

N o cuirass Cuirass alone - 10 cmHIO -20 cmH20

Functional residual capacity; I 2.25 (0.10) 2.19(0.11) 2.61* (0.14) 2.47* (0.12)
Vital capacity; 1 4.03 (0.19) 3.82 (0.14) 4.12 (0.16) 4.09 (0.15)
Total lung capacity; 1 5.12 (0.24) 4.9 (0.21) 5.2 (0.27) 5.24 (0.23)

*p < 0.05 from ‘no cuirass’.

Atiuesrhesirr, Volume 50, July 1995

606 A . J . Petros et al.
Span of 30 cmHzO References
[ I ] AL-SAADY NM, FERNANDO SSD, SINGERM, BENNETTED.
External high-frequency oscillation can replace intermittent
1.0 pressure ventilation in patients with acute respiratory
failure. Intensive Care Medicine 1992; 18 (Suppl. 2): S54, Abstr.
63.
[2] AL-SAADYNM, FERNANDO SSD, SIDHUVS, CUMMIN ARC,
HAYEK Z, BENNETT ED. High frequency chest wall oscillation in
normal subjects and in mechanically ventilated patients.
Abstracts: I 1 th Internationul Sjvnposium on Intensive Care and
Emergency medicine, Brussels: 1991; 65.
[3] PENNY DJ, HAYEK Z, RIGBYML, REDINGTON AN. Ventilation
with external high frequency oscillation around a negative
baseline increases pulmonary blood flow after the Fontan
operation. Cardiology in the Young 1992; 2: 277-80.

-
!=!
I
1
I I
2
I
3
[4] DILKES
HOLLAMBY
MG, MCNEILL JM, HILLAC, MONKS PS, MCKELVIE
RG. The Hayek oscillator: a new method of
ventilation in microlaryngeal surgery. Annals of Otology,
P,

0
-$
>
Frequency (Hz)
Rhilology and Luryngology 1993; 102: 455-8.
[5] DILKESMG, MCNEILL
P, HOLLAMBY
JM, HILLAC, MONKSPS, MCKELVIE
RG. The Hayek oscillator: a new method of
3
Mean chamber pressure -5 cmHzO ventilation in microlaryngeal surgery. Annuls of Otology,
5 1.0
\
Rhinology and Luryngology 1993; 102 455-8.
[6] CALVERLEY PMA, CHANCHK, VARTIAN V, ZIDULKAA.

o.81
\
High-frequency chest wall oscillation. Chest 1986; 8 9
21 8-23.
[7] HAYEK Z, RYANCA, EYALF, JONESR, ARMENGOL J, KINGG,
FINERNN. Comparison of high-frequency chest wall
compression with conventional mechanical ventilation in cats.
0.6 Criticul Cure Medicine 1987; 15: 6 7 6 8 1
[8] HAYEKZ, PELIOWSKY A, RYANCA, JONESR, FINERNN.
External high frequency oscillation in cats. Experience in the
0.4 1 normal lung and after saline lung lavage. American Review of
Respiratorv Diseuse 1986; 1 3 3 6 3 W .
[9] MENEELY GR, KALTREIDER NL. The volume of the lung
determined by helium dilution. Description of the method and
comparison with other procedures. Journal of Clinical
Investigation 1949; 28: 129.~39.
[lo] NUNNJF. Nunn's Applied Respiratory Physiology, 4th edn.
I I 1 1 I 1 I , Oxford: Butterworth, Heinemann 1993.
0 1 2 3 4 [l I ] SLUTSKY AS, DRAZEN JM, INGRAM RH, KAMMRD, SHAPIRO
Frequency (Hz) AH, FREDBERG JJ, LORING SH, LEHRJ. Effective pulmonary
ventilation with small-volume oscillations at high frequency.
Fig. 5. Tidal volume-frequency relationship nomogram constructed Science 1980; 2 0 9 609-10.
from mean VT at (a) increasing negative chamber pressure and fixed [12] LEHRJ, BARKYOUMB J, DRAZEN JM. Gas transport during high
span of 30 cmHzO and (b) differing span and fixed mean chamber frequency ventilation. Federation Proceedings 1981; 4 0 384
pressure of - 5 cmHzO. Dashed lines indicate (SD). (Abstr. 863).
[ 131 CHANG H K . Mechanisms of gas transport during ventilation by
high-frequency oscillation. Journal of Applied Physiology
a high frequency oscillator a t high frequencies (> 2 Hz). Respirutory Environmental E.xercise Phqlsiology 1984; 56:
It m a y also prove valuable in reducing pulmonary 553-63.
[14] HICKLINC; KG, HENDERSON SJ, JACKSON R. Low mortality
t r a u m a produced by conventional positive pressure associated with low volume pressure limited ventilation with
ventilation by reducing the peak pressures needed t o inflate permissive hypercapnia in severe adult respiratory distress
stiff lungs. syndrome. Intunsiw Cure Medicine 1990: 16: 372-7.

Anaesthesia, Volume 50, July 1995