Applied Ergonomics 34 (2003) 185–194

Positive pressure breathing during rest and exercise

E.A. den Hartoga,*, R. Heusb
a
Thermal Physiology Group, Department of Work Environment, TNO Human Factors, PO Box 23, 3769 ZG Soesterberg, The Netherlands
b
Product Ergonomics and Performance, TNO Industrial Technology, Eindhoven, The Netherlands
Received 30 August 2001; accepted 28 August 2002

Abstract

The requirements to maintain a positive pressure with respiratory protection during heavy exercise and the effects on ventilation
and feelings of discomfort were investigated. Eight male subjects participated, using the respirator system during rest and exercise at
about 80% of their individual maximum power. A blower was used at maximum and medium capacity and at two pressure levels (3
and 15 mbar). Additionally, the mouth pressure was used as a feedback for the blower.
The blower decreased the fraction of the breathing cycle with negative pressures from 50% (SD 4%) to 15% (SD 10%) during
exercise. Negative pressures occurred at all settings of the blower during exercise. Thus, the currently available commercial blower
systems do not supply a sufficient airflow to maintain a positive pressure during heavy exercise.
Positive pressure breathing did not affect the ventilation and the circulation. But the oxygen consumption was higher with the
blower and respirator than without.
r 2003 Elsevier Science Ltd. All rights reserved.

Keywords: Positive pressure breathing; Exercise; Performance

1. Introduction The positive pressure system should blow the air

through the filter under the respirator, without the
In the current military breathing protection system, inspiration being the driving force. A requirement to the
the wearer has to inhale his air through purification military use of the system is that it should be wearable
filters. As the number of biological and chemical agents by the soldier. Technical difficulties of such a system are
is increasing, the inspiratory resistance of the filters the power requirements of the blower, the weight of the
tends to increase. This trend of increasing inspiratory system and the capacity of the filter (as a high amount of
resistances may be counteracted by blowing clean air air has to be purified).
under the respirator. An additional advantage is that the An optimal positive pressure breathing apparatus
pressure under the respirator becomes positive. This should always maintain a minimal positive pressure
improves the NBC-protection of the soldier. Using a during the entire breathing cycle. The commercially
positive pressure breathing system, the protection factor available systems are known to maintain the positive
may be raised 10 times or more (Brinkhuis, 1982). The pressure during rest and low levels of exercise. But,
protection factor is defined as the ratio of the positive pressure may not be maintained during high
concentration of dangerous particles in the air outside levels of exercise. The high inspiratory peak flows
and inside the mask. In case of leakage with a positive (5–10 l=s), that are involved in breathing during heavy
pressure breathing apparatus, clean, filtered air will leak exercise, may cause periods of negative pressure below
out, in stead of the poisonous air leaking in when the mask, which will decrease the protection factor
normal full-face masks are worn. The odds of leakage of significantly. However, generation of high flows will
poisonous particles into the respirator, are therefore, result in a short filter life. A possible solution may be to
very small during positive pressure breathing. control the blower, so that it increases the flow only
when necessary. To design such respirator/blower
*Corresponding author. Tel.: +31-346-356-288; fax: +31-346-353- systems it is important to know what the design
977. parameters of such a system should be, in terms of air
E-mail address: [email protected] (E.A. den Hartog). flow, breathing frequency and pressure variations below

0003-6870/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 0 3 - 6 8 7 0 ( 0 2 ) 0 0 0 8 4 - 4
186 E.A. den Hartog, R. Heus / Applied Ergonomics 34 (2003) 185–194

the mask. In a number of studies it has been shown that 15 l=s at 0 mbar: At pressures of 30 mbar and higher,
there are no significant changes in breathing and blood the blower generates no airflow.
circulatory parameters, during positive pressure breath-
ing at rest, compared to normal breathing. However, 2.2. Feedback
during exercise some authors have found effects of
positive pressure breathing during exercise (Louhevaraa The blower control is used in two different conditions.
et al., 1984, 1985; Vroman et al., 1985). In this study we Firstly, a constant input voltage is used, resulting in an
aimed to maintain positive pressure during heavy air flow that is dependent on the pressure in the tube
exercise with a portable blower. Furthermore, we system (Constant condition). Therefore, with this
studied the physiological effects of the required air constant control voltage, the air flow varies with the
flows and (positive) pressures. breathing pattern. Secondly, a circuit with feedback is
used, in which the pressure at the mouth is compared to
a reference signal. The difference between these two
2. Methods signals is amplified and fed to the control of the blower.
With this circuit we aimed to keep the pressure under the
The experiments have been carried out using the set mask, in front of the mouth, nearly constant during the
up as shown in Fig. 1. The externally controlled blower breathing (Feedback condition).
generates an air flow in the U-shaped tube system. The
pressure in the tube system is controlled by an 2.3. Measurements
adjustable outlet valve. The air flow, generated by the
blower, is measured by a Fleisch flow meter (nr. 1, During the experiment the following variables were
Fleisch, Gould). The subject breathes out of and into the measured continuously (sample rate 5 Hz; Viewdac data
airflow in the U-shaped tube through the second flow acquisition system): Pressure under the respirator
meter. This second flow meter is part of the breathing (Pmask), Air flow through flowmeters 1 (Q’blower)
gas analysis equipment (Oxycon S; Mijnhardt, The and 2 (Q’breath), Control voltage to the blower
Netherlands). The airway pressure is measured under (Vcontrol) and the electric current that was used by
the mask, just in front of the mouth. the blower controller (Icontrol). Furthermore, a number
of physiological parameters were measured: Blood
2.1. Blower pressure (Part, FinaPress TNO-TPD, The Netherlands),
Heart rate (HR, sporttester Polar Vantage NV, Polar,
The blower (Minijammer, one stage brushless blower, The Netherlands). Oxygen consumption ðVO0 2 Þ; carbon
0
Ametek Rotron, Kent, Ohio, USA) requires 12 V DC as dioxide production ðVCO 2
Þ; ventilation rate (VE),
a stable power supply. The power generated by the breathing frequency (BF), expiration time (Te), Carbon
blower is controlled by a voltage ranging from 0 to 5 V: dioxide, Oxygen tension at the end of the breathing cycle
At maximal control voltage, the blower performs (PetCO2 and PetO2 ) and the respiratory quotient (RQ),
maximally. The airflow in the system is dependent on were all measured and stored by the Oxycon S:
the pressure just behind the blower; it is maximally
2.4. Comfort

12 V DC Power The subjects were also asked to give their comfort

Flowmeter 1 supply

Flowmeter 2
ratings on breathing and apnoea during the rest phase
Blower controller and the exercise phase. The ratings were scored on a
Blower
scale from 4 (breathing is easier than control) to +6
1: Constant
Mask control voltage (very tight in the chest, close to suffocating). The
Outlet valve 2: Feedback Variable
power supply
subjects were asked to give their ratings relative to the
Pressure signal
control condition. Therefore, negative values could be
Oxycon Amplifier
Pressure expected.
transducer

2.5. Conditions
Fig. 1. Scheme of the experimental set-up. The blower generates
airflow in the U-shaped tube system. The pressure in the tube system is By using the blower with and without feedback and
also regulated by an adjustable outlet valve. The subject breathes out by using different settings of the outlet valve (Fig. 1),
of and into the airflow in the U-formed tube. The airway pressure is
various conditions could be studied. At first the
measured below the mask, just in front of the mouth. The blower was
controlled in two ways, firstly, with a constant voltage and secondly, reference condition (N), in which the blower was part
the airway pressure was inversely fed back to the blower, thus of the circuit, but it was switched off. As there was a
generating a higher flow at lower airway pressures. small, but open connection to the outside air, this was
 E.A. den Hartog, R. Heus / Applied Ergonomics 34 (2003) 185–194 187

Table 1
Balanced scheme for the use of different conditions in all subjects

PP 1A 1B 2A 2B 3A 3B

1 FB N CLC CHO CLO CH

2 N FB CHO CHC CLC CLO
3 FB CLC CLO N CLC CHO
4 N CHO CHC FB CLO CLC
5 CLC CLO FB CHC CHO N
6 CHO CHC N CLO CLC FB
7 CLO CLC CHC N CHO FB
8 CHC CLO CHO CLC N FB

comparable to a condition with an industrial filter Exercise level

through which air has to be inhaled (resistance
0:18 kPa s=l measured at 5 l=s). In the next set of N Condition A Condition B
conditions the blower was controlled by a constant
voltage (C-conditions). In these conditions two degrees
of freedom exist: the control voltage of the blower and
the opening level of the outlet valve. We selected two
voltage levels: (L ¼ 2:5 V; H ¼ 4:0 V) and two levels of 0 3 8 10 15 20 25
the valve (Open, (almost) Closed). Time [min]
Phase: 1 2 3 4 5 6
In an extra condition (FB-conditions) the pressure at
the mouth was fed back to the controller of the blower. Fig. 2. Scheme of the protocol: The x-axis shows the time (min) and
the y-axis shows the level of exercise following the protocol. The
Thus, in total six conditions have been used in the subjects start with a rest period in the control condition (phase 1).
protocol of the experiments: N, FB, CLO, CLC, CHO After 3 min the first condition is set (A, phase 2). The subject starts to
and CHC (Table 1). cycle and the power is increased in 2 min to the desired level (phase 3):
85% of the individual maximum power. This level of exercise is
2.6. Exercise maintained for 10 min; half way the system is switched to a second
condition (B). The exercise period of 10 min (phases 4 and 5) is
followed by a 5 min recovery period in condition B (phase 6).
The aim of the experiments was to study the
requirements to maintain positive pressure under the
respirator at rest and at high levels of exercise. Prior to switched to a second condition (B). The exercise period
the experiments the maximum power of each subject was of 10 min (phases 4 and 5) was followed by a 5 min
determined in a graded exercise test on a bicycle recovery period in condition B (phase 6). A properly
ergometer (Lode, Excalibur, Groningen The Nether- balanced scheme was necessary to avoid order effects.
lands). To apply high levels of exercise we supposed that We used a scheme in which all conditions were used four
the subjects would be able to perform exercise for times in the first part (A) and four times in the second
10 min at 85% of their maximum power. All experi- part (B).
ments have been carried out on an equal bicycle
ergometer, in the climatic chamber at 201C; and 50% 2.8. Subjects
relative humidity.
Eight male subjects volunteered to participate in the
2.7. Protocol study. Their ages ranged from 25 to 51 years old. Prior
to the experiments an individual graded maximum
To reduce the number of experimental sessions we exercise test was performed on a bicycle ergometer to
decided to apply two different conditions during one determine each subjects’ maximum power and max-
session. In Fig. 2 the experimental protocol is schema- imum oxygen consumption ðVO0 2max Þ: VO0 2max ranged from
tically presented. Prior to the actual experiment, the 1666 to 4546 ml=min (Table 2). This protocol was
control condition (N) was always performed, in which approved by the local ethical committee of TNO Human
the blower was in the set up but not functioning. Then, Factors.
the blower was turned on in the required condition. The
subject started to cycle after 5 min and the power was 2.9. Statistics
increased to the desired level in 2 min (phase 3): 85% of
the individual maximum power. This level of exercise The measurement results have been analysed using a
was maintained for 10 min: Half way the system was repeated measures ANOVA (Statistica for Windows,
188 E.A. den Hartog, R. Heus / Applied Ergonomics 34 (2003) 185–194

Table 2
Anthropometric data and data from the maximum exercise test of all eight subjects. The last column shows the percentage of the maximum power at
which the experiments were performed. The rationale behind this is explained in Section 3

Subject Stature Weight Age Pmax VO0 2max Percentage

(m) (kg) (years) (W) (l/min) performed (%)

1 1.86 72 37 360 3115 85

2 1.88 86 31 387 2962 77.5
3 1.87 85 30 368 4546 80
4 1.90 82 29 308 2970 75
5 1.81 73 38 365 3390 85
6 1.92 82 51 328 1666 85
7 1.88 80 40 280 2113 75
8 1.68 64 28 256 2266 75

Statsoft Inc. 2000, USA), with the factors condition 3- phase 4), changing the condition during exercise
(1–8), session (1–4) and phase (2,4,5,6) as independent (phase 4- phase 5) and stopping exercise (phase 5-
variables. The phases 1 and 3 were not used in the phase 6). In Fig. 3 the blower was shut off shortly after
analysis; phase 1 was to get used to the system and phase 15 min to change from a condition with feedback (FB)
3 was a gradual (non-standardised) build up of the to a condition with a constant control voltage (CHC).
required exercise level. In the relevant phases 2, 4, 5 and At that moment, the pressure that would be present
6 all dependent variables have been analysed by during exercise without the blower (N), is seen briefly. In
averaging the data every minute. Fig. 3 the effect of the feedback on the pressure is also
Furthermore, at the low voltage setting there was demonstrated.
almost no pressure difference in the system between the During the rest periods, the pressure always remained
valve being open and the reference condition (C- positive when the blower was working. In Figs. 4A and
condition-Low voltage-Open valve; CLO: 0 to 1 mbar; B respectively, the average of the minimum and
N: 0 mbar). Therefore, CLO was eliminated from the maximum pressures during exercise, measured below
statistical analysis of the experiments and is not the respirator, are presented. During the high exercise
presented in the experimental results. At the high levels that we used in this study, we found negative
control voltage level both valve settings were used: values for the minimum pressure in all conditions in
(CHO: 3 mbar; CHC: 15 mbar). almost all subjects, even in condition CHC, in which the
When significant differences were identified, a Post- highest maximum pressures were found. To identify the
hoc Tukey HSD test was used to identify the actual extent to which the conditions exhibited negative
condition which showed the differences; differences with pressures, a probable measure of loss of protection,
po0:05 were denoted as significant. the area of the pressure curves below zero was analysed
(Fig. 5).
As an extra factor, the power use of the blower during
3. Results exercise has been analysed. Data on average use of
power and ‘peak’ power are shown in Fig. 6. Although
Because of the limited power of the blower it was not the peak power is higher in the feedback (FB) condition
possible to maintain a constant pressure exactly. During than in the constant (C) conditions, the average power
the first sessions, we found that not all subjects were able consumption is smaller in the FB-condition, about 40%.
to maintain an exercise level of 85% of their maximum No significant differences in the physiological para-
power for 10 min: For each subject independently, we meters were found between the blower conditions. In
identified the exercise level that they could sustain for Figs. 7 and 8 the results of some important physiological
10 min; and all consecutive sessions were performed at parameters are shown, the minute ventilation (Fig. 7)
that same level for that subject (Table 2). and the heart rate (Fig. 8) respectively. Only for the
Only three subjects were able to maintain a level of heart rate a significant difference was found between the
85% of their maximum power for 10 min; one subject first and the second part of exercise ðpo0:01Þ: This
performed at 80%, one at 78% and three at 75%. difference was probably due to the cumulative effect of
In Fig. 3 an example of a pressure recording under the the exercise at this high intensity. We expect that the
respirator during one entire experimental session is balanced design avoided an effect on the differences
presented. The six consecutive phases of the session can between the conditions.
be identified clearly in the figure: turning on the blower In Fig. 9 the measured oxygen consumption is plotted
(phase 1- phase 2), starting exercise (phase 2- phase against the oxygen consumption at the same intensity
 E.A. den Hartog, R. Heus / Applied Ergonomics 34 (2003) 185–194 189

Fig. 3. An example of the pressure below the respirator, plotted versus time, during an experimental session. The various phases of the session can be
distinguished clearly (cf. Fig. 2). At the transition from the first condition (TM) to the second condition (C3H) the blower was turned off shortly. At
that point the pressure level, obtained without the blower is shown.

45 6

40
2
Minimal pressure under respirator [mbar]

35
Peak pressure under respirator [mbar]

30 -2

25
-6
20

15 -10

10
-14
Non-Outlier Max
5
Non-Outlier Min

0 -18 75%
FB N CLC CHO CHC FB N CLC CHO CHC 25%
Median

(a) CONDITION (b) CONDITION

Fig. 4. The range of maximal (a) and minimal (b) pressures below the respirator. On the x-axis, the eight different conditions are plotted, the y-axis
represents the pressure. The plotted values are medians and ranges over the eight subjects, measured during the exercise phase.

level during the maximum exercise test, i.e. the oxygen 4. Discussion
consumption at 85% (or less if relevant) of the
maximum workload. From Fig. 9 it can be concluded As far as we know in all previous studies the lowest
that the oxygen consumption during exercise with a inspiratory pressure during exercise was always nega-
respirator is considerably higher than without a tive. The values of the positive pressure levels usually
respirator at the same exercise level. refer to the end-expiratory pressure. In this study we
In the comfort ratings we found that the condition FB have explored the requirements that would have to be
scored significantly better than the N condition, without set on a (wearable) blower if it would have to ensure a
blower. Unfortunately, the variance in the comfort data continuous positive pressure also during heavy exercise.
was quite large so that no other differences were found Furthermore, we studied whether the positive pressure
(Fig. 10). levels that were used had any deteriorating effect on the
190 E.A. den Hartog, R. Heus / Applied Ergonomics 34 (2003) 185–194

16 % 52 % 34 % 32 % 5 %
0

Negative area pressure curve [mbar s]

-2

-4

-6
Non-Outlier Max
Non-Outlier Min
75%
-8 25%
Median
Outliers
-10
FB N CLC CHO CHC Extremes

CONDITION

Fig. 5. The area of the pressure curve with negative pressure. The x-axis denotes the eight conditions, on the y-axis the product of the negative
pressure and the duration of the negative pressure is plotted. These values indicate the loss of protection due to the absence of the positive pressure.
The displayed values are the medians and the ranges of the eight subjects during exercise. The numbers in the graph denote the percentage of the
breathing cycle during which the pressure was negative. These latter values are 1 min averages over the eight subjects.

5.5
4.0

3.5
4.5

3.0

3.5
2.5

2.0
Imax [A]

Imean [A]

2.5
1.5

1.5 1.0

0.5
0.5 Non-Outlier Max
0.0 Non-Outlier Min
75%
25%
-0.5 -0.5
FB N CLC CHO CHC FB N CLC CHO CHC Mediaan

(a) CONDITION (b) CONDITION

Fig. 6. The mean (a) and maximal (b) electric currents of the blower. The values are plotted as the consumption of current in Ampere. As the power
supply to the blower is constant ð12 VÞ the electric current is proportional to the power consumption. The average electric current is computed over
the entire period, the maximal current is derived from the peak currents that were found during the inspiratory phases. The displayed values are the
medians and the ranges of the eight subjects during exercise.

performance of the subjects. Unfortunately the power of whole system (pressure transducer, feedback loop,
the blower was insufficient to provide enough air flow blower control): the increase of flow during inspiration
during heavy exercise to maintain the positive pressure. provided by the blower, lags the inspiration of the
The maximal flow that was generated by this blower subjects during exercise. This effect can be reduced by
during exercise was about 4:5 l=s: The maximum adjusting the dimensions of the tube system, so that the
performance of the blower in this experimental set-up pressure transducer itself is located near the mouth. In
was 5:5 l=s during tests prior to the experiments. In spite our experiments this distance was too large. However,
of this available extra capacity of the blower, the the maximum flow that was actually obtained in our
positive pressure was not maintained during heavy experiments is considerably higher than the flow in
exercise, not even in the condition with the highest commercially available systems. Such systems generate
pressures (CHC). The reason was the inertia of the flows of about 2–3 l=s: From our results it is clear
 E.A. den Hartog, R. Heus / Applied Ergonomics 34 (2003) 185–194 191

160

140

120

100
Ventilation [l/min]

80

60

40

20 Non-Outlier Max
Non-Outlier Min
0 75%
FB N CLC CHO CHC 25%
Median

CONDITION

Fig. 7. The ventilation (in l/min) of the eight different conditions during the exercise phase. No significant differences were found between the
conditions.

170

165

160
Heart rate [1/min]

155

150

145

1st Exercise period

140
FB N CLC CHO CHC 2nd Exercise period

CONDITION

Fig. 8. The average heart rate of the subjects in all conditions. The solid line shows the mean values during the first phase of exercise, the dashed line
shows the data during the second exercise phase. The significant average increase in heart rate from the 1st to the 2nd phase of exercise may be
subscribed to the effects of fatigue as it was only found in the heart rate data.

that such systems will not provide a continuous positive gical parameters has been studied. Most studies report
pressure during moderate and heavy exercise. If a that during rest, positive pressures of 5–10 mbar cause
high protection factor is necessary, a more power- no significant changes in various physiological para-
ful blower is required, or the wearer is allowed meters of ventilation and circulation (Arborelius et al.,
to perform at low levels of exercise only. However, 1983; Dahlback and Balldin, 1984; Verstappen et al.,
at the moment it is unknown how large the effect of 1986; Harber et al., 1989). Only at average mouth
short periods of negative pressure on the protection pressures of 15 mbar and higher, physiological effects
factor is. are reported on ventilation (Louhevaraa et al., 1984,
In the previous years a number of studies have been 1985) and circulation (Bjurstedt et al., 1979). Some
published on positive pressure breathing. In most papers studies have been performed during moderate or heavy
the effects of a certain type of respirator on physiolo- exercise (e.g. Arborelius et al., 1983; Dahlback, 1979;
192 E.A. den Hartog, R. Heus / Applied Ergonomics 34 (2003) 185–194

5000

4500

4000

Experimental V'O 2 [ml/min] 3500

3000

2500

2000

1500

1000

500

0
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Submaximal V'O 2 [ml/min]

Fig. 9. The oxygen consumption of the subjects during exercise is plotted versus the oxygen consumption at the same power level during the
maximum exercise test, performed prior to the experimental sessions. The solid line indicates the line of identity. The few points near the line can be
subscribed to case in which some leakage occurred during the positive pressure breathing, thus, underestimating the oxygen consumption. The
oxygen consumption between the blower conditions was not significantly different.

3
Shortness of breath-score

-1

Non-Outlier Max
-2
Non-Outlier Min
75%
-3
FB N CLC CHO CHC FB N CLC CHO CHC 25%
Rest Exercise Median

CONDITION

Fig. 10. The subjective scores for the feelings of shortness of breath for the eight conditions separate, for the rest period and the exercise period. The
score were allowed to range from 4 (breathing is much easier) to +6 (very tight in the chest).

Louhevaraa et al., 1984, 1985). Bjurstedt et.al. (1979) In the current experiments we have found no
have studied the effects of positive pressures between 15 significant effect of the increasing positive breathing
and 30 mbar: They found that such pressures signifi- pressure (N, CHO, CHC) on the ventilatory parameters.
cantly decreased ventilation and cardiac output, how- However, from the individual inspection of each curve
ever, these negative effects diminished during exercise. by eye, we did notice an increase in oxygen consumption
This is probably because during heavy exercise the each time the condition was changed to the ‘‘N’’
expiration is always active, also under normal circum- condition (no blower) and a decrease if it was changed
stances. Possibly, the extra back pressure to the from the ‘‘N’’ condition to another. This difference was
expiration during exercise is less relevant than during present in each subject, but did not appear significantly
rest, when the normal expiration is passive. in any statistic test. The reason was that the oxygen
 E.A. den Hartog, R. Heus / Applied Ergonomics 34 (2003) 185–194 193

consumption within a subject between sessions differed we have insufficient data to confirm Bentley’s results. In
more than the differences within a session from the ‘‘N’’ our data we did find a reduced discomfort in the
condition to the other conditions. This effect is likely to conditions TL and TM compared to the condition N,
be present in all physiological parameters and illustrates without blower. This would suggest a decreased pressure
the limited power of the study. difference between in- and expiration due to the
At equal levels of power on the bicycle, the oxygen feedback of the pressure. The variance in the data is
consumption was higher in experiments with the so large that it would require more subjects to determine
respirator than in the maximum exercise test that had significant differences between the other conditions.
been taken from the same people, without the respirator, Moreover, some subjects stated that the comfort scale
prior to the experiments (Fig. 9). The increased levels of was difficult to use. Therefore, in a future study an
oxygen consumption were not related to the pressure improved subjective scale is to be found that is easier to
level below the mask. Even in the neutral ‘‘N’’ condition use by the subjects.
the oxygen consumption was higher than during the The feedback of the pressure signal to the blower
maximum exercise test, while there was an inspiration controller leads to a decrease of the maximum pressures
resistance instead of an expiration resistance (i.e. the below the respirator (Fig. 3). The consequence was that
positive pressure). In a previous study from Heus et al. the blower requires a larger electrical peak current. This
(1999), it was also found that adding an inspiration is compensated during the rest of the breathing cycle,
resistance leads to considerably higher levels of oxygen leading to a lower average electrical current. The use of
consumption at each power level during a maximum the feedback leads to a lower energy consumption,
exercise test. The differences that have been found by which can be important if the blower has to run on
Heus are comparable to the values obtained in this batteries. Another advantage is that the average airflow
study. The inspiration resistance in ‘our’ N condition in the system is lower, which will increase the durability
ð0:18 kPa s=lÞ was close to the lowest resistance in Heus’ of the air filter. The disadvantage of the feedback loop is
study ð0:24 kPa s=lÞ: In our experiments five out of eight that it also requires a battery for the amplification.
subjects were unable to perform 10 min of exercise at However, the advantages of the feedback loop may
85% of their individual maximum power level. Under outweigh this disadvantage. Physiologically, the feed-
normal conditions working times up to 30 min are back has no large beneficial effects. The decreased peak
supposed to be possible (McArdle et al., 1996). This pressure may reduce discomfort, mainly during rest. But
agrees with the results of Raven et al. (1982), who also in the present study we found no significant effect of the
found that 50% of their subjects were not able to sustain feedback on ratings of discomfort. Probably such effects
exercise for 10 min at 80% of their maximum power. will only occur at average pressure levels of more than
They assumed that it was caused by the large pressure 15 mbar; or pressure differences of more than 20 mbar:
differences within the breathing cycle during positive In conclusion, the currently commercial available
pressure breathing (around 23 mbar). However, they blower systems that generate positive pressure during
stated that their results were inconclusive. In our breathing are not able to maintain a continuous positive
experiments we have not looked for a correlation pressure during moderate to heavy exercise. At the
between the pressure differences within the breathing pressure levels that have been used in this study, the
cycle and the ability to sustain exercise at 85% of their exercise performance of the subjects was decreased, as
maximum. To study such a relationship, the protocol only 3 out of 8 subjects were able to perform at 85% of
should have been changed and aimed at finding the their maximum for 10 min: The increased oxygen
power level that could be sustained for 10 min with consumption during positive pressure breathing seems
varying settings of the blower and valve of the system. comparable to the increased oxygen consumption
The average pressure difference in our experiments was during breathing with an inspiratory resistance.
17 mbar (SD 7 mbar). All studies seem to point out that
the maximum oxygen uptake is increased, but the
maximum power is decreased, during positive pressure Acknowledgements
breathing and during breathing with an inspiratory
resistance. However, we do not have enough data to The authors thank Jeroen van de Water, Lyda
identify at which levels of inspiratory resistance or Kistemaker and Walter van Dijk for their valuable
positive pressure the oxygen consumption starts to contributions to this project.
increase.
During positive pressure breathing the pressure in the
breathing system may lead to a sense of reduced load for References
inspiration (Wilson et al., 1989). But, Bentley et al.
(1973) pose that a clear sense of discomfort develops Arborelius, M., Dahlback, G.O., Data, P.G., 1983. Cardiac output
when pressure differences exceed 20 mbar: In our data and gas exchange during heavy exercise with a positive pressure
194 E.A. den Hartog, R. Heus / Applied Ergonomics 34 (2003) 185–194

respiratory protective apparatus. Scand. J. Work Environ. Health during exercise in well-trained men. Eur. J. Appl. Phys. 52,
9, 471–477. 340–345.
Bentley, R.A., Griffin, O.G., Love, R.G., Muir, D.C.F., Sweetland, Louhevaraa, V., Smolander, J., Tuomi, T., Korhonen, O., Jaakkola,
K.F., 1973. Acceptable levels of breathing resistance of respiratory J., 1985. Effects of an SCBA on breathing pattern, gas
apparatus. Arch. Environ. Health 27, 273–280. exchange and heart rate during exercise. J. Occup. Med. 27 (3),
Bjurstedt, H., Rosenhamer, G., Lindborg, B., Hesser, C.M., 1979. 213–216.
Respiratory and circulatory responses of sustained positive- McArdle, W.D., Katch, F.I., Katch, V.L., 1996. Exercise Physiology,
pressure breathing and exercise in man. Acta Physiol. Scand. 105, 4th Edition. Williams & Wilkins, Baltimore, USA.
204–214. Raven, P.B., Bradley, O., Rohm-Young, D., McClure, F.L., Skaggs,
Brinkhuis, B.H., 1982. Negative or positive pressure, an inventory of B., 1982. Physiological responses to ‘‘pressure-demand’’ respirator
the major advantages and disadvantages. De Veiligheid 12, 3–8 (in wear. Am. Ind. Hyg. Assoc. J. 43 (10), 773–781.
Dutch). Verstappen, F., Bloemen, L., van Putten, M., Reuvers, J., 1986. Self-
Dahlback, G.O., Balldin, U.I., 1984. Physiological effects of pressure contained respirators: effects of negative and positive pressure
demand masks during heavy exercise. Am. Ind. Hyg. Assoc. J. 45 demand types on physical exercise. Am. Ind. Hyg. Assoc. J. 47 (10),
(3), 177–181. 635–640.
Harber, P., Shimozaki, S., Barett, T., Losides, P., Fine, G., 1989. Vroman, N.B., Beckett, W.S., Permutt, S., Fortney, S., 1985.
Effects of respirator dead space, inspiratory resistance and Effect of positive-pressure breathing on cardiovascular and
expiratory resistance ventilatory loads. Am. J. Ind. Med. 16, thermoregulatory responses to exercise. J. Appl. Physiol. 58 (3),
189–198. 876–881.
Heus, R., Den Hartog, E.A., Kistemaker, J.A., Van Dijk, W.J., 1999. Wilson, J.R., Raven, P.B., Morgan, W.P., Zinkgraf, S.A., Garmon,
The influence of inspiration resistance on performance during R.G., Jackson, A.W., 1989. Effects of pressure-demand
graded exercise tests. Report TNO-TM 1999 A-58. respirator wear on physiological and perceptual variables during
Louhevaraa, V., Tuomi, T., Korhonen, O., Jaakkola, J., progressive exercise to maximal levels. Am. Ind. Hyg. Assoc. J. 50
1984. Cardiorespiratory effects of respiratory protective devices (2), 85–94.