CPAP Machine Performance and
Altitude*
Robert E. Fromm, Jr, MD, MPH, FCCP; Joseph Varon, MD;
Alex E. Lechin, MD; and Max Hirshkowitz, PhD
Study rationale and objective: Sleep-disordered breath¬
ing is commonly treated with nasally applied continu¬
ous positive airway pressure (CPAP). Typically, pres¬
sures are titrated to pneumatically splint the airway to
prevent its collapse in response to negative inspiratory
pressure. This investigation was prompted by several
patient complaints of sleep-related breathing diffi¬
culty associated with travel to high altitudes. CPAP
devices create pressure with fan-generated airflow;
therefore, CPAP performance should behave accord¬
ing to collective fan laws.
Measurements and results: In the
present study,
we
examined the effect of simulated altitude change on
four commercially available CPAP machines. Ma¬
chines were tested using anatomic airway mannequins
in an altitude chamber. We made three simulated as¬
cents to 12,000 feet with machines set at 5, 10, and 12
cm H2O sea level pressure equivalents. We measured
ments during ascent and descent. Mask pressures
varied systematically with changing altitude in three
machines. One machine, equipped with a pressure
regulation feature, maintained pressure within 1 mm
H2O at all pressure and altitude combinations.
Conclusions: Altitude significantly alters delivered
pressure according to predictions made by the fan
laws, unless a unit has pressure-compensating fea¬
tures. Clinicians should consider this factor when
CPAP is prescribed for patients who live or travel to
places located at significantly higher or lower eleva¬
tions than the titration site.
(CHEST 1995; 108:1577-80)
CPAP=continuous
positive airway pressure
pressure using water manometers at 2,000-foot incre¬
Key words: altitude; apnea; CPAP; sleep
afflicts more than 10% of
breathing
CJleep-disordered
the
American
It
are conducted to titrate
Overnight sleepandstudies
assess
CPAP
effectiveness. These
optimal pressure
are
performed in the
polysomnographicat procedures
ambient
pressure. Although sleep
sleep laboratory at
altitude,9 clinicians seldom
physiology changes
high
factor or
to.
^
middle-aged
population. usually
present, nasally applied continuous positive airway
pressure (CPAP) is the preferred treatment for sleepdisordered breathing.1,2 To be therapeutically benefi¬
cial, however, it requires patient acceptance and utili¬
zation. CPAP provides a constant, positive background
pressure in the pharyngeal airway.3 This constant
pressure presumably alleviates airway occlusion by
pneumatically
splinting the upper airway, increasing
the functional residual capacity, which in turn reflexively dilates the pharynx, or both.4"8
CPAP machines create pressure with fan-generated
airflow. Pressure adjustment involves either changing
fan speed or using a valve regulator to alter flow
velocity.
Typically, health-care providers set machines
to prescribed pressure using an external manometer
attached to the mask. Only recently have CPAP man¬
ufacturers incorporated pressure sensors and micro¬
processors to compensate for pressure changes.
results from partial or complete airway obstruction. At
*From the Sections of Cardiology, Pulmonary and Critical Care,
of Medicine Baylor College of Medicine, (Drs.
Department
Fromm and Lechin), Department of Anesthesiology and Critical
Care, The University of Texas M.D. Anderson Cancer Center (Dr.
Varon), and Sleep Disorders Center, Veterans Affairs Medical
Center, Baylor College of Medicine (Dr. Hirshkowitz), Houston.
Manuscript received
April 4, 1995; revision accepted July 17.
Dr. Hirshkowitz, VAMC Sleep Center 116A, 2002
Reprint
requests:
Holcombe Blvd, Houston, TX 77030
consider this
have any need However,
cities in North America situated at altitudes above
4,000 feet include the following: Flagstaff, Ariz; Albu¬
querque, NM; Reno, Nev; Klamath Falls, Ore;
Laramie, Wyo; and Mexico City. Furthermore, many
individuals travel to high-altitude locals for business or
Yosemite Village, Calif [7,569
pleasure (for example,
feet]). Finally, some sleep disorders centers in major
medical centers that service wide geographic areas
conduct laboratory CPAP titrations on
occasionally
patients who live at substantially different elevations.
For example, many patients from Mexico City (altitude
7,000 feet) come to the Texas Medical
approximately
Center in Houston for evaluation and treatment; this
includes patients with sleep-disordered breathing.
Symptom recurrence on follow-up can often be ex¬
plained byweight
gain, medication change, mask leaks,
allergies, alcohol
usage, or declining use of CPAP.
the
over
However,
years, we have encountered cases
of symptom recurrence that could not be explained by
such factors. One pattern we observed clinically was
that problems seemed to occur in some patients dur¬
ing travel to high elevation.
CHEST /108 / 6 / DECEMBER, 1995
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21725/ on 06/17/2017
1577
Table 1.Collective Fan Laws*
When These Conditions Are Fixed
The Laws Are:
The Equations Are:
1. Flow rate varies directly as the
speed ratio
2. Pressure varies as the square of
the speed ratio
3. Horsepower varies as the cube
of the speed ratio
4. Flow rate varies as the cube of
N2
And This Is a Variable
Impeller system density
Speed (N)
Speed density
Impeller diameter (D)
the fan diameter ratio
as the square of
the fan diameter ratio
6. Horsepower varies as the fifth
power of the fan diameter ratio
7. Horsepower and pressure vary
directly as the air density ratio
5. Pressure varies
Flow rate
speed
Density (p)
Weight flow
Density (p)
*-*(g)2
HP2=HPl(g)
P-P,(P1J
HP-H,'(S)
P2\
8. Air flow,
speed, and pressure
vary inversely as the density ratio
.(K)
9.
Horsepower varies inversely as
the square of the density ratio
*N=fan
Hft-HPxfg)
speed (revolutions/min); Q=flow rate (cubic ft/min); P=pressure (inches H2O); HP=horse power; D=fan diameter (inches); p=air density
(lbs/cubic feet)
changes in atmospheric pressure
produce alterations in gas density. These density
changes may affect the performance of fan-driven de¬
vices in accordance with the collective fan laws (Table
l).10 These laws describe the performance of a fandriven device according to gas flow rate, gas density,
and horsepower. Given constant power and variable
gas density, a fan-generated pressure is predicted by
the following equation:
As altitude varies,
P2=Pi x ds/di,
where Pi=pressure under condition 1, P2=pressure
under condition 2, di=density under condition 1, and
under condition 2.
d2=density
To our knowledge, no published studies are avail¬
able that have evaluated the effects of altitude on ei¬
ther CPAP performance or on the physiology of
sleep-disordered breathing. Therefore, as a first step,
we conducted this study to examine the effects of
simulated altitude change on commercially available
CPAP machine performance. This study served as an
empirical validation of performance predictions made
by the collective fan laws.
Methods
Four representative commercially available CPAP devices were
obtained for the study. Each was pretested to verify proper oper¬
ation. The brand and models used were the Respironics REMstar,
Puritan-Bennett Companion 318, Respcare II Sullivan, and Health-
dyne Tranquility. Of the four, only the Tranquility system incorpo¬
rated a pressure transducer and circuitry to regulate delivered
pressure (pressure-compensating). Multiple units from each man¬
ufacturer were not tested.
We obtained permission to use an altitude chamber located at
NASA/Johnson Space Center in Houston. This chamber was used
to simulate altitudes between sea level and 12,000 feet. The cham¬
ber provides temperature control during atmospheric pressure
changes. Each CPAP unit was connected to a Respironics, mediumsmall mask and fitted to an anatomic airway mannequin (Laerdahl;
Armonk, NY). A standard water pressure manometer was attached
to each mask and maintained in place throughout the experiment.
Three simulated altitude ascents to 12,000 feet were made.
Pressures were the same for all machines on each ascent. CPAP
machines were set at 5, 10, and 12 cm H2O pressure for the first,
second, and third ascents, respectively. Initial settings were made
at simulated sea level and allowed to equilibrate for a 10-min pe¬
riod to assure pressure stability. Water manometer pressure mea¬
surements were made at 2,000-foot increments during ascent and
descent. Independent measurements were made by two observers
of all devices at all altitudes and pressure combinations. We allowed
equipment to equilibrate at each altitude before each measurement.
Finally, chamber temperature was maintained at 22.2±1.1°C
throughout the experiment. This control was maintained notwith¬
standing the minimal effect on density of temperature fluctuation
within ambient range.
Analysis of variance was used to assess differences associated with
simulated altitude change, CPAP machines, and initial pressure
setting. Pairwise comparisons between machines were made with
Tukey tests. The relationship between predicted and measured
1578
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21725/ on 06/17/2017
Clinical Investigations
Table 2.Mean Pressure Measurements for Each Machine-Altitude-Pressure Combination
Altitude (xl,000 Feet)
Pressure,
cmH20
Device*
0
4.9
4.9
10.0
9.8
10.0
9.9
12.0
4.6
4.6
5.0
4.5
9.4
9.2
10.0
9.3
11.1
5.1
4.2
8.7
8.3
10.0
8.6
10.5
12.0
11.2
12.0
10.6
12.1
9.9
5.0
5.0
10
12
4.2
4.3
3.8
4.0
5.0
3.8
8.1
7.8
10.0
8.0
3.6
3.6
5.0
3.5
7.6
7.2
10.0
9.7
9.0
12.0
9.2
12.0
8.5
10
12
3.3
3.4
5.0
3.2
3.1
3.1
5.0
3.0
6.6
6.2
10.0
6.3
7.1
6.7
10.0
7.4
6.8
8.3
12.0
7.8
7.8
12.0
7.2
*l=Respcare II; 2=Companion 318; 3=Tranquility; and 4=REMstar.
pressures was assessed with regression analysis. Bias was assessed
after the methods of Bland and Altman.11 A p value of 0.05 or less
was accepted as significant and all tests were two tailed.
Results
Setup, calibration, and attachment of the anatomic
mannequin was accomplished without difficulty. How¬
ever, due to equipment failure, measurements from
the Companion 318 when set at 12 cm H2O pressure
could not be obtained. Four pressure readings were
obtained for all other machine, altitude, and pressure
combinations.
Delivered pressure varied significantly with chang¬
ing altitude for nonpressure-compensating CPAP ma¬
chines (p<0.0001). Table 2 shows mean pressure
measurements for each machine-altitude-pressure
combination. The pressure sensor-equipped machine
(Tranquility) maintained pressure at ± 1 mm H2O at all
altitude and pressure combinations. Small but statisti¬
reliable differences between noncompensating
cally
machines were found in delivered pressure as a func¬
tion of altitude.
Regression of collective fan law predictions (Fig 1)
against data collected from noncompensatory devices
produced
highly concordant results (r=0.99, p<0.0001).
The slope of the fitted regression line was not signif¬
from 1.0. Observed vs predicted
icantly different
are
pressures
plotted in difference
Figure 1 along with a line of
Mean
of predicted minus
(±SEM)
identity.
observed measurements (bias) was 0.12±0.019 cm
H2O. Figure 2 illustrates the difference between the
and each actual pressure measurement in
predicted
the manner used by Bland and Altman;11 the differ¬
ence is plotted against the pair mean.
8
6
8
10
Measured Pressure
Figure 1. Observed vs predicted CPAP pressures in cm H2O. The
line arising from the origin is the line of identity.
10
Pair Mean
Figure 2. The differences between predicted CPAP values and
the mean of the measured pressures are depicted in this graphic.
The line represents the mean difference (bias).
CHEST / 108 / 6 / DECEMBER, 1995
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21725/ on 06/17/2017
1579
Table 3.Changes in Density With Altitude Assuming
Constant
Standard
Altitude,
Feet
0
5,000
10,000
15,000
Temperature,
15.00
5.09
-4.81
-14.72
Temperature
Density Ratio
Temperature
Density Ratio
Standard
Temperature
1.0000
0.8320
1.0000
0.8617
0.6877
0.5643
0.6292
Constant
0.7385
Increase arousals from sleep has been documented.9,1'
The effect of reduced ambient pressure on the partial
pressure of oxygen may be an important effect beyond
the simple reduction in ambient pressure.
Conclusions
The delivered CPAP of noncompensating devices is
significantly affected
by changes in altitude. Pressure
devices
maintained delivered pres¬
sensor-equipped
the altitudes simulated in the present
throughout
The
collective
fan laws appear quite adequate
study.
for estimation of delivered pressure at altitude. These
estimates are easily calculated and can be used for ad¬
justing noncompensating CPAP machine for changes
in elevation. Until further studies are conducted,
clinicians may opt to prescribe pressure-compensating
CPAP machines to patients who frequently travel to
places that widely differ in elevation.
sure
Discussion
Untreated sleep-disordered breathing represents a
Pathophysiologic manifestations
include hypoxemia, cor pulmonale, pulmonary and
systemic hypertension, nocturnal dysrhythmias, and
increased risk of myocardial infarction and stroke.1213
Behavioral symptoms include snoring, hypersomnolence, depression, cognitive impairment, and erectile
failure. The serious medical, social, and psychologic
consequences make diagnosis and management of
serious health risk.
sleep-disordered
breathing crucial.1
Inattention to effects of altitude on CPAP machine
can introduce error into prescribing op¬
performance
timal pressure settings for some patients. Table 3
shows density ratios for various altitudes at standard
and constant temperature. Pressures delivered by
CPAP devices decrease as a function of altitude
increase unless they are equipped with pressure-com¬
pensating sensors and circuitry. The absolute decrease
in delivered CPAP is greater with higher CPAP
settings. Thus, patients requiring higher pressure may
be affected more. This observation's clinical impor¬
tance in vivo requires further study. Our study tested
CPAP performance on mannequins; we did not assess
the effect pressure changes potentially have on phys¬
iologic dynamics in patients with sleep-disordered
breathing.
The collective fan laws adequately estimate deliv¬
ered pressure at altitude. Deviations of predicted from
measured pressure were small and are unlikely to be
clinically important. The small but statistically signifi¬
cant delivered
differences between noncompressure
pensating machines as a function of altitude are like¬
wise of doubtful clinical
significance.
This study suggests that the symptom recurrence
associated with altitude changes reported by our
patients with sleep-disordered breathing may relate to
delivered mask pressure. However, other mechanisms
may also influence symptom recurrence. High altitude
can induce periodic breathing.14 Most lowlanders de¬
periodic breathing at elevations of 17,000 feet or
velop
more.5,16 Sleep is commonly disrupted in normal in¬
dividuals at high altitude; climbers frequently complain
of nonrefreshing sleep after a night at high altitude.
References
Kryger MH. Management of obstructive sleep apnea. Clin Chest
Med 1992; 13:481-92
2 Nino-Murcia G, McCann CC, Bliwise DL, et al. Compliance and
side effects in sleep apnea patients treated with nasal continuous
positive airway pressure. West J Med 1989; 150:165-69
3 Sullivan CE, Issa FG, Berthon-Jones M, et al. Reversal of
obstructive sleep apnoea by continuous positive airway pressure
applied through the nares. Lancet 1981; 1:862-65
4 Sanders MH. Nasal CPAP effect on patterns of sleep apnea.
Chest 1984; 86:839-44
5 Strohl KP, Redline S. Nasal CPAP therapy, upper airway muscle
activation, and obstructive sleep apnea. Am Rev Respir Dis 1986;
134:555-58
6 Hoffstein V, Zamel N, Phillipson EA. Lung volume dependence
of pharyngeal cross-sectional area in patients with obstructive
sleep apnea. Am Rev Respir Dis 1984; 130:175-78
7 Brown I, Taylor R, Hoffstein V. Obstructive sleep apnea reversed
by increased lung volume? Eur J Respir Dis 1986; 68:375-80
8 Series F, Cormier Y, Lampron N, et al. Increasing functional re¬
sidual capacity may reverse obstructive sleep apnea. Sleep 1988;
11:349-53
9 Reite M, Jackson D, Cahoon RL, et al. Sleep physiology at high
altitude. Electroenceph Clin Neurophysiol 1975; 38:463-71
10 Baumeister T, Avallone EA. Marks' standard handbook for me¬
chanical engineers. 9th ed. New York: McGraw, 1987
11 Bland J, Altman D. Statistical methods for assessing agreement
between two methods of clinical measurement. Lancet 1986;
1:307-10
12 Shepard JW. Hypertension, cardiac arrhythmias, myocardial in¬
farction, and stroke in relation to obstructive sleep apnea. Clin
Chest Med 1992; 13:437-58
13 Kales A, Vela-Bueno A, Kales JD. Sleep disorders: sleep apnea
and narcolepsy. Ann Intern Med 1987; 106:434-43
14 Weil JV. Sleep at high altitude. Clin Chest Med 1985; 6:615-21
15 Lenfant C. Time-dependent variations of pulmonary gas ex¬
change in normal men at rest. J Appl Physiol 1967; 22:675-84
16 Piriban I. An analysis of some short-term patterns of breathing in
man at rest. J Appl Physiol 1963; 166:425-34
17 Weil JV, Kryger MH, Scoggin CH. Sleep and breathing at high
altitude. In: Guilleminault C, Dement W, eds. Sleep apnea syn¬
dromes. New York: Alan R. Liss, 1978; 119-36
1
1580
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21725/ on 06/17/2017
Clinical
Investigations