DOI 10.1378/chest.93.5.911 1988;93;911-915Chest

 J Räsänen, J B Downs and M C Stock ventilation.pressure ventilation and airway pressure release Cardiovascular effects of conventional positive

  http://chestjournal.chestpubs.org/content/93/5/911

can be found online on the World Wide Web at: The online version of this article, along with updated information and services 

) ISSN:0012-3692http://chestjournal.chestpubs.org/site/misc/reprints.xhtml(without the prior written permission of the copyright holder.reserved. No part of this article or PDF may be reproduced or distributedChest Physicians, 3300 Dundee Road, Northbrook, IL 60062. All rights

ofbeen published monthly since 1935. Copyright1988by the American College is the official journal of the American College of Chest Physicians. It hasChest

 © 1988 American College of Chest Physicians by guest on July 11, 2011chestjournal.chestpubs.orgDownloaded from

1

2

DOG

CHEST 1 93 I 5 I MAY, 1988 911

Cardiovascular Effects of ConventionalPositive Pressure Ventilation and AirwayPressure Release Ventilation*

Jukka R#{228}s#{228}nen,M.D.;tJohn B. Downs, M.D.;1 and

M. Christine Stock, M.D. , F.C.C.P�

The hemodynamic sequelae of conventional positive pres-sure ventilation (CPPV), airway pressure release ventilation

(APRV), and spontaneous breathing were compared withcontinuous positive airway pressure (CPAP) in ten anesthe-tized dogs who had ventilatory failure with and withoutparenchymal lung injury. The APRV corrected respiratory

acidosis without significantly effecting arterial blood oxy-genation, venous admixture, cardiovascular function, or

tissue oxygen utilization. Application of CPPV precipitatedmarked depressions in blood pressure, stroke volume, and

N eed for mechanical ventilatory support in patients

with acute lung injury is variable. In many

patients, application of continuous positive airway

pressure (CPAP) produces sufficient improvement in

lung mechanics to allow spontaneous ventilation with

minimal work of �2 However, mechanical

augmentation ofalveolar ventilation cannot be avoided

in all patients, even with appropriate CPAP therapy.

Conventional positive pressure ventilation (CPPV)

impairs cardiac performance in animals and humans

who have normal circulatory function because it de-

creases systemic venous return.34 Adverse cardiovas-

cular effects ofmechanical ventilatory support may be

reduced by using intermittent mandatory ventilation,

which allows spontaneous respiratory efforts to de-

crease mean intrathoracic pressure. However, the

advantage is negligible if a high level of mechanical

ventilatory support is necessary.

Airway pressure release ventilation (APRV), a re-

cently introduced ventilatory modality, has been

shown to provide adequate alveolar ventilation to dogs

with normal or injured lungs and to humans with mild

lung injury following cardiopulmonary bypass.�7 An

APRV breathing circuit consists ofa CPAP system with

a pressure release valve added to the expiratory limb

(Fig 1). Augmentation ofalveolar ventilation is accom-

plished by transiently decreasing airway pressure from

*From the Department of Anesthesiology, Ohio State University,

Columbus; and the Department of Anesthesiology, Emory Uni-versity School of Medicine, Atlanta.

tCritical Care Research Fellow.tProfessor and Vice Chairman of Anesthesiology§Assistant Professor of Anesthesiology (Emory).Manuscript received June 18; revision accepted October 22.Reprint requests: Dr Down.s OSU Hospital, 410 West Tenth Avenue,Columbus 43017

cardiac output. A concomitant decrease in venous admix-tiire did not compensate for these adverse cardiovascular

effects. Deterioration of tissue oxygen delivery resulted inoxygen supply-demand imbalance during CPPV. The results

of this experimental study indicate that if ventilatoryaugmentation of subjects who require CPAP is desired,

APRV will enhance alveolar ventilation without compro-mising circulatory function and tissue oxygen balance,

whereas CPPV will impair cardiovascular function signifi-cantly.

CPAP to a lower level. Lung volume is thereby reduced

below functional residual capacity, and carbon dioxide

excretion is facilitated. When the release valve closes,

CPAP is re-established. During APR\� spontaneous

breathing may occur throughout the mechanical cycle.

However, controlled ventilation may be achieved using

APRV by increasing the rate or magnitude of airway

pressure release, and thereby, alveolar ventilation.

Previous comparisons of APRV and CPP\� using

similar mean airway pressures, have revealed no dif-

ferences in cardiovascular function.6’7 However, in dogswith lung injury, APRV produced more effective gas

exchange than CPPV6 Support for patients with acute

3

FIGURE 1. The breathing circuit. An oxygen-powered Venturi device

(1) provides a high continuous flow of gas that exits through one oftwo threshold resistor valves (2,3). Flow through one ofthe threshold

resistor valves 3) is regulated by a release valve (4) which is

controlled by a timer (5). The airway pressure pattern depends onthe opening pressures of the threshold resistors and the timing of

the release valve. The release valve is closed during spontaneous

breathing with CPAP

 © 1988 American College of Chest Physicians by guest on July 11, 2011chestjournal.chestpubs.orgDownloaded from

912 Cardiovascular Effects of Conventional Positive Pressure Ventilation (R#{228}s#{228}nen,Downs, Stock)

respiratory failure frequently requires addition of

ventilatory support to existing CPAP therapy, in which

case APRV would augment ventilation with lower

mean airway and intrathoracic pressures and, possibly,

with less cardiovascular interference than that ob-

served during CPPV This investigation compares

cardiopulmonary performance during spontaneous

breathing, CPP\� and APR\� all delivered with a similar

level of CPAP� in dogs with ventilatory failure, with

and without parenchymal lung injury.

SUBJECTS AND METHODS

Ten mongrel dogs were anesthetized intravenously with 25

mg/kg pentobarbital. Anesthesia was maintained with an intravenous

infusion of pentobarbital 4 mg/kg/h. The right femoral artery was

cannulated for continuous measurement of systemic blood pressure

and for arterial blood sampling. A pulmonary artery catheter was

inserted from the right external jugular vein, for measurement of

pulmonary artery pressure, right atrial pressure, and pulmonary

artery occluded pressure, and for sampling mixed venous blood.

Airway pressure was measured at the carmna with an air-filled

catheter which also was used intermittently for measurement of

end-tidal carbon dioxide tension. A 16-gauge, 20 cm long Teflon

catheter was inserted 10 cm into the right pleural space for

measurement ofpleural pressure.8

After instrumentation, the animals were connected to a breathing

circuit that permitted alternation between spontaneous breathing

with CPAI� APR�� and CPPV (Fig 1). Initiall�.; the CPAP level was

10 cmH2O and animals breathed spontaneously with an inspired

oxygen concentration (FIo�) of 0.30. A 15-minute equilibration

period was allowed before assessment of cardiopulmonary function.

Then, ventilation was controlled with either APRV or CPP\� applied

in random order. The APRV was delivered by releasing airway

pressure from 10 cm H2O to ambient for 1.5 5; CPPV was

accomplished by increasing airway pressure from 10 to 20 cmH2O

for 1.5 5. Ventilator rate was adjusted to prevent spontaneous

breathing with minimal hyperventilation and to assure similar

arterial blood carbon dioxide tension (PaCO,) during APRV and

CPPV Cardiopulmonary function was reassessed after 15 minutes

of ventilation with either mode.

After all three ventilatory modalities had been studied, the

animals were ventilated with APR� and acute lung injury was

induced by injecting 90 � of oleic acid into the right atrium.

Lung injury was allowed to develop for 90 minutes, during which

time the APRV rate was increased to maintain ventilatory control

with minimal hyperventilation. If20 APRV breaths per minute was

insufficient to prevent spontaneous ventilatory efforts, CPAP was

increased to allow a larger airway pressure release gradient and

tidal volume. The APRV rate then could be reduced to maintain an

airway pressure release time shorter than one halfofthe respiratory

cycle. The level of CPAP was not altered during the remainder of

the experiment. If a ventilator rate higher than 20 breaths per

minute was required to control ventilation during CPP\� the

inspiratory time was reduced to maintain an inspiratory-to-expira-

tory time ratio less than 1. The FIo, remained at 0.30. After lung

injury developed, APR\7 and CPPV were applied in the sequence

assigned at previous randomization. Then, mechanical ventilation

was discontinued, and the dogs breathed spontaneously with CPAP

Cardiopulmonary measurements were repeated after 15 minutes of

equilibration at each stage. Peak airway pressure during CPPV was

adjusted so that airway pressure change was similar to that created

by APRV Similarly, the frequency of positive pressure breaths was

adjusted to maintain PaCO2 equivalent to that measured during

APRV After all data were collected, animals were killed, and lung

injury was confirmed by gross postmortem inspection.

Arterial and mixed venous blood samples were analyzed promptly

for blood gas and pH values. Blood oxyhemoglobin saturations were

calculated with a computer program described by Ruiz et al.’

Thermodilution cardiac output was determined by averaging the

results of five measurements obtained with 5 ml of room-tempera-

ture 5 percent dextrose solution injected at a random moment in

the respiratory cycle. Systemic and pulmonary artery pressures,

transmural cardiac filling pressures, airway, pleural, and transpul-

monary pressures obtained from pressure tracings, were averaged

for one complete respiratory cycle during each ventilatory modality.

Variation in the measured variables between respiratory cycles was

minimal. Venous admixture (Qs�/Q�r), alveolar-arterial oxygen ten-

sion gradient, stroke volume, mean vascular pressures, pulmonary

and systemic vascular resistances, oxygen delivery, oxygen con-

sumption, and oxygen extraction ratio were calculated from standard

formulae.

The effects of the ventilatory modalities on cardiopulmonary

function were analyzed using a one-way, repeated measures analysis

of variance followed by Tukey’s test for multiple comparisons. A

difference was considered statistically significant if the probability

oftype a error was less than 5 percent.

RESULTS

The variables reflecting pulmonary function are

presented in Table 1, and those reflecting cardiovas-

cular function are shown in Table 2. Oleic acid

produced moderate to severe acute lung injury in all

dogs, as evidenced by significant deterioration of

oxygenation, increased QSP/QT, tachypnea, elevated

pulmonary vascular resistance, and morphologic

changes in the lungs. Patchy areas of subpleural

hemorrhage, and blood-tinged froth in the airways

were found by postmortem examination ofall animals.

Apart from increased pulmonary vascular resistance,

the cardiovascular effects of oleic acid were minimal

when the dogs breathed spontaneously.

Pulmonary Function

In this study, deep general anesthesia was used

deliberately to produce ventilatory failure and respi-

ratory acidosis during spontaneous breathing, an effect

augmented by the development of lung injury in the

second part of the investigation. Control of ventilation

with both APRV and CPPV corrected ventilatory

failure and produced normal mean PaCO2 and pHa

values both before and after induction of lung injury.

This was accomplished using a similar airway pressure

gradient and respiratory rate during APRV and CPPV

Therefore, peak airway pressure during CPPV was

twice that during APRV Mean airway and pleural

pressures were significantly elevated by CPP� but no

significant differences were observed in mean trans-

pulmonary pressure between the three ventilatory

modalities. The inspiratory-to-expiratory time ratio

during CPPV was 0. 47 ± 0. 12 when the dogs’ lungs

were normal, and 0.63 ± 0. 16 after induction of lung

injury.

Mechanical ventilation using APRV and CPPV con-

sistently improved arterial blood oxygenation, al-

 © 1988 American College of Chest Physicians by guest on July 11, 2011chestjournal.chestpubs.orgDownloaded from

Table 1-Variables Reflecting Pulmonary Function in Ten Dogs During Spontaneous Breathing with CPAP, During APRV,and During CPPV, before (NL) and After (ALl) Lung Injury

CPAP CPPV APRV

Respiratory rate NL 7 ± 5 13 ± 3t 1 1 ± 3t

(cpm)* ALl 42±28� 17±4t 14±2t

Peakairwaypressure NL 12±2 20±lt 10± it

(cm H20) ALl 15±3 26±5t 13±3t

Meanairwaypressure NL 9±1 13±lt 8±it

(cm H20) ALl 12±3� i6±4t 9±3t

Meanpleuralpressure NL -5±3 -2±3t -6±3t

(cm H20) AL! -3±3 0±3t -5±3t

Meantranspulmonary NL 14±2 15±4 14±3

pressure (cm H,O) ALl 15±3 16±3 14±3

PaCO2(mm Hg) NL

ALl

55± ii

55±13

39±4t

37±7t

37±5t

36±5t

pHa NL 7.25±07 7.35±.05t 7.37±.06t

ALl 7.22±09 7.37±.04t 7.37±.04t

PaO,(mmHg) NL

ALl

121±21

92±26�

147±i9t

104±26

147±21

82± 26

SaO2(%) NL

ALl

97±4

88±18

99±1

96±3

99±1

94±4

Q,�/Q,(%) NL 15±11 4±3 5±4

ALl 32±20� 15±7t 29± lit

*Spontaneous respiratory rate is reported during CPA1� ventilator rate during CPPV and APRV when spontaneous breathing was absent.

tStatistically significant difference compared to spontaneous breathing.

tStatistically significant difference compared to CPPV

§Statistically significant difference compared to spontaneous breathing without lung injury.

CHEST/93/5/MAY,1988 913

though this effect was statistically significant only for

Pa02 when the lungs were normal. Even though no

statistically significant difference in arterial blood

oxygenation was observed between APRV and CPP\�

venous admixture in injured lungs was significantly

lower during CPPV

Cardiovascular Function

No significant differences in cardiovascular function

were detected between APRV and spontaneous

breathing with CPAI� apart from slightly increased

systemic vascular resistance during APRV when the

lungs were normal. In contrast, CPPV decreased blood

pressure, stroke volume, cardiac output, and oxygen

delivery significantly, regardless of the presence or

absence of acute lung injury. Because oxygen con-

sumption remained unchanged, tissue oxygen extrac-

tion increased during CPPV The CPPV also was

associated with a tendency toward increased pulmo-

nary vascular resistance. Changing the ventilatory

modality did not influence heart rate or transmural

left and right ventricular filling pressures.

DISCUSSION

This investigation compared the hemodynamic ef-

fects of APR� CPP\� and spontaneous breathing,

administered with an unchanged level of CPAP The

results show that ventilation can be controlled using

APR\� without compromising cardiopulmonary func-

tion. In contrast, CPPV impaired circulatory function

and tissue oxygen balance.

Unlike any other mechanical ventilatory technique,

with the exception of external negative pressure de-

vices, the respiratory cycle during APRV is associated

with a decrease in intrathoracic pressure. Intrathoracic

pressure fluctuations produced by APRV closely mim-

icked those seen during spontaneous breathing with

CPAP (Fig 2). Theoretically, therefore, performance of

an initially normal, preload-dependent cardiovascular

system should be well maintained during transition

from spontaneous breathing to APR\� with a similar

level of CPAP In contrast, elevation of airway and

intrathoracic pressure during positive pressure breath-

ing likely would impair stroke volume, cardiac output,

and systemic oxygen 1�

In the current study, stroke volume decreased

markedly when control of ventilation was accom-

plished with CPPV compared to both APRV and

spontaneous breathing with CPAP Depression of car-

diac function during CPPV also was characterized by

hypotension and narrowing ofsystemic pulse pressure.

Since cardiac chamber volumes were not measured in

this investigation, the mechanisms of circulatory com-

promise cannot be assessed in detail. However, reduc-

tion in systemic venous return and mechanical corn-

pression of the heart by the inflated lungs are the

primary reasons for reduction ofcardiac output during

CPPV tO-12 Lung injury had little effect on the cardio-

vascular changes produced by application of CPPV

The slightly higher stroke volume during spontaneous

breathing compared to APRV was expected. The

experimental design dictated that respiratory acidosis

 © 1988 American College of Chest Physicians by guest on July 11, 2011chestjournal.chestpubs.orgDownloaded from

aw

IE

� io-w� -

U)(1)LU

CPAP APRV CPPV

���irtFIGURE 2. Changes in airway pressure (P,,,), pleural pressure (Pr,),

and transpulmonary pressure (shaded area), during spontaneous

breathing with CPAP� during APR\� and during CPPV

914 Cardiovascular Effects of Conventional Positive Pressure Ventilation (RSsSnen, Downs, Stock)

Table 2-Variables Reflecting Cardiovascular Function in Ten Dogs During Spontaneous Breathing with CPAP, DuringAPRV, and During CPPV, Before (NL) and After (ALl) Lung lnjury

CPAP CPPV APRV

Systolicbboodpressure NL 107±15 94±16 120±ii*

(mmHg) ALl 106±21 87±20t 1i6±l0*

Diastolic blood pressure NL 76± 11 71 ± 14 92±9*

(mmHg) ALl 74±i9 64±16 86±10*

Heartrate(bpm) NL

ALl

141±25

157±21

153±22

163±i5

145±33

162±20

Stroke volume (ml) NL

ALl

22 ± 7

19±5

14 ± 4t

ll±3t

i9 ± 3*

18±4*

Right atrial pressure NL 13±4 ii ±6 13±4

(mm Hg) AL! 13±4 10±5 13±4

PAocclusionpressure NL 14±4 12±5 14±4

(mm Hg) AL! 15±5 12±5 15±5

Systemic vascular NL 2170 ± 614 2633 ± 663 2874 ± 596t

resistance (dynes/cm’) AL! 2159 ± 684 2833 ± 854 2546 ± 635

Pulmonary vascular NL 159 ± 46 239 ± 74 160 ± 48

resistance (dynes/cm’) ALl 294 ± 86t 365 ± 134 256 ± 72*

Oxygen consumption NL 87 ± 30 89 ± 16 88 ± 20

(mI/mm) ALl 83±18 75±16 86± 19

Oxygen delivery NL 431 ± 117 323 ± 83t 401 ± 104

(mI/mm) AL! 367±97 251±54t 394�73*

Oxygen extraction NL 0.20± .04 0.28± .06t 0.22±04

ratio ALl 0.24±06 0.32±12t 0.22±94*

*Statistjodly significant difference compared to CPPV

tStatistically significant difference compared to spontaneous breathing.

tStatistically significant difference compared to spontaneous breathing without lung injury

would occur during spontaneous breathing. This may

have increased endogenous catecholamine levels,

which may explain why stroke volume was greater

during spontaneous breathing than during APRV

During normal spontaneous respiration, decrease in

intrathoracic pressure increases transpulmonary pres-

sure and lung volume. In contrast, the depression of

intrathoracic pressure that is created by APRV is

associated with decreased transpulmonary pressure

and deflation of the lung (Fig 2). Since efficiency of

gas exchange is related closely to mean lung volume,

particularly during acute lung injury, periodic lung

deflation below FRC theoretically could cause oxygen-

ation to deteriorate during APRV However, in this

study, arterial oxygenation was well maintained during

APRV Before induction oflung injury, mildly elevated

QSP/QT and hypoventilation caused arterial hypox-

emia. Augmentation ofventilation, using either APRV

or CPP� increased PaO2 by decreasing alveolar Pco2,

but did not change QSP/QT significantly. Although

cardiac output was reduced during CPP\� it did not

cause hypoxemia because Qsp!Q’r was minimal. Injec-

tion of oleic acid caused moderate to severe parenchy-

mal lung injury in all animals, and increased QSP/Qr

substantially, despite a higher level of CPAP Ventila-

tory support using APRV corrected the respiratory

acidosis, but did not alter Qs�!Qr, probably because

mean transpulmonary pressure decreased only mini-

mally. However, improvement in alveolar ventilation

corrected the rightward shift of the oxyhemoglobin

dissociation curve that was induced by respiratory

acidosis, and effected a net improvement in arterial

blood oxygenation by increasing oxyhemoglobin satu-

ration. In clinical practice, CPAP level would be

increased during APRV to improve arterial oxygena-

tion. In addition, the level to which airway pressure

falls during pressure release would be limited, in order

to prevent total lung collapse. These conditions were

not imposed in this experiment in order to maintain a

constant CPAP level during all three experimental

conditions, and to maintain similar airway pressure

gradients during ventilation with APRV and CPPV

Application of CPPV resulted in apparent improve-

ment in matching of ventilation and perfusion with

 © 1988 American College of Chest Physicians by guest on July 11, 2011chestjournal.chestpubs.orgDownloaded from

CHEST/93/5/MAY,1988 915

significant decrease in QSP!QT. It is possible that the

increase in transpulmonary pressure during the posi-

tive pressure breaths improved the ventilation of

previously underventilated lung units. Possible over-

estimation of pleural pressure at high lung volumes

may have led to an underestimation of transpulmonary

pressure during CPPV However, the reduction in

QSP/QT during CPPV may in part have been secondary

to reduced cardiac output and a preferential decrease

in perfusion to alveoli with a ventilation-to-perfusion

ratio close to � Our methodology is incapable of

determining the primary mechanism of QSP/QT re-

duction.

Because the respiratory and cardiovascular systems

serve to facilitate tissue gas exchange, tissue oxygen

dynamics ultimately determine the value of a ventila-

tory modality and guide its application. In the present

investigation, spontaneous breathing with CPAP and

controlled ventilation with APRV were associated with

similar oxygen delivery and oxygen extraction. During

CPP\� cardiovascular depression clearly outweighed

the advantages of slightly improved pulmonary func-

tion, which resulted in a net reduction in oxygen

delivery that necessitated increased oxygen extraction.

Therefore, our results indicate that if treatment of

ventilatory failure and reduction in the work of breath-

ing is attempted, addition of APRV breaths to CPAP

will accomplish the desired goal without interfering

with pulmonary oxygen transfer, circulatory function,

or tissue oxygen utilization. Application of conven-

tional continuous positive pressure ventilation also will

correct respiratory acidosis, but at the expense of high

peak airway pressure, impaired cardiovascular per-

formance, and compromised tissue oxygen balance.

ACKNOWLEDGMENTS: We gratefully acknowledge the technicalassistance of Roger Dzwonczyk, M.S.B.M.E., Deborah A. Fro-licher, B.S., and Michael R. Hodges.

REFERENCES

1 Katz JA, Marks JD. Inspiratory work with and without continu-

ous positive airway pressure in patients with acute respiratory

failure. Anesthesiology 1985; 63:598-607

2 R#{228}sanenJ, Heikkil#{228} J, Downs JB, Nikki P� Vaisanen IT, Viitanen

A. Continuous positive airway pressure by face mask in the

treatment ofcardiogenic pulmonary edema. Am J Cardiol 1985;

55:296-309

3 Cournand A, Motley HL, Werk#{246}L, Richards DW Physiological

studies ofthe effects ofintermittent positive pressure ventilation

on cardiac output in man. Am J Physiol 1948; i52:i62-74

4 Qvist J, Pontoppidan H, Wilson RS, Lowenstein E, Layer MB.

Hemodynamic responses to mechanical ventilation with PEEP:

the effect ofhypervolemia. Anesthesiology 1975; 42:45-55

5 Downs JB, Stock MC. Airway pressure release ventilation: a

new method of ventilatory support (Editorial). Crit Care Med

1987; 15:459-61

6 Stock MC, Downs JB, Frolicher DA. Airway pressure release

ventilation. Crit Care Med 1987; 15:462-66

7 Downs JB, Stock MC, R#{228}sanen J. Airway pressure release

ventilation (APRV): a new aproach to the management of acute

lung injury In: Vincent JL, ed. Update in intensive care and

emergency medicine. Heidelberg: Pringer Verlag, 1987:228-33

8 Downs JB. A technique for direct measurement of pleural

pressure. Cnt Care Med 1976; 4:207-10

9 Rui.z BC, Tucker WK, Kirby RR. A program for calculating

intrapulmonary shunts, blood-gas, and acid-base values with a

programmable calculator. Anesthesiology 1975; 42:88-131

10 Weber KT, Janicki JS, Hunter WC, Shroff S, Pearlman E,

Fishman AP The contractile behavior of the heart and its

functional coupling to the circulation. Progr Cardiovasc Dis

1982; 24:375-400

ii Robotham JL, Cherry D, Mitzner W, Rabson JL, Lixfeld W,

Bromberger-Barnea B. A re-evaluation of the hemodynamic

consequences of intermittent positive pressure ventilation. Cnt

Care Med 1983; 11:783-93

12 Marini JJ, Culver BH, Butler J. Mechanical effect of lung

distension with positive pressure on cardiac function. Am Rev

Respir Dis 1981; 124:382-86

13 Dantzker DR, Brook CJ, Dehart P� Lynch JI� Weg JG. Ventila-

tion-perfusion distributions in the adult respiratory distress

syndrome. Am Rev Respir Dis 1979; 120:1039-52

 © 1988 American College of Chest Physicians by guest on July 11, 2011chestjournal.chestpubs.orgDownloaded from

DOI 10.1378/chest.93.5.911 1988;93; 911-915Chest

J Räsänen, J B Downs and M C Stockairway pressure release ventilation.

Cardiovascular effects of conventional positive pressure ventilation and

 July 11, 2011This information is current as of

 

http://chestjournal.chestpubs.org/content/93/5/911Updated Information and services can be found at:

Updated Information & Services

http://chestjournal.chestpubs.org/content/93/5/911#related-urlsThis article has been cited by 4 HighWire-hosted articles:

Cited Bys

http://www.chestpubs.org/site/misc/reprints.xhtmlonline at: Information about reproducing this article in parts (figures, tables) or in its entirety can be foundPermissions & Licensing

http://www.chestpubs.org/site/misc/reprints.xhtmlInformation about ordering reprints can be found online:

Reprints

the right of the online article.Receive free e-mail alerts when new articles cite this article. To sign up, select the "Services" link to

Citation Alerts

slide format. See any online figure for directions. articles can be downloaded for teaching purposes in PowerPointCHESTFigures that appear in Images in PowerPoint format

 © 1988 American College of Chest Physicians by guest on July 11, 2011chestjournal.chestpubs.orgDownloaded from