DOI 10.1378/chest.106.6.1788 1994;106;1788-1792Chest

 D. Todd Loffert, David Ikle and Harold S. Nelson A Comparison of Commercial Jet Nebulizers

  http://chestjournal.chestpubs.org/content/106/6/1788

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

ISSN:0012-3692)http://chestjournal.chestpubs.org/site/misc/reprints.xhtml(

of the copyright holder.may be reproduced or distributed without the prior written permission Northbrook, IL 60062. All rights reserved. No part of this article or PDFby the American College of Chest Physicians, 3300 Dundee Road,

1994Physicians. It has been published monthly since 1935. Copyright is the official journal of the American College of ChestCHEST

 1994, by the American College of Chest Physicians by guest on May 7, 2010chestjournal.chestpubs.orgDownloaded from

A Comparison of Commercial JetNebulizers*D. Todd Loffert; David Ikle, Ph.D.; and Harold S. Nelson, M.D.

Seventeen commercially available jet nebulizers from15 commercial sources were studied (Acorn-I, Acorn-II,AquaTower, AVA-NEB, Cirrhus, Dart, DeVilbiss 646,Downdraft, Fan Jet, MB-5, Misty Neb, PARI LC JET,PARI-JET, Salter 8900, Sidestream, Updraft-II, WhisperJet). All nebulizers were filled with 2 ml of saline solu-tion plus 0.5 ml of albuterol and powered with the samesource (DeVilbiss PulmoAide). We compared total out-put (TO), time for total output (TTO), and percent outputin respirable range (PORR). The TO was obtained byweighing before nebulization and at the point of eight-fold decline in output. The ITO was calculated frominitiation of nebulization to the point of eightfolddecline in output. The PORR was measured by a laserparticle analyzer in continuous nebulization to the samepoint of abrupt drop in output. The TO varied from 0.98To 1.86 ml (p<0.0001) with the Acorn-I, Acorn-TI,Updraft-II, and Sidestream, significantly greater thanthe others (p<0.05). The TTO varied from 2.28 to 20.95min (p<0.0001). The AquaTower, PARI LC JET andPARI-JET, DeVilbiss, and Dart were significantlyshorter than the others (p<0.05). The PORR varied from21.89 to 71.95 percent (p<0.0001). The Sidestream wassignificantly greater than all others (p<0.05). The PARI

T nhaled medications for the treatment of asthmahave become the mainstay of therapy. The prin-

cipal means use to deliver inhaled particles arethrough jet nebulizers and fluorocarbon metered-dose inhalers (MDI). While the MDls are more con-venient,1 they cannot be used by young children andpresent problems to some individuals of all ages.2Furthermore, there are impending limits on the useof fluorocarbon-containing materials due to ozonedepletion.3 The jet nebulizer may then gain evengreater popularity.4Many models of small-volume jet nebulizers are

commercially available. Recommendations for med-ication doses by nebulizer treatment usually do notspecify a particular model, suggesting a perceptionthat they are functionally equivalent. There are,however, reports that suggest this is not true and thatdifferent models vary considerably in performance,

*From the National Jewish Center for Immunology and Respi-ratory Medicine, Denver.

Manuscript received November 8, 1993; revision accepted April19, 1994.Reprint requests: Dr. Nelson, National Jewish Center forImmunology and Respiratory Medicine, 1400 Jackson Street,Denver, CO 80206

LC JET and PARI-JET were, in turn, significantlygreater than the remaining models (p<0.05). To combinethese characteristics, we calculated respirable particledelivery rate (RPDR) by dividing TO by TTO and mul-tiplying by PORR. The RPDR varied from 0.03 ml/minto 0.26 ml/min (p<0.0001). The PARI LC JET (0.24 ml/min) and the PARI-JET (0.26 mg/min) had a RPDR thatwas significantly greater than the other models except theAquaTower, which, however, had a markedly variableperformance. The Sidestream (0.19 mg/ml) did not differsignificantly from the above group, nor from the DeVilbissand Downdraft. All other models had significantly loweroutputs (p<0.05). We conclude that the output character-istics of commercial nebulizers vary greatly and willimpact on the time required for treatment as well as thetotal amount of drug delivered to the lungs.

(Chest 1994; 106:1788-93)

MDI=metered-dose inhaler; PORR=percent output in re-spirable range; RPDR=respirable particle delivery rate;TO=total output; TTO=time for total output; VMD=vol-ume mean diameter

Key words: aerosol therapy; jet nebulizers

including the rate and extent of delivery5-9 and thepercent of output in the respirable range (<5 um).0-'13Furthermore, even within a single nebulizer model,there can be inconsistencies in unit performance.'4'15We report herein a study that was designed to ex-

plore the differences among the different models ofreusable jet nebulizers commercially available in theUnited States at the time of the study. We examined,as well, the performance consistency by studyingfour units of each model. We demonstrated differ-ences in volume output, time required to nebulizethat volume, and in the percent of particles in therespirable range of 1 to 5 gm.16 The results of thisstudy should allow clinicians to select nebulizers onthe basis of demonstrated performance rather thansolely on convenience and cost.

MATERIALS AND METHODSNebulizersThe 17 different nebulizers studies were the Acorn-I, Acorn-Il,

Fan Jet, Downdraft, and Whisper Jet (Marquest, Englewood,Colo), Aqua Tower (Medical Industries America, Adel, Iowa),AVA-NEB and Updraft-II (Hudson, Temecula, Calif), Cirrhus(DHD Medical Products, Canastota, NY), Dart (ProfessionalMedical Products, Greenwood, SC), DeVilbiss 646 (DeVilbiss

A Comparison of Commercial Jet Nebulizers (Loffert, Ikle, Nelson)1788

 1994, by the American College of Chest Physicians by guest on May 7, 2010chestjournal.chestpubs.orgDownloaded from

Health Care, Somerset, Pa), MB-5 (Mefar, Bovezzo, Italy), MistyNeb (Baxter, Valencia, Calif), PARI LC JET and PARI-JET(PARI Respiratory, Richmond, Va), Salter 8900 (Salter Labs,Arvin, Calif), and the Sidestream (Inspired Medical Products,Hopkinton, Mass). The nebulizers were studied using four unitsof each model, each run three times to determine intramodel andintermodel variability. All nebulizers studied were available in theUnited States. All were purchased through normal procurementchannels to preclude special pretesting. All models were repre-sented by the manufacturer to be reusable.

Volume FillEach nebulizer was filled with 2 ml of normal saline solution

and 0.5 ml of albuterol (Proventil Nebulizer Solution, ScheringCorp, Kenilworth, NJ). The initial and final weights wereprecisely determined on a balance (OHAUS Scale Corp, FlorhamPark, NJ).

Testing ConditionsThe driving pressure for the nebulizers was supplied by a

compressor (DeVilbiss PulmoAide 5610D, DeVilbiss Health Care,Somerset, Pa) at ambient conditions, temperature of 20 to 23.8°C(68 to 75°F) relative humidity of 20 to 54 percent, and an eleva-tion of 1.6 km (5,280 feet). This compressor, when not attachedto a nebulizer, was determined to have a gas flow rate of 12.3standard liters per minute. Nebulizer units were tested in randomorder.

Aerosol CharacterizationThe nebulizers were analyzed continuously with a high-volume

scattering aerosol spectrometry probe (model CSASP-100, PMS,Boulder, Colo). The aerosol was analyzed for particles in the rangeof 0.5 to 32 pm. In this device, particles pass through a laser beamand the light bent or refracted. Larger particles refract light at agreater angle and particles are sized by their angle of refraction.When two particles pass through the laser beam at the same timeor when a particle does not pass completely through the beam, theanalyzer detects these events and excludes these data fromcalculation.The probe was adapted for wet aerosol measurement with a

separation device, employing the same negative pressure, whichdiverted part of the aerosol cloud to preclude overload with co-incidence and oversizing of particles. Both the diverter and thesuction system connected to the probe allowed for a flow rate of16 L/min.The suction accelerates particles through a parabolic horn and

then through the sampling zone containing the laser. Samplingwas performed over three different size ranges sequentially at aninterval of 2 s. This allowed a complete measurement of the en-tire range every 6 s. Each of the three ranges of analysis contained15 channels of increasing incremental size (0.5 to 8.0 ,um at 0.5-pmincrements; 1.0 to 16.0 pm at 1.0-prm increments, and 2.0 to 32.0pum at 2.0-,m increments). There was overlap among ranges withthe 0.5-pum increments being the most precise. A computer pro-gram automatically discarded the values from the channels ineach range that overlapped with a more precise channel. Thisprogram also scaled the values from each range to 1.0-pm incre-ments so that comparisons between ranges could be easily plotted.The program combined the particles from each similar channelfor each time period analyzed and then converted the particlenumber into volumes by the formula V=pie4/3r3. A volume dis-tribution was determined by calculating the percentage of totalvolume in each size channel. From this profile, the percentage oftotal aerosol in the respirable range (1.0 to 5.0 pm) was calculated.

Time End Point

Nebulization was continued until an end point occurred that

was represented by an eightfold drop in total number of particlesread by the laser analyzer. This point correlated with the "sput-tering point" of the nebulizers. 817

Variables Studied1. Total Output (TO) in milliliters: This value was obtained by

subtracting from the prenebulization weight the weight of thenebulizer at the point of eightfold drop in total particles(sputtering). The initial volume fill was 2.5 ml.

2. Time of Total Output (TTO) in Seconds: This value was thetime required with continuous nebulization to reach the eightfolddrop in nebulizer particle output.

3. Percent of Particles in the Respirable Range (PORR): Thisvalue was the mean of the percent of particles in the range of 1to 5 pm during continuous nebulization to the point where thenebulizer output fell eightfold.

4. Respirable Rate (ml/min) (RPDR): This value gives weightto all three variables mentioned above. The TO was divided bythe TTO and that number was multiplied by the PORR to yieldthe delivery rate per minute in the 1- to 5-pum range.

Statistical AnalysisMeasurements of particle size were made in triplicate for each

of the four units of each of the 17 commercially available nebu-lizers. For analysis purposes, the means of the triplicates wereemployed, yielding four observations per nebulizer. For eachoutcome, the means were compared between nebulizers by one-way analysis of variance, and the variances were compared byBarlett's test for homogeneity of variance. If the variances weresignificantly different at the 0.05 level, the means were comparedusing the Welch adjustment to the analysis of variance for unequalvariances within groups. The individual nebulizers were com-pared using Fisher's Protected Least Significance Test at a levelof p<0.05. The results are summarized graphically using dia-mond plots, which are in the means and 95 percent confidenceintervals about the means using a pooled estimate of the variance.All analyses were performed (using JMP Version 2.0.5) on a per-sonal computer (Apple Quadra 700).

RESULTS

The results of the triplicate runs for four units ofeach of 17 models of jet nebulizers are summarizedin Figures 1 through 4.

VolumeThe TO to the point of eightfold decline in parti-

cle output for each of the 17 models of jet nebulizersis shown in Figure 1. Overall, both the means andvariances were significantly different (p<0.0001 foreach). The TO varied from 0.98 ml (Aquatower) to1.86 ml (Sidestream). The TO of the Sidestream (1.86ml) was significantly greater than all the others(p<0.05) and the TO of the Acorn 1 (1.63 ml), Acorn11 (1.61 ml), and the Updraft 11 (1.62 ml) were sig-nificantly greater than the remainder (p<0.05). TheTO from individual units of the AquaTower greatlyvaried.

TimeThe time to the point of eightfold drop in particle

output is shown in Figure 2. There was a highly sig-nificant difference among the means and variances

CHEST / 106 / 6 / DECEMBER, 1994 1789

 1994, by the American College of Chest Physicians by guest on May 7, 2010chestjournal.chestpubs.orgDownloaded from

20 -

15-

5-

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Nebulizer

Summary StatisticsNebulizer Name Mean Std Dev1 Acorn-t 1.63 0.082 Acorn-It 1.61 0.043 AquaTower 0.98 0.454 AVA-Neb 1.30 0.095 Cirrhus 1.20 0.126 Dart 1.34 0.107 DeVilbiss 646 1.28 0.208 Downdraft 1.08 0.099 Fan Jet 1.28 0.1410 MB-5 1.20 0.0511 Misty Neb 1.31 0.0912 PARI-JET 1.15 0.1213 PARI LCJET 1.16 0.0914 Salter 8900 1.23 0.0715 Sidestream 1.86 0.0416 Updraft-tt 1.62 0.0417 Whisper Jet 1.04 0.04

FIGURE 1. Total volume output to the point of eightfold declinein particle output. Initial fill for all nebulizers was 2.5 ml.Diamonds indicate 95 percent confidence intervals about themean (horizontal line in the diamond) based on pooled variances.Squares indicate the means of triplicate determinations with thefour units of that model. Names, means, and SDs of the modelsby number are given in the table below the figure. Overall dif-ferences between the means and variances were significant(p<O.OOOl).

within the group (p<0.0001 for each). The TTOvaried from 2.28 min (Downdraft) to 20.95 min(Acorn-II). The Downdraft (2.28 min), PARI-JET(2.39 min), AquaTower (2.47 min), and PARI LCJET (2.54 min) had total nebulization times that wereshorter than those of the remaining models (p<0.05).

Percent of Particles in Respirable RangeThe PORR is shown in Figure 3. There was a

highly significant difference in the means and vari-ances among the devices (p<.0001 for means,

p<0.005 for variances). The PORR varied from 21.89percent (Dart) to 71.94 percent (Sidestream). ThePORR of the Sidestream (71.94 percent) was signif-icantly greater than all others (p<0.05). The PORRof the PARI-JET (52.49 percent) and the PARI LCJET (52.28 percent) had a greater percent of particlesin the respirable range than all others except theAquaTower (p<0.05). Different units of the Aqua-Tower were highly variable in this regard.

Summary StatisticsNebulizer Name Mean Std Dev1 Acorn-I 11.72 1.622 Acorn-It 20.95 1.683 AquaTower 2.47 2.394 AVA-Neb 11.48 1.515 Cirrhus 8.77 1.816 Dart 3.37 0.647 DeVilbiss 646 3.34 1.148 Downdraft 2.28 0.389 Fan Jet 10.25 1.0010 MB-5 6.90 1.3711 Misty Neb 8.23 1.1412 PARI-JET 2.39 0.4213 PARI LC JET 2.54 0.2714 Salter 8900 6.92 0.7515 Sidestream 7.14 0.4716 Updraft-11 11.94 0.7117 Whisper Jet 8.82 1.22

FIGURE 2. Total time of continuous nebulization to the point ofeightfold decline in particle output. Diamonds indicate 95percent confidence intervals about the mean (horizontal line inthe diamond) based on pooled variances. Squares indicate themeans of triplicate determinations with the four units of thatmodel. Names, means, and SDs of the models by number aregiven in the table below the figure. Overall differences betweenthe means and variances were significant (p<0.0001).

Respiratory Particle Delivery Rate

The RPDR is shown in Figure 4. There was ahighly significant difference in the means and vari-ances among the groups (p<.0001 for means, p<0.005for variances. The RPDR varied from 0.03 ml/min(Acorn-II) to 0.28 ml/min (AquaTower). The RPDRfor the PARI-JET (0.26 ml/min) and PARI LC JET(0.24 ml/min) did not differ significantly from theAquaTower (0.29ml/min) and the Sidestream (0.19ml/min). However, the output among units of theAquatower was highly variable. The RPDR for theSidestream did not differ significantly from the Aqua-Tower, PARI-JET, and PARI LC JET or from theDeVilbiss or Downdraft. The remaining models allwere different at the p<0.05 level.

Volume Mean Diameter

The volume mean diameter (VMD) represents themidpoint in microns where 50 percent of the parti-cles are smaller and 50 percent are larger than theVMD value and is comparable to the mass medianaerodynamic diameter (MMAD). The VMD varied

A Comparison of Commercial Jet Nebulizers (Loffert, Ikle, Nelson)

6z

Nebulizer

()v-

1790

 1994, by the American College of Chest Physicians by guest on May 7, 2010chestjournal.chestpubs.orgDownloaded from

Nebulizer

Summary StatisticsNebulizer Name Mean Std Dev1 Acorn-t 42.10 5.692 Acorn-t1 44.54 4.763 AquaTower 46.24 15.204 AVA-Neb 40.04 2.475 Cirrhus 35.55 2.126 Dart 21.89 4.257 DeVilbiss 646 35.67 6.908 Downdraft 27.45 3.799 Fan Jet 40.80 4.6910 MB-5 32.01 1.8211 Misty Neb 35.25 2.2712 PARI-JET 52.49 3.7713 PARI LC JET 52.28 1.6914 Salter 8900 35.84 1.9815 Sidestream 71.95 2.7316 Updraft-It 40.16 4.4817 Whisper Jet 37.18 3.06

FIGURE 3. Percent of particle output in the respirable range (1.0to 5.0 ,m). Diamonds indicate 95 percent confidence intervalsabout the mean (horizontal line) based on pooled variances.Squares indicate the means of triplicate determinations with thefour units of that model. Names, means, and SDs of the modelsby number are given in the table below the figure. Overall dif-ferences between the means and variances were significant atp<0.0001 and p<0.005, respectively.

from 3.77 rtm (Sidestream) to 7.20 ,um (Dart). Thesmallest VMDs were Sidestream (3.76 Am), PARI LCJET (4.38 ,um), and PARI-JET (4.84 Mrm); each wassignificantly different from the others and from allthe remaining models (p<0.05). The extreme vari-ability of the Aquatower precluded meaningfulcomparison.

DISCUSSION

In this study, we have demonstrated a highly sig-nificant variation in the output characteristics of 17different reusable nebulizers, commercially availablein the United States and intended for delivery ofaerosol medication to the lower respiratory tract. Thisstudy represents an extension of previous studies thathave compared a smaller number of nebulizers, notall of which were available in the United States.5'4We have confirmed the reported variability in rateand extent of nebulization as well as the percent ofthe aerosol output in the respirable range (<5 Mm).We have also combined these variables into a singleparameter: the rate of delivery of respirable particles,which facilitates comparison of nebulizer models.

621

c-cc:La)Ia:

/ VAv LA \,'' I'l ''

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Nebulizer

Summary StatisticsNebulzer Name Mean Std Dev1 Acom-I 0.0594 0.00192 Acorn-I1 0.0344 0.00333 AquaTower 0.2860 0.19414 AVA-Neb 0.0452 0.00435 Cirrhus 0.0500 0.00656 Dart 0.0894 0.00837 DeVilbiss 646 0.1469 0.04178 Downdraft 0.1304 0.00499 Fan Jet 0.0507 0.003110 MB-5 0.0573 0.009411 Misty Neb 0.0570 0.001912 PARI-JET 0.2592 0.010313 PARI LCJET 0.2397 0.003014 Salter 8900 0.0642 0.001615 Sidestream 0.1905 0.017516 Updraft-11 0.0547 0.002417 Whfsper Jet 0.0446 0.0040

FIGURE 4. Delivery of particles in the respiratory range in milli-liters per minute. Diamonds indicate 95 percent confidenceintervals about the mean (horizontal line) based on pooled vari-ances. Squares indicate the means of triplicate determinationswith the four units of that model. Names, means, and SDs of themodels by number are given in the table below the figure. Over-all differences between the means and variances were significantat p<0.0001 and p<0.005, respectively.

This expression of nebulizer performance also variedmarkedly among the nebulizers studied.

Jet nebulizers are commonly used for the deliveryof bronchodilators, antimicrobial agents, and forbronchial challenge studies. The desirable character-istics for a nebulizer may vary among these differentindications.10'3 For treatment of acute bronchos-pasm or for economy of patient time, a high rate ofdelivery of bronchodilator medication is desirable.6For delivery of an expensive medication, a nebulizerthat delivers the greatest percent of medication fromthe nebulizer bowl may be most desirable, even if thedelivery rate is not maximal. For diagnostic and re-search challenge studies, the greatest possible repro-ducibility between different units of a particularmodel may be most important. The results of thisstudy suggest that some available nebulizers are notoptimal for any of these indications.

In this study, we only examined the delivery ofaerosol particles. We did not examine the delivery ofdrug. It has been shown that albuterol and othermedications are delivered to a lesser degree than thediluent, particularly as nebulization continues.17 This

CHEST / 106/ 6 / DECEMBER, 1994 1791

0

cn

aK:

 1994, by the American College of Chest Physicians by guest on May 7, 2010chestjournal.chestpubs.orgDownloaded from

is presumably due to the increasing contribution ofevaporation to the loss of diluent from the nebulizerbowl. While analysis for drug delivery might havevaried somewhat from that of particles, the principlesof evaporative loss of diluent have been shown toapply to nebulizers of various manufacturers. It isunlikely that this further analysis would have greatlyaltered the outcome of the study.

In summary, commercially available jet nebuliz-ers, all with the same initial fill, and all driven by thesame air compressor, varied greatly in the totaldelivery of fluid, the time required to deliver thisfluid, and the percent of the delivered fluid in therespirable range. This resulted in a variability of to-tal output of twofold, of the time required for neb-ulization of ninefold, and of drug delivery in therespiratory range per minute of eightfold. Theseperformance characteristics as well as the reproduc-ibility from unit to unit should be considered in se-lection of a jet nebulizer.14

REFERENCES1 Miller WC, Mason JW. Small-volume nebulizers versus me-

tered-dose inhalers [letter]. Chest 1993; 103:655-562 Kisch GL, Paloucek FP. Metered-dose inbalers and nebulizers

in the acute setting. Ann Pharmacother 1992; 26:92-53 Newhouse MT. Metered dose pressurized aerosols and the

ozone layer. Eur Respir J 1990; 3:1232-344 Newhouse MT. Emergency department management of life-

threatening asthma: are nebulizers obsolete? Chest 1993; 103:661-63

5 Kradjan WA, Lakshminarayan S. Efficiency of air compressor-driven nebulizers. Chest 1985; 87:512-16

6 Langford SA, Allen MD. Salbutamol output from two jet neb-ulizers. Respir Med 1993; 87:99-103

7 Hardy JG, Newman SP, Knoch M. Lung deposition from fournebulizers. Respir Med 1993; 87:461-65

8 O'Callaghan C, Clark AR, Milner AD. Why nebulise for morethan five minutes? Arch Dis Child 1989; 64:1270-73

9 Hess D, Horney D, Snyder T. Medication-delivery perfor-mance of eight small-volume, hand-held nebulizers: effects ofdiluent volume, gas flowrate, and nebulizer model. Respir Care1989; 34:717-23

10 Newman SP, Pellow PGD, Clay MM, Clarke SW. Evaluationof jet nebulizers for use with gentamicin solution. Thorax 1985;40:671-76

11 Clay MM, Pavia D, Newman SP, Clarke SW. Factors influ-encing the size distribution of aerosols from jet nebulisers.Thorax 1983; 38:755-59

12 Ferron GA, Kerrebijn KF, Weber J. Properties of aerosols pro-duced with three nebulizers. Am Rev Respir Dis 1976; 114:899-908

13 O'Doherty MJ, Thomas S, Page C, Barlow D, Bradbeer C,Nuhan TO, et al. Differences in relative efficiency of nebulis-ers for pentamidine administration. Lancet 1988; 2:1283-86

14 Alvine GF, Rodgers P, Fitzsimmons KM, Ahrens RC. Dispos-able jetnebulizers: how reliablearethey? Chest 1992; 101:316-19

15 Hollie MC, Malone RA, Skufca RM, Nelson HS. Extreme vari-ability in aerosol output of the DeVilbiss 646 jet nebulizer. Chest1991; 100:1339-44

16 Morrow PE. Aerosol characterization and deposition. Am RevRespir Dis 1974; 110:88-9

17 Malone RA, Hollie MC, Glynn-Barnhart A, Nelson HS. Opti-mal duration of nebulized albuterol therapy. Chest 1993;104:1114-18

A Comparison of Commercial Jet Nebulizers (Loffert, Ikle, Nelson)1792

 1994, by the American College of Chest Physicians by guest on May 7, 2010chestjournal.chestpubs.orgDownloaded from

DOI 10.1378/chest.106.6.1788 1994;106; 1788-1792Chest

D. Todd Loffert, David Ikle and Harold S. NelsonA Comparison of Commercial Jet Nebulizers

 May 7, 2010This information is current as of

 

& ServicesUpdated Information

8http://chestjournal.chestpubs.org/content/106/6/178Updated Information and services can be found at:

Citations

8#related-urlshttp://chestjournal.chestpubs.org/content/106/6/178articles:This article has been cited by 7 HighWire-hosted

Permissions & Licensing

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

Reprints http://www.chestpubs.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

Citation Alerts

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

formatImages in PowerPoint

format. See any online figure for directions. downloaded for teaching purposes in PowerPoint slide

articles can beCHESTFigures that appear in

 1994, by the American College of Chest Physicians by guest on May 7, 2010chestjournal.chestpubs.orgDownloaded from