PULMONARY HEMODYNAMICS

An Alternative Echocardiographic Method to EstimateMean Pulmonary Artery Pressure: Diagnostic and

Clinical Implications

Javier F. Aduen, MD, Ramon Castello, MD, Marcelo M. Lozano, MD, George N. Hepler, RDCS,Cesar A. Keller, MD, Francisco Alvarez, MD, Robert E. Safford, MD, Julia E. Crook, PhD,

Michael G. Heckman, MS, and Charles D. Burger, MD, Jacksonville, Florida

Background: The aim of this study was to evaluated an alternative echocardiographic method to calculatemean pulmonary arterial pressure (MPAP).

Methods: One hundred two patients were studied with simultaneous right-heart catheterization (RHC) andechocardiography. MPAP was calculated by adding the right ventricular–right atrial mean systolic gradientto right atrial pressure.

Results: The mean difference between MPAP calculated using this method and RHC-derived MPAP was�1.6mm Hg, less than that of traditional systolic pulmonary arterial pressure (SPAP; �6.4 mm Hg) and MPAP es-timated using the pulmonary regurgitation method (�13.9 mm Hg). The median absolute percentage differ-ence of the MPAP calculations relative to RHC was significantly less with this method than with thepulmonary regurgitation method (18% vs 71%; P < .001) and similar to the SPAP method (both 18%; P = .30).

Conclusion: MPAP calculated using the proposed method is as accurate as SPAP calculation and less variablethan previous methods, thus allowing widespread clinical use. (J Am Soc Echocardiogr 2009;22:814-819.)

Keywords: Doppler echocardiography, Mean pulmonary pressure, Pulmonary hypertension

Pulmonary hypertension (PH) is a hemodynamic abnormality that re-sults from a wide array of clinical conditions. The accurate measure-ment of mean pulmonary arterial pressure (MPAP) is essential forthe diagnosis of PH and is preferred to the measurement of systolicpulmonary arterial pressure (SPAP) to assess the severity and progres-sion of PH.1,2

Echocardiography, because of its noninvasive nature, is commonlyused to estimate pulmonary pressures. Its reproducibility and reliabil-ity to estimate SPAP have been well established.3-9 Right-heart cath-eterization (RHC) is the technique of choice and the gold standardfor the diagnosis and evaluation of patients with PH, because it pro-vides an accurate measurement of pulmonary pressures. However,its invasive nature makes it unsuitable for frequent, repeated use.

Different echocardiographic methods have been described to esti-mate MPAP. Systolic time intervals obtained from spectral Doppler ofthe pulmonary artery flow have shown that the time to peak pulmo-nary velocity has a logarithmic relationship with MPAP. The ratio of

From the Division of Pulmonary Medicine (J.F.A., C.A.K., F.A., C.D.B.), the Division

of Cardiovascular Diseases (R.C., G.N.H., M.M.L., R.E.S.), and the Biostatistics

Unit (J.E.C., M.G.H.), Mayo Clinic, Jacksonville, Florida.

This paper was presented in part at the American Thoracic Society International

Conference, San Francisco, CA, May 18-23, 2007.

Reprint requests: Ramon Castello, MD, Apex Cardiovascular Group, Florida Phy-

sicians Research Associates, 6428 Beach Boulevard, Jacksonville, FL 32216

(E-mail: castello.ramon@gmail.com).

0894-7317/$36.00

Copyright 2009 by the American Society of Echocardiography.

doi:10.1016/j.echo.2009.04.007

814

the time to peak pulmonary velocity to ejection time can also beused to estimate MPAP.10 These methods are tedious, and becauseof the small range of values of the systolic time intervals and their de-pendence on cardiac output and heart rate, their reproducibility islow, and correlation with invasively measured MPAP is suboptimal.11

Continuous-wave Doppler can be used to measure the pulmonaryregurgitation (PR) velocity and thus to calculate the pulmonaryartery–right ventricle diastolic gradient, which correlates well withMPAP.12,13 However, obtaining an adequate PR spectral Doppler sig-nal may be less feasible than obtaining a tricuspid regurgitation (TR)jet.14 Thus, MPAP is not routinely reported in most echocardiographylaboratories, which is a serious clinical drawback for clinicians manag-ing patients with PH.

We have developed an alternative echocardiographic method tocalculate MPAP, on the basis of a modification of the traditional echo-cardiographic method to calculate SPAP. The method is based on theaddition of right atrial (RA) pressure to the right ventricular (RV)–RAmean (instead of peak) systolic gradient. We hypothesized that thismethod, because of its closer relation to SPAP, would correlate wellwith MPAP. We retrospectively tested the hypothesis in a large groupof patients and found a rather close correlation.15 Thus, this study wasdesigned to prospectively validate this alternative method, comparingit with simultaneous RHC measurements.

METHODS

Patients

This was a prospective study in which we evaluated 117 patients whowere referred for RHC between May 2005 and June 2006 for the

Journal of the American Society of EchocardiographyVolume 22 Number 7

Aduen et al 815

evaluation of pulmonary pressure. All patients were considered eligi-ble for this study regardless of the indication for RHC. Demographicand clinical data were collected for each patient at study entry. Thestudy was approved by the Mayo Clinic Institutional Review Board,and all patients provided informed consent. The echocardiographicstudies were performed after the pulmonary artery catheters hadbeen appropriately positioned and calibrated.

Echocardiographic Examination

A Sonos 5500 (Philips Medical Systems, Andover, MA) or an Acu-son Sequoia 256 (Siemens Medical Solutions USA, Inc, MountainView, CA) ultrasound machine was used to perform the echocar-diographic examinations. A goal-directed 2-dimensional, pulsed,continuous, and color-flow Doppler examination was performed.The patient was placed in the decubitus position. RA pressurewas estimated echocardiographically on the basis of inferior venacava respiratory dynamics, as previously described.16 Color-flowDoppler was used to align the continuous-wave Doppler cursorwith the TR jet and to semiquantitate the severity of TR. A noni-maging continuous-wave Doppler transducer was occasionallyused when necessary.

All patients received an injection of agitated saline at the end of thestudy to enhance the TR jet signal. TR tracings were obtained in everypatient, both without saline and following the injection of agitatedsaline. Therefore, every patient had 2 sets of Doppler measurements.Saline injections were performed at the end of the study in a conven-tional form. Saline was agitated between 2 syringes and a stopcock.The bacteriostatic saline was mixed with 0.5 mL of air. Injectionwas performed through the proximal port of the pulmonary arterycatheter, which produces a strong opacification of the right atrium.Special care was taken to obtain continuous-wave Doppler signals af-ter most of the RA saline opacification had disappeared, to avoid or atleast minimize the classic linear artifacts induced by saline injection.All images were stored as 3-beat digital loops using the ProSolv sys-tem (ProSolv CardioVascular, Indianapolis, IN).

Adequate subcostal views for the estimation of RA pressure wereobtained by increasing the loop duration to 15 beats, enough to cap-ture respiratory changes. Optimal images could not be obtained in 15patients, so after their exclusion, the actual sample size was 102patients. Among these 15 patients, 7 had undergone surgery, includ-ing lung, heart, and liver transplantation.

Figure 1 Mean systolic RV-RA gradient calculated by tracingthe TR time-velocity integral.

Doppler Measurements

Measurements were performed offline from the ProSolv images, forwhich calibration is automatic. All measurements were performedby the same investigator (M.M.L.). This individual did not performthe echocardiographic studies, was unaware of the patient’s medicalhistories, and had no access to the RHC data. The highest TR velocityobtained from a sinus beat from 1 of 3 acoustic windows was mea-sured and traced to obtain the peak and mean systolic RV-RA gradi-ent. The mean gradient was calculated by tracing the TR time-velocityintegral (Figure 1). The cleanest signals with the least amount of noiseartifact were traced. The pulmonary artery catheter did not seem tocreate notable artifactual effects.

MPAP was calculated by adding the echocardiographically esti-mated RA pressure to the RV-RA mean systolic gradient. MPAP usingthe PR velocity method and SPAP were estimated as previously re-ported.12,13 For both pressures, the final values were obtained by add-ing the estimated RA pressures to the calculated gradients. Tracingsobtained before and after the injection of agitated saline were mea-sured in an identical manner.

To assess interobserver variability, the RV-RA mean systolic gradi-ent was measured independently by 2 different investigators(M.M.L., G.N.H.) in 32 of the 102 echocardiograms, 6 months apart.To assess intraobserver variability, one of the investigators (G.N.H.)remeasured the RV-RA mean systolic gradient in 20 patients 2months apart.

RHC

A 7Fr Swan-Ganz pulmonary artery catheter was placed, either in theintensive care unit or in a dedicated procedure room located in theheart and lung transplantation ward, following our institution’s estab-lished clinical protocol. Immediately after the echocardiographicstudy was done, the patient was positioned in either a supine ora Fowler position, and the pressure transducer was zeroed and leveledat the phlebostatic axis at the level of the mid right chest. SPAP, dia-stolic pulmonary arterial pressure (DPAP), and RA pressure wereobtained from an average of several beats obtained precisely duringend-expiratory time (with paper recording of the tracings). MPAPwas measured using the formula: SPAP � DPAP/3 + DPAP. RHChemodynamic data were collected independently by an individualunaware of the echocardiographic results.

Statistical Analysis

The mean and standard deviation (SD) for the difference between in-vasively and echocardiographically obtained pressures and the 95%confidence intervals (CIs) of those differences were calculated. TheSDs of the differences and the median absolute differences werealso determined. To account for the tendency of higher measures tohave more variation, the proposed method of calculating MPAPwas compared with other methods of determining MPAP or SPAP us-ing Wilcoxon’s signed-rank test. The absolute percentage differenceswere calculated by taking the absolute value of the differencebetween pressures divided by the invasively obtained pressures andexpressed as a percentage.

The lower and upper limits of agreement were estimated, as de-scribed by Bland and Altman,17 as the mean 6 2 SDs, with 95%CIs. Interpretation of limits of agreement assumed a normal distribu-tion of the differences between echocardiographically and invasivelyobtained pressures. Caution is warranted when comparing limits ofagreement for MPAP and SPAP. Because of the higher variabilityassociated with higher SPAP measurements, it is expected a priori

816 Aduen et al Journal of the American Society of EchocardiographyJuly 2009

that limits of agreement will be wider for SPAP than for MPAP. Re-ceiver operating characteristic curves were created to assess the diag-nostic utility of the two echocardiographically obtained methods ofcalculating MPAP, with PH defined as MPAP > 25 mm Hg fromRHC. The area under the curve (AUC) was estimated and comparedbetween methods using a c2 test. S-Plus version 8.0.1 (Insightful Cor-poration, Seattle, WA) was used for the statistical analysis.

RESULTS

The demographic characteristics and referral diagnoses of the patientsand the pulmonary and RA pressures obtained by RHC and echocar-diographic and Doppler studies with and without saline contrast areshown in Table 1. The Doppler TR jets were not analyzable and couldnot be used to estimate the pulmonary pressures in 7 of the 102 pa-tients (7%) without saline contrast and in 2 patients (2%) with salinecontrast. In turn, the Doppler PR jets were not analyzable in 23patients (23%) without saline contrast and in 17 patients (17%)with saline contrast.

The median absolute difference in the RV-RA mean systolic gradi-ent between two observers was 3 mm Hg (range, 0-12 mm Hg) andwas < 5 mm Hg in 77 of 102 patients (75%). For the intraobserver var-iability, the median absolute difference was 2 mm Hg (range, 0-9 mmHg) and was <4 mm Hg in 77 of 102 patients (75%).

Table 2 summarizes the differences between echocardiographicand invasively obtained pulmonary arterial pressures, along with theestimated limits of agreement and CIs with and without saline con-trast. Figures 2 to 4 show Bland-Altman plots17 of echocardiographicand invasively obtained pulmonary arterial pressures.

Table 1 Characteristics of 102 patients

Characteristic Value

Demographics

Men 54 (53%)

Age (y) 60.1 6 12.3

Body mass index (kg/m2) 27.8 6 6.4RHC referral diagnosis

Pulmonary arterial hypertension 25 (24%)

Chronic obstructive pulmonary disease 20 (20%)Pulmonary fibrosis 9 (9%)

End-stage liver disease with

portopulmonary

hypertension

14 (14%)

Postoperative evaluation 12 (12%)

Congestive heart failure 17 (17%)

Other 20 (20%)

RHC (mm Hg)MPAP 34.3 6 14.5

SPAP 53.0 6 23.7

RA pressure 9.3 6 5.3Echocardiography (mm Hg)

MPAP by RV-RA mean systolic gradient

without saline

33.6 6 14.9

MPAP by PR velocity without saline 21.7 6 13.0SPAP without saline 48.3 6 24.4

MPAP by RV-RA mean systolic gradient

with saline

40.9 6 18.2

MPAP by PR velocity with saline 29.2 6 13.5SPAP with saline 55.4 6 25.6

RA pressure 6.0 6 3.7

Data are expressed as number (percentage) or mean 6 SD.

The mean difference between the calculated MPAP measured byRV-RA mean systolic gradient without saline contrast and the RHC-measured MPAP was �1.6 mm Hg, with limits of agreement rangingfrom �16.9 to 13.7 mm Hg. The median absolute percentage differ-ence between echocardiographic (without saline) and RHC valueswas significantly smaller (P < .001) for the calculated MPAP (18%)compared with the estimated MPAP obtained using the PR method(71%) and virtually identical to the traditional SPAP (18%) (P = .30;Table 2).

The addition of saline contrast resulted in higher estimated pulmo-nary pressure values when the peak systolic velocity was used, thus im-proving SPAP estimation and MPAP estimation using the PR velocitymethod. Conversely, saline use did not improve agreement whenmean velocity methods were used, as in the alternative method.

The estimated AUC was higher (P = .001) for MPAP measured bythe RV-RA mean systolic gradient without saline (AUC, 0.92; 95% CI,0.87-0.97) than for MPAP obtained by the PR method without saline(AUC, 0.75; 95% CI, 0.65-0.86). Results were similar with the addi-tion of saline contrast, with respective estimates of AUCs of 0.90 and0.82, although the difference between the two was not statisticallysignificant (P = .31; Figure 5).

DISCUSSION

Our study shows satisfactory limits of agreement and correlation be-tween the echocardiographically and Doppler-calculated MPAP usingthe RV-RA mean systolic gradient and the RHC-measured MPAP.MPAP calculated using this alternative method is very comparablein reliability and accuracy with the traditional estimation of SPAP.Both methods have a median absolute percentage difference betweenechocardiographically and invasively obtained pressures of approxi-mately 18%, although the calculated MPAP has a lower bias (�1.6mm Hg) than SPAP (�6.4 mm Hg). The calculation of MPAP withour proposed method also appears more feasible and accurate than es-timating MPAP using the PR method, with better accuracy in diagnos-ing PH, as indicated by a higher AUC. In our experience, the additionof saline contrast improved the accuracy of peak systolic velocity–based methods but not that of mean systolic velocity–based methods.

In an elegant smaller study by Abbas et al,13 26 patients with echo-cardiographic and Doppler studies and invasive RHC measurementsperformed within 45 minutes of each other were evaluated. The au-thors compared MPAP estimated by the PR method and MPAP mea-sured by RHC and reported a mean difference of 0.8 6 5.6 mm Hgand limits of agreement of�10 to 12 mm Hg. In their study, saline con-trast was used as needed, whereas we used it in every patient to betterassess the additional value of using saline. In our larger simultaneousstudy, we also obtained a good correlation for MPAP obtained usingthe PR method, but the correlation was not as good as that reportedby Abbas et al.13 Furthermore, in agreement with Tramarin et al,14

we found it more feasible to obtain an adequate TR jet spectral Dopp-ler signal than a PR jet signal (94% vs 80% without saline contrast and98% vs 85% with saline contrast), thus increasing the applicability ofthe novel proposed method. Improved ultrasound technology, alongwith the routine use of 3 acoustic windows and the nonimaging trans-ducer, may explain our good results compared with those of previousstudies.18-21 Consistent with other reports,22,23 the addition of salinecontrast did improve the accuracy of the peak velocity–basedmethods such as the traditional SPAP estimation and that of the PRmethod to estimate MPAP. However, in our experience, saline wasnot helpful if a mean velocity–based method was used.

Journal of the American Society of EchocardiographyVolume 22 Number 7

Aduen et al 817

Table 2 Differences and limits of agreement between invasively and echocardiographically obtained pulmonary pressures

Echocardiographic estimate

minus RHC measurement

of MPAP and SPAP (mm Hg)

Echocardiographic estimate Mean difference (95% CI) SD of differences

Lower limit

of agreement (95% CI)

Upper limit

of agreement (95% CI)

Median absolute

percentage difference

MPAP calculatedby RV-RA mean

systolic gradient

Without saline �1.6 (�3.1 to �0.1) 7.7 �16.9 (�19.6 to �14.3) 13.7 (11.1 to 16.4) 18%

With saline 6.1 (4.0 to 8.1) 10.2 �14.3 (�17.9 to �10.7) 26.5 (22.8 to 30.1) 21%MPAP estimated

by PR velocity

Without saline �13.9 (�16.3 to 11.4) 11.6 �37.0 (�41.4 to �32.7) 9.3 (5.0 to 13.6) 71%*With saline �5.6 (�7.8 to �3.4) 10.6 �26.8 (�30.6 to �22.9) 15.6 (11.7 to 19.4) 27%*

SPAP

Without saline �6.4 (�8.6 to �4.2) 11.1 �28.5 (�32.3 to �24.7) 15.8 (11.9 to 19.6) 18%

With saline 1.5 (�1.2 to 4.3) 13.6 �25.7 (�30.5 to �20.8) 28.7 (23.9 to 33.5) 17%

*P < .05 vs MPAP by RV-RA mean systolic gradient without saline (Wilcoxon’s signed-rank test). The median absolute percentage difference was cal-

culated by taking the absolute value of the difference between the echocardiographic and RHC measures and dividing by the RHC measure.

Figure 2 MPAP by RHC and MPAP by echocardiography andDoppler using mean systolic RV-RA gradient, without (A) andwith (B) saline contrast.

Figure 3 MPAP by RHC and MPAP by echocardiography andDoppler using PR velocity, without (A) and with (B) saline con-trast.

When comparing two measurement techniques, the correlationcoefficient measures the strength of the relation between the twovariables, not the agreement between them.17 Yock and Popp3 re-ported a strong correlation (r = 0.93) between SPAP estimated by

Doppler and SPAP measured by RHC. Other studies4-9 have con-firmed that strong correlation. However, the level of agreementhas not been evaluated in most studies. Kim et al,8 who evaluated74 liver transplantation candidates, reported a mean difference of

818 Aduen et al Journal of the American Society of EchocardiographyJuly 2009

2.8 mm Hg between SPAPs, which is similar to our results but withan SD of 16.1 mm Hg, indicating a rather large dispersion in theirSPAP estimation. Denton et al,9 who evaluated 33 patients with sys-temic sclerosis, reported a mean absolute difference of 11.4 mm Hgbetween Doppler-estimated SPAP and SPAP estimated by RHC.Our mean absolute difference between Doppler-estimated andRHC-estimated SPAP was 7.5 mm Hg with saline and 8.5 mm Hgwithout saline contrast.

Most echocardiographic laboratories routinely report SPAP be-cause of its high reliability, reproducibility, and ease of acquisition.Previously described MPAP methods are perceived as tedious andtherefore have not gained widespread utilization. However, the useof MPAP is clinically advantageous over SPAP, because it has betterdiagnostic and prognostic value and because it is used to determinepulmonary vascular reactivity.1,2,21 It is currently accepted that al-though the Doppler estimates of SPAP correlate well with hemody-namic evaluations, only RHC can confirm PH diagnosis and assessprecise disease hemodynamics.24

In the past decade, several vasodilators capable of reducing pulmo-nary arterial pressure have become available for clinical use. Studiesusing these drugs have demonstrated an improvement in functionalstatus, hemodynamics, and survival of patients with PH, particularlythose with idiopathic pulmonary arterial hypertension.25-29 For thosepatients, an accurate method to determine MPAP noninvasively thatallows serial follow-up to assess their responses to therapy is of criticalimportance.

Ours is the largest study to date to prospectively compare invasiveand noninvasive calculation of MPAP in a simultaneous fashion. Thismethod, based on the mean RV-RA systolic gradient, has proven to be

Figure 4 SPAP by RHC and SPAP by echocardiography andDoppler without (A) and with (B) saline contrast.

accurate and applicable in most patients. Importantly, it is easy to useand represents only a minor variation (mean vs peak TR velocity) ofthe traditional echocardiographic SPAP calculation, which is obtainedin most echocardiographic laboratories and is required for laboratoryaccreditation. The transition to reporting MPAP in addition to SPAPhas been simple in our laboratory, with a notable clinical acceptanceon the part of the pulmonologists. We envision that the method canbe easily incorporated into most current echocardiographic protocols,given the similarities to the customarily obtained echocardiographicparameters.

In summary, the diagnosis of PH is best defined by elevation ofMPAP. We present here an alternative echocardiographic approachto calculate MPAP by adding the RV-RA mean systolic gradient toRA pressure. We found that this alternative method was at least com-parable with the traditional estimation of the SPAP.

Study Limitations

The method to estimate SPAP by adding RA pressure to the peak sys-tolic RV-RA gradient has a strong physiologic basis. The PR method toestimate MPAP using the peak early diastolic gradient relates closelyto DPAP, which is an integral component of the formula used to cal-culate MPAP. The relation between MPAP and its dependence onSPAP and DPAP have been challenged recently. It has been tradition-ally admitted that MPAP is twice as sensitive to DPAP as it is to SPAP.Recent studies, however, favor the use of SPAP only to estimateMPAP, without the need to include DPAP in the model. They haveshown a very linear strong correlation between SPAP and MPAP,30

so it has been suggested that both measurements may provide almostredundant clinical information.

Our study seems to confirm the same observation. The currentmethod is based on a variation of the traditional method, using themean instead of the peak gradient. The close relation with MPAP isa reflection of the close relation to SPAP. Methods that are basedmore on SPAP and DPAP may not be as robust and may be, bydefinition, more complex.31

Figure 5 Receiver operating characteristic curves measuringthe utility of diagnosing pulmonary hypertension by estimatingMPAP using either the RV-RA mean systolic gradient or thePR velocity method.

Journal of the American Society of EchocardiographyVolume 22 Number 7

Aduen et al 819

REFERENCES

1. McGoon M, Gutterman D, Steen V, Barst R, McCrory DC, Fortin TA, et al;American College of Chest Physicians. Screening, early detection, and di-agnosis of pulmonary arterial hypertension: ACCP evidence-based clinicalpractice guidelines. Chest 2004;126(suppl). 14S-34.

2. Barst RJ, McGoon M, Torbicki A, Sitbon O, Krowka MJ, Olschewski H,et al. Diagnosis and differential assessment of pulmonary arterial hyperten-sion. J Am Coll Cardiol 2004;43(suppl). 40S-7S.

3. Yock PG, Popp RL. Noninvasive estimation of right ventricular systolicpressure by Doppler ultrasound in patients with tricuspid regurgitation.Circulation 1984;70:657-62.

4. Chan KL, Currie PJ, Seward JB, Hagler DJ, Mair DD, Tajik AJ. Comparisonof three Doppler ultrasound methods in the prediction of pulmonaryartery pressure. J Am Coll Cardiol 1987;9:549-54.

5. Currie PJ, Seward JB, Chan KL, Fyfe DA, Hagler DJ, Mair DD, et al. Con-tinuous wave Doppler determination of right ventricular pressure: a simul-taneous Doppler-catheterization study in 127 patients. J Am Coll Cardiol1985;6:750-6.

6. Bidart CM, Abbas AE, Parish JM, Chaliki HP, Moreno CA, Lester SJ. Thenoninvasive evaluation of exercise-induced changes in pulmonary arterypressure and pulmonary vascular resistance. J Am Soc Echocardiogr2007;20:270-5.

7. Shapiro SM, Oudiz RJ, Cao T, Romano MA, Beckmann XJ, Georgiou D,et al. Primary pulmonary hypertension: improved long-term effects andsurvival with continuous intravenous epoprostenol infusion. J Am CollCardiol 1997;30:343-9.

8. Kim WR, Krowka MJ, Plevak DJ, Lee J, Rettke SR, Frantz RP, et al. Accuracyof Doppler echocardiography in the assessment of pulmonary hyperten-sion in liver transplant candidates. Liver Transplant 2000;6:453-8.

9. Denton CP, Cailes JB, Phillips GD, Wells AU, Black CM, Bois RM. Com-parison of Doppler echocardiography and right heart catheterization to as-sess pulmonary hypertension in systemic sclerosis. Br J Rheumatol 1997;36:239-43.

10. Kitabatake A, Inoue M, Asao M, Masuyama T, Tanouchi J, Morita T, et al.Noninvasive evaluation of pulmonary hypertension by a pulsed Dopplertechnique. Circulation 1983;68:302-9.

11. Posteraro A, Salustri A, Trambaiolo P, Amici E, Gambelli G. Echocardio-graphic estimation of pulmonary pressures. J Cardiovasc Med (Hagers-town) 2006;7:545-54.

12. Masuyama T, Kodama K, Kitabatake A, Sato H, Nanto S, Inoue M. Con-tinuous-wave Doppler echocardiographic detection of pulmonary regurgi-tation and its application to noninvasive estimation of pulmonary arterypressure. Circulation 1986;74:484-92.

13. Abbas AE, Fortuin FD, Schiller NB, Appleton CP, Moreno CA, Lester SJ.Echocardiographic determination of mean pulmonary artery pressure. AmJ Cardiol 2003;92:1373-6.

14. Tramarin R, Torbicki A, Marchandise B, Laaban JP, Morpurgo M, WorkingGroup on Noninvasive Evaluation of Pulmonary Artery Pressure; Euro-pean Office of the World Health Organization, Copenhagen. Dopplerechocardiographic evaluation of pulmonary artery pressure in chronic ob-structive pulmonary disease: a European multicentre study. Eur Heart J1991;12:103-11.

15. Bishop LJ, Lozano M, Aduen J, Burger CD, Safford R, Keller F, et al. Anovel echocardiographic method to estimate the mean pulmonary pres-

sure: correlation with right heart catheterization [abstract]. Proc AmThorac Soc 2005;2. abstract A57.

16. Ommen SR, Nishimura RA, Hurrell DG, Klarich KW. Assessment of rightatrial pressure with 2-dimensional and Doppler echocardiography: a simul-taneous catheterization and echocardiographic study. Mayo Clin Proc2000;75:24-9.

17. Bland JM, Altman DG. Statistical methods for assessing agreement be-tween two methods of clinical measurement. Lancet 1986;1:307-10.

18. Murata I, Kihara H, Shinohara S, Ito K. Echocardiographic evaluation ofpulmonary arterial hypertension in patients with progressive systemic scle-rosis and related syndromes. Jpn Circ J 1992;56:983-91.

19. Borgeson DD, Seward JB, Miller FA Jr, Oh JK, Tajik AJ. Frequency ofDoppler measurable pulmonary artery pressures. J Am Soc Echocardiogr1996;9:832-7.

20. Hinderliter AL, Willis PW IV, Barst RJ, Rich S, Rubin LJ, Badesch DB, et al;Primary Pulmonary Hypertension Study Group. Effects of long-term infu-sion of prostacyclin (epoprostenol) on echocardiographic measures ofright ventricular structure and function in primary pulmonary hyperten-sion. Circulation 1997;95:1479-86.

21. Trow TK, McArdle JR. Diagnosis of pulmonary arterial hypertension. ClinChest Med 2007;28:59-73.

22. Himelman RB, Struve SN, Brown JK, Namnum P, Schiller NB. Improvedrecognition of cor pulmonale in patients with severe chronic obstructivepulmonary disease. Am J Med 1988;84:891-8.

23. Himelman RB, Stulbarg M, Kircher B, Lee E, Kee L, Dean NC, et al;Noninvasive evaluation of pulmonary artery pressure during exercise bysaline-enhanced Doppler echocardiography in chronic pulmonary dis-ease. Circulation 1989;79:863-71.

24. Gaine SP, Rubin LJ. Primary pulmonary hypertension. Lancet 1998;352:719-25.

25. McLaughlin VV, Shillington A, Rich S. Survival in primary pulmonary hy-pertension: the impact of epoprostenol therapy. Circulation 2002;106:1477-82.

26. Simonneau G, Barst RJ, Galie N, Naeije R, Rich S, Bourge RC et al; Tre-prostinil Study Group. Continuous subcutaneous infusion of treprostinil,a prostacyclin analogue, in patients with pulmonary arterial hypertension:a double-blind, randomized, placebo-controlled trial. Am J Respir CritCare Med 2002;165:800-4.

27. Badesch DB, Tapson VF, McGoon MD, Brundage BH, Rubin LJ,Wigley FM, et al. Continuous intravenous epoprostenol for pulmonary hy-pertension due to the scleroderma spectrum of disease: a randomized,controlled trial. Ann Intern Med 2000;132:425-34.

28. Olschewski H, Simonneau G, Galie N, Higenbottam T, Naeije R, Rubin LJ,et al; Aerosolized Iloprost Randomized Study Group. Inhaled iloprost forsevere pulmonary hypertension. N Engl J Med 2002;347:322-9.

29. Rubin LJ, Badesch DB, Barst RJ, Galie N, Black CM, Keogh A, et al. Bosen-tan therapy for pulmonary arterial hypertension [published correctionappears in N Engl J Med 2002;346:1258]. N Engl J Med 2002;346:896-903.

30. Syyed R, Reeves JT, Welsh D, Raeside D, Johnson MK, Peacock AJ. Therelationship between the components of pulmonary artery pressure re-mains constant under all conditions in both health and disease. Chest2008;133:633-9.

31. Chemla D, Herve P. Estimation of mean pulmonary artery pressure: sim-pler than expected. Chest 2008;133:592-3.