Novel Echocardiographic Approach to the Accurate Measurement of Pulmonary Vascular Resistance Based on a Theoretical Formula in Patients With Left Heart Failure　– Pilot Study –

Abstract

Background: Several non-invasive methods for measuring pulmonary vascular resistance (PVR) have been proposed to date, but they remain empirical, lacking sufficient accuracy to be used in clinical practice. The aims of this study were to propose a novel echocardiographic measurement of PVR based on a theoretical formula and investigate the feasibilty and accuracy of this method in patients with heart failure.

Methods and Results: Echocardiography was performed in 27 patients before right heart catheterization. Peak tricuspid regurgitation pressure gradient (TRPG), pulmonary regurgitation pressure gradient in end-diastole (PRPG_ed), and cardiac output derived from the time-velocity integral and the diameter in the left ventricular outflow tract (CO_LVOT) were measured. PVR based on a theoretical formula (PVR_theo) was calculated as (TRPG–PRPG_ed)/3CO_LVOT in Wood units (WU). The results were compared with PVR obtained by right heart catheterization (PVR_cath) using linear regression and Bland-Altman analysis. Mean PVR_cath was 2.4±1.4 WU. PVR_theo correlated well with PVR_cath (r=0.83, P<0.001). On Bland-Altman analysis the mean difference was 0.1±0.7 WU. The limits of agreements were smaller than for other non-invasive estimations previously reported.

Conclusions: The new echocardiographic approach based on a theoretical formula provides a non-invasive and accurate assessment of PVR in patients with heart failure. (Circ J 2015; 79: 2408–2413)

Pulmonary hypertension resulting from heart failure carries a poor prognosis.^1,2 Although pulmonary vascular resistance (PVR) is an important pathophysiologic parameter in patients with heart failure as well as with pulmonary hypertension, its use has been limited because of the invasiveness of right heart catheterization required for measurement.^3–8 The invasive nature of right heart catheterization is an obstacle to the repeated measurement required for the assessment of therapeutic interventions, as well as the time course of the disease. Doppler echocardiography provides a simple and non-invasive means of assessing hemodynamics.^9–11 To date, several non-invasive methods have been proposed to estimate PVR,^12–14 but they remain empirical, lacking sufficient accuracy to be used in cardiology practice. In this study, we propose a new echocardiographic measurement of PVR based on a theoretical formula (PVR_theo). We also investigated whether PVR_theo had improved accuracy in clinical settings by comparing PVR_theo with PVR_cath, and also with earlier non-invasive methods in patients with heart failure.

Editorial p 2324

Methods

Subjects

We consecutively enrolled 27 patients with heart failure who underwent echocardiography and right heart catheterization on the same day. All patients underwent right heart catheterization using a thermodilution catheter within 6 h of echocardiography. Exclusion criteria consisted of tricuspid regurgitation grade >2, shunt disease, aortic stenosis or aortic regurgitation grade>2 to exclude inaccurate measurement of cardiac output (CO) with the thermodilution method.

Right Heart Catheterization

Right heart catheterization was performed using a thermodilution catheter (Goodman, Tokyo, Japan) with jugular venous access. The following pressure measurements were obtained: right atrial pressure (RAP), right ventricular pressure (RVP), pulmonary artery systolic pressure (PASP), pulmonary artery end-diastolic pressure (PADP), mean pulmonary artery pressure (MPAP) and right pulmonary capillary wedge pressure (PCWP). CO was determined using a standard thermodilution technique (mean of 3 consecutive measurements).^4,5 All patients were in a stable condition between echocardiography and right heart catheterization. Neither volume loading nor bolus injection of diuretics was done between echocardiography and right heart catheterization. PVR_cath was obtained using right heart catheterization as: (MPAP–PCWP)/CO.^4,5

Transthoracic Echocardiography

Comprehensive 2-D Doppler echocardiography was performed using a commercially available ultrasound apparatus (Artida or Aplio, Toshiba Medical Systems, Tochigi, Japan). We measured peak tricuspid regurgitation pressure gradient (TRPG), pulmonary regurgitation pressure gradient in end-diastole (PRPG_ed), and CO derived from the diameter and the time-velocity integral (TVI) in the left ventricular outflow tract (CO_LVOT) on echocardiography. PRPG_ed was measured using a continuous wave Doppler technique and pressure gradient was calculated using the simplified Bernoulli equation in end-diastole (Figure 1). In patients with atrial fibrillation, echocardiographic parameters were obtained when the preceding RR interval and the pre-preceding RR interval were equal to the average RR interval.

Figure 1.

Measurement of pulmonary regurgitation pressure gradient in end-diastole (PRPG_ed). PRPG_ed was estimated using the simplified Bernoulli equation at (dashed line) R wave on continuous-wave Doppler electrocardiography. Pulmonary regurgitation flow was easily demonstrable in most of the patients.

Theoretical Formula for PVR

PVR was defined as: (MPAP–PCWP)/CO.^4,5,12 Echocardiographic PVR (PVR_theo) was determined based on several theoretical assumptions: (1) pressure gradient was calculated from the Doppler-derived regurgitation velocity using the simplified Bernoulli equation;¹⁵ (2) mean blood pressure was weighted 2:1 in favor of the diastolic blood pressure: MPAP was derived from the estimated PADP and PASP according to the formula MPAP=DPAP+1/3(PASP–PADP);¹³ (3) PCWP was equal to PADP; and (4) the RV end-diastolic pressure and RAP were equal in end-diastole.¹⁶ Based on these assumptions, the following calculations were made: PADP=PRPG_ed+RAP; PASP=TRPG+RAP. TVI from a pulsed Doppler technique and the LV outflow tract (LVOT) diameter were calculated in the LVOT, and provided for CO calculation because there were no shunt diseases involved. Given that the RV outflow tract (RVOT) diameter is more difficult to measure than LVOT diameter, particularly in adults, we chose to measure CO in the LVOT rather than in the RVOT provided that there was no shunt disease involved.¹⁴ Using these parameters, PVR was therefore calculated as:

PVR=(MPAP–PCWP)/CO

={[PADP+(PASP–PADP)/3]–(PRPG_ed+CVP)}/CO

={[PRPG_ed+CVP+(TRPG+CVP–PRPG_ed–CVP)/3]–(PRPG_ed+CVP)}/CO

=(TRPG–PRPG_ed)/(3×CO_LVOT).

Finally, PVR_theo was defined as: (TRPG–PRPG_ed)/(3×CO_LVOT).

PVR was also calculated based on the earlier methods proposed by Abbas et al, Scapellato et al, Haddad et al and Kouzu et al in the same cohort of patients.^12,13,17,18 In brief, PVR by Abbas et al (PVR_Abbas) was based on the ratio of peak tricuspid regurgitant velocity (TRV) to TVI in the RV outflow tract. PVR by Scapellato et al (PVR_Scapellato) was based on a ratio of systolic time interval involving the pre-ejection period (PEP), acceleration time (AcT), ejection time, and total systolic time (TT): index=([PEP/AcT]/TT). PVR by Haddad et al (PVR_Haddad) was based on the index of SPAP/(HR×TVI), where SPAP is systolic pulmonary artery pressure and HR is heart rate. SPAP was estimated as RV systolic pressure, which equals the sum of the trans-tricuspid regurgitation gradient based on the simplified Bernoulli equation and the echocardiographically estimated RAP. PVR by Kouzu et al (PVR_Kouzu) was based on the ratio of TRPG to TVI.

This study was approved by the institutional review board, and written informed consent was obtained from each patient.

Statistical Analysis

All statistical analysis were performed using SPSS (SPSS version 22; Chicago, IL, USA). Categorical values are expressed as absolute and relative frequencies. Continuous values are expressed as mean ± SD. Linear regression analysis was generated between PVR_cath and the PVRs (PVR_theo, PVR_Abbas, PVR_Scapellato, PVR_Haddad, and PVR_Kouzu) non-invasively obtained by the earlier methods. Pearson’s correlation coefficients were then obtained. The 95% CI are reported when required. Bland-Altman analysis was carried out for invasive PVR and the non-invasively obtained PVRs.¹⁹

Results

Patient Characteristics

Of the 27 patients who underwent right heart catheterization, 5 patients (19%) were excluded because PRPG_ed was unmeasurable. Clinical characteristics are presented in Table 1. Referral diagnosis included ischemic heart disease (n=6), dilated cardiomyopathy (n=6), hypertrophic cardiomyopathy (n=2), valvular heart disease (n=2), hypertensive heart disease (n=2), tachycardia-induced cardiomyopathy (n=1), cardiac sarcoidosis (n=1), pulmonary hypertension (n=1), and other (n=1). Hemodynamic and echocardiographic characteristics are given in Table 2.

Table 1. Patient Characteristics

Characteristics	Results
Age (years)	65±12
Male	12 (55)
Height (cm)	161±11
Body weight (kg)	63±18
NYHA class
I	1 (5)
II	14 (64)
III	6 (27)
IV	1 (5)
Heart rate (beats/min)	68±14
Atrial fibrillation	6 (27)
BNP (pg/ml)	367±280

Data given as mean ± SD or n (%). BNP, brain natriuretic peptide; NYHA, New York Heart Association.

Table 2. Hemodynamic and Echocardiographic Characteristics

	Data
Hemodynamic parameters
Heart rate (beats/min)	71±16
PASP (mmHg)	32.2±16.8
PADP (mmHg)	15.1±12.1
MPAP (mmHg)	21.5±13.3
PCWP (mmHg)	11.8±10.6
RAP (mmHg)	4.2±4.6
PVR (WU)	2.4±1.4
CO (L/min)	4.0±0.8
Echocardiographic parameters
Heart rate (beats/min)	68±14
EF (%)	43.3±12.4
LAD (mm)	46.7±12.4
TRPG (mmHg)	24.8±6.0
PRPGed (mmHg)	4.7±2.9
COLVOT (L/min)	3.1±0.7

Data given as mean ± SD. CO, cardiac output; CO_LVOT, CO from the time-velocity integral at the LV outflow tract; EF, ejection fraction; LAD, left atrial diameter; MPAP, mean pulmonary artery pressure; PADP, pulmonary artery diastolic pressure; PASP, pulmonary artery systolic pressure; PCWP, pulmonary capillary wedge pressure; PRPG_ed, pulmonary regurgitation pressure gradient at end-diastole; PVR, pulmonary vascular resistances; RAP, right atrial pressure; TRPG, tricuspid regurgitation pressure gradient; WU, Wood units.

PVR Assessment on Echocardiography

Linear regression analysis indicated a significant correlation between PVR_cath and PVR_theo (r=0.83, 95% CI: 0.64–0.93, P<0.001, Figure 2). The equation of the regression line was: PVR_cath=1.16×PVR_theo+0.23. The 95% CI for slope and intercept included 1 and 0, respectively. On Bland-Altman analysis the mean difference was −0.1 Wood units (WU) ±1.6 WU (2 SD limit of agreement; Figure 3). In contrast, PVR_Abbas, PVR_Scapellato, PVR_Haddad and PVR_Kouzu had modest correlations with PVR_cath although this was statistically significant (PVR_Abbas: r=0.52, 95% CI: 0.13–0.773, P=0.013; PVR_Haddad: r=0.51, 95% CI: 0.12–0.77, P=0.13; PVR_Kouzu: r=0.68, 95% CI: 0.36–0.86, P<0.001). On Bland-Altman plots, PVR_Abbas, PVR_Scapellato, PVR_Haddad and PVR_Kouzu appeared to have larger differences in measurement than PVR_theo (Figure 3).

Figure 2.

Correlations between catheter-derived pulmonary vascular resistance (PVR_cath) and (A) PVR according to theoretical formula (PVR_theo), (B) PVR according to Abbas et al (PVR_Abbas), (C) PVR according to Kouzu et al (PVR_Kouzu), (D) PVR according to Haddad et al (PVR_Haddad), and (E) PVR according to Scapellato et al (PVR_Scapellato). Excellent correlation was observed between PVR_cath and PVR_theo. It is of note that the slope of the regression line was 1.15, and y intercept was −0.22. The 95% CI for slope and intercept included 1 and 0, respectively. AcT, acceleration time; BSA, body surface area; CO, cardiac output; HR, heart rate; PEP, pre-ejection period; PRPG_ed, pulmonary regurgitation pressure gradient in end-diastole; SPAP, systolic pulmonary artery pressure; TRPG, tricuspid regurgitation pressure gradient; TRV, tricuspid regurgitant velocity; TT, total systolic time; TVI, time-velocity integral.

Figure 3.

Bland-Altman analysis of the difference between catheter-derived pulmonary vascular resistance (PVR_cath) and (A) PVR according to theoretical formula (PVR_theo), (B) PVR according to Abbas et al (PVR_Abbas), (C) PVR according to Kouzu et al (PVR_Kouzu), (D) PVR according to Haddad et al (PVR_Haddad), and (E) PVR according to Scapellato et al (PVR_Scapellato). The mean difference in measurements was −0.1±1.6 WU between PVR_cath and PVR_theo. AcT, acceleration time; BSA, body surface area; CO, cardiac output; HR, heart rate; PEP, pre-ejection period; PRPG_ed, pulmonary regurgitation pressure gradient in end-diastole; SPAP, systolic pulmonary artery pressure; TRPG, tricuspid regurgitation pressure gradient; TRV, tricuspid regurgitant velocity; TT, total systolic time; TVI, time-velocity integral.

Intraobserver and interobserver variability of measurements for PVR_theo were also assessed in 10 randomly allocated patients. The mean intraobserver difference between the measurements was 0.10±0.53 WU. Intraclass correlation (ICC) was 0.89 (P<0.001). The mean difference between the 2 independent observers (T.K. and Y.M.) was 0.53±0.68 WU. ICC was 0.69 (P<0.001).

Discussion

We have proposed a new method for the non-invasive estimation of PVR based on a theoretical formula (PVR_theo). We found that (1) PVR_theo was significantly correlated with invasive PVR, with the regression line close to y=x in a cohort of patients with left heart failure; and (2) PVR_theo appeared to be accurate compared with earlier non-invasive methods.

PVR is calculated invasively as the ratio of transpulmonary pressure gradient to transpulmonary flow.^4,5 Catheter-derived measurement remains the gold standard, which hampers routine use at the bedside. Accordingly, several non-invasive methods have been proposed to date. Among them, non-invasive estimations of PVR by Abbas et al, Scapellato et al, Haddad et al and Kouzu et al have shown relatively good agreement with invasive PVR. Nonetheless, the accuracy of these non-invasive methods remains suboptimal with regard to use in cardiology practice.^14,17,18 Part of the difficulty was that those non-invasive PVRs remained empirical, and were not based on theory. Hence, in this study, we utilized theoretical formulae as far as possible in order to achieve accurate PVR measurement. Concurrently, we use measurements that were feasible on transthoracic echocardiography in the majority of adult patients. We used 3 parameters, TRPG, PRPG_ed and CO_LVOT, to calculate PVR_theo. These are well established echocardiographic parameters that are easy to measure. Tricuspid regurgitation exists in the majority of patients with pulmonary hypertension.²⁰ In addition, trivial-mild forms of pulmonary regurgitation are present in almost every patient.²¹ Although pulmonary regurgitation may be difficult to detect on Doppler echocardiography throughout diastole, it is relatively easy to detect if the measurement is limited to the end-diastole.

In general, there are 2 kinds of formulae for echocardiographically derived indices: theoretical formulae and empirical formulae. Theoretical formulae are those derived from physics theory. They are accurate as far as the theory can be applied to the particular conditions. The simplified Bernoulli equation to estimate pressure gradient is a good example.²² In contrast, empirical formulae are derived based on experience, that is, the statistical correlation between the parameters. They inherently include substantial errors of estimation. Examples include the estimation of mitral valve area in mitral stenosis using transmitral flow derived pressure half time,^23,24 as well as the estimation of mean pulmonary arterial pressure by the AcT of the RV outflow velocity.^25,26 In the present study, we exclusively utilized theoretical formulae for calculating PVR in order to minimize the error of estimation. Despite the use of several theoretical assumptions, we successfully estimated PVR by this method. It is noteworthy that 95% CI for slope and intercept included 1 and 0, in contrast to other earlier non-invasive estimation methods for PVR. For the best of our knowledge, this is the first repot of theoretically derived PVR estimation by echocardiography.

Clinical Implications

Because PVR_theo was found to have substantial accuracy, it can be used to investigate the various conditions of pathophysiology in heart failure and/or pulmonary hypertension in clinical settings.

Study Limitations

There are several limitations in this study. First, the measurement of PRPG_ed may encounter problems such as weak pulmonary regurgitant signal and improper Doppler angle. Nonetheless, even a trivial pulmonary regurgitation jet can often be measured if the measurement is limited to the end-diastole. Appropriate PRPG_ed was able to be measured in most of the present patients (81%). Second, in patients with tachycardia, mean blood pressure may not be obtained as weighted 2:1 from the peak and nadir pressures. Care should be taken to calculate PVR_theo under tachycardia. Third, PVR_theo was tested only in those presenting with relatively mild elevation of PVR due to left heart failure. High PVR as seen in reactive pulmonary hypertension was not included in this study. In patients with reactive pulmonary hypertension, PCWP may not be equal to PADP. Therefore, PVR_theo remains to be tested in those with pulmonary diseases. Fourth, CO_LVOT is not the same as CO obtained using thermodilution catheter in the presence of significant cardiac shunt or aortic valvulopathy. The use of CO_LVOT may not be a direct measure of right-sided CO. Nonetheless, we used CO_LVOT in the calculation of PVR_theo because the accuracy of measuring volumetric flow from the LVOT is higher than that measured from the right-sided vessels.¹⁴ Last, this pilot study involved a small sample size in a single center. A large-scale, multi-center study should be conducted to validate PVR_theo in a generalized population.

Conclusions

This new echocardiographic approach based on a theoretical formula provides a non-invasive and accurate assessment of PVR in patients with heart failure.

References

• 1.    Guazzi  M,  Borlaug  BA. Pulmonary hypertension due to left heart disease. Circulation 2012; 126: 975–990.
• 2.    Hirashiki  A,  Kondo  T,  Adachi  S,  Nakano  Y,  Shimazu  S,  Shimizu  S, et al. Prognostic value of pulmonary hypertension in ambulatory patients with non-ischemic dilated cardiomyopathy. Circ J 2014; 78: 1245–1253.
• 3.   JCS Joint Working Group. Guidelines for treatment of acute heart failure (JCS 2011): Digest version. Circ J 2013; 77: 2157–2201.
• 4.    McLaughlin  VV,  Archer  SL,  Badesch  DB,  Barst  RJ,  Farber  HW,  Lindner  JR, et al. ACCF/AHA 2009 expert consensus document on pulmonary hypertension: A report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents and the American Heart Association: Developed in collaboration with the American College of Chest Physicians, American Thoracic Society, Inc., and the Pulmonary Hypertension Association. Circulation 2009; 119: 2250–2294.
• 5.    Yancy  CW,  Jessup  M,  Bozkurt  B,  Butler  J,  Casey  DE Jr,  Drazner  MH, et al. 2013 ACCF/AHA guideline for the management of heart failure: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2013; 62: e147–e239, doi:10.1016/j.jacc.2013.05.019.
• 6.    Braunwald  E,  Colucci  WS. Vasodilator therapy of heart failure: Has the promissory note been paid? N Engl J Med 1984; 310: 459–461.
• 7.    Chemla  D,  Castelain  V,  Herve  P,  Lecarpentier  Y,  Brimioulle  S. Haemodynamic evaluation of pulmonary hypertension. Eur Resp J 2002; 20: 1314–1331.
• 8.    Rudski  LG,  Lai  WW,  Afilalo  J,  Hua  L,  Handschumacher  MD,  Chandrasekaran  K, et al. Guidelines for the echocardiographic assessment of the right heart in adults: A report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 2010; 23: 685–713; quiz 786–788.
• 9.    Eysmann  SB,  Palevsky  HI,  Reichek  N,  Hackney  K,  Douglas  PS. Two-dimensional and Doppler-echocardiographic and cardiac catheterization correlates of survival in primary pulmonary hypertension. Circulation 1989; 80: 353–360.
• 10.    Yeo  TC,  Dujardin  KS,  Tei  C,  Mahoney  DW,  McGoon  MD,  Seward  JB. Value of a Doppler-derived index combining systolic and diastolic time intervals in predicting outcome in primary pulmonary hypertension. Am J Cardiol 1998; 81: 1157–1161.
• 11.    D’Alto  M,  Romeo  E,  Argiento  P,  D’Andrea  A,  Vanderpool  R,  Correra  A, et al. Accuracy and precision of echocardiography versus right heart catheterization for the assessment of pulmonary hypertension. Int J Cardiol 2013; 168: 4058–4062.
• 12.    Abbas  AE,  Fortuin  FD,  Schiller  NB,  Appleton  CP,  Moreno  CA,  Lester  SJ. A simple method for noninvasive estimation of pulmonary vascular resistance. J Am Coll Cardiol 2003; 41: 1021–1027.
• 13.    Scapellato  F,  Temporelli  PL,  Eleuteri  E,  Corra  U,  Imparato  A,  Giannuzzi  P. Accurate noninvasive estimation of pulmonary vascular resistance by Doppler echocardiography in patients with chronic failure heart failure. J Am Coll Cardiol 2001; 37: 1813–1819.
• 14.    Roule  V,  Labombarda  F,  Pellissier  A,  Sabatier  R,  Lognone  T,  Gomes  S, et al. Echocardiographic assessment of pulmonary vascular resistance in pulmonary arterial hypertension. Cardiovasc Ultrasound 2010; 8: 21.
• 15.    Currie  PJ,  Seward  JB,  Reeder  GS,  Vlietstra  RE,  Bresnahan  DR,  Bresnahan  JF, et al. Continuous-wave Doppler echocardiographic assessment of severity of calcific aortic stenosis: A simultaneous Doppler-catheter correlative study in 100 adult patients. Circulation 1985; 71: 1162–1169.
• 16.    Sakai  K,  Nakamura  K,  Satomi  G,  Kondo  M,  Hirosawa  K. Evaluation of tricuspid regurgitation by blood flow pattern in the hepatic vein using pulsed Doppler technique. Am Heart J 1984; 108: 516–523.
• 17.    Haddad  F,  Zamanian  R,  Beraud  AS,  Schnittger  I,  Feinstein  J,  Peterson  T, et al. A novel non-invasive method of estimating pulmonary vascular resistance in patients with pulmonary arterial hypertension. J Am Soc Echocardiogr 2009; 22: 523–529.
• 18.    Kouzu  H,  Nakatani  S,  Kyotani  S,  Kanzaki  H,  Nakanishi  N,  Kitakaze  M. Noninvasive estimation of pulmonary vascular resistance by Doppler echocardiography in patients with pulmonary arterial hypertension. Am J Cardiol 2009; 103: 872–876.
• 19.    Bland  JM,  Altman  DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–310.
• 20.    Chan  KL,  Currie  PJ,  Seward  JB,  Hagler  DJ,  Mair  DD,  Tajik  AJ. Comparison of three Doppler ultrasound methods in the prediction of pulmonary artery pressure. J Am Coll Cardiol 1987; 9: 549–554.
• 21.    Masuyama  T,  Kodama  K,  Kitabatake  A,  Sato  H,  Nanto  S,  Inoue  M. Continuous-wave Doppler echocardiographic detection of pulmonary regurgitation and its application to noninvasive estimation of pulmonary artery pressure. Circulation 1986; 74: 484–492.
• 22.    Berger  M,  Haimowitz  A,  Van Tosh  A,  Berdoff  RL,  Goldberg  E. Quantitative assessment of pulmonary hypertension in patients with tricuspid regurgitation using continuous wave Doppler ultrasound. J Am Coll Cardiol 1985; 6: 359–365.
• 23.    Martin  RP,  Rakowski  H,  Kleiman  JH,  Beaver  W,  London  E,  Popp  RL. Reliability and reproducibility of two dimensional echocardiograph measurement of the stenotic mitral valve orifice area. Am J Cardiol 1979; 43: 560–568.
• 24.    Nichol  PM,  Gilbert  BW,  Kisslo  JA. Two-dimensional echocardiographic assessment of mitral stenosis. Circulation 1977; 55: 120–128.
• 25.    Dabestani  A,  Mahan  G,  Gardin  JM,  Takenaka  K,  Burn  C,  Allfie  A, et al. Evaluation of pulmonary artery pressure and resistance by pulsed Doppler echocardiography. Am J Cardiol 1987; 59: 662–668.
• 26.    Kitabatake  A,  Inoue  M,  Asao  M,  Masuyama  T,  Tanouchi  J,  Morita  T, et al. Noninvasive evaluation of pulmonary hypertension by a pulsed Doppler technique. Circulation 1983; 68: 302–309.