INTRODUCTION
Doppler echocardiography (DE) is a non-invasive
screening test for pulmonary artery hypertension (PH).
In the absence of pulmonary outflow obstruction, DE
provides an estimate of the right ventricular systolic
pressure (RVSP), which is equivalent to the systolic
pulmonary arterial pressures (sPAP). The right
ventricular systolic pressure, and thus the pulmonary
arterial pressure, can be estimated from the tricuspid
regurgitant (TR) jet velocity by using modified Bernoulli
equation: RVSP = 4V
2
+ RA pressure. In this equation,
'V' represents maximum TR jet velocity and RA denotes
right atrial pressure.
1
Tricuspid regurgitation is
mandatory for this calculation which is fortunately found
in 70% of normal subjects.
2
Right atrial pressure (RAP)
cannot be calculated with DE. However, it can be
estimated by looking at inferior vena cava (IVC)
diameter and its collapse with inspiration on two-
dimensional echocardiography.
3
Echocardiography can also provide information about
the cause and consequences of PH, including right and
left ventricular dimensions and function, heart valve
abnormalities, right ventricular ejection and left
ventricular filling characteristics, and presence of a
pericardial effusion which may have prognostic value.
4,5
However, right heart catheterization (RHC) is required
for a definitive diagnosis of PH. Currently, PH is defined
as mean pulmonary artery pressure (mPAP) of ≥ 25 mmHg
at rest obtained invasively through RHC.
6
The diagnostic
accuracy of DE in predicting PH has been evaluated
previously but results are conflicting.
7,8
To the best of authors' knowledge, no such study has
been done in Pakistan. The objective of this study was
to evaluate the correlation between RVSP and sPAP in
our settings and to study the impact of RAP on this
correlation in terms of correlation coefficient, error and
diagnostic accuracy.
Journal of the College of Physicians and Surgeons Pakistan 2016, Vol. 26 (4): 255-259 255
ORIGINAL ARTICLE
Correlation Between Doppler Echocardiography and Right Heart
Catheterization Derived Pulmonary Artery Pressures:
Impact of Right Atrial Pressures
Ikram Ahmed, Muhammad Masudul Hassan Nuri, Ahmad Nasiruddin Zakariyya,
Syed Mashud Ahmad and Mobasher Ahmed
ABSTRACT
Objective: To evaluate the correlation between Doppler echocardiography (DE) and right heart catheterization (RHC)
derived pulmonary artery pressures and to assess the impact of right atrial (RA) pressures on this correlation.
Study Design: Cross-sectional analytical study.
Place and Duration of Study: Cardiology Department, Tahir Heart Institute, Chenab Nagar, from June 2013 to December
2014.
Methodology: All patients undergoing RHC were included. Relevant data were collected from hospital database.
Continuous variables were expressed as the mean and SD or as the median and interquartile range where the
distributions were skewed. Pearson correlation coefficient and Bland-Altman method were used to correlate DE derived
right ventricular systolic pressure (RVSP) and RHC derived systolic pulmonary artery pressures (sPAP). Adjusted RVSP
was calculated by replacing default value of RA pressure (10 mmHg) with RHC derived mean RA pressure. Receiver
operating characteristic curve (ROC) was used to identify the best cut-off value of RVSP in predicting pulmonary
hypertension.
Results: Fifty-one patients completed the study protocol. Mean age of study population was 45.22 ±15.25 years with male
to female ratio of 1.47:1. Median error was 13 mmHg (7 to 20). Pearson correlation coefficient (r) between RVSP and sPAP
was 0.72. Bland-Altman method of correlation showed bias of +4.43 mmHg with 95% limits of agreement ranging from
-34.61 to +43.47. Using ROC curve, the best cut-off value of RVSP was greater than 52 mmHg with accuracy of 75%
(sensitivity: 81%, specificity: 69%) in predicting pulmonary hypertension. Adjusted RVSP showed only little improvement
in correlation (r = 0.75), adjusted error (13.65 ±13.05) and diagnostic accuracy (79%).
Conclusion: Doppler echocardiography can frequently overestimate pulmonary artery pressures. Though correctly
estimated RA pressure may improve this correlation, yet its contribution is only minimal.
Key Words: Doppler echocardiography. Cardiac catheterization. Pulmonary hypertension.
Department of Cardiology, Tahir Heart Institute, Chenab
Nagar, District Chiniot.
Correspondence: Dr. Ikram Ahmed, Consultant Cardiologist,
Department of Cardiology, Tahir Heart Institute, Chenab
Nagar, District Chiniot.
E-mail: ikramahmedrana@gmail.com
Received: January 26, 2015; Accepted: December 01, 2015.
METHODOLOGY
This cross-sectional analytical study was conducted
after approval from Hospital Review Board, at the
Cardiology Department, Tahir Heart Institute, Chenab
Nagar, Pakistan. Relevant data were collected from the
Hospital database. All patients who underwent RHC
from June 1, 2013 to December 31, 2014 were included.
Patients with right ventricular outflow tract obstruction,
non-availability of RVSP on DE and time interval greater
than 30 days between RHC and DE were excluded.
Echocardiographic examinations were performed on
commercially available ultrasound systems (Toshiba
Nemio XG) using 2.5 MHz transducer. Images were
obtained in left lateral decubitus for parasternal and
apical views. Right ventricular systolic pressure was
calculated from peak systolic TR velocity obtained with
continuous-wave Doppler (CW) using modified Bernoulli
equation: RVSP = 4V
2
+ RAP. RAP was taken as
10 mmHg in all patients. In order to control the
confounding effects of this constant value (10 mmHg),
adjusted RVSP was calculated by replacing this
constant with RHC derived RAP. Doppler estimations
were termed as over-estimated if RVSP ≥ 10 mmHg of
RHC derived sPAP, and under-estimated if RVSP ≤ 10
mmHg of RHC derived sPAP.
Right heart catheterization was performed through
right femoral vein with Toshiba Infinix-I system under
AP-projection. Five French Judkins Right 3.5 (JR 3.5)
diagnostic catheter was placed in main pulmonary
artery. Position of tip of catheter was confirmed by
injection of contrast media in main pulmonary artery.
Pressures from right ventricle, right atrium, sPAP and
mPAP were recorded. Pulmonary artery hypertension
(PH) was defined as mPAP ≥ 25 mmHg on RHC.
Statistical Package for Social Sciences (SPSS) version
20 was used for statistical analysis. Continuous
variables were expressed in means ± standard
deviations or, in situations where distributions were
skewed as median and interquartile range. Interquartile
range was expressed in brackets from median of
quartile 1 to median of quartile 3. Nominal data were
expressed in frequencies with percentages. Variable
“Error” was calculated by subtracting sPAP from RVSP
and converting negative values into same but positive
integer (e.g. -7 converted to +7). Another variable
“Adjusted Error” was calculated similarly but using
adjusted RVSP instead of RVSP. Kruskal-Wallis test was
used to compare medians of “Error” between different
operators. Significance level was set at p ≤ 0.05.
Pearson correlation coefficient and Bland-Altman
method were used to correlate DE derived RVSP, and
RHC derived sPAP.
9
Using receiver operating
characteristic (ROC) curve, the best cut-off value of
RVSP was identified in predicting PH. Same statistical
procedures were repeated with adjusted RVSP to obtain
value of correlation coefficient, adjusted error and
diagnostic accuracy.
RESULTS
Fifty-one (51) patients completed the study protocol.
Males were 30 (58.8%) and females were 21 (42.2%).
The mean age of the study population was 45.22 ±15.25
years. Mean time interval between DE and RHC was
8.57 ±8.21 days. The majority of the patients had
rheumatic heart disease (43%) and atrial septal defect
(33%). Out of 51 patients, 32 had PH.
Table I describes various parameters to correlate DE
derived RVSP, and RHC derived sPAP. Median RVSP
was 55 mmHg (45 to 70) while RHC derived median
sPAP was 48 mmHg (36 to 70). Median error committed
by the operators in estimating sPAP was 13 mmHg (7 to
20). Doppler derived RVSP was over-estimated in
23 (45%), under-estimated in 11 (22%) and found within
±10 mmHg limits in 17 (33%) patients. Pearson
correlation coefficient (r) between these two variables
was 0.72 (moderate correlation in statistical terms)
with p < 0.001. Bland-Altman analysis showed bias of
+4.43 mmHg and 95% limits of agreement (LOA)
ranging from -34.61 to +43.47. This wide range of LOA
in clinical context is indicative of inaccuracy. Median
error committed by different operators varied from 9 to
Ikram Ahmed, Muhammad Masudul Hassan Nuri, Ahmad Nasiruddin Zakariyya, Syed Mashud Ahmad and Mobasher Ahmed
256 Journal of the College of Physicians and Surgeons Pakistan 2016, Vol. 26 (4): 255-259
Table I: Correlation between RVSP and sPAP.
Echo operator Total
ABCD
n 17 11 13 10 51
RVSP - Median (IQR) 60 (50 to 70) 70 (50 to 100) 55 (45 to 80) 55 (45 to 61.25) 55 (45 to 70)
Adjusted RVSP - Median (IQR) 60 (46.5 to 67.5) 68 (42 to 99) 51 (41 to 83) 50.5 (38.25 to 64.75) 54 (43 to 71)
sPAP - Median (IQR) 55 (39.5 to 70) 53 (40 to 117) 38 (32 to 80) 36.5 (31.5 to 59.5) 48 (36 to 70)
Error - Median (IQR) 9 (5.5 to 15.5) 15 (10 to 31) 15 (8 to 30) 12.5 (6.25 to 21.25) 13 (7 to 20)
Adjusted error - Median (IQR) 8 (2.5 to 20.5) 10 (4 to 23) 13 (3.5 to 31.5) 8 (2 to 18) 9 (4 to 20)
Kruskal-Wallis test
(comparing median of Error) p = 0.37
Bias Mean +4.65 -1.73 +5.23 +9.8 +4.43
95% LOA -21.51 to +30.81 -52 to +49.05 -45.86 to +56.32 -14.67 to +34.28 -34.61 to +43.47
r (adjusted RVSP vs. sPAP) 0.52 (p=0.03) 0.81 (p=0.002) 0.67 (p=0.01) 0.86 (p=0.001) 0.75 (p<0.001)
r (RVSP vs. sPAP) 0.58 (p=0.01) 0.79 (p=0.004) 0.61 (p=0.02) 0.68 (p=0.03) 0.72 (p<0.001)
IQR = Interquartile range (quartile 1 to quartile 3); LOA = Bland-Altman 95% limits of agreement; r = Pearson correlation coefficient
Doppler echocardiography and right heart catheterization derived pulmonary artery pressures
Journal of the College of Physicians and Surgeons Pakistan 2016, Vol. 26 (4): 255-259
257
15mmHg. However, using Kruskal-Wallis test, this
variation was not found statistically significant (p = 0.37).
Using ROC curve, the best cut-off value of RVSP was
> 52 mmHg with diagnostic accuracy of 75% (sensitivity
81%, specificity 69%) in predicting PH. Area under curve
(AUC) was 0.81 (p < 0.001, 95% CI 0.68 to 0.93, Figure 1).
Same ROC curve analysis with adjusted RVSP
(AUC=0.83, p < 0.001, 95% CI 0.71 to 0.95, Figure 2)
showed the best cut-off value of adjusted RVSP > 48
mmHg with diagnostic accuracy of 79% (sensitivity 84%,
specificity 74%) in predicting PH (Table II). Impact of
RAP on this correlation was assessed in terms of r,
adjusted error and diagnostic accuracy. Median adjusted
error reduced slightly (error = 13, adjusted error = 9 mmHg).
Minimal improvement in r (0.72 to 0.75) and diagnostic
accuracy (75% to 79%) were noted.
DISCUSSION
Correlation between DE derived RVSP and RHC derived
sPAP was first described by Yock and Popp in 1984.
10
They described good correlation (r = 0.93). This is
followed by some studies of sample size ranging from 34
to 127 patients with good correlation.
11-13
However,
results of some other small studies questioned the
reliability of DE in estimating pulmonary artery
pressures.
8,14,15
Good correlation does not necessarily
mean that one test is an accurate substitute for another.
Most of the above mentioned studies used r and Bland-
Altman method to describe correlation between RVSP
and sPAP. It was found that each method has its own
merits and demerits. Another variable Error was
calculated to find error per case in estimating sPAP on
DE. Bland-Altman calculated bias by taking the mean of
difference between two variables (testing variables).
Mean value will be closer to zero as negative values
(under-estimated) will cancel positive values (over-
estimated). So Bias can tell us about over or under-
estimation but won't identify true magnitude of error in
study population. In this study, operator A showed good
correlation with narrow 95% LOA range (approximately
50) as compared to operator B, where 95% LOA range
was equal to 101. This better correlation by operator A
was also supported by decreased median Error
(operator A = 9, operator B = 15). However, the value of
r was better for operator B (Table I) suggesting that r
alone is not enough to describe such correlation. By
assessing this correlation through these three methods
simultaneously, the authors able to describe it as
moderately positive (r = 0.72) and inaccurate correlation
(95% LOA range = 77, median error = 13 mmHg) with
frequent over-estimation of RVSP on DE. Expertise of
different operators may influence the accuracy. So we
compared the medians of error by different operators but
did not find any significant difference.
Estimation of RAP is essential in calculation of RVSP
through modified Bernoulli equation. As stated earlier, in
this retrospective study, RAP was taken as 10 mmHg in
all patients. American Society of Echocardiography 2010
guidelines recommend that RAP ranges from 0 to
5 mmHg if IVC diameter in supine position is < 21 mm
and it collapses > 50% with inspiration (normal group).
When IVC diameter is > 20 mm and it collapses < 50%
then RAP is usually greater than 15 mmHg. Normal
Table II: Diagnostic accuracy of RVSP and adjusted RVSP in predicting
pulmonary hypertension.
Cut-off Value Sensitivity % Specificity % Accuracy %
RVSP > 47 mmHg 91 54 72
> 52 mmHg 81 69 75
> 57 mmHg 66 84 75
Adjusted RVSP > 43 mmHg 91 53 72
> 48 mmHg 84 74 79
> 53 mmHg 72 84 78
Figure 1: Receiver operating characteristics (ROC) curve analysis for DE
derived RVSP in predicting RHC derived pulmonary hypertension.
AUC= Area under curve.
Figure 2: Receiver operating characteristics (ROC) curve analysis for DE
derived Adjusted RVSP in predicting RHC derived pulmonary hypertension.
AUC = Area under curve.
diameter (≤ 20 mm) with subnormal collapse (≤ 50%) or
above normal diameter with normal collapse constitutes
intermediate group where RAP range is 5 to 10 mmHg.
Guidelines recommend midrange values (i.e. 3 for
normal and 8 for intermediate group).
3
Some studies
have evaluated the validity of the IVC parameters for the
accuracy of the estimation of RAP.
16-18
Most, but not all,
studies have demonstrated good correlations between
the IVC collapsibility index ([IVC max-IVC min] / IVC
max) and RAP (0.57 < r ≤ 0.76).
19,20
In this study, this
default value (10 mmHg) of RAP was replaced with RHC
derived RAP and then adjusted RVSP calculated.
Hence, adjusted RVSP is calculated by both DE and
RHC findings and cannot be attributed to DE alone. The
objective was not to discourage the estimation of RAP,
but to assess the impact of RAP on correlation between
RVSP and sPAP. Only minimal improvement was found
(Table I). This suggests that had RAP be estimated with
100% accuracy on DE, even then DE would have
remained inaccurate in estimating sPAP.
Greiner
et al. conducted a retrospective study on large
sample of unselected patients. Their results validate the
reliability of DE in estimating sPAP. They have
highlighted the causes of over- and under-estimation of
DE derived RVSP. Over-estimation was mainly due to
maximum TR velocity boundary artifacts. They
suggested that maximum velocity should be measured
at the best spectral-wave boundary, avoiding Doppler
artifacts (fringes). Incomplete spectral-wave envelope
was second common reason. They suggested that only
signals extended for at least half of the systole should be
measured, and incomplete or absent TR may be
avoided by increasing blood pool volume with a
strategy as simple as drinking a cup of water before
examination.
7
Fisher et al. conducted a prospective
study where time interval between DE and RHC was just
one hour. Their results are in line with our results, apart
from the fact that under-estimation was as common as
over-estimation.
8
This study has few limitations which should be kept in
mind while interpreting its results. It was a retrospective
collected data and images of DE recordings were not
available so the authors could not really look into the
causes of over- or under-estimation. Maximum time
interval between DE and RHC was set at 30 days.
Pulmonary pressures in patients with PH are known to
fluctuate significantly over the course of several hours.
21
Mean time interval in this study was 8.57 ±8.21 days.
The study population consisted of selected patients with
suspicion of PH on DE. Hence, evaluation of diagnostic
accuracy may be questioned. Sample size was small.
CONCLUSION
Doppler echocardiography is not very accurate in
estimating pulmonary artery pressures. Over-estimation
was more common than under-estimation. Correct
estimation of right atrial pressures may improve the
correlation between DE derived RVSP, and RHC derived
sPAP. However, this contribution is only minimal.
REFERENCES
1. Habib G, Torbicki A. The role of echocardiography in the
diagnosis and management of patients with pulmonary
hypertension. Eur Respir Rev 2010; 19:1-12.
2. Yoshida K, Yoshikawa J, Shakudo M, Akasaka T, Jyo Y, Takao
S, et al. Color Doppler evaluation of valvular regurgitation in
normals. Circulation 1988; 78:840-47.
3. Rudski LG, Lai WW, Afialo 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.
4. Fisher MR, Forfia PR, Mathai SC, Housten-Harris T, Hemnes
AR, Borlaug BA,
et al. Tricuspid annular displacement predicts
survival in pulmonary hypertension. Am J Respir Crit Care Med
2006; 174:1034-41.
5. Raymond RJ, Hinderliter AL, Willis PW, Ralph D, Caldwell EJ,
Williams W, et al. Echocardiographic predictors of adverse
outcomes in primary pulmonary hypertension. J Am Coll
Cardiol 2002; 39:1214-9.
6. Galiè N, Hoeper MM, Humbert H. Guidelines for the diagnosis
and treatment of pulmonary hypertension. Eur Heart J 2009;
30:2493-537.
7. Greiner S, Jud A, Aurich M, Hess A, Hilbel T, Hardt S,
et al.
Reliability of noninvasive assessment of systolic pulmonary
artery pressure by Doppler echocardiography compared to
right heart catheterization: analysis in a large patient
population. J Am Heart Assoc 2014; 3:1-8.
8. Fisher MR, Forfia PR, Chamera E, Housten-Harris T,
Champion HC, Girgis RE, et al. Accuracy of Doppler
echocardiography in the hemodynamic assessment of
pulmonary hypertension. Am J Respir Crit Care Med 2009;
179:615-21.
9. Bland JM, Altman DG. Statistical methods for assessing
agreement between two methods of clinical measurement.
Lancet 1986; 1:307-10.
10. Yock PG, Popp RL. Noninvasive estimation of right ventricular
systolic pressure by Doppler ultrasound in patients with
tricuspid regurgitation. Circulation 1984; 70:657-62.
11. Murua JA, Torres A, Figueroa A. Noninvasive determination of
pulmonary arterial systolic pressure by continuous wave
Doppler. Int J Cardiol 1987; 16:177-84.
12. Currie PJ, Seward JB, Chan KL, Fyfe DA, Hagler DJ, Mair DD,
et al. Continuous wave Doppler determination of right
ventricular pressure: a simultaneous Doppler-catheterization
study in 127 patients. J Am Coll Cardiol 1985; 6:750-6.
13. Berger M, Haimowitz A, Van TA, Berdoff RL, Goldberg E.
Quantitative assessment of pulmonary hypertension in
patients with tricuspid regurgitation using continuous wave
Doppler ultrasound. J Am Coll Cardiol 1985; 6:359-65.
14. D'Alto M, Romeo E, Argiento P, D'Andrea A, Vanderpool R,
Ikram Ahmed, Muhammad Masudul Hassan Nuri, Ahmad Nasiruddin Zakariyya, Syed Mashud Ahmad and Mobasher Ahmed
258 Journal of the College of Physicians and Surgeons Pakistan 2016, Vol. 26 (4): 255-259
Correra A, et al. Accuracy and precision of echocardiography
versus right heart catheterization for the assessment of
pulmonary hypertension. Int J Cardiol 2013; 168:4058-62.
15. Rich JD, Shah SJ, Swamy RS, Kamp A, Rich S. Inaccuracy of
Doppler echocardiographic estimates of pulmonary artery
pressures in patients with pulmonary hypertension:
Implications for clinical practice. Chest 2011; 139:988-93.
16. Kircher BJ, Himelman RB, Schiller NB. Noninvasive estimation
of right atrial pressure from the inspiratory collapse of the
inferior vena cava. Am J Cardiol 1990; 66:493-6.
17. Nagueh SF, Kopelen HA, Zoghbi WA. Relation of mean right
atrial pressure to echocardiographic and Doppler parameters
of right atrial and right ventricular function.
Circulation 1996;
93:1160-69.
18. Brennan JM, Blair JE, Goonewardena S, Ronan A, Shad D,
Vasslwala S,
et al. Reappraisal of the use of inferior vena cava
for estimating right atrial pressure. J Am Soc Echocardiogr
2007; 20:857-61.
19. Simonson JS, Schiller NB. Sonospirometry: a new method for
noninvasive estimation of mean right atrial pressure based on
two-dimensional echocardiographic measurements of the
inferior vena cava during measured inspiration. J Am Coll
Cardiol 1988; 11:557-64.
20. Moreno FL, Hagan AD, Holmen JR, Pryor TA, Strickland RD,
Castle CH. Evaluation of size and dynamics of the inferior vena
cava as an index of right-sided cardiac function.
Am J Cardiol
1984; 53:579-85.
21. Rich S, D'Alonzo GE, Dantzker DR, Levy PS. Magnitude and
implications of spontaneous hemodynamic variability in
primary pulmonary hypertension. Am J Cardiol 1985; 55:159-63.
Doppler echocardiography and right heart catheterization derived pulmonary artery pressures
Journal of the College of Physicians and Surgeons Pakistan 2016, Vol. 26 (4): 255-259
259