Assessment of Pulmonary Artery Pressures by Various Doppler Echocardiographic Parameters and its Correlation with Cardiac Catheterization in Patients with Pulmonary Hypertension.

BACKGROUND: Measuring pulmonary artery pressures is a routine index in Doppler echocardiography to diagnose, risk stratify, and prognosticate patients with pulmonary hypertension (PH). There are numerous methods in use to measure it in routine clinical practice.
OBJECTIVE: The objective of this study was to assess the correlation between the commonly used Doppler-derived parameters such as tricuspid regurgitation (TR)-derived systolic pulmonary artery pressure (SPAP), pulmonary regurgitation (PR)-derived mean pulmonary artery pressure (MPAP), and right ventricular outflow tract acceleration time (RVOT AcT) with right heart catheterization (RHC) data which are the gold standard.
MATERIALS AND METHODS: In this analytical study, we prospectively measured echo and angiogram parameters such as TR-derived SPAP, PR-derived MPAP, and RVOT AcT and studied its association with RHC data of thirty patients for a span of 2 years. Right ventricular AcT was also included in the study. Their relationship was displayed using Bland-Altman scatter plots. P < 0.05 was considered as statistically significant.
RESULTS: Although both TR-derived SPAP and PR-derived MPAP had a moderate correlation with RHC-acquired data, the agreement between them was poor. RVOT AcT showed a strong inverse correlation with invasive MPAP.
CONCLUSION: Among the three Doppler methods that were assessed to measure pulmonary pressures, RVOT AcT had a strong correlation with MPAP. RVOT AcT of <80 ms had a high sensitivity to detect severe PH (defined as MPAP >45 mmHg). Hence, it is recommended to include AcT as a routine measure in the armamentarium of echocardiographic parameters used in patients with PH. Copyright:

INTRODUCTION

Pulmonary hypertension (PH) is defined hemodynamically as an increase in mean pulmonary artery pressure (MPAP) ≥25 mmHg at rest as measured by right heart catheterization (RHC).[1] It is a progressive disease in many and can lead to right heart failure and death if untreated. The current guidelines suggest a series of investigations in the assessment of suspected PH patients to diagnose, frame the clinical group of PH, determine the hemodynamic profile, and finally, for prognostication. Among these, two modalities are widely used: Doppler echocardiography (DE) being noninvasive is used as a screening and monitoring tool, whereas RHC is still the gold standard to diagnose and guide treatment in patients with PH.[2]

Systolic pulmonary artery pressure (SPAP) measurement by continuous-wave Doppler using peak tricuspid regurgitation (TR) jet velocity was first described by Yock and Popp.[3] Since then, there had been numerous studies and meta-analyses on the correlation between Doppler- and catheterization-derived pulmonary pressures with varied inferences. Several of them had reported a good correlation between RHC and DE in measuring SPAP.[4] However, a few other studies had reported a poor correlation between them.[5] Data have been presented about severe TR and severe PH affecting the accuracy of SPAP measured by DE.[6]

Measurement of MPAP scores more hemodynamic significance than SPAP and can be noninvasively calculated from the peak pulmonary regurgitant velocity (PRV).[7] Doppler-derived MPAP showed a good correlation with catheterization-derived pressure.[8] Unfortunately, a majority of patients lack an appropriate Doppler profile of pulmonary regurgitation (PR), whereas it can be used in the selected subset of patients with significant PH.

Right ventricular acceleration time (RV AcT), defined as the time to peak flow velocity in milliseconds, is an alternative Doppler technique proposed to estimate pulmonary pressure. It had been reported to have a significant negative correlation with mean pulmonary pressure.[9] The advantage of this measure is that it can be used nonselectively in almost all of the patients screened for PH.

Most of the studies dwelling on the accuracy of Doppler parameters of PH are retrospective, and we could not find a single study that analyzed all three abovesaid measures together in a selected cohort of patients. The purpose of this study was to prospectively evaluate the accuracy of the available Doppler measures in a cohort of patients with PH referred to our center for evaluation/treatment and in whom RHC was performed.

MATERIALS AND METHODS

This study was conducted in the Department of Cardiology, Mahatma Gandhi Medical College and Research Institute, Puducherry, a tertiary health care center. The study was carried out for a span of 2 years from February 2017 to February 2019. Consecutive patients who were referred for evaluation/intervention (percutaneous and surgical) with echo-derived SPAP more than 50 mmHg were asked to participate in the study after providing informed consent. The Institute Human Ethics Committee approved the conduct of our study. The patients who did not have a traceable TR and PR Doppler profile and the patients with severe right ventricular (RV) dysfunction were excluded. Patients included in the study underwent RHC within 6 h of a transthoracic DE.

We evaluated three Doppler methods, namely TR velocity-derived pulmonary artery systolic pressure (PASP), PR-derived MPAP, and pulmonary flow index in patients undergoing RHC. Echocardiograms were performed using a PHILIPS IE33 echocardiogram machine with a 3 MHz transducer by a single cardiologist who was blinded to the catheterization data. Images were acquired in standard views for the primary analysis. Right atrial pressure (RAP) was estimated by recording the inferior vena cava size and its change with respiration as per the American Society of Echocardiography recommendations in the absence of right ventricular outflow tract (RVOT) obstruction.[10] The peak TR velocity of the jet was measured at end-expiration using continuous-wave Doppler in apical four-chamber (A4C) view or in another view where maximum TR jet velocity was obtained.

PASP was calculated with the following equation:[11] PASP = 4 × (peak TR jet velocity) + estimated RAP (in the absence of RVOT obstruction).

Mean pulmonary artery (PA) pressure was estimated with modified Bernoulli equation from the peak diastolic velocity of PR jet obtained from multiple echocardiographic views and adding RAP, MPAP = 4 × (PDPRV) + RAPMPAP = 4 × (peak diastolic PR jet velocity) + estimated RAP.

Similarly, PA diastolic pressure (PADP) was estimated with modified Bernoulli equation from the end-diastolic velocity of PR jet and adding RAP, PADP = 4 × (end-diastolic PR jet velocity) + estimated RAP.

RVOT AcT-derived MPAP was calculated by estimating the RVOT AcT in a short-axis view at the aortic level. It was obtained using pulsed-wave Doppler by placing the sample volume just below the pulmonary cusp in the RVOT. RVOT AcT was defined as the interval between the onset of systolic pulmonary arterial flow and peak flow velocity. RVOT AcT-derived MPAP was calculated using this equation, MPAP = 79−(0.45 × RVOT AcT).

RV functions were assessed using RV fractional area change, tricuspid annular plane systolic excursion (TAPSE), and RV peak systolic annular velocity(S') denoted as RVS'. TAPSE was obtained by placing M mode cursor in the A4C view through the tricuspid annulus and measuring the amount of longitudinal motion of annulus in peak systole. TAPSE <16 mm was considered as abnormal. RVS′ was recorded in A4C view with tissue Doppler mode highlighting the RV free wall and pulse Doppler sample volume being placed in the tricuspid annulus region of RV free wall. The highest systolic velocity was taken as S′. Those with a value of it < 10 cm/s was considered to have RV dysfunction.

RHC was performed without any sedation by an interventionist who was blinded to the echocardiographic data. After achieving venous access either through the antecubital vein or femoral vein, seven French sheaths were secured in place. A multipurpose catheter was taken into the PA, and the PA pressures (systolic, diastolic, and MPAP) were noted. Patients were categorized based on MPAP into three groups. Normal PAP was defined as MPAP <26 mmHg. Mild PH was defined as MPAP from 26 to 35 mmHg, moderate PH as MPAP between 36 and 45 mmHg, while severe PH as defined as MPAP being >45 mmHg.[12]

Statistical analysis

The quantitative data obtained were entered in Microsoft Excel and analyzed using SPSS software version 16.0 (IBM Corp., Armonk, NY, USA). The quantitative measures (echo and angiogram parameters) were summarized as mean and standard deviation/median and interquartile range. Qualitative variables were presented as numbers and percentages. The bivariate relationship between the echocardiographic and angiogram parameters was expressed as Pearson's correlation coefficient (r). Their relationship was displayed using Bland–Altman scatter plots. P < 0.05 was considered as statistically significant.

RESULTS

Fifty-six consecutive RHC were performed on patients with DE-derived SPAP >50 mmHg during the study period. Twenty-three of them did not have a satisfactory PR Doppler profile to calculate MPAP and hence were excluded, and another three had poor image quality to interpret. All these patients underwent an echocardiogram before RHC on the same morning with a mean time lag of 2 h between them. The clinical diagnosis, echocardiographic, and hemodynamic data are listed in Table 1. Three-fourths of the patients were females and 90% were between 20 and 60 years of age. A majority had PH secondary to heart disease and all were in sinus rhythm. Out of 30 patients, 2 had MPAP <25 mmHg, 9 had mild PH (25–35 mmHg), 9 had moderate PH, and 10 had severe PH (>45 mmHg). Our study group had preserved RV function with a mean TAPSE of 19.1 ± 0.97.

Table 1

Clinical, Doppler echocardiography, and hemodynamic profile of study patients

Characteristics	Mean±SD
Age (years)	32.7±14.8
Sex, female (%)	22 (73.3)
Interval between DE and RHC (h)	2
Doppler echo data (mmHg)
TR Vmax, m/s	4±0.5
TR PG	66.7±16.8
TR-derived SPAP	75.2±17.4
RAP	8.5±2.3
Peak PR velocity	2.4±0.6
PR-derived MPAP	48.6±10.9
RVOT ACT (m/s)	71.6±23.1
RVOT ACT-derived MPAP	46.5±10.4
Diagnosis
Rheumatic heart disease	13 (43.3)
Atrial septal defect	9 (30)
Patent ductus arteriosus	2 (6.7)
Primary pulmonary arterial hypertension	5 (16.7)
Cor. pulmonale	1 (3.3)
Right heart catheterization data (mmHg)
SPAP	67.3±23.5
DPAP	30.7±14.8
MPAP	43.3±14.6

DE: Doppler echocardiography, RHC: Right heart catheterization, TR Vmax: Tricuspid regurgitation jet maximum velocity, TR PG: Tricuspid regurgitation pressure gradient, SPAP: Systolic pulmonary artery pressure, DPAP: Diastolic pulmonary artery pressure, MPAP: Mean pulmonary artery pressure, RAP: Right atrial pressure, PR: Pulmonary regurgitation, RVOT ACT: Right ventricular outflow tract acceleration time. SD: Standard deviation

Accuracy of echocardiographic methods

Tricuspid regurgitation-derived systolic pulmonary artery pressure

Linear regression analysis of SPAP arrived by DE and RHC [Figure 1a] showed a moderate correlation (r = 0.403, P = 0.027) but poor agreement. The Bland–Altman analysis revealed a positive bias of 7.9 mmHg with 95% limits of agreement ranging from − 37.0 to 52.84 [Figure 1b]. Only 16.7% (n = 5) of the Doppler estimates were accurate (defined as SPAP estimate within 10 mmHg of the value from catheterization). A total of 22 (73.3%) patients had more than 20% difference between DE- and RHC-derived SPAP values. Out of these 22, 16 (72.7%) Doppler-derived values were an overestimate and 6 (27.3%) were an underestimate.

Figure 1

(a) Linear regression analysis plot of echocardiography and right heart catheterization of systolic pulmonary artery pressure (n = 30, r = 0.403, P= 0.027). (b) Bland–Altman plot of systolic pulmonary artery pressure measured by echocardiography and right heart catheterization

Given our inclusion criteria, all patients had TR. The patients were grouped into three strata based on TR severity as mild or less, moderate, and severe.[13] One-third of our patients (n = 10) had severe TR. Subgroup analysis did not show any improvement in correlation coefficients. Severe PH is defined as SPAP >65 mmHg by DE. A cutoff value of Doppler-estimated SPAP >65 mmHg (n = 22) to detect severe PH (defined invasively as MPAP >45 mmHg) was accurate only in 41% (n = 9) of the study participants.

Pulmonary regurgitation-derived mean pulmonary artery pressure

Data obtained depicted only a moderate correlation (r = 0.429, P = 0.018) between invasive MPAP and 4 PRV2+ RAP by linear regression analysis [Figure 2a]. The Bland–Altman analysis showed a poor agreement with a bias of 5.3 mmHg with 95% limits of agreement ranging from −24.0 to 34.7 mmHg [Figure 2b]. Only 6 out of 30 (20%) Doppler estimates were accurate. A total of 24 (80%) patients had more than 20% difference between DE- and RHC-derived MPAP values. More than 20% overestimation by Doppler was noted in 16 patients (61.5%) and >20% underestimation in 8 patients (30.8%). Since we did not record the severity of PR, further subgroup analysis could not be performed similarly to TR.

Figure 2

(a) Linear regression analysis plot of echocardiography and right heart catheterization of mean pulmonary artery pressure (n = 30, r = 0.429, P = 0.018). (b) Bland–Altman plot of mean pulmonary artery pressure measured by echocardiography using pulmonary regurgitation velocity and right heart catheterization

Right ventricular outflow tract acceleration time

The heart rate in the study group ranged between 60 and 110 beats/min. AcT of RVOT showed a significant inverse correlation with invasive MPAP (r = −0.671, P < 0.001) by regression analysis [Figure 3]. RVOT AcT <80 ms correctly identified severe PH (invasive MPAP >45 mmHg) in 90% of the study sample. However, 3 out of 20 patients with MPAP <45 mmHg too had RVOT AcT <80 ms. Only one patient out of these 30 patients with an AcT >80 ms had severe PH by RHC.

Figure 3

Correlation between mean pulmonary artery pressure and acceleration time (n= 30, r= −0.671, P< 0.001)

DISCUSSION

Estimation of pulmonary pressures using DE is widely applied given its easy availability and noninvasive nature. Although recommended by many guidelines, few studies have raised doubts about its reliability in measuring SPAP.[514] The well-designed prospective studies were performed in the last decade by Fisher et al.[5](n = 65) on consecutive patients with PH, with a mean SPAP of 41.4 mmHg, and Rich et al. (n = 183)[14] where simultaneous measurements by DE and RHC were done. Both the studies concluded that DE estimates of SPAP were inaccurate and should not be relied upon to diagnose PH. In our study, SPAP arrived by DE and RHC showed a moderate correlation but poor agreement.

There are several methods described to measure pulmonary pressures using DE. However, the most commonly used in the descending order of frequency are TR jet gradient-derived SPAP, PR peak velocity-derived MPAP, and RVOT AcT-derived MPAP.[15] To the best of our knowledge, ours is the first prospective study to collectively compare the accuracy of all these Doppler-derived parameters with catheterization data in the same cohort.

Both DE and RHC were performed on the same day within 4-h gap with a mean time of 120 min. No specific therapy was instituted during this time lag. Despite these optimized settings, Doppler estimates of SPAP differed significantly from the invasive ones. The Bland–Altman analysis depicted the tendency to both overestimate and underestimate SPAP. Although overestimation was more frequent, the magnitude of underestimation was large (24.9 mmHg vs. 20.9 mmHg).

The initial interest for Doppler-estimated SPAP was driven by several small studies[3] in which results were obtained in a research setting,[16] those conditions that are not easy to replicate in our routine schedule. In a meta-analysis by Finkelhor et al.,[17] the overall weighted mean correlation coefficient of 32 studies included was 0.68 and it improved to 0.79 for prospective studies. Although good correlation coefficients were reported, it did not translate to a good agreement,[18] and hence, frequent misclassification of PH groups using DE has been documented.[19] The present study was conducted to investigate the relationship between RHC and Doppler measures such as TR-derived SPAP, PR-derived MPAP, and RVOT AcT in a cohort of patients with PH.

In studies by Fei et al.,[6] and Ozpelit et al.,[20] severe TR and severe PH independently influenced the accuracy of Doppler-derived SPAP. They postulated that broad and laminar flow is noted in severe TR and hence TR pressure gradient measured using the simplified Bernoulli equation may be erroneous as the inertial component cannot be avoided. The other explanation could be an overestimation of RAP due to congested vena cava in severe TR. In our study, we did not aim to identify the technical factors responsible for these discrepancies.

In spite of data showing a good correlation between RVOT AcT and MPAP[9] and its logarithm (r = 0.82 and 0.88, respectively), reluctance persists among echocardiographers to include this in routine clinical practice. One probable reason to be so is the excessive utility of measuring SPAP from TR jet. Our data show that among the studied parameters, AcT has the best correlation. According to the study by Dabestani et al., the correction for heart rate is not recommended. Hence, we did not correct it for heart rate due to two reasons: first, all our patients had a rate between 60 and 110, and second, the correction for RV ejection time did not improve its diagnostic accuracy.[21]

RVOT AcT <80 ms had a high sensitivity to detect severe PH (MPAP >45 mmHg). Our study adds this important piece of information which is critical as the guidelines recommend invasively measured MPAP as the gold standard in diagnosing and treating PH. These data are in line with the study by Tossavainen et al.[22] where AcT <90 ms was a strong noninvasive predictor of a pulmonary vascular resistance of >3 wood units. It could also differentiate pre- and postcapillary hypertension, thus aiding in the classification of PH.

Echocardiograms were performed by a qualified cardiologist with 5 years of experience and who reports more than 7000/year, thus meeting the standard of care. The reasons for TR- and PR-derived pressures to be inaccurate compared to AcT are numerous. There are many technical glitches inherent to the Doppler method in measuring TR Vmax such as incomplete spectral wave envelope, maximal velocity boundary artifacts (fringes), the spectral gain being too soft and setting high-velocity range[4]. One third of included patients had only mild TR or less rendering the reliance on TR Vmax unfeasible.[23] Whereas, there are not many sources of error in measuring AcT, if RV function is preserved as corrections for heart rate are not necessary as per published studies.[24] Careful alignment of the beam with the long axis of the RVOT is the only essential step in the use of AcT, thus potentially reducing the sources of error.

Our study is subject to the limitations inherent with a small sample size, but we intended to do a prospective study in routine clinical circumstances rather than analyzing retrospective data. In addition, the measurements were derived in a nonsimultaneous way which may not be ideal.

CONCLUSION

This study assessed three Doppler methods of estimating PA pressures. RVOT AcT showed the best correlation with invasive pressures among the studied parameters and a value of <80 ms and had a high sensitivity to detect severe PH by RHC. Measuring AcT is feasible more often than TR Vmax or PR-derived PA pressures, and hence, it is recommended to include it in routine practice.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.