The aim of this study was to evaluate right ventricular (RV) regional systolic function and dyssynchrony in patients with pulmonary hypertension (PH) using real-time three-dimensional echocardiography.
Real-time three-dimensional echocardiographic images were acquired to obtain RV regional (inflow, body, and outflow) ejection fraction (EF) and time to minimum systolic volume in 70 patients with PH and 26 normal controls. Pulmonary artery systolic pressure (PASP) and pulmonary vascular resistance measured by echocardiography in all subjects and by right heart catheterization in 17 patients were recorded.
Inflow EF and global EF were significantly lower in patients with PH than in controls ( P < .05). Body EF was significantly decreased in patients with moderate (PASP, 50–69mm Hg) and severe (PASP ≥ 70 mm Hg) PH ( P < .05). Outflow EF was significantly lowered in patients with severe PH ( P < .001). The standard deviation of regional time to minimum systolic volume corrected by heart rate was significantly prolonged in patients with severe PH ( P < .05). Inflow EF and global EF were negatively correlated with PASP ( r = −0.731 and r = −0.769, respectively, P < .001) and with pulmonary vascular resistance ( r = −0.789 and r = −0.801, P < .001).
In patients with PH, RV inflow and global systolic function was impaired in inverse relationship with PASP and pulmonary vascular resistance. RV systolic synchronicity was impaired in patients with severe PH. Evaluation of RV regional systolic function using real-time three-dimensional echocardiography may play a potential role in the noninvasive assessment of the severity of PH.
Right ventricular (RV) function is an important determinant of prognosis in patients with pulmonary hypertension (PH). Regional analysis of RV function is of practical importance in clinical settings because it provides valuable information for deeper understanding of RV function. Echocardiography is the most widely used method in the assessment of RV systolic function for its high availability and reproducibility. Two-dimensional echocardiography was mainly used in the assessment of RV regional systolic function in the longitudinal direction in previous studies. However, the complete RV structure is difficult to display in one two-dimensional view with the inflow and outflow sections perpendicular to each other. Furthermore, longitudinal shortening represents only one direction of RV deformation, and regional transverse wall contraction also provides important information on RV function. Therefore, there are certain limitations to two-dimensional echocardiography in the evaluation of the RV regional systolic function.
Real-time three-dimensional (3D) echocardiography (RT3DE) is a well-developed echocardiographic technique with the advantage of displaying the 3D anatomy of the right ventricle despite its irregular chamber shape, reflecting the effect of both longitudinal and transverse movements on RV systolic function simultaneously. This advantage makes it superior to conventional two-dimensional methods in RV functional assessment. Previous studies have validated the accuracy and reproducibility of RT3DE in the evaluation of RV global volume and systolic function. Quantification of RV global volume and systolic function by RT3DE has been found to be feasible and accurate in patients with PH. Recently, several groups of researchers have focused on assessing RV regional function using RT3DE in ischemic heart disease and repaired tetralogy of Fallot, reporting distinct features for RV regional systolic function in different clinical settings when measured by RT3DE. However, the quantification of RV regional systolic function using RT3DE has not been sufficiently addressed in patients with PH, nor has the relationship between regional systolic function parameters on RT3DE and hemodynamic data been fully explored. Thus, the purposes of the present study were to assess RV regional systolic function and dyssynchrony in patients with PH using RT3DE and to explore the relationship between RV regional systolic functional parameters and RV hemodynamic data.
A total of 70 consecutive patients with PH (22 men; mean age, 41 ± 15 years) were enrolled in this study. Patients were admitted to the Division of Cardiology or the Division of Cardiac Surgery at Zhongshan Hospital of Fudan University from December 2010 to April 2011. Forty-seven patients were diagnosed with pulmonary arterial hypertension (six idiopathic, 27 with atrial septal defects, eight with ventricular septal defects, and six with patent ductus arteriosus), and 23 patients were diagnosed with PH with left-sided valvular heart disease. The diagnosis of PH was made according to the American College of Cardiology and American Heart Association 2009 expert consensus document on PH. Exclusion criteria for patient selection were age < 18 or > 60 years, atrial fibrillation, and inadequate transthoracic echocardiographic window. Twenty-six healthy subjects were included as a control group.
Conventional echocardiography and RT3DE were performed in all patients and controls. According to pulmonary artery systolic pressure (PASP) measured by echocardiography, patients with PH were classified into groups with mild PH (PASP, 40–49 mm Hg), moderate PH (PASP, 50–69 mm Hg), and severe PH (PASP ≥ 70 mm Hg). Seventeen of the 70 patients (24%) underwent right heart catheterization (RHC) <24 hours after echocardiographic image acquisition. RHC-derived hemodynamic parameters including PASP and pulmonary vascular resistance (PVR) were recorded. This study was approved by the ethics committee of Zhongshan Hospital, affiliated to Fudan University.
Conventional echocardiographic images of all subjects were obtained using an iE33 ultrasound machine with an S5-1 transducer (Philips Medical Systems, Andover, MA). Heart rate was recorded and left ventricular ejection fraction (EF) was calculated using the biplane method. RV end-diastolic longitudinal and midcavity diameters were measured from a right ventricle–focused apical four-chamber view. RV end-diastolic and end-systolic areas were measured to calculate RV fractional area change (FAC). PASP was estimated from the peak continuous-wave Doppler velocity of the tricuspid regurgitation (TRV) plus right atrial pressure according to the modified Bernoulli equation, PASP (mm Hg) = 4 × TRV 2 + right atrial pressure, where right atrial pressure was assessed by inferior vena cava diameter and collapsibility. The time-velocity integral was determined from pulsed-wave Doppler in the RV outflow tract. Estimated PVR was calculated using the equation PVR (Wood units) = TRV/time-velocity integral × 10 + 0.16, as suggested by Abbas et al. With the pulsed-wave Doppler sample volume positioned at the lateral margin of the tricuspid annulus in the apical four-chamber view, the peak tissue Doppler systolic velocity of the tricuspid annulus was measured and the RV myocardial performance index (MPI) was calculated as the ratio of isovolumic time divided by ejection time of the right ventricle.
RT3DE was performed in all subjects after the two-dimensional examination using an iE33 ultrasound machine with an X3-1 matrix-array transducer (Philips Medical Systems). RV 3D images were acquired in a full-volume set in the apical four-chamber view. The acquisition of RV 3D images was conducted twice, with a time interval of within 5 min, and the acquired images were stored on a compact disc.
Postprocessing of real-time 3D echocardiographic images was performed using a TomTec workstation with four-dimensional analysis software (EchoView, TomTec Imaging Systems GmbH, Munich, Germany). The software analyzed real-time 3D echocardiographic images in a semiautomatic way. With the observer manually tracing end-diastolic and end-systolic frames in the sagittal, four-chamber, and coronal views obtained from real-time 3D images, the software automatically traced the RV inner border during one heart cycle, recording RV chamber volume changes, calculating RV global end-diastolic volume (EDV), end-systolic volume, stroke volume, and EF, and developing an immediate report. At the same time, the 3D chamber of the right ventricle was divided into three parts, the inflow, body, and outflow compartments, on the basis of the RV surface landmarks by the software automatically ( Figure 1 ). These landmarks were defined at 50% of the distance between the point of the tricuspid annular border, the apical point, and the point of the pulmonary annular border. We converted the digital record of RV regional volumes over one heart cycle into a spreadsheet to form a volume-time curve ( Figure 2 ). The maximum and minimum values of regional volumes were detected as regional EDV and end-systolic volume. Regional stroke volume and EF were then calculated as the difference and the percentage change of EDV and end-systolic volume, respectively. Furthermore, we detected RV regional time to minimum systolic volume (Tmsv) and Tmsv corrected by heart rate (Tmsv%) by reading the digital record of RV regional volumes over one heart cycle to calculate RV systolic dyssynchrony indices, including the standard deviations of Tmsv and Tmsv% in three RV segments.
Using the methods described above, 20 randomly chosen real-time 3D echocardiographic images were analyzed by two observers and by one observer with a minimum time interval of 2 weeks for the measurement of regional EF and dyssynchrony indices. The measurements were performed on the same cardiac cycle of the same image. All readers were blinded to previous measurements. Interobserver and intraobserver variability was assessed using the coefficient of variation (the absolute difference between repeated measurements as a percentage of their mean value) and Bland-Altman analysis.
Statistical analysis was performed using SPSS version 16.0 (SPSS, Inc., Chicago, IL). Categorical data, presented as numbers, were compared using χ 2 tests. Continuous variables are expressed as mean ± SD. Conventional and real-time 3D echocardiographic parameters of RV dimensions and systolic function were compared between groups using Student’s unpaired t tests. Differences among three or more groups were assessed using one-way analysis of variance. Correlations between two variables were evaluated using Pearson’s rank correlation coefficient. All tests were two sided, and P values <.05 were considered statistically significant.
Clinical and Conventional Echocardiographic Characteristics
Conventional echocardiographic images were successfully obtained and analyzed in all subjects. Clinical and conventional echocardiographic characteristics of the study population are summarized in Table 1 . There were no significant differences in age, gender, peak tissue Doppler systolic velocity of the tricuspid annulus, and left ventricular EF between the PH groups and the control group. Heart rates in the moderate and severe PH groups were higher than in the control group. RV end-diastolic midcavity diameter and RV end-systolic area were higher in all PH groups than in the control group. RV end-diastolic longitudinal diameter and RV end-diastolic area were higher in the severe PH group than in the control group. FAC was lower in all PH groups than in the control group and was the lowest in the severe PH group. MPI was prolonged in all the PH groups compared with the control group and changed most significantly in the severe PH group. TRV/time-velocity integral and PVR were higher in all PH groups than in the control group and were highest in the severe PH group.
|Control||Mild PH||Moderate PH||Severe PH|
|Variable||( n = 26)||( n = 12)||( n = 23)||( n = 35)|
|Age (y)||44 ± 11||41 ± 17||44 ± 14||38 ± 16|
|Heart rate (beats/min)||69 ± 11||73 ± 10||78.0 ± 14.7 ∗||86.1 ± 16.7 ∗ † ‡|
|Left ventricular EF (%)||68.2 ± 4.3||65.8 ± 10.5||66.0 ± 8.9||67.1 ± 7.6|
|RV end-diastolic longitudinal diameter (mm)||69.0 ± 5.9||76.9 ± 12.7||74.9 ± 18.3||77.4 ± 9.4 ∗|
|RV end-diastolic midcavity diameter (mm)||31.0 ± 5.7||40.1 ± 9.7 ∗||44.0 ± 13.8 ∗||45.6 ± 8.9 ∗|
|RV end-diastolic area (cm 2 )||15.6 ± 4.8||22.5 ± 9.8||27.0 ± 4.8 ∗||31.4 ± 18.4 ∗|
|RV end-systolic area (cm 2 )||7.5 ± 2.4||13.5 ± 7.6 ∗||14.8 ± 7.9 ∗||20.5 ± 7.6 ∗ † ‡|
|FAC (%)||50.1 ± 8.8||42.4 ± 8.9 ∗||43.5 ± 12.8 ∗||31.8 ± 10.3 ∗ † ‡|
|Peak tissue Doppler systolic velocity of the tricuspid annulus (cm/sec)||13.4 ± 3.9||12.5 ± 2.8||14.3 ± 3.9||12.1 ± 2.8|
|MPI||0.39 ± 0.09||0.52 ± 0.17 ∗||0.54 ± 0.15 ∗||0.68 ± 0.18 ∗ † ‡|
|PASP (mm Hg)||27.7 ± 3.6||43.7 ± 2.7 ∗||58.0 ± 5.7 ∗ †||107.9 ± 26.5 ∗ † ‡|
|TRV/time-velocity integral||0.11 ± 0.02||0.15 ± 0.06 ∗||0.18 ± 0.09 ∗||0.38 ± 0.14 ∗ † ‡|
|PVR (Wood units)||1.26 ± 0.15||1.69 ± 0.58 ∗||1.85 ± 0.99 ∗||4.00 ± 1.44 ∗ † ‡|
Real-Time 3D Echocardiographic Measurements
Real-time 3D echocardiographic images of the right ventricle were successfully analyzed in 68 of the 70 patients (97%) and 25 of the 26 controls (96%). Image quality was inadequate to be analyzed in two patients and one control because of unsatisfactory echocardiographic windows. The mean times for image acquisition and analysis were 3 ± 1 min and 6 ± 2 min, respectively. RV global and regional EFs measured by RT3DE were correlated with FAC, peak tissue Doppler systolic velocity of the tricuspid annulus, and MPI in all subjects ( r = 0.24–0.58, P < .05).
Real-time 3D echocardiographic parameters regarding RV regional systolic function and dyssynchrony are summarized in Table 2 . Compared with controls, inflow EF and global EF were lower in all PH groups ( P < .05), while body EF was decreased in the moderate and severe PH groups ( P < .05) and outflow EF was changed in the severe PH group ( P < .001). Regional and global EFs in the severe PH group were lowest among the three PH groups. Regional and global EDVs in the PH groups were larger than in the controls. The standard deviations of Tmsv% in the mild and moderate PH groups were similar to that in the control group and significantly prolonged in the severe PH group ( P < .05). Figure 2 shows that the three RV compartments reached the minimum regional systolic volume more simultaneously in a normal control than in a patient with severe PH. Differences of other real-time 3D echocardiographic parameters between the patients and control groups were not significant.
|Control||Mild PH||Moderate PH||Severe PH|
|Variable||( n = 25)||( n = 12)||( n = 23)||( n = 33)|
|Body EF (%)||52.9 ± 8.2||47.6 ± 8.7||42.4 ± 5.3 ∗||27.8 ± 9.6 ∗ † ‡|
|Body EDV (mL)||21.7 ± 7.0||54.3 ± 27.8 ∗||52.0 ± 41.7 ∗||57.5 ± 29.4 ∗|
|Outflow EF (%)||55.7 ± 11.1||53.3 ± 8.9||50.55 ± 12.8||39.4 ± 13.9 ∗ † ‡|
|Outflow EDV (mL)||18.4 ± 5.3||33.7 ± 17.3 ∗||38.2 ± 26.0 ∗||34.8 ± 15.6 ∗|
|Inflow EF (%)||66.4 ± 6.8||59.6 ± 7.6 ∗||58.3 ± 8.1 ∗||41.3 ± 11.4 ∗ † ‡|
|Inflow EDV (mL)||38.8 ± 9.8||65.5 ± 33.8 ∗||82.0 ± 41.9 ∗||76.5 ± 27.0 ∗|
|Global EF (%)||60.5 ± 6.4||54.8 ± 6.0 ∗||53.4 ± 7.0 ∗||36.4 ± 9.4 ∗ † ‡|
|Global EDV (mL)||79.1 ± 18.8||153.3 ± 59.5 ∗||178.3 ± 101.6 ∗||168.6 ± 67.9 ∗|
|Standard deviation of Tmsv (msec)||6.3 ± 4.5||8.7 ± 8.4||12.1 ± 12.4||13.6 ± 13.6|
|Standard deviation of Tmsv% (%)||0.7 ± 0.5||1.2 ± 0.9||1.47 ± 1.58||2.0 ± 2.1 ∗|