Comprehensive Assessment of Right Ventricular Function in Patients with Pulmonary Hypertension with Global Longitudinal Peak Systolic Strain Derived from Multiple Right Ventricular Views Sudarshan Rajagopal, MD, PhD, Daniel E. Forsha, MD, Niels Risum, MD, PhD, Christoph P. Hornik, MD, Abby D. Poms, RRT, Terry A. Fortin, MD, Victor F. Tapson, MD, Eric J. Velazquez, MD, Joseph Kisslo, MD, and Zainab Samad, MD, Durham, North Carolina; Hvidovre, Denmark; Beverly Hills, California
Background: Right ventricular (RV) function is a strong predictor of mortality in pulmonary hypertension (PH), but two-dimensional (2D) echocardiography–derived assessments of RV function that could aid in risk assessment and management of patients with PH are of limited utility. RV longitudinal peak systolic strain (RVLS) derived from 2D speckle-tracking echocardiography is a relatively novel method for quantifying RV function but typically is derived from a single apical four-chamber view of the right ventricle and may have inherent limitations. The objective of this study was to determine the utility of regional and global RVLS calculated from multiple views of the right ventricle to comprehensively assess RV function in a cohort of patients with PH. Methods: Regional and global RVLS were obtained from multiple views of the right ventricle (centered on the right ventricle–focused apical position) in 40 patients with PH, defined as a mean pulmonary artery pressure $ 25 mm Hg, most of whom also had pulmonary capillary wedge pressures # 15 mm Hg and were thus defined as having pulmonary arterial hypertension. This was compared with other 2D echocardiography– derived parameters of RV function and functional parameters. Results: Global RVLS calculated from multiple views had a superior correlation with 6-min walk distance compared with other parameters of RV function, including tricuspid annular plane systolic excursion, RV myocardial performance index, and fractional area change. Although global RVLS calculated from multiple views displayed a similar correlation with 6-min walk distance as global RVLS calculated from a single fourchamber view, analysis of regional strains provided by multiple views identified distinct patterns of RV dysfunction, consisting of global, free wall, or septal dysfunction, that were associated with specific clinical characteristics. Conclusions: Global RVLS derived from multiple right ventricle–focused views yields a comprehensive quantitative assessment of regional and global RV function that correlates moderately with functional parameters and may be useful in the assessment of PH. Distinct patterns of regional RV dysfunction are associated with different clinical characteristics. (J Am Soc Echocardiogr 2014;-:---.) Keywords: Pulmonary hypertension, Right ventricular function, Speckle-tracking echocardiography, Strain, 6-minute walk distance
From the Department of Medicine, Duke University Medical Center, Durham, North Carolina (S.R., A.D.P., T.A.F., E.J.V., J.K., Z.S.); Department of Pediatrics, Duke University Medical Center, Durham, North Carolina (D.E.F., C.P.H.); Department of Cardiology, Hvidovre Hospital, Hvidovre, Denmark (N.R.); and Department of Medicine, Cedars-Sinai Medical Center, Beverly Hills, California (V.F.T.). Dr Samad received research grant support from the American Society of Echocardiography (Morrisville, NC). Reprint requests: Sudarshan Rajagopal, MD, PhD, Duke University Medical Center, Box 3126, Durham, NC 27710 (E-mail: [email protected]. edu). 0894-7317/$36.00 Copyright 2014 by the American Society of Echocardiography. http://dx.doi.org/10.1016/j.echo.2014.02.001
Pulmonary arterial hypertension (PAH), a subset of pulmonary hypertension (PH), is a disease of the pulmonary vasculature that leads to right ventricular (RV) dysfunction and failure. RV function is of critical importance in the prognosis of PAH. Hemodynamic parameters, including right atrial pressure and cardiac index, both reflecting RV function,1 and pro–brain natriuretic peptide, associated with right heart failure, are all important prognostic biomarkers.2 Patients with PAH are routinely followed using echocardiography, but the conventional twodimensional (2D) echocardiographic assessment of the right ventricle does not include a clear quantitative assessment of function but rather a quantification of RV dimensions and a qualitative assessment of contractility.3 This is due both to the complex geometry of the right ventricle and a poor understanding of its mechanical functioning compared with that of the left ventricle.4 Quantitative assessments of 1
2 Rajagopal et al
RV function would aid in clinical decision making, as echocardioAP4 = Four-chamber right graphic assessments of RV funcventricular equivalent view tion are routinely used to guide the level of aggressiveness of AP3 = Three-chamber right PAH therapy.5 ventricular equivalent view To this end, a number of 2D AP2 = Two-chamber right echocardiography–derived paventricular equivalent view rameters for RV functional assessFAC = Fractional area change ment, such as tricuspid annular plane systolic excursion (TAPSE), LV = Left ventricular RV fractional area change (FAC), MPI = Myocardial and RV myocardial performance performance index index (MPI), have been proposed, but all have inherent PA = Pulmonary artery strengths and weaknesses3 PAAT = Pulmonary artery (Figure 1). RV longitudinal peak acceleration time systolic strain (RVLS) is a relatively novel approach for quantiPAH = Pulmonary arterial hypertension fying RV function and, when calculated using 2D specklePH = Pulmonary hypertension tracking echocardiography, is REVEAL = Registry to angle independent and yields a Evaluate Early and Long-Term quantitative assessment of RV Pulmonary Arterial systolic function. Moreover, Hypertension Disease recent studies have demonstrated Management relationships between RVLS and RV = Right ventricular outcomes and response to therapy in patients with PAH.6-9 RVLS = Right ventricular However, a continued limitation longitudinal peak systolic of RVLS from a single RV view strain is that it does not yield a truly 6MWD = 6-min walk distance ‘‘global’’ view of RV function.3 Recently, an approach for a global TAPSE = Tricuspid annular assessment of RV function has plane systolic excursion been developed that uses 3 right 2D = Two-dimensional ventricle–focused views analogous to their apical views of the left ventricle,10 to allow a full reconstruction of the right ventricle in an 18-segment or a 17-segment model (Figure 2) with a calculation of a true ‘‘global RVLS.’’ In this work, we compared global RVLS between normal controls and patients with PH, correlated global RVLS with other parameters of RV function (including TAPSE, MPI, and FAC) and with functional status as assessed by 6-min walk distance (6MWD), and identified patterns of regional RV dysfunction in patients with PH. Abbreviations
Journal of the American Society of Echocardiography - 2014
while one underwent repeat heart catheterization that demonstrated no evidence of PH on no medical therapy. The remaining 40 patients with PH, along with a group of 40 previously described controls,10 were included in our final analysis. This normal population was a cohort of young volunteers (mean age, 28.9 6 9.1 years) without any history of cardiovascular disease, with a minimum age of 18 years at the time of the study, normal echocardiographic findings, including anatomy, left ventricular (LV) ejection fraction $ 50%, RV FAC $ 35%, and TAPSE $ 16 mm. Exclusion criteria for the control population included any abnormal echocardiographic findings or prolonged QRS duration. Subjects used as normal controls provided informed consent for research echocardiography. For patients with PH whose images were obtained during clinically indicated echocardiographic assessments, a waiver of consent was approved for the Duke Echocardiography Lab Database by the Duke University Medical Center Institutional Review Board. This study was approved by the Duke University Medical Center Institutional Review Board. Echocardiography All echocardiographic studies were performed using a Vivid E9 scanner with a 3.5-MHz probe (GE Vingmed Ultrasound AS, Horten, Norway). A full standard echocardiographic examination, including grayscale images optimized for 2D strain analysis (50–90 frames/sec) including three right ventricle–focused apical views,10 was performed. The three apical RV views were equivalent to the imaging planes of the twochamber, three-chamber, and four-chamber LV apical views with the transducer angled rightward (Figure 1). View optimization often required repositioning of the transducer toward the left anterior axillary line. The resulting four-chamber-equivalent view has four chambers (right atrium and ventricle, left ventricle and atrium), the twochamber-equivalent view has three (right atrium, right ventricle, and RV outflow tract), and the three-chamber equivalent view has two chambers (right atrium and ventricle) (Figures 2A–2C). All echocardiographic examinations were performed either on the same day as a 6-min walk test or within 3 months of 6-min walk test in patients with PH. Six-minute walk tests were not done in three patients with PH and were not performed in any of the normal controls. Right heart catheterization was typically not performed near the time of echocardiography, except in those patients with new diagnoses of PH. Offline analysis was performed using EchoPAC PC version BT11 (GE Vingmed Ultrasound AS) by a single experienced reader, and analysis was confirmed by a separate experienced reader; interreader variability for this methodology has been shown to be low, with intraobserver variability (2 6 6%) and interobserver variability (1 6 9%) for global peak strain (percentage mean difference [bias] 6 coefficient of variation).10
METHODS Study Population Clinically indicated echocardiographic studies incorporating the three views of the right ventricle (Figure 1) were performed on 42 consecutive patients at Duke University Medical Center (Durham, NC) with diagnoses of PH, defined as a mean pulmonary artery (PA) pressure $ 25 mm Hg, between May and December 2012. These patients had diagnoses of PAH (mean PA pressure $ 25 mm Hg and PCWP # 15 mm Hg),11 exercise-induced PAH (normal mean PA pressure at rest but mean PA pressure $ 30 mm Hg with exercise),12 or PH out of proportion to diastolic dysfunction (mean PA pressure $ 25 mm Hg, PCWP > 15 mm Hg, and pulmonary vascular resistance significantly higher than 3 Wood units).11 Of these patients, one did not have images of sufficient quality for strain calculations,
2D Echocardiography–Derived Parameters of RV Function TAPSE was determined from an M-mode through the lateral tricuspid annulus by calculating the amount of longitudinal motion of the annulus at peak systole13 (Figure 1A). PA acceleration time (PAAT) was calculated from a spectral Doppler image obtained by placing a pulsed Doppler sample volume at the pulmonary valve annulus. PAAT was calculated as the time from the onset of systolic pulmonary arterial flow to peak flow velocity14 (Figure 1B). RV MPI was calculated as the RV isovolumic time divided by the ejection time using the pulsed Doppler method.3 Isovolumic time was calculated as the duration of tricuspid regurgitation from continuous-wave Doppler across the tricuspid valve minus the ejection time from a single representative beat. Ejection time was calculated as the duration of RV
Journal of the American Society of Echocardiography Volume - Number -
Rajagopal et al 3
Figure 1 Common 2D echocardiographic approaches used in the assessment of RV function and PH: (A) TAPSE, (B) PAAT, (C) MPI, and (D) FAC. For full details, refer to ‘‘Methods.’’ EDA, End-diastolic area; ESA, end-systolic area; ET, ejection time; TRd, duration of tricuspid regurgitation.
Figure 2 Right ventricle–focused apical views used for calculation of global longitudinal strain: (A) AP4, (B) AP2, and (C) AP3 equivalent views. Walls and chambers are labeled for clarity. Calculation of strain in these views allowed the construction of a 17-segment model of the right ventricle as shown (D). ANT, Anterior; ANT-SEPT, anteroseptal; EF, ejection fraction; GS, global strain; INF-SEPT, inferoseptal; LA, left atrium; LV, left ventricle; POST, posterior; RA, right atrium; RV, right ventricle; RVOT, right ventricular outflow tract; SEPT, septal.
outflow on pulsed Doppler across the RV outflow tract from a single representative beat (Figure 1C). Care was taken to use beats with similar RR intervals to minimize errors in calculation. RV FAC was calculated as ([RV end-diastolic area RV end-systolic area]/RV end-diastolic area) 100. The RV endocardium was traced in systole and diastole from the annulus, along the free wall, to the apex and back along the interventricular septum using the apical four-
chamber view. Attempts were made to trace the free wall beneath trabeculations (Figure 1D). RVLS Two-dimensional speckle-tracking echocardiographic analysis was performed from the right ventricle–focused four-chamber RV
4 Rajagopal et al
equivalent (AP4), three-chamber RV equivalent (AP3), and twochamber RV equivalent (AP2) views as previously described10 using EchoPAC PC version BT11, which has previously been validated for LV strain10 (Figures 2 and Supplementary Figure1; available at www.onlinejase.com). Images were optimized for strain analysis by decreasing depth to improve resolution and decreasing sector width to improve frame rate to 50 to 90 frames/sec. The reference point for a single cardiac cycle was placed at the beginning of the QRS complex. Pulmonic valve closure was determined from the pulsed-wave Doppler profile of the RVoutflow tract. For 2D speckle-tracking echocardiography, the endocardial border was traced in end-systole, and the region of interest was adjusted to exclude the pericardium. The quality of the speckle-tracking was assessed by EchoPAC and confirmed visually from 2D images and from the strain traces. Segments with persistent inadequate tracking despite attempts at improving border definition and region of interest were excluded from analysis (34 of 720 segments [4.7%]). RVLS was calculated for each region in these views, allowing calculation of a 17-segment bull’s-eye plot (Figure 2D) and global RVLS. In this method, three right ventricle–focused apical views equivalent to the four-chamber, two-chamber, and three-chamber LV views (termed the AP4, AP2, and AP3 views in this study) are obtained by translation of the probe medially (Figure 2). These views allow a comprehensive reconstruction of the RV chamber in a 17-segment bull’s-eye plot. For one patient, significant differences in RR interval between the AP2, AP3, and AP4 views precluded a calculation of global RVLS and a bull’s-eye plot. Examples of strain curves and bull’s-eye plots from a normal control and a patient with PH are shown in Supplementary Figure 1. The longitudinal strain of the RV free wall was calculated as the average of the free wall strains and the longitudinal strain of the RV septum was calculated as the average of the septal strains in the three RV views. All strain and other 2D echocardiography–derived parameter analyses were performed blinded to 6MWD and other clinical data. Statistical Analysis Unless otherwise stated, specified data are presented as mean 6 SD or as percentages. We compared 2D echocardiography–derived parameters (regional and global RVLS, TAPSE, FAC, MPI, and PAAT) between patients with PH and controls using two-tailed, unpaired MannWhitney tests. Correlations between pairs of 2D echocardiography– derived assessments of RV function were evaluated using Pearson’s correlation coefficients. Correlations between 2D echocardiography– derived RV functional parameters, RVLS, and 6MWD were examined using linear regression models. Clustering analysis was used to identify patterns of regional RVLS. Hierarchical clustering of the 17-segment regional RVLS arrays from 39 patients in whom global longitudinal strain was performed in JMP 10 (SAS Institute Inc, Cary, NC). Missing values were imputed as either the average of all other segments or, if there were significant differences in strains between different segments, the average of its nearest neighbors. Data were standardized by their segment means and standard deviations, and distances were calculated using Ward’s minimum variance method. A visual assessment of the hierarchical clusters to assess similarities between different patterns was used to determine the cutoff point for the number of clusters. We compared clinical characteristics between the different clusters of regional RV strain using one-way analysis of variance followed by pairwise multiple comparison tests. We used Bonferroni’s method to correct for multiple testing. Statistical analyses were performed using Prism 5.0 (GraphPad Software, San Diego, CA).
Journal of the American Society of Echocardiography - 2014
Table 1 Characteristics of study population (n = 40) Characteristic
Clinical Women Age (y) BMI (kg/m2) Symptom/disease duration (y) WHO FC I II III IV 6MWD (m) 6MWD (% predicted) REVEAL risk score PH cause Group 1 Idiopathic Connective tissue disease Hereditary Congenital heart disease PVOD HIV Portopulmonary Out-of-proportion DD (group 2) Significant lung disease (group 3) CTEPH (group 4) Sarcoidosis (group 5) Exercise-induced (other) Medications ERA PDE5 inhibitor Prostacyclin, inhaled Prostacyclin, IV Combination therapy None Other studies Serum creatinine (mg/dL) NT-proBNP (ng/L) DLCO Most recent catheterization RA pressure (mm Hg) Cardiac index (L/min/m2) Mean PA pressure (mm Hg) PVR (Wood units)
Value
34 (85%) 58 6 14 30.3 6 8.4 4 (2–7) 4 (10%) 25 (63%) 10 (25%) 1 (3%) 376 6 130 75 6 23 7.7 6 2.6 31 (78%) 14 (35%) 7 (18%) 4 (10%) 2 (5%) 2 (5%) 1 (2.5%) 1 (2.5%) 2 (5%) 2 (5%) 3 (7.5%) 1 (2.5%) 1 (2.5%) 23 (58%) 16 (40%) 7 (18%) 8 (20%) 21 (53%) 6 (15%) 1.0 6 0.4 490 (124–1,232) 56 6 27 11 6 6 2.4 6 0.7 50 6 15 9.5 6 5.3
BMI, Body mass index; CTEPH, Chronic thromboembolic PH; DLCO, diffusion capacity for carbon monoxide; DD, diastolic dysfunction; ERA, endothelin receptor antagonist; FC, functional class; HIV, human immunodeficiency virus; IV, intravenous; PDE5, phosphodiesterase 5; NT-proBNP, N-terminal pro-brain natriuretic peptide; PVOD, pulmonary veno-occlusive disease; PVR, pulmonary vascular resistance; RA, right atrial; WHO, World Health Organization. Data are expressed as mean 6 SD, median (interquartile range), or number (percentage).
RESULTS Characteristics of the Study Population Baseline characteristics of the PH cohort are presented in Table 1. As shown, the demographics of this population are similar to registries of
Rajagopal et al 5
Journal of the American Society of Echocardiography Volume - Number -
Table 2 Global and regional RVLS from the PH cohort compared with a control population of young individuals with no significant medical problems10 Variable
Patients with PH (n = 40)
Controls (n = 41)
P
15.6 6 4.6
23.8 6 2.3
Background: Right ventricular (RV) function is a strong predictor of mortality in pulmonary hypertension (PH), but two-dimensional (2D) echocardiography–derived assessments of RV function that could aid in risk assessment and management of patients with PH are of limited utility. RV longitudinal peak systolic strain (RVLS) derived from 2D speckle-tracking echocardiography is a relatively novel method for quantifying RV function but typically is derived from a single apical four-chamber view of the right ventricle and may have inherent limitations. The objective of this study was to determine the utility of regional and global RVLS calculated from multiple views of the right ventricle to comprehensively assess RV function in a cohort of patients with PH. Methods: Regional and global RVLS were obtained from multiple views of the right ventricle (centered on the right ventricle–focused apical position) in 40 patients with PH, defined as a mean pulmonary artery pressure $ 25 mm Hg, most of whom also had pulmonary capillary wedge pressures # 15 mm Hg and were thus defined as having pulmonary arterial hypertension. This was compared with other 2D echocardiography– derived parameters of RV function and functional parameters. Results: Global RVLS calculated from multiple views had a superior correlation with 6-min walk distance compared with other parameters of RV function, including tricuspid annular plane systolic excursion, RV myocardial performance index, and fractional area change. Although global RVLS calculated from multiple views displayed a similar correlation with 6-min walk distance as global RVLS calculated from a single fourchamber view, analysis of regional strains provided by multiple views identified distinct patterns of RV dysfunction, consisting of global, free wall, or septal dysfunction, that were associated with specific clinical characteristics. Conclusions: Global RVLS derived from multiple right ventricle–focused views yields a comprehensive quantitative assessment of regional and global RV function that correlates moderately with functional parameters and may be useful in the assessment of PH. Distinct patterns of regional RV dysfunction are associated with different clinical characteristics. (J Am Soc Echocardiogr 2014;-:---.) Keywords: Pulmonary hypertension, Right ventricular function, Speckle-tracking echocardiography, Strain, 6-minute walk distance
From the Department of Medicine, Duke University Medical Center, Durham, North Carolina (S.R., A.D.P., T.A.F., E.J.V., J.K., Z.S.); Department of Pediatrics, Duke University Medical Center, Durham, North Carolina (D.E.F., C.P.H.); Department of Cardiology, Hvidovre Hospital, Hvidovre, Denmark (N.R.); and Department of Medicine, Cedars-Sinai Medical Center, Beverly Hills, California (V.F.T.). Dr Samad received research grant support from the American Society of Echocardiography (Morrisville, NC). Reprint requests: Sudarshan Rajagopal, MD, PhD, Duke University Medical Center, Box 3126, Durham, NC 27710 (E-mail: [email protected]. edu). 0894-7317/$36.00 Copyright 2014 by the American Society of Echocardiography. http://dx.doi.org/10.1016/j.echo.2014.02.001
Pulmonary arterial hypertension (PAH), a subset of pulmonary hypertension (PH), is a disease of the pulmonary vasculature that leads to right ventricular (RV) dysfunction and failure. RV function is of critical importance in the prognosis of PAH. Hemodynamic parameters, including right atrial pressure and cardiac index, both reflecting RV function,1 and pro–brain natriuretic peptide, associated with right heart failure, are all important prognostic biomarkers.2 Patients with PAH are routinely followed using echocardiography, but the conventional twodimensional (2D) echocardiographic assessment of the right ventricle does not include a clear quantitative assessment of function but rather a quantification of RV dimensions and a qualitative assessment of contractility.3 This is due both to the complex geometry of the right ventricle and a poor understanding of its mechanical functioning compared with that of the left ventricle.4 Quantitative assessments of 1
2 Rajagopal et al
RV function would aid in clinical decision making, as echocardioAP4 = Four-chamber right graphic assessments of RV funcventricular equivalent view tion are routinely used to guide the level of aggressiveness of AP3 = Three-chamber right PAH therapy.5 ventricular equivalent view To this end, a number of 2D AP2 = Two-chamber right echocardiography–derived paventricular equivalent view rameters for RV functional assessFAC = Fractional area change ment, such as tricuspid annular plane systolic excursion (TAPSE), LV = Left ventricular RV fractional area change (FAC), MPI = Myocardial and RV myocardial performance performance index index (MPI), have been proposed, but all have inherent PA = Pulmonary artery strengths and weaknesses3 PAAT = Pulmonary artery (Figure 1). RV longitudinal peak acceleration time systolic strain (RVLS) is a relatively novel approach for quantiPAH = Pulmonary arterial hypertension fying RV function and, when calculated using 2D specklePH = Pulmonary hypertension tracking echocardiography, is REVEAL = Registry to angle independent and yields a Evaluate Early and Long-Term quantitative assessment of RV Pulmonary Arterial systolic function. Moreover, Hypertension Disease recent studies have demonstrated Management relationships between RVLS and RV = Right ventricular outcomes and response to therapy in patients with PAH.6-9 RVLS = Right ventricular However, a continued limitation longitudinal peak systolic of RVLS from a single RV view strain is that it does not yield a truly 6MWD = 6-min walk distance ‘‘global’’ view of RV function.3 Recently, an approach for a global TAPSE = Tricuspid annular assessment of RV function has plane systolic excursion been developed that uses 3 right 2D = Two-dimensional ventricle–focused views analogous to their apical views of the left ventricle,10 to allow a full reconstruction of the right ventricle in an 18-segment or a 17-segment model (Figure 2) with a calculation of a true ‘‘global RVLS.’’ In this work, we compared global RVLS between normal controls and patients with PH, correlated global RVLS with other parameters of RV function (including TAPSE, MPI, and FAC) and with functional status as assessed by 6-min walk distance (6MWD), and identified patterns of regional RV dysfunction in patients with PH. Abbreviations
Journal of the American Society of Echocardiography - 2014
while one underwent repeat heart catheterization that demonstrated no evidence of PH on no medical therapy. The remaining 40 patients with PH, along with a group of 40 previously described controls,10 were included in our final analysis. This normal population was a cohort of young volunteers (mean age, 28.9 6 9.1 years) without any history of cardiovascular disease, with a minimum age of 18 years at the time of the study, normal echocardiographic findings, including anatomy, left ventricular (LV) ejection fraction $ 50%, RV FAC $ 35%, and TAPSE $ 16 mm. Exclusion criteria for the control population included any abnormal echocardiographic findings or prolonged QRS duration. Subjects used as normal controls provided informed consent for research echocardiography. For patients with PH whose images were obtained during clinically indicated echocardiographic assessments, a waiver of consent was approved for the Duke Echocardiography Lab Database by the Duke University Medical Center Institutional Review Board. This study was approved by the Duke University Medical Center Institutional Review Board. Echocardiography All echocardiographic studies were performed using a Vivid E9 scanner with a 3.5-MHz probe (GE Vingmed Ultrasound AS, Horten, Norway). A full standard echocardiographic examination, including grayscale images optimized for 2D strain analysis (50–90 frames/sec) including three right ventricle–focused apical views,10 was performed. The three apical RV views were equivalent to the imaging planes of the twochamber, three-chamber, and four-chamber LV apical views with the transducer angled rightward (Figure 1). View optimization often required repositioning of the transducer toward the left anterior axillary line. The resulting four-chamber-equivalent view has four chambers (right atrium and ventricle, left ventricle and atrium), the twochamber-equivalent view has three (right atrium, right ventricle, and RV outflow tract), and the three-chamber equivalent view has two chambers (right atrium and ventricle) (Figures 2A–2C). All echocardiographic examinations were performed either on the same day as a 6-min walk test or within 3 months of 6-min walk test in patients with PH. Six-minute walk tests were not done in three patients with PH and were not performed in any of the normal controls. Right heart catheterization was typically not performed near the time of echocardiography, except in those patients with new diagnoses of PH. Offline analysis was performed using EchoPAC PC version BT11 (GE Vingmed Ultrasound AS) by a single experienced reader, and analysis was confirmed by a separate experienced reader; interreader variability for this methodology has been shown to be low, with intraobserver variability (2 6 6%) and interobserver variability (1 6 9%) for global peak strain (percentage mean difference [bias] 6 coefficient of variation).10
METHODS Study Population Clinically indicated echocardiographic studies incorporating the three views of the right ventricle (Figure 1) were performed on 42 consecutive patients at Duke University Medical Center (Durham, NC) with diagnoses of PH, defined as a mean pulmonary artery (PA) pressure $ 25 mm Hg, between May and December 2012. These patients had diagnoses of PAH (mean PA pressure $ 25 mm Hg and PCWP # 15 mm Hg),11 exercise-induced PAH (normal mean PA pressure at rest but mean PA pressure $ 30 mm Hg with exercise),12 or PH out of proportion to diastolic dysfunction (mean PA pressure $ 25 mm Hg, PCWP > 15 mm Hg, and pulmonary vascular resistance significantly higher than 3 Wood units).11 Of these patients, one did not have images of sufficient quality for strain calculations,
2D Echocardiography–Derived Parameters of RV Function TAPSE was determined from an M-mode through the lateral tricuspid annulus by calculating the amount of longitudinal motion of the annulus at peak systole13 (Figure 1A). PA acceleration time (PAAT) was calculated from a spectral Doppler image obtained by placing a pulsed Doppler sample volume at the pulmonary valve annulus. PAAT was calculated as the time from the onset of systolic pulmonary arterial flow to peak flow velocity14 (Figure 1B). RV MPI was calculated as the RV isovolumic time divided by the ejection time using the pulsed Doppler method.3 Isovolumic time was calculated as the duration of tricuspid regurgitation from continuous-wave Doppler across the tricuspid valve minus the ejection time from a single representative beat. Ejection time was calculated as the duration of RV
Journal of the American Society of Echocardiography Volume - Number -
Rajagopal et al 3
Figure 1 Common 2D echocardiographic approaches used in the assessment of RV function and PH: (A) TAPSE, (B) PAAT, (C) MPI, and (D) FAC. For full details, refer to ‘‘Methods.’’ EDA, End-diastolic area; ESA, end-systolic area; ET, ejection time; TRd, duration of tricuspid regurgitation.
Figure 2 Right ventricle–focused apical views used for calculation of global longitudinal strain: (A) AP4, (B) AP2, and (C) AP3 equivalent views. Walls and chambers are labeled for clarity. Calculation of strain in these views allowed the construction of a 17-segment model of the right ventricle as shown (D). ANT, Anterior; ANT-SEPT, anteroseptal; EF, ejection fraction; GS, global strain; INF-SEPT, inferoseptal; LA, left atrium; LV, left ventricle; POST, posterior; RA, right atrium; RV, right ventricle; RVOT, right ventricular outflow tract; SEPT, septal.
outflow on pulsed Doppler across the RV outflow tract from a single representative beat (Figure 1C). Care was taken to use beats with similar RR intervals to minimize errors in calculation. RV FAC was calculated as ([RV end-diastolic area RV end-systolic area]/RV end-diastolic area) 100. The RV endocardium was traced in systole and diastole from the annulus, along the free wall, to the apex and back along the interventricular septum using the apical four-
chamber view. Attempts were made to trace the free wall beneath trabeculations (Figure 1D). RVLS Two-dimensional speckle-tracking echocardiographic analysis was performed from the right ventricle–focused four-chamber RV
4 Rajagopal et al
equivalent (AP4), three-chamber RV equivalent (AP3), and twochamber RV equivalent (AP2) views as previously described10 using EchoPAC PC version BT11, which has previously been validated for LV strain10 (Figures 2 and Supplementary Figure1; available at www.onlinejase.com). Images were optimized for strain analysis by decreasing depth to improve resolution and decreasing sector width to improve frame rate to 50 to 90 frames/sec. The reference point for a single cardiac cycle was placed at the beginning of the QRS complex. Pulmonic valve closure was determined from the pulsed-wave Doppler profile of the RVoutflow tract. For 2D speckle-tracking echocardiography, the endocardial border was traced in end-systole, and the region of interest was adjusted to exclude the pericardium. The quality of the speckle-tracking was assessed by EchoPAC and confirmed visually from 2D images and from the strain traces. Segments with persistent inadequate tracking despite attempts at improving border definition and region of interest were excluded from analysis (34 of 720 segments [4.7%]). RVLS was calculated for each region in these views, allowing calculation of a 17-segment bull’s-eye plot (Figure 2D) and global RVLS. In this method, three right ventricle–focused apical views equivalent to the four-chamber, two-chamber, and three-chamber LV views (termed the AP4, AP2, and AP3 views in this study) are obtained by translation of the probe medially (Figure 2). These views allow a comprehensive reconstruction of the RV chamber in a 17-segment bull’s-eye plot. For one patient, significant differences in RR interval between the AP2, AP3, and AP4 views precluded a calculation of global RVLS and a bull’s-eye plot. Examples of strain curves and bull’s-eye plots from a normal control and a patient with PH are shown in Supplementary Figure 1. The longitudinal strain of the RV free wall was calculated as the average of the free wall strains and the longitudinal strain of the RV septum was calculated as the average of the septal strains in the three RV views. All strain and other 2D echocardiography–derived parameter analyses were performed blinded to 6MWD and other clinical data. Statistical Analysis Unless otherwise stated, specified data are presented as mean 6 SD or as percentages. We compared 2D echocardiography–derived parameters (regional and global RVLS, TAPSE, FAC, MPI, and PAAT) between patients with PH and controls using two-tailed, unpaired MannWhitney tests. Correlations between pairs of 2D echocardiography– derived assessments of RV function were evaluated using Pearson’s correlation coefficients. Correlations between 2D echocardiography– derived RV functional parameters, RVLS, and 6MWD were examined using linear regression models. Clustering analysis was used to identify patterns of regional RVLS. Hierarchical clustering of the 17-segment regional RVLS arrays from 39 patients in whom global longitudinal strain was performed in JMP 10 (SAS Institute Inc, Cary, NC). Missing values were imputed as either the average of all other segments or, if there were significant differences in strains between different segments, the average of its nearest neighbors. Data were standardized by their segment means and standard deviations, and distances were calculated using Ward’s minimum variance method. A visual assessment of the hierarchical clusters to assess similarities between different patterns was used to determine the cutoff point for the number of clusters. We compared clinical characteristics between the different clusters of regional RV strain using one-way analysis of variance followed by pairwise multiple comparison tests. We used Bonferroni’s method to correct for multiple testing. Statistical analyses were performed using Prism 5.0 (GraphPad Software, San Diego, CA).
Journal of the American Society of Echocardiography - 2014
Table 1 Characteristics of study population (n = 40) Characteristic
Clinical Women Age (y) BMI (kg/m2) Symptom/disease duration (y) WHO FC I II III IV 6MWD (m) 6MWD (% predicted) REVEAL risk score PH cause Group 1 Idiopathic Connective tissue disease Hereditary Congenital heart disease PVOD HIV Portopulmonary Out-of-proportion DD (group 2) Significant lung disease (group 3) CTEPH (group 4) Sarcoidosis (group 5) Exercise-induced (other) Medications ERA PDE5 inhibitor Prostacyclin, inhaled Prostacyclin, IV Combination therapy None Other studies Serum creatinine (mg/dL) NT-proBNP (ng/L) DLCO Most recent catheterization RA pressure (mm Hg) Cardiac index (L/min/m2) Mean PA pressure (mm Hg) PVR (Wood units)
Value
34 (85%) 58 6 14 30.3 6 8.4 4 (2–7) 4 (10%) 25 (63%) 10 (25%) 1 (3%) 376 6 130 75 6 23 7.7 6 2.6 31 (78%) 14 (35%) 7 (18%) 4 (10%) 2 (5%) 2 (5%) 1 (2.5%) 1 (2.5%) 2 (5%) 2 (5%) 3 (7.5%) 1 (2.5%) 1 (2.5%) 23 (58%) 16 (40%) 7 (18%) 8 (20%) 21 (53%) 6 (15%) 1.0 6 0.4 490 (124–1,232) 56 6 27 11 6 6 2.4 6 0.7 50 6 15 9.5 6 5.3
BMI, Body mass index; CTEPH, Chronic thromboembolic PH; DLCO, diffusion capacity for carbon monoxide; DD, diastolic dysfunction; ERA, endothelin receptor antagonist; FC, functional class; HIV, human immunodeficiency virus; IV, intravenous; PDE5, phosphodiesterase 5; NT-proBNP, N-terminal pro-brain natriuretic peptide; PVOD, pulmonary veno-occlusive disease; PVR, pulmonary vascular resistance; RA, right atrial; WHO, World Health Organization. Data are expressed as mean 6 SD, median (interquartile range), or number (percentage).
RESULTS Characteristics of the Study Population Baseline characteristics of the PH cohort are presented in Table 1. As shown, the demographics of this population are similar to registries of
Rajagopal et al 5
Journal of the American Society of Echocardiography Volume - Number -
Table 2 Global and regional RVLS from the PH cohort compared with a control population of young individuals with no significant medical problems10 Variable
Patients with PH (n = 40)
Controls (n = 41)
P
15.6 6 4.6
23.8 6 2.3
