© 2014, Wiley Periodicals, Inc. DOI: 10.1111/echo.12812
Echocardiography
ORIGINAL INVESTIGATION
Right Ventricular Strain before and after Pulmonary Thromboendarterectomy in Patients with Chronic Thromboembolic Pulmonary Hypertension Nicholas Marston, M.D.,* Jason P. Brown, M.D.,† Nicholas Olson, M.D.,* William R. Auger, M.D.,* Michael M. Madani, M.D.,* Darrin Wong, M.D.,* Ajit B. Raisinghani, M.D.,* Anthony N. DeMaria, M.D.,* and Daniel G. Blanchard, M.D.* *School of Medicine and Sulpizio Cardiovascular Center, University of California San Diego, La Jolla, California; and †Kaiser Permanente, San Diego, California
Background: Right ventricular (RV) function is significantly impaired in patients with chronic thromboembolic pulmonary hypertension (CTEPH). Two-dimensional speckle tracking RV strain and strain rate are novel methods to assess regional RV systolic function in CTEPH patients before and after pulmonary thromboendarterectomy (PTE). Our goal was to (1) assess baseline longitudinal strain and strain rate of the basal RV free wall in CTEPH and (2) measure early changes in RV strain and strain rate after PTE. Methods: We performed echocardiography on 30 consecutive patients with CTEPH referred for PTE with adequate pre- and post-PTE strain imaging. Strain and strain rate were assessed 6.4 4.5 days before and 9.1 3.9 after PTE. Results: Basal RV free wall strain and time to peak strain—but not basal RV strain rate and time to peak strain rate—changed significantly after PTE. Unexpectedly, basal RV strain became less negative, from 24.3% to 18.9% after PTE (P = 0.005). Time to peak strain decreased from 356 to 287 msec after PTE (P < 0.001). Preoperatively, RV strain correlated with pulmonary vascular resistance (PVR) and mean pulmonary artery pressure (mPAP) but this relationship was not evident postoperatively. Furthermore, the change in RV strain did not correlate with the change in mPAP or PVR. Conclusions: In patients with CTEPH, RV basal strain paradoxically became less negative (i.e., relative systolic shortening decreased) following PTE. This change in RV strain could be due to intraoperative RV ischemia and/or postoperative stunning. Thus, RV basal strain cannot be used as a surrogate marker for surgical success early after PTE. (Echocardiography 2014;00:1–7) Key words: pulmonary hypertension, right ventricular function, strain, strain rate imaging, pulmonary embolism Chronic thromboembolic pulmonary hypertension (CTEPH) is an increasingly recognized complication in post–pulmonary embolism patients, with an incidence ranging from 1% to 3.8%.1–4 CTEPH is the result of clot persistence in the pulmonary arteries and subsequent intimal hyperplasia of the adjacent pulmonary vasculature.5,6 These changes lead to an increase in pulmonary vascular resistance (PVR) and mean pulmonary artery pressure (mPAP), progressive right ventricular (RV) dilation, and eventually right heart failure. Mortality is high if left untreated.7–9 Pulmonary thromboendarterectomy (PTE) has been shown to normalize the pulmonary pressures in up to 80% of patients at rest,9 leading to improvement in symptoms and quality of life. Address for correspondence and reprint requests: Daniel G. Blanchard, M.D., UCSD Sulpizio Cardiovascular Center, 9444 Medical Center Drive, #7411, La Jolla, CA 92037. Fax: 858 657-5012; E-mail: [email protected]
Newer echocardiographic techniques such as two-dimensional (2D) speckle tracking provide an opportunity to better quantify RV function using strain and strain rate. The purpose of this study was to assess strain and strain rate of the basal RV free wall before and after PTE. Methods: Patients: Thirty consecutive CTEPH patients undergoing PTE with adequate pre- and postoperative echocardiographic images were enrolled in this study. All patients had confirmed CTEPH (Dana Point Class IV) diagnosed by pulmonary VQ scan and pulmonary angiography. Each patient had preand post-PTE echocardiograms with 2D speckle tracking analysis. The average timing of the echocardiograms was 6.4 4.5 days before and 9.1 3.9 days after PTE. Right heart catheterization (RHC) was performed 6 5 days before surgery and on the first postoperative day. mPAP, 1
Marston, et al.
PVR, central venous pressure, and cardiac output (CO) were obtained. The UCSD institutional research review committee approved this study. Pulmonary Thromboendarterectomy: All operations were performed at UCSD Medical Center using techniques described previously.10 PTE is a complex procedure that begins with a median sternotomy and pericardiotomy to gain access to the heart and major vessels. The patient is initially placed on cardiac bypass and then cooled until cardiac arrest is achieved. Incisions are first made in the right pulmonary artery, endarterectomy is then performed starting from the main pulmonary artery and continuing all the way to the subsegmental branches. Next, the same technique is applied to the left pulmonary artery. Upon completion of both the left and right endarterectomies, cardiopulmonary bypass is reinitiated and rewarming is begun.
Echocardiography: All echocardiograms were performed using Vivid E7 cardiovascular ultrasound system (GE VingMed, Horton, Norway). Examinations included measurements of the RV size and function, and estimated PA pressures. Left ventricular measurements were also recorded, including end-systolic and end-diastolic diameters and ejection fraction. All measurements were made in accordance to the American Society of Echocardiography recommendations.11 Speckle Tracking Analysis: 2D speckle tracking analysis was performed using standard grayscale 2D images in the apical fourchamber view. The RV myocardium and interventricular septum were separated into 6 standard segments as presented in Figure 1 (left). Only RV free wall segmental strain was analyzed. A frame rate of at least 40 frames per second (fps) was used for all images to optimize performance of
Figure 1. Representative freeze-frame image of RV strain speckle tracking. Upper left frame: speckle tracking superimposed on the right ventricular (RV). Lower left frame: peak systolic strain of RV and septal segments. Upper right frame: segmental strain analysis of the same example. Lower right frame: 2D strain map.
2
RV Strain Following Pulmonary Thromboendarterectomy
the software (Echo-Pac 6.1). Using a single frame from end-systole, the RV free wall segments were manually mapped by marking the endocardial border and the width of the myocardium. These points were then mapped throughout the cardiac cycle using the automated tracking algorithm creating a tracing as displayed in Figure 1 (upper right). The RV free wall myocardial segments were then independently analyzed as poor, average, or good by the software’s automated tracking program. Only images rated as good in the RV basal free wall myocardial segment both before and after PTE were included in this study. We chose to focus on measurements from the RV basal free wall segment as this is functionally and anatomically the furthest segment away from the left ventricle, and thus least likely to be affected by changes in LV systolic or diastolic function. Statistical Analysis: A paired two-tailed Student’s t-test was used to evaluate differences in RV basal free wall strain, strain rate, time to peak strain, and time to peak strain rate following PTE. 95% confidence intervals and P-values were calculated to establish significance. RV basal strain, strain rate (SR), and time to peak strain were compared to RHC parameters using linear regression software (http://www.Wessa.net).12 Comparison plots included PVR versus strain, SR, and time to peak strain as well as change in PVR versus change in strain, SR, and time to peak strain. Results: Data from pre- and post-PTE transthoracic echocardiography and RHC are presented in Table I. Statistically significant reductions were seen in
TABLE I
PVR (950 550 to 31 160 [dynes-sec]/cm5 [11.9 6.9 to 0.4 2.0 Woods units], P < 0.001) and mPAP (44 15 to 29 9 mmHg, P < 0.001). CO significantly increased following PTE (3.9 1.0 to 5.0 1.0 L/min, P < 0.001). LV ejection fraction did not change significantly after PTE (67 8% to 65 8%). TR velocity decreased from 4.2 0.8 m/sec pre-PTE to 3.3 0.5 m/sec after PTE (P < 0.001). Strain and strain rate data are presented in Table II. RV basal strain became less negative, changing from 24.3 8% to 18.9 6% (P = 0.005). For comparison, a group of normal controls (n = 10) was also evaluated. RV basal strain in this population was 30.8 7.0% (P = 0.03 compared to study group). Strain rate in the study population was unchanged at 1.7 0.6 before and after PTE (P = 0.8). Time to peak strain decreased from 356 72 to 287 70 msec (P < 0.001). Time to peak strain rate did not change significantly (157 61 to 137 50 msec, P = 0.2). Two additional reviews of RV strain data found an observer variability of 9%. Linear regression data for RV strain variables and RHC parameters (PVR and mPAP) are shown in Table III along with R-values as a measure of linear fit. Preoperative PVR correlated with strain and strain rate (R = 0.55, P = 0.002; R = 0.47, P = 0.01, respectively) but not time to peak strain. Preoperative mPAP correlated with strain (R = 0.5, P = 0.006) but not strain rate or time to peak strain. Postoperatively these significant relationships were no longer observed. Furthermore, change in PVR and mPAP did not correlate with change in strain, strain rate, or time to peak strain. Scatter plots demonstrating the pre- and postoperative relationships between PVR and strain (Fig. 2), mPAP and strain (Fig. 3), and PVR and strain rate (Fig. 4) are shown.
Hemodynamic and Echocardiographic Data before and after PTE
Age (years) Mean PA pressure (mmHg) PVR (dyne-sec)/cm5 LVEDD (cm) LVESD (cm) LV ejection fraction (%) Fractional shortening (%) Cardiac output (L/min)
PValue
Pre-PTE
Post-PTE
51 17 44 15
29 9
Echocardiography
ORIGINAL INVESTIGATION
Right Ventricular Strain before and after Pulmonary Thromboendarterectomy in Patients with Chronic Thromboembolic Pulmonary Hypertension Nicholas Marston, M.D.,* Jason P. Brown, M.D.,† Nicholas Olson, M.D.,* William R. Auger, M.D.,* Michael M. Madani, M.D.,* Darrin Wong, M.D.,* Ajit B. Raisinghani, M.D.,* Anthony N. DeMaria, M.D.,* and Daniel G. Blanchard, M.D.* *School of Medicine and Sulpizio Cardiovascular Center, University of California San Diego, La Jolla, California; and †Kaiser Permanente, San Diego, California
Background: Right ventricular (RV) function is significantly impaired in patients with chronic thromboembolic pulmonary hypertension (CTEPH). Two-dimensional speckle tracking RV strain and strain rate are novel methods to assess regional RV systolic function in CTEPH patients before and after pulmonary thromboendarterectomy (PTE). Our goal was to (1) assess baseline longitudinal strain and strain rate of the basal RV free wall in CTEPH and (2) measure early changes in RV strain and strain rate after PTE. Methods: We performed echocardiography on 30 consecutive patients with CTEPH referred for PTE with adequate pre- and post-PTE strain imaging. Strain and strain rate were assessed 6.4 4.5 days before and 9.1 3.9 after PTE. Results: Basal RV free wall strain and time to peak strain—but not basal RV strain rate and time to peak strain rate—changed significantly after PTE. Unexpectedly, basal RV strain became less negative, from 24.3% to 18.9% after PTE (P = 0.005). Time to peak strain decreased from 356 to 287 msec after PTE (P < 0.001). Preoperatively, RV strain correlated with pulmonary vascular resistance (PVR) and mean pulmonary artery pressure (mPAP) but this relationship was not evident postoperatively. Furthermore, the change in RV strain did not correlate with the change in mPAP or PVR. Conclusions: In patients with CTEPH, RV basal strain paradoxically became less negative (i.e., relative systolic shortening decreased) following PTE. This change in RV strain could be due to intraoperative RV ischemia and/or postoperative stunning. Thus, RV basal strain cannot be used as a surrogate marker for surgical success early after PTE. (Echocardiography 2014;00:1–7) Key words: pulmonary hypertension, right ventricular function, strain, strain rate imaging, pulmonary embolism Chronic thromboembolic pulmonary hypertension (CTEPH) is an increasingly recognized complication in post–pulmonary embolism patients, with an incidence ranging from 1% to 3.8%.1–4 CTEPH is the result of clot persistence in the pulmonary arteries and subsequent intimal hyperplasia of the adjacent pulmonary vasculature.5,6 These changes lead to an increase in pulmonary vascular resistance (PVR) and mean pulmonary artery pressure (mPAP), progressive right ventricular (RV) dilation, and eventually right heart failure. Mortality is high if left untreated.7–9 Pulmonary thromboendarterectomy (PTE) has been shown to normalize the pulmonary pressures in up to 80% of patients at rest,9 leading to improvement in symptoms and quality of life. Address for correspondence and reprint requests: Daniel G. Blanchard, M.D., UCSD Sulpizio Cardiovascular Center, 9444 Medical Center Drive, #7411, La Jolla, CA 92037. Fax: 858 657-5012; E-mail: [email protected]
Newer echocardiographic techniques such as two-dimensional (2D) speckle tracking provide an opportunity to better quantify RV function using strain and strain rate. The purpose of this study was to assess strain and strain rate of the basal RV free wall before and after PTE. Methods: Patients: Thirty consecutive CTEPH patients undergoing PTE with adequate pre- and postoperative echocardiographic images were enrolled in this study. All patients had confirmed CTEPH (Dana Point Class IV) diagnosed by pulmonary VQ scan and pulmonary angiography. Each patient had preand post-PTE echocardiograms with 2D speckle tracking analysis. The average timing of the echocardiograms was 6.4 4.5 days before and 9.1 3.9 days after PTE. Right heart catheterization (RHC) was performed 6 5 days before surgery and on the first postoperative day. mPAP, 1
Marston, et al.
PVR, central venous pressure, and cardiac output (CO) were obtained. The UCSD institutional research review committee approved this study. Pulmonary Thromboendarterectomy: All operations were performed at UCSD Medical Center using techniques described previously.10 PTE is a complex procedure that begins with a median sternotomy and pericardiotomy to gain access to the heart and major vessels. The patient is initially placed on cardiac bypass and then cooled until cardiac arrest is achieved. Incisions are first made in the right pulmonary artery, endarterectomy is then performed starting from the main pulmonary artery and continuing all the way to the subsegmental branches. Next, the same technique is applied to the left pulmonary artery. Upon completion of both the left and right endarterectomies, cardiopulmonary bypass is reinitiated and rewarming is begun.
Echocardiography: All echocardiograms were performed using Vivid E7 cardiovascular ultrasound system (GE VingMed, Horton, Norway). Examinations included measurements of the RV size and function, and estimated PA pressures. Left ventricular measurements were also recorded, including end-systolic and end-diastolic diameters and ejection fraction. All measurements were made in accordance to the American Society of Echocardiography recommendations.11 Speckle Tracking Analysis: 2D speckle tracking analysis was performed using standard grayscale 2D images in the apical fourchamber view. The RV myocardium and interventricular septum were separated into 6 standard segments as presented in Figure 1 (left). Only RV free wall segmental strain was analyzed. A frame rate of at least 40 frames per second (fps) was used for all images to optimize performance of
Figure 1. Representative freeze-frame image of RV strain speckle tracking. Upper left frame: speckle tracking superimposed on the right ventricular (RV). Lower left frame: peak systolic strain of RV and septal segments. Upper right frame: segmental strain analysis of the same example. Lower right frame: 2D strain map.
2
RV Strain Following Pulmonary Thromboendarterectomy
the software (Echo-Pac 6.1). Using a single frame from end-systole, the RV free wall segments were manually mapped by marking the endocardial border and the width of the myocardium. These points were then mapped throughout the cardiac cycle using the automated tracking algorithm creating a tracing as displayed in Figure 1 (upper right). The RV free wall myocardial segments were then independently analyzed as poor, average, or good by the software’s automated tracking program. Only images rated as good in the RV basal free wall myocardial segment both before and after PTE were included in this study. We chose to focus on measurements from the RV basal free wall segment as this is functionally and anatomically the furthest segment away from the left ventricle, and thus least likely to be affected by changes in LV systolic or diastolic function. Statistical Analysis: A paired two-tailed Student’s t-test was used to evaluate differences in RV basal free wall strain, strain rate, time to peak strain, and time to peak strain rate following PTE. 95% confidence intervals and P-values were calculated to establish significance. RV basal strain, strain rate (SR), and time to peak strain were compared to RHC parameters using linear regression software (http://www.Wessa.net).12 Comparison plots included PVR versus strain, SR, and time to peak strain as well as change in PVR versus change in strain, SR, and time to peak strain. Results: Data from pre- and post-PTE transthoracic echocardiography and RHC are presented in Table I. Statistically significant reductions were seen in
TABLE I
PVR (950 550 to 31 160 [dynes-sec]/cm5 [11.9 6.9 to 0.4 2.0 Woods units], P < 0.001) and mPAP (44 15 to 29 9 mmHg, P < 0.001). CO significantly increased following PTE (3.9 1.0 to 5.0 1.0 L/min, P < 0.001). LV ejection fraction did not change significantly after PTE (67 8% to 65 8%). TR velocity decreased from 4.2 0.8 m/sec pre-PTE to 3.3 0.5 m/sec after PTE (P < 0.001). Strain and strain rate data are presented in Table II. RV basal strain became less negative, changing from 24.3 8% to 18.9 6% (P = 0.005). For comparison, a group of normal controls (n = 10) was also evaluated. RV basal strain in this population was 30.8 7.0% (P = 0.03 compared to study group). Strain rate in the study population was unchanged at 1.7 0.6 before and after PTE (P = 0.8). Time to peak strain decreased from 356 72 to 287 70 msec (P < 0.001). Time to peak strain rate did not change significantly (157 61 to 137 50 msec, P = 0.2). Two additional reviews of RV strain data found an observer variability of 9%. Linear regression data for RV strain variables and RHC parameters (PVR and mPAP) are shown in Table III along with R-values as a measure of linear fit. Preoperative PVR correlated with strain and strain rate (R = 0.55, P = 0.002; R = 0.47, P = 0.01, respectively) but not time to peak strain. Preoperative mPAP correlated with strain (R = 0.5, P = 0.006) but not strain rate or time to peak strain. Postoperatively these significant relationships were no longer observed. Furthermore, change in PVR and mPAP did not correlate with change in strain, strain rate, or time to peak strain. Scatter plots demonstrating the pre- and postoperative relationships between PVR and strain (Fig. 2), mPAP and strain (Fig. 3), and PVR and strain rate (Fig. 4) are shown.
Hemodynamic and Echocardiographic Data before and after PTE
Age (years) Mean PA pressure (mmHg) PVR (dyne-sec)/cm5 LVEDD (cm) LVESD (cm) LV ejection fraction (%) Fractional shortening (%) Cardiac output (L/min)
PValue
Pre-PTE
Post-PTE
51 17 44 15
29 9
