© 2014, Wiley Periodicals, Inc. DOI: 10.1111/echo.12561
Echocardiography
Left Ventricular Twist and Ventricular–Arterial Coupling in Hypertensive Patients Hong-Won Shin, M.D., Ph.D., Hyungseop Kim, M.D., Ph.D., Jeong-Eun Lee, M.D., In-Cheol Kim, M.D., Hyuck-Jun Yoon, M.D., Ph.D., Hyoung-Seob Park, M.D., Yun-Kyeong Cho, M.D., Ph.D., Chang-Wook Nam, M.D., Ph.D., Seung-Ho Hur, M.D., Ph.D., Yoon-Nyun Kim, M.D., Ph.D., and Kwon-Bae Kim, M.D., Ph.D. Division of Cardiology, Department of Internal Medicine, Keimyung University Dongsan Medical Center, Daegu, Korea
Background: Left ventricular (LV) twist is usually influenced by LV hypertrophy resulting from hypertension or vascular stiffness. Vascular stiffness would increase arterial elastance (Ea), whereas LV end-systolic stiffness (Ees) could be influenced by LV hypertrophy. Therefore, in hypertensive patients, we assessed the extent to which ventricular–arterial coupling (VAC; Ea/Ees) affects LV twist, which may be a compensatory mechanism for systolic dysfunction. Methods: Hypertensive patients (n = 128) and healthy controls (n = 40) underwent conventional and speckle tracking echocardiography including LV twist. Ea and Ees were estimated noninvasively by echocardiography. Patients were divided into 3 tertiles according to the twist angle. Univariate and multivariate regression analyses were performed to test the influence of VAC on twist. Results: Patients in the lowest LV twist tertile had larger LV end-systolic volume, lower ejection fraction, lesser mid-wall fractional shortening (MWFS), and higher LV mass index (LVMI), compared to those with the highest tertile. They showed the lower septal tissue Doppler velocity, and global longitudinal and circumferential strain. With regard to VAC, Ea was similar among 3 groups, but Ees was significantly decreased in patient with lower tertile, resulting in increased VAC (1.1 0.2 vs. 0.9 0.1 vs. 0.7 0.1, P < 0.001). While LV twist showed significant correlations with Ees, MWFS, and LVMI, VAC (b = 14.92, P < 0.001) was most associated with twist in a multivariate analysis. Conclusions: LV twist was significantly associated with VAC in accordance with LV function; LV twist and VAC decreased progressively as LV systolic function deteriorated, while being enhanced during the well-compensated phase. (Echocardiography 2014;31:1274–1282) Key words: ventricular–arterial coupling, stiffness, elastance, echocardiography Hypertensive hypertrophy frequently displays the decreased left ventricle (LV) diastolic and systolic function which were evidenced by LV strain or tissue Doppler velocity.1–3 However, as for evaluation of cardiac function, LV rotation or twist could provide an important role as a part of the compensatory mechanism for decreased systolic function.4 Thus, LV twist is usually enhanced, particularly in hypertensive hypertrophy or heart failure with preserved EF (HFPEF).5,6 There has been also increasing recognition that prolonged hypertension appears to be related with LV hypertrophy and geometric remodeling, accompanying changes of LV twist.3,7,8 Considering the interrelationship of LV and arterial system, hypertension could have profound effect on LV hypertrophy and vascular stiffAddress for correspondence and reprint requests: Hyungseop Kim, M.D., 56 Dalseong-no, Jung-gu, Keimyung University Dongsan Medical Center, Daegu 700-712, Korea. Fax: +8253-250-7034; E-mail: [email protected]
1274
ness, respectively; hypertension, per se, could seem to worsen LV end-systolic stiffness (Ees) by LV hypertrophy, and to increase arterial elastance (Ea) by vascular stiffness.9,10 Consequently, Ees is a major equivalence of LV systolic performance, whereas effective Ea functions as a measure of impedance. Thus, the Ea/Ees ratio could serve as a reliable index of ventricular–arterial coupling (VAC) and this evaluation may be of clinical importance in gathering the information for detecting preclinical LV dysfunction.11–13 However, despite this importance, little is known of the relationship between LV twist and vascular/ventricular stiffness linked with hypertension. In this study, we hypothesized that the changes of LV twist would be associated with VAC changes and thus, LV twist would be decreased progressively as systolic dysfunction becomes far more advanced in hypertensive patients. Accordingly, we sought to analyze the relationship between VAC and LV twist in hypertensive patients compared with healthy controls
LV Twist and Ventricular–Arterial Coupling
to better understand the interactions between hypertension and LV twist. Methods: Study Population: From March 2011 to February 2012, we prospectively recruited the patients referred to our echocardiographic laboratory for evaluation of hypertension. Patients were enrolled according to the following criteria: (1) age >18 years; (2) hypertension defined as office blood pressure (BP) ≥140 mmHg and/or diastolic BP ≥90 mmHg in a sitting position in at least 3 measurements or previously diagnosed hypertension with antihypertensive medication; (3) normal or preserved LV ejection fraction (LVEF) ≥55%; and (4) normal sinus rhythm. The patients were excluded if they had cardiomyopathies, secondary hypertension, diabetes mellitus, a history of coronary artery disease or overt HF, renal impairment, peripheral arterial disease, or valvular heart disease. For the current analyses, nine patients were excluded because of inadequate poor echocardiographic imaging. Therefore, a total of 128 patients met the inclusion criteria during the period of enrollment. As a control group, 40 normotensive healthy volunteers who did not have history of cardiovascular disease were recruited. The study was approved by our institutional ethical committee, and each study patient gave a written informed consent. Echocardiography: Echocardiography was performed using a Vivid 7 ultrasound (GE-VingMed, Horten, Norway), including tissue Doppler imaging (TDI) according to the recommendations of the American Society of Echocardiography. To assess LV short-axis images, the short-axis basal view was obtained at the mitral valve tips. As caudally as possible, a short-axis apical view was captured at the point that LV cavity obliteration was noticed at the end-systole. LV mass index (LVMI) was derived from LV mass indexed to body surface area, and LV mass was calculated as follows:14,15 LVmass ðgmÞ ¼ 0:8 f1:04½ðLVEDD þ PWT þ IVWTÞ3 LVEDD3 g þ 0:6 where LVEDD is LV end-diastolic dimension, PWT is posterior wall thickness (PWT), and IVWT is interventricular wall thickness. Mid-wall fractional shortening (MWFS) was measured using the following formula:15,16
MWFS ¼ ½ðLVEDD þ IVWT=2 þ PWT=2Þ ðLVESD þ Inner shellÞ=½ðLVEDD þ IVWT=2 þ PWT=2Þ 100 where LVESD is LV end-systolic dimension. Inner shell ¼ ½ðLVEDD þ IVWT=2 þ PWT=2Þ3 LVEDD3 þ LVESD3 1=3 LVESD The patients were subdivided into 4 subgroups of the LV geometric patterns on LVMI and relative wall thickness (RWT)15; normal geometry (RWT ≤ 0.42, LVMI ≤ 95 g/m2 for women and 115 g/m2 for men), concentric remodeling (RWT > 0.42, LVMI ≤ 95 g/m2 for women and 115 g/m2 for men), concentric hypertrophy (RWT > 0.42, LVMI > 95 g/m2 for women and 115 g/m2 for men), and eccentric hypertrophy (RWT ≤ 0.42, LVMI > 95 g/m2 for women and 115 g/m2 for men). Ventricular–Arterial Coupling: Ea was obtained as the end-systolic pressure/ stroke volume (SV), with end-systolic pressure estimated noninvasively as systolic pressure 9 0.9, and SV calculated as 0.785 9 LV outflow tract diameter2 9 time velocity integral of LV outflow tract. The Ees was obtained using BP, SV, EF, and an estimated normalized Ees at arterial end-diastole which was derived from the noninvasive single-beat method:17 Ees ¼ ½diastolic BP ðENd systolic BP 0:9Þ= ðENd SVÞ where ENd was the predicted normalized LV elastance at the onset of ejection and estimated from a group-averaged value adjusted for individual contractile/loading effects, using the ratio of aortic preejection time to total systolic time. Accordingly, VAC was expressed as the Ea/Ees ratio. Speckle Tracking: All strain measurements were performed off line using software EchoPAC version 7.0. The apical two-, four-, and three-chamber views together with the 3 short-axis views at basal, mid, and apical levels were obtained with frame rates >50– 70 frames/sec. We gathered the data of longitudinal and circumferential strain, as well as rotation. Global longitudinal strain (GLS) was obtained by averaging all values of regional or segmental peak strain in each apical view, and global circumferential strain (GCS) by averaging the six myocardial segments at 3 short-axis levels. With the use of two-dimensional speckle tracking software to generate LV twist angle and rate
1275
Shin, et al.
automatically, LV twist angle was calculated as the net absolute angle between apical and basal differences in LV rotation (Fig. 1). Peak twist rate was obtained and time to peak rotation of apex and base as well as time to peak LV twist were all measured in the analysis. To adjust heart rate differences in interpatient for time parameter of twist mechanics, the time was normalized to the percentage of each systolic and diastolic period; time 0% at the onset of systolic phase and at the onset of diastolic ejection time (coincidence of aortic valve close), respectively. Diastolic untwist rate was measured at the apex and base level. Peak early untwist rate was defined as the net difference between apical and basal peak untwist rates during diastole. Statistical Analyses: Data analyses were performed with the Statistical Package for Social Science software for Windows 12.0 (SPSS Inc., Chicago, IL, USA). All values are presented as the mean standard deviation for continuous variables, and frequencies for discrete variables. Between-group differences of continuous variables were analyzed using one-way analysis of variance, and categorical variables were analyzed using a chi-square test. Univariate correlations were determined using Pearson’s coefficient for continuous variables. Multiple regression analysis was performed to explore the determinants of LV twist. To assess the incremental value of VAC to the multivariable logistic regression model in discriminating the highest LV
twist tertile group, we constructed receiver-operating characteristic (ROC) curves and compared C-statistics after computing each modeling’s probability of highest LV twist tertile. The calibration was tested with the Hosmer–Lemeshow statistic for goodness-of-fit of the model. P-values were two sided, and a P-value
Echocardiography
Left Ventricular Twist and Ventricular–Arterial Coupling in Hypertensive Patients Hong-Won Shin, M.D., Ph.D., Hyungseop Kim, M.D., Ph.D., Jeong-Eun Lee, M.D., In-Cheol Kim, M.D., Hyuck-Jun Yoon, M.D., Ph.D., Hyoung-Seob Park, M.D., Yun-Kyeong Cho, M.D., Ph.D., Chang-Wook Nam, M.D., Ph.D., Seung-Ho Hur, M.D., Ph.D., Yoon-Nyun Kim, M.D., Ph.D., and Kwon-Bae Kim, M.D., Ph.D. Division of Cardiology, Department of Internal Medicine, Keimyung University Dongsan Medical Center, Daegu, Korea
Background: Left ventricular (LV) twist is usually influenced by LV hypertrophy resulting from hypertension or vascular stiffness. Vascular stiffness would increase arterial elastance (Ea), whereas LV end-systolic stiffness (Ees) could be influenced by LV hypertrophy. Therefore, in hypertensive patients, we assessed the extent to which ventricular–arterial coupling (VAC; Ea/Ees) affects LV twist, which may be a compensatory mechanism for systolic dysfunction. Methods: Hypertensive patients (n = 128) and healthy controls (n = 40) underwent conventional and speckle tracking echocardiography including LV twist. Ea and Ees were estimated noninvasively by echocardiography. Patients were divided into 3 tertiles according to the twist angle. Univariate and multivariate regression analyses were performed to test the influence of VAC on twist. Results: Patients in the lowest LV twist tertile had larger LV end-systolic volume, lower ejection fraction, lesser mid-wall fractional shortening (MWFS), and higher LV mass index (LVMI), compared to those with the highest tertile. They showed the lower septal tissue Doppler velocity, and global longitudinal and circumferential strain. With regard to VAC, Ea was similar among 3 groups, but Ees was significantly decreased in patient with lower tertile, resulting in increased VAC (1.1 0.2 vs. 0.9 0.1 vs. 0.7 0.1, P < 0.001). While LV twist showed significant correlations with Ees, MWFS, and LVMI, VAC (b = 14.92, P < 0.001) was most associated with twist in a multivariate analysis. Conclusions: LV twist was significantly associated with VAC in accordance with LV function; LV twist and VAC decreased progressively as LV systolic function deteriorated, while being enhanced during the well-compensated phase. (Echocardiography 2014;31:1274–1282) Key words: ventricular–arterial coupling, stiffness, elastance, echocardiography Hypertensive hypertrophy frequently displays the decreased left ventricle (LV) diastolic and systolic function which were evidenced by LV strain or tissue Doppler velocity.1–3 However, as for evaluation of cardiac function, LV rotation or twist could provide an important role as a part of the compensatory mechanism for decreased systolic function.4 Thus, LV twist is usually enhanced, particularly in hypertensive hypertrophy or heart failure with preserved EF (HFPEF).5,6 There has been also increasing recognition that prolonged hypertension appears to be related with LV hypertrophy and geometric remodeling, accompanying changes of LV twist.3,7,8 Considering the interrelationship of LV and arterial system, hypertension could have profound effect on LV hypertrophy and vascular stiffAddress for correspondence and reprint requests: Hyungseop Kim, M.D., 56 Dalseong-no, Jung-gu, Keimyung University Dongsan Medical Center, Daegu 700-712, Korea. Fax: +8253-250-7034; E-mail: [email protected]
1274
ness, respectively; hypertension, per se, could seem to worsen LV end-systolic stiffness (Ees) by LV hypertrophy, and to increase arterial elastance (Ea) by vascular stiffness.9,10 Consequently, Ees is a major equivalence of LV systolic performance, whereas effective Ea functions as a measure of impedance. Thus, the Ea/Ees ratio could serve as a reliable index of ventricular–arterial coupling (VAC) and this evaluation may be of clinical importance in gathering the information for detecting preclinical LV dysfunction.11–13 However, despite this importance, little is known of the relationship between LV twist and vascular/ventricular stiffness linked with hypertension. In this study, we hypothesized that the changes of LV twist would be associated with VAC changes and thus, LV twist would be decreased progressively as systolic dysfunction becomes far more advanced in hypertensive patients. Accordingly, we sought to analyze the relationship between VAC and LV twist in hypertensive patients compared with healthy controls
LV Twist and Ventricular–Arterial Coupling
to better understand the interactions between hypertension and LV twist. Methods: Study Population: From March 2011 to February 2012, we prospectively recruited the patients referred to our echocardiographic laboratory for evaluation of hypertension. Patients were enrolled according to the following criteria: (1) age >18 years; (2) hypertension defined as office blood pressure (BP) ≥140 mmHg and/or diastolic BP ≥90 mmHg in a sitting position in at least 3 measurements or previously diagnosed hypertension with antihypertensive medication; (3) normal or preserved LV ejection fraction (LVEF) ≥55%; and (4) normal sinus rhythm. The patients were excluded if they had cardiomyopathies, secondary hypertension, diabetes mellitus, a history of coronary artery disease or overt HF, renal impairment, peripheral arterial disease, or valvular heart disease. For the current analyses, nine patients were excluded because of inadequate poor echocardiographic imaging. Therefore, a total of 128 patients met the inclusion criteria during the period of enrollment. As a control group, 40 normotensive healthy volunteers who did not have history of cardiovascular disease were recruited. The study was approved by our institutional ethical committee, and each study patient gave a written informed consent. Echocardiography: Echocardiography was performed using a Vivid 7 ultrasound (GE-VingMed, Horten, Norway), including tissue Doppler imaging (TDI) according to the recommendations of the American Society of Echocardiography. To assess LV short-axis images, the short-axis basal view was obtained at the mitral valve tips. As caudally as possible, a short-axis apical view was captured at the point that LV cavity obliteration was noticed at the end-systole. LV mass index (LVMI) was derived from LV mass indexed to body surface area, and LV mass was calculated as follows:14,15 LVmass ðgmÞ ¼ 0:8 f1:04½ðLVEDD þ PWT þ IVWTÞ3 LVEDD3 g þ 0:6 where LVEDD is LV end-diastolic dimension, PWT is posterior wall thickness (PWT), and IVWT is interventricular wall thickness. Mid-wall fractional shortening (MWFS) was measured using the following formula:15,16
MWFS ¼ ½ðLVEDD þ IVWT=2 þ PWT=2Þ ðLVESD þ Inner shellÞ=½ðLVEDD þ IVWT=2 þ PWT=2Þ 100 where LVESD is LV end-systolic dimension. Inner shell ¼ ½ðLVEDD þ IVWT=2 þ PWT=2Þ3 LVEDD3 þ LVESD3 1=3 LVESD The patients were subdivided into 4 subgroups of the LV geometric patterns on LVMI and relative wall thickness (RWT)15; normal geometry (RWT ≤ 0.42, LVMI ≤ 95 g/m2 for women and 115 g/m2 for men), concentric remodeling (RWT > 0.42, LVMI ≤ 95 g/m2 for women and 115 g/m2 for men), concentric hypertrophy (RWT > 0.42, LVMI > 95 g/m2 for women and 115 g/m2 for men), and eccentric hypertrophy (RWT ≤ 0.42, LVMI > 95 g/m2 for women and 115 g/m2 for men). Ventricular–Arterial Coupling: Ea was obtained as the end-systolic pressure/ stroke volume (SV), with end-systolic pressure estimated noninvasively as systolic pressure 9 0.9, and SV calculated as 0.785 9 LV outflow tract diameter2 9 time velocity integral of LV outflow tract. The Ees was obtained using BP, SV, EF, and an estimated normalized Ees at arterial end-diastole which was derived from the noninvasive single-beat method:17 Ees ¼ ½diastolic BP ðENd systolic BP 0:9Þ= ðENd SVÞ where ENd was the predicted normalized LV elastance at the onset of ejection and estimated from a group-averaged value adjusted for individual contractile/loading effects, using the ratio of aortic preejection time to total systolic time. Accordingly, VAC was expressed as the Ea/Ees ratio. Speckle Tracking: All strain measurements were performed off line using software EchoPAC version 7.0. The apical two-, four-, and three-chamber views together with the 3 short-axis views at basal, mid, and apical levels were obtained with frame rates >50– 70 frames/sec. We gathered the data of longitudinal and circumferential strain, as well as rotation. Global longitudinal strain (GLS) was obtained by averaging all values of regional or segmental peak strain in each apical view, and global circumferential strain (GCS) by averaging the six myocardial segments at 3 short-axis levels. With the use of two-dimensional speckle tracking software to generate LV twist angle and rate
1275
Shin, et al.
automatically, LV twist angle was calculated as the net absolute angle between apical and basal differences in LV rotation (Fig. 1). Peak twist rate was obtained and time to peak rotation of apex and base as well as time to peak LV twist were all measured in the analysis. To adjust heart rate differences in interpatient for time parameter of twist mechanics, the time was normalized to the percentage of each systolic and diastolic period; time 0% at the onset of systolic phase and at the onset of diastolic ejection time (coincidence of aortic valve close), respectively. Diastolic untwist rate was measured at the apex and base level. Peak early untwist rate was defined as the net difference between apical and basal peak untwist rates during diastole. Statistical Analyses: Data analyses were performed with the Statistical Package for Social Science software for Windows 12.0 (SPSS Inc., Chicago, IL, USA). All values are presented as the mean standard deviation for continuous variables, and frequencies for discrete variables. Between-group differences of continuous variables were analyzed using one-way analysis of variance, and categorical variables were analyzed using a chi-square test. Univariate correlations were determined using Pearson’s coefficient for continuous variables. Multiple regression analysis was performed to explore the determinants of LV twist. To assess the incremental value of VAC to the multivariable logistic regression model in discriminating the highest LV
twist tertile group, we constructed receiver-operating characteristic (ROC) curves and compared C-statistics after computing each modeling’s probability of highest LV twist tertile. The calibration was tested with the Hosmer–Lemeshow statistic for goodness-of-fit of the model. P-values were two sided, and a P-value
