Multidirectional Global Left Ventricular Systolic Function in Normal Subjects and Patients with Hypertension: Multicenter Evaluation Daniel A. Morris, MD, Kyoko Otani, MD, Tarek Bekfani, MD, Kiyohiro Takigiku, MD, Chisato Izumi, MD, Satoshi Yuda, MD, Konomi Sakata, MD, Nobuyuki Ohte, MD, Kazuaki Tanabe, MD, Katharina Friedrich, MD, York K€ uhnle, MD, Satoshi Nakatani, MD, Yutaka Otsuji, MD, Wilhelm Haverkamp, MD, Leif-Hendrik Boldt, MD, and Masaaki Takeuchi, MD, Berlin, Germany; Kitakyushu, Azumino, Tenri, Sapporo, Tokyo, Nagoya, Izumo, and Suita, Japan
Background: The aim of this multicenter study was to determine the normal ranges and the clinical relevance of multidirectional systolic parameters to evaluate global left ventricular (LV) systolic function. Methods: Three hundred twenty-three healthy adult subjects prospectively included at 10 centers and a cohort of 310 patients with hypertension were analyzed. Multidirectional global LV systolic function was analyzed using two-dimensional speckle-tracking echocardiography by means of two indices: longitudinalcircumferential systolic index (the average of longitudinal and circumferential global systolic strain) and global systolic index (the average of longitudinal, circumferential, and radial global systolic strain). Results: The ranges of values of the multidirectional systolic parameters in healthy subjects were 21.22 6 2.22% for longitudinal-circumferential systolic index and 29.71 6 5.28% for global systolic index. In addition, the lowest expected values of these multidirectional indices were determined in this population (calculated as 1.96 SDs from the mean): 16.86% for longitudinal-circumferential systolic index and 19.36% for global systolic index. Concerning the clinical relevance of these measurements, these indices indicated the presence of subtle LV global systolic dysfunction in patients with hypertension, even though LV global longitudinal systolic strain and LV ejection fraction were normal. Moreover, in these patients, functional class (dyspnea [New York Heart Association classification]) was inversely related to both the longitudinal-circumferential index and the global systolic index. Conclusions: In the present multicenter study analyzing a large cohort of healthy subjects and patients with hypertension, the normal range and the clinical relevance of multidirectional systolic parameters to evaluate global LV systolic function have been determined. (J Am Soc Echocardiogr 2014;27:493-500.) Keywords: Strain, Speckle-tracking echocardiography, Systolic function, Left ventricular
Left ventricular (LV) ejection fraction (LVEF) is the most common echocardiographic parameter to assess systolic LV function.1 However, this volumetric measurement has several limitations, such University Hospital, Berlin, Germany (D.A.M., T.B., K.F., Y.K., From the Charite W.H., L.-H.B.); University of Occupational and Environmental Health, School of Medicine, Kitakyushu, Japan (K.O., Y.O., M.T.); Nagano Children’s Hospital, Azumino, Japan (K.T.); Tenri Hospital, Tenri, Japan (C.I.); Sapporo Medical University School of Medicine, Sapporo, Japan (S.Y.); Kyorin University School of Medicine, Tokyo, Japan (K.S.); Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan (N.O.); Shimane University Faculty of Medicine, Izumo, Japan (K.T.); and Osaka University Graduate School of Medicine, Suita, Japan (S.N.). Drs Boldt and Takeuchi contributed equally to this study. University Hospital, Campus Reprint requests: Daniel A. Morris, MD, Charite Virchow-Klinikum, Department of Cardiology, Augustenburger Platz 1, 13353 Berlin, Germany (E-mail: [email protected]). 0894-7317/$36.00 Copyright 2014 by the American Society of Echocardiography. http://dx.doi.org/10.1016/j.echo.2014.01.017
as high load dependence and low sensitivity to determine subtle LV systolic dysfunction.2-9 Because of these limitations, research into new and more sensitive systolic parameters has been of significant interest in recent years.10-15 Several studies using two-dimensional (2D) speckle-tracking echocardiography (STE) have demonstrated that despite normal LVEF, many patients in different clinical settings have longitudinal systolic dysfunction of the left ventricle.11-15 Thus, systolic analysis of the left ventricle using global longitudinal systolic strain has been suggested as a new standard assessment for global LV systolic function.16,17 Nevertheless, more recent studies have demonstrated that the global systolic function of the left ventricle is the result of multidirectional contractions in longitudinal, circumferential, and radial directions.18-25 In addition, several studies have found heterogeneous systolic alterations of the left ventricle (in the longitudinal, radial, and circumferential directions) in diverse clinical settings such as diabetes, hypertension, LV hypertrophy, coronary artery disease, and heart failure.18-25 Therefore, analyzing only a unidirectional function (i.e., longitudinal, radial, or circumferential) would merely be a partial approach to assess true global LV systolic function. 493
494 Morris et al
Recent findings in a small group of patients suggested that LV = Left ventricular using multidirectional systolic parameters such as a global systolic LVEF = Left ventricular index (average of longitudinal, ejection fraction circumferential, and radial global STE = Speckle-tracking systolic strain),25 it is possible to echocardiography detect an alteration in multidirectional global LV contractile func2D = Two-dimensional tion in patients with heart failure.25 However, despite these advances in multidirectional parameters to determine true global LV systolic function, the normal ranges and the clinical relevance of these multidirectional indices remain poorly understood. Therefore, the aim of this multicenter study was to establish the range of values of these multidirectional measurements in the healthy population and to demonstrate if these multidirectional indices provide improved detection of LV global systolic dysfunction in patients with a common condition such as hypertension. Abbreviations
METHODS Healthy Population We enrolled healthy subjects $18 years of age prospectively included at nine centers in Japan and one center in Germany. These subjects were part of the Japanese Ultrasound Speckle Tracking of the Left Ventricle Research Project,26 which enrolled healthy volunteers subjects at different university hospitals and was endorsed by the Japanese Society of Echocardiography.26 Healthy subjects were defined as all those with absence of disease or cardiovascular risk factors such as obesity (body mass index $ 30 kg/m2), diabetes (fasting plasma glucose $ 126 mg/dL), hypertension (systolic and diastolic blood pressure $ 140/90 mm Hg), and hypercholesterolemia (fasting plasma low-density lipoprotein cholesterol $ 160 mg/dL); no medications; and normal results on echocardiography according to the diagnostic criteria of the American Society of Echocardiography.1,27,28 The ethics committee of each of the hospitals approved this research project, and informed consent was obtained from all subjects.26 Hypertensive Patient Cohort To determine the clinical relevance of multidirectional systolic parameters to evaluate global LV systolic function, we analyzed a cohort of patients with hypertension included in previous studies of our research group.29,30 Hypertension was defined as systolic and diastolic blood pressure $ 140/90 mm Hg and/or use of antihypertensive medications. To avoid causes of LV myocardial dysfunction other than hypertension, patients with coronary artery disease were excluded (namely, those with unstable angina or non– ST-segment elevation myocardial infarctions, ST-segment elevation acute myocardial infarctions, coronary artery bypass graft surgery, chronic stable angina, or evidence of myocardial ischemia assessed by stress echocardiography). Moreover, with the purpose of excluding causes of dyspnea or LV dysfunction other than hypertension, patients with the following characteristics were excluded: (1) severe pulmonary disease, defined as pulmonary pathology with requirement for supplemental oxygen or need for treatment with corticoids; (2) severe kidney disease, defined as estimated glomerular filtration rate < 30 mL/min/1.73 m2 for $3 months, history of renal transplantation, or severe acute renal failure with requirement for dialysis; (3) severe chronic liver disease or history of liver transplanta-
Journal of the American Society of Echocardiography May 2014
tion; (4) congenital heart disease; (5) pericardial disease characterized by moderate or severe pericardial effusion (echo-free space in enddiastole $ 5 mm) or constrictive pericarditis; (6) cardiomyopathy; and (7) valvular heart disease, defined as mild, moderate, or severe mitral or aortic stenosis, moderate or severe nonfunctional mitral or tricuspid regurgitation, and moderate or severe aortic regurgitation (according to the diagnostic criteria of the guidelines for the management of patients with valvular heart disease of the American College of Cardiology).27 Furthermore, to avoid underestimations of myocardial and mitral annular measurements, patients with valvular heart surgery, mitral annular calcification ($5 mm), cardiac pacing, and poor 2D quality in one or more myocardial segments of the left ventricle (not suitable for analysis by 2D STE in the apical fourchamber, two-chamber, and long-axis views) were also excluded. In addition, to avoid mistakes or large variations in the myocardial measurements of the left ventricle due to variability of the R-R interval, patients with atrial or ventricular arrhythmias were also excluded. Transthoracic Echocardiography All subjects were examined at rest in the left lateral decubitus position using the Vivid 7 or Vivid E9 ultrasound system (GE Healthcare, Little Chalfont, United Kingdom). LV diameters, LV volumes, LV mass, LVEF (determined using Simpson’s method), and LV diastolic function were assessed as recommended by the American Society of Echocardiography.1,28 All echocardiographic measurements using 2D STE, Doppler, and conventional 2D echocardiography were calculated as the averages of three measurements. In addition, to avoid large variations in the myocardial analyses of the left ventricle, subjects with poor 2D imaging quality in one or more segments of the left ventricle were excluded from this study. Two-Dimensional STE The myocardial analyses by 2D STE were performed offline and blinded to the clinical characteristics of the subjects using EchoPAC version 113.0 (GE Healthcare).25,26 The analyses of LV longitudinal systolic strain were performed in the whole myocardium in the basal, middle, and apical segments in the apical four-chamber, twochamber, and long-axis views (i.e., 18 segments of the left ventricle).25,26,31 The average value of peak systolic strain in the longitudinal direction from 18 LV segments was called LV global longitudinal systolic strain.25,26,31 Measurements of LV circumferential and radial systolic strain were performed in the entire myocardium in the three short-axis views of the left ventricle (basal, middle, and apical levels).25,26,31 The average values of peak systolic strain in the circumferential and radial directions from 18 LV segments were called LV global circumferential and radial systolic strain.25,26,31 Multidirectional Global Systolic Function or Myocardial Systolic Performance of the Left Ventricle Using 2D STE and the same 18-segment LV model used for the aforementioned measurements of LV global longitudinal, radial, and circumferential systolic strain, we assessed the multidirectional global systolic function of the left ventricle by means of two indices25: (1) longitudinalcircumferential systolic index = average of longitudinal and circumferential global systolic strain, and (2) global systolic index = average of longitudinal, circumferential, and radial global systolic strain. Examples of how to perform these multidirectional measurements are shown in Figure 1.
Journal of the American Society of Echocardiography Volume 27 Number 5
Morris et al 495
Figure 1 Analysis of multidirectional global systolic function or performance of the left ventricle by means of the global systolic index and the longitudinal-circumferential systolic index. The global systolic index is simply calculated as the average of longitudinal, circumferential, and radial global systolic strain, assuming, for practical purpose, all strain values as positive; thus, in this case, global systolic index = (38 + 22 + 24)/3 = 28. Moreover, it is important to know that specifically the global systolic index can also be calculated as [(global radial systolic strain) (global longitudinal systolic strain + global circumferential systolic strain)]/3; namely, to average negative and positive values, one must subtract the negative value of the global longitudinal and circumferential strain from the positive value of the global radial stain; thus, in the same case, global systolic index = {38 [(22) + (24)]}/3 = [38 (46)]/3 = 84/3 = 28. The longitudinal-circumferential systolic index is simply calculated as the average of the longitudinal and circumferential global systolic strain; thus, in this case, longitudinal-circumferential systolic index = {[(22) + (24)]/2} = [(46)/2] = 23. Statistical Analysis Continuous data are expressed as mean 6 SD and dichotomous data as percentages. Differences in continuous variables between groups were analyzed using Student’s t test. Categorical variables were compared using c2 and Fisher’s exact tests as appropriate.
Comparisons among three or more groups were analyzed using one-way analysis of variance. The relationships of the global systolic index and the longitudinal-circumferential systolic index with continuous variables were analyzed using a simple regression analysis. In addition, to determine the variables with the strongest associations
496 Morris et al
Journal of the American Society of Echocardiography May 2014
Table 1 Clinical characteristics of the cohort of healthy subjects and patients with hypertension Variable
Clinical Characteristics Age (y) Women Body mass index (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Conventional LV measurements LVEF (%) LV mass (g/m2) Mitral early diastolic inflow velocity (E) (cm/sec) Mitral E/A ratio Lateral early diastolic mitral annular velocity by DTI (e0 ) (cm/sec) Septal early diastolic mitral annular velocity by DTI (e0 ) (cm/sec)
Healthy subjects (n = 323)
Patients with hypertension (n = 310)
37.4 6 12.9 43.3% 22.2 6 2.5 118.7 6 10.3 70.7 6 8.9
69.5 6 9.6 39.4% 28.3 6 4.8 138.1 6 21.9 80.2 6 12.4
63.9 6 5.5 75.5 6 15.3 79.3 6 15.7 1.6 6 0.5 13.8 6 2.9 11.3 6 2.2
61.3 6 6.8 113.2 6 28.0 71.2 6 20.5 1.0 6 0.6 5.5 6 1.6 7.4 6 1.7
DTI, Doppler tissue imaging. Data are expressed as mean 6 SD or as percentages.
with these multidirectional indices, we performed a multivariate stepwise forward linear regression analysis. Following the recommendations on chamber quantification of the American Society of Echocardiography,1 the lowest expected values of both the global systolic index and the longitudinal-circumferential systolic index were calculated as 1.96 SDs from the mean. With the purpose of determining the reproducibility of the global systolic index and the longitudinal-circumferential systolic index, we analyzed the intraobserver and interobserver variability of these indices in 20 randomly selected subjects. All statistical analyses were performed using StatView version 5.0 (SAS Institute Inc, Cary, NC) and SPSS version 19.0 (IBM, Armonk, NY). Differences were considered statistically significant at P < .05. RESULTS Clinical Characteristics of the Cohort of Healthy Subjects and Patients with Hypertension A total of 340 healthy subjects met the eligibility criteria during the study period. However, in this group of individuals, multidirectional global LV systolic function evaluated by the longitudinalcircumferential systolic index and the global systolic index could not be analyzed in 17 subjects because of inadequate 2D imaging quality for an analysis by 2D STE in one or more segments of the left ventricle (feasibility, 95%). Thus, 323 healthy adult subjects (221 Asian and 102 European) with adequate imaging quality for an analysis by 2D STE were studied and analyzed. Clinical characteristics and conventional LV measurements of these subjects are shown in Table 1. Concerning the cohort of patients with hypertension, a total of 340 patients were initially included. Thirty patients had inadequate 2D imaging quality for an analysis by 2D STE (feasibility, 91.2%). Accordingly, 310 patients with hypertension (duration of hypertension, 5.5 6 2.5 years) were analyzed. Clinical characteristics and conventional LV measurements of these patients are shown in Table 1. Range of Values of Multidirectional LV Systolic Indices in Healthy Subjects The range of values in healthy subjects of the multidirectional systolic parameters (longitudinal-circumferential systolic index and global sys-
tolic index) to evaluate global LV systolic function are shown in Table 2. In addition, we determined the lowest expected value of these multidirectional indices in this population (calculated as 1.96 SDs from the mean), which is displayed in Table 2. The range of values of each component of these multidirectional parameters (i.e., longitudinal, radial, and circumferential global systolic strain) are also presented in Table 2.
Distribution of Multidirectional LV Systolic Indices According to Age and Gender in the Healthy Population The multidirectional systolic parameters to determine global LV systolic function did not differ significantly among the different age groups (see Table 3). In this regard, both the longitudinalcircumferential systolic index and the global systolic index neither increased nor decreased considerably with age (see Table 3). Furthermore, there were no important differences in the values of the longitudinal-circumferential systolic index between women and men (see Table 3).
Interrelations of Multidirectional LV Systolic Parameters in Healthy Subjects In the analysis of factors that could influence the measurement of global LV systolic function using multidirectional parameters, we found that in healthy subjects, both the global systolic index and the longitudinal-circumferential systolic index were slightly influenced by clinical variables such as age, body surface area, and LV mass (see Table 4). Furthermore, to analyze a possible race variation in the evaluation of the global LV systolic function by means of these multidirectional systolic indices, we analyzed a subgroup of 200 Asian (Japanese) and European (German) healthy subjects of similar age and using the same software package (EchoPAC version 113.0). There were no significant differences in the values of global LV systolic function between Asians (n = 100; age, 37.7 6 11.2 years; global systolic index, 29.44 6 4.92%) and Europeans (n = 100; age, 35.9 6 12.8 years; global systolic index, 28.13 6 5.26%) (P = .07).
Morris et al 497
Journal of the American Society of Echocardiography Volume 27 Number 5
Table 2 Unidirectional and multidirectional LV systolic parameters using 2D STE Variable
Unidirectional LV systolic function LV global longitudinal systolic strain (%) Mean 6 SD Lowest expected value* LV global circumferential systolic strain (%) Mean 6 SD Lowest expected value* LV global radial systolic strain (%) Mean 6 SD Lowest expected value* Multidirectional LV systolic function Longitudinal-circumferential systolic index (%) Mean 6 SD Lowest expected value* Global systolic index (%) Mean 6 SD Lowest expected value*
Healthy subjects (n = 323)
Patients with hypertension (n = 310)
21.23 6 2.03 17.25
17.46 6 3.58 10.44
21.21 6 3.25 14.84
18.01 6 3.68 10.79
46.68 6 13.54 20.14
35.72 6 15.58 5.18
21.22 6 2.22 16.86
17.74 6 3.11 11.64
29.71 6 5.28 19.36
23.73 6 6.31 11.36
Data are expressed as mean 6 SD. The differences between healthy subjects and subjects with hypertension were statistically significant for all measurements (P < .0001). *The lowest expected value was calculated as 1.96 SDs from the mean.
Clinical Relevance of Multidirectional Systolic Parameters to Evaluate Global LV Systolic Function Concerning the clinical relevance of the longitudinal-circumferential systolic index and the global systolic index to evaluate global LV systolic function, these indices indicated the presence of subtle LV global systolic dysfunction in patients with hypertension, even though LV global longitudinal systolic strain and LVEF were normal (see Figure 2). In addition, these multidirectional parameters also revealed the presence of subtle LV global systolic dysfunction in patients with hypertension with normal LV global circumferential systolic strain as well as in those with normal LV global radial systolic strain (rate of subtle LV global dysfunction determined by a low longitudinal-circumferential systolic index, 23.3% and 34.1%, or by a low global systolic index 21.3% and 19.5%, respectively, using the lowest expected values of these indices in healthy subjects to determine a low index). Furthermore, in patients with hypertension, these multidirectional indices were significantly lower in symptomatic than in asymptomatic patients (as well as in comparison with healthy subjects; see Tables 2, 3, and 5). In agreement, functional class (dyspnea [New York Heart Association classification]) was inversely related to both the longitudinal-circumferential index and the global systolic index in patients with hypertension (see Table 5). Regarding the interrelations of the longitudinal-circumferential index and the global systolic index in patients with hypertension, these multidirectional parameters were minimally affected by age, gender, and other clinical variables such as systolic blood pressure (see Tables 3 and 4). On the other hand, both the longitudinalcircumferential index and the global systolic index were inversely related to LV mass and significantly correlated with LVEF in these patients (see Table 4). Reproducibility of Multidirectional Systolic Parameters to Evaluate Global LV Systolic Function The reproducibility of the longitudinal-circumferential systolic index and the global systolic index was adequate, with low intraobserver and interobserver variability (absolute mean differences for
longitudinal-circumferential systolic index, 0.33 6 0.27% and 0.45 6 0.40%, respectively; absolute mean differences for global systolic index, 0.42 6 0.36% and 0.88 6 0.49%, respectively). DISCUSSION In the present multicenter study analyzing a large cohort of healthy subjects and patients with hypertension, we have determined the normal range and the clinical relevance of multidirectional systolic parameters to analyze global LV systolic function. Volumetric versus Myocardial Analyses of the Left Ventricle LVEF remains the main echocardiographic parameter to assess the systolic function of the left ventricle.1 However, it has been demonstrated that this volumetric measurement has several limitations, such as high load dependence and low sensitivity to detect subtle LV myocardial systolic dysfunction.2-9 Several studies have shown that this method is highly influenced by preload and afterload changes.2-5,32,33 In addition, recent studies using 2D STE have shown the low sensitivity of LVEF to identify early myocardial systolic alteration of the left ventricle in different clinical settings.11-15,18-24 In agreement with these reports, we demonstrate that multidirectional myocardial parameters derived from the average of longitudinal, circumferential, and radial systolic strain (i.e., the longitudinalcircumferential systolic index and global systolic index) identified global LV systolic dysfunction in patients with hypertension despite normal LVEF. Furthermore, we found that these myocardial indices were minimally affected by age, gender, and other clinical variables such as race and systolic blood pressure (an indirect parameter of LV afterload). Therefore, we consider that the analysis of global LV systolic function using multidirectional parameters with 2D STE has several advantages over LVEF, which makes these multidirectional measurements essential methods to assess the true or intrinsic global contractile function of the left ventricle.
498 Morris et al
Journal of the American Society of Echocardiography May 2014
Table 3 Distribution of multidirectional LV systolic parameters according to age and gender
Variable
Longitudinal-circumferential systolic index Age distribution 18–59 y $60 y P Gender distribution Women Men P Global systolic index Age distribution 18–59 y $60 y P Gender distribution Women Men P
Healthy subjects (n = 323)
Table 4 Interrelations of multidirectional LV systolic parameters in healthy subjects and patients with hypertension
Patients with hypertension (n = 310)
Longitudinalcircumferential systolic index Variable
21.24 6 2.21% 20.94 6 2.27% .5350
17.18 6 2.97% 17.85 6 3.13% .1606
21.39 6 2.26% 21.09 6 2.17% .2279
17.98 6 2.95% 17.58 6 3.21% .2686
29.68 6 5.23% 30.09 6 6.08% .7273
22.03 6 6.02% 24.07 6 6.32% .0345
28.93 6 5.32% 30.30 6 5.19% .0207
24.98 6 6.72% 22.92 6 5.90% .0048
Data are expressed as mean 6 SD. The differences in the longitudinal-circumferential systolic index and global systolic index between healthy subjects and subjects with hypertension were statistically significant in all groups of age and gender (P < .05).
Unidirectional versus Multidirectional Approach for Assessing Global LV Systolic Function Because of the limitations of LVEF to identify subclinical LV systolic dysfunction,6-9 research into new and more sensitive systolic parameters has been of significant interest in recent years.10-15 Several studies using 2D STE have demonstrated that despite normal LVEF many patients in different clinical settings have longitudinal systolic dysfunction of the left ventricle.11-15 Thus, systolic analysis of the left ventricle using global longitudinal systolic strain has been suggested as a new standard assessment of global LV systolic function.16,17 Nevertheless, more recent studies have demonstrated that global LV systolic function is the result of multidirectional contractions in longitudinal, circumferential, and radial directions.18-25 In addition, several studies have found heterogeneous systolic alterations of the left ventricle (namely, in longitudinal, radial, and circumferential directions) in diverse clinical settings, such as diabetes, hypertension, LV hypertrophy, coronary artery disease, and heart failure.18-25 Therefore, analyzing only a unidirectional function (i.e., longitudinal, radial, or circumferential) would merely be a partial approach to assess true global LV contractile function. Accordingly, using multidirectional systolic parameters such as the longitudinal-circumferential systolic index or the global systolic index, we could determine true global LV systolic function or myocardial systolic performance of the left ventricle. In this respect, in the present study, we demonstrate that the longitudinal-circumferential systolic index and the global systolic index detected global LV systolic dysfunction in patients with hypertension despite normal LV global longitudinal systolic strain. Furthermore, we found that the functional class (dyspnea [New York Heart Association classification]) in patients with hypertension
Healthy subjects (n = 323) Age Systolic blood pressure Mitral E/e0 septal-lateral ratio LVEF LV mass Body surface area Hypertensive Patients (n= 310) Age Systolic blood pressure Mitral E/e0 septal-lateral ratio LVEF LV mass Body surface area
r
0.09 0.03 0.11
P
.0773 .4973 .0502
Global systolic index r
0.08 0.09 0.03
P
.1445 .0852 .5947
0.18* .0013 0.19* 0.06 .2873 0.05 0.25 < .0001 0.07
.0006 .3436 .1878
0.01 0.09 0.30
.6766 .1243 .0519
.9985 0.02 .1932 0.11 < .0001 0.12
0.39* .0013 0.23* < .0001 0.31* < .0001 0.18* .0231 0.13 .0926 0.27 .0003
*Variable with the strongest associations with the longitudinalcircumferential index and the global systolic index in a forward stepwise multivariate regression analysis.
was inversely related to these multidirectional indices. Hence, we believe that the multidirectional systolic indices addressed in the present study meet the conditions to be considered valuable measurements of the global systolic function of the left ventricle. Determination of Normal LV Global Systolic Function Although recent studies analyzing healthy subjects had identified the normal range of the unidirectional contractile function of the left ventricle (i.e., longitudinal, circumferential, or radial),26,34-38 the normal range of multidirectional global LV systolic function or myocardial systolic performance (i.e., the result or the average of longitudinal, circumferential, and radial function) had not yet been determined. For these reasons, the objective of the present multicenter study was to analyze a large cohort of healthy subjects to determine the normal range of multidirectional global LV contractile function. In this regard, in 323 healthy subjects, we determined the range of values of multidirectional global LV systolic function, which was analyzed using multidirectional systolic indices with adequate feasibility and reproducibility. In addition, we showed that using the lowest percentile of healthy subjects of these multidirectional parameters, it was possible to detect global LV systolic dysfunction in patients with hypertension even when conventional and unidirectional LV measurements, such as LVEF and LV global longitudinal strain, were normal. Limitations Some considerations should be taken into account when analyzing global LV systolic function by means of multidirectional systolic parameters using 2D STE. Recent studies have shown that 2D speckle-tracking echocardiographic values of the left ventricle vary among different
Morris et al 499
Journal of the American Society of Echocardiography Volume 27 Number 5
Figure 2 Global LV systolic dysfunction evaluated by the longitudinal-circumferential systolic index and the global systolic index in patients with hypertension, despite normal LVEF and normal LV global longitudinal systolic strain. To define normal LV global longitudinal systolic strain, low longitudinal-circumferential systolic index, and low global systolic index, the lowest expected values of these measurements in the healthy population were used as cutoffs (namely, 17.25%, 16.86%, and 19.36%, respectively; see Table 2). Normal LVEF was defined as $55% by the biplane Simpson’s method.
Table 5 Worsening of symptomatic status linked to the deterioration of multidirectional LV systolic function in patients with hypertension NYHA functional class Index
I (n = 212)
II (n = 67)
III and IV (n = 31)
P (ANOVA)
Global systolic index Longitudinal-circumferential systolic index
25.2 6 6.1% 18.9 6 2.5%
20.7 6 5.7%* 15.2 6 2.5%*
19.9 6 5.3% 15.1 6 2.7%†
Background: The aim of this multicenter study was to determine the normal ranges and the clinical relevance of multidirectional systolic parameters to evaluate global left ventricular (LV) systolic function. Methods: Three hundred twenty-three healthy adult subjects prospectively included at 10 centers and a cohort of 310 patients with hypertension were analyzed. Multidirectional global LV systolic function was analyzed using two-dimensional speckle-tracking echocardiography by means of two indices: longitudinalcircumferential systolic index (the average of longitudinal and circumferential global systolic strain) and global systolic index (the average of longitudinal, circumferential, and radial global systolic strain). Results: The ranges of values of the multidirectional systolic parameters in healthy subjects were 21.22 6 2.22% for longitudinal-circumferential systolic index and 29.71 6 5.28% for global systolic index. In addition, the lowest expected values of these multidirectional indices were determined in this population (calculated as 1.96 SDs from the mean): 16.86% for longitudinal-circumferential systolic index and 19.36% for global systolic index. Concerning the clinical relevance of these measurements, these indices indicated the presence of subtle LV global systolic dysfunction in patients with hypertension, even though LV global longitudinal systolic strain and LV ejection fraction were normal. Moreover, in these patients, functional class (dyspnea [New York Heart Association classification]) was inversely related to both the longitudinal-circumferential index and the global systolic index. Conclusions: In the present multicenter study analyzing a large cohort of healthy subjects and patients with hypertension, the normal range and the clinical relevance of multidirectional systolic parameters to evaluate global LV systolic function have been determined. (J Am Soc Echocardiogr 2014;27:493-500.) Keywords: Strain, Speckle-tracking echocardiography, Systolic function, Left ventricular
Left ventricular (LV) ejection fraction (LVEF) is the most common echocardiographic parameter to assess systolic LV function.1 However, this volumetric measurement has several limitations, such University Hospital, Berlin, Germany (D.A.M., T.B., K.F., Y.K., From the Charite W.H., L.-H.B.); University of Occupational and Environmental Health, School of Medicine, Kitakyushu, Japan (K.O., Y.O., M.T.); Nagano Children’s Hospital, Azumino, Japan (K.T.); Tenri Hospital, Tenri, Japan (C.I.); Sapporo Medical University School of Medicine, Sapporo, Japan (S.Y.); Kyorin University School of Medicine, Tokyo, Japan (K.S.); Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan (N.O.); Shimane University Faculty of Medicine, Izumo, Japan (K.T.); and Osaka University Graduate School of Medicine, Suita, Japan (S.N.). Drs Boldt and Takeuchi contributed equally to this study. University Hospital, Campus Reprint requests: Daniel A. Morris, MD, Charite Virchow-Klinikum, Department of Cardiology, Augustenburger Platz 1, 13353 Berlin, Germany (E-mail: [email protected]). 0894-7317/$36.00 Copyright 2014 by the American Society of Echocardiography. http://dx.doi.org/10.1016/j.echo.2014.01.017
as high load dependence and low sensitivity to determine subtle LV systolic dysfunction.2-9 Because of these limitations, research into new and more sensitive systolic parameters has been of significant interest in recent years.10-15 Several studies using two-dimensional (2D) speckle-tracking echocardiography (STE) have demonstrated that despite normal LVEF, many patients in different clinical settings have longitudinal systolic dysfunction of the left ventricle.11-15 Thus, systolic analysis of the left ventricle using global longitudinal systolic strain has been suggested as a new standard assessment for global LV systolic function.16,17 Nevertheless, more recent studies have demonstrated that the global systolic function of the left ventricle is the result of multidirectional contractions in longitudinal, circumferential, and radial directions.18-25 In addition, several studies have found heterogeneous systolic alterations of the left ventricle (in the longitudinal, radial, and circumferential directions) in diverse clinical settings such as diabetes, hypertension, LV hypertrophy, coronary artery disease, and heart failure.18-25 Therefore, analyzing only a unidirectional function (i.e., longitudinal, radial, or circumferential) would merely be a partial approach to assess true global LV systolic function. 493
494 Morris et al
Recent findings in a small group of patients suggested that LV = Left ventricular using multidirectional systolic parameters such as a global systolic LVEF = Left ventricular index (average of longitudinal, ejection fraction circumferential, and radial global STE = Speckle-tracking systolic strain),25 it is possible to echocardiography detect an alteration in multidirectional global LV contractile func2D = Two-dimensional tion in patients with heart failure.25 However, despite these advances in multidirectional parameters to determine true global LV systolic function, the normal ranges and the clinical relevance of these multidirectional indices remain poorly understood. Therefore, the aim of this multicenter study was to establish the range of values of these multidirectional measurements in the healthy population and to demonstrate if these multidirectional indices provide improved detection of LV global systolic dysfunction in patients with a common condition such as hypertension. Abbreviations
METHODS Healthy Population We enrolled healthy subjects $18 years of age prospectively included at nine centers in Japan and one center in Germany. These subjects were part of the Japanese Ultrasound Speckle Tracking of the Left Ventricle Research Project,26 which enrolled healthy volunteers subjects at different university hospitals and was endorsed by the Japanese Society of Echocardiography.26 Healthy subjects were defined as all those with absence of disease or cardiovascular risk factors such as obesity (body mass index $ 30 kg/m2), diabetes (fasting plasma glucose $ 126 mg/dL), hypertension (systolic and diastolic blood pressure $ 140/90 mm Hg), and hypercholesterolemia (fasting plasma low-density lipoprotein cholesterol $ 160 mg/dL); no medications; and normal results on echocardiography according to the diagnostic criteria of the American Society of Echocardiography.1,27,28 The ethics committee of each of the hospitals approved this research project, and informed consent was obtained from all subjects.26 Hypertensive Patient Cohort To determine the clinical relevance of multidirectional systolic parameters to evaluate global LV systolic function, we analyzed a cohort of patients with hypertension included in previous studies of our research group.29,30 Hypertension was defined as systolic and diastolic blood pressure $ 140/90 mm Hg and/or use of antihypertensive medications. To avoid causes of LV myocardial dysfunction other than hypertension, patients with coronary artery disease were excluded (namely, those with unstable angina or non– ST-segment elevation myocardial infarctions, ST-segment elevation acute myocardial infarctions, coronary artery bypass graft surgery, chronic stable angina, or evidence of myocardial ischemia assessed by stress echocardiography). Moreover, with the purpose of excluding causes of dyspnea or LV dysfunction other than hypertension, patients with the following characteristics were excluded: (1) severe pulmonary disease, defined as pulmonary pathology with requirement for supplemental oxygen or need for treatment with corticoids; (2) severe kidney disease, defined as estimated glomerular filtration rate < 30 mL/min/1.73 m2 for $3 months, history of renal transplantation, or severe acute renal failure with requirement for dialysis; (3) severe chronic liver disease or history of liver transplanta-
Journal of the American Society of Echocardiography May 2014
tion; (4) congenital heart disease; (5) pericardial disease characterized by moderate or severe pericardial effusion (echo-free space in enddiastole $ 5 mm) or constrictive pericarditis; (6) cardiomyopathy; and (7) valvular heart disease, defined as mild, moderate, or severe mitral or aortic stenosis, moderate or severe nonfunctional mitral or tricuspid regurgitation, and moderate or severe aortic regurgitation (according to the diagnostic criteria of the guidelines for the management of patients with valvular heart disease of the American College of Cardiology).27 Furthermore, to avoid underestimations of myocardial and mitral annular measurements, patients with valvular heart surgery, mitral annular calcification ($5 mm), cardiac pacing, and poor 2D quality in one or more myocardial segments of the left ventricle (not suitable for analysis by 2D STE in the apical fourchamber, two-chamber, and long-axis views) were also excluded. In addition, to avoid mistakes or large variations in the myocardial measurements of the left ventricle due to variability of the R-R interval, patients with atrial or ventricular arrhythmias were also excluded. Transthoracic Echocardiography All subjects were examined at rest in the left lateral decubitus position using the Vivid 7 or Vivid E9 ultrasound system (GE Healthcare, Little Chalfont, United Kingdom). LV diameters, LV volumes, LV mass, LVEF (determined using Simpson’s method), and LV diastolic function were assessed as recommended by the American Society of Echocardiography.1,28 All echocardiographic measurements using 2D STE, Doppler, and conventional 2D echocardiography were calculated as the averages of three measurements. In addition, to avoid large variations in the myocardial analyses of the left ventricle, subjects with poor 2D imaging quality in one or more segments of the left ventricle were excluded from this study. Two-Dimensional STE The myocardial analyses by 2D STE were performed offline and blinded to the clinical characteristics of the subjects using EchoPAC version 113.0 (GE Healthcare).25,26 The analyses of LV longitudinal systolic strain were performed in the whole myocardium in the basal, middle, and apical segments in the apical four-chamber, twochamber, and long-axis views (i.e., 18 segments of the left ventricle).25,26,31 The average value of peak systolic strain in the longitudinal direction from 18 LV segments was called LV global longitudinal systolic strain.25,26,31 Measurements of LV circumferential and radial systolic strain were performed in the entire myocardium in the three short-axis views of the left ventricle (basal, middle, and apical levels).25,26,31 The average values of peak systolic strain in the circumferential and radial directions from 18 LV segments were called LV global circumferential and radial systolic strain.25,26,31 Multidirectional Global Systolic Function or Myocardial Systolic Performance of the Left Ventricle Using 2D STE and the same 18-segment LV model used for the aforementioned measurements of LV global longitudinal, radial, and circumferential systolic strain, we assessed the multidirectional global systolic function of the left ventricle by means of two indices25: (1) longitudinalcircumferential systolic index = average of longitudinal and circumferential global systolic strain, and (2) global systolic index = average of longitudinal, circumferential, and radial global systolic strain. Examples of how to perform these multidirectional measurements are shown in Figure 1.
Journal of the American Society of Echocardiography Volume 27 Number 5
Morris et al 495
Figure 1 Analysis of multidirectional global systolic function or performance of the left ventricle by means of the global systolic index and the longitudinal-circumferential systolic index. The global systolic index is simply calculated as the average of longitudinal, circumferential, and radial global systolic strain, assuming, for practical purpose, all strain values as positive; thus, in this case, global systolic index = (38 + 22 + 24)/3 = 28. Moreover, it is important to know that specifically the global systolic index can also be calculated as [(global radial systolic strain) (global longitudinal systolic strain + global circumferential systolic strain)]/3; namely, to average negative and positive values, one must subtract the negative value of the global longitudinal and circumferential strain from the positive value of the global radial stain; thus, in the same case, global systolic index = {38 [(22) + (24)]}/3 = [38 (46)]/3 = 84/3 = 28. The longitudinal-circumferential systolic index is simply calculated as the average of the longitudinal and circumferential global systolic strain; thus, in this case, longitudinal-circumferential systolic index = {[(22) + (24)]/2} = [(46)/2] = 23. Statistical Analysis Continuous data are expressed as mean 6 SD and dichotomous data as percentages. Differences in continuous variables between groups were analyzed using Student’s t test. Categorical variables were compared using c2 and Fisher’s exact tests as appropriate.
Comparisons among three or more groups were analyzed using one-way analysis of variance. The relationships of the global systolic index and the longitudinal-circumferential systolic index with continuous variables were analyzed using a simple regression analysis. In addition, to determine the variables with the strongest associations
496 Morris et al
Journal of the American Society of Echocardiography May 2014
Table 1 Clinical characteristics of the cohort of healthy subjects and patients with hypertension Variable
Clinical Characteristics Age (y) Women Body mass index (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Conventional LV measurements LVEF (%) LV mass (g/m2) Mitral early diastolic inflow velocity (E) (cm/sec) Mitral E/A ratio Lateral early diastolic mitral annular velocity by DTI (e0 ) (cm/sec) Septal early diastolic mitral annular velocity by DTI (e0 ) (cm/sec)
Healthy subjects (n = 323)
Patients with hypertension (n = 310)
37.4 6 12.9 43.3% 22.2 6 2.5 118.7 6 10.3 70.7 6 8.9
69.5 6 9.6 39.4% 28.3 6 4.8 138.1 6 21.9 80.2 6 12.4
63.9 6 5.5 75.5 6 15.3 79.3 6 15.7 1.6 6 0.5 13.8 6 2.9 11.3 6 2.2
61.3 6 6.8 113.2 6 28.0 71.2 6 20.5 1.0 6 0.6 5.5 6 1.6 7.4 6 1.7
DTI, Doppler tissue imaging. Data are expressed as mean 6 SD or as percentages.
with these multidirectional indices, we performed a multivariate stepwise forward linear regression analysis. Following the recommendations on chamber quantification of the American Society of Echocardiography,1 the lowest expected values of both the global systolic index and the longitudinal-circumferential systolic index were calculated as 1.96 SDs from the mean. With the purpose of determining the reproducibility of the global systolic index and the longitudinal-circumferential systolic index, we analyzed the intraobserver and interobserver variability of these indices in 20 randomly selected subjects. All statistical analyses were performed using StatView version 5.0 (SAS Institute Inc, Cary, NC) and SPSS version 19.0 (IBM, Armonk, NY). Differences were considered statistically significant at P < .05. RESULTS Clinical Characteristics of the Cohort of Healthy Subjects and Patients with Hypertension A total of 340 healthy subjects met the eligibility criteria during the study period. However, in this group of individuals, multidirectional global LV systolic function evaluated by the longitudinalcircumferential systolic index and the global systolic index could not be analyzed in 17 subjects because of inadequate 2D imaging quality for an analysis by 2D STE in one or more segments of the left ventricle (feasibility, 95%). Thus, 323 healthy adult subjects (221 Asian and 102 European) with adequate imaging quality for an analysis by 2D STE were studied and analyzed. Clinical characteristics and conventional LV measurements of these subjects are shown in Table 1. Concerning the cohort of patients with hypertension, a total of 340 patients were initially included. Thirty patients had inadequate 2D imaging quality for an analysis by 2D STE (feasibility, 91.2%). Accordingly, 310 patients with hypertension (duration of hypertension, 5.5 6 2.5 years) were analyzed. Clinical characteristics and conventional LV measurements of these patients are shown in Table 1. Range of Values of Multidirectional LV Systolic Indices in Healthy Subjects The range of values in healthy subjects of the multidirectional systolic parameters (longitudinal-circumferential systolic index and global sys-
tolic index) to evaluate global LV systolic function are shown in Table 2. In addition, we determined the lowest expected value of these multidirectional indices in this population (calculated as 1.96 SDs from the mean), which is displayed in Table 2. The range of values of each component of these multidirectional parameters (i.e., longitudinal, radial, and circumferential global systolic strain) are also presented in Table 2.
Distribution of Multidirectional LV Systolic Indices According to Age and Gender in the Healthy Population The multidirectional systolic parameters to determine global LV systolic function did not differ significantly among the different age groups (see Table 3). In this regard, both the longitudinalcircumferential systolic index and the global systolic index neither increased nor decreased considerably with age (see Table 3). Furthermore, there were no important differences in the values of the longitudinal-circumferential systolic index between women and men (see Table 3).
Interrelations of Multidirectional LV Systolic Parameters in Healthy Subjects In the analysis of factors that could influence the measurement of global LV systolic function using multidirectional parameters, we found that in healthy subjects, both the global systolic index and the longitudinal-circumferential systolic index were slightly influenced by clinical variables such as age, body surface area, and LV mass (see Table 4). Furthermore, to analyze a possible race variation in the evaluation of the global LV systolic function by means of these multidirectional systolic indices, we analyzed a subgroup of 200 Asian (Japanese) and European (German) healthy subjects of similar age and using the same software package (EchoPAC version 113.0). There were no significant differences in the values of global LV systolic function between Asians (n = 100; age, 37.7 6 11.2 years; global systolic index, 29.44 6 4.92%) and Europeans (n = 100; age, 35.9 6 12.8 years; global systolic index, 28.13 6 5.26%) (P = .07).
Morris et al 497
Journal of the American Society of Echocardiography Volume 27 Number 5
Table 2 Unidirectional and multidirectional LV systolic parameters using 2D STE Variable
Unidirectional LV systolic function LV global longitudinal systolic strain (%) Mean 6 SD Lowest expected value* LV global circumferential systolic strain (%) Mean 6 SD Lowest expected value* LV global radial systolic strain (%) Mean 6 SD Lowest expected value* Multidirectional LV systolic function Longitudinal-circumferential systolic index (%) Mean 6 SD Lowest expected value* Global systolic index (%) Mean 6 SD Lowest expected value*
Healthy subjects (n = 323)
Patients with hypertension (n = 310)
21.23 6 2.03 17.25
17.46 6 3.58 10.44
21.21 6 3.25 14.84
18.01 6 3.68 10.79
46.68 6 13.54 20.14
35.72 6 15.58 5.18
21.22 6 2.22 16.86
17.74 6 3.11 11.64
29.71 6 5.28 19.36
23.73 6 6.31 11.36
Data are expressed as mean 6 SD. The differences between healthy subjects and subjects with hypertension were statistically significant for all measurements (P < .0001). *The lowest expected value was calculated as 1.96 SDs from the mean.
Clinical Relevance of Multidirectional Systolic Parameters to Evaluate Global LV Systolic Function Concerning the clinical relevance of the longitudinal-circumferential systolic index and the global systolic index to evaluate global LV systolic function, these indices indicated the presence of subtle LV global systolic dysfunction in patients with hypertension, even though LV global longitudinal systolic strain and LVEF were normal (see Figure 2). In addition, these multidirectional parameters also revealed the presence of subtle LV global systolic dysfunction in patients with hypertension with normal LV global circumferential systolic strain as well as in those with normal LV global radial systolic strain (rate of subtle LV global dysfunction determined by a low longitudinal-circumferential systolic index, 23.3% and 34.1%, or by a low global systolic index 21.3% and 19.5%, respectively, using the lowest expected values of these indices in healthy subjects to determine a low index). Furthermore, in patients with hypertension, these multidirectional indices were significantly lower in symptomatic than in asymptomatic patients (as well as in comparison with healthy subjects; see Tables 2, 3, and 5). In agreement, functional class (dyspnea [New York Heart Association classification]) was inversely related to both the longitudinal-circumferential index and the global systolic index in patients with hypertension (see Table 5). Regarding the interrelations of the longitudinal-circumferential index and the global systolic index in patients with hypertension, these multidirectional parameters were minimally affected by age, gender, and other clinical variables such as systolic blood pressure (see Tables 3 and 4). On the other hand, both the longitudinalcircumferential index and the global systolic index were inversely related to LV mass and significantly correlated with LVEF in these patients (see Table 4). Reproducibility of Multidirectional Systolic Parameters to Evaluate Global LV Systolic Function The reproducibility of the longitudinal-circumferential systolic index and the global systolic index was adequate, with low intraobserver and interobserver variability (absolute mean differences for
longitudinal-circumferential systolic index, 0.33 6 0.27% and 0.45 6 0.40%, respectively; absolute mean differences for global systolic index, 0.42 6 0.36% and 0.88 6 0.49%, respectively). DISCUSSION In the present multicenter study analyzing a large cohort of healthy subjects and patients with hypertension, we have determined the normal range and the clinical relevance of multidirectional systolic parameters to analyze global LV systolic function. Volumetric versus Myocardial Analyses of the Left Ventricle LVEF remains the main echocardiographic parameter to assess the systolic function of the left ventricle.1 However, it has been demonstrated that this volumetric measurement has several limitations, such as high load dependence and low sensitivity to detect subtle LV myocardial systolic dysfunction.2-9 Several studies have shown that this method is highly influenced by preload and afterload changes.2-5,32,33 In addition, recent studies using 2D STE have shown the low sensitivity of LVEF to identify early myocardial systolic alteration of the left ventricle in different clinical settings.11-15,18-24 In agreement with these reports, we demonstrate that multidirectional myocardial parameters derived from the average of longitudinal, circumferential, and radial systolic strain (i.e., the longitudinalcircumferential systolic index and global systolic index) identified global LV systolic dysfunction in patients with hypertension despite normal LVEF. Furthermore, we found that these myocardial indices were minimally affected by age, gender, and other clinical variables such as race and systolic blood pressure (an indirect parameter of LV afterload). Therefore, we consider that the analysis of global LV systolic function using multidirectional parameters with 2D STE has several advantages over LVEF, which makes these multidirectional measurements essential methods to assess the true or intrinsic global contractile function of the left ventricle.
498 Morris et al
Journal of the American Society of Echocardiography May 2014
Table 3 Distribution of multidirectional LV systolic parameters according to age and gender
Variable
Longitudinal-circumferential systolic index Age distribution 18–59 y $60 y P Gender distribution Women Men P Global systolic index Age distribution 18–59 y $60 y P Gender distribution Women Men P
Healthy subjects (n = 323)
Table 4 Interrelations of multidirectional LV systolic parameters in healthy subjects and patients with hypertension
Patients with hypertension (n = 310)
Longitudinalcircumferential systolic index Variable
21.24 6 2.21% 20.94 6 2.27% .5350
17.18 6 2.97% 17.85 6 3.13% .1606
21.39 6 2.26% 21.09 6 2.17% .2279
17.98 6 2.95% 17.58 6 3.21% .2686
29.68 6 5.23% 30.09 6 6.08% .7273
22.03 6 6.02% 24.07 6 6.32% .0345
28.93 6 5.32% 30.30 6 5.19% .0207
24.98 6 6.72% 22.92 6 5.90% .0048
Data are expressed as mean 6 SD. The differences in the longitudinal-circumferential systolic index and global systolic index between healthy subjects and subjects with hypertension were statistically significant in all groups of age and gender (P < .05).
Unidirectional versus Multidirectional Approach for Assessing Global LV Systolic Function Because of the limitations of LVEF to identify subclinical LV systolic dysfunction,6-9 research into new and more sensitive systolic parameters has been of significant interest in recent years.10-15 Several studies using 2D STE have demonstrated that despite normal LVEF many patients in different clinical settings have longitudinal systolic dysfunction of the left ventricle.11-15 Thus, systolic analysis of the left ventricle using global longitudinal systolic strain has been suggested as a new standard assessment of global LV systolic function.16,17 Nevertheless, more recent studies have demonstrated that global LV systolic function is the result of multidirectional contractions in longitudinal, circumferential, and radial directions.18-25 In addition, several studies have found heterogeneous systolic alterations of the left ventricle (namely, in longitudinal, radial, and circumferential directions) in diverse clinical settings, such as diabetes, hypertension, LV hypertrophy, coronary artery disease, and heart failure.18-25 Therefore, analyzing only a unidirectional function (i.e., longitudinal, radial, or circumferential) would merely be a partial approach to assess true global LV contractile function. Accordingly, using multidirectional systolic parameters such as the longitudinal-circumferential systolic index or the global systolic index, we could determine true global LV systolic function or myocardial systolic performance of the left ventricle. In this respect, in the present study, we demonstrate that the longitudinal-circumferential systolic index and the global systolic index detected global LV systolic dysfunction in patients with hypertension despite normal LV global longitudinal systolic strain. Furthermore, we found that the functional class (dyspnea [New York Heart Association classification]) in patients with hypertension
Healthy subjects (n = 323) Age Systolic blood pressure Mitral E/e0 septal-lateral ratio LVEF LV mass Body surface area Hypertensive Patients (n= 310) Age Systolic blood pressure Mitral E/e0 septal-lateral ratio LVEF LV mass Body surface area
r
0.09 0.03 0.11
P
.0773 .4973 .0502
Global systolic index r
0.08 0.09 0.03
P
.1445 .0852 .5947
0.18* .0013 0.19* 0.06 .2873 0.05 0.25 < .0001 0.07
.0006 .3436 .1878
0.01 0.09 0.30
.6766 .1243 .0519
.9985 0.02 .1932 0.11 < .0001 0.12
0.39* .0013 0.23* < .0001 0.31* < .0001 0.18* .0231 0.13 .0926 0.27 .0003
*Variable with the strongest associations with the longitudinalcircumferential index and the global systolic index in a forward stepwise multivariate regression analysis.
was inversely related to these multidirectional indices. Hence, we believe that the multidirectional systolic indices addressed in the present study meet the conditions to be considered valuable measurements of the global systolic function of the left ventricle. Determination of Normal LV Global Systolic Function Although recent studies analyzing healthy subjects had identified the normal range of the unidirectional contractile function of the left ventricle (i.e., longitudinal, circumferential, or radial),26,34-38 the normal range of multidirectional global LV systolic function or myocardial systolic performance (i.e., the result or the average of longitudinal, circumferential, and radial function) had not yet been determined. For these reasons, the objective of the present multicenter study was to analyze a large cohort of healthy subjects to determine the normal range of multidirectional global LV contractile function. In this regard, in 323 healthy subjects, we determined the range of values of multidirectional global LV systolic function, which was analyzed using multidirectional systolic indices with adequate feasibility and reproducibility. In addition, we showed that using the lowest percentile of healthy subjects of these multidirectional parameters, it was possible to detect global LV systolic dysfunction in patients with hypertension even when conventional and unidirectional LV measurements, such as LVEF and LV global longitudinal strain, were normal. Limitations Some considerations should be taken into account when analyzing global LV systolic function by means of multidirectional systolic parameters using 2D STE. Recent studies have shown that 2D speckle-tracking echocardiographic values of the left ventricle vary among different
Morris et al 499
Journal of the American Society of Echocardiography Volume 27 Number 5
Figure 2 Global LV systolic dysfunction evaluated by the longitudinal-circumferential systolic index and the global systolic index in patients with hypertension, despite normal LVEF and normal LV global longitudinal systolic strain. To define normal LV global longitudinal systolic strain, low longitudinal-circumferential systolic index, and low global systolic index, the lowest expected values of these measurements in the healthy population were used as cutoffs (namely, 17.25%, 16.86%, and 19.36%, respectively; see Table 2). Normal LVEF was defined as $55% by the biplane Simpson’s method.
Table 5 Worsening of symptomatic status linked to the deterioration of multidirectional LV systolic function in patients with hypertension NYHA functional class Index
I (n = 212)
II (n = 67)
III and IV (n = 31)
P (ANOVA)
Global systolic index Longitudinal-circumferential systolic index
25.2 6 6.1% 18.9 6 2.5%
20.7 6 5.7%* 15.2 6 2.5%*
19.9 6 5.3% 15.1 6 2.7%†
