Author + information
- Received July 22, 2016
- Revision received October 12, 2016
- Accepted October 20, 2016
- Published online November 6, 2017.
- Nicholas Cauwenberghs, MSca,
- Judita Knez, MDa,b,
- Jan D’hooge, PhDc,
- Lutgarde Thijs, MSca,
- Wen-Yi Yang, MDa,
- Fang-Fei Wei, MDa,
- Zhen-Yu Zhang, MDa,
- Jan A. Staessen, MD, PhDa and
- Tatiana Kuznetsova, MD, PhDa,∗ ()
- aResearch Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium
- bHypertension Division, Department of Internal Medicine, University Clinical Centre Ljubljana, Ljubljana, Slovenia
- cDivision of Cardiovascular Imaging and Dynamics, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium
- ↵∗Address for correspondence:
Dr. Tatiana Kuznetsova, Research Unit of Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Campus Sint Rafaël, Kapucijnenvoer 35, Block D, box 7001, B-3000 Leuven, Belgium.
Objectives We assessed to what extent arterial properties measured at baseline and follow-up predict longitudinal alterations in echocardiographic indexes reflecting left ventricular (LV) structure and function.
Background Serial imaging studies are needed to clarify the relation of changes in LV structure and function to arterial stiffness.
Methods In 607 participants (50.7% women; mean age 50.7 years), using echocardiography and Doppler imaging, we measured LV dimensions, transmitral blood flow, and mitral annular tissue velocities at baseline and after 4.7 years. Using applanation tonometry, we assessed central pulse pressure (cPP) and carotid-femoral pulse wave velocity (PWV) at baseline. We regressed longitudinal changes in LV indexes on the arterial stiffness parameters and reported standardized effect sizes as a fraction of SD of LV change.
Results After full adjustment, longitudinal increase in LV septal (standardized effect size: +14.4%; p = 0.0018) and posterior wall (+12.6%; p = 0.0027) thickness was associated with higher baseline PWV, whereas LV internal diameter (-12.4%; p = 0.012) decreased during follow-up with PWV. Consequently, greater increase in relative wall thickness was associated with higher baseline PWV (+17.2%; p <0.0001). Participants with higher baseline PWV had a greater risk to develop or retain LV concentric remodeling during follow-up (odds ratio 1.35; p = 0.028). In addition, in women, baseline cPP predicted a greater increase in LV mass (+22.8%; p = 0.0009) and E/e′ ratio (+36.1%; p <0.0001).
Conclusions Progression to LV concentric remodeling pattern was associated with higher baseline PWV. In women, cPP predicted worsening of LV diastolic function. Our study highlights the importance of arterial properties as mediator of LV concentric remodeling in men and women, and diastolic dysfunction in women.
The heart and vascular properties determine the capacity of the body to adapt cardiac output, regulate blood pressure (BP), and respond to changes in pre- and afterload. When cardiac and vascular properties are matched, maximal cardiac work, power, and chamber efficiency are achieved to maintain BP and cardiac output within a physiological range.
Arterial stiffening is a common feature of aging and is exacerbated by cardiovascular risk factors such as hypertension and diabetes mellitus (1). In a stiff aorta, changes in aortic systolic pressure during left ventricular (LV) ejection are greater due to increased forward wave propagation and impaired aortic reservoir function (2–4). Because the heart needs to generate greater force to extend the stiffer arteries, its energy expenditure increases. To confront the increased systolic afterload, the heart might adapt by adverse LV remodeling (5–7). Indeed, previous cross-sectional studies in patients and in the general population showed an association of LV remodeling and/or dysfunction with arterial properties (8–15). However, the cross-sectional design of these studies did not allow inferring causality. On the other hand, data on the longitudinal changes of LV structure and function in relation to arterial characteristics are sparse. Serial imaging studies are keys to elucidate the impact of arterial properties on early LV remodeling and dysfunction that forerun overt heart failure (HF). In the framework of a longitudinal population study, we therefore investigated to what extent arterial properties as assessed by applanation tonometry predict alterations in echocardiographic indexes of LV structure and diastolic function.
The Ethics Committee of the University of Leuven approved the FLEMENGHO (Flemish Study on Environment, Genes, and Health Outcomes) study. From August 1985 until December 2005, we randomly recruited a family-based population sample from a geographically defined area in northern Belgium as described elsewhere (9). From 2005 to 2009 we invited 1,031 former participants for a re-examination at our field center, including echocardiography and arterial tonometry (Figure 1). We obtained written informed consent from 828 subjects (participation rate, 80.3%). To study changes in LV structure and function, we invited these people for a follow-up examination, on average 5 years after their first cardiovascular examination. We excluded 147 participants because they died (n = 25), were lost to follow-up (n = 19), or declined the follow-up invitation (n = 103). For this analysis, we additionally excluded 74 subjects because of atrial fibrillation at baseline (n = 8) or at follow-up (n = 4), an artificial pacemaker (n = 3), or insufficient quality of the echocardiographic recordings (n = 19) or tonometry (n = 40). Thus, current analysis included 607 participants (Figure 1).
To ensure steady state, echocardiographic and arterial measurements were obtained consecutively and after the subjects had rested for at least 15 min in the supine position.
A detailed echocardiographic protocol is provided in the Online Appendix. Briefly, an experienced physician (T.K.) performed both ultrasound examinations in accordance to the recent recommendations (16) and as described previously (17) using a Vivid7 Pro and Vivid E9 (GE Vingmed, Horten, Norway) interfaced with a 2.5- to 3.5-MHz phased-array probe. With the subjects in partial left decubitus, the observer obtained images along the parasternal long and short axes and from the apical 4- and 2-chamber long-axis views, together with a simultaneous electrocardiogram (ECG) signal. All recordings included at least 5 cardiac cycles and were digitally stored for off-line analysis.
One observer (T.K.) analyzed the echocardiograms blinded to the participants’ characteristics. Digitally stored images were post-processed using EchoPac software (GE Vingmed). Measurements were averaged over at least 3 heart cycles for statistical analysis. From the long-axis parasternal view, LV internal diameter and interventricular septal and posterior wall thickness were measured at end-diastole from the 2-dimensionally guided M-mode tracings. End-diastolic LV dimensions were also used to derive LV mass using an anatomically validated formula. Relative wall thickness (RWT) was calculated as (interventricular septum + posterior wall)/LV internal diameter at end-diastole. LV concentric remodeling was defined as RWT >0.42.
Transmitral Doppler flow signals were used to measure peak early (E) and late (A) diastolic velocities and E/A ratio. From pulsed-wave tissue Doppler imaging (TDI) recordings, we measured the early (e′) and late (a′) diastolic peak velocities of the mitral annulus displacement at 4 acquisition sites. We calculated the E/e′ ratio by dividing transmitral E peak by e′ averaged from the 4 acquisition sites.
A detailed protocol of the arterial phenotyping is provided in the Online Appendix. We collected carotid, femoral, and radial arterial waveforms using a high-fidelity SPC-301 micromanometer (Millar Instruments, Houston, Texas) interfaced with a computer running SphygmoCor version 7.1 (AtCor Medical, New South Wales, Australia). The pulse waves were calibrated by the supine brachial BP measured immediately before the tonometric recordings. From the radial signal, SphygmoCor software calculated the aortic pulse wave by means of a validated transfer function using brachial diastolic BP and mean arterial pressure (MAP). MAP was defined as: diastolic BP + 0.40 x (systolic BP − diastolic BP). The software returned the central systolic BP and the pressure at the first (P1) and second (P2) peak or shoulder of the central waveform. Central pulse pressure (cPP) was calculated as the difference between central systolic and diastolic pressure. Augmentation pressure (AP) was the difference between P2 and P1. Aortic pulse wave velocity (PWV) was the ratio of the carotid-sternal-femoral distance (in meters) to the transit time of the pressure wave (in seconds).
A complete description of other measurements is provided in the Online Appendix.
For database management and statistical analysis, we used SAS software version 9.4 (SAS Institute, Cary, North Carolina). We compared changes in means and proportions by means of a paired Student t test and McNemar test, respectively. Statistical significance was a 2-sided significance level <0.05. We calculated longitudinal changes in echocardiographic indexes by subtracting the baseline from the follow-up measurement. We performed stepwise regression with the forward selection to determine clinical correlates of the 4.7 years change in LV structure and diastolic function parameters. The baseline characteristics considered as covariables in stepwise regression were baseline LV parameter, age, sex, body mass index (BMI), smoking, heart rate, brachial systolic and diastolic BP, MAP, and antihypertensive drug intake. Stepwise models also included changes in these covariables. We set the p values for variables to enter and to stay in the stepwise regression models at 0.05.
By use of a mixed model, we assessed multivariable-adjusted associations between longitudinal changes in LV structural and diastolic indexes and baseline arterial properties or their changes. All statistical models were adjusted for follow-up duration, baseline LV parameter, sex, age, BMI, heart rate, and MAP. We additionally adjusted for changes in these covariables if they were selected in the stepwise analyses. We reported effect sizes on an absolute scale as the multivariable-adjusted regression coefficients per 1-SD increase in the arterial indexes at baseline, and on a relative scale as a percentage of the standardized effect size (i.e., the absolute effect size divided by the SD of the echocardiographic changes multiplied by 100). Using multiple logistic regressions, we explored whether baseline PWV was associated with development of LV concentric remodeling.
Characteristics of participants
From the 607 participants (50.7% women), 247 (40.7%) were hypertensive at baseline of whom 143 (23.6%) were on antihypertensive drug treatment. The mean age at baseline was 50.7 (SD, 14.3) years. The median follow-up was 4.7 years (5th to 95th percentile, range 3.7 to 5.4 years). Tables 1 and 2⇓⇓ show the clinical, arterial, and echocardiographic characteristics of the study participants by examination phase. AP, cPP, and PWV increased significantly over time (p ≤ 0.025 for all) (Table 2, Figure 2).
Correlates of change in LV structure and diastolic function indexes
Correlates of change in aortic and LV structure indexes
During follow-up, aortic root diameter, LV outflow tract (LVOT) diameter and LV septal and posterior wall thickness significantly increased (p < 0.0001 for all), whereas LV internal diameter slightly decreased (p = 0.045) (Table 2, Figure 2). Subsequently, RWT and LV mass index increased by 6.1% and 5.1%, respectively (p < 0.0001 for both).
Online Table 1 lists the correlates of the 4.7 years change in indexes of aortic and LV structure. Major correlates included baseline LV structure parameter, sex, baseline age, BMI, and hemodynamic factors such as heart rate and MAP as well as changes over time in these variables. Start or maintenance of antihypertensive treatment during follow-up did not determine changes in LV structure indexes.
Correlates of change in LV diastolic function indexes
During follow-up, transmitral and TDI early and late diastolic velocities as well as E/A ratio significantly decreased, whereas E/e′ ratio increased (p < 0.0001 for all) (Table 2, Figure 2). Baseline age, BMI, and hemodynamic factors such as heart rate and MAP as well as changes over time in these variables were major determinants of longitudinal change in LV Doppler diastolic indexes (Online Table 2).
Online Figures 1 and 2 show the histograms of change in LV indexes during follow-up.
Associations between change in LV structure and diastolic function indexes and baseline arterial properties
Multivariable-adjusted estimates (95% confidence interval [CI]) associated with a 1-SD increase in baseline AP (+9 mm Hg), cPP (+15 mm Hg), and PWV (+1.8 m/s) are presented for the changes in LV structure indexes in Table 3 and in diastolic Doppler indexes in Table 4.
After full adjustment, longitudinal increases in LVOT diameter (standardized effect size: +15.3%; p = 0.0037), LV septal (+14.4%; p = 0.0018), and posterior wall thickness (+12.6%; p = 0.0027) were significantly associated with higher baseline PWV, whereas LV internal diameter (−12.4%; p = 0.012) decreased during follow-up with baseline PWV (Table 3). Hence we observed a greater increase in RWT with higher baseline PWV (+17.2%; p < 0.0001) (Table 3). Figure 3 shows the multivariable-adjusted changes in RWT by deciles of baseline PWV. Moreover, LV posterior wall thickness (+8.8%; p = 0.044) and LV mass index (+13.7%; p = 0.0092) increased during follow-up with baseline cPP (Table 3).
In multivariable-adjusted models, a higher cPP at baseline was associated with less decrease in E/A ratio (+14.7%; p = 0.0015), greater decrease in TDI a′ (−11.5%; p = 0.014), and greater increase in E/e′ ratio (+20.1%; p = 0.0002) (Table 4). Higher baseline AP was also associated with greater increase in E/e′ ratio (+17.2%; p = 0.011).
Additional adjustment for antihypertensive drug intake did not substantially alter the multivariable-adjusted associations between longitudinal changes in LV indexes and baseline arterial properties (Online Table 3).
Sex-specific associations between change in LV structure and diastolic function indexes and baseline arterial properties
Sex-specific multivariable-adjusted standardized effect sizes (95% CI) associated with a 1-SD increase in baseline AP, cPP, and PWV are presented in Online Table 4.
In both men and women, we observed similar associations of longitudinal increases in LV wall thickness and RWT with baseline PWV (p value for sex interaction ≥0.38) (Figures 4A to 4C).
In women, a higher cPP at baseline was independently associated with greater longitudinal increase in LV mass index (+22.8%, p = 0.0009) and E/e′ ratio (+36.1%, p < 0.0001) and a smaller decrease in E peak (+16.5%, p = 0.018) and E/A ratio (+18.2%, p = 0.018) (Figures 4D to 4F). Similarly, in women, higher baseline AP was associated with greater longitudinal increase in E/e′ ratio (+30.7%, p = 0.0053) (Online Table 4). On the other hand, in men cPP at baseline did not correlate with the change over time in any of the echocardiographic LV indexes (p ≥ 0.14 for all) (Online Table 4).
Progression to LV concentric remodeling in relation to baseline PWV
At the follow-up examination, the prevalence of LV concentric remodeling significantly increased from 19.3% (n = 117) to 24.9% (n = 151; p = 0.0013). In unadjusted analyses, baseline PWV was higher in participants who developed LV concentric remodeling (mean PWV, 8.25 m/s; n = 73) or remained in the LV concentric remodeling group (8.55 m/s; n = 78) during follow-up as compared to participants who did not develop (7.15 m/s; n = 417) or even reversed LV concentric remodeling over time (7.36 m/s; n = 39) (Online Table 5). Overall, participants with higher PWV at baseline had a greater risk to develop or retain LV concentric remodeling during the follow-up period (odds ratio [OR]: 1.35 [1.03 to 1.77]; p = 0.028).
Associations between changes in LV indexes and arterial properties
After full adjustment, longitudinal changes in LV structure indexes was not significantly associated with ΔPWV (Online Table 6), whereas a greater longitudinal increase in cPP was independently related to less decrease in E and e′ peaks and E/A ratio (p ≤ 0.0032) (Online Table 7).
In this longitudinal study, we investigated whether arterial properties can predict changes in echocardiographic indexes of LV structure and diastolic function. The key findings of this study can be summarized as follows: 1) in both men and women, longitudinal changes in LV structural indexes reflecting adverse LV remodeling were associated with higher baseline PWV; 2) in categorical analyses, a higher baseline PWV was associated with a significantly higher risk to develop or retain LV concentric remodeling; and 3) in women, higher cPP at baseline predicted the longitudinal increase in LV mass and E/e′ ratio (i.e., worsening of LV diastolic function).
The structure and function of the heart alter with aging and in response to cardiovascular risk factors such as hypertension and diabetes mellitus. The LV tends to remodel concentrically over the adult life course. In line with our findings, Cheng et al. (18) reported that RWT increased in 4062 Framingham participants during a 16-year period due to increasing LV wall thickness and decreasing LV internal dimensions. Similarly, Eng et al. (19) showed an increase in LV mass-to-volume ratio after 9.4 years of follow-up in 2935 participants from the MESA (Multi-Ethnic Study of Atherosclerosis) study. Worsening of LV geometry is associated with increased risk of cardiovascular outcome (20–22). Recently, Framingham investigators reported that exposure to multiple cardiovascular risk factors, such as elevated brachial systolic BP and greater BMI, were associated with the development of abnormal LV geometry (20). However, in previous publications the authors did not investigate the impact of arterial stiffness on the natural history of LV geometry.
Arterial stiffness plays an important role in the development of HF. At systole, the heart generates a forward pressure wave that is reflected at various sites in the arterial system. On the other hand, as suggested recently, entrapment of reflected waves in the periphery might limit the influence of peripheral reflected waves on central BP (23). In fact, the augmentation of central BP following arterial stiffening may largely originate from increased forward wave propagation and decreased proximal aortic reservoir function (2–4). In the long-term, the chronically increased cardiac loading might lead to LV remodeling, increases LV oxygen requirements, and eventually causes HF. Indeed, prospective studies showed that a higher carotid-femoral PWV, reflecting increased aortic stiffness, is associated with a higher risk of HF beyond traditional cardiovascular risk factors (24–26).
Serial imaging studies are keys to clarify the role of arterial properties in progression of adverse LV remodeling. So far, previous studies described the relation between LV structure and arterial stiffness in a cross-sectional manner. For instance, in 100 subjects free of overt cardiac disease, Redheuil et al. (8) reported the association between increased LV mass-to-volume ratio and structural alterations in the aortic arch reflecting higher proximal aortic stiffness and PWV assessed by magnetic resonance imaging. Similarly, a recent cross-sectional magnetic resonance imaging study in 2,093 MESA participants showed that higher aortic arch PWV was associated with LV concentric remodeling (15). In our longitudinal study, we observed that the increase in RWT during 4.7 years of follow-up was associated with higher PWV at baseline and this association was independent of important covariables. In contrast to a cross-sectional study reporting an independent relation between RWT and PWV in postmenopausal women only (27), we observed a similar longitudinal increase in LV wall thickness with higher baseline PWV in both men and women. We also showed in the categorical analyses that a greater arterial stiffness at baseline was associated with a higher risk to develop or retain LV concentric remodeling during follow-up.
In parallel to changes in cardiac geometry, LV diastolic function tends to worsen over the adult life course (17). Conventional echocardiography and TDI allow the detection of subclinical deterioration of LV diastolic function (28). Impaired myocardial relaxation is characterized by decreased transmitral early and enhanced late diastolic velocities. On the other hand, combining early transmitral flow velocity with mitral annular velocity (E/e′ ratio) allows evaluation of elevated LV filling pressure. We previously showed in a general population sample that higher brachial systolic BP at baseline predicted worsening in LV diastolic function as reflected by increased E/e′ (17). Similar association was also reported by Ghosh et al. (29). However, these longitudinal reports did not explore the impact of central hemodynamic on LV diastolic dysfunction.
Two cross-sectional community-based studies reported that moderate LV diastolic dysfunction was associated with a decreased aortic compliance (9,10). When stratifying by sex, other cross-sectional studies found this association in elderly women (30,31). The sex-dependent relation between LV diastolic function and cPP might be explained by the higher aortic stiffness and pulsatile hemodynamic load in women compared to men (9,30).
In our longitudinal analysis, we found that the increase in LV mass and E/e′ ratio during follow-up was independently related to baseline cPP in women but not in men. Therefore, our longitudinal observation supports the hypothesis that LV diastolic function in women is more vulnerable to the detrimental effects of increased central pulsatility. This finding might explain why women are more likely to develop HF with preserved ejection fraction than men.
This study must be interpreted within the context of its potential limitations and strengths. First, echocardiographic measurements are prone to measurement errors due to signal noise, acoustic artefacts, and angle dependency. However, a single experienced observer recorded and centrally post-processed all echocardiographic images using a highly standardized imaging protocol. Second, our findings could only be interpreted in the light of previous cross-sectional observations as longitudinal observations addressing the issue of association of LV structure and function with arterial properties are currently lacking. Finally, our study population only included white European participants, limiting the generalizability of our findings to other ethnicities.
Our study described in a general population the association of longitudinal changes in LV structure and diastolic function with various indexes of arterial stiffness as assessed by applanation tonometry. We found that having or developing LV concentric remodeling was associated with higher baseline PWV. In women, higher cPP predicted worsening of LV diastolic function. Our study underscored the importance of arterial properties as a mediator of LV concentric remodeling in men and women, and diastolic dysfunction in women. Strategies to reduce arterial stiffness and improve central hemodynamics might have additional value to prevent or delay subclinical alterations in LV structure and function.
COMPETENCY IN MEDICAL KNOWLEDGE: Serial imaging studies are keys to elucidate the impact of arterial properties on early LV remodeling and dysfunction that forerun overt HF. Our study described the association of longitudinal changes in LV structure and diastolic function with various indexes of arterial stiffness as assessed by applanation tonometry.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: Our study underscored the importance of arterial properties as a mediator of LV concentric remodeling in men and women, and diastolic dysfunction in women.
TRANSLATIONAL OUTLOOK: Strategies to reduce arterial stiffness and improve central hemodynamics might have additional value to prevent or delay subclinical alterations in left ventricular structure and function.
The European Union (grants HEALTH-2011-278249-EUMASCARA, and ERC Advanced Grant-2011-294713-EPLORE) supported the Studies Coordinating Centre (SCC, Leuven, Belgium). The SCC also received grants from the Fonds voor Wetenschappelijk Onderzoek Vlaanderen, Brussels, Belgium (grants G.0880.13, G. 0881.13, and 11Z0916N). Dr. D’hooge collaborates with GE and Phillips; and has received consultant fees from SuperSonic Imagine. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- augmentation pressure
- blood pressure
- central pulse pressure
- ejection fraction
- heart failure
- left ventricular
- mean arterial pressure
- pulse wave velocity
- relative wall thickness
- tissue Doppler imaging
- Received July 22, 2016.
- Revision received October 12, 2016.
- Accepted October 20, 2016.
- 2017 American College of Cardiology Foundation
- Wang J.J.,
- O'Brien A.B.,
- Shrive N.G.,
- Parker K.H.,
- Tyberg J.V.
- Davies J.E.,
- Baksi J.,
- Francis D.P.,
- et al.
- Schultz M.G.,
- Davies J.E.,
- Hardikar A.,
- et al.
- Chen C.H.,
- Nakayama M.,
- Nevo E.,
- Fetics B.J.,
- Lowell Maughan W.,
- Kass D.A.
- Kawaguchi M.,
- Hay I.,
- Fetics B.,
- Kass D.A.
- Redheuil A.,
- Yu W.C.,
- Mousseaux E.,
- et al.
- Eren M.,
- Gorgulu S.,
- Uslu N.,
- Celik S.,
- Dagdeviren B.,
- Tezel T.
- Ohyama Y.,
- Ambale-Venkatesh B.,
- Noda C.,
- et al.
- Kuznetsova T.,
- Thijs L.,
- Knez J.,
- et al.
- Cheng S.,
- Xanthakis V.,
- Sullivan L.M.,
- et al.
- Eng J.,
- McClelland R.,
- Gomes A.,
- et al.
- Lieb W.,
- Gona P.,
- Larson M.G.,
- et al.
- Cohn J.N.,
- Ferrari R.,
- Sharpe N.
- Vlachopoulos C.,
- Aznaouridis K.,
- Stefanadis C.
- Ben-Shlomo Y.,
- Spears M.,
- Boustred C.,
- et al.
- Tsao C.W.,
- Lyass A.,
- Larson M.G.,
- et al.
- Ghosh A.K.,
- Hughes A.D.,
- Francis D.,
- et al.
- Coutinho T.,
- Borlaug B.A.,
- Pellikka P.A.,
- Turner S.T.,
- Kullo I.J.
- Shim C.Y.,
- Park S.,
- Choi D.,
- et al.