Background
In aortic stenosis (AS), the combination of risk factors can progressively lead to an increased arterial rigidity, which can be evaluated by the carotid artery and aortic stiffness (β index). The aim of this study was to investigate the relationship between carotid and aortic β index, left ventricular (LV) function, plasma brain natriuretic peptide (BNP) level, and symptoms in patients with AS.
Methods
Comprehensive echocardiography including Doppler tissue imaging of the mitral annulus was performed in 53 patients with AS (aortic valve area < 1.2 cm 2 ) and preserved LV ejection fractions (≥50%). Carotid β index was automatically derived from ultrasound wall tracking of the right carotid artery. The mitral E/e′ ratio was used to estimate LV filling pressures.
Results
Carotid β index was higher in women than in men and was significantly correlated with age ( P < .0001), diastolic arterial pressure ( P = .046), pulse pressure ( P = .006), and systemic arterial compliance ( P = .001). Interestingly, carotid β index was significantly correlated with E/e′ ratio ( P < .0001) and plasma BNP level ( P = .011). In multivariate regression analysis, carotid β index was an independent predictor of E/e′ ratio ( P < .0001) and of BNP level ( P = .02). Moreover, carotid β index was significantly higher in symptomatic patients ( P = .009). Aortic β index was significantly correlated with carotid β index ( P < .0001), E/e′ ratio ( P = .004), and BNP ( P < .001) and was significantly higher in symptomatic patients ( P = .037).
Conclusions
In patients with moderate to severe AS and preserved LV ejection fractions, the presence of increased carotid artery and aortic stiffness, assessed using carotid and aortic β index, is independently associated with elevated LV filling pressures, BNP level, and symptoms.
Calcific aortic stenosis (AS) is the most common valvular disease in Western countries, and its prevalence increases with population ageing. AS is characterized by an active degenerative process that shares similarities with atherosclerosis. Currently, AS is no longer considered an isolated aortic valve disease but rather a complex disease in which the central actors are the left ventricle (ability to adapt to the increased afterload), the valve (severity of valvular obstruction), and the vascular system (reduced arterial compliance). As a matter of fact, patients with the same degrees of valvular stenosis can have differing prognoses depending on different degrees of left ventricular (LV) functional compromise or alterations in vascular afterload.
In patients with AS, the combination of risk factors (aging process, atherosclerosis, hypertension, etc) can progressively lead to an increased arterial rigidity, which can be evaluated by the assessment of local arterial stiffness at specific sites. Using two-dimensional imaging to measure aortic diameters in patients with AS, it appears that the increased aortic rigidity is independently correlated with LV systolic and diastolic function and brain natriuretic peptide (BNP) levels. However, measurements of arterial stiffness at different sites of the vascular tree do not seem to be interchangeable, even between the aorta and carotid artery. They are both elastic arteries, but the impacts of different cardiovascular risk factors on their wall properties are not uniform.
So far, in patients with AS, the relationship between carotid arterial stiffness and LV function, BNP, and symptoms has not yet been evaluated. Because of the low sampling rate of B-mode images, the accuracy of two-dimensional imaging for the assessment of carotid stiffness remains limited. Conversely, ultrasound wall tracking allows more accurate evaluation of vascular diameters, producing precise waveforms of changes during the cardiac cycle. This recent technology can be easily applied to the carotid artery, which is known to be a frequent site affected by the atherosclerotic process.
The aim of this study was to evaluate the impact of carotid artery and aortic stiffness on LV function, BNP release, and clinical status in a series of patients with moderate to severe AS and preserved LV ejection fraction.
Methods
Patient Population
The present study included a total of 53 patients (mean age, 75 ± 10 years; 27 men [51%]) who underwent comprehensive echocardiography in our Heart Valve Clinic from March 2010 to December 2011 and who fulfilled the following inclusion criteria: moderate to severe AS, defined as an aortic valve area ≤ 1.2 cm 2 ; preserved LV ejection fraction (≥50%); dimension of the ascending aorta < 40 mm or ≤ 21 mm/m 2 ; no significant atherosclerosis of the right carotid artery; and sinus rhythm. Patients with more than mild concomitant mitral valve dysfunction were excluded, as were patients with concomitant aortic insufficiency more than mild in degree. Twenty-six patients had already been included in our previous study of aortic stiffness evaluation. The following clinical data were collected: age, gender, hypercholesterolemia (total cholesterol > 190 mg/dL use of lipid-lowering therapy), current smoking, diabetes mellitus, systemic arterial hypertension (blood pressure ≥ 140/90 mm Hg or use of antihypertensive treatment), and previous evidence of coronary artery disease (presence of ≥50% coronary artery stenosis on angiography, previous revascularization, or previous myocardial infarction). Information regarding current medications was also obtained. The relevant institutional review board approved the protocol, and all patients gave written informed consent.
Measurement of Carotid Artery and Aortic Stiffness
Subjects were studied after resting supine for >10 min. Systolic and diastolic blood pressures were measured in the right arm with using an arm-cuff sphygmomanometer at the time of examination. The common right carotid artery was scanned using a Hitachi-Aloka machine (Prosound α7 version 1.1; Aloka, Tokyo, Japan) and a linear-array probe. The change in diameter of the vessel was measured as the difference between the displacement waveforms of the anterior and posterior walls, using the e-tracking technique, with the cursors set manually to track the media-adventitia boundaries in the arterial wall approximately 1 cm proximal to the carotid sinus. At least 10 sec of consecutive cardiac cycles were recorded for every patient. During offline analysis, carotid artery stiffness (β index) was automatically derived from the average of five cardiac cycles manually selected by the physician and according to the established formula : β index = ln(Ps/Pd)/[(Ds − Dd)/Dd], where Ps and Pd are systolic and diastolic blood pressure, and Ds and Dd are the maximal and minimal diameters of the right common carotid artery ( Figure 1 ). Two sets of measurements were performed and averaged for each patient.
The same method was used to record aortic stiffness, 1 cm above the sinotubular junction by two-dimensionally guided M-mode transthoracic echocardiography in the parasternal long-axis view, as previously described by our group.
Echocardiographic Measurement
After the assessment of carotid artery stiffness, all patients underwent comprehensive Doppler echocardiographic examinations. Standard echocardiographic views were obtained using second-harmonic imaging. M-mode, two-dimensional, color Doppler, pulse-wave, and continuous-wave Doppler data were recorded for each patient and were stored in digital format on a dedicated workstation for offline analysis. For each measurement, at least two cardiac cycles were averaged. LV end-diastolic and end-systolic volumes and ejection fraction were measured using the biapical Simpson’s disk method. Continuous-wave Doppler was used to measure the aortic transvalvular maximal velocities; peak and mean gradients were calculated using the simplified Bernoulli equation. Aortic valve area was calculated using the continuity equation (velocity-time integral method). Stroke volume was calculated using the Doppler method as follows: 0.785 × (LV outflow tract diameter) 2 × LV outflow tract velocity-time integral. Peak E-wave and A-wave velocities of mitral inflow were measured using pulsed-wave Doppler. Pulsed-wave tissue Doppler was used to measure systolic (s′) and early diastolic (e′) medial mitral annular velocities. The E/e′ ratio was then calculated as an estimate of LV filling pressures.
Global LV Afterload
To estimate the global LV afterload, valvuloarterial impedance was calculated as the sum of systolic arterial pressure and mean transaortic pressure gradient divided by the stroke volume index. The ratio between stroke volume index and brachial pulse pressure was used as an indirect measure of total systemic arterial compliance and normal values were defined when >0.6 mL/m 2 /mm Hg. Systemic vascular resistance was estimated as the ratio between (mean arterial pressure × 80) and cardiac output.
Plasma BNP
Venous blood samples were obtained before echocardiographic examinations, after resting supine for >10 min. Chilled ethylenediaminetetraacetic acid tubes were centrifuged immediately at 4,000 rpm (4°C) for 15 min. Separated plasma samples were processed by immunofluorescence assay (Biosite; Beckman Coulter, San Diego, CA). Interassay and intra-assay variation was 5% and 4%, respectively. The assay detection limit was 1 pg/mL.
Symptomatic Status and Risk Score
Symptomatic status was obtained for each patient with a careful evaluation of patient’s history and hospital medical records. Dyspnea was graded according to the New York Heart Association (NYHA) functional class. Patients were classified as symptomatic in the presence of NYHA class ≥ II, angina, and/or history of syncope. To differentiate and exclude symptoms potentially related to coronary artery disease, all patients underwent stress echocardiography at least once in the previous 6 months. None of the patients included in the study had wall motion abnormalities at rest or during exercise. A risk score was calculated for all patients according to the following formula: [peak transvalvular velocity (m/sec) × 2] + [ln(BNP) × 1.5] + 1.5 if female, as previously described by Monin et al .
Statistical Analysis
Data are expressed as mean ± SD or as percentages unless otherwise specified. Data on β index and BNP were skewed and were thus logarithmically transformed. Log BNP and log β index values were used in correlation and regression analyses as appropriate. Relationships between different parameters were assessed by linear correlation analysis and Pearson’s correlation coefficient. To determine the impact of carotid artery stiffness on LV diastolic function, LV filling pressure, BNP plasma level, and symptoms, stepwise linear or logistic multiple regression analyses were performed. Variables with P values < .10 on univariate analysis were incorporated into the multiple regression models, with special care to avoid collinearity among a subset of several variables measuring the same phenomenon. Two-sided P values < .05 were considered significant. Continuous and nominal variables were compared using Student’s t test. Carotid β index was compared between symptomatic and asymptomatic patients using a one-way analysis of variance followed by Tukey’s test. All statistical analyses were performed using Statistica version 6 (StatSoft Inc, Tulsa, OK).
Reproducibility Analysis
Interobserver and intraobserver variability for measurement of carotid β index was determined from the analysis of 12 randomly selected patients by two independent readers blinded to previous measurements. During offline analysis, each reader was able to select the preferred five cardiac cycles, and carotid β index was then automatically derived. Two sets of measurements were performed for each patient and averaged. Absolute difference between repeated measurements was calculated and expressed as the percentage of their mean value for both interobserver and intraobserver results. Moreover, data were compared using intraclass correlation coefficients.
Results
Patient Characteristics
Table 1 reports demographic and clinical variables, while Table 2 depicts echocardiographic characteristics of the study population. The origin of AS was calcific in 48 patients (91%) and bicuspid in five (9%). Seventeen patients (32%) were symptomatic (dyspnea in 16, angina in three, syncope in three, combined symptoms in five). Carotid β index distribution is reported in Figure 2 .
