Left Ventricular Function Assessed by One-Point Carotid Wave Intensity in Newly Diagnosed Untreated Hypertensive Patients

Authors


Abstract

Objectives

To investigate whether newly diagnosed untreated hypertensive patients show higher left ventricular (LV) contractility, as assessed by traditional echocardiographic indices and carotid wave intensity (WI) parameters, including amplitude of the peak during early (W1) and late systole (W2).

Methods

A total of 145 untreated hypertensive patients were compared with 145 age- and sex-matched normotensive subjects. They underwent comprehensive echocardiography and WI analysis. WI analysis was performed at the level of the common carotid artery. The diameter changes were the difference between the displacement of the anterior and posterior walls, with the cursors set to track the media-adventitia boundaries 2 cm proximal to the carotid bulb and calibrated by systolic and diastolic BP. Peak acceleration was derived from blood flow velocity measured by Doppler sonography with the range-gate positioned at the center of the vessel diameter. WI was based on the calculation of (dP/dt)×(dU/dt), where dP/dt and dU/dt were the derivatives of BP (P) and velocity (U) with respect to time. One-point pulse wave velocity (PWVβ) and the interval between the R wave on ECG and the first peak of WI (R-W1), using a high definition echo-tracking system implemented in the ultrasound machine (Aloka), were also derived.

Results

After adjustment for body weight, heart rate, and physical activity, the two groups had similar general characteristics and diastolic function. However, hypertensives showed significantly higher LV mass, LV ejection fraction (LVEF), circumferential and LV end-systolic stress, and one-point PWV as well as W1 (13.646 ± 7.368 vs 9.308 ± 4.675 mmHg m/s3, P =.001) and W2 (4.289 ± 2.017 vs 2.995 ± 1.868 mmHg m/s3, P =.001). Hypertensives were divided into tertiles according to LVEF: W1 (11.934 ± 5.836 vs 11.576 ± 5.857 vs 17.227 ± 8.889 mmHg m/s3, P <.0001) was higher in the highest LVEF tertile along with relative wall thickness, midwall fractional shortening, endocardial fractional shortening, and R-W1.

Conclusions

Newly diagnosed hypertensives show increased LVM and LV contractility, including carotid WI parameters and R-W1 values, as compared with normotensive subjects, but no differences in LV diastolic function.

Abbreviations
BP

blood pressure

BSA

body surface area

cESS

circumferential end-systolic stress

CIs

confidence intervals

LV

left ventricular

LVEF

LV ejection fraction

LVM

LV Mass

PWVβ

parameter pulse wave velocity derived from the β stiffness parameter

RF

radio frequency

WI

wave intensity

The impact of systolic and diastolic blood pressure (BP) on target organ damage and cardiovascular events is well established.[1] The left ventricle may be a target organ for hypertension, and left ventricular (LV) hypertrophy is a manifestation of target organ damage from hypertension that predicts adverse cardiovascular events in both hypertensive patients and the general population.[1-9] Hypertension is a complex syndrome where BP elevation is only one sign of multiple underlying pathophysiological abnormalities. Among these, excessive sympathetic activity, which is present in a large proportion of patients in the early phase of hypertension, may be responsible for high BP, heart rate, cardiac output with increased LV systolic performance, and vascular resistance and only marginal changes in LV structure. A normalization of the sympathetic tone with a parallel increase in vascular resistance (hypertrophic vessels) and LV hypertrophy occur during the course of hypertension.

The aim of the present study was to evaluate whether newly diagnosed untreated hypertensive patients show higher LV contractility than normotensive controls, as assessed by traditional echocardiographic indices and the more recent ventricular-arterial coupling parameters such as wave intensity (WI).

Materials and Methods

A total of 145 untreated hypertensive patients were prospectively enrolled and compared with 145 age- and sex-matched normotensive subjects. The study population was recruited from two centers (Department of Cardiology, T. Marciniak Hospital, Wroclaw, Poland and Division of Cardiology, Sant'Antonio Hospital, San Daniele del Friuli, Udine, Italy).

The study was approved by the institutional ethics committee and informed consent was obtained from all participants.

All subjects underwent physical examination and anthropometry; they also completed questionnaires about their medical history, family history of hypertension, physical activity, and dietary habits, including coffee intake, alcohol use, and cigarette smoking.[10, 11] Physical activity was assessed using a standardized questionnaire.[10]

Hypertensive patients were enrolled when they visited the outpatient clinic for recent onset of high BP, with values confirmed on three separate consecutive visits (systolic BP ≥ 140 mmHg and/or diastolic BP ≥ 90 mmHg). None of the patients presented with hypertension exceeded grade 2, according to the ESH/ESC guidelines.[12]

To generate a healthy sample, participants were excluded if they had diabetes mellitus, nephropathy, cardiovascular disease, or any chronic condition requiring medication.[13] BP measurements with an oscillometric semiautomatic sphygmomanometer and heart rate recordings were performed twice in the right arm after a 10 minute rest in the supine position in a quiet room; the two measurements were taken just before the transthoracic echocardiographic exam and before the carotid study (30 minutes after the first measurement). Phase V Korotkoff sounds were considered as DBP, except for subjects with sounds tending towards zero, in whom phase IV was taken.

Echocardiographic Measurements

Standardized transthoracic and Doppler echocardiographic examinations were performed under continuous electrocardiographic recording using commercially available equipment in all subjects (Alpha 10; Aloka Co, Ltd, Tokyo, Japan), in accordance with the American Society of Echocardiography recommendations.[14] Specific views included parasternal long- and short-axis views (at the mitral valve and papillary muscle levels) and apical four-, two-, and three-chamber views. Pulsed and continuous wave Doppler interrogation was performed on all four cardiac valves. Two-dimensionally guided M-mode measurements of the left ventricle (LV end-diastolic and end-systolic diameters, interventricular septal, and posterior wall thickness) were obtained from the parasternal short-axis view, with the patient in the left lateral position. LV mass (LVM) was calculated using the Penn convention as 1.04 [(interventricular septal thickness in diastole + LV internal diameter in diastole + posterior wall thickness in diastole)[3] – (LV internal diameter in diastole)[3]] – 13.6, and was indexed to body surface area (BSA).[15] Relative wall thickness was determined as (interventricular septal thickness in diastole + posterior wall thickness in diastole)/LV internal diameter in diastole.[16] Left atrial maximal volume was measured at the point of mitral valve opening using the biplane area-length method and corrected for BSA.[17]

Echocardiographic indices of LV performance were obtained as reported in Table 1.

Table 1. Echocardiographic Indices of Left Ventricular Performance Used in the Present Study
ParameterFormula
  1. Hs, combined septal and posterior wall thickness at end-systole (IVSTs+PWTs); IVSTd, interventricular septal thickness at end-diastole; IVSTs, interventricular septal thickness at end-systole; LVIDd, left ventricular internal diameter at end-diastole; LVIDs, left ventricular internal diameter at end-systole; PWTd, posterior wall thickness at end-diastole; PWTs, posterior wall thickness at end-systole; SBP, systolic blood pressure.

LV ejection fraction (EF)Monoplane Simpson method[14]
LV short-axis fractional shortening (FS)[(LVIDd-LVIDs)/LVIDd]×100 (%)[18]
LV end-systolic stress (ESS).334×SBP×LVIDs/[PWTs×(1+PWTs/LVIDs)]
Midwall fractional shortening (MWFS)(LVIDd + PWTd/2 + IVSTd/2) – (LVIDs + Hs/2)/(LVIDd + PWTd/2 +IVSTd/2)
Circumferential ESS (cESS)[SBP×(LVIDs/2)2]× {[1+(LVIDs/2+LV-PWTs)2/(LVIDs/2+LV-PWTs/2)2]}/(LVIDs/2+LV-PWTs)2-(LVIDs/2)2 [20]

Color Doppler Analysis

Valvular regurgitation was quantified from color Doppler imaging and categorized as absent, minimal (within normal limits), mild, moderate, or severe.[14]

Doppler-derived LV diastolic inflow was recorded in the apical four-chamber view by placing the sample volume at the level of the tips of the mitral valve and E, A peak velocities (in meters per second), and their ratio and E-wave deceleration time (the time elapsed between peak E velocity and the point at which the extrapolated deceleration slope of the E velocity crosses the zero baseline) were derived. Pulsed wave Doppler tissue imaging was performed in the four-chamber view at the septal and lateral mitral annular level and the mean of the two measurements was considered. The peak velocity of myocardial systolic wave (Sm), early peak (Em) and atrial (Am) diastolic wave (in centimeters per second), and the E/Em ratio were recorded.[21, 22]

Each parameter was assessed in three to five consecutive cardiac cycles and the mean values were obtained for the purposes of our study. All measured variables were re-estimated offline with an image processing workstation with the software ComPACS (v.10.5.8, MediMatic, Genoa, Italy).

Wave Intensity Analysis

WI was calculated as the product of the rate of change in BP (dP/dt) and blood flow velocity (dU/dt). WI was defined as (dP/dt)(dU/dt), where dP/dt and dU/dt were the derivatives of BP (P) and velocity (U) with respect to time of the blood flow.[23] WI was obtained at the level of the left common carotid artery before the bifurcation using an Aloka Alpha 10 ultrasound color Doppler machine implemented with a high-resolution echo-tracking algorithm (R-track) with a sampling rate of 1kHz on single line to detect vessel diameter changes and a system with real-time WI analysis software (Aloka). A wide-band multifrequency 5-13 MHz linear array probe was used. During a cardiac cycle, the typical pattern of carotid arterial WI exhibits two sharp positive peaks. The first peak (W1) represents a forward compression wave [also expressed as (max dP/dt)2c; ρ = blood density and c= pulse wave intensity] and is associated with flow acceleration and an increase in aortic pressure; it defines the relationship between a given compliance of the arterial conduit and the central organ.[24, 25] The second peak (W2) is a forward expansion wave [ρc (max dU/dt)[2]] and relates to the ability of the left ventricle to stop aortic blood flow; it correlates with the time constant of LV relaxation.[23, 24, 26] In addition, early diastolic relaxation is quantified by the rate of pressure decay and is influenced by late systolic arterial loading.[24, 26]

Arterial diameter varies throughout the cardiac cycle along with systolic and diastolic BP in the brachial artery, which was used for the calculation of the pressure waveform. The diameter changes of the vessel were measured as the difference between the displacement waveforms of the anterior and posterior walls, using the echo-tracking algorithm with the cursors set to track the media-adventiatia boundaries in the arterial wall approximately 2 cm proximal to the carotid bulb and calibrated by systolic and diastolic BP. The echo-tracking system, phase shift measurement based on digitalized data of broadband radio frequency (RF) echos, accurately measures the changes in arterial diameter. Peak acceleration (dU/dt) was derived from blood flow velocity measured by Doppler sonography with the range-gate positioned at the center of the vessel diameter. WI parameters (W1 and W2) were automatically calculated[23, 24, 27] (Figure 1).

Figure 1.

Left: assessment of common carotid artery with simultaneous display of change in diameter using the echo-tracking technique. Long-axis view, B-mode and M-mode.

Right: representative output of wave intensity. W1 and W2 values, R-1st: R-W1, 1st-2nd: W1-W2. Pulse wave velocity (PWV) is defined as PWVβ.

Reproducibility and variability of WI indexes have been proven to be clinically acceptable.[28-33] Inter- and intraobserver variability was evaluated in 15 randomly selected subjects in one center (San Daniele del Friuli, Udine, Italy). Intraobserver intrasession variability was evaluated from three repeated measurements in one session. Intraobserver intersession variability was studied by one observer by performing two sessions on different days in the same subject, with an interval between sessions from 2 to 7 days. Interobserver intrasession variability was evaluated by two observers who measured WI in each subject consecutively.

Ten consecutive beats were ensemble-averaged to obtain a representative waveform. A stiffness parameter pulse wave velocity derived from the β stiffness parameter (PWVβ) was also calculated from measured data.[33] Temporal indices were also automatically derived: the interval between the R wave on ECG and the first peak of WI (R−W1), corresponding to the pre-ejection period, and the interval between the first and second peaks (W1−W2), corresponding to the ejection time.

Statistical Analysis

Data are expressed as mean ± SD. Differences between hypertensive and normotensive subjects were tested with unpaired Student's t-test and chi-square test, either unadjusted or adjusted for body weight, heart rate, and physical activity. Hypertensive patients were divided into tertiles according to LVEF values, and one-way ANOVA was used to test differences among and within groups. Simple correlation and stepwise forward multiple linear regression analyses were carried out to weigh the independent effects of the potential determinants (independent variables: age, gender, pulse pressure, heart rate, LVEF, LVM/BSA, relative wall thickness, E wave, E/A, Em, E/Em, stroke volume/BSA, one-point carotid PWVβ). To test the inter-intra observer variability of WI and common carotid artery measurements, Pearson's bivariate two-tailed correlation test and Bland-Altman analysis were used. In particular Bland-Altmann was calculated as the difference of the two measurements plotted versus the mean of the measurements. All statistical analyses were performed using SYSTAT for Windows release 12.0 (Systat Software Inc, Chicago, IL).

Results

WI reproducibility obtained by the same observer or by two different observers during the same session was satisfactory. For intraobserver variability, Pearson's correlations were r = .88 (P < .0001) and r = .81 (P = .0003) for W1 and W2, respectively; and the Bland-Altman 95% confidence intervals (CIs) were 1.13 (−.34 to 2.60 mmHg m/s3) for W1 and −.12 (−.61 to .43 mmHg m/s3) for W2. For maximum and minimum common carotid diameters, Pearson's correlations were r = .87 (P = .0005) and r = .84 (P = .0011), respectively. The Bland-Altman 95% CIs were .20 (−.64– 1.03 mm) and −.23 (−.70—1.16 mm) for maximum and minimum carotid diameters respectively. For interobserver variability, Pearson's correlations were r = .82 (P = .0003) and r = .82 (P = .0003) for W1 and W2, respectively; and the Bland-Altman 95% CIs were −.2 (−1.5 to 1.0 mmHg m/s3) for W1 and −.1 (−.61 to .41 mmHg m/s3) for W2. For maximum and minimum common carotid diameters, Pearson's correlations were r = .89 (P = .0002) and r = .87 (P = .0005), respectively. The Bland-Altman 95% CIs were −.06 (−.82–.71 mm) and −.04 (−.87–.79 mm), respectively. Test-retest reproducibility or repeatability was r = .89 (P < .0001) and r = .89 (P < .0001) for W1 and W2, respectively; and the Bland-Altman 95% CIs were −.11 (−1.6 to 1.37 mmHg m/s3) for W1 and −.05 (−.49 to .39 mmHg m/s3) for W2. For maximum and minimum common carotid diameters, Pearson's correlations were r = .80 (P = .006) and r = .79 (P = .008), respectively; and the Bland-Altman 95% CIs were .12 (−.98–1.22 mm) and .17 (−1.01–1.36 mm), respectively.[27]

The general characteristics of the study population are reported in Table 2. Hypertensives were heavier, less physically active and had higher LDL-cholesterol, triglyceride, and glucose, and lower HDL-cholesterol levels. Echocardiographic indices for both hypertensive and normotensive subjects are summarized in Table 3. After adjustment for heart rate, weight and physical activity, hypertensive patients still showed higher LVM/BSA, LVEF, circumferential end-systolic stress (cESS), and LV end-systolic stress than normotensive controls and no differences were observed in diastolic parameters, except for A wave that was higher in hypertensives. WI parameters were higher but R-W1 or pre-ejection time was significantly lower in hypertensives than in normotensives. Also systemic vascular resistance and one-point carotid PWVβ were higher in hypertensives than in normotensives (Table 3).

Table 2. General Characteristics of the Study Population
 Hypertensives (n=145)Normotensives (n=145)PP (body weight, physical activity)
  1. BMI, body mass index; BP, blood pressure; BSA, body surface area; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

Age (years)51.9 ± 12.150 ± 10.1.17-
Gender (M/F)96/4991/54.5-
Smoker (no/yes)111/14107/22.2 
Coffee intake (no/yes)9/9915/114.5 
Physical activity (yes/no)50/7967/78.04-
Weight (kg)81.9 ± 13.171.5 ± 12.0.0001-
BSA (m2)1.9±.21.9±.3.8-
BMI (kg/m2)27.9 ± 4.624.5 ± 3.1.0001-
Systolic BP (mmHg)156.8 ± 18.3122.4 ± 20.9.0001.0001
Diastolic BP (mmHg)88.8 ± 11.875.0 ± 8.8.0001.0001
Pulse pressure (mmHg)65.8 ± 16.447.2 ± 10.1.0001.0001
Mean BP (mmHg)110.6 ± 12.191 ± 9.3.0001.0001
Heart rate (bpm)70.9 ± 12.565.0 ± 12.7.0001.075
Laboratory    
Glucose (mg/dL)98.8 ± 18.492.6 ± 15.2.006.2
Total cholesterol (mg/dL)215.7 ± 44.9214.4 ± 37.2.7.02
HDL-cholesterol (ml/dL)54.5 ± 14.963.1 ± 16.2.0001.6
LDL-cholesterol (ml/dL)129.4 ± 45.7114.9 ± 57.5.02.01
Triglycerides (ml/dL)125.3 ± 86.894.8 ± 48.001.8
Table 3. Echocardiographic Characteristics of the Study Population
 Hypertensives (n=145)Normotensives (n=145)PP (heart rate, body weight, physical activity)
  1. BSA, body surface area; cESS, circumferential end-systolic stress; CO, cardiac output; IVSTd, interventricular septal thickness at end-diastole; LVIDd, left ventricular internal diameter at end-diastole; LVEDV, left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LV ESS, left ventricular end-systolic stress; LVM, left ventricular mass; MWFS, midwall fractional shortening; PWTd, posterior wall thickness at end-diastole; PWVβ, pulse wave velocity; RWT, relative wall thickness; SV, stroke volume.

  2. a

    No adjustment for body weight.

Left atrium    
Left atrial volume (ml)47.9 ± 15.346.2 ± 21.2.47.8
Left ventricle    
LVM/BSA (g/m2)105.6 ± 29.593.4 ± 24.7.0001.0001a
RWT36.5 ± 7.334.2 ± 5.9.003.1
LVIDd (mm)51.4 ± 5.250.0 ± 5.0.02.6
IVSTd (mm)9.3 ± 2.28.5 ± 1.5.0001.2
PWTd (mm)9.3 ± 1.78.5 ± 1.4.0001.06
LVEDV/BSA (ml/m2)110.5 ± 41.387.8 ± 28.4.039 
LVEF (%)64.9 ± 6.562.9 ± 6.4.014.008
Fractional shortening (%)39.5 ± 6.638.0 ± 6.8.052.1
MWFS (%)21.3 ± 5.420.3 ± 5.1.3
cESS (103/dynes/cm2)137.9 ± 41.3121.9 ± 39.001.006
LV ESS (103dynes/cm2)65.9 ± 23.459.5 ± 22.2.02.048
SV/BSA (ml/m2)37.7 ± 11.138.4 ± 9.9.5.9a
CO/BSA (ml/min/m2)2663 ± 919.92381 ± 715.5.04.2a
E/A1.3 ± 0.71.6 ± 0.6.001.9
E cm/s63.3 ± 1665.1 ± 16.3.9
A cm/s57 ± 2146.1 ± 16.0001.004
E/Em6.6 ± 2.26.3 ± 1.9.1.7
Wave intensity    
W1 (mmHg m/s3)13 646 ± 73689308 ± 4675.0001.001
W2 (mmHg m/s3)4289 ±20172995 ± 1868.0001.001
R-W1 (ms)99.4 ± 17.2108.5 ± 16.2.0001.004
W1-W2 (ms)267.8 ± 61.4275.6 ± 60.3.3.8
Left common carotid diameter    
Maximum (mm)7.2±.96.9±.9.001.2
Minimum (mm)6.7±.96.4±.9.0001.2
One-point carotid PWVβ (m/s)6.5 ± 1.65.42 ± 1.0.0001.0001

Given that differences between groups were related to LV function, hypertensive patients were divided into tertiles according to LVEF. No differences were found within tertiles in hemodynamic and laboratory variables, whereas diastolic function parameters were similar among the three groups (Table 4). A trend towards higher relative wall thickness and midwall fractional shortening was evident in hypertensive patients in the highest LVEF category. They also showed higher LV fractional shortening but lower cESS, LV end-systolic stress, LVM/BSA, LV end-diastolic diameter, and left atrial volume (Table 5). W1 increased significantly in parallel with LVEF, whereas an opposite trend was observed for R-W1. W2 did not change among the three subgroups of hypertensive patients (Table 5).

Table 4. General Characteristics of Hypertensive Patients Divided Into Tertiles According to Left Ventricular Ejection Fraction Values
 Group 1 LVEF 57.4 ± 3.5% (n=45)Group 2 LVEF 63.1 ± 1.4% (n=46)Group 3 LVEF 72 ± 3.5% (n=47)P
  1. BMI, body mass index; BP, blood pressure; BSA, body surface area; LVEF, left ventricular ejection fraction.

Age (years)53.0 ± 11.050.6 ± 13.351.9 ± 13.3.6
Gender (M/F)32/1335/1128/19.2
Weight (kg)80.7 ± 14.280.9 ± 13.682.8 ± 10.9.6
BSA (m2)1.9±.21.9±.21.9±.2.9
BMI (kg/m2)27.2 ± 3.927.7 ± 5.528.2 ± 3.9.5
Systolic BP (mmHg)155.2 ± 19.5152.6 ± 19.2156.5 ± 16.6.6
Diastolic BP (mmHg)88.1 ± 13.790.6 ± 11.787.4 ± 9.5.4
Pulse pressure (mmHg)67.1 ± 16.162.0 ± 16.269.1 ± 16.1.1
Mean BP (mmHg)110.4 ± 13.9111.3 ± 12.4110.5 ± 9.7.9
Heart rate (bpm)70.1 ± 11.769.4 ± 12.673.7 ± 13.6.2
Table 5. Echocardiographic Characteristics of Hypertensive Patients Divided Into Tertiles According to Left Ventricular Ejection Fraction Values
 Group 1 LVEF 57.4 ± 3.5% (n=45)Group 2 LVEF 63.1 ± 1.4% (n=46)Group 3 LVEF 72 ± 3.5% (n=47)P
  1. BSA, body surface area; cESS, circumferential end-systolic stress; CO, cardiac output; HR, heart rate; IVSTd, interventricular septal thickness at end-diastole; LVIDd, left ventricular internal diameter at end-diastole; LVEDV, left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LV ESS, left ventricular end-systolic stress; LVM, left ventricular mass; MWFS, midwall fractional shortening; PWTd, posterior wall thickness at end-diastole; PWVβ, pulse wave velocity; RWT, relative wall thickness; SV, stroke volume.

Left atrium    
Left atrial volume52.2 ± 19.846.1 ± 11.245.9 ± 12.9.1
Left ventricle
LVM/BSA (g/m2)110.7 ± 26.9105.4 ± 31.9101.6 ± 30.6.3
RWT35.0 ± 6.635.9 ± 6.738.5 ± 8.1.05
LVIDd (mm)53.2 ± 4.551.6 ± 5.149.6 ± 5.4.003
IVSTd (mm)9.1 ± 1.49.3 ± 2.49.5 ± 2.5.6
PWTd (mm)9.3 ± 1.69.2 ± 1.79.4 ± 1.7.8
Fractional shortening (%)37.8 ± 7.138.1 ± 7.742.0 ± 6.1.007
MWFS (%)20.1 ± 5.320.5 ± 4.423.2 ± 5.9.01
cESS (103/dynes/cm2)153.7 ± 50.8133.6 ± 29.7126.7 ± 36.9.004
LV ESS (103/dynes/cm2)74.7 ± 28.663.0 ± 15.959.2 ± 20.5.003
SV/BSA (ml/m2)36.7 ± 8.238.1 ± 10.341.2 ± 8.9.06
CO/BSA (ml/min/m2)1945 ± 15792389 ± 10182298 ± 1029.5
E/A1.01.2 ± 0.61.4±.81.2 ± 06.3
E cm/s59.7 ± 14.763.8 ± 1.666 ± 16.9.1
A cm/s56.1 ± 2152.8 ± 18.861.7 ± 22.8.1
E/Em6.4 ± 26.3 ± 2.27.1 ± 2.4.2
Wave intensity
W1 (mmHg m/s3)11 934 ± 583611 576 ± 585717 227 ± 8889.0001
W2 (mmHg m/s3)4225 ± 21284052 ±18724616 ±2082.4
R-W1 (ms)102.5 ± 14.0102.8 ± 13.494.5 ± 16.4.01
W1-W2 (ms)264.1 ± 24.4266.1 ± 23.5266.1 ± 48.2.6
Left common carotid diameter    
Maximum (mm)7.28±.897.13±.887.32 ± 1.02.4
Minimum (mm)6.82±.936.64±.96.82 ± 1.01.5
One-point carotid PWVβ (m/s)6.3 ± 1.96.3 ± 1.56.8 ± 1.4.2

A simple correlation was tested between W1 and W2 and age, gender, pulse pressure, heart rate, LVEF, LVM/BSA ratio, relative wall thickness, E wave, E/A, Em, E/Em, stroke volume/BSA, maximum and minimum carotid diameter, and one-point carotid PWVβ. In normotensive subjects, a significant correlation was found between W1 and pulse pressure (r = .547, P < .0001), heart rate (r = .27, P < .0001), LVEF (r = .29, P = .027) and between W2 and pulse pressure (r = .32, P = .016), heart rate (r = .24, P = .004) or one-point carotid PWVβ (r = .3, P = .036). In hypertensive patients, only W1 significantly correlated with pulse pressure (r = .478, P < .0001). No correlation was found between carotid diameters (minimum and maximum diameter) either in normotensives or hypertensives. On multiple regression analysis, age correlated negatively and pulse pressure and heart rate positively with W1, whereas maximum carotid diameter correlated negatively and pulse pressure positively with W2 (Table 6).

Table 6. Backward Multiple Regression Analysis in Hypertensive and Normotensive Subjects of the Dependent Variables W1 and W2
 R2βP
  1. BSA, body surface area; LVEF, left ventricular ejection fraction; PWV, pulse wave velocity; RWT, relative wall thickness; SV, stroke volume.

Hypertensives   
W1.426  
Age (years) −.356.0001
Gender −1.98 
Pulse pressure (mmHg) .54.0001
Heart rate (bpm) .33.0001
LVEF (%) 015.01
W2.078  
Age (years) −.17.032
Pulse pressure (mmHg) .24.006
SV/BSA (ml/m2) −.158.042
RWT .235.004
Maximum common carotid diameter (mm) −.19.013
Normotensives   
W1.49  
Age (years) −.292.0001
Gender −.19.005
Pulse pressure (mmHg) .528.0001
Heart rate (bpm) .241.001
LVEF (%) .173.01
W2.289  
Gender −.183.03
Pulse pressure (mmHg) .22.008
Heart rate (bpm) .28.001
Maximum common carotid diameter (mm) −.34.0001
PWV (m/s) .231.009

Discussion

The natural history of sustained hypertension became clearer with Lund-Johansen's work, when he investigated the hemodynamic features of patients with early hypertension, presenting with minimal BP elevation and no target organ damage. The patients had a hyperkinetic circulation in the early phase,[34] but they developed established hypertension thereafter, characterized by increased peripheral resistance and normal cardiac output.[35] Additional features were later described by Amarena and Julius,[36] who found that subjects in the early stage of hypertension were overweight with elevated cholesterol, triglyceride, and plasma insulin levels and decreased HDL-cholesterol.[37]

In the present study, conducted in newly diagnosed untreated hypertensive patients, we found that hypertensives had higher LVM/BSA but similar diastolic function, after adjustment for confounding factors, and a better LV performance than normotensive controls, including better performance according to WI parameters. On the other hand, hypertensive patients had increased LV end-systolic stress, systemic vascular resistance, and local arterial stiffness. Among hypertensive subgroups divided into tertiles according to LVEF (the most common index of LV function), a trend towards LV remodeling, less LV end-systolic stress, higher W1, and lower pre-ejection time was observed among those in the highest LVEF category. Hypertensive patients were also heavier and showed increased total and LDL-cholesterol levels.

Our patients were in the early stage of hypertension and were aware of their status for a mean period of 7 months. At this early phase, increased LVM/BSA, vascular resistance and LV contractility were observed, as determined by LVEF, ventricular-arterial coupling parameters (W1 and W2) and R-W1 or shorter pre-ejection time. W1 represents a forward compression wave and is associated with flow acceleration and increased aortic pressure; its magnitude correlates with the magnitude of LV pressure rise.[38] W2 is a forward expansion wave and relates to the ability of the left ventricle to stop aortic blood flow; it correlates with the time constant of LV relaxation[23, 25] and is influenced by late systolic arterial loading. R-W1 corresponds to the LV pre-ejection period and W1-W2 corresponds to the ejection time. A prolonged pre-ejection period and a shortened ejection time were reported as significantly correlated with decreased LV systolic function.[39] Although no differences in diastolic function were observed between groups, A wave and E/A ratio were significantly higher and lower, respectively, in hypertensives when crude diastolic E wave, A wave and E/A parameters were considered. However, after adjustment for weight, heart rate and physical activity, only A wave remained significantly different between groups. The increase in A wave might reflect early diastolic dysfunction and, in the early stage of hypertension, it is mainly influenced by age and heart rate in an inverse fashion, as previously reported by Palatini et al[40] and confirmed by our results.

In our study arterial stiffness (one-point carotid PWVβ) and LV end-systolic stress were found to be higher in hypertensive patients than in normotensive controls. Although increased W2 values in hypertensives may be somewhat surprising,[28, 41] the W2 formula itself may account for this finding given that hypertensive patients show higher pulse wave velocity. Increased contractility and reduced LV stress in hypertensive subjects are consistent with previous studies[40, 42] that used either traditional or innovative parameters of LV function.[43, 44] Opposite to our findings, Du et al[45] analyzed WI according to different LV geometry patterns in a population of untreated hypertensives and found similar W1, W2 and R-W1 in patients with LV remodeling. As compared with our findings, the different study population may account for the differences observed in WI parameters. Our hypertensive patients were aware of their status for a few months, probably after a long period of high-normal BP. They also had a tendency towards higher heart rate and LV remodeling, suggesting a “hyperkinetic phase” associated with increased LV contractility.[40]

Smaller end-diastolic dimensions and increased LV wall thickness and remodeling have been reported in patients with normal or increased LV contractility. De Simone et al[46] demonstrated that a hyperkinetic LV function, as assessed by endocardial fractional shortening, may mask a subgroup of hypertensive patients with depressed LV myocardial function, LV hypertrophy, and LV concentric remodeling. Conversely, LVM and relative wall thickness were within normal range in our group of hypertensive patients. In a series of hypertensive stage I subjects, Palatini et al[40] observed elevated plasma norepinephrine concentrations in addition to LV remodeling. These authors concluded that more complete LV emptying during systole occurred in these patients to compensate for lower end-diastolic volume.

Hypertensive subjects in the highest LVEF category showed increased LV mechanical work generated by the myocardium along with increased inotropic activity, as demonstrated by the correlation of W1 with Doppler-derived dP/dt index[38] and shorter R-W1.[39] Experimental and clinical data have shown that the first derivative of LV pressure (dP/dt) is an important function of the inotropic state of the myocardium.[47]

In both hypertensive and normotensive subjects, arterial stiffness (pulse pressure) and age (negatively correlated) were found to be the most powerful independent determinants of W1. The negative relation with age was consistent with previous findings from our group[28] and from Borlotti et al.[48] Aging is responsible for the impairment of LV function,[49, 50] including WI. In particular, arterial stiffness per se may result in a significant impairment of WI also considering the basic mathematical formulation of W1 [W1=max dP/dt)2c (ρ=blood density and c=pulse wave intensity)].

In hypertensives, W2 was independently correlated with pulse pressure and relative wall thickness, suggesting that arterial stiffness and LV remodeling may have a greater effect than alterations in ventricular relaxation, at least in the early stage of hypertension. In normotensives, W2 was correlated with arterial stiffness and heart rate. In both groups W2, also considered the pre-relaxation time, was negatively correlated with maximum carotid diameter. Increased luminal diameter may be considered a deterioration of the arterial wall due to fractures of the load-bearing elastin fibers in response to the tensile stress.[51] Increased luminal diameter may be considered a deterioration of the arterial wall due to fractures of the load-bearing elastin fibers in response to the tensile stress.[51]

In the present study, inter- and intraobserver variability as well as repeatability of WI and carotid diameter analysis were satisfactory, although a bit lower than what reported by others.[32]

Limitations

Several limitations should be acknowledged. First, peripheral BP measurements are known to overestimate central BP, especially in young populations. Brachial rather than central BP was measured, as reported by Sugawara et al[29] who demonstrated a good linear relationship between carotid arterial pressure and diameter. Second, this technique for WI measurement is based on the assumption of a relatively linear relationship between pressure and diameter, which requires calibration of BP measurements to obtain arterial pressure waveforms, different from the method based on measurement of blood flow velocity and diameter as described by Borlotti et al.[48] Third, brachial BP measurements were taken just before starting carotid artery evaluation, assuming that hemodynamic parameters did not change significantly.

In conclusion, our results suggest that, in the early phase of hypertension, diastolic function does not differ between hypertensive and normotensive subjects, though hypertensives have increased LVM and LV contractility, including WI parameters and carotid arterial stiffness. Among hypertensive patients, those with a tendency towards LV remodeling showed increased LV contractility, including W1 and R-W1 values, but no changes in W2, pre-relaxation time and carotid arterial stiffness.

Ancillary