Background
Subclinical carotid atherosclerosis has been associated with impaired left ventricular (LV) function and the development of heart failure. Whether impaired LV function is related primarily to increased intima-media thickness (IMT) or burden of plaque disease or both remains to be determined.
Methods
A total of 2,279 subjects without clinical cardiovascular disease recruited from the London Life Sciences Prospective Population cohort study were studied. Carotid ultrasonography and transthoracic echocardiography were performed on all subjects. Carotid IMT and plaque scores were measured, and their relationships with LV volumes, LV ejection fraction, myocardial LV longitudinal function (Sa and Ea velocities), and LV filling pressure (E/Ea ratio) were assessed before and after adjustment for covariates.
Results
Compared with those without carotid artery disease, subjects with either increased IMT and/or presence of plaque disease had identical Sa velocities (both 9.0 cm/sec), lower Ea velocities (8.7 vs 9.9 cm/sec, P < .001) and higher E/Ea ratios (8.4 vs 7.6, P < .001). After multiple linear regression analysis, increasing IMT remained independently related to reduced Ea velocity ( P < .001) but not LV ejection fraction, Sa velocity, or E/Ea ratio. In a separate adjusted analysis, subjects with severe burdens of carotid plaque disease (more than five plaques) had reduced LV ejection fractions (β = −2.9; 95% confidence interval [CI], 1.0 to 4.8, P = .003), attenuated Sa velocities (β = −0.79; 95% CI, −1.2 to −0.3, P = .003), attenuated Ea velocities 2 (β = −0.79; 95% CI, −1.3 to −0.2, P = .007), and increased E/Ea ratios (β = 0.84; 95% CI, 0.2 to 1.5, P = .009) compared to individuals without carotid plaques.
Conclusion
These findings demonstrate that subclinical carotid plaque disease rather than IMT is more closely related to LV systolic function and LV filling pressure. These data support the application of carotid ultrasonography beyond cardiovascular disease risk prediction, while providing insight into potential mechanisms underlying the development of subclinical LV dysfunction.
The recognition of left ventricular (LV) dysfunction in asymptomatic individuals is of paramount importance, because not only does this phase of disease foretell the development of incident congestive heart failure, but its treatment has been shown to delay the onset of overt heart failure symptoms. This has prompted the search for “risk factors” of incipient myocardial dysfunction, examples of which include the burden of conventional cardiovascular disease risk factors, the presence of novel risk factors, abnormalities in LV geometry, increased LV mass, and the degree of subclinical atherosclerosis. Increased intima-media thickness (IMT) and the presence of atheromatous plaques are forms of carotid artery disease that are both regarded as surrogate markers of atherosclerosis. Although these two phenotypes are related, the pathophysiology underlying intima-media thickening and plaque formation are not necessarily similar. Hypertrophy of the medial layer of the arterial wall can occur either as a response to hypertension or as a manifestation of normal aging, whereas plaque formation represents the maturation of the atherosclerotic process. Accordingly, it has been suggested that these two types of arterial disease have distinct relationships to cardiac disease and vascular events.
Recent studies have suggested that increased carotid IMT is also associated with both incipient myocardial systolic dysfunction and the development of clinical heart failure. However, these studies have been unable to discern whether impaired LV function or clinical heart failure events are related primarily to intima-media thickening, burden of plaque disease, or both. Therefore, we assessed separately the relationships of carotid IMT and carotid plaque disease with echocardiographic parameters of LV systolic and diastolic function in a large cohort of subjects without clinical cardiovascular disease.
Methods
Subjects were recruited between August 2004 and November 2007 from the London Life Sciences Prospective Population (LOLIPOP) study. LOLIPOP is a population-based study of about 30,000 Indian Asian and European white men and women recruited from the lists of 58 general practitioners in West London. The methodology for this substudy has been described previously. Briefly, we randomly selected 2,315 Indian Asian and European white subjects, aged 35 to 74 years, who were free from clinical cardiovascular disease. Consenting subjects underwent physical assessments, including blood pressure (BP) determination, anthropometric measurements (height, weight, and waist/hip ratio), and electrocardiography. Subjects were then invited to undergo echocardiography and carotid ultrasonography and to provide fasting plasma and serum samples for biochemical analysis, including total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and glucose. The study was approved by the Northwick Park Hospital and Ealing Hospital research ethics committees.
Echocardiography
Transthoracic two-dimensional echocardiography was performed by experienced sonographers using a digital commercial harmonic imaging ultrasound system with an S3 3-MHz phased-array transducer (Philips iE33; Philips Medical Systems, Best, The Netherlands) at a single center. LV dimensions were obtained in the parasternal short-axis view with measurement of interventricular septal thickness in diastole, LV dimension in diastole, LV dimension in systole, and LV posterior wall thickness in diastole. LV mass was calculated using the Devereux formula and indexed to height to provide LV mass index. LV end-diastolic and end-systolic volumes indexed to body surface area were measured using Simpson’s apical biplane rule. Tracing of the LV contour was performed carefully to exclude papillary muscles and trabeculations, as recommended by the American Society of Echocardiography. LV ejection fraction (EF) was automatically calculated following acquisition of the LV volumes using Simpson’s method. Transmitral spectral Doppler was performed to obtain mitral inflow peak E-wave and peak A-wave velocities. Left atrial volume was calculated from three measurements of left atrial dimension using a validated formula : π/6(PLAX × A4C1 × A4C2), where PLAX is left atrial dimension measured in the parasternal long-axis view, and A4C1 and A4C2 are measurements of the long axis and short axis in the apical four-chamber view. Left atrial volume was then indexed to body surface area.
Doppler Tissue Imaging
Myocardial velocities were measured online using a standard pulse-wave Doppler technique acquired during a breath hold over two consecutive cardiac cycles using low-velocity, high-intensity myocardial signals at a high frame rate (>150 MHz). The imaging angle was adjusted to ensure as near parallel alignment of the beam as possible with the myocardial segment of interest. The sample volume was placed at the junction of the LV wall with the mitral annulus of the septal and lateral myocardial segments from the apical four-chamber view and inferior and anterior myocardial segments from the apical two-chamber view. Peak velocities during systole (Sa) and early diastole (Ea) were measured online from all four mitral annular site segments and averaged. Estimated LV filling pressure was derived from the ratio of transmitral E velocity to Ea velocity.
Carotid Ultrasonography
Carotid duplex scans were conducted using a high-resolution, nonharmonic B-mode ultrasound system (Philips iE33) with an 11-MHz to 3-MHz transducer. The proximal, mid, and distal common carotid artery (CCA), bifurcation of the CCA (CCA-Bif), and proximal portion of the internal and external carotid arteries were systematically interrogated in long-axis and short-axis views. Gated cine loops of two cardiac cycles and still images during diastole of the CCA and CCA-Bif were digitally acquired for offline analysis of IMT. Measurements of IMT were taken at the far wall of the CCA and CCA-Bif (about 1 cm proximal to the carotid bulb) during end-diastole using a semiautomated edge detection algorithm (QLAB version 5; Philips Medical Systems). A region of interest 1 cm in length was placed parallel to the vessel wall, allowing the software to detect the luminal-intimal and medial-adventitial interfaces at the far wall of the vessel, enabling determination of the IMT. Values for mean IMT were derived from three measurements at the distal, mid, and proximal CCA and three measurements from the CCA-Bif, for both the left and right arteries. If there was evidence of carotid plaque formation, care was taken not to include the plaque in the IMT measurement. The CCA, CCA-Bif, and internal and external carotid arteries were then systematically interrogated in short-axis and long-axis views for the detection of atherosclerotic plaque formation using color Doppler imaging. A plaque was defined as a focal structure encroaching into the arterial lumen by ≥0.5 mm or a distinct area of intima-medial thickening ≥50% greater than the adjacent wall or >1.5 mm in thickness.
Statistical Analysis
Clinical and echocardiographic parameters were primarily stratified according to those subjects with either normal carotid arteries versus those with evidence of carotid artery disease, defined as the presence of either a mean IMT measurement > 75th age-specific percentile values obtained from our healthy reference cohort and/or presence of plaque disease. In this cohort, we did not observe a relationship between ethnicity and the burden of carotid atherosclerosis, as previously published. Data are presented as mean ± SD. Student’s t tests were performed to compare differences in continuous variables and χ 2 tests to assess differences in categorical variables between the two groups. Parameters of LV volumes and function were then stratified according to (1) different patterns of carotid artery disease, namely, those subjects having no evidence of carotid disease, increased IMT alone, plaque disease alone, or the presence of both increased IMT and plaque disease, and (2) carotid plaque burden, as derived from each subject’s plaque score, defined as the total number of plaques identified in both their carotid arteries. Subjects with plaque scores of 0, 1 or 2, 3 to 5, and >5 were classified as having either no, mild, moderate, or severe plaque burdens, respectively.
Linear regression was then performed to assess the relationship of IMT with LV volumes, LV EF, Sa velocity, Ea velocity, and E/Ea ratio. The first model adjusted for the variables of age, race, gender, presence of type 2 diabetes, systolic BP, diastolic BP, use of antihypertensive medications, and LV mass. To assess whether the relationships of IMT with LV volumes and with LV function was influenced by subclinical atherosclerosis, the presence of carotid plaque disease was entered as a covariate into the second model. Further adjusted analyses were performed to assess the relationship of the severity of carotid plaque burden with LV volumes and function using analysis of covariance. The effect of plaque burden on LV volumes and function was adjusted for age, race, gender, presence of type 2 diabetes, systolic BP, diastolic BP, use of antihypertensive medications, and LV mass. For post hoc comparisons of least square means arising in the analysis of covariance models, the α level for rejection of the null hypothesis was adjusted after Bonferroni correction on the basis of the number of comparisons being made (for three comparisons, P = .05/3 = .0167, rounded up to .02). Given the correlation between the end points (measures of LV function), a full Bonferroni correction was avoided, which assumes independence of the tests being performed. Statistical analyses were performed using SPSS version 15 (SPSS, Inc., Chicago, IL).
Results
Of the 2,315 subjects enrolled into the study, 2,279 had adequate data entry of carotid artery disease and LV function parameters. Of these subjects, 1,339 (59%) were found to have evidence of carotid artery disease (either increased IMT and/or evidence of plaque disease). Carotid plaque disease was present in 1,027 subjects (45%), of whom 674 (66%) were classified as having mild degrees of plaque disease (one or two plaques), 310 (30%) as having moderate degrees of plaque disease (three to five plaques), and 43 (4%) as having severe degrees of plaque disease (more than five plaques). The cohort consisted of 1,595 men (70%) and 1,025 European white subjects (45%).
Compared to nondiseased carotid arteries, the presence of increased IMT, plaque disease, or both was associated with a greater burden of cardiovascular risk factors ( Table 1 ), such as higher prevalences of male gender, type 2 diabetes, treatment for hypertension, and history of smoking. These subjects were also older and had higher systolic and diastolic BPs, greater body mass indexes, lower high-density lipoprotein cholesterol, higher low-density lipoprotein cholesterol, and higher fasting glucose compared with controls. Structural and hemodynamic parameters reflecting LV function were also significantly associated with carotid artery disease ( Table 2 ). Although LV EFs were slightly lower in those with carotid disease, the difference was statistically significant ( P = .03), however, Sa velocity was identical between the two groups. Abnormalities in the morphophysiologic parameters of LV diastolic function, such as LV mass, left atrial volume, Ea velocity, and the E/Ea ratio, were all significantly related to the presence of carotid disease. Table 3 illustrates the demographic and clinical characteristics according to burden of plaque disease. As expected, individuals with more plaques were older and more likely to be hypertensive, to be diabetic, to have higher serum glucose, and to have higher LV mass. However, there was no discernible relationship between plaque burden and dyslipidemia.
| Variable | Normal carotid arteries ( n = 934) | Carotid disease (either IMT > 75th percentile and/or presence of plaque) ( n = 1,345) | P |
|---|---|---|---|
| Age (years) | 54 ± 9 | 59 ± 9 | <.001 |
| Men (%) | 61 | 77 | <.001 |
| European white (%) | 44 | 46 | .59 |
| BMI (kg/m 2 ) | 27 ± 4 | 28 ± 4 | .001 |
| Systolic BP (mm Hg) | 127 ± 17 | 136 ± 19 | <.001 |
| Diastolic BP (mm Hg) | 80 ± 10 | 81 ± 10 | <.001 |
| Treated for hypertension (%) | 18 | 29 | <.001 |
| Type 2 diabetes (%) | 9 | 17 | <.001 |
| Glucose (mmol/L) | 5.4 ± 1.4 | 5.8 ± 2.0 | <.001 |
| Total cholesterol (mmol/L) | 5.4 ± 1.0 | 5.5 ± 1.1 | .16 |
| HDL cholesterol (mmol/L) | 1.4 ± 0.4 | 1.31 ± 0.33 | <.001 |
| LDL cholesterol (mmol/L) | 3.3 ± 0.8 | 3.4 ± 0.9 | .031 |
| Ever smoker (%) | 27 | 36 | <.001 |
| Variable | Normal carotid arteries ( n = 934) | Carotid disease (either IMT > 75th percentile and/or presence of plaque) ( n = 1,345) | P |
|---|---|---|---|
| LVM (g) | 158 ± 49 | 177 ± 54 | <.001 |
| LAVI (mL/m 2 ) | 16.2 ± 5.4 | 17.7 ± 6.4 | <.001 |
| EDVI (mL/m 2 ) | 38.3 ± 9.9 | 38.4 ± 10.1 | .17 |
| ESVI (mL/m 2 ) | 14.5 ± 4.7 | 14.8 ± 5.2 | .81 |
| LV EF (%) | 62.3 ± 5.8 | 62.0 ± 6.0 | .03 |
| Sa velocity (cm/sec) | 9.0 ± 1.7 | 9.0 ± 1.7 | .85 |
| Ea velocity (cm/sec) | 9.9 ± 2.3 | 8.7 ± 2.1 | <.001 |
| E/Ea ratio | 7.6 ± 2.1 | 8.4 ± 2.4 | <.001 |
| Variable | None (0 plaques) ( n = 1,252) | Mild (1-2 plaques) ( n = 674) | Moderate (3-5 plaques) ( n = 310) | Severe (>5 plaques) ( n = 43) | P |
|---|---|---|---|---|---|
| Age (years) | 54 ± 10 | 59 ± 9 | 64 ± 8 | 66 ± 5 | <.001 |
| Systolic BP (mm Hg) | 128 ± 17 | 135 ± 20 | 141 ± 19 | 144 ± 17 | <.001 |
| Diastolic BP (mm Hg) | 80 ± 10 | 82 ± 10 | 82 ± 11 | 80 ± 11 | .003 |
| Treated for hypertension (%) | 18 | 27 | 43 | 60 | <.001 |
| BMI (kg/m 2 ) | 27 ± 4 | 27 ± 4 | 28 ± 4 | 27 ± 4 | .19 |
| Type 2 diabetes (%) | 10 | 15 | 25 | 38 | <.001 |
| Glucose (mmol/L) | 5.5 ± 1.5 | 5.8 ± 2.0 | 6.1 ± 1.9 | 7.0 ± 4.0 | <.001 |
| Total cholesterol (mmol/L) | 5.4 ± 1.1 | 5.5 ± 1.0 | 5.5 ± 1.2 | 5.0 ± 1.1 | .045 |
| HDL cholesterol (mmol/L) | 1.3 ± 0.3 | 1.3 ± 0.4 | 1.3 ± 0.3 | 1.3 ± 0.2 | .06 |
| LDL cholesterol (mmol/L) | 3.4 ± 0.9 | 3.4 ± 0.9 | 3.5 ± 1.0 | 2.9 ± 0.9 | .007 |
| LVM (g) | 162 ± 49 | 176 ± 55 | 187 ± 55 | 193 ± 49 | <.001 |
The tissue Doppler parameters of LV function were further stratified according to different patterns of carotid artery disease ( Figure 1 ). Compared with subjects without evidence of carotid artery disease ( n = 940 [41%]), Sa velocity was not significantly influenced by either the presence of increased IMT alone ( n = 319 [14%]), the presence of plaque disease alone ( n = 575 [25%]), or the presence of both increased IMT and plaque formation ( n = 446 [20%]). The presence of carotid artery disease, whether in the form of increased IMT, plaque disease, or both, was significantly related to impaired early diastolic function in a dose-dependent manner (9.3, 8.8, and 8.4 cm/sec, respectively, vs 9.9 cm/sec for controls, P < .001 for all comparisons). However, LV filling pressure was significantly elevated only in those subjects with manifest plaque disease whether in the absence or presence of increased IMT (E/Ea 8.4 and 8.7, respectively, vs 7.6, P < .001 both comparisons). Neither LV volumes nor LV EF were influenced by increased IMT, presence of plaque disease, or both. The effect of plaque burden severity on LV longitudinal function, LV EF, and LV filling pressure was assessed in a separate analysis ( Figures 2 and 3 ). Severe plaque burden was associated with significantly attenuated Sa velocity compared with subjects without plaque disease. The presence of plaque disease was associated with reduced Ea velocity and increased LV filling pressure in a dose-dependent manner (analysis of variance P < .001). As with longitudinal systolic function, LV EF was reduced only in those individuals with severe plaque burdens, compared with controls (59.0% vs 62.1%, P < .001). Increasing plaque burden was associated with greater LV end-systolic volume index (analysis of variance P = .01) and with greater LV end-diastolic volume index (analysis of variance P = .008).
Adjusted Analyses
To assess whether independent relationships existed between the degree of IMT and LV volumes or LV systolic function, multiple linear regression analyses were performed ( Table 4 ). In model 1, IMT was independently associated with reduced Ea velocity ( P = .001) and increased E/Ea ratio ( P = .04) but not with LV volumes, LV EF, or Sa velocity. After entering the presence of carotid plaque disease into the second model, IMT remained independently associated with Ea velocity, but the relationship with E/Ea ratio became weaker and borderline nonsignificant ( P = .05). Further adjusted analyses were performed to assess whether the severity of plaque disease was related with LV volumes and LV systolic function ( Table 5 ). Using analysis of covariance, LV EF and Sa velocity were both attenuated only in individuals with evidence of severe plaque burden, which reached the α value for significance after Bonferroni correction for multiple comparisons of P = .02. LV end-systolic volume index remained significantly greater in those with severe burden of plaque disease, whereas LV end-diastolic volume index was no longer related to plaque burden. Ea velocity was also significantly impaired in subjects with mild or severe plaque burden ( P = .006 and P = .01, respectively) compared with subjects with no plaque disease. Although the E/Ea ratio increased with increasing plaque burden, the effect did not reach the adjusted statistical significance until a severe burden of plaque disease was present.
