|Home | About | Journals | Submit | Contact Us | Français|
Low circulating levels of high density lipoprotein cholesterol (HDL-C) are associated with increased risk for cardiovascular events. HDL-C has a variety of poorly understood atheroprotective effects, including altering lipid metabolism and reducing inflammation. Increased arterial stiffness is an important predictor of subsequent cardiovascular risk. Therefore, in the current study, we sought to determine whether HDL-C levels are associated with carotid arterial stiffness. In addition we examined potential correlates of this association such as inflammatory factors, cardio-respiratory fitness and body fat percentage.
Carotid artery β stiffness was measured by ultrasound in 47 (23 yrs old) healthy pre-hypertensive men. Low HDL-C was defined as <1.0 mmol/L. Body fat was measured by air displacement plethysmograpy. Cardio-respiratory fitness was measured using a maximal exercise test with metabolic gas analysis and inflammatory markers consisted of C-reactive protein (CRP), white blood cell (WBC) count, and absolute neutrophil count.
Men with low HDL-C had significantly higher carotid artery stiffness, CRP, WBC count, neutrophil count, body fat, fasting glucose and lower cardio-respiratory fitness (p<0.05). Co-varying for cardio-respiratory fitness, % body fat, and glucose had no effect on group differences in carotid artery stiffness. Co-varying for inflammatory markers resulted in groups having similar carotid artery stiffness.
Pre-hypertensive men with low HDL-C have higher carotid artery stiffness when compared with those with higher HDL-C. The detrimental effects of low HDL-C on large artery stiffness in pre-hypertensive men may be mediated by inflammation and not by cardio-respiratory fitness or body fat levels.
Low levels of high density lipoprotein cholesterol (HDL-C) are associated with higher CVD risk 1-4. In vitro examinations note that HDL-C has a favorable effect on the vasculature and in vivo examinations have confirmed an association between HDL-C and vascular function 5-7. Vascular stiffness is now acknowledged as an independent risk factor for future cardiovascular morbidity and mortality 8,9. Pre-hypertensive individuals demonstrate marked vascular stiffening 10, a known antecedent to the development of essential and isolated systolic hypertension 11,12. Although several correlates of vascular structure and function in pre-hypertensives have been examined 13, the influence of blood lipids, markers of inflammation, cardio-respiratory fitness and body fat percentage on arterial stiffness in this cohort remains poorly understood. Hence the primary purpose of this study was to compare carotid artery stiffness in young, otherwise healthy, pre-hypertensive men with low versus higher HDL-C. A secondary purpose was to examine the influences of inflammation, cardio-respiratory fitness, and % body fat on this association.
Forty-seven young healthy men volunteered for this study. All subjects were free of cardiovascular, metabolic, renal, or respiratory disease and none smoked. Subjects were considered pre-hypertensive if systolic BP/ diastolic BP was 120 to 139/80 to 89 mmHg. Subjects were not diabetic (fasting blood glucose < 7 mmol/L) or hypercholesterolemic (total cholesterol > 6.2 mmol/L) and all were not taking medications of any kind. Family history of hypertension was recorded by questionnaire. All subjects were recruited from a university student population and none were engaged in a formal exercise training regimen. Subjects were separated into two groups using a HDL cut-point of 1.0 mmol/L. This cut-point was chosen based on the National Cholesterol Education Program Adult Treatment Panel III definition of “low” HDL cholesterol 14. All subjects gave written consent and this study was approved by the Institutional Review Board.
Subjects were tested on two separate occasions. The first visit consisted of a fasting blood draw followed by measures of body composition. During the second visit, subjects were required to rest in the supine position for a minimum of ten minutes, after which time vascular measures were conducted. All subjects were at least 3-hours post prandial and did not exercise or consume caffeine or alcohol for 24-hours prior to testing.
A subset of subjects reported back to the lab for a third visit for repeatability assessment. Vascular measures were carried in out in 34 men at the same time of day to reduce influence of diurnal variation.
Brachial blood pressure (BP) was measured in the supine position using an automated oscillometric cuff following established guidelines 15.
Carotid artery diameter was measured by ultrasonography (SSD-5500, Aloka, Tokyo, Japan). The cephalic portion of carotid artery was imaged in a longitudinal section, 1-2 cm proximal to the bifurcation, using a high frequency (7.5 MHz) linear array probe. Carotid artery pressure waveforms were attained using applanation tonometry (Millar Instruments, Houston, TX) and calibrated against brachial mean arterial and diastolic pressure. This technique has been shown to record a pressure wave with harmonic content that does not differ from that of an intra-arterially recorded wave 16. Heart rate (HR) was derived from ECG with a single lead CM5 configuration. β-stiffness index (β) was calculated as a means of adjusting arterial compliance for changes in distending pressure as follows:
where D1 and D0 are the maximum (systolic) and minimum (diastolic) diameters, and P1 and P0 are the highest (systolic) and lowest (diastolic) carotid pressures. The pressure-strain elastic modulus (epsilon, Ep) of the vessel, as a measure of local arterial elasticity, was calculated as follows:
Additionally, a one-point pulse wave velocity was calculated from the stiffness parameter, previously described in detail, as PWV = (β P/2ρ)1/2 where ρ is blood density (assumed constant). We have previously demonstrated high measurement repeatability in our laboratory. The ICC for measures of carotid stiffness in a subset of subjects from this investigation, collected on two separate days, was 0.86-0.94 (β-stiffness ICC = 0.94, ε ICC = 0.90, PWV ICC = 0.86).
Peak oxygen consumption (VO2 peak) was assessed using a graded cycle ergometry protocol until volitional fatigue. Expired gases were analyzed using a Quark b2 breath-by-breath metabolic system (Cosmed, Rome Italy).
Body composition (% body fat) was determined using whole body air displacement plethysmography (Bod Pod, Life Measurement Inc., Concord CA). Height and weight was measured using a stadiometer (to the nearest 0.5 cm) and a beam balance platform scale, respectively. Body mass index (BMI) was calculated as weight (kg) divided by height (m) squared.
All blood draws were carried out first thing in the morning with subjects in a fasted state. Fasting glucose was assessed via an oxygen rate method using a Beckman Coulter oxygen electrode (Beckman Coulter, Villapointe, France). Total cholesterol, HDL-C and triglycerides were measured using enzymatic techniques. LDL-C was calculated using the Friedewald formula. White blood cell (WBC) count and absolute neutrophil count were measured using a quantitative automated hematology analyzer (Sysmex XE-2100, Sysmex Corp, Kobe, Japan). Circulating levels of high sensitivity C-reactive protein (hsCRP) was measured by ELISA using a commercially available kit (Diagnostic Automation Inc., Calabasas, CA). Samples with absorbance values exceeding the standard curve were re-run on a separate plate following appropriate dilution. The intra-class correlation between two C-reactive protein measurements during the same study visit for all subjects was 0.99. The intra-class correlation in a subset of subjects collected on 2 separate days was 0.90. The intra-assay and inter-assay CV is 2.3-7.5% and 2.5-4.1% respectively.
Glomerular filtration rate (GFR) was estimated from serum creatinine measurements. The following formula, developed from the Modification of Diet in Renal Disease Study was used to estimate GFR: GFR [ml·min−1 x (1.73 m2)] = 186 x (sCR)−1.154 * (age)−0.203 x (1.210 if African American).
Group differences were then assessed using analysis of variance. Analysis of covariance was performed to examine group differences after adjusting for potential confounders (body fat, VO2peak, fasting blood glucose, LDL-C, triglycerides, CRP, WBC). Since neutrophils are a subtype of WBC, we did not co-vary for absolute neutrophil count due to issues of co-linearity. Chi-square tests were used to compare categorical variables (family history of hypertension). Associations between variables of interest were assessed with univariate and partial correlations. Values are mean ± SD. Statistical significance was set at p<0.05. Statistical analyses were performed using SPSS 12.0 (SPSS, Chicago, IL).
There was a significant negative association between HDL-C and carotid beta-stiffness, PWV, and elastic modulus (Table 2). According to partial correlations, co-varying for body fat, cardi-orespiratory fitness, LDL-C, triglycerides and glucose had no effect on the association between HDL-C and carotid beta-stiffness (−0.539, p<0.05), PWV (−0.537, p<0.05), or elastic modulus (−0.507, p<0.05). However, co-varying for CRP and WBC abolished the association between HDL-C and carotid beta-stiffness (−0.207, p>0.05), PWV (−0.217, p>0.05), and elastic modulus (−0.209, p>0.05).
Subjects with HDL-C < 1.0 mmol/L and those with HDL-C ≥; 1.0 mmol/L did not differ in age, height, systolic BP, diastolic BP, total cholesterol, LDL cholesterol, triglycerides and family history of hypertension (Table 1). The group with “low” HDL-C had significantly lower HDL-C (observed power = 1.0, ηp2 = 0.46, F-statistic = 37.5) and cardio-respiratory fitness (observed power = 0.52, ηp2 = 0.09, F-statistic = 4.2). The group with “low” HDL-C also had significantly greater body weight (observed power = 0.56, ηp2 = 0.09, F-statistic = 4.7), body mass index (observed power = 0.85, ηp2 = 0.17, F-statistic = 9.3), body fat (observed power = 0.85, ηp2 = 0.17, F-statistic = 9.3), fasting blood glucose (observed power = 0.63, ηp2 = 0.11, F-statistic = 5.5), white blood cell count (observed power = 0.77, ηp2 = 0.14, F-statistic = 7.6), absolute neutrophil count (observed power = 0.87, ηp2 = 0.19, F-statistic = 9.9) and C-reactive protein (observed power = 0.96, ηp2 = 0.25, F-statistic = 14.4).
Subjects with “low” HDL-C had significantly higher carotid artery β-stiffness (observed power = 0.86, ηp2 = 0.18, F-statistic = 9.8; Figure 1a). When ANCOVA was performed with cardio-respiratory fitness, body fat, LDL-C, triglycerides and fasting glucose entered as co-variates, adjusted means remained significantly different (4.7 vs 3.7, p<0.05). When ANCOVA was performed with CRP and WBC count entered as co-variates, adjusted means were no longer different (4.3 vs 4.0, p>0.05).
Subjects with “low” HDL-C had significantly higher carotid artery pulse wave velocity (observed power = 0.82, ηp2 = 0.16, F-statistic = 8.6; Figure 1b). When ANCOVA was performed with cardio-respiratory fitness, body fat, LDL-C, triglycerides and fasting glucose entered as co-variates, adjusted means remained significantly different (4.6 vs 4.2 m/s, p<0.05). When ANCOVA was performed with CRP and WBC count entered as co-variates, adjusted means were no longer different (4.5 versus 4.3 m/s, p>0.05).
Subjects with “low” HDL-C had significantly higher carotid artery elastic modulus (observed power = 0.72, ηp2 = 0.13, F-statistic = 6.7; Figure 1c). When ANCOVA was performed with cardio-respiratory fitness, body fat, LDL-C, triglycerides and fasting glucose entered as co-variates, adjusted means remained significantly different (61.2 vs 48.9 kPa, p<0.05). When ANCOVA was performed with CRP and WBC count entered as co-variates, adjusted means were no longer different (56.1 vs 52.7 kPa, p>0.05).
Current estimates suggest that 70 million Americans aged 20 years or older are pre-hypertensive 13,17. Pre-hypertensive individuals are eleven times more likely to develop hypertension 18. Subsequently, cardiovascular risk is significantly elevated in this population 19,20. In the present study, there was an inverse association between HDL-C and carotid artery stiffness. Pre-hypertensive men with HDL-C < 1.0 mmol/L had higher carotid artery stiffness when compared to their pre-hypertensive peers with HDL-C > 1.0 mmol/L despite equivalent total cholesterol and LDL-C. This relationship was not affected by cardio-respiratory fitness, body fat, BMI, LDL-C, triglycerides and fasting glucose. Interestingly, when taking into account levels of CRP and WBC count, group differences in carotid artery stiffness no longer prevailed. Thus, the detrimental effect of low HDL-C on the vasculature may be related to a heightened state of inflammation.
Low HDL-C may be prevalent in 66% of the population 21. It is commonly accepted that low levels of HDL-C are associated with higher CVD risk, independent of LDL-C level. Our findings extend the detrimental effect of low HDL-C to the vasculature of young apparently healthy pre-hypertensive men by acknowledging an association between low HDL-C and large artery stiffness, a known independent CVD risk factor and antecedent to future development of hypertension.
Elevated C-reactive protein (CRP) and white blood cell (WBC) count, markers of subclinical low-grade inflammation, have been shown to predict CVD events and contribute to vascular dysfunction 22-28. CRP and WBC count have also each been shown to be independently associated with the risk of developing hypertension 29-33. Pre-hypertensive populations have higher CRP and WBC count compared with normotensive subjects, suggesting that pre-hypertension might be a pro-inflammatory condition 34,35.
Both acute and chronic inflammatory conditions are associated with increased arterial stiffness 26,36-40. CRP has been shown to induce production of other inflammatory cytokines, decrease tetrahydrobiopeterin (a vital cofactor required for nitric oxide production), downregulate and/or uncouple endothelial nitric oxide synthase, decrease prostacyclin release, increase gene expression/activity of the elastolytic enzyme matrix metalloproteinase-9, and upregulate endothelin-1 41-44. This can reduce vasodilation, enhance vasoconstriction, increase collagen formation, and increase elastin breakdown. WBCs play a role in vascular injury through proteolytic enzymes and plugging of microvessels, which may create rheological abnormalities leading to increased vascular resistance 45. Neutrophil rolling, recruitment and extravasion, considered the initial steps in the vascular inflammatory cascade, leads to subsequent neutrophil-mediated endothelial cell damage and lysis. In the present study, pre-hypertensive men with low HDL-C had higher WBC count, higher neutrophil count and higher CRP, all suggestive of a heightened inflammatory state. The summative effect may produce a structurally and functionally stiffer vessel. Co-varying for CRP and WBC count in the present study abolished the group differences in carotid artery stiffness. Our findings suggest that the damaging effect of low HDL-C on large artery properties in pre-hypertensive men may be mediated by its inflammatory effects.
This is the first study to specifically examine the influence of cardio-respiratory fitness and body fat levels on the relation between HDL-C on large artery properties in men with pre-hypertension. It has been suggested that the detrimental effect of low HDL-C on the vasculature and inflammatory status may be secondary to levels of cardio-respiratory fitness and/or body fatness 5. Both low fitness and high body fat are associated with low HDL-C, heightened inflammation and concomitant vascular dysfunction 46-49. In the present study, pre-hypertensive men with low HDL-C had slightly lower cardio-respiratory fitness and higher body fat than pre-hypertensive men with high HDL-C. However, these factors do not appear responsible for the higher carotid artery stiffness in men with low HDL-C as adjusting for these factors had no effect on group differences in carotid artery stiffness. Thus, the negative effects of HDL cholesterol on carotid artery stiffness may not be due to the direct influence of cardio-respiratory fitness or body fat.
We noted a clustering of risk factors in men with low HDL-C that may also categorize them as having pre-metabolic syndrome. Men with low HDL-C had higher BMI/body fat, higher glucose, higher LDL-C and higher triglycerides. However, adjusting for these risk factors had no effect on group differences in carotid artery stiffness. Although these variables were not directly associated with measures of carotid artery stiffness, they were associated with inflammatory markers. Thus it remains plausible that pre-metabolic syndrome risk factors indirectly modulate vascular stiffness via inflammation (i.e. release of inflammatory cytokines from adipose tissue stores). It has been shown that inflammation reduces HDL-C via several mechanisms including increased activity of endothelial lipase, increased activity of soluble phospholipase A2, and replacement of apoA-1 in HDL-C with serum amyloid A 50. Indeed, when in a state of systemic inflammation, HDL-C may even become pro-oxidant 51. Thus, an alternative interpretation of the present findings may be that the high inflammatory state in men with pre-hypertension and high body fat instigated low HDL-C levels.
The sample size used in the present investigation is small. Findings may be due to other factors associated with low HDL-C that were not measured in this study (i.e. insulin resistance). Given the cross-sectional nature of our study, we cannot determine causality and statistically adjusting for group differences in various risk factors does not prove a definitive physiologic link. These findings are restricted to young, male cohort and may not extrapolated to older individuals or women.
This study was supported by predoctoral student research grants from the American Heart Association and the American College of Sports Medicine.