Peak Shear and Peak Flow Mediated Dilation: A Time Course Relationship

Abstract

Purpose

This study explored the temporal relationship between peak shear stress (Shear) and flow-mediated dilation (FMD) in children and adults.

Methods

Shear and brachial artery diameter were tracked following reactive hyperemia in 122 children and 350 adults using ultrasound imaging.

Results

Peak Shear, Shear area under the curve (Shear_AUC), and Peak FMD were significantly larger in children than adults. The time to peak Shear (Shear_TTP) and time to peak FMD (FMD_TTP) were significantly slower in children while there was no significant difference in time from Shear_TTP to FMD_TTP between children and adults.

Conclusion

Our findings demonstrate that children, compared to adults, have a slower shear stimulus and FMD response but that the time interval separating these events is similar. These differences in timing could be due to changes in vascular dynamics with age, including reduced smooth muscle cell responsiveness, and other factors. Despite differences in timing, the interval from peak Shear to peak FMD was similar in children and adults.

Introduction

Ultrasound imaging of the brachial artery following reactive hyperemia, or flow mediated dilation (FMD), is the most widely accepted non-invasive technique for measuring endothelial function^1. Furthermore, endothelial dysfunction is known to represent atherosclerotic risk in both adults² and children³. A number of other variables, such as peak shear stress can also be measured during FMD studies and have been found to play a pivotal role in the progression and mitigation of cardiovascular disease⁴ (CVD). It is possible that these other variables measured during FMD studies may prove to be biomarkers of cardiovascular disease. Shear stress is also recognized as the main stimulus that elicits the dilation of the brachial artery that is observed during FMD studies^5. Shear stress is commonly estimated during FMD studies as the shear rate⁶. This shear stimulus has potent effects on endothelial cells including initiating paracrine signaling and changing morphological features⁷. The vast majority of studies examining the relationship of shear stress and FMD have focused on the magnitude of brachial artery dilation in response to reactive hyperemia^8–12, however few have explored the time course of peak shear and peak dilation of the brachial artery. Of the few studies examining the time course of FMD and related components only one included children¹³ while two have examined young adults^14,15.

Thijssen et al.¹³ investigated the relationship of arterial shear stress and the magnitude of FMD in children and adults and found significant differences in shear rate area-under-the-curve (AUC) while finding no differences in FMD time to peak. Black et al.¹⁴ sought to explain the importance of measuring the time course of FMD. They found time to peak diameter was faster in young versus older adults. Stoner et al.¹⁵ displayed peak and time integrated shear rates independently predict FMD in young adult males which highlights the importance to consider time-course variables during FMD studies. Despite these findings, the temporal association of shear stress and FMD still remains unclear. In addition, the effect of age on these relationships has not been fully investigated.

To our knowledge, no studies have examined the time-course relationship between arterial shear stress and FMD in a large sample of children and adults. It is important to characterize how children and adults differ in order to better understand the effects of aging and CVD risk factors on the vasculature. Therefore, the purpose of the present study was to examine this temporal relationship between time to peak shear stress (Shear_TTP) and time to peak FMD (FMD_TTP) in children and adults.

METHODS

Study population

A total of one hundred twenty two healthy children (mean age 10.7 ± 0.2) range 6–15 years old; 48 female, 74 male) and three hundred fifty healthy adults (age 39.2 ± 0.1 range 33–57 years old; 195 female, 155 male) from the community who participated in a longitudinal cardiovascular risk study¹⁶ were used in this analysis. Smokers as well as those taking any medications were excluded from analysis to reduce these potential confounding factors. The University of Minnesota Institutional Review approved the study design and protocol. All adult participants gave written consent; parental consent and child assent was obtained from the child/adolescent participants. All information regarding participant’s information was protected in accordance with the Health Insurance and Accountability Act (HIPAA).

Physical and Vascular Assessments

Height was measured with a wall-mounted stadiometer (Ayrton, Model S100, Prior Lake, MN, USA), and weight was measured using an electronic scale (ST Scale-Tronix,(White Plains, NY, USA). Body mass index (BMI) was calculated as the weight (kg) divided by the height (meters) squared. Waist-hip ratio (WHR) was calculated as the ratio of the smallest circumference (cm) between the last palpable rib and iliac crest to the largest circumference over the gluteus maximus.

Seated blood pressure was measured two separate times with a random-zero sphygmomanometer on the right arm. The average of two systolic blood pressure (SBP) and the fifth phase Korotkoff diastolic blood pressure (DBP) measurements were analyzed and recorded.

Fasting lipid profile (total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglyceride (TG), and insulin were measured using standard procedures at the Fairview Diagnostic Laboratories at the Fairview-University Medical Center (Minneapolis, MN), a Centers for Disease Control and Prevention–certified laboratory.

Vascular testing was conducted in the Clinical and Translational Science Institute (CTSI) of the University of Minnesota. Vascular structure and function were measured in a quiet, temperature-controlled environment (22–23 °C). Vascular images were obtained using a conventional ultrasound scanner (Acuson, Sequoia 512, Siemens Medical Solutions USA, Inc. Mountain View, CA) with an 8.0–15mHz linear array transducer held in place by a stereotactic arm (2–10 cm proximal to the antecubital fossa). This system was interfaced with a standard personal computer equipped with a data acquisition card for attainment of radio frequency ultrasound signals from the scanner. Arterial occlusion was accomplished by placing an occlusion cuff distal to the elbow and inflated to 200 mmHg for 5-min.¹⁷ All arterial images were triggered and captured at the R wave of the electrocardiogram (end-diastolic diameter) beginning 20-sec. pre cuff release and continued for 180-sec. post cuff release (total imaging time = 200-sec.). Continuous flow measurements were achieved during brachial diameter acquisition by placing a pulse-wave Doppler gate near the vessel ROI with an ionization angle of 60 degrees. The real time images were digitized and stored on a personal computer for later off-line analysis using electronic wall and pulse wave tracking software (Vascular Research Tools 5, Medical Imaging Application, LLC, Iowa City, IA, USA). All image files were averaged over a 10 sec period and peak dilation during the each study was defined as the greatest percent change from resting baseline brachial artery diameter. Peak blood flow (Flow_peak) was used to calculate shear rate (Flow_peak divided by brachial artery diameter). A trained sonographer performed all digital image analysis. FMD repeatability studies in our laboratory found the coefficient of variation was 11.1%, demonstrating good reproducibility when measured at least 7 days apart. Figure 1 displays a graphical representation of the key vascular parameters over the time course of the measurement.

Figure 1.

Graphical representation of the key vascular parameters over the time course of the measurement in a child. Flow mediated dilation (FMD) is represented by open circles. Shear rate (Shear) is represented by closed circles.

Statistical analysis

Data were analyzed using open source statistical package “R” (version 2.15.1, Vienna Austria) and recorded as mean ± standard error (SEM). A two-sided alpha level of 0.05 was used to define statistical significance. Variables showing a non-normal distribution were log-transformed prior to analysis. T-tests were performed to compare baseline differences between age groups.

Main study variables (age and gender) and all independent variables (vascular and blood factors) were added to regression models in order to test correlates while adjusting for colinearity. Additionally, interaction statements for baseline diameter and age and gender were added to the model for proper adjustment. Models for each age group were constructed for each dependent variable (FMD_TTP and Shear_TTP). Stepwise multiple regression analysis was implemented to indicate significant associations. The initial P-value cut-off for eliminating independent variables from the model was 0.50. As further reductions were considered a lower P-value was used until model fit was significantly compromised.

RESULTS

Table 1 shows the physical characteristics for both the adults and children. As expected, adults were significantly taller and weighed more than children (P=0.0001). Similarly, adults had a significantly (P=0.0001) greater WHR, BMI, SBP and DBP. Fasting insulin, fasting glucose, total cholesterol, triglycerides, and LDL were significantly higher in adults compared to children while there was no difference in HDL levels (Table 2).

Table 1.

Physical characteristics

Variable	Group
	Children	Adults
n (Female, Male)	122 (48, 74)	350 (195, 155)
Age (years)	10.7 ± 0.2	39.2 ± 0.1
Waist-Hip Ratio	82.5 ± 0.5	88.7 ± 0.5
Weight (kg)	43.3 ± 1.6	86.0 ± 1.5
Height (cm)	145.8 ± 1.5	171.5 ± 1.0
Body mass index (kg/m2)	19.5 ± 0.4	29.1 ± 0.4
Systolic blood pressure (mmHg)	113 ± 1	125 ± 1
Diastolic blood pressure (mmHg)	58 ± 1	70 ± 1

Data are mean ± SEM.

Table 2.

Blood variables

	Group		p-value
Variable	Children	Adults
High-density lipoproteins (mmol/L)	0.53 ± 0.01	0.50 ± 0.01	0.1
Low-density lipoproteins (mmol/L)	0.91 ± 0.02	1.1 ± 0.02	< 0.0001
Triglycerides (mmol/L)	0.79 ± 0.03	1.40 ± 0.06	< 0.0001
Total cholesterol (mmol/L)	4.1 ± 0.06	4.8 ± 0.05	< 0.0001
Fasting glucose (mmol/L)	4.2 ± 0.07	5.7 ± 0.08	< 0.0001
Fasting insulin (pmol/L)	29.4(25.2, 34.2)	33.4(30.6, 37.2)	0.1

Data are mean ± SEM. P-value < 0.05 demonstrates significant differences between means.

Fasting Insulin was log transformed and values represent back transformed geometric mean (95% Confidence interval).

Table 3 displays vascular measures for adults and children. Peak FMD was significantly larger in children compared to adults. FMD_AUC was also significantly larger in children than adults. Shear rate AUC from cuff release to peak FMD (Shear_AUC) was greater in children. Shear_TTP was significantly slower in children than adults and FMD_TTP was also slower in children than adults. Interestingly the time from Shear_TTP to FMD_TTP was the same between children and adults respectively.

Table 3.

Vascular measures by group

Variable	Group
	Children	Adults	P-value
Baseline Diameter (mm)	3.07 ± 0.04	3.79 ± 0.04	< 0.0001
Peak FMD (% from baseline)	7.25 ± 0.28	6.47 ± 0.18	0.022
Peak Shear rate (sec−1)	314 ± 8.38	278 ± 4.92	0.0002
ShearAUC (sec−1)	13231 ± 531	9907 ± 237	< 0.0001
ShearAUC 0-peak flow (sec−1)	3231 ± 135	2569 ± 67	< 0.0001
FMDAUC (% sec−1)	715 ± 35.2	602 ± 20.5	0.008
FMD Time to Peak (sec)	57.2 (53.3, 61.4)	52.8 (50.5, 55.1)	0.027
Time to Peak Shear (sec)	11.8 (11.1, 12.6)	10.6 (10.2, 11.0)	0.004
FMDTTP-ShearTTP (sec)	42.9 (38.9, 47.4)	40.2 (38.0, 42.6)	0.27

Variables represented: flow mediated dilation (FMD), shear area under the curve (Shear_AUC), Shear_AUC from cuff deflation to peak flow (0-peakflow), FMD under the curve (FMD_AUC), FMD time to peak (FMD_TTP), Shear Time to peak (Shear_TTP). Baseline Diameter, Peak FMD, Shear_AUC, and FMD_AUC are mean ± SEM, FMD time to peak and Time to Peak shear values are back-transformed and represent geometric mean (95% CI), P-value < 0.05 demonstrates significant differences between means (two sided).

Regression models

Upon model reduction, the emergent associated variables to FMD_TTP in the adult model were gender (β=1.23, P=0.007) and baseline diameter (β=0.42, P=0.02). A test for interaction yielded a significant (P=0.001) interaction effect for gender and baseline diameter. The reduced FMD_TTP model for children revealed BMI (β=0.03, P=0.009) and gender (β=−0.18, P=0.009) to be significantly associated with FMD_TTP. No significant predictors were found in the reduced adult Shear_TTP model. The reduced child Shear_TTP model indicated age (β=−0.26, P=0.02) to be the only significant predictor upon adjustment, however an interaction of age and baseline diameter was evident (β=0.07, P=0.04). Additionally, no significant predictors were related to the difference between FMD_TTP and Shear_TTP in neither the child nor adult models.

DISCUSSION

In the present study, children displayed slower Shear_TTP and FMD_TTP compared to adults. However the time from peak stimulus (shear) and peak response (FMD) was similar in both children and adults. This observation demonstrates that once the peak stimulus is achieved a predictable time to peak FMD may be expected in younger and older individuals.

Thijssen et al.¹³ reported no significant difference in FMD_TTP between children and older adults, which differ from the present finding that FMD_TTP is significantly slower in children than adults. This finding may have been influenced by a smaller sample size compared to the present study or the fact that the Thijssen et al. (2009) adult population was older (mean age 58 yrs vs. 39 yrs. in the current study). However they found that peak FMD was greater in children than adults, which agree with our findings. Though their study focused on differing AUC timeframes, Thijssen et al.¹³ reported FMD_TTP, but they did not report Shear_TTP in absolute terms (only in relation to AUC) which makes comparison with the current study difficult.

Based on the regression models, baseline arterial diameter is the most common variable associated with the dependent variables (FMD_TTP and Shear_TTP). Arterial diameter is inversely related to FMD in adults¹⁸ and children.¹⁹ The interaction between baseline diameter and gender or age within both the child and adult FMD_TTP models are similar to results reported by others.^20–22 However, gender was significantly associated with FMD_TTP in both models: girls displayed faster FMD_TTP whereas adult females showed a slower FMD_TTP compared to males (reference group) when adjusted for age and baseline diameter. These differences could possibly be due to differences in smooth muscle function.^23–25 Additionally, since a smaller diameter is associated with a greater increase in blood flow velocity and shear rate ^18,19, it may possibly explain the slower time to peak in children.

BMI was also associated with FMD_TTP in children (β=0.03, P=0.009) despite the fact that the mean BMI was within the normal range (BMI = 19.5, 75^th percentile). Interestingly, BMI was not a significant predictor in the adult model even though the mean BMI was in the overweight range (BMI = 29.1). This may indicate, that like FMD ^19,26, FMD_TTP may be influenced by adiposity in children. However, since our data are cross-sectional in nature, this hypothesis should be further investigated.

The most novel finding of this study is that, even though children took longer to reach peak shear as well as peak FMD, there was no difference in the time from peak shear rate to peak FMD compared with adults. This observation demonstrates that once the brachial artery reaches peak hemodynamic stimulus via reactive hyperemia, there may be an expected time to maximal dilation regardless of age. The major implication of this finding is that deviation from this expected time course may indicate vascular abnormalities considering that the participants in this study were healthy despite the adult population being clinically overweight.

Since the clinical significance of FMD and Shear time-courses are yet to be determined, future studies examining FMD should not only report Shear_TTP and FMD_TTP, but also the difference between them (FMD_TTP–Shear_TTP). Furthermore, the time from Shear_TTP to FMD_TTP should be studied in groups of children and adults with known cardiovascular risk factors to test the hypothesis that this time course may change as a result of factors such as obesity and type 2 diabetes mellitus. Time course variables in relation to Tanner stages should also be studied to further explore the relationship between development and vascular dynamics.

Strengths of this study include the large sample size as well as highly developed techniques to capture both FMD and shear in children and adults. A limitation of the current study is lack of Tanner stage data. The ability to stratify children by pubertal development may have uncovered potential developmental effects on FMD and shear stress time courses. Additionally endothelium-independent dilation was not accounted for in the current study. Having this information would have helped understand possible smooth muscle differences due to aging.

In conclusion, this is the first study to demonstrate that the time course between Shear_TTP and FMD_TTP are similar in children and adults. However the manner in which children accomplish peak responses is different. Children exhibit significantly slower time to peak shear rate as well as slower time to peak FMD compared to adults. Age, gender, baseline brachial artery diameter, and BMI appear to influence the time course of FMD_TTP. Slowing of the peak dilation response could be due to changes in the vascular function with age, smooth muscle cell responsiveness, and other factors.