M. H. Friedman Biomedical Engineering Center, The Ohio State University, Columbus, OH 43210
C. B. Bargeron D. D. Duncan Applied Physics Laboratory, The Johns Hopkins University, Laurel, MD 20723
G. M. Hutchins School of Medicine, The Johns Hopkins University, Baltimore, MD 21218
F. F. Mark Applied Physics Laboratory, The Johns Hopkins University, Laurel, MD 20723
Effects of Arterial Compliance and Non-Newtonian Rheology on Correlations Between Intimal Thickness and Wall Shear A minimally diseased (mean intimal thickness = 56 jim) human aortic bifurcation was replicated in rigid and compliant flow-through casts. Both casts were perfused with physiological flow waves having the same Reynolds and unsteadiness numbers; the pulse pressure in the compliant cast produced radial strains similar to those expected from post-mortem measurements of the compliance of the original tissue. The compliant cast was perfused with a Newtonian fluid and one whose rheology was closer to that of blood. Wall shear rate histories were estimated from near-wall velocities obtained by laser Doppler velocimetry at identical sites in both casts. Intimal thickness was measured at corresponding sites in the original vessel and linear regressions were performed between these thicknesses and several normalized shear rate measures obtained from the histories. The correlations showed a positive slope—that is, the intima was thicker at sites exposed to higher shear rates—consistent with earlier results for relatively healthy vessels, but their significance was often poor. There was no significant effect of either model compliance or fluid rheology on the slopes of the correlations of intimal thickness against any normalized shear rate measure.
Introduction There is considerable indirect evidence that vascular fluid dynamics plays a role in the development and progression of arterial disease (e.g., Schettler et al., 1983; Yoshida et al., 1988). This role presumably reflects the response of the vessel wall to the adjacent flow field. To illuminate this response, we seek relationships between fluid mechanical variables at points along the margin of the arterial lumen and the morphology at corresponding sites in the arterial wall. The primary morphologic variable that we have used to date has been intimal thickness, because there appears to be a correspondence between the sites of early intimal thickening in relatively healthy humans and the locations of advanced disease later in life; this suggests that intimal thickening may be an early phase of the atherosclerotic process. Our procedure has been to prepare flow-through casts of human vascular segments obtained at autopsy; both aortic bifurcations and coronary artery segments have been used. These replicas are then perfused with physiologically realistic flows. Fluid velocities are measured at multiple sites near the walls of these replicas by laser Doppler velocimetry (LDV). The time-dependent velocities are then processed to estimate wall shear rates, and correlations are sought between measures derived from the wall shear histories and the corresponding Contributed by the Bioengineering Division for publication in the JOURNAL OF BIOMECHANICAI ENGINEERING. Manuscript received by the Bioengineering Division December 16, 1991; revised manuscript received May 6, 1992.
intimal thicknesses (Friedman et al., 1981). The slopes of these correlations appear to depend on the mean intimal thickness of the vessel (Friedman et al., 1986): for aortic bifurcations in which the mean intimal thickness was less than about 400 /xm, the intima was thicker where the shear was relatively high, while for a specimen whose mean intimal thickness was larger than this amount, the relation between thickness and shear rate was reversed. The inverse relation between thickness and shear rate in vessels with somewhat thicker intimas was also seen for a human coronary artery segment (Friedman et al., 1987) and in a similar investigation by Ku et al. (1985), who obtained their hemodynamic data in a model based on one by Bharadvaj et al. (1982) that typified the human carotid bifurcation. From these data, summarized in Fig. 1, it was inferred (Friedman et al., 1986) that, probably because of competing shear-dependent processes in the arterial wall, the intima at sites exposed to relatively high or unidirectional shears thickened first, but with the passage of time, the greatest thicknesses were ultimately achieved at sites exposed to lower or more oscillatory shear environments. These conclusions formed the basis for a subsequent model of the dependence of intimal thickness and thickening rate on the flow-generated stress at the vessel wall (Friedman, 1989). All of the fluid dynamic data used to draw the foregoing inferences and to parameterize the model were obtained from rigid replicas perfused with a Newtonian working fluid. It is
Journal of Biomechanicai Engineering
AUGUST 1992, Vol. 114/317 -
Copyright © 1992 by ASME Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 12/13/2013 Terms of Use: http://asme.org/terms
1
A82,
i AB-10
*
*B" T AB-12 §
I
f ii t»-AB-15
i
AB-ST
AB-8
M
AEf-20 T
Pressure
1
"
1'
1AB-16
C-2 (coronary) Ku et at (carotid) T
"
-
0 q = ms m f l >
"
AB-5
+ b
-
-
i
"
- Flow
\ 223 E E Mean 126 98
^ - J\ i> H Y^nf ;W
/
0
w
\
I
1 2 Time (s)
+74
Y~-c
+ 12
i Xf
0 -13.8
|
3
Fig. 2 Inlet flow and pressure waves for Newtonian flow in the compliant cast
-
the range of published values for human arteries (Langewouters et al., 1984). The tissue was retained for subsequent morphometry. Fig. 1 Slopes (m) of linear regressions of intimal thickness (
C. B. Bargeron D. D. Duncan Applied Physics Laboratory, The Johns Hopkins University, Laurel, MD 20723
G. M. Hutchins School of Medicine, The Johns Hopkins University, Baltimore, MD 21218
F. F. Mark Applied Physics Laboratory, The Johns Hopkins University, Laurel, MD 20723
Effects of Arterial Compliance and Non-Newtonian Rheology on Correlations Between Intimal Thickness and Wall Shear A minimally diseased (mean intimal thickness = 56 jim) human aortic bifurcation was replicated in rigid and compliant flow-through casts. Both casts were perfused with physiological flow waves having the same Reynolds and unsteadiness numbers; the pulse pressure in the compliant cast produced radial strains similar to those expected from post-mortem measurements of the compliance of the original tissue. The compliant cast was perfused with a Newtonian fluid and one whose rheology was closer to that of blood. Wall shear rate histories were estimated from near-wall velocities obtained by laser Doppler velocimetry at identical sites in both casts. Intimal thickness was measured at corresponding sites in the original vessel and linear regressions were performed between these thicknesses and several normalized shear rate measures obtained from the histories. The correlations showed a positive slope—that is, the intima was thicker at sites exposed to higher shear rates—consistent with earlier results for relatively healthy vessels, but their significance was often poor. There was no significant effect of either model compliance or fluid rheology on the slopes of the correlations of intimal thickness against any normalized shear rate measure.
Introduction There is considerable indirect evidence that vascular fluid dynamics plays a role in the development and progression of arterial disease (e.g., Schettler et al., 1983; Yoshida et al., 1988). This role presumably reflects the response of the vessel wall to the adjacent flow field. To illuminate this response, we seek relationships between fluid mechanical variables at points along the margin of the arterial lumen and the morphology at corresponding sites in the arterial wall. The primary morphologic variable that we have used to date has been intimal thickness, because there appears to be a correspondence between the sites of early intimal thickening in relatively healthy humans and the locations of advanced disease later in life; this suggests that intimal thickening may be an early phase of the atherosclerotic process. Our procedure has been to prepare flow-through casts of human vascular segments obtained at autopsy; both aortic bifurcations and coronary artery segments have been used. These replicas are then perfused with physiologically realistic flows. Fluid velocities are measured at multiple sites near the walls of these replicas by laser Doppler velocimetry (LDV). The time-dependent velocities are then processed to estimate wall shear rates, and correlations are sought between measures derived from the wall shear histories and the corresponding Contributed by the Bioengineering Division for publication in the JOURNAL OF BIOMECHANICAI ENGINEERING. Manuscript received by the Bioengineering Division December 16, 1991; revised manuscript received May 6, 1992.
intimal thicknesses (Friedman et al., 1981). The slopes of these correlations appear to depend on the mean intimal thickness of the vessel (Friedman et al., 1986): for aortic bifurcations in which the mean intimal thickness was less than about 400 /xm, the intima was thicker where the shear was relatively high, while for a specimen whose mean intimal thickness was larger than this amount, the relation between thickness and shear rate was reversed. The inverse relation between thickness and shear rate in vessels with somewhat thicker intimas was also seen for a human coronary artery segment (Friedman et al., 1987) and in a similar investigation by Ku et al. (1985), who obtained their hemodynamic data in a model based on one by Bharadvaj et al. (1982) that typified the human carotid bifurcation. From these data, summarized in Fig. 1, it was inferred (Friedman et al., 1986) that, probably because of competing shear-dependent processes in the arterial wall, the intima at sites exposed to relatively high or unidirectional shears thickened first, but with the passage of time, the greatest thicknesses were ultimately achieved at sites exposed to lower or more oscillatory shear environments. These conclusions formed the basis for a subsequent model of the dependence of intimal thickness and thickening rate on the flow-generated stress at the vessel wall (Friedman, 1989). All of the fluid dynamic data used to draw the foregoing inferences and to parameterize the model were obtained from rigid replicas perfused with a Newtonian working fluid. It is
Journal of Biomechanicai Engineering
AUGUST 1992, Vol. 114/317 -
Copyright © 1992 by ASME Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 12/13/2013 Terms of Use: http://asme.org/terms
1
A82,
i AB-10
*
*B" T AB-12 §
I
f ii t»-AB-15
i
AB-ST
AB-8
M
AEf-20 T
Pressure
1
"
1'
1AB-16
C-2 (coronary) Ku et at (carotid) T
"
-
0 q = ms m f l >
"
AB-5
+ b
-
-
i
"
- Flow
\ 223 E E Mean 126 98
^ - J\ i> H Y^nf ;W
/
0
w
\
I
1 2 Time (s)
+74
Y~-c
+ 12
i Xf
0 -13.8
|
3
Fig. 2 Inlet flow and pressure waves for Newtonian flow in the compliant cast
-
the range of published values for human arteries (Langewouters et al., 1984). The tissue was retained for subsequent morphometry. Fig. 1 Slopes (m) of linear regressions of intimal thickness (
