057 coronary endothelial shear stress profiling

Coronary Endothelial Shear Stress ProfilingIn-Vivo to Predict Progression of

Atherosclerosis and In-Stent Restenosis in Man

Peter H. Stone, M.D.Ahmet U. Coskun, Ph.D.Scott Kinlay, M.D., Ph.D., Maureen E. Clark, M.S.Milan Sonka, Ph.D.Andreas Wahle, Ph.D.,

Olusegun J. Ilegbusi, Ph.D.Yerem Yeghiazarians, M.D.Jeffrey J. Popma, M.D.Richard E. Kuntz, M.D., M.S.Charles L. Feldman, Sc.D.

Cardiovascular Division, Brigham & Women’s Hospital, Harvard Medical School;Department of Mechanical, Industrial and Manufacturing Engineering,

Northeastern University;Department of Electrical and Computer Engineering, University of Iowa

Abstract - 1

The focal and eccentric nature of CAD must be related to local hemodynamic factors. The endothelium is uniquely capable of controlling local arterial responses by transduction of hemodynamic shear stress. Low or reversed shear stress (< ~10 dynes/cm2) leads to plaque development and progression. Physiologic shear stress (~10 - 30 dynes/cm2) is vasculoprotective, maintaining normal vascular morphology. Increased shear stress(> ~ 30 dynes/cm2) promotes outward remodeling and platelet aggregation.

Characterization of shear stress along the coronary artery may allow for prediction of progression of atherosclerosis and vascular remodeling.

Abstract - 2Current methodologies cannot provide adequate information

concerning the micro-environment of the coronary arteries. We developed a unique system using intravascular ultrasound (IVUS), biplane coronary angiography, and measurements of coronary blood flow, to present the artery in accurate 3-D space, and to produce detailed characteristics of intravascular flow, ESS, and arterial wall and plaque morphology.

We observed that over 6 mo followup, areas of low ESS demonstrated plaque progression, areas of physiologic ESS remained quiescent, and areas of increased ESS developed outward remodeling.

The technology may be invaluable to study the impact of pharmacologic or device interventions on the natural history of coronary disease.

Fundamental Nature of the Problem

• Although all portions of the coronary arterial tree are exposed to the same systemic risk factors,

atherosclerosis is focal and eccentric• Each coronary artery has many different

obstructions in different “stages” of evolution:– There is not a “wave-front” of vulnerability

and consequent rupture.

Varying Degrees of CAD Lesion Severity in a Single Coronary Artery

Fundamental Nature of the Problem• Coronary atherosclerotic obstructions behave differently based on the

degree of luminal obstruction and morphology:– Lesions > 50-75% obstruction Angina Pectoris– Lesions < 50% obstruction Rupture,superimposed

thrombus, MI, death

These small, potentially lethal lesions are, therefore, These small, potentially lethal lesions are, therefore, “clinically silent” until they rupture.“clinically silent” until they rupture.

• It would be of enormous value to identify minor obstructions It would be of enormous value to identify minor obstructions which were progressing and/or evolving towards which were progressing and/or evolving towards “vulnerability” since they could be treated before rupture “vulnerability” since they could be treated before rupture occurred, thereby averting an acute coronary syndrome.occurred, thereby averting an acute coronary syndrome.

Nature of Progression of Atherosclerosis

• The only truly local phenomena which could lead to varying local vascular responses are endothelial shear stresses (ESS)

• Local ESS variations are critical:– Low ESS and disturbed flow (< 6-10 dynes/cm2)

• Causes atheroma; pro-thrombotic, pro-migration, pro-apoptosis– Physiologic shear stress and laminar flow (10-30 dynes/cm2)

• Vasculoprotective, anti-thrombotic, anti-migration, pro-survival– High shear stress and turbulent flow (> 30 dynes/cm2)

• Promotes platelet activation, thrombus formation, and probably plaque rupture

• Until now, Until now, in vivoin vivo determination of intracoronary flow velocity and determination of intracoronary flow velocity and endothelial shear stress has not been possible.endothelial shear stress has not been possible.

The Detrimental Effect of Low Shear Stress on Endothelial Structure and Function

Low shear stresses and disturbedlocal flow (< ~ 6 dynes/cm2)are atherogenic:

(Malek, et al. JAMA 1999; 282:2035)

• Cell proliferation, migration• Expression of vascular adhesion molecules, cytokines, mitogens

• Monocyte recruitment and activation

• Procoagulant and prothrombotic state• Local oxidation

Promotes:

The Effect of Physiologic Shear Stress onEndothelial Structure and Function

Physiologic shear stress (~15-50 dynes/cm2) isvasculoprotective:

(Malek, et al. JAMA 1999; 282:2035)

• Enhances endothelial quiescence - decreases proliferation

• Enhances vasodilation

• Enhances anti-oxidant status

• Enhances anti-coagulant and anti-thrombotic status

Overview of Intracoronary Flow Profiling System

Patient • Coronary angiography• Intracoronary ultrasound• Coronary flow (TIMI Frame Count)

Acquire image data

3D reconstructionof lumen, EEL, Plaque

Generation of grid for ComputationalFluid Dynamics

Numericalcomputation

Determination oflocal velocity vectorsand shear stress

Application of vascular data to

patient care

Prediction ofrestenosis

Prediction ofCAD progression

Intracoronary Flow Profiling Methods• The intracoronary ultrasound (ICUS) “core” is positioned in the relevant section of

the artery and a biplane angiogram is recorded using dilute contrast.• ICUS is performed with controlled pull-back at 0.5 mm/sec with biplane

angiography. ECG is simultaneously recorded for “gating.”• A dynamic programming technique extracts the lumen and EEL outline from the

ICUS at end-diastolic frames and re-aligns them. • The ICUS frames are realigned in 3-D space perpendicular to the ICUS core image.• The reconstructed lumen is divided into computational control volumes comprising

0.3 mm thick slices along the segment, 40 equal intervals around the circumference, and 16 intervals in the radial direction.

• Dividing the blood into small “cubes” on the grid, the Navier-Stokes equations of fluid flow are solved numerically using an iterative procedure (Computational Fluid Dynamics).

• Shear stress at the wall is obtained by multiplying viscosity by the velocity gradient at the wall.

Selected ICUS frames

Total number of frames 100-200/arterial segment

Measurements of Lumen, Outer Vessel Wall, and Plaque by IVUS

(DeFranco. AJC 2001; 88 [Suppl]: 7M)

• Lumen

• Outer Vessel Wall = Area within EEM

• Plaque = Intimal-Medial Thickness

Stacking of ICUS frames

Top half-plane

Reconstructed Lumen

Creation of Computational Mesh

640 Cells per cross-section

3mm

Representative Example of3-D Reconstruction of Coronary Artery

RAO projection LAO projection

Example of 3-D Reconstruction ofCoronary Artery

Solid line passing through the centroid of the lumen defines a pathlinePerpendicular distance between pathline and lumen border defines local lumen radius, perpendicular distance between EEL border and pathline defines the local EEL radiusDifference between local EEL and lumen radii defines local plaque thickness

Original angiogram ofa portion of an artery

studied

Composite reconstruction of portion of the arterial segment,consisting of outer arterial wall, plaque, and lumen:

Isolated view of reconstructed outer arterial wall:

Isolated view of reconstructed lumen:

Isolated view of reconstructed atherosclerotic plaque:

Example of 3-D Reconstruction of Arterial Segment

Velocity Field Presented As ALongitudinal Section

Coronary Endothelial Shear Stress

wyuWSS

dynes/cm2

[Artery is displayed as if it were cut and opened longitudinally, as a pathologist would view it.]

Reproducibility Studies ofIntra-coronary Flow Profiling Measurements

Cardiac catheterization and coronary angiography– Patients studied completely with ICUS pullback and

biplane angiography (“Test A”)– All catheters removed, and after a few minutes, entire

procedure repeated (“Test B”):• catheters reinserted• angle, skew, table height reproduced to mimic the

initial procedure– All calculations performed to measure lumen, outer

vessel, plaque morphology, and endothelial shear stress

Reproducibility of 3-D Coronary Artery Reconstruction

“Test A” and “Test B” Performed Separately

Lumen Radius[mm]

EEL Radius[mm]

Plaque Thickness[mm]

Endothelial SS[dynes/cm2]

r = 0.96 r = 0.95 r = 0.91 r = 0.88

Grid divided into 2,560-10,640 areas/artery (average 5,900/artery)

Each p < 0.0001(Coskun, et al. JACC 2002, 39; 44A)

Arte

rial S

egm

ent L

engt

h (m

m)

In-Vivo Determination of the Natural Historyof Restenosis and Atherosclerosis

• First pilot study of its kind in the world• Complete intra-coronary flow profiling at index

catheterization and repeated at 6-month followup• 10 patients enrolled:

– Followup catheterization completed in 8 patients• one refused recath; one had clinical event prior to

recath

Pilot Study of Natural History of Progression of Coronary Atherosclerosis and In-Stent Restenosis

Effect of Candesartan vs. Felodipine

Con

sent

and

Ran

dom

ize

Identification ofappropriate CADsubstrate:-PTCA/stent-obstruction < 50% in adj artery, not revascularized

Cath # 1

Cath # 2

EnterBWH System

Candesartan activeFelodipine placebo

Candesartan placeboFelodipine active

Titration to BP < 140/90 mmHg(Outpatient visits)

Time Line: Hours Time 0 Mo 1 Mo 2 Mo 3 Mo 6

Preliminaryidentificationof hypertensivepatient

Inclusion Criteria:• Hypertension• CAD requiring stent• Additional minor CAD

Pilot Study of Natural History of Progression of Coronary Atherosclerosis and In-Stent Restenosis

Followup Status:One patient refused repeat catheterizationOne patient developed acute coronary syndrome

and required urgent cath and restentingSerial Study Cohort: 8 patients

Native CAD Endpoints: 6 patients with serial studies5 Felodipine and 1 patient Candesartan

Restenosis Endpoints: 6 patients with serial studies3 Candesartan and 3 Felodipine

Pilot Study of Candesartan to Reduce CoronaryIn-Stent Restenosis and

Progression of AtherosclerosisPatient Population: 10 patients

9 men; 1 womanMean age: 60.8 years (range 37-83 years)Concomitant medications: B-blockers, statins, and aspirin (all patients)Mean fasting lipids: Total cholesterol: 156 mg/dl

LDL cholesterol: 95 mg/dlHDL: 36 mg/dlTriglycerides: 150 mg/dl

Blood Pressure: Baseline: 156/89 mmHg

Followup: 137/78 mmHg

Example of Coronary Atherosclerosis Progression Over 6-Month Period

(Stone, et al. JACC 2002, 39: 217A)

Plaque Thickness [mm] Lumen Radius [mm] EEL Radius [mm] ESS [dynes/cm2]

Arte

ry le

ngth

[mm

]

Plaque ThicknessIncreases in Areasof Low ESS

Lumen RadiusDecreases inAreas of IncreasedPlaque Thickness

EEL RadiusIncreases in Distal Areas

ESS Increasesin Areas ofPlaque Increaseand Decreases in Distal Areas

Example of Coronary Artery“Outward Remodeling” Over 6-Month Period

Lumen Radius[mm]

EEL Radius[mm]

Plaque Thickness[mm]

Endothelial SS[dynes/cm2]

Lumen radiusenlarges

Outer vessel radiusenlarges

Plaque thicknessdoes not change

ESS returnsto normal values

(Stone, et al. JACC 2002, 39: 217A)

Arte

ry S

egm

ent L

engt

h (m

m)

Example of Instent RestenosisOver 6-Month Period

Lumen Radius[mm]

EEL Radius[mm]

Plaque Thickness[mm]

Endothelial SS[dynes/cm2]

Lumen radiussmaller withinstent,larger outsideof stent

Outer vesselradiusenlarges

Plaque thickenswithin stent,no change outsidestent

Endothelialshear stress increaseswithin stent,normalizes outsidestent

(Kinlay, et al. JACC 2002, 39: 5A)

Arte

ry S

egm

ent L

engt

h (m

m)

Example of No Change in Stented Segment Over 6-Month Period

Lumen Radius [mm] EEL Radius [mm] Plaque Thickness [mm] ESS [dynes/cm2]

Ar te

r y S

egm

ent L

e ngt

h ( m

m)

(Kinlay, et al. JACC 2002, 39: 5A)

Conclusions• This methodology allows for the first time in man the systematic

and serial in vivo investigation of the natural history of CAD and consequent vascular responses.

• There are different and rapidly changing behaviors of different areas within a coronary artery in response to different ESS environments.

• The methodology can evaluate in detail the ESS that are responsible for the development and progression of CAD, as well as the remodeling that occurs in response to CAD.

• The technology may be invaluable to study the impact of pharmacologic or device interventions on these natural histories

References• Asakura T, Karino T. Flow patterns and spatial distribution of atherosclerotic lesions in

human coronary arteries. Circ 1990; 66: 1045-66.• Nosovitsky VA, et al. Effects of curvature and stenosis-like narrowing on wall shear stress in

a coronary artery model with phasic flow. Computer and Biomed Res 1997; 9: 575-580.• Malek A, et al. Hemodynamic shear stress and its role in atherosclerosis. JAMA 1999; 282:

2035-42.• Ward M, et al. Arterial remodeling. Mechanisms and clinical implications. Circ 2000; 102:

1186-91.• Ilegbusi O, et al. Determination of blood flow and endothelial shear stress in human

coronary artery in vivo. J Invas Cardiol 1999; 11: 667-74.• Feldman CL, et al. Determination of in vivo velocity and endothelial shear stress patterns

with phasic flow in human coronary arteries: A methodology to predict progression of coronary atherosclerosis. Am Heart J 2002; 143: (in press).

• Feldman CL, Stone PH. Intravascular hemodynamic factors responsible for progression of coronary atherosclerosis and development of vulnerable plaque. Curr Opin in Cardiol 2000; 15: 430-40.

References• Coskun AU, et al. Reproducibility of 3-D lumen, plaque and outer

vessel reconstructions and of endothelial shear stress measurements in vivo to determine progression of atherosclerosis. JACC 2002; 39: 44A.

• Stone PH, et al. Prediction of sites of progression of native coronary disease in vivo based on identification of sites of low endothelial shear stress. JACC 2002; 39: 217A.

• Kinlay S, et al. Endothelial shear stress identified in vivo within the stent is related to in-stent restenosis and remodeling of stented coronary arteries. JACC 2002; 39: 5A.

• Feldman CL, et al. In-vivo prediction of outward remodeling in native portions of stented coronary arteries associated with sites of high endothelial shear stress at the time of deployment. JACC 2002; 39: 247A.