novel uses for ultrasound as both an imaging and ... · thesis title: novel uses for ultrasound as...
TRANSCRIPT
Novel Uses for Ultrasound as Both an Imaging and Therapeutic Tool in
the Characterization and Percutaneous Revascularization of Chronic
Total Occlusion
By
Amandeep Singh Thind
A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy
Graduate Department of Medical Biophysics
University of Toronto
Amandeep Singh Thind
Doctor of Philosophy Thesis
Department of Medical Biophysics, University of Toronto
Sunnybrook Health Sciences Centre, S639-2075 Bayview Avenue
Toronto, Ontario, M4N 3M5 Canada
© Copyright Amandeep Thind 2011
ii
Abstract
Thesis Title: Novel Uses for Ultrasound as Both an Imaging and Therapeutic Tool in the Characterization
and Percutaneous Revascularization of Chronic Total Occlusion
Amandeep Singh Thind, Doctor of Philosophy, Department of Medical Biophysics, University of Toronto,
2011.
Revascularization of Chronic Total Occlusions (CTO) by percutaneous coronary interventions is limited by
low success rates, primarily due to difficulty in guidewire crossing. There are a number of contributing
factors that make guidewire crossing challenging. Two of the most significant impediments are: a)
inability to adequately visualize the CTO to appropriately plan a pathway to the distal lumen, and b)
difficulty in physically crossing the rigid endcap at the proximal end of CTO without using stiff wires.
Moreover, there is a significant knowledge gap in the composition of CTOs, and the consequent impact
of that composition on crossability.
This thesis presents tools and techniques to help mitigate the current shortcomings, while shedding new
light on CTO composition and maturation. The tools and techniques presented herein are based upon
ultrasound approaches with the intent of eventually developing these strategies into catheter based
solutions.
Recent studies have suggested that the presence of microvessels in CTO may provide a preferred
pathway for guidewire crossing. However, due to limited resolution and a lack of soft tissue contrast in
angiography, microvessels within CTO cannot generally be detected by in-vivo angiographic techniques,
and when they are visualized, it is unknown whether or not they are intraluminal. In this thesis, high
frequency ultrasound with Power Doppler overlays is shown to be capable of detecting and tracking
iii
transluminal recanalization channels using an in vivo porcine model of CTO. It is also shown that
ultrasound is a more sensitive technique to detect and map these channels than MRI. Furthermore,
features of microvasculature in CTOs that had not previously been seen are presented.
A technique was then developed to facilitate guidewire crossing through the proximal endcap, also
known as the proximal fibrous cap (PFC). In order to assess the ease with which a probe is able to
perforate the PFC, a system was designed and to measure the force required for PFC puncture. This
system was validated by examining the required puncture forces for CTOs of different ages. It was
shown that CTOs less than 6 weeks in age are significantly easier to puncture than those greater than 12
weeks. This coincides with differences in composition, with the presence of softer materials at the
earlier time point, such as thrombus and proteoglycans compared to stiffer fibrotic materials which
predominate at late timepoints.
After development and validation of a reliable technique to measure ease of PFC puncture, the efficacy
of therapies designed to modify PFC compliance could be assessed. The use of ultrasound mediated
microbubble (UMM) disruption to act as an adjuvant to accelerate collagenase therapy in CTO was
examined. A significant reduction in puncture force and an increase in the amount of collagen degraded
was achieved using a combined UMM + collagenase treatment compared with collagenase therapy
alone and UMM treatment alone.
iv
To Jeevan,
whose arrival gave me the inspiration to complete this work
v
Acknowledgements
The work presented in this thesis would not have been possible without the help of many people who
provided a combination of guidance, support, motivation, compassion and entertainment.
First I would like to thank my supervisor, Dr. F. Stuart Foster for providing me with an opportunity to
complete this work with the freedom to make mistakes, learn from them, and ultimately become an
independent scientist. My supervisory committee members, Drs. Graham Wright, Alex Vitkin, and
Bradley Strauss provided me with a great deal of guidance in a constructive and productive manner.
I was very fortunate to be able to make numerous extremely fruitful scientific collaborations. Without a
doubt, the work presented in this thesis would not have been possible without the Strauss lab. Starting
with Dr. Strauss, who despite his unbelievably busy schedule, was always willing to have a friendly
discussion. Within the lab, Michelle Ladouceur-Wodzak was absolutely vital to the animal experiments,
and always brought a cheery attitude. Aaron Teitelbaum was a patient teacher, and taught me how to
perform a number of procedures for which I am grateful. I also had excellent collaborations with Drs.
Raffi Karshafian and David Goertz, who provided expertise in therapeutic ultrasound, as well as excellent
insights on academia and research as a whole.
Dr. Brian Courtney proved to be an excellent mentor and provided me with the opportunity for a very
exciting side project to my thesis work, and I feel fortunate to have him as a colleague and friend.
I would also like to acknowledge the members of the CTO team for helping to guide me through the
early stages of my graduate career. In particular, I appreciate the discussions and support I received
from Nigel Munce, General Leung, and Kevan Anderson.
vi
I would like to very emphatically thank all of the members of the Foster and Burns labs. I can’t think of a
better environment and group of colleagues to spend almost 6 years of my career with. The ultrasound
lab environment very much starts with Kasia Harasiewicz. She is truly the heart and soul of the lab, and
was always the first person I went to talk to about anything – good or bad. The members of the
ultrasound group are really what kept me sane through my time at Sunnybrook, and I’d particularly like
to thank in no particular order: John Hudson, Kogee Leung, Shawn Stapleton, Robin Castelino, and Toby
Lam. You were a blast to be around from the beginning to the end, were instrumental in keeping the lab
fun, and making sure that I don’t stop believing. I will miss our daily chats about anything and everything
when I leave the lab.
I have also been very lucky to have such a supportive family. My brother, sister, and parents have been
there for me since the day I was born, never doubting me in any endeavours I have embarked upon. I’ve
also recently seen many additions to the family, and I am privileged to have in-laws that treat me like
I’ve been part of the family my whole life.
Most importantly, I’d like to thank my wife and best friend, Loveleen. I met her during the first year of
graduate school, and she has turned my life upside down. She has been with me every step of the way
to provide me support and encouragement to pick me up no matter how down I was. I look forward to
the next chapter in our lives together, and all the joys and challenges that come along with being a
husband and father.
vii
Table of Contents
Abstract ......................................................................................................................................................... ii
Acknowledgements ....................................................................................................................................... v
Table of Contents ........................................................................................................................................ vii
List of Figures ................................................................................................................................................ x
List of Tables ............................................................................................................................................... xv
List of Acronyms & Abbreviations ............................................................................................................... xvi
1 Introduction .......................................................................................................................................... 1
1.1 Motivation ..................................................................................................................................... 1
1.2 Occlusive Vascular Disease ........................................................................................................... 1
1.2.1 Anatomy of a healthy artery ................................................................................................. 1
1.2.2 Atherosclerosis ...................................................................................................................... 2
1.2.3 Clinical Presentation and Treatment .................................................................................... 5
1.3 Chronic Total Occlusion ................................................................................................................ 7
1.3.1 Clinical presentation & Benefits to revascularizing CTOs ..................................................... 7
1.3.2 Progression to CTO ................................................................................................................ 9
1.3.3 Maturation of CTO ................................................................................................................ 9
1.3.4 Impediments to PCI ............................................................................................................. 11
1.3.5 Animal Models of Chronic Total Occlusion ......................................................................... 12
1.3.6 CTO Crossing Strategies ...................................................................................................... 16
1.4 Collagen and Collagenase ........................................................................................................... 18
1.4.1 The Collagen Molecule ........................................................................................................ 18
1.4.2 Collagenase ......................................................................................................................... 19
1.5 Ultrasound .................................................................................................................................. 20
1.5.1 Imaging ................................................................................................................................ 20
1.5.2 Blood flow imaging ............................................................................................................. 25
1.5.3 Ultrasound mediated microbubbles (UMM) ...................................................................... 26
1.6 Thesis aims .................................................................................................................................. 28
1.6.1 Specific Aims of the Thesis .................................................................................................. 30
2 Microvascular Study of Chronic Total Occlusion in a Porcine Model ................................................. 32
viii
2.1 Introduction ................................................................................................................................ 32
2.2 Methods ...................................................................................................................................... 33
2.2.1 CTO Model .......................................................................................................................... 33
2.2.2 High Frequency Ultrasound (Micro-ultrasound) ................................................................. 34
2.2.3 MRI ...................................................................................................................................... 36
2.2.4 MicroCT (μCT) ..................................................................................................................... 36
2.2.5 Histological Processing ........................................................................................................ 37
2.3 Results ......................................................................................................................................... 38
2.3.1 Histology ............................................................................................................................. 38
2.3.2 Micro-ultrasound (μUS) ..................................................................................................... 39
2.3.3 MRI ...................................................................................................................................... 42
2.3.4 Intraluminal Microvessel CTO Features .............................................................................. 43
2.4 Discussion .................................................................................................................................... 47
2.5 Conclusions ................................................................................................................................. 50
3 A Novel Method for Measurement of Proximal Fibrous Cap Puncture Force in Chronic Total
Occlusions and its application ..................................................................................................................... 52
3.1 Introduction ................................................................................................................................ 52
3.2 Materials & Methods .................................................................................................................. 53
3.2.1 CTO Model .......................................................................................................................... 53
3.2.2 Removal of CTO for Ex-Vivo Testing .................................................................................... 53
3.2.3 Puncture Force Testing ....................................................................................................... 54
3.2.4 Histology ............................................................................................................................. 58
3.2.5 Statistical Analysis ............................................................................................................... 58
3.3 Results ......................................................................................................................................... 59
3.4 Discussion .................................................................................................................................... 62
3.4.1 Potential applications ......................................................................................................... 65
3.5 Conclusions ................................................................................................................................. 66
4 The use of Ultrasound Mediated Contrast agents as an adjuvant for collagenase therapy in Chronic
Total Occlusion ............................................................................................................................................ 67
4.1 Introduction ................................................................................................................................ 67
4.2 Materials and methods ............................................................................................................... 68
4.2.1 CTO Model .......................................................................................................................... 68
ix
4.2.2 Sample collection ................................................................................................................ 68
4.2.3 Treatment groups ............................................................................................................... 70
4.2.4 Standard Treatment ............................................................................................................ 72
4.2.5 Treatment Duration ............................................................................................................ 72
4.2.6 Modified Acoustic Setup ..................................................................................................... 72
4.2.7 Collagenase ......................................................................................................................... 73
4.2.8 Acoustic parameters ........................................................................................................... 73
4.2.9 Biochemical Assays ............................................................................................................. 75
4.2.10 Puncture Force Test ............................................................................................................ 77
4.2.11 Statistical Analysis ............................................................................................................... 77
4.3 Results ......................................................................................................................................... 77
4.3.1 Standard treatment duration .............................................................................................. 77
4.3.2 Extended treatment duration ............................................................................................. 80
4.3.3 Modified acoustic setup ...................................................................................................... 80
4.4 Discussion .................................................................................................................................... 81
4.4.1 Clinical Relevance of Study ................................................................................................. 84
4.5 Conclusions ................................................................................................................................. 84
5 Summary and Future work ................................................................................................................. 86
5.1 Summary and Discussion ............................................................................................................ 86
5.2 Future work: In vivo collagenase studies .................................................................................... 87
5.2.1 Materials & Methods .......................................................................................................... 87
5.2.2 Results & Discussion ........................................................................................................... 89
5.3 Future work: Microspheres loaded with VEGF ........................................................................... 90
5.4 Future work: Collagenase dynamics studies ............................................................................... 93
5.5 Future work: Catheter based technology ................................................................................... 94
5.5.1 Imaging ................................................................................................................................ 95
5.5.2 Therapy ............................................................................................................................... 97
5.5.3 Image Guided Therapeutics ................................................................................................ 98
5.5.4 Compliance Testing ............................................................................................................. 98
5.6 Concluding remarks .................................................................................................................. 100
6 References ........................................................................................................................................ 102
x
List of Figures
Figure 1-1 - The anatomy of a normal artery shown in cross- section. An artery is a three-layered
structure. ................................................................................................................................... 2
Figure 1-2 - Histology of OVD. Panel A shows a Hemotoxylin & Eosin stained slide with intraplaque
haemorhage (Hem) and a calcium deposit (Ca2+). Panel B shows an example of Thin Cap
Fibrous Atheroma. The fibrous cap is labelled FC, and the necrotic core is labelled NC. Figure
adapted from(12). ..................................................................................................................... 5
Figure 1-3 - Angiogram of a CTO before and after revascularization. Panel A shows CTO in Left Anterior
Descending (LAD) artery marked with white arrow. Panel B shows result of successful
recanalization. Figure adapted from (30). ................................................................................. 8
Figure 1-4 - Cross-sectional views of human coronary CTOs. Panel A shows an elastic van Gieson stain of
a lipid rich lesion, with the solid black arrow showing significant cholesterol deposit. These
are generally considered to be soft and strong candidates for successful PCI. Panel B shows
a complex CTO stained with a Movat Pentachrome, containing Microvessels (MV), a Necrotic
Core (NC), as well as significant amounts of collagen (stained yellow). Panel C shows an H&E
stain of a CTO with a large calcium deposit (curved solid arrow). Panels A & C adapted from
(32) and Panel B adapted from (34). Bars in A & C represent 1266µm. ................................ 11
Figure 1-5 - Longitudinal views of the Proximal Fibrous Cap. Panel A shows an Elastic Trichrome stain of a
12 week old rabbit femoral artery occlusion. The residual Lumen (L) is shown along with the
Proximal Fibrous Cap (PFC). Panel B shows a Movat Pentachrome stain of a similar rabbit
CTO. The Lumen (L) is shown with a thrombus formed at the site of the plaque. The PFC is
shown in purple. The Media (M), IEL and EEL are also shown. ............................................... 12
Figure 1-6 - Schematic of a focused ultrasound transducer and associated beam. Depending on the
strength of focus, or F-number, the width of the focus will change. This directly affects the
lateral resolution. .................................................................................................................... 24
Figure 2-1 - Angiograms of right leg of porcine subject. The vessel that was occluded is the superficial
branch of the femoral artery. A) shows the region before deposition of polymer. Vessel to be
occluded located in yellow ellipse. B) shows the region immediately after the deposition of
the polymer, with yellow ellipse indicating region of occlusion. ............................................ 35
Figure 2-2 - Histological images of CTO arteries containing intraluminal microvessels. A) shows an H&E
stained vessel containing a single large microvessel (MV). This vessel was perfused with
Microfil (MF) just prior to sacrifice, which can be seen as black portions within microvessels.
The plaque (P) and media (M) are also marked. Bar represents 500μm. B) shows an Elastic
Trichrome stain of an occluded artery with a blood filled microvessel (MV). Here, the plaque
(P), media (M), and Internal (IEL) and External (EEL) Elastic Lamina regions are clearly shown
for easier delineation of layer boundaries. Bar represents 500μm. C) shows the vessel in B
with a PE-Cy5 fluorescent labelled CD31 stained antibody imaged at 5x. The microvessel
(MV) has endothelialized and consequently fluoresces. The elastic layers (IEL and EEL)
autofluoresce and are also seen. D) shows a 20x image of the same sample shown in C),
showing in more detail the microvessel (MV) and the IEL. ..................................................... 40
xi
Figure 2-3 - Cross sectional in vivo B-scan images of a porcine CTO Artery. The artery is flanked on either
side by a vein. A) shows standard B-scan ultrasound image including veins (V), occlusive
plaque (O), and medial layer of artery (M) which appears as a dark, echolucent band. Bar
represents 1mm. Image in B) shows ultrasound B-scan with Power Doppler overlay. Power
Doppler wire frame encompasses occluded artery and one adjacent vein. Regions in colour
represent areas with blood flow. An intraplaque microvessel (MV) is shown which is not
visible in A) and is similar in size to those shown in Figure 2-2. Bar represents 1mm. C) shows
a Pulsed Wave Doppler graph of flow in the centre of the microvessel shown in B) revealing
dampened pulsatile flow within the microvessel. .................................................................. 41
Figure 2-4 - Longitudinal axis in vivo B-scan & MR images of a porcine CTO Artery. The artery is the same
as shown in Figure 2-3. A) shows a reconstructed long axis B-scan ultrasound image
including occlusive plaque (O) and medial layer of artery (M). Image in B) shows ultrasound
B-scan with Power Doppler overlay. Power Doppler readings from outside artery are as a
result of motion at the skin surface. An intraplaque microvessel (MV) is shown. C) shows a
contrast enhanced T1 weighted MR image of coronal view of the same vessel. The vessel
contains a faint signature of a microvessel (MV) running inbetween the adjacent veins (V).
Bar represents 1cm. ................................................................................................................ 42
Figure 2-5 - MRI Images of CTO cross section. A) shows a T1 weighted image of an occluded vessel
flanked by veins in the same vessel shown in Figure 2-3. Upon comparison with Power
Doppler imaging and histology, the area marked (MV) is presumed to be a microvessel. The
corresponding phase contrast image shown in B) is not sufficiently sensitive to detect flow in
the region. ............................................................................................................................... 43
Figure 2-6 - Corkscrew pattern in microvessels. A) shows a 3D rendering of a microvessel (MV) with a
partial corkscrew morphology, imaged in vivo. Portion of adjacent vein (V) also shown. Bar
represents 1mm. B) shows a rendered isosurface of a similar microvessel ........................... 44
Figure 2-7 - Crescent shaped morphology. A) shows an ultrasound cross sectional enface view of a 3D
rendered microvessel (MV) in a characteristic crescent shape acquired in vivo. Bar
represents 1mm B) shows an H&E cross sectional slice of a similar morphology with a
crescent shaped microvessel. Tearing artifact (A) from processing is also marked. Bar
represents 500μm. C) shows a magnified view of the cross section shown in B). ................. 45
Figure 2-8 - Branching from within occlusion. Part A shows an in vivo cross sectional ultrasound B-scan
image with a Power Doppler overlay. Microvessel (MV) in occlusion connects to branch (BR)
outside of vessel. Adjacent vein (V) is also depicted. Bar represents 1mm. B) shows similar
morphology seen with μCT rendered isosurface of a vessel running parallel to the axis of the
artery in the same animal. Dashed green dashed line indicates similar planar section to that
shown from A). ........................................................................................................................ 46
Figure 3-1 - Puncture force measurement setup schematic. The stent at the proximal end is clamped and
the vessel is secured at the distal end using the vessel securing pin. The puncture probe is
placed just above the Proximal Fibrous Cap and clamped into the 3 jawed chuck. The
Microtester arm is slowly lowered such that the probe advances into the PFC. As the probe
moves into the cap, the force pushing back on the probe increases until the PFC is
xii
punctured, at which point the force drops precipitously. The force-displacement information
is read from the load cell and stored electronically. ............................................................... 55
Figure 3-2 - Photographs of puncture test setup. Panel a) shows a front view of the sample holder. The
V-shaped groove where the vessel is placed is shown (black arrow). The stent is then held
using the stent clamp (magenta arrow). The clamp is tightened using the clamp adjustment
screws (brown arrow). Panel b) shows a photograph of the specimen holder from a top
view. The vessel is laid flat in the V-groove and the stent is clamped (blue arrow). A 20G
needle is used to hold the tautly pulled vessel (red arrow) and secured with set screws
(turquoise arrow). The puncture needle is gently placed into the stent (green arrow). Panel
c) shows a photograph of the entire setup. The microtester arm (orange arrow) is lowered
slowly so that the probe assembly (maroon arrow) is advanced into the vessel held in the
fluid tank (purple arrow). Position is controlled by 2- axis stage (yellow arrow). .................. 56
Figure 3-3 - PFC puncture probe and guidewire. The probe used to puncture the PFC is shown in two
views (top and middle) next to a 0.014” guidewire. The bevel cut probe is of similar profile
to a 0.014” guidewire (bottom) in one plane, while remaining completely rigid along the
shaft. ........................................................................................................................................ 57
Figure 3-4 - Sample force displacement curve of puncture from 6 week old CTO. Force rises to peak prior
to puncture (~0.85N), followed by dropoff afterwards. ......................................................... 59
Figure 3-5 - Longitudinal section of a punctured vessel stained with Movat pentachrome. The trajectory
of the puncture probe is seen as the void in this longitudinal section of the proximal fibrous
cap of a CTO. Residual PFC shown in yellow ellipse. Internal Elastic Lamina (IEL), External
Elastic Lamina (EEL), and Media (M) are shown. Bar = 100µm. .............................................. 60
Figure 3-6 - Summary of puncture force tests comparing early occlusive lesions (<=6 weeks) and later
occlusions (>= 12 weeks).Panel a) shows a comparison of mean puncture force values. There
is a significant increase in the mean puncture force value at >=12 week time points
compared to early occlusions (<=6 weeks). Panel b) shows the percentage of lesions in each
group with a puncture force below a cutoff of 1N. * - p < 0.01 vs. >=12 weeks. ................... 61
Figure 3-7 - Mean puncture forces at different timepoints. Panel a) shows that there is a trend towards
increasing puncture force with increasing age of the occlusion. Panel b) shows the
percentage of lesions that require <1N of force for puncture at each timepoint. * - p < 0.001
vs 15 week. † - p < 0.01 vs. 15 week. ...................................................................................... 63
Figure 4-1 – Proposed clinical scenario. Panel a) shows the current clinical protocol. An over-the-wire
(OTW) balloon is advanced to the proximal segment of the CTO and the balloon is inflated.
Collagenase is injected through the wire port of the balloon. Panel b) shows the proposed
clinical protocol. As above, an OTW balloon is advanced to the proximal segment of the CTO.
Now, microbubbles are injected through the wire port. The microbubbles are disrupted by
an ultrasound transducer. In the image, the ultrasound transducer is attached to the
catheter delivering the microbubbles. However, the transducer may also be placed outside
the skin surface and is used as such in the present study. After microbubbles are disrupted,
balloon is inflated and collagenase is injected as with the current clinical protocol. ............. 69
Figure 4-2 - Schematic illustration for treatment groups. Panel a) shows “Control” group using only a
saline infusion, while b) shows the “Collagenase only group”. Panels C & D show groups
xiii
using ultrasound to disrupt microbubbles. c) shows “Ultrasound Only” group, and d) shows
“Both Treatments” group. ....................................................................................................... 71
Figure 4-3 - Ex vivo acoustic setup. The vessel was secured into the sample container and placed onto
the adjustable stage so that the PFC was located at the focus of the ultrasound transducer. A
balloon catheter was advanced into the stent and the balloon was inflated to localize the
proximal end of the lesion. Microbubbles were injected via the wire port of the catheter and
disrupted in the acoustic beam. .............................................................................................. 74
Figure 4-4 - Pulse Sequence for acoustic treatments. A 500ms disruption pulse, consisting of 500 pulses
of acoustic energy (each pulse was 16 cycles at 1MHz) separated by 1ms was used to disrupt
the bubbles in the acoustic field. These pulses were sent every 5s for a total of 120s in order
to allow fresh bubbles to flow into the field after a disruption sequence. ............................. 75
Figure 4-5 - Biorad protein assay data for standard treatment duration. There is no statistically
significant difference in the total protein released in any of the groups. († - p > 0.05). ........ 78
Figure 4-6 - Hydroxyproline released into medium for standard treatment duration. Both the combined
treatment and collagenase only groups showed significant increases in hydroxyproline
release compared to ultrasound only and control. Hydroxyproline in the combined
treatment group was significantly higher than the collagenase only group. († - p > 0.05, ‡ -
p<0.05,* - p< 0.01)................................................................................................................... 79
Figure 4-7 - Puncture force test results for standard treatment duration. The difference between the
Both Treatments group and the Collagenase Only group is statistically significant. The
difference between all other groups is not statistically significant. († - p > 0.05, ‡ - p<0.05) 80
Figure 4-8 - Extended treatment duration results. The extended treatment group showed a large
increase in the amount of hydroxyproline released – significant. The extended treatment
group also showed a drop in the required puncture force, which was significant. (‡ -
p<0.05,* - p< 0.01)................................................................................................................... 81
Figure 4-9 - Modified acoustic setup results. Hydroxyproline and Puncture Force assays are shown. In
both cases, there is no statistically significant difference between the two treatment
configurations. († - p > 0.05) ................................................................................................... 82
Figure 5-1 - Schematic setup of in-vivo collagenase therapy. The imaging beam and the therapy beam
are confocally aligned. The imaging beam is used to locate the PFC. The therapy transducer
is used to disrupt the microbubbles at the confocal region. .................................................. 89
Figure 5-2 - Puncture test results from in vivo study. The combined treatment group showed a
significantly lower puncture force compared to the collagenase only group. ‡ - p < 0.05. .... 90
Figure 5-3 – Data for puncture force testing using local injections of various solutions. Panel a) shows
puncture force values of all 3 groups. † - Data is not statistically significant. Panel b) shows
the proportion of injections less than cutoff point of 1N. ...................................................... 92
Figure 5-4 – Movat pentachrome sections from each treatment group Left panel shows VEGF treated
sample with increased vascularity. Centre panel shows BSA Microspheres injection only.
Right panel shows sample from NIV group. Histology courtesy of Dr. Bradley Strauss. ........ 93
Figure 5-5 – Typical individual crossectional B-scan image of a heavily remodelled rabbit CTO artery.
Artery is located within ellipse, and is not easily delineated using standard imaging
techniques. .............................................................................................................................. 96
xiv
Figure 5-6 - Combined IVUS-OCT image of an atherosclerotic plaque. IVUS on far left shows good
boundary contrast between media and intimal plaque. Calcium is shown as shadow from 6
o'clock to 7 o'clock. OCT image, second from left, shows much higher resolution in fibrous
plaque. Histology, second from right, shows a fibrous plaque with large calcium chunk (5
o’clock to 7 o’clock) stained with a Movat Pentachrome. µCT, far right, shows calcification (5
o’clock to 7 o’clock). ................................................................................................................ 97
Figure 5-7 - Comparison of forward viewing micro ultrasound and OCT images in CTO. Top left shows a
reconstructed cross-section of CTO with 40MHz ultrasound imaging. Top right shows similar
image generated with OCT. OCT image shows finer detail, and detects microchannels not
visible with B-scan ultrasound image alone. Bottom left shows longitudinal image of a
different CTO vessel performed with the same systems. Ultrasound image shows vastly
superior penetration depth. OCT images courtesy of Dr. Nigel Munce. ................................. 99
xv
List of Tables
Table 1-1 - Summary of animal CTO models ............................................................................................... 15
Table 2-1 - Summary of vessel imaging by modality .................................................................................. 34
Table 2-2 - Microvascular characteristics of CTO arteries from histology .................................................. 39
Table 2-3 - Vessel imaging details. Boxes marked with “•” denote presence of feature, boxes marked
with “×” denote that feature was not detected, and N/A implies that technique was not
attempted on sample. ............................................................................................................... 46
Table 4-1 - Summary of treatment groups. Main treatment groups listed in boldface, subgroups listed in
standard text. ............................................................................................................................ 71
xvi
List of Acronyms & Abbreviations
µCT – Micro-Computed Tomography
µUS – Micro Ultrasound
AMI – Acute Myocardial Infarction
BSA – Bovine Serum Albumin
CABG – Coronary Artery Bypass Graft
CCAC - Canadian Council on Animal Care
CFD – Colour flow Doppler
CSA – Cross-sectional area
CTO – Chronic Total Occlusion
CW – Continuous Wave
dB – Decibel
DC – Detergent Compatible
DOTA - Tetraazacyclododecane-tetraacetic acid
DOF – Depth of Field
ECM – Extracellular Matrix
EEL – External Elastic Lamina
FC – Fibrous Cap
FLIVUS – Forward Looking Intravascular Ultrasound
FWHM – Full width half maximum
GLY - Glycine
H&E – Hematoxylin & Eosin
HDL – High Density Lipoprotein
HYP - Hydroxyproline
IEL – Internal Elastic Lamina
xvii
IPH – Intraplaque Haemorrhage
IVUS – Intravascular Ultrasound
L - Lumen
L-PLA - Levo Poly Lactic Acid
LAD – Left Anterior Descending
LDL – Low Density Lipoprotein
M – Media
MRI – Magnetic Resonance Imaging
MI – Mechanical Index
MMP – Matrix metalloproteinase
MV - Microvessel
NC – Necrotic core
NIV – Non-intervened
NS – Not Significant
OCR – Optical coherence reflectometry
OCT – Optical coherence tomography
OTW – Over- the- wire
OVD – Occlusive vascular disease
PCI – Percutaneous Coronary Intervention
PD – Power Doppler
PFC – Proximal Fibrous Cap
PLGA – Poly(lactide-co-glycolide acid)
PRF – Pulse Repetition Frequency
PRO - Proline
PW – Pulsed Wave
xviii
RF – Radiofrequency
RMV – Realtime Microvisualization
SEM – Standard Error of the Mean
SMC – Smooth Muscle Cells
SONAR - Sound Navigation and Ranging
TCFA - Thin Cap Fibrous Atheroma
TIMI - Thrombolysis in Myocardial Infarction
t-PA – Tissue Plasminogen Activator
UCA – Ultrasound Contrast Agent
UMM – Ultrasound Mediated Microbubbles
VH – Virtual Histology
VV – Vaso Vasorum
1
1 Introduction
1.1 Motivation
Chronic Total Occlusions (CTO), or persistent complete stenoses of arteries, are often referred to as the
final frontier for angioplasty. A bevy of devices and techniques have been developed specifically to solve
the problem of CTO, yet they remain a significant challenge, and the majority of CTO specific devices are
no longer on the market. Only recently has significant effort been put into understanding the structure,
composition, and maturation of CTOs. This knowledge may lead to alternative techniques and strategies
for CTO.
Two of the most significant difficulties in successful angioplasty in CTO that remain unmitigated are: a)
the inability to adequately visualize the CTO to find pathways to cross lesions with guidewires, and b)
the inability to cross lesions with conventional guidewires because of the hard physical properties of
mature CTOs. This thesis will examine the use of ultrasound to image preferred pathways through CTOs,
as well as using ultrasound mediated microbubbles to help change the compliance of CTOs by assisting
enzyme therapy to facilitate guidewire crossing.
1.2 Occlusive Vascular Disease
1.2.1 Anatomy of a healthy artery
A healthy artery has a well-defined trilaminar structure, as shown in Figure 1-1. The innermost layer is
referred to as the intima. At birth, this is a very thin 1-2 cell thick layer, predominantly consisting of
endothelial cells. This layer is in direct contact with blood flowing through the vessel lumen. The
smooth surface of the cells provide a low friction conduit for blood to flow, and the expression of
various molecules on the surface of the cells, such as heparin, allow blood to remain in its liquid state for
2
extended periods of time(1). The intima is bounded radially by a thin, fenestrated elastic layer known as
the internal elastic lamina (IEL). Outside of the IEL lies the media. In elastic arteries, which tend to be the
vessels with very high flow rates such as the aorta, the media consists of concentric layers of smooth
muscle cells and highly elastic extracellular matrix (ECM). Smaller arteries, such as the coronary vessels,
tend to be predominantly muscular and have elastic tissue and smooth muscle cells co-residing instead
of being organized concentrically in layers. The outer boundary of the media is delineated by the
external elastic lamina (EEL). This is again a thin, fenestrated elastic layer. Outside of the EEL lies the
outermost layer of an artery, known as the adventitia. The adventitia contains loose connective tissue,
fibroblasts and mast cells. Importantly, the vaso vasorum also resides here. The vaso vasorum is a
network of small blood vessels that provide a blood supply to the cells of the artery itself.
Figure 1-1 - The anatomy of a normal artery shown in cross- section. An artery is a three-layered structure.
1.2.2 Atherosclerosis
Atherosclerosis is a condition in which the wall of an artery thickens as a result of an inflammatory
process that is initiated with the deposition of fatty materials. The term atherosclerosis comes from the
combination of two words: 1) atheroma - the Greek term for “lump of gruel” and 2) sclerosis, the
3
hardening or stiffening of a structure. The disease has been documented in humans as far back as the
Egyptian empire, with evidence of disease seen in mummies(2). Up until the end of the 19th century,
atherosclerosis was viewed as an inevitable passive buildup in inanimate pipes. The Osler Textbook of
Medicine went so far as to say “In the make-up of the machine, bad material was used in the tubing”(3).
It has, however, become quite clear that this is not the case. Atherosclerosis is now considered to be an
inflammatory driven disease(4). The resulting partial or complete blockage of a blood vessel is referred
to as Occlusive Vascular Disease (OVD).
The preliminary stage of atherogenesis is believed to be the deposition of small lipoproteins, particularly
Low Density Lipoprotein (LDL) cholesterol, into the intima(5). Normally, these particles are able to pass
freely in and out of the arterial wall. However, with increased concentration, these particles are more
easily bound to proteoglycans in the intima, where they are retained for an extended period. These
bound lipoprotein molecules tend to be more susceptible to oxidative changes (6,7) in what is termed
response-to-retention. Endothelial cells in the vessel wall recognize these oxidized lipoproteins as a sign
of danger, and produce chemokines while expressing adhesion molecules on the blood facing surface.
These chemokines act to attract monocytes and cause them to proliferate and mature into
macrophages(8). These macrophages proceed to ingest the lipids, forming what are referred to as
“foam cells”. Similarly, T lymphocytes are attracted to the subintimal region. These lipid-filled
inflammatory cells comprise fatty streaks, which are the first step in the development of an atheroma.
Next, lesion growth is triggered largely by the migration and proliferation of smooth muscle cells (SMC)
into the intimal area. These SMCs are recruited by chemoattractants secreted by macrophages. They
then begin to replicate rapidly and produce an ECM as a way of isolating the developing plaque from the
bloodstream. The ECM is rich in proteoglycans and fibrillar collagens (mostly types I and III). The
deposition of these fibrous components over the plaque is referred to as a fibrous cap (FC). The
4
presence of this cap often results in the death of a significant proportion of the foam cells which are
separated from their nutrient supply. This in turn leads to the release of free lipids into the plaque
region from the necrotic foam cells. This region is referred to as the necrotic core (NC).
Angiogenesis is also prominent in atherosclerosis. There are functional reasons for this including: a) as a
response to hypoxia from cells within the necrotic core - analogous to angiogenesis in tumours, and b)
to provide inflammatory cells an easy pathway into and out of the lesion as they respond to
chemoattractants summoning them to the developing atheroma. Similar to the neovessels that form in
tumours, the microvessels in atherosclerosis tend to be poorly organized and can be prone to leakage.
This leakage is referred to as Intraplaque Haemorrhage (IPH), whereby blood is allowed to enter the
plaque. An example of IPH is shown in Figure 1-2. This can also promote rapid plaque growth, largely
driven by the presence of free cholesterol released from necrotic macrophages and/or the cell
membranes of lysed red blood cells from the IPH (9). This free cholesterol has been implicated with
lesion instability(10), as it results in a further increase in both the level of inflammation and the size of
the necrotic core. The collection of inflammatory cells acts to further degrade the fibrous cap over the
plaque. The thickness of the FC depends on a balance between the secretion of new ECM constituents
(secreted predominantly by SMCs) and the activity of matrix mellatoproteinases (MMP), which are a
family of enzymes that break down the ECM (secreted predominantly by macrophages and SMCs). Caps
which are particularly thin (<65 um) are at high risk for plaque rupture(11). This feature is referred to as
Thin Cap Fibrous Atheroma (TCFA). When the thin cap is disrupted, the blood from the lumen of the
vessel is exposed to a myriad of prothrombogenic factors present in the necrotic core. The resultant
thrombus formation leads to a rapid obstruction of the vessel and the sudden loss of blood flow to
downstream tissue, often resulting in what is known as an Acute Myocardial Infarction (AMI), or heart
attack. Lesions with TCFA are responsible for the majority of AMI. An example of TCFA is shown in Figure
1-2.
5
Figure 1-2 - Histology of OVD. Panel A shows a Hemotoxylin & Eosin stained slide with intraplaque haemorhage (Hem) and a calcium deposit (Ca
2+). Panel B shows an example of Thin Cap Fibrous Atheroma. The fibrous cap is labelled FC, and the
necrotic core is labelled NC. Figure adapted from(12).
Plaque rupture does not inevitably result in AMI. Rather, there is a balance between pro and anti
thrombotic factors that often lead to partial vessel blockage. Lesion growth, both from the perspective
of atherosclerotic plaque progression and thrombus formation, tends to be episodic, advancing in fits
and starts, as opposed to a steady linear or exponential growth as was previous believed(12). Lesions
initially tend to grow outwardly, meaning that the disease often progresses significantly before
significantly reducing the size of the blood vessel lumen and restricting blood flow(13).
Various behavioural and environmental factors have been implicated with causing OVD. Smoking and
diabetes are seen as the most significant risk factors for developing disease, while hypertension,
hypercholesterolemia, and others are seen as also being significant (14).
1.2.3 Clinical Presentation and Treatment
There are three main approaches to treating occlusive vascular disease. The first is with the use of
lifestyle changes in combination with medication. This is applicable for nearly all patients and is the
mainstay of therapy for milder cases of disease, or for patients that are deemed unable to undergo
more invasive interventions. This therapy often includes the use of one or more of the following: A
6
vasodilator (Nitroglycerin) to dilate the blood vessels and increase blood flow to the myocardium (i.e.
increasing supply), while relaxing the veins, thus decreasing the blood flow returning to the heart. Anti-
platelet agents (Aspirin, Plavix) reduce the risk of blood clots; and statins (eg. atorvastatin) decrease
blood cholesterol levels. A plethora of other medications are given on a case by case basis.
The second approach to treating occlusive vascular disease, especially in the coronary arteries, is known
as Coronary Artery Bypass Graft (CABG). CABG is an extremely invasive surgical procedure that typically
requires a sternotomy in order to expose the heart to the surgeon. The heart is then cannulated and
attached to a heart-lung machine, which oxygenates the blood and pumps it through the circulation,
allowing the surgeon to arrest the heart in order to perform the bypass. Blood flow is shunted away
from the diseased segment of the coronary artery with vessels harvested from other parts of the
patient’s body, most commonly the saphenous veins or the left internal mammary artery. These
procedures are very expensive and are associated with significant recovery times.
The third approach is known as Percutaneous Coronary Intervention (PCI). PCI involves crossing through
the narrowed, or stenosed, artery with a small guide wire (~360um diameter) and inflating a balloon in
the artery to open the stenosed segment and facilitate increased blood flow. Often, a stent is placed to
help prevent restenosis.
The past several decades have seen drastic improvements in PCI. Developments in catheter and
guidewire technology have increased success rates well into the 90% range for partially stenosed
vessels, while greatly reducing the risks of adverse effects in unsuccessful cases. The advent of drug
eluting stents has greatly reduced rates of restenosis following PCI, a problem that previously plagued
the procedures(15).
7
When possible, PCI interventions are preferred over CABG surgical procedures. Angioplasty procedures
are less traumatic to the patient, result in shorter hospitalization and recovery times, and result in fewer
cerebrovascular complications, such as stroke(16).
PCI is guided using fluoroscopy. The interventionalist injects a radiopaque dye to visualize the
vasculature in order to identify the presence of a narrowing within an artery. Although fluoroscopy
provides only projection imaging, it has proven to be very useful in PCI of partially stenosed arteries.
However, in cases where an artery is completely occluded, fluoroscopy is often not sufficient for
guidance(17), as will be detailed in the subsequent section.
1.3 Chronic Total Occlusion
Although various definitions of Chronic Total Occlusion (CTO) exist in literature, CTO of an artery is
generally described as occurring when the artery is completely blocked such that it inhibits blood flow
through the channel. The flow is either stopped completely, representing a “true” occlusion
corresponding to Thrombolysis In Myocardial Infarction (TIMI) grade 0, or blood is allowed to penetrate,
but does not perfuse the distal coronary bed (TIMI grade 1), and does not have a angiographically
discernable lumen(18,19). Definitions of CTO also include a minimum age of occlusion, varying from
greater than 2 weeks(20) to requiring occlusions to be at least 3 months(21). A consensus document
from a panel of CTO experts agreed that 3 months is an appropriate time period to be considered a
CTO(22).
1.3.1 Clinical presentation & Benefits to revascularizing CTOs
It has been shown that the majority of AMI are as a result of plaque rupture, and not as a result of
CTO(22). Patients with coronary CTOs generally present with recurring chest pain (stable angina) which
is most commonly encountered under duress. This is due to presence of stunned or hibernating
8
myocardium that is viable, but is poorly perfused via collaterals and unable to meet the increased
demands of stress situations (exercise, emotional upset).
CTO is a very common occurrence in patients with OVD. Approximately one third of all patients
undergoing coronary angiography are diagnosed as having at least one CTO(23). Despite the common
occurrence of CTO, only approximately 12-15% of PCI attempts are on lesions classified as CTOs (24).
While there has been doubt raised as to the benefits of opening CTOs in all patients(25), there remain
several clear, significant benefits that come from selective CTO revascularization. Revascularization of
CTO has shown improvements in ventricular function(26), as well as possible increases to long term
survival rates(27,28). While it is has not been conclusively shown that there is a definite increase to long
term survival rates, it has been shown that the presence of CTO increases mortality in infarct in non-CTO
arteries(29). Paired angiograms showing a CTO prior to and following successful revascularization of CTO
with PCI are shown in Figure 1-3.
Figure 1-3 - Angiogram of a CTO before and after revascularization. Panel A shows CTO in Left Anterior Descending (LAD) artery marked with white arrow. Panel B shows result of successful recanalization. Figure adapted from (30).
9
1.3.2 Progression to CTO
CTOs generally develop under one of two mechanisms. First, a CTO can develop as a result of
atherosclerotic plaque rupture followed by bi-directional thrombus formation. Second, CTO can form as
an atherosclerotic plaque proceeds to gradually occlude an entire lumen. The constituents of an early
stage CTO generally include thrombus and lipid rich components, which are eventually replaced by
collagen and calcium deposits. Also, both mechanisms often occur concurrently - with a vessel largely
occluded via atherosclerotic progression and thrombus formation closing the residual lumen.
1.3.3 Maturation of CTO
As CTOs age, several of their characteristics change drastically over time. The changing characteristics
have significant effects on the ability of an operator to perform PCI on these lesions.
1.3.3.1 CTO Constituents
Despite the common clinical occurrence of CTO, very little published literature on the pathophysiology
of CTO exists. Currently, only two publications of studies on the histology of CTOs with more than 10
human samples are available (31,32). The study by Srivasta et al showed that early lesions tend to
contain components that are believed to be soft (lipid or thrombus laden), while older lesions tend to be
much harder, largely acellular, and rich in fibrous components. Old lesions also tend to contain calcium
deposits. These features are all shown in Figure 1-4.
While histopathological studies on human CTOs are very limited, and what is available has some
selection bias, there has been useful study done on various animal models of CTO. Particularly, Jaffe et
al have performed a thorough characterization of a rabbit model of a thrombus- based CTO. This model
10
was examined at 4 different time points: 2 weeks, 6 weeks, 12 weeks, and 18-24 weeks. Some of the
highlights of the composition of the lesions are listed below:
The 2 week time point showed the presence of a significant amount of organized thrombus,
which was almost completely replaced by the 6 week time point and beyond.
The inflammatory response was seen to be extremely high at the 2 week period, dropping off
precipitously beyond that point.
In terms of vessel size, the 2 week time point showed vessels of similar size to the patent vessel,
while vessels at later time points had negatively remodelled significantly.
2 week lesions showed only minor signs of a proteoglycan matrix, while 6 week lesions showed
lesions rich in proteoglycans
Collagen was not found in 2 week lesions, but was found in increasing amounts in with age.
Lipids were almost non-existent at 2 weeks, increasing over time and peaking at 12 weeks.
1.3.3.2 Neovascularization
As with partially stenosed atherosclerotic plaques, neovascularisation is also present in CTO. There are 3
distinct recognized forms of neovascularisation in CTO. First, an increase in the vasculature in the VV
occurs as an inflammatory response. These channels are oriented both radially in the lesions, as well as
running parallel to the parent vessel. This network helps reroute some of the blood that was running
through the parent vessel to keep the distal tissue partially perfused. Also present is neovascularisation
that occurs in the intimal plaque. These vessels connect to the VV and are believed to be responsible for
IPH. The third form of neovessels present in the plaques is what is referred to as recanalization
channels, which may be formed de novo and run predominantly along the same longitudinal channel
and within the original vessel(33).
11
Figure 1-4 - Cross-sectional views of human coronary CTOs. Panel A shows an elastic van Gieson stain of a lipid rich lesion, with the solid black arrow showing significant cholesterol deposit. These are generally considered to be soft and strong candidates for successful PCI. Panel B shows a complex CTO stained with a Movat Pentachrome, containing Microvessels (MV), a Necrotic Core (NC), as well as significant amounts of collagen (stained yellow). Panel C shows an H&E stain of a CTO with a large calcium deposit (curved solid arrow). Panels A & C adapted from (32) and Panel B adapted from (34). Bars in A & C represent 1266µm.
1.3.3.3 Proximal Fibrous Cap
It has been long understood that there is particular difficulty in crossing the proximal and distal ends of a
CTO. Interventional cardiologists have anecdotally reported a distinct “crunch” associated with crossing
what have been often termed “endcaps”(22). There has been dispute over what these endcaps actually
are, with some suggesting that they are small calcifications, while others suggested they were collagen
rich. Jaffe et al were the first to show the appearance of a fibrous endcap, which was termed the
Proximal Fibrous Cap (PFC)(35). The PFC is a particularly densely packed collagen rich region that is
largely acellular, and is shown in Figure 1-5.
1.3.4 Impediments to PCI
CTO is the most commonly cited reason for referral to CABG procedures, representing approximately
68% of such cases(36). The large discrepancy between the common occurrence of CTO and its
infrequent treatment with PCI is a result of low success rates of PCI in CTO patients. Although success
rates vary widely in literature, they are consistently considerably lower than success rates in PCI of
partially stenosed vessels. The high failure rates are compounded by the fact that PCI is only attempted
in cases that the interventionalist believes will be successful, and still yields poor results.
12
Figure 1-5 - Longitudinal views of the Proximal Fibrous Cap. Panel A shows an Elastic Trichrome stain of a 12 week old rabbit femoral artery occlusion. The residual Lumen (L) is shown along with the Proximal Fibrous Cap (PFC). Panel B shows a Movat Pentachrome stain of a similar rabbit CTO. The Lumen (L) is shown with a thrombus formed at the site of the plaque. The PFC is shown in purple. The Media (M), IEL and EEL are also shown.
There are several characteristics of CTO that appear to be indicators of positive and negative outcomes
for PCI. Some of the features that are correlated with negative outcomes are: CTO age (>3 months), long
occlusions ( >15 mm), blunt entrance, no antegrade flow, the presence of bridging collaterals, the
presence of calcium, and the presence of fibrous tissue(37).
Low success rates are generally attributed to two main factors. First, CTO presents a large guidance
problem for angioplasty. Since, by definition, no angiographically detectable contrast is allowed to pass
through the lesion, the interventionalist is not able to accurately track the path of the vessel lumen
through which to pass a guidewire(38). Second, the composition of CTO lesions makes lesions difficult to
cross with a guidewire largely because of the presence of large areas of calcification and the presence of
the PFC.
1.3.5 Animal Models of Chronic Total Occlusion
13
Numerous animal models have been developed to mimic Chronic Total Occlusion for various different
purposes. Predominantly, the work has been focused in rabbit and porcine species models, although
other animals have been used, such as a rat model of carotid occlusion(39). However, carotid arteries
are of significantly different character than coronary vessels. Others have used ameroid constrictors to
slowly limit the blood supply with extrinsic constriction(40). We will focus on femoral and coronary
models with intrinsic methods of creating occlusion.
Using porcine models of CTOs affords the opportunity to work with vessels of similar calibre and
structure to the equivalent vessels in humans. It allows the use of larger intravascular devices, and more
clinically applicable techniques. However, since housing, transportation and animal care costs create
significant practical hurdles; large sample sizes are often not possible in pig models. Conversely, since
rabbits are of significantly lower cost, they are useful for studies of larger sample sizes, although the
calibre of vessels is somewhat smaller than those in humans.
A major criticism of all animal CTO models developed to date is the lack of an atherosclerotic base. All
models rely on a thrombotic occlusion of the vessel. In the vast majority of clinical cases, an atheroma is
present prior to the vessel becoming completely stenosed via a thrombotic event. Another criticism is
that most animal models are femoral artery models even though the vessels of primary interest are
coronary arteries. While the femoral arteries are similar to the coronary vessels in terms of structure,
they add another element of variation. Taken together, the use of animal femoral arteries without a
thrombotic may be seen as significantly different from human coronary arteries with an atherosclerotic
base. However, the models do indeed show several similarities to the human disease, as will be
described.
1.3.5.1 Thrombin injection
14
A rabbit model of CTO has been developed (35,41,42) with the use of a direct thrombin injection. CTOs
are created in the main femoral artery. The artery is surgically isolated and clamped in two places,
approximately 1 cm apart. 100 IU of thrombin is directly injected into the artery. Clamps are maintained
for 60 min to ensure persistent occlusion.
This model has been shown to exhibit many of the characteristics of the human disease, including the
development of microvasculature, both from the vaso vasorum as well as recanalization channels. Also,
this model shows many of the features of human CTO as it matures, from being thrombus and
proteoglycan-rich at the early time points, to collagen-rich at late timepoints. Furthermore, the presence
of the PFC has been shown in this model. This model has been characterized with more rigour than any
other CTO model(35).
Some of the drawbacks for this model are that it requires an exogenous injection of thrombus as a lesion
trigger, as well as the fact that intraluminal calcium is not seen in lesions from this model.
1.3.5.2 Polymer plug model
Prosser et al developed a model using the percutaneous implantation of a bioabsorbable Levo poly lactic
acid (L-PLA) polymer plug(43). These lesions have been successfully created in porcine coronary and
femoral arteries. This model shows microvessel-rich lesions with recanalization channels(44). The main
drawback to this model is that a non-native polymer is introduced to create the blockage.
1.3.5.3 Polymer with embedded Apatite
In order to develop a calcified model of CTO, Suzuki et al implanted bioabsorbable poly(DL-lactide-co-
glycolide acid) (PLGA) plugs coated with apatite into the femoral arteries of both rabbits and pigs and
allowed them to develop into CTOs(45,46). They were able to successfully create CTOs with
15
microcalcifications with no residual polymer. The main drawbacks to this model are that a non-native
polymer is introduced to create the blockage and that the model has not been thoroughly characterized.
1.3.5.4 Stent model
Song et al (47) used the placement of a copper stent as a nidus for the creation of CTO. Copper is an
immunogenic element, and creates a significant immune response from porcine subjects. The lesions
created showed fresh red thrombus and prominent inflammation at the 1 week time point, followed by
organized thrombus at the 4 week time point, with some evidence of calcification near the stent struts.
At the 8 week time point, there was significant deposition of fibrotic components at the proximal and
distal ends, with organized thrombus in between with more well organized calcium.
The main drawback to this model is the requirement of a stent to be in placed in the vessel, which adds
significant difficulty for device testing, therapy, and imaging. Also, this model often results in only
partial occlusion. The animal models of CTO are summarized in Table 1-1 below:
Table 1-1 - Summary of animal CTO models
Model Species Comments
Thrombin injection Rabbit Well characterized Presence of PFC Use of exogenous clotting agent No calicification
Polymer plug Porcine Use of exogenous polymer to stop blood flow Clotting via endogenous agent No calcification
Polymer plug + Apatite Porcine / Rabbit Not well characterized Use of exogenous polymer to stop blood flow Clotting via endogenous agent Calcification
Stent model Porcine Inflammation driven process Stent makes imaging & device testing difficult No calcification Lumen often left unoccluded Not well characterized
16
1.3.6 CTO Crossing Strategies
PCI in CTOs generally confer a success rate in the 75% range (48), which is far below the approximate
success rate of 95% in non-occlusive lesions. Furthermore, CTOs are normally only attempted for PCI
when they are deemed likely to succeed, so success rates are inflated by selection bias. A number of
devices and techniques have been developed for crossing CTOs, the majority of which have not seen
significant adoption and are no longer used. An overview of select techniques and devices is discussed
below.
1.3.6.1 Devices
The development of CTO specific specialty guidewires has provided interventionalists with a large array
of tools for the specific obstacles encountered. For instance, the development of stiff wires with finely
tapered ends has been very useful for easing PFC puncture and for entering microchannels. The
development of hydrophilic wires has found use in many situations, including the STAR technique, as
described below.
A number of ablation / mechanical disruption techniques have also been used. The CROSSER™ is a high
frequency piezoelectric oscillator made by FlowCardia operating in the 20kHz range used to
mechanically disrupt fibrocalcific plaque(49). The Frontrunner™ is a blunt microdissection catheter that
uses a pair of retracting jaws to pull apart plaque and advance itself following a path of least resistance
through a CTO. While this device has shown reasonable efficacy (77% successful in previously failed
cases), it often enters the subintimal space, and can be risky to use, relegating it to use exclusively in
peripheral vessels (50).
Since visualization of CTOs is a significant challenge, various groups have tried imaging solutions to
provide real-time guidance through lesions. The Safecross™ system uses Optical Coherence
17
Reflectometry (OCR) to visualize the boundary of the vessel wall to warn the interventionalist that they
are close to exiting the vessel, while using a radiofrequency (RF) ablator to cross fibrocalcific plaque(51).
Volcano Corporation has also recently acquired a company called Novelis Medical, who have developed
a partially forward looking intravascular ultrasound (FLIVUS) device to visualize the vessel wall, and an
RF dessicator to ablate in the opposite direction of the wall. The aforementioned have not been well
characterized in either animal or human clinical trials.
1.3.6.2 Techniques
It has become particularly popular for several groups, beginning in Japan, to attempt CTO crossings using
retrograde approaches. In these cases, the operator will follow septal sidebranches or large collaterals
to the distal end of the lesions, and attempt to cross the lesion from the distal end to the proximal end
(52). This is normally attempted once the anterograde approach is unsuccessful. Often, this approach is
combined with an anterograde approach using multiple wires, and creating a common channel through
the body of the occlusion.
A relatively risky but often effective technique is the Subintimal Tracking and Re-entry (STAR)
technique(38). This involves the interventionalist intentionally entering the subintimal space with a
hydrophilic guidewire and traversing the lesion outside the plaque, while remaining within the
adventitial boundary. The lumen is then re-entered distal to the CTO. This technique works on the
notion that the tissue in the subintimal space is more compliant and easier to cross than the fibrous and
calcium rich body of the lesion. However, there is a small but definite risk of vessel perforation and
relatively high restenosis rates associated with this technique.
A relatively new technique referred to as the open sesame technique(53) has also been used. This
technique is used when there is a sidebranch just proximal to a hard PFC. This is a very problematic and
common morphology in CTO. The guidewire has a tendency to slide into the sidebranch as opposed to
18
through the PFC. In this technique, either a guidewire or a small balloon is placed into the sidebranch.
This is thought to cause a mechanical shift in the hard plaque to make it easy to enter, as well as
blocking off an undesirable pathway for the wire.
1.3.6.3 Pharmacological approaches
Attempts have also been made to modify the CTO cap using pharmacological approaches. Abbas et
al(54) used a preprocedural infusion of tissue plasminogen activator (t-PA) 8 hours prior to PCI in CTO
patients with a previously failed PCI in an attempt to soften lesions. A success rate of 54% was reported.
Carlino et al(55) recently published a study using a direct injection of contrast and nitroglycerine into a
CTO to open up microchannels. This provides both a guidance advantage, since microvascular pathways
through the occlusion are revealed, as well as dilating the microchannels to provide a wider path for the
guidewire to pass. There is a significant risk of vessel dissection and occasionally perforation with this
technique, and is to be used with caution.
Finally, Strauss et al (41,56) used a localized injection of a collagenase enzyme just proximal to the CTO
to help break down and soften the PFC. These studies showed significant softening of the PFC, but the
effect required 18-72 hours to be effective. This technique has recently begun human trials in a dose
escalation study. This work is germane to this thesis, and will be discussed in more detail in Chapter 4.
1.4 Collagen and Collagenase
1.4.1 The Collagen Molecule
Collagen is the most abundant protein in the human body(57). It has extremely high tensile strength,
and thus proves very useful in providing structural support in the ECM as connective tissue. The collagen
molecule is composed of long chains of amino acids, each chain consisting of approximately 1000 amino
acids, and having a mass of approximately 32 kDa. Each chain is helically wound about itself in what is
19
referred to as a minor helix, and then all three chains are bound together in a triple (major) helix. Each
chain is made up of a repetition of small subunits, each containing the amino acid glycine (GLY) at every
third position. Proline (PRO) and hydroxyproline (HYP) - which is similar to PRO, except with a hydroxyl
group on one of the carbon atoms in the structure, are also found throughout the strands. A typical
sequence of amino acids goes: GLY-X-Y, where X and Y are any amino acids. X is often PRO and Y is often
HYP. HYP is found almost exclusively in collagen, and makes up approximately 11-14% of collagen by
mass(58). For this reason, it is often used as a marker for collagen and collagen degradation. However, it
is also found in much lower quantities in elastin.
A complete collagen molecule, or fibril is approximately 300nm in length. Fibrils are cross-linked
together with covalent bonds to form indefinitely large structures. There are a total of 29 known
different types of collagen, differentiated by the compositions of their α-chains. Type I collagen is the
most commonly found collagen type throughout the body, appearing in the skin, vasculature, bone, and
scar tissue. Blood vessels also contain a fair amount of Type III collagen, which appears as reticulated
fibre. Type I collagen is made up of two identical α1(I) chains and one α2(I), while Type III collagen is
made up of 3 identical α1(III) chains(59).
1.4.2 Collagenase
Collagen tends to be a very rigidly bound molecule that is resistant to degradation by standard tissue
proteases(60). Since there are a number of biological applications where extracellular matrix would
need to be degraded, it follows that there is a need for endogenous mechanisms to achieve this. A
subgroup of the family of enzymes known as matrix metalloproteinases (MMP) is able to degrade
collagen. Members of this subgroup are termed collagenases.
20
1.4.2.1 Human Collagenase
Until 1962, no endogenous mammalian molecules were found that were capable of cleaving the triple
helical structure of collagen. A few of the MMPs in humans, particularly MMP1, MMP8, and MMP13
have since been found to be capable of degrading collagen. MMP1 is the most abundant and
ubiquitous. Human collagenase tends to cleave Type I collagen at a point called the ¾ position of the
molecule, and is capable of cleaving all three strands in the helix at this one locus, leaving two segments,
one of 71 kDa and the other of 24 kDa, both of which can be further digested by trypsin and other non-
specific proteases(60,61).
1.4.2.2 Bacterial Collagenase
Prior to 1960, only enzymes found in bacteria were known to be capable of degrading collagen. The
most well-characterized bacterial collagenase is derived from Clostridium Histolyticum. This enzyme has
proven to be extremely effective at degrading all types of collagen. It is able to cleave collagen
molecules predominantly at sequences that include GLY–PRO-X-GLY-PRO-Y, with cleavage occurring
between the X-GLY bond(62). X and Y represent any amino acid. This sequence is very common in
collagen molecules, allowing for scission into numerous small polypeptide chains. This allows for
complete and rapid digestion of collagen without the aid of other proteases, making it very attractive for
use in therapeutic applications where collagen is to be degraded with exogenous delivery of
collagenase.
1.5 Ultrasound
1.5.1 Imaging
Ultrasound has been used as a medical imaging modality for several decades (63). Ultrasound imaging as
it is used clinically is based on the principle of pulse-echo. This is similar to the concept of Sound
21
Navigation and Ranging (SONAR), where pressure waves are transmitted into the water and the echoes
reflected from objects in the water are received. The distance from the object is determined by:
Equation 1-1
where d represents the distance between the source and the object, co represents the speed of sound in
water, and t represents the time between the pulse being sent and the echo being received. As the
pressure wave propagates into the medium, it encounters boundaries between different objects along
its propagation path. As it encounters a boundary between two different propagation mediums, a
portion of the energy in the wave is reflected, and a portion of the energy is propagated into the next
medium. A medium can be characterized by its acoustic impedance. An object’s acoustic impedance (Z)
is given by the equation:
Equation 1-2
where is the density of the material and is the speed of sound in the material.
Acoustic impedance is a measure of the amount of force it takes to induce a given particle velocity in the
medium and is a bulk property of a material. The proportion of incident energy reflected from a surface,
assuming normal incidence is given by the reflection coefficient (R):
Equation 1-3
where Z1 is the acoustic impedance of the initial medium and Z2 is the acoustic impedance of the
medium being transmitted into. The proportion of incident energy transmitted into the second medium
is given by the transmission coefficient (T):
Equation 1-4
22
Equation 1-5
1.5.1.1 Ultrasound Transducers
An ultrasound signal is generated by a device known as an ultrasound transducer. This is an
electromechanical device that is used for both transmission and reception of acoustic pulses. Clinical
ultrasound transducers generally work using the piezoelectric effect. When a piezoelectric material is
compressed or expanded, an electric field is produced. Conversely, if an electric field is applied to a
piezoelectric material, the material will correspondingly experience a constriction or expansion. It
therefore follows that to create an acoustic wave, or pressure wave, an electric field is applied to an
ultrasound transducer. The wave interacts with various tissue boundaries, sending reflections back to
the transducer. These reflections cause an expansion/constriction in the transducer, which leads to the
generation of an electric field which can be detected and processed for display.
1.5.1.2 Imaging tradeoffs
Selecting which ultrasound parameters need to be used for a particular imaging application requires the
optimization of a series of tradeoffs. One tradeoff that must be made is resolution vs. penetration
depth. Lower frequency acoustic waves are able to penetrate more deeply into tissue than high
frequency waves, but lead to lower resolution images. Another tradeoff that must be dealt with is
sensitivity vs. pulse bandwidth. A broadband pulse (i.e. a short pulse in time) is essential for high
resolution in the axial direction (the direction of wave propagation). The lateral resolution, or the
resolution in the direction perpendicular to wave propagation can also be optimized depending upon
the application. Acoustic beams can be focused to improve lateral resolution. A spherically focussed
transducer is shown in Figure 1-6.
23
Lateral resolution is defined by how narrow the beam waist is at the focus. The lateral resolution is often
reported as the -6 dB (decibel) beamwidth. This is the width of the beam at a point 6 dB below the peak,
often referred to as the Full Width at Half Maximum (FWHM). For a spherically focused transducer, this
value can be estimated using the equation(64):
Equation 1-6
where Rlat represents the -6dB beamwidth, r represents the radius of curvature of the transducer, λ
represents the wavelength of the acoustic wave (inversely proportional to the frequency), and a
represents the aperture, or diameter of the transducer face for a circular shaped aperture. F represents
the F-number of the transducer, and is the ratio between the radius of curvature and aperture size. A
transducer with a lower F-number will have a better lateral resolution at the focus than an equivalent
transducer of higher F-number.
24
Figure 1-6 - Schematic of a focused ultrasound transducer and associated beam. Depending on the strength of focus, or F-number, the width of the focus will change. This directly affects the lateral resolution.
However, there is a tradeoff that comes with higher lateral resolution as well. Along the beam axis, the
pressure both before and beyond the focus is lower than the peak (which occurs at the focal point). The
FWHM along the beam axis is referred to as the Depth of Field (DOF) and is a quantity that relates to the
axial depth range for which the transducer receives adequate signal for imaging. This quantity can be
estimated for a spherically focused transducer as(64):
Equation 1-7
Conversely with lateral resolution, DOF increases with a higher F-number.
25
1.5.2 Blood flow imaging
A technique to detect moving structures, particularly blood cells, is vital for ultrasound as an imaging
modality. It allows for the estimation of blood flow, detection of microvessels, and the imaging of
stenotic vessels. Various techniques are used for blood flow imaging, the simplest of which depends on
the Doppler Effect. The Doppler Effect, a theory first presented by Christian Andreas Doppler in 1843,
states that the frequency of a wave being transmitted from a moving source, or being received by a
moving receiver, will be shifted by an amount proportional to the velocity of the moving object(65). The
shift in frequency is approximated by the following expression:
Equation 1-8
where f represents the received frequency, f0 represents the transmitted frequency, v represents the
relative velocity between the transmitter and the target, c represents the speed the sound wave travels
in the medium, and ϴ represents the angle between the acoustic beam and the moving object. The
Doppler Effect is only directly used in Continuous Wave (CW) ultrasound imaging systems, meaning that
detected frequency shifts can come from anywhere along the acoustic beam path, and cannot be
localized to small regions in space.
1.5.2.1 Pulsed wave Doppler
Pulsed wave Doppler (PW) systems allow the user to gate windows of tissue regions, sending short
pulses down the selected signal line. Only moving scatterers from the selected region are considered,
allowing the identification of flow in a particular region on an image. In this case, it is not the Doppler
shift that is detected, but rather a phase difference between successive received pulses, most commonly
estimated using an autocorrelation function(65).
26
1.5.2.2 Colour & Power Doppler
Although PW systems are useful in providing data about flow, or velocity at a single location, it is often
desirable to understand the nature of flow through an entire region in space. This is done by measuring
PW shifts over an entire ultrasound image, and displaying them as mean particle velocity in each region.
This is referred to as Colour Flow Doppler (CFD). By convention, scatterers detected as moving towards
the transducer are coloured red (with varying shades representing different velocities), and those
moving away from the transducer are coloured blue(66).
While CW, PW and CFD modes all measure velocities, another technique, known as Power Doppler (PD)
is often used. This technique essentially measures the total volume of moving scatterers in a region of
interest. This is done by first filtering out all non-moving objects using a user-modifiable wall filter, and
then integrating the remaining power in a pulsed scheme similar to what is used in PW or CFD. PD is
significantly more sensitive to slow flow than CFD since it sums the power components from the entire
Doppler spectrum instead of trying to find a mean value(67). Also, since blood velocities are not
measured using this technique, a lower Pulse Repetition Frequency (PRF) can be used(66).
1.5.3 Ultrasound mediated microbubbles (UMM)
Small, microscopic encapsulated bubbles have been used as clinical contrast agents for ultrasound
imaging for over 20 years. The bubbles consist of a gas core surrounded by a shell, usually either a lipid
or polymer. The bubbles are generally <10µm in diameter.
1.5.3.1 Contrast Imaging
As a result of the large difference in acoustic impedance between the gas core of microbubbles and
water based surrounding tissue, microbubbles oscillate very actively in an acoustic field, producing a
very apparent signal on ultrasound images. Furthermore, microbubbles, if driven by adequately high
acoustic pressures, oscillate non-linearly. This property provides the basis for several techniques specific
27
to detecting microbubbles, and isolating them from surrounding tissue(68-70). Maximum oscillation
occurs at the resonance frequency of the bubble.
Primarily, UMM are currently used clinically in echocardiography for a number of applications, including
visualizing ventricles and endocardial borders(71), as well as measuring myocardial blood flow(72) and
vascular stenoses(73). They also have several applications in radiology, including screening and staging
of liver tumours(74).
1.5.3.2 Therapy with microbubbles
A particularly unique characteristic of microbubbles compared with contrast agents of other modalities
is that they can be disrupted by the acoustic pulses used to image them. This characteristic gives rise to
a number of issues, both beneficial and detrimental. There are several different modes of disruption.
One mode, which is specific to polymer shelled bubbles, involves the shell of the bubble being cracked,
allowing the gas core to leak out and form a free gas bubble(75). These free gas bubbles will eventually
diffuse into the fluid, and this diffusion may be assisted by the acoustic field. A second mode involves
bubble fragmentation into smaller bubbles. A third mode, which is of particular importance to the
studies detailed in Chapter 4, is known as inertial cavitation. This occurs when the bubble experiences a
significant and rapid radial expansion, followed by a violent collapse. This collapse results in extreme
temperatures and pressures. Surrounding structures, including tissues may be damaged through the
production of free radicals. A phenomenon known as microjetting also occurs when cavitation occurs
next to a rigid boundary. These microjets are small high velocity streams of fluid that are directed into
the boundary, and can cause significant bioeffects(76).
Microbubbles are more easily disrupted under certain acoustic conditions. The concept of Mechanical
Index (MI) is often used to determine the risk of cavitation events. MI is defined as:
28
Equation 1-9
where MI is the Mechanical Index, Pneg represents the peak rarefactional pressure in MPa, and f0
represents the centre frequency of the transducer in MHz. However, recent studies have suggested that
for cavitation in microbubbles, MI is an oversimplification of parameters, since it does not take into
account pulse lengths and PRF, which have been shown to have significant effects on cavitation
thresholds(77). Since the accurate modelling of shell parameters is very difficult, most studies on
cavitation have been performed experimentally. In general, high peak rarefactional pressure, high PRF,
and low frequency acoustic pulses tend to facilitate microbubble disruption. For the contrast agent
Definity™ specifically, at 1MHz, with a dilution of 6µL/1mL, and a pulse length of 10 cycles, and a PRF of
1KHz, 95% of bubbles were disrupted at 0.2MPa(78).
While care needs to be taken to avoid various bioeffects in most clinical scenarios, there are a number
of situations where controlled bioeffects can be beneficial. Microbubble disruption has been shown to
create small pores in cell membranes to transiently increase drug uptake in an effect known as
sonoporation (79,80). Microbubbles have also been used to temporarily open up the blood brain barrier
to allow drug delivery into the brain(81-83). A third application is the use of UMM to increase the
penetration or access of a drug or enzyme to its target or substrate. UMM have been used to increase
the effectiveness of thrombolytic agents, such as tissue plasminogen activator (t-PA,) for clot lysis in
stroke applications(84,85). While the mechanism is not completely understood, it appears that the
microbubbles are able to create small perforations in the clot to increase the available surface area for t-
PA to act, as well as increasing the penetration of t-PA into the clot through microstreaming and
microjetting. This is currently a very active area of research.
1.6 Thesis aims
29
The focus of this thesis is to examine novel ways in which ultrasound may be used as an aid for
revascularizing Chronic Total Occlusions with percutaneous approaches. Beyond this primary objective,
we also aim through the use of animal models, to increase understanding of the pathology of CTO as
they develop and mature.
Certainly, many standard b-mode imaging or disruptive high energy standalone ultrasound approaches
have been attempted to guide and/or mechanically assist percutaneous revascularizations. However,
these approaches have been largely unsuccessful. The techniques proposed here generally build on pre-
existing techniques that have had some success, rather than attempting to provide a standalone
solution to the CTO problem.
In chapter one, the necessary background for appropriate understanding and contextualization of the
techniques and ideas proposed in the later chapters was provided, beginning with an overview of
occlusive vascular disease and its eventual progression into CTO. This led to discussion and justification
of the need for percutaneous revascularization for CTO. The difficulties posed by CTO and current
strategies for overcoming those difficulties were also discussed, one of which is an experimental enzyme
therapy technique using collagenase which will be expanded upon later in this thesis. In order to provide
adequate background on this therapeutic technique, the structure of the collagen molecule and two
different collagenase enzyme formulations was covered. As this thesis is based upon ultrasound
techniques to aid CTO guidewire crossing, some of the fundamentals of medical ultrasound were
examined including basic imaging principles, blood flow detection, and the use of contrast agents as
both imaging and therapeutic tools.
Chapter two details a study of microvascular formation in a porcine model of CTO. The study aims to
show the presence of longitudinal recanalization channels in CTOs, and the ability of high frequency
30
ultrasound to detect them in vivo. Also included is a comparison with MRI detection of similar
microchannels. Histology and ex vivo µCT are used as correlates.
Chapter three shows the design and use of a novel method of quantitatively determining how difficult it
is to puncture the proximal fibrous cap of CTOs. This technique is applied to study the proximal cap
puncture force as a function of CTO age. This technique will be a vital tool in assessing the therapeutic
effect of strategies to change the compliance of CTOs. While this is not an ultrasound based technique,
it will help us further characterize CTO, and is necessary for the subsequent work presented in Chapter
four.
Chapter four introduces a novel adjunct to the use of collagenase as a therapeutic tool for softening the
proximal end of CTO. The use of ultrasound mediated microbubbles to increase the ability of the
collagenase enzyme to soften the proximal cap is shown. The effect of softening is measured using both
biochemical assays and the puncture force measurement described in Chapter three.
Chapter five provides an overall summary to the thesis as well as discussion on the possibilities that exist
for expanding the work presented in Chapters two through four.
1.6.1 Specific Aims of the Thesis
a) To examine a porcine model of Chronic Total Occlusion to further understand various patterns
of neovascular formation.
b) To investigate the characteristics of recanalization channels in CTO to determine if they are
simply unoccluded regions, or true functional channels.
c) To study the ability of high frequency ultrasound to detect neovessels in CTO in vivo
d) To design and build a system for reliably quantifying the force required to puncture the proximal
fibrous cap of CTOs.
31
e) To investigate the changes in the compliance of the proximal end of a CTO as it matures from an
early thrombus rich lesion to a fibrotic lesion
f) To determine if the use of ultrasound mediated microbubbles can enhance collagenase therapy
for softening the PFC of CTOs.
32
2 Microvascular Study of Chronic Total Occlusion in a Porcine Model1
2.1 Introduction
Although great strides have been made over the last few decades in percutaneous treatment of lesions
in coronary artery disease (CAD) and peripheral arterial disease (PAD), several challenges still remain.
Chronic total occlusions (CTOs) are defined as complete occlusions > 3 months in age. They remain a
significant limitation to PCI with reported success rates of 60-70% in CAD(22) and 80% in PAD(86).
Although successful CTO revascularization significantly improves symptoms of angina and claudication,
may improve LV function(26) and possibly mortality(27), the expected low success rate discourages
operators from attempting percutaneous revascularization with only 12% of coronary CTOs and 13% of
peripheral CTOs being treated percutaneously(87). The principle cause for unsuccessful angioplasty in
cases of CTO is the inability to cross the lesion with a guidewire to reach the distal segment of the
artery. Failed guidewire crossing can generally be attributed to two main factors. First, the proximal and
distal ends of the lesion tend to contain densely packed fibrous tissue, creating a barrier that is
physically difficult to penetrate with a guidewire(22). Second, guidance poses a large problem for
angioplasty in CTO. Current imaging techniques do not provide adequate contrast between the walls of
the artery and occlusive plaque resulting in a significant risk of vessel wall perforation(38). Furthermore,
adequate imaging of CTO features that may influence treatment strategy is not achieved by fluoroscopy.
Recent animal studies have indicated that specific lesion characteristics may portend higher success
rates in CTO interventions. In particular, the presence of intraluminal microvessels seem to be a
favourable predictor of successful guide-wire crossing, possibly by serving as a pathway for advancing
1 This chapter is adapted from the following publication: Thind AS, Leung G, Munce NR, Graham JJ, Anderson KJ,
Dick AJ, Strauss BH, Wright GA, Foster FS. Investigation of micro-ultrasound for microvessel imaging in a model of
chronic total occlusion. Ultrason Imaging 2007;29(3):167-81.
33
the guidewire across the occlusion(34). These microvessels are present in approximately 80% of all CTOs
(31,32). An imaging tool that is able to detect microvessels within CTOs in real-time in vivo – specifically
those that run parallel to the parent vessel providing a path for guidewire passage, also known as
recanalization channels - may provide an effective method for guiding interventional procedures.
Angiography, the current imaging technique used in percutaneous revascularizations, cannot reliably
detect microvessels due to lack of resolution limited to ~300µm.
The aim of the current study is to evaluate the ability of micro-ultrasound (μUS) to accurately identify
microvessels in a porcine CTO model. In this study, we examine microvascular characteristics and
features of our CTO model ex vivo with cross-sectional histology and µCT. These features are compared
with those seen in vivo with μUS and MRI with their respective flow sensitive techniques. Various
interesting features of CTO microvascular formations are also shown and discussed.
2.2 Methods
2.2.1 CTO Model
The polymer plug porcine model of CTO was briefly described in 1.3.5. The polymer plugs used in this
study have a donut-like shape and are approximately 3mm in diameter. After percutaneous placement,
the polymer plug swells, becoming a nidus for an acutely occlusive thrombus that later organizes into a
fibrotic CTO containing a variable number of microvessels (34). In the present study, occlusions were
induced in a superficial branch of the femoral artery of the animals. The superficial femoral artery (SFA)
is located approximately 2-3mm from the surface of the skin in swine at their most superficial location.
Occlusions at this depth are palpable transcutaneously. Placing the occlusion at this depth was ideal for
both transcutaneous high frequency ultrasound imaging, as well as for MR imaging. For the purposes of
this study, intraluminal microvessels were defined as blood carrying structures in the longitudinal axis of
the artery for a distance of greater than 5mm. The occlusions were studied at one week after induction
34
and at 8 weeks, immediately prior to sacrifice. After sacrifice, samples were analyzed for histological
features.
Attempts were made to place polymer plugs in both the right and left femoral arteries of 6 animals, with
a total of 9 CTOs successfully created (Table 2-1). The percutaneous placement of polymer plugs was
guided angiographically using an OEC 9800 Digital Mobile Imaging System capable of real-time 1k x 1k
imaging. Angiograms were performed before and immediately after polymer placement, as well as just
prior to sacrifice, as shown in Figure 2-1. The arteries tended to occlude back to the nearest major
branch proximal to the deployment location of the polymer. All occlusions were persistent after 8
weeks.
Table 2-1 - Summary of vessel imaging by modality
Modality Left (n= )
Right (n= )
Total (n= )
CTOs created
3 6 9
MRI 1 6 7
μUS 2 6 8
μCT 0 2 2
Histology 2 5 7
IHC 0 2 2
Legend: n – Number of vessels studied
2.2.2 High Frequency Ultrasound (Micro-ultrasound)
In vivo ultrasound imaging was performed using a Visualsonics Vevo 770 high frequency micro-
ultrasound (μUS) imaging system. A Real-time Microvisualization (RMV)-704 transducer operating at 40
MHz with a 100% two-way bandwidth and an F-number of 2.0 was used for imaging. This transducer can
35
achieve resolutions of 40μm axially and 80μm laterally. One of the salient features of ultrasound is its
ability to detect motion – specifically blood flow – with great sensitivity using Power Doppler(67).
Figure 2-1 - Angiograms of right leg of porcine subject. The vessel that was occluded is the superficial branch of the femoral artery. A) shows the region before deposition of polymer. Vessel to be occluded located in yellow ellipse. B) shows the region immediately after the deposition of the polymer, with yellow ellipse indicating region of occlusion.
The animals were anaesthetized and placed in the supine position. The transducer was coupled to the
skin of the animal using heated ultrasound gel. The imaging probe was connected to a Visualsonics 3D
motor, which was attached to a stabilization arm. The artery was imaged with 300 B-Scan images
acquired in succession and spaced 50μm apart. This spacing was achieved with the aid of a computer
controlled linear translation motor. A 3D image reconstruction was generated from individually acquired
frames. Individual B-Scan images were acquired in Power Doppler mode with a clutter filter threshold
set at 2.0mm/s. As with all Doppler techniques, the angle subtended between the transducer and the
vessel was required to be non-perpendicular, and was controlled using the stabilization arm. Image
processing and reconstruction was performed using Visualsonics Vevo software. Once 3D images were
acquired, microvessels were isolated for emphasis and rendered in 3D to show continuity.
36
High frequency ultrasound imaging was performed at the 1 week time point and just prior to sacrifice in
8 of the 9 CTO arteries. One case was excluded due to technical difficulties associated with scanning
system.
2.2.3 MRI
MR imaging was performed using a 3T GE EXCITE scanner. A single turn, custom built surface coil with
dimensions of 5x3 cm was placed directly over the palpable lesion. Two dimensional gradient echo
phase contrast and T1 Imaging were performed with the following imaging parameters TR/TE/alpha =
55ms/7.2ms/25. 15.6 kHz read bandwidth and a velocity encoding gradient of 15 cm/s. An acquisition
matrix of 256x192 over a 3 cm field of view and 14 three millimetre thick slices yielding voxel
dimensions of 0.11 x 0.15 x 3 mm was acquired in 5 minutes. Flow images were acquired in each
orthogonal axis and a composite flow map along with a magnitude image were reconstructed.
Images were also acquired prior to and two minutes after the injection of an intravascular contrast
agent (Clariscan, GE Healthcare USA). A linear fill, 3D fast spoiled gradient echo sequence was used with
the following image acquisition parameters (TR/TE/Alpha = 9.1/2.8/30) 31.25 kHz read bandwidth. A
three dimensional volume encoded with a 320x320x26 acquisition matrix over an 8x8x1.56 cm field of
view yielded voxel dimensions of 0.25x0.25x0.6mm. A total of 6 averages were taken of the entire
volume to increase SNR for a total imaging time of 8 minutes 47 seconds.
MR imaging was performed on 7 of the 9 CTOs available. In 2 cases of bilateral femoral artery CTOs, MR
imaging could only be performed in one CTO.
2.2.4 MicroCT (μCT)
Prior to sacrifice, selected samples were perfused with Microfil™ (Carver Michigan, USA) to demonstrate
the location of microvessels at high spatial resolution using μCT. Microfil™ is a radiopaque polymer that
37
was injected into vessels under approximately 80mmHg. It was allowed to harden in the vessels and
provided a roadmap of vasculature, making it effective as a demonstration of microvessel continuity.
Samples were scanned with a MS8 cone-beam CT (GE Healthcare, London, Canada), which is capable of
resolving structures to a resolution of 17μm(88) (both the 17μm and 35μm modes were used in
imaging). The samples were fixed with formalin and then embedded in agar to ensure a rigid orientation
during scanning.
2.2.5 Histological Processing
After sacrifice, the occluded arteries were excised and placed in a 10% buffered formalin solution at 40C
until histological processing was performed. The arteries were embedded in paraffin and cross-sections
were prepared every 5mm. 4μm thick sections were stained first with H&E to identify cellular
components (pink) and nuclei (purple). Subsequently, select samples were also stained with elastic
trichrome to identify elastic tissue (black), collagen (blue), as well as muscle and blood (red). All
histology slides were scanned using a white light slide scanner (Aperio Technologies, Vista, CA) under
20x magnification to allow for high resolution and large field-of-view images. The following parameters
were measured in 47 arterial cross-sections: total arterial Cross sectional area (CSA), lumen CSA and
intraluminal microvascular area. In addition, the ratios of lumen CSA: total vessel CSA and microvessel
CSA: lumen CSA and the mean intraluminal microvessel number were computed. The total vessel CSA
was defined as the area bounded by the EEL. The lumen CSA was defined as the area bounded by the
IEL. Intraluminal microvascular area was defined as the area of all intraluminal vascular structures (lined
by endothelial cells and/or containing microfil). Intraluminal microvessels were defined as vascular
structures extending more than 5mm along the longitudinal axis of the vessel.
In two CTOs, microvessels were identified by immunofluorescensce using porcine CD31, an endothelial-
specific marker (Product MCA1746PE, Serotec). Antibody was conjugated with Cy5 to produce
38
fluorescent emission at 670nm with excitation at 480nm. The slides were imaged under confocal
microscopy with an Argon laser emitting at 488nm.
Histological analysis was performed in 7 CTOs. In two additional cases, the CTO arteries were not
suitable for histologic processing.
2.3 Results
2.3.1 Histology
Intraluminal microvessels were present in 5 of 7 CTO vessels examined. The mean total artery CSA was
2.93mm2, with an equivalent diameter of 1.86mm (assuming a circular configuration). The mean lumen
CSA was 0.93mm2. This corresponds to the lumen composing approximately 32% of the total vessel
area. The average area of intraluminal microvessels was 1.10x10-2 mm2, corresponding to an equivalent
diameter of approximately 9.8x10-2mm. There was an average of 5.5 intraluminal microvessels per
section, covering about 1.5% of the area of the occluded lumen. An intact IEL was seen in 6 of the 7
vessels that were processed histologically. Characteristics of vessels studied are detailed in Table 2-2.
Representative histology with H&E and elastic trichrome stains were used as correlates for the imaging
techniques. Figure 2-2 shows examples of resultant histological images. Figure 2-2A shows an H&E
stained cross-section of a vessel that has been perfused with Microfil. Microfil (MF) can be seen as black
regions showing microvessels (MV) within the plaque (P), the media (M), and also in the vaso vasorum
(VV) located in the adventitia. Figure 2-2B shows a different vessel stained with an Elastic Trichrome,
with well-defined layers of the artery. The IEL and EEL are seen bounding the plaque (P) and the media
(M) respectively. Figures 2-2C and 2-2D show confocal microscope images of the vessel in Figure 2-2B at
5x and 20x respectively with a CD31 stain. This staining indicates that the cells lining the microvessels
are, in fact, endothelial.
39
Table 2-2 - Microvascular characteristics of CTO arteries from histology
Vessel # 1 2 3 4 5 6 7 Total
Number of
cross sections
10 3 3 12 2 5 12 47
Artery Area
(mm2)
2.64 1.42 1.24 4.26 5.74 3.81 1.42 2.93
Artery
Diameter (mm)
1.83 1.34 1.26 2.33 2.70 2.20 1.34 1.86
Lumen Area
(mm2)
0.83 0.33 0.37 1.32 2.42 0.89 0.36 0.93
Lumen
Diameter (mm)
1.03 0.65 0.69 1.30 1.76 1.06 0.68 1.02
Lumen Ratio
(%)
31.3 23.6 29.8 31.0 42.2 23.3 25.6 31.8
Endoluminal
MV area (mm2)
1.50e-2
3.04e-3
1.43e-2
4.03e-2
5.98e-4
3.01e-3
9.37e-4
1.10e-2
Average
Endoluminal
MV count
0.3 2.7 1.7 6.6 14.5 5.8 6.9 5.5
MV Diameter
(mm)
1.38e-1
6.22e-2
1.35e-1
2.26e-1
2.76e-2
6.20e-2
3.45 e-2
9.80 e-2
MV Ratio (%) 1.8 0.9 3.9 3.0 0.0 0.3 0.3 1.5
2.3.2 Micro-ultrasound (μUS)
The characteristic pattern of the occluded vessel lying between two immediately adjacent veins made
the lesions relatively simple to follow. A cross-sectional B-scan ultrasound image of an occlusion is
shown in Figure 2-3A. The media is clearly seen as a dark band around the occlusion. The characteristic
veins are seen adjacent. Micro-ultrasound in all 8 imaged arteries clearly differentiated between
occlusive plaque and arterial media.
40
Figure 2-2 - Histological images of CTO arteries containing intraluminal microvessels. A) shows an H&E stained vessel containing a single large microvessel (MV). This vessel was perfused with Microfil (MF) just prior to sacrifice, which can be seen as black portions within microvessels. The plaque (P) and media (M) are also marked. Bar represents 500μm. B) shows an Elastic Trichrome stain of an occluded artery with a blood filled microvessel (MV). Here, the plaque (P), media (M), and Internal (IEL) and External (EEL) Elastic Lamina regions are clearly shown for easier delineation of layer boundaries. Bar represents 500μm. C) shows the vessel in B with a PE-Cy5 fluorescent labelled CD31 stained antibody imaged at 5x. The microvessel (MV) has endothelialized and consequently fluoresces. The elastic layers (IEL and EEL) autofluoresce and are also seen. D) shows a 20x image of the same sample shown in C), showing in more detail the microvessel (MV) and the IEL.
Power Doppler image overlays were used to locate microvessels on all imaged arteries. Figure 2-3A
shows a B-scan image, while Figure 2-3B shows the same image with a Power Doppler overlay.
Using the 3D reconstructions of sequentially acquired cross sectional images, recanalization channels >
5mm in length were observed in 6 of the 8 vessels imaged. This compared well to histology, where 5 of 7
vessels showed these channels (with ultrasound, the same 5 of 7 vessels also showed these channels). A
long axis slice of such a reconstruction is shown in Figure 2-4, both with and without a Power Doppler
overlay (Figure 2-4A and 2-4B respectively). The microvessel seen in the Power Doppler overlay image is
41
seen traversing through the image plane. Segments of the microvessel that are not seen in frame may
either be out of plane or may not be detected due to a Power Doppler phenomenon known as the
spatio-temporal artifact (89). Flow profiles were examined with pulsed wave Doppler readings showing
pulsatile flow of at least 0.5 cm/s (peak) before they were deemed microvessels. An example of such a
Doppler reading is shown in Figure 2-3C, showing dampened pulsatile flow with a 2 cm/s peak velocity.
Although microvessel diameters cannot accurately be measured directly with ultrasound using Power
Doppler analysis, it is likely that vessels being detected are of diameters as low as ~30μm (90).
Figure 2-3 - Cross sectional in vivo B-scan images of a porcine CTO Artery. The artery is flanked on either side by a vein. A) shows standard B-scan ultrasound image including veins (V), occlusive plaque (O), and medial layer of artery (M) which appears as a dark, echolucent band. Bar represents 1mm. Image in B) shows ultrasound B-scan with Power Doppler overlay. Power Doppler wire frame encompasses occluded artery and one adjacent vein. Regions in colour represent areas with blood flow. An intraplaque microvessel (MV) is shown which is not visible in A) and is similar in size to those shown in Figure 2-2. Bar represents 1mm. C) shows a Pulsed Wave Doppler graph of flow in the centre of the microvessel shown in B) revealing dampened pulsatile flow within the microvessel.
42
Figure 2-4 - Longitudinal axis in vivo B-scan & MR images of a porcine CTO Artery. The artery is the same as shown in Figure 2-3. A) shows a reconstructed long axis B-scan ultrasound image including occlusive plaque (O) and medial layer of artery (M). Image in B) shows ultrasound B-scan with Power Doppler overlay. Power Doppler readings from outside artery are as a result of motion at the skin surface. An intraplaque microvessel (MV) is shown. C) shows a contrast enhanced T1 weighted MR image of coronal view of the same vessel. The vessel contains a faint signature of a microvessel (MV) running in-between the adjacent veins (V). Bar represents 1cm.
2.3.3 MRI
Similarly, the characteristic pattern of the occluded vessel lying between two immediately adjacent
veins made the lesions relatively easy to locate under MRI. All 7 arteries imaged with MRI allowed for
clear differentiation between the occlusive lumen and the arterial medial layer. Contrast enhanced MRI
detected intraluminal microvessels in 3 of 7 vessels imaged with an example shown in Figure 2-4C.
Figure 2-5A shows a T1 weighted MR image of the same vessel shown in Figures 2-3 and 2-4. Many of
the same features are visible – the occluded vessel and the veins, with greater contrast but lower
resolution than the corresponding micro-ultrasound image. In addition, a region that appears to contain
a well-formed microvessel can be seen. The use of phase contrast imaging is a common method for
detecting motion with MRI. Figure 2-5B shows a phase contrast image of a cross section of the same
43
occlusion shown in Figure 2-3 and its adjacent veins. In the phase contrast image, dark regions represent
velocity in one direction, and bright regions represent velocity in the opposite direction. The two dark
regions in the image show significant blood flow in the veins. However, within the artery, no discernable
flow is apparent despite the fact that a microvessel is known to be present.
Figure 2-5 - MRI Images of CTO cross section. A) shows a T1 weighted image of an occluded vessel flanked by veins in the same vessel shown in Figure 2-3. Upon comparison with Power Doppler imaging and histology, the area marked (MV) is presumed to be a microvessel. The corresponding phase contrast image shown in B) is not sufficiently sensitive to detect flow in the region.
2.3.4 Intraluminal Microvessel CTO Features
Features are mentioned only if they appeared in at least two samples, and were seen under at least two
modalities.
44
2.3.4.1 Corkscrew pattern
2 of the 9 vessels studied showed corkscrew-like pattern as shown in Figure 2-6 under µUS and μCT.
Figure 2-6A shows a 3D rendering from reconstructed µUS B-scans along with a portion of one of the
adjacent veins for reference. Figure 2-6B shows a 35μm resolution μCT image of a similar effect.
Figure 2-6 - Corkscrew pattern in microvessels. A) shows a 3D rendering of a microvessel (MV) with a partial corkscrew morphology, imaged in vivo. Portion of adjacent vein (V) also shown. Bar represents 1mm. B) shows a rendered isosurface of a similar microvessel
2.3.4.2 Crescent shaped channel
In 5 of the 9 vessels studied, a microvascular pattern resembling a crescent shape was seen in cross-
section under μUS and/or histology. These channels extended for lengths of greater than 1 cm
longitudinally. This feature is shown in Figure 2-7. Figure 2-7A shows a 3D rendering of an ultrasound
image reconstruction with a Power Doppler overlay. Figure 2-7B shows an H&E stained cross-section of
a similar pattern. The entire crescent shape remains intraluminal. Figure 2-7C shows a 20x magnification
of the slide shown in Figure 2-7B, showing a lining of organized endothelial cells around the channel.
Figure 2-7D shows a μCT image of this feature as well as showing connections to the vaso vasorum
45
around the periphery. A planar section taken through the dashed green line represents a similar section
to that shown in Figure 2-7A and Figure 2-7B.
Figure 2-7 - Crescent shaped morphology. A) shows an ultrasound cross sectional enface view of a 3D rendered microvessel (MV) in a characteristic crescent shape acquired in vivo. Bar represents 1mm B) shows an H&E cross sectional slice of a similar morphology with a crescent shaped microvessel. Tearing artifact (A) from processing is also marked. Bar represents 500μm. C) shows a magnified view of the cross section shown in B).
46
2.3.4.3 Communication outside artery
In 3 of 9 arteries, longitudinally running microvessels were seen to branch off and communicate with
vessels outside of the artery under μUS. Figure 2-8 shows this feature. A μUS B-scan image with a Power
Doppler overlay is shown in Figure 2-8A. The microvessel running through the vessel is seen to branch
outside of the artery’s boundaries. A rendered μCT isosurface of the same vessel shown in Figure 2-8A is
shown in Figure 2-8B showing a microvessel with multiple branch points leading to the vaso vasorum. A
planar section taken through the dashed green line represents a similar section to that shown in Figure
2-8A.
Figure 2-8 - Branching from within occlusion. Part A shows an in vivo cross sectional ultrasound B-scan image with a Power Doppler overlay. Microvessel (MV) in occlusion connects to branch (BR) outside of vessel. Adjacent vein (V) is also depicted. Bar represents 1mm. B) shows similar morphology seen with μCT rendered isosurface of a vessel running parallel to the axis of the artery in the same animal. Dashed green dashed line indicates similar planar section to that shown from A).
Results for all modalities are detailed in Table 2-3.
47
Table 2-3 - Vessel imaging details. Boxes marked with “•” denote presence of feature, boxes marked with “×” denote that feature was not detected, and N/A implies that technique was not attempted on sample.
Vessel # Histology µUS MRI
MV Cr Co Cs MV Cr Co Cs MV Cr Co Cs
1 N/A N/A · × × ×
2 × × × × × × × × × × × ×
3 · · × × · × × × N/A
4 · × × × · · · · · · × ×
5 · × × × · · · · · · · ×
6 × × × × × × × × × × × ×
7 · · × × · · × ×
×
×
×
×
8 · · × × · · · · × × × ×
9 N/A · · × × × × × ×
Totals: 5 3 0 0 6 5 3 2 3 2 1 0
2.4 Discussion
Micro-ultrasound and magnetic resonance imaging provided excellent anatomical visualization of a
superficial model of chronic total occlusion in pigs. The essential feature of ultrasound in imaging CTO in
vivo was its ability to image blood flow velocities as low as 2mm/s. This allowed for excellent detection
of intraluminal microvessels that are within the penetration depth of the transducer, as shown in table
3. However, since the penetration depth of ultrasound at the frequencies used is only ~5-10mm in the
tissues present in CTOs, only a small region in front of the transducer could be studied. High frequency
ultrasound as described here is currently unsuitable for in vivo coronary application. However, the
extension of the principles studied here to intravascular ultrasound (IVUS) would overcome the
penetration depth limitation. Efforts are in progress to develop forward-looking IVUS to assist guidance
48
of coronary interventions. To date, the majority of this work has been focused on maintaining
intraluminal position of the guidewire by detecting the boundaries of the blood vessel using B-scan
imaging and providing instant feedback to the operator. However, the addition of Doppler or other
blood flow detection techniques – whose feasibility was demonstrated here – to identify intraluminal
recanalization channels could facilitate guidewire crossing by revealing potential pathways. As a more
immediate solution, conventional Doppler IVUS probes could be placed in veins running parallel to
occluded vessels and pulled back to form 3D images detailing lesion vasculature.
In contrast to μUS, MRI provided excellent imaging at all points along the CTO, including those well
below the surface of the skin. However, contrast enhanced MRI was not able to detect microvessels as
consistently as μUS, presumably because the sensitivity to blood flow is lower, and only larger
microvessels were detected (Table 2-3).
Although the peak velocities are often relatively high (1-5 cm/s), phase contrast images were not able to
detect small microvessels. This likely occurred because the mean blood velocities were considerably
lower in the pulsatile flow patterns found in the microvessels of this model. Gating these acquisitions to
the cardiac cycle may improve sensitivity to the time varying flow patterns.
Moreover, future work will need to determine whether the identification of specific morphologic
features by μUS (e.g. crescent-shaped vessels, corkscrew patterns, and connections between vaso
vasorum and intraluminal microvessels) are correlated with successful crossability of CTO. This would
also enhance pre-procedural planning and intraprocedural guidance.
Although histology was likely the most effective technique for definitively determining the presence of
longitudinal microchannels, it was less successful in determining the presence of extraluminal
communication and corkscrew patterns in the occlusions. This can be attributed to the relatively sparse
49
sampling frequency (every 5mm) of histology along the length of the vessels, causing several of these
features to be missed.
In order to be translated into a useful clinical tool, both respective modalities (μUS and MRI) need to be
applied using intravascular approaches.
Intravascular ultrasound (IVUS) has been used to verify stent deployment and characterize plaque
geometry. More recently, forward looking ultrasound(91,92) has been investigated for use in guiding
CTO procedures(93). The majority of these studies have been centred around guiding procedures using
B-scan imaging, with axial resolutions on the order of 100μm reported(94) in 30MHz transducers, with
potential for higher resolutions at the frequencies used in this study. Also, IVUS transducer
development is underway in our lab for further improvements. Vessel boundary detection with B-scan
imaging can be very difficult in the compromised vessel structure present in CTOs to the point where
guidance with B-scan imaging alone is insufficient(41). Early versions of forward looking Doppler probes
have also been developed(95). One potential point of concern is that the motion of the coronary vessels
may cause difficulty in using low threshold clutter filtering schemes with Power Doppler. However, it
may be the case that the intravascular approach can overcome this limitation since the catheter motion
may be synchronized with bulk cardiac motion. Furthermore, this limitation would likely be overcome
with more sensitive techniques, such as contrast agent based harmonic imaging(96).
Intravascular MR coils provide an attractive means of enhancing signal-to-noise ratio (SNR) of MR
images. These receiver coils, typically wound around the tips of catheters, provide a very small noise
sensitivity volume. Combined with very specific placement of these coils close to the region of interest,
SNRs that are significantly higher than those of the externally placed coils used in this study can be
achieved. Many authors have shown the utility and ability to visualize a guidewire(97) and segments of
the introducing hardware under MRI. Additionally, several authors have also shown some initial human
50
in-vivo intravascular experiences using these catheters to image abdominal(98), iliac (99), as well as the
very technically difficult, coronary arteries(100). One of the technical challenges that still remains to be
overcome is the difficulty in obtaining appropriate coil sensitivity patterns to visualize far enough
surrounding the catheter coil, while still keeping these coils small enough to be manoeuvrable. This
geometry problem remains an ongoing area of research(101).
2.4.1.1 Limitations
This model does provide several important features of human CTOs, particularly intraluminal
microvessels and a fibrotic occlusive lumen. However, there are some limitations to the model. First,
since it is induced by the addition of a polymer, the presence of the polymer itself is a deviation from the
natural process of CTO development. The polymer has been seen histologically to persist for longer than
8 weeks. Moreover, the CTO lesions in this model do not demonstrate calcification which may occur in
human coronary CTOs. Finally, since the branch point of the femoral artery is > 1cm below the surface of
the skin, the proximal end of the occlusion is not visible under μUS, as this is beyond the penetration
capabilities of the transducer used.
The ultrasound techniques demonstrated in the study would not likely find direct clinical translation
beyond application in the peripheral vasculature, as imaging is performed transcutaneously, However,
the imaging principles used can be applied to intravascular approaches.
With regards to extending the approach to intravascular and coronary applications, it should be noted
that the presence of a guidewire and catheter in the vessel will limit the flow through microvessels,
making detection more difficult.
2.5 Conclusions
51
The use of a porcine CTO model has provided the opportunity to study the feasibility of using μUS and
MRI to detect microvessels in occluded lesions in vivo. μUS has been shown to be excellent for the
detection of small microvessels (6 of 8 vessels studied contained identifiable recanalization channels).
This may be valuable in providing realtime guidance by translating the technique to an intravascular
ultrasound approach. While MRI has been shown to be less sensitive for identifying small microvessels
(3 of 7 vessels studied contained identifiable intraluminal microvessels), it is excellent for global lesion
imaging and shows excellent potential for detecting vascular volumes. Both modalities may have
complementary roles in the characterization and treatment of CTOs.
52
3 A Novel Method for Measurement of Proximal Fibrous Cap Puncture
Force in Chronic Total Occlusions and its application2
3.1 Introduction
Chronic total occlusions (CTOs), defined as arteries with TIMI (Thrombolysis In Myocardial Infarction)- 0
or 1 flow that have been known to have been completely occluded for ≥ 3 months duration (22,102),
represent a major challenge for revascularization by percutaneous coronary interventions (PCI) due
primarily to two problems: 1) difficulties in navigation due to inadequate visualization and 2) inability to
physically cross the lesions with currently available equipment, including guidewires and crossing
devices. There are numerous obstacles preventing successful navigation of guidewires due to the
heterogeneous composition of CTOs(19).
Advancing age of the CTO has long been recognized as an important predictor of failure to cross the CTO
with a guidewire during PCI(103). Early occlusions (< 6 weeks old) tend to contain thrombus,
proteoglycans and/or lipid-rich tissue, which are easier to cross with softer guidewires(32). However, as
lesions age, the softer components are replaced by dense fibrous tissue and calcification, and
consequently are much more difficult to cross. This necessitates the use of stiffer, specialty guidewires
and crossing devices, which results in an increase in complication rates due to dissections and even
vessel perforation.
A recent study was performed to examine the compositional changes occurring during the maturation of
CTOs in a rabbit model(35). This study showed that collagen-rich fibrotic tissue increases with the age of
the CTO and replaces tissue rich in proteoglycans and thrombus remnants that predominate at early
2 This chapter is adapted from a manuscript currently in press at Eurointervention as: Amandeep S. Thind PhD,
Bradley H. Strauss MD PhD, Aaron A. Teitelbaum MD MSc, Raffi Karshaffian PhD, Michelle Ladouceur, Cari M.
Whyne PhD, David E. Goertz PhD, F. Stuart Foster PhD. A Novel Method for Measurement of Proximal Fibrous Cap
Puncture Force in Chronic Total Occlusions: The Effect of Increasing Age.
53
stages of occlusion. Also identified was a specific geographic site at the entrance of the CTO where a
particularly densely-packed collagen matrix is deposited. This location, the PFC, has long been
recognized as a major barrier to guidewire crossing in clinical angioplasty.
To further understand the physical properties of lesions that would affect the ability of guidewires to
successfully enter the CTO, we have developed a novel methodology to reproducibly quantify the force
required to puncture the fibrous cap in an ex-vivo setting. This measurement can be used to further
characterize CTO age and composition and could conceivably provide important information to assess
therapies to ease guidewire crossing in percutaneous interventions of CTOs.
3.2 Materials & Methods
3.2.1 CTO Model
All animal care and handling were performed in accordance with the guidelines specified by the
Canadian Council on Animal Care (CCAC) for the Care and Use of Laboratory Animals. Approval for
experiments was obtained from St. Michael’s and Sunnybrook Hospital Animal Care Committees.
Arterial occlusions were created in 31 Male New Zealand white rabbits (Charles River Canada, St
Constant, Quebec) (44 arteries), weighing 3.0 to 3.5 kg by injection of thrombin solution (100 IU,
Millipore catalogue # 82-036-3, Kankakee, Ill) into an isolated femoral artery segment, as previously
described (35,41). Ligatures were maintained up to 60 minutes to ensure a persistent occlusion. Animals
were then returned to their cages and fed a regular diet. Animals were sacrificed at 2, 6, 12 or 15 weeks
[n= 6, 8, 7, 23 respectively] following creation of the CTO. Presence of occlusions was verified
sonographically at 2 weeks.
3.2.2 Removal of CTO for Ex-Vivo Testing
54
Prior to sacrifice, animals were anaesthetized and an angiogram was performed to verify persistence of
the occlusion, and to assess lesion location and characteristics. To assist in the mounting of the arterial
segment on the apparatus, a balloon-mounted stent was deployed in the patent artery immediately
adjacent to the proximal end of the CTO. Following sacrifice with euthanyl, the CTO vessel was carefully
excised with the stent in place for proper positioning in the apparatus. The vessel was kept in 0.9%
saline and immediately used for puncture force testing.
3.2.3 Puncture Force Testing
In order to give a reliable measurement of the PFC puncture force, a custom setup was built (Figure 3-1,
Figure 3-2). The sample was laid flat in a V-shaped channel running throughout the length of the sample
holder. The stent was carefully clamped at the proximal end using a custom machined holder, as shown
in Figure 3-2. The vessel was gently stretched to maintain vessel length and to ensure that the stent
remained secure in the clamp. The distal end of the vessel was secured by a 20G needle that ensured
that the vessel was kept taut. The puncture probe was then carefully advanced through the patent
portion of the artery that contained the stent until it reached the site of the proximal entrance of the
occlusion. The probe used was the inner mandrel from an 18G spinal needle and was chosen for the
rigidity of its shaft, its ease of use, its approximately equivalent profile to a standard 0.014” coronary
angioplasty guidewire in one plane, as well as its ability to enter the tapered entrance of the CTO due to
its bevelled profile (Figure 3-3). The probe had an outer diameter of 700µm along the shaft, which was
tapered to approximately 100µm at the distal tip.
55
Figure 3-1 - Puncture force measurement setup schematic. The stent at the proximal end is clamped and the vessel is secured at the distal end using the vessel securing pin. The puncture probe is placed just above the Proximal Fibrous Cap and clamped into the 3 jawed chuck. The Microtester arm is slowly lowered such that the probe advances into the PFC. As the probe moves into the cap, the force pushing back on the probe increases until the PFC is punctured, at which point the force drops precipitously. The force-displacement information is read from the load cell and stored electronically.
56
a
b c
Figure 3-2 - Photographs of puncture test setup. Panel a) shows a front view of the sample holder. The V-shaped groove where the vessel is placed is shown (black arrow). The stent is then held using the stent clamp (magenta arrow). The clamp is tightened using the clamp adjustment screws (brown arrow). Panel b) shows a photograph of the specimen holder from a top view. The vessel is laid flat in the V-groove and the stent is clamped (blue arrow). A 20G needle is used to hold the tautly pulled vessel (red arrow) and secured with set screws (turquoise arrow). The puncture needle is gently placed into the stent (green arrow). Panel c) shows a photograph of the entire setup. The microtester arm (orange arrow) is lowered slowly so that the probe assembly (maroon arrow) is advanced into the vessel held in the fluid tank (purple arrow). Position is controlled by 2- axis stage (yellow arrow).
57
Figure 3-3 - PFC puncture probe and guidewire. The probe used to puncture the PFC is shown in two views (top and middle) next to a 0.014” guidewire. The bevel cut probe is of similar profile to a 0.014” guidewire (bottom) in one plane, while remaining completely rigid along the shaft.
The sample holder was then mounted into a fluid chamber filled with saline heated to 37oC. The
chamber was attached to a servoelectric uniaxial materials testing system (800LE with a WMC-5F 25N
load cell, Test Resources) with a high degree of positional accuracy, load sensitivity and motion control.
The sample container was positioned on the 800LE system using a 2-axis stage, which was used to
position the vessel directly below the linear actuator to ensure direct vertical motion of the probe with
respect to the specimen. The probe was attached to the actuator via a 3- jawed chuck so that the probe
tip was located just above and not in contact with the PFC. The load cell was zeroed in this position to
account for the weight of the chuck and the clamped puncture probe.
58
The actuator was lowered in displacement control at a constant rate of 0.05mm/s into the occlusion.
The rate was chosen such that it was lower than the slowest rate seen in literature (104) to ensure
adequate time for tissue to respond to probe displacement. As the probe slowly advances into the
occlusion, the load begins to increase as the vessel and proximal fibrous cap both begin to stretch. The
load eventually reaches a peak value at maximal stretch, followed by a rapid dropoff when puncture has
occurred. Load and displacement data is collected at a sampling rate of 4 Hz. A sample curve showing a
punctured occlusion at the 6 week time point is presented in Figure 3-4. The first peak on the force-
displacement curve is defined as the puncture point.
3.2.4 Histology
After the puncture test was completed, selected vessels were removed from the apparatus and
processed for histology to determine the exact pathway taken by the probe. Samples were stored for 2
weeks in 10% formalin with the puncture probe embedded within tissue to preserve the path of the
probe. After careful removal of the probe, the vessel was embedded in paraffin for longitudinal
sectioning. The blocks were serially cut at every 100µm and every 4th slide was stained with Movat-
pentachrome stain. If necessary, intermediate slides were also stained.
3.2.5 Statistical Analysis
Data are expressed as mean ± Standard Error of the Mean (SEM). To compare the puncture forces at
each time point, a two sample t-test was used. A p-value of less than 0.05 was considered to be
statistically significant.
59
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 2 4 6 8 10 12
Forc
e (N
)
Displacement (mm)
Figure 3-4 - Sample force displacement curve of puncture from 6 week old CTO. Force rises to peak prior to puncture (~0.85N), followed by dropoff afterwards.
3.3 Results
The puncture test was successfully performed in 44 occlusions. A longitudinal histology section of a 12-
week-old punctured PFC is shown in Figure 3-5, showing the path of the puncture probe and the residual
PFC. In 6 cases, puncture data could not be used due to technical issues such as: stent deployment
(sidebranch rather than main vessel in 1 case, incomplete stent deployment in 1 case, inadvertent
crossing of soft occlusions with the guidewire during stent placement in two cases [at 2 weeks and at 6
weeks]; misalignment of the puncture probe in one case [resulting in vessel wall perforation], and in 1
case, manual dissection of the vessel prior to puncture test caused damage distal to the stent. Data from
these cases was excluded.
60
Figure 3-5 - Longitudinal section of a punctured vessel stained with Movat pentachrome. The trajectory of the puncture probe is seen as the void in this longitudinal section of the proximal fibrous cap of a CTO. Residual PFC shown in yellow ellipse. Internal Elastic Lamina (IEL), External Elastic Lamina (EEL), and Media (M) are shown. Bar = 100µm.
The mean puncture forces for early occlusive lesions (<=6 weeks) are compared to late occlusions (>=12
weeks) are shown in Figure 3-6. There was a statistically significant increase in puncture force at the late
61
timepoints with a mean value of 1.45N compared to the early time points puncture force of 0.72N.
Figure 3-6 - Summary of puncture force tests comparing early occlusive lesions (<=6 weeks) and later occlusions (>= 12 weeks).Panel a) shows a comparison of mean puncture force values. There is a significant increase in the mean puncture force value at >=12 week time points compared to early occlusions (<=6 weeks). Panel b) shows the percentage of lesions in each group with a puncture force below a cutoff of 1N. * - p < 0.01 vs. >=12 weeks.
62
When a cutpoint of 1N was used to represent the threshold between a “soft” lesion and a “stiff” lesion,
for 86% of the early time point lesions the puncture point was below the cutpoint, compared to only
30% from the late time points. There was a trend of increasing puncture force with increasing occlusion
age (Figure 3-7). A Spearman’s rank test shows a correlation factor of 0.55, indicating that there is a
trend towards increasing puncture force with age. The mean values for puncture forces at 2, 6, 12, and
15 weeks were 0.61N, 0.78N, 1.21N and 1.52N, respectively. There was a statistically significant
difference between the 2 week lesions and 15 week lesions, as well as between the 6 week lesions and
15 week lesions. Using a cutoff point of 1N to delineate stiff lesions from soft lesions, there were
marked differences in the percentage of samples with puncture points below the 1N value as follows:
100% at 2 weeks, 75% at 6 weeks, 43% at 12 weeks, and 26% at 15 weeks.
3.4 Discussion
There are two important observations in this study. First, we have shown the utility of a systematic,
reliable technique to quantify the force required to puncture the proximal fibrous cap of CTOs in an ex-
vivo setting. Second, the results indicate that there are significant differences in the amount of force
required to puncture CTOs of different ages, which corresponds to previously published differences in
occlusion composition. Specifically the force required to puncture the proximal fibrous cap increases
significantly in response to occlusion age.
It has been shown in several biologic systems that a burst or puncture point can be determined by
examining a force-displacement curve. Schober et al(104) studied the force-displacement curve in a
burst testing setup to show the relationship between the use of different probe profiles and discussed
the interpretation of the force-displacement curve while examining fetal membranes. Saito et al(105)
studied the use of a needle setup to automate needle puncture in rabbit ear veins for blood sampling.
63
Figure 3-7 - Mean puncture forces at different timepoints. Panel a) shows that there is a trend towards increasing puncture force with increasing age of the occlusion. Panel b) shows the percentage of lesions that require <1N of force for puncture at each timepoint. * - p < 0.001 vs 15 week. † - p < 0.01 vs. 15 week.
64
The data clearly shows for the first time that there is an increase in the amount of force required to
puncture occlusions according to the occlusion age. Specifically, this change in puncture force was
apparent as CTO age increased from <=6 weeks to >=12 weeks. Both lower time points (2 weeks and 6
weeks) had similar values, although it was previously shown that there are differences in composition at
these time points. The 2-week time point still contains mainly thrombotic tissue with some
proteoglycan-rich areas present within the occlusion. These substances are known to be soft, which is
confirmed by the low puncture force. At 6 weeks, when the microvascular area has been shown to be at
its peak, the tissue still contains large amounts of proteoglycan-rich material with very small amounts of
collagen present within the CTO. However at 12 weeks and beyond, there is a marked involution of the
microvascular network within the CTO, accompanied by increases in collagen and regression of the
proteoglycan material. Thus, the 12-week-old CTO (and later time points) are characterized by avascular,
fibrotic tissue, particularly at the proximal fibrous cap. This is also precisely the time period where
success rates in crossing the CTO have been shown to precipitously decline. With this new technique to
measure puncture force, we now have a mechanical correlate to the histologic changes previously
described in this model. This technique provides an objective means to quantify the difficulty in crossing
CTOs and potentially an assay to assess interventions to alter the CTO composition. In our model, it
appears that 1N is an appropriate cutpoint to distinguish between soft early stage occlusions and the
harder later time points, although this will require further validation in future studies. Further studies
are also required to determine whether interventions can alter the puncture forces in older CTOs and
whether the presence of calcification, which is not characteristic of this model, would substantially
increase the puncture force.
Although this technique appears to be a very promising tool for CTO research, there are still some
caveats. The degree of variation at each time point is one issue. There are several possible explanations.
First, there is the possibility of heterogeneity either due to the composition or the morphology of the
65
lesions. It has been shown that most lesions in this model of CTO have a tapered entrance to the CTO
and the angle of entry may play a role in the resultant puncture force. Secondly, there may be some
variability in the rate at which fibrosis develops, such that some lesions develop a fibrous cap much
earlier than others. Furthermore, it has been suggested (34) that the presence of longitudinal
microchannels may provide a path for guidewire crossing. This could be due to their capacity to provide
a direct path through the lesion (in very large microchannels), or that perivascular tissues adjacent to
the microchannel may be more compliant than those in the fibrotic areas. In both instances, this could
provide a softer path across the entrance of the CTO instead of the stiff proximal fibrous cap. This would
imply that it may be more useful to classify lesion-crossing difficulty on the basis of composition,
morphology and mechanical properties (such as puncture force) rather than on age alone. This will
require further validation, particularly with in-vivo diagnostic imaging and mechanical interrogation
systems.
3.4.1 Potential applications
The ability to study the force-displacement curves and to determine both the forces to puncture the
proximal fibrous cap as well as the characteristics of the curves beyond that point for determining tissue
compliance could be useful in a number of relevant applications. Since it is desirable to use conventional
softer guidewires rather than stiffer specialty guidewires or crossing devices to cross a CTO, agents that
could affect the compliance of CTOs, and particularly, the proximal fibrous cap, are under investigation.
Strauss et al have had success using a localized delivery of collagenase to digest and soften the collagen
rich proximal fibrous cap (41,56). A second approach is to increase the vascular network at the proximal
end of lesions by promoting the growth of microvasculature, which may in turn lead to increased
crossability due to the proliferation of softer tissues and possible low resistance pathways. While the
vascularity of a lesion can be assessed using imaging tools, the mechanical properties of increased
66
vasculature on facilitating guidewire crossing cannot currently be directly measured using imaging
techniques. In both of the above applications, the ability to measure puncture force allows for a simple
and direct way to measure the efficacy of the treatments and associated required doses.
While the present study used a bevel cut needle as the puncture probe, any arbitrary probe that could
be mounted into the 3-jawed chuck could be used. A future study of various guidewires with different
tip profiles and compositions in a similar setup could prove useful in studying the interactions of
different guidewires with not only the proximal fibrous cap, but also other geographic locations,
including the body of the CTO and the distal fibrous cap. There also exists the possibility of coupling a
guidewire in series with a load cell to measure puncture forces in an in vivo situation.
3.5 Conclusions
This study demonstrates a novel technique for quantitatively measuring the amount of force required to
puncture the proximal fibrous cap of CTOs in a systematic, repeatable way. This technique provides the
first objective evidence of the increased puncture forces required to cross the proximal fibrous cap with
increasing occlusion age. This technique could be useful for assessment of therapeutic manipulations of
CTOs such as collagenase. Moreover, it could be used to test the characteristics of various guidewires
and their interactions with CTOs. Finally, it could be potentially adapted to provide in-vivo assessment of
crossing characteristics of the proximal fibrous cap during percutaneous interventions.
67
4 The use of Ultrasound Mediated Contrast agents as an adjuvant for
collagenase therapy in Chronic Total Occlusion3
4.1 Introduction
Chronic Total Occlusions, or CTOs, are defined angiographically as arteries with TIMI (Thrombolysis in
Myocardial Infarction) grade 0/1 flow that have been occluded for at least 3 months duration(22). As
these lesions age, their composition changes from a relatively soft thrombotic material to a hard,
predominantly collagenous occlusion(35). The presence of collagen is correlated with the age of the
occlusion, which is a predictor of guidewire crossing(106). It has also been shown that the entrance into
the CTO consists of a particularly densely-packed, collagen-rich region termed the Proximal Fibrous Cap
(PFC)(35). The PFC appears to be a significant impediment to advancing a guidewire into the body of the
CTO.
In an attempt to alter the guidewire crossing characteristics of the PFC, studies have shown improved
procedural results following local delivery of a bacterially-derived collagenase formulation(41,56). In
these studies, the response to collagenase was dependent on waiting a certain period of time after
collagenase delivery prior to the guidewire crossing attempt. Efforts to advance guidewires immediately
following collagenase delivery did not show improvement. However, guidewire crossing rates
significantly improved when the waiting period was extended to between 18 and 72 hours(41).
Refinement of the procedure to shorten the waiting period to a few hours to enable same day guidewire
3 This chapter is adapted from a manuscript in preparation for submission as: Amandeep S. Thind PhD, Bradley H.
Strauss MD PhD, Raffi Karshaffian PhD, Aaron A. Teitelbaum MD, Michelle Ladouceur, Muhammed Ali Akbar,
Brendan Rosen, Michael Bohnen, Cari M. Whyne PhD, David E. Goertz PhD, F. Stuart Foster PhD. The Use of
Ultrasound Mediated Contrast Agents as an Adjuvant for Collagenase Therapy in Chronic Total Occlusion
68
crossing would be desirable to shorten the hospitalization stay. This procedure is shown schematically in
Figure 4-1a.
One potential technique to accelerate the collagenase effect is to use UMM as an adjuvant. In the
current study, we report on the use of therapeutic ultrasound for accelerating the effects of locally
delivered collagenase in an ex vivo model. The proposed clinical scenario is shown in Figure 4-1b.
4.2 Materials and methods
4.2.1 CTO Model
All animal care and handling were performed in accordance with the guidelines specified by the
Canadian Council on Animal Care (CCAC) for the Care and Use of Laboratory Animals and were approved
by the Institutional Animal Care and Use Committee. Chronic total occlusions were created in rabbit
femoral arteries using a thrombin injection model as previously described(42). This technique produces
CTOs with several features present in the human disease, including the presence of microchannels(34), a
proximal fibrous cap, and a fibrous core(35).
4.2.2 Sample collection
CTOs were allowed to age for 12 weeks and were verified for occlusion characteristics angiographically
prior to sacrifice. The 12-week time point is characterized by significant collagen accumulation and
marked increases in puncture force compared to earlier time periods. It has also been shown that 12
week old CTOs, both in humans and in the rabbit femoral artery CTO model, have low success rates for
guidewire crossing(22,41,102). Prior to sacrifice and removal of the CTO, a stent was deployed
immediately proximal to each CTO lesion. This allowed us to easily identify the proximal entrance into
the CTO, acted as a cannula for therapy, and also acted as a brace to support the extracted CTO vessel in
69
the testing apparatus. Vessels containing the CTO were harvested and placed in 0.9% saline solution and
treated immediately after sacrifice.
Figure 4-1 – Proposed clinical scenario. Panel a) shows the current clinical protocol. An over-the-wire (OTW) balloon is advanced to the proximal segment of the CTO and the balloon is inflated. Collagenase is injected through the wire port of the balloon. Panel b) shows the proposed clinical protocol. As above, an OTW balloon is advanced to the proximal segment of the CTO. Now, microbubbles are injected through the wire port. The microbubbles are disrupted by an ultrasound transducer. In the image, the ultrasound transducer is attached to the catheter delivering the microbubbles. However, the transducer may also be placed outside the skin surface and is used as such in the present study. After microbubbles are disrupted, balloon is inflated and collagenase is injected as with the current clinical protocol.
70
4.2.3 Treatment groups
Arterial samples were separated into 4 main treatment groups (Figure 4-2): Control [Group I] (n=6);
Collagenase-only [Group II] (n=15); Ultrasound mediated microbubbles [Group III] (n=5); and combined
collagenase and ultrasound mediated microbubbles [Group IV] (n=19). The collagenase-only group was
further subdivided into two subgroups: collagenase-only standard treatment time (2.5 hrs) [Group IIa]
(n=10) and collagenase-only extended treatment time (5 hrs) [Group IIb] (n=5). The combined
collagenase-UMM group was also separated into two subgroups: sequential treatment [Group IVa]
(n=11) and concurrent treatment [Group IVb] (n=8). The groupings are summarized in Table 4-1.
A number of possible control cases were omitted to minimize the number of samples used. For
instance, any or all of the following groups could have been used for further controls: a) No therapy at
all (i.e. no saline injections), b) ultrasound only (with no microbubbles),or c) collagenase with
microbubbles in the absence of ultrasound. Control case a) was rejected in favour of Group I because
Group I includes any potential effects arising from the fluid injections to the PFC. Control case b) was
rejected since it was inferred that samples treated using ultrasound in the absence of microbubbles
would have a less significant effect than samples treated with ultrasound mediated microbubbles
[Group III] under the same acoustic conditions. If results had shown a significant effect with Group III,
then case b) would have been performed. Finally, control case c) was rejected in favour of Group II since
the main goal of the study was to determine if there was incremental benefit that could be achieved
using a combined treatment compared to the collagenase only treatment as currently used in practice,
which is represented by Group II.
71
Figure 4-2 - Schematic illustration for treatment groups. Panel a) shows “Control” group using only a saline infusion, while b) shows the “Collagenase only group”. Panels C & D show groups using ultrasound to disrupt microbubbles. c) shows “Ultrasound Only” group, and d) shows “Both Treatments” group.
Table 4-1 - Summary of treatment groups. Main treatment groups listed in boldface, subgroups listed in standard text.
Group designation Group title Group description n =
Group I Control Samples treated with only saline. Total
treatment duration 2.5 hours
6
Group II Collagenase only Samples treated with collagenase 15
Group IIa Standard treatment
duration
Treatments were performed for 2.5h total 10
Group IIb Extended treatment
duration
Treatments were performed for 5h total 5
Group III Ultrasound Mediated
Microbubbles
Samples treated with UMM. Total treatment
duration 2.5 hours
5
Group IV Combined treatment Samples treated with both collagenase and
UMM. Total treatment duration 2.5 hours
19
Group IVa Sequential UMM treatment given prior to collagenase
infusion
11
Group IVb Concurrent UMM treatment given in the presence of
collagenase
8
72
4.2.4 Standard Treatment
Each arterial sample was placed in a custom built treatment reservoir with 10mL of solution, with the
contents of the solution depending upon treatment group. Prior to treatment, an OTW (Long Cobra,
SciMed) balloon was advanced into the stent and inflated with the balloon tip ~5-10mm from the CTO
entrance to localize the proximal region. Depending on the treatment group, 3 infusions of either 500 µL
of saline or microbubble solution were delivered via the central wire port of the OTW balloon. One
infusion was made every 2 minutes. Next, one infusion of 500µL of either saline or collagenase was
delivered to the CTO site. The balloon was left inflated for 30 minutes, then deflated and removed. The
sample was incubated at 37oC for an additional 2 hours, resulting in a total treatment time of 2.5 hours.
500µL of supernatant solution was extracted from the space proximal to the PFC. The sample was then
taken for biomechanical testing. The supernatant solution was frozen for later biochemical analysis.
4.2.5 Treatment Duration
In the initial collagenase study(41), collagenase only therapy was ineffective at very short waiting
periods (~1hr). Therefore to assess the time dependency and potential interactions between ultrasound
and collagenase, we studied two treatment times: (1) standard treatment duration (2.5 hours, including
30 minutes of balloon inflation) [Group IIa] and (2) an extended treatment duration [Group IIb], (5
hours, including 1 hr of balloon inflation).
4.2.6 Modified Acoustic Setup
The order in which operations are conducted from an acoustic perspective may have a significant effect
on bioeffects, and therefore, may have a significant effect on the efficacy of ultrasound/microbubble
therapy. Therefore, two different sequences of UMM and collagenase were assessed: 1 - microbubbles
73
injected and disrupted prior to collagenase therapy [Group IVa] and 2 – microbubbles and collagense
injected together with the microbubbles disrupted in the presence of collagenase [Group IVb]. The
former potentially allows the disruption of microbubbles to create small fenestrations in the plaque to
provide better access to the substrate for the collagenase enzyme once it is injected. This group was
termed “sequential treatment”. The latter technique has the potential additional benefit of giving the
collagenase a mechanical push further into the plaque as opposed to simply creating pathways for it to
travel. This group was termed “concurrent treatment”.
4.2.7 Collagenase
Type VII-S bacterial collagenase derived from Clostridium histolyticum (Sigma, c2399) was used at a
concentration of 1000 Collagen Digestion Units (CDU) in 300uL of Phosphate Buffered Saline (PBS).
Unlike human or other mammalian collagenase, collagenase from Clostridium histolyticum cleaves
collagen at numerous sites, leaving several very small peptides(60).
4.2.8 Acoustic parameters
The treatment reservoirs were designed to include acoustically transparent windows to allow ultrasound
energy in and out of the reservoirs without significant reverberations so as to avoid standing waves
(Figure 4-3). Ultrasound treatments were conducted with a 1MHz centre frequency transducer, with an
aperture size of 3.8cm and an F-number of 2 (Valpey Fisher). The peak negative pressure was measured
at 1.7 MPa in hydrophone tests.
74
Figure 4-3 - Ex vivo acoustic setup. The vessel was secured into the sample container and placed onto the adjustable stage so that the PFC was located at the focus of the ultrasound transducer. A balloon catheter was advanced into the stent and the balloon was inflated to localize the proximal end of the lesion. Microbubbles were injected via the wire port of the catheter and disrupted in the acoustic beam.
The microbubbles used were Definity™ (Bristol-Meyers Squibb), which is a clinically approved agent in
North America. The bubbles are approximately the same size as red blood cells (peak volume fraction
count of ~3µm), making them an excellent intravascular agent. Once constituted according to the
manufacturer’s specifications, the microbubbles were diluted to a 10% volume fraction.
Samples in the Ultrasound Only (Group III) and Both UMM + collagenase treatments group (Group IV)
were exposed to 3 treatments, each lasting 120s. A disruption sequence, 508ms long was used to drive
the microbubbles into cavitation. This sequence consisted of 500 individual pulses of acoustic energy,
each pulse being 16µs long, with 1ms between successive pulses. There were 5s between disruption
pulses to allow for bubbles to flow into the acoustic field. The pulse sequence used is shown in Figure
4-4.
75
16µs 1ms
500 ms
5 s
120 s
Disruption pulse
Disruption pulse
Figure 4-4 - Pulse Sequence for acoustic treatments. A 500ms disruption pulse, consisting of 500 pulses of acoustic energy (each pulse was 16 cycles at 1MHz) separated by 1ms was used to disrupt the bubbles in the acoustic field. These pulses were sent every 5s for a total of 120s in order to allow fresh bubbles to flow into the field after a disruption sequence.
4.2.9 Biochemical Assays
The supernatant solution extracted at the end of treatment was used to determine the effect of the
different treatments. Two different assays were performed to examine the degradation products
released at the site of the occlusion. First, a Biorad Detergent Compatible (DC) protein assay was
performed to estimate the total quantity of protein products released into solution. This is a
colorimetric assay based on a method developed by Lowry et al(107) and modified to be detergent
compatible. Briefly, 50µL of supernatant solution was placed in a tube, mixed with 500µL of Biorad
reagent A, and 2000µL of Biorad reagent B. The sample was vortexed and left for 15 minutes to allow
the colour to develop. Bovine serum albumin (BSA) standards were prepared in concentrations ranging
76
from 0.2 to 2.5 g/L. These standards were exposed to the same reagents as the samples as described
above. Absorbances were measured at 750nm using a KC4 plate reader. This assay, which measured the
total protein in solution, was used to determine if any increased collagen degradation was due to a non-
specific increase in all proteins in solution or due to a specific increase in collagen degradation alone.
Second, a hydroxyproline assay was carried out to determine the quantity of products in the
supernatant solution released from collagen degradation. Hydroxyproline is an amino acid that is found
predominantly in connective tissue collagen(108) and is used as a quantitative measure for collagen
degradation. The assay is colourimetric and the protocol used was a modified version of the techniques
described by Reddy et al(109) and those described by Edwards et al(110). For this assay, 50µL of the
supernatant solution was extracted and placed into a 2mL screw capped tube with a rubber O-ring in the
cap (Ultident) containing 100µL of 6N HCl. Similarly, standards from pure hydoxyproline (Sigma) for
concentrations up to 120µg/mL were made up and placed in the 2mL screw capped tubes containing
HCl. The tubes were heated at 120oC for 4 hours to hydrolyze the protein in the solution. After heating,
the tubes were placed in a speed-vacuum (Savant) and dried for 4 hours. Once dried, 450uL of
chloramine-T solution (made up from 0.127g chloramine-T (Sigma), 2mL of 50% n-propanol, diluted to
10mL with Acetate buffer) was added to each tube and allowed to react for 25 min at room
temperature. Acetate buffer was made in batches from 120g sodium acetate, 46g citric acid, 12mL
acetic acid, 34g NaOH, and brought to 1L by dilution with autoclaved water. Then, 500µL of Ehrlich’s
reagent (made up from 1.5g p-dimethyl-amino-benzaldehyde (Sigma), 6mL n-propanol, 2.6mL perchloric
acid, diluted to 10mL with autoclaved water) was added and allowed to incubate at 65oC for 20 min.
200µL of each reacted sample was pipetted into a 96 well plate and taken to a plate reader to have
absorbance values at 550nm measured. Samples were compared to standards to estimate
hydroxyproline concentrations, which were used as a measure for collagen degradation into
77
supernatant solution. All supernatant samples were performed in triplicate and averaged, while
standards were performed in duplicate.
4.2.10 Puncture Force Test
Following treatment, a puncture force test was performed on all samples to directly measure the
puncture force required to penetrate the treated PFC. The test was performed as described in 3.2.3.
4.2.11 Statistical Analysis
Data are reported as mean ± standard error of the mean (SEM). Groups were compared pairwise using
two sample T-tests, making no assumption of equal variance between groups. A p-value < 0.05 was
considered significant.
4.3 Results
4.3.1 Standard treatment duration
4.3.1.1 Supernatant Protein assays
There were no significant differences in total protein in the supernatant between the four groups as
determined by the Biorad DC assay (Figure 4-5).
There were, however, significant differences in hydroxyproline [collagen] content between the groups
(Figure 4-6).
78
0.000
0.500
1.000
1.500
2.000
2.500
Both Treatments Collagenase Only Ultrasound Only Control
Tota
l P
rote
in (
g/L
)
Treatment
†
Figure 4-5 - Biorad protein assay data for standard treatment duration. There is no statistically significant difference in the total protein released in any of the groups. († - p > 0.05).
In both the Control and Ultrasound Only [Groups I & III] cases, minimal hydroxyproline levels were
detected in the supernatant (0.004 and 0.003 g/L respectively). The collagenase-only treatment [Group
IIa] showed a marked increase in hydroxyproline concentration, approximately tenfold greater than the
control group (0.030 vs. 0.004 g/L respectively, p < 0.01). The addition of the ultrasound treatment to
the collagenase therapy [Group IV] resulted in a significant 2-fold increase in hydroxyproline level
compared to the collagenase-only treatment [Group IIa] (0.065 vs. 0.030 g/L, p < 0.05).
79
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0.080
Both Treatments Collagenase Only Ultrasound Only Control
Hyd
roxy
pro
line
Co
nce
ntr
atio
n (
g/L
)
Treatment
‡
*
†
Figure 4-6 - Hydroxyproline released into medium for standard treatment duration. Both the combined treatment and collagenase only groups showed significant increases in hydroxyproline release compared to ultrasound only and control. Hydroxyproline in the combined treatment group was significantly higher than the collagenase only group. († - p > 0.05, ‡ - p<0.05,* - p< 0.01).
4.3.1.2 Puncture Force test
The results for the puncture test are shown in Figure 4-7. The peak force required to puncture a CTO
was significantly reduced in the combined collagenase-ultrasound treatment [Group IV] compared to all
three other groups (p < 0.05). There were no statistically significant differences between Control [Group
I], Collagenase only [Group IIa], and Ultrasound + microbubbles [Group III] amongst one another.
80
0.000
0.200
0.400
0.600
0.800
1.000
1.200
1.400
1.600
Both Treatments Collagenase Only Ultrasound Only Control
Pu
nct
ure
Fo
rce
(N
)
Treatment
‡
†
Figure 4-7 - Puncture force test results for standard treatment duration. The difference between the Both Treatments group and the Collagenase Only group is statistically significant. The difference between all other groups is not statistically significant. († - p > 0.05, ‡ - p<0.05)
4.3.2 Extended treatment duration
Extending the collagenase-only treatment time from 2.5 hours [Group IIa] to 5 hours [Group IIb] had
significant effects in both the hydroxyproline assay the puncture force test (Figure 4-8). The extended
treatment arm showed >4-fold increase in hydroxyproline (0.135 vs. 0.30g/L, p < 0.01) and a reduction
in the puncture force (0.44 N versus 0.80 N, p < 0.05), compared to the standard duration arm.
4.3.3 Modified acoustic setup
The sequential [Group IVa] and concurrent [Group IVb] groups showed similar hydroxyproline levels
(0.065 vs. 0.062 g/L, p=Not Significant (NS)) and puncture forces (0.300N vs 0.388N, p=NS) (Figure 4-9).
81
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Collagenase (2.5 hr) Collagenase (5 hr)
Pu
nct
ure
Fo
rce
(N
)
‡
0.000
0.020
0.040
0.060
0.080
0.100
0.120
0.140
0.160
Collagenase (2.5 hr) Collagenase (5 hr)
Hyd
roxy
pro
line
Co
nce
ntr
atio
n (
g/L
)*
a b
Figure 4-8 - Extended treatment duration results. The extended treatment group showed a large increase in the amount of hydroxyproline released – significant. The extended treatment group also showed a drop in the required puncture force, which was significant. (‡ - p<0.05,* - p< 0.01)
4.4 Discussion
The major finding of this study is that combined collagenase-ultrasound mediated microbubble therapy
increases the degradation of collagen and decreases the puncture force required to cross the PFC of a
CTO compared to collagenase-alone. The benefit of the combined collagenase-UMM treatment appears
to be predominantly to accelerate the softening effect since prolonging the waiting time in the
collagenase-only group results in similar degradation of collagen and changes in the puncture force. This
effect may have significant clinical relevance by shortening the waiting time after collagenase
administration and allowing a single day procedure rather than the current requirement of a two-day
procedure.
82
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0.080
0.090
Both Treatments (Concurrent)
Both Treatments (Sequential)
Hyd
roxy
pro
line
Co
nce
ntr
atio
n(g
/L)
†
0.000
0.100
0.200
0.300
0.400
0.500
0.600
Both Treatments (Concurrent)
Both Treatments (Sequential)
Pu
nct
ure
Fo
rce
(N)
†
a b
Figure 4-9 - Modified acoustic setup results. Hydroxyproline and Puncture Force assays are shown. In both cases, there is no statistically significant difference between the two treatment configurations. († - p > 0.05)
The collagenase effect was shown to be selective for collagen since there were no significant differences
between the total protein levels in any of the treatment groups. This suggests that there is a low
background level of protein release related to the testing conditions, which is not disturbed by either
the collagenase or UMM treatments. Therefore, any increase in the hydroxyproline content for a given
treatment can be attributed to an increase in collagen degradation, and not just to an increase in
nonspecific protein. As expected, the control group (Group I) showed negligible amounts of degraded
collagen in the supernatant. Similarly, the ultrasound-only group (Group III) showed minimal
hydroxyproline levels, as there was no enzyme present to degrade the collagenous PFC. The
collagenase-only group showed a nearly 10-fold increase in the amount of collagen released into the
supernatant, indicating the presence of large amounts of collagen in the proximal fibrous cap as
83
previously shown by our group. There was a further doubling in the hydroxyproline levels in the
combined collagenase-ultrasound group (Group IV).
The effect of ultrasound appears to be synergistic with collagenase since the UMM only group (Group III)
showed minimal collagen release, similar to the controls. While the mechanism for these
ultrasound/microbubble interactions are not fully understood, a plausible explanation is that
microjetting resulting from bubble cavitation creates mechanical disturbances within the plaque,
exposing an increased surface area of the substrate to enzyme therapy.
The puncture force tests are important as a mechanical correlate to the hydroxyproline assay data. This
study complements earlier studies showing that collagenase therapy can improve guidewire crossing
success rates by degrading the PFC. Thus, our data in this study further reinforces the mechanism by
which collagenase works in CTOs: collagen is degraded from the PFC and thereby lowers the puncture
force required to cross the occlusion to levels typically associated with chronologically younger (and
more easily crossed) occlusions. Combining UMM with collagenase (Group IV) lowered the puncture
force by a factor of 2 compared to the collagenase only group (Group IIa), which suggests that a
combined therapy is a promising adjunct to collagenase therapy.
Within the collagenase-only group, the reduction in puncture force was evident in the 5 hour treated
group (Group IIb), but not the 2.5 hour group (Group IIa). This is also consistent with our previous results
that local delivery of collagenase followed by a short waiting period did not facilitate guide-wire
crossing(41).
Although the hydroxyproline content in the supernatant was higher in the extended collagenase
treatment (Group IIb) compared to the combined therapy used in Group IV (0.135 vs. 0.063 g/L,
respectively), this was not accompanied by a consequent reduction in the puncture force required to
cross the lesions (0.440 vs. 0.345 N, respectively, p=NS). This suggests that although there is a
84
relationship between the collagen assay and the puncture force test, there may be a threshold-type
effect present whereby additional collagen degradation beyond a certain point would not further
facilitate guidewire crossing and could potentially cause adverse effects to the vessel wall. Furthermore,
the very low puncture force demonstrated in this study provides further support for the use of UMM
since a similar effect on puncture force was achieved with half of the collagenase exposure time and
consequently, less potential for degradative effects on deeper layers of the vessel wall.
An additional important finding of this study is that there were no significant differences whether the
collagenase was delivered after microbubble therapy or concurrently with microbubble therapy. This
was a very desirable result, since delivery of microbubbles through a balloon catheter while the balloon
is inflated can be problematic. Injections of microbubbles through long, narrow lumens can cause
microbubble collapse, rendering them unusable for UMM therapy. Since it appears that the option to
deliver the bubbles without balloon inflation exists, it is likely that in vivo studies using UMM would use
the sequential treatment (Group IVa) configuration.
4.4.1 Clinical Relevance of Study
Accelerating collagenase activity without increasing the enzyme dose should lower the potential risk of
undesirable side effects of high dose collagenase, such as soft tissue bruising (5). The feasibility of the
procedure in terms of patient and physician acceptance and cardiac catheterization lab efficiency is
improved by the ability to perform both parts of the procedure (collagenase delivery and guidewire
crossing) in a single day rather than over two days as the current clinical protocol maintains. Moreover,
it provides further support for the potential of therapeutic ultrasound in accelerating and enhancing
enzyme therapies.
4.5 Conclusions
85
Local delivery of collagenase therapy into chronic total occlusion has previously been shown to
successfully and significantly assist guidewire crossing when given adequate dwell time. The effect of
collagenase therapy can be accelerated by using ultrasound mediated microbubble disruption by
increasing the total amount of collagen degraded in the lesion, resulting in a decreased puncture force
of the Proximal Fibrous Cap. Combined ultrasound-collagenase treatment appears to be a promising
modification of collagenase therapy by shortening the required treatment period and potentially
allowing for a single day therapy for both treatment and revascularization.
86
5 Summary and Future work
5.1 Summary and Discussion
This thesis has explored novel diagnostic and therapeutic approaches to solve the problem of chronic
total occlusions in cardiovascular disease.
In Chapter 2, the ability of micro-ultrasound to detect in-vivo recanalization channels in a porcine model
of CTO was demonstrated. These channels were shown for the first time to be continuous through large
portions of the lesion, to experience extraluminal communication with sidebranches, and to be lined
with an endothelial layer.
In Chapter 3, the development of a robust technique for measuring the amount of force required to
puncture the PFC of CTOs using an arbitrary puncture probe was shown. This system was used to show
that CTO puncture force increases with lesion age. This is a technique that will prove to be extremely
valuable for characterizing CTOs, guidewires, and perhaps most importantly, assessing the efficacy of
therapies that aim to modify the compliance of the PFC, as shown in Chapter 4.
In Chapter 4, ultrasound mediated microbubbles were used to accelerate the effect of collagenase
therapy in an ex-vivo study. We showed that it is possible to significantly increase the amount of
collagen degraded as well as decreasing the amount of force required to puncture the lesions using
ultrasound as an adjuvant to collagenase therapy.
Overall, new tools were introduced that can both help characterize CTOs to obtain a further
understanding of their characteristics and maturation, as well as tools to help assist guidewire crossing
either by finding more desirable paths or by changing lesion compliance to make crossing easier.
87
One point of interest which appears to be a discrepancy is the difference in puncture forces between
the control group in chapter 4, and the 12 week group in chapter 3. One would expect the two groups to
have very similar puncture values, but they are indeed quite different. The value for the control group
was 0.71N vs. 1.22N (p > 0.05) for the 12 week group. While the difference between the two groups is
not statistically significant, it does appear at first glance to be unusual. Furthermore, this is true to some
extent for all the groups in chapter 4 vs. chapter 3. It appears as though the groups in chapter 4 have
lower puncture thresholds than one would expect from the results in chapter 3. One main difference
between the samples from the biomechanical study in chapter 3 and the ex-vivo treatment study in
chapter 4 is the presence of a series of injections into the cap. The treatments all involved OTW balloon
wire port injections. As previously mentioned, it has been shown that injections directly into the PFC can
open up microchannels and facilitate guidewire crossing(55), and we may be seeing this phenomenon
here. Another difference is that the samples from the treatment groups (chapter 4) were kept in saline
for a much longer time, with the potential for PFC delamination from the IEL, resulting in easier
puncture.
5.2 Future work: In vivo collagenase studies
The work in chapter 4 of the thesis showed a promising technique using UMM as an adjuvant for
collagenase therapy by performing the treatment in a well controlled ex-vivo study. In order to further
the study towards clinical realization, it is vital to show efficacy in an in vivo scenario. A study has been
performed translating this work in the same animal model, but performed in vivo. This study will be
described briefly herein.
5.2.1 Materials & Methods
CTOs were constructed in 16 rabbits for an in vivo study using the model described in Section 1.3.5.1.
Rabbits were separated into 2 groups: collagenase only and combined UMM + Collagenase treatment.
88
Occlusions were verified angiographically prior to the procedure to assess lesion persistence, location,
entry characteristics, as well as neovasculature.
5.2.1.1 Ultrasound Treatment
An OTW balloon was advanced to the proximal end of the occlusion such that it blocked off the largest
possible side branch, but was still ~1cm away from the PFC to avoid vessel perforation from injection.
The leg to be treated was cut down so that the CTO was <5mm from the skin surface. A metal grid was
placed over the lesion to act as a fiducial marker for both angiography and ultrasound imaging. An L12-5
probe on the Phillips iU-22 system was used to find the lesion, guided by fiducials. Definity™
microbubbles were diluted in a 10:1 ratio and drawn up in 1mL syringes. The agent was carefully
injected in 100-200µL boluses, and imaged in a low Mechanical Index (MI) contrast mode to precisely
locate the PFC.
A therapy transducer (F-4, 1MHz centre frequency, 1.1 MPa peak negative pressure) was setup to be
confocally aligned with the imaging probe. DefinityTM diluted to 10% in saline were injected in 100-200
µL boluses. They were disrupted using the therapy transducer to achieve a similar therapeutic effect as
shown in Chapter 4. Between 10 and 20 therapeutic sequences were performed for each rabbit. Each
therapeutic disruption pulse included 100 pulses of 50µs duration were sent at a 1KHz PRF. A schematic
of the setup is shown in Figure 5-1. Collagenase only samples were treated with a similar number of
saline injections to account for any effect from the injection itself.
5.2.1.2 Collagenase Treatment
After treatment with microbubbles / saline (depending on treatment group), all samples were treated
with collagenase. The OTW balloon was inflated to localize the region proximal to the PFC. 1000 CDU of
collagenase from Clostridium Histolyticum was diluted into 330µL of saline and drawn up in a 1mL
89
syringe. The collagenase was injected carefully through the wire port of the OTW balloon and flushed
with 660µL of saline. The balloon remained inflated for a total of 1hr. Upon balloon deflation, the animal
was stented and sacrificed after an additional 30-90 minutes. The vessel was carefully harvested for
biomechanical testing, as described in Chapter 3.
Figure 5-1 - Schematic setup of in-vivo collagenase therapy. The imaging beam and the therapy beam are confocally aligned. The imaging beam is used to locate the PFC. The therapy transducer is used to disrupt the microbubbles at the confocal region.
5.2.2 Results & Discussion
The results from biomechanical testing are shown in Figure 5-2. The combined treatment showed a
significant reduction in the amount of force required to puncture the lesions (1.019N for Collagenase
treatment vs. 0.606N for combined, p=0.03). This implies that we can indeed transfer the therapeutic
effect achieved in Chapter 4 to an in vivo situation. Even though the transducer used in the in vivo study
was not as strongly focused as that used ex-vivo, it was capable of causing inertial cavitation in the
90
microbubbles. It was chosen for its wider focal zone since there was significantly more uncertainty in the
location of the PFC in this study vs. the ex-vivo study.
Figure 5-2 - Puncture test results from in vivo study. The combined treatment group showed a significantly lower puncture force compared to the collagenase only group. ‡ - p < 0.05.
5.3 Future work: Microspheres loaded with VEGF
While the use of enzyme therapy is clearly beneficial in softening the PFC of CTO, there are other
promising techniques under investigation for modifying the compliance of CTO. It has been posited that
the soft tissues surrounding microvessels provides a softer barrier than the collagen rich PFC(34). It then
follows that inducing microvascular growth in the proximal region will allow for easier guidewire
crossing. There is work currently underway testing the use of Vascular Endothelial Growth Factor (VEGF)
91
as a means for softening lesions. In a study of n=52 rabbit femoral artery occlusions, VEGF was loaded
into Bovine Serum Albumin (BSA) microspheres (on the order of 40µm in diameter) to provide a slow
release. The injection is performed in the same manner as that shown in Figure 4-1A, except with an
infusion of VEGF in place of collagenase. 12 week old samples were separated into 3 groups: VEGF
treated (n=15), samples treated with microspheres only (n=13), and those left untreated, or a non-
interventional group (NIV) (n=23). Animals were sacrificed 3 weeks after treatment. Preliminary results
for puncture force testing of these samples is shown in Figure 5-3. While not statistically significant, it
does appear that the use of VEGF does slightly decrease the mean puncture force compared to the NIV
group. However, and of significant interest, when we look at the proportion of lesions that are soft (in
this case, a cutoff of 1N is used), we see that there is a very large difference between both the VEGF
groups and the microspheres only group when compared to the NIV group. More than half of the
samples in these groups fall below the 1N threshold. This suggests that in select cases, significant
softening does occur. However, what is more interesting is that it does not appear to matter whether or
not VEGF is used, but rather the effect is dominated by the presence of the microspheres. This effect is
appreciated by examining the histology in Figure 5-4. Left panel shows the effect of VEGF but no
embedded microspheres. Centre panel shows a sample from the microspheres only group.
Microspheres appear to have replaced native ECM, possibly accounting for the reduced puncture force
in many samples. The right panel shows a sample from NIV group. While this is an interesting
preliminary result, significant rigour is required to tease out the mechanism of this replacement and
whether it may be a viable tool for lesion softening.
92
0%
10%
20%
30%
40%
50%
60%
VEGF Microspheres NIV
Pu
nct
ure
Fo
rce
<1
N
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
VEGF Microspheres NIV
Pu
nct
ure
Fo
rce
(N)
†a
b
Figure 5-3 – Data for puncture force testing using local injections of various solutions. Panel a) shows puncture force values of all 3 groups. † - Data is not statistically significant. Panel b) shows the proportion of injections less than cutoff point of 1N.
93
Figure 5-4 – Movat pentachrome sections from each treatment group Left panel shows VEGF treated sample with increased vascularity. Centre panel shows BSA Microspheres injection only. Right panel shows sample from NIV group. Histology courtesy of Dr. Bradley Strauss.
5.4 Future work: Collagenase dynamics studies
One of the long standing questions of the collagenase technique, and for that matter, others like it
relates to the dynamics of the enzyme. We inject 1mL of collagenase into less than 1mL of space
between the balloon catheter and the occlusion. Certainly there is some combination of leakage through
sidebranches, as well as some penetration into the occlusion itself, but we do not currently have a clear
understanding of a) where the collagenase goes and b) how long it stays there. We are able to obtain
some sense of the residency of a fluid in the proximal space by performing angiographic contrast
injections through wire port of the OTW balloon and periodically checking for the persistence of contrast
on fluoroscopic images. While this is useful in some respects, it is also very limited for a number of
reasons:
a) X-ray contrast and collagenase solutions have very different viscosities and adhesion properties, so
one cannot expect that they will behave similarly;
b) Collagenase has a particular affinity for its substrate, suggesting that it will be more likely to remain in
the region of the collagen rich PFC compared with X-ray contrast;
94
c) The ability to visualize the presence of contrast in an angiogram is very limited by the concentration of
the contrast agent in a given pixel - it is relatively insensitive to flow in small channels;
d) Vessels in this configuration are extremely susceptible to dissection from injection through the
balloon’s wire port. Contrast, due to its viscosity, is particularly dangerous, as a large amount of pressure
is needed for injection. As a result, contrast injections need to be minimized.
There is work currently underway to address these limitations. A study is attempting to bind molecules
such as collagenase to an MRI contrast agent, gadolinium, via a chelate molecule -
tetraazacyclododecane-tetraacetic acid (DOTA) as a way of tracking molecular location over time. Bound
collagenase will then be injected through the wire port of an OTW balloon into a rabbit femoral artery
model of CTO in the same fashion as has been performed in the studies described within this thesis. T1
maps of the animals are generated prior to injection, 1 hour following injection, 2 hours following
injection, as well as 24 hours after injection. While this pilot study is very useful in showing the proof of
concept, as well as to gain an appreciation for how much collagenase remains over a 24 hour period, a
study looking at much more frequent timepoints immediately after treatment with the rabbit in the
same position would allow for better understanding of which parameters lead to different penetrations
and residency times of collagenase.
5.5 Future work: Catheter based technology
The ultimate goal of all the studies presented here would be to adapt them to catheter based solutions.
It is true that imaging and therapeutics can both be performed adequately through the use
transcutaneous probes for peripheral vessels, since there are usually adequate acoustic imaging and
treatment windows present for most peripheral vessels. Furthermore, peripheral vessels are generally
within a few cm of the skin surface, so reasonably high frequency (10-12MHz, and potentially even
higher) imaging probes can be used to obtain adequate imaging resolution.
95
This is not, however, true of the coronary vessels, where the more significant demand for these devices
lies. Limited imaging windows and attenuation through tissue make transcutaneous imaging and
therapy extremely difficult in coronary vessels.
5.5.1 Imaging
There has been significant effort within the medical device community to develop intravascular imaging
solutions to CTO, none of which have been particularly clinically successful. Since the only clinically
available IVUS is sideviewing, it does not provide adequate visualization of CTOs. A forward looking
imaging device is required. While some attempts at forward looking imaging have been attempted, none
have found commercial success. Compounding the problem, as shown in Figure 5-5, 2D ultrasound
visualization of a heavily remodelled CTO is often insufficient. There is a need for something beyond B-
scan ultrasound imaging. There are a number of potential solutions currently under investigation. One
potential solution is RF analysis of ultrasound data obtained from CTO imaging. Volcano Corporation has
developed a technique(111), commercially referred to as Virtual Histology (VH) IVUS. While this
technique has had commercial success, it has been questioned in the scientific community in terms of its
accuracy (112,113). The ultimate imaging solution may require both forward viewing imaging as well as
a technique beyond standard B-scan imaging.
A multimodality approach is also attractive, since imaging modalities with different contrast mechanisms
may provide complimentary information. The most logical pairing for intravascular imaging is the
combination of IVUS and Optical Coherence Tomography (OCT)(114). OCT is essentially an optical
analogue of ultrasound, and is also amenable to intravascular imaging. It provides significantly higher
resolution than IVUS, but has minimal penetration depth. Pilot studies have been performed co-
registering IVUS and OCT images(115) in the standard cross-sectional orientation, shown in Figure 5-6
96
for an atherosclerotic plaque. Benchtop studies of µUS compared with OCT of human peripheral CTOs
are shown in Figure 5-7.
Figure 5-5 – Typical individual crossectional B-scan image of a heavily remodelled rabbit CTO artery. Artery is located within ellipse, and is not easily delineated using standard imaging techniques.
Approaches to combine these techniques into a single device are under development. The proposed
device would include a technique for forward looking, 3D, combined IVUS-OCT imaging. The device
could also include Doppler techniques to detect microchannels. This technique is, however, still in its
developmental infancy.
97
Figure 5-6 - Combined IVUS-OCT image of an atherosclerotic plaque. IVUS on far left shows good boundary contrast between media and intimal plaque. Calcium is shown as shadow from 6 o'clock to 7 o'clock. OCT image, second from left, shows much higher resolution in fibrous plaque. Histology, second from right, shows a fibrous plaque with large calcium chunk (5 o’clock to 7 o’clock) stained with a Movat Pentachrome. µCT, far right, shows calcification (5 o’clock to 7 o’clock).
5.5.2 Therapy
In order to effectively provide localized UMM therapy, several conditions are necessary. Firstly,
significant acoustic pressure needs to be present at the site of therapy. This requires that the
attenuation for the path from the acoustic source to the therapy site be minimal. This implies that
having the therapy transducer as close as possible to the PFC is desirable. Secondly, a tightly focused
beam is required to ensure disruption in only desired areas to minimize collateral damage. Thirdly,
accurate knowledge of the treatment location in space is required. Focused acoustic beams can drop off
quickly in both the lateral and axial directions, and treatment will be completely ineffective in the
absence of sufficient acoustic power in the desired location. All three of the above points favour the
development of a catheter based solution to CTO therapy. On a catheter, the attenuation would be
minimal since the probe would be very close to the treatment location. A small aperture device would
create a small beam, and minimize the collateral disruption. Moreover, having the therapy probe
immediately adjacent to the lesion would simplify lesion targeting, and an angiogram could be used for
guidance. Finally, having a therapeutic probe directed forward beyond the end of the catheter could
benefit from using primary radiation force on the bubbles to give the bubbles a push into the PFC prior
to disruption. This could be achieved by using long low-intensity pulses to push the bubbles away from
98
the acoustic source towards the PFC. A short, high intensity pulse would then disrupt the bubbles to
achieve the therapeutic effect.
There are however, significant engineering challenges inherent in building a catheter based therapeutic
ultrasound system. One limitation is aperture size. Low frequency devices tend to require larger
aperture sizes and probe thickness. A compromise would have to be made with regards to therapeutic
effect (achieved with low frequencies) and device size (minimized with high frequencies).
5.5.3 Image Guided Therapeutics
The coupling of a catheter-based ultrasound imaging probe (as described in 5.5.1) with a catheter-based
ultrasound therapy probe (as described in 5.5.2) could provide lesion characterization using the imaging
tool, followed by therapy. The imaging tool could then be brought in to assess the efficacy of the
therapy.
5.5.4 Compliance Testing
The force testing apparatus as described in this thesis was useful as an ex vivo testing mechanism, used
for characterizing maturation and evaluating enzyme treatment efficacy. However, there are limitations
inherent with performing tests using the system presented here. Firstly, ex-vivo testing removes the
vessel from its native environment, and while the saline used for treatments is relatively biological, the
absence of blood may have an effect on puncturing CTOs. Secondly, the system implemented in this
thesis uses motion in a single axis to perform the puncture. In reality, CTO crossings are often done with
far more complex movements, which may include torsion being applied to the guidewire, as well as
translation in multiple axes. If one could capture the forces applied to a guidewire using the true range
of motions performed in the catheterization laboratory, it would provide a more realistic
characterization.
99
Figure 5-7 - Comparison of forward viewing micro ultrasound and OCT images in CTO. Top left shows a reconstructed cross-section of CTO with 40MHz ultrasound imaging. Top right shows similar image generated with OCT. OCT image shows finer detail, and detects microchannels not visible with B-scan ultrasound image alone. Bottom left shows longitudinal image of a different CTO vessel performed with the same systems. Ultrasound image shows vastly superior penetration depth. OCT images courtesy of Dr. Nigel Munce.
There is commercial test equipment available that could be used to perform this type of
characterization. One system, produced by Machine Solutions Inc. (Model# IDTE 2000) allows the
accurate measurement and display of real-time force-displacement data that is oriented horizontally
100
and could be coupled to an arbitrary guidewire within an arbitrary sheath with minimal modification,
and could be used in vivo. While this system is also currently set up to perform only linear translation, it
is highly configurable, and algorithms have been developed (Hansen Medical, Intellisense™), that could
be used to track forces applied by human manipulated guidewires. The efficacy of these algorithms,
however, has not been validated.
It would also be of significant value to have the ability to compare lesion composition and morphology
to puncture force. This is not possible with the setup described here, since the puncture disrupts the
lesion, rendering lesion characterization through histology impractical. This may be overcome through
the use of lesion characterization imaging techniques prior to puncture. Conversely, one could measure
compliance non-invasively using an elastography technique (i.e. ultrasound elastography or MRI
elastography). Compliance measurements could then be compared to composition as determined by
histology.
5.6 Concluding remarks
Chronic total occlusions have been a long standing problem for interventional cardiologists, and have
lead to a large graveyard of abandoned devices, many originally thought to be one step all-
encompassing solutions. It appears that CTO treatment is too complicated to solve with any single
device, and appears to require a multipronged approach. One of the main outstanding issues is lesion
characterization. This requires a better understanding of lesion pathology and maturation, and this
thesis has attempted to shed some light upon this issue. Once an understanding of overall disease
progression is attained, in vivo diagnostics – be it X-ray angiography, IVUS, OCT, MRI, CT, or some
combination therein, will be required to determine the optimal strategy based on case by case lesion
characteristics. Furthermore, the development of an array of therapeutic approaches according to the
lesion specific information gathered appears to be necessary. The CTO problem is more complicated
101
than many device manufacturers had hoped, and will require an equally complex set of approaches for
its optimal clinical resolution.
102
6 References
1. Braunwald E, Zipes DP, Libby P. 2001. Heart disease : a textbook of cardiovascular medicine. Philadelphia: Saunders.
2. Leibowitz J. 1970. The History of Coronray Heart Disease. Berkley: University of California Press. 3. Osler W. 1892. The principles and practice of medicine : designed for the use of practitioners
and students of medicine. New York: D. Appleton and Co.; xvi, 1079, 8 p. p. 4. Ross R. Atherosclerosis--an inflammatory disease. New England Journal of Medicine
1999;340(2):115-26. 5. Kruth HS. The fate of lipoprotein cholesterol entering the arterial wall. Current Opinion in
Lipidology 1997;8(5):246-52. 6. Williams KJ, Tabas I. The response-to-retention hypothesis of atherogenesis reinforced. Current
Opinion in Lipidology 1998;9(5):471-4. 7. Williams KJ, Tabas I. The response-to-retention hypothesis of early atherogenesis.
Arteriosclerosis, Thrombosis & Vascular Biology 1995;15(5):551-61. 8. Libby P. Atherosclerosis: The New View. Scientific American 2002;286(5):46-55. 9. Kolodgie FD, Burke AP, Nakazawa G, Cheng Q, Xu X, Virmani R. Free cholesterol in
atherosclerotic plaques: where does it come from? Current Opinion in Lipidology 2007;18(5):500-7.
10. Kolodgie FD, Gold HK, Burke AP, Fowler DR, Kruth HS, Weber DK, Farb A, Guerrero LJ, Hayase M, Kutys R and others. Intraplaque hemorrhage and progression of coronary atheroma. New England Journal of Medicine 2003;349(24):2316-25.
11. Virmani R, Burke AP, Kolodgie FD, Farb A. Vulnerable plaque: the pathology of unstable coronary lesions. Journal of Interventional Cardiology 2002;15(6):439-46.
12. Virmani R, Kolodgie FD, Burke AP, Farb A, Schwartz SM. Lessons from sudden coronary death: a comprehensive morphological classification scheme for atherosclerotic lesions. Arterioscler Thromb Vasc Biol 2000;20(5):1262-1275.
13. Glagov S, Weisenberg E, Zarins CK, Stankunavicius R, Kolettis GJ. Compensatory enlargement of human atherosclerotic coronary arteries. New England Journal of Medicine 1987;316(22):1371-5.
14. Hirsch AT, Haskal ZJ, Hertzer NR, Bakal CW, Creager MA, Halperin JL, Hiratzka LF, Murphy WR, Olin JW, Puschett JB and others. ACC/AHA 2005 guidelines for the management of patients with peripheral arterial disease (lower extremity, renal, mesenteric, and abdominal aortic): executive summary a collaborative report from the American Association for Vascular Surgery/Society for Vascular Surgery, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, Society of Interventional Radiology, and the ACC/AHA Task Force on Practice Guidelines (Writing Committee to Develop Guidelines for the Management of Patients With Peripheral Arterial Disease) endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation; National Heart, Lung, and Blood Institute; Society for Vascular Nursing; TransAtlantic Inter-Society Consensus; and Vascular Disease Foundation. Journal of the American College of Cardiology 2006;47(6):1239-312.
15. Hoye A, Tanabe K, Lemos PA, Aoki J, Saia F, Arampatzis C, Degertekin M, Hofma SH, Sianos G, McFadden E and others. Significant reduction in restenosis after the use of sirolimus-eluting stents in the treatment of chronic total occlusions. J Am Coll Cardiol 2004;43(11):1954-1958.
103
16. Selnes OA, Goldsborough MA, Borowicz LM, McKhann GM. Neurobehavioural sequelae of cardiopulmonary bypass. Lancet 1999;353(9164):1601-1606.
17. Cordero H, Warburton KD, Underwood PL, Heuser RR. Initial experience and safety in the treatment of chronic total occlusions with fiberoptic guidance technology: optical coherent reflectometry. Catheter Cardiovasc Interv 2001;54(2):180-187.
18. Cannon CP, Battler A, Brindis RG, Cox JL, Ellis SG, Every NR, Flaherty JT, Harrington RA, Krumholz HM, Simoons ML and others. American College of Cardiology key data elements and definitions for measuring the clinical management and outcomes of patients with acute coronary syndromes. A report of the American College of Cardiology Task Force on Clinical Data Standards (Acute Coronary Syndromes Writing Committee). J Am Coll Cardiol 2001;38(7):2114-2130.
19. Sionis DG, Tolis V, Michalis L. Chronic Total Coronary Occlusions: A Review of their Special Features and the Existing Techniques for Percutaneous Treatment. Hellenic J Cardiol 2003;44:136-142.
20. Werner GS, Emig U, Mutschke O, Schwarz G, Bahrmann P, Figulla HR. Regression of collateral function after recanalization of chronic total coronary occlusions: a serial assessment by intracoronary pressure and Doppler recordings. Circulation 2003;108(23):2877-2882.
21. Stewart JT, Denne L, Bowker TJ, Mulcahy DA, Williams MG, Buller NP, Sigwart U, Rickards AF. Percutaneous transluminal coronary angioplasty in chronic coronary artery occlusion. J Am Coll Cardiol 1993;21(6):1371-1376.
22. Stone GW, Kandzari DE, Mehran R, Colombo A, Schwartz RS, Bailey S, Moussa I, Teirstein PS, Dangas G, Baim DS and others. Percutaneous recanalization of chronically occluded coronary arteries: a consensus document: part I. Circulation 2005;112(15):2364-2372.
23. Kahn JK. Angiographic suitability for catheter revascularization of total coronary occlusions in patients from a community hospital setting. Am Heart J 1993;126(3 Pt 1):561-564.
24. Williams DO, Holubkov R, Yeh W, Bourassa MG, Al-Bassam M, Block PC, Coady P, Cohen H, Cowley M, Dorros G and others. Percutaneous coronary intervention in the current era compared with 1985-1986: the National Heart, Lung, and Blood Institute Registries. Circulation 2000;102(24):2945-2951.
25. Boden WE, O'Rourke RA, Teo KK, Hartigan PM, Maron DJ, Kostuk WJ, Knudtson M, Dada M, Casperson P, Harris CL and others. Optimal medical therapy with or without PCI for stable coronary disease. N Engl J Med 2007;356(15):1503-16.
26. Danchin N, Angioi M, Cador R, Tricoche O, Dibon O, Juilliere Y, Cuilliere M, Cherrier F. Effect of late percutaneous angioplastic recanalization of total coronary artery occlusion on left ventricular remodeling, ejection fraction, and regional wall motion. Am J Cardiol 1996;78(7):729-735.
27. Noguchi T, MD SM, Morii I, Daikoku S, Goto Y, Nonogi H. Percutaneous transluminal coronary angioplasty of chronic total occlusions. Determinants of primary success and long-term clinical outcome. Catheter Cardiovasc Interv 2000;49(3):258-264.
28. Safley DM, House JA, Marso SP, Grantham JA, Rutherford BD. Improvement in survival following successful percutaneous coronary intervention of coronary chronic total occlusions: variability by target vessel. Jacc: Cardiovascular Interventions 2008;1(3):295-302.
29. Claessen BE, van der Schaaf RJ, Verouden NJ, Stegenga NK, Engstrom AE, Sjauw KD, Kikkert WJ, Vis MM, Baan J, Jr., Koch KT and others. Evaluation of the effect of a concurrent chronic total occlusion on long-term mortality and left ventricular function in patients after primary percutaneous coronary intervention. JACC Cardiovasc Interv 2009;2(11):1128-34.
30. Christ G, Glogar D. Successful recanalization of a chronic occluded left anterior descending coronary artery with a modification of the retrograde proximal true lumen puncture technique: the antegrade microcatheter probing technique. Catheter Cardiovasc Interv 2009;73(2):272-5.
104
31. Katsuragawa M, Fujiwara H, Miyamae M, Sasayama S. Histologic studies in percutaneous transluminal coronary angioplasty for chronic total occlusion: comparison of tapering and abrupt types of occlusion and short and long occluded segments. J Am Coll Cardiol 1993;21(3):604-611.
32. Srivatsa SS, Edwards WD, Boos CM, Grill DE, Sangiorgi GM, Garratt KN, Schwartz RS, Jr Holmes D. Histologic correlates of angiographic chronic total coronary artery occlusions: influence of occlusion duration on neovascular channel patterns and intimal plaque composition. J Am Coll Cardiol 1997;29(5):955-963.
33. Dible H. Organization and canalization in arterial thrombosis. Journal of Pathology and Bacteriology 1958;75(1):1-7.
34. Strauss BH, Segev A, Wright GA, Qiang B, Munce N, Anderson KJT, Leung G, Dick AJ, Virmani R, Butany J. Microvessels in chronic total occlusions: pathways for successful guidewire crossing? J Interv Cardiol 2005;18(6):425-436.
35. Jaffe R, Leung G, Munce NR, Thind AS, Leong-Poi H, Anderson KJ, Qi X, Trogadis J, Nadler A, Shiff D and others. Natural history of experimental arterial chronic total occlusions. Journal of the American College of Cardiology 2009;53(13):1148-58.
36. Bourassa MG, Roubin GS, Detre KM, Sopko G, Krone RJ, Attabuto MJ, Bjerregaad P, Bolling S, Herman MV, Frye R. Bypass Angioplasty Revascularization Investigation: patient screening, selection, and recruitment. Am J Cardiol 1995;75(9):3C-8C.
37. Puma JA, Sketch MH, Jr., Tcheng JE, Harrington RA, Phillips HR, Stack RS, Califf RM. Percutaneous revascularization of chronic coronary occlusions: an overview. J Am Coll Cardiol 1995;26(1):1-11.
38. Colombo A, Mikhail GW, Michev I, Iakovou I, Airoldi F, Chieffo A, Rogacka R, Carlino M, Montorfano M, Sangiorgi GM and others. Treating chronic total occlusions using subintimal tracking and reentry: the STAR technique. Catheter Cardiovasc Interv 2005;64(4):407-411.
39. Lavinsky D, Arterni NS, Achaval M, Netto CA. Chronic bilateral common carotid artery occlusion: a model for ocular ischemic syndrome in the rat. Graefes Arch Clin Exp Ophthalmol 2006;244(2):199-204.
40. Litvak J, Siderides LE, Vineberg AM. The experimental production of coronary artery insufficiency and occlusion. Am Heart J 1957;53(4):505-18.
41. Strauss BH, Goldman L, Qiang B, Nili N, Segev A, Butany J, Sparkes JD, Jackson ZS, Eskandarian MR, Virmani R. Collagenase plaque digestion for facilitating guide wire crossing in chronic total occlusions. Circulation 2003;108(10):1259-1262.
42. Murphy TP, Dorfman GS, Esparza AR, Duwaji MS, Smith WJ. Arteriosclerosis obliterans in a rabbit model. Investigative Radiology 1992;27(12):1059-63.
43. Prosser L, Agrawal CM, Polan J, Elliott J, Adams DG, Bailey SR. Implantation of oxygen enhanced, three-dimensional microporous L-PLA polymers: a reproducible porcine model of chronic total coronary occlusion. Catheterization & Cardiovascular Interventions 2006;67(3):412-416.
44. Thind AS, Leung G, Munce NR, Graham JJ, Anderson KJ, Dick AJ, Strauss BH, Wright GA, Foster FS. Investigation of micro-ultrasound for microvessel imaging in a model of chronic total occlusion. Ultrason Imaging 2007;29(3):167-81.
45. Suzuki K, Saito N, Zhang G, Conditt G, McGregor J, Flynn AM, Leahy D, Glennon P, Leon MB, Hayase M. Development of a novel calcified total occlusion model in porcine coronary arteries. J Invasive Cardiol 2008;20(6):296-301.
46. Suzuki Y, Oyane A, Ikeno F, Lyons JK, Yeung AC. Development of animal model for calcified chronic total occlusion. Catheter Cardiovasc Interv 2009;74(3):468-75.
47. Song W, Lee J, Kim H, Shin J, Oh D, Tio F, Wong SC, Hong MK. A new percutaneous porcine coronary model of chronic total occlusion. J Invasive Cardiol 2005;17(9):452-4.
105
48. Suero JA, Marso SP, Jones PG, Laster SB, Huber KC, Giorgi LV, Johnson WL, Rutherford BD. Procedural outcomes and long-term survival among patients undergoing percutaneous coronary intervention of a chronic total occlusion in native coronary arteries: a 20-year experience. J Am Coll Cardiol 2001;38(2):409-14.
49. Melzi G, Cosgrave J, Biondi-Zoccai GL, Airoldi F, Michev I, Chieffo A, Sangiorgi GM, Montorfano M, Carlino M, Colombo A. A novel approach to chronic total occlusions: the crosser system. Catheter Cardiovasc Interv 2006;68(1):29-35.
50. Loli A, Liu R, Pershad A. Immediate- and short-term outcome following recanalization of long chronic total occlusions (> 50 mm) of native coronary arteries with the Frontrunner catheter. J Invasive Cardiol 2006;18(6):283-5.
51. Garcia-Garcia HM, Kukreja N, Daemen J, Tanimoto S, van Mieghem C, Gonzalo N, van Weenen S, van der Ent M, Sianos G, de Feyter P and others. Contemporary treatment of patients with chronic total occlusion: critical appraisal of different state-of-the-art techniques and devices. EuroIntervention 2007;3(2):188-96.
52. Saito S. Different strategies of retrograde approach in coronary angioplasty for chronic total occlusion. Catheter Cardiovasc Interv 2008;71(1):8-19.
53. Saito S. Open Sesame Technique for chronic total occlusion. Catheter Cardiovasc Interv;75(5):690-4.
54. Abbas AE, Brewington SD, Dixon SR, Boura JA, Grines CL, O'Neill WW. Intracoronary fibrin-specific thrombolytic infusion facilitates percutaneous recanalization of chronic total occlusion. J Am Coll Cardiol 2005;46(5):793-8.
55. Carlino M, Latib A, Godino C, Cosgrave J, Colombo A. CTO recanalization by intraocclusion injection of contrast: the microchannel technique. Catheter Cardiovasc Interv 2008;71(1):20-6.
56. Segev A, Nili N, Qiang B, Charron T, Butany J, Strauss BH. Human-grade purified collagenase for the treatment of experimental arterial chronic total occlusion. Cardiovascular Revascularization Medicine 2005;6(2):65-9.
57. Di Lullo GA, Sweeney SM, Korkko J, Ala-Kokko L, San Antonio JD. Mapping the ligand-binding sites and disease-associated mutations on the most abundant protein in the human, type I collagen. J Biol Chem 2002;277(6):4223-31.
58. Neuman RE, Logan MA. The determination of hydroxyproline. J Biol Chem 1950;184(1):299-306. 59. Alberts B, Alexander Johnson, Julian Lewis, Martin Raff, Keith Roberts, Walter. P. 2002.
Molecular Biology of the Cell. New York, NY: Garland Science; 1463 p. 60. Harris ED, Jr., Krane SM. Collagenases (first of three parts). New England Journal of Medicine
1974;291(11):557-63. 61. Harris ED, Jr., Krane SM. Collagenases (second of three parts). New England Journal of Medicine
1974;291(12):605-9. 62. Seifter S, Harper E. Collagenases. Methods in enzymology 1970;19:613-635. 63. Wild J. The use of ultrasonic pulses for the measurement of biologic tissues and the detection of
tissue density changes. Surgery 1950;27(2):183-8. 64. Cobbald RSC. 2006. Foundations of Biomedical Ultrasound: Oxford University Press; 832 p. 65. Evans DH. 1989. Doppler Ultrasound: Physics, Instrumentation, and Clinical Applications: John
Wiley & Sons. 66. McDicken WNH, P.R. 2000. Clinical Doppler Ultrasound: Harcourt Publishers Limited. 67. Rubin JM, Bude RO, Carson PL, Bree RL, Adler RS. Power Doppler US: a potentially useful
alternative to mean frequency-based color Doppler US. Radiology 1994;190(3):853-856. 68. Simpson DH, Chin CT, Burns PN. Pulse inversion Doppler: a new method for detecting nonlinear
echoes from microbubble contrast agents. IEEE Trans Ultrason Ferroelectr Freq Control 1999;46(2):372-82.
106
69. Leavens C, Williams R, Foster FS, Burns PN, Sherar MD. Golay pulse encoding for microbubble contrast imaging in ultrasound. IEEE Trans Ultrason Ferroelectr Freq Control 2007;54(10):2082-90.
70. Needles A, Couture O, Foster FS. A method for differentiating targeted microbubbles in real time using subharmonic micro-ultrasound and interframe filtering. Ultrasound Med Biol 2009;35(9):1564-73.
71. Al-Mansour HA, Mulvagh SL, Pumper GM, Klarich KW, Foley DA. Usefulness of harmonic imaging for left ventricular opacification and endocardial border delineation by optison. Am J Cardiol 2000;85(6):795-9, A10.
72. Wei K, Jayaweera AR, Firoozan S, Linka A, Skyba DM, Kaul S. Quantification of myocardial blood flow with ultrasound-induced destruction of microbubbles administered as a constant venous infusion. Circulation 1998;97(5):473-83.
73. Lindner JR, Wei K, Kaul S. Imaging of Myocardial Perfusion with SonoVuetrade mark in Patients with a Prior Myocardial Infarction. Echocardiography 1999;16(7, Pt 2):753-760.
74. Leen E, Moug SJ, Horgan P. Potential impact and utilization of ultrasound contrast media. Eur Radiol 2004;14 Suppl 8:P16-24.
75. Bevan PD, Karshafian R, Tickner EG, Burns PN. Quantitative measurement of ultrasound disruption of polymer-shelled microbubbles. Ultrasound Med Biol 2007;33(11):1777-86.
76. Wu J, Nyborg WL. Ultrasound, cavitation bubbles and their interaction with cells. Adv Drug Deliv Rev 2008;60(10):1103-16.
77. King DA, Malloy MJ, Roberts AC, Haak A, Yoder CC, O'Brien WD, Jr. Determination of postexcitation thresholds for single ultrasound contrast agent microbubbles using double passive cavitation detection. J Acoust Soc Am;127(6):3449-55.
78. Yeh CK, Su SY. Effects of acoustic insonation parameters on ultrasound contrast agent destruction. Ultrasound Med Biol 2008;34(8):1281-91.
79. Larkin JO, Casey GD, Tangney M, Cashman J, Collins CG, Soden DM, O'Sullivan GC. Effective tumor treatment using optimized ultrasound-mediated delivery of bleomycin. Ultrasound Med Biol 2008;34(3):406-13.
80. Karshafian R, Bevan PD, Williams R, Samac S, Burns PN. Sonoporation by ultrasound-activated microbubble contrast agents: effect of acoustic exposure parameters on cell membrane permeability and cell viability. Ultrasound Med Biol 2009;35(5):847-60.
81. Hynynen K, McDannold N, Vykhodtseva N, Jolesz FA. Noninvasive MR imaging-guided focal opening of the blood-brain barrier in rabbits. Radiology 2001;220(3):640-6.
82. Hynynen K, McDannold N, Vykhodtseva N, Jolesz FA. Non-invasive opening of BBB by focused ultrasound. Acta Neurochir Suppl 2003;86:555-8.
83. Hynynen K. Macromolecular delivery across the blood-brain barrier. Methods Mol Biol 2009;480:175-85.
84. Meairs S, Culp W. Microbubbles for thrombolysis of acute ischemic stroke. Cerebrovascular Diseases 2009;27 Suppl 2:55-65.
85. Tachibana K, Tachibana S. Albumin microbubble echo-contrast material as an enhancer for ultrasound accelerated thrombolysis. Circulation 1995;92(5):1148-50.
86. Mossop PJ, Amukotuwa SA, Whitbourn RJ. Controlled Blunt Microdissection for Percutaneous Recanalization of Lower Limb Arterial Chronic Total Occlusion: A Single Center Experience. Catheterization & Cardiovascular Interventions 2006;68(2):304-10.
87. 2002. Medtech Insight 4; 117 p. 88. Marxen M, Thornton MM, Chiarot CB, Klement G, Koprivnikar J, Sled JG, Henkelman RM.
MicroCT scanner performance and considerations for vascular specimen imaging. Medical Physics 2004;31(2):305-13.
107
89. Cherin E, Williams R, Needles A, Liu G, White C, Brown AS, Zhou YQ, Foster FS. Ultrahigh frame rate retrospective ultrasound microimaging and blood flow visualization in mice in vivo. Ultrasound in Medicine & Biology 2006;32(5):683-91.
90. Goertz DE, Christopher DA, Yu JL, Kerbel RS, Burns PN, Foster FS. High-frequency color flow imaging of the microcirculation. Ultrasound Med Biol 2000;26(1):63-71.
91. Gatzoulis L, Watson RJ, Jordan LB, Pye SD, Anderson T, Uren N, Salter DM, Fox KA, McDicken WN. Three-dimensional forward-viewing intravascular ultrasound imaging of human arteries in vitro. Ultrasound in Medicine & Biology 2001;27(7):969-82.
92. Light ED, Smith SW. Two dimensional arrays for real time 3D intravascular ultrasound. Ultrasonic Imaging 2004;26(2):115-28.
93. Chen J-J, Esashi M, Haga Y. Small diameter ultrasound imager for intraluminal forward-looking inspection. 2005. p 4 pp.
94. Pinto TL, Waksman R. Clinical applications of optical coherence tomography.[see comment]. Journal of Interventional Cardiology 2006;19(6):566-73.
95. Gatzoulis L, Ramnarine KV, Pye SD, Anderson T, Newby DE, Hoskins PR, McDicken WN. Doppler colour flow imaging and flow quantification with a novel forward-viewing intravascular ultrasound system. Ultrasound in Medicine & Biology 2003;29(1):53-64.
96. Goertz DE, Needles A, Burns PN, Foster FS. High-frequency, nonlinear flow imaging of microbubble contrast agents. IEEE Transactions on Ultrasonics Ferroelectrics & Frequency Control 2005;52(3):495-502.
97. Dick AJ, Raman VK, Raval AN, Guttman MA, Thompson RB, Ozturk C, Peters DC, Stine AM, Wright VJ, Schenke WH and others. Invasive human magnetic resonance imaging: feasibility during revascularization in a combined XMR suite. Catheterization & Cardiovascular Interventions 2005;64(3):265-74.
98. Weiss CR, Georgiades C, Hofmann LV, Schulick R, Choti M, Thuluvath P, Bluemke DA, Arepally A. Intrabiliary MR imaging: assessment of biliary obstruction with use of an intraluminal MR receiver coil. Journal of Vascular & Interventional Radiology 2006;17(5):845-53.
99. Larose E, Yeghiazarians Y, Libby P, Yucel EK, Aikawa M, Kacher DF, Aikawa E, Kinlay S, Schoen FJ, Selwyn AP and others. Characterization of human atherosclerotic plaques by intravascular magnetic resonance imaging. Circulation 2005;112(15):2324-31.
100. Botnar RM, Bucker A, Kim WY, Viohl I, Gunther RW, Spuentrup E. Initial experiences with in vivo intravascular coronary vessel wall imaging. Journal of Magnetic Resonance Imaging 2003;17(5):615-9.
101. Hillenbrand CM, Elgort DR, Wong EY, Reykowski A, Wacker FK, Lewin JS, Duerk JL. Active device tracking and high-resolution intravascular MRI using a novel catheter-based, opposed-solenoid phased array coil. Magnetic Resonance in Medicine 2004;51(4):668-75.
102. Stone GW, Reifart NJ, Moussa I, Hoye A, Cox DA, Colombo A, Baim DS, Teirstein PS, Strauss BH, Selmon M and others. Percutaneous recanalization of chronically occluded coronary arteries: a consensus document: part II. Circulation 2005;112(16):2530-2537.
103. Dong S, Smorgick Y, Nahir M, Lotan C, Mosseri M, Nassar H, Gotsman MS, Hasin Y. Predictors for successful angioplasty of chronic totally occluded coronary arteries. Journal of Interventional Cardiology 2005;18(1):1-7.
104. Schober EA, Kusy RP, Savitz DA. Resistance of fetal membranes to concentrated force applications and reconciliation of puncture and burst testing. Annals of Biomedical Engineering 1994;22(5):540-8.
105. Saito H, Togawa T. Detection of needle puncture to blood vessel using puncture force measurement. Medical & Biological Engineering & Computing 2005;43(2):240-4.
108
106. Stone GW, Rutherford BD, McConahay DR, Johnson WL, Jr., Giorgi LV, Ligon RW, Hartzler GO. Procedural outcome of angioplasty for total coronary artery occlusion: an analysis of 971 lesions in 905 patients.[see comment]. Journal of the American College of Cardiology 1990;15(4):849-56.
107. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193(1):265-75.
108. Nemethy G, Scheraga HA. Stabilization of collagen fibrils by hydroxyproline. Biochemistry 1986;25(11):3184-8.
109. Reddy GK, Enwemeka CS. A simplified method for the analysis of hydroxyproline in biological tissues. Clinical Biochemistry 1996;29(3):225-9.
110. Edwards CA, O'Brien WD, Jr. Modified assay for determination of hydroxyproline in a tissue hydrolyzate. Clinica Chimica Acta 1980;104(2):161-7.
111. Nair A, Kuban BD, Tuzcu EM, Schoenhagen P, Nissen SE, Vince DG. Coronary plaque classification with intravascular ultrasound radiofrequency data analysis. Circulation 2002;106(17):2200-6.
112. Thim T, Hagensen MK, Wallace-Bradley D, Granada JF, Kaluza GL, Drouet L, Paaske WP, Botker HE, Falk E. Unreliable Assessment of Necrotic Core by VHTM IVUS in Porcine Coronary Artery Disease. Circ Cardiovasc Imaging.
113. Granada JF, Wallace-Bradley D, Win HK, Alviar CL, Builes A, Lev EI, Barrios R, Schulz DG, Raizner AE, Kaluza GL. In vivo plaque characterization using intravascular ultrasound-virtual histology in a porcine model of complex coronary lesions. Arterioscler Thromb Vasc Biol 2007;27(2):387-93.
114. Courtney BK, Li BH, Leung AS, Soong A, Munding CE, Thind AS, Munce NR, Lee H, Yang VXD, Strauss BH and others. 2010.Images of Human Coronary Atherosclerosis From a Hybrid Intravascular Ultrasound and Optical Coherence Tomography Catheter. TCT. Washington.
115. Li BH, Leung ASO, Munding CE, Thind AS, Munce NR, Lee H, Strauss BH, Foster FS, Courtney aBK. 2009.Hybrid Ultrasound and Optical Imaging Catheter for Vascular Disease Characterization. American Heart Association. Orlando, FL.