Keywords
aneurysm - dissection - paraplegia - PEARS - TEVAR - prophylaxis - cure
Introduction
The concept of novel endovascular “net” prosthesis may provide safe and early treatment
of the root pathology of aneurysms instead of the current approach of prophylaxis
or cure of aneurysmal complications. Since 1994, we have reported[1] experimental results of an endovascular “net” prosthesis to strengthen the aortic
wall without interference with collateral branch perfusion. Branch vessel is the main
source of technical problems and complications of current endovascular procedures.
During these nearly three decades, we conceived and mechanically tested this device,
both ex vivo[1] and in vivo, in a swine aortic model[2] ([Fig. 1]). We explored a variety of “net” prosthesis configurations to find those most effective
and reliable from a strictly mechanical point of view. We also assessed the most appropriate
mesh dimensioning and configuration of the polyester layers, as well as miniaturization
of the delivery apparatus to render it suitable for use in the human patient.
Fig. 1 Past experiences[1] showed that polypropylene “Net” had been embodied deep into the vascular wall down
to reach the media and appears to be on the predictably physiopathological path of
any foreign material away from the blood stream. The net threads of the part not encased
up to the media are anyway entirely wrapped by neointima. (A) The net prosthesis. (B) The wired net sectioned with the vascular wall exposing the inner vascular lumen.
(C) The vascular wall after removal of the nitinol wires. Blue outlined squares histology
of the (C) blue selection.
The purpose of this article is to illustrate the principles and implications of this
new approach, to present our latest endovascular “net” prosthesis devices, and to
focus on details learned along this experimental path.
Background
Looking at the aortic surgery evolution timeline over the past century, some facts
in the technological domain of the surgical art could perhaps appear as curious and
eventually meaningful.
One of the first attempts at the treatment of an abdominal aortic aneurysm by cellophane external partial wrapping was performed (1948) on the most famous theoretical physicist of all time (Einstein).
The Carrel vascular suture technique had been developed 46 years earlier (1902).[3] The first appropriate successful application of personalized mechanical support of the aorta (precision
therapy Personalized External Aortic Root Support [PEARS]) was done in 2004.[4]
On the contrary, anatomical or functional full substitution of the aortic aneurysm tract by a polyester (and/or later on by polytetrafluoroethylene)
graft, first introduced in 1951 by DeBakey, is still currently performed, by open
or endovascular means, the latter being first applied by Volodos[5] in 1984 and popularized in the western countries by Parodi et al.[6] It must be acknowledged, however, that this current clinical path had been first
performed, even in its hybrid setting, in 1982 by H.G. Borst, by his conventional
elephant trunk technique.[7]
Both open and endovascular full substitution techniques (except the Borst Conventional Elephant Trunk), however, are hampered by potential impairment of the collateral branches of the
substituted tract, whose perfusion must generally be restored, either surgically or
by appropriate further endovascular procedures. Replacement grafts cannot generally
preserve perfusion of the all-important spinal segmental arteries, which are often
included in the substituted tract.[8] Flow deprivation induces important anatomical changes in the spinal cord arterial
perfusion network, which[9]
[10] may result in intraoperative or later postoperative paraplegia, despite eventual
development of collateral circulation.
Whatever may be the possible causative factors (degenerative, genetic, infective,
inflammatory, or traumatic)[11]
[12] and the related pathophysiological events (proteolysis of the structural components
of the aortic wall, inflammation, and abnormal biomechanical forces), aortic aneurysm
formation is associated with impairment of two critical structural elements of the
aortic wall: elastin and collagen.[11]
[12] Overall, the gradual reduction of the aortic wall strength causes the most evident
feature of aortic aneurysm: dilatation, which, in fact, was first suggested in the
external aortic support attempted in the case of Einstein. On the contrary, rupture
and dissection, the true clinically relevant adverse events of aortic aneurysm, although
beginning with weakness of the aortic wall layers, imply a continuity solution at
the inner aortic surface.
New Target
However, from a merely mechanical point of view, these pathogenic contexts could perhaps
indicate total substitution of an aortic aneurysm, either surgically or by endovascular
procedures, such an approach may represent overtreatment. It may rather be preferable to strengthen the aortic wall, by addressing the inner[13] rather than the outer aortic surface.[14] This hypothesis is made obviously easier after the widespread diffusion of endovascular
techniques.
From the mere structural mechanic point of view, in fact, inner network support, being
applied directly at the origin and peak of the systolic stress,[15] allows more accurate configuration and dimensioning for optimal strengthening support.
Moreover, our “net,” positioned inside the aorta in stable contact with the intimal wall, is both spontaneously colonized and encased by neointima (as
usual for the inner surface of any vascular prosthesis). Also, our very first version
of the “net” is spontaneously moved deeper into the aortic wall ([Fig. 1]), perhaps along the natural pathophysiology path of any intraluminal foreign material.
Our original experimental swine model, in fact, showed migration of roughly one-half
of the circumference of the “net” tube into contact with the media layer after a few
weeks,[2] well beyond the intimal inner surface ([Fig. 1]). Although we first thought that the complete “net” neointimal wrapping may have
required more time for complete deep incorporation, the real explanation could be
instead that the maximal diameter of that “net” conduit, virtually inextensible in
that first polypropylene configuration, was simply too small to fully adhere and then
to be moved deeper all around the inner aortic surface.
This reasoning prompts a very important consideration: contrary to external support
devices,[4]
[14]
[16] the inner aortic diameter stabilization of our “net” devices does not rely on the
maximal diameter of the endovascular “net” prosthesis conduit. Rather, the mechanical strengthening unit of our device, in fact, is each and every single mesh whose threads, colonized by
fibroblasts, collagen, and other cells, circumscribe and then stabilize the small
inner aortic wall area, which also connects to the six confining meshes. Overall,
the aortic strengthened area is actually independent from the diameter of the “net”
conduit, being stabilized by the threads and the accompanying cellular proliferation
process.
This mechanism of aortic stabilization offers a very great advantage compared with
the external wrapping of the aorta, allowing a generous oversizing of the maximal
geometric diameter of the “net” conduit so that it may achieve the contact with more
dilated aortic areas, without accompanying danger or detriment.
Device Prototypes
The latest device consists essentially of three main parts: the polyester network
layer(s) ([Fig. 2], bottom), the support wireframe ([Figs. 2], top and [3A]) (consisting of a single thin nitinol wire defining a double-crossing spiral loop
personalized ([Fig. 3B, C]) at each “net” prosthesis segment, throughout the conduit), and the apparatus for
insertion, deployment, and in situ release of the “net” prosthesis ([Fig. 4]). Device configurations and possible constructive materials may vary according to
the vascular location, the local anatomy of the vascular tract, and, perhaps, some
other details that further experience would eventually suggest.
Fig. 2 Bottom: The 2 polyester layers (A and B) possible combinations for the prototypes in this experimental phase. From the strict
structural mechanics point of view a single polyester layer either with 5 (A) or 2.5 (B) mm meshes embodied into the aortic wall with the nitinol wireframe could largely
withstand any intravascular stress that may occur.[2] Top: These two prototypes, in the mono-A mesh layers configuration (mandatorily
always personalized on the patient computed tomography [CT] scan images), outline
the extremes that face this new approach identifying respectively the starting point
(left), i.e., giant aortic deformation, and the final goal of this research project
(right), i.e., the full prevention of genetic aneurysm and very, very early stabilization
of any other aortic aneurysms. Accurate personalization of the device allows the use
of a single layer. Contrariwise, current endovascular prosthesis the “net” does not
recanalize the blood stream then no/minimal stress at the ends; it is the geometrical
shape itself that prevents significant “net” dislocation.
Fig. 3 (A) Prostheses with double thinner mesh (B + B) layers as well as stronger (central,
shorter) or thinner (external, longer) nitinol wire (smaller or larger red-black circles)
that may allow to better comply with inner aortic wall pathology, being perhaps the
stronger nitinol wire more appropriate in presence of atherosclerotic pathology and
the lighter perhaps more suitable in Marfan and/or in dissection cases. (B, C) The process of personalization of both components of the expandable “net” device
is quite simple on the simulation of three-dimensional (3D) printed from patient computed
tomography (CT) scan images by pulling out the single nitinol wire at its both ends
(B, red arrows).
Fig. 4 Early prototype of expandable prosthesis for the arch in a bank simulation on the
swine model. 1 and 2, the expanded device in double layer; 3, the nitinol wireframe
retracted acting simultaneously at each of the 8 nitinol double crossing loops; 4,
the device is positioned into the aortic arch; 5, the nitinol wireframe was expanded
and the handle of the expanding mechanism withdrawn; 6, appearance of the deployed
prosthesis at the distal end. The same mechanical principle for retraction and expansion
can be applied, with minor modifications required for control of the selective relevant
different amplitude of some loops, for the endovascular positioning of all the nitinol/“net”
conduits illustrated.
The single or double large mesh (≥ 5 mm) layer ([Fig. 2], lower strip: A, A + A) does not significantly impact the blood flow rate of any
aortic collaterals and can be expected to be fully embodied into the aortic wall after
having been in contact for 2 to 5 months. Together with the nitinol wireframe, these
elements constitute a structural support to the vascular wall that can withstand forces
largely exceeding those applied by the bloodstream under clinical circumstances.[2]
On the contrary, devices with multiple layer meshes (≤ 1 mm) together with other layers
with meshes of various diameters form conduits that could eventually become fully
blood-sealed over time. The rate of blood sealing may be pharmacologically controllable
before full embodiment into the aortic wall.
It must be emphasized that the primary aim of our approach is to integrate into the aortic wall a nitinol-polyester network that restores (or preserves) aortic physiological strength,
whose impairment has been (or will be) the cause of the aneurysm. This is very different
from providing a new endovascular conduit that recanalizes the blood flow away from
the vascular wall, as does any current open or endovascular prosthesis. The “net”
prosthesis allows early aortic wall stabilization, ideally without any interference
with the perfusion of any collateral branch.
The device configuration may differ in terms of the amplitude of “net” meshes, the
number of layers, and the diameter of the nitinol wire whose double-crossing spiral
loops define the prosthesis tube shape ([Figs. 2] and [3A]).
It must be emphasized that the nitinol network that defines the tubular shape and
keeps the polyester “net” conduit in contact with the intima aortic surface, as well
as their respective different diameters, must be personalized based on the three-dimensional model taken from the patient computed tomography (CT) scan data.
The choice of a single- or double-layer device relies on the accuracy of the CT-based
personalization ([Fig. 3B, C]). A double layer would be appropriate only when there is a doubt about the complete
adhesion of the “net” to the intima aortic surface throughout the entire aneurysm.
Comments and Perspectives
Comments and Perspectives
Our method and device rely on integration into the aortic wall, with the mesh ultimately
sequestered from the bloodstream inside the intimal layer and in contact with the
media ([Fig. 1]). The polyester-nitinol framework compensates for the primary aortic wall structural
defect,[10]
[11] without interferences with collateral branch circulation. When clinically proven
after refinements in technical details, the endovascular “net” prosthesis will induce
radical changes in indications and timing of treatment. In fact, our method corrects
the aortic wall structural defect by reestablishing a new structural framework, a
mechanical concept already currently in use in other clinical conditions (e.g., hernia
repairs).
It must be very clear that the aim of this method is not to offer an alternative therapy
to all the aneurysms at the stage where the treatment according to current guidelines
is required. On the contrary, in the selected cases (e.g., genetically triggered diseases),
the application of this novel therapy may be warranted.
Earlier treatment for endovascular stabilization may be indicated as well when other
family members have manifested complications of aortic disease. The “net” treatment
could be fully endovascular or part of an open procedure, for example, for aortic
root stabilization at the time of aortic valve replacement ([Fig. 5]). Also, this new endovascular variety of precision personalized surgery could be applied as an adjunct to PEARS,[4]
[14]
[16]
[17] which is necessarily external to not involve open-heart surgery. The endovascular
“net” prosthesis could be positioned up to the arch from femoral access before, during,
or after PEARS ([Fig. 6], left section), thus providing early bimodal aorta stabilization.
Fig. 5 (A) Although been expressly devised as an endovascular procedure this new method may
be applied also in virtually any aortic open approach to complete at once the aortic
strengthening of the entire extension of the pathology. The prosthesis shown in this
very first prototype consists of two layers (A-mesh outside, B-mesh inside) both along
connected to the nitinol wireframe. This “net” prosthesis in a hypothesized open thoracic
configuration for the descending aorta shows its great adaptability since it preserves
the regular lumen amplitude and shape either in variable length (1), in extreme bending
(2), as well as in torsion (3); moreover, its tip can be controlled to enhance its
progression into the aorta (4). A*: When personalization cannot be performed or whenever it may appear appropriate one
or two, quite oversized (B) (small mashes) layers can be externally added, connected only at the prosthesis
ends and free to float in between to enhance contact with the aneurysmal wall.
Fig. 6 Left: In Marfan and other genetic disorders the full aortic stabilization via transfemoral
access can ideally complete PEARS, necessarily open for aortic valve stabilization
but without open heart surgery. (A, B) The two “net” prostheses prototypes hypothesized to stabilize descending aorta along
the aortic arch open prosthetic substitution. After having positioned the “net” the
connection of the proximal end of the descending aorta and with the aortic arch can
be done by expandable device type III,[23 ](top) or by expandable device[24]
[25] type I; the latter does not require the full section of the aortic stump and the
entire procedure can be completed really very, very quickly (video, https://youtu.be/ZEwzqQevgXw, 16th World Congress WSCTS, Ottawa 2006). The anastomosis with the supra-aortic trunks
here is represented as performed with expandable devices type I.
This new endovascular method ideally provides the correction of the structural defect[11]
[12] within the aortic wall with virtually no impact on perfusion of aortic collateral
branches.
This can then be viewed as the true therapy of the pathology causing the aneurysm—therapy
that can be applied virtually without risk at a very early stage. This would also
address those not so rare cases of aneurysms experiencing rupture or dissection before
their diameter meets the current criteria for aortic substitution.[18]
[19]
[20]
[21]
Endovascular devices implementing this structural support principle, tailored to the
vascular tract shape and dynamics, can be hypothesized anywhere in the aorta and,
perhaps, eventually also in any other arterial districts.
The essential requisite for the “net” embodiment into the vascular wall is its stable contact with the intimal wall for a period of ⅖ months. This requires accurate configuration
of the “net” prosthesis personalized on patient's CT images; specifically, the diameter of the nitinol wire loops must
be predicted on the diameter as the corresponding aortic segment; the “net” conduit
that should be 15 to 25% larger.
In addition, when appropriate (because of imperfect personalization of the “net” prosthesis
or particularly unfavorable anatomical configuration), further compensation may be
provided by a second 15 to 25% larger, external, polyester layer without nitinol wireframe
support, simply sutured together only at both ends of the conduit ([Figs. 3] and [5A]*) and allowed to float freely to enhance contact with any part of the intimal surface
of the aneurysm.
The “net” device should be implicitly immune to endoleak issues of conventional stents.
Also, any nitinol wire break in the “net” device should be of little or no consequence.
Although it may appear difficult to correctly keep the “net” device in position (left
[Fig. 2]), its geometrical configuration itself implicitly prevents its dislocation even
independently from its full embodiment into the aortic wall.
Prevention of age-related aortic elongation (which is significantly consistently observed
after thoracic endovascular aortic repair[22]) is implicit in the working principle of the “net” prosthesis on its target, that
is, the stabilization of the aortic wall by strengthening its structure at its media
layer. The nitinol/polyester network could act in containing the diameter as well
as the length of the aorta where embodied.
A possibly relevant point to be settled by clinical experience concerns the potential
effects of our meshes crossing the collateral branches origin. However, both mesh
configurations, even together ([Fig. 2], lower section A + A, B + B, A + B), probably would have little impact on blood
flow that could produce turbulence and, perhaps, enhance thrombus formation and microembolization.
In our system, it would be simple, if there is a requirement of “stretching” the “net”
components in the vicinity of an important branch by endovascular dilation of the
mesh elements of that site.
Use of the “net” prosthesis devices is not appropriate in cases of bleeding from full-thickness
rupture in any aortic tract. However, in aortic dissection, particularly Type B, the
“net” approach could be, perhaps, a safer option than current surgery or endovascular
procedures, as the “net” technique can fully preserve spinal cord perfusion.
A very important additional feature is that the “net” technique can be combined with
virtually any aortic open approach, through the operative field ([Figs. 5] and [6A, B]) or by endovascular means ([Fig. 6], left), to provide full aortic strengthening to the pathologic zone(s).
One of the first hypothesized examples could be the open arch-ascending aorta prosthetic
replacement where the descending aorta can be fully stabilized with devices as shown
in [Figs. 5] and [6] ([Video 1]).
Video 1 Device of [Fig. 5].
“Net” refinements could include markers allowing to verify appropriate “net” integration
into the intimal aortic layer and, perhaps, deeper into the media wall during the
postoperative course.
Moreover, It is quite possible that our longitudinally compliant “net” construct may
lengthen to accommodate the growth of young subjects. Growing swine is an ideal experimental
model to quickly verify if that can occur in a predictable, controlled process that
could hopefully open new perspectives in the earlier management of genetic aortic
aneurysms.
