From the Department of Medical Biophysics, University of Western Ontario,
London, Ontario, Canada (H.M.F., P.B.C.), and the Heart Institute, Good
Samaritan Hospital and Department of Medicine, University of Southern
California, Los Angeles (P.W.).
Methods—Sixteen segments of arteries from the circle of Willis,
including bifurcations, were pressure distended, fixed, and sectioned
in 1 of 3 orthogonal planes. We measured the 3-dimensional organization
of collagen at the flow divider by using the polarized light
microscope. An electron microscopy study performed in tandem provided
measurements on the collagen fibril diameters and packing density.
Results—Orientation data of the collagen fabric were obtained
from sections from 3 different cutting planes. The tunica media of all
bifurcations had an alignment that was primarily circumferential, and
the medial gap (medial defect) was distinguishable at the apex of all
bifurcations. The subendothelial layer was thin at the
apex but thicker and more disorganized distally. Adventitial collagen
showed little organization except for a high degree of alignment along
the apex. Results from the electron microscopy study showed densely
packed collagen fibrils of uniform diameter at the apex, compared with
slightly smaller and less densely packed fibrils nearby.
Conclusions—In the region of the medial gap, a narrow band of
highly aligned tendonlike collagen running in the direction of the
ridge of the flow divider was a consistent finding. This
structure would provide strength and stability to the vessel and is
inconsistent with the concept of an inherent defect in the
structure of bifurcations.
Brain arteries are muscular arteries, with the characteristic dominance
of circumferentially organized smooth muscle cells in the tunica media
and a distinct internal elastic lamina.9 10 The
adventitia is a layered collagen fabric with a strong dominance of
circumferential fibers adjacent to the media and a wide range of
orientations in the outer layers.10 This
characteristic structure of cylindrical segments of the artery
continues into the junction region, with transitions in the tissue
structure above and below the plane of the bifurcation, where the trunk
vessel widens.4 At the curving ridge of the flow
divider, the normal layered fabric of the wall becomes significantly
altered. A well-established feature of the cerebral artery bifurcation
is a region at the apex in which the media is absent. This is commonly
known as the "medial defect,"3 11 terminology
that may be misleading, because this region is almost always
present in cerebral artery bifurcations and therefore is unlikely
to be an acquired defect. Other terms used have been "medial
gap"9 or "raphe."12
In our study of fenestrations (regions of incomplete fusion of arteries
at the embryological stage resulting in a length of duplication of the
artery), we investigated the structure at the regions of divergence and
convergence.13 We demonstrated how the influence
of flow factors is revealed by the contrast of the built-up, layered
subendothelium at the distal end of the fenestration
compared with the minimal structure at the leading edge, or flow
divider.
In the course of this work, we have demonstrated that there is a
collagen band, or sling, that runs through the bifurcation region in
the direction of the flow divider. This collagen reveals itself under
polarized light as strikingly aligned, like tendon, with densely
packed, uniform fibrils that may serve as a tendon of the bifurcation.
The position of this band of collagen coincides with the medial gap,
the region devoid of media.
The 3-dimensional alignment of collagen fibers was measured by using a
Zeiss polarized light microscope with a rotating universal stage
attachment.15 16 The universal stage is mounted
on the main rotating stage of the microscope and permits the tissue
slide to be tipped and rotated in oblique planes, with all rotational
displacements calibrated. Thus, the 2 angles that define the
orientation of a fiber of collagen in 3-dimensional space, ie, the
azimuthal angle (in the plane of the stage) and the elevation angle
(out of the plane of the stage), are recorded for each fiber
measured within the tissue section. Key components of the universal
stage are 2 glass hemispheres with an index of refraction of 1.55,
which are located above and below the glass slide, so that the incident
light is always normal to the glass. Interfaces between the slide and
these hemispheres are coated with glycerol, with an index of refraction
of 1.47, that serves to minimize any reflection and refraction. The
fiber or group of fibers chosen for measurement is
Methods of analysis of the 3-dimensional orientation data are
well established.10 13 For graphical
presentation, the data were plotted on Lambert equal-area
projections. These plots are similar to polar projection maps
used to display entire hemispheres of the earth. Each data point on the
Lambert projection represents the intersection, at the
surface of the hemisphere, of the direction of a line originating at
the center. Each region of concentration of data points on the
projection shows organization of the collagen fibers, with the
amount of coalignment indicated by the group concentration of the
points. The projection may be rotated about any axis, so that data
taken from the bifurcations from any of the 3 sectioning planes can be
compared. We analyzed the 3-dimensional orientation data by
using Fisher spherical statistics17 18 to provide
values of the mean orientation of all sets of fibers and of their
circular standard deviation (CSD). The CSD is a measure of dispersion
of 3-dimensional alignments, analogous to the standard deviation in a
gaussian distribution, and is defined as the solid angle about the mean
orientation that encloses 63% of the data.
A complementary assessment was made of the apical collagen fibers by
using transmission electron microscopy. For these measurements standard
procedures were followed.19 Three middle cerebral
artery bifurcations from 2 autopsies (additional to those used in the
orientation studies) were pressure fixed at
physiological pressure with Karnofsky's perfusion
fluid. The vessels were trimmed so that only the apex and some tissue
from the daughter vessels remained. Samples were stained in 1%
OsO4 in a 0.1 mol/L s-collidine buffer, pH 7.4,
for 1 hour at 4°C; immersed in 2% aqueous uranyl acetate for 1 hour;
dehydrated in alcohol; and embedded in Epon 812. Thin sections were cut
on a Sorval MT 1 ultramicrotome equipped with glass knives. The
sections were mounted on unfilmed grids, stained with lead citrate, and
examined with an AEI 801 or JEOL 100CX transmission electron microscope
at 80 kV. The sections used were taken midplane longitudinally through
the apex, with the intent that the collagen fibers would be cut in
cross section.
Fibril diameters were measured from high-contrast photomicrographs
(magnification x60 000) of 2 regions of the bifurcations—the apical
ridge, which was also the region of the medial gap, and the nearby
tunica adventitia at a distance of
Cross Sections
Longitudinal Perpendicular
Tunica Adventitia
The mean CSD of the 8 vessels was 7.8° for the adventitial
layer, with the mean azimuthal orientation tracking the planar
curvature of the section, and the mean elevation angle being within
3° of the plane of the section for 5 of the bifurcations. The other 3
had mean elevation angles of 8°, 20°, and 28° out of the section
plane. We interpreted these sections as having been cut at an angle
that was oblique to the plane of the flow divider.
Tunica Media
The same statistical analysis was applied to the data
from the tunica media, and in this case, 6 of the vessels had a mean
CSD of 10.5°, whereas the other 2 had considerably greater CSD values
(29.3° and 38.4°). These latter 2 were the vessels containing the
very large area in which the media was absent, so that the regions of
measurement were at a distance from the apex.
Subendothelium
The mean CSD for all 8 vessels was 35.8°, indicating a
large scatter in the fiber orientations, with a weak central tendency.
Results from this layer were very inconsistent in both the CSD
values and the mean orientations. In some vessels, the orientations of
the subendothelial fibers separated into 2 main
directions, having 1 group of fibers aligned along the direction of the
apical ridge and the other concordant with the main alignment
perpendicular to the ridge, ie, in the direction circumferential to the
daughter vessel.4
Longitudinal Planar
Electron Microscopy
The second part of the transmission electron microscopy study was the
comparison of fibril density, or volume fraction, in the 2 regions. The
total number of test points was 1368 at the apex and 3287 in the
adjacent adventitia. Fibrils of the apical region occupied a volume
fraction of 62%, whereas fibrils in the nonapical region occupied
46%. In the adventitial region, the range of measurements (variability
of the packing density) was
The bifurcation region bears the same transmural pressure as does a
straight length of artery, yet there appears to be little structural
accommodation of the added loading in the wall. In the straight
lengths, the principal stress is circumferential (hoop) stress,
The microscopic studies of this research provide strong evidence
for an apical "tendon" in 4 ways: (1) The transmission electron
microscopy sections, though unsuited for quantitative orientation
studies, emphasize the uniformity of size of the cross-sectioned
fibrils and their high volume fraction. The measured fibril sizes are
similar in mean value and variation to those of Merrilees et
al,26 who reported fibril diameters of 50 and 53
nm in the inner and outer adventitia, respectively, of human
coronary artery. (2) We viewed the bifurcation in the
cross-sectional plane, which shows the daughter vessels joined at the
apical ridge (or flow divider). Polarized light microscopy revealed the
collagen to be coherent and parallel to the line of contact between the
daughter vessels. In this plane, any unexpected transverse fibers would
stand out microscopically; however, none were observed. (3) Collagen,
when viewed in longitudinal perpendicular sections, had straightened
fibers tracking along the apical ridge. (4) When viewed in longitudinal
sections parallel to the plane of the bifurcation (the conventional
view for bifurcations), the coherent alignment of the thin band of
collagen was striking. With the correct tilt on the universal stage,
the collagen is aligned vertically and can then be visualized as a
small black region within the adjacent brightly birefringent
adventitia, which remains dark as the universal stage is rotated. This
confirms that all of the fibers of this region are highly coaligned in
the direction parallel to the optic axis of the microscope. A summary
of the CSD values from the different cutting planes (the Table
Our results, in combination with the findings of other studies, provide
a fuller understanding of wall mechanics in the region of brain artery
bifurcations. We observed collagen fiber structure in the medial gap to
be similar to that of tendon and quite unlike that of the rest of the
adventitia. Because of this band of collagen, at any given pressure the
medial gap will be stiffer than the rest of the adventitia. Macfarlane
et al23 27 reported that the apex of the
bifurcation was able to maintain its general shape even at negative
pressures and that the radius of curvature at the apex changed only
moderately as the pressure increased over a wide range. Evidence for a
strong band of collagen in the bifurcation apex, rather than indicating
a region of weakness, points to its being a region of high strength;
however, the immediately adjacent vessel wall may be left vulnerable to
forces of hemodynamics and blood pressure. The
formation of aneurysms is therefore most likely to be initiated
at the discontinuity at the edge of the band, where distension of the
vessel could occur. Forbus1 reported on the
mechanical failure of middle cerebral artery bifurcations exposed to
excessive pressure and found that in no case did rupture occur at the
apex but was always nearby in the general region of the bifurcation.
Stehbens28 had noted that the site of formation
of aneurysms in human cerebral arteries was often in the
daughter vessel, just distal to the medial gap. More recent studies by
Hazama and colleagues29 30 have a complementary
theme. Their work on experimentally induced cerebral aneurysms
in animals showed that the vessel wall developed grooves directly
adjacent to the apex, some of which later developed into saccular
aneurysms.
Our quantitative morphological investigation provides new information
about the bifurcation region of brain arteries that, in the vast
majority of cases, are mechanically stable. The mechanical
vulnerability of this region to the formation of aneurysms has
still to be explained.
Received January 28, 1998;
revision received April 8, 1998;
accepted May 20, 1998.
2.
Hassler O. Morphological studies on the large cerebral
arteries: with reference to the aetiology of subarachnoid
haemorrhage. Acta Psychiatr Neurol Scand.
1961;154(suppl):1–145.
3.
Glynn LE. Medial defects in the circle of Willis and
their relation to aneurysm formation. J Pathol
Bacteriol. 1940;51:212–222.
4.
Walmsley JG, Campling MR, Chertkow HM.
Interrelationships among wall structure, smooth muscle orientation, and
contraction in human major cerebral arteries. Stroke. 1983;14:781–790.
5.
Ferguson GG. Physical factors in the initiation,
growth and rupture of human intracranial saccular aneurysms.
J Neurosurg. 1972;37:666–677.[Medline]
[Order article via Infotrieve]
6.
Steiger HJ. Pathophysiology of development and
rupture of cerebral aneurysms. Acta Neurochir
(Wien). 1990;(suppl 48):1–57.
7.
Sekhar LN, Heros RC. Origin, growth and rupture of
saccular aneurysms: a review. Neurosurgery. 1981;8:248–260.[Medline]
[Order article via Infotrieve]
8.
Canham PB, Ferguson GG. A mathematical model for the
mechanics of saccular aneurysms. Neurosurgery. 1985;17:291–295.[Medline]
[Order article via Infotrieve]
9.
Stehbens WE. Pathology of Cerebral Blood
Vessels. St Louis, Mo: CV Mosby Co; 1972.
10.
Finlay HM, McCullough L, Canham PB. Three-dimensional
collagen organization of human brain arteries at different transmural
pressures. J Vasc Res. 1995;32:301–312.[Medline]
[Order article via Infotrieve]
11.
Stehbens WE. Medial defects of the cerebral arteries of
man. J Pathol Bacteriol. 1959;78:179–185.[Medline]
[Order article via Infotrieve]
12.
Stehbens WE. Pathology and pathogenesis of intracranial
berry aneurysms. Neurol Res. 1990;12:29–34.[Medline]
[Order article via Infotrieve]
13.
Finlay HM, Canham PB. The layered fabric of cerebral
artery fenestrations. Stroke. 1994;25:1799–1806.[Abstract]
14.
Rinne J, Hernesniemi J, Niskanen M, Vapalahti M.
Analysis of 561 patients with 690 middle cerebral artery
aneurysms: anatomic and clinical features as correlated to
management outcome. Neurosurgery. 1996;38:2–11.[Medline]
[Order article via Infotrieve]
15.
Canham PB, Finlay HM, Dixon JG, Ferguson SE. Layered
collagen fabric of cerebral aneurysms quantitatively assessed
by the universal stage and polarized light microscopy. Anat
Rec. 1991;231:579–592.[Medline]
[Order article via Infotrieve]
16.
Canham PB, Finlay HM, Tong SY. Stereological
analysis of the layered collagen of human intracranial
aneurysms. J Microsc. 1996;183:170–180.[Medline]
[Order article via Infotrieve]
17.
Tarling DH. Statistical analyses. In:
Principles and Applications of Paleomagnetism.
London, England: Chapman and Hall; 1971:72–85.
18.
Fisher NI, Lewis T, Embleton BJJ. Statistical
Analysis of Spherical Data. New York, NY: Cambridge
University Press; 1987:101–151.
19.
Ohara PT, Buck RC. Contact guidance in vitro: a light,
transmission and scanning electron microscopic study. Exp Cell
Res. 1979;121:235–249.[Medline]
[Order article via Infotrieve]
20.
Wiebel ER, Kistler GS, Scherle WF. Practical
stereological methods for morphometric cytology. J Cell
Biol. 1966;30:23–38.
21.
Finlay HM, Dixon JG, Canham PB. Fabric organization of
the subendothelium of the human brain artery by
polarized-light microscopy. Arterioscler Thromb. 1991;11:681–690.
22.
Canham PB, Talman EA, Finlay HM, Dixon JG. Medial
collagen organization in human arteries of the heart and brain by
polarized light microscopy. Connect Tissue Res. 1991;26:121–134.[Medline]
[Order article via Infotrieve]
23.
Macfarlane TWR, Canham PB, Roach MR. Shape changes at
the apex of isolated human cerebral bifurcations with changes in
transmural pressure. Stroke. 1983;14:70–76.
24.
Fung YC. Biomechanics: Mechanical Properties of
Living Tissues. 2nd ed. New York, NY: Springer-Verlag;
1993:242–320.
25.
Sherebrin MH, Song SH, Roach MR. Mechanical anisotropy
of purified elastin from the thoracic aorta of dog and sheep. Can
J Physiol Pharmacol. 1983;61:539–545.[Medline]
[Order article via Infotrieve]
26.
Merrilees NJ, Tiang KM, Scott L. Changes in collagen
fibril diameters across artery walls including a correlation with
glycosaminoglycan content. Connect Tissue
Res. 1987;16:237–257.[Medline]
[Order article via Infotrieve]
27.
Macfarlane TWR, Roach MR, Chan K. The geometry of
human cerebral bifurcations: effect of static distending pressure.
J Biomech. 1980;13:265–277.[Medline]
[Order article via Infotrieve]
28.
Stehbens WE. Histopathology of cerebral
aneurysms. Arch Neurol. 1963;8:272–285.
29.
Kojima M, Handa H, Hashimoto N, Kim C, Hazama F. Early
changes of experimentally induced cerebral aneurysms in rats:
scanning electron microscope study. Stroke. 1986;17:835–841.
30.
Kim C, Cervs-Navarro J, Kikuchi H, Hashimoto N,
Hazama F. Alterations in cerebral vessels in experimental animals and
their possible relationship to the development of aneurysms.
Surg Neurol. 1992;38:331–337.[Medline]
[Order article via Infotrieve]
© 1998 American Heart Association, Inc.
Original Contributions
Collagen Organization in the Branching Region of Human Brain Arteries
Abstract
Top
Abstract
Introduction
Methods
Results
Discussion
References
Background and Purpose—Unruptured
saccular aneurysms are relatively common, occurring in 4% to
9% of autopsies. Their development at the apex region of brain artery
bifurcations is attributed to a combination of structural factors and
the effect of blood pressure. Collagen is a primary tension-bearing
fabric of the vessel wall, and our purpose was to examine its
3-dimensional alignment at arterial branches.
Key Words: biophysics • cerebral aneurysms • cerebral arteries • collagen
Introduction
Top
Abstract
Introduction
Methods
Results
Discussion
References
The branching region
of brain arteries has long been known to be the site of formation of
saccular aneurysms, balloonlike structures capable of gradual
enlargement and catastrophic rupture.1 2 3 This
region is mechanically and structurally complex, with curvatures of the
vascular wall that must bear the tensions arising from the distending
forces of blood pressure.4 5 We undertook mapping
of the collagen organization as a 3-dimensional, layered fabric that
provides the structural framework for the arterial
bifurcations. By focusing on collagen, we are acknowledging its role as
the strong component of the wall that may be a factor in the
development of aneurysmal lesions if it yields to blood
pressure forces.6 7 8
Methods
Top
Abstract
Introduction
Methods
Results
Discussion
References
For the collagen orientation measurements, 16 branching regions
of major brain arteries from the circle of Willis were obtained from 13
autopsies (7 male and 6 female) between the ages of 42 and 77 years.
Our intent was to obtain samples of bifurcations from those vessels
known to be sites of aneurysm formation. Causes of death were
post heart transplant in 1, post lung transplant in 1, myocardial
infarction in 3, dementia in 3, pneumonia in 1, and liver cirrhosis in
2. For 2 of the 13 cases, we were not able to ascertain the cause of
death. We saw no evidence of atherosclerosis on gross
examination of the cerebral vessels that we used. Segments of artery,
including the bifurcation, were cannulated, pressure distended, and
fixed in 10% neutral buffered formalin at a
physiological pressure of either 110 or 120
mm Hg. Twelve of the vessels used were bifurcations of the middle
cerebral artery, vessels most frequently associated with
aneurysms,14 and 4 were bifurcations of
the posterior cerebral artery. The branching regions were embedded in
paraffin and sectioned at either a 5- or 7-µm thickness. Three
different sectioning planes were used, so that the orientation of the
fibers within the bifurcations could be explored thoroughly (Figure 1
). We sectioned 4 bifurcations in the
plane of the cross section of the main proximal vessel (Figure 2a); 8 in the longitudinal direction
perpendicular to the plane of the bifurcation, which would include the
ridge of the apical divider in profile (Figure 2b); and 4 in the
longitudinal transverse plane of the bifurcation (Figure 2c). Collagen
has natural birefringence that can be enhanced by the use of
appropriate stains. For most of the studies, we used picrosirius red, a
strongly enhancing stain for collagen. Gomori's silver impregnation,
Masson's trichrome, and Verhoeff's elastin stains were used to
highlight structures other than collagen.
View larger version (24K):
[in a new window]
Figure 1. Schematic of bifurcation region of cerebral artery
indicating 3 orthogonal sectioning planes. Upper diagram shows
characteristic cross section immediately distal to the apex.
View larger version (29K):
[in a new window]
Figure 2. Profile of representative sections
showing 3 main layers of the artery wall and how they appear at the
apical region of the bifurcation with the sectioning planes used. 
4x4 µm
and must be centered between the 2 glass hemispheres. 0.4 mm from the gap. From
photomicrographs of 3 bifurcations, a total of 1100 fibril diameters
were measured from each of the 2 regions. The volume fraction of the
fibrils was determined by point counting20 by
using a transparent overlay with a 4-mm grid. Measurements were taken
from 3 bifurcations by using 3 photomicrographs taken at the medial gap
and 7 taken a small distance away from the gap (magnification
x35 000).
Results
Top
Abstract
Introduction
Methods
Results
Discussion
References
Examination of variously stained sections from the region of the
bifurcations revealed the following. An internal elastic lamina was
present in all vessels, including the apex; this was seen most
clearly from the 2 longitudinal cutting planes. It was sometimes wavy
and in all of the sections had an appearance of having small
discontinuities. There was no external elastic lamina. The tunica media
was composed mostly of smooth muscle, with a small amount of collagen.
In most cases there was little tunica intima, but regions of thickened
intima were seen locally in 7 sections, some from all 3 cutting planes.
These are probably indications of the intimal pads frequently seen in
association with bifurcations.
When the main proximal vessel is cut in right cross section, the
initial sections are round. As a cut toward the bifurcation is made,
the sections widen into an oval and then in quick succession to a
figure 8 and to 2 round sections belonging to the daughter branches
closest to the apex (Figure 1
). Subsequent sections are at various
angles of oblique cross section for the 2 daughter vessels, depending
on their branch angle and curving path away from the junction region.
Four bifurcations were sectioned in this way at 7-µm thickness, and
we focused on those sections just proximal to just distal to the apical
region (Figure 2a). From each bifurcation, 20 measurements of the
3-dimensional alignment of collagen fibers were made in the region of
the medial gap, the apex, and 20 measurements were made from the
adventitial collagen spaced around the 2 daughter vessels. In all
cases, the mean orientation of the collagen was in the circumferential
direction around the daughter vessels and was close to the plane of the
microscope section. The CSDs of these data about the mean direction for
the apical region and the adventitia away from the apex are given in
the Table, indicating a striking
difference in coherence. The photomicrographs of Figure 3a show the flow divider just before the
edge of the apex, with the intensely bright collagen visible at both
sides as it approaches the apical ridge. Figure 3b, 40 µm
farther along the bifurcation, shows the highly aligned tendonlike
collagen of the apex, with the much less birefringent
subendothelium above and below, toward the lumen of the
daughter vessels. The values of CSDs for the media and the
subendothelium in cross section in the Table are from a
previous report.21
View this table:
[in a new window]
Table 1. Mean Dispersion (CSD) of Collagen Alignments Within Each
Region
View larger version (78K):
[in a new window]
Figure 3. Photomicrographs of cross-sectioned
bifurcation, taken with circularly polarized light. 3a and 3b show
region just before the flow divider with a brightly birefringent band
of collagen on each side. Complete band is shown at the apex in 3c and
3d, 40 µm distally.
Sections cut in this plane are able to show the true profile of
the apical flow divider (Figure 2b). The positioning for sectioning of
these bifurcations in the paraffin blocks was important for ensuring
that the section plane was along the apical ridge. This goal
established a challenge for sectioning because many of the bifurcations
have daughter vessels that are dissimilar with respect to their
diameters and the angles of divergence from the axis of the parent
artery. As a consequence, we were not equally successful at revealing
the apical ridge for all bifurcations. We made measurements from 8
bifurcations in this plane and found considerable variability in the
appearance of the medial gap. Two of the vessels had a region 1
mm long in which the media was absent, and in 3 cases the gap was
extremely short, <100 µm, giving the appearance of an intrusion
of the adventitia toward the subendothelium. These
features may have been due to misalignment of the section plane. In all
of the bifurcations there was a segment of the adventitia at the apex,
coinciding with the region of the medial gap, in which the collagen was
seen in the plane of the section to be straight and highly coaligned.
Using the sections in this plane, we made measurements from the
adventitia, tunica media, and subendothelium. For each
of these layers, 5 readings were taken from each of 10 separate regions
that were evenly spaced around the ridge of the flow divider. Spherical
statistics were used to analyze the combined 50 orientation
measurements of each layer, and the results are summarized in the
Table.
Our third sectioning plane was the longitudinal planar (Figure 2c), which is the one most commonly used to reveal branching vessels.
The main artery, branches, and bifurcation are seen in profile. We took
readings from 4 of these bifurcation sections, all of which had a
region of medial gap at the apex. The sections selected for
measurements were those in which the gap was the most pronounced, to be
as close as possible to the midregion of the divider. Twenty
orientation measurements were made from each of the 3 main layers
adjacent to the region of the apex. Readings were made from both sides
of the apex in the subendothelium, which in 3 of the
vessels was very thin, and from both sides of the medial gap in the
media. The readings from the adventitia were made from the region close
to the apex. In addition to these measurements, we detected a small
region of highly aligned collagen that stood out at the apical edge of
the adventitia, ie, in the medial gap. This region is 10 to 25
µm across and consists of collagen oriented perpendicular to the
section plane. A set of 20 measurements was made from each of these 4
regions (subendothelium, media, adventitia, and apical
ridge). Analysis by spherical statistics from 4 bifurcations
gave mean CSDs as shown in the Table. The CSD value of 4.5° at the
apex in this plane is lower than the values of 7.3° and 7.8°
obtained from the other 2 planes, for which measurements were made over
a greater length of the apical collagen band. With polarized light, the
very highly aligned, birefringent collagen of the apex region, if
oriented perpendicular to the microscope axis, is shown as an area that
remains dark when the stage is rotated through 360°. If the alignment
is off by as little as a few degrees from the perpendicular, then the
region appears bright when seen with polarized light. In 1 of the
sections examined, the dark region of perpendicular collagen was
visible with the microscope stage in its horizontal position, and in
the other 3, the region could be identified only when the universal
stage was tilted. Lambert projections of the data from 1 of the
planar sections are shown in Figure 4.
The data are presented so that fibers oriented perpendicular to
the microscope section appear at the center of the plot. The strong
coalignment of the apex collagen is shown by the tight grouping of
fiber orientations (CSD=3.8°), whereas in the adventitia and the
media, there is a greater dispersion of fiber orientations (CSD=37.7°
and 7.8°, respectively). In the subendothelium, 2
main directions were often observed.
View larger version (14K):
[in a new window]
Figure 4. Lambert equal-area projections illustrating
representative fiber orientations and distributions of
measurements from 1 midplane longitudinal tissue section of a middle
cerebral artery bifurcation. Projections provide a graphical,
quantitative view of average alignment and dispersion. Note highly
coherent alignment of the set of fibers from the small region at the
apex (CSD of 3.7°) compared with widely dispersed data from the
adventitia (CSD of 37.8°). Media also has relatively high alignment
of data (CSD of 7.8°), and data of the subendothelial
layer show characteristic separation into 2 main directions (combined
CSD of 34.1°).
The electron microscopy study was conducted in 2 parts. The first
was the measurement of the diameter of collagen fibrils at the apex of
the bifurcation and at a distance (0.4 mm) from the apex.
Although small, this distance was far enough from the apex so that
there was a layer of tunica media underlying the adventitia, the
dimension of the apical collagen band being 50 µm. The
cutting plane was longitudinal planar, so that at the apex the
direction of the fibrils would be perpendicular to the section plane.
The mean diameter of the fibrils of the apex from a total of 1100
measurements from 3 bifurcations was 49.9±7.2 nm and of the adjacent
adventitia was 52.3±8.4 nm, a difference that was not significant. 3 times greater than in the apex. The
transmission electron photomicrograph of the apical region in Figure 5 shows aligned fibrils that are very
densely packed. In the photomicrograph of the near-apex adventitia,
variations in the fibril diameters and orientations as well as in the
packing density are displayed. It was noted that the entire area shown
in Figure 5 is smaller than the zone of measurement for a single
orientation reading from the universal stage.
View larger version (164K):
[in a new window]
Figure 5. Transmission electron photomicrographs of collagen
fibrils of apex and near-apex adventitia (x40 000). Fibrils in apex
region are of uniform size and coaligned, compared with those of the
adventitia 0.4 mm from the apex.
Discussion
Top
Abstract
Introduction
Methods
Results
Discussion
References
The striking feature of collagen in the medial gap is its highly
aligned organization, much like that of tendon under load. Fibers and
fibrils run parallel to each other and, at the apex, track through the
perpendicular to the long axis of the parent artery. Polarized light
microscopy is possibly the best microscopic technique for studying
apical collagen. The method is ideally suited for detection of the
region of coaligned fibers and also for making comparative measurements
on their directional organization. A complementary study uses the high
resolution of transmission electron microscopy to compare the fibril
diameters and packing density of the apical and near-apical
collagen. 
C, and the key structural layers, the media
and adventitia, have mainly circumferentially directed collagen and
smooth muscle fibers.10 22 The lesser presence of
longitudinal fibers in the outer adventitia is able to bear the lower
longitudinal stress,
L=C/2. The
challenge of providing a stable bifurcation is illustrated in Figure 6. Laplace's law for thin, convex
vessels with principal radii of curvature
R1 and R2
provides the relationships among wall tension T, transmural
pressure P, and curvatures R1
and R2:
P=T(1/R1+1/R2).23
For a right cylinder, 1/R2=0 and
P=T/R, which relates to
circumferential stress by the equation
C=T/t, t
being wall thickness. Not only is R2
negative at the apical region of a bifurcation but also for brain
arteries, R2 is smaller in magnitude than
R1, thus making the Laplace relationship
untenable. The modified Laplace's law for unequal tensions
P=T1/R1+T2/R2
brings out an important factor for brain artery branch regions. Because
R2 is negative,
T1 must therefore be substantially higher
than T2, possibly by a factor of 5 or 10
times that for the branches sectioned in the plane of the bifurcation.
This argument of basic mechanics illustrates 2 points: (1) there is a
requirement for extra strength at the apex, a need that is satisfied by
a tendonlike structure of collagen running along the apical ridge and
(2) the high tension of this structure would straighten any natural
waviness of unloaded fibers. (The other load-bearing fabric, elastin,
is only a minor contributor to wall mechanics in the region of
bifurcations, mainly because in comparison to collagen, its stiffness
is less by a factor of 1000 and its tensile strength less by a
factor of 100.24 25 )
View larger version (14K):
[in a new window]
Figure 6. Representation of tension forces
T1 and T2 acting
on the flow divider. ) shows
how much more organized the apical collagen is than that of the other
regions.
Acknowledgments
This work was supported by the Heart and Stroke Foundation
of Ontario. The authors would like to especially acknowledge Dr Robert
Buck for his contributions concerning electron microscopy. We also
acknowledge Jan Dixon for her histological assistance
and John Lloyd who helped as a summer research student. We would like
to thank the Pathology Departments of the London Health Science Center
and St Joseph's Hospital, London, for their help in obtaining
tissue.
Footnotes
Reprint requests to H.M. Finlay, Department of Medical Biophysics, University of Western Ontario, London, Ontario, N6A 5C1 Canada.
References
Top
Abstract
Introduction
Methods
Results
Discussion
References
1.
Forbus WD. On the origin of miliary
aneurysms of the superficial cerebral arteries. Bull
Johns Hopkins Hosp. 1930;47:239–284.
This article has been cited by other articles:
 |
|
 |
|
  T. C. Gasser, R. W Ogden, and G. A Holzapfel Hyperelastic modelling of arterial layers with distributed collagen fibre orientations J R Soc Interface, February 22, 2006; 3(6): 15 - 35. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Canham and H. M. Finlay Morphometry of Medial Gaps of Human Brain Artery Branches Stroke, May 1, 2004; 35(5): 1153 - 1157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. van Gijn and G. J. E. Rinkel Subarachnoid haemorrhage: diagnosis, causes and management Brain, February 1, 2001; 124(2): 249 - 278. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||