вход по аккаунту


Observation on the microcirculatory architecture of the rat liver.

код для вставкиСкачать
Observation on the Microcirculatory Architecture
of the R a t Liver
Department of Gastroenterology, Division of Medicine, W a l t e r Reed Army
Institute o f Research, W a l t e r Reed A r m y Medical Center,
W a s h i n g t o n , D. C .
Microcirculation of rat livers was studied on preparations perfused
with silicone rubber. Silicone rubber provided a n excellent perfusicn medium for the
study of the special arrangement of hepatic microcircuIation.
As demonstrated by Gershbein and Elias ('54), a great portion of the hepatic
venous tree of the rat liver receives sinusoidal channels and thus this portion is located centrilobularly.
In the portal venous system of the rat liver, not only the distributing veins but
also the conducting veins give rise to inlet venules regularly to adjacent peripheral
Rich capillary networks of the periductal plexus which receive their afferent
channels from the hepatic arteries and empty their blood either into the portal veins
or into the adjacent peripheral sinusoids are demonstrated in the portal canals. Less
prominent fine capillaries are distributed in the walls of portal veins and i n the connective tissue components of larger portal canals. Existence of a more direct type of
anastomoses between the hepatic arteries and portal veins through capillary networks
i n the medium and large sized portal canals is also indicated. Evidence for the existence of intralobular arteriolar terminals were not obtained and the present observation indicated that the hepatic arteries terminate at the periphery of the lobule.
The rat has been undoubtedly the most
commonly used experimental animal for
the study of liver diseases. In this respect,
a clear understanding of rat liver morphology is valuable and often essential for
the proper interpretation and evaluation
of experimental results obtained with this
animal. Existence of some anatomical differences between human and rat livers has
been pointed out (Gershbein and Elias,
'54; Elias and Popper, '55); and their significance in pathogenesis of certain experimental liver diseases in this animal was
suggested (Elias and Popper, '55).
Recently a silicone rubber perfusion
material ' has been used to accomplish excellent filling of microcirculatory beds with
the subsequent preservation of microvascular architecture (Sobin et al., '62, '63;
Demis and Brim, '65). We adapted this
technique in order to re-examine the microstructure of normal rat liver mainly from
the standpoint of hepatic vascular arrangement.
Male albino rats (Walter Reed Carworth
Farm Strain) weighing approximately 300
gm were used.
ANAT. REC., 156: 157-174.
The perfusion technique described by
Demis and Brim ('65) was generally followed. However, for the preparatory perfusion, physiological saline solution containing 0.01 M sodium citrate was used.
For the perfusion of the silicone rubber
via the hepatic artery, a polyethylene tube
(no. 60) was inserted into the abdominal
aorta cranially so that the tip of the tube
reached just below the opening of the celiac
artery. The aorta at the site just above the
diaphragm and the main portal vein were
double ligated and the rubber was injected
into the aorta through the polyethylene
tube. The polyethylene tube was placed in
the main portal vein for perfusion via the
portal vein. For perfusion via the hepatic
vein, the polyethylene tube was inserted
into the right atrium down the inferior
vena cava so that the tip of the tube was
located in the hepatic vein at its opening
to the inferior vena cava. The inferior
vena cava was then double ligated at the
site just caudal to the opening of the
hepatic vein, and the rubber was injected
into the hepatic vein through the tube.
1 Silicone Products Department, General Electric
Laboratories, Waterfard, New York.
The properties of the silicone rubber
were described in detail by Sobin and his
co-workers ('62, '63). For one color perfusion, silicone rubber, RTV 201, of the
RTV 200 series was used exclusively. RTV
201 has a rusty red color and excellent
quality for filling microvascular beds. For
independent perfusion of two different vascular systems, silicone rubber, RTV 11,
was used for injection of the second vascular system. However, the silicone rubber, RTV 11, was considerably inferior to
RTV 201 in filling the capillary beds. A
mixture of silicone rubber, RTV 201, and
Ferro-colors * (Blue Silicone paste, V-1232)
in a ratio of approximately 9: 1 was also
used as the perfusion material of the second vascular system. This combination
caused some decrease of the vulcanizing
potency of the mixture. After the infused
rubber vulcanized (usually 4 to 6 hours
at room temperature or overnight at 4"C),
the liver was removed, cut into slices 50
to 200 v thick with a frozen section microtome, cleared with glycerin, mounted on
microscopic slides with glycerin and studied under a microscope with reflected light,
Hepatic veins. The hepatic venous
tree with its surrounding sinusoids was
best demonstrated in the preparations in
which the silicone rubber was injected via
the hepatic vein, figures 1, 2, 3. As observed by Gershbein and Elias ('54), the
portion which receives sinusoids in the rat
liver is not limited to the smallest terminal
radicles of the hepatic venous tree but
also extends to hepatic veins of considerably larger diameters. Thus, a great portion of the hepatic venous tree is centrilobular in location (Elias and Popper, '55).
Since central veins generally denote terminal radicles of the hepatic venous tree, more
rigid definition of the term central vein is
necessary when hepatic venous system of
the rat liver is described. Accordingly, we
define the central veins of the rat liver
as the tributaries of the hepatic vein which
are centrilobular in location and receive
surrounding sinusoidal channels directly,
regardless of their sizes. In this regard,
the smallest radicles of the hepatic venous
tree which correspond to the central veins
of the human liver are called primary cen-
tral veins in comparison to central veins
of the larger sizes. Two primary central
veins either converge at acute angles and
form a central vein of larger diameter or
they enter adjacent large central veins or
hepatic veins directly, figures 1, 2, 3. We
observed that hepatic veins of up to 600 p
in diameter possessed distinct characteristics of central vein in the rat liver.
Terminal twigs of the portal veins are
frequently seen ending near large central
veins and anastomosing to them through
sinusoids, figure 2. In the surrounding
area of a large hepatic vein rather short
non-branching and branching central veins
are seen receiving sinusoidal channels and
entering the vein, figure 3 .
Hepatic sinusoids. The sinusoids of a
lobule are seen as richly ailastornotic delicate vascular channels arranged around
the central vein, figure 4. Because of the
characteristics of the hepatic venous system of the rat liver, central veins among
different lobules may show considerable
variations in cross sections. When a lobule
contains more than one central vein in a
cut surface, it usually indicates that the
veins are cut near the site where they join.
These central veins are connected to each
other by sinusoids of the hepatic tissue
which is placed among them.
The sinusoids in the area surrounding
the portal canals constitute peripheral
sinusoids which are particularly rich in
anastomotic channels showing a honeycomb pattern. The sinusoids in the mid
and central zones, radial sinusoids, are
less anastomotic and more radially arranged toward the central vein.
Portal vein. A branching pattern of
portal veins is shown in figures 5, 6.
Along their course, the portal veins give
rise to inlet venules at right angles
(Knisely et al., '48; Elias, '49) which immediately branch into richly anastomotic
peripheral sinusoids of adjacent hepatic
lobules, figures 7, 8. The terminal twigs
of the portal veins also branch into sinusoids, figure 2.
Hartroft ('52) and Elias and Popper
('55) indicated that, in the rat liver, the
areas around large portal triads receive
blood indirectly from remote smaller por2
Fern, Corporation, 4150 East 56th Street, Cleve-
land 5 , Ohio.
tal triads and that this area is, in the
functional sense, "nonportal" similar to
the central zone. Certainly we did not see
smaller distributing veins which run
alongside conducting veins in the rat liver
as being described by Elias and his COworkers in man and other species of animals (Elias, '49; Elias et al., '55). The
definitions made by Elias ('49) of conducting veins as the larger portal veins
which do not give rise to inlet venules
and of distributing veins as the smaller
portal veins which give rise to inlet venules, though convenient and useful for
the general descriptive purpose, do not
apply exclusively. As pointed out by Rappaport ('63), conducting veins also give
rise to venules to supply a periportal cuff
of tissue, thus, distributing portal blood
quite regularly to neighboring tissue. This
type of short afferent veins and venules
coming out of a conducting vein at right
angles and branching into neighboring
sinusoids is frequently seen around a
large portal canal of the rat liver, figures
7, 8. At the same time, the areas around
large portal canals receive additional
blood supply from the drainage of periductal plexuses which are particularly
prominent in these areas. Our present
study did not indicate that deficiency of
portal blood distribution exists in the
areas around large portal triads. Capillary
networks in the portal canals, periductal
plexuses, were well demonstrable in the
preparations in which the perfusion was
carried out via the portal vein, figures 10,
1 1 . Periductal plexuses are seen as elongated basket-like capillary networks, lying
along portal canals. In large portal canals,
the plexuses are formed in double layers;
the inner layer having the network of
fine capillaries and representing the submucosal plexus and the outer layer having
the network of coarser capillaries and
representing subepithelial plexus with capillary connections between the channels
of these two layers (Elias and Petty, '53).
Small branches of portal veins, the internal roots of a portal vein, connect the
plexuses and other capillaries with main
portal veins of the portal canal, figures 9,
11, (Aunap, '31; Elias and Petty, '53). In
an area where the periductal plexuses are
composed of particularly rich capillaries,
groups of venules, i.e. radicular portal
venules, connect the plexuses with adjacent peripheral sinusoids, figures 10, 11,
(Olds and Stafford, '30; Elias and Petty,
Heputic artery. In the perfusion preparations in which the portal venous route
was used, the hepatic arteries in the portal canals remained unperfused except for
perhaps those which lie near rich periductal capillary networks. On the other
hand, in the preparations in which the
hepatic arterial route was used, not only
the hepatic arterial tree but also the portal venous tree were perfused. As seen in
figures 5, 6, the hepatic arteries run and
ramify more or less in the same manner
as the portal veins. Usually one or two
hepatic arteries of much smaller calibres
run together with one portal vein. The
hepatic arteries frequently give rise to
small terminal twigs in the surrounding
areas along their courses and these terminal twigs branch into sinusoids in a
tree-like fashion.
The problem concerning the actual arterial terminations and the way they join
the portal veins still remains controversial.
Difficulties in studying endings of hepatic
arteries and arterioles by perfusion techniques arise in that the injection of a perfusion material via the hepatic artery invariably results in the appearance of a
considerable amount of the material in the
portal venous system through the channels of capillary anastomoses between the
two vascular systems in the portal canals.
This frequently obscures fine arterial and
arteriolar terminals. Injection of the rubber via the hepatic artery in our present
study also caused rather selective perfusion of the sinusoids of many central zones
leaving the peripheral sinusoids relatively
unperfused. This is probably the same
phenomena observed by Chrzonszczewsky
(1866) and later confirmed by Elias ('49).
In our preparations, no definite connections between the perfused vessels of central sinusoids and portal canals are
observed indicating that the filling of central sinusoids does not result from the vessels of the portal canals. As pointed out
by Elias and Petty ( ' 5 3 ) , the perfusion of
the central zones when injection was made
via the hepatic artery probably occurs by
the back flow of the perfusion material
from the hepatic vein through collateral
channels in the perihepatic tissues.
In our two color perfusion technique,
the liver was first perfused by silicone
rubber, RTV 201, via the hepatic artery
while animals were still alive. Then either
silicone rubber, RTV 11, or the mixture of
silicone rubber, RTV 20 1 , and blue silicone
paste was injected via the portal vein. In
these preparations, the portal venous tree
and practically all the sinusoidal channels
were filled with the material injected via
the portal vein while the hepatic arterial
tree retained silicone rubber, RTV 201.
Terminal arteries are seen branching in a
tree-like fashion. These branches end
rather abruptly and connect to the channels represented by the material perfused
through the portal vein at the periphery of
the lobules. Despite intentional search,
we failed to observe microvessels of the
hepatic arterial origin entering intralobularly as described by Elias (’49); Elias
and Petty (’53); and Tajiri (60). This
indicated to us that the hepatic arteries terminate and join the portal venous channels at the periphery of the lobules.
The hepatic arteries also provide radicular branches to the periductal plexuses in
the portal canals. In addition, less prominent arteriolar and capillary branchings independent to the periductal plexuses are
seen occasionally in large portal canals.
These capillaries are seen distributed in
the walls of large portal veins and other
tissue components of portal canals (Tajiri,
’60). However, this was an inconsistent
finding and was not clearly demonstrated
in medium and small sized portal canals.
Communications between portal veins
and hepatic arteries through capillary networks such as shown in figure 12 probably
represent a more direct type of anastomosis
between the two vascular systems. The
capillary network seen in the figure does
not show characteristics of the periductal
plexus and it is connected by short
branches to both hepatic artery and portal
vein. This type of capillary network is
seen between a hepatic artery and portal
vein in large and medium sized portal
Our observations confirmed the centrilobular location of the greater portion of
the hepatic venous tree of rat liver as observed by Gershbein and Elias ( 5 4 ) . Thus,
the concept which views a hepatic lobule
as an architectural unit does not apply to
the rat liver. Instead, the lobules of rat
liver can be pictured as long cylindrical
masses of liver tissue which branch along
the tributaries of the hepatic vein. This
explains the observation of rat liver on
histological preparations, which usually
shows poor delineation of individual hepatic lobules, variable sizes of central veins
and presence of a few central veins in a
field which appear to be one hepatic lobule. In diffuse central zone necrosis
caused by certain hepatic poisons such as
carbon tetrachloride, the same pathologic
changes are observed in the central zones
surrounding larger central veins as well as
in the central zone surrounding small
primary central veins indicating that central veins of larger sizes have the same
functional significance as the small ones.
Hepatic arteries and portal veins anastomose through capillary networks while
they run together in the portal canals.
The most prominent capillary networks
are periductal plexuses which are related
to one of the major hepatic functions,
bile secretion. Less prominent capillaries
distribute in vascular walls and connective
tissue components of larger portal canals.
In addition, capillary networks, figure 12,
which do not fit into the characteristics of
the periductal plexus were observed particularly in large and medium sized portal
canals. The functional significance of this
type of anastomoses is not clear, but it is
possible that partial arterialization of portal blood takes place through these channels.
There is difference of opinion as to the
actual arterial terminations and the way
they join the portal vein. The problem
seems to center on whether the terminal
hepatic arteries or arterioles end at the
periphery of the lobule emptying their
blood in the peripheral sinusoids or
whether they extend intralobularly in the
form of either “hepatic arterioles” or arterial sinusoids.” Wakim and Mann (’42,
’53) observed their so-called “arterial si-
nusoids” by their transillumination tech- uniform mixing of the arterial blood with
nique. However, the existence of “arterial the portal blood. Our present observations
sinusoids” as well as other observations on the microvasculature in the rat liver
made by them were questioned recently by generally agree with those in the rabbit
Nakata (’64). In the transillumination liver made by Reeve’s and his co-workers
technique, by exposing the liver under the indicating that hepatic microcirculatory
microscope, artifacts which affect the nor- systems in these two species of animals
mal blood flow of the hepatic vascular are similar.
channels may be easily introduced (Nakata
et al., ’60). In this regard, the blood flow in
the portal venous system is more vulnerable Aunap, E. 1931 Uber der Verlauf der Arteria
Hepatica in der Leber. Z. Micro. Anat. Forsch.,
to any effects than the hepatic arterial sys25: 238-251.
tem because of its low blood pressure. In
G . R., and B. T. Mayes 1930 Ligaaddition, microcirculation in the edges and Cameron,
tion of the hepatic artery. J. Path. and Bact.,
subcapsular areas of the liver which are
332: 799-831.
the principal areas studied by the transil- Chrzonszczewsky, N. 1866 Zur Anatomie und
Physiologie der Leber. Virchow’s Arch., 35:
lumination technique is considerably differ153-165.
ent from that in other areas because small
D. J., and J. Brim 1965 A method of
branches of three hepatic vascular systems
preparing three dimensional casts of the microrun perpendicular to the surface and reach
circulation of the skin. J. Invest. Dcrm., 45:
the subcapsular area where they form
Elias, H. 1949 A re-examination of the strucanastomoses.
ture of the mammalian liver, 11. The hepatic
Demonstration of intralobular arterioles
lobule and its relation to the vascular and biliby Elias (’49), Elias and Petty (’53) and
ary systems. Am. J. Anat., 85: 379-456.
Tajiri (’60) by perfusion techniques is Elias, H., and D. Petty 1953 Terminal dismore difficult to reconcile with the obsertribution of the hepatic artery. Anat. Rec.,
116: 9-18.
vations made by others (Olds and Stafford,
’30; Cameron and Mayes, ’30; Knisely et Elias, H., and H. Popper 1955 Venous distribution in livers. Comparison i n man and
al., ’48 and Reeves et al., ’66) who failed
experimental animals and application to the
to show any indications of intralobular
morphogenesis of cirrhosis. A.M.A. Arch. Path.,
arterial terminals. Unfortunately, those
59: 332-340.
investigators who observed intralobular Gershbein, L. L., and H. Elias 1954 Observations on the anatomy of the rat liver. Anat.
arteriolar terminals did not specify the
Rec., 120: 85-98.
extent of distribution of such arterioles
Hartroft, W. S. 1952 Morphology of the liver,
within the lobule or the functional sigin Transaction of the 11th Conference of Liver
nificance of this type of blood supply. If
Injury, Josiah Macy, Jr., Foundation, New
York, p. 169.
this is the principal way that the hepatic
arteries terminate, uniform mixing and Knisely, M. H., E. H. Bloch and L. Warner
1948 Selective phagocytosis. 1. Microscopic
distribution of arterial blood and smooth
observations concerning the regulation of the
dissipation of arterial blood pressures withblood flow through the liver and other organs
in the lobule may not be facilitated unless
and the mechanism and rate of phagocytic
removal of particles from the blood. Det. Kong.
the arteriolar terminals are distributed uniDans. Videnskab. Biol. Skr. 4, 7: 1-93.
formly within the lobule.
K. 1964 Symposium on intrahepatic
Our present results also failed to demon- Nakata,
microcirculation: Methods f o r the study of instrate intralobular arterioles and our two
tra- or extra-lobular circulation; hemodynamic
color perfusion preparations strongly sugand morphological aspects. I. Mlcroscopic observation of the intralobular blood circulation
gested that hepatic arterial terminals end
and the pressure gradient of hepatic vascular
at the periphery of the lobules. The honeypathway. Jap. Circ. J., 28: 24-27.
comb pattern of the peripheral sinusoids Nakata, K., G . F. Leong and R. W. Brauer 1960
appears particularly suited for buffering
Direct measurement of blood pressures in minute vessels of the liver. Am. J. Physiol., 199:
high arterial pressures which may be
1181-1 188.
transmitted as the arterial blood enters
sinusoids and joins with the portal blood. Olds, J. M., and E. S . Stafford 1930 On the
manner of anastomosis of the hepatic and porAt the same time, rich anastomotic chantal circulations. Bull. Johns Hbpkins Hasp.,
nels of peripheral sinusoids may achieve
47: 176-185.
Rappaport, A. M. 1963 Anatomic considerations
in Diseases of the Liver. L. Schiff, ed. J. B.
Lippincott Company, Philadelphia, p. 22.
Reeves, J. T., J. E. Leathers and C. Boatright
1966 Microradiography of the rabbit’s hepatic
microcirculation. The similarity of the hepatic
portal and pulmonary arterial circulations.
Anat. Rec., 154: 103-119.
Sobin, S. S . , W. G. Frasher, Jr. and H. M. Tremer
1962 Vasa vasorum of the pulmonary artery
of the rabbit. Circ. Rescarch, 11: 257-263.
Sobin, S . S., W. G. Frasher, Jr., H. M. Tremer
and G. G. Hadley 1963 The microcirculation
of the tracheal mucosa. Angiology, 14: 165-170.
Tajiri, S. 1960 The terniinal distribution of the
hepatic artery. Acta Med. Okayama, 14: 215224.
Wakim, K. G., and F. C. Mann 1942 The intrahepatic circulation of blood. Anat. Rec., 82:
1953 The blood supply of the normal
liver. Proc. Staff Meet. Mayo Clinic, 28: 218227.
1 A rat liver perfused via the hepatic vein. Central veins ( c ) and
radial sinusoids surrounding them are well perfused, while portal
veins ( p ) are seen in the areas where smusoids are poorly perfused.
A primary central vein ( p c ) is seen in the left lower corner extending
to join the other to form a larger central vein ( X 30).
A r a t liver perfused via the hepatic vein. A terminal twig ( t p ) of
a n axial distributing vein is seen anastomosing to a large central vein
( c ) through sinusoids. One primary central vein (pc) is seen entering the larger central vein. p, portal vein ( X 60).
Tatsuo Hase and Jack Brim
A rat liver perfused via the hepatic vein. A hepatic vein ( h v ) in
the left is seen receiving branching and non-branching central veins
( c ) . Silicone rubber filled i n the hepatic vein escaped on sectioning.
p, portal vein ( X 25).
A rat liver perfused via the portal vein. A cross section of a hepatic
lobule shows several portal canals with thcir portal veins ( p ) at the
periphery. Peripheral sinusoids are richly anastomotic compared to
less anastornotic radial sinusoids. The poorly perfused central veins
( c ) in the center of the lobule show some irregularity because of
their convergence just beneath the cut surface. ( x 50).
Tatsuo Hase and Jack Brim
A rat liver perfused via the hepatic artery. The branching patterns
of the portal venous and hepatic arterial trees are shown. p, portal
vein; a, hepatic artery ( X 25).
6 A rat liver perfused via the hepatic artery. A higher magnification
of figure 5 showing branches of portal vein ( p ) and hepatic artery
( a ) . (x60).
Tatsuo Hase and Jack Brim
A rat liver perfused via the portal vein. Two distributing veins (dv)
branching out of a conducting vein (cv) are seen. Inlet venules are
indicated by arrows. Note the conducting veins (right lower corner)
giving rise to inlet venules. Central veins ( c ) and their surrounding
sinusoids are relatively poorly perfused. ( X 50).
8 A rat liver perfused via the portal vein. A conducting vein (cv) is
seen giving rise to small short veins and venules (arrows) at right
angles. These short veins are the same i n charactcristics to inlet
venules in that they immediately branch into neighboring peripheral
sinusoids. Central veins ( c ) are seen i n relatively poorly perfused
areas. ( x 50).
Tatsuo Hase and Jack Brim
9 A rat liver perfused via the portal vein. A small portal vein branch
is seen connecting capillaries in a portal canal to portal veins ( p ) as
a n internal root (i). c, central vein ( x 50).
10 A rat liver perfused via the portal vein. Radicular portal venules ( r )
are seen connecting a periductal plexus ( d ) to peripheral sinusoids.
p, portal vein; c, central vein ( X 50).
Tatsuo Hase and Jack Brim
A rat liver perfused via the portal vein. A long capillary network of
periductal plexus ( d ) is seen with internal roots of portal veins ( i )
and radicular portal venules ( r ) . p, portal vein; c, central vein ( X 50).
12 A rat liver perfused via the hepatic artery. A n anastomosis between
a portal vein ( p ) and hepatic artery ( a ) through a capillary network
is seen. Arrows point t o the connections between the capillary network and the vessels. ( X 60).
Tatsuo Hase and Jack Brim
Без категории
Размер файла
1 329 Кб
architecture, microcirculation, observations, rat, live
Пожаловаться на содержимое документа