вход по аккаунту


Microvascular networks in tympanic membrane malleus periosteum and annulus perichondrium of neonatal mongrel dog A vasculoanatomic model for surgical considerations.

код для вставкиСкачать
Microvascular Networks in Tympanic Membrane, Malleus Periosteum,
and Annulus Perichondrium of Neonatal Mongrel Dog: A
Vasculoanatomic Model for Surgical Considerations
Department of Anatomy and Cellular Biology, The Medical College of Wisconsin, Milwaukee, Wisconsin 53226
India-ink-imaged blood-vessel networks in cleared tympanic membranes and adnexa from
ten neonatal dogs were examined microscopically and
photographed. The major significance of the study lies
in documentation of a dual source of arterial supply, a
bilaminar relationship of arterial and venous plexuses
intrinsic to the tympanic membrane, and a consistent
major venous pathway relative to a definite locus (pars
flaccida of the membrane). Illustration of all three types
of blood pathways (arteries, veins, and capillaries) provides new vasculoanatomic data that are essential to
ear surgery, specifically-to myringotomies and myringoplasties. A comparison was made between dog and
human tympanic membrane structures and their arterial supplies. Close similarities suggested that dog tympanic membrane might serve as a suitable model for
development of innovative surgical procedures and as a
model for rehearsal of difficult techniques. The results
of this investigation provide a valuable caveat to
There is a paucity of information concerning mammalian tympanic membrane microangiology. Illustrative
and descriptive accounts of microvessel networks in the
human ear made visible by radiographic perfusate
(Saini, 1964), and by silicone perfusate (Anson and Donaldson, 1981)provide the only current documentation of
human tympanic membrane microvasculature. These
angiographic studies clearly indicated a central source
of blood supply to the human tympanic membrane but
did not unequivocally demonstrate an additional peripheral source, primarily due to incomplete vascular perfusion or to the obscuring of vessel pathways by overiying
dense tissue.
The present study concerns the microvascular network of the tympanic membrane and its adnexa in young
mongrel dogs. Observations were compared with traditional descriptive accounts of human tympanic membrane vascular patterns. Similarities between the
structure and vascular supply of dog and human tympanic membranes described in this study suggest that
dog tympanic membrane could serve as a model for
development of new surgical procedures and for rehearsal of difficult surgical techniques on human tympanic membranes.
0 1988 ALAN R.LISS, INC.
Ten neonatal mongrel dogs between 7 and 10 days of
age were killed by intraperitoneal pentothal overdose.
They were wrapped in wet toweling to prevent dehydration and refrigerated at 5°C for 48 hr to allow sufficient
time for them to pass into and partially out of rigor. The
right and left common carotid arteries were subsequently exposed by dissection, and the upper limbs of a
Y-shaped cannula were inserted into their lumina and
ligated. A 50-ml syringe filled with India ink diluted
with water (1:lO)was inserted into the free end of the
cannula. Vascular perfusion was accomplished by manual operation of the syringe; short, alternating, “pushpull”-type plunger strokes were used to simulate pumping action of the heart and associated systolic and diastolic pressures, but at exaggerated levels (Maher and
Swindle, 1962). This method urges the perfusate successively into arteries, then capillaries, and finally veins.
Perfusion pressure was not monitored. When all blood
vessels of the head and neck region were filled (based
upon experience), perfusion was stopped. The stem of
the cannula was clamped before the syringe was removed t o prevent retrograde flow and perfusate loss. If
perfusate loss is not prevented, such loss will cause
incomplete filling of some vessels, especially
An ill effect of exaggerated perfusion pressure is vasodilation of small vessels, but this can be somewhat
overcome. Upon completion of the perfusion, the animals were wrapped again in moist toweling and refrigerated at 5°C for 48 hr. During this time, some dilated
vessels contract passively because much of the excess
water in vascular lumina has an opportunity to pass
through the vessel walls and into perivascular tissue
spaces, leaving the solids (carbon particles) behind. Other
vessels may remain dilated, however, and these dilations are due primarily to large, dense deposits of carbon
particles in their lumina. Although some vessel calibers
remain dilated by this perfusion method, the distribution pattern is not altered. While the technique appears
to be quite satisfactory for vessel pattern and distribution studies, as in the present one, it is not recommended
Received January 19,1988.Accepted July 1, 1988.
The extrinsic arterial source
The stylomastoid branch of the posterior auricular
artery enters the stylomastoid canal and emits neural
and mucosal branches. Neural branches (arterii nervorum) supply the facial nerve, while mucosal branches
supply tympanum mucosa, mastoid antrum, and semicircular canals. The stylomastoid artery emits a prominent posterior tympanic branch that courses on the
surface of the annulus (Fig. 1) where it emits ramiscules
that form a microplexus in the perichondrium of the
annulus. Branches from this microplexus extend into
the tympanic membrane (Figs. 2, 3) where its fibrous
middle stratum blends with the fibrous perichondrium
of the annulus. The perichondrial microplexus also receives miniscule branches from the tympanic branch of
the maxillary artery.
The intrinsic arterial source
Fig. 1. Tympanic membrane, viewed from its outer surface at the
membrane-annulus junction. A, Annulus; B, posterior tympanic artery
and its branches in the perichondrium of the annulus; C, venous plexus
in the cutaneous stratum.
for vessel caliber measurements because of the vasodilatory effect.
Necropsied tympanic membranes and their adnexa
(malleus and annulus) were cleared by immersing them
sequentially in increasing concentration of water:
alcohol, alcohol:xylene, and xy1ene:methylsalicylate.
Pure methylsalicylate was the final clearing agent, and
it also served as a permanent preserving medium (Spalteholz, 1914).Images of the three blood-vesseltypes were
viewed with a stereomicroscope while each specimen
was immersed in methylsalicylate. Combinations of
transmitted and incident light sources were used and
adjusted to obtain appropriate illumination of microscopic fields for photography. Additional microdissections were performed on all specimens to improve the
visibility of vascular beds obscured by overlying dense
tissue. A 35-mm camera and Panatomic-X film were
used to photograph microscopic fields at magnifications
ranging from x 5 to x50. No attempt was made to record
accurate magnifications, but microvessel relationships
and patterns of arrangement are clearly illustrated
Arteries of Dog Tympanic Membrane and Adnexa
Dog tympanic membrane, like that of the human,
receives arterial supply from intrinsic and extrinsic
sources. The extrinsic source stems from the stylomastoid branch of the posterior auricular artery, while intrinsic sources stem from deep auricular and anterior
tympanic branches of the maxillary artery.
The deep auricular branch from the first part of the
maxillary artery-en route to the pars flaccida-emits
rami to the membrane and cutaneous portions of the
external auditory canal. The artery at the pars flaccida
forms anterior and posterior branches that course in the
periosteum of the manubrium of the malleus at its posterior margin (Figs. 4-6). Here these major periosteal
branches emit miniscule branches that form a microplexus in the periosteum of the manubrium of the malleus (Fig. 6). Branches from this microplexus extend into
the tympanic membrane where the fibrous middle stratum of the membrane blends with the fibrous periosteum of the malleus (Figs. 43).
Miniscule arteries from the malleus periosteal microplexus and annulus perichondrial microplexus extend
into the fibrous stratum of the tympanic membrane and
form radial branches paralleling the direction of the
radial fibers of that stratum. In their aggregate, inkimaged branches intrinsic to the membrane give the
appearance of wagon wheel spokes. Each radially arranged artery is flanked by at least one but more often
two miniscule veins. Paired flanking veins (Venae comites) are frequently connected to one another by very
short branches crossing their companion artery at right
angles (Figs. 2,3,7).
Veins of the Tympanic Membrane and Its Adnexa
Throughout the body, venous radicals stem from capillary beds and compose the first order of collectingveins.
The origin of the tympanic membrane venous plexus is
no exception to this basic vascular concept. Successive
orders of convergencies ultimately form two distinct venous routes that course separately from the tympanic
membrane. A principal route extends into the middle
ear via the pars flaccida and joins the middle ear venous
plexus. Minor routes join cutaneous veins of the external
auditory meatus at the membrane-meatus junction. Exit
portals for veins coursing from tympanic membrane are
not the same as entrance portals for arteries coursing to
tympanic membrane. Veins-to-artery companionships
(venae comites) were observed only in the pars radialis
of the tympanic membrane (Fig. 3). Elsewhere, venous
and arterial plexuses form a bilaminar relationship. The
former is peripheral and within the cutaneous stratum
while the latter is deep and within the fibrous stratum
(Figs. 5,7).
Fig. 2. Overview of neonatal dog tympanic membrane and adnexa,
seen from the internal surface and illustrating ink-imaged vessel networks. The cutaneous stratum was removed above the horizontal black
line to permit viewing of radially arranged arteries and flanking veins
(venae comites) in the fibrous middle stratum (A). B, Venous plexus in
the cutaneous stratum; C, manubrium of the malleus; D, annulus.
Venous radicals stemming from capillaries in the annulus perichondrium, malleus periosteum, and tympanic membrane cutaneous stratum converge. A succession of convergencies forms the principal collecting
vein of the tympanic membrane (Fig. 8). The principal
vein forms a loop paralleling the long axis of the manubrium of the malleus (Figs. 2, 7). The two ends of the
venous loop exit the area at the pars flaccida and course
inwardly to join the middle ear venous plexus. Miniscule
veins exit the tympanic membrane at its junction with
the cutaneous stratum of the external canal.
The principal vein in dog tympanic membrane and its
adnexa course into the middle ear via the pars flaccida
(Figs. 2,7). But whether this relationship is the same in
the human tympanic membrane is not certain because
venous networks intrinsic t o the structure have not been
Fig. 3. The fibrous stratum of the tympanic membrane viewed from
the internal surface, enlarged from top left portion of Fig. 2. A, Peri-
chondiral network of the annulus; B, radial arteries and companion
veins; C, a portion of the capillary network at the epithelial base of
the cutaneous stratum.
demonstrated satisfactorily. Although veins in the cutaneous stratum of dog tympanic membrane often vary
in number, size, and pattern of distribution, the location
of the principal vein (parallel to the long axis of the
malleus) was consistent in the 20 specimens examined.
Capillaries are not distributed uniformly throughout
the three major strata of tympanic membrane. The inner (mucosal) stratum is relatively free of capillaries. In
the middle (fibrous) stratum, capillaries appear t o be
limited to areas at junctions of the fibrous stratum with
the fibrous periosteum of the malleus and with the fibrous perichondrium of the annulus (Fig. 3). The outer
(cutaneous) stratum contains a greater proportion of capillaries per unit area than the other two strata. Here,
the ink-imaged capillary network is found in neat array
and evenly distributed at the epithelial base of cutaneous stratum (Fig. 3). The inferior topography of the
epithelial base of cutaneous stratum is flat. Unlike the
Fig. 4. The lower midthird of the tympanic membrane, viewed from
its internal surface. The malleus and its fibrous covering and vessels
are in place. Images of vessel pathways from these fibrous coverings
are seen coursing into the tympanic membrane. A, Annulus; B, portion
of the venous plexus in the cutaneous stratum; C, anterior and posterior branches of the deep auricular artery coursing in the periosteum
of the malleus. The area outlined at upper left is shown at higher
magnification in Figure 5.
bility of outcome when similar procedures are undertaken on human subjects. Since the structure and
arterial supply of dog tympanic membrane appear to be
very similar to those of human tympanic membrane, it
seems reasonable to suggest that dog tympanic memIt is generally believed that extrapolations derived brane could provide a suitable biological model for surfrom experimental surgical data enhance the predicta- gical rehearsal and experimentation. Furthermore, this
cutaneous stratum of the external auditory meatus, it is
not fitted with grooves and sockets corresponding to
ridges and papillae of connective tissue.
Fig. 5. View from the internal surface of the tympanic membrane.
A, Arteries in the vicinity of the periosteum of the malleus extending
into the tympanic membrane fibrous stratum; B,confluence of flanking veins (venae comites) and their collecting vein, parallel to the long
axis of the malleus; C, a part of the venous plexus in the cutaneous
stratum. The capillary network at the epithelial base of the cutaneous
stratum, although somewhat out of focus, is also visible.
model provides opportunities for postsurgical histologic
studies that are not available following human ear
This study clearly indicates that anterior and posterior branches of the deep auricular artery form a network in the periosteum of the manubrium, and branches
from this network extend into the fibrous stratum of the
tympanic membrane in a radial fashion. The study also
shows that the posterior tympanic branch of the stylo-
mastoid artery forms a network in the perichon-&ium of
the annulus and that branches of this network also
extend into the fibrous stratum of the tympanic membrane in a radial fashion. As a consequence, we can
conclude that the tympanic membrane in dogs is supplied from two sources: 1) centrally, by ramifications
from the malleus periosteal network, and 2) circumferentially, by ramifications from the annulus perichondrial network. Concerning the direction of arterial flow
Fig. 6. Lower portion of an excised malleus, viewed from its lateral
surface. Posterior branch of the deep auricular artery (A), and microplexus (B) formed from it. Radial branches (partially interrupted) extend from the periosteal microplexus of the malleus (see Fig. 7).
to the human tympanic membrane provided by Saini
(1964)and Anson and Donaldson (1981) clearly indicated
only a central source of supply. Their photographic evidence seemingly supports their observations, but imaged vessels at the circumference of the membrane and
in the perichondrium of the annulus were obscured by
overlying dense tissue. Consequently any continuities of
the annulus perichondrial network with networks of
vessels intrinsic to the tympanic membrane that may
be present were not visible.
The fact that the tympanic membrane in dog has two
major sources of supply is particularly relevant to a
myringocentesis. In this procedure, incisions of the tympanic membrane are usually made starting at the bottom of the drum in a backward and upward direction
parallel to the annulus border. Accordingly, the incision
extends from 6 o’clock, passes through 7, 8, 9, and 10
o’clock, and stops at 11 o’clock. This incision pathway
also avoids interception of the chorda tympani nerve
and the incudostapedial articulation. Currently advocated procedures for a myringocentesis might compromise structures at the tympanic membrane circumference if the tympanic membrane were supported by
only one arterial source of supply. On the other hand, if
the human tympanic membrane were supplied by two
sources as it is in dog, that portion of the membrane
marginal t o the annulus would be very adequately supplied by branches from the perichondrial network of the
The numerous veins so prominent in the pars cutaneous of the dog tympanic membrane are not constant
in number, size, or location. Network arrangements vary
but the venous exit portal from the membrane is constant (i.e., via the pars flaccida). Assuming that venous
plexus variations are present in the human tympanic
Fig. 7. Radial arteries and flanking veins (venae comites) in the
fibrous stratum, as seen from the inner surface of the tympanic membrane. Note that the these vessels extend radially from annulus peri-
chondrium (A) and manubrium periosteum (B). The principal vein (C)
is parallel to the long axis of the malleus (D).
A portion of the venous
plexus of the cutaneous stratum is shown (El.
in radial arteries, it is centrifugal from the central source
and centripetal from the circumferential source.
A dual arterial supply to the human tympanic membrane is yet in doubt. Illustrations of arteries intrinsic
Fig. 8. The cutaneous stratum from the exterior surface of tympanic membrane at its junction with the annulus (A). B, Converging
veins that stem from capillary networks in the annulus perichondrium
and cutaneous stratum of the tympanic membrane; C, a large venous
branch en route to principal vein coursing parallel to the long axis of
the manubrium.
membrane and that an acute or chronic passive venous
congestion were present, surgeons should be alert to the
possibility of significancevenous bleeding during mryingocentesis. This extravasation will course internally as
well as externally. Unaspirated blood may collect in the
epitympanic recess, expecially in its small compartments. During infectious processes these same compartments often harbor pathogenic organisms and debris
that are difficult to eliminate.
The walls and contents of the human tympanic cavity
are supplied by three major and three minor sources.
The major sources are 1) the anterior tympanic branch
of the maxillary artery, 2) the stylomastoid branch of
the occipital artery, and 3) the posterior auricular artery. Minor sources include 1)the petrosal and superior
tympanic branches of the middle meningeal artery, 2)
branches from the ascending pharyngeal artery, and 3)
the artery of the pterygoid canal.
Only the major arterial sources of supply to the tympanic membrane and its adnexa in neonatal mongrel
dog are illustrated and described in this paper. These
arteries were consistently present in every specimen
examined. However, these findings do not guarantee
that the same is true for all strains of dog or for humans.
It would be prudent to assume that branches of minor
arterial sources could have a primary role in supplying
the tympanic membrane in instances of aberrant development or growth, and that differences in primacy of
supply may be present in different age groups or races
of humans. Consequently, surgeons who contemplate an
otoplasty are alerted to the possibility that arterial and
venous pathway variations may be present in normally
developed tympanic membranes but even more so in
abnormally developed ones.
Clearly, inadvertent section of a primary arterial supply source can compromise recovery. Indeed, a serious
deprivation of blood supply can result in necrosis or
sphacelus of reconstructed flaps or grafts.
Five types of tympanoplasty and their variations were
described by Miglets et al. (1986). It is interesting to
note that the malleus and incus are missing in their
type 111. Since earlier angiographic illustrations indicated that the tympanic membrane is supplied mainly
by arteries stemming from the periosteal network of the
malleus, and since this structure may be missing, the
question remains: what artery or arteries supply the
tympanic membrane in that condition? Angiograms provided in this paper clearly illustrate that the arterial
network intrinsic to the tympanic membrane stems from
networks of arteries in the periostium of the malleus
and perichondrium of the annulus.
Illustrations of arterial and venous pathway variations in instances of normal and abnormal growth of the
human ear would provide essential vasculoanatomic
data for consideration prior to ear surgery on infants
and young children. It is unfortunate that these data
have not been provided. It is suggested that methods
described in this paper, or perhaps similar ones, could
be used to examine microvascular networks intrinsic to
tympanic membranes and adnexa from spontaneous human abortuses and thus provide these missing data.
Illustrations contained in this paper were presented at
the 100th Annual Meeting of the American Association
of Anatomists, Washington, D.C., May 10-14, 1987 (Anatomical Record, 1987, Volume 219, Number 2).
Anson, B.J., and J.A. Donaldson 1981 Surgical Anatomy of the Temporal bone, 3rd ed. W.B. Saunders, Philadelphia, p. 710.
Maher. W.P. 1987 Arterial. venous. and cauillarv networks intrinsic to
selected structures of the middle ear, -as &died in neonatal dog.
Anat. Rec. 218:86A (Abstract).
Maher, W.P., and P.F. Swindle 1962 Submucosal blood vessels of the
palate. Dent. Prog., 2167-180.
Miglets, A.W., M.H. Paparella, and W.H. Saunders 1986 Atlas of Ear
Surgery, 4th ed. C.V. Mosby, St. Louis, p. 368.
Saini, V.K. 1964 Vascular pattern of the human tympanic membrane.
Arch. Otolaryngol,, 79:193-196.
Spalteholz, W.S. 1914 Uber das Durchsichtigmachen von Menschlichen
und Tierchen Praparaten und seine Theoretischen Bedingungen,
2nd ed. S. Hertzel, Leipzig. (Cited in: The Microtomists Vade
Mecum, 10th ed. (J.B. Gatenby and T.S. Painter, eds. Blakiston’s,
1937, Philadelphia.)
Без категории
Размер файла
1 206 Кб
periosteum, mode, network, perichondrium, mongrel, vasculoanatomic, membranes, annulus, tympani, microvascular, surgical, malleus, considerations, neonatal, dog
Пожаловаться на содержимое документа