The spinal arachnoid villi of the monkeys Cercopithecus aethiops sabaeus and Macaca irus.код для вставкиСкачать
The Spinal Arachnoid Villi of the Monkeys Cercopithecus crethiops scrbcreus and Mcrccrccr irus' KEASLEY WELCH AND MICHAEL POLLAY Division of Neurosurgery, University of Colorado Medical Center and Veterans Administration Hospital, Denver, Colorado The spinal arachnoid villi have been studied in the monkey by the examination of serial sections of spinal nerve roots, particular attention being paid to the relations of the arachnoid to the dura mater and to the veins which are regularly applied to the emerging roots. Although the roots vary in their content of specialized arachnoid formations, in many, columns and clusters of arachnoidal cells occupy spaces between collagen bundles in the dura mater. In a few, leptomeningeal tissue extends completely through the dura. In approximately one of each six roots examined, arachnoidal tissue formed part of the wall of and projected into a vein associated with the root. This relationship to veins is considered an especially compelling point in dispelling the doubt which has been cast upon the conception of spinal arachnoid villi as specialized structures. ABSTRACT It is widely appreciated that at the sites of exit of nerve roots from the spinal theca, there are often found leptomeningeal proliferations which resemble the arachnoid villi of the cerebral meninges. Whether these are specialized structures as proposed by Elman ('23), who first described them, or simply unspecific formations or reactions of perhaps pathological significance, as implied by others, remains an unresolved question. The subject has been re-investigated by study of the spinal nerve roots and related meninges and blood vessels in two species of monkey. In the African green monkey (Cercopithecus aethiops sabaeus) and the Cynamologus monkey (Macaca irus) the evidence that there are indeed specialized structures bearing a close resemblance to arachnoid villi of the cerebral meninges is unmistakable. METHODS Serial sections of 32 nerve roots, of which 31 were lumbar and one cervical, from 15 monkeys were studied. These were cut at 5 LI from formalin fixed and paraffin embedded material and stained with hematoxylin and eosin. In two anesthetized animals colored latex was injected into the inferior vena cava under pressure in order to outline the venous plexuses in and surrounding the nerve roots, and in four animals direct observations of the vasculature of several roots were made through a dissecting microscope during life. OBSERVATIONS Venous plexuses associated with spinal nerve roots. A venous plexus is regularly found to be associated with each spinal nerve root and its sheath but the size and extent of the plexus varies widely from one root to the next. The plexus is in and attached to the dural sleeve of the nerve root and the connective tissue covering the spinal nerve root ganglion. Deeper branches are entirely within the dura mater; the more supeficial are applied to the dura. There does not seem to be a regular or characteristic pattern of distribution of the veins; in some roots, they are more concentrated on the axillary side and in others, toward the upper border. The portions of the plexus which are entirely within the dura are very thinwalled; often there is only endothelium applied directly to the connective tissue of the dura. The more superficial veins have walls of their own, especially on their external aspect, but these sometimes have only an endothelial layer applied to dura 1 Aided by research g p n t (B-627)from the National Institute of Neurqloglcal Ihseases and Blmdness, United States Pubhc Health Servlce. 43 44 XEASLEY WELCH AND MICHAEL POLLAY mater where vein and dura are apposed (figs. 1, 2, 3). The plexus receives blood from a segmental vein from within the subarachnoid space and from veins of the dorsal root ganglion. At levels where there is a particularly large segmental vein leaving the subarachnoid space the plexus is likely to be more rich than at other levels. Drainage is through veins leaving the spinal canal accompanying the spinal nerve root and by numerous and irregular connections with epidural veins. Direct observation was made through a dissecting microscope of a number of nerve roots during life and no vein was ever observed to contain clear fluid. The leptomeninges associated with spinal nerve roots. Clusters of arachnoidal cells are often found in the spinal leptomeninges and such nests are particularly common at the very apex of the subarachnoid space where the arachnoid is reflected back along the nerve root. In this situation, and sometimes more proximally, these proliferated formations come to occupy the interstices of the dura mater of the root sleeve and the connective tissue capsule of the spinal root ganglion. Rarely, there is complete penetration of the dura mater so that the arachnoidal cells are found in an epidural location. A thinwalled arachnoid diverticulum penetrating the dura mater and coming into relation with epidural fat, was observed on one occasion (fig. 4). Other arachnoidal formations consisting of arachnoidal cells intermixed with fibers come into relation with venous channels of the plexus applied to the root, form parts of the walls of such vessels and actually project into them (figs. 2, 5 ) . The villus depicted in figure 2 is, of those we have examined, the largest, the most complex and the most suitable for detailed study. The vein which receives this villus is thin-walled. That portion applied to the dura mater consists only of an endothelial layer, while the portion presenting to the epidural tissue shows a thin layer of collagenous connective tissue external to the endothelium. The defect in the dura mater, through which the arachnoidal tissue protrudes, measures 220 K. At the edges of the dura mater bordering the defect the connective tissue of the dura does not, at most points, end abruptly. For a variable distance back from the edge, perhaps 20 CI,looser arachnoidal tissue is intermingled with the connective tissue bundles of the dura and some of the latter are extended for a short distance into the villus or for a distance toward its dome. The structure protrudes 130 p into the vein. The villus itself is made up of mesothelial cells applied to fibers outlining, between them, many spaces. The largest of these spaces measures approximately 16 II in diameter, the average about one-fourth of this. From the surface of the villus, several invaginations protrude deeply into the structure. The covering is the same as the walls of the spaces within. The simplest formations associated with veins were only a few cell layers deep occupying a portion of the wall of a vein. The nerve roots are not equally endowed with spinal arachnoid villi. In only 5 of the 32 roots studied was arachnoidal tissue found to form a portion of the wall of, or to project into a vein. DISCUSSION Elman (’23) studied the microscopic anatomy of the meninges at the site of exit of spinal nerve roots of the dog. He discovered that at the distal extent of the arachnoid, where it is obliquely reflected from its position adjacent to the dura mater back along the intradural portion of the emerging nerve roots, there often occurred clusters or columns of arachnoidal cells. These were attached to and blended with the dura mater and even penetrated into the interstices of the root sleeve. Upon subarachnoid injection of iron ammonium citrate and potassium ferrocyanide and subsequent acid fixation, he showed granules of Prussian blue to be precipitated amongst the mesothelial cells of the clusters, in the dura mater in that neighborhood, in the tissue surrounding the nerve roots and in some veins. He did not describe a connection between the spinal arachnoid “granulations” and the adjacent veins but pointed out that large veins are numerous in their neighborhood. The rich venous plexuses associated with the spinal theca at the sites of emerg- SPINAL ARACHNOID VILLI ence of nerve roots were later studied in cats by Wislocki and Kubie (Wislocki, '32). By the intravascular injection of India ink or carmine gelatin, they outlined branches of segmental arteries giving rise to a capillary network in the dura mater. This capillary bed was found to drain into veins which accompany the segmental arteries, but as they reached the dural sleeve of the nerve root they enlarged and entered a series of sinuses in the dural sheath. These, in turn, were described to drain into epidural veins. Wislocki thought it probable that the veins of the plexus might bear a functional relationship to the leptomeninges similar to that borne by the venous sinuses of the cranial dura although the spinal veins had not been described to receive arachnoid villi. In the rat, Woollam and Millen ('58) described massive venous plexuses associated with and intimately applied to the emerging spinal nerve roots. These venous plexuses they found to be supplied by veins from the subarachnoid space and from the spinal root and ganglion. They described also some of these veins to receive arachnoid villi which had pierced the dura mater and actually projected into the vascular lumen. An issue with respect to the significance of the villi has been raised especially from the study of human material where the emphasis has been largely upon pathological change. Hassin ('30) found overgrowth of arachnoid and the formation of arachnoidal protrusions, some of which remained subdural while others penetrated into the dura mater. These formations he found to be exaggerated under certain pathological conditions. He did not describe relationships of the proliferated arachnoid to veins. Unwilling to ascribe to the structures any role in the drainage of cerebrospinal fluid, he conceded their existence, albeit as a manifestation of some general propensity of the leptomeninx to proliferate and invade the dura. Rexed and Wennstrom ('59) described proliferations of the spinal leptomeninges associated with nerve roots leading, in some cases, to cyst formation which compressed nerve roots or occupied the spinal root ganglion. Some of the arachnoidal proliferations actually penetrated 45 completely through the dura mater. No relation between the arachnoid and the adjacent veins was demonstrated. Rexed and Wennstrom considered whether the arachnoid penetrations of the dura might represent spinal arachnoid villi but dismissed this possibility on the grounds that such penetrations were not regularly seen in the absence of other proliferative change. Thus, they included the existence of these formations with the other changes they described as being reflections of pathological response to an inflammatory process in the meninges. Our own observations in the monkey are compatible with those described in other species with the exception that we have not encountered changes which could be considered pathological. The existence of groups of arachnoidal cells occupying tissue spaces of the dura mater and capsule of the ganglion seem not dissimilar to the observations originally made by Elman ('23) for the dog and by Hassin ('30) in human tissue, while the extension of these arachnoidal formations all the way through the dura was also seen in human material by Rexed and Wennstrom ('59). The relationship to veins, seen by Woollam and Millen ('58) in the rat and shown also in our own material, appears to be of crucial importance in any consideration of the significance of arachnoid villi of the spinal meninges. Simple proliferation of tissue in response to a pathological agent would not be expected to lead to invasion of venous walls. The establishment of relationships between these structures and the lumina of veins confirms the suspicion expressed long ago by Wislocki ('32) and, although direct evidence cannot be offered, there is sufficient resemblance between some of these and the villi of the cerebral meninges (Welch and Friedman, '60) to suggest a similar function. In this connection, it may be pointed out that the spinal arachnoid villi are so situated that they might conduct, to the sites where they have been found, the various tracer particles which have been employed in the study of cerebrospinal fluid pathways. The studies made before 1948were reviewed by Brierly and Field ('48). Introduced into the subarachnoid space of rabbits, particles of 46 KEASLEY WELCH AND MICHAEL POLLAY DURA MATER DORSAL NERVE ROOT E Fig. 1 Diagrammatic representation of the kinds of meningeal and vascular relationships found in spinal nerve roots. At A growth of arachnoidal cells within the dura mater is represented. B represents complete penetration of dura mater by leptomeninx and C protrusion of arachnoidal tissue into a vein. At D and E arachnoidal proliferations are shown. India ink reached the epidural connective tissue and regional lymphatics (Brierly and Field, ’48). They were also found within spinal arachnoid villi of the rat by Woollam and Millen (‘58). A pathway into the blood was shown for colloidal palladium by Howarth ( ’ 5 2 )and Howarth and Cooper (’55) in cats, and, although the authors believed the passage to be directly through vascular walls, the spinal arachnoid villi were not studied. LITERATURE CITED Brierley, J. B., and E. J. Field 1948 The connexions of the spinal subarachnoid space with the lymphatic system. J. Anat., 82: 153-166. Elman, R. 1923 Spinal arachnoid granulations with especial reference to the cerebrospinal fluid. Bull. Johns Hopkins Hosp., 34: 99-104. Hassin, G. B. 1930 Villi (Pacchionian bodies) of the spinal arachnoid. A.M.A. Arch. Neurol. Psychiat., 23: 65-78. Howarth, F. 1952 Observations on the passage of a colloid from cerebrospinal fluid to blood and tissues. Brit. J. Pharmacol., 7: 573-580. Howarth, F., and E. R. A. Cooper 1955 The fate of certain foreign colIoids and crystalloids after subarachnoid injection. Acta Anat., 25: 112-140. Rexed, B. A., and K. G. Wennstrom 1959 Arachnoidal proliferation and cystic formation in the spinal nerve root pouches of man. J. Neurosurg., 16: 73-84. Welch, K., and V. Friedman 1960 The cerebrospinal fluid valves. Brain., 83.454-469. Wislocki, G. B. 1932 in “Special Cytology.” Ed. by Edmund V. Cowdry, 2nd edition, New York, Paul B. Hoeber, Vol. 3, p. 1483. Woollam, D. H. M., and J. W. Millen 1958 Observations on the production and circulation of the cerebrospinal fluid in Ciba Foundation Symposium on the Cerebrospinal Fluid. G. E. W. Wolstenholme, editor, Boston, Little Brown and Company, pp. 124-146. SPINAL ARACHNOID VILLI PLATE 1 Keasley Welch and Michael Pollay 2 3 An arachnoid villus protruding into a venous channel associated with a nerve root sleeve is shown. H & E X 200. AV, arachnoid villus; V, vein; DM, dura mater; NR, nerve root. Higher power view of a portion of the villus shown in figure 2. H & E x 640. 47 SPINAL ARACHNOID VILLI Keasley Welch and Michael Pollay 4 5 48 PLATE 2 A diverticulum of arachnoid (arrow) completely penetrates the dura mater and comes into relation with epidural fat. H & E X 200. NR, nerve root; DM, dura mater. A villus composed of both compact and loose arachnoidal tissue forms a wall of a space which in serial sections is connected with epidural veins. H & E X 200. V, vein; DM, dura mater; NR, nerve root.