THE ANATOMICAL RECORD 230:3-21 (1991)

On the Question of a Subdural Space D.E. HAINES Department of Anatomy, The University of Mississippi Medical Center, Jackson, Mississippi 39216

ABSTRACT The structure of the meninges, with particular attention to the architecture of the inner portions of the dura mater and the arachnoid mater, has been reviewed in reference to the probable existence of a “subdural” space. The dura is composed of fibroblasts and large amounts of extracellular collagen. The innermost part of the dura is formed by the dural border cell layer. This layer is characterized by flattened cells with sinuous processes, extracellular spaces containing an amorphous material, and the presence of junctions between its cells. The dural border cell layer is continuous with the inner (meningeal) portions of the dura and may be attached to the underlying arachnoid by an occasional cell junction. The arachnoid consists of an outer part, the arachnoid barrier cell layer, and an inner portion, the arachnoid trabeculae which bridge the subarachnoid space. Arachnoid barrier cells are electron-lucent, closely apposed to each other, and joined by many cell junctions; in this layer there is little extracellular space and essentially no intercellular material. Arachnoid trabecular cells cross the subarachnoid space in a random manner, have extracellular collagen associated with their flattened processes, and form structures of variable shapes and sizes. There is no evidence of an intervening space between the arachnoid barrier cell layer and the dural border cell layer that would correlate with what has been called the subdural space. When a tissue space is created in this general area of the meninges it is the result of tissue damage and represents, in most instances, a cleaving open of the dural border cell layer. In this situation, extracellular spaces in the dural border cell layer are enlarged, cell junctions are separated, and it is probable that cell membranes are damaged. A survey of reports describing the morphology of the inner and outer capsule of so-called subdural hematomas in humans reveals that dural border cells are found in both parts of the capsule. Also, experimental infusion of blood into this portion of the meninges in animals frequently dissects open the dural border cell layer. These data support the view that what has been called a subdural hematoma is most frequently a lesion found within the layer formed by dural border cells. It is suggested that the so-called subdural space is not a “potential” space since the creation of a cleft in this area of the meninges is the result of tissue damage. In this respect it shares no similarities with legitimate potential spaces (i.e., serous cavities) found a t other locations in the body. It is concluded that there is no evidence of a subdural space (actual or potential) in the region of the dura-arachnoid junction and it is suggested that the term spatiurn subdurale be removed from Nomina Anatomica. The general structure and relationships of the cranial and spinal meninges are well known (see Williams et al., 1989) and it is universally recognized that a space of variable dimensions exists between the arachnoid and pia, the subarachnoid space (SAS). The SAS contains cerebrospinal fluid (CSF), has enlarged regions called cisterns, is fed by the choroid plexuses of the ventricular system via openings in the fourth ventricle, and accesses the venous system primarily through arachnoid villi/granulations. On the other hand, there is a difference of opinion concerning the probability of actual, or potential, spaces associated with other segments of the meninges. This review is primarily concerned with the structure of the men0 1991 WILEY-LISS, INC

inges at the dura-arachnoid interface and specifically addresses the concept of the so-called “subdural” space. HISTORICAL PERSPECTIVE ON A SUBDURAL SPACE

Early thinking on the probability of an organized space between the arachnoid and dura was influenced by the elegantly detailed studies of Key and Retzius

Received June 11, 1990; accepted August 6, 1990. Address reprint requests to Dr. D.E. Haines, Department of Anatomy, The University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216.

4

D.E. HAINES

(1875-1876). These authors not only described the structure of the meninges with special emphasis on the arachnoid, SAS, cisterns, and the pia mater, but also offered experimental evidence that a substance injected into the SAS did not mix with another substance injected into the presumed subdural space. They also reported that this space was a t least partially lined by a mesothelium. Quinche, in 1872 (cited in Weed, 1917a), had suggested that fluids moved from the subdural space into the SAS but not in the reverse direction, while Hill, in 1896 (cited in Weed, 1917a1, concluded that fluids moved freely in both directions between these two compartments. Langdon (18911, subsequent to a macroscopic examination of the meninges in fetal and adult human specimens, concluded that the arachnoid was a closed serous sac. He described parietal (next to the dura) and visceral arachnoid layers separated by the “arachnoid cavity proper”; this cavity appears to correlate, in general, with the subdural space of other early studies. Building on these data, investigators in the early part of this century offered what, a t the time, was considered compelling evidence that a f hid-containing space existed between the dura and arachnoid (Cushing, 1914; Weed, 1914a,b, 1916, 1917a,b, 1920, 1923, 1938; Wegefarth and Weed, 1914; Mallory, 1920; Penfield, 1923, 1924; see also Woollard, 1924; Leary and Edwards, 1933). In developmental studies Weed (1916, 1917a) reported that the mesenchymal layers forming the dura and arachnoid were fused in pig embryos up to about 50 mm but that these two layers could be “ . . . separated . . .” in larger specimens. A thin, distinct line of dense mesenchymal cells was seen at the dura-arachnoid interface (Weed, 1916, 1938) and interpreted as representing “ . . . the outer surface of the arachnoid . . .” and “. . . the inner surface of the dura mater.” This structure, as reported by Weed (1916), appears to correlate with the “dural limiting layer” as described by O’Rahilly and Miiller (1986) and Andres (1967). Weed (1916, 1917a) also noted that “. . . points offusion . . .,’ remained between the dura and arachnoid, such as in the arachnoid villi, and he chose to interpret this as basically representing an incomplete formation of the subdural space. The spaces which seemed to appear spontaneously between the arachnoid and dura in preparations of pig embryos larger than 50 mm and the fact that in such specimens the dura could be easily separated from the arachnoid was interpreted (Weed, 1916,1917a) as indicating the initial appearance of the subdural space (cavum subdurale). A further correlate of the advent of this cavity was a “. . . thickening . . .” on the inner surface of the dura and the appearance of what was interpreted as a mesothelial cell layer a t this point. Key and Retzius (1875-1876) had described a mesothelial-like layer of cells on the inner aspect of the dura; this concept was substantiated in developmental studies (Weed, 1916, 1917a) and gave rise to the idea that the subdural space was a mesothelial-lined fluidcontainig cavity (Weed, 1914b, 1917b, 1920,1938; Penfield, 1924). Regarding the origin of the fluid presumably found in the subdural space Weed (1914b) specifically noted that it was “. . . the product of the activity of the cells lining this serous cavity . . .” and not subarachnoid fluid that had passed through the arachnoid membrane. Although not disputing the ex-

istence of a subdural space, Mallory (1920) concluded that the inner surface of the dura was lined by fibroblasts and that the outer surface of the arachnoid “. . . develops a layer of cells resembling endothelium; the so-called dural endothelium . . .”. In spite of earlier suggestions of fluid movement out of the subdural space, or between it and the SAS (see Weed, 1917a for review), the results of experimental studies in which different substances were simultaneously injected into the presumed subdural space and the SAS (Key and Retzius, 1875-1876) or only into the SAS (Weed, 1914a,b, 1923; Woollard, 1924) revealed a lack of communication between these two cavities. This was especially the case when dyes that were injected into the SAS did not appear in the area of the subdural space (Weed, 1914a; Woollard, 1924). In this regard Penfield (1923) described a case of internal hydrocephalus in a male child caused by what he interpreted as fluid accumulation in the subdural space. During treatment f h i d was simultaneously removed from the presumed subdural space and from the lumbar cistern. Subdural fluid was yellow to amber in color (CSF was clear), had a higher specific gravity, higher protein and nitrogen content than CSF, and contained more cells. Penfield (1923) concluded that subdural and subarachnoid fluids were of different compositions and that their simultaneous retrieval indicated that they occupied distinct, but separate, spaces. Along this same vein, Cushing (1914) had earlier noted that the products of ‘‘. . . suppuration . . .” could be frequently found in the SAS, but were not found in the subdural space. In a related study, Penfield (1924) fixed and froze heads of dogs, sawed them into slices, removed the brain slices with their arachnoid-pia and SAS containing frozen CSF, and then collected what was assumed to be frozen subdural fluid from the outer surface of the arachnoid and inner surface of the dura. These crystals, and resultant fluid, were light yellow in color and formed a “. . . drop or two to 1cc in dogs of 5 kilograms.” Penfield (1924)estimated the thickness of this ice crystal layer (and consequently the size of the subdural space) to be 0.5-1.0 mm. Subsequent to the detailed descriptions of Key and Retzius (1875-1876) and the embryological investigations of Weed (1916, 1917a) the subdural space was usually defined as a mesothelial-lined space (Weed, 1917b, 1920, 1938). The lining on the dural side was described as flattened cells and processes while that on the arachnoid side was either flattened or cuboidal. Leary and Edwards (1933) offered a counterpoint to this view. These authors compared scrapings of the pericardial, pleural, and peritoneal cavities with scrapings of the outer surface of the arachnoid and inner surface of the dura. Samples from the arachnoid-dura consisted of irregular populations of fibroblast-like cells (confirming the earlier comments of Mallory, 1920), while samples from the other cavities were obvious continuous mesothelial sheets quite similar to each other but quite different from the presumed lining cells of the subdural space. Leary and Edwards (1933) concluded that the inner dura was lined by fibroblasts, the outer arachnoid was not covered by mesothelial cells, and that it was “. . . evident that the subdural space does not correspond to other serous cavities.” Weed (1938) summarized the majority view of the

THE QUESTION O F A SUBDURAL SPACE

time by noting that a “. . . flattened polygonal mesothelium . . .” lined the inner surface of the dura, the outer surface of the arachnoid was covered by a mesothelium although i t was not always flattened, fluid was present in the intervening space [yellow in color-after Penfield (1923, 1924)], and that the arachnoid and dura “. . . constitute definitive membranes separated by a fluid-containing subdural space.” The text books of the time, in general, reflected this opinion. For example, Strong and Elwyn (1943) stated “Between the dura and the arachnoid is the capillary subdural space [emphasis theirs] filled with fluid. . . . It has no communication with the subarachnoid space.” A CONTEMPORARY VIEW OF THE MENINGES

This review is concerned primarily with whether or not a n organized cleft, the so-called subdural space (spatium subdurale, Nomina Anatomica, 1989), exists a t the dura-arachnoid junction. While admittedly a very small part of the overall anatomy of the human, this issue is directly related to well-known clinical manifestations seen following the collection of blood or other fluids in this area (e.g., Rosenberg, 1980; Rowland, 1984). Although attention is centered on the dura-arachnoid junction the geneal features of other parts of the meninges will be briefly considered. Dura Mater

The dura mater (Figs. 1, 2) is a thick layer of fibroblasts and extracellular collagen that is intimately adherent to the surface of the inner table of the skull. This attachment is especially tenacious in the base of the cranial cavity and along suture lines but somewhat less so in other areas (Allen and DiDio, 1977). Throughout its thickness the dura has flattened fibroblasts, large amounts of extracellular collagen that run in oblique directions as well as in planes which are perpendicular to each other, and few or no cell junctions (Pease and Schultz, 1958; Andres, 1967; Klika, 1967; Waggener and Beggs, 1967; Rhodin, 1974; Nabeshima e t al., 1975; Peters et al., 1991; Allen and DiDio, 1977; McLone, 1980; Orlin et al., 1991). Although they vary in shape depending on their location, fibroblasts of the dura contain essentially the same organelles; a Golgi apparatus, granular endoplasmic reticulum, free ribosomes, mitochondria, and occasional filaments, vesicles, or fat droplets. While there is not a distinct border between inner and outer parts of the dura there are some general features that characterize periosteal vs. meningeal portions. The periosteal dura (Fig. 1) is adherent to the surface of the inner table of the skull. It contains fibroblasts which are less elongate and have a moderately clear cytoplasm containing organelles, large oval nuclei with a modest amount of chromatin, and thick mats of interlaced collagen fibrils. The meningeal dura is internal to, but continuous with, the periosteal layer. Fibroblasts of this portion of the dura (Figs. 1, 2) have elongated processes that appear to be more distinctly arranged into layers, slightly darker cytoplasm, and oval shaped nuclei with a slightly more condensed chromatin. Small blood vessels are present in both portions of the dura. Although there are few morphologically distinct cell junctions in these layers, the occurrence of large amounts of interlacing collagen gives the

5

periosteal and meningeal layers of the dura great strength. Schachenmayr and Friede (1978-human) suggested that some cells of the inner (meningeal)dura may have intermediate junctions and that such cellto-cell connections decrease in number a s one moves outward within the dura. Waggener and Beggs (1967) and Schachenmayr and Friede (1978) have also reported that elastic fibers and some microfibrils may also be present in certain areas of the dura. Dural Border Cell Layer

Located internal to the meningeal dura is a unique population of modified fibroblasts that form what Nabeshima et al. (1975) called the dural border cell layer (Fig. 1). Waggener and Beggs (1967) referred to these cells a s “dense dural cells,” “cells . . . of the medial dural border,” or “bordering cells.” This layer is continuous externally with the meningeal dura and internally with the arachnoid (see below) and has morphological featureslrelationships that clearly differentiate it from these adjacent layers. Dural border cells (DBCs) are flattened and have long sinuous processes that may branch and interdigitate with each other (Figs. 1-4); these processes extend for considerable distances and form tiers of varying thicknesses (Nabeshima e t al., 1975; Schachenmayr and Friede, 1978, 1979; Yamashima and Friede, 1984; Alcolado e t al., 1988). Cells of essentially the same morphology have been described in the “superficial zone” (Rascol and Izard, 1976) or “outermost part” (Lopes and Mair, 1974a) of the arachnoid, at the “medial” border of the spinal dura (Waggener and Beggs, 1967), and specified as a layer of “subdural cells” (Akashi, 1972) or as a n “intermediate cellular layer” (Himango and Low, 1971) between the arachnoid and dura. Alcolado et al. (1988) specified this cell layer as the “subdural mesothelium” and these cells form what Orlin et al. (1991) have recently described as the “subdural compartment.” In spite of the different terms used, the descriptions given by these various investigators strongly suggest that all are referring to a comparable population of cells.’ This pattern of thin, flattened cell processes results in the creation of extracellular spaces of varying sizes and irregular shapes (Figs. 1-4) that contain a non-fibrillar material variably described a s flocculant, amorphous, grainy, granular, or fuzzy. This material does not have the morphological features of collagen or elastic fibers, and Akashi (1972) suggested that it may be a mucopolysaccharide. This pattern of sinuous attenuated cell processes sep’The general appearance of dural border cells and the somewhat exaggerated sizes of their extracellular spaces in human samples may be partially related to the inherent difficulties of fixing human material for electron microscopy. Furthermore, it is acknowledged that the method of preparationifixation may affect the relative translucency of dural border cells. On the other hand, animal material is usually well fixed (via perfusion) and the anatomical relationships more reliably seen. In spite of these variables, there is agreement that dural border cells in both human and animal samples are flattened, have long sinuous processes, are surrounded by extracellular spaces of variable sizes that contain a non-fibrillar material, and are continuous with the dura and occasionally attached to the arachnoid. Consequently, even when taking into account the fixation problems related to human material, the similarities of the dural border cell layer between human and animal samples are much more evident than are the differences.

6

D.E. HAINES

L

a,

s c

E

3

Q

V

L

Q

1

]Pi,

----- Elastic

Fibers

Basement Membrane

X

G a p Junctions Desmosomes

>r,

Tight Junctions Fig. 1

Muter

THE QUESTION OF A SUBDURAL SPACE

7

Fig. 2. Photomicrograph of the inner dura (D), the DBC layer, the ABC layer, and portions of the SAS, from a human female. The transitional zone (TZ) represents the point at which the DBC layer is continuous with the inner (meningeal) dura (see text). Note the characteristic appearance of the cells making up the DBC layer (compare

with Figs. 3, 4) and their close apposition to ABCs. Collagen (Col) is found free in the SAS as well as in close association with trabecular cells (AT). Some cell junctions (arrowheads) are evident in this view. X 12,400. (Reproduced from Schachenmayr and Friede, 1979, with permission of the publisher.)

Fig. 1 . Semidiagrammatic representation of the meninges emphasizing the main structural features of each layer. The large dots indicate areas where collagen is present and are not intended to specify the orientation or size of individual fibrils. In the normal situation collagen is oriented in many directions parallel to the long axes of the dura and collagen fibrils associated with trabecular cells are usually somewhat smaller than those of the dura. Note the characteristic appearance of the dura, the DBC layer, and the ABC layer. The DBC layer is continuous with the dura. Cells of this layer are also attached to each other and, occasionally, to the underlying ABC layer by cell junctions: there is no intervening cleft that would substantiate the existence of a subdural space. See text for additional discussion.

arated by an amorphous material in enlarged extracellular spaces is obvious in human material (Fig. 2) (Lopes and Mair, 1974a; Rascol and Izard, 1976; Schachenmayr and Friede, 1978, 1979; Yamashima and Friede, 1984; Alcolado et al., 1988), is present in the macaque (Fig. 3) (Nabeshima et al., 1975), and is seen to varying degrees in other animals (mice, rat, guinea pig, rabbit, pig, cat, dog) (Fig. 4) (Andres, 1967; Klika, 1967; Waggener and Beggs, 1967; Akashi, 1972; Nabeshima et al., 1975; Haines and Frederickson,

8

D.E. HAINES

Fig. 3. Photomicrograph of a small part of the dura (D), the DBC layer, the ABC layer, and the loosely-arranged portion of the arachnoid located internal to the ABC layer from a Macaque monkey. While originally labeled as “arachnoid by the authors, these latter structures are specified here as part of the arachnoid trabeculae (AT, see text). The DBC layer has flattened sinuous cell processes and extracellular spaces containing a n amorphousiflocculent material (small arrow; see also Figs. 2 , 4 ) . The ABC layer has tightly apposed cells with larger processes (see also Fig. 4), essentially no extracellu-

lar spaces, and in Macaque a “dense intercellular gap” (asterisks) is found in this layer. Note the segment of basement membrane (BM) on the inner aspect of part of the ABC layer. Structures interpreted as trabecular cell processes (AT) are attached to the barrier layer, irregular in their configuration, internal to the basement membrane, and they envelope collagen. Examples of cell junctions are clear in this low power view (arrowheads) and one tight junction is indicated by the large arrow, x 20,000. (Reproduced from Nabeshima et al., 1975, with permission of the publisher.)

1991; Orlin et al., 1991). Although they referred to the homologue of DBCs as the “subdural mesothelium” Pease and Schultz (1958-rat) noted that the processes of these cells were flattened but did not specifically

describe significant amounts of amorphous extracellular material in this area. Also, Oda and Nakanishi (1984-mouse)described this region of the meninges as composed of flattened closely-apposed cells with only

THE QUESTION OF A SUBDURAL SPACE

9

Fig. 4. Photomicrograph of the inner dura (D), the cells comprising the DBC layer, and the ABC layer in the guinea pig. The SAS with a trabecular cell (AT) is also shown. Note that the cytoplasm of the ABCs is electron-lucent when compared with the somewhat darker appearance of the cells of the DBC layer. Although DBCs in animals have processes that are less tortuous than in humans (compare with

Figs. 2, 7) they still tend to be flattened, somewhat sinuous, and sometimes electron dense. More importantly, the ABC and DBC layers are continuous with the inner parts of the dura and there is no evidence of a subdural space. x 27,500. (From Waggener and Beggs, 1967, with permission of the publisher.)

a n occasional narrow extracellular dilatation; in general this observation is similar to DBCs but these authors did not specifically designate such a cell layer. The continuity of the DBC layer with the meningeal dura (Figs. 1-3; see also Fig. 9), referred to as a transitional zone by Schachenmayr and Friede (1978) (Fig. 2), is characterized by varying amounts of the amorphous material of the DBC layer intermixed with some elastic fibers, the appearance of collagen fibrils in the meningeal dura (Schachenmayr and Friede, 1978, 19791, and the complete lack of a basement membrane a t this interface (Waggener and Beggs, 1967; Lopes and Mair, 1974a; Rascol and Izard, 1976; Yamashima and Friede, 1984). Consequently, the DBC layer is structurally continuous with the meningeal dura and is considered the innermost part of the dura (Fig. 1). Orlin et al. (1991) offer a n alternative interpretation by suggesting that this layer of flattened cells, the DBC layer of the present review, forms a separate meningeal layer designated by them a s the “subdural

compartment.” Collagen, elastic fibrils, and some microfibrils, while present at this junction (Fig. l ) , are noticeably lacking within the DBC layer itself. Klika (1967) reported some collagen between the “dark cells” of the outer arachnoid; based on their morphology these cells appear to be similar to DBCs. The cytoplasm and nuclei of DBCs (or of the homologue cell type) are sometimes described as being electron dense in preparations of human material (Figs. 1, 2) (Lopes and Mair, 1974a; Rascol and Izard, 1976; Schachenmayr and Friede, 1978, 1979; Yamashima and Friede, 1984; Alcolado et al., 1988)and in macaca (Klika, 1967) and rabbit (Akashi, 1972); this characteristic is less pronounced in many animal species especially in the face of optimal fixation (Fig. 3) (Pease and Schultz, 1958; Waggener and Beggs, 1967; Himango and Low, 1971; Nabeshima et al., 1975; see also Oda and Nakanishi, 1984; Haines and Frederickson, 1991; Orlin et al., 1991). (For comments related to fixation see footnote 1.)In spite of the dense appearance of

10

D.E. HAINES

the cytoplasm of these cells in some reports, organelles such as mitochondria, a small Golgi apparatus, granular endoplasmic reticular, ribosomes, microtubules, filaments, and occasional lipid droplets and small vesicles have been recognized (Andres, 1967; Akashi, 1972; Lopes and Mair, 1974a; Rascol and Izard, 1976; Schachenmayr and Friede, 1978). It should be noted that even in the absence of dark cytoplasm the morphological feature of these cells and their intervening extracellular spaces are distinctive of this layer. Although fewer in number, especially when compared with the arachnoid (see below), morphologically distinct cell junctions interconnect cells of the dural border layer with each other (Andres, 1967; Lopes and Mair, 1974a; Nabeshima et al., 1975; Rascol and Izard, 1976; Schachenmayr and Friede, 1978,1979; Yamashima and Friede, 1984) and occasional junctions have been reported between DBCs and cells of the underlying arachnoid (Akashi, 1972; Lopes and Mair, 1974a; Schachenmayr and Friede, 1978; Yamashima and Friede, 1984) (Figs. 1, 2). While usually described as being seen only occasionally, there is agreement that such junctions at both locations consist mainly of desmosomes and infrequently of gap junctions or intermediate junctions. Cells of the dural border layer do not appear to have organized junctional complexes with the fibroblasts of the meningeal dura, this continuum being characterized by an interweaving of an amorphous material and connective tissue fibers (Figs. 1,2). In summary, the DBC layer represents the innermost part of the dura; it is continuous with the meningeal dura, and is closely apposed to the underlying arachnoid. The layer itself is characterized by flattened cells with long processes, the presence of extracellular spaces containing a non-fibrillar amorphous material, a lack of extracellular connective tissue fibers, and the occurrence of some distinct cell-to-cell junctions within this layer and between this layer and the underlying arachnoid. Arachoid Barrier Cell Layer and the Arachnoid Trabeculae

The arachnoid portion of the meninges consists of two main parts (Figs. 1,3; see also Fig. 71, these being a layer of closely packed cells that is apposed to the DBC layer, the arachnoid barrier cell layer (so-named by Nabeshima et al., 1975), and an inner portion of loosely-arranged cells that are found in and sometimes bridge the SAS, the arachnoid trabeculae. Arachnoid barrier cells (ABCs), in contrast to cells of the DBC layer (Figs. 1-4), are larger, have closely-apposed processes that may form tiers of several cells thick, and have electron-lucent cytoplasm (Nabeshima et al., 1975; Schachenmayr and Friede, 1978,1979; Yamashima and Friede, 1984; Orlin et al., 1991). Although not specifically designated as ABCs (this term was introduced in 1975),other investigators have described cells of essentially identical morphology in this particular part of the arachnoid (Pease and Schultz, 1958; Andres, 1967; Klika, 1967; Waggener and Beggs, 1967; Himango and Low, 1971; Akashi, 1972; Lopes and Mair, 1974a; Rascol and Izard, 1976; Oda and Nakanishi, 1984; Alcolado et al., 1988). In addition to its characteristic translucency, the cytoplasm of ABCs contains numerous mitochondria (Figs. 3, 4; see also Fig. 9), a prominent Golgi apparatus, granular endoplasmic reticulum, free ribosomes, microtubules, filaments/tono-

filaments (many in association with desmosomes), and, infrequently, lipid droplets, lyosomes, and some small vesicles (Andres, 1967; Klika, 1967; Waggener and Beggs, 1967; Akashi, 1972; Rascol and Izard, 1976; Schachenmayr and Friede, 1978; McLone, 1980; Oda and Nakanishi, 1984; Yamashima and Friede, 1984; Alcolado et al., 1988; Orlin et al., 1991). The nuclei of these cells (Fig. 1) are also translucent, large, oval or elongate in shape, and there is tendency for the chromatin to be concentrated at the inner surface of the nuclear envelope (e.g., Andres, 1967; Waggener and Beggs, 1967;Schachenmayr and Friede, 1978; Alcolado et al., 1988; Orlin et al., 1991). In concert with the unique appearance of ABCs and the fact that their processes are closely apposed, is the occurrence of numerous morphologically distinct junctions between the cells of this layer (Fig. 1).It has been reported that ABCs may be joined to each other by numerous desmosomes (belt and spot), gap junctions, and tight junctions (Figs. 1, 5, 6); some intermediate junctions and hemidesmosomes may also be occasionally seen (Andres, 1967; Klika, 1967; Waggener and Beggs, 1967; Akashi, 1972; Tripathi, 1973; Lopes and Mair, 1974a,b; Nabeshima et al., 1975; Rascol and Izard, 1976; Schachenmayr and Friede, 1978, 1979; McLone, 1980; Drobrovol’skii, 1984; Oda and Nakanishi, 1984; Yamashima and Friede, 1984; Alcolado et al., 1988; Orlin et al., 1991). Although numerous, the sequencing of different types of junctions between cells of the arachnoid barrier layer appears to be random (Fig. 1).For example, there may be a series of desmosomes, a series of tight junctions (Fig. 5), or desmosomes may alternate with tight junctions along a given segment of apposed cell membranes. In any case, there does not appear to be a predictable repeating sequence of junctions between the cells of the arachnoid barrier layer. Oda and Nakanishi (1984-mouse) described an occasional tight junction between cells of the “inner” arachnoid; based on several morphological features these may be homologous to trabecular cells of other forms. There is, however, a consensus that tight junctions are unique to the ABC layer and are not found, to any significant degree, between cells in other parts of the meninges. Nabeshima et al. (1975) reported that the occurrence of many tight junctions (Fig. 5) between ABCs gave this cell layer its “barrier” characteristic and that these structures could serve to prevent the movement of large molecular weight substances (and possibly certain ions, see Bennett, 1969)out of the SAS. In addition to their close apposition to each other, cells of the arachnoid barrier layer were also apposed to the processes of overlying DBCs (Figs. 1-4; see also Fig. 9). While markedly fewer in number than that seen within the ABC layer, desmosomes, gap junctions, and intermediate junctions have been reported between ABCs and DBCs (Akashi, 1972; Lopes and Mair, 1974a; Schachenmayr and Friede, 1978, 1979; Yamashima and Friede, 1984). In this respect, the inner surface of the dura (the DBC layer) is not only continuous with, but is tenuously attached to (Fig. l),the outer surface of the arachnoid (the ABC layer). The close packing of ABC cell processes and the appearance of many cell junctions largely exclude the presence of any significant extracellular space in this layer (Figs. 1,3,4).Consequently, collagen and elastic fibrils are not usually found between the translucent

THE QUESTION OF A SUBDURAL SPACE

Fig. 5. Photomicrograph showing tight junctions (T)between cells of the arachnoid barrier layer (compare with Fig. 1) and a gap junction (G)between the barrier cell layer and a probable arachnoid trabecular cell in the rat. Although there are inherent difficulties in perfusion fixation of avascular membranes such as the ABC layer, this illus-

11

tration clearly shows the general relationships of cell processes and examples of cell junctions from this cellular layer. Note the lack of any significant extracellular space and the lack of extracellular connective tissue. ~60,000.(Reproduced from Nabeshima et al., 1975, with permission of the publisher.)

cells characteristic of the ABC layer (Waggener and These descriptions are remarkably similar to random Beggs, 1967; Nabeshima et al., 1975; Rascol and Izard, shapes of trabecular cells and indeed may actually rep1976; Schachenmayr and Friede, 1978, 1979; Oda and resent this cell population rather than cells of the Nakanishi, 1984; Yamishima and Friede, 1984; Orlin barrier layer. Arachnoid cells located immediately inet al., 1991). Authors that have described collagen in ternal to the ABC layer are loosely arranged and have relation to the arachnoid usually refer to this sub- thin extensions which surround bundles of collagen stance being present in portions of the arachnoid that (Figs. 1,3,7; see also Fig. 9). This particular part of the form the border of the SAS (e.g., Pease and Schultz, arachnoid has been variably described as its “inner 1958; Lopes and Mair, 1974a; Alcolado et al., 1988). layer,” “-aspect,” or “-part” (Lopes and Mair, 1974a;

12

D.E. HAINES

Fig. 6. Photomicrograph of a desmosome (rnaculu adherens-spot desmosome) between two cells in the arachnoid barrier layer of the monkey (Macaque). x 120,000. (Reproduced from Nabeshima et al., 1975, with permission of the publisher.)

Nabeshima et al., 1975; Oda and Nakanishi, 1984; Alcolado et al., 1988),or simply discussed in more general terms (Pease and Schultz, 1958; Andres, 1967; Waggener and Beggs, 1967; Himango and Low, 1971; Akashi, 1972; Rascol and Izard, 1976; Schachenmayr and Friede, 1978; Yamashima and Friede, 1984). Orlin et al. (1991) have recently suggested that this part of the arachnoid is a morphologically distinct region. Based on the branching pattern of these cells, the presence of collagen-containing lacunae, and the orientation of collagen fibrils, Orlin et al. (1991) have tentatively designated this part of the arachnoid as the “arachnoid reticular layer.” The evidence from the literature on animal and human meninges, however, suggests that these cells and their associated collagen, are directly continuous with, and may help to form, the trabeculae that bridge the SAS. Furthermore, the organelles, cell junctions, and general morphology of these cells are fundamentally the same as trabecular cells. Also, CSF found in the SAS undoubtedly bathes this loosely arranged portion of the arachnoid. In view of these morphological characteristics, the present review considers this part of the arachnoid as those outer portions of the trabeculae that are in continuity with the ABC layer and does not designate it as a separate layer (Fig. 1). Future studies will undoubtedly offer additional insights concerning this point. While many investigators have described an electron dense line or material a t some locations between ABCs and DBCs (e.g., Andres, 1967; Waggener and Beggs, 1967; Morse and Low, 1972), only Lopes and Mair (1974a) have suggested that an “interrupted” basement membrane may be present a t this point. Confirmation of a basement membrane between the ABC and DBC layers has not been provided in subsequent studies. The internal aspect of the ABC layer, that being the portion which borders on the SAS, is characterized by the presence of a basement membrane, trabecular cells of the SAS which attach to the ABC layer, and the occurrence of collagen usually in close relation to trabecular cells (Figs. 1, 3, 7). In humans, a basement

membrane is found on that portion of the arachnoid barrier layer that borders on the SAS (Rascol and Izard, 1976; Schachenmayr and Friede, 1978,1979; Yamashima and Friede, 1984); Lopes and Mair (1974ahuman) describe this membrane as “interrupted.” In animals, this basement membrane (Fig. 3) has been variably described as either relatively continuous or somewhat incomplete (Andres, 1967; Klika, 1967; Akashi, 1972; Nabeshima et al., 1975; Oda and Nakanishi, 1984). In either case (human or animal) there seems to be a consensus that a basement membrane is present on the internal surface of the ABC layer or its homologue. Processes of arachnoid trabecular cells (Figs. 1,3) are attached to ABCs by gap junctions, desmosomes, intermediate junctions, and possibly some hemidesmosomes (Lopes and Mair, 1974a; Nabeshima et al., 1975; Schachenmayr and Friede, 1978,1979; Alcolado et al., 1988). As they attach to the ABCs, trabecular cell processes perforate the basement membrane which, in turn, tends to form “cuffs” around such attachments (Fig. 1)(Andres, 1967; Schachenmayr and Friede, 1978, 1979). The arachnoid trabeculae are composed of modified fibroblasts with flattened, tortuous processes that bridge the SAS in a random fashion (Figs. 1-4,7). Trabecular cell processes, in addition to their association to ABCs, are also attached to each other, as well as to pial cells located on the surface of the brain or extending over vessels (Figs. 1, 3, 7), by desmosomes, gap junctions, and some intermediate junctions (Andres, 1967; Nabeshima et al., 1975; Schachenmayr and Friede, 1978, 1979; Oda and Nakanishi, 1984; Alcolado et al., 1988). The shapes and orientations of trabecular cells, as they cross the SAS, vary widely. While some may form thin strands of relatively few processes, others may form larger “filiform” or “sheet-like” structures (Figs. 2-4, 7) (Pease and Schultz, 1958; Schachenmayr and Friede, 1978, 1979; Alcolado et al., 1988; Nicholas and Weller, 1988). These more organized condensations of leptomeningeal processes coalesce to form the boundaries of most of the cisterns of the SAS (Yasargil et al., 1976; Yasargil, 1984). Allen

THE QUESTION OF A SUBDURAL SPACE

13

Fig. 7.Photomicrograph of the SAS from a human female, showing arachnoid trabeculae (AT) and collagen (Col) free in the SAS and associated with folds in the membranes of trabecular cells. A small part of the ABC layer and the contiguous DBC layer are seen in the upper right; note the lack of an intervening space. In this low power

view cell junctions can be recognized between DBCs, ABCs, and AT (arrowheads). Note that the arachnoid trabeculae may be formed by the apposition of the process of more than one trabecular cell. x 12,400. (Reproduced from Schachenmayr and Friede, 1979, with permission of the publishers.)

and DiDio (1977) have suggested that there are three types of trabecular (leptomeningeal) processes: intermembranous processes that span the SAS between pia and arachnoid (barrier cells), intertrabecular processes that extend between trabeculae, and vasomembranous processes that extend between blood vessels and the pia andlor arachnoid. Extracellular collagen is closely associated with the trabecular cells of the SAS (Figs. 1, 2, 7). While these extracellular fibrils may be found free in the SAS adjacent to trabecular cells, they are more frequently surrounded by thin folds of trabecular cell processes (Pease and Schultz, 1958; Nelson et al., 1961; Andres, 1967; Klika, 1967; Waggener and Beggs, 1967; Nabeshima et al., 1975; Rascol and Izard, 1976; Lopes and Mair, 1974a; Schachenmayr and Friede, 1978,

1979; Oda and Nakanishi, 1984; Yamashima and Friede, 1984; Alcolado et al., 1988). Also, the SAS may be partially obliterated at some points by a close apposition of the ABC layer with the pia mater. In such regions the trabeculae and their collagen may be compacted between these two layers and a recognizable SAS obliterated. In summary, the arachnoid consists of a layer of closely packed cells (the ABC layer) that is continuous with the dural border layer, and an inner part of more loosely-arranged cells that contribute to the formation of the trabeculae that bridge the SAS. Barrier cells are characterized by their translucent cytoplasm and nuclei, tightly arranged cell processes, the occurrence of many cell junctions (desmosomes, and gap and tight junctions), little or no extracellular space, essentially

14

D.E. HAINES

no extracellular connective tissue fibers, and the presence of a basement membrane on their internal surface. Tight junctions appear to be unique to this layer. Trabecular cells attach to the internal surface of the barrier layer and to pial cells located on the surface of the brain or those covering vessels in the SAS, frequently have flattened processes, are tortuous in their configuration, and usually have collagen either close to cells in the SAS or enclosed in thin folds of trabecular cell processes. Pia Mater

The pia mater is composed of flattened cells with long processes that closely follow the surface of the brain and spinal cord and may, in association with trabecular cells, also surround vessels in the SAS (Fig. 1). Pial cells are basically modified fibroblasts that are quite similar in their general morphology to arachnoid trabecular cells. Their cytoplasm is relatively translucent and contains some mitochondria, granular endoplasmic reticulum, free ribosomes, a small Golgi apparatus, and occasional vesicles; the nuclei of these cells are small and oval with peripherally located chromatin. In addition, the processes of pial cells can be very long and may variably form more than one layer, a single layer, or be discontinuous at certain points thereby exposing the basement membrane on the surface of the neural tissue to CSF (Pease and Schultz, 1958; Nelson et al., 1961; Andres, 1967; Waggener and Beggs, 1967; Morse and Low, 1972; Lopes and Mair, 197413; Nabeshima et al., 1975; Cragg, 1976; Allen and Didio, 1977; Schachenmayr and Friede, 1979; McLone, 1980; Hutchings and Weller, 1986; Alcolado et al., 1988). While the processes of pial cells may closely appose each other, morphologically distinct cell junctions have not been frequently reported. When identified, they are described as intermediate junctions, gap junctions, desmosomes, and possibly an occasional tight junction (Waggener and Beggs, 1967; Schachenmayr and Friede, 1979; Dobroval’skii, 1984; Hutchings and Weller, 1986; Alcolado et al., 1988). On the other hand, there is wide-spread agreement that pial cell processes are separated from the glial surface of the neural tissue by a basement membrane (Fig. 1) and that pial cells may either touch this glial basement membrane or be separated from it by a small (subpial) space containing variable amounts of collagen (Pease and Schultz, 1958; Andres, 1967; Waggener and Beggs, 1967; Morse and Low, 1972; Lopes and Mair, 1974b; Nabeshima et al., 1975; Cragg, 1976; Schachenmayr and Friede, 1979; Dobrovol’skii, 1984; Oda and Nakanishi, 1984; Alcolado et al., 1988). This structural relationship of pial cells to neural surface (a small space containing connective tissue fibrils plus a basement membrane) is markedly similar to that found a t the interface of leptomeningeal cell processes with the smooth muscle cells making up the walls of vessels in the SAS. At this latter point the leptomeningeal process is separated from the smooth muscle cell by a small extracellular space that contains the basement membrane of the smooth muscle cell and may contain connective tissue fibrils (Frederickson and Low, 1969). In addition to collagen, small vessels or occasional macrophages may be found between the glial basement membrane and the

overlying pial cell processes (Morse and Low, 1972; Nabeshima et al., 1975; Cragg, 1976). A POTENTIAL SPACE?

As reviewed above there is compelling evidence to support the view that, in the normal condition, the cells of the arachnoid barrier layer appose and are occasionally attached to the cells of the dural border layer; there is no intervening (subdural) space. There is, in fact, a continuum of cellular layers, some with variable amounts of extracellular material, from the inner surface of the arachnoid barrier cell layer to the point where the periosteal dura is attached to the inner table of the skull. Evidence of a thin fluid-filled and mesothelial-lined cleft that would correlate with the subdural space (spatium subdurale, Nomina Anatomica, 1989) is lacking. A representative survey’ of the extant teaching literature (Table 1) reveals that the majority (36/42) of textbooks and atlases in gross anatomy, histology, and neuroscience state categorically that a subdural space is present between the arachnoid and dura. In a minority of this sample (6/42) the probable existence of a subdural space is clearly qualified or specifically denied (Table 1). Two points mentioned in at least half (17-19/36) of those texts which describe a subdural space are, first, that it is a “potential” space and, second, that it may contain a small amount of fluid. In view of the fact that this cleft is frequently described as a “potential subdural space” (Table 1)it is appropriate to consider more precisely what may be meant by the term “potential space” and to arrive at a definition that could be more broadly applicable and useful. A true potential space is one that may be created without disrupting the normal structure/functional integrity of the tissues involved in the creation of the space (Fig. 8A,B). For example, the pleural and peritoneal cavities are frequently described as “potential spaces” and indeed they are true potential spaces. These cavities are lined by a mesothelium which is covered by small amounts of a serous fluid. When structures covered by visceral pleura (or peritoneum) come in direct contact with structures covered by parietal pleura (or peritoneum) the “space” of the cavity is obliterated and the parietal and visceral mesothelial layers abutt against each other (Fig. 8A,B). This obliterated space may subsequently be enlarged, for example, if the lung collapses, without damaging the mesothelium or its subjacent connective tissue base. Indeed, these spaces may be repeatedly created and obliterated without resulting in tissue damage or requiring tissue repair (Fig. 8A,B). Consequently, the interface of the visceral and parietal layers represents a true “potential” space (Fig. 8B). This is clearly not the case as regards the formation of the so-called subdural space. On the other hand if the normal structural relationships of the tissues involved in the creation of a space are violated by the disruption of cell junctions, artificial enlargment of extracellular spaces, or thc probable tearing of cell membranes, then the space created is ‘This sample of extant textbooks and atlases is meant to be representative of the three general areas of anatomy and is not intended as a n exhaustive survey.

15

T H E QUESTION OF A SUBDURAL SPACE

TABLE 1. Summary of comments concerning, or descriptions of, the subdural space from contemporary textbooks or atlases in gross anatomy (GA), histology (H), and neuroscience (NS). The listings are in chronological and alphabetical order by the year in each discipline beginning with the most recent dates Akesson et al., 1990

GA

Hall-Craggs, 1990

GA

Basmajian and Slonecker, 1989

GA

Leeson and Leeson, 1989

GA

Lindner, 1989

GA

Netter, 1989 Platzer, 1989 Williams et al., 1989

GA GA GA

9 Berkovitz and Moxham, 1988

GA

10 Rohen and Yokochi, 1988 11 Stern, 1988 12 Woodburne and Burkel, 1988

GA GA GA

13 Clemente, 1987 14 Liebgott, 1986

GA GA

15 ORahilly, 1986

GA

16 Romanes, 1986

GA

17 Snell, 1986

GA

18 Moore, 1985

GA

19 Hollinshead and Rosse, 1985

GA

20 Telford and Bridgman, 1990 21 Junqueira et al., 1989 22 Ross et al., 1989 23 Amenta, 1987 24 Cormack, 1987 25 Wheater and Burkitt, 1987

H

26 Fawcett, 1986

H

27 Geneser, 1986

H

“Subdural space. This lies between the dura and the arachnoid membranes. . .” + illustration. [The arachnoid] “. . . is separated from the dura mater by the subdural space . . .” “The subdural space exists between the meningeal dura and the arachnoid. Bleeding by cerebral veins . . . may expand this potential space.” “The arachnoid is separated from the dura by a capillary subdural space that contains a thin film of tissue fluid.” + illustration. “The subdural space lies between the dura mater and the arachnoid.” [It] “. . . is ordinarily filled with small amounts of lymph-like material . . .” “Subdural space” in illustrations. “Subdural space” in illustrations. “The subdural space between the dura and arachnoid maters is a potential space containing a film of serous fluid between the surfaces of apposed membranes.” [It] “. . . is traversed by isolated trabeculae . . .” “The arachnoid mater closely lines the dura mater, being separated from it by a potential space called the subdural space. This space contains a thin film of serous fluid and is probably in continuity with the lymph spaces of cranial and spinal nerves.” “Subdural space” in illustrations. “ . . . the subdural space (between dura and arachnoid) . . .” “Its inner surface is lined by squamous cells and is separated from the spinal arachnoid membrane by a capillary interval, the subdural space. This contains a film of fluid . . . to moisten the apposed surfaces of the membranes.” + illustration. “Subdural space” in illustrations. “Subdural space. This is another potential space immediately below the dura. It is normally obliterated by pressure of the underlying arachnoid layer.” “The arachnoid . . . is seDarable from the dura bv a Dotential subdural space . . .” “The arachnoid is seDarated from the dura mater bv a bursa-like. capillary space (th; subdural space) containing a“fi1m of fluid. ’ This forms a sliding plane where movement is possible . . .” + illustrations. [The arachnoid] “. . . is separated from the dura by a potential space, the subdural space . . .” “The subdural space is deep to the dura mater, i.e., between the dura and arachnoid. It is also a potential space with only a thin film of subdural fluid in it. It is not obliterated except at places where it is pierced by arteries, veins, and nerves.” “The smooth outer surface of the arachnoid lies against the inner surface of the dura, separated from it only by a slitlike subdural space that contains enough fluid to keep the adjacent surfaces moist.” + illustration. “The arachnoid is a delicate . . . thin, connective tissue component in contact with the dura . . .” “The dura is always separated from the arachnoid by a thin space, the subdural space.” “ . . . the arachnoid mater, is attached to the dura mater . . .” [The dura] “. . . is separated from the arachnoid by the very narrow subdural space.” + illustration. “The potential space between its [dura] inner surface and the outer surface of the arachnoid is called the subdural space . . . it normally contains a slight amount of fluid . . .” + illustration. “The dura is closely applied to, but not connected with, the arachnoid layer and a potential space, the subdural space, containing a minute amount of fluid separates the two layers.” + illustration. “Between the dura mater and the arachnoid, the subdural space is a serous cavity containing a minimum of fluid: actually, it is scarcely more than a potential space.” + illustration. “The deep surface [of the dura] is separated from the arachnoid by the subdural space, which contains fluid.” + illustration. (continued) I



-

16

D.E. HAINES

TABLE 1. Summary of comments concerning, or descriptions of, the subdural space from contemporary textbooks or atlases in gross anatomy (GA), histology (H), and neuroscience (NS). The listings are in chronological and alphabetical order by the vear in each discipline beginning with the most recent dates (continued) 28 Martin, 1989

NS

29 Barr and Kiernan, 1988

NS

30 deGroot and Chusid, 1988

NS

31 Montemurro and Bruni, 1988

NS

32 Neiuwenhuys et al., 1988 33 Nolte, 1988

NS NS

34 Guyton, 1987

NS

35 Haines, 1987

NS

36 Kiernan, 1987

NS

37 Snell, 1987

NS

38 Daube et al., 1986

NS

39 Noback and Demarest, 1986

NS

40 Romero-Sierra, 1986

NS

41 Kandel and Schwartz, 1985

NS

42 Carpenter and Sutin, 1983

NS

“The arachnoid mater adjoins but is not tightly bound to the dura mater, thus allowing a potential space to exist between them.” “The avascular arachnoid is separated from the dura mater by a film of fluid in the subdural space.” “The dura . . . is separated from the thin arachnoid by a potential compartment, the subdural space which normally contains but a few drops of cerebrospinal fluid.” “The arachnoid mater is a delicate membrane . . . closely adherent to the inner layer of the dura mater being separated from it by a narrow subdural space. This space, which contains a thin film of fluid, is in reality more of a potential space than a real one.” “Spatium subdurale” in illustrations. “No space exists on either side of the dura . . . one side is attached to the skull and the other side adheres to the arachnoid. However, two potential spaces, the epidural and subdural spaces are associated with the dura.” “The arachnoid is a delicate structure loosely attached to the inner surface of the dura mater.” [The arachnoid is] “. . . attached to dura in living condition (no actual subdural space) . . . only a potential one.” “The arachnoid is applied to the inside of the dura mater.” “Potential subdural space” on illustration. “The arachnoid . . . is separated from the dura by a potential space, the subdural space, filled by a film of fluid . . .” “ . . . beneath the dura mater is the subdural space.” [The] epidural and subdural spaces are potential spaces in which illustration. blood or pus may accumulate.” “The subdural space is the potential thin space located between the inner dura mater and the arachnoid. The film of fluid in the subdural space is not cerebrospinal fluid.” “The dura apposes to the arachnoid but does not attach itself; between these layers there is what is called a potential subdural space . . .” + table. “The arachnoid mater . . . adjoins but is not tightly bound to the dura mater, thus allowing a potential space to exist between them. This potential space, called the subdural space, is important clinically.” “This layer [innermost dura] is in close contact with the arachnoid, suggesting that the subdural space is a potential space rather than an actual space.” illustration.

+

+

artifactuallpathological in its origin and not truely “potential” in nature (Figs. 8C, 9A,B). In other words, the space is not preexisting but is formed by abnormal forceslconditions that separate constituent parts of the tissue involved. This is clearly the process of tissue shearing (Figs. 8C, 9A,B) that takes place in the development of what has been called the subdural space. In this regard Orlin et al. (1991) have offered experimental evidence that blood infused into the area of the dura-arachnoid junction invades the cellular layer identified by them as the “subdural compartment” (DBC layer of this review). These authors report that erythrocytes and platelets, while sometimes being found between the dura and DBCs or between the DBC and ABC layers, were most frequently found insinuated among the cells of the DBC layer (their subdural compartment). Under their experimental conditions Orlin et al. (1991) reported that blood was not only found among DBCs (subdural compartment) but further noted that these meningeal cells also appeared modified (thinner, irregularly shaped), had widened intercellular spaces, and frequently separated the underlying ABC layer or the overlying dura from the infused

blood. In addition, Waggener and Beggs (1967, their Fig. 21, Nabeshima et al. (1975, their Fig. 51, and Yamashima and Friede (1984) (Fig. 9A,B) have reported, in both animals and human samples, that a so-called subdural space may occasionally appear as a preparation artifact. These authors noted that, in such instances, this space appeared in response to a splitting open of the DBC layer (Fig. 9A,B) and not as separation of the DBCs from either the dura or from the arachnoid (ABC layer). It is, however, important to note that even in the face of such technical problems the morphological characteristics of DBCs and ABCs are fundamentally the same as when the DBC layer is maintained intact (Fig. 9B, compare with Fig. 2). The appearance of the so-called subdural space is the result of tissue damage and is not due to the expansion of a patent (or temporarily obliterated) preexisting space. Consequently, this space is neither “actual” nor “potential”; it is, in fact, non-existent in the normal situation. The occurrence of few cell junctions’ lack of extracellular connective tissue fibers, and presence of extracellular spaces of variable sizes between DBCs (Fig. 1) makes this particular layer the weak plane

THE QUESTION OF A SUBDURAL SPACE

17

the result of a disruption of cell junctions, possible damage to cell membranes, and subsequent enlargement of extracellular spaces within the layer composed of dural border cells. In this regard ORahilly and Miiller (1986) have noted that this separation is “ . . . intradural rather than subdural . . .”. An extension of this reasoning would suggest that subdural hematomas are actually intradural hematomas. Unfortunately this terminology (intradural) would imply that such tissue separations may take place a t any point within the dura; this is certainly not the case. While the collection of blood or fluids in pathological/traumatic situations (e.g., Rosenberg, 1980; Rowland, 1984) does not take place into a preexisting space, it does take place primarily into a disrupted DBC layer and in doing so creates a space where none existed. Consequently, so-called subdural hematomas are actually dural border hematomas. This latter terminology precisely identifies the location of the lesion within a specific morphological portion of the meninges. CAPSULE OF THE “SUBDURAL HEMATOMA’

Meninaeal Dura

Fig. 8. Semidiagrammatic representation of the formation of a true potential space (A,B) vs. the process of tissue damage that results in what has classically been called a “subdural” hematoma. A true potential space can be alternately formed and obliterated without tissue damage (A,B). On the other hand, the formation of a so-called ‘‘subdural” hematoma is the result of forces (large arrow) which shear open the DBC layer; cell junctions and probably cell membranes are damaged and the extracellular spaces of the DBC layer are opened up. Note that the lesion (large arrow in C) progresses into the DBC layer and that some of these cells may be found external to the lesion in association with the dura and others found internal to the lesion in association with the ABC. See text for additional discussion.

of the meninges (Nabeshima et al., 1975; Rascol and Izard, 1976; Friede and Schachenmayr, 1978; Schachenmayr and Friede, 1978, 1979; Yamashima and Friede, 1984; Orlin et al., 1991). In contrast, the meningeal dura with its mats of extracellular collagen, and the arachnoid barrier layer with its closelyarranged cells joined by many junctions and its subjacent collagen-containing trabeculae are, when compared with the DBC layer, more resistant to tearing. Studies in animals (Waggener and Beggs, 1967; Nabeshima et al., 1975; see also Orlin et al., 1991) and in humans (Fig. 9A) (Rascol and Izard, 1976; Friede and Schachenmayr, 1978; Schachenmayr and Friede, 1978, 1979; Yamashima and Friede, 1984) have concluded that the creation of a subdural space is actually

If, as has been suggested (Waggener and Beggs, 1967; Nabeshima et al., 1975; Rascol and Izard, 1976; Friede and Schachenmayr, 1978; Schachenmayr and Friede, 1978, 1979; Yamashima and Friede, 1984; Orlin et al., 1991), the formation of a so-called subdural hematoma is the result of a shearing open of the DBC layer (Figs. 8C, 9A); evidence of the cell type characteristic of this layer should be seen in the capsule surrounding this lesion in human material. There is general agreement that the capsule of subdural hematomas consists of an outer thicker portion adjacent to the dura and an inner thinner part that is apposed to the ABC layer (Apfelbaum et al., 1974; Friede and Schachenmayr, 1978; Yamashima and Yamamoto, 1985; Kawano et al., 1988; Yamashima et al., 1985, 1989). The outer membrane may be vascularized by proliferating capillaries from the dura and is usually composed of a variety of cells including fibroblasts, mast cells, eosinophils, myofibroblasts, and some cells which at the electron microscopic level resemble smooth muscle cells (Sato and Suzuki, 1975; Kawano and Suzuki, 1981; Yamashima and Yamamoto, 1984; Kawano et al., 1988; Yamashima et al., 1985,1989). In addition to elongated fibroblasts, the outer part of the capsule contains some flattened cells with long sinuous processes that are remarkably similar to DBCs (Friede and Schachenmayr, 1978; Yamashima and Friede, 1984; Kawano et al., 1988). The inner portion of the capsule has few or no capillaries, many slender spindle shaped fibroblasts, and many other cells with long, flattened, branching and interdigitating processes (Apfelbaum et al., 1974; Friede and Schachenmayr, 1978; Yamashima and Yamamoto, 1985; Kawano et al., 1988).These latter cells share many morphological similarities with those of the DBC layer and have been specified as that particular cell type (Friede and Schachenmayr, 1978; Yamashima and Friede, 1984; Yamashima and Yamamoto, 1985; Kawano et al., 1988). Friede and Schachenmayr (1978) noted that tight junctions were lacking between the cells forming inner and outer portions of the capsule but that intermediate junctions, desmosome-like junctions, and an occasional gap junction were present (see also Yamashima and Friede,

18

D.E. HAINES

Fig. 9.

THE QUESTION OF A SUBDURAL SPACE

1984; Kawano et al., 1988). Tight junctions are also lacking between cells of the DBC layer in normal samples. In addition, DBCs found in inner and outer portions of the “subdural” hematoma capsule have an enlarged granular endoplasmic reticulum, more free ribosomes, and numerous mitochondria (Friede and Schachenmayr, 1978; Yamashima and Friede, 1984; Yamashima and Yamamoto, 1985) when compared with normal specimens. This evidence of a heightened state of cell activity explains, at least partially, the extracellular fibrillar substance (collagen, elastic fibrils, microfilaments) found among cells forming the inner and outer aspects of the capsule (Sato and Suzuki, 1975; Friede and Schachenmayr, 1978; Kawano and Suzuki, 1981; Yamashima and Friede, 1984; Yamashima and Yamamoto, 1985). While present around DBCs in the pathological state, such extracellular material is lacking in this particular cell layer in the normal condition (see above). Data from studies on the capsule of “subdural” hematomas from human material support the view that this lesion is formed within the DBC layer but is accompanied by a proliferation of fibroblasts, DBCs and capillaries, an invasion of other cell types, and the deposition of connective tissue fibers. Friede and Schachenmayr (1978) have suggested that the inner and outer portions of the capsule of the “subdural” hematoma “ . . . correspond to an excessively proliferated and thickened layer of dural border cells,” while Kawano et al. (1988) have concluded that “. . . the subdural hematoma . . . is formed within the split dural border cell layer.” The exprimental data of Orlin et al. (1991) also support this fundamental concept. The available evidence from studies on normal animal and human meninges, from investigations on pathological specimens, and from animal experiments, supports the view that the so-called “subdural” hematoma is not subdural in its location but is mainly formed by a splitting open of the DBC layer. CONCLUSIONS

Two points merit comment. First, the available evidence supports the view that a natural biological space is not present at the dura-arachnoid junction. Second, when a space does appear in this region it is the result of pathological/traumatic processes that result in tissue damage. Consequently, there is no substantive contemporary data to support the position that a subdural space exists in the normal state. Furthermore, to refer to the subdural space as a “potential space” is a misnomer; it is certainly not comparable, in any sense to bona fide “potential” spaces (e.g., pleural, peritoneal, etc.) found at other locations in the body. In addition, with the exception of small amounts of amorphous ex-

19

tracellular material between cells of the dural border layer, there is no evidence of a fluid plane at the duraarachnoid interface. Such fluid is undoubtedly the product of local tissue reaction when their constituent parts are sheared. While it is not possible to ascertain how the conclusion was reached that the subdural space was lined by a mesothelium, the structure and characteristics of the DBC layer itself offer a clue. When this layer is cleaved open (e.g., Waggener and Beggs, 1967; Nabeshima et al., 1975; Friede and Schachenmayr, 1978; Schachenmayr and Friede, 1978; Yamashima and Friede, 1984; Orlin et al., 1991) flattened DBCs may remain attached to the inner surface of the meningeal dura and to the outer surface of the arachnoid barrier layer (Figs. 8C, 9A,B). In light microscopic preparations as well as in electron microscopy this could give the impression of a space lined on either side by flattened cells with long thin processes; such a relationship is reminiscent of a narrow mesothelial lined cavity. The idea that such a cavity contained a small amount of fluid was not only based on historical reports, but was a natural extension of the view that this space had a mesothelial lining. Although a number of names have been proposed or used, the terms “dural border cells”3 and “arachnoid barrier cells” appear to be established descriptors of those cells found at the dura-arachnoid interface and are logical and appropriate extensions of parent terms (i.e., dura, arachnoid) that are widely used and universally accepted. Furthermore, there is abundant evidence (reviewed above) which documents the existence of such cells, gives details of their individual and characteristic morphology, and describes the relationships of the meningeal layers formed by these cells. While it has been suggested the ABC layer and the DBC layer collectively form the “interface layer” (e.g., Schachenmayr and Friede, 1978), the application of an additional term to this particular area of the meninges seems unncessary. Such a term provides no additional insights into, or explanations of, the structural/functional characteristics of these layers. The available data indicate that a subdural space does not exist within the meninges. When a space is created in this general area of the meninges, it is, first and foremost, artifactualtpathological in its genesis and, second, not in a subdural location but located primarily within a structurally weak plane represented by the DBC layer. It is suggested that the term spatium subdurale be dropped from Nomina Anatomica (1989) as there is no evidence to support the view that it is a legitimate and naturally occurring space in the body. ACKNOWLEDGMENTS

Fig. 9. Photomicrographs of the inner dura (D), the DBC layer, the ABC layer, and what are interpreted as outer portions of some arachnoid trabeculae (AT) from humans. A: In this specimen a split occurred within the DBC layer during preparation (large arrow). In the detail (B) note that thin processes of DBCs remained on the outer surface of the ABC layer and the inner surface of the dura (arrowheads). Even though an artifactual space is present in the DBC layer, the characteristic appearance of DBC possesses, the adjacent dura, and the ABC layer is obvious (9A,B, compare with Fig. 2). Scale = 1 pm. (Reproduced from Yamashima and Friede, 1984, with permission of the publisher.)

I am indebted to Drs. John Beggs (Barrows Neurological Institute), Reinhard L. Freide (Gottingen), 31t should be noted that Nabeshima et al. (1975) coined the specific term “dural border cells”. However, Waggener and Beggs (1967) had earlier referred to the identical population of meningeal cells as “dural cells”, “cells . . . of the medial dural border” or “bordering cells”. Based on the marked similarity of the terms applied to this part of the meninges by these investigators it would appear that the term “dural border cells” has priority over other contemporary designations.

20

D.E. HAINES

George F. Martin (Ohio State), Keith L. Moore (Toronto), and Richard G. Frederickson (University of Washington and Bastyr College) for their helpful comments and suggestions. My colleagues in the Department of Neurosurgery at the University of Mississippi (Drs. Ossama Almefty and Orlando Andy) have also offered insights. In extensive conversations concerning the general topic of the subdural space, Dr. Kirsten Osen (Oslo) has provided valuable insights that have influenced this paper. I would also like to thank Drs. Beggs, Friede, Milton Brightman, and Tetsumori Yamashima for supplying me copies of previously published photographs and the publishers of the journals from which I have borrowed photographs for their permission to do so. Dr. Yoshihiro Yamamoto, a Fellow in Neurosurgery (at the University of Mississippi), translated Japanese articles for me; I greatly appreciate his kindness. Mr. Larry Bird gave valuable editorial assistance, Mr. Tim Vickmark and Mr. Glen Hoskins did the photography, Ms. Gail Rainer typed the manuscript, and Ms. Janet Brown typed the table.

Frederickson, R.G., and F.N. Low 1969 Blood vessels and tissue space associated with the brain of the rat. Am. J . Ant., 125t123-146. Friede, R.L., and W. Schachenmayr 1978 The origin of subdural neomembranes. 11. Fine structure of neomembranes. Am. J . Pathol., 92t69-84. Geneser, F. 1986 Textbook of Histology. Lea & Febiger, Philadelphia, pp. 323-324. Guyton, A.C. 1987 Basic Neuroscience: Anatomy and Physiology. W.B. Saunders Company, Philadelphia, p. 33. Haines, D.E. 1987 Neuroanatomy: An Atlas of Structures, Sections, and Systems. 2nd Edition. Urban & Schwarzenberg, Baltimore, pp. 41-42. Haines, D.E., and R.G. Frederickson 1991 The meninges. In: Meningiomas. 0. Al-Mefty, ed. Raven Press, New York, pp. 9-25. Hall-Craggs, E.C.B. 1990 Anatomy a s a Basis for Clinical Medicine. Urban & Schwarzenberg, Baltimore, p. 465. Himango, W.A., and F.N. Low 1971 The fine structure of a lateral recess ofthe subsarachnoid space in the rat. Anat. Rec., 171tl-20. Hollinshead, W.H., and C. Rosse 1985 Textbook of Anatomy. 4th Edition. Harper & Row, Philadelphia, pp. 921-926. Hutchings, M., and R.O. Weller 1986 Anatomical relationships of the pia mater to cerebral blood vessels in man. J . Neurosurg., 65; 316-325. Junqueira, L.C., J. Carneiro, and R.O. Kelley 1989 Basic Histology. 6th Edition. Appleton & Lange, Nonvalk, pp. 190-191. Kandel, E.R., and J.H. Schwartz 1985 Principles of Neural Science. 2nd Edition. Elsevier, New York, p. 247. Kawano, N., M. Endo, M. Saito, and K. Yada 1988 Origin of the LITERATURE CITED capsule of chronic subdural hematoma, a n electron microscopic study. No Shineki Geka. 16:747-752 (Original in Japanese). Akashi, Y. 1972 Electron microscope studies on the fine structure of Kawano, N., and K. Suzuki 1981 Presence of smooth-muscle cells in the arachnoid membrane on the base of the rabbit brain. Acta the subdural neomembrane. J . Neurosurg., 54t646-651. Anat. Nippon., 47t285-297 (Original in Japanese). Key, A,, and G. Retzius 1875-1876 Studien in der Anatomie des Akesson, E.J., J.A. Loeb, and L. Wilson-Pauwels 1990 Thompson's Nervensystems und des Bindegewebes. Samson and Waller, Core Textbook of Anatomy. 2nd Edition. J.P. Lippincott ComStockholm. pany, Philadelphia, pp. 339-342. Alcolado, R., R.O. Weller, E.P. Parrish, and D. Garrod 1988 The cra- Kiernan, J.A. 1987 Introduction to Human Neuroscience. J.B. Lippincott Company, Philadelphia, pp. 211-215. nial arachnoid and pia mater in man: Anatomical and ultrastrucKlika, E. 1967 The ultrastructure of meninges in vertebrates. Acta tural observations. Neuropathol. Appl. Neurobiol., 14tl-17. Univ. Carol. [Medl (Praha), 13~53-71. Allen, D.J., and L.J.A. DiDio 1977 Scanning and transmission electron microscopy of the encephalic meninges in dogs. J . Submi- Langdon, F.W. 1891 The arachnoid of the brain. J . Comp. Neurol., 1 :205-210. crosc. Cytol. 9r1-22. Leary, T., and E.A. Edwards 1933 The subdural space and its linings. Amenta. P.S. 1987 Elias-Paulv's Histology and Human MicroanatArch. Neurol. Psychiatry, 29t691-701. omy. 5th Editiqn. John Wiley & Sons:-New York, pp. 154-160. Andres, K.H. 1967 Uber die Feinstruktur der Arachnoidea und Dura Leeson, C.R., and T.S. Leeson 1989 Human Structure. B.C. Decker Inc., Toronto, pp. 105-110. mater von Mammalia. Z. Zellforschung, 79:272-295. Liebgott, B. 1986 The Anatomical Basis of Dentistry. C.V. Mosby Apfelbaum, R.I., A.N. Guthkelch, and K. Shulman 1974 ExperimenCompany, St. Louis, pp. 259-261. tal production of subdural hematomas. J. Neurosurg., 40~336Lindner, H.H. 1989 Clinical Anatomy. Appleton & Lange, Norwalk, 346. pp. 13-15. Barr, M.L., and J.A. Kiernan 1988 The Human Nervous System, An Anatomical Viewpoint. 5th Edition. J.P. Lippincott Company, Lopes, C.A.S. and W.G.P. Mair 1974a Ultrastructure of the arachnoid membrane in man. Acta Neuropathol., 28:167-173. Philadelphia, pp. 380-382. Lopes, C.A.S., and W.G.P. Mair 1974b Ultrastructure of the outer Basmajian, J.V., and C.E. Slonecker 1989 Grant's Method of Anatcortex and the pia mater in man. Acta Neuropathol., 28:79-86. omy, A Clinical Problem-Solving Approach. 11th Edition. Mallory, F.B. 1920 The type cell of the so-called dural endothelioma. Williams & Wilkins, Baltimore. J . Med. Res., 41:349-364. Bennett, M.V.L. 1969 Electrical impedance of brain surfaces. Brain Martin, J.H. 1989 Neuroanatomy, Text and Atlas. Elsevier, New Res., 15t584-590. York, pp. 19-20. Berkovitz, B.K.B., and B.J. Moxham 1988 A Textbook of Head and McLone, D.G. 1980 The subarachnoid space: a review. Child's Brain, Neck Anatomy. Year Book Medical Publishers, Inc., London, pp. 6:113-130. 390-393. Montemurro, D.G., and J.E. Bruni 1988 The Human Brain in DissecCarpenter, M.B., and J. Sutin 1983 Human Neuroanatomy. Williams tion. 2nd Edition. Oxford University Press, New York, p. 22. & Wilkins, Baltimore, pp. 5-7. Moore, K.L. 1985 Clinically Oriented Anatomy. 2nd Edition. Clemente, C.D. 1987 Anatomy, A Regional Atlas of the Human Body. Williams & Wilkins, Baltimore, pp. 615-616, 859-864. 3rd Edition. Urban & Schwarzenberg, Baltimore, figs. 572, 573. Morse, D.E., and F.N. Low 1972 The fine structure of the pia mater of Cormack, D.H. 1987 Ham's Histology. 9th Edition. J.B. Lippincott the rat. Am. J. Anat., 133:349-368. Company, Philadelphia, pp. 366-367. Nabeshima, S., T.S. Reese, D.M.D. Landis, and M.W. Brightman 1975 Cragg, B.G. 1976 Ultrastructural features of human cerebral cortex. Junctions in the meninges and marginal glia. J. Comp. Neurol., J . Anat., 121t331-362. 164:127-170. Cushing, H. 1914 Studies on the cerebro-spinal fluid. J. Med. Res., Nelson, E., K. Blinzinger, and H. Hager 1961 Electron microscopic 3Itl-19. observations on subarachnoid and perivascular spaces of the SyrDaube, J.R., T.J. Reagan, B.A. Sandok, and B.F. Westmoreland 1986 ian hamster brain. Neurology, 11t285-295. Medical Neurosciences: An Approach to Anatomy, Pathology, Netter, F.H. 1989 Atlas of Human Anatomy. Ciba-Geigy Corporation, and Physiology by Systems and Levels. 2nd Edition. Little, Summit, NJ, plates 94, 96. Brown and Company, Boston, pp. 28-30. Nicholas, D.S., and R.O. Weller 1988 The fine anatomy of the human deGroot, J., and J.G. Chusid 1988 Correlative Neuroanatomy. 12th spinal meninges. J . Neurosurg., 69t276-282. Edition. Appleton and Lange, Nonvalk, pp. 210-212. Nieuwenhuys, R., J . Voogd, and C. van Huijzen 1988 The Human Dobrovol'skii, G.F. 1984 Ultrastructure of the meninges. Neurosci. Central Nervous System, A Synopsis and Atlas. 3rd Edition. Behav. Physiol., 14t100-110. Springer-Verlag, Berlin, pp. 57, 62. Fawcett, D.W. 1986 Bloom and Fawcett, A Textbook of Histology. Noback, C.R., and R.J. Demarest 1986 The Nervous System, Intro11th Edition. W.B. Saunders Company, Philadelphia, pp. 357duction .- and Review. 3rd Edition. McGraw-Hill, New York, pp. 359. 43-4'1. 1 -

21

THE QUESTION OF A SUBDURAL SPACE Nolte, J. 1988 The Human Brain, An Introduction to its Functional Anatomy. 2nd Edition, C.V. Mosby Company, St. Louis, pp. 3443. Nomina Anatomica 1989. 6th Edition. Churchill Livingstone, New York. Oda, Y., and I. Nakanishi 1984 Ultrastructure of the mouse leptomeninx. J. Comp. Neurol., 225r448-457. ORahilly, R. 1986 Anatomy, A Regional Study of Human Structure. 5th Edition. W.B. Saunders Company, Philadelphia, pp. 604609. ORahilly, R., and F. Miiller 1986 The meninges in human development. J . Neuropathol. Exp. Neurol., 45.588-608. Orlin, J.R., K. Osen and T. Hovig 1991 Subdural compartment in pig: A morphologic study with blood and horseradish peroxidase infused subdurally. Anat. Rec. 230:22-37. Pease, D.C., and R.L. Schultz 1958 Electron microscopy of rat cranial meninges. Am. J . Anat., 102t301-321. Penfield, W.G. 1923 Subdural effusion and internal hydrocephalus. Am. J. Dis. Child., 26r383-390. Penfield, W.G. 1924 The cranial subdural space (a method of study). Anat. Rec., 28r173-175. Peters, A., S.L. Palay, and H.D. Webster 1991 The Fine Structure of the Nervous Svstem: Neurons and Their Sumortine - Cells. Oxford University Press, New York, pp. 395-406.Platzer, W. 1989 Pernkopf Anatomy, Atlas of Topographic and Applied Human Anatomy. 3rd Edition. Urban & Schwarzenberg, Baltimore, pp. 47, 155, 157, 163, 192, 339, 341. Rascol, M.M., and J.Y. Izard 1976 The subdural neurothelium of the cranial meninges in man. Anat. Rec., 186:429-436. Rhodin, J.A.G. 1974 Histology, A Text and Atlas. Oxford University Press, New York, pp. 322-326. Rohen, J.W., and C. Yokochi 1988 Color Atlas of Anatomy. 2nd Edition. Igaku-Shoin, New York, pp. 83, 210. Romanes, G.J. 1986 Cunningham’s Manual of Practical Anatomy. Vol. 3, Head and Neck and Brain. 15th Edition. Oxford University Press, New York, pp. 211-214. Romero-Sierra, C. 1986 Neuroanatomy, A Conceptual Approach. Chruchill Livingstone, New York, pp. 35-37. Rosenberg, R.N. 1980 Neurology. In: The Science and Pratice of Clinical Medicine. Vol. 5. J.M. Dietschy, ed. Grune & Stratton, Philadelphia, pp. 371-381. Ross. M.H.. E.J. Reith. and L.J. Romrell 1989 Histolom, A Text and Atlas. 2nd Edition. Williams & Wilkins, BaltimoFi,.p. 264. Rowland, L.P. 1984 Merritt’s Textbook of Neurology. 7th Edition. Lea & Febiger, Philadelphia, pp. 145-198, 277-299. Sato, S., and J. Suzuki 1975 Ultrastructural observations of the capsule of chronic subdural hematoma in various clinical stages. J . Neurosurg., 43569-578. Schachenmayr, W., and R.L. Friede 1978 The origin of subdural neomembranes. I. Fine structure of the dura-arachnoid interface in man. Am. J. Pathol., 9253-68. Schachenmayr, W., and R.L. Friede 1979 Fine structure of arachnoid cysts. J . Neuropathol. Exp. Neurol., 38:434-446. Snell, R.S. 1986 Clinical Anatomy for Medical Students. 3rd Edition. Little, Brown and Company, Boston, pp. 798-801. Snell, R.S. 1987 Clinical Neuroanatomy for Medical Students. 2nd Edition. Little, Brown and Company, Boston, pp. 499-501. Stern, J.T., J r . 1988 Essentials of Gross Anatomy. F.A. Davis Company, Philadelphia, p. 397. Strong, O.S., and A. Elwyn 1943 Human Neuroanatomy. Williams & Wilkins Company, Baltimore, p. 81-86. Telford, I.R., and C.F. Bridgman 1990 Introduction to Functional Histology. Harper & Row, New York, pp. 237-238. Tripathi, R.C. 1973 Ultrastructure of the arachnoid mater in relation to outflow of cerebrospinal fluid. Lancet 2:8-11. Waggener, J.D., and J. Beggs 1967 The membranous coverings of neural tissues: An electron microscopy study. J . Neuropathol. Exp. Neurol., 26r412-426. Weed, L.H. 1914a Studies on cerebro-spinal fluid. No. 11. The theories of drainage of cerebro-spinal fluid with a n analysis of the methods of investigation. J. Med. Res., 26r21-49. Weed, L.H. 1914b Studies on cerebro-spinal fluid. No. 111. The pathways of escape from the subarachnoid spaces with particular reference to the arachnoid villi. J. Med. Res., 3151-91. Weed, L.H. 1916 The formation of the cranial subarachnoid spaces. Anat. Rec., 10t475-481. Weed, L.H. 1917a The Development of the Cerebro-spinal Spaces in Pig and in Man. Carnegie Institute of Washington 225, Contributions to Embryology, No. 14. Weed, L.H. 1917b An anatomical consideration of the cerebro-spinal fluid. Anat. Rec., 12t461-496.

Weed, L.H. 1920 The cells of the arachnoid. Johns Hopkins Hosp. Bull., 31r343-357. Weed, L.H. 1923 The absorption of cerebrospinal fluid into the venous system. Am. J . Anat.,31:191-221. Weed, L.H. 1938 Meninges and cerebrospinal fluid. J . Anat., 72t183215. Wegefarth, P., and L.H. Weed 1914 Studies on cerebro-spinal fluid. No. VII. The analogous processes of the cerebral and ocular fluids. J . Med. Res., 31t167-176. Wheater, P.R., and H.G. Burkitt 1987 Functional Histology, A text and Colour Atlas. 2nd Edition. Churchill Livingstone, New York, p. 112. Williams, P.L., R. Warwick, M. Dyson, and L.H. Bannister 1989 Gray’s Anatomy. 37th Edition. Churchill Livingstone, New York, pp. 1088-1090. Woodburne, R.T., and W.E. Burke1 1988 Essentials of Human Anatomy. 8th Edition. Oxford University Press, New York, UP. 312314,341-343. Woollard, H.H. 1924 Vital staining of the leptomeninges. J. Anat., 58r89-100. Yamashima, T., and R.L. Friede 1984 Light and electron microscopic studies on the subdural space, the subarachnoid space and the arachnoid membrane. Neurol. Med. Chir. (Tokyo), 24r737-746 (Original in Japanese). Yamashima, T., T. Kubota, and S. Yamamoto 1985 Eosinophil degranulation in the capsule of chronic subdural hematomas. J . Neurosurg., 62; 257-260. Yamashima, T., 0. Tachibana, M. Hasegawa, H. Nitta, and J. Yamashita 1989 Liberation of eosinophil granules in the inner capsule of chronic subdural hematomas. -Neurochirurgia, 32:168171. Yamashima, T., and S. Yamamoto 1984 How do vessels proliferate in the capsule of a chronic subdural hematoma? Neurosurgery, 15; 672-678. Yamashima, T., and S. Yamamoto 1985 The origin of inner membranes in chronic subdural hematomas. Acta Neuropathol., 67: 219-225. Yasargil, M.G. 1984 Microneurosurgery: Microsurgical Anatomy of the Basal Cisterns and Vessels of the Brain, Diagnostic Studies, General Operative Techniques and Pathological Considerations of the Intracranial Aneurysms. Vol. 1. Georg Thieme Verlag, Stuttgart, pp. 5-168. Yasartril. M.G.. K. Kasdaelis. K.K. Jain, and H-P. Weber 1976 Anatokical observationsof the subarachnoid cisterns of the brain during surgery. J . Neurosurg., 44r298-302.

__

NOTE ADDED IN PROOF

In a recent study Frederickson (The Subdural Space Interpreted as a Cellular Layer of Meninges, Anat. Rec. 230:38- 51, 1991) described the electron microscopy of the meninges with particular attention to those cells found in the region of the dura-arachnoid junction. Internal to the dura Frederickson identified a morphologically distinct area composed of cells with flattened processes, comparatively translucent cytoplasm, minimal cytoplasmic organelles, and only “occasional” cell junctions. An additional characteristic of this cellular layer is the occurrence of enlarged extracellular spaces containing a “dense granular . . . material’’. The structural features of this cellular layer (termed “light cells” by Frederickson based on their staining characteristics) are, in their essentials, fundamentally the same as the dural border cell layer as described by others. Frederickson also showed that when the arachnoid and dura were reflected from each other the plane of separation is through the light cell (dural border cell) layer and that fragments of these cells remained attached to the inner surface of the dura and the outer surface of the arachnoid. This study concluded that a subdural space, as classically identified, is not present and that the appearance of a space in this area of the meninges is the result of mechanicallpathological processes which split open the “light cell” (dural border cell) layer.

On the question of a subdural space.

The structure of the meninges, with particular attention to the architecture of the inner portions of the dura mater and the arachnoid mater, has been...
2MB Sizes 0 Downloads 0 Views