- Email: [email protected]
On the flow dynamics of cerebrospinal fluid
neurology, psychiatry and brain research 21 (2015) 96–103
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/npbr
On the flow dynamics of cerebrospinal fluid§ Karl Bechter a,b,*, Patrick R. Hof c, Helene Benveniste d a
Clinic for Psychiatry and Psychotherapy II, Ulm University, Bezirkskrankenhaus Günzburg, Ludwig-Heilmeyer-Str. 2, D-89312 Günzburg, Germany b Ulm University, Department Psychosomatics/Psychotherapy, Ludwig-Heilmeyer-Str. 2, D-89312 Günzburg, Germany c Fishberg Department of Neuroscience, Hess Center for Science and Medicine, 1470 Madison Avenue, 10th Floor, 10-118, Mount Sinai School of Medicine, New York, NY 10029, United States d Department of Anesthesiology, Stony Brook Medicine, Stony Brook, NY 11794, United States
article info
abstract
Article history:
Despite a not insignificant number of anatomical and experimental studies describing the
Available online 9 March 2015
distribution and movement of the cerebrospinal fluid several questions were answered controversially, leaving room for objections and doubts. Some of these controversies I have
Keywords:
tried to address by new experiments. Before going on to describe these studies, some short
Cerebrospinal fluid
anatomical notes on the membranes that cover the central nervous system may be
CSF outflow
appropriate.
CSF and psychomotor activity CSF flow
Bichat described the anatomical membranes as follows: in between the fibrous lining of the brain-vertebral cavity, the dura mater and the vascularized coat of the spinal cord (the pia mater) a space covered by a serous skin is interposed, the parietal part of which is integrated with the inner side of the dura mater; and the visceral part in contrast is detachable from the underlying pia mater. The watery fluid of the brain-spinal-cord cavity discovered by Cotugno received more attention by Magendie, who initially placed it [the fluid] within the serous space described by Bichat, convincing himself [Magendie] however later, that it [the fluid] is present in between the visceral sheet of the arachnoid and the pia mater. © 2015 Elsevier GmbH. All rights reserved.
Despite a not insignificant number of anatomical and experimental studies describing the distribution and movement of the cerebrospinal fluid several questions were answered controversially, leaving room for objections and doubts. Some of these controversies I have tried to address by new experiments. Before going on to describe these studies, some short anatomical notes on the membranes that cover the central nervous system may be appropriate.
Bichat described the anatomical membranes as follows: in between the fibrous lining of the brain-vertebral cavity, the dura mater and the vascularized coat of the spinal cord (the pia mater) a space covered by a serous skin is interposed, the parietal part of which is integrated with the inner side of the dura mater; and the visceral part in contrast is detachable from the underlying pia mater. The watery fluid of the brain-spinalcord cavity discovered by Cotugno received more attention by
§ Published in 1872 by Dr. H. Quincke, Berlin, in KB Reichert Arch Anat Physiol Wiss Med, 153–177. Translated by Karl Bechter, Patrick R. Hof and Helene Benveniste. * Corresponding author at: Ulm University, Department Psychosomatics/Psychotherapy, Ludwig-Heilmeyer-Str. 2, D-89312 Günzburg, Germany. Tel.: +49 822196 2540; fax: +49 822196 2736. http://dx.doi.org/10.1016/j.npbr.2015.01.001 0941-9500/© 2015 Elsevier GmbH. All rights reserved.
neurology, psychiatry and brain research 21 (2015) 96–103
Magendie, who initially placed it [the fluid] within the serous space described by Bichat, convincing himself [Magendie] however later, that it [the fluid] is present in between the visceral sheet of the arachnoid and the pia mater. Later on, researchers agreed with this assumption and only recently did Henle1 present this point-of-view in his handbook of anatomy, by describing the subarachnoid connective tissue with its mesh-like spaces as a physiological water-filled connective tissue (Virchow) of unusually loose quality; towards the inside it faces the vascular skin (the pia mater), and towards the outside the so-called arachnoidea visceralis. The latter is adhering – typically tightly – to the inner side of the dura mater. Given that the majority of anatomists now agree that the cerebrospinal fluid is located mainly subarachnoidally, there are nevertheless different views about the presence of fluid in the actual arachnoidal sac (in between dura and arachnoidea visceralis). While Luschka [and] Reichert assume an admittedly very small volume of fluid in that very place, Ecker totally refutes that notion in the live animal. This contradiction arises, I assume, from the fact that some [anatomists] preferentially focus on the scenario at the level of the spinal cord, while others focus on the skull. At least in live animals (dogs, cats, rabbits) I found that in the spinal cord the arachoid2 always closely adheres to the dura mater, because during incision of the latter the arachnoid in most cases will be concomitantly injured and the subarachnoid fluid will flow out. One would only rarely exclusively incise the fibrous skin [dura mater]; in this case, the arachnoid protrudes like a hernia due to the fluid content within its meshed spaces; the basis of the wound however is staying dry, because in between arachnoid and dura no fluid is present. Also in dead animals, the view into the spinal canal through the foramen magnum reveals that free connective tissue strings are spawning from the pia to the arachnoid; the latter, however adheres to the dura and can only after overcoming a certain resistance be separated from it. When trying via a puncture to make an injection, I never once accessed the interstice in between the arachnoid and dura, but always ended up in the subarachnoid space. The situation is different at the level of the skull. Here the connective tissue strings between pia and arachnoid are shorter and more robust, the meshed filled spaces (with the exception of the points assigned by Magendie as ``confluents''), narrower; the arachnoid does not stick to the dura, but is separated from it by a capillary fluid layer, therefore in the skull, opening of the dura itself without injuring the [underlying] arachnoid is with some caution relatively easy. [Even] with the dura intact, the distance between both membranes is wide enough to insert a rounded rectangular bent cannula and administer an injection into the actual serous sac in between arachnoid and dura, without injuring the arachnoid. Hence, in the brain, the arachnoid is disconnected from the dura by a capillary fluid layer, whereas at the level of the spinal cord this serous space is non-existent, and instead [in the spinal cord] the arachnoid is tightly adherent to the dura mater. [In the spinal cord], the numerous connective 1
Band III. 2. S. 312. For the sake of brevity I assign herewith the (anatomically exclusively isolated) arachnoidea visceralis. 2
97
tissue bands (denticulate ligament, nerve roots sheets) running from the pia to the dura may have induced this closer relationship. The transition from one anatomical scenario to the other starts at the back rim of the cerebellum and at the medulla oblongata. Experiments reported below will show whether a communication between the subarachnoidal and the arachnoidal mesh spaces, exists. I now continue with the description of experiments where an emulsion of cinnabar was injected at various places into the cavity of the spinal cord of live animals. [For these experiments] I used, as do painters, the finest pulverized cinnabar granulated with sugar water, diluted with five- to ten-fold the volume of water. In a series of animal experiments (exclusively dogs, – cats, rabbits are too small) the cinnabar emulsion was injected into the subarachnoid spaces of the spinal cord cavity via a cannula. Usually this was done by incising the skin above the upper lumbar vertebra and by dissecting the musculature from one side of one or two spinal processes; and when arriving at the vertebral arches, a cannula was passed through the [ligamentum flavum] close to the midline, 1 cm deep, and the cinnabar emulsion was injected. When not considering completely unsuccessful trials, in which case the needle [altogether] missed the spinal canal, successful placement of the tip of the cannula (at the intended spot) depended on pure chance and training. An injury of the spinal cord was usually not an issue, because from such a needle stick the cinnabar may have nevertheless reached the subarachnoidal spaces. Not rarely did it leak through the small, local incision associated with the penetration of the dura mater or – in the case of minute misplacement of the cannula during the injection – directly into the fat-rich loose tissue between the dura and the periost [epidural space]; sometimes it accumulated exclusively in this space; and never advanced further into the space between the dura mater and the arachnoid. After surgery, the skin wound was closed and the animal was left to recover. The behaviour of the animals now was completely different (not considering the after effects of narcosis, which the dogs received in some cases), usually a weakness of one or both lower extremities was observed, presumably due to an injury of the spinal cord. In some dogs this weakness dissipated completely or incompletely over the ensuing days or weeks. The seemingly healthy animals were euthanized after a period of one week to three months (after the injection). In other experiments, if the animals exhibited strong restlessness the spinal cord was more severely injured by the needle stick, [myelitis developed leading to complete paraplegia]; under these circumstances, the animals were usually euthanized after 2–4 days. In other cases although the animals appeared to be feeling well after the injection, they would die during the next 12–24 h. Examination of the brain and spinal cord always revealed dispersion of cinnabar within the subarachnoid connective tissue and in the tissue of the pia mater. In ten out of twelve cases it had travelled to the skull where it preferentially accumulated at the base of the brain, specifically at the site where pia and arachnoid are separated from one another by larger meshed spaces. Beyond, it was observed on [the surface of] the entire brain and spinal nerves if their roots, were
98
neurology, psychiatry and brain research 21 (2015) 96–103
running in the cerebrospinal cavity; the most abundant accumulation was observed at the site where the nerves left the sac of the dura mater (in the skull meaning at the portals into the bony canals). The distribution of cinnabar along the spinal nerves varied from case to case for no apparent reason that could be determined. In a number of cases the cinnabar advanced beyond the cerebrospinal cavity. In about half of the experiments it was observed along the intercostal nerves as far as to where the rami communicans branch off to the sympathetic chain, or even some millimetres beyond. This particular distribution pattern could be observed without any extra preparation after the lungs were removed due to the transparency of the pleura. At the level of the lumbar nerves the cinnabar could in several cases be seen as far away as in the lumbar plexus in between the psoas, and additionally in the sacral plexus beyond its entry into the pelvis. Underneath the brain, the olfactory nerve was not accompanied by cinnabar beyond the cribriform plate. In contrast, cinnabar was consistently observed along the optic nerve, in one case even when the dye was found to be only sparsely dispersed within the pia at the basis of the brain and also not observed along the other brain nerves. The cinnabar accumulation was always most abundant close to the entrance of the optic nerve into the eye, and from there on onwards to the optic foramen it gradually faded, parts of the nerve being completely free of dye. Of the remaining brain nerves the trigeminal nerve in particular has to be mentioned, its ganglion in its intracranial position also being in most cases dyed by cinnabar. Beyond the bony canals I never could demonstrate cinnabar along the brain nerves (except along the above-mentioned optic nerve), also not along the hypoglossus [nerve], [in spite of data from] Key and Retzius3 [who] by artificial injection [noted that] dye travelled [along the hypoglossus nerve] until its entrance into the tongue. In the majority of cases cinnabar was also found in the arachnoidal sheet of the carotid at the point of exit from the cavernous sinus, in the cortical substance of the posterior upper part of the large cervical lymphatic glands, [and] sometimes smaller volumes [were also observed] in the submaxillary lymphatic glands. On the inside, the dura mater was always free of cinnabar, however at certain spots along the venous sinus reddish dyed points or nodules were present, [a topic] on which we shall come back to later. These were the pathways by which the cinnabar after injection into the subarachnoidal spaces of the spinal cord was travelling, and which usually could be visualized with the naked eye. In which tissular elements and spaces it had deposited remains to be described. If the injection was copious and occurred one or few days before the anatomical examination, free cinnabar would be found within the connective tissue meshes, partly in irregular and only microscopically visible accumulations, partly in larger [deposits confined to] pseudomembranes, in which the dye granules were pasted onto one another by a fine granular amorphous substance (fibrin?). In those collections one found more or less copious cinnabar-containing lymphocytes; these 3
Virchow-Hirsch Jahresbericht 1870. Centralbl. f. die medic. Wissenschaft 1871.
were [also] observed in association with free cinnabar distributed in more compact tissue of the pia. Frequently lymphocytes were present within the sheets around the small arteries, even more frequent however, they were [present] independent of these. The number of lymph bodies in the connective tissue of the pia and the subarachnoidal meshed spaces were undoubtedly in many cases increased [compared to normal], in that a beginning meningitis was present. In addition, cinnabar was found in rounded or irregularly shaped cells, a bit larger than lymphocytes, heterogeneously distributed in the subarachnoid tissue, sometimes in solitary presentation, sometimes in groups, and which presumably have to be considered some type of connective tissue cell. The longer the animal lived, the more the free cinnabar seemed to fade and to transit intracellularly. In this regard, the pia and the subarachnoid of the brain and the spinal cord behave rather similarly. Transport of cinnabar into the brain or the spinal cord parenchyma was also microscopically never confirmed. The choroid plexus was in the majority of cases free of dye; in only two cases did the choroid plexus of all four ventricles contain cinnabar and then it presented in a particular way, with remarkable little dye in the stroma, but was instead overwhelmingly present in the epithelium. In one case, in which the choroid plexus was not stained, there was a clot of cinnabar-containing lymphocytes in the fourth ventricle. Around nerve roots both from the brain and the spinal cord, the cinnabar was found in the lymphoid cells (lymphoiden Zellen), observed in lymph bodies (Lymphkörper) along the surface [of the nerve], the free dye always occurring in more or less large heaps very close to the sites where the nerves exited the dura mater. Where cinnabar was found beyond these places, it seemed always to be included in round cells of the nerve sheet. Cinnabar was never found deposited into the trunk of the nerves, or in the trigeminal ganglion and spinal ganglia. The lymph glands contained free cinnabar as well as within the round cells. In vain did I search for cinnabar in the nasal mucosa and the spleen. In the labyrinth cinnabar was found only in one out of five cases investigated, in the tympanic membrane of the cochlea. The cinnabar reached during the injection not only the subarachnoid space but also beyond (or exclusively) in the tissue situated between the dura and the periost, where it distributed from the back plane of the vertebral bodies (more or less farther), often as far as to the neck; it was often guided even further away by the current of the fluid [CSF]; it appeared in the lumbar lymph glands situated along the aorta, in the subclavian and mediastinal glands, and also at the inner plane of the intercostal muscles, so that it shone through the pleura. While the initial course of cinnabar transport from the subarachnoid space appears to keep the exact course of the intercostal nerves, its general movement out of the vertebral canal is less clear, progressing more diffusely and a little bit farther into the intercostal spaces below the pleura, sometimes seemingly following the course of the azygos vein. In the cellular tissue of the vertebral canal the cinnabar is partly contained in lymph bodies, partly freely in the tissue, even after weeks. Because neither during life nor after death
neurology, psychiatry and brain research 21 (2015) 96–103
The most plausible cause of distribution of cinnabar in live animals to be considered might be the up-and-down revolving (respiratory) movement of the subarachnoid fluid as was [originally] proposed by Magendie. Some researchers (e.g., Leyden)4 doubted that such a movement of fluid5 [CSF] was normal and interpreted it to be a consequence of the opening of the brain/spinal cord cavities, yet this argument does not apply to the majority of my experiments, because during the injections with a cannula in the lumbar vertebral trunk, the normal anatomy was not altered at all, and in the skull, the small opening was immediately closed by a wooden pin, and only in those cases in which the dura had lost the respective bony supporting plane near the trepanation site, a slightly more compliant area was established. The distribution of cinnabar was however the same for all of the procedures. Incidentally, in children a membranous closure of the skull at some places is normal. Moreover, in adults as well the skull does not represent a completely stiff enclosure, like would a metal or glass bottle with only a single opening; rather, the multiple holes of the skull are filled out by smooth tissue, associated with some compliance, making it possible to change the capacity of its cavity. Within the sheet of the optical nerve we clearly see a protuberance of the subarachnoid spaces with its flexible walls. The way the respiratory movement influences the [CSF] fluid has been clearly described by Ecker6; the flexible venous plexus of the spinal cord canal sinks more strongly with inspiration, and is expanding during the expiration when compared to the stiff-walled sinus of the skull; by this [one can infer that] during expiration the cerebrospinal fluid will be driven from the spinal cord to the skull, whereas during inspiration the reverse is taking place. Only on one point do I have to contradict Ecker, in that by this (movement) a fluctuating filling and emptying of the brain ventricles [is to be also considered], and accordingly then the cinnabar constantly had to be found in the ventricular spaces. It may well be that the weight of the cinnabar contributes in itself somehow to its accumulation at the base of the brain, nevertheless, from its distribution, we will overall recognize the pathways of the fluid's currents within the subarachnoid spaces. It can be expected that such CSF currents occur preferentially in the more spacious parts of the skull – in the so-called subarachnoid sinus (``confluents'', Magendie), and that from there only weak [CSF] currents occur to the hemisphere convexity. When the injection happened over a hemisphere, the dye distributed preferentially to the quadrigeminal plate and the base [of the brain]; only smaller amounts of dye reached the convexity of the contralateral hemisphere consistently. The ascending current from spinal cord to brain appears in general to be stronger than the descending, because the
the cinnabar did not induce major inflammatory changes, one feels permitted to consider that its distribution pattern represents natural pathways of [CSF] flow, which the products of pathological deposits from the vertebral canal, as in the case of caries [=osteomyelitis as the likely meaning; the authors] of the vertebral bones, will also preferentially follow. In a second series of experiments the cinnabar emulsion was injected into the skull cavity, in rabbits and cats, as well as in dogs. To this end, the dura was opened, using a small trocar [6 mm diameter] or a simple pointed tool, and afterwards the opening was closed by a wooden pin. The animals were examined (after spontaneous death or euthanasia) a few hours to eight days after the surgery. If the cinnabar emulsion was injected into the skull by a straight cannula, it advanced unavoidably through the arachnoid and pia into the brain, the dye could thus advance into the subarachnoid as well as into the arachnoid spaces. When using a very thin, cannula bent in a quarter of a circle, it was possible to introduce this under the exposed dura (the opening inflected towards it) and to avoid injuring the arachnoid as taught by the [previous] sessions; as such, the dye was injected only into the actual arachnoidal space (between dura and visceral layer of the arachnoidea). After a few days most of the dye disappears to be found in the subarachnoid spaces and in the pia of the brain, exactly as if [the dye] had been directly injected into this space. The dye is deposited preferentially at the base of the brain, as well as at the side of the convexity of the brain adjacent to the injury; beyond this, it is found at the exit points of the cranial nerves, in the carotid sheet, and in the cervical lymph glands. In the majority of cases it advances also into the vertebral canal and distributes in a variable manner, often as far as the cauda equina, depositing in the pia mater and at nerve roots, as described in the first series of experiments. Eventually, the distribution of the dye is the same whether it is injected into the subarachnoid spaces of the spinal cord or into those of the brain; the only difference is quantitative, inasmuch as a more copious deposition in the first case is found in the skull and in the other case, in the spinal cord canal. From these experiments I believe that the following conclusions can be drawn. 1. There exists a relationship between the subarachnoid spaces of the brain and the spinal cord. 2. In the live animal, the current of the subarachnoid fluid flows in a reverse direction from the back to the front. Dye particles, in the [subarachnoid] fluid, will be transported from the spinal cord to the brain as well as from the brain to the spinal cord. Evidence supporting that this transport process is not exclusively mediated by migration of lymph bodies is based on the fact that in many experiments deposits of fine [free] cinnabar were found at places remote from the injection site. Control experiments in dead animals demonstrated that following the injection itself or following passive movements (to and fro swinging of the legs among others) the cinnabar never distributed as diffusely as in living animals; when injecting the same amounts around the lumbar vertebra, cinnabar reached at most the cervical marrow, and if injecting the skull, at most the medulla oblongata.
99
4
Virchow's Jahresbericht XXXVIL. I confirm that I am speaking only about a movement of the fluid, not about a movement of the brain or the spinal cord itself. Such a one is, at least for the brain, disproven by the experiments with insertion of a glass window (Donders, Leyden and others). 6 Physiologische Untersuchung über die Bewegung des Gehirns und Rückenmarks. Stuttgart 1843. 5
100
neurology, psychiatry and brain research 21 (2015) 96–103
progression of the cinnabar in the former direction was more substantial and consistent than in the latter. 3. Also, after injection into the arachnoidal spaces of the skull, the dye advanced into the subarachnoidal spaces of the brain and the spinal cord, indicating that these two spaces must be communicating. In the live animal the CSF flow has to be preferentially from the arachnoidal to the subarachnoidal spaces, as with injections into the latter in the spinal cord, cinnabar was never found in the space between the dura and arachnoid. 4. The outflow pathways of the CSF. A fraction of CSF leaves the brain/spinal cord cavity along the [cranial and peripheral] nerves. This assumption appears to me to be explained by the consistent accumulation of dye where the nerves exit the dura mater/arachnoid. Free dye was transported by the CSF current to these exits, and dye accumulating within cells was most likely taken up by these `on route', because there were typically very few cells, which probably represent migrating lymphocytes. At the sites where the nerves exit, the dura and/or arachnoidal membranes appear to be more tightly adhering to the nerves, because beyond this point of exit the cinnabar was never observed along the [cranial] nerves,7 only occasionally along the lumbar nerves, and most frequently along the intercostal nerves. Hence, pathways seem to exist which allow CSF to track along nerves beyond their exit points from the brain and spinal cord, although only under certain circumstances do these pathways permit solutes to pass. In case of a meningitis, purulent matter might easily access these pathways and by exerting pressure on the nerves may cause inflammatory disturbances. In addition, the flow of CSF is passing through the cervical lymph glands, the upper posterior portion of which in the live dog has a special relationship to pia and arachnoid of the brain and upper cervical cord, as well as to the maxillary lymph glands. Whether the subarachnoidal spaces need be consequently considered as a lymph space or only as an appendix of the lymphatic vascular system, may be left to anatomists to decide. Another outflow pathway for the arachnoidal and subarachnoidal CSF was found by Key and Retzius8 in the Pacchioni' granulations. According to them these granulations do not simply attach to the dura mater, but are, as Trolard9 described, free protrusions, coated exclusively by an epithelia, and localized partly in the veinous sinuses, and partly in their lateral protuberances. In my experiments these structures were also marked, when the injection occurred in the vertebral column or the skull, by a strong cinnabar staining; these were regularly present in the superior sagittal sinus as well as in the transverse sinus; I also found at times similar structures in the cavernous sinus. According to their structure, described in more detail by Key and Retzius as similar to that of lymph glands, these granulations seem to function as a filtration system, which may well let fluid pass while however retaining large 7 A special case is, the N. opticus, the behavior of which has to be discussed separately. 8 A. a. O. 9 Arch. Gen. 1870. Mars.
particles. Had any substantial number of dye particles passed through these, they would have arrived into the blood stream, and consequently would have plausibly been deposited in the spleen; never however have I succeeded to find cinnabar in this organ. The cinnabar did not pass the vascular network in the dura described by von Boehm10 in my experiments. Thus one has to ascribe to the subarachnoidal fluid a double movement: probably secreted by the blood vasculature it is flowing under a certain pressure on defined pathways into the lymphatic vasculature. Which of these pathways are preferentially used, is as variable as the blood stream within the vasculature, depending on time and situations; in general the distribution of cinnabar suggests that the draining pathways out of the skull are preferred compared to that of the spinal cord. The pressure, exerted upon the fluid, is varying over time and differs at different locations as a to-and-fro movement is taking place; the respiratory movements are effective as a pump on the venous blood, the lymph stream within the pleura, and so on, which, together with the pressure propagated from the arteries, is pushing the fluid through its draining pathways. It is clear that these constant movements of the fluid within the continuously connected meshed spaces has to favor the distribution of pathological products; thus, we see not rarely a purulent meningitis distributing with high velocity from one point all over the whole central nervous system. Nevertheless, purulent infiltration of the pia mater localized to certain areas can presumably be explained by the fact that the dye particles are usually suspended more freely within the fluid compared to purulent matter, the latter sticking to the walls of the meshed-like space and therefore not easily cleared away by the fluid currents [compared to the free dye particles]. At least I found, that especially in cases where the distribution of cinnabar was more localized, an inflammatory process had developed, typically at the site of injection; rather continuous red-colored purulent fibrinous masses filled up the subarachnoidal space over a certain distance; microscopically the cinnabar was partly found in pus accumulations, partly in between such pus bodies inlaid within a fine granular substance. Generally, inflammation of the pia after cinnabar injection was however rare, particularly when the marrow [we think that they used the term ``marrow'' for white matter, similarly brain and spinal cord; the authors] itself was little affected; it occurred more frequently after injections into the skull, with injury of the outer layers being more considerable. Some animals were euthanized only 2–3 months after the injection; the wide deposition of dye described above had not disturbed their well-being. The wide distribution of dye seems to me to reflect the advantageous nature of the cerebrospinal fluid. As is well known, it is according to its anatomical distribution and to its chemical nature, most closely connected to the fluid of edematous tissues and contains, in contrast to the fluids of serous cavities like the pleura and the peritoneum, few solid 10
Virch. Arch. Bd. 47.
neurology, psychiatry and brain research 21 (2015) 96–103
By injections partly from the arachnoideal space, partly from the eye bulb, Schwalbe16 demonstrated a system of hollow spaces in this location, which are interconnected and communicate with the skull cavity. Schwalbe has the space in between the inner (firmly adhering to the nerve) and the outer (loose, mobile) sheet of the optic nerve, the subvaginal space, communicate with the arachnoideal space, which is tube-like, ending blindly at the bulb and interlaced by beams like the subarachnoideal spaces (!); Its walls is covered with an epithelia. A second lymphatic space, supravaginal, situated in between the outer sheet and the M. retractor bulb, is funnel-shaped, being related anteriorly to the Tenon's space and posteriorly again with the subarachnoideal room; the structure and place of the latter are not very clear from the description by Schwalbe, because he explicitly assigns the outer sheet of the optic nerve as a continuation of the dura mater. Key and Retzius differentiate the two tube-like spaces at the optic nerve, the inner of those communicating with the subarachnoideal space, the outer with the arachnoideal (subdural) space, both being interconnected. H. Schmidt17 tried, based on Schwalbe's and his own injection experiments, to explain the development of the papilledema in brain diseases by an accumulation of fluid in the sheet and a resulting entrapment of the optic nerve. Finally Manz18 made this last point in particular an issue of experimental and patho-anatomic examinations and agreed with Schmidt's explanation. He found a stronger accumulation of serum in the optic sheet whenever the intracranial pressure is increased, and even when ophthalmic alterations are not yet visible; in one case of hemorrhagic pachymeningitis. He saw an infiltration of the optic nerve sheet with blood, which well seemed to have been carried there from the skull cavity. My experiments demonstrate now that a transfer of subarachnoideal flow into the optic nerve sheet has to be considered a constant and essentially normal event. This is because the dye, which was mixed to this fluid in vivo, was nearly exclusively found in the optic nerve sheet, whether the injection had happened in the skull or deep below in the spine; even in case of very minor injections in the spine the dye appeared in the optic nerve, while in other cranial nerves not dye was observed, only the large meshed spaces at the brain basis containing some; the vicinity of the latter to the optic nerve's origin apparently favors the movement of the fluid to this area. Yet after a few hours a mass-like accumulation of cinnabar in the optic sheet could occur; a part of the dye is then always free; only in smaller amounts may it be taken up by endothelial cells, which coat the sheet's spaces and the connective tissue strings crossing it. It is found most constantly and in greater amounts in the ``subvaginal space'' (Schwalbe) at its blind end, close to its entrance into the eyeball; sometimes the more centrally situated part of the optic nerve is even free.
elements, namely little protein and no fibrinogen; the formation of clots, which in such cavities may easily trap the cinnabar and isolate it, cannot occur in CSF. With regard to the continuous relationships of the subarachnoid spaces of the brain and the spinal cord the experiments presented here confirm the observations of Magendie11 and Luschka,12 which were recently confirmed by Axel Key and Retzius.13 Later authors have described the relationships of these named spaces with other parts. However, because they used non-physiological injection methods in living and dead animals, it may easily have been performed by using stronger pressure so that a passage of fluid into wrong pathways occurred. They could have accessed spaces, indeed naturally communicating with the arachnoideal and the subarachnoideal spaces, which laid within the trajectories of the physiological fluid movement but upstream from the natural flow, and thus did not serve as draining pathways [under normal conditions]. Either is excluded in my experiments because only little fluid (0.2–1, rarely some cubic centimeters) was injected at once. The forces used during the injection therefore could not widely distribute the dye, its transport remaining left to the normal forces effective in the living body. This demonstrates therefore the natural CSF current pathways within the cerebrospinal cavity with certainty, as well as its outflow pathways. That in various situations these take different directions and therefore resulting in some disparities, has been mentioned above when describing the experiments. However in none of my experiments the dye, compared to Key's und Retzius' experiments, advanced into the central canal of the spinal cord, in the perivascular rooms of the brain14 and the spinal cord parenchyma, so it is therefore quite probable that under normal circumstances these spaces pour their contents into the subarachnoid spaces, and thus do not receive flow from there. The same is valid for the lymphatic vasculature of the olfactory mucosa, as well as for Tenon's space, the perichoroidal space, and the lumbar glands, in which Key, Retzius, and Schwalbe15 saw the injection mass advancing from the arachnoideal space. Here the possibility remains that under certain circumstances these paths may sometimes serve as collateral draining pathways. In contrast to Key and Retzius [who] injected nerve sheets I have only in part (see above) and only exceptionally found some dye; perhaps the current is strong enough to carry the dye so far only in the case of heavier pressure. Some points touched only shortly thus far are to be discussed in more detail: 1. The relationships of the optic nerve and its surroundings to the skull cavity.
11
Rech. Phys. et Chim. Sur le liq. Céphalorachidien. Paris. 1842. Die Adergeflechte des menschlichen Gehirns. Berlin. 1855. 13 A. a. O. 14 Only in one case, where presumably the cannula happened to be placed below the pia, there were cinnabar containing lymph bodies in the area of the injection after three days in the perivascular space of the cerebral cortex. 15 Archiv f. mikroskop. Anat. Bd. VI.
101
12
16 17 18
A. a. O. Arch. f. Ophthalmol. XV. 1869. Arch. f. Ophthalmol. Bd. 16. 1870.- Arch. f. klin. Medic. IX. 1871.
102
neurology, psychiatry and brain research 21 (2015) 96–103
Cinnabar was also found in the supravaginal space, yet more rarely, partly free, partly enclosed into cells; in some cases it had even progressed into the posterior aspect of Tenon's space; I never could found it in the suprachoroid space. The separation between supra- and subvaginal spaces appears not to be very strict, because I often found cinnabar-containing cells in the outer optic nerve sheet; there seems therefore that a communication exists between them and with the subarachnoideal space through the connective tissue bundles in between them; I never saw examples that would fit with the description of a arachnoideal and subarachnoideal sheet according to Key and Retzius, after simple injection [in these spaces}, so that I cannot decide whether the actual arachnoideal space (subdural space according to Key) is continuing towards the optic nerve, or whether such communication is the fact of only by the subarachnoideal space. The subvaginal space appears completely separated anteriorly near the eyeball, as was revealed by the injections. In the retina, cinnabar was never found, as in the optic nerve itself. The space between the nerve fibers injected by H. Schmidt would therefore serve only for influx into the hollow space of the optic sheet. One may wonder which of either currents occurring in the CSF may carry the dye into the optic sheet; that the respiratory to and fro movement of the fluid has to be involved, was already considered as probable, because the optic sheet represents the largest protuberance of the cerebrospinal cavity with compliant walls.19 Nevertheless depending on the predominant pressure conditions here (and towards the Tenon's space) a continuous outflow might take place as well, as in other cranial nerves. The fact that an exchange of fluid in between the optic sheet and the subarachnoideal spaces is physiologic suggests that the relationships between diseases of the brain and of the optic nerve are closer than previously assumed. Propagation towards the optic nerve sheet is rather likely in case of hemorrhage into or in between the meninges, or in case of purulent infiltration of the weak brain envelopes. Frequently, but always unsuccessfully, I attempted to observe the initial stages of a papilledema after cinnabar injections; apparently the resulting pressure increase within the skull cavity were not sufficient to induce the phenomena observed by Manz after injection; or these may have happened (see below) so rapidly as to result in death. 2. The fluid of the subarachnoideal spaces of brain and spinal cord seems rarely to advance Into the connective tissue stroma of the choroid plexus, which is continuous with the meshed tissues of the arachnoid, as cinnabar was found there only in three of about 20 experiments of this type. – This is in agreement with the observation that in purulent meningitis, even widely distributed, purulent infiltration of the plexus chorioidei is found extremely rarely.
19
I want especially to remark again that for the other nerves at the eyebulb cinnabar was never found beyond the holes of the skull.
The brain ventricles never contained free cinnabar even in case of near complete distribution of the dye; only occasionally some was found in the ventricles as flake-like purulent deposits, by which it was presumably transported there. A fluid current into the ventricles can therefore not take place according to this negative finding just as little as an alternating in-and-out flow movement as assumed by Magendie, Ecker, and others to represent a side effect of the current of the actual subarachnoid fluid that depends on respiration.20 The possibility of an opening of the IV. ventricle towards the posterior side remains the only possibility left for a continuous fluid current out of the IV. ventricle into the subarachnoideal space; this is a possibility, which is supported by the opinion of those anatomists, who suggested the choroid plexus to represent a secretory organ. For a closer focus on this point I tried to inject cinnabar emulsion (0.08–0.3 cc.) directly in the lateral ventricles in a number of living dogs, by introducing through a narrow borehole a cannula until to a certain depth; after the injection was performed, the defect in the bone was closed by a wooden pin, the skin sewed and after 1–25 days an anatomical examination was performed. Although even in successful experiments, an injury of the subarachnoideal tissues at the convexity and at the plexus itself was unavoidable, I believe nevertheless to be able to conclude from the distribution of the cinnabar in the four ventricles, as well as in the arachnoideal tissues of the medulla oblongata and the neighboring structures, that mainly it may have advanced through the third ventricle and the aqueduct of Sylvius into the fourth ventricle and beyond into the subarachnoideal tissue. Accordingly, it seems that a natural fluid current is taking place in the described direction, seemingly induced by a secretion of the choroid plexus, the pathological increase of which we see in children's hydrocephalus. It appears that open pathways must exist along which the fluid carrying cinnabar is advancing from the ventricles into the subarachnoideal spaces. In humans such a communication in the fourth ventricle was described by Magendie, Luschka and others, in the form of an opening (sometimes multiple) of several lines diameter (Magendie's foramen), also reported by others such as Burdach, and Reichert21 as the tela chorioidea posterior being a continuous membrane, completely tightening the ventricle; in the horse Luschka also agrees with this description. Magendie's statements about the situation in animals are mutually conflicting. Because I was not myself successful in discovering in dogs a preexisting well-bordered opening in the tela chorioidea posterior, my experiment having however demonstrated the existence of connecting pathways, these would have to be sought in the interstices of the connective tissue bands, which constitutes the pia. The variability in the strength, robustness, and arrangement of these connective tissue bundles among various species and 20 The observation of Magendie, who observed at the opening of the ventricles a rising and falling of fluid synchronous with breathing, does not contradict this. 21 Bau des menschlichen Gehirns II. S. 53.
neurology, psychiatry and brain research 21 (2015) 96–103
individuals of the same species, may explain the discrepancies in the statements of other authors. A far reaching significance of these variations is in any case not to be assumed; the foramen of Magendie is nothing more than a variable hole of the connective tissue, which is sometimes not even patent. Also in the surrounding of the great vein of Galen, where the subarachnoideal tissue is more loose (``confluent supérieur'' of Magendie) the ventricle seems to communicate with the subarachnoideal spaces (not to be mistaken with the so-called Bichat's hole, which was considered a communication towards the actual arachnoideal space but turned out to be an artifact). Twice after injections into the lateral ventricle cinnabarcontaining lymph bodies were found in the central canal of the spinal cord, even as far as to the lumbar region. I cannot explain with certainly how in three cases the cinnabar could advance into the epithelia of the choroid plexus after subarachnoideal injections. It was such an aggregation in those cases, that the plexus appeared completely reddish with the naked eye; yet it was found only sparsely in singular lymph bodies in the tissue of the plexus. Because the cinnabar was not taken up by the epithelia of the plexus after the injections into the ventricular cavities, it is probable that it had advanced in those cases not from the free but from the lower side into the epithelial cells. Henle described in humans yellow or reddish dyed bodies in these cells, which he believed to be derived from red blood cells and which would have similarly moved in from below into the epithelial cells. This may represent a rather similar event as in my experiments with the cinnabar. 3. In some (4) experiments death occurred after cinnabar injections in the vertebral column (in two dogs) as in the skull (in one cat in between dura and arachnoid, and in one dog in the right lateral ventricle), during the first day between the 6th and the 20th hour. In all these cases the cinnabar was widely distributed, and was deposited in all brain and spinal nerve's roots. Copious lymph cells in the meshes of the pia made me suspect a beginning of, but definitely not a severe, meningitis. No remarkable lesions of the brain and spinal cord itself were seen. The observation over the first hours after the injection did not suggest that so rapid a death should be expected; to the contrary the animals seemed to be very little affected by the surgery and were even feeding some; their movements were completely free, only in one dog, in which the spinal cord was a little injured, the lower legs appeared to be paretic. Unfortunately these animals were not observed during the unexpected death (usually during the night); the assistant reported only about one dog, that the animal was strongly dyspneic shortly before death. What could explain death in those cases? Given that any local lesion in the central nervous system was missing and because of the shortness of time of the injection, one should think first of a sudden increase in the pressure within the brain and spinal cords cavities, – caused by the obstruction of the outflow pathways of the CSF. The injected fluid itself could not have caused it because of its minute volume (1 cc.; in one of the very large dogs even only 0.3 cc.) and no symptoms of increased pressure were
103
directly seen; however as the cinnabar follows the movements of the CSF and distributes everywhere being carried to the outflow places, it can happen very simply that these ways became obstructed; yet such obstruction may remain in the majority of cases only partial, as a sufficient outflow occurs at other outflow locations, under certain circumstances the outflow volumes may be less the secreted volumes; because of the minor compliance of the walls of skull and vertebral cavity an increased pressure may develop centrally, may disturb blood supply and lead to death. The more rapid the obstruction of many outflow pathways the easier this might occur, resulting in fewer possibilities of compensatory enlargement of other outflow regions. Relevant tension of the brain's envelopes and anemia of the brain may not necessarily be found anatomically, because after death part of the fluid may easily transudate into the surrounding tissues. Therefore, although not strictly proven, it appears that an increase in intracranial pressure from the obstruction of the fluid outflow in these cases probably caused the rapid death. It is nevertheless possible, that the pressure of the dye in a specific location (e.g. at a single nerve like the vagus) may also have contributed to death. 4. With regard to the tissular elements it was mentioned above that the cinnabar was contained only in lymph cells and in other similar, but larger, connective tissue cells of variable shape, in the subarachnoideal space as well as in the subvaginal space of the optic nerve. In the actual epithelia of the dura or the arachnoid, the dye was never detected, although frequently, enough cinnabar-containing cells were sitting above the epithelial layer. Little cinnabar was found in large spindle-like cells, which in younger animals form the connective tissue beams of the subarachnoideal tissues, and in the pale epithelial-like, ordered cells, which coat the connective tissue meshed spaces of this membrane. The extraordinary softness and paleness of these structures requires a special treatment to make them visible. I rinsed the end of the subarachnoideal space on the exposed spinal cord with intact dura mater of a freshly killed animal with a silver solution (1:400), followed by a cooking salt solution (1:100), and then hung the spinal cord in alcohol, after the bag of the dura mater was tied at the lower site and at the upper end a glass cannula with a funnel was affixed, so that the subarachnoideal space was widely preserved by the pressure of the alcohol column over several inches. By this procedure a rather nice microscopic picture of silver-brown connective tissue meshwork turned out. Microscopically, there appeared on the connective tissue beams, like at the inner plane of the arachnoid and the outer plane of the pia, the most variable forms of the known silver net, inside of which at some places the weakly brown protoplasma and indications of a core were appearing. I could never assess cinnabar contents in these structures interpreted by anatomists as endothelia; the dye was found always in connective tissue cells or lymph cells, which were laying in another level. The above described experiments were performed at the department of Anatomy here, the use of which was allowed to me by Mr. Geheimrath Reichert in the most liberal way.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/npbr
On the flow dynamics of cerebrospinal fluid§ Karl Bechter a,b,*, Patrick R. Hof c, Helene Benveniste d a
Clinic for Psychiatry and Psychotherapy II, Ulm University, Bezirkskrankenhaus Günzburg, Ludwig-Heilmeyer-Str. 2, D-89312 Günzburg, Germany b Ulm University, Department Psychosomatics/Psychotherapy, Ludwig-Heilmeyer-Str. 2, D-89312 Günzburg, Germany c Fishberg Department of Neuroscience, Hess Center for Science and Medicine, 1470 Madison Avenue, 10th Floor, 10-118, Mount Sinai School of Medicine, New York, NY 10029, United States d Department of Anesthesiology, Stony Brook Medicine, Stony Brook, NY 11794, United States
article info
abstract
Article history:
Despite a not insignificant number of anatomical and experimental studies describing the
Available online 9 March 2015
distribution and movement of the cerebrospinal fluid several questions were answered controversially, leaving room for objections and doubts. Some of these controversies I have
Keywords:
tried to address by new experiments. Before going on to describe these studies, some short
Cerebrospinal fluid
anatomical notes on the membranes that cover the central nervous system may be
CSF outflow
appropriate.
CSF and psychomotor activity CSF flow
Bichat described the anatomical membranes as follows: in between the fibrous lining of the brain-vertebral cavity, the dura mater and the vascularized coat of the spinal cord (the pia mater) a space covered by a serous skin is interposed, the parietal part of which is integrated with the inner side of the dura mater; and the visceral part in contrast is detachable from the underlying pia mater. The watery fluid of the brain-spinal-cord cavity discovered by Cotugno received more attention by Magendie, who initially placed it [the fluid] within the serous space described by Bichat, convincing himself [Magendie] however later, that it [the fluid] is present in between the visceral sheet of the arachnoid and the pia mater. © 2015 Elsevier GmbH. All rights reserved.
Despite a not insignificant number of anatomical and experimental studies describing the distribution and movement of the cerebrospinal fluid several questions were answered controversially, leaving room for objections and doubts. Some of these controversies I have tried to address by new experiments. Before going on to describe these studies, some short anatomical notes on the membranes that cover the central nervous system may be appropriate.
Bichat described the anatomical membranes as follows: in between the fibrous lining of the brain-vertebral cavity, the dura mater and the vascularized coat of the spinal cord (the pia mater) a space covered by a serous skin is interposed, the parietal part of which is integrated with the inner side of the dura mater; and the visceral part in contrast is detachable from the underlying pia mater. The watery fluid of the brain-spinalcord cavity discovered by Cotugno received more attention by
§ Published in 1872 by Dr. H. Quincke, Berlin, in KB Reichert Arch Anat Physiol Wiss Med, 153–177. Translated by Karl Bechter, Patrick R. Hof and Helene Benveniste. * Corresponding author at: Ulm University, Department Psychosomatics/Psychotherapy, Ludwig-Heilmeyer-Str. 2, D-89312 Günzburg, Germany. Tel.: +49 822196 2540; fax: +49 822196 2736. http://dx.doi.org/10.1016/j.npbr.2015.01.001 0941-9500/© 2015 Elsevier GmbH. All rights reserved.
neurology, psychiatry and brain research 21 (2015) 96–103
Magendie, who initially placed it [the fluid] within the serous space described by Bichat, convincing himself [Magendie] however later, that it [the fluid] is present in between the visceral sheet of the arachnoid and the pia mater. Later on, researchers agreed with this assumption and only recently did Henle1 present this point-of-view in his handbook of anatomy, by describing the subarachnoid connective tissue with its mesh-like spaces as a physiological water-filled connective tissue (Virchow) of unusually loose quality; towards the inside it faces the vascular skin (the pia mater), and towards the outside the so-called arachnoidea visceralis. The latter is adhering – typically tightly – to the inner side of the dura mater. Given that the majority of anatomists now agree that the cerebrospinal fluid is located mainly subarachnoidally, there are nevertheless different views about the presence of fluid in the actual arachnoidal sac (in between dura and arachnoidea visceralis). While Luschka [and] Reichert assume an admittedly very small volume of fluid in that very place, Ecker totally refutes that notion in the live animal. This contradiction arises, I assume, from the fact that some [anatomists] preferentially focus on the scenario at the level of the spinal cord, while others focus on the skull. At least in live animals (dogs, cats, rabbits) I found that in the spinal cord the arachoid2 always closely adheres to the dura mater, because during incision of the latter the arachnoid in most cases will be concomitantly injured and the subarachnoid fluid will flow out. One would only rarely exclusively incise the fibrous skin [dura mater]; in this case, the arachnoid protrudes like a hernia due to the fluid content within its meshed spaces; the basis of the wound however is staying dry, because in between arachnoid and dura no fluid is present. Also in dead animals, the view into the spinal canal through the foramen magnum reveals that free connective tissue strings are spawning from the pia to the arachnoid; the latter, however adheres to the dura and can only after overcoming a certain resistance be separated from it. When trying via a puncture to make an injection, I never once accessed the interstice in between the arachnoid and dura, but always ended up in the subarachnoid space. The situation is different at the level of the skull. Here the connective tissue strings between pia and arachnoid are shorter and more robust, the meshed filled spaces (with the exception of the points assigned by Magendie as ``confluents''), narrower; the arachnoid does not stick to the dura, but is separated from it by a capillary fluid layer, therefore in the skull, opening of the dura itself without injuring the [underlying] arachnoid is with some caution relatively easy. [Even] with the dura intact, the distance between both membranes is wide enough to insert a rounded rectangular bent cannula and administer an injection into the actual serous sac in between arachnoid and dura, without injuring the arachnoid. Hence, in the brain, the arachnoid is disconnected from the dura by a capillary fluid layer, whereas at the level of the spinal cord this serous space is non-existent, and instead [in the spinal cord] the arachnoid is tightly adherent to the dura mater. [In the spinal cord], the numerous connective 1
Band III. 2. S. 312. For the sake of brevity I assign herewith the (anatomically exclusively isolated) arachnoidea visceralis. 2
97
tissue bands (denticulate ligament, nerve roots sheets) running from the pia to the dura may have induced this closer relationship. The transition from one anatomical scenario to the other starts at the back rim of the cerebellum and at the medulla oblongata. Experiments reported below will show whether a communication between the subarachnoidal and the arachnoidal mesh spaces, exists. I now continue with the description of experiments where an emulsion of cinnabar was injected at various places into the cavity of the spinal cord of live animals. [For these experiments] I used, as do painters, the finest pulverized cinnabar granulated with sugar water, diluted with five- to ten-fold the volume of water. In a series of animal experiments (exclusively dogs, – cats, rabbits are too small) the cinnabar emulsion was injected into the subarachnoid spaces of the spinal cord cavity via a cannula. Usually this was done by incising the skin above the upper lumbar vertebra and by dissecting the musculature from one side of one or two spinal processes; and when arriving at the vertebral arches, a cannula was passed through the [ligamentum flavum] close to the midline, 1 cm deep, and the cinnabar emulsion was injected. When not considering completely unsuccessful trials, in which case the needle [altogether] missed the spinal canal, successful placement of the tip of the cannula (at the intended spot) depended on pure chance and training. An injury of the spinal cord was usually not an issue, because from such a needle stick the cinnabar may have nevertheless reached the subarachnoidal spaces. Not rarely did it leak through the small, local incision associated with the penetration of the dura mater or – in the case of minute misplacement of the cannula during the injection – directly into the fat-rich loose tissue between the dura and the periost [epidural space]; sometimes it accumulated exclusively in this space; and never advanced further into the space between the dura mater and the arachnoid. After surgery, the skin wound was closed and the animal was left to recover. The behaviour of the animals now was completely different (not considering the after effects of narcosis, which the dogs received in some cases), usually a weakness of one or both lower extremities was observed, presumably due to an injury of the spinal cord. In some dogs this weakness dissipated completely or incompletely over the ensuing days or weeks. The seemingly healthy animals were euthanized after a period of one week to three months (after the injection). In other experiments, if the animals exhibited strong restlessness the spinal cord was more severely injured by the needle stick, [myelitis developed leading to complete paraplegia]; under these circumstances, the animals were usually euthanized after 2–4 days. In other cases although the animals appeared to be feeling well after the injection, they would die during the next 12–24 h. Examination of the brain and spinal cord always revealed dispersion of cinnabar within the subarachnoid connective tissue and in the tissue of the pia mater. In ten out of twelve cases it had travelled to the skull where it preferentially accumulated at the base of the brain, specifically at the site where pia and arachnoid are separated from one another by larger meshed spaces. Beyond, it was observed on [the surface of] the entire brain and spinal nerves if their roots, were
98
neurology, psychiatry and brain research 21 (2015) 96–103
running in the cerebrospinal cavity; the most abundant accumulation was observed at the site where the nerves left the sac of the dura mater (in the skull meaning at the portals into the bony canals). The distribution of cinnabar along the spinal nerves varied from case to case for no apparent reason that could be determined. In a number of cases the cinnabar advanced beyond the cerebrospinal cavity. In about half of the experiments it was observed along the intercostal nerves as far as to where the rami communicans branch off to the sympathetic chain, or even some millimetres beyond. This particular distribution pattern could be observed without any extra preparation after the lungs were removed due to the transparency of the pleura. At the level of the lumbar nerves the cinnabar could in several cases be seen as far away as in the lumbar plexus in between the psoas, and additionally in the sacral plexus beyond its entry into the pelvis. Underneath the brain, the olfactory nerve was not accompanied by cinnabar beyond the cribriform plate. In contrast, cinnabar was consistently observed along the optic nerve, in one case even when the dye was found to be only sparsely dispersed within the pia at the basis of the brain and also not observed along the other brain nerves. The cinnabar accumulation was always most abundant close to the entrance of the optic nerve into the eye, and from there on onwards to the optic foramen it gradually faded, parts of the nerve being completely free of dye. Of the remaining brain nerves the trigeminal nerve in particular has to be mentioned, its ganglion in its intracranial position also being in most cases dyed by cinnabar. Beyond the bony canals I never could demonstrate cinnabar along the brain nerves (except along the above-mentioned optic nerve), also not along the hypoglossus [nerve], [in spite of data from] Key and Retzius3 [who] by artificial injection [noted that] dye travelled [along the hypoglossus nerve] until its entrance into the tongue. In the majority of cases cinnabar was also found in the arachnoidal sheet of the carotid at the point of exit from the cavernous sinus, in the cortical substance of the posterior upper part of the large cervical lymphatic glands, [and] sometimes smaller volumes [were also observed] in the submaxillary lymphatic glands. On the inside, the dura mater was always free of cinnabar, however at certain spots along the venous sinus reddish dyed points or nodules were present, [a topic] on which we shall come back to later. These were the pathways by which the cinnabar after injection into the subarachnoidal spaces of the spinal cord was travelling, and which usually could be visualized with the naked eye. In which tissular elements and spaces it had deposited remains to be described. If the injection was copious and occurred one or few days before the anatomical examination, free cinnabar would be found within the connective tissue meshes, partly in irregular and only microscopically visible accumulations, partly in larger [deposits confined to] pseudomembranes, in which the dye granules were pasted onto one another by a fine granular amorphous substance (fibrin?). In those collections one found more or less copious cinnabar-containing lymphocytes; these 3
Virchow-Hirsch Jahresbericht 1870. Centralbl. f. die medic. Wissenschaft 1871.
were [also] observed in association with free cinnabar distributed in more compact tissue of the pia. Frequently lymphocytes were present within the sheets around the small arteries, even more frequent however, they were [present] independent of these. The number of lymph bodies in the connective tissue of the pia and the subarachnoidal meshed spaces were undoubtedly in many cases increased [compared to normal], in that a beginning meningitis was present. In addition, cinnabar was found in rounded or irregularly shaped cells, a bit larger than lymphocytes, heterogeneously distributed in the subarachnoid tissue, sometimes in solitary presentation, sometimes in groups, and which presumably have to be considered some type of connective tissue cell. The longer the animal lived, the more the free cinnabar seemed to fade and to transit intracellularly. In this regard, the pia and the subarachnoid of the brain and the spinal cord behave rather similarly. Transport of cinnabar into the brain or the spinal cord parenchyma was also microscopically never confirmed. The choroid plexus was in the majority of cases free of dye; in only two cases did the choroid plexus of all four ventricles contain cinnabar and then it presented in a particular way, with remarkable little dye in the stroma, but was instead overwhelmingly present in the epithelium. In one case, in which the choroid plexus was not stained, there was a clot of cinnabar-containing lymphocytes in the fourth ventricle. Around nerve roots both from the brain and the spinal cord, the cinnabar was found in the lymphoid cells (lymphoiden Zellen), observed in lymph bodies (Lymphkörper) along the surface [of the nerve], the free dye always occurring in more or less large heaps very close to the sites where the nerves exited the dura mater. Where cinnabar was found beyond these places, it seemed always to be included in round cells of the nerve sheet. Cinnabar was never found deposited into the trunk of the nerves, or in the trigeminal ganglion and spinal ganglia. The lymph glands contained free cinnabar as well as within the round cells. In vain did I search for cinnabar in the nasal mucosa and the spleen. In the labyrinth cinnabar was found only in one out of five cases investigated, in the tympanic membrane of the cochlea. The cinnabar reached during the injection not only the subarachnoid space but also beyond (or exclusively) in the tissue situated between the dura and the periost, where it distributed from the back plane of the vertebral bodies (more or less farther), often as far as to the neck; it was often guided even further away by the current of the fluid [CSF]; it appeared in the lumbar lymph glands situated along the aorta, in the subclavian and mediastinal glands, and also at the inner plane of the intercostal muscles, so that it shone through the pleura. While the initial course of cinnabar transport from the subarachnoid space appears to keep the exact course of the intercostal nerves, its general movement out of the vertebral canal is less clear, progressing more diffusely and a little bit farther into the intercostal spaces below the pleura, sometimes seemingly following the course of the azygos vein. In the cellular tissue of the vertebral canal the cinnabar is partly contained in lymph bodies, partly freely in the tissue, even after weeks. Because neither during life nor after death
neurology, psychiatry and brain research 21 (2015) 96–103
The most plausible cause of distribution of cinnabar in live animals to be considered might be the up-and-down revolving (respiratory) movement of the subarachnoid fluid as was [originally] proposed by Magendie. Some researchers (e.g., Leyden)4 doubted that such a movement of fluid5 [CSF] was normal and interpreted it to be a consequence of the opening of the brain/spinal cord cavities, yet this argument does not apply to the majority of my experiments, because during the injections with a cannula in the lumbar vertebral trunk, the normal anatomy was not altered at all, and in the skull, the small opening was immediately closed by a wooden pin, and only in those cases in which the dura had lost the respective bony supporting plane near the trepanation site, a slightly more compliant area was established. The distribution of cinnabar was however the same for all of the procedures. Incidentally, in children a membranous closure of the skull at some places is normal. Moreover, in adults as well the skull does not represent a completely stiff enclosure, like would a metal or glass bottle with only a single opening; rather, the multiple holes of the skull are filled out by smooth tissue, associated with some compliance, making it possible to change the capacity of its cavity. Within the sheet of the optical nerve we clearly see a protuberance of the subarachnoid spaces with its flexible walls. The way the respiratory movement influences the [CSF] fluid has been clearly described by Ecker6; the flexible venous plexus of the spinal cord canal sinks more strongly with inspiration, and is expanding during the expiration when compared to the stiff-walled sinus of the skull; by this [one can infer that] during expiration the cerebrospinal fluid will be driven from the spinal cord to the skull, whereas during inspiration the reverse is taking place. Only on one point do I have to contradict Ecker, in that by this (movement) a fluctuating filling and emptying of the brain ventricles [is to be also considered], and accordingly then the cinnabar constantly had to be found in the ventricular spaces. It may well be that the weight of the cinnabar contributes in itself somehow to its accumulation at the base of the brain, nevertheless, from its distribution, we will overall recognize the pathways of the fluid's currents within the subarachnoid spaces. It can be expected that such CSF currents occur preferentially in the more spacious parts of the skull – in the so-called subarachnoid sinus (``confluents'', Magendie), and that from there only weak [CSF] currents occur to the hemisphere convexity. When the injection happened over a hemisphere, the dye distributed preferentially to the quadrigeminal plate and the base [of the brain]; only smaller amounts of dye reached the convexity of the contralateral hemisphere consistently. The ascending current from spinal cord to brain appears in general to be stronger than the descending, because the
the cinnabar did not induce major inflammatory changes, one feels permitted to consider that its distribution pattern represents natural pathways of [CSF] flow, which the products of pathological deposits from the vertebral canal, as in the case of caries [=osteomyelitis as the likely meaning; the authors] of the vertebral bones, will also preferentially follow. In a second series of experiments the cinnabar emulsion was injected into the skull cavity, in rabbits and cats, as well as in dogs. To this end, the dura was opened, using a small trocar [6 mm diameter] or a simple pointed tool, and afterwards the opening was closed by a wooden pin. The animals were examined (after spontaneous death or euthanasia) a few hours to eight days after the surgery. If the cinnabar emulsion was injected into the skull by a straight cannula, it advanced unavoidably through the arachnoid and pia into the brain, the dye could thus advance into the subarachnoid as well as into the arachnoid spaces. When using a very thin, cannula bent in a quarter of a circle, it was possible to introduce this under the exposed dura (the opening inflected towards it) and to avoid injuring the arachnoid as taught by the [previous] sessions; as such, the dye was injected only into the actual arachnoidal space (between dura and visceral layer of the arachnoidea). After a few days most of the dye disappears to be found in the subarachnoid spaces and in the pia of the brain, exactly as if [the dye] had been directly injected into this space. The dye is deposited preferentially at the base of the brain, as well as at the side of the convexity of the brain adjacent to the injury; beyond this, it is found at the exit points of the cranial nerves, in the carotid sheet, and in the cervical lymph glands. In the majority of cases it advances also into the vertebral canal and distributes in a variable manner, often as far as the cauda equina, depositing in the pia mater and at nerve roots, as described in the first series of experiments. Eventually, the distribution of the dye is the same whether it is injected into the subarachnoid spaces of the spinal cord or into those of the brain; the only difference is quantitative, inasmuch as a more copious deposition in the first case is found in the skull and in the other case, in the spinal cord canal. From these experiments I believe that the following conclusions can be drawn. 1. There exists a relationship between the subarachnoid spaces of the brain and the spinal cord. 2. In the live animal, the current of the subarachnoid fluid flows in a reverse direction from the back to the front. Dye particles, in the [subarachnoid] fluid, will be transported from the spinal cord to the brain as well as from the brain to the spinal cord. Evidence supporting that this transport process is not exclusively mediated by migration of lymph bodies is based on the fact that in many experiments deposits of fine [free] cinnabar were found at places remote from the injection site. Control experiments in dead animals demonstrated that following the injection itself or following passive movements (to and fro swinging of the legs among others) the cinnabar never distributed as diffusely as in living animals; when injecting the same amounts around the lumbar vertebra, cinnabar reached at most the cervical marrow, and if injecting the skull, at most the medulla oblongata.
99
4
Virchow's Jahresbericht XXXVIL. I confirm that I am speaking only about a movement of the fluid, not about a movement of the brain or the spinal cord itself. Such a one is, at least for the brain, disproven by the experiments with insertion of a glass window (Donders, Leyden and others). 6 Physiologische Untersuchung über die Bewegung des Gehirns und Rückenmarks. Stuttgart 1843. 5
100
neurology, psychiatry and brain research 21 (2015) 96–103
progression of the cinnabar in the former direction was more substantial and consistent than in the latter. 3. Also, after injection into the arachnoidal spaces of the skull, the dye advanced into the subarachnoidal spaces of the brain and the spinal cord, indicating that these two spaces must be communicating. In the live animal the CSF flow has to be preferentially from the arachnoidal to the subarachnoidal spaces, as with injections into the latter in the spinal cord, cinnabar was never found in the space between the dura and arachnoid. 4. The outflow pathways of the CSF. A fraction of CSF leaves the brain/spinal cord cavity along the [cranial and peripheral] nerves. This assumption appears to me to be explained by the consistent accumulation of dye where the nerves exit the dura mater/arachnoid. Free dye was transported by the CSF current to these exits, and dye accumulating within cells was most likely taken up by these `on route', because there were typically very few cells, which probably represent migrating lymphocytes. At the sites where the nerves exit, the dura and/or arachnoidal membranes appear to be more tightly adhering to the nerves, because beyond this point of exit the cinnabar was never observed along the [cranial] nerves,7 only occasionally along the lumbar nerves, and most frequently along the intercostal nerves. Hence, pathways seem to exist which allow CSF to track along nerves beyond their exit points from the brain and spinal cord, although only under certain circumstances do these pathways permit solutes to pass. In case of a meningitis, purulent matter might easily access these pathways and by exerting pressure on the nerves may cause inflammatory disturbances. In addition, the flow of CSF is passing through the cervical lymph glands, the upper posterior portion of which in the live dog has a special relationship to pia and arachnoid of the brain and upper cervical cord, as well as to the maxillary lymph glands. Whether the subarachnoidal spaces need be consequently considered as a lymph space or only as an appendix of the lymphatic vascular system, may be left to anatomists to decide. Another outflow pathway for the arachnoidal and subarachnoidal CSF was found by Key and Retzius8 in the Pacchioni' granulations. According to them these granulations do not simply attach to the dura mater, but are, as Trolard9 described, free protrusions, coated exclusively by an epithelia, and localized partly in the veinous sinuses, and partly in their lateral protuberances. In my experiments these structures were also marked, when the injection occurred in the vertebral column or the skull, by a strong cinnabar staining; these were regularly present in the superior sagittal sinus as well as in the transverse sinus; I also found at times similar structures in the cavernous sinus. According to their structure, described in more detail by Key and Retzius as similar to that of lymph glands, these granulations seem to function as a filtration system, which may well let fluid pass while however retaining large 7 A special case is, the N. opticus, the behavior of which has to be discussed separately. 8 A. a. O. 9 Arch. Gen. 1870. Mars.
particles. Had any substantial number of dye particles passed through these, they would have arrived into the blood stream, and consequently would have plausibly been deposited in the spleen; never however have I succeeded to find cinnabar in this organ. The cinnabar did not pass the vascular network in the dura described by von Boehm10 in my experiments. Thus one has to ascribe to the subarachnoidal fluid a double movement: probably secreted by the blood vasculature it is flowing under a certain pressure on defined pathways into the lymphatic vasculature. Which of these pathways are preferentially used, is as variable as the blood stream within the vasculature, depending on time and situations; in general the distribution of cinnabar suggests that the draining pathways out of the skull are preferred compared to that of the spinal cord. The pressure, exerted upon the fluid, is varying over time and differs at different locations as a to-and-fro movement is taking place; the respiratory movements are effective as a pump on the venous blood, the lymph stream within the pleura, and so on, which, together with the pressure propagated from the arteries, is pushing the fluid through its draining pathways. It is clear that these constant movements of the fluid within the continuously connected meshed spaces has to favor the distribution of pathological products; thus, we see not rarely a purulent meningitis distributing with high velocity from one point all over the whole central nervous system. Nevertheless, purulent infiltration of the pia mater localized to certain areas can presumably be explained by the fact that the dye particles are usually suspended more freely within the fluid compared to purulent matter, the latter sticking to the walls of the meshed-like space and therefore not easily cleared away by the fluid currents [compared to the free dye particles]. At least I found, that especially in cases where the distribution of cinnabar was more localized, an inflammatory process had developed, typically at the site of injection; rather continuous red-colored purulent fibrinous masses filled up the subarachnoidal space over a certain distance; microscopically the cinnabar was partly found in pus accumulations, partly in between such pus bodies inlaid within a fine granular substance. Generally, inflammation of the pia after cinnabar injection was however rare, particularly when the marrow [we think that they used the term ``marrow'' for white matter, similarly brain and spinal cord; the authors] itself was little affected; it occurred more frequently after injections into the skull, with injury of the outer layers being more considerable. Some animals were euthanized only 2–3 months after the injection; the wide deposition of dye described above had not disturbed their well-being. The wide distribution of dye seems to me to reflect the advantageous nature of the cerebrospinal fluid. As is well known, it is according to its anatomical distribution and to its chemical nature, most closely connected to the fluid of edematous tissues and contains, in contrast to the fluids of serous cavities like the pleura and the peritoneum, few solid 10
Virch. Arch. Bd. 47.
neurology, psychiatry and brain research 21 (2015) 96–103
By injections partly from the arachnoideal space, partly from the eye bulb, Schwalbe16 demonstrated a system of hollow spaces in this location, which are interconnected and communicate with the skull cavity. Schwalbe has the space in between the inner (firmly adhering to the nerve) and the outer (loose, mobile) sheet of the optic nerve, the subvaginal space, communicate with the arachnoideal space, which is tube-like, ending blindly at the bulb and interlaced by beams like the subarachnoideal spaces (!); Its walls is covered with an epithelia. A second lymphatic space, supravaginal, situated in between the outer sheet and the M. retractor bulb, is funnel-shaped, being related anteriorly to the Tenon's space and posteriorly again with the subarachnoideal room; the structure and place of the latter are not very clear from the description by Schwalbe, because he explicitly assigns the outer sheet of the optic nerve as a continuation of the dura mater. Key and Retzius differentiate the two tube-like spaces at the optic nerve, the inner of those communicating with the subarachnoideal space, the outer with the arachnoideal (subdural) space, both being interconnected. H. Schmidt17 tried, based on Schwalbe's and his own injection experiments, to explain the development of the papilledema in brain diseases by an accumulation of fluid in the sheet and a resulting entrapment of the optic nerve. Finally Manz18 made this last point in particular an issue of experimental and patho-anatomic examinations and agreed with Schmidt's explanation. He found a stronger accumulation of serum in the optic sheet whenever the intracranial pressure is increased, and even when ophthalmic alterations are not yet visible; in one case of hemorrhagic pachymeningitis. He saw an infiltration of the optic nerve sheet with blood, which well seemed to have been carried there from the skull cavity. My experiments demonstrate now that a transfer of subarachnoideal flow into the optic nerve sheet has to be considered a constant and essentially normal event. This is because the dye, which was mixed to this fluid in vivo, was nearly exclusively found in the optic nerve sheet, whether the injection had happened in the skull or deep below in the spine; even in case of very minor injections in the spine the dye appeared in the optic nerve, while in other cranial nerves not dye was observed, only the large meshed spaces at the brain basis containing some; the vicinity of the latter to the optic nerve's origin apparently favors the movement of the fluid to this area. Yet after a few hours a mass-like accumulation of cinnabar in the optic sheet could occur; a part of the dye is then always free; only in smaller amounts may it be taken up by endothelial cells, which coat the sheet's spaces and the connective tissue strings crossing it. It is found most constantly and in greater amounts in the ``subvaginal space'' (Schwalbe) at its blind end, close to its entrance into the eyeball; sometimes the more centrally situated part of the optic nerve is even free.
elements, namely little protein and no fibrinogen; the formation of clots, which in such cavities may easily trap the cinnabar and isolate it, cannot occur in CSF. With regard to the continuous relationships of the subarachnoid spaces of the brain and the spinal cord the experiments presented here confirm the observations of Magendie11 and Luschka,12 which were recently confirmed by Axel Key and Retzius.13 Later authors have described the relationships of these named spaces with other parts. However, because they used non-physiological injection methods in living and dead animals, it may easily have been performed by using stronger pressure so that a passage of fluid into wrong pathways occurred. They could have accessed spaces, indeed naturally communicating with the arachnoideal and the subarachnoideal spaces, which laid within the trajectories of the physiological fluid movement but upstream from the natural flow, and thus did not serve as draining pathways [under normal conditions]. Either is excluded in my experiments because only little fluid (0.2–1, rarely some cubic centimeters) was injected at once. The forces used during the injection therefore could not widely distribute the dye, its transport remaining left to the normal forces effective in the living body. This demonstrates therefore the natural CSF current pathways within the cerebrospinal cavity with certainty, as well as its outflow pathways. That in various situations these take different directions and therefore resulting in some disparities, has been mentioned above when describing the experiments. However in none of my experiments the dye, compared to Key's und Retzius' experiments, advanced into the central canal of the spinal cord, in the perivascular rooms of the brain14 and the spinal cord parenchyma, so it is therefore quite probable that under normal circumstances these spaces pour their contents into the subarachnoid spaces, and thus do not receive flow from there. The same is valid for the lymphatic vasculature of the olfactory mucosa, as well as for Tenon's space, the perichoroidal space, and the lumbar glands, in which Key, Retzius, and Schwalbe15 saw the injection mass advancing from the arachnoideal space. Here the possibility remains that under certain circumstances these paths may sometimes serve as collateral draining pathways. In contrast to Key and Retzius [who] injected nerve sheets I have only in part (see above) and only exceptionally found some dye; perhaps the current is strong enough to carry the dye so far only in the case of heavier pressure. Some points touched only shortly thus far are to be discussed in more detail: 1. The relationships of the optic nerve and its surroundings to the skull cavity.
11
Rech. Phys. et Chim. Sur le liq. Céphalorachidien. Paris. 1842. Die Adergeflechte des menschlichen Gehirns. Berlin. 1855. 13 A. a. O. 14 Only in one case, where presumably the cannula happened to be placed below the pia, there were cinnabar containing lymph bodies in the area of the injection after three days in the perivascular space of the cerebral cortex. 15 Archiv f. mikroskop. Anat. Bd. VI.
101
12
16 17 18
A. a. O. Arch. f. Ophthalmol. XV. 1869. Arch. f. Ophthalmol. Bd. 16. 1870.- Arch. f. klin. Medic. IX. 1871.
102
neurology, psychiatry and brain research 21 (2015) 96–103
Cinnabar was also found in the supravaginal space, yet more rarely, partly free, partly enclosed into cells; in some cases it had even progressed into the posterior aspect of Tenon's space; I never could found it in the suprachoroid space. The separation between supra- and subvaginal spaces appears not to be very strict, because I often found cinnabar-containing cells in the outer optic nerve sheet; there seems therefore that a communication exists between them and with the subarachnoideal space through the connective tissue bundles in between them; I never saw examples that would fit with the description of a arachnoideal and subarachnoideal sheet according to Key and Retzius, after simple injection [in these spaces}, so that I cannot decide whether the actual arachnoideal space (subdural space according to Key) is continuing towards the optic nerve, or whether such communication is the fact of only by the subarachnoideal space. The subvaginal space appears completely separated anteriorly near the eyeball, as was revealed by the injections. In the retina, cinnabar was never found, as in the optic nerve itself. The space between the nerve fibers injected by H. Schmidt would therefore serve only for influx into the hollow space of the optic sheet. One may wonder which of either currents occurring in the CSF may carry the dye into the optic sheet; that the respiratory to and fro movement of the fluid has to be involved, was already considered as probable, because the optic sheet represents the largest protuberance of the cerebrospinal cavity with compliant walls.19 Nevertheless depending on the predominant pressure conditions here (and towards the Tenon's space) a continuous outflow might take place as well, as in other cranial nerves. The fact that an exchange of fluid in between the optic sheet and the subarachnoideal spaces is physiologic suggests that the relationships between diseases of the brain and of the optic nerve are closer than previously assumed. Propagation towards the optic nerve sheet is rather likely in case of hemorrhage into or in between the meninges, or in case of purulent infiltration of the weak brain envelopes. Frequently, but always unsuccessfully, I attempted to observe the initial stages of a papilledema after cinnabar injections; apparently the resulting pressure increase within the skull cavity were not sufficient to induce the phenomena observed by Manz after injection; or these may have happened (see below) so rapidly as to result in death. 2. The fluid of the subarachnoideal spaces of brain and spinal cord seems rarely to advance Into the connective tissue stroma of the choroid plexus, which is continuous with the meshed tissues of the arachnoid, as cinnabar was found there only in three of about 20 experiments of this type. – This is in agreement with the observation that in purulent meningitis, even widely distributed, purulent infiltration of the plexus chorioidei is found extremely rarely.
19
I want especially to remark again that for the other nerves at the eyebulb cinnabar was never found beyond the holes of the skull.
The brain ventricles never contained free cinnabar even in case of near complete distribution of the dye; only occasionally some was found in the ventricles as flake-like purulent deposits, by which it was presumably transported there. A fluid current into the ventricles can therefore not take place according to this negative finding just as little as an alternating in-and-out flow movement as assumed by Magendie, Ecker, and others to represent a side effect of the current of the actual subarachnoid fluid that depends on respiration.20 The possibility of an opening of the IV. ventricle towards the posterior side remains the only possibility left for a continuous fluid current out of the IV. ventricle into the subarachnoideal space; this is a possibility, which is supported by the opinion of those anatomists, who suggested the choroid plexus to represent a secretory organ. For a closer focus on this point I tried to inject cinnabar emulsion (0.08–0.3 cc.) directly in the lateral ventricles in a number of living dogs, by introducing through a narrow borehole a cannula until to a certain depth; after the injection was performed, the defect in the bone was closed by a wooden pin, the skin sewed and after 1–25 days an anatomical examination was performed. Although even in successful experiments, an injury of the subarachnoideal tissues at the convexity and at the plexus itself was unavoidable, I believe nevertheless to be able to conclude from the distribution of the cinnabar in the four ventricles, as well as in the arachnoideal tissues of the medulla oblongata and the neighboring structures, that mainly it may have advanced through the third ventricle and the aqueduct of Sylvius into the fourth ventricle and beyond into the subarachnoideal tissue. Accordingly, it seems that a natural fluid current is taking place in the described direction, seemingly induced by a secretion of the choroid plexus, the pathological increase of which we see in children's hydrocephalus. It appears that open pathways must exist along which the fluid carrying cinnabar is advancing from the ventricles into the subarachnoideal spaces. In humans such a communication in the fourth ventricle was described by Magendie, Luschka and others, in the form of an opening (sometimes multiple) of several lines diameter (Magendie's foramen), also reported by others such as Burdach, and Reichert21 as the tela chorioidea posterior being a continuous membrane, completely tightening the ventricle; in the horse Luschka also agrees with this description. Magendie's statements about the situation in animals are mutually conflicting. Because I was not myself successful in discovering in dogs a preexisting well-bordered opening in the tela chorioidea posterior, my experiment having however demonstrated the existence of connecting pathways, these would have to be sought in the interstices of the connective tissue bands, which constitutes the pia. The variability in the strength, robustness, and arrangement of these connective tissue bundles among various species and 20 The observation of Magendie, who observed at the opening of the ventricles a rising and falling of fluid synchronous with breathing, does not contradict this. 21 Bau des menschlichen Gehirns II. S. 53.
neurology, psychiatry and brain research 21 (2015) 96–103
individuals of the same species, may explain the discrepancies in the statements of other authors. A far reaching significance of these variations is in any case not to be assumed; the foramen of Magendie is nothing more than a variable hole of the connective tissue, which is sometimes not even patent. Also in the surrounding of the great vein of Galen, where the subarachnoideal tissue is more loose (``confluent supérieur'' of Magendie) the ventricle seems to communicate with the subarachnoideal spaces (not to be mistaken with the so-called Bichat's hole, which was considered a communication towards the actual arachnoideal space but turned out to be an artifact). Twice after injections into the lateral ventricle cinnabarcontaining lymph bodies were found in the central canal of the spinal cord, even as far as to the lumbar region. I cannot explain with certainly how in three cases the cinnabar could advance into the epithelia of the choroid plexus after subarachnoideal injections. It was such an aggregation in those cases, that the plexus appeared completely reddish with the naked eye; yet it was found only sparsely in singular lymph bodies in the tissue of the plexus. Because the cinnabar was not taken up by the epithelia of the plexus after the injections into the ventricular cavities, it is probable that it had advanced in those cases not from the free but from the lower side into the epithelial cells. Henle described in humans yellow or reddish dyed bodies in these cells, which he believed to be derived from red blood cells and which would have similarly moved in from below into the epithelial cells. This may represent a rather similar event as in my experiments with the cinnabar. 3. In some (4) experiments death occurred after cinnabar injections in the vertebral column (in two dogs) as in the skull (in one cat in between dura and arachnoid, and in one dog in the right lateral ventricle), during the first day between the 6th and the 20th hour. In all these cases the cinnabar was widely distributed, and was deposited in all brain and spinal nerve's roots. Copious lymph cells in the meshes of the pia made me suspect a beginning of, but definitely not a severe, meningitis. No remarkable lesions of the brain and spinal cord itself were seen. The observation over the first hours after the injection did not suggest that so rapid a death should be expected; to the contrary the animals seemed to be very little affected by the surgery and were even feeding some; their movements were completely free, only in one dog, in which the spinal cord was a little injured, the lower legs appeared to be paretic. Unfortunately these animals were not observed during the unexpected death (usually during the night); the assistant reported only about one dog, that the animal was strongly dyspneic shortly before death. What could explain death in those cases? Given that any local lesion in the central nervous system was missing and because of the shortness of time of the injection, one should think first of a sudden increase in the pressure within the brain and spinal cords cavities, – caused by the obstruction of the outflow pathways of the CSF. The injected fluid itself could not have caused it because of its minute volume (1 cc.; in one of the very large dogs even only 0.3 cc.) and no symptoms of increased pressure were
103
directly seen; however as the cinnabar follows the movements of the CSF and distributes everywhere being carried to the outflow places, it can happen very simply that these ways became obstructed; yet such obstruction may remain in the majority of cases only partial, as a sufficient outflow occurs at other outflow locations, under certain circumstances the outflow volumes may be less the secreted volumes; because of the minor compliance of the walls of skull and vertebral cavity an increased pressure may develop centrally, may disturb blood supply and lead to death. The more rapid the obstruction of many outflow pathways the easier this might occur, resulting in fewer possibilities of compensatory enlargement of other outflow regions. Relevant tension of the brain's envelopes and anemia of the brain may not necessarily be found anatomically, because after death part of the fluid may easily transudate into the surrounding tissues. Therefore, although not strictly proven, it appears that an increase in intracranial pressure from the obstruction of the fluid outflow in these cases probably caused the rapid death. It is nevertheless possible, that the pressure of the dye in a specific location (e.g. at a single nerve like the vagus) may also have contributed to death. 4. With regard to the tissular elements it was mentioned above that the cinnabar was contained only in lymph cells and in other similar, but larger, connective tissue cells of variable shape, in the subarachnoideal space as well as in the subvaginal space of the optic nerve. In the actual epithelia of the dura or the arachnoid, the dye was never detected, although frequently, enough cinnabar-containing cells were sitting above the epithelial layer. Little cinnabar was found in large spindle-like cells, which in younger animals form the connective tissue beams of the subarachnoideal tissues, and in the pale epithelial-like, ordered cells, which coat the connective tissue meshed spaces of this membrane. The extraordinary softness and paleness of these structures requires a special treatment to make them visible. I rinsed the end of the subarachnoideal space on the exposed spinal cord with intact dura mater of a freshly killed animal with a silver solution (1:400), followed by a cooking salt solution (1:100), and then hung the spinal cord in alcohol, after the bag of the dura mater was tied at the lower site and at the upper end a glass cannula with a funnel was affixed, so that the subarachnoideal space was widely preserved by the pressure of the alcohol column over several inches. By this procedure a rather nice microscopic picture of silver-brown connective tissue meshwork turned out. Microscopically, there appeared on the connective tissue beams, like at the inner plane of the arachnoid and the outer plane of the pia, the most variable forms of the known silver net, inside of which at some places the weakly brown protoplasma and indications of a core were appearing. I could never assess cinnabar contents in these structures interpreted by anatomists as endothelia; the dye was found always in connective tissue cells or lymph cells, which were laying in another level. The above described experiments were performed at the department of Anatomy here, the use of which was allowed to me by Mr. Geheimrath Reichert in the most liberal way.