Journal of Neuro-Ophthalmology, December 1999, Volume 19, Issue 4

Lymphatic Capillaries in the Meninges of the Human Optic Nerve

OBJECTIVE: Although many anatomical studies of the orbit and the optic nerve have been performed, lymphatic capillaries in the dura of the human optic nerve have never been reported. This study was performed to determine whether or not lymphatic capillaries are present in the dura of the human optic nerve. MATERIALS AND METHODS: This postmortem study was carried out in seven subjects without ocular disease. The subjects were obtained no later than 6 hours after death, following qualified consent for autopsy. The dura of the human optic nerve was studied with light microscopy, scanning electron microscopy, and transmission electron microscopy. In some cases, india ink was injected into the subarachnoid space as a marker. RESULTS: Lymphatic capillaries in the dura of the human optic nerve were morphologically demonstrated with histological criteria (fenestrated endothelium, lack of a basal membrane, and absence of blood cells in the lumen of the vessels). The highest concentration of lymphatic capillaries was found in the bulbar part of the dura behind the ocular globe. Using light microscopy and transmission electron microscopy, ink was seen within the lumen of the lymphatic capillaries. The dura itself was not stained with the marker. CONCLUSION: The presence of lymphatic capillaries in the dura of the human optic nerve was demonstrated with light microscopy, transmission electron microscopy, and scanning electron microscopy.

Journal of Neuro- Ophthalmology 19( 4): 222- 228, 1999. © 1999 Lippincott Williams & Wilkins, Inc., Philadelphia Lymphatic Capillaries in the Meninges of the Human Optic Nerve H. Esriel Killer, M. D., Hubert R. Laeng, M. D., and Peter Groscurth, M. D. Objective: Although many anatomical studies of the orbit and the optic nerve have been performed, lymphatic capillaries in the dura of the human optic nerve have never been reported. This study was performed to determine whether or not lym­phatic capillaries are present in the dura of the human optic nerve. Materials and Methods: This postmortem study was carried out in seven subjects without ocular disease. The subjects were obtained no later than 6 hours after death, following qualified consent for autopsy. The dura of the human optic nerve was studied with light microscopy, scanning electron microscopy, and transmission electron microscopy. In some cases, india ink was injected into the subarachnoid space as a marker. Results: Lymphatic capillaries in the dura of the human optic nerve were morphologically demonstrated with histological cri­teria ( fenestrated endothelium, lack of a basal membrane, and absence of blood cells in the lumen of the vessels). The highest concentration of lymphatic capillaries was found in the bulbar part of the dura behind the ocular globe. Using light micros­copy and transmission electron microscopy, ink was seen within the lumen of the lymphatic capillaries. The dura itself was not stained with the marker. Conclusion: The presence of lymphatic capillaries in the dura of the human optic nerve was demonstrated with light micros­copy, transmission electron microscopy, and scanning electron microscopy. Key Words: Cerebrospinal fluid papilledema- Lymphatic channels- Optic nerve meninges- Subarachnoid space. The optic nerve ( ON), a white matter tract of the cen­tral nervous system ( CNS), extends from the cranial cav­ity through the optic foramen and into the orbit. The ON, an integral part of the CNS, bears an envelope of men-ingothelial cells and is surrounded by cerebrospinal fluid ( CSF). The subarachnoid space ( SAS) is lined with a Manuscript received February 17, 1999; accepted May 10, 1999. From the Department of Ophthalmology ( H. E. K.) and Pathology ( H. R. L.), Teaching Hospital for the University of Basel and Bern, Kantonsspital Aarau, Switzerland; and the Institute of Anatomy ( P. G.), University of Zurich, Switzerland. Supported in part by an unrestricted grant to the Department of Ophthalmology and Visual Sciences, Montefiore Medical Center, Al­bert Einstein College of Medicine, New York, New York, from Re­search to Prevent Blindness, Inc., New York, New York, U. S. A. Address correspondence and reprint requests to H. Esriel Killer, M. D., Kantonsspital Aarau, CH- 5001, Aarau, Switzerland. fenestrated layer of meningothelial cells known as the neurothelium. The connective tissue of the meninges contains collagen fibrils and elastic fibers ( 1,5,8,21,32). It also harbors a dense network of arteries, veins, and unmyelinated nerves ( 5,8). The arachnoid villus is con­sidered to be the major site of CSF absorbtion ( 15,16). Other sites and mechanisms of CSF absorption have been proposed in the literature ( 5,14,23,24). Schwalbe was the first author to suggest that CSF drainage into lymphatic channels is a possible CSF outflow pathway ( 10). To find evidence of a CSF draining pathway within the meninges of the intraorbital part of the human optic nerve, we studied the ultrastructure of the dura of the human ON before and after injecting india ink into the SAS of the ON. MATERIALS AND METHODS This postmortem study was carried out in seven sub­jects without ocular disease. The subjects were obtained no later than 6 hours after death, following qualified consent for autopsy. Specimen Preparation and Labelling On removal of the orbital roof, the ocular globes, to­gether with the optic nerves and the chiasm, were care­fully dissected in situ from surrounding tissues. The optic nerves were ligated with a 6.0 silk suture proximal to the optic chiasm. Subsequently, the fixative ( either neutral buffered 4% formalin or 2.5% glutaraldehyde) was in­jected slowly into the SAS with a 19- gauge needle. Spe­cial care was taken to avoid high- injection pressure, to minimize the risk of creating artefacts. In two cases, to label the lymphatic vessels within the dura, india ink dissolved in 8% formalin ( vokvol = 1: 1) was injected slowly under low pressure into the SAS at the level of the midorbital segment of the ON. The intact specimens were fixed in 4% formalin by immersion for 1 to 7 days before further dissection. Light Microscopy From each eye, a single piece including the midorbital and bulbar segment of the ON ( Fig. 3A) was processed for paraffin blocks and cut in sections 5- 8 fxm thick. The 222 LYMPHATIC CAPILLARIES 223 stains included haematoxylin and eosin, van Gieson elas-tin, and Masson trichrome. Scanning Electron Microscopy ( SEM) For injection and immersion fixation, 2.5% glutaral-dehyde solved in 0.05 mol/ L cacodylate buffer was use. Transverse sections of the midorbital and bulbar seg­ments were dehydrated in an acetone series, dried by the critical point method ( C02), mounted on aluminum stubs, and sputtered with gold ( approximately 30 nm). The specimens were analyzed with an SEM 505 ( Philips, Einthoven, the Netherlands) at an accelerated voltage of 20 kV. Transmission Electron Microscopy ( TEM) After injection of 2% glutaraldehyde ( 0.1 mol/ L cac­odylate buffer) into the SAS, the globe and optic nerve were further fixed for at least 1 week by immersion in the same solution. Subsequently, small fragments ( approxi­mately 1 mm3) were cut from the optic nerve ( midorbital and bulbar segments) and postfixed for 1 to 2 days in 1 % Os04 ( 0.1 mol/ L sodium phosphate buffer). The speci­mens then were dehydrated in an alcohol series and em­bedded into epon. Semithin sections were cut from each block ( approximately 1 ( xm) and stained with toluidine blue to identify the meninges. Ultrathin sections ( ap­proximately 50 nm) were contrasted with uranyl acetate and lead citrate and studied with a CM 100 transmission electron microscope ( Philips, Einthoven, the Nether­lands). RESULTS Scanning Electron Microscopy In cross sections of the optic nerve, the arachnoid was found partially detached from the dura, forming an arti­ficial subdural space ( Figs. 1A and IB). This artefact was caused by shrinkage of the specimen during SEM prepa­ration. However, in each cross section, areas were found with the arachnoid lining in close contact to the dura, thus allowing detailed analysis of normal morphology of the SAS. Distinct differences in the SEM appearance of the SAS were detectable between the midorbital and bulbar segments of the optic nerve ( Fig. 1). In the midorbital segment, the SAS was bridged by coarse arachnoidal pillars that occasionally showed blood vessels running from dura to pia, and vice versa ( Fig. 1 A). In the bulbar segment of the ON, the SAS was significantly widened ( Fig. IB). The SAS contained a delicate network of branched trabeculae connecting the dural and pial sur­face of the arachnoid lining. At higher magnification, the surface of the arachnoid cells appeared smooth, with no microvilli or other cell processes. However, small oval clefts ( 0.1 to 0.3 | xm) were found in the dural level of the arachnoid layer ( Fig. 1C). The number of lymphatic capillaries varied dis­tinctly between the optic nerve segments. Lymphatic capillaries rarely occurred in the SAS of the midorbital part but were often detectable in the bulbar segment of the ON. FIG. 1. Scanning electron microscopy appearance of optic nerve and adjacent meninges ( cross sections). A: Midorbital segment with a few arachnoid pillars spanning between the dura and the pia of the narrow subarachnoid space; original magnification x12. B: Bulbar segment displaying wide subarachnoid space with deli­cate network of trabeculae; original magnification x12. C: High magnification of arachnoid surface lining the dura, displaying multiple pores; original magnification x500. J Neuro- Ophthalmol, Vol. 19, No. 4, 1999 224 H. E. KILLER ETAL. FIG. 2. Transmission electron microscopy morphology of lym­phatic capillaries found in the dura. The extremely flat endothelial cells lack a basal lamina and are in direct contact with the mod­erately electron- dense extracellular matrix. The perinuclear cyto­plasm of an endothelial cell displays few organelles and single lipid droplets ( A, asterisk). Small intercellular pores ( B, arrow) are usually found along the endothelial lining; original magnification A x5500, B x6000. Transmission Electron Microscopy The dura was examined carefully by TEM to establish the morphology of the various vessel types. Blood cap­illaries with continuous endothelium supported by a well- defined basal lamina, as well as arterioles, small arteries, and veins were detectable, although not very frequently. In addition, lymphatic capillaries were found and could easily be distinguished from blood capillaries by their typical ultrastructure ( Fig. 2). The lymphatic capillaries were usually contiguous with collagen fiber bundles and embedded into an amorphous, moderately electron- dense extracellular matrix. The lymphatic cap­illaries were lined by extremely flat endothelial cells that lacked a basal lamina. The oval, heterochromatin- rich nucleus of the endothelial cells was surrounded by a thin layer of cytoplasm with few organelles. Occasionally, small aggregates of lipid droplets and lipofuscine gran- FIG. 3. Labelling experiment, light microscopy. A: Gross mor­phology of india ink injected specimen. Note the concentric wid­ening of the bulbar segment of the optic nerve. The lines indicate the levels where specimens were taken for light and electron microscopy. B: Midorbital segment of the optic nerve: india ink is clearly visible at the surface of the arachnoidal layer ( arrows) and within slit- like vessels of the dura ( arrow head); original magnifi­cation x20. C: Bulbar portion of the optic nerve displaying dense network of labelled vessels within the dura. Asterisk indicates the lumen of the subarachnoid space; original magnification x40. J Neuro- Ophthalmol, Vol. 19, No. 4, 1999 LYMPHATIC CAPILLARIES 225 x- v */&&?^€* f* ^ « l S I S ^ ^ f c ^ . © ''•--- 3 ^'^ v^' \ ThOx * . : • . •< • • • . ? : • » . * tf » i ^ f . 5 . FIG. 4. Labelling experi­ment, transmission elec­tron microscopy appear­ance. A: Arachnoid layer: ink particles are detectable in the subarachnoid space ( arrows) as well as in the intercellular space between arachnoidal cells ( arrow heads). Note the distinct in­tercellular cleft between the luminal cells, which is ap­parently the site of escape of injected ink; original magnification x25,000. B: Lymphatic capillary in the dura. Ink particles ( arrows) are visible in the extracellu­lar space ( E), in interendo-thelial pores ( P), and in the lumen ( L) of the lymphatic capillary; original magnifi­cation x25,000. ules were found in the perinuclear cytoplasm. The endo­thelial lining of the lymphatic capillaries was interrupted by small interendothelial pores, often found where the extracellular matrix was in direct contact with the elec­tron- translucent lumen of the vessel ( Fig. 2B). Labelling Experiments India ink was injected into the SAS to reveal a pos­sible CSF drainage function of the lymphatic capillaries found in the dura. Gross inspection of the injected speci­mens clearly showed the dye as black deposits in the SAS beneath the dura ( Fig. 3A). The ink deposits local­ized anteriorly at the level of the lamina cribrosa and the sclera remained unstained. Furthermore, differences in the extension of SAS between the midorbital and the bulbar part of the optic nerve became distinctly more pronounced. The bulbar segment regularly showed a spherical widening and a blind end at the level of the lamina cribosa, reflecting a cul de sac shape. With light microscopy, the india ink appeared as a black- stained deposit within the SAS and at the surface of the arachnoid layer ( Figs. 3B and C). Labelling was further detectable within slit- like lymphatic channels of the adjacent dura. The stained lymph channels mainly were found in the inner third of the dura. Prevalence and distribution of lymphatics varied distinctly between the J Neuro- Ophthalmol, Vol. 19, No. 4, 1999 226 H. E. KILLER ET AL. orbital segments studied. In the midorbital segment, a few unbranched lymph channels were usually detectable ( Fig. 3B), whereas the dura of the bulbar segment dis­played a complex network of labelled lymphatic vessels ( Fig. 3C). To follow possible lymphatic drainage routes of india ink, we studied the injected specimens with TEM ( Fig. 4). The india ink was easily detectable as electron- dense granules of approximately 30 nm in diameter. Careful examination of the arachnoid layer revealed circum­scribed areas where ink particles could be found spread­ing through intercellular clefts between the arachnoid cells ( Fig. 4A). In addition, ink particles were detectable in the extracellular space of the dura, in close vicinity to endothelial cells of the lymphatic capillaries ( Fig. 4B). Labelling was also found in the interendothelial pores and in the lumens of the lymphatic capillaries ( Fig. 4B). DISCUSSION The ON represents an integral white- matter tract of the CNS. The ON can be divided into the intraorbital, the canalicular, and the intracranial portions. The bulbar seg­ment, or ampulla, is part of the intraorbital portion and is attached to the sclera of the ocular globe. The ON is surrounded by a meningeal envelope. The SAS of the ON communicates distal of the intracanalicular portion, with the chiasmal cystern and ends blind in the bulbar portion at the level of the lamina cribrosa, resembling a cul de sac. Cerebrospinal fluid inflow into the orbital portion of the SAS of the ON is supplied from the chi­asmal cystern, and a reverse flux is to be expected, pro­vided that CSF escape by alternative routes is insignifi­cant. There is general agreement that the choroid plexus epithelium and the ependymal cells of the ventricular system are the principal sources of CSF production ( 15, 16,26). An important site of CSF drainage into the major dural sinus is in the archnoid villus, which can be found in the optic nerve of humans and monkeys ( 15,33). Other sites of absorption, including the choroid plexus ( 23,24), the capillaries, the intercellular space of the brain ( 5), and the lymphatic channels have been discussed in sci­entific literature ( 10,11,17). The idea of a direct CSF penetration from the SAS through the arachnoid mem­brane and into the dura has found little acclaim in the past. In 1869, Schwalbe was the first author to demon­strate communication of the cranial SAS with the cervi­cal lymphatic system. He introduced Prussian blue " un­der constant unspecific pressure" into the cranial SAS of rabbits, thereby claiming that lymphatic channels were the major drainage pathway for CSF ( 10). Schwalbe, however, did not perform histologic studies to provide morphologic proof for his hypothesis. In a 1989 study, Mc Getrick et al. ( 37) failed to pro­vide evidence for lymphatic vessels posterior to the con­junctiva in a monkey model. Exit of macromolecules from the subarachnoid space at the termination of the optic nerve via " open channals" was described by Erlich ( 38). Brinker ( 40) provided evidence for CSF outflow along the optic nerve in rats, cats, dogs, and monkeys, after dye injection into the cisterna magna. Using en­zyme histologic, light microscopic, and electron micro­scopic studies, Sherman et al. ( 39) demonstrated lym­phatic vessels in the conjunctiva, the extraocular muscles, the lacrimal gland, and the archnoid trabeculae. In 1994, Zenker et al. ( 9,11) published the concept of diffuse absorption of CSF through the spinal meninges in the rat; cationized ferritin injections into the SAS of rat dorsal roots that showed active transport of this tracer into the surrounding dura were used. The morphologic counterpart for this observation was the presence of lym­phatic clefts in the dura matter of the meningeal funnels in the rat ( 9). Foldi demonstrated lymphatic vessels in the lacrimal system, the conjunctiva, and the cornea ( 4), as well as in the skull base ( 13,18,19). He suggested that lymph drainage occurred from orbital structures into the nodal system of the neck. Experimental support for his concept of CNS lymphatics resulted from his observation of relief of papilledema upon blockage of the stellate ganglion in dogs. The dogs previously were subjected to ligation of lymph vessels and of their expected tributary lymph nodes in the neck ( 2,3,6,7,12). Brierly ( 14) intro­duced ink into the cranial SAS that appeared in mid thoracic and cervical lymph nodes as well as in the lum­bar and sacral nerve roots. Although CSF drainage into lymphatics has been shown to exist in the CNS ( 34,35, 36,40), it was never shown to exist in the human optic nerve; the existence of a lymphatic system in the optic nerve was strongly opposed by Hayreh ( 20), who denied the existence of lymphatic vessels in the entire central nervous system. To investigate the structures possibly involved in CSF drainage in the meninges of the intraorbital portion of the human ON, we performed the present morphologic study. We were able to demonstrate that ink enters into lymphatic capillaries in the dura of the human ON upon injection of the SAS, possibly via slit- like pores in the neurothelial layer of the arachnoid membrane, indicating a functional drainage system for CSF from the SAS of the ON into the dura. Because the blind end of the SAS in the bulbar segment of the ON resembles a cul de sac at the level of the lamina cribrosa, CSF turnover is ex­pected to be very slow in that narrow compartment, es­pecially if there is no drainage system in this minute compartment. Our discovery of intradural lymphatic cap­illaries, which are predominantly located in the bulbar part of the ON, and the results of our tracer studies in­dicate CSF drainage from the SAS of the bulbar part of the ON into the meninges of the ON. Little is known about the CSF pressure in the SAS of the optic nerve. Until now, the concept of a homoge­neous pressure in the entire CSF compartment, including ventricles, SAS, and cysterns, has not been seriously challenged. Direct measurements of CSF pressure in the SAS of the ON are technically difficult to perform and tend to be unreliable because of the small size and tra-beculation of this compartment ( 25). Indirect methods, such as ultrasound studies of the diameter of the ON J Nenro- Ophlhalmol, Vol. 19, No. 4, 1999 LYMPHATIC CAPILLARIES 227 under increasing volume pressure, may be a promising approach in the future ( 26,27). Inividual sheath elasticity and individual degree of trabeculation and septae be­tween the arachnoid and pia in the SAS, however, may limit the value of such studies. To prevent accumulating CSF pressure in the bulbar part of the ON, which may cause papilledema and prob­ably pressure- related malperfusion of the feeding pial arteriolar system of the ON ( 22), a functional CSF out­flow system appears to be desirable. In addition to arach­noid villi in the meninges of the ON ( 33), our morpho­logic findings provide another anatomic basis for such a drainage system. Intradural lymphatic vessels further may help to understand phemomena such as unilateral papilledema ( 28,29), pseudotumor cerebri without pap­illedema ( 30), pseudotumor cerebri with normal intra­cranial pressure ( 31), or the pathophysiology of the ret­rograde part of axonal degeneration in normal tension glaucoma ( private communication with J. Flammer, De­partment of Ophthalmology, University of Basel). This study may give further histologic evidence for the route of CSF outflow from the SAS of the optic nerve that has been described in the literature ( 38,41). Additional stud­ies are necessary to define the physiologic purpose of lymphatic capillaries in the meninges of the human optic nerve. Acknowledgements The authors thank P. Gerber ( Institute of Pathology, Kan-tonsspital Aarau, Switzerland) for careful preparation of the optic nerves used in this study, as well as P. Rosenbaum, M. D., ( Albert Einstein College of Medicine, Bronx, New York, NY), who prepared some of the nerves used in this study for electron microscopy studies. We also thank R. M. Burde ( Albert Einstein College of Medicine, New York, NY) for continuous inspiration, as well as for reading this manuscript. Special thanks to Professor W. Zenker, who inspired us with his previous publications. REFERENCES 1. Greaney MJ. The fine anatomy and ultrastructure of human optic nerve meninges. Poster presented at: Annual Congress of the Royal College of Ophthalmologists; 1995; Birmingham, England. 2. Foldi M, Foldi E, Van Husen H. Lymphostatische Retinopathie beim primarem Lymphodem. In Odem. Jahresband 1986. ( eds. Prov. Vorstand der Gesellschaft Deutschsprachiger Lympholo-gen). Perimed Fachbuch- Verlagsgesellschaft mbH, D- 8520 Erlan-gen. Perimed Fachbuch- Verlagsgesellschaft Erlangen. 3. Foldi M. Lymphostatische Enzephalopathie und Ophthalmopathie. Lehrbuch der Lymphologie fur Meclizine!' und Physiolhciapeuten, Gustav Fischer Verlag Stuttgart Jena New York 1993. 4. Foldi M, Casley- Smith JR. Lymphangiology. New York: Schat-tauer Verlag Stuttgart, 1983. 5. Zenker W. Htillen des Zentralnervensystems, Ventrikelausklei-dung und Liquor cerebrospinalis. In Benninghoff Anatomie Band 2, 1994 Urban & Schwarzkopf, Munchen, Wien, Baltimore. 6. Varkonyi T, Polgar J, Zoltan OT, Csillik B, Foldi M. Lymphostatic retinal heamangiopathy. Experientia 1970; 26: 67- 8. 7. Foldi M, Szenes T, Kalian A. Experimental lymphography by means of a subarachnoidal injection of lipiodol ultrafluid. Experi­entia 1967; 23: 455- 7. 8. Andres KH. Uber die Feinstruktur der Arachnoidea und Dura mater von Mammalia. Zeitschrift fiir Zellforschung 1967; 79: 272- 95. 9. Zenker W, Bankoul S, Braun JS. Morphological indications for considerable diffuse absorption of cerebrospinal fluid in spinal meninges particularly in the areas of meningeal funnels. Anat Em­bryo! 1994; 189: 243- 58. 10. Schwalbe G. Der Arachnoidalraum ein Lymphraum und sein Zusammenhang mit dem Perichorioidalraum. Centralblatt fur medicinische Wissenschaften 1869; 30: 465- 7. 11. Schwalbe G. Uber Lymphbahnen der Netzhaut und des Glaskor-pers. Ber. Ges. Wiss. Leipzig, Math. Phys. Klasse, 1872; 24: I42. 12. Foldi M, Szeghy G, Csanda E. Relief of papilledema by blockage of the stellate ganglion. Lancet 1962; 512. 13. Foldi M. Die Rolle der Lymphzirkulaltion im Saftekreislaluf des Auges und des Zentralnervensystems. Archiv fill" Kreislauffor-schung. Arch. Kreisl.- Forsch 1963; 41: 186- 212. 14. Brierley JB, Field EJ. The connexions of the spinal sub- arachnoid space with the lymphatic system. J Anat 1948; 82: 153- 66. 15. Weed LH. The absorbtion of cerebrospinal fluid into the venous system. Am. J Anat 1923; 31: 191- 221. 16. McComb JG. Recent research into the nature of cerebrospinal fluid formation and absorbtion. J Neurosurg 1983; 59: 369- 83. 17. Hoffmann E, Thiel W. Untersuchungen vermeintlicher und wirkli-cher Abflusswege aus dem Subdural- und Subarachnoidalraum. Zeitschr. fur Anat. und Entwicklungsgesch 1956; 119: 283- 301. 18. Foldi M, Gellert A, Kozma M. New contributions to the anatomi­cal connections of the brain and the lymphatic system. Acta Anat 1966; 64: 498- 505. 19. Foldi M, Csillik B, Zlotan OT. Lymphatic drainage of the brain. Experientia 1968; 24: 1283- 7. 20. Hayreh SS. Pathogenesis of oedema of the optic disc. Doc Oph­thalmol 1966; 24: 289- 411. 21. Klika E. The ultrastructure of meninges in vertebrates. Acta Univ Carol [ Med] 1967; 13: 53- 71. 22. Neetens A. Vascular supply of the optic nerve. A practical survey for the clinician. J Newoophthalmol 1994; 14: 113- 20. 23. Hashimoto PH, Gotow T, Ichimura T. Visualization of the cere­brospinal fluid drainage into Galen's vein. Arch. Histol. jap 1985; 48( 2): 173- 81. 24. Spector R, Johanson CE. The mammalian choroid plexus. Set Am 1989; 261: 48- 53. 25. Liu D, Kahn M. Measurement on relationship of subarachnoidal pressure of the optic nerve to intracranial pressure in fresh cadav­ers. Am J Ophthalmol 1993; 116: 548- 56. 26. Hansen HC, Helmke K. Validation of the optic nerve sheath re­sponse to changing cerebrospinal fluid pressure: ultrasound find­ings during intrathecal infusion tests. J Neurosurg 1997; 87: 34- 40. 27. Hansen HC, Helmke K. The subarachnoid space surrounding the optic nerves. An ultrastructural study of the optic nerve sheath. Surg Radiol Anat 1996; 18: 323- 8. 28. Strominger MB, Weiss GB, Fehler MF. Asymptomatic unilateral papilledema in pseudotumor cerebri. J Neuroophthalmol 1992; 12 ( 4): 238- 41. 29. Sedwick LA, Burde RM. Unilateral asymmetric optic disk swelling with intracranial abnormalities. Am J Ophthalmol 1983; 96: 484- 7. 30. Marcelis J, Silberstein SD. Idiopathic intracranial hypertension without papilledema. Arch Neurol 1991; 48: 392- 9. 31. Green JP, Newman NJ. Stowe Z. " Normal pressure" pseudotumor cerebri. J of din Neuro- ophthalmol 1996; 16( 4): 241- 6. 32. Anderson DR. Ultrastructure of meningeal sheaths. Normal human and monkey optic nerves. Arch Ophthalmol 1969; 82: 659- 74. 33. Shanthaveerappa TR, Bourne GH. Arachnoid villi in the optic nerve of man and monkey. Exp Eye Res 1964; 3: 31- 5. 34. Bradbury MWB, Cserr HF, Westrop RJ. Drainage of cerebral in­terstitial fluid into deep cervical lymph of the rabbit. Am J Physiol 1981; 240: 329- 36. 35. Bradbury MWB, Cole DF. The role of the lymphatic system in drainage of cerebrospinal fluid and aqueous humor. ./ Physiol 1980; 299: 353- 65. 36. Bradbury M. Lymphatics and the central nervous system. Trends Neurosci 1981; 4: 100- 1. 37. McGetrick JJ, Wilson DG, Dortzcnbach RK. A search for lym- J Neuro- Oplillialmol, Vol. 19, No. 4, 1999 228 H. E. KILLER ETAL. phatic drainage of the monkey orbit. Arch Ophthalmol 1989; 107: 255- 60. 38. Ehrlich SS, McComb J, Hyman S. Ultrastriicture of the orbital pathway for cerebrospinal fluid drainage in rabbits. J Neurosurg 1989; 70: 926- 31. 39. Sherman DD, Gonnering RS, Wallow HL. Identification of orbital lymphatic: Enzyme histochemical light microscopic and electron microscopic studies. Ophthal Plast Reconstr Surg 1993; 9: 153- 69. 40. Brinker T, Liidemann W, Berens von Rautenfeld D. Dynamic properties of lymphatic pathways for the absorbtion of cerebrospi­nal fluid. Arch Neuropathol 1997; 94: 493- 8. 41. Brinker T, Botel C, Samii M. The perineural pathway of cerebro­spinal fluid absorption in the cervical lymphatic system. In: Nagai H, Kamiya K, Ishii K, eds. Morphological findings in rats, dogs and monkeys. Intracranial Pressure IX. Tokyo, Japan: Springer; 1994: 132- 5. J Neuro- Ophthalmol, Vol. 19, No. 4, 1999