| OCR Text |
Show Papilledema: The Vexing Issues Jonathan D. Trobe, MD Abstract: Papilledema has long been recognized as a valuable sign of increased intracranial pressure (ICP). But because papilledema is based on interruption of the energy-dependent process of axoplasmic flow, it appears late after a rise in ICP. Papilledema is usually present in chronically high ICP but sometimes asymmetrically in the 2 eyes and rarely in 1 eye only. Distinguishing it from other optic neuropathies that produce elevated optic discs is challenging, especially in the chronic phase, when visual function may be impaired. Papilledema is often an unrecognized cause of optic disc edema in inflammatory and compressive meningeal disorders that interfere with cerebrospinal fluid (CSF) passage through the arachnoid granulations. Its detection is particularly critical in patients with noncompliant ventricles or extraventricular blockage of cerebrospinal flow because imaging may fail to disclose conventional signs of high ICP. Therefore, patients with indwelling CSF shunts, tuberous sclerosis, chronic granulomatous meningitis, or meningiomatosis should be periodically examined for papilledema so that timely ICP-lowering measures can be instituted to pre-serve vision. Journal of Neuro-Ophthalmology 2011;31:175-186 doi: 10.1097/WNO.0b013e31821a8b0b 2011 by North American Neuro-Ophthalmology Society William Fletcher Hoyt, MD, trained more than 70 fellows and coauthored the biblical textbook Clin-ical Neuro-Ophthalmology, bringing pathophysiologic rigor to a discipline that had been rich in anecdote but thin in science. His acolytes, many of whom became the leaders of the next generation of neuro-ophthalmologists, are still looking over their shoulders at the master. I am grateful for the invitation to deliver the ninth Hoyt Lecture in his honor and on a subject that greatly interested him. Papilledema, or optic disc edema resulting from high intracranial pressure (ICP), has always intrigued physicians not merely because it may be the only accessible sign of high ICP but because it behaves differently from other forms of optic disc edema. Visual function in papilledema is rela-tively spared in the early stages. But if high ICP continues unabated, permanent visual impairment may occur. Who will suffer this consequence and why the optic nerves should be so selectively vulnerable to high ICP are puzzles that remain unsolved. Even the mechanism by which high ICP gives rise to papilledema is still a mystery. I will review the development of ideas about the path-ophysiology of papilledema and show how they apply to the vexing clinical issues. IDEAS ABOUT PAPILLEDEMA Before the 20th century, physicians labeled all forms of optic disc edema ‘‘optic neuritis.'' But it was apparent that the optic disc edema associated with brain tumors lasted relatively long and produced relatively little visual impair-ment at outset as compared to other types of optic neu-ropathy. Yet if the brain tumors were unattended, disabling visual loss might occur much later, when the optic discs were flattening and turning pale. Accordingly, early in the 20th century, the term ‘‘papilledema'' was adopted to distinguish this form of optic disc edema (1,2). Physicians of various specialties put forward their ideas about how high ICP caused papilledema (1,2). Among the imaginative but ultimately incorrect postulates were com-pression of the veins of the cavernous sinus, transudation of parenchymal brain edema or third ventricular cerebrospinal William Fletcher Hoyt, MD, received his medical education at the University of California, San Francisco (UCSF), followed by a year's study at the Wilmer Institute, Johns Hopkins University, under the mentorship of Frank B. Walsh, MD. Dr. Hoyt returned to UCSF in 1958 to found the neuro-ophthalmology service. During a 36-year academic career-all of it at UCSF-he authored 266 journal articles, co-authored (with Frank B. Walsh, MD) the third edition of Clinical Neuro- Ophthalmology, and trained 71 neuro-ophthalmology fellows. In 1983 he received the title of Honorary Doctor ofMedicine from the Karolinska Institute. Currently he serves as Professor Emeritus of Ophthalmology, Neurology, and Neurosurgery at UCSF. In recognition of his contributions, in 2001 the North American Neuro-Ophthalmology Society (NANOS), in conjunction with the American Academy of Ophthalmology, initiated the Hoyt Lecture to be delivered each year at the Annual Meeting William F. Hoyt, MD of the American Academy of Ophthalmology. Departments of Ophthalmology and Visual Sciences and Neurology, University of Michigan Kellogg Eye Center, Ann Arbor, Michigan. The author reports no conflicts of interest. Address correspondence to Jonathan D. Trobe, MD, University of Michigan Kellogg Eye Center, 1000Wall Street, Ann Arbor, MI 48109; E-mail: jdtrobe@med.umich.edu Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 175 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. fluid (CSF) into the optic chiasm and thence into the optic nerves, leakage of vessels within the optic discs from va-somotor instability, and a toxic substance within the CSF. In 1910 (3) and 1911 (4), Leslie Paton, an ophthalmol-ogist, and Gordon Holmes, a neurologist, reported the results of the first definitive light microscopic study of papilledema. It came from 50 eyes and optic nerves of 39 patients who had died of brain tumors at the National Hospital for the Par-alysed and Epileptic in London (now known as the National Hospital for Neurology and Neurosurgery, Queen Square). In this landmark publication, the authorsmade 3 important observations: 1) the optic discs and the retrolaminar and distal retrobulbar nerves contained excessive edema, which they as-sumed was extra-axonal, 2) the prelaminar axons were ‘‘vari-cose'' and ‘‘fragmented'' where they were most deviated by the optic disc swelling, and 3) the central retinal vein was dilated in the retrolaminar and prelaminar optic nerve but flattened in the subarachnoid space around the optic nerve (Fig. 1). From these observations, they concluded that papilledema came from compression of the subarachnoid portion of the central retinal vein by high ICP. This mechanical process, they believed, led to dilation of the vein's optic disc segment. Leakage from this dilated vein and its feeding capillaries caused prelaminar optic disc swelling and bending of axons, which sometimes led to their fracture. The Paton-Holmes venous compression doctrine held sway for most of the 20th century. However, critics asked why retinal edema, hemorrhages, and cotton wool spots-features associated with occlusion of the central retinal vein-were not usually evident with the papilledema. Moreover, optic disc edema is known to be an unimpressive finding even in severe central retinal vein occlusion (2). FIG. 1. The Paton-Holmes light microscopic study of the eyes of 39 patients who had died of brain tumors. A. Cross section of representative swollen optic disc. Prelaminar optic disc edema; the authors assumed it was extra-axonal. B. Prelaminar swollen and fragmented axon (arrow). C. Postlaminar dilated central retinal vein (arrow). D. Collapsed central retinal vein (arrow) in the subarachnoid space behind the eye. The authors posited that the high ICP caused the retrobulbar vein to collapse, leading to dilation of the perilaminar vein, its leakage, and secondary optic disc swelling. They believed that the swelling induced extreme bending of the axons and their breakage. (Modified from Paton and Holmes (4).) 176 Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. In 1948 came the discovery that organelles normally flow back and forth between cell bodies and their synaptic terminals in an energy-dependent process called axoplasmic transport (5). In 1976 and 1977, a flurry of publications reported electron microscopic studies on the anatomy and physiology of optic nerve axons and axoplasmic transport in acute glaucoma, ocular hypotony, and increased ICP (6-9). In all 3 conditions, electron microscopy showed edema of the optic disc that was mostly intra-axonal. Studies of axoplasmic flow within the optic nerve, as measured by the progress of tritiated leucine injected into the vitreous cavity and in-corporated into retinal ganglion cells, showed that axoplasm was arrested in the region of the lamina cribrosa (Fig. 2). Considering that these 3 conditions share a high pressure gradient across the lamina cribrosa-albeit in opposite directions in acute glaucoma and ocular hypotony/increased ICP-investigators reasoned that this abnormal pressure gradient caused or contributed to the axoplasmic pileup. As an explanation for papilledema, the doctrine of ve-nous hypertension gave way to the doctrine of axoplasmic stasis. But still unanswered was the question of how high ICP caused this axoplasmic stasis. Was it by direct compression of axons (mechanical theory) or by reduced perfusion of axons (ischemic theory) (1,2,10-13) (Fig. 3)? Although the mechanical theory has generally been favored, there is compelling support for the ischemic theory. For example, experimental occlusion of the ciliary arteries leads to axoplasmic stasis and axonal swelling (14,15). Fluorescein angiography in patients with papilledema shows delayed filling of the optic disc and peripapillary choroidal vessels (11). Patients with arteriosclerosis seem to be espe-cially prone to optic neuropathy in the setting of chronic idiopathic intracranial hypertension (IIH) (16). There are some tantalizing similarities between the optic neuropathy of papilledema and the optic neuropathy of acute systemic hypotension. First, there is a shared proclivity to affect the inferior arcuate nerve fiber bundles and to spare the maculopapillar bundles. Second, just as papilledema may be the only important neurologic finding in chronic high ICP, so ischemic optic neuropathy may be the only important neurologic finding in acute systemic hypotension (17). These phenomena converge on the ciliary arterial circle, which might be the site of acute hypoperfusion in systemic hypotension and chronic hypoperfusion in sus-tained high ICP (Fig. 3). Why might the ciliary arterial circle be so vulnerable to low blood pressure and high ICP? Because it competes with the choroidal circulation, which draws off a voluminous flow to furnish oxygen to the highly metabolic visual transduction process and to dissipate the heat that trans-duction generates. Under normal circumstances, there is enough arterial blood flow to go around. But if systemic blood pressure falls or subarachnoid pressure rises, the optic nerve might suffer from ‘‘choroidal steal.'' Notably, one postulate as to how optic nerve sheath fenestration protects the optic nerve in papilledema is that it forms a scar around the nerve that isolates the ciliary arterial circle from trans-mission of high subarachnoid pressure (12). The issue about whether axons lose function from compression or ischemia also applies to primary open-angle glaucoma (POAG), where it remains unresolved. Papilledema and POAG share a pressure gradient across the lamina cribrosa and visual field loss that affects principally the arcuate bundles in the nasal field with sparing of visual acuity. But there is a profound difference between papilledema and POAG: the optic neuropathy of chronic papilledema does not include excavation of the neuroretinal rim tissue. Thus, there must be something else besides chronic ischemia to the pathogenetic mech-anism for POAG (18). The relationship between axoplasmic stasis and optic nerve dysfunction is uncertain. The optic nerve is evidently FIG. 2. The Tso-Hayreh electronic microscopic studies of papilledema in rhesus monkeys whose intracranial pressure had been elevated by implantation of an intracranial balloon. A. Prelaminar optic disc shows some very distended axons (asterisks). (Modified from Tso and Hayreh (7), Arch Ophthalmol. 1977;95:1448-1457, copyright 1977 American Medical Association. All rights reserved.) B. Autoradiograph of optic nerve from a rhesus monkey eye enucleated 6 hours after intravitreal injection of tritiated leucine. Silver grains (arrows) have accumulated in perilaminar region as compared to control eyes (not shown). (Modified from Tso and Hayreh (8), Arch Ophthalmol. 1977;95:1458-1462, copyright 1977 American Medical Association. All rights reserved.) Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 177 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. able to conduct visual signals adequately as long as the stasis is not severe, given that visual function is relatively preserved in papilledema in the early stages-even when there is considerable optic disc elevation. Axoplasmic stasis is not unique to papilledema, having been found or surmised in many other optic neuropathies (19). But in papilledema, the axoplasm presumably accu-mulates more slowly as optic nerve dysfunction usually takes a longer time to develop. THE VEXING ISSUES Is papilledema a reliable clinical indicator of a recent rise in intracranial pressure? No-at least not in humans. In one monkey experiment, ICP elevation induced by subarachnoid balloon inflation led to the development of papilledema in 30% within 24 hours and in 90% within 120 hours. The faster the balloon was inflated and the higher the ICP, the greater the like-lihood and degree of papilledema (20). But in humans, papilledema lags. Among 37 patients who had documented high ICP from acute intracranial hemorrhage in trauma or ruptured aneurysm and who were examined for several days in an intensive care unit, only 1 patient had papilledema. Five patients had peripapillary hemorrhages, a phenomenon known as Terson syndrome (21). Other studies have affirmed that fewer than 20% of patients examined within a few days of sustaining head trauma (22) or ruptured aneurysm (23) have papilledema. The shortcoming in the evidence about the prevalence of papilledema in acutely high ICP is that authors do not consistently report how long they continued per-forming ophthalmoscopy in these patients. The delay in the appearance of papilledema in acutely elevated ICP is consistent with the idea that papilledema is not based on dilated veins, as is the case in Terson syn-drome, but rather on interruption of the metabolic process that mediates axoplasmic flow. Based on examination of patients in a neurointensive care unit for several decades, I believe that Terson syndrome is more common than pap-illedema within the first few days of a sudden rise in ICP. I have also found papilledema to be generally absent in hy-drocephalic children who sustain acute shunt failure, a topic that requires further study. Is papilledema a reliable indicator of chronically high ICP? Yes, but there are no published studies that correlate papilledema with ICP. Older studies found that papilledema occurred in 50%-80% of brain tumors (1). That range is now probably an overestimate as modern brain imaging detects tumors before ICP has risen enough to cause papilledema. FIG. 3. The mechanical and ischemic theories of optic nerve axoplasmic stasis. A. Normal ICP: The optic nerve axons are of normal caliber (a) because axoplasm flows normally through the laminar optic nerve. The arterial supply of the laminar optic nerve (b) is adequate. B. In-creased ICP: mechanical theory of axoplasmic stasis. The high ICP compresses the optic nerve at its junction with the eye and distends its prelaminar axons (a) as their axoplasm piles up at the scleral lamina. The arterial supply (b) is not affected. C. Increased ICP: ischemic theory of axoplasmic stasis. The high ICP compresses the ciliary circulation and reduces the arterial supply to the laminar optic nerve (b), which competes with the high flow to the choroid. The optic nerve axons are distended (a) as chronic ischemia interferes with the metabolic process of axoplasmic flow. CSF, cerebrospinal fluid. 178 Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. In a study of 252 brain tumors, Paton (1) found that papilledema appears more rapidly in cerebellar tumors than cerebral tumors. That observation fits with my experience, namely that a very large amount of supratentorial mass effect is necessary to cause papilledema, whereas a relatively inconspicuous tumor may cause elevated ICP and papil-ledema if it blocks convexity arachnoid granulations (24), ventricular CSF egress (25,26), or dural venous sinuses (27). The papilledema of chronically high ICP may be of different severity in the 2 eyes. It may be entirely absent from 1 eye, but rarely from both eyes! The frequency of strictly unilateral papilledema has never been adequately documented, but in my experience, it occurs in fewer than 5% of cases. This phenomenon is attributed to anatomic variations that impede transmission of ICP through the optic canal to the distal optic nerve. How often is papilledema completely absent in chronically high ICP? Studies of IIH indicate that about 6% of such patients lack papilledema (IIH without papilledema or IIHWOP) (28). The weakness of this evidence is that without papilledema as an anchor, the diagnosis of IIH depends on the accuracy of the lumbar puncture opening FIG. 4. Postdecompression optic neuropathy. A. Optic discs show a mixture of swelling and pallor. B. Precontrast T1 sagittal MRI discloses hydrocephalus from a suprasellar ganglioglioma (arrow). C. Following surgery, optic discs eventually became flat and pale and vision did not recover (right eye: finger counting, left eye: no light perception). Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 179 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. pressure measurement, which is egregiously error-ridden (29). Are these patients suffering from chronic migraine or tension headache and generating falsely high opening pressures because of poorly performed lumbar punctures? Reversal of headache with ICP-lowering treatment in IIHWOP, often used as support for the diagnosis, is not convincing. Can patients develop optic neuropathy from chronically high ICP if they do not have papilledema? No. In a series of 20 patients with IIHWOP (28), many had constricted visual fields, but all were judged to be ‘‘nonphysiologic,'' that is, faked. Apparently, optic neu-ropathy will not occur unless the optic disc swells to the point of being visible on ophthalmoscopy. Thus, IIHWOP remains an issue of headache, not vision loss. Can a sudden rise in ICP cause vision loss from optic neu-ropathy without first causing papilledema? Such a phenomenon has never been properly docu-mented, but there are hints that it may rarely occur. In a study of the long-term visual outcome in 30 patients with Terson syndrome (30), 2 patients developed optic disc pallor and markedly impaired vision. In those patients, papilledema was not described but may have been hidden behind peripapillary retinal hemorrhages. Although persistent visual loss in Terson syndrome in that series was attributed mostly to epiretinal membrane, cystic macul-opathy, or macular hole, a contribution from optic neuropathy may have been overlooked. I have encountered a patient who developed severe vision loss and optic disc pallor after postpartum dural sinus thrombosis. A retinal specialist had examined the patient several times during the first 14 days after the thrombosis and found no papilledema! I urge my colleagues to look out for this phenomenon and to report it if they come upon it. Can papilledema linger after ICP has normalized? Yes. Because papilledema is based on axoplasmic stasis, which is a failure of an energy-dependent metabolic process, it lags not only on the way up but also on the way down. FIG. 5. Nonpapilledematous optic disc swelling in malignant hypertension. A. Bilateral hemorrhagic optic disc edema is evident in these photographs performed with a hand-held camera at presentation in a 14-year-old boy with acute glomerulonephritis. Ophthalmoscopy also disclosed serous retinal detachments and cotton wool spots and Elschnig spots. B. Six months after successful treatment, standard fundus photographs show normal-appearing optic discs and small scattered retinal pigment epithelial defects (arrows). Visual fields were normal. (Modified from Besirli et al (47), with permission.) 180 Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. The attendant visual deficit may linger as well. As a conse-quence, errors in patient management may occur, and I have made them. For example, a young man developed transient ob-scurations of vision in the left eye and had asymmetric optic disc elevation and nerve fiber bundle visual field loss worse in the symptomatic eye. MRI showed triventriculomegaly from aqueductal stenosis owing to a presumed low-grade tectal astrocytoma. The patient underwent third ventriculostomy. Three weeks after the procedure, the visual obscurations had ceased, but the papilledema was unchanged and the visual fields had only slightly improved. I suggested that the ven-triculostomy had failed and recommended ICP monitoring. It was normal. Eight weeks later, the papilledema and visual fields had finally recovered. The correct diagnosis, which had eluded me, was prompt resolution of high ICP and delayed resolution of papilledema. In many cases, papilledema never goes away-even months after ICP has normalized. A 40-year-old woman shunted for hydrocephalus in infancy had undergone several shunt revisions for recurrent headache. She came under my care for a new headache with features similar to those that had triggered previous shunt revisions. Visual acuity was normal, visual fields were unreliable, and optic discs showed mild elevation. A thorough evaluation showed no new ventricular enlargement, a functioning shunt reservoir, and a normal opening pressure and normal CSF formula on lumbar puncture. Six weeks later, the optic discs were unchanged and visual function remained normal. The diagnosis was residual optic disc elevation, per-haps related to surface gliosis but not to high ICP. She was treated symptomatically for headache and improved. Can vision loss from optic neuropathy develop after ICP has normalized in patients who previously had papilledema? Yes. Vision loss from optic neuropathy can develop immediately after ICP has been normalized or months to years later, long after papilledema has disappeared. The immediate type of visual loss has been amply documented in patients who undergo surgery for brain tumor or CSF diversion for high ICP. Such patients have always had a mixture of optic disc elevation and pallor, together with optic nerve dysfunction, before the decom-pressive procedure occurred (31). I encountered this sad phenomenon in a 10-year-old boy with hydrocephalus from a suprasellar ganglioglioma who presented with slowly progressive binocular visual loss and best-corrected visual acuities of 20/400 in the right eye and finger counting in the left eye. Visual fields were severely compromised, and optic discs displayed a mixture of elevation and pallor. One day after ventriculoperitoneal shunting and partial tumor re-moval, visual acuities fell to finger counting in the right eye and no light perception in the left eye. Months later, visual acuity was unchanged and optic discs had flattened and become profoundly pale (Fig. 4). The explanation for this ‘‘postdecompression optic neuropathy'' is uncertain. Perhaps tumor decompression and restoration of normal ICP disturb compensatory blood flow to the optic nerve. To prevent this phenomenon, preoperative administration of ICP-lowering medication has been advocated, but its efficacy is unproven (31). Optic nerve dysfunction can also occur progressively or develop suddenly months to years after decompression, even when ICP is normal (32). For example, an 18-year-old girl FIG. 6. Detection of papilledema may be critical: shunted neonatal hydrocephalus. A. T2 axial MRI in a 39-year-old man with hydrocephalus shunted in infancy who complained of new headache and vision loss. Because there was no obvious ventriculomegaly, high ICP was discounted. But examination disclosed bilateral optic disc edema, so the patient underwent ventriculoperitoneal shunt revision. B. Postoperative T2 axial MRI disclosed some reduction in ventricular size. Vision improved slightly but considerable optic neuropathy persisted. (Modified from Katz et al (43), with permission.) Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 181 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. with chronic optic disc edema and severely compromised visual fields from IIH underwent ventriculoperitoneal shunting under my care. Visual function improved after the shunt and remained stable until 18 months later, when she suddenly lost all vision in 1 eye. Both optic discs were pro-foundly pale. ICP monitoring was normal, as performed with an intraparenchymal Codman ICP Monitoring System (33). In this delayed form of postdecompression optic neuropathy, I presume that damaged axons are prone to die, much as muscles are in the postpolio syndrome. I know of no effective way to guard against this tragic phenomenon, but I recommend measures to avoid systemic hypotension. Can we reliably predict who is at risk for irreversible optic neuropathy from papilledema? No. In one IIH study, the amount of papilledema appeared to be an independent risk factor, as judged by the fact that the eye with the greater amount of papilledema had the worse visual field. But there was poor correlation be-tween the amount of papilledema and the visual field loss (34). Other putative risk factors in IIH, not necessarily independent of each other, are the amount of preexisting optic nerve dysfunction (35), superimposed optic disc atrophic features (35), narrowed retinal arterioles (35), sustained systemic hypertension (16), weight gain, anemia (36), and African American race (37). Transient obscura-tions of vision and the height of the lumbar puncture opening pressure, which are intuitive risk factor candidates, have not been adequately dispelled as contributory (38). We lack reliable data on the prevalence of clinically meaningful visual loss in papilledema. Data on this point come entirely from the studies of IIH in referral centers. Reviews of these data affirm that on automated static perimetry, nearly all patients with IIH have visual field defects initially affecting the inferior nasal visual field (39,40). But how often is the visual loss disabling? One FIG. 7. Detection of papilledema may be critical: decompensated hydrocephalus. A. Axial FLAIR MRI. It shows peri-ependymal capping (arrows) and multiple cavernomas (white arrowheads). Because ventricles were not larger than on MRI performed 1 year earlier, the patient's symptoms were attributed to mental illness and the ventriculomegaly was considered ‘‘compensated.'' B. Vision loss became more severe, and ophthalmologic examination showed papilledema. ICP monitoring showed very high ICP. After placement of a ventriculoperitoneal shunt, the patient returned to baseline neurologic and vision status. C. Postoperative MRI shows smaller ventricles. D. Resolution of papilledema. (Modified from Mizrachi et al (50), with permission.) 182 Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. study of 57 IIH patients at a Philadelphia eye hospital reported that 25% had severe visual loss, but that figure is likely inflated by accrual of sicker patients (16). A long-term study of IIH in England found poor long-term vision from optic neuropathy in 17% of patients, but at least half of them had poor visual function at outset (41). A study of 54 IIH patients at 4 Israeli medical centers followed for 2-28 years (average, 6.2 years) (42) and a study of 68 patients in 2 London hospitals followed for an average of 4 years (38) reported that 6% had severe visual field loss on final examinations (42). In common with other studies, I have found that most patients with severe visual loss from papilledema already have advanced visual loss when the papilledema is first detected. Shunts or optic nerve sheath decompressions performed on my patients have nearly always occurred within hours to days of the first clinical encounter! Among my patients, I have found that severe visual loss from papilledema occurs in 3 groups: 1) Group 1: young women with IIH who have florid and/or atrophic papilledema and severe optic neuropathy at diagnosis, 2) Group 2: children with long-standing obstructive hydrocephalus from aque-ductal stenosis or brain tumors who have atrophic papil-ledema and severe optic neuropathy at diagnosis, and 3) Group 3: children or adults shunted for hydrocephalus in infancy who have not been regularly monitored by oph-thalmologists and who eventually appear for consultation when vision fails from chronically high ICP owing to shunt malfunction. Ophthalmoscopy typically discloses atrophic papilledema. In this third group, brain imaging usually does not show ventriculomegaly because shunted hydrocephalus is associated with stiff ventricular walls (43). Patients in all 3 groups present late because the visual loss of chronic papilledema proceeds silently and because chronic high ICP often causes no headache. Symptoms are ignored by children, dismissed by their parents, or attrib-uted to other causes by their physicians. I recommend intensified instruction in ophthalmoscopy to pediatricians and internists and periodic ophthalmologic examination of patients at high risk for papilledema, such as those with indwelling shunts, tuberous sclerosis (proclivity for obstructing giant cell astrocytomas), and neurofibromatosis (proclivity for meningiomatosis). Can papilledema be reliably distinguished ophthalmoscopically from congenitally elevated optic discs? No. Although there are reasonable ophthalmoscopic clues to congenitally anomalous optic disc elevation (drusen, dome-shaped elevation, anomalous surface vessels, and clear peripapillary nerve fiber layer), there are many cases in which one simply cannot tell if the optic disc elevation is congenital or acquired. Such diagnostic dif-ficulty should not be surprising, given that anomalous optic discs contain axons that are crowded into small scleral canals (44). Although congenitally elevated optic discs associated with buried drusen can be readily iden-tified by ultrasound or CT, the confusing cases I encounter have minimal optic disc elevation without drusen. Fluo-rescein angiography and optical coherence tomography probably cannot reliably distinguish minimal papilledema from a congenitally anomalous elevated optic disc (45). If I find no visual field loss or manifestations of a neurologic illness, I perform fundus photography, defer brain imag-ing, and re-examine for optic disc changes after several months. With this approach, I do not believe that I have overlooked high ICP. FIG. 8. Detection of papilledema may be critical: hydro-cephalus from subependymal giant cell astrocytoma in tuberous sclerosis. A. Fundus photographs show a mix-ture of optic disc pallor and swelling in a 17-year-old girl. Visual acuity was hand movements in both eyes. B. Pre-contrast axial CT shows ventriculomegaly from a partially calcified mass arising in the region of the foramen of Monro with features typical of a subependymal giant cell astrocytoma. Partial surgical excision and ven-triculoperitoneal shunting did not improve vision. Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 183 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Can papilledema be distinguished from other acquired optic neuropathies? Not always. Yes, there are some ophthalmoscopic signs that allow a presumptive diagnosis of other types of optic neuropathy, such as segmental and pallid edema in ischemic optic neuropathy, peripapillary telangiectasia in Leber hereditary optic neuropathy, and optociliary shunt vessels in juxta-ocular optic nerve sheath meningioma. But unless these signs are present, an ophthalmoscopic diagnosis of papilledema is difficult. After all, the optic disc edema of all acquired optic neuropathies arises from axoplasmic stasis (19). Therefore, distinguishing papilledema from other optic neuropathies with optic disc edema is typically based on 2 nonophthalmoscopic criteria: 1) binocular involvement and 2) relatively preserved visual function for the amount of nonatrophic optic disc edema. But these criteria will often fail to make a clear distinction. Papilledema can be mon-ocular. Binocular optic disc edema can occur in inflammatory, ischemic, diabetic, neoplastic, and hyper-tensive optic neuropathies. Relative preservation of visual function may occur in those conditions. As papilledema becomes chronic, visual dysfunction develops. These facts create clinical errors as follows. A 30-year-old mildly overweight woman developed blurred vision in both eyes. Both optic discs were mildly swollen, and visual fields showed mild inferior nasal nerve fiber bundle defects. Brain MRI was interpreted as normal. Lumbar puncture showed an opening pressure of 35 cm H2O and a normal cerebrospinal formula. A diagnosis of IIH was made elsewhere, and the patient was treated with acetazolamide. When vision continued to worsen, I ex-amined her and made the same observations, but an orbit-centered MRI showed enhancement of both optic nerves, and a chest CT showed hilar adenopathy. Bronchoscopic biopsy disclosed granulomas. The diagnosis was changed to sarcoidosis, the patient was treated with corticosteroid, and vision gradually recovered. The mistake was interpreting bilateral chronic optic disc elevation as papilledema when it was likely due to inflammation. The opposite scenario can also occur. A 20-year-old man with acquired immunodeficiency syndrome developed cryptococcal meningitis, but no opening pressure had been performed on lumbar puncture. He complained of diminished vision in both eyes. Brain imaging showed enhancement of the optic nerves and meninges; there was no ventriculomegaly. On ophthal-moscopy, the optic discs were elevated. The explanation for the visual loss was cryptococcal infiltration of the optic nerves. When vision declined precipitously, he underwent another lumbar puncture that showed a markedly elevated opening pressure. Ventriculoperitoneal shunting improved headache and vision, but as papilledema turned into optic disc pallor, he was left with considerable optic nerve dys-function. The error was in attributing vision loss entirely to optic nerve infiltration and in not recognizing papilledema as an important contributor. This last case exemplifies a common situation, namely that optic disc edema may be the result of a mixed mechanism-inflammation of the optic nerve and increased ICP. Such a mixed mechanism often occurs in meningitis, where the optic nerves may be directly attacked and the arachnoid granulations blocked to cause high ICP. In viral meningitis, the high ICP is usually well tolerated, but in bacterial, fungal, protozoal, or neoplastic meningitis, it may be vision threatening. CSF diversion is critical. Does papilledema occur in malignant hypertension or in hy-pertensive encephalopathy? No. This is a common misconception. Optic disc edema certainly occurs in malignant hypertension, but it represents leakage of serumfromincompetent optic disc arterioles owing to protective vasoconstriction followed by autoregulatory breakthrough vasodilation (46) (Fig. 5). Vascular leakage from excessive autoregulation underlies hypertensive reti-nopathy (cotton wool spots, surface hemorrhages, and hard exudates), which is always present if the optic disc is swollen in malignant hypertension (46). The vasogenic edema of malignant hypertension is not enough to cause very high ICP, which occurs only when the brain becomes massively infarcted, a rare phenomenon now that there is earlier in-tervention with powerful blood pressure-lowering regimens (48,49). But the vasogenic edema may be enough to cause a mild elevation of the opening pressure. Unless the vasogenic edema of malignant hypertension turns into infarction in the optic disc, visual function will be relatively preserved, a feature that will tempt a mistaken impression of papilledema, particularly if the opening pressure is elevated (48). I have encountered patients who have undergone lumbar puncture prompted by the finding of bilateral optic disc edema in the setting of malignant hypertension. The lumbar puncture has shown an elevated opening pressure and led to ill-advised placement of a ventriculoperitoneal shunt. I suggest that the optic disc edema of malignant hypertension be interpreted as a manifestation of high blood pressure rather than of high ICP. The appropriate intervention is gradual lowering of blood pressure, not CSF diversion or optic nerve sheath fenestration. Can papilledema be present even if brain imaging shows no clear signs of high ICP? Yes. The signs that radiologists look for in diagnosing high ICP-enlarged or enlarging ventricles, periependymal signal alteration, sulcal or cisternal obliteration-are not very sensitive to high ICP. First, patients with shunted hydrocephalus typically have stiff ventricular walls that do not expand with pressure (43). Second, patients whose obstruction to CSF outflow is in the arachnoid granulations or dural venous sinuses usually do not develop ven-triculomegaly because there is no pressure gradient between the intraventricular and extraventricular spaces (27). Third, 184 Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. patients with parenchymal edema from head trauma or encephalitis or tight skulls from syndromic craniosynostosis may lack obvious radiologic signs of high ICP. Here is an example of how this issue plays out in clinical practice. A 39-year-old man who had undergone ventriculoper-itoneal shunting at age 12 for hydrocephalus developed new headache and vision loss 27 years later. Although papil-ledema was noted, shunt malfunction was initially dismissed as a diagnosis because MRI did not show definite signs of high ICP. The shunt was eventually revised, at which time ventricular pressure was very high. Following the revision, papilledema resolved to pallor but the patient had lost a substantial amount of vision (43) (Fig. 6). Brain ventricles may be dilated yet the ICP is normal, a condition called ‘‘compensated hydrocephalus.'' If there are no other compelling imaging signs of high ICP, radi-ologists typically look for an enlargement of ventricular size over time before suspecting high ICP. Here is an example of how imaging can be misleading, and the finding of papil-ledema can be critical. A 54-year-old woman developed a confusional state, but her symptoms were disregarded because she carried an earlier diagnosis of bipolar disorder (50). She had un-dergone brain imaging a year earlier in follow-up of a known diagnosis of Osler-Weber-Rendu disease pro-ducing multiple brain cavernomas. Brain MRI showed ventriculomegaly but no change in the ventricular size as compared to an MRI performed a year earlier. The ven-triculomegaly was considered compensated. But months later, she became incoherent, complained of blurred vision, and was found to have papilledema. ICP monitoring showed very high ICP. Ventriculoperitoneal shunting re-stored baseline mental status and normal vision and elim-inated papilledema (Fig. 7). The search for papilledema is especially important in patients who are prone to high ICP because they have an indwelling CSF shunt, a subependymal giant cell astrocytoma in tuberous sclerosis (Fig. 8), chronic granulomatous men-ingitis, or meningiomatosis. Such patients may be unaware of progressive visual loss from high ICP. Their imaging may be insensitive to high ICP, and their tests of shunt function may be unreliable. Finding papilledema may be vision saving and excluding it may spare them ICP monitoring or needless shunt revision (24-26,43). REFERENCES 1. Duke-Elder S, Scott GI. System of Ophthalmology, Vol XII. London, UK: Henry Kimpton, 1971:32-66. 2. Walsh FB, Hoyt WF. Clinical Neuro-Ophthalmology, 3rd edition. Baltimore, MD: Williams & Wilkins Company, 1969: 567-641. 3. Paton L, Holmes G. The pathology of papilloedema. Brain. 1911;33:389-432. 4. Paton L, Holmes G. The pathology of papilloedema. Trans Ophthal Soc U K. 1911;31:117-132. 5. Weiss P, Hiscoe HB. Experiments on the mechanism of nerve growth. J Exp Zool. 1948;107:315-395. 6. Minckler DM, Tso MO, Zimmerman LE. A light microscopic, autoradiographic study of axoplasmic transport in the optic nerve head during ocular hypotony, increased intraocular pressure, and papilledema. Am J Ophthalmol. 1976;82: 741-767. 7. Tso MO, Hayreh SS. Optic disc edema in raised intracranial pressure. III. A pathologic study of experimental papilledema. Arch Ophthalmol. 1977;95:1448-1457. 8. Tso MO, Hayreh SS. Optic disc edema in raised intracranial pressure. IV. Axoplasmic transport in experimental papilledema. Arch Ophthalmol. 1977;95:1458-1462. 9. Minckler DS, Bunt AH. Axoplasmic transport in ocular hypotony and papilledema in the monkey. Arch Ophthalmol. 1977;95:1430-1436. 10. Schirmer CM, Hedges TR III. Mechanisms of visual loss in papilledema. Neurosurg Focus. 2007;23:1-8. 11. Hayreh SS. Optic disc edema in raised intracranial pressure. V. Pathogenesis. Arch Ophthalmol. 1977;95: 1553-1565. 12. Sanders MD. The Bowman Lecture. Papilloedema: the pendulum of progress. Eye (Lond). 1997;11:267-294. 13. Wirtschafter JD, Rizzo FJ, Smiley BC. Optic nerve axoplasm and papilledema. Surv Ophthalmol. 1975;20:157-189. 14. Radius RL. Optic nerve fast axonal transport abnormalities in primates. Occurrence after short posterior ciliary artery occlusion. Arch Ophthalmol. 1980;98:2018-2022. 15. McLeod D, Marshall J, Kohner EM. Role of axoplasmic transport in the pathophysiology of ischaemic disc swelling. Br J Ophthalmol. 1980;64:247-261. 16. Corbett JJ, Savino PJ, Thompson HS, Kansu T, Schatz NJ, Orr LS, Hopson D. Visual loss in pseudotumor cerebri: follow-up of 57 patients from five to 41 years and a profile of 14 patients with permanent severe visual loss. Arch Neurol. 1982;39:461-474. 17. Johnson MW, Kincaid MC, Trobe JD. Bilateral retrobulbar optic nerve infarctions after blood loss and hypotension. A clinicopathologic case study. Ophthalmology. 1987;94: 1577-1584. 18. Minckler DS, Spaeth GL. Optic nerve damage in glaucoma. Surv Ophthalmol. 1981;26:128-148. 19. Sjostrand J. A dynamic view of the retinal ganglion cell and its transport. A mini-review of the role of axonal transport dysfunction in the pathophysiology of optic neuropathies. Acta Ophthalmol. 1981;59:785-797. 20. Hayreh MS, Hayreh SS. Optic disc edema in raised intracranial pressure. I. Evolution and resolution. Arch Ophthalmol. 1977;95:1237-1244. 21. Steffen H, Eifert B, Aschoff A, Kolling GH, Volcker HE. The diagnostic value of optic disc elevation in acute elevated intracranial pressure. Ophthalmology. 1996;103: 1229-1232. 22. Selhorst JB, Gudeman SK, Butterworth JF IV, Harbison JW, Miller JD, Becker DP. Papilledema after acute head injury. Neurosurgery. 1985;16:357-363. 23. Fahmy JA. Papilledema associated with ruptured intracranial aneurysms. Acta Ophthalmol. 1972;50: 793-802. 24. Pribila JT, Ronan SM, Trobe JD. Multiple intracranial meningiomas causing papilledema and visual loss in a patient with nevoid basal cell carcinoma syndrome. J Neuroophthalmol. 2008;28:41-46. 25. Dotan SA, Trobe JD, Gebarski SS. Visual loss in tuberous sclerosis. Neurology. 1991;41:1915-1917. 26. Chong DY, Hirunwiwatkul P, McKeever PE, Trobe JD. Papilledema in obstructive hydrocephalus caused by giant cell astrocytoma of tuberous sclerosis. J Neuroophthalmol. 2007;27:50-54. 27. Kim AW, Trobe JD. Syndrome simulating pseudotumor cerebri caused by partial transverse venous sinus obstruction in metastatic prostate cancer. Am J Ophthalmol. 2000;129:254-256. Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 185 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. 28. Digre KB, Nakamoto BK, Warner JE, Langeberg WJ, Baggaley SK, Katz BJ. A comparison of idiopathic intracranial hypertension with and without papilledema. Headache. 2009;49:185-193. 29. Digre KB, Friedman DI. An overweight young woman with new headache and normal-appearing optic discs. J Neuroophthalmol. 2010;30:85-90. 30. Schultz PN, Sobol WM, Weingeist TA. Long-term visual outcome in Terson syndrome. Ophthalmology. 1991;98: 1814-1819. 31. Beck RW, Greenberg HS. Post-decompression optic neuropathy. J Neurosurg. 1985;63:196-199. 32. Wilkes BN, Siatkowski RM. Progressive optic neuropathy in idiopathic intracranial hypertension after optic nerve sheath fenestration. J Neuroopthalmol. 2009;29: 281-283. 33. Warden KF, Alizai AM, Trobe JD, Hoff JT. Short-term continuous intraparenchymal intracranial pressure monitoring in presumed idiopathic intracranial hypertension. J Neuroophthalmol. 2011 April 7. Epub ahead of print. 34. Wall M, White WN. Asymmetric papilledema in idiopathic intracranial hypertension: prospective interocular comparison of sensory visual function. Invest Ophthalmol Vis Sci. 1998;39:134-142. 35. Holmes G. Intracranial hypertension. Br J Ophthalmol. 1937;21:337-342. 36. Biousse V, Rucker JC, Vignal C, Crassard I, Katz BJ, Newman NJ. Anemia and papilledema. Am J Ophthalmol. 2003;135:437-446. 37. Bruce BB, Preechawat P, Newman NJ, Lynn MJ, Biousse V. Racial differences in idiopathic intracranial hypertension. Neurology. 2008;70:861-867. 38. Orcutt JC, Page NGR, Sanders MD. Factors affecting visual loss in benign intracranial hypertension. Ophthalmology. 1984;91:1303-1312. 39. Wall M. Idiopathic intracranial hypertension. Neurol Clin. 2010;28:593-617. 40. Thurtell MJ, Bruce BB, Newman NJ, Biousse V. An update on idiopathic intracranial hypertension. Rev Neurol Dis. 2010; 7:1-13. 41. Rowe FJ, Sarkies NJ. Assessment of visual function in idiopathic intracranial hypertension: a prospective study. Eye (Lond). 1998;12:111-118. 42. Kesler A, Hadayer A, Goldhammer Y, Almog Y, Korczyn AD. Idiopathic intracranial hypertension: risk of recurrences. Neurology. 2004;63:1737-1739. 43. Katz DM, Trobe JD, Muraszko KM, Dauser RC. Shunt failure without ventriculomegaly proclaimed by ophthalmic findings. J Neurosurg. 1994;81:721-725. 44. Mullie MA, Sanders MD. Scleral canal size and optic nerve head drusen. Am J Ophthalmol. 1985;99:356-359. 45. Heidary G, Rizzo JF III. Use of optical coherence tomography to evaluate papilledema and pseudopapilledema. Semin Ophthalmol. 2010;25:198-205. 46. Hayreh SS, Servais GE, Virdi PS. Fundus lesions in malignant hypertension. V. Hypertensive optic neuropathy. Ophthalmology. 1986;93:74-87. 47. Besirli CG, Sudhakar P, Wesolowski J, Trobe JD. Serous retinal detachment in hypertensive posterior reversible encephalopathy syndrome. AJNR Am J Neuroradiol. 2011 Mar 10. Epub ahead of print. 48. Chang GY, London ZN. Hypertensive encephalopathy. Medlink Neurol. 2010. 49. Chester EM, Agamanolis DP, Banker BQ, Victor M. Hypertensive encephalopathy: a clinicopathologic study of 20 cases. Neurology. 1978;28:928-939. 50. Mizrachi IB, Trobe JD, Gebarski SS, Garton HJ. Papilledema in the assessment of ventriculomegaly. J Neuroophthalmol. 2006;26:260-263. 186 Trobe: J Neuro-Ophthalmol 2011; 31: 175-186 The Ninth Hoyt Lecture Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. |
