Journal of Neuro-Ophthalmology, September 2000, Volume 20, Issue 3

Neuro-Ophthalmology of the Pregeniculate Afferent Visual System

Journal of Neuro- Ophthalmology 20( 3): 173- 191, 2000. © 2000 Lippincott Williams & Wilkins, Inc., Philadelphia Neuro- Ophthalmology of the Pregeniculate Afferent Visual System: January- December 1999 Laura J. Balcer, MD, MSCE, and Steven L. Galetta, MD As neuro- ophthalmology enters a new millennium, significant advances continue in the evaluation, diagno­sis, and treatment of disorders of the pregeniculate af­ferent visual system. This review includes highlights from January through December 1999. The following table of contents has been provided for ease of reference to the topics presented: 1) EVALUATION OF THE AFFERENT VISUAL SYSTEM a) Visual Field Testing b) Anisocoria and the Afferent Pupillary Defect c) The Pulfrich Phenomenon 2) NEURO- OPHTHALMOLOGY AND THE RETINA a) Vigabatrin and Outer Retinal Dysfunction b) Cancer- associated Retinopathies c) Neuroretinitis d) Carotid Artery Dissection e) Retinal Artery Occlusion f) Amaurosis Fugax 3) THE OPTIC NERVE a) Optic Nerve Compression by the Carotid Artery b) Optic Nerve Head Blood Flow c) Anterior Ischemic Optic Neuropathy d) Posterior Ischemic Optic Neuropathy e) Giant Cell Arteritis: Arteritic Ischemic Optic Neurop­athy and Ocular Ischemic Syndrome f) Optic Neuritis and Multiple Sclerosis g) Optic Atrophy and Polyneuropathy h) Idiopathic Intracranial Hypertension ( and its Impost-ers) i) Leber Hereditary Optic Neuropathy j) Optic Disc Drusen k) Toxic Optic Neuropathies 1) Traumatic Optic Neuropathy Manuscript received May 2, 2000; accepted June 14, 2000. From the Division of Neuro- Ophthalmology, the Departments of Neurology and Ophthalmology, Hospital of the University of Pennsyl­vania, Scheie Eye Institute, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania. Address correspondence and reprint requests to Steven L. Galetta, MD, Department of Neurology, 3 East Gates, 3400 Spruce Street, Philadelphia, PA 19104. 4) THE OPTIC CHIASM AND BEYOND a) Visual Evoked Potential Evidence of Chiasmal Hy­poplasia b) Pituitary Adenoma c) Hereditary Chiasmal Optic Neuropathy d) Langerhans Cell Histiocytosis EVALUATION OF THE AFFERENT VISUAL SYSTEM Visual Field Testing Testing time and patient fatigue are often important factors to consider when selecting automated visual field testing programs for patients with afferent visual loss. Khoury et al. ( 1) compared the 24- 2 and 30- 2 programs for the Humphrey Field Analyzer ( Zeiss Humphrey Sys­tems, Dublin, CA) in patients with nonglaucomatous op­tic neuropathies to determine whether the 24- 2 strategy could provide information on visual field loss compa­rable to that yielded by the 30- 2 program, with shorter testing time and less test- retest variability. The authors found that the 24- 2 testing strategy provides information on visual field defects comparable to that for the 30- 2 strategy, with less testing time and variability. However, caution must be used in the interpretation of 24- 2 fields, especially in patients with papilledema ( as in idiopathic intracranial hypertension [ IIH]) or in those with subtle defects; evaluation with the 30- 2 strategy is recom­mended for such patients. Anisocoria and the Afferent Pupillary Defect The degree to which anisocoria, in the absence of optic nerve or retinal pathology, may produce an afferent pu­pillary defect ( APD) has important implications for the evaluation of patients with afferent visual loss. Lam and Thompson ( 3) used infrared videography to examine the relation of induced anisocoria ( using mydriatic drops) to the presence and magnitude of the APD. Experiments were performed in normal subjects and in patients with known APDs caused by optic nerve disease. Examina­tion of eight normal subjects in darkness revealed that for every 1 mm of induced anisocoria, 0.14 log units of APD was produced in the eye with the smaller pupil. Six nor­mal subjects were also examined under room lighting conditions; the magnitude of the induced APD in the smaller pupil was less compared with those examined in 173 174 L. J. BALCER AND S. L. GALETTA darkness (< 0.4 log units 2 hours after dilation). Among 24 patients with known APDs and optic nerve disease, induced anisocoria tended to shift the APD toward the undilated ( smaller) pupil. However, the authors found that to induce a clinically significant APD ( defined as 5: 0.3 log units), a large degree of anisocoria must be present ( at least 2 mm). A number of additional factors, including light intensity and relative degree of asymme­try of optic nerve function ( input asymmetry), may also influence the magnitude of anisocoria- induced APDs. The Pulfrich Phenomenon Similar to the APD, the Pulfrich phenomenon may be induced by conditions that impair optic nerve conduc­tion. This phenomenon is most commonly elicited by the swinging pendulum test, in which patients with unilateral or asymmetric optic neuropathies perceive the path of the pendulum to be ellipsoid rather than flat within a plane perpendicular to the visual axis. Such patients may also experience a spontaneous Pulfrich phenomenon, mani­fested by visual symptoms during daily activities ( 3). To alleviate the symptoms of the Pulfrich phenomenon, Dia­per et al. ( 3) suggest the use of a neutral density filter in a spectacle or contact lens placed in front of the " good" eye. NEURO- OPHTHALMOLOGY AND THE RETINA Vigabatrin and Outer Retinal Dysfunction Visual field constriction and other neuro- ophthalmol-ogic signs have been shown to be prevalent among pa­tients taking the antiepileptic drug vigabatrin ( 4- 8). Sev­eral reports on this important topic were published in the ophthalmologic and neurologic literature during 1999. Daneshvar et al. ( 4) investigated the clinical, perimetric, and electrophysiologic findings for 41 consecutive pa­tients. Among these patients, 12 were found to have evi­dence of peripheral visual field constriction on 60- 4 pe­rimetry. Daily doses of vigabatrin among the 12 patients with visual field constriction ranged from 3,000 to 4,000 mg, with treatment durations of 9 to 75 months. Inter­estingly, eight of these patients were visually asymptom­atic, while problems among the four symptomatic pa­tients included tunnel vision, missing objects, and bump­ing into objects. Most patients ( 8/ 12) had visual acuities of 20/ 20 OU; mild bilateral optic disc pallor was noted in four patients. Bilateral visual field constriction was dem­onstrated by 30- 2 perimetry in 9 of 12 patients, and by blue- on- yellow perimetry in 8 of 9 patients tested. Field defects were similar to those demonstrated in Figure 1. Among the 10 patients who underwent electrophysi­ologic testing, the electroretinogram ( ERG) results were abnormal in four patients, and electrooculogram ( EOG) results were abnormal in five patients. Delayed visual evoked potentials were demonstrated in three of ten pa­tients. Because the scotopic b- wave, reflective of Muller cell function, was depressed in all of the patients with ERG abnormalities, the authors postulate that Muller cell toxicity in the peripheral retina caused by vigabatrin may be responsible for the observed visual field constriction. Given the high incidence of visual field constriction among patients administered vigabatrin ( 12/ 41 consecu­tive patients in this series alone), visual field examina­tions ( covering the peripheral 60° of the visual field) before initiation of vigabatrin therapy and at regular in­tervals thereafter are recommended ( 4). Careful assess­ments and discussions of the potential risks and benefits of therapy must also be undertaken with each patient. The prevalence of visual field constriction among pa­tients on vigabatrin monotherapy was likewise high in a series reported by Kalviainen et al. ( 5). They compared the prevalence of visual field constriction among 32 adult patients on vigabatrin ( duration, 29- 119 months) with that seen among 18 patients on carbamazepine monotherapy and 18 healthy control subjects. Among the 32 patients on vigabatrin monotherapy, all of whom were visually asymptomatic, 13 ( 40%) demonstrated concen­tric visual field constriction by Goldmann kinetic perim­etry. Because none of the healthy controls ( 0/ 18) nor patients on carbamazepine monotherapy ( 0/ 18) were found to have visual field defects, vigabatrin mono­therapy was significantly associated with visual field constriction in this study ( P = 0.0001, X2 test). Among those in the vigabatrin group who had visual field loss, constriction was classified as severe in three patients ( defined as extension of intact field to less than 50° in the temporal meridian, I4e isopter). This study ( 5) provides further evidence that vigabatrin monotherapy is associ­ated with peripheral visual field constriction and that visual field testing and careful counseling must be un­dertaken before and during therapy. Neuro- ophthalmologic and electrophysiologic abnor­malities also were more prevalent among patients receiv­ing vigabatrin compared with controls in an investigation by Miller et al. ( 6). Among 32 vigabatrin- treated patients and 11 control patients ( matched by age, seizure fre­quency, and anticonvulsant therapies other than vi­gabatrin), nearly 50% of the vigabatrin group had bilat­eral peripheral visual field constriction by kinetic or static perimetry. Visual acuities were 20/ 20 or better among all of the control patients ( median 20/ 15 OU), while 12 of the 32 vigabatrin patients ( 37.5%) had visual acuities ranging from 20/ 25 to 20/ 60 in one or both eyes. Retinal abnormalities, absent among control patients, were found in 23 patients in the vigabatrin group; such findings included retinal arterial narrowing, epiretinal membrane formation, irregular sheen or abnormal pig­mentation in the macula, and reduction in the peripapil­lary nerve fiber layer. Among ERG parameters, oscilla­tory potential amplitudes were severely depressed in all patients in the vigabatrin group ( P < 0.001 OU), as were photoreceptor sensitivity ( a- wave; P - 0.005), and b-waves for both rod and cone systems ( P = 0.001- 0.003 OU). The reduction in cone ERG amplitudes, particu­larly cone flicker responses, correlated strongly with the degree of visual field constriction as measured by kinetic perimetry ( OD, r = 0.64; P = 0.003; OS, r = 0.72; P = 0.001). The findings of this study ( 6) were important not only in providing further evidence for neuro-ophthalmologic abnormalities in patients taking vi- / Neuro- Ophthalmol, Vol. 20, No. 3, 2000 PREGENICULA TE REVIEW 1999 175 Left Eye Right Eye White- on- White 30- 2 White- on- White 60- 4 Blue- on- Yellow 30- 2 - t3CT FIG. 1. Central 30- 2, peripheral 60- 4, and blue- on- yellow perimetry for a patient taking vigabatrin. Visual fields demonstrate con­striction OU. Reprinted with permission from: Danesh-var H, Racette L, Coupland SG, et al. Symp­tomatic and asymptomatic visual loss in pa­tients taking vigabatrin. Ophthalmology 1999; 106; 1792- 8. gabatrin, but also in confirming that the extent of visual field constriction may be related to the degree of retinal electrophysiologic dysfunction. Electrophysiologic dysfunction in the retinas of pa­tients taking vigabatrin was likewise demonstrated by Arndt et al. ( 7) in a study of 20 patients. Visual symp­toms were again uncommon ( reported by 5 of the 20 patients) and included constriction ( 2 patients) and blurred vision ( 3 patients). Visual acuities were normal ( 20/ 20) in 32 of 40 eyes tested; acuities in the remainder of the group were 20/ 25 ( 3 eyes), 20/ 30 ( 4 eyes), and 20/ 100 ( 1 eye with amblyopia). Twelve patients had vi­sual field constriction ( bilateral in eight patients). The EOG results ( EOG light/ dark ratio) were abnormal in 14 patients, with marked impairment in both eyes of 13 patients. Electroretinogram oscillatory potentials were impaired in both eyes of ten patients. Overall, impair­ment of retinal electrophysiologic function was demon­strated in 17/ 20 patients ( 85%), while visual field con­striction was noted in only 12/ 20 patients ( 60%). In this study, electrophysiologic testing, particularly EOG, dem­onstrated retinal dysfunction more commonly than did visual field testing, and it may therefore provide a more sensitive method for detecting retinal abnormalities in patients taking vigabatrin. Visual field constriction may be associated with vigabatrin therapy in children, as well ( 8). Two patients, aged 10 and 15 years, had visual field defects by Goldmann perimetry. These defects, observed after 33 to 57 months of therapy, were similar to those reported in adult patients ( concentric constriction). Cancer- associated Retinopathies Gittinger and Smith ( 9) implicated the bipolar cell as the major site of the paraneoplastic dysfunction in cuta­neous melanoma- associated retinopathy. They presented the clinical, electrophysiologic, and retinal histopatho­logic findings for a 59- year- old man who died of meta­static melanoma. The patient developed progressive vi­sual loss five years after removal of a cutaneous malig­nant melanoma from his forearm. In the interim, he had J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 176 L. J. BALCER AND S. L. GALETTA also been found to have metastases to the retroperito-neum and brain. Visual acuities were 20/ 30 OD and 6/ 200 OS. Goldmann visual fields demonstrated con­striction OU with a central scotoma OS. Color vision was markedly abnormal, and optic disk pallor was noted. A full- field flash ERG demonstrated markedly reduced b-wave amplitudes. Subsequent histopathologic examina­tion of both retinas revealed marked reduction in the thickness of the inner nuclear layer and in the density of the bipolar neuron nuclei. Photoreceptor nuclei in the outer nuclear layer, however, were normal. Gliosis in both the ganglion and bipolar cell layers was detected by glial fibrillary acidic protein ( GFAP) stain. Prominent gliosis was also noted in the optic nerves, with preser­vation of axonal structure. These histopathologic studies provide evidence, consistent with clinical, electrophysi­ologic, and immunologic data, that implicates the bipolar cell as the primary site of dysfunction in melanoma-associated retinopathy. The involvement of the 23- kDa retinal antigen with cases of cancer- associated retinopathy has been well es­tablished. Recently, eight patients with unexplained pro­gressive visual loss ( over months to, years) in one or both eyes were found to produce autoantibodies to a novel 22- kDa neuronal antigen present within optic nerve and retina ( 10). Seven of the eight patients had abnormal ERG results; optic disc pallor ( six patients) and retinal vessel attenuation ( five patients) were also noted. The progressive visual loss stabilized in one patient after im­munosuppressive therapy ( cyclosporine, azathioprine, and prednisone). While autoantibodies to the 22- kDa an­tigen were found to be a common link within this group of patients with progressive visual loss, the authors cau­tion that a direct role for such antibodies in the patho­genesis of optic nerve and retinal disease remains to be demonstrated. Such autoantibodies may represent a marker for optic nerve or retinal damage and may prove useful for identifying patients with unexplained visual loss for whom further immunologic investigation is ap­propriate ( 10). Neuroretinitis The clinical manifestations of infectious neuroretinitis were highlighted by several reports ( 11- 14). Solley et al. ( 11) reviewed the clinical and photographic records for 24 patients ( 35 eyes) with ocular cat- scratch disease to evaluate the prevalence of various posterior segment findings. Retinal or choroidal white lesions ( discrete le­sions measuring 50- 3,000 m in diameter) were the most common findings, present in 29 of 35 eyes ( 83% of eyes). These lesions were variable in location, occurring in superficial retina ( 30%), deep retina ( 49%), full-thickness retina ( 14%), and choroid ( 7%). Disc edema ( 46% of eyes), macular star formation ( 43% of eyes), and mild vitreous inflammation ( 20% of eyes) were also seen. Vascular occlusive events occurred in a small num­ber of patients ( 14% of eyes); these included branch retinal artery occlusion ( four eyes) and branch retinal vein occlusion ( one eye). Initial visual acuities ranged from 20/ 15 to counting fingers ( median 20/ 40). Final median visual acuities were 20/ 20 among both treated and untreated patients after a mean follow- up of 4.5 months. Isolated foci of retinitis and choroiditis were therefore the most common ocular manifestations of cat-scratch disease in this series ( 11), although disc edema, macular star formation, and other signs of neuroretinitis were also frequent. Although uncommon, branch retinal artery occlusion with severe visual loss may occur in cat- scratch neurore­tinitis. Gray et al. ( 12) reported the case of a 20- year- old man with fever, malaise, lymphadenopathy, and de­creased vision OD after known exposure to a cat scratch. Examination OD revealed hand motions vision, a 3+ APD, and inflammatory cells in the anterior chamber and vitreous. Ischemic hemiretinal whitening, a macular star, and a peripapillary angiomatous lesion were seen on ophthalmoscopy. Fluorescein angiography confirmed the presence of a branch retinal artery occlusion. Serum IgM titers to Bartonella henselae were 1: 80 ( five times higher than normal). The vision did not improve after a 7- day course of oral clindamycin therapy, but it did respond to a combination of doxycycline and rifampin. A 1- week prednisone taper was also given. Visual acuity improved to 20/ 400, with resolution of the retinal whitening, macu­lar star, and angiomatous lesion. This patient's findings and course emphasize that although cat- scratch neurore­tinitis is generally associated with mild visual loss and good visual recovery, severe and permanent loss of vi­sion associated with uncommon retinal- vascular findings may occur. Another rare neuro- ophthalmologic manifestation of cat- scratch disease, peripheral facial nerve palsy, was reported by Thompson et al. ( 13) in a 40- year- old woman with neuroretinitis. She developed painless visual loss OD 5 weeks after a kitten scratch and flu- like symptoms. Five days later, a right peripheral facial palsy developed; the patient declined oral corticosteroid therapy. The au­thors ( 13) propose that peripheral facial nerve palsy in the setting of cat- scratch disease, previously reported only in pediatric patients, may occur on the basis of nerve inflammation ( vasculitis) or through inflammation and enlargement of the parotid gland. Tuberculosis should also be considered in the differ­ential diagnosis for patients with neuroretinitis, as illus­trated by a case described by Stechschulte et al. ( 14). A 43- year- old woman with a history of exposure to tuber­culosis noted " shadows" OD for 2 months. Examination revealed a visual acuity of 20/ 30 in the affected right eye with optic disc edema, subretinal exudates, and a macu­lar star. Syphilis serologies, angiotensin- converting en­zyme level, and chest radiograph were normal. One year later, she developed decreased vision OS. Visual acuities were 20/ 25 OD and 20/ 200 OS; there was no APD. Fun-duscopic examination revealed optic disc edema, vitre­ous inflammation, and multifocal choroiditis OU. An exudative retinal detachment involving the fovea was present OS. The purified protein derivative skin test re­sult was positive ( 15 mm of induration); a repeated chest x- ray and results of spinal fluid analysis were normal. J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 PREGENICULATE REVIEW 1999 177 The patient was treated with isoniazid and rifampin; this resulted in rapid resolution of the choroidal infiltrates and retinal detachment. Although isolated neuroretinitis, as occurred in this patient, represents an unusual presen­tation of tuberculosis infection, this entity must be con­sidered in patients with neuroretinitis, particularly in those with a history of exposure ( 14). Carotid Artery Dissection Kerry ( 15) reviewed the neuro- ophthalmologic histo­ries and findings for 28 patients with spontaneous ex­tracranial carotid artery dissection. Each patient under­went a complete neuro- ophthalmologic examination at the time of diagnosis. Carotid dissection was confirmed by cerebral angiography in 5 patients, and by magnetic resonance imaging ( MRI), MR angiography, and Dopp-ler ultrasonography in the remaining 23 patients. Neuro-ophthalmologic findings in this series, present in all but one patient, included oculosympathetic paresis ( 23 pa­tients), amaurosis fugax ( 2 patients, with episodes pro­voked by sitting up from a supine position in 1 patient), homonymous hemianopsia ( 2 patients), posterior isch­emic optic neuropathy with monocular field defect ( 1 patient), and sixth nerve palsy ( 1 patient). To minimize the potential risks of cerebral ischemia, prompt suspi­cion, recognition, and treatment of carotid artery dissec­tion is recommended ( 15). If the dissection is discovered in the first 2 weeks after onset, most authorities would anticoagulate the patient ( heparin/ warfarin) for 3 to 6 months. However, there is no solid data supporting the use of anticoagulation over aspirin. Long- term aspirin therapy is usually given after the course of anticoagula­tion. While headache, retroorbital pain, and neck and jaw pain are commonly present in the setting of carotid artery dissection, Lee et al. ( 16) presented an unusual case with painless monocular visual loss and central retinal artery occlusion. A 38- year- old man developed transient mild stiffness of the left posterior neck while playing volley­ball; this did not persist. Five days later, he experienced sadden visual loss OS. Funduscopic examination re­vealed diffuse retinal whitening and a macular cherry- red spot; no emboli were seen. T2- weighted axial MR im­ages demonstrated a low- signal flow void surrounded by a circular hyperintense bright signal at the petrous por­tion of the left internal carotid artery, consistent with dissection. The diagnosis was confirmed by cerebral an­giography. Improvement of both retinal artery filling ( by fluorescein angiography) and the patient's vision ( from count- fingers to - 20/ 30) were noted 4 months after an­ticoagulation. The authors ( 16) suspect that either hemo­dynamic compromise or embolic occlusion led to sudden visual loss in this patient with carotid artery dissection. Although his visual loss was " painless," it is possible that the episode of transient ipsilateral neck stiffness five days earlier represented the onset of a traumatic dissec­tion. Carotid artery dissection should therefore be con­sidered in young patients with acute monocular visual loss, even in the absence of prominent pain or headache ( 16). Retinal Artery Occlusion Optic disc morphology after experimental central reti­nal artery occlusion in Rhesus monkeys was studied by Jonas et al. ( 17). These authors compared optic disc fea­tures of eight eyes ( eight monkeys) with experimental central retinal artery occlusion with those of eight healthy eyes and eight eyes with experimental high-pressure glaucoma. Serial optic disc photographs were evaluated morphometrically for 1) area and shape of the neuroretinal rim, and 2) presence and area of parapapil-lary chorioretinal atrophy. After a mean period of 62 days after central retinal artery occlusion ( and mean du­ration of 24.4 months for increased intraocular pressure in glaucomatous eyes), no significant differences in neu­roretinal rim area or area of parapapillary chorioretinal atrophy were noted for central retinal artery occlusion eyes before versus after occlusion; similar results were obtained for comparison of central retinal artery occlu­sion eyes versus healthy eyes. The neuroretinal rim was significantly smaller for glaucomatous eyes versus cen­tral retinal artery occlusion eyes ( P < 0.0006, Wilcoxon Mann- Whitney test); the area of parapapillary atrophy was also significantly larger in the glaucomatous eyes ( P = 0.01) compared with central retinal artery occlusion eyes. The authors ( 17) concluded that experimental cen­tral retinal artery occlusion does not markedly change the size and shape of parapapillary atrophy and the neu­roretinal rim, and that these features may be useful for differentiating glaucomatous optic nerve damage from nonglaucomatous optic atrophy that occurs after retinal nerve fiber damage. Aggressive treatment for central retinal artery occlu­sion is often recommended for patients who present acutely, given the frequent occurrence of severe and ir­reversible visual loss. Rumelt et al. ( 18) treated 11 pa­tients with unilateral central retinal artery occlusion of fewer than 48 hours' duration using an aggressive step­wise systematic regimen. Five additional patients were treated in an " arbitrary nonsystematic manner." The two groups were compared with respect to visual outcome. The aggressive treatment regimen ( performed in 11 pa­tients) included ocular massage, sublingual isosorbide dinitrate, intravenous acetazolamide, intravenous manni-tol or oral glycerol, anterior chamber paracentesis, intra­venous methylprednisolone followed by streptokinase, and retrobulbar tolazoline. After each step, retinal artery flow was visualized using a three- mirror contact lens; improvement was determined upon the reestablishment of continuous laminar flow, an increase in the width of the blood column, and disappearance of fragmented flow. When improvement of retinal flow was noted, the above regimen was stopped. Treatment of the " nonsys­tematic" group consisted of intravenous acetazolamide, methylprednisolone with or without streptokinase, or an­terior chamber paracentesis; no assessments of retinal flow were made in these patients, although both groups were admitted for 24- hour observation. Improvement of visual acuity and retinal flow was noted between 1 and 7 days after treatment in 8 of the 11 aggressively treated J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 178 L. J. BALCER AND S. L. GALETTA patients. In contrast, none of the five patients in the " non-systematic" group had improvement of visual acuity. In­terestingly, 10 of 11 patients in the aggressive treatment group underwent all steps in the regimen before the time when improvement was noted. In the aggressive treat­ment group, all of the patients ( 8/ 11) who had improve­ment of visual acuity had an onset- treatment interval of 12 hours or less; this interval was 13 hours or greater among the three patients who did not improve. Among the five patients in the " nonsystematic" group, none of whom experienced visual improvement, the onset-treatment interval ranged from 2 to 6 hours. As the au­thors point out, larger studies with random assignment of patients to treatment groups will be needed to determine the effectiveness of various regimens for acute central retinal artery occlusion, and the extent of treatment nec­essary in this setting ( 18). In a retrospective noncomparative case series, Richard et al. ( 19) studied the use of recombinant tissue plasmin­ogen activator ( rTPA) in the treatment of retinal arterial occlusion. Twenty- five of 51 patients treated ( 47%) showed an improvement of more than 2 lines of visual acuity. No change in visual acuity occurred in 12 patients ( 22%). The mean occlusion time was 14 hours ( range, 3- 50) and there was no correlation between occlusion time and visual outcome. The authors ( 19) conclude that rTPA provides a reasonable treatment for patients with retinal artery occlusion. Furthermore, rTPA may be use­ful in the lysis of secondary thromboemboli that form at sites of nonthrombotic embolic occlusion ( i. e., calcium or cholesterol emboli). In an accompanying editorial, Hayreh ( 20) criticizes the paper by Richards et al. ( 19) on the basis of its retrospective uncontrolled nature. Fur­thermore, Hayreh demonstrates from his own series that the natural history of this disorder may not be very dif­ferent from that of patients treated with rTPA. Clearly, a prospective clinical trial of rTPA in retinal artery occlu­sion is needed to settle the issue of its therapeutic benefit. Until such results are available, clinicians and patients will have to weigh their options based on anecdotal data. Amaurosis Fugax Recurrent amaurosis fugax often occurs in the setting of carotid artery atherosclerotic disease or stenosis. For this reason, noninvasive imaging of the carotid system is indicated for patients with amaurosis fugax and other forms of transient ischemic attack to determine if a sig­nificant stenosis is present. In patients without hemody-namically significant carotid artery stenosis, other etiol­ogies, such as hypoperfusion or ocular small vessel dis­ease, may produce recurrent amaurosis fugax. McKibbin and Verma ( 21) investigated the potential role of these factors in 15 patients who presented with recurrent amaurosis fugax in the absence of hemodynamically sig­nificant carotid stenosis. Patients and age- matched con­trols were studied using duplex scanning of the extra­cranial carotid system; pulsatile ocular blood flow, ocu­lar pulse amplitude, and ocular perfusion pressure were recorded and calculated using the Ocular Blood Flow Tonograph ( OBF Labs, UK). While median ocular per­fusion pressure and pulsatile ocular blood flow were not significantly different between the patient control groups, pulse amplitude was significantly greater in the control group ( P = 0.01, Wilcoxon test). According to the authors ( 20), the significant difference in pulse am­plitude provides evidence for a potential role for ocular small vessel disease in the pathogenesis of recurrent amaurosis fugax in patients without hemodynamically significant carotid artery stenosis. Reduced compliance or narrowed lumen within ocular small vessels ( particu­larly ciliary vessels) are postulated as potential mecha­nisms for the observed reduction in pulse amplitude for patients in this study with recurrent amaurosis fugax. THE OPTIC NERVE Optic Nerve Compression by the Carotid Artery Compression of the optic nerve by the supraclinoid carotid artery is an uncommon cause of progressive uni­lateral or bilateral optic neuropathy. The development of MRI techniques, however, has facilitated diagnosis of this entity in patients with otherwise unexplained pro­gressive visual loss. The patterns of visual loss and optic atrophy that may develop in this setting were recently discussed by Jacobson ( 22). He reviewed the neuro-ophthalmologic and MR imaging records for 18 patients ( 24 eyes) to characterize the clinical features and course of MRI- defined optic nerve compression by the supra­clinoid carotid artery. At the time of diagnosis, patients ranged in age from 28 to 86 years ( median, 72); 8 were women and 10 were men. Evidence for unilateral optic neuropathy was demonstrated in 12 patients, while 6 had bilateral optic neuropathies. The most common patterns of visual field loss were those reflecting nerve fiber bundle injury ( superior and inferior arcuate defects, al-titudinal defects); visual field loss was present in all pa­tients examined. Visual acuities in affected eyes ranged from 20/ 15 to light perception. Visual acuity loss in most of these patients had progressed slowly over a 5- to 20- year period. Despite possible clinical similarities be­tween optic nerve compression by the carotid artery and normal- pressure glaucoma, Jacobson ( 22) pointed out several important features that may help distinguish these two entities. Patients with MRI- defined optic nerve compression by the carotid artery had not only nerve fiber bundle visual field defects and optic disc excava­tion ( which are seen in glaucoma), but also demonstrated pallor of the neuroretinal rim, impairment of visual acu­ity, and abnormal color vision. The presence of these features should prompt MR imaging with careful atten­tion to the relation of the optic nerves and chiasm to the supraclinoid carotid arteries, in addition to the usual im­aging protocols to exclude other compressive lesions. Optic Nerve Head Blood Flow Laser Doppler flowmetry serves a potential role in the diagnosis and monitoring of retinal and optic nerve head blood flow in patients with anterior ischemic optic neu­ropathy. As outlined by Petrig et al. ( 23), the optic nerve head is supplied by 1) the posterior ciliary arteries within J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 PREGENICULA TE REVIEW 1999 179 the deeper layers, and 2) the central retinal artery in the superficial layers. To determine which circulation is ac­tually measured by the laser Doppler flowmetry tech­nique, Petrig et al. ( 22) performed a study of optic nerve head blood flow in the eyes of Rhesus monkeys. These authors measured blood flow in optic nerve tissue at baseline and after experimental occlusion of the posterior ciliary arteries ( 9 eyes), central retinal artery ( 12 eyes), and both posterior ciliary and central retinal arteries combined ( 9 eyes). Compared with laser Doppler flow­metry measurements at baseline, optic nerve head blood flow was significantly reduced after occlusion of the cen­tral retinal artery ( by 39%, P < 0.001), but not after occlusion of the posterior ciliary artery alone ( flow re­duced by 17%, difference not statistically significant). Combined occlusion of the posterior ciliary and central retinal arteries resulted in a significant reduction of optic nerve head blood flow compared with baseline ( by 57%, P < 0.0005), but this decrease in flow was not signifi­cantly different from that produced by central retinal artery occlusion alone. These results ( 23) indicate that laser Doppler flowmetry measurements reflect most strongly the blood flow within the most superficial layers of the optic nerve head; these layers are supplied by the central retinal artery. Because the posterior ciliary artery circulation represents the primary lesion site in anterior ischemic optic neuropathy, techniques that selectively measure flow in this circulation are needed. Cigarette smoking is a known risk factor for systemic atherosclerotic vascular disease. Its effects on the optic nerve head and choroidal circulations, however, have not been evaluated. Tamaki et al. ( 24) used the laser speckle method ( a fundus camera equipped with a diode laser and image sensor) to determine tissue blood flow veloc­ity in the optic nerve head and choroidal circulations of healthy habitual smokers. Nine healthy habitual cigarette smokers ( mean age, 28 ± 4 years; number of cigarettes per day, 27 ± 10; length of smoking history, 10 ± 4 years) participated in the study. Mean blood flow velocity in the optic nerve head circulation ( normalized blur) was sig­nificantly higher at 1 to 5 minutes and at 25 to 45 minutes after cigarette smoking compared with values obtained after sham smoking ( P < 0.05, paired t- test). Nonsignif­icant increases in choroidal blood flow velocity were also noted. These results indicate that cigarette smoking increases blood flow velocity in the optic nerve head circulation and possibly in the choroidal circulation, as measured by the laser speckle method. The authors sug­gest that compensatory increases in tissue blood flow to maintain the oxygen supply after smoking may be re­sponsible for the findings of this study. The long- term effects of cigarette smoking on changes in optic nerve head blood flow over time remain to be investigated. Anterior Ischemic Optic Neuropathy Potential risk factors, precipitants, and pathogenic mechanisms for nonarteritic anterior ischemic optic neu­ropathy ( NAION) were again topics for investigation and discussion during 1999 ( 25- 31). Salomon et al. ( 25) per­formed a case- control study with retrospective ascertain­ment to examine the potential role for genetic and ac­quired thrombophilias in the pathogenesis of NAION. Sixty- one consecutive patients diagnosed with NAION between 1984 and 1997 were examined for evidence of thrombophilias ( blood samples drawn at time of study), including testing for protein C, protein S, antithrombin III, lupus anticoagulant, anticardiolipin antibodies, and three recently described prothrombotic polymorphisms ( factor V G1691 A, factor II G20210A, and methylene-tetrahydrofolate reductase [ MTHFR] C677T). Medical histories were also examined for potential arterioscle­rotic risk factors, including hypertension, diabetes mel-litus, hypercholesterolemia (> 260 mg/ dL), ischemic heart disease, arrhythmias, arterial and venous throm­boembolism, and smoking. Testing for thrombophilias and assessment of medical history were also performed for a control group of 90 consecutive patients of similar age and gender to the NAION group. Control patients had presented to the same institution for diagnoses other than NAION, retinal vein occlusion, or retinal artery oc­clusion. Median ages were similar for the NAION and control groups ( 62 years, range 34- 85 for NAION group; 66 years, range 31- 85 for control group). Among the genetic and acquired thrombophilic markers examined, none were significantly associated with NAION. Several vascular risk factors, however, were associated with NAION in this study, including ischemic heart disease, hypercholesterolemia, and diabetes mellitus. Patients with both ischemic heart disease and hypercholesterol­emia demonstrated additive risk for NAION. None of the vascular risk factors, however, were significant predic­tors of second eye involvement during follow- up. As stated in an accompanying editorial by Feldon ( 26), ge­netic and acquired thrombophilias may represent poten­tial triggers for the development of NAION in suscep­tible patients, despite the fact that the study by Salomon et al. ( 25) did not demonstrate an association with NAION in the population examined. Vasculopathic con­ditions, particularly ischemic heart disease, hypercholes­terolemia, and diabetes, may represent important medical risk factors for NAION. Despite our increasing knowl­edge of potential risk factors for NAION, however, much remains unknown regarding the exact underlying patho­genesis of this disorder, as eloquently summarized in an editorial by Lessell ( 27). In a recent report by Fivgas and Newman ( 28), nasal decongestants were implicated as possible precipitants for NAION in susceptible patients. A 43- year- old woman developed bilateral sequential NAION; each epi­sode occurred several hours after the use of oxymetazo-line, a sympathomimetic nasal spray. The patient had a history of hypertension and diabetes mellitus. The epi­sode of painless visual loss occurred on the morning after the patient's first use of oxymetazoline. Visual acuity in the affected right eye several days later was 20/ 400; a right afferent pupillary defect and disc edema were noted. Visual field examination showed an inferior alti-tudinal defect OD. MRI revealed subtle right optic nerve enhancement, but no evidence of white matter demye- J Neuro- Ophthalmol, Vol. 20, No. 3. 2000 180 L. J. BALCER AND S. L. GALETTA linating lesions. Results of spinal fluid analysis, carotid Doppler studies, and serologic tests were normal. Several weeks later, the patient used oxymetazoline nasal spray for the second time. She again awoke the following morning with painless visual loss, this time OS. Exami­nation revealed visual acuities of 20/ 400 OD and 20/ 70 OS. Residual slight disc elevation and pallor were noted OD; the left optic disc was swollen with exudates and scattered hemorrhages. Inferior altitudinal defects were present OU. This patient's clinical findings were most consistent with sequential NAION, including painless vi­sual loss, inferior altitudinal defects, and disc edema. Several potential risk factors were present in this patient, including hypertension, diabetes, and crowded optic discs. The occurrence of visual loss within hours after her only two uses of oxymetazoline nasal spray strongly suggests a temporal causal relationship for this sympa­thomimetic drug and NAION in this patient. As sug­gested by the authors ( 28), the vasoconstrictive effects of oxymetazoline likely aggravated the patient's other risk factors, leading to optic nerve head ischemia. Transient hypotension is another known precipitant for NAION in patients at risk. Jackson et al. ( 29) reported a patient who developed bilateral NAION in the setting of continuous ambulatory peritoneal dialysis. This 58- year- old man had a history of nephrocalcinosis and con­sequent renal failure. There was no history of hyperten­sion, diabetes, or hypercholesterolemia. The patient was evaluated for a two- day history of painless visual loss OU; his symptoms were worse on standing. Examination revealed visual acuities of 20/ 24 OD and 20/ 200 OS; there was marked reduction of color vision OS and a left APD. The optic discs were swollen with dilated capil­laries and flame- shaped hemorrhages. Visual fields were remarkable for an inferior altitudinal defect OS, with generalized constriction OD. Although blood pressure measurements in the office were within an acceptable range ( 122/ 68- 130/ 86), 24- hour ambulatory blood pres­sure monitoring revealed evidence of early morning hy­potension, with mean pressures of 94/ 53 ( lowest reading, 91/ 41) from midnight to 6: 30 A. M. The combination of anemia, crowded discs, and hypotension may have been contributing factors for this patient's visual loss. The authors suggest the use of 24- hour ambulatory blood pressure monitoring to document hypotension in patients at risk. To investigate the possibility of an association be­tween ophthalmic ischemic disorders, including NAION, and hearing loss, Hayreh et al. ( 30) evaluated 583 con­secutive patients. These patients comprised the following groups, according to diagnosis: NAION ( 81 patients), normal- tension glaucoma ( 36 patients), primary open-angle glaucoma ( 138 patients), other types of glaucoma ( 142 patients), ocular arterial occlusive disorders ( 22 pa­tients), retinal vein occlusion ( 89 patients), ocular vas­culitis ( 42 patients), and miscellaneous ( 33 patients with no history of ocular disease or with nonischemic ocular disorders). None of the ocular diagnoses were signifi­cantly associated with hearing loss independent of age and gender. Although formal audiography was not rou­tinely performed to assess hearing loss as part of this study, the results of this exploratory investigation ( 30) argue against a strong association of hearing loss and ocular ischemic disorders in this predominantly older population ( median age, 68 years). The prevalence of family history of NAION among first- and second- degree relatives of patients with NAION was investigated by Wang et al. ( 31). These authors mailed a short survey questionnaire to 148 con­secutive patients whose records had indicated a diagnosis of NAION. Follow- up phone called were placed to pa­tients who reported family members with NAION. Medi­cal records of affected family members were reviewed for the following NAION criteria: sudden onset of pain­less visual loss, arcuate or altitudinal visual field loss, sedimentation rate < 40 mm/ h or negative results of a temporal artery biopsy, and ophthalmoscopic appearance of optic neuropathy. Among all patients with NAION surveyed, 79 ( 53%) returned questionnaires; among these patients, 4 ( 2.7%) reported relatives with NAION. Among the four families, nine patients ( of the ten iden­tified) gave consent for review of their medical records. Familial cases appeared to have an earlier age of onset of first NAION, with mean ages of 55.0 ± 14.9 years in the familial group ( n = 9) versus 63.3 ± 10.5 years in the isolated NAION group ( n = 144). The familial group also had a higher incidence of second eye involvement ( 56% vs. 15% by 4 years after initial NAION). Although the prevalence of NAION in the setting of affected sib­lings was higher in this study ( 2.7%) than the previously estimated prevalence of NAION in the general popula­tion ( 0.04%), the authors caution that ascertainment bias, or an increased likelihood for patients to have responded to the questionnaire if they had affected relatives, may have at least partially affected the results. The prevention of second eye involvement in patients with a first episode of NAION remains a hot topic of discussion. A retrospective study by Salomon et al. ( 32) recently suggested potential benefits for aspirin therapy in this setting. These authors reviewed the records of 52 patients who presented with NAION between 1984 and 1997; this was the same patient cohort that was investi­gated in a study of thrombophilic and vascular risk fac­tors for NAION ( 25: see above). Among this patient group ( n = 52), 16 developed second eye involvement during follow- up, with a mean interval of 95.7 months between the first and second episodes. Thirty- six patients took aspirin after the first episode of NAION; 16 patients did not take aspirin. Although the differences did not reach statistical significance, a higher proportion of pa­tients who did not take aspirin had second eye involve­ment ( 8/ 16 [ 50%] patients; mean interval to second event, 63.4 months) compared with those who took 100 mg aspirin per day ( second event in 3/ 8 [ 38%] patients; mean interval, 80.3 months) and those who took 325 mg aspirin per day ( second even in 5/ 18 [ 18%] patients; mean interval, 156.5 months). These differences were observed despite a higher prevalence of other potential J Neuro- Ophthalmol Vol. 20, No. 3, 2000 PREGENICULATE REVIEW 1999 181 risk factors ( such as hypertension, diabetes, ischemic heart disease, hypercholesterolemia, smoking, and crowded optic discs) among the 325 mg aspirin group. These findings are interesting, but the retrospective na­ture of the study and small sample size make it difficult to draw any definitive conclusions regarding the effec­tiveness of aspirin in NAION. Posterior Ischemic Optic Neuropathy Posterior ischemic optic neuropathy is an uncommon entity that may occur after severe blood loss. Alexan-drakis and Lam ( 33) reported a patient with bilateral posterior ischemic optic neuropathy that developed dur­ing spinal surgery. This 68- year- old woman with a his­tory of insulin- dependent diabetes of 43 years' duration underwent an L3- L5 spinal decompression and fusion procedure with general anesthesia. The patient was placed in the prone position for the surgery. Eight hours were required for the procedure, and the estimated blood loss was 3 L ( she was transfused with 5 units packed red blood cells). The postoperative hemoglobin concentra­tion was 12.0 g/ dL, and intraoperative systolic blood pressures ranged from 95 to 110 mm Hg. After surgery, the patient awoke with complete blindness ( no light per­ception) OU. Pupils were markedly sluggish to light without an APD. External examination revealed moder­ate periorbital edema and conjunctival chemosis. Nasal fullness of the optic discs was present ( attributed to drusen, which were confirmed by ultrasound); no disc edema or pallor was noted. The retinas were remarkable for mild background diabetic retinopathy. Diffuse pallor of the optic discs developed over the ensuing weeks, and there was no recovery of vision. Extensive testing and examination in this case therefore localized the visual pathway dysfunction to the retrobulbar optic nerves. The authors suspect that severe selective hypoperfusion of the retrobulbar optic nerves occurred in this patient in the setting of prolonged surgery, severe blood loss, perior­bital edema with venous stasis, and longstanding diabe­tes mellitus. Another surgical procedure associated with rare com­plications of visual loss is blepharoplasty. A patient who developed posterior ischemic optic neuropathy after lower lid blepharoplasty was recently described by Good et al. ( 34). A 68- year- old woman underwent carbon di­oxide laser therapy and cosmetic lower lid blepharo­plasty. The evening after surgery, the patient developed marked bilateral periorbital swelling and bruising; she was unable to open the right eye. The following day, she noted rapid visual loss OD to no light perception. An examination confirmed no light perception vision OD and 20/ 20 acuity OS. Marked periorbital swelling and subconjunctival hemorrhages were noted, particularly OD. An MRI scan 5 days after surgery demonstrated abnormal signal within the right retrobulbar fat that was consistent with hemorrhage and diffuse edema. The right intraorbital optic nerve had abnormal high signal on T2- weighted images ( 7 mm in length) extending posteriorly from 13 mm behind the globe; this signal abnormality involved the full thickness of the optic nerve on coronal T2- weighted images. The clinical findings and MR im­aging were consistent with infarction of the posterior portion of the intraorbital optic nerve. Increased intraor­bital pressure and edema with compromise of the cen­tripetal pial plexus, the blood supply to the affected seg­ment of the optic nerve, were implicated- as causative factors for the rapid visual loss experienced by this pa­tient. The authors recommend careful monitoring for postoperative increases in intraorbital pressure in pa­tients undergoing blepharoplasty and other surgical pro­cedures. Giant Cell Arteritis: Arteritic Ischemic Optic Neuropathy/ Ocular Ischemic Syndrome The presence of enhancement of the optic nerve on MR imaging usually argues for an inflammatory or in­filtrative process. Such enhancement of the optic nerve, however, may occur in patients with arteritic ischemic optic neuropathy, as demonstrated in a report of three patients by Lee et al. ( 35). All three patients had biopsy-proven giant cell arteritis and ranged in age from 80 to 86 years. Two patients presented with acute bilateral severe visual loss and pallid optic disc edema; the third patient had sequential episodes of visual loss, including a branch retinal artery occlusion followed 3 months later by pos­terior ischemic optic neuropathy in the contralateral eye. Two patients experienced constitutional symptoms be­fore vision loss, including anorexia, scalp tenderness, jaw claudication, and fatigue. MRI scans of the brain and orbits in each case demonstrated bilateral optic nerve enhancement. These cases indicate that disruption of the blood- brain barrier, as demonstrated by optic nerve en­hancement, may be a consequence of giant cell arteritis. Optic nerve enhancement in the setting of arteritic AION has been only rarely documented; however, arteritic AION should be considered in elderly patients with optic nerve enhancement and acute visual loss ( 35). Ocular ischemic syndrome is another uncommon but devastating consequence of giant cell arteritis. Bilateral ocular ischemic syndrome is even more rare in this set­ting, but, as demonstrated in a report by Hwang et al. ( 36), it may occur in patients with giant cell arteritis, despite early aggressive treatment with high- dose intra­venous corticosteroids. A 76- year- old woman developed double vision, headache, and jaw claudication. Sudden complete loss of vision OD occurred 2 weeks later. Ex­amination revealed no light perception OD. Folds of the Descemet membrane were noted OD, and mild anterior chamber flare was present OU. Intraocular pressures were 3 mm Hg OD and 8 mm Hg OS. Abduction OD and adduction OS were diminished. The right optic disc was swollen. The erythrocyte sedimentation rate was 67 mm/ h. Treatment with intravenous methylprednisolone ( 250 mg every 6 hours for 3 days) was initiated; giant cell arteritis was confirmed by temporal artery biopsy. Severe stenosis of the right internal carotid artery was also noted by Doppler imaging. Despite treatment, the patient re­turned 1 week after discharge ( on 60 mg prednisone/ d) with signs of ocular ischemic syndrome OS, including corneal edema, uveitis, reduced visual acuity ( 20/ 160), J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 182 L. J. BALCER AND S. L. GALETTA pupillary distortion, and ocular hypotony ( intraocular pressure, 1 mm Hg). Visual acuity OS deteriorated to no light perception, despite retreatment with high- dose in­travenous methylprednisolone and heparin. This report demonstrates that bilateral ocular ischemic syndrome may occur as a rare manifestation of giant cell arteritis. As suggested by the authors, other factors, including hy­pertension, anemia, and vascular stenosis, may have con­tributed to the severe and refractory nature of this pa­tient's clinical course. The question of whether the performance of bilateral temporal artery biopsies increases diagnostic yield in pa­tients with suspected giant cell arteritis was addressed in a retrospective study by Boyevet al. ( 37). They reviewed the pathologic reports from 908 consecutive temporal artery biopsy specimen received at the Wilmer Ocular Pathology Laboratory during 1968 to 1996. These speci­men ( n := 908) were from 758 patients, and they repre­sented the following categories: 1) 300 specimen from bilateral simultaneous biopsies ( 150 patients), 2) 72 specimen from bilateral sequential biopsies ( 36 patients), and 3) 536 specimen from unilateral biopsies ( 536 pa­tients). Among the 186 patients who underwent bilateral ( simultaneous or sequential) temporal artery biopsies, the diagnoses were identical for both biopsies in 176 pa­tients. In the remaining ten patients, four had no artery present on one side, while six had different diagnoses for the two biopsies. The final diagnosis in five of the six cases was giant cell arteritis; one patient, for whom one biopsy revealed mild nonspecific arteritis, was diagnosed with prostate cancer. Based on these results, bilateral biopsies will yield the same results for each side approxi­mately 97% of the time. While this number is somewhat reassuring, the performance of a second biopsy should be considered in inconclusive cases, particularly in light of the potential complications of both giant cell arteritis and its treatment ( 37). Optic Neuritis and Multiple Sclerosis The Optic Neuritis Treatment Trial ( ONTT) has been perhaps the most important and influential clinical trial in neuro- ophthalmology. The impact of this trial on the practices of ophthalmologists and neurologists was re­cently studied by Trobe et al. ( 38). These authors sought to evaluate whether the practice patterns of ophthalmolo­gists and neurologists have been affected by results of the ONTT and how specific recommendations based on ONTT results have been interpreted and acted upon by practicing physicians. A random sample of 987 ophthal­mologists and 900 neurologists were mailed a 4- page, 24- question survey. These physicians were chosen at random from memberships of the American Academy of Ophthalmology and American Academy of Neurology; the numbers of surveys mailed were chosen in order to detect an absolute difference of 15% between the two specialties with respect to responses to questions on the treatment of optic neuritis. Among the 1,887 surveys mailed, 865 physicians returned questionnaires with us­able responses ( i. e., questionnaire was returned and phy­sician was not retired or unreachable), yielding an overall response rate of 47% ( 53% for ophthalmologists, 40% for neurologists). Among the 509 ophthalmologists and 356 neurologists who responded to the survey, 357 ( 70%) and 257 ( 72%), respectively, had encountered at least 1 patient during the previous year with optic neu­ritis. Within this group of physicians, 202/ 357 ophthal­mologists ( 57%) and 244/ 257 neurologists ( 95%) had initiated or carried out complete treatment for patients with optic neuritis; analyses of survey responses were based upon this group of 446 physicians. The reported changes in treatment practices for optic neuritis based upon the initial ONTT publication were presented ( 38A). Most dramatic were the percentages of both ophthalmolo­gists ( 90%) and neurologists ( 95%) reporting changes in their use of oral prednisone; 100% of physicians from both groups who reported such a change indicated that they now use less oral prednisone. Ophthalmologists were generally less inclined to treat acute optic neuritis; 53% of ophthalmologists versus 16% of neurologists re­ported choosing " no treatment" > 50% of the time (" usu­ally" or " always") for patients with optic neuritis ( Table 1). Significant differences between ophthalmologists and neurologists were also found with respect to reasons for prescribing intravenous methylprednisolone, as demon­strated in Table 2. Among physicians who reported use of intravenous methylprednisolone for > 50% of cases of optic neuritis, a significantly higher proportion of neu­rologists ( 59%) versus ophthalmologists ( 38%, P = 0.0029) reported " improvement of 1- year visual out­come" as a very important reason for use of intravenous methylprednisolone, despite ONTT results indicating no difference in visual outcome between treatment groups. Finally, reasons for ordering MRI scans also differed between neurologists and ophthalmologists, with signifi­cantly more ophthalmologists ( 36%) than neurologists ( 7%, P = 0.0001) reporting use of MRI to decide wheth­er to treat patients with acute optic neuritis, and a sig­nificantly greater proportion of neurologists ( 65%) ver­sus ophthalmologists ( 45%, P = 0.0001) ordering MRI scans to determine prognosis for developing future neu­rologic events of multiple sclerosis ( MS). These results provide important evidence that practitioners may change their practices as a result of major collaborative clinical trials. However, as illustrated by the high per­centages of physicians, particularly neurologists, who re­port that they now prescribe intravenous methylprednis­olone for reasons of " improving 1- year visual outcome," the recommendations of such trials may be embraced without a complete understanding of the results. The in-terspecialty differences demonstrated in this survey, par­ticularly the greater tendency for neurologists to treat patients with intravenous methylprednisolone, may re­flect differing perspectives on the implications of acute optic neuritis and its role as a potential first manifestation of MS, even in tht setting of a normal MRI. Important results from the ONTT have also continued to emerge. A recent series of articles ( 39- 42) reported the findings of analyses of visual field defects observed in the ONTT, including comparisons of central versus / Neuro- Ophthalmol, Vol. 20, No. 3, 2000 PREGENICULATE REVIEW 1999 183 TABLE 1. Current treatment of optic neuritis* Prednisone alone Ophthalmologists ( n = 180) Neurologists ( n = 236) Intravenous methylprednisolone ± prednisone Ophthalmologists ( n = 160) Neurologists ( n = 171) No treatment Ophthalmologists ( n = 181) Neurologists ( n = 183) Never ( 0%) 74 68 8 1 3 22 Sometimes ( l%- 49%) 18 28 52 17 44 62 Usually ( 50%- 90%) 6 3 34 53 49 14 Always ( 100%) 2 1 6 28 4 2 Reprinted with permission from Trobe JD, Sieving PC, Guire KE, Fendrick AM. The impact of the Optic Neuritis Treatment Trial on the practices of ophthalmologists and neurologists. Ophthalmology 1999; 106: 2047- 53. * Values are percentages, n, number of specialists who supplied complete information about this aspect of their treatment practices. peripheral and focal versus global defects. Keltner et al. ( 39) compared the results of peripheral kinetic visual field testing ( Goldmann I3e and II4e isopters) and central static perimetry ( Humphrey Field Analyzer, program 30- 2) for 448 patients enrolled in the ONTT. They sought to determine 1) whether peripheral or central field testing was more sensitive to visual field loss and recovery, and 2) whether testing of the far peripheral visual field pro­vided additional useful information beyond that provided by the results of central static perimetry. For both af­fected and fellow eyes at all visits, central static perim­etry yielded abnormal results more frequently than did peripheral kinetic visual field testing. At the baseline visit, 97.1% of affected eyes had abnormal Humphrey mean deviations, while 69.9% had abnormal peripheral kinetic perimetry results. In terms of recovery of visual field defects, 80% of Goldmann isopters ( I3e and II4e) that were abnormal at baseline had returned to normal by day 30 of the trial, while central static fields were slower to recover, returning to normal in only 70% of affected eyes by week 19. The authors ( 39) therefore concluded that for affected eyes in acute optic neuritis, central static perimetry demonstrates abnormalities more frequently than does peripheral kinetic visual field testing. This is true both at baseline and during the recovery phase. Re­covery of the far periphery of the visual field is also more rapid and complete. While Goldmann kinetic perimetry provides useful additional information regarding the far periphery of the visual field in cases of severe central visual loss, automated central static perimetry appears to provide effective visual field monitoring for patients with acute optic neuritis ( 39). Whether particular areas of the visual field are pref­erentially affected in patients with acute optic neuritis was investigated by Fang et al. ( 40). These authors also sought to determine whether patients with apparently " localized" visual field defects also demonstrated more diffuse involvement of the entire visual field. Central static perimetry data ( Humphrey Field Analyzer, pro­gram 30- 2) from the affected and fellow eyes of 440 ONTT patients at the baseline ( entry) visit were re­viewed. Average thresholds were calculated for the en­tire 30- 2 visual field and for test locations contained within concentric rings spaced 3, 9, 15, 21, and 27° from fixation. Among patients with diffuse visual field loss ( 206 patients [ 61%]), a relatively equal diminution of visual threshold was present throughout the entire 30- 2 field. Depressions of visual threshold were greatest cen­trally among patients with localized central and cecocen-tral scotomas ( 37 patients [ 11%]). However, depression of the entire tested visual field was also present in these patients, even among those whose central defects were mild (< 6 dB depression). In fact, the degree of peripheral depression was greatest for those with the most severe central defects, indicating a relation between the severity of central versus peripheral loss. This study ( 40) there- TABLE 2. Reasons for prescribing intravenous methylprednisolone* Percent who consider this reason very important Reason Ophthalmologists (%) Neurologists (%) Most effective way to improve 1- year visual outcome Most effective way to hasten visual recovery Most effective way to reduce future neurologic events of MS Patient preference Consultants recommend it Colleagues are using it 38 46 66 24 20 13 59 72 48 15 11 9 0.0029 0.0002 0.0114 0.09 NS NS Reprinted with permission from Trobe JD, Sieving PC, Guire KE, Fendrick AM. The impact of the Optic Neuritis Treatment Trial on the practices of ophthalmologists and neurologists. Ophthalmology 1999; 106: 2047- 53. * Among practitioners who report prescribing this agent in 50% or more of cases. MS, multiple sclerosis; NS, not significant. J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 184 L. J. BALCER AND S. L. GALETTA fore provides evidence that optic neuritis may affect the entire central visual field, even in patients with appar­ently localized defects. Patterns of visual recovery within affected portions of the central visual fields of patients with optic neuritis ( affected eyes) also were investigated by Fang et al. ( 41). Data from the initial and 6- month Humphrey 30- 2 visual fields of 383 patients from the ONTT was reviewed. Improvement in average threshold from the initial to 6- month visit was calculated for each patient. Potential patterns of recovery among regions of the central field were assessed by comparing measurements of average threshold within the areas bounded by the concentric rings. The greatest degree of recovery was noted for the area about fixation ( 89% improvement of threshold for central 3°). The degree of recovery also decreased with increasing eccentricity from fixation ( 83% for points lo­cated 27° from fixation); this association was statistically significant ( P < 0.05, analysis of variance). Patients with localized and diffuse visual field defects demonstrated similar percentages of improvement between the initial visit and 6 months ( 86% for localized vs. 85% for diffuse defects). The return of visual function in patients with optic neuritis, based on central static perimetry results, is extensive (> 80%) and does not appear to differ with respect to initial visual field defect type ( localized vs. diffuse loss). As indicated by the authors, the greater degree of recovery observed in the central portions of the visual field may relate to a greater redundancy of axons subserving the areas near fixation. Important results from the preceding three studies of visual field testing in the ONTT ( 39^ 11) are well summarized and thoughtfully discussed in an accompanying editorial by Arnold ( 42). One very important aspect of clinical trials and other research studies is the generalizability of their results across a variety of patient populations. Recently pub­lished results of a randomized controlled trial in Japan ( 43- 44) were consistent with those obtained in the North American ONTT, indicating more rapid recovery but no long- term differences in visual outcome for patients with acute optic neuritis treated with intravenous methylpred-nisolone versus a control drug ( mecobalamin). The trial in Japan began in 1991 and included 66 patients from 22 clinical centers. Inclusion criteria were as follows: aged 14 to 55 years; acute symptoms indicative of unilateral optic neuritis of unknown or demyelinating origin; visual symptoms of 14- days duration or less; APD in the af­fected eye; and normal or swollen optic disc in the af­fected eye. Participants were randomized to receive ei­ther intravenous methylprednisolone ( 1 g/ d for 3 days) followed by a 7- to 10- day oral corticosteroid taper, or intravenous mecobalamin ( 500 | xg/ d for 3 days) followed by oral mecobalamin for at least 7 days ( control group). Mecobalamin, which promotes recovery of experimen­tally induced neuronal degeneration, was used as a con­trol drug for this trial because the study group considered it unethical to use a placebo treatment. Ocular or perio­cular pain was reported by 56% of participants in this trial ( vs. 92% in the ONTT). Four patients ( 5.6% vs. 13% in the ONTT) had clinically definite MS at the time of onset of visual loss, and periventricular white matter le­sions were seen on MRI scans in nine patients ( 14%). These differences likely reflect the lower incidence of MS in Japan compared with North American popula­tions. A high prevalence of disc swelling Was noted ( 50% vs. 35% in the ONTT), but this may be attributable to the common use of fluorescein angiography as a tool for diagnosing disc swelling in Japanese ophthalmology clinics ( 44). Despite these potential differences between the populations studied for the Japanese trial versus the ONTT, the results were similar in terms of visual recov­ery and visual outcome. The methylprednisolone group demonstrated more rapid recovery of visual function, particularly for visual acuity at 1 week ( P = 0.03, t- test), Humphrey visual field mean deviation at 4 weeks ( higher proportion of normal fields in treatment group; P = 0.01, X2 test), and color vision at 1 week { P = 0.01, t- test). Recovery of contrast sensitivity at several differ­ent spatial frequencies ( Vision Contrast Test System method) was significantly more rapid in the methylpred­nisolone group, with higher proportions of normal scores in this group at 1 week ( P < 0.01, X2 test), 2 weeks ( P < 0.05), and 4 weeks ( P < 0.05). At 12 weeks and 1 year after the initiation of treatment, however, visual function test scores were essentially the same for both groups. The results of these studies ( 43- 44) were therefore in agree­ment with those of the ONTT, showing a more rapid recovery of visual function in the methylprednisolone group, but no effect on long- term visual outcomes. Be­cause the incidence of MS in the Japanese population is very low, the development of clinically definite MS after monosymptomatic optic neuritis was not examined ( 43). Optic neuritis is a frequent clinical manifestation of MS, but it is also a prominent feature of neuromyelitis optica ( Devic syndrome). The clinical features in a large series of patients with neuromyelitis optica ( NMO) were recently reviewed by Wingerchuk et al. ( 45). These au­thors reviewed the medical records and MRI studies for all patients diagnosed with NMO at the Mayo Clinic, Rochester, Minnesota, between 1950 and 1997 ( 45). Pa­tients who met the " strict" case definition of NMO had bilateral optic neuritis and myelitis, the index ( first) events, occurring within 2 years of one another, without symptomatic disease outside of the optic nerve and spi­nal cord; other patients were considered to have NMO but to not meet " strict" case criteria. The clinical course for each patient was classified as either monophasic or relapsing. Patients with monophasic ( n = 23) and re­lapsing ( n = 48) NMO differed with respect to several demographic and neurologic characteristics. Patients with relapsing NMO were significantly older ( mean age, 39 vs. 29; P = 0.009, Wilcoxon rank sum test), more often female ( female to male ratios, 40: 8 vs. 11: 12; P = 0.008, X2 test), more frequently had a history of autoim­mune disease ( 30% vs. 0%; P = 0.003, Fisher exact test), and more frequently had optic neuritis as an index event ( 48% vs. 26%; P = 0.001, exact X2 test for pooled data). Index events were associated with more severe J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 PREGENICULATE REVIEW 1999 185 neurologic and visual impairment in the monophasic group. However, respiratory failure caused by acute cer­vical myelitis occurred in 16 patients ( 33%; 15 died) with a relapsing course, but only in 2 ( 9%) monophasic patients, both of whom recovered. The 5- year survival rate was consequently lower in the relapsing group ( 68% vs. 90%; P = 0.06, log- rank test). At last available fol­low- up, significantly higher proportions of relapsing pa­tients had visual acuities of 20/ 200 or worse in one or both eyes ( 60% vs. 22%; P = 0.014, Wilcoxon rank sum test) or had permanent monoplegia or paraplegia ( 52% vs. 31%; P - 0.086). The presence of more than 50 white blood cells/ mm3 in the cerebrospinal fluid ( CSF) was a distinguishing feature from " typical" MS fre­quently present in patients with NMO; these patients also tended to have normal initial brain MRI scan results and lesions extending over three or more vertebral segments on spine MRI examinations. Patients with NMO there­fore differ in important ways from patients with " typical" MS, based on clinical, CSF, and neuroimaging features. Optic neuritis in children has several clinical features that may distinguish it from the typical adult form. Brady et al. ( 46) reviewed the clinical spectrum of pediatric optic neuritis in a retrospective cohort, including present­ing signs and symptoms, neuroimaging findings, CSF abnormalities, associated systemic disease, and visual outcomes. The authors identified 25 patients ( 39 affected eyes) who were evaluated for unilateral or bilateral optic neuritis between 1991 and 1997. Patients ranged in age from 21 months to 18 years, and the mean follow- up for the group was 11 months. As suggested by previous re­ports and clinical experience, bilateral disease was com­mon in this pediatric age group, occurring in 14/ 25 pa­tients ( 56%). Visual loss was generally severe at presen­tation, with 33/ 39 eyes ( 84%) having visual acuities of 20/ 200 or worse. Recovery of visual acuity to 20/ 40 or better was seen in only 76% of eyes ( 30/ 39), while 6/ 39 eyes ( 15%) had visual acuities at last follow- up of 20/ 200 or worse. Brain and orbital MRI scans were per­formed in 23/ 25 patients ( 92%); findings included nor­mal MRI results ( 5 patients, 20%), isolated optic nerve enhancement ( 4 patients), and multiple white matter le­sions ( 15 patients). Among the five patients with normal MRI scans, 6 of 6 affected eyes recovered to 20/ 40 or better. Younger patients ( 6 years of age or younger) were also observed to have bilateral disease more often ( 6 of 7 patients, 85%), and to have better visual recovery. Among children less than 7 years of age, 12 of 13 af­fected eyes ( 92%) had recovery to 20/ 40 or better, com­pared with 18 of 26 affected eyes ( 50%) in children 7 years or older. Larger studies may confirm that bilateral disease and younger age are associated with a better visual prognosis in children with optic neuritis. Although visual prognosis is often good, this study ( 46) showed that disabling visual loss may remain in up to 22% of pediatric patients with optic neuritis. A much less common ocular disorder that may occur in patients with MS is uveitis. A group of patients with both MS and a history of uveitis was recently described by Biousse et al. ( 47). Patients having both diagnoses were identified from patient databases from an MS clinic ( n = 1098) and a uveitis clinic ( n = 1530) in Paris, France. Patients were classified on the basis of MS dis­ease severity ( Expanded Disability Status Scale score; EDSS), visual acuity, type of uveitis ( anterior, posterior, intermediate/ pars planitis, panuveitis), and presence of periphlebitis ( venous sheathing) on dilated funduscopy. The authors identified 28 patients ( of 2,628 patients) with both MS and a history of uveitis; the mean age was 47 years ( range, 28- 67), and 20 patients were women. The prevalence of patients with both diagnoses ( combin­ing both clinics) was 1% ( 28 of 2628); 12 patients were identified from the MS clinic ( 12 of 1,098, 1.1%), and 16 patients were from the uveitis clinic ( 16 of 1,530, 1%). The most common types of uveitis were pars planitis ( 10 of 28, 35.7%) and panuveitis ( 11 of 28, 39.3%). Anterior uveitis ( 4 of 28, 14.3%) and posterior uveitis ( 3 of 28, 10.7%) in isolation were less common. Eleven patients ( 39.3%) had retinal periphlebitis ( venous sheathing). Uveitis occurred before ( 39.3% of patients, 9.8 years) and after ( 57.1%, 9.1 years) the onset of neurologic symptoms. This large series ( 47) shows that the associa­tion of these two disorders ( 1% in this study) may be much less common than previously thought. Optic Atrophy and Polyneuropathy Chronic inflammatory demyelinating polyneuropathy ( CIDP) and multifocal motor neuropathy involve demy-elination in the peripheral nervous system. However, scattered reports of central nervous system demyelina-tion in patients with these disorders have been increasing in the literature, and clinical features in these patients may be indistinguishable from those of MS. Lee et al. ( 48) recently described two patients who developed optic neuropathy in the setting of CIDP ( one patient) and mul­tifocal motor neuropathy ( one patient). The first patient was a 57- year- old woman with CIDP who developed acute visual loss OS several months after the onset of pain, paresthesias, weakness, and areflexia in the arms and legs. Examination 6 months after the onset of visual loss revealed visual acuities of 20/ 20 OD and 20/ 25 OS. The visual field OS demonstrated a dense nasal arcuate defect, and the VEP PI00 was delayed OS ( normal OD). Results of an MRI scan were normal, and CSF exami­nation revealed an elevated protein ( 66 mg/ dL). Over the next several months, visual acuity OS worsened to 20/ 50, but it improved again after a course of intravenous cor­ticosteroids. Visual loss developed OD, and an MRI showed mild enhancement of the right optic nerve. The second patient, a 52- year- old woman with multifocal motor neuropathy, likewise experienced the acute onset of visual loss OD ( visual acuity 20/ 80) 1 year after di­agnosis of her neuropathy. The visual loss was associated with pain on eye movement and optic nerve enhance­ment by MRI. Similar to the first patient, this patient experienced a relapsing- remitting course of optic neu­ropathy and visual loss. Neither patient, however, dem­onstrated evidence of more widespread demyelination. Despite its uncommon occurrence, clinicians should be J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 186 L. J. BALCER AND S. L. GALETTA aware of the possible association of optic neuropathy and chronic acquired demyelinating polyneuropathies ( 48). Idiopathic Intracranial Hypertension ( and Its Imposters) In adults, IIH predominantly affects young obese women. Obesity, a known risk factor for IIH, as demon­strated by case- control studies, may be present in ^ 90% of adult women and > 66% of adult men with IIH. Case series, reviews, and clinical experience have indicated that children with IIH are less likely to be obese, and that obesity may be even less common in younger children. The relation between age and obesity in children with IIH was recently examined by Balcer et al. ( 49). They reviewed the records for all consecutive pediatric pa­tients aged 17 years or younger with IIH seen at two medical centers between 1993 and 1997. Patients were categorized as obese if their weight at the time of diag­nosis of IIH was > 120% of their ideal body weight for height and sex. Forty patients were identified for whom both height and weight data were available; 5 patients without such data were excluded from the obesity analy­sis. Median age at the time of initial evaluation was 14 years ( range, 3- 17); 31 of 40 patients ( 78%) were girls. Eighty percent had headache, and 14 patients ( 35%) had sixth nerve palsies at presentation. Visual acuities were 20/ 20 or better in both eyes of 22 patients ( 50%), while 8 patients ( 20%) had visual acuities of 20/ 50 or worse acuity in one or both eyes. Papilledema was present bi­laterally in all patients. Older children were more likely to be obese; the proportions of obese patients across the three age tertiles were 6 of 14 patients ( 43%) in the 3 to 11 year age group, 13 of 16 patients ( 81%) in the 12 to 14 year age group, and 9 to 10 patients ( 90%) among patients aged 15 to 17 years. The trend of increasing proportions of obese patients across age groups was sta­tistically significant ( P = 0.01, chi- square test for trend). A significant association between increasing age and in­creasing probability of obesity was demonstrated by lo­gistic regression analysis ( P = 0.003). Further statistical analysis demonstrated that the greater likelihood of obe­sity among older children did not appear to depend solely on the higher proportions of female patients in the older age groups. The results of this study ( 49) provided strong evidence that, within the pediatric age group, younger patients with IIH are less likely to be obese than are older children and adults. However, larger collaborative stud­ies will be needed to examine the potential roles of meta­bolic, hormonal, or other physiologic factors in deter­mining the relation between age and obesity in pediatric IIH. Potential alternative mechanisms for the develop­ment of IIH in young nonobese children also remain to be investigated. Cinciripini et al. ( 50) also reported the clinical features of IIH in pediatric patients. They reviewed the records of children aged 11 years and younger who were diagnosed with IIH at two medical centers over a 3- year period. Ten patients were identified; four were girls ( 40%). Diplopia ( four patients), stiff neck ( four patients), and strabismus ( eight patients, with sixth nerve palsy in six patients) were the most common presenting signs and symptoms. Visual field defects were noted in 11 of 13 eyes ( 85%). Resolution of papilledema and sixth nerve palsy were generally rapid, occurring over a mean period of 4.7 months and 1.6 months, respectively. Interestingly, only one patient in this group of young pediatric patients (< 11 years; mean age, 7.3) was obese; this patient was an 11- year- old girl. This study emphasizes the relatively uncommon occurrence of obesity among children with IIH in the youngest age groups, as well as the frequent occurrence of diplopia and sixth nerve palsy as present­ing features of IIH in this patient population. Fourth nerve palsy may also occur as a nonspecific sign of increased intracranial pressure in children with IIH. Speer et al. ( 51) described the cases of three pa­tients, aged 8 to 15 years ( 1 boy, 2 girls), who had fourth nerve palsies as a presenting manifestation of IIH. All three children had accompanying symptoms and signs of increased intracranial pressure, including headache, tran­sient visual obscurations, visual field defects, and papil­ledema. In addition, all had binocular vertical diplopia with clinical signs characteristic of fourth nerve palsy, including a hypertropia of the affected eye that increased with contralateral gaze and ipsilateral head tilt. Exami­nation of the oldest patient ( 15 years old) confirmed the presence of excyclotorsion of the affected eye by double Maddox rod testing. Because the fourth nerve palsies resolved within 1 to 2 days in all three children, this sign is likely to represent a nonspecific sign of raised intra­cranial pressure in the setting of IIH ( 51). Because signs and symptoms of hypervitaminosis A mimic those of IIH, vitamin A is an attractive candidate for a mediator of IIH. Jacobson et al. ( 52) tested the hypothesis that serum vitamin A levels are elevated in patients with IIH compared with control subjects. Serum retinol levels and serum retinyl ester concentrations in 16 women with IIH and 70 healthy controls ( also women) were measured. Average daily vitamin A ingestion was also estimated using a survey instrument. Patients with IIH ranged in age from 18 to 42 years ( median age, 28); controls were aged 17 to 53 years ( median age, 36). Median serum retinol levels were significantly higher among the patient group ( 751 | xg/ L for patients vs. 530 p. g/ L for controls; P < 0.001, Wilcoxon rank sum test). This was true even after adjusting for age and body mass index ( P < 0.001, analysis of covariance). Serum retinyl ester concentrations were not significantly different be­tween the two groups ( P = 0.32, Wilcoxon rank sum test). There were also no significant differences in the calculated amounts of vitamin A ingested ( based on the questionnaire) between IIH patients and controls, al­though as the authors pointed out, small numbers may have limited the statistical power for this comparison. The authors concluded that elevated serum retinol con­centration is associated with IIH, independent of the po­tential influence of obesity. Possible mechanisms to ex­plain this difference may include differences in amounts of vitamin A ingested, differences in metabolism, or in- J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 PREGENICULATE REVIEW 1999 187 creased sensitivity to the effects of vitamin A in patients with IIH. Patients with cerebral venous thrombosis may present with isolated intracranial hypertension and a clinical pic­ture that mimics IIH; this was emphasized in a recent report by Biousse et al. ( 53). They reviewed the records of 160 consecutive patients with cerebral venous throm­bosis who presented between 1975 and 1998. Those with isolated intracranial hypertension were distinguished from those who had other neurologic signs and symp­toms at presentation. Isolated intracranial hypertension was present in 59 patients ( 37%); among these patients, 93% had headache ( including 8 without papilledema) and 86% had papilledema ( including 4 without head­ache). Nine patients ( 15%) had unilateral ( 7 patients) or bilateral ( 2 patients) sixth nerve palsies. Involvement of more than one venous sinus was seen by neuroimaging in 35 of 59 patients ( 59%). Cortical vein thrombosis was present in only 2 of 59 patients with isolated intracranial hypertension ( 3.5%), but was present in 38 of 101 pa­tients ( 38%) with other neurologic signs and symptoms at presentation ( P < 0.001, X2 test). Results of computed tomography ( CT) scans with contrast ( performed in 85% of patients) were more frequently normal among patients with isolated intracranial hypertension ( 54% vs. 16%; P < 0.001, X2 test), as was CSF content ( 75% vs. 41%; P < 0.001, X2 test). The " empty triangle sign" was present in 23% to 24% of each group. Complete recovery was seen in 93% ( 55 of 59 patients) of those with isolated intracranial hypertension, while only 67% ( 68 of 101 patients) of those who presented with other neurologic signs and symptoms experienced complete recovery ( P < 0.001, X2 test). Permanent neurologic sequelae remained in 26% of patients who presented with neurologic symp­toms, and death occurred in 6%. In addition to the fact that patients with central venous thrombosis may present with a clinical picture that mimics IIH ( including normal CT scan results and CSF content), the authors add that MRI with MR venography should be performed in pa­tients with apparent isolated intracranial hypertension. This recommendation is particularly important given the fact that the prognosis and outcome for central venous thrombosis may be variable, and early recognition is therefore of great importance. Another entity which may cause signs and symptoms that mimic IIH is spinal tumor. Haslbeck et al. ( 54) re­ported the case of a patient with a cauda equina paragan­glioma who presented with signs and symptoms of in­tracranial hypertension. This 45- year- old woman devel­oped acute diplopia. She had a 5- year history of frontal headache and had also experienced sciatic pain. Exami­nation revealed bilateral optic disc swelling and a right sixth nerve palsy. Results of MRI and CT scans of the brain were normal. Lumbar puncture demonstrated 35,000 red cells/ mm3 with a protein concentration of 4,500 mg/ dL; the opening pressure was 33 cm water. Cerebral angiography did not reveal any source for in­tracranial hemorrhage. Repeat lumbar puncture 1 month later revealed a profile identical to that of the first CSF analysis, with continued elevation of the opening pres­sure. Spine MRI showed an enhancing mass extending from L3 to the filum terminale. The tumor was resected, and pathology revealed grade I paraganglioma. These tumors are benign and are of neuroectodermal origin; when present in the nervous system, they- are located commonly in the lumbar region. Cerebrospinal fluid mal­absorption caused by elevated protein content was im­plicated as the most likely cause of intracranial hyper­tension in this case. Spinal tumors should be included in the differential diagnosis for causes of bilateral disc swelling and intracranial hypertension, particularly in the setting of abnormal CSF with elevated protein or evi­dence of hemorrhage. Leber Hereditary Optic Neuropathy Patients with Leber hereditary optic neuropathy ( LHON) usually have normal orbital MRI findings. Va-phiades and Newman ( 55) recently described the case of a patient with bilateral visual loss associated with the LHON 3460 mitochondrial DNA mutation; orbital MR imaging in this patient demonstrated bilateral optic nerve enhancement. An 18- year- old man developed painless progressive visual loss OD, followed by a similar pattern of visual loss OS 2 weeks later. Visual acuities on ex­amination were count fingers OD and 20/ 80 OS. Color vision was also abnormal ( 0/ 14 Ishihara plates correct OD, 7.5/ 14 OS), and large central scotomas were present OU on Goldmann perimetry. Thorough evaluation for potential causes of inflammatory and infectious optic neuropathies was unrevealing. Mitochondrial DNA test­ing demonstrated a mutation at nucleotide position 3460, consistent with the diagnosis of LHON. The MRI find­ings in this patient with LHON are consistent with the fact that the optic nerve pathology in LHON is not en­tirely limited to the prelaminar portion of the optic nerve ( 55). Although this finding has not been previously re­ported, the presence of optic nerve enhancement on MRI in patients with otherwise unexplained optic neuropa­thies should prompt consideration of LHON in addition to infectious, inflammatory, and neoplastic etiologies. Leber hereditary optic neuropathy has also been infre­quently associated with a MS- like illness. Most of these patients have been women, and the LHON mutations that have been associated with the MS- like illness have in­cluded the 11778 mutation, and, less frequently, the 3460 mutation. An unusual case of a 28- year- old man with LHON and an MS- like illness was reported by Bhatti and Newman ( 56); this patient harbored the 14484 mutation. There was no known family history of visual loss or neurologic disease. The patient developed bilateral visual loss that progressed relentlessly for several years. His course was also punctuated by the development of mul­tiple episodes of neurologic signs and symptoms. Results of MRI scans showed possible enhancement of the left optic nerve, as well as multiple white matter lesions in the brain and upper cervical spinal cord. Spinal fluid analysis during the initial acute episode of neurologic symptoms revealed a mildly elevated protein ( 47 mg/ dL) and a lymphocytic pleocytosis ( 20 cells/ mm3); a second J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 188 L. J. BALCER AND S. L. GALETTA lumbar puncture also showed elevated protein and an elevated immunoglobulin G concentration. Despite his visual loss, VER PI00 latencies were normal. Mitochon­drial DNA analysis was negative for point mutations at positions 11778 and 3460, but was positive for the 14484 mutation. This young man therefore demonstrated both clinical and paraclinical evidence for demyelinating dis­ease, as well as DNA evidence for LHON. Although, as the authors ( 55) point out, the predisposition to the two diseases may be coincidental, the presence of the mito­chondrial DNA mutation associated with LHON may influence the clinical course of visual loss in MS, and vice- versa. The mutations may render the optic nerve more susceptible to damage by demyelination; alterna­tively, demyelination itself in the setting of MS and optic neuritis may influence the threshold for expression of the LHON mutation ( 56). Optic Disc Drusen It has long been taught that the examination of family members may be useful in strengthening suspicion for the diagnosis of anomalous optic discs or disc drusen. Antcliff and Spalton ( 57) recently conducted an analysis of the families of several patients with optic disc drusen to examine the prevalence of disc drusen and related optic disc anomalies among family members. The au­thors ( 57) identified patients with optic disc drusen from a photographic database; these patients and their family members were invited to participate in the study. Each participant underwent a full clinical examination, includ­ing Snellen visual acuity, dilated ophthalmoscopy, B-scan ultrasonography, and optic disc photography. Optic discs were analyzed for the presence of drusen, the pres­ence or absence of an optic cup, and the presence of anomalous vessels ( defined as trifurcation of the arteri­oles within or adjacent to the optic disc or the presence of cilioretinal vessels). Seven patients ( all with bilateral disc drusen) and 27 family members were examined. Among the relatives of patients with optic disc drusen, only one ( 3.7%) had evidence of disc drusen ( buried, but confirmed by ultrasound). Other optic disc anomalies, however, were common among family members, with 30 of 53 eyes ( 57%) demonstrating anomalous vessels and 26 of 53 eyes ( 49%) having no optic cup. While the authors found the low prevalence of optic disc drusen among relatives of affected patients somewhat surpris­ing, they suggest that what may actually be inherited is a dysplasia of the optic disc and its blood supply; this may, in turn, predispose the formation of optic disc drusen in some patients. Toxic Optic Neuropathies Both NAION and amiodarone- induced optic neurop­athy occur in patients with cardiac disease and other vascular risk factors. To better define the distinction be­tween amiodarone- induced optic neuropathy and NAION, Macaluso et al. ( 58) reported the clinical char­acteristics of 73 patients with amiodarone- induced optic neuropathy and contrasted these with typical features of NAION. Based on their observations of published case reports ( 16 patients) and reports from agencies, such as the National Registry of Drug Induced Ocular Side Ef­fects, the authors ( 58) found certain features to be helpful in distinguishing amiodarone- induced optic neuropathy. In contrast to typical NAION, amiodarone- induced optic neuropathy is characterized by insidious onset, slow pro­gression of visual loss, bilateral involvement, and pro­tracted optic disc swelling that may persist for months with resolution after the discontinuation of medication. Although these features are helpful in distinguishing the two entities, the occurrence of visual loss or optic neu­ropathy in a patient taking amiodarone should prompt immediate examination. Close surveillance is recom­mended for patients taking amiodarone, with examina­tions before the start of medication and at intervals of 6 months thereafter ( 58). Traumatic Optic Neuropathy The treatment of traumatic optic neuropathy is a topic of debate, and the optimal management of this disorder is not well defined. High- dose corticosteroids ( in doses ranging from 60 mg to 7 g/ d), surgical decompression, and observation without treatment have all been advo­cated in the literature. A recent study by Levin et al. ( 59) compared visual acuity outcomes among patients in three nonrandomized treatment groups, including 1) no treat­ment ( n = 9), 2) corticosteroid treatment ( n = 85), and 3) optic canal decompression surgery ( n = 33). This was a nonrandomized interventional study, and treatments ( as well as corticosteroid doses) were determined based on the clinical judgment of the individual participating phy­sicians. Data analysis was planned only for those patients who had vision assessment within 3 days of injury and who received intervention within 7 days. Final visual outcome analysis was based upon the group of patients that completed 1- month follow- up assessments ( n = 104). Improvement of visual acuity by ^ 3 lines was observed in 57% of the untreated group, 52% of the corticosteroid group, and 32% of the surgery group; these proportions of patients with improvement were sta­tistically similar ( P = 0.22). Even after adjustment for baseline visual acuity, no significant differences in visual acuity were noted between the surgery group and the untreated ( P = 0.86) and steroid groups ( P - 0.25). No clear benefit was found for either corticosteroid therapy or optic canal decompression in this study. The potential limitations of the study design, however, including se­lection bias, were extensively discussed by the authors. THE OPTIC CHIASM AND BEYOND Visual Evoked Potential Evidence of Chiasmal Hypoplasia Thompson et al. ( 60) reported the cases of five chil­dren with varying degrees of chiasmal hypoplasia and aplasia, all of whom had a characteristic pattern of crossed asymmetry of the monocular flash visual evoked potential ( VEP) ( Fig. 2). These patients presented be­tween 2 and 7 months of age with pendular or see- saw nystagmus. Optic disc appearance ranged from normal ( one patient) to hypoplastic ( three patients); one patient J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 PREGENICULATE REVIEW 1999 v 189 left eye stimulation FIG. 2. Schema illustrating the visual pathway fiber distribution in achiasmia and albinism and its relationship to occipital asymme­try of the monocular visual- evoked potential. An upward deflec­tion represents a positivity. When the left eye is stimulated in achiasmia, the left hemisphere is more active than the right, and a negative peak N is recorded on the left occipital electrode ( l. occ; i. e., ipsilateral to the stimulated eye). In albinism, when too many fibers cross at the chiasm, stimulating the left eye elicits greater activity in the right hemisphere, and typically a negative peak on the right occipital electrode is seen contralateral to the stimulated eye. In both cases, the occipital asymmetry reverses when the right eye is stimulated, a " crossed asymmetry." In the normal condition with both hemispheres being similarly activated, the supposed dipole sheets summate to produce the largest re­sponse ( a positive peak) over the midoccipital electrode ( m. occ). Reprinted with permission from: Thompson DA, Kriss A, Chong K, et al. Visual- evoked potential evidence of chiasmal hypoplasia. Ophthalmology 1999; 106; 2354- 61. had bilateral colobomatous " morning glory" discs. Re­sults of MRI scans revealed chiasmal hypoplasia in three patients ( one of whom had only a thin strand of tissue connecting the optic nerves posteriorly) and absence of the chiasm ( aplasia) in two patients. Other midline ab­normalities were present in four patients, including de­ficiency of the anterior falx, septo- optic dysplasia, cleft lip and palate, holoprosencephaly, nasosphenoidal en-cephalocele, and frontonasal dysplasia. Developmental delay, ranging from moderate to severe, was also present in these children. Results of mixed rod- cone ERGs were normal in all patients. However, a common finding among this group of children with chiasmal hypoplasia and aplasia was a crossed asymmetry of the monocular flash VEP occipital distribution ( Fig. 2). This finding, v/ hich is consistent with a paucity or absence of crossing fibers at the optic chiasm, is marked by activation of the occipital electrode ipsilateral to the stimulated eye ( posi­tive peak of the VEP at ipsilateral occipital electrode); this asymmetry reverses (" crossed asymmetry") when the contralateral eye is stimulated. In patients with nor­mal degrees of crossing of fibers at the chiasm, the posi­tive peak of the VEP is present at the midline occipital electrode. Monocular flash VEP is therefore a useful test in infants with nystagmus; findings of crossed asymme­try should prompt MR imaging, even in children who have no other evidence of midline defects or optic disc anomalies. Pituitary Adenoma " The perils of a sneeze," particularly in the setting of a pituitary macroadenoma, were described in a report by Wein and Gans ( 61). A 51- year- old woman developed severe left supraorbital pain and visual loss OS after a sneeze. Examination revealed visual acuities of 20/ 20 OD and 20/ 100 OS, with a left afferent pupillary defect. Superior bitemporal defects and a small paracentral sc­otoma OS were noted on Goldmann perimetry. Results of MRI demonstrated a large intra-, supra-, and parasel-lar mass with superior displacement of the optic chiasm and third ventricle. More dramatic, however, was the presence of a collection of air ( pneumocephalus) be­tween the lateral aspect of the mass and the medial por­tion of the left temporal lobe. Erosion of the tumor through the sphenoid sinus roof, as demonstrated on CT scan, was thought to be responsible for the entry of air into the brain, thereby demonstrating " the perils of a sneeze." The authors propose a ball- valve and " inverted soda pop bottle" mechanism for the development of pneumocephalus in this patient ( therefore, air was able to enter, but could not escape after the sneeze). This case represents a highly unusual presentation of a pituitary adenoma ( null cell type) with spontaneous pneumatoc­ele. Hereditary Chiasmal Optic Neuropathy Pituitary adenoma and other compressive lesions of the optic chiasm are the most common cause of bitem­poral visual field defects. Pomeranz and Lessell ( 62) de­scribed three siblings with nearly identical bitemporal visual field defects and optic atrophy. In all three sib­lings, static perimetry revealed bilateral temporal visual field defects involving predominantly the superior fields. The authors ( 62) suggested a possible localization to the optic chiasm in these patients given the tendency for the field defects to respect the vertical meridian. Results of MRI scans of the brain and sella were normal in all cases; blood test results for known mitochondrial mutations as­sociated with Leber hereditary optic neuropathy were negative in one sibling. Although there was no known family history of optic atrophy or visual loss, the pres­ence of the disc pallor and visual field defects among all three siblings of a single family suggests the possibilities of autosomal dominant inheritance of a new mutation, mitochondrial disease, or autosomal recessive inheri­tance with occurrence in 3 of 3 siblings ( a 1 in 48 chance). As the authors suggest, bitemporal visual field defects have been identified among patients with heredi­tary optic neuropathies, including dominant optic atro­phy. Langerhans Cell Histiocytosis Job et al. ( 63) described a rare case of Langerhans cell histiocytosis with chiasmal and optic nerve involvement but without detectable systemic involvement. Langer­hans cell histiocytosis is a disorder of the mononuclear phagocytic system characterized pathologically by the accumulation of Langerhans cells in the skin, bones, lungs, lymph nodes, spleen, and liver. The patient was a J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 190 L. J. BALCER AND S. L. GALETTA 42- year- old woman with a history of hypersomnolence, panhypopitutarism, and visual loss. Examination re­vealed visual acuities of 20/ 100 ( searching) OD and 20/ 40 OS. Color vision was also impaired, and Humphrey visual field testing demonstrated bitemporal hemianopic scotomas with extension inferiorly. An MRI scan showed an enhancing lesion in the hypothalamus and optic chiasm. Enhancing brain parenchymal lesions, un­changed from a previous MRI performed 5 years earlier, were also noted. Biopsy of a left frontal brain lesion demonstrated a spindle and epithelioid histiocytic neo­plasm with inflammation and reactive astrocytes. Immu-nohistologic stains for S- 100, CD- 68, and CD- la pro­teins confirmed the diagnosis of Langerhans cell histio­cytosis. Further studies for systemic involvement had negative results. The patient underwent treatment with corticosteroids and radiation therapy. Visual acuities at follow- up were 20/ 80 OD and 20/ 60 OS. Although in­volvement of the central nervous system by Langerhans cell histiocytosis is well documented, such involvement of the central nervous system and visual pathways in isolation is rare. This entity should be included in the differential diagnosis for infiltrative lesions of the ante­rior visual pathways; however, more common disorders, such as sarcoidosis, must also be excluded in this setting. REFERENCES 1. Khoury JM, Donahue SP, Lavin PJM, et al. Comparison of 24- 2 and 30- 2 perimetry in glaucomatous and nonglaucomatous optic neuropathies. J Neuroophthalmol 1999; 19: 100- 8. 2. Lam BL, Thompson HS. An anisocoria produces a small relative afferent pupillary defect in the eye with the smaller pupil. J Neu­roophthalmol 1999; 19: 153- 9. 3. Diaper CJM, Dutton GN, Heron G. The Pulfrich phenomenon: its symptoms and their management. J Neuroophthalmol 1999; 19: 12. 4. Daneshvar H, Racette L, Coupland SG, et al. Symptomatic and asymptomatic visual loss in patients taking vigabatrin. Ophthal­mology 1999; 106: 1792- 8. 5. Kalviainen R, Nousiainen I, Mantyjarvi M, et al. Vigabatrin, a gabaergic antiepileptic drug, causes concentric visual field defects. Neurology 1999; 53: 922- 6. 6. Miller NR, Johnson MA, Paul SR, et al. Visual dysfunction in patients receiving vigabatrin: Clinical and electrophysiologic find­ings. Neurology 1999; 53: 2082- 7. 7. Arndt CF, Derambure P, Defoort- Dhellemmes S, et al. Outer reti­nal dysfunction in patients treated with vigabatrin. Neurology 1999; 52: 1201- 5. 8. Vanhatalo S, Paakkonen L. Visual field constriction in children treated with vigabatrin. Neurology 1999; 52: 1713- 4. 9. Gittinger JW, Smith TW. Cutaneous melanoma- associated para­neoplastic retinopathy: histopathologic observations. Am J Oph­thalmol 1999; 127: 612- 4. 10. Keltner JL, Thirkill CE. The 22- kDa antigen in optic nerve and retinal diseases. J Neuroophthalmol 1999; 19: 71- 83. 11. Solley WA, Martin DF, Newman NJ, et al. Cat scratch disease: Posterior segment manifestations. Ophthalmology 1999; 106: 1546- 53. 12. Gray AV, Reed JB, Wendel RT, et al. Bartonella henselae infec­tion associated with peripapillary angioma, branch retinal artery occlusion, and severe vision loss. Am J Ophthalmol 1999; 127: 223- 4. 13. Thompson PK, Vaphiades MS, Saccente M. Cat- scratch disease presenting as neuroretinitis and peripheral facial palsy. J Neu­roophthalmol 1999; 19: 240- 1. 14. Stechschulte SU, Kim RY, Cunningham ET. Tuberculous neurore­tinitis. J Neuroophthalmol 1999; 19: 201- 4. 15. Kerty E. The ophthalmology of internal carotid artery dissection Acta Ophthalmol Scand 1999; 77: 418- 21. 16. Lee SK, Kwon SU, Hyosook A, et al. Acute isolated monocular blindness and painless carotid artery dissection. Neurology 1999- 53: 1155- 6. 17. Jonas JB, Hayreh SS. Optic disk morphology in experimental cen­tral retinal artery occlusion in Rhesus monkeys. Am J Ophthalmol 1999; 127: 523- 30. 18. Rumelt S, Dorenboim Y, Rehany U. Aggressive systematic treat­ment for central retinal artery occlusion. Am J Ophthalmol 1999- 128: 733- 8. 19. Richard G, Lerche RC, Knospe V, et al. Treatment of retinal ar­terial occlusion with local fibrinolysis using recombinant tissue plasminogen activator. Ophthalmology 1999; 106: 768- 73. 20. Hayreh SS. Retinal arterial occlusion with LIF using rTPA ( letter). Ophthalmology 1999; 106: 1236- 8. 21. McKibbin M, Verma D. Recurrent amaurosis fugax without hae-modynamically significant ipsilateral carotid stenosis. Acta Oph­thalmol Scand 1999; 77: 224- 6. 22. Jacobson DM. Symptomatic compression of the optic nerve by the carotid artery: clinical profile of 18 patients with 24 affected eyes identified by magnetic resonance imaging. Ophthalmology 1999; 106: 1994- 2004. 23. Petrig BL, Riva CE, Hayreh SS. Laser Doppler flowimetry and optic nerve head blood flow. Am J Ophthalmol 1999; 127: 413- 25. 24. Tamaki Y, Araie M, Nagahara M, et al. Acute effects of cigarette smoking on tissue circulation in human optic nerve head and cho-roid- retina. Ophthalmology 1999; 106: 564- 9. 25. Salomon O, Huna- Baron R, Jurtz S, et al. Analysis of prothrom-botic and vascular risk factors in patients with nonarteritic anterior ischemic optic neuropathy. Ophthalmology 1999; 106: 739^ 2. 26. Feldon SE. Anterior ischemic optic neuropathy: trouble waiting to happen. Ophthalmology 1999; 106: 651- 2. 27. Lessell S. Nonarteritic anterior ischemic optic neuropathy: enigma variations. Arch Ophthalmol 1999; 117: 386- 8. 28. Fivgas GD, Newman NJ. Anterior ischemic optic neuropathy fol­lowing the use of a nasal decongestant. Am J Ophthalmol 1999; 127: 104- 6. 29. Jackson TL, Farmer CKT, Kingswood C, et al. Hypotensive isch­emic optic neuropathy and peritoneal dialysis. Am J Ophthalmol 1999; 128: 109- 11. 30. Hayreh SS, Podhajsky P, Zimmerman MB. Ocular and optic nerve head ischemic disorders and hearing loss. Am J Ophthalmol 1999; 128: 606- 11. 31. Wang MY, Sadun F, Levin LB, et al. Occurrence of familial non­arteritic anterior ischemic optic neuropathy in a case series. J Neu­roophthalmol 1999; 19: 144- 7. 32. Salomon O, Huna- Baron R, Steinberg DM, et al. Role of aspirin in reducing the frequency of second eye involvement in patients with non- arteritic anterior ischemic optic neuropathy. Eye 1999; 13: 357- 9. 33. Alexandrakis G, Lam BL. Bilateral posterior ischemic optic neu­ropathy after spinal surgery. Am J Ophthalmol 1999; 127: 354- 5. 34. Good CD, Cassidy LM, Moseley IF, et al. Posterior optic nerve infarction after lower lid blepharoplasty. J Neuroophthalmol 1999; 19: 176- 9. 35. Lee AG, Eggenberger ER, Kaufman DI, et al. Optic nerve en­hancement on magnetic resonance imaging in arteritic ischemic optic neuropathy. J Neuroophthalmol 1999; 19: 235- 7. 36. Hwang J- M, Girkin CA, Perry JD, et al. Bilateral ocular ischemic syndrome secondary to giant cell arteritis progressing despite cor­ticosteroid treatment. Am J Ophthalmol 1999; 127: 102- 4. 37. Boyev LR, Miller NR, Green WR. Efficacy of unilateral versus bilateral temporal artery biopsies for the diagnosis of giant cell arteritis. Am J Ophthalmol 1999; 128: 211- 5. 38. Trobe JD, Sieving PC, Guire KE, et al. The impact of the Optic Neuritis Treatment Trial on the practices of ophthalmologists and neurologists. Ophthalmology 1999; 106: 2047- 53. 38A. Beck RW, Cleary PA, Anderson MM, et al. A randomized, con­trolled trial of corticosteroids in the treatment of acute optic neu­ritis. N Engl J Med 1992; 326: 581- 8. J Neuro- Ophthalmol, Vol. 20, No. 3, 2000 PREGENICULATE REVIEW 1999 } g\ 39. Keltner JL, Johnson CA, Spurr JO, et al., for the Optic Neuritis Study Group. Comparison of central and peripheral visual field properties in the Optic Neuritis Treatment Trial. Am J Ophthalmol 1999; 128: 543- 53. 40. Fang JP, Donahue SP, Lin RH. Global visual field involvement in acute unilateral optic neuritis. Am J Ophthalmol 1999,128: 554- 65. 41. Fang JP, Lin RH, Donahue SP. Recovery of visual field function in the Optic Neuritis Treatment Trial. Am J Ophthalmol 1999; 128: 566- 72. 42. Arnold AC. Visual field defects in the Optic Neuritis Treatment Trial: central vs. peripheral, focal vs. global. Am J Ophthalmol 1999; 128: 632- 4. 43. Wakakura M, Mashimo K, Oono S, et al. Multicenter clinical trial for evaluating methylprednisolone pulse treatment of idiopathic optic neuritis in Japan. Jpn J Ophthalmol 1999; 43: 133- 8. 44. Wakakura M, Minei- Higa R, Oono S, et al. Baseline features of idiopathic optic neuritis as determined by a multicenter treatment trial in Japan. Jpn J Ophthalmol 1999; 43: 127- 32. 45. Wingerchuk DM, Hogancamp WF, O'Brien PC, et al. The clinical course of neuromyelitis optica ( Devic's syndrome). Neurology 1999; 53: 1107- 14. 46. Brady KM, Brar AS, Lee AG, et al. Optic neuritis in children: clinical features and visual outcome. J AAPOS 1999; 3: 98- 103. 47. Biousse B, Trichet C, Bloch- Michel E, et al. Multiple sclerosis associated with uveitis in two large clinic- based series. Neurology 1999; 52: 179- 81. 48. Lee AG, Galetta SL, Lepore FE, et al. Optic atrophy and chronic acquired polyneuropathy. J Neuroophthalmol 1999; 19: 67- 9. 49. Balcer LJ, Liu GT, Forman S, et al. Idiopathic intracranial hyper­tension: relation of age and obesity in children. Neurology 1999; 52: 870- 2. 50. Cincirpini GS, Donahue S, Borchert MS. Idiopathic intracranial hypertension in prepubertal pediatric patients: characteristics, treatment, and outcome. Am J Ophthalmol 1999; 127: 178- 82. 51. Speer C, Pearlman J, Phillips PH, et al. Fourth cranial nerve palsy in pediatric patients with pseudotumor cerebri. Am J Ophthalmol 1999; 127: 236- 7. 52. Jacobson DM, Berg R, Wall M, et al. Serum vitamin A concen­tration is elevated in idiopathic intracranial hypertension. Neurol­ogy 1999; 53: 1114- 8. 53. Biousse V, Ameri A, Bousser M- G. Isolated intracranial hyperten­sion as the only sign of cerebral venous thrombosis. Neurology 1999; 53: 1537- 42. 54. Haslbeck KM, Eberhardt KEW, Nissen U, et al. Intracranial hy­pertension as a clinical manifestation of cauda equina paragan­glioma. Neurology 1999; 52: 1297- 8. 55. Vaphiades MS, Newman NJ. Optic nerve enhancement on orbital magnetic resonance imaging in Leber's hereditary optic neuropa­thy. J Neuroophthalmol 1999; 19: 238- 9. 56. Bhatti MT, Newman NJ. A multiple sclerosis- like illness in a man harboring the mtDNA 14484 mutation. J Neuroophthalmol 1999; 19: 28- 33. 57. Antcliff RJ, Spalton DJ. Are optic disc drusen inherited? Ophthal­mology 1999; 106: 1278- 81. 58. Macaluso DC, Shults WT, Fraunfelder FT. Features of amio-darone- induced optic neuropathy. Am J Ophthalmol 1999; 127: 610- 2. 59. Levin LA, Beck RW, Joseph MP, et a)., the International Optic Nerve Trauma Study Group. Ophthalmology 1999; 106: 1268- 77. 60. Thompson DA, Kriss A, Chong K, et al. Visual- evoked potential evidence of chiasmal hypoplasia. Ophthalmology 1999; 106: 2354- 61. 61. Wein FB, Gans MS. The perils of a sneeze. J Neuroophthalmol 1999; 19: 128- 30. 62. Pomeranz HD, Lessell S. A hereditary chiasmal optic neuropathy. Arch Ophthalmol 1999; 117: 128- 31. 63. Job OM, Schatz NJ, Glaser JS. Visual loss with Langerhans cell histiocytosis: Multifocal central nervous system involvement. J Neuroophthalmol 1999; 19: 49- 53. J Neuro- Ophthalmol, Vol. 20, No. 3, 2000