SURVEY OF OPHTHALMOLOGY VOLUME 44 • NUMBER 4 • JANUARY–FEBRUARY 2000 CLINICAL CHALLENGES PETER SAVINO, EDITOR An Unusual Homonymous Visual Field Defect Francine Wein, MD and Neil R. Miller, MD The Johns Hopkins Medical Institution, The Wilmer Eye Institute, Neuro-Ophthalmology Unit, Baltimore, Maryland, USA Comments by Michael S. Vaphiades, DO (In keeping with the format of a Clinical Pathologic Conference, the abstract and key words appear at the end of the article.) Case Report. A general ophthalmologist requested an opinion regarding a visual field defect found in a 75-year-old man who realized, while completing a form, that he could not see in the inferior right field of either eye. The patient’s past medical history was significant for thrombophlebitis, benign prostatic hypertrophy, osteoarthritis in the hip, and seasonal allergies. He had a remote smoking history and admitted to one to two alcoholic drinks per day. His medications included doxazocin, aspirin, and loratadine. Best-corrected visual acuity was 20/20 in both eyes, and color vision was 9/10 with the Hardy-RandRittler pseudoisochromatic plates. He had equal pupils that were normally reactive to light and near stimuli. The results of visual field testing with static perimetry (24-2 strategy with a Humphrey perimeter) are shown in Fig. 1. The optic nerves were normal on funduscopic examination. Comments Michael Vaphiades, DO, Department of Ophthalmology and Neurology, University of Arkansas for Medical Sciences, Little Rock, Arkansas, USA The clinical history and examination are consistent with a patient who has a lesion of the left retrochiasmal visual pathways. The lack of a relative afferent pupillary defect (RAPD) and normal-appearing optic nerves makes an optic tract lesion unlikely. The lateral geniculate nucleus (LGN) is next in the retrochiasmal visual pathway. A major difference between an optic tract and an LGN lesion is the lack of an RAPD in LGN lesions. This is because of the pupillomotor fibers exiting from the optic tract before reaching the LGN.14 In addition, the patient’s visual field abnormality is a relatively incongruous right homonymous wedge-shaped defect consistent with a homonymous horizontal sectoranopia (HHS). This type of defect is classically seen with LGN lesions and may be either congruous or incongruous.15 This Based on the appearance of the visual fields, where do you think the lesion is located? What investigations, if any, would you recommend? 324 © 2000 by Elsevier Science Inc. All rights reserved. 0039-6257/00/$–see front matter PII S0039-6257(99)00116-2 325 HOMONYMOUS VISUAL FIELD DEFECT Fig. 1. Visual field defects as detected by static perimetry (Humphrey perimetry, 24-2 strategy) in the left (left) and right (right) eyes. would also support the LGN as the location of the lesion.3,9,17 If this case represents an LGN lesion, it would have to be small, given the proximity of the ipsilateral thalamus (sensory fibers) and pyramidal tract (motor fibers) and the patient’s lack of a contralateral hemianesthesia or hemiparesis. Such a lesion may occur as a result of ischemia to the posterior lateral choroidal artery, which perfuses the central portion of the LGN, resulting in an HHS.9 These lesions are rare. Also, LGN lesions are often associated with optic nerve pallor. The pattern of optic nerve pallor is similar to that of an optic tract lesion with generalized pallor of the ipsilateral optic nerve and band- or wedge-shaped pallor of the optic nerve contralateral to the lesion.14 However, optic nerve pallor is not present in the acute stages. The next possible location is the optic radiations. This again would have to be a small, strategically placed lesion to spare the other parietal or temporal lobe structures. The last and most likely location of the lesion is the occipital lobe. Lesions in this location are not typically associated with other neurologic deficits. Patients with homonymous hemianopic visual field defects who are otherwise neurologically intact generally harbor occipital lobe lesions. What does not favor an occipital lobe lesion is the apparent incongruity of the HHS. A complete lesion of visual pathway nervous tissue causes complete homonymous hemianopic visual field defects. If there is partial interference of fibers such that less than 120⬚ of visual field is affected, then the more posterior the lesion, the more congruous the hemianopia. The caveat is that the automated visual field only tests the central 24⬚ of visual field. Finally, the cause of the lesion should be addressed. The lesions causing HHS include ischemia, vascular malformations, trauma, and astrocytomas.8 Ischemic infarctions are more common in the occipital lobe than other lesions.14 Since the present patient most likely has an occipital lobe lesion, he probably has had an ischemic infarction in the posterior cerebral artery territory of the left occipital lobe. The patient should undergo gadoliniumenhanced magnetic resonance imaging (MRI) of the brain with thin cuts through the posterior visual pathways. If during the study an infarction is noted, he should also have magnetic resonance angiography of the posterior circulation. Before the patient leaves for the MRI, he should have his vital signs checked, cardiac auscultation performed, and receive laboratory testing to help screen for vasculopathic risk factors. Case Report (Continued) The initial interpretation of the perimetric findings was that they showed a right homonymous horizontal sectoranopia (HHS) consistent with a lesion 326 Surv Ophthalmol 44 (4) January–February 2000 WEIN AND MILLER How do you explain the disparity between the static and kinetic perimetry? Comments (Continued) Fig. 2. MRI, T2-weighted axial view, shows changes consistent with infarct in mesial aspect of the left occipital lobe (arrow). of the left lateral geniculate nucleus (LGN). Accordingly, an MRI of the brain with special attention to the region of the LGN was obtained. The brain MRI revealed a hyperintense signal, compatible with an infarct, in the left mesial occipital lobe. No lesion was seen in the left LGN (Fig. 2). How do you explain the visual fields? What would you do next? Comments (Continued) The visual field defect is an apparent HHS caused by an infarction of the left mesial occipital lobe. An HHS is a rare visual field defect, and perhaps more information regarding the visual fields would be gained if a larger portion of the visual fields were tested. Therefore, I would perform Goldmann kinetic perimetry. Case Report (Continued) Kinetic perimetry was performed, which showed that the apparent HHS was actually an homonymous hemianopia with incomplete sparing of the temporal crescent (Fig. 3). This patient had a left mesial occipital lobe infarction that spared the most anterior portion of the visual (striate) cortex. This anterior region of the mesial occipital lobe corresponds to an unpaired portion of the contralateral temporal visual field in the form of a temporal crescent. This results from the unique arrangement of the visual fibers and their relationship to the visual fields. The temporal retina sees images in the nasal visual field and the nasal retina sees images from the temporal visual field. The temporal visual field of each eye is larger than its corresponding homonymous nasal field because of unpaired nasal fibers. The temporal crescent becomes important in the localization of the central nervous system lesions producing visual field defects, as in this case. Most of our detailed mapping of the retinoptic organization of the striate cortex came from the study of injured soldiers during World War I, with revision of this map in 1991 by Horton and Hoyt.12 They correlated visual field defects with magnetic resonance images of patients with occipital lobe lesions and determined that the temporal crescent is represented at the rostral end of the striate cortex.12 This area constitutes less than 10% of its total surface and is supplied by the parieto-occipital artery.4,12,16 The temporal crescent extends at its widest portion from the peripheral 60⬚ to 110⬚ at the horizontal meridian of the visual field.18 Lesions affecting this anterior portion of the striate cortex will cause a monocular contralateral temporal visual field defect in the shape of a temporal crescent. This is also called a half-moon syndrome and has value in localizing monocular visual field defects.4,18 Conversely, if the infarction spares this anterior 8% to 10% of the striate cortex, as in this patient, there will be a contralateral homonymous hemianopia with a spared monocular temporal crescent.12 The temporal crescent is not detected on conventional 30⬚ full-threshold static computerized automated perimetry.4,12,13 Although the central 30⬚ detects visual field defects caused by lesions of the central visual pathways in 83% of cases, it provides no information on the remaining 17%12,13 and may lead to an erroneous localization of the lesion. Kinetic visual field testing using Goldmann perimetry makes it possible to test the peripheral temporal visual field beyond the central 30⬚ up to the temporal 90⬚.1 The temporal crescent extends to 110⬚, but the Goldmann perimeter plots isopters temporally only to 90⬚.1,4,18 The basic premise of kinetic perimetry is to present the patient with kinetic and static stimuli HOMONYMOUS VISUAL FIELD DEFECT 327 Fig. 3. Visual field defects as detected by kinetic perimetry (Goldmann perimetry) in the left (left) and right (right) eyes. Note that the apparent homonymous horizontal sectoranopia is actually an incomplete homonymous hemianopia with sparing of most of the temporal crescent in the right eye. of various sizes and intensities to generate a visual field composed of isopters (line connecting all points of a given stimulus). The examiner usually starts the testing using the I4e stimulus (I ⫽ size, 2 ⫽ coarse intensity adjustment, e ⫽ fine intensity adjustment). Although others have reported that the I4e isopter extends up to 90⬚ temporally and 60⬚ nasally in each eye,1,5 in my experience it usually extends about 70⬚ temporally and 50⬚ nasally. The isopters may be larger in younger patients and more constricted in older patients, with a larger decrease in the temporal when compared with the nasal half of the visual field with age.5 If these isopters are abnormally constricted, a larger stimulus (V4e) is used to plot the remaining temporal 90⬚ and nasal 60⬚ of visual field.1,10 With the use of this method of visual field testing, a peripheral temporal crescent can be mapped and a more accurate localization of the lesion can be made. Authors’ Discussion A horizontal homonymous sectoranopia is an uncommon visual field defect first described by Holmes in 193111 and infrequently reported in the literature. This field defect is most commonly caused by damage to the central portion of the LGN, which is supplied by the lateral posterior choroidal artery.6,15 The undisturbed anterolateral and anteromedial parts of the LGN, supplied by the distal portion of the anterior choroidal artery, allow for preservation of vision in the superior and inferior portions of the same hemifield. Small vascular lesions tend to produce congruous HHS with steeply sloping borders, whereas infiltrative or compressive lesions generally produce incongruous defects.7,11,15 Although initially thought to be specific for LGN lesions,6,7,15 HHS is also said to occur with lesions of the retrogeniculate pathway. Most of the lesions affect the optic radiations,3,8,17 but three have been ascribed to damage of the occipital cortex.9 For example, Grossman et al described a left HHS in a patient with an ischemic infarction in the occipital cortex.9 The authors postulated that selective occlusion of a penetrating branch of the right calcarine artery in the area of the calcarine fissure was responsible for the visual field defect. One pathologically confirmed case of a retrogeniculate lesion producing an HHS was reported by Holmes, who described a patient with an intracranial hemorrhage from an arteriovenous malformation located between the calcarine cortex and the occipital horn of the lateral ventricle, sparing the cortex but destroying a narrow band of white matter carrying the middle portion of the optic radiations.11 The fibers above and below the hematoma destined for the superior and inferior banks of the calcarine cortex were noted to be intact. The neuroanatomic correlate of our patient’s visual field defect as identified by automated perimetry was initially thought to be the left LGN; however, kinetic perimetry, performed after an MRI showed only a left occipital lesion and no evidence of an LGN lesion. It revealed that the defect was actually a highly congruous, homonymous hemianopia with sparing of the inferior—and to a lesser extent, the 328 Surv Ophthalmol 44 (4) January–February 2000 WEIN AND MILLER superior—aspects of the right temporal crescent, a pattern specific for occipital disease. This defect was likely the result of calcarine artery occlusion. The homonymous field defect appeared incongruous—and more typical of an HHS—with automated perimetry, in large part because only the central 24⬚ was tested. Thus, sparing of the temporal crescent was not identified until kinetic perimetry was performed, which also showed the defect to be congruous. Interestingly, of the two reports in the literature (comprising five cases of HHS) that included both kinetic and static perimetry,2,8 the defects also appeared more congruous with kinetic perimetry than with static perimetry. Grochowicki and Vighetto postulated two reasons for this disparity: artificial enlargement of the central 30⬚ by automated perimetry and accentuation of the asymmetry in mapping of the blind spot.8 Indeed, in all five cases, the visual field defects were greater in the eye with the temporal field loss than in the eye with nasal field loss. In our patient, however, the defect was greater in the eye with the field loss (i.e., the left eye). This distinction, along with the lack of optic atrophy, suggests that the occipital lobe, rather than the LGN, was the site of the lesion. Our case emphasizes the potential for confusion between true HHS and incomplete homonymous hemianopias with sparing of the temporal crescent, particularly when only automated perimetry is performed. 4. Chavis PS, al-Hazmi A, Clunie D, Hoyt WF: Temporal crescent syndrome with magnetic resonance correlation. J Neuroophthalmol 17:151–5, 1997 5. Egge K: The visual field in normal subjects. Acta Ophthalmol Suppl 169:1–64, 1984 6. Frisén L: Quadruple sectoranopia and sectorial optic atrophy: a syndrome of the distal anterior choroidal artery. J Neurol Neurosurg Psychiatry 42:590–4, 1979 7. Frisén L, Holmegaard L, Rosencrantz M: Sectorial optic atrophy and homonymous, horizontal sectoranopia: a lateral choroidal artery syndrome?. J Neurol Neurosurg Psychiatry 41:374–80, 1978 8. Grochowicki M, Vighetto A: Homonymous horizontal sectoranopia: report of four cases. Br J Ophthalmol 75:624–8, 1991 9. Grossman M, Galetta SL, Nichols CW, Grossman RI: Horizontal homonymous sectoral field defect after ischemic infarction of the occipital cortex. Am J Ophthalmol 109: 234–6, 1990 10. Harrington DO, Drake MV: The Visual Fields, Text and Atlas of Clinical Perimetry. St. Louis, CV Mosby, 1990, ed 6, pp 13–70 11. Holmes G: A contribution to the cortical representation of vision. Brain 54:570–9, 1931 12. Horton JC, Hoyt WF: The representation of the visual field in human striate cortex: a revision of the classic Holmes map. Arch Ophthalmol 109:816–24, 1991 13. Landau K, Wichmann W, Valavanis A: The missing temporal crescent. Am J Ophthalmol 119:345–9, 1995 14. Miller NR, Newman NJ: Topical diagnosis of lesions in the visual sensory pathway, in Miller NR, Newman NJ (eds): Walsh & Hoyt’s Clinical Neuro-Ophthalmology, Vol 1. Baltimore, Williams and Wilkins, 1998, ed 5, pp 237–386 15. Shacklett DE, O’Connor PS, Dorwart RH, et al: Congruous and incongruous sectoral visual field defects with lesions of the lateral geniculate nucleus. Am J Ophthalmol 98:283–90, 1984 16. Smith CG, Richardson WF: The course and distribution of the arteries supplying the visual (striate) cortex. Am J Ophthalmol 61:1391–6, 1966 17. Smith RJ: Horizontal sector hemianopia of non-traumatic origin. Br J Ophthalmol 54:208–10, 1970 18. Walsh TJ: Temporal crescent or half-moon syndrome. Ann Ophthalmol 6:501–5, 1974 References 1. Anderson DR: Testing of the Field of Vision. St Louis, CV Mosby, 1982, pp 8–69 2. Borruat FX, Maeder P: Sectoranopia after head trauma: evidence of lateral geniculate body lesion on MRI. Neurology 45:590–2, 1995 3. Carter JE, O’Connor P, Shacklett D, Rosenberg M: Lesions of the optic radiations mimicking lateral geniculate nucleus visual field defects. J Neurol Neurosurg Psychiatry 48:982–8, 1985 Dr. Vaphiades’ research was supported in part by an unrestricted grant from Research to Prevent Blindness, Inc., New York, New York. Reprints are not available. Abstract. A 75-year-old man suddenly became aware of an inferior right homonymous visual field defect. Although static perimetry suggested a lesion of the left lateral geniculate nucleus, kinetic perimetry indicated that the presumed homonymous horizontal sectoranopia noted on static perimetry was actually an incomplete homonymous hemianopia with incomplete sparing of the temporal crescent. The location of the lesion was subsequently confirmed by magnetic resonance imaging. This case shows the value of kinetic perimetry in assessing homonymous visual field defects. (Surv Ophthalmol 44:324– 328, 2000. © 2000 by Elsevier Science Inc. All rights reserved.) Key words. homonymous hemianopia • homonymous horizontal sectoranopia • lateral geniculate nucleus • occipital lobe • temporal crescent