Rarely is a nonneoplastic disorder responsible for a bitemporal visual field defect, and in those cases, mechanical or radiation injury, inflammation, infection, demyelination, intoxication, infarction, or hypoplasia of the optic chiasm are the mechanisms. We have encountered a family in which 3 siblings had a unique, presumably hereditary neuropathy manifested by a chiasmatic visual field defect.
The 3 white patients whose histories are presented below represent all of the children of a nonconsanguineous marriage between a healthy mother of Puerto Rican descent and a healthy father of Irish descent.
A healthy 28-year-old woman was advised to have an ophthalmic examination because an abnormality in her sister's optic nerves had been discovered. The patient's only eye problem had been fully correctable myopia and astigmatism. Her neurological and endocrine histories showed no abnormalities and an ophthalmologist found that her visual acuity was 20/20 OU, but she had pallor of her optic nerves and a bitemporal visual field defect. A computed tomographic scan of the brain showed no abnormalities. She remained asymptomatic and was referred for evaluation at age 35 years.
Her best-corrected visual acuity was 20/15 OU. She missed several of the Ishihara test plates with each eye, but she made no errors when tested with the Farnsworth D-15 panel. There was a bilateral superotemporal quadrantanopsia (Figure 1). Her corneal nerves seemed unusually prominent and there were white dots scattered in the cortex of both lenses. Both optic discs were atrophic but of normal size. Findings from the remainder of her examination showed no abnormalities. Magnetic resonance imaging scan of the brain and orbits with gadolinium injection and full-field and foveal electroretinograms showed no abnormalities. Blood test results for 4 of the mitochondrial mutations known to be associated with Leber hereditary optic neuropathy (codons 11778, 14484, 15257, and 3468) were normal. Her condition remains unchanged 5 years later.
Static automated visual fields from patient 1 show bilateral superotemporal defects respecting the vertical and horizontal meridians.
The older brother of patient 1 was found to have abnormal optic nerves when he went to an optometrist at age 27 years to have his contact lenses checked. He was in good general health and his only eye problem had been myopia and astigmatism that was fully correctable with lenses. An ophthalmologist to whom he was referred found that his visual acuity was 20/20 OU with slight posterior subcapsular cataracts and pale optic discs. Automated perimetry showed defects near fixation. Findings from the neurological examination, 100-Hue color testing, magnetic resonance imaging scanning, and pattern visual evoked response testing showed no abnormalities. He was examined again at age 33 years by a neuro-ophthalmologist. Best-corrected visual acuity was 20/20 OU. On Ishihara color plate testing, he missed 2 with his right eye and 1 with his left. Slitlamp examination revealed mildly thickened corneal nerves and posterior subcapsular cataracts in both eyes. There was no relative afferent pupillary defect. Humphrey visual field testing showed central dropout with a superotemporal loss approaching the vertical meridian in each eye (Figure 2). Both optic nerves were pale but of normal size.
Static automated visual fields from patient 2 show bilateral superotemporal defects respecting the vertical and horizontal meridians.
The younger sister of patient 1 allegedly had poor vision since childhood and a "lazy eye" that was treated with exercises. Migraine headaches had been present since age 15 years. As early as age 27 years she was noted to have elevated prolactin levels but no endocrine symptoms. Ophthalmic examination then revealed a visual acuity of 20/25 OD and 20/20 OS, 9 of 14 Ishihara color plates correct with each eye; enlarged blind spots and loss of superotemporal visual field and bilateral optic atrophy on kinetic visual field were noted during testing. Computed tomographic scans showed no abnormalities at ages 27 and 30 years. At age 32 years, Humphrey visual field testing showed superotemporal defects respecting the vertical meridian. Magnetic resonance imaging revealed a small, hyperintense, intrasellar lesion in the left side of the pituitary gland without suprasellar extension, presumed to be a pituitary cyst. When examined by a neuro-ophthalmologist at age 37 years, her visual acuity was 20/25 OU. She correctly identified 2 of the 14 Ishihara color test plates with each eye. There were bitemporal visual field defects on static visual field testing and both optic discs were atrophic but of normal size. A second magnetic resonance imaging scan showed no abnormalities.
She was examined by a neuro-ophthalmologist 2 years later. Her best-corrected visual acuity was 20/25 OD and 20/20 OS. On color vision testing, she correctly identified 3.5 of the 14 plates with her right eye and 2.5 of the 14 plates with her left. Slitlamp examination showed no abnormalities. Humphrey visual fields revealed bilateral superotemporal field defects (Figure 3). The mean deviation was unchanged compared with the automated visual fields done at age 32 years. There was diffuse optic nerve pallor of both optic nerve heads with temporal excavation. The optic discs were normal sized.
Static automated visual fields from patient 3 show bilateral superotemporal defects respecting the vertical and horizontal meridians.
The father's family history is notable only for cataracts. The mother's family history is notable for siblings who died at ages 14 and 12 years of a neurological disease characterized initially by stuttering and loss of balance, progressing to bed confinement, and aphasia. Further details of their illnesses are unavailable. Both parents were free of systemic disease, had normal corrected visual acuities, Humphrey automated visual field results, color vision (Ishihara color plates), and optic nerve heads. Patient 1 has 2 daughters, a 12-year-old and a 9-year-old , both of whom are healthy. Both had neuro-ophthalmic examinations, testing visual acuity, color vision, and visual fields, as well as visualization of the optic nerves, all of which showed no abnormalities. The older daughter of patient 1 had strabismic amblyopia that was successfully treated with patching. Patient 2 has an 18-month-old son, who is healthy, and patient 3 has a 9-year-old son and a 2-year-old daughter, both of whom are healthy.
We suspect that the 3 siblings described in this article have a unique optic neuropathy that is manifested exclusively or predominantly at the level of the optic chiasm based on the presence of typical chiasmatic visual field defects, and is probably inherited as an autosomal recessive trait. While we recognize that centrocecal scotomas can be mistaken for chiasmatic bitemporal defects and that automated perimetry can give the false impression that a temporal defect comes to, but does not cross, the vertical midline, nevertheless we are confident that our patients' defects were chiasmatic. Centrocecal defects would not give a defect confined to the superior quadrant and would not be likely to spare Snellen visual acuities in every patient. The issue of the midline boundary can be resolved in 2 of the patients because testing on several occasions with Goldmann perimetry demonstrated that their visual field defects "respected" the vertical meridian. Two of the siblings had a pure median chiasmal syndrome when first evaluated and their visual acuity and color sense during many years of observation showed no abnormalities. Although the visual field defect in patient 3 is bitemporal, her slightly reduced visual acuity in 1 eye and her bilateral dyschromatopsia imply that she also has some involvement of uncrossed axons.
It is highly unlikely that the disorder was not inherited as a mendelian dominant trait since neither parent was affected. Only in the event that an affected man other than the putative father sired all 3 siblings could this conclusion be confounded. An illness involving an entire sibship but sparing the parents could be genetically or nongenetically inherited. Nongenetic, static, congenital optic neuropathies in multiple siblings can, albeit rarely, result from an abnormal uterine environment. Segmental optic nerve hypoplasia in the offspring of mothers with diabetes mellitus is one of the few examples of this phenomenon; however, nongenetic inheritance of this kind cannot be ruled out in our cases. Mutations in mitochondrial DNA can be transmitted through unaffected maternal carriers as in Leber hereditary optic neuropathy. Leber hereditary optic neuropathy was ruled out in patient 1 and, without special molecular investigations and extensive information about multiple generations of the family, this possibility cannot be evaluated. Autosomal recessive mendelian inheritance would be the mechanism most likely to explain the occurrence of the optic neuropathy in this family.
It is impossible to determine if the visual defects were congenital or acquired. In all 3 patients, the neuro-ophthalmic abnormalities were discovered incidentally when each patient was asymptomatic. Two of the patients had normal findings on routine examinations (without perimetry) many years before discovery of the optic disc pallor, but it is possible that the pallor was present but missed. None of the siblings has shown a decline in visual function under observation following discovery of the defect. In certain congenital optic neuropathies, the optic disc is hypoplastic. None of our patients had discernible hypoplasia, but subtle degrees of hypoplasia can be hard to recognize. If acquired postnatally, the latest the disorder could have begun was in the third decade of life.
Bitemporal visual field defects have been identified in hereditary optic neuropathies. Raaf and Bair1 described the findings in patients from a family with Leber hereditary optic neuropathy. Two siblings, both with impaired visual acuity, had visual field defects with bitemporal characteristics. Weiner et al2 reported a sporadic case of Leber hereditary optic neuropathy in a woman whose first symptom was a temporal blur. On her initial examination she had reduced visual acuity in one eye but normal visual acuity in the other with a bitemporal visual field defect, densest adjacent to fixation. Manchester and Calhoun3 reported bitemporal visual defects in several members of 2 generations of a family with dominantly inherited optic atrophy. Although 2 of the patients were asymptomatic, all of them had subnormal visual acuity. Votruba et al,4 in their extensive and intensive evaluation of patients with dominantly inherited optic atrophy, found that 33 of 50 study patients had superotemporal visual field defects when tested with automated perimetry. It is impossible to determine if any them were asymptomatic or had a pure chiasmatic syndrome.
How might a congenital bitemporal visual field defect eventuate from a genetic disorder? Recent research has revealed several possible molecular bases for chiasmal maldevelopment.5- 12 Retinal ganglion cell axons are believed to respond to specific cues in their environment during development as they migrate toward the optic chiasm. Cell surface proteins on neurons or glia in the chiasmal region may help to direct axonal processes to their appropriate destinations in the ipsilateral or contralateral optic tract and lateral geniculate nucleus. The loss of signals and/or growth factors that maintain the survival of fibers destined to cross in the chiasm may lead to their selective loss and subsequently to bitemporal visual field defects. There are even cell surface proteins such as the protein product of the gene roundabout that are believed to repel axons from crossing midline structures.10 It is possible that the absence of this protein at a stage where retinal ganglion cell axons from the inferonasal regions of the ipsilateral and contralateral eyes are approaching the midline structures of the optic chiasm may preclude survival of the retinal ganglion cells. Alternatively, loss of factors that maintain or influence neuronal metabolism or synaptic connections between retinal ganglion cells and then postsynaptic targets in the lateral geniculate nucleus may be responsible for selective loss of inferonasally located retinal ganglion cells.
We believe that the siblings described in this article have a novel hereditary chiasmal optic neuropathy that is recessively inherited. Although there are several possible mechanisms for selective loss of crossing axonal fibers in the optic chiasm, it is unclear which may be affected in this hereditary optic neuropathy. Current research in progress studying the molecular cues involved in the control of axonal crossing at midline structures may contribute to the identification and treatment of the hereditary genetic defect responsible for this phenotype.
Corresponding author: Simmons Lessell, MD, Neuro-ophthalmology Unit Massachusetts Eye and Ear Infirmary, Department of Ophthalmology, Harvard Medical School, 243 Charles St, Boston, MA 02114.
Pomeranz HD, Lessell S. A Hereditary Chiasmal Optic Neuropathy. Arch Ophthalmol. 1999;117(1):123-124. doi: