The visual course subsequent to the development of Leber hereditary optic neuropathy (LHON) is variable.1 Although the loss of vision is permanent in most patients, there is ample documentation that some patients' vision improves, in some cases as long as 10 years later.2 The prospect for recovery is related to the particular mitochondrial DNA (mt DNA) mutation responsible for the neuropathy. Only 4% of patients with the most prevalent mt DNA mutation(11778) have visual improvement, 3 whereas 37.8% of those with the 14484 mutation, found in 15% of pedigrees, have some visual recovery.4 In contrast to the sizable proportion of patients with LHON whose vision improves, we are aware of only 1 published case of late worsening in a patient with LHON (see below).5
Institutional review board approval was obtained for the case record reviews, with informed consent waived. The 3 patients whose cases form the basis of this report were identified retrospectively among the patients with LHON evaluated in the neuro-ophthalmic specialty practices of 2 academic hospitals.
In 1 patient who refused genetic testing the diagnosis of LHON was made based on clinical and laboratory findings, including the presence of the characteristic circumpapillary telangiectatic microangiopathy. In the other 2 patients an mt DNA mutation was identified.
In March 1979, a 21-year-old Mexican man, on chancing to cover his left eye, discovered he had impaired vision in his right eye. He had been healthy and was well nourished. He drank small quantities of alcohol on weekends and did not smoke or use illicit or prescription drugs. There was no family history of any eye or neurological disorders. He had no pain then or later. An ophthalmologist found that his visual acuities were limited to finger counting at 8 in OD and to 20/30 OS. Using Ishihara pseudoisochromatic color plates, color vision was absent in the right eye, and he could only identify 3 of 8 test plates with the left eye. He had a right relative afferent pupillary defect. There was a constricted peripheral field with a central scotoma in the right eye, but no visual field defect in the left eye by Goldmann kinetic perimetry. Neither disc appeared atrophic, but there was a bilateral circumpapillary telangiectatic microangiopathy. Myelinated nerve fibers were present at the lower pole of the left disc. Neurological examination revealed no other abnormalities and fluorescein angiography showed only the peripapillary microangiopathy. His electrocardiogram was remarkable for a short P-R interval and right bundle branch block. Cranial computed tomographic (CT) scanning of the head, chest radiographs, electroencephalography, cerebrospinal fluid analysis, complete blood count, and serological tests for syphilis and systemic lupus erythematosus gave normal results. Despite treatment with corticotropin and hydroxocobalamin, his visual acuity decreased to 1/70 OD and 1/200 OS 11 months after he was first seen by us. Goldmann perimetry revealed dense central scotomata in both eyes, with spared islands of vision superiorly and inferiorly in the left eye (Figure 1, A). Both optic nerves were pale, but a swath of myelinated fibers remained inferior to the left optic disc. Five and 7 years later, examination findings remained unchanged.
Case 1. Goldmann visual field perimetry (left eye). A, An abnormal visual field at age 22 years that subsequently remained stable. B, Visual field at age 29 years after symptomatic loss in the visual field in the left eye.
At age 29, without any change in his habits or general health, he noticed a further loss of visual field in his left eye. His acuities were unchanged, but there had been further visual field loss in the left eye (Figure 1, B), and the myelinated nerve fibers had disappeared from the left optic fundus.
In 1989, a 17-year-old, previously healthy adolescent Japanese girl began to see tiny pink flashes in each eye. These increased over the next 6 months at which time vision began to painlessly decrease, first in the right eye and then in the left. There was no family history of any eye or neurological diseases. She was well nourished and abstemious. Findings from cranial CT scan, magnetic resonance imaging, and neurological examination were all normal. She failed to improve with corticosteroid therapy and visual acuities declined to 20/400 OU. A blood test showed that she had the 11778mt DNA mutation. From ages 19 to 23 years she smoked cigarettes, but there was no change in her vision until age 25. At that time she complained of glare sensitivity and noted that her vision temporarily decreased in both eyes during menses. She had started treatment with fluoxetine hydrochloride(Prozac) for depression. Shortly thereafter her vision began to decline painlessly, stabilizing months later at light perception OD and counting fingers at 4 ft OS.
In December 1957, a 35-year-old white man developed gradual painless loss of vision in his left eye. The visual acuity decreased to 20/200 OS with a central scotoma. The patient was treated with intravenous corticotropin. His visual acuity OS improved over the next 6 months to 20/40. In September 1958, the visual acuity OD deteriorated to only hand motions. Tangent screen testing showed a temporal visual field defect. One month later the visual acuity OS relapsed and decreased to 20/400. Analysis of his cerebrospinal fluid gave normal results. A pneumoencephalogram showed increased air in the suprasellar cisterns, suggesting chiasmal atrophy. Both optic discs were slightly pale. A craniotomy in October 1958 revealed no abnormalities.
Over the next decade his visual acuity improved, reaching 20/30 OD and 20/100 OS. The acuities remained unchanged for the next 32 years and the patient was able to work in his customary occupation as a printer. A cranial CT scan in 1989 showed no abnormalities. A cataract was removed from his right eye in 1995.
In 2000, at age 78, his visual acuity decreased to 20/800 OD over a period of several months. The vision also declined to the same level in his left eye. He was then in excellent general health and his eye examination results showed no ocular basis for the loss of vision. A blood test revealed that he had the 14484 mt DNA mutation. Humphrey perimetry showed that he had a large dense central scotoma in the right eye and temporal hemianopia in the left eye. Compared with prior tangent screen and Goldmann visual field examinations, the pattern of field loss had not changed since the initial attack of LHON. However, small central areas of recovered field were lost again when visual function relapsed.
Late, recurrent visual loss appears to occur in at least a small subset of patients with LHON. In the first patient, despite the absence of affected family members and the lack of genetic testing, the diagnosis of LHON seems very likely. The patient had a sequential, bilateral, painless optic neuropathy with circumpapillary telangiectatic microangiopathy, severe unremitting visual loss, and normal findings on neurological examination CT scan, and cerebrospinal fluid analysis. Skeptics might ascribe the late change in his visual field defect to intertest variability, but the loss of myelinated fibers from the fundus reinforces our conclusion that the patient indeed suffered secondary loss of visual function years after his condition had clinically stabilized. In the third patient, there were 3 attacks of LHON in the right eye and 2 in the left over a long period. This case demonstrates that multiple attacks of LHON can occur independently in each eye, with intercurrent improvement. No external factors were identified to account for remissions or relapses. Although there was no family history of LHON in our second and third cases, the diagnosis was firmly established by genetic testing.
To our knowledge, the only other published case of recurrent visual loss in a patient with LHON was reported by Chuman et al.5 Their patient, an alcohol and tobacco abuser, first experienced a decrease in visual acuity at age 38, with acuity declining to 20/30 OD and to 20/300 OS. With the cessation of smoking and drinking his visual acuity recovered to 20/20 OU. From age 38 until age 53, he experienced another 6 cycles of loss and recovery of visual function. His optic discs were described as small and hyperemic with blurred margins, and there was a circumpapillary telangiectatic microangiopathy. A blood test showed that he had the 11778 mt DNA mutation. One very unusual aspect of this patient's case is that he did not develop optic atrophy despite 7 episodes of neurogenic visual loss. Perhaps this patient's visual loss was due to somethingother than LHON. Although he had the peripapillary microangiopathy typical of LHON, this can be found in carriers of the LHON mt DNA mutations who retain normal vision.6- 9 It may be that this patient suffered recurrent episodes of nutritional or toxic amblyopia to which the mt DNA mutation made him susceptible.
Since it is well established that the mt DNA mutation, while necessary, is not sufficient to cause visual loss, efforts have been made to identify other genetic or epigenetic factors that play a role in this disease. Some authors have speculated that epigenetic toxic or nutritional factors, such as alcohol and tobacco use, are responsible for the loss of vision in LHON.10- 12 There is reason to doubt the importance of these particular exogenous factors13 in most cases of LHON, and in none of our cases could the initial episode of visual loss be related to any obvious exogenous factors.
Whatever causes a patient with the LHON genotype to express the phenotype, it is very unlikely that the exposure occurs only once in the patient's lifetime. Yet cases of recurrent or secondary visual loss are patently rare in LHON. Why is this? The subsequent loss might be missed in patients who were left with profound, unremitting visual loss after the first attack, if meticulous testing of acuity and visual fields were not maintained. However, this could not explain the rarity in those patients whose vision improves after the initial episode. Since, following the initial decrease in vision, most patients retain some visual function, one might expect that secondary visual loss should be the rule and not the exception in patients harboring one of the mt DNA mutations of LHON. The apparent rarity of recurrent visual loss in LHON remains an etiopathogenic conundrum.
The authors have no proprietary or commercial interest in any materials or products mentioned in this study.
Corresponding author: Simmons Lessell, MD, MEEI Ninth Floor, 243 Charles St, Boston, MA 02114 (e-mail: firstname.lastname@example.org).
Newman-Toker DE, Horton JC, Lessell S. Recurrent Visual Loss in Leber Hereditary Optic Neuropathy. Arch Ophthalmol. 2003;121(2):288-291. doi:10.1001/archopht.121.2.288