Neurofibromatosis type 2 (NF2), previously known as central neurofibromatosis, is defined clinically by the presence of bilateral acoustic nerve schwannomas and genetically by a mutation in the long arm of chromosome 22.1 Schwannomas of other cranial and spinal nerves are often present, as well as meningiomas. Patients typically have hearing loss in the second or third decade of life. About 50% of cases are inherited in an autosomal dominant fashion.
Among the many ophthalmic signs that have been described in NF2,1- 4 papilledema has not been highlighted. We report 2 cases in which the dire consequences of sustained elevated intracranial pressures were not recognized in time to prevent visual loss.
A 31-year-old man reported a 2-year history of worsening bilateral hearing loss and dizziness in 1992. Bilateral sensorineural hearing was documented. Results of a neurological examination were otherwise normal. A magnetic resonance imaging (MRI) scan disclosed masses coating the cerebral convexities (Figure 1, A), the acoustic nerves, gasserian ganglia bilaterally, the right hypoglossal canal, the posterior tuberculum sella, and planum sphenoidale. These findings were considered diagnostic of sporadic NF2. A lumbar puncture revealed a normal opening pressure and the cerebrospinal fluid (CSF) formula was normal. Results of an ophthalmologic examination, including visual acuity, visual fields, and ophthalmoscopy, were entirely normal.
Case 1. A, Magnetic resonance imaging scan in 1992 shows diffuse thickening of the convexity dura by meningioma. B, Magnetic resonance imaging scan in 1997 shows interval growth of convexity meningioma.
In late 1993, mild bilateral optic disc edema was noted without any new symptoms or abnormalities of visual function (Figure 2, A). Another MRI scan showed no changes. He was diagnosed as having papilledema related to impaired CSF outflow owing to obstruction by diffuse convexity meningiomatosis. Because he had normal visual function, no further diagnostic or therapeutic intervention was undertaken. Subsequent semiannual ophthalmologic examinations during the next 2 years showed no new findings.
Case 1. A, In 1993, mild optic disc edema is present (left), but visual fields are normal (right). B, In September 1997, optic disc edema has become more pronounced (left), and early visual field loss has emerged (right). C, In December 1997, optic disc edema has worsened (left), and visual fields are markedly constricted (right).
After being lost to ophthalmologic follow-up for 2 years, he returned in September 1997 with a 2-week history of orthostatic transient visual obscurations. Visual acuity was still 20/20 OU, but automated visual fields showed generalized depression and mildly enlarged blind spots in both eyes, with an inferior nerve fiber bundle defect in the left eye Figure 2, B). The optic disc edema was more pronounced in each eye. An MRI scan showed interval growth of the cerebral dural convexity masses Figure 1, B). There was no venous sinus thrombosis or ventriculomegaly.
He was treated with acetazolamide (2 g/d) and furosemide (40 mg/d) but his visual acuity, visual fields, and optic disc edema worsened during the next 2 months. By December 1997, visual acuity had fallen to 20/80 OD and 20/40 OS. Visual fields now showed marked constriction in both eyes. Optic disc edema was even more severe (Figure 2, C). A lumbar puncture revealed an opening pressure of 32 cm H2O with a normal CSF formula. In the month that followed, he underwent optic nerve sheath fenestration in the right eye, high-dose intravenous methylprednisolone treatment, and a lumboperitoneal shunt, all of which failed to halt the downward spiral of visual loss. By February 1998, visual acuity had declined to hand motions OU. The optic discs were pale and flat (Figure 3). A shunt tap confirmed normalization of the opening pressure to 17 cm H2O.
Case 1. In 1998, 3 weeks after ventriculoperitoneal shunting, the optic disc edema has been replaced by disc pallor.
A 26-year-old man with a remote diagnosis of sporadic NF2 began complaining in 1997 of progressive visual loss in the left eye during the previous year and the recent onset of transient visual obscurations in that eye. His right eye had earlier been blinded by a retinal detachment.
He was totally deaf owing to bilateral acoustic schwannomas and had earlier undergone partial resections of a foramen magnum meningioma, a right frontal meningioma, and a left acoustic schwannoma. The acoustic surgery had resulted in left facial and trigeminal neuropathy with resulting neurotrophic keratitis in the left eye. A right acoustic schwannoma had remained unoperated. A brainstem stimulator, placed in the medulla in 1989 to provide some hearing, precluded further use of MRI.
Ophthalmologic examination in 1997 disclosed visual acuities of no light perception OD and 20/50 OS. Pupils measured 5 mm in dim illumination; the right did not react to direct light and the left was sluggish. A right afferent pupillary defect was present. Because of his deafness, neck weakness, and quadriplegia, visual fields could be assessed only by confrontation, showing a small temporal island in the left eye. Neurotrophic keratitis in the left eye had caused partial opacification of the cornea with pannus. Ophthalmoscopy of the left eye faintly disclosed chronic disc edema with some atrophy. The right fundus revealed a chronic total retinal detatchment with no view of the optic nerve.
A computed tomographic scan of the brain showed considerable enlargement of the ventricles relative to an MRI in 1988 (Figure 4, A and B). Because of the foramen magnum lesion, lumbar puncture was considered contraindicated for fear of provoking brainstem herniation. The patient's progressive visual loss was attributed to papilledema associated with chronic obstructive hydrocephalus from the posterior fossa masses. In preparation for a CSF shunting procedure, the patient was given acetazolamide (2 g/d) to avoid the visual loss associated with surgically induced sudden intracranial hypotension. Several weeks later, the patient's visual acuity had fallen to 20/200 OS.
Case 2. A, Magnetic resonance imaging scan in 1988 shows normal-sized ventricles. B, Computed tomographic scan in 1997 (before shunting operation) shows interval ventricular enlargement. C, Computed tomographic scan in 1998 (after shunting operation) shows reduction in ventricular size.
The patient underwent ventriculoperitoneal shunting that revealed an opening CSF pressure of 33 cm H2O. Four weeks after shunting, visual acuity had improved to 20/70 OS. Confrontation visual fields showed persistent nasal loss with a preserved temporal island. Fundus examination disclosed significantly less optic disc edema in the left eye with mild pallor. A postoperative computed tomographic scan revealed normalization of ventricular size (Figure 4, C).
Our cases highlight chronic papilledema as a cause of severe visual loss in NF2. This manifestation has been mentioned in previous reports, but without emphasis or explanation of mechanism. For example, Bouzas et al2 mentioned that 4 (7.4%) of 54 patients with NF2 had progressive visual field constriction from long-standing papilledema, one of whom had loss of central acuity in one eye to 20/160. Kaye et al4 and Landau and Yasargil3 separately made passing references to 2 patients with NF2 and papilledema secondary to intracranial tumors. Furthermore, the National Institutes of Health guidelines2 include an initial ophthalmic assessment for detecting the presence of juvenile cataracts and for the "photography of papilledema." No explanation accompanies this recommendation.
In our first patient, increased intracranial pressure was likely caused by diffuse convexity meningiomatosis that interfered with CSF outflow across the arachnoid granulations. In this communicating hydrocephalus, the fact that there was no pressure gradient across the ventricular outflow routes explains why the ventricles were not enlarged. A similar mechanism has been described in a case of childhood meningiomatosis.5 Dural sinus drainage may also have been compromised by the meningiomas, but was not seen on MRI or documented by angiography. To the best of our knowledge, these mechanisms of increased intracranial pressure in NF2 have not been previously described.
In our second patient, papilledema was due to compression of the brainstem by tumor. This led to an obstruction of CSF outflow through the fourth ventricle. In this noncommunicating hydrocephalus, there is a pressure gradient between the ventricles and subarachnoid space, leading to ventriculomegaly. We speculate that this may be the more common cause of papilledema in NF2, referred to previously by other authors.3,4
The early recognition and treatment of papilledema in patients with NF2 can be difficult. As in our second patient, media opacification from coexistent neurotrophic keratitis can inhibit the detection of disc edema. Once recognized, it may be difficult to decide at what point the potential for developing chronic atrophic papilledema outweighs the risks of intervention. Whether early surgery in these patients is warranted is unknown. Regular ophthalmologic follow-up is essential in preventing visual morbidity. Visual loss can be particularly devastating to these patients, as they often have hearing impairment and other neurological disabilities.
Reprints: Jonathan D. Trobe, MD, W. K. Kellogg Eye Center, 1000 Wall St, Ann Arbor, MI 48105 (e-mail: email@example.com).
Thomas DA, Trobe JD, Cornblath WT. Visual Loss Secondary to Increased Intracranial Pressure in Neurofibromatosis Type 2. Arch Ophthalmol. 1999;117(12):1650-1653. doi: