In this issue of the Journal of Neuro-Ophthalmology, optic nerve sheath fenestration (ONSF) was performed for 2 different indications. One was acquired hyperopia with choroidal folds (AHCF) (1) and the other was idiopathic intracranial hypertension (IIH) where the subgroup of patients with an opening pressure (OP) greater than 50 cm H2O was compared with the group with an OP less than 50 cm H2O (2). For the AHCF indication and the subgroup of patients with IIH with an OP >50 cm water, the procedure “failed.” This might be an appropriate time to ask, “What is the current status of ONSF?”
Historically, the first described OSNF dates back to 1872 when deWecker (3) reported 2 patients who likely had brain tumors with “neuroretinitis,” extensive vision loss, headaches, and a progressive course. With the eye deviated upwards and inwards, a specially designed instrument with a spatula and a sheathed neurotome was introduced transconjunctivally in the inferotemporal fornix, without any anesthetic, and the optic nerve sheath was blindly incised approximately 1 cm behind the globe. The first patient improved from no light perception (NLP) to light perception (LP) in the operated eye, whereas the second patient did not recover any vision. Both patients were relieved of their headaches, however.
In 1887, Carter (4) modified the procedure by detaching the lateral rectus muscle and incising the sheath under direct visualization along the length of the optic nerve. The first of his 3 reported patients did not improve from LP in the operated eye, but improvement from 20/50 to 20/40 was noted in the nonoperated eye. The second patient had relief of headache with improved vision in both eyes following left ONSF (right eye improved from NLP to hand movements (HM) while left eye improved from counting fingers (CF) to 10/100). The last patient was NLP in each eye and following left ONSF had relief of headache with vision of HM, right eye, and CF at 8 inches, left eye (5).
There was not much interest in ONSF until Hayreh (6) demonstrated its effectiveness in dealing with experimental papilledema in monkeys. Surgical approaches to the optic nerve include a medial orbitotomy (7) (after a Krönlein orbitotomy to enhance exposure), transconjunctival medial orbitotomy with detachment of the medial rectus muscle (8), lateral orbitotomy (9) (with and without removal of the lateral orbital wall) and, most recently, via the superomedial lid crease (10). A recent survey showed that the medial transconjunctival approach with a window vs. a slit fenestration is most popular (11).
There have been 2 theories based on histopathological studies by which the procedure is thought to work. One is occlusive fibrosis [using slits (12,13) or a window (14)], and the other is egress of fluid through a fistula similar to a glaucoma filtering operation (using slits) (15,16). There have not been any studies to show whether there is a therapeutic difference between slits or a window. Neither have there been any studies to show any therapeutic difference where along the course of the optic nerve the fenestration is performed. The immediate retrobulbar segment has a distinct enlargement of the subarachnoid space containing trabeculae (accessible via a medial or superomedial orbitotomy) while further along the intraorbital optic nerve, the subarachnoid space is smaller with broad septae and trabeculations (17).
The overall pressure-lowering effect of either a cerebrospinal fluid (CSF) diversion procedure or ONSF and the 3 compartment model have been challenged clinically and radiographically. Clinically, the presence of persistent optic disc edema in light of an apparently functioning shunt or successful ONSF argues that CSF is not distributed evenly with a continuous flow (18). Computed tomographic cisternography has demonstrated that the lowest concentration of contrast-loaded CSF was in the subarachnoid space in the optic nerve immediately behind the globe. In addition, CSF concentrations of lipocalin-like prostaglandin D synthase, a substance toxic to astrocytes, were highest in the same area leading to speculation that papilledema does not result from raised intracranial pressure alone but, in some instances, compartmentation of the subarachnoid space around the optic nerve promotes a toxic environment (18,19).
Regarding ONSF, it is axiomatic that correct therapy is dependent on correct diagnosis. In general, why do interventions (surgical or nonsurgical) fail? It could be the right operation for the wrong reasons, the wrong operation for the right reasons, the correct operation with technical failure, or simply related to the definition of “failure” which rather than improvement might simply mean lack of deterioration.
Considering the articles published in this issue of the Journal of Neuro-Ophthalmology, was ONSF the correct surgical procedure for the patient described by Adesina et al (1)? The diagnosis of AHCF was probably correct although no previous refractions were given (to show progressive hyperopia). Results of the cycloplegic refraction are more consistent with latent hyperopia. No axial lengths were obtained but the choroidal folds could possibly be explained by the hyperopia alone. The discs were minimally, if at all edematous, and may in fact have been anomalous. Optical coherence tomography (OCT) and fluorescein angiography would have helped to determine whether the discs were truly swollen (20). OCT with shape analysis of the peripapillary retinal pigment epithelium has not been reported with AHCF but provides another opportunity to study and possibly follow the therapeutic response (21,22). There was minimal if any change in the appearance of the optic discs following bilateral ONSFs. If ONSFs were to decompress the optic nerve sheath, it could theoretically reduce the flattening effect on the back of the globe. Although unlikely to be abnormal, the window specimen of dura could have been examined histopathologically (11), which might have helped to shed light on the pathogenesis of AHCF. One also might expect changes in the refraction (less hyperopia) or axial length (lengthening) postoperatively, but these variables were not measured.
The article by Robinson et al (2) retrospectively compared two groups with IIH. One subgroup had an OP >50 cm H2O and the other had an OP <50 cm H2O. The subgroup with the higher OP was associated with a 3-fold higher risk of “failure of ONSF to protect vision.” OP >50 cm H2O is a previously undescribed prognostic parameter that bears consideration in future management of patients with IIH. All ONSFs were performed by medial transconjunctival approach, but the use of slits vs. a window was not specified. Robinson et al found that high OP was correlated with worse initial visual acuity. They also found that poor initial visual acuity had the strongest correlation with poor outcomes and need for shunt (higher correlation than with the OP). Was high OP an independent risk factor for a poor outcome or was it directly related to the initial visual acuity? The authors could consider a mediation analysis. Other questions regarding the article include whether or not there was a correlation between OP and degree of papilledema? Was there any correlation between grade of papilledema and visual outcome? What were the differences in surgical technique among surgeons? Were the procedures unilateral or bilateral? How did they determine who would undergo ONSF and when?
Over the past year, JJ Chen MD, PhD and I have seen 6 patients who underwent unilateral ONSF (lateral orbitotomy, no bone flap, creation of a 3 × 5 mm window), three of whom had an OP >50 cm H2O. Two of these patients had IIH and the third likely had a combination of IIH and underlying congenital hydrocephalus. The first patient was a 17-year-old woman with a 1-month history of blurred vision and headaches. Grade 4 papilledema and choroidal folds were noted bilaterally. LP showed an OP of 55 cm H2O. She was diagnosed with IIH and started on high-dose acetazolamide. Within 1 week, her visual acuity dropped from 20/40 to 20/400 in the right eye, and from 20/20 to 20/40 in the left eye with a new right relative afferent pupillary defect (RAPD). Visual field examination of the right eye showed a cecocentral scotoma and the left eye had inferior greater than superior depression. The patient underwent right ONSF, and 2 months later, the visual acuity was 20/30 in each eye with normalization of her color vision. The papilledema had resolved and the right disc was mildly atrophic. The dosage of acetazolamide was being tapered and was currently 1 g daily.
The second patient was a 22-year-old woman with IIH who presented with a 1-month history of neck, base of skull, and bilateral eye pain with blurred vision. Visual acuity was 20/40, right eye, and 20/60, left eye. Color vision was normal but there was a left RAPD. Grade 4 papilledema and choroidal folds were present bilaterally. Visual fields were unreliable. OP was greater than 53 cm H2O. She was prescribed acetazolamide (2 g daily) for 2 weeks which improved her headaches. Signs of optic neuropathy persisted and because visual field testing was unreliable, left ONSF was performed. Two months postoperatively, visual acuity was 20/25, right eye, and 20/50, left eye. Optic disc edema resolved bilaterally with mild left optic atrophy. The acetazolamide dose was being tapered and was currently at 1 g daily.
The third patient was a 38-year-old woman born 3 months prematurely with hydrocephalus requiring ventriculoperitoneal (VP) shunting at age 18 months. She had done well until 3 months before our evaluation at which time she noticed bilateral visual blurring and transient visual obscurations. Visual acuity was 20/50, right eye, and 20/25, left eye. She was noted to have grade 3 papilledema in both eyes. Visual fields showed an enlarged blind spot in both eyes and an inferior arcuate defect in the left eye. Brain magnetic resonance imaging showed mild ventriculomegaly and the presence of a VP shunt. OP was 61 cm H2O. The patient was thought to have papilledema from a combination of IIH and her underlying congenital hydrocephalus. She initially was treated with acetazolamide and topiramate was added as her papilledema persisted. The patient noted a subjective decline in vision in her left eye. She underwent left ONSF, and 6 months postoperatively, the visual acuity was 20/20, right eye, and 20/30, left eye. Optic disc edema resolved and topiramate was discontinued while acetazolamide was being tapered, currently at 1 g daily.
These 3 cases illustrate 3 separate reasons for ONSF. The first patient had fairly fulminant IIH and required urgent ONSF to salvage her vision. The second patient had severe papilledema with unreliable fields and, therefore, ONSF was performed because it would be difficult to evaluate for worsening vision. The third patient had persistent papilledema and visual complaints despite medical therapy for many months. It is unlikely that she could have discontinued her medications successfully, thus ONSF was performed to allow her to taper off her medications.
Treatment for IIH tends to be institution and surgeon dependent. Fonseca et al (23) retrospectively compared the visual outcomes in 41 surgically treated patients from the Johns Hopkins University and 33 had at least 3 months of follow-up to be included in the analysis. There were 14 patients who underwent ONSF (12 unilateral and 2 bilateral) and 19 who underwent a shunting procedure. The ONSF group had a median preoperative papilledema grade of 4 in 12 of 14 patients and 3 patients later underwent a shunting procedure for persistent papilledema. The shunt group had 4 patients with some degree of optic atrophy and of the 13 of 15 patients with data, and the median preoperative papilledema grade was 2. One patient developed a postoperative shunt infection and 8 patients required shunt revisions (2 persistent papilledema and 6 persistent headache). The decision process between ONSF and shunting was not specified, but the conclusion was that “both procedures effectively reduced papilledema and are associated with improved visual function as measured by visual acuity testing and automated perimetry in patients with pseudotumor cerebri.” Interestingly, the authors speculated about the economic impact of the 2 procedures in that ONSF is generally an outpatient procedure that does not have to be repeated but may not relieve headache whereas shunting procedures require hospitalization and have a high incidence of repeat procedures. The Neuro-Ophthalmology Research Disease Investigator Consortium has proposed a prospective surgical IIH treatment trial to compare ONSF, CSF shunting procedure, and medication for patients with moderate to severe vision loss from IIH. Hopefully, this study will provide many answers regarding the best treatment options for patients presenting with severe IIH.
Where does all this currently leave us regarding ONSF? We at Mayo Clinic still advocate ONSF for medically unresponsive IIH with vision threatening disease. It has the ability to successfully decompress the papilledema, even bilaterally in many cases of unilateral surgery, and does not have the risk of spontaneous failure that is seen with current shunting operations. ONSF is especially recommended in the cases of pediatric IIH, if headaches are not problematic, unilateral optic disc edema of unknown etiology, brain tumors with chronic papilledema and visual loss, complicated meningitis cases with visually threatening disc edema where spread of either tumor or infection (24) through the shunt is a concern or as a salvage procedure if needed following a failed shunt.
It is my opinion that Keltner's 1988 editorial (25) on ONSF is still germane almost 20 years later. It is a safe and effective procedure.
1. Adesina OO, Warner JE, Patel BC. Optic nerve fenestration in a patient with the syndrome of acquired hyperopia and choroidal folds. J Neuroophthalmol. 2016;36:294–298.
2. Robinson ME, Moreau A, O'Meilia R, Pagteilan J, Ding K, Siatkowski RM, Farris BK. The relationship between optic nerve sheath decompression failure and intracranial pressure in idiopathic intracranial hypertension. J Neuroophthalmol. 2016;36:246–251.
3. DeWecker L. On incision of the optic nerve in cases of neuroretinitis. Int Ophthalmol Cong Rep. 1872;4:11–14.
4. Carter RB. On retrobulbar incision of the optic nerve in cases of swollen disc. Brain. 1887;10:199–209.
5. Carter RB. Operation of opening the sheath of the optic nerve for the relief of pressure. Br Med J. 1889;1:399–401.
6. Hayreh SS. Pathogenesis of edema of the optic disc (papilloedema): a preliminary report. Br J Ophthalmol. 1964;48:522–542.
7. Smith JL, Hoyt WF, Newton TH. Optic nerve sheath decompression for relief of chronic monocular choked disc. Am J Ophthalmol. 1969;68:633–639.
8. Galbraith JEK, Sullivan JH. Decompression of the perioptic meninges for relief of papilledema. Am J Ophthalmol. 1973;76:687–692.
9. Tse DT, Nerad JA, Anderson RL, Corbett JJ. Optic nerve sheath fenestration in pseudotumor cerebri. A lateral orbitotomy approach. Arch Ophthalmol. 1988;106:1458–1462.
10. Pelton RW, Patel BC. Superomedial lid crease approach to the medial intraconal space. A new technique for access to the optic nerve and central space. Ophthal Plast Reconstr Surg. 2001;17:241–253.
11. Sobel RK, Syed NA, Carter KD, Allen RC. Optic nerve sheath fenestration: current preferences in surgical approach and biopsy. Ophthal Plast Reconstr Surg. 2015;31:310–312.
12. Davidson SI. A surgical approach to plerocephalic disc oedema. Trans Ophthalmol Soc U K. 1970;89:669–690.
13. Davidson SI. The surgical relief of papilledema. In: Cant JS, ed. The Optic Nerve. London, United Kingdom: Henry Kimpton, 1972:174–179.
14. Hayreh SS. Pathogenesis of oedema of the optic disc. Doc Ophthalmol. 1968;24:289–411.
15. Keltner JL, Albert DM, Lubow M, Fritsch E, Davey LM. Optic nerve decompression: a clinical pathologic study. Arch Ophthalmol. 1977;95:97–104.
16. Tsai JC, Petrovich MS, Sadun AA. Histopathological and ultrastructural examination of optic nerve sheath decompression. Br J Ophthalmol. 1995;79:182–185.
17. Killer HE, Laeng HR, Flammer J, Groscurth P. Architecture of arachnoid trabeculae, pillars, and septa in the subarachnoid space of the human optic nerve: anatomy and clinical considerations. Br J Ophthalmol. 2003;87:777–781.
18. Killer HE, Jaggi GP, Miller NR. Papilledema revisited: is its pathophysiology really understood? Clin Experiment Ophthalmol. 2009;37:444–447.
19. Killer HE, Jaggi GP, Miller NR, Huber AR, Landolt H, Mironov A, Meyer P, Remonda L. Cerebrospinal fluid dynamics between the basal cisterns and the subarachnoid space of the optic nerve in patients with papilloedema. Br J Ophthalmol. 2011;95:82282–82287.
20. Jacobson DM. Intracranial hypertension and the syndrome of acquired hyperopia with choroidal folds. J Neuroophthalmol. 1995;15:178–185.
21. Sibony P, Kupersmith MJ, Rohlf FJ. Shape analysis of the peripapillary RPE layer in papilledema and ischemic optic neuropathy. Invest Ophthalmol Vis Sci. 2011;52:7987–7995.
22. Sibony P, Strachovsky M, Honkanen R, Kupersmith MJ. Optical coherence tomography shape analysis of the peripapillary retinal pigment epithelium layer in presumed optic nerve sheath meningiomas. J Neuroophthalmol. 2014;34:130–136.
23. Fonseca PL, Rigamonti D, Miller NR, Subramanian PS. Visual outcomes of surgical intervention for pseudotumor cerebri: optic nerve sheath fenestration vesus cerebrospinal fluid diversion. Br J Ophthalmol. 2014;98:1360–1363.
24. Garrity JA, Herman DC, Imes R, Fries P, Hughes CF, Campbell RJ. Optic nerve sheath decompression for visual loss with acquired immunodeficiency syndrome and cryptococcal meningitis and crpytococcal meningitis with papilledema. Am J Ophthalmol. 1993;116:472–478.
25. Keltner JL. Optic nerve sheath decompression. How does it work? Has its time come? Arch Ophthalmol. 1988;106:1365–1369.