Retrolaminar Migration of Intraocular Silicone Oil

Migration of intravitreal silicone to the retrolaminar optic nerve was detected pathologically in 1983, symptomatic migration to the subarachnoid space of the optic nerve was reported in 1994, and asymptomatic intraventricular silicone was first seen radiographically in 1999. Since then, little advance has been made in understanding this phenomenon despite numerous case reports. Although some authors have restricted their attention to cases of intraventricular silicone, we believe that these represent part of a clinical spectrum and that all cases with retrolaminar silicone should be considered. The pathophysiology of silicone migration may have significant implications for the management of patients after vitrectomy.; Two patients were evaluated by the authors. An internet-based literature review was conducted, beginning with the key search terms 'intraventricular, intracranial, subarachnoid, or optic nerve silicone,' and 'complications of vitrectomy or intravitreal silicone.' Further searches cascaded from the initial search results. An additional 24 cases of retrolaminar migration of silicone oil were found and summarized. The relevant anatomy and pathophysiology were reviewed, with attention to additional information from enucleation studies, as well as to gaps in the current understanding of this process.; Retrolaminar migration of silicone oil may be more common than previously thought, especially in at-risk patient groups, and may be associated with visual and neurologic symptoms. Some impressions regarding the cause and significance of this syndrome seem incorrect. Although this process is likely linked to postoperative elevations of intraocular pressure, the exact mechanisms of silicone entry into the subarachnoid space remain undefined. A number of anatomic factors may influence the movement of silicone from the orbit and in the various compartments of the subarachnoid space and ventricular system, resulting in variability of clinical presentations and radiologic findings. Implications for clinical decision making and directions for further research are discussed.; Greater awareness on the part of treating physicians, systematic study of at-risk populations, and advances in imaging technology will allow further insight into this phenomenon.

State-of-the-Art Review Section Editors: Valérie Biousse, MD Steven Galetta, MD Retrolaminar Migration of Intraocular Silicone Oil Rance A. Boren, MD, Carson D. Cloy, Ankur S. Gupta, MD, Vinay N. Dewan, MD, R. Nick Hogan, MD, PhD Background: Migration of intravitreal silicone to the retrolaminar optic nerve was detected pathologically in 1983, symptomatic migration to the subarachnoid space of the optic nerve was reported in 1994, and asymptomatic intraventricular silicone was first seen radiographically in 1999. Since then, little advance has been made in understanding this phenomenon despite numerous case reports. Although some authors have restricted their attention to cases of intraventricular silicone, we believe that these represent part of a clinical spectrum and that all cases with retrolaminar silicone should be considered. The pathophysiology of silicone migration may have significant implications for the management of patients after vitrectomy. Evidence Acquisition: Two patients were evaluated by the authors. An internet-based literature review was conducted, beginning with the key search terms "intraventricular, intracranial, subarachnoid, or optic nerve silicone," and "complications of vitrectomy or intravitreal silicone." Further searches cascaded from the initial search results. An additional 24 cases of retrolaminar migration of silicone oil were found and summarized. The relevant anatomy and pathophysiology were reviewed, with attention to additional information from enucleation studies, as well as to gaps in the current understanding of this process. Results: Retrolaminar migration of silicone oil may be more common than previously thought, especially in atrisk patient groups, and may be associated with visual and neurologic symptoms. Some impressions regarding the cause and significance of this syndrome seem incorrect. Although this process is likely linked to postoperative elevations of intraocular pressure, the exact mechanisms of silicone entry into the subarachnoid space remain undefined. A number of anatomic factors may influence the movement of silicone from the orbit and in the various compartments of the subarachnoid space and ventricular system, resulting in variability of clinical presentations and radiologic findings. Implications for clinical decision making and directions for further research are discussed. Brownwood Regional Medical Center (RB, CC), Brownwood, Texas; and Department of Ophthalmology (AG, VD, RNH), University of Texas Southwestern Medical Center, Dallas, Texas. The authors report no conflicts of interest. Address correspondence to Rance Boren, MD, PO Box 610, Brownwood, TX 78604; E-mail: [e-mail]. Boren et al: J Neuro-Ophthalmol 2016; 36: 439-447 Conclusion: Greater awareness on the part of treating physicians, systematic study of at-risk populations, and advances in imaging technology will allow further insight into this phenomenon. Journal of Neuro-Ophthalmology 2016;36:439-447 doi: 10.1097/WNO.0000000000000440 © 2016 by North American Neuro-Ophthalmology Society R etrolaminar migration of silicone oil is a recognized complication of intravitreal silicone injection, with at least 24 patients reported since 1983 (1-25) (Table 1). A 2014 review of the neuronal complications of silicone oil, including 8 cases of intracranial silicone, concluded that the phenomenon was benign, that preexisting glaucoma and congenital abnormalities (optic disc pits) were the main risk factors for its development, and that in their absence no special preoperative consideration was needed. There was little discussion of postoperative management (26). In addition to an extensive literature review, we report 2 additional cases. REPORT OF CASES Patient One An 82-year-old woman with Type II diabetes mellitus on aspirin and clopidogrel for coronary artery disease was hospitalized with a large pretibial hematoma. Her hemoglobin dropped from 13.6 to 9.7 mg/dL. She became confused and brain computed tomography (CT) showed hyperdense areas in the frontal horns of both lateral ventricles, as well as in the right eye and the nondependent portion of the right temporal horn (Fig. 1). Hounsfield units (HU) were 95 in the right eye and 75 in the intraventricular lesions. The initial radiologic diagnosis for the intracranial abnormalities was one of residual Pantopaque vs atypical intraventricular hemorrhage. Consultation with the patient's ophthalmologist confirmed a history of multiple previous retinal detachments in the right eye because of proliferative retinopathy, 439 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Reference, yr Location of Silicone at Presentation* Duration of Tamponade Preop/Postop Glaucoma Viscosity (Centistokes) Underlying Retinal Pathology Initial Signs & Symptoms/Clinical Course Enucleation for sympathatic ophthalmia Enucleation for painful glaucoma Postop visual loss Enucleation for painful glaucoma HIV complications Ni et al (1) Optic nerve 12 yr No/yes NA RD Shields and Eagle (2) Optic nerve 3 yr No/yes NA Traumatic RD Tang et al (3) Wenkel and Naumann (4) Williams et al (5) and Eller et al (6) Papp et al (7) Optic nerve sheath Optic nerve 1d 3.5 yr No/no No/yes NA NA RD Aphakic RD Optic nerve and chiasm; bilateral lateral ventricles Optic nerve and subarachnoid space 15 mo No/yes 5,000 CMV retinitis 2 mo No/yes NA Diabetic retinopathy Taherian et al (8) Fangtian et al (9) Optic nerve and subarachnoid space Suprasellar cistern; 3rd, 4th, and right lateral ventricles Bilateral lateral ventricles Optic nerve and chiasm Optic chiasm and tract; bilateral lateral ventricles Optic nerve and chiasm; bilateral lateral ventricles Optic nerve; right lateral ventricle Optic chiasm; suprasellar region Optic nerve, chiasm, and tract; bilateral lateral ventricles Optic nerve; 4th ventricle; right lateral ventricle Left lateral ventricle 3rd ventricle; bilateral lateral ventricles Optic nerve; suprasellar region; right lateral ventricle Bilateral lateral ventricles Optic nerve, chiasm, and tract; bilateral lateral Ventricles Optic nerve and chiasm Third ventricle Optic nerve; bilateral lateral ventricles 3 wk 8 mo No/yes No/yes 1,000 5,000 Macular hole Diabetic retinopathy Enucleation for painful glaucoma Postop visual loss Syncope and "dizziness" 15 mo 1 yr 5.5 yr No/yes No/yes No/yes NA 5,000 NA Diabetic retinopathy RD Optic pit Hypoglycemia/AMS Chiasmal field defect Persistent headaches NA NA/NA NA Diabetic retinopathy Nausea and headaches NA 8 yr 15 yr NA/NA NA/yes NA/NA NA NA NA Diabetic retinopathy RD Diabetic retinopathy Headache/leg weakness Chiasmal field defect Headache after a fall 10 yr NA/NA NA Diabetic retinopathy Headache and dizziness NA 1 yr 20 yr NA/NA No/yes NA/NA NA NA NA NA Diabetic retinopathy RD Seizures Persistent headaches Headaches NA NA NA/NA NA/NA NA NA NA RD 9 mo 10 yr 3 yr No/yes NA/NA No/yes NA NA 1,300 Diabetic retinopathy Diabetic retinopathy RD Yu and Apk (10) Eckle et al (11) Kuhn et al (12) Chen et al (13) Tatewaki et al (14) Lee et al (15) Jabbour et al (16) Boren et al: J Neuro-Ophthalmol 2016; 36: 439-447 Chang et al (17) Campbell et al (18) Hruby et al (19) Cosgrove et al (20) Sato et al (21) Chaio et al (22) Kim et al (23) Swami et al (24) Mathis et al (25) Headache and AMS Concussion Hypertensive urgency Sepsis and AMS Stroke *Except for the pathologically proven cases, "Optic Nerve" silicone may be subarachnoid and/or intraneural. The exact location of silicone in the suprasellar/chiasmal region cannot be confirmed in every case; the authors' descriptions were used. NA, not available or unclear; CMV, cytomegalovirus; HIV, human immunodeficiency virus; RD, retinal detachment without mention of additional pathology; AMS, altered mental status. State-of-the-Art Review 440 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. TABLE 1. Reported cases of retrolaminar migration of silicone oil State-of-the-Art Review FIG. 1. Patient 1. Unenhanced axial computed tomography demonstrates silicone oil within the right eye and right temporal horn (A) and in the frontal horns of both lateral ventricles (B). with silicone oil placed after a vitrectomy 9 years earlier. There was no history of myelography. The diagnosis of asymptomatic intraventricular silicone oil was made. The patient's mental status changes were attributed to hospital-acquired delirium and gradually resolved. At follow-up after hospitalization her intraocular pressure (IOP) was 42 mm Hg in the right eye. Further review showed that her IOP in the right eye had been intermittently elevated since her retinal surgery and that she had been blind in that eye for several years. migration and the underlying mechanisms. Silicone oil has been documented in the optic nerve sheath 24 hours after intravitreal injection, within the optic nerve by 1 month, and in the lateral ventricle by 8 months (3,9,27). Conversely, a prospective MRI study of 19 patients undergoing silicone extraction after successful treatment of retinal detachment (mean duration of tamponade: 3 months) failed to demonstrate extraocular silicone, despite the use of silicone-sensitive sequences (28). Patient Two A 42-year-old man with Type II diabetes mellitus presented to the emergency department with 3 days of painless vision loss in the right eye because of vitreous hemorrhage. Visual acuity was 20/400, right eye and no light perception, left eye. Brain magnetic resonance imaging (MRI) revealed silicone along the left optic nerve as far as the chiasm (Fig. 2), and small amounts in both lateral ventricles. Sixteen months previously, the patient underwent vitrectomy with silicone tamponade in the left eye for total retinal detachment secondary to proliferative diabetic retinopathy. Over the first postoperative year, he had IOPs of 36, 50, and 65 mm Hg associated with progression to total visual loss in the left eye. PREVALANCE Infiltration of silicone into the retrolaminar optic nerve was first demonstrated pathologically by Ni et al in 1983 (1). Involvement of the optic nerve may or may not be a precondition for further movement into the subarachnoid space, but the 2 may coexist. The frequencies of both these complications are unknown, as are the time course of the Boren et al: J Neuro-Ophthalmol 2016; 36: 439-447 FIG. 2. Patient 2. Axial fluid attenuated inversion recovery magnetic resonance imaging reveals silicone oil in the left eye and extending along the left optic nerve to the chiasm (arrow). 441 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. State-of-the-Art Review Based on pathologic studies of silicone optic neuropathy, it seems likely that there is a correlation with both the length of tamponade and the presence of ocular hypertension. A retrospective assessment of 74 enucleated eyes by Budde et al (29) showed presumed silicone vacuoles in 19% of optic nerve specimens. Duration of tamponade was greater in the affected eyes (3.5 vs 1.5 years). Clinical information was available for 10/14 affected eyes; all had postoperative glaucoma. A study of 8 enucleated eyes found similar vacuoles only in the 2 patients with postoperative glaucoma (30) In a series of 36 enucleated eyes, 2 patients with significantly IOP developed presumed optic nerve infiltration within 1 month (27). The status of the other affected patients cannot be determined from the article, although glaucomatous pathology was reported in 56% of eyes and only 25% of nerves demonstrated apparent silicone infiltration. Another series of 9 eyes revealed silicone in the optic nerve of 1 patient (11%) (31). The overall rate of presumed silicone infiltration in these studies was 20%. For comparison, the incidence of Schnabel cavernous degeneration (see below) in a series of 155 enucleated glaucomatous eyes was 1.3%, and in eyes with both glaucoma and uveal melanoma the rate was 11.6%. (32). Enucleation studies introduce obvious selection bias that may magnify the effect of glaucoma. There are no estimates of the frequency of silicone infiltration in patients who do not come to enucleation, nor are there pathologic studies of patients in whom intracranial silicone has been documented. Subarachnoid (perineural) silicone has only been demonstrated pathologically in 1 eye with optic nerve infiltration (7). CLINICAL FINDINGS From a neurologic standpoint, the natural history of migration of silicone seems benign in most cases; many were detected incidentally during evaluation of nonspecific symptoms such as headaches or dizziness. However, a patient with chronic headaches, intraventricular silicone, and elevated cerebrospinal fluid (CSF) pressure (27 cm H2O) has been reported (19). This patient lost sight despite normal postoperative IOPs and eventually underwent enucleation after developing painful neovascular glaucoma (maximum IOP = 52 mm Hg) several months later. He developed headaches 5 years after his initial surgery and failed conservative treatment before obtaining symptomatic relief after ventriculoperitoneal shunting. Associated visual problems are more common. Tang et al (3) reported a case of immediate postoperative visual loss and perineural subarachnoid silicone oil. The patient was treated with optic nerve decompression within 24 hours but had no significant visual improvement over several months. Another patient with sudden postoperative (1 week) visual loss was found to have an optic nerve sheath that was "distended with silicone oil" on MRI but due to a delayed 442 presentation no treatment was proposed (8). Eckle et al (11) reported a patient who had undergone repair of a retinal detachment in his left eye. One year later, with acuity of 20/20, right eye and light perception, left eye, he suddenly developed temporal visual field loss in the right eye. This was attributed to silicone oil pooling in the "optic nerve sheath" at the chiasm. The patient underwent craniotomy with chiasmal decompression and improvement in the right visual field. Acuity in the left eye remained light perception. The left optic nerve was described as "hollowed out" (11). In another case of monocular temporal field defect of the untreated eye, the field loss resolved over a month in association with radiographically documented migration of silicone from the suprasellar region to lateral ventricle (15). A patient with immediate postoperative periorbital pain, ptosis, orbital, and meningeal enhancement, inflammatory spinal fluid, and presumed silicone oil in the left lateral ventricle was described by Espinoza et al (33). In this case, the intraventricular abnormality was in the mid-portion of the left lateral ventricle near the foramen of Monro, and, on the basis of the single image provided, may reflect CSF pulsation artifact rather than intraventricular silicone (34). This patient was excluded from our review of cases. Most of the other patients with a detailed ophthalmologic history described a progressive loss of vision in the treated eye before the discovery of intracranial silicone oil. In some cases this could have been due to glaucoma or retinal disease. However, in 3 patients the authors specifically commented on the loss of vision despite normal or minimally elevated IOPs (1,12,19). A report of visual outcomes after silicone injection has suggested that a significant number of patients with good visual potential may have decline in vision related to the oil itself (35). Microcystic macular changes detected by optical coherence tomography (OCT) have been described in eyes after silicone oil injection. These retinal changes are similar to those reported from other causes, including Leber hereditary optic neuropathy and multiple sclerosis (36). These macular changes may be secondary to a silicone-related optic neuropathy. However, in a rabbit model, loss of retinal neurons with optic atrophy has been described in the absence of elevated IOP, silicone infiltration of the optic nerve, or subarachnoid silicone (37). The diagnosis of intraventricular silicone may be made on the basis of an unenhanced CT, with the findings being the globular and nondependent nature of the radiologically hyperdense abnormalities, features that are due to the low specific gravity and high viscosity and interfacial tension of silicone oil (38). Numerous patients have been scanned in the prone position to document buoyancy of silicone (5,9,12,17-19,22). Asymptomatic movement of silicone from the optic nerve/chiasm to the lateral ventricles, and between lateral ventricles, has been noted on serial scans (13,17). Movement of silicone between the third and lateral ventricles has been demonstrated over 48 hours (24). Boren et al: J Neuro-Ophthalmol 2016; 36: 439-447 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. State-of-the-Art Review Therefore, it should be remembered that imaging findings reflect only a snapshot of a dynamic process. Silicone in the optic nerve or sheath appears as a tubular hyperdensity but may be difficult to appreciate without dedicated orbital views, fortuitous slice alignment, or silicone-sensitive sequences (14,25). The main differential diagnosis on CT is hemorrhage, although an isolated globule in the third ventricle mimicked a colloid cyst (24). Silicone oil near the chiasm may resemble an aneurysm or subarachnoid blood in the basal cisterns (23). Several patients underwent additional testing or were referred to neurosurgical centers for these findings (13,16,18,23). Intraocular silicone was originally found to have a density of 110-130 HU (39). Including our first patient, the average of reported intraventricular lesions was 82 HU (range 55-98 HU) (10,13,14,17,18,24,25). This is still higher than typically seen with hemorrhage (30-60 HU), which tends to be dependent and form fluid levels. Pantopaque has a higher specific gravity than CSF and when seen intracranially, it classically forms multiple lesions near the skull base (40). The MRI findings with ocular silicone have been well characterized and intraventricular silicone has been found to follow the ocular silicone on all sequences (39,41). In addition, a characteristic chemical shift artifact may be detected (5,39). Two of the reported patients were found to have intracranial silicone after silicone removal from the eye or after enucleation (9,19). PROPOSED MECHANISMS OF SILICONE MIGRATION FROM THE VITREOUS CAVITY Both pathology and advanced imaging techniques have revealed the presence of silicone inclusions in all retinal layers, the choroid, and optic nerve (1,2,42-44). This suggests the possibility that infiltration of silicone into the retina could be followed by movement along the retinal nerve fiber layer into the optic nerve. However, it is generally assumed that the silicone infiltrates the optic nerve directly through the lamina cribrosa (along the axis of the entering axons) facilitated by postoperative pressure elevations. Hyaluronic acid-filled cavities in the optic nerve (Schnabel cavernous degeneration) were initially associated with glaucoma, and the classical paradigm is that the mucopolysaccharide in the optic nerves reflects infiltration by vitreous (45). Shields and Eagle (2) found "optically empty" vacuoles in the optic nerve of a silicone-treated eye and postulated that migrating silicone replaced vitreous in such cases, using the term "pseudo-Schnabel" degeneration to describe this process. Subsequent studies using energy dispersive X-ray analysis confirmed the presence of silicone in these vacuoles (7,46). Although intraneural silicone oil originates in the vitreous cavity, it has been suggested based on histochemical Boren et al: J Neuro-Ophthalmol 2016; 36: 439-447 studies that the hyaluronic acid seen in Schnabel degeneration may be produced in situ by Type II astrocytes as a response to tissue damage (47). Similar pathologic findings have been observed in cases of ischemic optic neuropathy without glaucoma, and a large autopsy study suggested that cavernous degeneration may develop because of chronic ischemia independent of IOP (48,49). Therefore, the exact relationship between these 2 syndromes remains unclear. Williams et al (5) postulated that infiltration of the optic nerve by silicone was followed by coalescence into larger vacuoles, posterior migration, and eventual entry of silicone into the subarachnoid space of the optic nerve. Although it seems unlikely that this mechanism could be a major factor in patients who present in the immediate postoperative period, it has been shown that Schnabel pathology may develop within 4 days in an animal model of acute glaucoma, even without structural changes in the lamina cribrosa (50). Rupture through the periphery of the lamina cribrosa with direct passage of silicone to the subarachnoid space was proposed by Fangtian et al (9) whose patient had significant optic disc cupping. (9) Detailed assessments of the anatomic changes in the lamina cribrosa in glaucomatous eyes have shown that thinning and cupping of this structure may combine to bring the posterior lamina into direct contact with the pia (51,52). Laminar holes and dehiscences have been demonstrated in glaucomatous eyes by OCT (53). These defects may provide a route for silicone movement. Congenital optic nerve pits also have been implicated in migration of silicone into the subarachnoid space. Kuhn et al (12) reported a 15-year-old girl with bilateral optic pits who lost vision 2-3 years after silicone placement in her left eye despite only minimally elevated IOP (26 mm Hg). Three years later she developed "strong persistent headaches," and was found to have an IOP of 36 mm Hg in the left eye and silicone in the ventricular system on MRI. Optic nerve pits are rare, and are commonly associated with serous retinal detachments (54). Therefore, it is unclear whether the case reported by Kuhn et al represents a chance association or an additional route of silicone migration. To our knowledge, silicone migration has not been reported in patients with retinal detachment related to coloboma or morning glory syndrome (55). Movement of silicone through the perivascular space could be involved in passage from the retina to the optic nerve or from the nerve to the subarachnoid space. Pathologic studies typically include no more than 1 cm of optic nerve and infiltration of oil has been noted up to the surgical margins (2,4,29). The central retinal artery and vein pierce the optic nerve just beyond this point and may provide the most direct conduit from the optic nerve to the subarachnoid space (Fig. 3). This route has been proposed as an explanation for some cases of Terson syndrome (56). Transpial penetrating vessels supplying the chiasm or optic tract could conceivably play a similar role intracranially. 443 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. State-of-the-Art Review FIG. 3. Schematic representation of the anatomy and blood supply of the optic nerve. Note the proximity (D) of vitreous cavity to the subarachnoid space surrounding the optic nerve. The hatched area indicates the communication between the perivascular spaces and the subarachnoid space. Modified from Hayreh S. Anatomy and physiology of the optic nerve head. Trans Am Acad Ophthalmol Otolaryngol. 1974;8:OP240-OP254. A, arachnoid; C, choroid; CRA, central retinal artery; Col. Br., Collateral branches; CRV, central retinal vein; D, dura; LC, lamina cribrosa; OD, optic disc; ON, optic nerve; P, pia; PCA, posterior ciliary artery; PR and PLR, prelaminar region; R, retina; RA, retinal arteriole; S, sclera; SAS, subarachnoid space. Chiao et al (22) noted that these mechanisms of silicone movement are not mutually exclusive, and there is insufficient evidence to support one over the other. It is possible that rupture through the periphery of the lamina cribrosa or dissection along the central retinal vessels explains cases with rapid subarachnoid spread of silicone oil, but that gradual infiltration of the optic nerve followed by posterior migration (with or without slow leakage into the CSF) also could occur (Fig. 4). Alternatively, owing to the lack of a physiologic communication between the posterior chamber and the subarachnoid space, an "active transport" system has been postulated to explain silicone migration. Knect et al (57) reported a negative study of silicone maintained under pressure in cadaveric eyes in support of this theory. However, a short-duration study in biologically inert eyes is of limited application. Although the term "active transport" was attributed to Budde and Papp, both authors stated that postoperative ocular hypertension was the driving force behind silicone movement (7,29,58). Because silicone-containing macrophages have been observed throughout the orbit and in some cases of optic nerve infiltration their involvement in the "active transport" of silicone has been proposed (7,59). Macrophages have been noted in inflammatory tissue surrounding breast implants and silicone has been noted in macrophages in lymph nodes after both arthroplasty and mammoplasty (60). However, in the study by Budde et al (29), siliconeladen macrophages were seen in only 3/14 cases (21%) 444 FIG. 4. Diagram of the optic nerve and chiasm demonstrating possible routes of silicone migration. Yellow arrows: intraneural/subpial silicone; purple arrows: subarachnoid silicone. (29). Therefore, it seems more likely that the presence of macrophages/microglia is a response to silicone infiltration of the optic nerve, rather than the cause, and that their role in role in silicone movement is negligible. INTRACRANIAL SPREAD OF SILICONE Once silicone gains access to the subarachnoid space of the optic nerve, several factors may influence further movement. The compartmentalized nature of the optic nerve subarachnoid space surrounding the orbital segment of the optic nerve and the choke point at the canalicular segment (Fig. 5) could lead to arrest of silicone migration (61). Owing to the buoyancy of silicone, movement away from the eye may even have a tendency to reverse during supine sleep. Also, dural lymphatics have been proposed as a significant route of CSF reabsorption in the optic nerve (62,63). These might be expected to clear small quantities of silicone from this segment. Owing to a combination of these factors, there may be cases in which subarachnoid silicone never reaches the intracranial compartment. The optic nerve enters the cranial cavity through the optic canal. There the subarachnoid space of the optic nerve is in direct communication with the subarachnoid space of Boren et al: J Neuro-Ophthalmol 2016; 36: 439-447 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. State-of-the-Art Review FIG. 5. Scanning electron micrograph of the optic nerve in the sagittal plane demonstrating narrowing of the subarachnoid space in the canilicular segment (arrow) (A), and the complex trabecula and septa bridging the subarachnoid space (B). Reproduced with permission from Killer H, Jaggi G, Miller N, Huber A, 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 papilledema. Br J Ophthalmol. 2011;95:822-827. the brain, and ultimately with the ventricular system. Eller et al (6) postulated that individual microbubbles of silicone are carried to the ventricles and coalesce there (6). However, in at least 1 case there was apparent collection of silicone in the suprasellar cistern (15). This, along with the other clinical and radiologic observations described above, suggests that the high viscosity of silicone favors its migration as a large bolus in some cases. There are numerous trabecula in the cerebral subarachnoid space, and the individual basal cisterns are separated by sheets of arachnoid, which must be circumvented before subarachnoid silicone can reach the fourth ventricle. Historically, delays in migration of air attributable to these structures were seen during the performance of air cisternography (64). Even the foramina of Luschka and Magendie have been found to be obstructed by a membrane in some patients (64,65). Taken together, these factors suggest that anatomic barriers to silicone migration may account for variability between patients. Finally, as it goes against both gravity and the classic model of CSF flow, the entry of silicone into the ventricular system adds additional support to the emerging understanding of CSF flow as a pulsatile bidirectional process (66,67). In many cases, intracranial silicone lies within the substance of the optic nerve and chiasm. Our second patient (Fig. 2) and others (Table 1) illustrate this. In the case reported by Eckle et al (11), silicone at the chiasm was surgically proven to be subpial. This indicated that silicone can travel long distances along the anterior visual pathway. In 2 patients, silicone extended the length of the optic tract (16,22). Therefore, the pia may be a partial to barrier silicone movement into the subarachnoid space. Both patients had intraventricular silicone; whether it entered the subarachnoid space in the orbit or intracranially is unknown. These 2 cases also suggest a tendency for silicone to be reflected ipsilaterally at the chiasm, either following the nondecussating fibers or possibly filling space created by axonal degeneration. Boren et al: J Neuro-Ophthalmol 2016; 36: 439-447 The possibility of blockage of CSF outflow is supported by the case described by Hruby et al (19). The presence of communicating hydrocephalus would seem to imply widespread diffusion of silicone in the CSF. Further support for this theory exists in the eye, where emulsified silicone in the trabecular meshwork can cause glaucoma (68,69). One experimental animal model for obstructive hydrocephalus involves intraventricular injection of silicone into the cisterna magna and basal cisterns (70). This raises the possibility that transient "nonspecific" symptoms, such as headache, could occur as silicone moves through the ventricular system. CONCLUSION Optic nerve and intracranial complications of silicone injections may arise more frequently in the future because of an aging population and an increase in the number of indications for silicone usage (71,72). Awareness of the radiographic appearance of intracranial silicone may avoid unnecessary testing. Although the migration of silicone intracranially is rare in proportion to all ocular silicone injections, the more important issue may be the cases in which silicone remains over the long-term, especially in association with postoperative glaucoma. This patient population is poorly characterized (72) and likely the true prevalence is under-appreciated. Furthermore, it is possible that cases of progressive visual loss associated with intraneural or subarachnoid silicone are incompletely evaluated or simply attributed to glaucoma. The contribution of these processes to long-term visual outcomes in retinal detachment remains unknown, but may be significant. The case reported by Eckle et al (11) suggested that optic nerve fenestration or neurosurgical drainage may have a role in some cases of postoperative visual loss. Numerous authors have called for more aggressive management of postoperative glaucoma, imaging of patients with postoperative visual loss, and removal of silicone whenever 445 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. State-of-the-Art Review possible (3,7,10,19). It also seems prudent to consider the possibility of silicone migration during discussions with patients regarding silicone implantation and removal. The absence of preoperative glaucoma or optic pits is not protective. Clearly, much more needs to be learned about the routes of dissemination of intraocular silicone and the clinical implications of its spread along the anterior visual pathway and to the subarachnoid space and the ventricular system. Conclusions regarding the prevalence and natural history of this phenomenon are hampered by the lack of awareness, absence of a convenient, validated, in vivo diagnostic tool, and by fragmentation of care. Additional detailed case reports would be useful. Studies using OCT or other advanced imaging modalities on a larger population of patients with long-term silicone tamponade, especially those with elevated IOPs, would provide additional information regarding the prevalence and natural history of this disorder and the contribution of silicone neuropathy to visual outcomes after retinal detachment (73-75). An animal model incorporating silicone injection and induced glaucoma could yield pathologic confirmation of the theories discussed in this review (76). STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: R. Boren, C. Cloy; b. Acquisition of data: R. Boren, A. Gupta, V. Dewan; c. Analysis and interpretation of data: R. Boren, A. Gupta, V. Dewan, R. N. Hogan. Category 2: a. Drafting the manuscript: R. Boren, C. Cloy; b. Revising it for intellectual content: R. Boren, C. Cloy, A. Gupta, V. Dewan, R. N. Hogan. Category 3: a. Final approval of the completed manuscript: R. Boren, C. Cloy, A. Gupta, V. Dewan, R. N. Hogan. REFERENCES 1. Ni C, Wang W, Albert D, Schepens C. Intravitreous silicone injection. Histopathologic findings in a human eye after 12 years. Arch Ophthalmol. 1983;101:1399-1401. 2. Shields C, Eagle R. Psuedo-Schnabel's cavernous degeneration of the optic nerve secondary to intraocular silicone oil. Arch Ophthalmol. 1989;107:714-717. 3. Tang R, Carter K, Woodward J, Pardo G, Sabates F, Heravi M, Hayman L. Silicone and blindness: computed tomographic imaging. J Neuroophthalmol. 1994;14:155-156. 4. Wenkel H, Naumann G. Retrolaminar infiltration of optic nerve with intraocular tamponade following silicone oil instillation [in German]. Klin Monbl Augenheilkd. 1999;214:120-122. 5. Williams R, Beatty R, Kanal E, Weissman J. MR imaging of intraventricular silicone: case report. Radiology. 1999;212: 151-154. 6. Eller A, Friberg T, Mah F. Migration of silicone oil into the brain: a complication of intraocular silicone oil for retinal tamponade. Am J Ophthalmol. 2000;140:685-688. 7. Papp A, Toth J, Kerenyi T, Jackel M, Suveges I. Silicone oil in the subarachnoidal space-a possible route to the brain? Pathol Res Pract. 2004;200:247-252. 8. Taherian K, Bishop F, Woon W, Nelson M. Silicone oil endotamponade-is it safe? Eye. 2004;18:649-650. 9. Fangtian D, Rongping D, Lin Z, Weihong Y. Migration of intraocular silicone into the cerebral ventricles. Am J Ophthalmol. 2005;140:156-158. 10. Yu J, Apte R. A case of intravitreal silicone oil migration to the central nervous system. Retina. 2005;25:791-793. 446 11. Eckle D, Kampik A, Hintschich C, Haritoglou C, Tonn J, Uhl E, Lienemann A. Visual field defect in association with chiasmal migration of silicone oil. Br J Ophthalmol. 2005;89:918-920. 12. Kuhn F, Kover F, Szabo I, Mester V. Intracranial migration of silicon oil from an eye with optic pit. Grafes Arch Clin Exp Ophthalmol. 2006;244:1360-1362. 13. Chen J, Nidecker A, Aygun N, Gujar S, Gandhi D. Intravitreal silicone oil migration into the subarachnoid space and ventricles: a case report and review of literature. Eur J Radiol Extra. 2011;78:e81-e83. 14. Tatewaki Y, Kurihara N, Sato A, Suzuki I, Ezura M, Takahashi S. Silicone oil migrating from intraocular tamponade into the ventricles: case report with magnetic resonance imaging findings. J Compt Assist Tomogr. 2011;35:43-45. 15. Lee S, Shin K, Jo Y. Temporal hemianopsia of healthy eye in a patient with contralateral silicone oil filled eye. J Korean Ophthalmol Soc. 2011;52:633-638. 16. Jabbour P, Hanna A, Rosenwasser R. Migration of silicone oil in the cerebral intraventricular system. Neurologist. 2011;17:109-110. 17. Chang C, Chang H, Toh C. Intraventricular silicone oil. J Neurosurg. 2013;1127-1129. 18. Campbell G, Milbourne S, Salman U, Khan M. Ocular silicone oil in the lateral cerebral ventricle. J Clin Neurosci. 2013;20:1312-1313. 19. Hruby P, Poley P, Terp P, Thorell W, Margalit E. Headaches secondary to intraventricular silicone oil successfully managed with ventriculoperitoneal shunt. Retin Cases Brief Rep. 2013;7:288-290. 20. Cosgrove J, Djoukhadar I, Warren D, Jamieson S. Migration of intraocular silicone oil into the brain. Pract Neurol. 2013;13:418-419. 21. Sato S, Bathia G, Policeni B. An 87 year old woman with headaches and change in mental status. Case of the Week. Am J Neuroradiol. 2014. Available at: http://www.ajnr.org/site/ home/cow/08182014.xhtml. Accessed March 11, 2015. 22. Chiao D, Ksendzovsky A, Buell T, Sheehan J, Newman S, Wintermark M. Intraventricular migration of silicone oil: a mimic of traumatic and neoplastic etiology. J Clin Neurosci. 2015;22:1205-1207. 23. Kim H, Hong H, Park J, Lee A. Intracranial migration of intravitreal silicone oil: a case report. Clin Neuroradiol. 2015; 26:93-95. 24. Swami M, Bhootra K, Shah C, Mevada B. Intraventricular silicone oil mimicking a colloid cyst. Neurol India. 2015;63: 564-566. 25. Mathis S, Boissonnot M, Tasu J, Simonet C, Ciron J, Neau P. Intraventricular Silicone Oil-A Case Report. Vol 95. Baltimore, MD: Medicine, 2016:e2359. 26. Grzyboski A, Pieczynski J, Acaso F. Neuronal complications of intraventricular silicone oil: an updated review. Acta Ophthalmol. 2014;92:201-204. 27. Knorr H, Seltsam A, Holback L, Naumann G. Intraocular silicone oil tamponade: a clinico-pathologic study of 36 enucleated eyes. Ophthalmologe. 1996;93:130-138. 28. Kiilgaard J, Milea D, Logager V, la Cour M. Cerebral migration of silicone oil: an MRI Study. Acta Ophthalmol. 2011;89: 522-525. 29. Budde M, Cursiefen C, Holbach L, Naumann G. Silicone oilassociated optic nerve degeneration. Am J Ophthalmol. 2001; 131:392-394. 30. Biswas J, Verma A, Davda M, Ahuja S, Pushparaj V. Intra-ocular tissue migration of silicone oil after silicone oil tamponade: a histopathological study of enucleated silicone oil-filled eyes. Indian J Ophthalmol. 2008;56:425-428. 31. Wickham L, Asaria R, Alexander R, Luthert P, Charteris D. Immunopathology of intraocular silicone oil: enucleated eyes. Br J Ophthalmol. 2007;91:253-257. 32. Kanavi M, Azari A, Potter H, Lee V, Albert D. The occurrence and proposed significance of Schnabel cavernous degeneration in Uveal Melanoma. JAMA Ophthalmol. 2014;132:600-604. 33. Espinosa P, Espinosa P, Smith C, Berger J. Contralateral optic neuritis following silicone oil endotamponade for retinal detachment. J Neurol Neurophysiol. 2011;2:112. Available at: Boren et al: J Neuro-Ophthalmol 2016; 36: 439-447 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. State-of-the-Art Review 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. http://dx.doa.org/10.4172/2155-9562.1000112. Accessed November 3, 2015. Bakshi R, Caruthers S, Janardhan V, Wasay M. Intraventricular CSF pulsation artifact on fast fluid-attenuated inversionrecovery MR images: analysis of 100 consecutive normal studies. AJNR Am J Neuroradiol. 2000;21:503-508. Christensen U, la Cour M. Visual loss after use of intraocular silicone oil associated with thinning of inner retinal layers. Acta Ophthamologica. 2012;90:733-737. Shalchi Z, Mahroo O, Schunmugam M, Mohamed M, Sullivan P, Williamson T. Spectral domain optical coherence tomography findings in long-term silicone oil-related visual loss. Retina. 2015;35:555-563. Papp A, Kiss E, Timar O, Szabo E, Berecki A, Toth J, Pali J. Long-term exposure of the rabbit eye to silicone oil causes optic atrophy. Brain Res Bull. 2007;74:130-133. Barca F, Caporossi T, Rizzo S. Silicone Oil: different physical properties and clinical applications. Biomed Res Int. 2014;2014:502143. Mathews V, Elster A, Barker P, Buff B, Haller J, Greven C. Intraocular silicone oil: in vitro and in vivo MR and CT characteristics. Am J Neuroradiol. 1994;15:343-347. Ginat D. Pantopaque (Myodil, Iodophenylundecyclic acid). In: Ginat D, Small J, Schaefer P, eds. Neuroimaging Pharmacopoeia. Cham, Switzerland: Springer International Publishing, 2015:111-118. Herrick R, Hayman A, Maturi R, Diaz-Marchan P, Tang R, Lambert M. Optimal imaging protocol after intraocular silicone oil tamponade. Am J Neuroradiol. 1998;19:101-108. Leaver P, Grey R, Garner A. Silicone oil injection in the treatment of massive pre-retinal retraction. II. Late complications in 93 eyes. Br J Ophthalmol. 1979;63:361-367. Odrobina D, Laudanska-Olszewska I. Analysis of the time and location of silicone oil emulsification by spectral domain optical coherence tomography after silicone oil tamponade. Biomed Res Int. 2014;2014:372045. Mrejen S, Sato T, Fisher Y, Spaide R. Intraretinal and intraoptic nerve head silicone oil vacuoles using adaptive optics. Ophthalmic Surg Lasers Imaging Retina. 2014;45:71-73. Lampert P, Vogel M, Zimmerman L. Pathology of the optic nerve in experimental acute glaucoma: electron microscopic studies. Invest Ophthalmol Vis Sci. 1968;7:199-203. Saitoh A, Taniguchi H, Gong H, Ohira A, Amemiya T, Baba T. Long term effect on optic nerve of silicone oil tamponade in rabbits: histological and EDXA findings. Eye. 2002;16:171-176. Gong H, Ye W, Freddo T, Hernandez R. Hyaluronic acid in the normal and glaucomtous optic nerve. Exp Eye Res. 1997;64:587-595. Knox D, Kerrison J, Green W. Histopathologic studies of ischemic optic neuropathy. Trans Am Ophthalmol Soc. 2000;98:203-222. Giarelli L, Falconieri G, Cameron D, Pheley A. Schnabel cavernous degeneration: a vascular change of the aging eye. Arch Pathol Lab Med. 2003;127:314-1319. Zimmerman L, de Venecia G, Hamasaki D. Pathology of the optic nerve in experimental acute glaucoma. Invest Ophthalmol Vis Sci. 1967;6:109-125. Jonas J, Berenschtein E, Holbach L. Anatomic relationship between lamina cribosa, intraocular space, and cerebrospinal fluid space. Invest Opthalmol Vis Sci. 2003;44:5189-5195. Downs C, Roberts M, Sigal I. Glaucomatous cupping of the lamina cribrosa: a review of the evidence for active progressive remodeling as a mechanism. Exp Eye Res. 2011;93:133-140. You J, Park S, Su D, Teng C, Leibmann J, Ritch R. Focal lamina cribrosa defects associated with glaucomatous rim thinning and acquired pits. JAMA Ophthalmol. 2013;131: 314-320. Tittler E, Khaderi K, Sadun A. Optic disc pitting and associated serous macular detachment. Retin Physician. 2010; Article ID 104948. Available at: http://www.retinalphysician.com/ articleviewer.aspx?articleID=104948. Accessed November 3, 2015. Boren et al: J Neuro-Ophthalmol 2016; 36: 439-447 55. Ramezani A, Dehghan M, Rostami A, Ahmadieh H, Entazari M, Soheilian M, Azarmina M, Moradian S, Yaseri M. Outcomes of retinal detachment surgery in eyes with chorioretinal coloboma. J Ophthalmic Vis Res. 2010;5:240-245. 56. Sakamoto M, Nakamura K, Shibata M, Yokoyama K, Matsuki M, Ikeda T. Magnetic resonance imaging findings of Terson's syndrome suggesting a possible vitreous hemorrhage mechanism. Jpn J Ophthalmol. 2010;54:135-139. 57. Knecht P, Groscurth P, Ziegler U, Laeng H, Jaggi G, Killer H. Is silicone oil neuropathy caused by high intraocular pressure alone? A semi-biological model. Br J Ophthalmol. 2007;1:1293-1295. 58. Budde M, Cursiefen C, Holbach L, Naumann G. Author's reply. Am J Ophthalmol. 2002;133: 430. 59. Agrawal R, Soni M, Biswas J, Sharma T, Gopal L. Silicone oilassociated optic nerve degeneration. Am J Ophthalmol. 2002;133:429-430. 60. Lin R, DiLeonardo M, Jacoby R. Silicone lymphadenopathy: a case report and review of the literature. Am J Dermatopathol. 1993;15:82-84. 61. Killer H, Laeng H, Flammer J, Groscurth P. Architecture of the 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. 62. Killer H, Jaggi G, Flammer J, Miller N, Huber A, Mironov A. Cerebrospinal fluid dynamics between the intracranial and the subarachnoid space of the optic nerve. Is it always bidirectional? Brain. 2007;130:514-520. 63. Johnston M, Papaiconomou C. Cerebrospinal fluid transport: a lymphatic perspective. News Physiol Sci. 2002;17:227-230. 64. Yasragil M. Microneurosurgery. Vol 1. New York, NY: ThiemeStratton Inc, 1984:5-52. 65. Sharifi M, Ungier E, Ciszek B, Krajewski P. Microsurgical anatomy of the foramen of Luschka in the cerebellopontine angle, and its vascular supply. Surg Radiol Anat. 2009;31:431-437. 66. Brinker T, Stopa E, Morrison J, Klinge P. A new look at cerebrospinal fluid circulation. Fluids Barriers CNS. 2014; 11:10. Available at: http://www.fluidsbarrierscns.com/ content/11/1/10. Accessed November 3, 2015. 67. Oreskovic D, Klarica M, Commentary: a new look at cerebrospinal fluid movement. Fluids Barriers CNS. 2014;11:16. 68. Valone J, McCarthy M. Emulsified anterior chamber silicone oil and glaucoma. Ophthalmology. 1994;101:1908-1912. 69. Ichhpujani P, Jindal A, Katz L. Silicone oil induced glaucoma: a review. Graefes Arch Clin Exp Ophthalmol. 2009;247: 1585-1593. 70. Wisniewski H, Weller R, Terry R. Experimental hydrocephalus produced by the subarachnoid infusion of silicone oil. J Neurosurg. 1969;31:10-14. 71. Shah G, Ahmad B, Kiss S. Controversies in Vitreoretinal Surgery: Is Scleral Buckling an Important Mainstay in the Treatment of Retinal Detachment in 2014? Retina Today, Wayne, PA. 2014:45-50. 72. Stappler T, Morphis G, Irigoyen C, Heimann H. Is there a role for long-term silicone oil tamponade for more than twelve months in vitreoretinal surgery? Ophthalmologica. 2011;226 (suppl 1):36-41. 73. Park S, Ritch R. High resolution in vivo imaging of the lamina cribrosa. Saudi J Ophthalmol. 2011;25:363-372. 74. Ohno-Matsui K, Akiba M, Moriyama M, Ishibashi T, Tokoro T, Spaide R. Imaging retrobulbar subarachnoid space around optic nerve by swept-source optical coherence tomography in eyes with pathologic myopia. Invest Ophthalmol Vis Sci. 2011;52:9644-9650. 75. Ohno-Matsui K, Hirakata A, Inoue M, Akiba M, Ishibashi T. Evaluation of congenital optic disc pits and optic disc colobomas by swept-source optical coherence tomography. Invest Ophthalmol Vis Sci. 2013;54:7769-7778. 76. Smedowski A, Pietrucha-Dutczak M, Kaarniranta K, Lewin-Kowalik J. A rat experimental model of glaucoma incorporating rapid-onset elevation of intraocular pressure. Sci Rep. 2014;4:5910. 447 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.