**4. Anterior versus posterior traumatic optic neuropathy**

Traumatic optic neuropathy is the name given to the syndrome of an optic neuropathy after head or ocular trauma in the absence of other causes [40]. Like any other optic neuropathy, there are variable degrees of visual acuity and visual field loss and an afferent pupillary defect if unilateral or significantly asymmetric.

Traumatic optic neuropathy is either anterior or posterior and within each category can either be direct or indirect. Trauma to the anterior optic nerve usually injures the central retinal artery and vein, which enter or exit the nerve approximately 10 mm posterior to the globe. This vascular injury often results in retinal infarct. Hemorrhages are usually the result of severing the pial vessels with or without disk edema and rarely manifestations of central retinal or branch artery occlusion, central retinal vein occlusion, or anterior ischemic optic neuropathy. Axonal injury in the posterior optic nerve does not cause any acute effects on the disk, nerve fiber layer, or retinal ganglion cell layers. Axonal transport abnormalities posteriorly do not affect the more anterior nerve fibers, and so disk edema is not seen in posterior traumatic optic neuropathy. For these reasons, isolated posterior traumatic optic neuropathy is associated with a normal fundus examination at presentation. Only after a few weeks, we can see the structural signs of optic neuropathy evident, namely disk pallor and thinning of the retinal nerve fiber layer. A particular type of posterior traumatic optic neuropathy is when there is injury to the chiasm, in which case, there may be unilateral or bilateral temporal visual field defects respecting the vertical meridian. Rare chiasmal injury can be seen with posterior avulsion of the optic nerve, for example, traumatic enucleation, or penetration from a foreign body.

#### **4.1. Direct anterior traumatic optic neuropathy**

Direct anterior traumatic optic neuropathy is defined when there is penetration of the optic nerve by a foreign body or projectile. Anterior direct optic nerve injuries result from medial penetrating orbital trauma that damages the anterior optic nerve, for example, a knife transecting the optic nerve just posterior to the globe. This is because the optic nerve course transverses the medical part of the deep orbit and is not protected there by the bones or the eye. Posterior direct optic nerve injuries result from penetrating orbital or head trauma more posteriorly, for example, a bullet that passes just anterior to the chiasm. Direct injuries tend to produce severe and immediate visual loss, with little likelihood of recovery. The reason for this presumably is that a major element in these injuries is transection injury to retinal ganglion cell axons, which causes instantaneous loss of axonal conduction and an inability to regenerate axons later.

#### **4.2. Indirect anterior traumatic optic neuropathy**

○ Auto antibodies against aquaporin-4

46 Causes and Coping with Visual Impairment and Blindness

○ Urinary methylmalonate excretion

if unilateral or significantly asymmetric.

**4.1. Direct anterior traumatic optic neuropathy**

○ Human T-lymphotropic virus type 1 (HTLV-1) serology

**4. Anterior versus posterior traumatic optic neuropathy**

○ *Treponema pallidum* hemagglutination assay (TPHA), long-chained fatty acids

Traumatic optic neuropathy is the name given to the syndrome of an optic neuropathy after head or ocular trauma in the absence of other causes [40]. Like any other optic neuropathy, there are variable degrees of visual acuity and visual field loss and an afferent pupillary defect

Traumatic optic neuropathy is either anterior or posterior and within each category can either be direct or indirect. Trauma to the anterior optic nerve usually injures the central retinal artery and vein, which enter or exit the nerve approximately 10 mm posterior to the globe. This vascular injury often results in retinal infarct. Hemorrhages are usually the result of severing the pial vessels with or without disk edema and rarely manifestations of central retinal or branch artery occlusion, central retinal vein occlusion, or anterior ischemic optic neuropathy. Axonal injury in the posterior optic nerve does not cause any acute effects on the disk, nerve fiber layer, or retinal ganglion cell layers. Axonal transport abnormalities posteriorly do not affect the more anterior nerve fibers, and so disk edema is not seen in posterior traumatic optic neuropathy. For these reasons, isolated posterior traumatic optic neuropathy is associated with a normal fundus examination at presentation. Only after a few weeks, we can see the structural signs of optic neuropathy evident, namely disk pallor and thinning of the retinal nerve fiber layer. A particular type of posterior traumatic optic neuropathy is when there is injury to the chiasm, in which case, there may be unilateral or bilateral temporal visual field defects respecting the vertical meridian. Rare chiasmal injury can be seen with posterior avulsion of the optic nerve, for example, traumatic enucleation, or penetration from a foreign body.

Direct anterior traumatic optic neuropathy is defined when there is penetration of the optic nerve by a foreign body or projectile. Anterior direct optic nerve injuries result from medial penetrating orbital trauma that damages the anterior optic nerve, for example, a knife transecting the optic nerve just posterior to the globe. This is because the optic nerve course transverses the medical part of the deep orbit and is not protected there by the bones or the eye. Posterior direct optic nerve injuries result from penetrating orbital or head trauma more posteriorly, for example, a bullet that passes just anterior to the chiasm. Direct injuries tend to produce severe and immediate visual loss, with little likelihood of recovery. The reason

○ HIV serology

○ *Mycoplasma* serology

This is diagnosed when traumatic optic neuropathy occurs without a history of foreign body. It occurs in anterior indirect injuries, which associated with sudden rotation of the globe from blunt trauma. Examples include a digit trauma to the globe or falling and hitting the eye on the corner of a table. Anterior indirect traumatic optic neuropathy can cause partial or total avulsion of the optic nerve, with associated peripapillary hemorrhage.

#### **4.3. Posterior indirect traumatic optic neuropathy**

Posterior indirect injury is the most common cause of traumatic optic neuropathy. It results from blunt head trauma that transmits a concussive force to the optic nerve, resulting in contusion at the optic canal. There may be little or no evidence of significant head trauma; a fall from a bicycle may suffice. In other cases, there is multisystem trauma or significant brain injury. Loss of consciousness occurs in 40–72% of patients with traumatic optic neuropathy. Motor vehicle and bicycle accidents are the most frequent causes of traumatic optic neuropathy, accounting for 17–63% of cases. Traumatic optic neuropathy may be iatrogenic, especially after maxillofacial or endoscopic surgery as a result of inadvertent direct injury to the optic nerve or transmitted force fracturing the optic canal. The common site of posterior indirect optic nerve injury is at the optic canal; the intracranial optic nerve is the next most common site of injury. There may or may not be bone fractures. Despite being most common, posterior indirect traumatic optic neuropathies fortunately occasionally have the most favorable prognosis, its spontaneous visual recovery sometimes occurring at variable times after injury. Presumably, the injury causes concussion and focal blockade of axonal conduction without loss of its structural integrity. Once there is healing of the edema or other molecular events blocking conduction, axonal function can return. The severity of initial visual loss in patients with traumatic optic neuropathy varies from no light perception to 20/20, with sometimes only a visual field defect as functional evidence of disease. An afferent pupillary defect is always present and is the major clue for the diagnosis in the presence of otherwise normal eye. Patients with very poor vision (e.g., light perception only or no light perception) are less likely to improve, regardless of therapy, than patients with vision better than light perception. The reason is likely that severe injury causes axonal transection, membrane disruption, or cytoskeletal disorganization, any of which can lead to axonal dissolution and irreversible loss of conduction of visual information. In some cases, the visual loss only begins several hours to days after the injury. If this happens, the possibility of an intrasheath hemorrhage should be entertained, and neuroimaging should be repeated.

#### **4.4. Neuroimaging**

The diagnosis is radiological. It is essential in the evaluation of a patient with traumatic optic neuropathy not only for demonstrating correlative signs of injury but also detection of pre-existing structural lesions and coincident intracranial effects of trauma, e.g., hematomas or carotid cavernous fistulas. CT scanning is superior to magnetic resonance imaging (MRI) in delineating fractures of bone. It is critical that CT be performed with very thin sections that are aimed to the optic canal, and reconstructions performed, particularly in the coronal plane. About 20 to 50% of patients with posterior traumatic optic neuropathy have evidence of an optic canal fracture by neuroimaging, and sometimes, the clue is a small loss of contour of bone. Although the displacement on neuroimaging may be small, it is possible that at the time of injury, there was a much larger displacement of the bone into the canal. Even in the absence of a fracture, blood in the sphenoid sinus should raise suspicion for optic nerve injury. MRI is better for imaging soft tissue, particularly the intracranial optic nerve and chiasm, and may be useful for delineating intrasheath hemorrhage that occurs at the orbital portion from penetrating injury (anterior direct TON). It is critical that MRI only be performed after a metallic intracranial, intraorbital, or intraocular foreign body has been ruled out by CT scanning or conventional radiography. If CT is used for screening, care should be taken to use thin slices and no interslice skip.

to the treatment of spinal cord injury is not uniform [46, 47]. Furthermore, animal and cell culture data suggest that high doses of methylprednisolone may actually be toxic for the retinal ganglion cell and/or its axon [48–50]. Finally, the Corticosteroid Randomization After Significant Head Injury (CRASH) trial demonstrated that 48 hours of mega-dose methylprednisolone significantly increased the risk of death after head injury [51], with a hazard ratio at

Visual Loss in Neuro-Ophthalmology http://dx.doi.org/10.5772/intechopen.80682 49

The authors concluded that "These final results still provide clear evidence that treatment

Decompression of the optic canal is usually achieved through the transethmoidal route, most commonly via an external ethmoidectomy or endonasally [53]. The canal is then decompressed inferomedially from the superior lateral wall of the sphenoid sinus, with care taken to avoid the carotid artery. Although the canal can also be decompressed through an intracranial approach, the former is less invasive. However, if surgery in the area is being performed for other reasons necessitating unroofing of the canal, then an argument can be made that decompression of the canal should be done through this approach. However, there is also no evidence that optic canal decompression is efficacious. A recent Cochrane review concluded that there is no conclusive evidence that any particular form of surgical decompression improves the visual outcome in TON. The decision to proceed with surgery in TON remains controversial and each case needs to be assessed on its own merits. The final decision will inevitably reflect a combination of clinical judgment, the availability of local surgical expertise, and the patient's perception of the possible risks and benefits. If surgery is to be considered, it should only be performed in centers with experience with the procedure. Because of the possibility that the carotid may be iatrogenic injured, there should be informed consent regarding the risk of death or stroke. Surgery should not be performed on an unconscious patient because of the difficulty in assessing visual function. Observation of traumatic optic neuropathy may improve without any treatment. There are no convincing randomized control trials to show a treatment benefit in traumatic optic neuropathy, and a nonrandomized concurrent comparative study did not demonstrate clear differences between treatments and observation. Therefore, when a patient cannot give informed consent for corticosteroid or surgical therapy, some neuro-ophthalmologists may simply observe the patient as none of these treatments

with corticosteroids following head injury affords no material benefit."

6 months of 1.15 (95% CI 1.07–1.24) [52].

**4.7. Optic canal decompression**

have been proved to be superior.

\*Address all correspondence to: erath@netvision.net.il

2 Faculty of Medicine, Bar-Ilan University, Safad, Israel

1 Department of Ophthalmology, Galilee Medical Center, Nahariya, Israel

**Author details**

Eitan Z. Rath1,2\*

#### **4.5. Treatment of traumatic optic neuropathy**

In anterior and direct traumatic optic neuropathy, there is no evidence that treatment of anterior optic injuries or direct optic nerve injuries is efficacious. In the former, the concurrent vascular injuries cause direct ischemia and infarction to the neural retina and/or optic nerve head, and the time until irreversible neuronal death is measured in minutes to hours. In the latter, there is often sufficient direct axonal trauma to disrupt the integrity of the axon, up to and including its transection, and in the central nervous system of mammals, this is a point of no return for neuronal function. An exception is anterior traumatic optic neuropathy associated with neuroimaging evidence of an enlarged optic nerve sheath. In these cases, an optic nerve sheath fenestration should be performed in the hopes of evacuating an intrasheath hematoma.

#### **4.6. Treatment of posterior indirect traumatic optic neuropathy**

With respect to posterior indirect traumatic optic neuropathy, the three commonly used approaches that have been used are very high doses ("mega doses") of corticosteroids [41], decompression of the optic canal, and observation alone; there is insufficient evidence from good quality randomized trials to guide decision-making on how to treat traumatic optic neuropathy. Because visual function often spontaneously improves in this disease, clinical trials are particularly necessary for physicians to select therapies based on evidence. Megadose corticosteroids experimental models of white matter trauma in animals showed that doses of 15–30 milligrams per kilogram of intravenous methylprednisolone are protective for injured neurons [41]. The NASCIS 2 and 3 studies found that patients treated within 8 hours of spinal cord injury with a loading dose of 30 milligrams per kilogram of intravenous methylprednisolone load followed by 5.4 ml/kg/hr continuous infusion for 48 hours had a better outcome than control patients [42, 43]. Extrapolating these results to traumatic optic nerve injury, it was thought reasonable to believe that similar doses should be used for injury to this comparable central nervous system white matter structure. However, over the years, there has been controversy about interpretation of the NASCIS data [44, 45], and its application to the treatment of spinal cord injury is not uniform [46, 47]. Furthermore, animal and cell culture data suggest that high doses of methylprednisolone may actually be toxic for the retinal ganglion cell and/or its axon [48–50]. Finally, the Corticosteroid Randomization After Significant Head Injury (CRASH) trial demonstrated that 48 hours of mega-dose methylprednisolone significantly increased the risk of death after head injury [51], with a hazard ratio at 6 months of 1.15 (95% CI 1.07–1.24) [52].

The authors concluded that "These final results still provide clear evidence that treatment with corticosteroids following head injury affords no material benefit."

#### **4.7. Optic canal decompression**

structural lesions and coincident intracranial effects of trauma, e.g., hematomas or carotid cavernous fistulas. CT scanning is superior to magnetic resonance imaging (MRI) in delineating fractures of bone. It is critical that CT be performed with very thin sections that are aimed to the optic canal, and reconstructions performed, particularly in the coronal plane. About 20 to 50% of patients with posterior traumatic optic neuropathy have evidence of an optic canal fracture by neuroimaging, and sometimes, the clue is a small loss of contour of bone. Although the displacement on neuroimaging may be small, it is possible that at the time of injury, there was a much larger displacement of the bone into the canal. Even in the absence of a fracture, blood in the sphenoid sinus should raise suspicion for optic nerve injury. MRI is better for imaging soft tissue, particularly the intracranial optic nerve and chiasm, and may be useful for delineating intrasheath hemorrhage that occurs at the orbital portion from penetrating injury (anterior direct TON). It is critical that MRI only be performed after a metallic intracranial, intraorbital, or intraocular foreign body has been ruled out by CT scanning or conventional radiography. If CT

is used for screening, care should be taken to use thin slices and no interslice skip.

**4.6. Treatment of posterior indirect traumatic optic neuropathy**

In anterior and direct traumatic optic neuropathy, there is no evidence that treatment of anterior optic injuries or direct optic nerve injuries is efficacious. In the former, the concurrent vascular injuries cause direct ischemia and infarction to the neural retina and/or optic nerve head, and the time until irreversible neuronal death is measured in minutes to hours. In the latter, there is often sufficient direct axonal trauma to disrupt the integrity of the axon, up to and including its transection, and in the central nervous system of mammals, this is a point of no return for neuronal function. An exception is anterior traumatic optic neuropathy associated with neuroimaging evidence of an enlarged optic nerve sheath. In these cases, an optic nerve sheath fenestration should be performed in the hopes of evacuating an intrasheath

With respect to posterior indirect traumatic optic neuropathy, the three commonly used approaches that have been used are very high doses ("mega doses") of corticosteroids [41], decompression of the optic canal, and observation alone; there is insufficient evidence from good quality randomized trials to guide decision-making on how to treat traumatic optic neuropathy. Because visual function often spontaneously improves in this disease, clinical trials are particularly necessary for physicians to select therapies based on evidence. Megadose corticosteroids experimental models of white matter trauma in animals showed that doses of 15–30 milligrams per kilogram of intravenous methylprednisolone are protective for injured neurons [41]. The NASCIS 2 and 3 studies found that patients treated within 8 hours of spinal cord injury with a loading dose of 30 milligrams per kilogram of intravenous methylprednisolone load followed by 5.4 ml/kg/hr continuous infusion for 48 hours had a better outcome than control patients [42, 43]. Extrapolating these results to traumatic optic nerve injury, it was thought reasonable to believe that similar doses should be used for injury to this comparable central nervous system white matter structure. However, over the years, there has been controversy about interpretation of the NASCIS data [44, 45], and its application

**4.5. Treatment of traumatic optic neuropathy**

48 Causes and Coping with Visual Impairment and Blindness

hematoma.

Decompression of the optic canal is usually achieved through the transethmoidal route, most commonly via an external ethmoidectomy or endonasally [53]. The canal is then decompressed inferomedially from the superior lateral wall of the sphenoid sinus, with care taken to avoid the carotid artery. Although the canal can also be decompressed through an intracranial approach, the former is less invasive. However, if surgery in the area is being performed for other reasons necessitating unroofing of the canal, then an argument can be made that decompression of the canal should be done through this approach. However, there is also no evidence that optic canal decompression is efficacious. A recent Cochrane review concluded that there is no conclusive evidence that any particular form of surgical decompression improves the visual outcome in TON. The decision to proceed with surgery in TON remains controversial and each case needs to be assessed on its own merits. The final decision will inevitably reflect a combination of clinical judgment, the availability of local surgical expertise, and the patient's perception of the possible risks and benefits. If surgery is to be considered, it should only be performed in centers with experience with the procedure. Because of the possibility that the carotid may be iatrogenic injured, there should be informed consent regarding the risk of death or stroke. Surgery should not be performed on an unconscious patient because of the difficulty in assessing visual function. Observation of traumatic optic neuropathy may improve without any treatment. There are no convincing randomized control trials to show a treatment benefit in traumatic optic neuropathy, and a nonrandomized concurrent comparative study did not demonstrate clear differences between treatments and observation. Therefore, when a patient cannot give informed consent for corticosteroid or surgical therapy, some neuro-ophthalmologists may simply observe the patient as none of these treatments have been proved to be superior.
