Traumatic Olfactory Nerve Injury

The literature reports a wide range of percentages for traumatic olfactory nerve dysfunction. Injury to the olfactory nerve varies from 4% to 60%.² Clinical diagnosis of traumatic olfactory nerve dysfunction depends on patient age and their recognition of the dysfunction. The likelihood of traumatic olfactory dysfunction increases with age.

Self-assessment may under- or over-estimate the incidence of olfactory nerve dysfunction because of the lack of awareness of olfactory function. Only 40% of patients with olfactory dysfunction were aware of their deficit.² Most dysfunction is present immediately after injury, but delayed onset has been observed. It is thought that delay is secondary to excessive fibrous tissue around a fractured cribriform plate; there is little radiographic or histologic evidence, however, to prove this.

The probability of traumatic olfactory nerve dysfunction does not depend on the severity of the traumatic head injury, as recorded by the Glasgow coma scale (GCS).² The severity of dysfunction, however, is related to the severity of head injury. Traumatic olfactory nerve dysfunction is also related to whether the anterior skull base and/or frontal lobe are involved. Radiographic evidence of anterior skull base or frontal lobe injuries is, however, absent in 24% of cases.² Patients with skull base fractures present not only with olfactory nerve dysfunction, but can also present with rhinorrhea.³ Coup and contracoup injuries that occur with frontal and occipital trauma are the most common type of injuries producing olfactory nerve dysfunction.⁴ There are literature reports of minor head injury associated with olfactory nerve dysfunction.

Traumatic olfactory nerve dysfunction involves damage to the olfactory epithelium containing the receptor neurons in the nasal mucosa and/or mechanical stretching/shearing of unmyelinated nerve rootlets passing through the cribriform plate to form the olfactory bulb. Olfactory information is processed by the olfactory tract via the olfactory bulb to the entorhinal cortex (primary sensory areas) and the orbitofrontal cortex (secondary sensory areas). Therefore, olfactory dysfunction after traumatic head injury may not only be a result of shearing of the olfactory nerve filaments, but also by damage to frontotemporal cortical structures. Experimental animal models and human histopathologic studies in mild traumatic brain injury patients have shown axonal damage, mainly located in the frontal lobes and olfactory bulbs.⁵ Impairment of olfactory stimuli processing leads to distortion of olfactory sensation known as “dysosmia.” Patients complaining of dysosmia have abnormal regeneration of injured olfactory pathways. Recent animal studies demonstrate that recovery of the olfactory system varies with the severity of injury and that dexamethasone treatment may have therapeutic value by reducing injury-associated edema.⁶

There are a variety of tests used for olfactory nerve dysfunction. These include the Brief Smell Identification Test, testing using Sniffin’ Sticks, and chemosensory-evoked potentials. Another widely available test is the 40-odor University of Pennsylvania Smell Identification Test (UPSIT), which has been administered to over 180,000 people in Europe and North America.⁷ Hypometabolism in the orbitofrontal cortex and the medial prefrontal cortex were found in positron emission tomography (PET) studies in patients with post-traumatic anosmia.⁸ In head injury patients with anosmia or hyposmia, coexistence of olfactory bulb and tract damage and frontal encephalomalacia on magnetic resonance imaging (MRI) has also been found.⁹

Treatment of traumatic olfactory nerve dysfunction is usually conservative. Approximately one third of patients have significant recovery with conservative management.¹⁰ Unlike the other cranial nerves, the olfactory nerve contains olfactory ensheathing cells rather than Schwann cells. The olfactory pathway continuously rebuilds itself throughout life. More invasive treatment, including removal of the olfactory bulbs and tracts, is reserved for patients with severe dysosmia. Case reports demonstrated patients with severe dysosmia causing significant anorexia.

Traumatic Optic Nerve Injury

Traumatic optic nerve injuries occur in 0.5% to 5% of head injuries.¹¹ More recent surveys of craniofacial trauma data suggest an incidence of traumatic optic nerve injury in 2% to 5%.¹²

Optic nerve injury can be caused by primary and secondary mechanisms. Primary injury is by permanent axonal injury at the moment of impact from mechanical shearing, contusion, and ischemic necrosis of nerve axons. Secondary mechanisms are due to apoptosis, edema, and cell death, incorporating a variety of mechanisms leading to further axonal damage after the initial impact.^13–16

Traumatic optic nerve injury can also be divided into direct and indirect injuries. Direct injuries occur with penetrating objects into the orbit causing direct damage the optic nerve. Indirect traumatic optic nerve injury is a closed injury resulting from the transmitted impact to the optic nerve.¹⁷

Direct traumatic optic nerve injuries typically have a worse prognosis in comparison to indirect injuries.¹⁸ Direct optic nerve injury anterior to the central retinal artery entry point disturbs the retinal circulation due to the associated vascular injury. Injuries posterior to the entry point maintain normal circulation. Partial or complete optic nerve avulsion can result in a partial or complete hemorrhagic ring at the optic nerve head.^19–20 Computer modeling of direct optic nerve trauma has shown that the main site of stress is at the insertion of the nerve to the sclera and opposite to the side of impact.²¹ Nerve damage at this point leads to optic disc swelling.²² Rotation of the nerve at its insertion point to the sclera contributes the most damage in direct traumatic injury.²¹ A sudden increase in intraocular pressure is more likely to injure the optic nerve head.²¹ The intracranial segment of the optic nerve can also be injured against the falciform ligament.²³

Indirect optic nerve injury can be intracanalicular, within the optic canal, or intracranial affecting the optic nerve or chiasm. Studies utilizing laser interferometry demonstrated forces applied to the frontal bone are transferred to the optic canal.²⁴ Deceleration forces on the facial bones deform the sphenoid bone, with energy being transferred to the optic nerve.^25,26 Indirect iatrogenic injury has been reported from heat transmission during electric cautery of the posterior ethmoidal artery during treatment of epistasis, and mechanical or heat transmission injuries during transnasal endoscopic surgery. Orbital hemorrhage can cause an intraorbital compartment syndrome, a potentially reversible cause of traumatic optic neuropathy.

Injury to the optic nerve can cause significant disability. Patients typically present initially with an afferent papillary defect and visual field defect. Significant visual loss is almost always accompanied by an afferent papillary defect.²⁷ Diagnosis of traumatic optic nerve injury can be challenging and can be delayed weeks to months. This is attributed to multiple factors including severe craniofacial fractures, traumatic brain injury, and multi-system trauma requiring mechanical ventilation and sedation. Retinal examination is usually not helpful in diagnosing optic nerve injuries, as optic atrophy does not become apparent for 3 to 4 weeks.²⁸ A retinal exam is useful in the diagnosis of retinal hemorrhages and other ocular trauma.

The initial clinical assessment should document the time interval between injury and the visual examination. Delayed visual loss was reported in 10% of patients included in the International Optic Nerve Trauma Study.²⁹

In patients with altered mental status, a careful history should be obtained. Multiple facial bone fractures and facial ecchymosis imply significant facial trauma to which the clinician should suspect traumatic optic neuropathy, especially injuries caused by indirect and secondary mechanisms. The first reliable sign of a traumatic optic neuropathy is the relative afferent pupillary defect (APD). The relative afferent pupillary deficit is helpful in the diagnosis of unilateral traumatic optic nerve injury. It is elicited with the swinging flashlight test. The light shines into a normal eye stimulating the ipsilateral pupil to constrict and the contralateral pupil to constrict consensually. There is less pupillomotor stimulation reaching the brain stem when the light shines into the injured optic nerve to the uninjured side, so the pupillary response is diminished. If both optic nerves are normal, the pupillomotor stimulation remains the same with an equal pupillary constriction in both eyes. However, if the light is instead swung from the normal eye to the injured eye, the pupillomotor response from the injured eye is less and both pupils will dilate. This test is useful for detecting unilateral optic nerve injury in an unresponsive patient. A relative afferent pupillary defect is difficult to detect in bilateral optic nerve injury, and hence, is less helpful.

Visual-evoked potentials (VEP) can be useful in the diagnosis of bilateral optic nerve injury. The test can be limited by various medications and sedation. Sedation can abolish the VEP. Also, responses usually remain unchanged until deterioration beyond 20/200. Hence, the test is only diagnostic when the evoked potential is basically not reportable.

Flash visual–evoked potentials (FVEP) have been used to predict the degree of visual function in patients limited by sedation and multiple facial trauma as patient cooperation is not required to obtain reliable recordings. Utilizing the ratio of FVEP between a patient’s normal eye and injured eye in unilateral optic nerve injury has been helpful in predicting the extent of nerve injury. Utilizing the ratio helps to eliminate any confounding factors (e.g., sedation). Patients whose ratio was at least 0.5 had visual acuity of at least 20/30 in the affected eye. It is not certain whether early or late evoked potential testing makes any difference.

Patients with direct disruption of the optic nerve will have minimal, if any, recovery. Indirect or secondary injury can be treated and prevented. Treatment includes observation, medical treatment with steroids and optic canal decompression or nerve sheath fenestration. Conservative management (i.e., observation) can achieve a 40% to 60% recovery rate.²⁸ Prognostic factors have been studied. Initial visual acuity is the most influential indicator of good outcome.²⁹ Poor prognostic factors include blood in the posterior ethmoidal cells, age over 40 years, loss of consciousness, and absence of recovery within 48 hours of steroid treatment.³⁰ Blood in the posterior ethmoidal cells reflects the severity of injury, as the frontal bones transfer energy to the optic canal region.²⁴

A combination of medical and surgical treatment can be utilized. Both medical and surgical treatments remain controversial and are not without risks. High-dose steroids are thought to suppress the secondary inflammatory cascade leading to potential edema and inflammatory responses. Surgical decompression releases tension off of the nerve. Compression can be from bone fragments, hematoma, or edema.

Steroid treatment is based on the National Acute Spinal Cord Injury studies. High-dose corticosteroids were translated from acute spinal cord injury to optic nerve injury. The International Optic Nerve Trauma Study (IONTS) revealed no clear benefit for either high-dose corticosteroid therapy or surgical decompression when compared to observation alone,³¹ reporting 54% improvement after 3 months with steroid therapy, and 57% improvement with observation. Rat models have not shown histologic evidence of either improved axonal regeneration or prevention of further degeneration.³² The IONTS study reported that mega-doses of steroids within the first few hours of injury most likely provides the maximum benefit out of all treatment modalities studied. Adverse effects of steroids should be considered in light of other injuries.^30,31

Surgery should be considered with any obvious nerve compression or continual visual loss despite steroid therapy.^28,33,34 Timing of surgery is controversial. Surgical decompression has similar outcomes up to 4 months after the injury in patients who are not completely blind.³⁵

Evaluation of optic canal fractures can be visualized with high resolution computed tomography (CT) scan. Fracture of the optic canal potentially results in direct impingement on the nerve by bony fragments or associated hematoma and/or optic nerve edema (Fig. 204-1). Intrasheath hematoma should be treated with optic nerve fenestration. Perisheath hematoma can, however, be confused with an intrasheath hematoma. Optic nerve decompression has been performed by a variety of approaches, including transnasal trans-sphenoidal/transethmoidal, transpalpebral, and supraorbital transcranial. There is no statistical difference between steroids only or steroids plus surgical decompression.³⁶

FIGURE 204-1 Axial computed tomography demonstrating optic canal fracture with compression of the canal by bony elements.

Potential adverse effects of surgical decompression include injury to the globe and extraocular muscles, injury to the ethmoidal arteries resulting in orbital hematoma and further compression, cerebrospinal fluid rhinorrhea, infection, and cosmetic scarring.

Endoscopic decompressive approaches have proven to maintain or improve vision in 50% of cases, and have shown to have fewer adverse outcomes.³⁶ Endoscopic transnasal approaches are recently been favored due to the anatomic proximity of the optic canal to the sphenoid sinus, lack of external scars, preservation of olfaction, decreased morbidity, and faster recovery time. Endoscopic decompression can be considered earlier than conventional surgery, especially in patients with poor prognosis.³⁶

Oculomotor Nerve

Oculomotor nerve palsy due to closed head injury is uncommon (5% to 15%).^37–39 Injury to multiple ocular cranial nerves including oculomotor, trochlear, and abducens, can occur with post-traumatic, carotid-cavernous fistula. The oculomotor nerve exits the midbrain into the interpeduncular fossa after its rootlets traverse the red nucleus. The nerve separates into two divisions upon entering the orbit. The superior division innervates the superior rectus and levator palpebrae superioris. The inferior division innervates the medial and inferior recti and the inferior oblique muscles. Preganglionic parasympathetic innervation of the eye runs off of a branch of the nerve to the inferior oblique muscle. The complex anatomy of the oculomotor nerve results in a spectrum of presentations. In the midbrain, subnuclei within the oculomotor nuclei specifically innervate individual ocular muscles. The superior rectus muscle receives input from the contralateral subnuclei. Both levator palpebrae superioris muscles receive bilateral input.

Oculomotor nerve injury may result from distraction of the nerve, avulsion, and compression or displacement of the oculomotor nerve by traumatic intracerebral hematoma or hemorrhagic contusion.^40,41 Oculomotor nerve injury can be associated with severe traumatic head injury.³⁹ Ophthalmoplegia may result from downward displacement of the brain stem at the time of impact, and direct injury to the pupillomotor fibers on the ventromedial surface of the third nerve at the posterior petroclinoid ligament.⁴² Disturbance of the oculomotor nerve’s blood supply and biochemical changes from head injury can also contribute to the mechanism of injury.

Diagnosis of an oculomotor nerve injury can be challenging, especially in patients with orbital edema and/or ecchymosis. An adequate exam must be made, excluding any extraocular influences on ocular movement. Waiting for edema to resolve is usually required. Localization of oculomotor nerve injuries is based on the anatomic specifics of the nerve and nuclei. Involvement of the cerebral peduncle will not only present with an oculomotor nerve palsy, but also a contralateral hemiplegia and tremor. Involvement of the nuclei itself will present with not only ipsilateral oculomotor weakness (except superior rectus, which would be contralateral), but also bilateral ptosis and contralateral superior rectus weakness.

Oculomotor nerve injury is associated with a lower GCS when compared to other traumatic cranial neuropathies.⁴³ Patients were involved mainly in motor vehicle accidents and had higher number of temporal lobe abnormalities (43%) on CT and MRI. Patients with bilateral traumatic oculomotor nerve palsies or unilateral oculomotor paresis were associated with severe injuries.³ The prognosis of traumatic oculomotor palsy is poor and full recovery is uncommon. A prolonged period (up to years) of healing is usually anticipated.⁴⁴

Trochlear Nerve

The trochlear nerve is the smallest and longest of the ocular motor nerves. It runs at the free tentorial edge around the midbrain after decussating around the dorsal midbrain. Contracoup injury can occur when the nerve is compressed against the tentorium. Unilateral trochlear nerve injury can occur after frontolateral impact whereas bilateral trochlear nerve injury can occur from midfrontal impact.⁴⁵ The presence of a Horner syndrome or an afferent pupillary defect should alert the clinician to a nuclear cause of the trochlear nerve palsy. The trochlear nucleus is in close proximity to the sympathetic fibers and the pupillomotor fibers. Trochlear nerve palsies can also be challenging to diagnosis in the comatose patient. There is large variation in the amount of head trauma that translates into trochlear nerve injury.⁴

Patients report diplopia in the vertical or oblique plane. The three-step test can be used to diagnose a unilateral palsy. There is a hyperdeviation of the affected eye, hyperdeviation when looking to the contralateral side, and hyperdeviation when the head is tilted to the affected side.

Eye patches and prisms are the immediate mode of treatment. After 1 year without complete improvement, eye muscle surgery, prism use, or both is indicated.

Trigeminal Nerve

Trigeminal nerve injuries not only causes significant neurosensory deficits and facial pain, but can cause significant comorbidities due to changes in eating habits from muscular denervation of masticator muscles or altered sensation of the oral mucosa. Approximately 70% of patients with trigeminal nerve injuries complain about paresthesias.⁴⁶ Ten to fifteen percent present with neuropathic pain confirmed with nerve block testing. The development of neuropathic pain is similar in susceptibility as peripheral nerves.⁴⁷

Risk factors include direct maxillofacial trauma with facial bone fractures, injury secondary to local anesthetic injections, and during surgical intervention for facial trauma repair. There is an association between trigeminal nerve injury and facial bone fractures. Facial fractures affect the peripheral branches of the trigeminal nerve. With penetrating injuries including gunshot wounds, any portion of the trigeminal nerve can be affected. Traumatic trigeminal nerve injury associated with fractures of the upper third of the face and temporal bone is rare.⁴⁸ Fractures extending from the temporal bone to the clivus can cause injury to the trigeminal root or ganglion. Forceful impact to the posterior third of the skull may crush the petrous apex against the dorsal sella leading to injury of the trigeminal ganglion. Facial fractures directly involving the trigeminal nerve and dislocated fractures have a higher incidence of traumatic trigeminal nerve injury than nondisplaced fractures. Traumatic trigeminal nerve injury was found in 88.2% of dislocated fractures, 54.4% of nondislocated fractures, and 100% of fractures with direct nerve involvement.⁴⁹ Nondisplaced midfacial fractures had the highest incidence of trigeminal nerve impairment, but had the best prognosis out of all trigeminal nerve impairment associated with facial fractures.

Trigeminal nerve injuries are also a well known risk after oral and dental procedures, especially the inferior alveolar nerve during mandibular procedures.⁵⁰ The second most common nerve injured in dental procedures is the lingual nerve. Injuries to the long buccal nerve, greater palantine, and nasopalantine are usually clinically insignificant. Tay and Zuniga reported a higher incidence of females presenting with trigeminal nerve injuries. Lower third molar surgery was the procedure most commonly associated with trigeminal nerve injury. Patients typically presented 3 to 9 months with subjective functional problems, but minimal signs of self-injury to the hypesthetic or hyperpathic oral mucosa.

Initial evaluation begins with the neurologic exam when there are suspicions of trigeminal nerve trauma. Patients with neurosensory dysfunction of the trigeminal nerve should be evaluated serially for up to 3 months.⁵¹ Patients with unacceptable recovery at 3 months should be offered exploration and/or repair of the nerve.

If trigeminal nerve trauma is due to a fracture, open reduction with decompression is indicated unless there are other contraindications. Closed reductions of fractures continue to show deformity of the nerve and attenuation of nerve at the site of the fracture despite gentle reduction of fractured elements.⁵¹ This is due to reactive osseous proliferation causing subsequent compression of the nerve. Decompression by enlarging surrounding bone and foramina prevents nerve compression secondary to post-traumatic nerve edema and ossification of surrounding bone. Nerve repair produces significant improvement or complete recovery in 86% of patients.⁵¹ Similar results are achieved when patients undergo surgical repair up to 6 to 9 months after the initial injury.

Abducens Nerve

Abducens nerve palsy is uncommon following traumatic brain injury, with an incidence of 1% to 2.7%.⁵² Postmortem examination in severe head trauma revealed abducens nerve injury at the dural entry point (Dorello’s canal), petrous apex and the lateral wall of the ICA.⁵³ The petrous bone is the most common of the three causing stretching of the nerve. Other theories include neck hyperextension causing stretch of the nerve as it passes through Dorello’s canal. Isolated abducens nerve palsies without traumatic brain injury have been described in the literature.^54,55 There was an association with fractures of the cervical vertebrae.

Abducens nerve injury has been reported after lumbar puncture.⁵⁶ The mechanism is thought to be downward sagging of the brain after intracranial hypotension resulting in traction of the nerve. Patients usually recover from this.

The recovery rate after conservative management is high, 71%, with palsy or paresis.⁵⁷ Predictors of nonrecovery include the failure to recover by 6 months from onset, complete paresis or bilateral involvement.⁵⁸ Botulinum treatment does show significantly improved outcomes with 73% showing recovery when given within 3 months of injury.⁵⁷

Facial Nerve

Facial nerve injury usually results from blunt or penetrating trauma to the petrous portion of the temporal bone. Approximately 5% of patients who suffered head injuries have temporal bone fractures.⁶¹ Facial nerve weakness can be partial or complete; it can manifest immediately or in a delayed fashion. Late presentation is secondary to pressure effect from hemorrhage, edema, or granulation tissue.