A case of back surgery and ensuing visual loss

Recognizing which vascular bed is compromised is the key to successful intervention.

The patient is a 60-year-old African-American man with a history of cervical myelopathy (C3-C4) from severe arthritis who underwent a posterior cervical fusion with lateral mass screws. His past medical history is significant for asthma and osteoarthritis. Past surgical history is significant for two lower back surgeries in 2001 and 2002. His home medications include Proventil and Naprosyn. He has no known drug allergies. Intraoperatively, he was placed facedown with his head supported by steel blunt pins and a halo support. At the end of the surgical procedure, a pin was found compressing his left orbit, and a laceration was noted over the left brow. The surgery lasted approximately 2.5 hours. Upon awakening from anesthesia, he complained of decreased vision and pain in the left eye.

When seen by the ophthalmology service in consultation, the patient was still quite drowsy. His external exam was significant for a superficial laceration above the left brow with sutures intact. He had mild preseptal swelling of his left upper lid with mild proptosis and some increased resistance to retropulsion. His vision at near without correction was 20/70 in the right eye and bare light perception in the left eye. His pupils were 4 mm and minimally reactive in the right eye and 4.5 mm and fixed in the left eye. He demonstrated full ductions in the right eye but a seemingly frozen globe in the left eye. IOPs were 14 mm Hg and 16 mm Hg by Tonopen. His anterior exam was otherwise not remarkable. Fundoscopic exam revealed flat discs, cup to disc ratio of 0.4 in both eyes and discs that were normal in color, capillarity and contour. The macula, vessels and periphery were otherwise normal in both eyes.

MRI of the orbits later that evening revealed significant diffuse edema of the retrobulbar fat with a strand of retrobulbar heme. There was no evidence of intrasheath hemorrhage of the optic nerve or compression of the nerve within the canal. MRA and MRV were within normal limits.

What is the differential diagnosis of this case at its onset, when first seen by the ophthalmology service before imaging?

Based on the brow laceration and clinical findings, the patient could have had a prolonged direct compressive force applied to his globe, resulting in an apparent insult to both the optic nerve and the extraocular muscles, along with some evidence of a compartment syndrome. As such, our differential started with a lesion compromising blood flow and neural function. Specifically, our concern was for retrobulbar hemorrhage or edema.

The external brow laceration could have extended deeply enough to cause further bleeding behind the globe. Retrobulbar hemorrhage is a well-known complication of eyelid and periorbital surgery. In this patient, the compressive force of the malpositioned blunt pin could also have effects within the orbit even if it never directly pressed on the globe. Anderson and colleagues studied the effects of a static load at the supraorbital ridge. Transmission of force from this ridge into the orbital roof could cause deformation or fracture with resultant orbital hemorrhage and edema. This hemorrhage or edema in the bony retrobulbar compartment would compress the vascular supply to the optic nerve and compromise the vascular and neural supply to the extraocular muscles.

Although there is a dual blood supply to most ocular tissues, there are cases of vascular compromise causing visual loss and ophthalmoplegia. Hollenhorst and colleagues described perioperative orbital infarction in 1954. They hypothesized that intraoperative arterial and venous compression can lead to an orbital congestion and ischemia. Van Stavern and Gorman described a case of orbital infarction syndrome (ischemia of all intraocular and orbital structures) after prone orbital compression in an unconscious, facedown 36-year-old cocaine user. It is unlikely that ophthalmic artery occlusion in isolation could cause both optic nerve dysfunction and ophthalmoplegia due to the collateral supply from the external carotid artery. However, it is reasonable to assume that in some patients collateral supply could be compromised from direct compression, vasoconstrictive agents, intraoperative blood loss, arrhythmia or anomalous anastomoses.

Also of concern to us would be intraorbital pressure from edema and/or blood accompanied by another insult such as globe rupture, orbital fracture, optic nerve sheath hematoma, periosteal hematoma, central retinal artery occlusion, central retinal vein occlusion or acute glaucoma.

The presence of possible multiple cranial nerve pathologies also brings into question the status of the cavernous sinus. Ophthalmoplegia due to third, fourth and sixth nerve dysfunction is common. Optic nerve involvement due to anterior extension is also possible. Therefore, cavernous sinus pathology, especially intraoperative thrombosis or ruptured aneurysm and cavernous carotid fistula need to be investigated with MRI.

A stroke of the midbrain causing all of these symptoms may be hypothetically possible. However, given the unilateral involvement and lack of further neurological sequelae, this becomes much less likely.

Table 1: Fifty Reported cases of prone position spinal procedure resulting in visual deficits Source: Katz B.

How would one narrow the differential diagnosis and come to a working hypothesis?

In general, this patient would need a full ophthalmologic exam including assessment of skin and corneal sensation, visual acuity, pupillary responses, resistance to retropulsion, Hertel exophthalmometry, IOP, extraocular motility, forced ductions, anterior and dilated fundoscopic exam.

Of particular importance are assessments for proptosis and resistance to retropulsion. In a case such as this, IOP may be misleading. It is not necessarily a reflection of retrobulbar pathology. In fact, it may be normal even with significant retrobulbar pathology. This is particularly true in cases where there may be compromise of the ciliary body. The presence of resistance to retropulsion, even in the face of normal IOP, should point to some space occupying orbital lesion.

Narrowing the hypothesis would also be aided by a review of the surgical chart assessing for intraoperative hypotension, arrhythmia or blood loss, which could be contributory to further decreased perfusion pressure to ocular tissues.

If the patient is judged to be stable enough, then a CT scan of the head and orbit would provide additional diagnostic information regarding presence of blood, edema and/or fracture. Again in this case, even if gross blood or dramatic edema is not present on imaging, this does not rule out a retrobulbar compressive pathology. A clinical exam that indicated an orbital compartment syndrome should dictate therapy.

An MRI of the brain and orbits is indicated to evaluate for optic nerve sheath hematoma, retrobulbar edema or stroke. This is particularly important as the presence of an optic nerve sheath hematoma is a surgical emergency and a potentially reversible cause of vision loss. MRA and MRV could demonstrate aneurysm, thrombosis or fistula in the cavernous sinus.

Repeat examinations would need to be done when the patient is fully awake and serially thereafter. Follow-up imaging may also be indicated if there is continuance or worsening of symptoms.

What are the possible appropriate therapeutic maneuvers to perform, on what time frame should they be done, and what is the rationale for each?

In this case, where there is a high suspicion of postoperative orbital space-occupying pathology, the priority must be placed on decompression in the hope of salvaging as much visual function as possible. Even immediately postoperatively, there may already have been severe damage from prolonged intraoperative compression. Treatment of IOP is an adjunct only, as this is not the primary mechanism of insult that is causing visual loss. Medical and surgical orbital by lateral canthotomy/cantholysis are the mainstays of successful therapy here. Decompression has the potential to reverse vision loss, especially when interventions are begun promptly. Furthermore, Borruat and colleagues presented evidence that reversal of an ischemic insult could aid in the resolution of ophthalmoplegia. A study by Cook of traumatic optic neuropathies demonstrated some treatment (steroids, surgical canal decompression or both) to be superior to no treatment. There may be some pathophysiologic similarities between a traumatic optic neuropathy and the damage seen in this patient.

Lateral canthotomy/cantholysis: This is classically indicated in patients with resistance to retropulsion, increased IOP, decreased visual acuity and afferent pupillary defect. Patients that do not meet all of these criteria are also candidates if clinical suspicion of pressure-related symptomatology remains high. Neural tissue is very sensitive to hypoxia, and therefore this maneuver should be performed immediately when there is evidence of a retrobulbar process that has compromised visual acuity. Return of vision to a totally blind eye as a result of retrobulbar hemorrhage has been shown after surgical decompression 4 hours from insult.

High-dose corticosteroids: Efficacy of use is largely based on studies of antioxidant and anti-inflammatory effects of high-dose steroids status post spinal cord injury. They are thought to limit both edema and free radical formation. Spoor and colleagues postulated that corticosteroids decrease intraneural or extraneural edema and thereby increase blood flow. These are typically started immediately and continued for at least the first 48 hours after reduced visual acuity. This can be combined with the use of IV mannitol.

Paracentesis: Paracentesis is often thought to be the first required intervention; this is a mistaken assessment, as it is not compromise of the retinal circulation that is causing the visual loss. Paracentesis would be efficacious in cases of high IOP. Although there is evidence of correlation between elevated IOPs and anterior ischemic optic neuropathy, this patient’s visual loss is caused by compromise to vascular supply within the orbit, not the globe itself. Optimizing blood flow to ocular tissue is the objective, and if a marginal increase in intraorbital blood flow can be achieved acutely by lowering IOPs, then the action has merit. Elevation of the bed may also be worthwhile.

Overall, it should be reiterated that in cases where the clinical exam demonstrates a clinically significant orbital compartment syndrome, decompression and thereby reinstitution of blood flow is the primary goal.

What is the importance and interpretation of normal IOP in this setting?

IOP is not a reflection of the intraorbital pressure. Rather, as Katz and colleagues demonstrated in their study of retrobulbar hemorrhage patients, there can be complete visual impairment exclusively due to the elevation of pressure in an enclosed orbital compartment.

During extended prone positioning of surgery, IOP is increased significantly (40 mm Hg intraoperatively). Further, in this patient it is likely that IOP may have been much higher if the pin was directly compressing the globe. The effect of the high intraoperative IOP and resultant vascular congestion may compromise the production of aqueous for a period of time. This would be worsened if there were intraoperative complications such as hypotension, arrhythmia or blood loss anemia. Postoperatively, IOP could be within normal limits (pt supine, removal of offending pin, ciliary body compromise); however, significant damage could already have been done. Posttraumatic orbital edema could continue to accumulate postoperatively and further compromise the pial vessels, the short posterior ciliary arteries or the intra-canalicular nerve in the face of normal IOPs.

What does the literature have to say about cases such as this?

Overall, perioperative vision loss is a rare occurrence (approximately 1 in 61,000), but it happens more commonly (perhaps up to 50 times) with spinal procedures (1 in 1,100). Review of the orthopedic and anesthesiology literature reveals several case reports and surveys. It seems that the etiology of postoperative vision loss is typically multifactorial, with patient positioning, direct trauma, coagulopathies, embolism, co-morbidities, hypotension, arrhythmia and blood loss all contributing. In this case, it seems that patient positioning and direct compression was the major problem. The ophthalmoplegia and optic nerve dys- function can best be explained by widespread ocular tissue ischemia stemming from a prolonged compressive insult, with the resultant edema causing an orbital compartment syndrome.

Literature describing a patient who is unilaterally blind and ophthalmoplegic after a prone surgical procedure is scant. A survey of the literature did not reveal a patient who perioperatively had visual loss and a frozen globe that was thought to be due primarily to a compressive etiology for approximately 50 years. Most of this can be attributed to better intraoperative position monitoring; however, a portion of this could be due to disincentives to reporting these events. Table 1 summarizes 50 reported cases of prone position spinal procedures resulting in visual deficits.

As seen in the table, limited mobility was reported in five of the cases, all of which had some degree of proptosis. Proptosis was either absent or not mentioned in the remaining cases. There is not enough data here, however, to define a relationship.

Bhatti and colleagues described a patient with light perception vision and ophthalmoplegia after a shoulder procedure. However, the patient was in a sitting position with no evidence of compressive insult.

The rarity of prolonged orbital compression with resulting blindness and ophthalmoplegia has precluded a large study of therapeutics and outcomes. However, given the known sensitivity of neural tissue to hypoxia, treatment of patients that display symptoms and/or exam findings consistent with a compartment syndrome should lead the treating team to actively maneuver to decrease intraorbital pressure in hopes of salvaging some function. Relying on IOP or imaging modalities alone to dictate treatment could result in costly delays in orbital decompression, thereby increasing morbidity.

For Your Information:

• Tarek Persaud, MD, can be reached at the department of ophthalmology of the George Washington University, 2150 Pennsylvania Ave. NW, Washington, DC 20037; 202-741-2825; fax: 202-741-2821; e-mail: tarekpersaud@hotmail.com.

References:

• Basic and Clinical Science Course: Orbits, Eyelids and Lacrimal System. San Francisco: American Academy of Ophthalmology; 2003-2004:(7)228-229.
• Anderson RL, Panje WR, Gross CE. Optic nerve blindness following blunt forehead trauma. Ophthalmology. May 1982;89(5):445-455.
• Basic and Clinical Science Course: Neuro-Ophthalmology. San Francisco: American Academy of Ophthalmology; 2003-2004:(5)245.
• Hollenhorst RW, Suien HJ, Benoit CF. Unilateral blindness occurring during anesthesia for neurosurgical procedures. Arch Ophthalmol. 1954;52:819-830.
• Van Stavern GP, Gorman M. Orbital infarction after cocaine use. Neurology. August 27, 2002;59(4):642-643.
• Wang BH, Robertson BC, Girotto JA, et al. Traumatic optic neuropathy: a review of 61 patients. Plast Reconstr Surg. June 2001;107(7):1655-1664.
• Katz B, Herschler J, Brick DC. Orbital haemorrhage and prolonged blindness: a treatable posterior optic neuropathy. Br J Ophthalmol. August 1983;67(8):549-553.
• Borruat FX, Bogousslavsky J, Uffer S, Klainguti G, Schatz NJ. Orbital infarction syndrome. Ophthalmology. April 1993;100(4):562-568.
• Cook MW, Levin LA, Joseph MP, Pinczower, EF. Traumatic optic neuropathy: a meta-analysis. Arch Otolaryngol Head Neck Surg. 1996;122:389.
• Spoor TC, Hartel WC, Lensink DB, Wilkinson MJ. Treatment of traumatic optic neuropathy with corticosteroids. Am J Ophthalmol. December 1990;100:665-669.
• Williams EL, Hart WM, Tempelhoff R. Postoperative ischemic optic neuropathy. Anesth Analg. 1995;80:1018-1029.
• Cheng MA et al. The effect of prone positioning on intraocular pressure in anesthetized patients. Anesthesiology. December 2001;85(6);1351-1355.
• Roth S, Thisted RA, Erickson JP, Balck S, Schreider BD. Eye injuries after nonocular surgery: a study of 60,965 anesthetics from 1988 to 1992. Anesthesiology. 1996;85:1020-1027.
• Roth S, Barach P. Postoperative visual loss: still no answers — yet. Anesthesiology. September 2001;95(3).
• Lee LA, Lam AM. Unilateral blindness after prone lumbar spine surgery. Anesthesiology. September 2001;95(3):793-795.
• Stevens, WR, Glazer PA, Kelley SD, Lietman TM, Bradford DS. Ophthalmic complications after spinal surgery. Spine. 1997;22(12):1319-1324.
• Cheng MA, Sigurdson W, Tempelhoff R, Lauryssen C. Visual loss after spine surgery: a survey. Neurosurgery. March 2000;46(3):625-631.
• Myers MA, Hamilton SR, Bogosian AJ, Smith CH, Wagner TA. Visual loss as a complication of spine surgery. A review of 37 cases. Spine. 1997;22(12):1325-1329.
• Dilger JA, Tetzlaff JE, Bell GR, Kosmorsky GS, Agnor RC, O‘Hara Jr JF. Ischaemic optic neuropathy after spinal fusion. Can J Anaesth. 1998;45(1):63-66.
• Hoski JJ, Eismont FJ, Green BA. Blindness as a complication of intraoperative positioning. J Bone Joint Surg. August 1993;75-A(8):1231-1232.
• Grossman W, Ward WT. Central retinal artery occlusion after scoliosis surgery with a horseshoe headrest. Spine. 1993;18(9):1226-1228.
• Wolfe SW, Lospinuso MF, Burke SW. Unilateral blindness and patient positioning for spinal surgery. Spine. 1992;17(5):600-604.
• Warner ME, Warner MA, Garrity JA, MacKenzie RA, Warner DO. The frequency of perioperative vision loss. Anesth Analg. 2001;93:1417-1421.
• Bhatti TM, Enneking FK. Visual loss and ophthalmoplegia after shoulder surgery. Anesth Analg. March 2003;96:899-902.