Traumatic optic neuropathy (TON) describes an injury to the optic nerve following either blunt or penetrating trauma. Clinical features of TON include vision loss, dyschromatopsia, visual field defects and a relative afferent pupil defect (except in the case of a symmetric bilateral TON). TON can be classified as either indirect or direct. Indirect TON results from trauma to the head or face causing energy from the force of impact to be transmitted to the bony structures that convey the optic nerve. Sheering forces act to compromise either the nerve itself or the pial vascular. Direct TON describes damage caused by mechanical stresses applied directly to the nerve from avulsion or laceration or impingement of the nerve from various causes, including penetrating foreign body, displaced fracture fragment or optic canal fracture.
TON remains an important cause of vision loss following traumatic head injury. Little consensus exists among neuro-ophthalmologists regarding the appropriate treatment for TON. Various management strategies have been investigated, including observation, corticosteroid treatment, surgical decompression of the optic canal and combinations of steroids and surgical decompression. In addition, various novel neuroprotective therapies have been proposed. To date, no study has conclusively shown the superiority of any single treatment with regard to favourable visual outcomes. Low incidence, variable treatment algorithms, variable corticosteroid dosage ranges and a natural history that may result in spontaneous recovery without intervention compound the difficulty in achieving consensus among practitioners. This study aims to review recent literature pertaining to the treatment of TON and to discuss current controversies surrounding the treatment of TON.
LACK OF CONSENSUS
In 1999, Levin et al.  reported the outcomes of a nonrandomized, unmasked, comparative interventional study, the International Optic Nerve Trauma Study (IONTS), which compared visual outcomes of patients with indirect TON who were treated within 7 days of initial injury with either steroids (at doses ranging from 60 mg to greater than 5400 mg) or surgical decompression (often in conjunction with corticosteroids) versus patients who were simply observed. No statistically significant difference in final visual acuity was observed among the three groups. The authors conclude that there was no significant benefit to either surgical decompression or corticosteroid treatment, when compared with observation alone. With evidence lacking to support a standard-of-care, treatment for TON has largely followed the preferred practice patterns of clinicians. With estimates of spontaneous visual recovery following TON ranging from 40 to 60% [2,3][2,3] some authors have advocated no treatment .
The use of steroids in the treatment of TON remains controversial. With a paucity of Class I evidence, clinicians have been forced to glean insights largely from retrospective, observational studies. Various steroid regimens, dose ranges and treatment durations have been proposed. Methylprednisolone dosages ranging from low dose (60 mg daily) to megadose (greater than 5400 mg daily) have been suggested. The most recent Cochrane review from 2013  includes only one randomized control trial published in 2007 of 31 patients with indirect TON randomized to receive either high-dose intravenous steroids or placebo within 7 days of TON diagnosis . This study found no statistically significant difference in mean final logMAR visual acuity at 3 months between the high-dose corticosteroid group and placebo. On the basis of this and our review of the literature, we did not find convincing evidence that steroids represent any additional benefit over observation alone. Furthermore, the side effect profile of high-dose corticosteroids is far from benign and includes gastrointestinal ulcers, impaired glucose metabolism, acute psychosis, impaired healing and increased risk of infection and sepsis at high doses [6,7][6,7].
There also appears to be mounting evidence that steroids used in the setting of traumatic brain or spinal cord injuries may have negative effects on patient survival and neurologic outcomes. The Corticosteroid Randomization After Significant Head Injury (CRASH) Study from 2004 found an increased risk of death within 14 days in patients with traumatic brain injury (TBI) randomized to receive high-dose steroids . Data from the National Acute Spinal Cord Injury Study (NASCIS) suggest that steroid treatment initiated more than 8 h after initial injury was associated with significantly poorer neurologic outcomes than placebo . As TON often occurs in the context of concomitant head or multisystem trauma, the observations of these studies should give pause to the ophthalmologist deciding to treat TON patients with corticosteroids.
Some authors still advocate surgical decompression of the optic canal and optic nerve as a preferred treatment for all TON. It seems that the preference for surgery is derived largely from the preferred practice patterns of individual clinicians or institutions rather than from robust clinical evidence. The support for partial optic canal decompression is derived from numerous, mostly small, retrospective case series in the literature. These case series are subject to inherent biases and inconsistencies that make both the authors’ conclusions and comparison among series difficult to interpret.
For example, there are at least four surgical techniques described for optic canal decompression (transcranial approaches and various extracranial approaches, including transethmoidal, endonasal and sublabial). One can find descriptions of the merits of each of these approaches in the literature [10–12][10–12][10–12]. These small, nonrandomized, retrospective case series are subject to both observer and selection bias. The lack of treatment randomization results in a tendency towards offering surgery to those who either have failed conservative therapy or presented initially with severe vision loss. Furthermore, the lack of a control group makes it difficult to determine whether or not surgical intervention – and the risks associated with it – offers any benefit over observation alone. We found no recent randomized controlled trials investigating surgery for TON and no conclusive evidence exists that any type of surgical decompression improves visual outcomes in either direct or indirect TON. Even in cases of optic canal fracture, with the possibility of a bone fragment impinging directly on the optic nerve, current evidence suggests that the potential for visual recovery may be limited regardless of surgical or medical intervention [13,14][13,14]. Potentially serious surgical complications associated with optic canal decompression surgery include postoperative CSF leak and accidental dural exposure, and the risk does not appear to outweigh the benefit for surgical decompression for TON.
COMBINATION OF CORTICOSTEROIDS AND SURGICAL DECOMPRESSION
Although the IONTS remains the largest, multicentre, prospective study of TON to date, its conclusions regarding steroid treatment versus surgical decompression arms are complicated by the fact that, in the surgery group, 32 of 33 patients received corticosteroids in addition to surgery.
Ropposch et al.  recently reported on their experience in a retrospective cohort of 42 patients with direct or indirect TON treated with optic nerve decompression surgery. In this cohort, 21 patients (50%) were treated with steroids in addition to optic nerve decompression, 15 patients (36%) did not receive steroids and in six patients (14%), there was inadequate documentation to determine whether or not steroids had been used. Again, variable steroid doses were included with nine of 21 patients receiving a high-dose corticosteroid regimen corresponding to an initial dose of greater than 500 mg of methylprednisolone, while 12 of 21 patients received a lower dosage. Comparing visual acuity outcomes at a follow-up interval ranging from 2 to 12 months after decompression surgery, the authors found that just 29% of patients treated with the combination of decompression surgery and steroids experienced visual improvement, compared with 53% of patients treated with decompression surgery alone.
The authors conclude that, ‘additional steroids had no beneficial effect on visual outcomes’. This study suffers from biases inherent to the retrospective nature of the study (observer bias, and so on), small sample size (n = 42) and significant loss to follow-up (17% or seven of 42 patients). Nevertheless, this study provides a context for understanding the role of corticosteroids in conjunction with optic nerve decompression surgery. It should be noted that this study included cases of both direct and indirect TON, whereas the IONTS included only cases of nonpenetrating, indirect TON.
DIMINISHED FUNCTIONAL OPTIC NERVE RESERVE FOLLOWING TRAUMA
Guy et al.[16▪] reported their experience with second optic nerve injuries following initial visual recovery from TON. This retrospective case series studied 131 patients treated for direct and indirect TON within an 18-year period at a single oculoplastic surgery tertiary referral centre. Twelve patients in this series were treated for a second, ipsilateral TON with the mean length of time between initial and second injury being 7 years. The mean loss of visual acuity following second nerve injury was 5.6 lines on the Snellen chart. Ten of twelve patients (83.3%) received optic canal decompression in addition to high-dose corticosteroids, while two were treated with high-dose corticosteroids only (methylprednisolone 1.5 mg/kg, followed by 5.4 mg/kg/h for up to 72 h). Two of 12 patients (16.7%) treated with corticosteroids experienced no improvement in visual acuity, while two patients (16.7%) treated with the combination of optic canal decompression and high-dose steroids experienced ‘minimal’ vision improvement of one line of Snellen visual acuity. Two patients treated with combination surgery and steroids (16.7%) experienced improvement of two lines of Snellen acuity.
A statistically significant improvement in visual acuity was noted following treatment for the initial TON (P = 0.004). However, the presenting visual acuity following second optic nerve injury tended to be at or below the acuity noted after the first injury. No substantive improvement was observed following treatment for the second insult (P = 0.05). Treatments for the initial TON were variable among patients with second ipsilateral TON and included limited surgical decompression of the optic canal in one of 12 patients (8.3%), methylprednisolone per the NASCI-II study  in five of 12 patients (41.7%), other corticosteroid regimen in four of 12 patients (33.3%) or unknown treatment in three of 12 (25.0%). This study is subject to biases inherent to retrospective analyses and suffers from potential referral bias, as a relatively large proportion of patients presenting with primary TON underwent surgical decompression. Of 115 patients presenting with primary TON, 91 patients (79.1%) underwent surgical decompression, while 24 patients who ‘declined surgery’ were treated with corticosteroids alone. The authors observed an improvement in visual acuity in 75 patients (82.4%) who underwent decompression surgery, in contrast to 10 patients (41.6%) who were treated with corticosteroids alone.
This series highlights important considerations for clinicians to consider in the treatment of TON. It suggests that regardless of initial treatment, there may be changes to the optic nerve following initial trauma that render it more susceptible to second nerve injury and more recalcitrant to recovery from subsequent insults. One might infer from this study that neither corticosteroid treatment nor surgical decompression convey any meaningful, durable neuroprotective effects.
NOVEL NEUROPROTECTIVE STRATEGIES
Historically, treatment strategies for TON have focused largely on minimizing mechanical damage to the optic nerve, either by surgically decompressing the bony canal surrounding the nerve or by using corticosteroids in an attempt to limit inflammation and reduce swelling. Increasing attention has shifted recently towards the use of novel neuroprotective therapies in TON, and initial reports are intriguing.
A pilot study investigating the effect of intravenous erythropoietin on visual improvement in patients with TON recently published by Entezari et al.[17▪] describes preliminary data from a single arm of a multicentre, randomized clinical trial comparing different treatments for TON. The pilot study consists of a case series of 18 eyes of 18 patients diagnosed with indirect TON. Patients were treated with three doses of 20 000 IU of intravenous erythropoietin (EPO) daily for 3 days within 14 days of the initial trauma. This study included only cases of indirect TON, and all study participants underwent neurosurgical evaluation and were excluded if an indication for surgical decompression was found. Visual function was assessed at baseline and at 1 and 3 months following treatment. The authors found a statistically significant improvement in LogMAR visual acuity compared with baseline logMAR acuity at both 1 and 3 months following treatment with EPO. Baseline mean LogMAR acuity improved from 2.21 ± 0.97 to a mean of 1.48 ± 1.29 at 1 month (P = 0.001) and to a mean of 1.31 ± 1.27 at 3 months (P < 0.001). A trend towards improved colour vision in patients treated with EPO at both 1 and 3 months following treatment was also observed. However, this trend did not achieve statistical significance. The lack of both randomization and a relevant control group makes it difficult to interpret this study's conclusions, particularly in the context of a disease that has a relatively high rate of spontaneous improvement without intervention.
Although the prospect of a novel neuroprotective therapy is exciting, the hope of a prospective, multicentre, randomized, placebo-controlled study of various treatment approaches for TON may prove even more enticing to the neuro-ophthalmic community as such little Class I evidence currently exists to guide treatment decisions.
Vision loss from TON may be profound and permanent. There is currently little evidence from randomized controlled trials or other methodologically rigorous studies to help clinicians provide treatment recommendations. Publications demonstrating results from small, retrospective case series are subject to inherent biases that limit the extrapolation of study findings to the general population and call into question the validity of study conclusions. There is a high spontaneous improvement rate in TON. In addition, there is no current reliable evidence that the two most commonly employed treatment strategies – corticosteroids or surgical optic canal decompression – offer any visual benefit over observation alone. At the present time, when we see patients with indirect TON, we do not recommend corticosteroids and optic canal decompression, as they incur additional risk without proven benefit.
Financial support and sponsorship
This work was supported by an Unrestricted Grant from Research to Prevent Blindness, New York, New York, USA.
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
1. Levin LA, Beck RW, Joseph MP, et al. The treatment
of traumatic optic neuropathy
: the International Optic Nerve Trauma Study. Ophthalmology 1999; 106:1268–1277.
2. Chou PI, Sadun AA, Chen YC, et al. Clinical experiences in the management of traumatic optic neuropathy
. Neuroophthalmol 1996; 16:325–336.
3. Yu-Wai-Man P, Griffiths PG. Steroids for traumatic optic neuropathy
. Cochrane Database Syst Rev 2013; CD006032.
4. Volpe F, Nicholas LL. How should patients with indirect traumatic optic neuropathy
be treated. J Neuroophthalmol 2011; 31:169–174.
5. Entezari M, Rajavi Z, Sedighi N, et al. High-dose intravenous methylprednisolone in recent traumatic optic neuropathy
: a randomized double-masked placebo-controlled clinical trial. Graefes Arch Clin Exp Ophthalmol 2007; 245:1267–1271.
6. Bracken M, Shepard M. A randomized, controlled trial of methylprednisolone or naloxone in the treatment
of acute spinal-cord injury. Results of the Second National Acute Spinal Cord Injury Study. N Engl J Med 1990; 322:1405–1411.
7. Bracken MB, Shepard MJ, Holford TR, et al. Administration of methylprednisolone for 24 or 48 h or tirilazad mesylate for 48 h in the treatment
of acute spinal cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury. JAMA 1997; 277:1597–1604.
8. Edwards P, Arango M, Balica L, et al. Final results of MRC CRASH, a randomised placebo-controlled trial of intravenous corticosteroid in adults with head injury-outcomes at 6 months. Lancet 2005; 365:1957–1959.
9. Bracken MB, Shepard MJ, Collins WF, et al. Methylprednisolone or naloxone treatment
after acute spinal cord injury: 1-year follow-up data. Results of the second National Acute Spinal Cord Injury Study. J Neurosurg 1992; 76:23–31.
10. Joseph MP, Lessell S, Rizzo J, Momose KJ. Extracranial optic nerve decompression for traumatic optic neuropathy
. Arch Ophthalmol (Chicago, Ill 1960) 1990; 108:1091–1093.
11. He Z, Li Q, Yuan J, et al. Evaluation of transcranial surgical decompression of the optic canal as a treatment
option for traumatic optic neuropathy
. Clin Neurol Neurosurg 2015; 134:130–135.
12. Emanuelli E, Bignami M, Digilio E, et al. Posttraumatic optic neuropathy: our surgical and medical protocol. Eur Arch Otorhinolaryngol 2014; [Epub ahead of print].
13. Yu Wai Man P, Griffiths PG. Surgery
for traumatic optic neuropathy
. Cochrane Database Syst Rev 2013; CD005024.
14. Rajiniganth MG, Gupta AK, Gupta A, Bapuraj JR. Traumatic optic neuropathy
: visual outcome following combined therapy protocol. Arch Otolaryngol Head Neck Surg 2003; 129:1203–1206.
15. Ropposch T, Steger B, Meço C, et al. The effect of steroids in combination with optic nerve decompression surgery
in traumatic optic neuropathy
. Laryngoscope 2013; 123:1082–1086.
16▪. Guy WM, Soparkar CNS, Alford EL, et al. Traumatic optic neuropathy
and second optic nerve injuries. JAMA Ophthalmol 2014; 132:567–571.
Little information is known about patients who present with a second ipsilateral traumatic optic nerve injury following an initial TON. Here, Guy et al. [16▪] suggest that there is loss of functional optic nerve reserve following trauma and find that the potential for visual recovery after second episode of TON may be limited regardless of treatment.
17▪. Entezari M, Esmaeili M, Yaseri M. A pilot study of the effect of intravenous erythropoetin on improvement of visual function in patients with recent indirect traumatic optic neuropathy
. Graefes Arch Clin Exp Ophthalmol 2014; 252:1309–1313.