2.2 Surgery
Surgery was performed under general anesthesia with patient placed supine in an Anti-Trendelenburg position. After proper decongestion of nasal cavities wide middle meatal antrostomy and ethmoidectomy were done and proceeded with identification of the natural sphenoidal ostium. A wide sphenoidotomy was done with resection of anterior sphenoid wall superiorly upto the skull base, laterally up to the lamina papyracea and medially upto the nasal septum. Once inside the sinus, the critical landmarks on the sphenoidal lateral wall were identified: the optic nerve, the internal carotid artery (ICA), and between them, the optic carotid recess. The prominence of the optic canal sometimes can be seen inside the last cell of the posterior ethmoid (Onodi cell), that can be easily detected with the pre-operative CT scan of paranasal sinuses. Unlike conventional endoscopic sinus surgeries, these landmarks were not “spot on” due to distorted anatomy as a result of fracture and fracture segments. Hence, entire mucosal covering of sphenoid sinus was denuded for better visualization and evaluation. Within the sinus, areas of traumatic dehiscence of ICA canal (cavernous and clinoid segments) were looked upon. Such dehiscences were more commonly found in well pneumatised sphenoid sinus. Surgery proceeds with delineation of entire lamina papyracea; which in most cases were found to be fractured. Decompression begins by removing the lamina papyracea at the midpoint between posterior wall of maxillary sinus and orbital apex. approximately 10-15mm anterior to the face of the sphenoid. Special care was taken not to injure the peri-orbita and the underlying extraocular muscles; if the periorbita are injured, fat protrudes into the field of view( if it occurs. it can be reduced by bipolar cautery). After the posterior lamina papyracea has been removed, periorbita was followed posteriorly to where they converge at the orbital apex. The thick bone between posterior ethmoid and sphenoid at the orbital apex is known as the optic tubercle. The Annulus of Zinn is attached to the superior, inferior and medial margins of the orbital junction with the optic canal and can be identified at the level of the optic tubercle. At this point, a micro-drill with a diamond burr, equipped with an irrigation system, was used to remove the optic tubercle, performing 1800 in circumference angle exposing the annulus of Zinn. Decompression was performed on entire length of optic canal within the sphenoid sinus till the point where the nerve curves from an oblique to a transverse direction in its posterior course, indicating that it was nearing the chiasmal region and well posterior to the encasing bony canal (Fig. 2). Finally, the optic nerve sheath and the annulus of Zinn are incised with a sickle knife, as these two structures may contribute to pressure on the optic nerve. This step of fenestration was done in our study only in selected cases; 1) when an intrasheath hematoma was suspected 2) when the optic nerve appears grossly edematous or an impression of bulging optic nerve was obtained after decompression 3) when there was a traumatic breach in the nerve sheath inflicted by a fracture segment and 4) lateral displacement of the bony optic canal with impingement on the optic nerve as well as cases with papillary edema and/or bleeding. The incision was performed in a longitudinal way on the upper-medial part, reducing the risk of damaging the ophthalmic artery, usually situated in the inferior-medial part of the optic nerve/canal (Fig. 3). In cases where we placed nerve sheath incisions, a mucosal graft was placed to prevent potential CSF leak. Also mucosal grafts from nasal cavity were placed over traumatic dehiscence of carotid canal; if any. Nasal packings were avoided as far as possible. But if any coexistent CSF leak was repaired, then packing was done using Surgicel® and AbGel®©™ without any contact with decompressed optic nerve.
After the surgical procedure, high-dose systemic administration of steroids was maintained every 8 hours for 24 hours. An ophthalmological examination was performed 24 hours after surgery.
Post operative improvement of vision was assessed 24 hours following surgery. Post operative visual improvement was classified as complete improvement, partial improvement and no improvement (Table II).
RESULTS
When surgical intervention was made within 72 hours, there was complete improvement in 59% cases; 25% cases had partial improvement and 16% had no improvement.
When done between 3 and 7 days, 31% had partial improvement and 69% had no improvement.
Beyond 7 days, there was 0% improvement (Table II)
We encountered 13 cases of intra operative CSF leaks all of which were traumatic and were not apparent before surgery. All these were repaired using fascial graft using tensor fascia lata at the time of surgery itself.
Nerve sheath fenestrations were done in 5 selected cases as discussed above of which 2 (40%) patients had partial improvement of vision. There were no associated CSF leaks in cases where fenestration of sheath was done. However a mucosal graft was placed over the site of fenestration to prevent the chances of delayed CSF leaks.
Post operative bed side visual assessment were done 24 hours after surgery and daily till the patient was stable to be shifted to ophthalmology OPD for detailed evaluation. Earliest signs of improvement in vision were found 72 hours after surgery. Visual improvement had continued to occur for upto 8 weeks after surgery in our study.
DISCUSSION
Traumatic optic neuropathy that failed to recover with medical management should ideally be intervened as early as possible. Optic nerve being a part of the central nervous system is incapable of regeneration and if not “rescued” at the earliest, it can result in irreversible blindness. The influence of the interval between trauma and surgery is controversial. Theoretically, improvement in vision following surgery would be better when patient receives surgery as early as possible (within 72 hours). However, in a study, Dhaliwal et al. found vision improvement of 57% in patients receiving surgery within 3 days post-trauma, 58% in patients who underwent surgery between 3 and 7 days after the trauma, and 51% in the group who underwent surgery more than 7 days after the trauma12. The authors thus concluded that timing is not a prognostic factor in vision improvement. These contradictory results could be due to an absent consensus on protocols for EOND. Secondly, in this study types and severity of injury were not discussed. According to pathophysiological principles, increased pressure on the optic nerve leads to ischemia and to deterioration in visual acuity that might be irreversible after 24–48 hours13, 14.
The possible complications of optic nerve decompression are bleeding of the sphenopalatine artery, ethmoidal arteries, ophthalmic artery, internal carotid artery, injury to the optic nerve and / or chiasm, medial rectus muscle, CSF leak, pneumoencephalus or death.
There is significant heterogeneity in the literature considering visual improvement following EOND for traumatic optic neuropathy as reviewed by Wang et al, 15. Visual improvement was found in 35% to 82% of the time.16, 12, 17-29. Rates of improvement in patients presenting with no LP range from 0% to 62%17,19,23,26,29 , whereas rates of improvement in those with partial vision loss range from 50% to 100%,19,23,28,29. In a large systematic review of traumatic ON, rates of improvement were decreased in those with No LP preoperatively (41% vs. 84-93% in patients with partial vision loss)12. Although improvement in vision is seen in the majority of cases, there is insufficient follow-up data to establish whether the improvements seen in these patients are clinically meaningful. There is no quantitative assessment of visual improvement following decompression in the literature. Surgery can be considered beneficial only if the patient regains useful vision. Hence we have devised a method of quantifying the visual improvement as visual improvement scale (VIS) as follows: Complete improvement, partial improvement and no improvement. A patient is said to have complete improvement if his vision improves upto 4/6 or above after surgery; partial improvement if his vision improves upto 3/6 and no improvement if there is no change in vision post surgery (Table I).
Finally, unlike facial nerve, optic nerve cannot be repaired. Optic nerve which develops from central nerve system as out-pouching of the diencephalon (optic stalks) is incapable of regeneration. Whereas facial nerve is a mixed nerve which develops from second pharyngeal arch . Facial nerve injuries are initially treated with decompression and if it fails various other options are available ( direct facial- facial nerve suture, facial nerve interpositional graft, hypoglossal- facial – jump- nerve anastomosis, dynamic muscle transfer and sling plasties)30. But optic nerve once atrophied cannot be repaired. Hence it is always advisable to surgically intervene as early as possible to salvage the vision in clinically indicated, properly selected patients. And it is justified to state that “Time lost is vision lost”.
CONCLUSION
Traumatic optic neuropathy (TON) denotes an acute injury to the optic nerve (ON) secondary to direct or indirect trauma. Direct trauma occurs usually due to penetrating injury
causing shearing or hematoma of the optic nerve. Indirect TON is usually due to blunt head trauma and is caused by increased intracanalicular pressure from the injury with vascular ischemia and interruption of neurofeedback mechanisms leading to blindness28, 31. To date no standard treatment protocol has been developed regarding traumatic optic neuropathy. However, our study shows that the timing is the most important prognostic factor in traumatic optic neuropathy. Best results are obtained when decompression is done within 72 hours. Since most of these patients have associated head injury, which needs neurosurgical intervention; it is advisable to perform optic nerve decompression in the same sitting to ensure better prognosis. Degree of vascular compromise is another determining factor for visual restoration. Unfortunately we do not still have a tool to assess the degree of optic nerve ischemia.
Also proper selection of patient for decompression plays a vital role. The overall condition/ prognosis of the patient as well as the extend of head injury of the patient has to be considered before doing a “heroic” attempt to restore the vision. Our study did not show any added benefits with nerve sheath incision routinely. Also pre operative visual status have important role in the post operative improvement. Patient with light perception and above seems to benefit more from the surgery than patients with no perception of light.