Abstract
BACKGROUND
Congenital optic canal stenosis causing compressive optic neuropathy is a rare disorder that presents unique diagnostic and treatment challenges. Endoscopic endonasal optic nerve decompression (EOND) has been described for optic nerve compression in adults and adolescents but has never been reported for young children without pneumatized sphenoid sinuses. The authors describe preoperative and intraoperative considerations for three patients younger than 2 years of age with congenital optic canal stenosis due to genetically confirmed osteopetrosis or chondrodysplasia.
OBSERVATIONS
Serial ophthalmological examinations, with a particular focus on object tracking ability, fundoscopic examination, and visual evoked potential trends in preverbal children, are important for detecting progressive optic neuropathy. The lack of pneumatization of the sphenoid sinus presents unique challenges and requires the surgical creation of a sphenoid sinus with the use of neuronavigation to determine the limits of bony exposure given the lack of easily identifiable anatomical landmarks such as the opticocarotid recess. There were no perioperative complications.
LESSONS
EOND for congenital optic canal stenosis is safe and technically feasible even given the lack of pneumatization of the sphenoid sinus in young patients. The key operative step is surgically creating the sphenoid sinus through careful bony removal with the aid of neuronavigation.
Keywords: endoscopic endonasal transsphenoidal approach, optic nerve decompression, optic canal, pediatric, case report
ABBREVIATIONS: 3D = three-dimensional, EOND = endoscopic endonasal optic nerve decompression, CSF = cerebrospinal fluid, CT = computed tomography, MRI = magnetic resonance imaging, VEP = visual evoked potential
Congenital optic canal stenosis causing compressive optic neuropathy is a rare disorder that presents a unique set of challenges for anterior skull base surgeons. Traditional surgical approaches to decompress the optic canal consist of craniotomies, extranasal transethmoidal approaches, and transorbital approaches; however, endoscopic endonasal optic nerve decompression (EOND) has been more widely adopted in recent decades given the advances in endoscopic technology and more favorable morbidity profile compared to traditional approaches.1 Much of the literature reporting on surgical decompression of the optic canal has been in adult patients with traumatic optic neuropathy.1–5 In this demographic, recent studies have highlighted that residual vision, shorter timescale of vision loss, and early timing of surgery are associated with favorable visual outcomes after EOND.3–5
On the other hand, outcomes after surgical decompression for nontraumatic optic neuropathy are less frequently reported. The few studies on this subject have reported favorable outcomes even in cases of severe preoperative vision loss; however, disparate etiologies such as meningiomas, fibrous dysplasias, and pituitary adenomas are often grouped together, making it difficult to extrapolate their findings to other disease processes.6–8 Even fewer studies exist for nontraumatic optic neuropathy in pediatric patients. Case reports and case series have described some favorable outcomes after EOND for nontraumatic etiologies of optic neuropathy such as fibrous dysplasia and hyperostosis in adolescent patients.9, 10 To our knowledge, EOND for younger patients who have not yet had sphenoid sinus pneumatization has never been reported. Here, we illustrate three cases of patients younger than 2 years of age without sphenoid sinus pneumatization who underwent EOND for congenital optic canal stenosis. While long-term follow-up is limited in these cases, we highlight key components of the preoperative workup and the surgical considerations in this patient population.
Illustrative Cases
Case 1
History
A 22-month-old female with TCIRG1-related autosomal recessive osteopetrosis had been born at term (39 weeks’ gestation) via cesarean section to a 30-year-old female (gravida 1, para 0) via natural pregnancy. She had had no maternal exposures and had passed all prenatal screening tests. She had been discharged to home after 5 days.
At 3 weeks of age, she was admitted for workup for feeding difficulties and poor weight gain. Routine chest and abdominal radiographs revealed increased femur bone density, but otherwise she had a normal neurological examination and no dysmorphic features. Ophthalmological examination revealed reactive pupils and hypoplastic-appearing nerves but no frank pallor. She had limited blink to light responses bilaterally. Computed tomography (CT) revealed sclerosis of the skull base, enlargement of the anterior clinoid processes, and narrowing of the optic canals bilaterally (Fig. 1).
FIG. 1.
Case 1. Preoperative and postoperative CT and MRI. Thin arrows indicate the sphenoid bone; thick arrows, the right optic canal.
At 2 months of age, ophthalmological examination showed a visual evoked potential (VEP) amplitude of 15% of normal with some roving eye movements. She was also found to have optic nerve dysplasia (20% smaller than normal for age) with pallor and atrophy. She underwent unrelated donor stem cell transplantation at 3 months of age for TCIRG1-related autosomal recessive osteopetrosis, complicated by rejection.
At 4 months of age, she had an increased ventricle size and extra-axial spaces. Her anterior fontanelle was full, and a lumbar puncture showed an elevated opening pressure, so a reservoir was placed. She underwent intermittent cerebrospinal fluid (CSF) removal for several weeks, with stability of head circumference measurements. Ophthalmological examination revealed mild to moderate pallor bilaterally. By 6 months of age, she was not tracking and did not shield her eyes from bright lights. At 12 months of age, her pupils were sluggishly reactive at 6 mm, and ophthalmological examination revealed pallor and a further decline in her VEP amplitude. She underwent a second unrelated donor stem cell transplantation.
At 18 months of age, the patient would only reach for objects if they were very bright in an environment of low light. VEP amplitude showed diminished but residual vision in both eyes (13% on the left and 5% on the right). Magnetic resonance imaging (MRI) and CT showed bilateral narrowing of the optic canals (Figs. 1 and 2). She underwent evaluation by the neurosurgery and otolaryngology teams and was deemed a candidate for EOND for bilateral optic nerve decompression.
FIG. 2.
Case 1. Three-dimensional reconstructions of preoperative and postoperative CT. The thin arrow indicates the sellar floor; the thick arrows indicate the right optic canal.
Surgical Management
Preoperatively, a fine-cut head CT and brain MRI with and without contrast were performed to aid in frameless stereotactic neuronavigation. Her weight was 11.7 kg. After the induction of general anesthesia and endotracheal tube placement, Afrin-soaked pledgets were inserted into the nostrils, and the patient was prepped and draped.
Otolaryngology began the surgery using a 0˚ 2.9-mm endoscope to visualize each nasal opening, and the inferior turbinates were lateralized bilaterally. The inferior halves of the middle turbinates were removed. The superior turbinates and sphenoethmoidal recess on each side were identified, and the ethmoids were opened. A posterior septoplasty was performed to open the rostrum of the sphenoid sinus. A septal flap was raised in the submucoperichondrial and subperiosteal plane on the right side, and the cartilage just anterior to the bony cartilaginous junction was divided to elevate a right-sided septal flap. The cartilage and bone posterior to this incision posterior to the nasal bones were incised and taken down, and the cartilage and bone over the rostrum were also taken down. The sphenoid on each side was identified, and the intrasinus septum, which consisted of cartilaginous and cancellous bone, was removed. The remaining portions of the middle turbinates were removed, and the ethmoids were opened posteriorly. A posterior septectomy was performed to allow binasal manipulation.
The neurosurgery team then used a small, rough, diamond drill bit to remove the mostly cancellous bone in the sphenoid sinus until the dura of the sella and the floor of the anterior cranial fossa were identified. The bone of the right medial posterior orbits was removed with a drill and small curettes. The optic canal was then unroofed along its medial and inferior margins. The dissection was then carried to the left posterior medial orbit where additional bone was removed, and the bone of the sella and anterior fossa region also extended farther to the left. The left optic canal was unroofed along its medial and inferior margins. A number of endoscopes were used including 0˚ and 30˚ angles of both the 2.9- and 4-mm diameter variety.
The septum was closed anteriorly, and nasal splints were placed. Gelfoam and a hemostatic matrix were placed. The patient was awakened, extubated, and taken to the pediatric intensive care unit. Estimated blood loss was 150 ml.
Clinical Outcome
On postoperative day 1, the patient underwent head CT, which showed decompression of the bilateral optic canals (Figs. 1 and 2). She was discharged home on postoperative day 4. Endoscopic endonasal debridement of the sinus cavity and removal of splints were performed in the office 1 week after surgery.
Six months after surgery, VEP amplitude improved to 19% on the left and 15% on the right. She was reported to be shielding her eyes more from the sun, and she had more pupil reactivity. She still had roving eye movements and stable diffuse pallor. Three years after surgery, MRI showed continued decompression of the optic nerves (Fig. 1). Five years after surgery, she was learning braille reading and writing.
Case 2
History
A 22-month-old female with TCIRG1-related autosomal recessive osteopetrosis had been born at term (39 weeks’ gestation) via spontaneous delivery to a 25-year-old female (gravida 1, para 0) via natural pregnancy. She had passed all screening tests and had been discharged home shortly after birth.
At 7 months of age, she presented with macrocephaly and bulging anterior fontanelle with enlarged ventricles. She underwent ventriculoperitoneal shunt placement at 8 months of age. Postoperative shunt radiographs showed increased bone density throughout the skull, ribs, and long bones. The metaphyses of the bilateral humeri appeared frayed. After genetic testing confirming TCIRG1-related autosomal recessive osteopetrosis, she underwent stem cell transplantation at 10 months of age.
At 8 months of age, ophthalmology examination showed reduced dark-adapted rod and rod/cone electroretinogram responses. VEP amplitude was 18% on the right and 51% on the left. At 12 months of age, she had increased pallor of both optic nerves and decreased VEP amplitude, 23% on the right and 14% on the left.
At 13 months of age, she had a prolonged hospital admission for bacteremia and cytomegalovirus viremia in the setting of acute graft-versus-host disease complicated by pancytopenia and pericardial effusion.
At 19 months of age, ophthalmology examination showed worsening optic nerve pallor and worsening VEP, 7% on the right and 19% on the left. She was still able to track large objects despite some nystagmus bilaterally. Head CT and MRI of the orbits showed severe compression of the bilateral orbital apices with fluid-filled optic nerve sheaths suggestive of edema (Figs. 3 and 4). At 22 months of age, VEP amplitude was 22% on the right and 24% on the left. Her weight was 6.86 kg.
FIG. 3.
Case 2. Preoperative and postoperative CT and MRI. Note the hypoplastic optic nerve and hyperintensity surrounding the optic nerves, indicating fluid-filled optic nerve sheaths in the middle panel. The thin arrow indicates the sphenoid bone; the thick arrows indicate the right optic canal. SPACE = sampling perfection with application-optimized contrast using different flip angle evolutions.
FIG. 4.
Case 2. Three-dimensional reconstructions of preoperative and postoperative CT. The thin arrow indicates the sellar floor; the thick arrows indicate the right optic canal.
Surgical Management
An operative approach similar to that in case 1 was taken for this patient. However, nasal splints were not placed. Estimated blood loss was 20 ml.
Clinical Outcome
On postoperative day 1, the patient underwent head CT that showed postsurgical changes in the sphenoid sinus and decompression of bilateral optic canals (Figs. 3 and 4). She was discharged home on postoperative day 5. One month after surgery, her VEP amplitude was 38% on the right and 6% on the left. However, over the next few months, she had numerous hospitalizations for recurrent neutropenic fevers and failure to thrive. She ultimately experienced complications of graft-versus-host disease versus graft rejection and died 2 months after surgery.
Case 3
History
A 23-month-old male with X-linked chondrodysplasia punctata 1 had been born at term (37 weeks’ gestation) via spontaneous delivery to a 20-year-old female (gravida 4, para 1) via natural pregnancy. Spinal anomalies were noted on prenatal ultrasound. Postnatal radiography demonstrated stippled epiphyses and calcifications in the lower sacrum and pelvis.
At 7 months of age, ophthalmology examination showed left strabismic amblyopia, bilateral high hyperopia, and bilateral mild optic nerve dysplasia. At 15 months of age, the patient was noted to have pallor of the left optic disc. At 17 months of age, head CT showed left-sided optic canal stenosis and MRI showed left orbital apex narrowing with thinning of the left optic nerve, which was unmyelinated (Fig. 5). At 20 months of age, VEP amplitude was 14% on the left. The right eye was tracking, but the left eye was not. He was evaluated and was deemed a candidate for left-sided optic canal decompression. His weight was 9.2 kg.
FIG. 5.
Case 3. Preoperative CT and MRI. The thin arrow indicates the sphenoid bone; the thick arrows indicate the left optic canal.
Surgical Management
A similar approach was taken for exposure. However, in this case of unilateral decompression, the lamina papyracea on the left side was drilled until the posterior periorbita was identified. Then using a combination of the drill and a curette, the optic nerve was decompressed in an anterior to posterior direction from the posterior aspect of the globe to the sella. We confirmed 180˚ of bony removal around the left optic nerve with intraoperative navigation. Splints were not placed. The estimated blood loss was 20 ml.
Clinical Outcome
The patient was discharged home on post-operative day 1. Eight months after surgery, he continued to have diminished vision in the left eye; however, he was able to see bright lights and the movement of objects. Fundoscopic examination and VEP recording have not been performed at the time of writing.
Patient Informed Consent
The necessary patient informed consent was obtained in this study.
Discussion
Observations
Ophthalmological Examination in Infants
Although there is inherent variability in an individual child’s milestones, children should be able to follow large objects and toys and recognize their mother by 2–3 months of age and should be reaching for and grasping objects by 5–6 months of age.11 Not shielding the eyes from bright light, failure to track objects in normal lighting, and not blinking to threat should raise concern about visual perception. Unilateral findings should also raise concern. A dilated fundoscopic examination can reveal structural abnormalities such as optic nerve pallor, as seen in our patients. VEP is a technique that measures an electroencephalographic response to a visual stimulus and has been used in hopes of providing objective measurements of visual acuity especially in nonverbal infants. Zheng et al.12 listed several studies of normal visual development in infants using VEP; however, they also described a need for established norms of VEP acuity in infants and children. In addition, test-retest variability exists in each component of the VEP test (sweep, latency, and amplitude), which may limit interpretability for any individual subject.13 Therefore, VEP results must be used with that variability in mind. In our patients, there was a general trend for VEP amplitude to decrease prior to surgery. After surgery, our first patient had an improvement of VEP amplitude but was unable to read well enough to not learn how to read and write in braille. Our second patient’s right eye had an improvement in VEP amplitude, but the left eye did not. Unfortunately, she passed away from reasons unrelated to surgery, so longer follow-up was not possible. Our third patient continued to have reduced left-sided vision at 8 months but was able to see bright lights and object movement.
Surgical Considerations and Anatomy
For EOND specifically, the size of the nostrils and diameter of the endoscope must be considered. We found that the 2.9-mm endoscope is well suited for children around 2 years of age. Imaging findings for candidates for EOND should demonstrate orbital apex narrowing with or without thinning of the optic nerve or optic nerve sheath edema. The lack of pneumatization of the sphenoid sinus in this age group must be considered. The literature varies on the age at onset of initial pneumatization,14, 15 with reports ranging from 6 months to 4 years, starting at the anterior part of the sphenoid bone. Pneumatization then continues until the early teenage years. Because of the lack of pneumatization in our patients, it was not possible to identify the usual landmarks seen in adult transsphenoidal surgery. Our exposure included surgically creating a sphenoid sinus through careful drilling of the cancellous sphenoid bone in the midline until the dura of the sella was reached. The posterior ethmoid cells may be opened to create additional working room. The bony removal of the tuberculum sellae can define the upper limit of the exposure to define the optic nerves laterally. It is imperative to consider the location of the carotid arteries, which are typically inferior and anterior to the optic nerves. Incomplete pneumatization of the sinuses makes it difficult to visualize standard anatomic landmarks such as the opticocarotid recess.16 Patients may undergo unilateral decompression for unilateral pathology to avoid iatrogenic injury to the contralateral nerve. In these cases, the lamina papyracea can be drilled until the posterior periorbita is identified, and then this corridor can be carefully expanded towards the midline in a superior and medial direction along the length of the optic nerve to reach the sellar dura. A diamond drill bit with irrigation was used to allow for appropriate visualization and removal of excess heat from drilling. Stereotactic neuronavigation was imperative to safely define the depth and width of bony exposure. We used an electromagnetic frameless stereotaxy system and navigated using preoperative thin-cut bone window head CT and brain T1-weighted MRI with gadolinium contrast. No patients suffered any immediate surgical complications such as CSF leakage, hemorrhage, optic nerve injury, vascular injury, or surgical site infection.
Counseling
Prior to surgery, families were counseled regarding the expectations of the visual outcome after EOND. Based on the limited literature, the visual acuity and visual fields may be preserved in the majority of patients following EOND. However, there is no literature regarding performing surgery in this age group (< 2 years), so its benefits are not clear. Families were counseled that the intention of surgery was to preserve any residual visual acuity and visual fields, but it is difficult to translate or correlate physical examination and VEP information in this preverbal age group with any predictive value to visual outcomes at school age (e.g., necessity for learning braille). Therefore, it is not clear whether this outcome was expected or unexpected. Families were counseled regarding the natural history of visual outcomes without surgery, which are known, and they made an individualized decision regarding whether or not to pursue the operation.
Lessons
Candidacy for EOND
Detailed and serial ophthalmological histories and examinations are crucial to identify patients not meeting visual milestones. The trend of VEP amplitudes may be a useful adjunct measure of visual function in preverbal children, keeping in mind inherent variability between measurements. To illustrate this point, the patient in case 2 had a slight increase in the VEP amplitude measurement at 22 months of age, but because no improvements in object tracking were noted to correlate with this, she proceeded to have surgery to preserve visual function. Optic canal stenosis on CT and MRI is apparent when there are concurrent findings of orbital apex narrowing and either optic nerve atrophy or edema within the optic nerve sheath.
Technical Considerations of EOND
There are a variety of endoscope options for pediatric sinus operations, but in our experience, a 2.9-mm-diameter endoscope provides a good balance between usability and size profile and is technically feasible to be used in infants starting at approximately the age of 2 years. For patients who have not yet had sphenoid sinus pneumatization, EOND can be safely performed by surgically creating an operative corridor through the sphenoid bone. Accurate intraoperative neuronavigation is crucial to defining the limits of depth and width of bony exposure given the absence of usual anatomic landmarks. Key operative steps and technical considerations are careful exposure of the midline dura, bony removal in an anterior-to-posterior and lateral-to-medial direction along the optic nerve, and the proximities and courses of the optic nerve dura and the internal carotid artery.
This study is limited by the small number of cases and limited follow-up. However, these were the only cases of EOND performed for congenital optic canal stenosis at our institution given the rarity of this condition. Although we did not have any immediate peri-operative complications, the efficacy of this procedure is unclear since do not have long-term follow-up on two of the three patients. Further study in the form of a multi-institutional collaboration is needed to better understand the risks and benefits of EOND for this indication.
Disclosures
Dr. Kim reported personal fees for consulting from Monteris Medical and grants from Stryker for a dural substitute study outside the submitted work.
Author Contributions
Conception and design: Schneider, Chicoine, Limbrick. Acquisition of data: Yang, Schneider, Chicoine, Limbrick. Analysis and interpretation of data: Yang, Schneider, Kim, Limbrick. Drafting the article: Yang, Limbrick. Critically revising the article: all authors. Reviewed submitted version of manuscript: Yang, Chicoine, Kim, Limbrick. Approved the final version of the manuscript on behalf of all authors: Yang. Administrative/technical/material support: Limbrick. Study supervision: Chicoine, Kim, Limbrick.
Correspondence
Peter H. Yang: Washington University in St. Louis, MO. peter.yang@wustl.edu.
References
- 1.Luxenberger W, Stammberger H, Jebeles JA, Walch C. Endoscopic optic nerve decompression: the Graz experience. Laryngoscope. 1998;108(6):873-882. [DOI] [PubMed] [Google Scholar]
- 2.Levin LA, Beck RW, Joseph MP, Seiff S, Kraker R. The treatment of traumatic optic neuropathy: the International ptic erve Trauma Study. Ophthalmology. 1999;106(7):1268-1277. [DOI] [PubMed] [Google Scholar]
- 3.Sun J, Cai X, Zou W, Zhang J. Outcome of endoscopic optic nerve decompression for traumatic optic neuropathy. Ann Otol Rhinol Laryngol. 2021;130(1):56-59. [DOI] [PubMed] [Google Scholar]
- 4.Li H, Zhou B, Shi J, Cheng L, Wen W, Xu G. Treatment of traumatic optic neuropathy: our experience of endoscopic optic nerve decompression. J Laryngol Otol. 2008;122(12):1325-1329. [DOI] [PubMed] [Google Scholar]
- 5.Yang QT, Zhang GH, Liu X, Ye J, Li Y. The therapeutic efficacy of endoscopic optic nerve decompression and its effects on the prognoses of 96 cases of traumatic optic neuropathy. J Trauma Acute Care Surg. 2012;72(5):1350-1355. [DOI] [PubMed] [Google Scholar]
- 6.Berhouma M, Jacquesson T, Abouaf L, Vighetto A, Jouanneau E. Endoscopic endonasal optic nerve and orbital apex decompression for nontraumatic optic neuropathy: surgical nuances and review of the literature. Neurosurg Focus. 2014;37(4):E19. [DOI] [PubMed] [Google Scholar]
- 7.Pletcher SD, Metson R. Endoscopic optic nerve decompression for nontraumatic optic neuropathy. Arch Otolaryngol Head Neck Surg. 2007;133(8):780-783. [DOI] [PubMed] [Google Scholar]
- 8.Bulters DO, Shenouda E, Evans BT, Mathad N, Lang DA. Visual recovery following optic nerve decompression for chronic compressive neuropathy. Acta Neurochir (Wien). 2009;151(4):325-334. [DOI] [PubMed] [Google Scholar]
- 9.Baba T, Minamida Y, Mikama T, Koyanagi I, Houkin K. Entrapment neuropathy of the optic nerve due to hyperostosis associated with congenital anemia. J Neurosurg. 2005;103(5):917-919. [DOI] [PubMed] [Google Scholar]
- 10.Behbahani M, Fernando S, Peng S, et al. Endoscopic endonasal optic nerve decompression: treatment of fibrous dysplasia in a pediatric population. J Neurosurg Pediatr. 2023;31(2):179-185. [DOI] [PubMed] [Google Scholar]
- 11.Scharf RJ, Scharf GJ, Stroustrup A. Developmental milestones. Pediatr Rev. 2016;37(1):25-37. [DOI] [PubMed] [Google Scholar]
- 12.Zheng X, Xu G, Zhang K, et al. Assessment of human visual acuity using visual evoked potential: a review. Sensors. 2020;20(19):5542. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Prager TC, Zou YL, Jensen CL, Fraley JK, Anderson RE, Heird WC. Evaluation of methods for assessing visual function of infants. J AAPOS. 1999;3(5):275-282. [DOI] [PubMed] [Google Scholar]
- 14.Jang YJ, Kim SC. Pneumatization of the sphenoid sinus in children evaluated by magnetic resonance imaging. Am J Rhinol. 2000;14(3):181-185. [DOI] [PubMed] [Google Scholar]
- 15.Szolar D, Preidler K, Ranner G, et al. The sphenoid sinus during childhood: establishment of normal developmental standards by MRI. Surg Radiol Anat. 1994;16(2):193-198. [DOI] [PubMed] [Google Scholar]
- 16.Wu T, Chandy Z, Ference E, Lee JT. Endoscopic skull base surgery in the pediatric population. Curr Treat Options Allergy. 2021;8(3):274-284. [Google Scholar]