Acute Enophthalmos After Lumbar Puncture in a Patient with Type 1 Neurofibromatosis Related Sphenoid Wing Dysplasia

ABSTRACT

A 19-year-old woman with type 1 neurofibromatosis related sphenoid wing dysplasia underwent a lumbar puncture (LP) after an unexplained syncopal event. The cerebrospinal fluid work-up was unremarkable. However, 30-hours post LP, she developed marked enophthalmos ipsilateral to the calvarial dysplasia. The enophthalmos gradually resolved within 72 hours. We surmise that the patient’s enophthalmos was due to an LP induced decrease of intracranial pressure in conjunction with the orbital bone anomaly.

Introduction

Type 1 neurofibromatosis (NF-1) is a genetic condition consisting of a combination of several characteristics including cutaneous café au lait spots, optic gliomas, Lisch nodules and calvarial bone defects. Although rare, the most notable calvarial defect in NF-1 patients is unilateral sphenoid wing dysplasia, which may occur in up to 11% of this population.^1,2 Such dysplasia leaves patients with orbital instability, which can result in structural defects. Enophthalmos, or abnormal posterior displacement of the globe in the orbit, has been reported in several NF-1 related cases with sphenoid wing dysplasia.^2–4 In accordance with a previous case by Ye et al., who observed an NF-1 patient with enophthalmos post thoracentesis-induced meningocele leakage, we report a case of lumbar puncture (LP) induced enophthalmos in an NF-1 patient with sphenoid wing dysplasia. We surmise our patient’s enophthalmos was due to a presumed decrease in intracranial pressure (ICP) post-LP in conjunction with a pre-existing sphenoid wing dysplasia.

Case report

A 19-year-old African American woman with a medical history significant for NF-1 presented to hospital after collapsing while leisurely walking outside. She denied any previous syncopal events, blurred vision, diplopia, dizziness, headache, incontinence, weakness, sensory loss, nor history of drug or alcohol use prior to this incident. She denied taking any medications, including anti-epileptics. She had been diagnosed with NF-1 5 years earlier after a single epileptic event, which led to the discovery of a right-sided sphenoid wing dysplasia on imaging. In addition, her family history was positive for NF-1 in her mother. The evaluation was notable only for mild right ankle pain secondary to the fall. An electrocardiogram demonstrated normal sinus rhythm. Aside from the patient’s previously known NF-1 related sphenoid wing dysplasia, a non-contrast computed tomography (CT) scan of the head was negative for acute abnormality (Figure 1A). She was subsequently admitted to the hospital for a further work-up. An electroencephalogram revealed no epileptiform activity. No acute changes were appreciated on magnetic resonance imaging (MRI) at this time. CT angiography of the brain revealed a 2 mm infundibular dilatation at the origin of the left ophthalmic artery. No vasculitic findings were observed.

Figure 1.

(A) Two serial image slices from computed tomography (CT) scan without contrast taken prior to the lumbar puncture (LP). Right sphenoid wing dysplasia is appreciated. An axis has been drawn between bilateral zygomatic bones demonstrating the globe position prior to intervention. (B) Two serial image slices from the CT without contrast taken after the LP. When assessing the globe relative to the inter-zygomatic axis, a posterior displacement of the right globe is appreciated when compared with the left globe. Similarly, the right eye has migrated posteriorly as compared to its pre-LP state in Figure 1A. Images in Figures 1A,B slightly vary based on the patient’s head tilt during scanning.

A fluoroscopic-guided LP was conducted to assess the patient for any other causative aetiologies. Thirteen millilitres of clear cerebrospinal fluid (CSF) were collected and revealed 0/mm³ red blood cells, 1/mm³ white blood cells, a mildly elevated protein level of 88 mg/dL protein (normal range15-45 mg/dL), but negative oligoclonal bands. An opening pressure was not measured. In addition, she had a negative laboratory work-up including complete blood count, inflammatory markers and microbiology. The appropriate care was administered at bedside with minimal physical activity for the duration of her stay and she was deemed fit for discharge within 20 hours of the LP with appropriate outpatient follow-up.

She returned to the emergency department 30 hours post-LP due to right-sided headache, blurred vision, diplopia and tearing of the right eye. She denied any trauma, dizziness, difficulty walking or back pain surrounding her LP site. A family member remarked that her eye had looked “sunken-in” and advised her to return to the hospital.

External examination of the right eye (OD) revealed mild ptosis with enophthalmos when compared to the left eye (OS) (Figure 2). Additionally, the right eye was noted to have increased lacrimation and tenderness just inferior to the superior orbital rim. External examination of the left eye was unremarkable. Her visual acuity without corrective lenses using a Rosenbaum near card was found to be 20/50 OD and 20/25 OS. Her intraocular pressures were within normal limits in each eye. Her pupils were equal, round and reactive to light bilaterally without any signs of an afferent pupillary defect. She was found to have a new onset right monocular diplopia. Her extraocular movements were full bilaterally with pain during infraduction OD. A CT scan of her head demonstrated the dysplastic right sphenoid bone along with new right-sided enophthalmos (Figure 1B) and lacrimal gland enlargement. Laboratory studies including erythrocyte sedimentation rate, C-reactive protein, anti-neutrophil cytoplasmic antibody, anti-nuclear antibody, rheumatoid factor and complete blood count came back within normal ranges. Aside from right-sided enophthalmos, it was determined she was also experiencing a concurrent ipsilateral dacryoadenitis. Intravenous (IV) fluids were started, along with IV methylprednisolone and IV ampicillin/sulbactam to cover potential inflammatory and infectious aetiologies of the dacryoadenitis, respectively.

Figure 2.

External photograph taken after the lumbar puncture demonstrating notable enophthalmos of the right eye compared with the left eye.

After 2 days she received a follow-up MRI of the head (Figure 3), which revealed a corrected, slightly proptotic right globe and an improving dacryoadenitis. She was transitioned from IV steroids to an oral steroid taper, and the antibiotics were stopped. Her visual acuity without corrective lenses improved to 20/20 in each eye. Both her pupils and intraocular pressures remained stable for the duration of her inpatient follow-up, and the previously noted monocular diplopia improved. Supraduction OD remained slightly limited without pain, likely secondary to her proptotic state. All other extraocular movements were intact and painless. Mild fullness was noted at the right lacrimal gland, which had improved from previously.

Figure 3.

T1-weighted magnetic resonance imaging withcontrast taken after treatment with intravenous fluids, steroids and antibiotics demonstrating mild anterior displacement of the right eye orbital contents as noted from the inter-zygomatic axis. Right-sided dacryoadenitis can also be appreciated.

Discussion

Sphenoid wing dysplasia is defined as partial or complete absence of the sphenoid’s greater wing.^1,2,5 Although rare, sphenoid wing dysplasia is a known complication of NF-1 and has been reported in up to 11% of cases.^1,2 This orbital defect was first described by Binet et al. in 1969 as a congenital osteo-dysgenesis of the sphenoid, however modern hypotheses suggest age-related destruction through an overall progressive disease course.^6–8 While it remains unclear if the pathophysiology of sphenoid wing dysplasia is osteogenic versus osteolytic in nature, the dysplasia has been found to be asymptomatic in most cases and typically discovered as an incidental finding.^1,8

There is structural importance of the greater and lesser wings of the sphenoid bone as they serve in both anatomic support to the walls of the orbit as well as intracranial conduits for nervous and vascular components. Such structures include the optic nerve and ophthalmic artery and therefore compromise of the sphenoid bone may put patients at higher risk of visual complications. One ophthalmological complication of sphenoid wing dysplasia, which we report here is enophthalmos, or posterior displacement of the globe within the orbit.^5,9 Through relative expansion of the orbit itself, the globe and surrounding orbital contents migrate posteriorly.⁵ Patients can present with a wide range of signs and symptoms depending on the aetiology, including pain, blurred vision, headache, and nausea, among other non-specific symptoms.⁵

Enophthalmos may be evident clinically, however modalities such as Hertel exophthalmometry or imaging may be required for diagnnosis. While the Ophthalmology team in this case did not have access to an exophthalmometer at the time of inpatient consultation, the progression of enophthalmos and its subsequent reversal was clearly demonstrated on serial neuroimaging. Although variations in head positioning during pre-LP and post-LP radiological capture contribute to deviation between image slices and measurement, the post-LP neuroimage (Figure 1B) demonstrates a relative posterior migration of the right globe contents when contrasted with the pre-LP image (Figure 1A).

Similarly, Ye et al. reported a case of acute enophthalmos after thoracentesis-induced meningocele leakage of an NF-1 patient with sphenoid wing dysplasia. After removal of approximately 280 mL CSF via a thoracentesis, they reported a near instantaneous enophthalmos ipsilateral to the calvarial bone dysgenesis, as well as a 1 week period before resolution of symptoms. Though their mechanism of extraction allowed for larger volumes of CSF removal, this case further supports the association between sphenoid wing dysplasia and acute enophthalmos in the setting of diminished intracranial pressure (ICP).⁴ Due to the transient nature of symptoms in both our case, as well as the case described by Ye et al., it is reasonable to assume that the spontaneous resolution of symptoms may be attributed to CSF volume restoration, occurring at an estimated rate of 20 mL/hr.¹⁰

Both Hwang et al. and Meyer et al. reported cases of bilateral enophthalmos in the setting of ventriculoperitoneal shunting.^11,12 These cases further demonstrate how relative intracranial hypotension may result in anatomical shifts of orbital contents. Moreover, our case can serve as an example of how anomalies in cranial anatomy may predispose patients to such outcomes.

While our patient did have a new onset right-sided dacryoadenitis upon presentation, we feel her posterior globe displacement was more strongly associated with the shift of intracranial pressure, as opposed to this glandular inflammation. To our knowledge and from lack of evidence in the literature, dacryoadenitis and acute enophthalmos do not have a known correlation. Contrastingly, dacryoadenitis can be associated with proptosis.¹³ This may explain her mildly proptotic state once the more appreciable low ICP-induced enophthalmos had resolved. Though her microbiology and inflammatory laboratory results were negative, further studies including lacrimal gland biopsy may have shed light on the cause of this likely incidental finding.

It is difficult to conclude whether or not there was a small leak, which could have led to a CSF volume loss greater than the measured 13 mL. Orthostatic headache following LP can serve as a clinical tool when assessing for intracranial hypotension.¹⁴ Though our patient presented with a right-sided headache after the fluoroscopic-guided LP, it was unilateral and persistent regardless of posture, which can perhaps be attributed to the acutely distorted orbital anatomy. Unfortunately, the opening pressure were not measured during the LP, which could have helped in the assessment of this patient’s predisposition to intracranial hypotension. The remainder of the work-up demonstrated no other complications pathognomonic with CSF leakage.

Conclusion

To the best of our knowledge, this is the first case report of a patient with NF-1 related sphenoid wing dysplasia developing enophthalmos following an LP. Our observations have allowed us to depict enophthalmos as a rare, yet reasonably anticipated side effect of LP in this patient population. Thus, caution should be taken when caring for these patients particularly in the setting of LP and other instances of ICP shifts.

Funding Statement

The authors reported there is no funding associated with the work featured in this article.

Disclosure statement

No potential conflict of interest was reported by the authors.

Patient consent

The patient orally consented to the utilised images and the publication of this case. This case does not present any information which could lead to the identification of the patient.