2026 Volume 13 Pages 161-165
Sphenoid wing dysplasia is a recognized diagnostic feature of neurofibromatosis type 1, yet it rarely occurs in isolation. We present the case of a girl who had pulsatile exophthalmos since birth and was diagnosed with sphenoid wing dysplasia at 1 year and 2 months of age, with no clinical or familial evidence of neurofibromatosis type 1 identified during long-term follow-up. Progressive herniation of the temporal lobe into the left orbit prompted surgical intervention at age 12. Using a 3-dimensional printed skull base model, a patient-specific, hand-bent titanium mesh was preoperatively fabricated and implanted via a frontotemporal approach to reconstruct the superior orbital wall and restore separation between the cranial and orbital compartments. Postoperatively, the pulsatile exophthalmos resolved completely, and a transient abduction deficit recovered within 6 months. Follow-up imaging confirmed stable mesh positioning and durable compartment separation, with satisfactory cosmetic results maintained at 2 years. This case expands the clinical spectrum of sphenoid wing dysplasia by illustrating a neurofibromatosis type 1-negative phenotype and supports 3-dimensional model-assisted, patient-specific titanium mesh reconstruction as an effective strategy for correcting spheno-orbital defects and eliminating pulsatile exophthalmos. Accumulation of similar cases is needed to refine surgical indications, optimize timing, and assess the long-term durability of implant materials.
Sphenoid wing dysplasia (SWD) is a distinctive osseous abnormality and a diagnostic criterion for neurofibromatosis type 1 (NF1),1) occurring in approximately 3%-11% of affected individuals.2) By contrast, isolated SWD without clinical or familial evidence of NF1 is rare.3-6)
Progressive expansion of the orbital and middle cranial fossae caused by this dysplastic process facilitates herniation of the temporal lobe into the orbit, presenting clinically as pulsatile exophthalmos. Because ongoing herniation can lead to visual impairment and aesthetic deformity, surgical reconstruction of the bony defect is often indicated.4,7)
We describe a unique case of NF-1-independent SWD, addressed through implantation of a 3-dimensional (3D) model-assisted, patient-specific titanium mesh.
A Japanese girl was diagnosed with SWD at 1 year and 2 months of age because of left pulsatile exophthalmos present since birth, which paradoxically appeared as enophthalmos during hyperventilation episodes (e.g., crying), accompanied by a soft mass in the left temporal region. No other NF1-related manifestations, including café-au-lait spots, Lisch nodules, or plexiform neurofibromas, were observed. The patient had no family history of NF1. Genetic counseling regarding the benefits of molecular analysis for NF1 gene variants was offered on multiple occasions; however, the family declined genetic testing after careful consideration, citing concerns about psychosocial implications for the child.
The patient was followed with periodic magnetic resonance imaging, which demonstrated progressive herniation of the temporal lobe into the orbit (Figure 1a-c). At age 12, progressive worsening of exophthalmos over the preceding several years, combined with the patient's increasing awareness of facial asymmetry and growing psychological readiness for surgery, prompted referral for surgical repair. Visual acuity and extraocular movements remained intact at this time. Computed tomography (CT) revealed a large orbital roof defect measuring 3.3 × 3.8 cm and a separate pterional skull defect measuring 1.8 × 3.8 cm (Figure 1d-f). The distance between the temporal orbital rim and corneal apex measured 20.5 mm on the right and 14.5 mm on the left.
Preoperative imaging.
a-c: Serial axial T2-weighted MRI at ages 1 (a), 6 (b), and 11 years (c), demonstrating progressive temporal lobe herniation into the left orbit.
d-f: 3D-CT at age 12 showing dysplastic sphenoid bone with extensive defects in the middle cranial fossa floor and posterolateral orbital wall (frontal, lateral, inferior views).
3D-CT: 3-dimensional computed tomography; MRI: magnetic resonance imaging
A 3D-printed skull base model was generated from CT DICOM data, and a titanium mesh (0.6-mm thickness; Muranaka Medical Instruments Co., Ltd., Osaka, Japan) was hand-bent to match the defect before sterilization (Figure 2a). Under general anesthesia, a lumbar cerebrospinal fluid drain was placed. A zigzag incision was made behind the hairline (Figure 2b), and the skin flap and temporal muscle were elevated with care to avoid injury to the temporal dura at the site of the bony defect (Figure 2c). A frontotemporal craniotomy was performed to expose the posterior orbital wall and the temporal lobe dura. Careful blunt dissection of the temporal dura from the periorbita was conducted to expose the extent of the bone defect. The patient-specific titanium mesh was seated to recreate the superior orbital wall and secured to the anterior cranial fossa with a single screw (Figure 2d). The bone flap was replaced with absorbable plates, and an additional titanium mesh was applied to cover the pterional defect (Figure 2e).
Custom-made titanium mesh preparation and intraoperative findings.
a: 3D-printed skull model and the hand-bent mesh tailored to the orbital roof defect.
b: Patient positioned supine with the head turned 45° contralaterally; zigzag skin incision posterior to the hairline.
c: Operative field after elevation of the skin flap and temporalis.
d: Seating the mesh to reconstruct the superior orbital wall after frontotemporal craniotomy and careful dural periorbital dissection.
e: Final reconstruction with the bone flap secured by absorbable plates and an additional mesh covering the pterional defect.
3D: 3-dimensional
During the first postoperative week, the exophthalmos transiently worsened and was accompanied by keratitis sicca, which resolved spontaneously within the following week. Marked abduction limitation of the left eye was noted; however, it fully recovered within 6 months. The pulsatile exophthalmos resolved completely, with the orbital rim-corneal apex distance measuring 14.8 mm on the right and 15.2 mm on the left at 6 months postoperatively. Follow-up imaging confirmed correct mesh placement and complete separation of intraorbital and intracranial spaces, with a satisfactory cosmetic result (Figure 3). At 2 years postoperatively, the patient remains free from recurrence or new neurological deficits.
Postoperative findings.
a, b: Postoperative 3D-CT (lateral and inferior views) confirming proper placement of orbital-roof and pterional meshes.
c: Axial T2-weighted MRI at 1 year showing resolution of temporal lobe herniation.
d, e: Preoperative (d) and 6-month postoperative (e) facial photographs showing resolution of exophthalmos and satisfactory cosmetic outcome.
3D-CT: 3-dimensional computed tomography; MRI: magnetic resonance imaging
This NF1-negative presentation expands the recognized clinical spectrum of SWD. Although molecular confirmation was not obtained due to family preference, the probability of NF1 in this patient is low based on the following observations: (1) absence of any National Institutes of Health or revised diagnostic criteria for NF1 during more than 10 years of follow-up, (2) no family history of NF1 across 3 generations, and (3) lack of associated features commonly seen in NF1-related SWD, such as plexiform neurofibromas or dural ectasia adjacent to the bony defect. While segmental or mosaic NF1 cannot be entirely excluded without genetic analysis, the isolated nature of the sphenoid abnormality and prolonged clinical surveillance make this diagnosis unlikely. The pathogenesis of truly isolated SWD remains unclear; proposed mechanisms include primary mesodermal developmental anomalies affecting the sphenoid ossification center or yet-unidentified genetic variants influencing cranial base morphogenesis independent of the neurofibromin pathway. Additional case accumulation, ideally with genetic analyses, will be essential to clarify the underlying etiology.
Surgical approaches to SWD can be broadly categorized as transconjunctival8-10) or transcranial.4,5,7,11-16) While the transorbital route is less invasive, the transcranial approach offers several advantages: it provides broader exposure of the bony defect, facilitates visualization and preservation of critical structures such as the optic nerve, and allows more controlled dissection between the periorbita and dura mater. In our case, the transient ocular abduction deficit likely resulted from either mechanical restriction of the lateral rectus muscle due to limited space at the superior orbital fissure or minor traction injury to the abducens nerve during dural-periorbital dissection. Similar transient palsies have been reported13) and generally resolve spontaneously, as observed here.
Choice of reconstruction material is equally critical. Autologous calvarial or rib grafts avoid the risks associated with foreign bodies but involve donor-site morbidity, prolonged operative time, and potential graft resorption, sometimes requiring reoperation.8) Non-custom titanium mesh is durable and does not resorb, but can be challenging to contour precisely. In contrast, a 3D model-assisted, patient-specific mesh offers excellent anatomical conformity, minimizing intraoperative adjustments and effectively restoring the cranio-orbital barrier.7,11,13,15) Long-term issues with titanium mesh include tissue ingrowth through pores and infection;14,17) such risks may be reduced by overlaying lyophilized dura12,13) or using titanium-reinforced porous polyethylene.14) Although favorable outcomes lasting up to 19 years have been reported,12) systematic long-term data remain scarce.
The optimal timing of surgical intervention remains controversial, with published cases ranging from 2 to 30 years of age.10,13) Some advocate early repair to prevent progression;10) however, concerns exist regarding surgical risks in infancy and the potential need for revision surgery due to craniofacial growth.18) In our case, surgery was deferred until age 12 based on several considerations. First, visual function remained intact throughout the observation period, allowing for elective rather than urgent intervention. Second, waiting until near completion of craniofacial growth reduced the likelihood of requiring revision surgery due to implant-growth mismatch. Third, the patient's psychological maturity at age 12 enabled her to participate meaningfully in the treatment decision and to cooperate with postoperative care. This individualized approach underscores the importance of balancing disease progression, anatomical factors, and psychosocial readiness when determining surgical timing for SWD.
Regarding postoperative orbital position, the surgical technique described in this report is not intended to elevate or reposition the orbit itself; therefore, vertical globe asymmetry is expected to persist. However, reduction of exophthalmos may lessen the apparent downward sagging of the previously protruded globe, thereby modestly improving the perceived vertical globe alignment, which may contribute to a satisfactory cosmetic outcome.
This report has limitations. First, the absence of genetic testing precludes definitive exclusion of mosaic NF1, though clinical probability remains low as discussed above. Second, the 2-year follow-up period, while demonstrating stable results, may be insufficient to assess long-term implant performance, particularly given the patient's young age. Continued surveillance is planned to monitor for delayed complications such as implant migration or recurrent herniation.
In conclusion, intracranial reconstruction using a 3D model-assisted, patient-specific, hand-bent titanium mesh restored separation between the orbit and the middle cranial fossa and eliminated pulsatile exophthalmos in NF1-negative SWD. Functional and cosmetic outcomes were durable for 2 years. Additional cases with extended follow-up will help refine indications, optimize timing, and assess the long-term performance of reconstructive materials.
Taku Sugiyama contributed to the conceptualization of the study. Keisuke Ohmae and Taku Sugiyama drafted the manuscript, and Taku Sugiyama critically revised the manuscript. All authors contributed to data acquisition and interpretation. All authors approved the version to be published and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
All authors have no conflict of interest.
Author Miki Fujimura is one of the Editorial Board members of this Journal. This author was not involved in the peer review or decision-making process for this manuscript.
Written informed consent for publication has been obtained from the parents of the patient and the patient herself.
This study was conducted in accordance with the Declaration of Helsinki.
