Premoulded Custom Implants for Spheno-Orbital Reconstruction: A Novel Multidisciplinary Approach

Objective: Spheno-orbital hyperostotic tumors require extensive bony resection to achieve the desired cosmetic, visual, and oncological outcome. The method for orbital reconstruction after extensive resection is controversial. We report the use of premolded custom orbital and cranial implants after combined orbital and skull base tumor resection.

Patients and Methods: A series of six patients who underwent resection of spheno-orbital tumors (five meningioma and one fibrous dysplasia) between August 2012 and July 2014. All patients underwent helical 3D CT planning imaging and the intended resection was defined in a 3D model to remove all involved bone including the orbital rim where necessary. The reconstructive implant is then fashioned based on the contours of the contralateral side. The implant material was dependent on the anticipated need for adjuvant radiotherapy either polyetheretherketone (PEEK) or titanium. Complete resection was planned in all patients, including removal of the involved orbital rim, anterior clinoidectomy, and optic canal resection where needed. Surgery was undertaken by a combined team of neurosurgeons, ophthalmic, and maxillofacial surgeons.

Results: All patients presented with visual deficit and proptosis. Five cases were female and the median age at the time of operation was 49.5 years. Current median follow-up is 8 months (range, 2–24 months). Overall, two patients had previous resections of meningioma, representing either with recurrence (one) or failure of the previous prosthesis (one).

At the latest follow-up, two patients had ongoing minor visual symptoms (one has persistent mild proptosis—10 months postprocedure, one patient has mild vertical diplopia—2 months postprocedure which is well controlled with prisms. All others have no ongoing visual symptoms and all patients have a normal optic nerve function.

A gross total resection was achieved in all cases. Overall, three patients had PEEK implants and three patients had titanium. The implants were either a two orthree3 piece reconstruction. There were no intraoperative problems with implant placement. The overall cosmetic outcome was good and there were no cases of visual deficit postoperatively. One patient developed transient ptosis which resolved in 4 weeks. There were no infections or CSF leak.

Conclusion: Radical resection of spheno-orbital hyperostotic tumors can be achieved with good cosmetic outcome and minimal morbidity, using custom premolded rigid implants for reconstruction. The meticulous preoperative 3D planning and multidisciplinary skull base team has provided excellent functional, oncological, and cosmetic outcomes. The majority of these cases were performed on patients with meningioma, a condition which is known to recur. An advantage of PEEK implants in this context is reduced artifact on subsequent imaging and reduced radiotherapy beam scatter where adjuvant treatment is needed.