Background: Traditionally, the standard of care for sinonasal and ventral skull base malignancies involving the orbit was radical excision with orbital exenteration. With the increased visualization during endoscopic endonasal resection of these tumors, the potential for orbital preservation while maintaining appropriate oncologic outcomes is increasingly recognized. However, it is crucial to identify cases in which comparable outcomes can be achieved with orbit-sparing procedures. Secondary to the differing definitions of orbital invasion in the literature, vast array of pathology and varying neoadjuvant/adjuvant treatment options, no widely accepted treatment paradigm currently exists.
Objective: To retrospectively review oncologic outcomes and orbital preservation status in patients with sinonasal malignancies with orbital involvement to identify clinical predictors of recurrence and exenteration and to correlate preoperative MRI/CT findings that predict oncologically sound orbital preservation.
Materials and Methods: The sinonasal malignancy and exenteration databases were queried at a tertiary academic institution. Exclusion criteria included benign pathology, surgical biopsy or decompression of lesion, primarily external approach or age <18 years. Data regarding prior treatment, intraoperative appearance of invasiveness, pathology, margin status, adjuvant treatment, orbital function, and recurrence was collected. In addition, the preoperative CT and MRI were reviewed by a neuroradiology attending and a radiology resident who were both blinded to oncologic and orbital preservation status. The imaging was independently evaluated for characteristics of orbital involvement such as erosion of lamina papyracea, periorbita, extraocular muscles (EOM), and loss of fat planes.
Results: A total of 82 patients were identified with 64 patients undergoing endoscopic endonasal resection with orbital preservation while 18 underwent primary orbital exenteration. Of the 64 patients who underwent EEA with orbital preservation, 54.7% patients ultimately had orbital preservation without disease, 23.4 and 14.1% had orbital preservation with distant and local disease recurrence at last follow-up, respectively. Exenteration was ultimately performed in 3% of this population. Univariate analysis was performed on CT and MRI characteristics using orbital preservation without disease, orbital preservation with disease or exenteration as endpoints. Compared with no or possible involvement of EOM, definite EOM involvement was associated with patients undergoing exenteration on both CT (p = 0.01) and MRI (p = 0.046). In contrast, definite involvement of lamina papyracea and periorbita on MRI was not associated with exenteration and in fact the majority of these patients (57.7 and 57.9%, respectively) had orbital preservation without disease.
Conclusion: There is a trend in the field toward orbital preservation as opposed to orbital exenteration in the correct clinical setting. In our orbital preservation cohort, the majority of patients at the time of last visit were without disease. Radiologic findings of definite EOM, intraconal fat and orbital apex involvement were predictive of patients who would undergo exenteration; however, interestingly, lamina papyracea erosion and periorbital involvement were not associated with exenteration suggesting that adequate margins can be achieved endoscopically when lamina papyracea and/or periorbita are involved. With a greater understanding of disease-specific factors including preoperative radiologic invasiveness, pathology, and adjuvant treatment, the correct patients can be chosen as good candidates for orbital preservation surgery.
