Outcomes of orbital exenteration for craniofacial lesions

BACKGROUND Orbital exenteration (OE) is an ablative procedure used in the management of malignancies of the orbit of either primary or secondary origin. Publications evaluating this procedure have suffered from small patient numbers, heterogeneity of pathologies, and poor patient follow-up. The purpose of this study was to assess patient outcomes in a large cohort of patients undergoing OE at a tertiary cancer center. METHODS A retrospective review was conducted of 180 consecutive patients who underwent OE at the authors' institution. Overall survival (OS) was the primary end point measured in the study. Time to locoregional recurrence (progression-free survival [PFS]) and disease-free survival were secondary end points. RESULTS Between the years 1993 and 2011, 180 consecutive patients received OE for craniofacial malignancy at the authors' institution. The median follow-up for the cohort was 9.7 years (116 months). The median OS was 73 months, and the median PFS was 96 months. The presence of perineural invasion was associated with shorter OS (P = .01) and PFS (P < .01). Magnetic resonance imaging was predictive of perineural invasion (P < .01). Positive margins were associated with shorter PFS than negative margins (P < .01) but with no change in OS (P = .15). The overall complication rate was 15%. The major complication rate (Clavien-Dindo 3b or greater) was 2.8% (n = 5), and there was 1 death observed (0.6%). CONCLUSIONS Used judiciously in the setting of a multidisciplinary management plan, OE for tumor control is a safe therapy. LAY SUMMARY Between the years 1993 and 2011, 180 consecutive patients received orbital exenteration for craniofacial malignancy at the MD Anderson Cancer Center. The median follow-up for the cohort was 9.7 years. The presence of perineural invasion was associated with shorter overall survival (P = .01) and progression-free survival (P < .01). Magnetic resonance imaging was predictive of perineural invasion (P < .01). Positive margins were associated with shorter progression-free survival than negative margins (P < .01). The overall complication rate was 15%. The major complication rate (Clavien-Dindo 3b or greater) was 2.8% (n = 5).