Back to Program

PURPOSE:
Our goal was to develop an updated and comprehensive algorithm for reconstruction following skull base resection based on data from the 10-year period that followed our initial report.
METHODS:
Subjects included 250 patients undergoing skull base reconstruction from 2000 to 2009. Reconstructive outcomes, including results following facial nerve repair and static reanimation, were analyzed.
RESULTS:
Reconstructions included 39 local or regional pedicled flaps and 211 microvascular free flaps. Free flaps were usually selected over pedicled flaps for patients with a history of prior surgery, radiation, and chemotherapy (p=0.003, p<0.001, p=0.04, respectively). Skull base reconstructions were performed for 36 (14%) Region I (anterior) defects, 39 (16%) Region II (middle) defects, 124 (50%) Region III (posterior or lateral) defects, and 51 (21%) defects that involved more than one region (Regions I+II or II+III). Pedicled flaps were used most frequently in Region III (76% of pedicled flaps). The most common pedicled flap was the temporalis muscle flap, which was used in 90% of pedicled flap reconstructions. The most common free flaps were the anterolateral thigh and rectus abdominis myocutaneous free flaps, which were used in 62% and 18% of free flap reconstructions, respectively.
Complications occurred in 30% of patients overall. Risk factors for complications included: age>60 years (p<0.0001), one or more major medical comorbidities (p<0.001), prior surgery (p<0.0001), prior radiation (p<0.0001), and preoperative chemotherapy (p<0.0001). There were no significant differences in the rate of overall complications between patients reconstructed with a pedicled flap and patients reconstructed with a free flap (p=0.70). The total operative time was shorter for cases involving reconstruction with regional flaps compared to free flaps (7.9 hours vs. 11.2 hours, p<0.0001), as were intensive care unit (ICU) and hospital stays (1.1 days vs. 3.3 days, p<0.0001, and 5.0 days vs. 8.9 days, p<0.0001, respectively). The mean follow-up time was 30.2 months.
The facial nerve was resected in 89 patients (36%). A nerve graft repair was performed in 30 patients. At least 70% of repaired facial nerves showed some sign of recovery by 6 months. Recovery was less likely in the setting of preoperative weakness, postoperative radiation, and age ≥60 years (odds ratio of 0.38, 0.44, and 0.75, respectively) but not significantly so (p=0.60, p=0.59, and p=1.00, respectively). Static reanimation procedures were performed in 82 patients (92% of those with facial paralysis) with a 2% rate of revision surgery for complications or inadequate correction.
CONCLUSION:
Based on the largest series of skull base reconstructions to date, we recommend using regional flaps for small defects based on minimal donor site morbidity, and shorter operative times, ICU, and hospital stays, particularly in Region III. For extensive defects, and in cases involving prior surgery, radiation, or chemotherapy, free flaps are preferred. The anterolateral thigh and rectus abdominis myocutaneous free flaps were useful for large defects, especially those involving multiple regions and when multiple skin paddles were required. Facial nerve repair should be attempted in every patient whenever feasible, even in the setting of preoperative weakness, anticipated postoperative radiation, or advanced age.