Advantages of the Endoscopic Endonasal Approach versus the Transoral Approach for Odontoid Resection

Background: The transoral approach has been the predominant technique for odontoid resection over the past three decades. This approach passes through the posterior oral pharynx and typically requires temporary restriction of oral intake requiring a nasogastric feeding tube. The endoscopic endonasal approach allows resection of the odontoid through the posterior nasopharynx. This approach may allow sooner resumption of oral feeding postop, simplify airway management, and decrease ICU stay.

Objective: To compare the transoral versus the endoscopic endonasal approach for odontoid resection.

Methods: A retrospective chart review of all patients undergoing odontoid resection followed by posterior spinal fusion between 1997 and 2018 by the senior author was conducted. Sixteen patients were identified: 7 had an endoscopic endonasal resection and 9 had a transoral resection. All but one patient underwent posterior fusion surgery the same day as the odontoid resection. The medical record was reviewed for clinical and radiographic outcomes. Statistical significance was p < 0.05.

Results: The median age of patients at surgery was 44.5 years. Chiari I malformation with a retroflexed odontoid (31%) and basilar invagination (31%) were the most common diagnoses. Gait instability/quadriparesis (44%) and numbness/tingling in extremities (44%) were the most common presenting symptoms. A Dobhoff feeding tube was placed on 8 patients (89%) who underwent a transoral resection and one patient (14%) who underwent an endoscopic resection. The Dobhoff was replaced by a PEG feeding tube 16 days postop on 1 patient who underwent a transoral resection. Complications occurred in 4 patients (44%) who underwent a transoral resection. They included infection surrounding the odontoid resection cavity, postop pneumonia with pulmonary embolism, one revision of the posterior fusion hardware, and one reintubation with replacement of the Dobhoff feeding tube. One complication occurred in one patient (14%) who underwent endoscopic surgery, which was the requirement of two additional anterior surgeries on the residual C2 vertebrae due to cranial settling 8 years after the first surgery.

Postoperative patients who underwent endoscopic surgery spent significantly less time in the ICU compared with patients who underwent transoral surgery (2.2 vs. 4.2 days; p = 0.02). Patients who underwent endoscopic versus transoral surgery tended to spend less time intubated (median 1 [1–1.75] vs. 3 [2–5] days) and tended to have a shorter LOS (median 6 [5–6] vs. 8 [7.5–14] days), though this did not reach statistical significance. Neurological outcome was classified as improved, unchanged, or worse. Of the patients who underwent a transoral approach, 3 patients (33%) were improved, 4 patients (44%) were unchanged, 1 patient (11%) was worse, and 1 was lost to follow-up. Of the patients who underwent an endoscopic approach, 6 patients (86%) were improved, and 1 patient (14%) was unchanged.

Conclusion: Endoscopic endonasal odontoid resection through the nasopharynx warrants further investigation but may help reduce length of stay, intubation time, length of ICU stay, the need for a feeding tube, complication rates and improve patient outcomes.