Abstract
Objective:
Odontoid resection may be indicated in ventral cervicomedullary compression. Surgical approaches have traditionally utilized a transoral technique, which may carry significant morbidity. Endonasal odontoid resection represents a new technique that provides direct, minimal access to the odontoid process with maximal preservation of anatomic structures. Limited evidence-based data exist to guide clinical decisions.
Methods:
We performed a MEDLINE database search for the years 1950–2010 to identify relevant studies. Comparisons were made between endonasal and transoral cohorts for patient and tumor characteristics and postoperative complications. Clinical outcomes were classified into improved, stable, or deteriorated groups. Statistical analyses of categorical variables were carried out using Mann-Whitney U, chi-square, and Fisher's exact tests.
Results:
Forty studies, involving 764 patients, were included. The endonasal cohort had a significantly lower rate of soft palate incision (0.0% vs. 33.6%,
P
= 0.005). There was no difference in complete odontoid resection between cohorts. In the endonasal cohort, there were no cases of postoperative soft palate repair, atlanto-occipital instability, wound dehiscence, CSF leak, or meningitis. Patients in the endonasal group had a shorter time to extubation (
P
< 0.0001) and oral feeding (
P
= 0.0025). There was a significantly higher percentage of patients with improved outcome in the endonasal cohort (100% vs. 73.3%,
P
= 0.009).
Conclusion:
Our systematic meta-analysis provides evidence-based support for the endonasal approach as a safe and effective alternative to traditional transoral surgery for odontoid lesions. Indeed, results in appropriately selected cases appear to be superior for a variety of outcome measures. Further investigations comparing cost and length of stay are warranted.