Surgical margins and primary site resection in achieving local control in oral cancer resections.

OBJECTIVES/HYPOTHESIS Evaluate effectiveness of resection of oral cavity cancer with a standardized approach for margin evaluation. Primary end points were local control and survival. STUDY DESIGN Retrospective, nonrandomized, single institution. METHODS One hundred eight patients who underwent surgery for oral cancer were evaluated using specific anatomical pathology criteria. Frozen section was performed with the surgeon and pathologist agreeing where on the specimen the frozen sections should be taken in most cases. RESULTS Ninety-one patients (84.3%) had frozen sections taken from the specimen, eight from the tumor bed, and nine had none taken at the time of surgery. Overall local recurrence rate was 18.5%, 25% in patients who had margins taken from the tumor bed and 17.6% when taken from the specimen. Twenty-nine patients had margins ≥5 mm, 53 <5 mm and clear, and 14 positive re-resected to negative with local recurrence rates of 3.4%, 26.4%, and 28.6%, respectively. The radial distance of the resection margin was shown to have an impact on overall survival (hazard ratio [HR] = 3.59, 95% confidence interval [CI] = 1.12-11.57), disease-free survival (HR = 7.00, 95% CI = 1.89-25.95), and local recurrence-free survival (HR = 28.80, 95% CI = 3.00-276.82). CONCLUSIONS Assessing margins from the resection specimen rather than the tumor bed consistently predicts local control. There is a statistical improvement in local control, disease-free, and overall survival with increasing radial margin distance from the tumor, and 5 mm should be agreed upon as the definition of a clear resection margin. Frozen sections can be used to revise positive or close resection margins intraoperatively with improved outcomes. LEVEL OF EVIDENCE 4.