Background: An ipsilateral neck dissection is mandatory during initial treatment stages II-IV oral carcinomas. However, no consensus exists whether or not to perform an elective contralateral neck dissection.
Methods: Five hundred thirteen consecutive cases of squamous cell carcinoma (269 tongue, 135 floor of the mouth, 44 inferior gingiva, 65 retromolar trigone) were reviewed. Tumor stages were: 69 T1, 227 T2, 217 T3-T4, 263 N0, 250 N1-N3. A total of 563 neck dissections were performed in 448 patients. Univariate and multivariate analysis of risk factors were performed using logistic regression.
Results: Two hundred twenty-three patients (49.8%) had positive nodes in the specimen (182 ipsilateral, 36 bilateral, 5 contralateral). Contralateral neck recurrences occurred in 38 cases (33 not submitted to a contralateral neck dissection initially). Multivariate logistic regression analysis demonstrated that clinical stage (p = .0001), tumor crossing midline (p = .0011), and floor of the mouth involvement (p = .0236) were the most important predictors of contralateral metastasis.
Conclusion: The contralateral side of the neck is a common and potentially preventable site of recurrence in tumors of the oral cavity. The multivariate model obtained discriminates patients with low and high risk (more than 20%) of contralateral metastasis. The application of this mathematical model can be useful for the indication of contralateral neck dissections, because not all tumors crossing midline are associated to a high risk (stages I and II tumors not involving the floor of the mouth) and not all tumors not crossing midline are at low risk (stages III and IV tumors involving the floor of the mouth).