Prospective validation of the Martinez-Gimeno clinicopathologic scoring system (MGSS) for evaluating risk of cervical lymph node metastases of squamous cell carcinoma of the oral cavity

Abstract

Background: The prediction of neck metastasis from oral squamous cell carcinoma is an unresolved issue. The Martinez-Gimeno Scoring System (MGSS) is an index designed as a predictive aid that may be implemented after analysis of the tumor resection.

Methods: A double-blind cohort study was designed and carried out in a tertiary hospital. Seventy-nine consecutive patients diagnosed with squamous cell carcinoma of the oral cavity requiring local resection and neck dissections were included. After tumor resection, the MGSS was applied, and patients were then classified into one of four groups on the basis of the MGSS score: group I, 7 to 12 points; group II, 13 to 16 points; group III, 17 to 20 points; and group IV, 21 to 30 points. After this, pathologic analysis of the neck specimen was performed to detect metastasis. The results, histopathologic analysis of the neck specimens, and previously tabulated MGSS scores were then compared, and on the basis of their results an estimation of the predictive validity of MGSS of neck metastases was obtained. A correlational analysis was performed, comparing metastases to a set of predictive factors, including the MGSS score. A logistic binary regression model that used metastases as the dependent variable was adjusted in an attempt to estimate odd ratios for these factors.

Results: Twenty-six cases displayed metastases. The metastatic nodes measured from 0.5 to 7 cm and were smaller than 1.5 cm in 46% of the cases. The rates of metastases in the four groups were as follows: group 1, 0%; group II, 21%; group III, 50%; and group IV, 67%. MGSS had a sensitivity of 100% (95% confidence interval [CI], 98% to 100%) for predicting metastases from oral squamous cell carcinoma, a specificity of 55% (95% CI, 44% to 66%), an overall efficiency of 73% (95% CI, 63% to 83%), a positive predictive value of 59% (95% CI, 48% to 70%), and a negative predictive value of 100% (95% CI, 98% to 100%). The correlation between various predictive factors and the presence of metastases showed a value of r = .87 (p = .001) with the MGSS score. No correlation was found between the MGSS score and the size of the metastatic lymph node. In the logistic regression model, the MGSS score displayed an odds ratio of 3.5 (95% CI, 1.9-6.3; p = .00001) for the presence of metastases in patients with squamous cell carcinoma of the oral cavity.

Conclusion: The MGSS is a useful index for the prediction of neck node metastases that is applied at the start of the analysis of the tumor resection in patients with squamous cell carcinoma of the oral cavity.