Objectives: To investigate preoperative MRI evaluation of the features of the mylohyoid muscle (MM) predictive of its infiltration in oral squamous cell carcinoma (OSCC) treatment planning, defining the most appropriate sequences to study its deep extension into the floor of the mouth (FOM). Materials and methods: We applied a 7-point score to retrospectively evaluate preoperative imaging of patients who underwent surgery for OSCC over 11 years. The results were compared with histopathological findings using Spearman's rank coefficient. Receiver operating characteristic curves were employed to assess the score's ability to predict MM infiltration, determining optimal thresholds for sensitivity, specificity, and predictive values. The Mann-Whitney U-test confirmed that infiltration judgments did not overlap around this threshold. Cohen's K statistical coefficient was used to evaluate the interobserver agreement. Results: Fifty-two patients (mean age 66.4 ± 11.9 years, 36 men) were evaluated. Histopathological examination found MM infiltration in 21% of cases (n = 11), with 90% classified in the highest Score categories. A score > 4 proved to be the best cut-off for predicting the risk of MM infiltration, with a sensitivity of 91% (CI: 0.57-0.99), specificity 61% (CI: 0.45-0.76), PPV 38% (CI: 0.21-0.59), and NPV 96% (CI: 0.78-0.99). At the subsequent single-sequence assessment, the TSE-T2wi had the highest diagnostic accuracy, with sensitivity 90% (CI: 0.57-0.99), specificity 70% (CI: 0.53-0.82), PPV 45% (CI: 0.25-0.67), and NPV 96% (CI: 0.80-0.99). Conclusion: The 7-point score is a promising predictor of safe surgical margins for MM in OSCC treatment, with the particular benefit of T2-weighted sequences. Clinical relevance statement: Our scoring system for tumor infiltration of MM, which is easy to use even for less experienced radiologists, allows for uniformity in radiological language, thereby ensuring crucial preoperative information for the surgeon. Key points: The relationship of the MM to an oral lesion may impact surgical planning. As the score increases, there is a greater incidence of infiltration in the MM. Our score system improves radiologists' reporting for MM involvement by tumor.
MRI-based assessment of the mylohyoid muscle in oral squamous cell carcinoma, a 7-point scoring method
Radin, E.;Marcuzzo, A. V.;de Groodt, J.;Degrassi, F.;Ramella, V.;Tirelli, G.;Ukmar, M.;Cova, M. A.
2024-01-01
Abstract
Objectives: To investigate preoperative MRI evaluation of the features of the mylohyoid muscle (MM) predictive of its infiltration in oral squamous cell carcinoma (OSCC) treatment planning, defining the most appropriate sequences to study its deep extension into the floor of the mouth (FOM). Materials and methods: We applied a 7-point score to retrospectively evaluate preoperative imaging of patients who underwent surgery for OSCC over 11 years. The results were compared with histopathological findings using Spearman's rank coefficient. Receiver operating characteristic curves were employed to assess the score's ability to predict MM infiltration, determining optimal thresholds for sensitivity, specificity, and predictive values. The Mann-Whitney U-test confirmed that infiltration judgments did not overlap around this threshold. Cohen's K statistical coefficient was used to evaluate the interobserver agreement. Results: Fifty-two patients (mean age 66.4 ± 11.9 years, 36 men) were evaluated. Histopathological examination found MM infiltration in 21% of cases (n = 11), with 90% classified in the highest Score categories. A score > 4 proved to be the best cut-off for predicting the risk of MM infiltration, with a sensitivity of 91% (CI: 0.57-0.99), specificity 61% (CI: 0.45-0.76), PPV 38% (CI: 0.21-0.59), and NPV 96% (CI: 0.78-0.99). At the subsequent single-sequence assessment, the TSE-T2wi had the highest diagnostic accuracy, with sensitivity 90% (CI: 0.57-0.99), specificity 70% (CI: 0.53-0.82), PPV 45% (CI: 0.25-0.67), and NPV 96% (CI: 0.80-0.99). Conclusion: The 7-point score is a promising predictor of safe surgical margins for MM in OSCC treatment, with the particular benefit of T2-weighted sequences. Clinical relevance statement: Our scoring system for tumor infiltration of MM, which is easy to use even for less experienced radiologists, allows for uniformity in radiological language, thereby ensuring crucial preoperative information for the surgeon. Key points: The relationship of the MM to an oral lesion may impact surgical planning. As the score increases, there is a greater incidence of infiltration in the MM. Our score system improves radiologists' reporting for MM involvement by tumor.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
