br Thirty five patients had
Thirty-five patients had positive Kainic acid nodes in the postoperative pathological report, 16 of whom were in the N1 stage and 19 in the N2 stage. During a median 13-month follow-up period (range 7–39 months), 15 patients developed cervical lymph node metastasis,
and 2 patients had local recurrence. Eleven patients died, and no pa-tients were lost to follow-up. The OS rate was 88.1%, and the overall DSS rate was 93.1%. The cutoﬀ value of the MRI depth that could identify nodal metastasis was 8 mm. The cutoﬀ values for OS and DSS were both 11 mm (Fig. 7a-c). In unadjusted Cox proportional hazard models, tumor DOI > 8 mm, age > 60 years, sex, tumor site, T stage, pathological nodal status, and tumor morphology were included to determine the independent eﬀect on survival. The variables without significance, which included sex and age > 60 years, were removed from the final analysis (Table 6). After adjusting for these variables, only the tumor site was found to be predictive of both OS (p = 0.021) and DSS (p = 0.02), followed by DOI > 8 mm (OS, p = 0.085; DSS, p = 0.086). Dorsal tongue was found to be an individually significant factor associated with survival (OS, p = 0.026; DSS, p = 0.031) (Table 7).
The AJCC staging system has been widely adopted and applied. The latest eighth edition of the AJCC staging system adds the depth of tumor invasion to the T stage of oral cancer as a staging criterion. A clinically examined or pathologically measured DOI greater than 5 mm is stage T2, and a depth greater than 10 mm is stage T3. The depth of tumor invasion can be obtained by preoperative imaging examination and analysis of intraoperative and postoperative pathological sections. However, analyzing the consistency among invasion depths at the three time points and whether the same grading criteria apply to each time point yields interesting results. False-positive results on clinical images (such as MRI) and atrophy of postoperative pathological sections have been reported . The former can lead to a decline in staging, and the latter can lead to excessive staging. The consistency and the specific diﬀerences between the three time points have become a key issue and represent the main problem that tubal ligation study aims to address. One of the core concepts of the AJCC staging system is that “the general TNM principle of selecting the less advanced attribute should always be ob-served when the clinician
MRI is an ideal method for soft tissue imaging and has been widely used for the preoperative evaluation of tongue cancer patients. However, −4.6- to 3.19-mm false-positive results have been reported, which should warrant greater attention because the clinical staging criteria use thresholds of only 5 and 10 mm. Some scholars have re-searched this problem. Yesuratnam et al.  designed a prospective study comparing preoperative MRI measurements of tumor thickness with measurements obtained from postoperative pathological sections in 81 patients. The results showed that the mean diﬀerence between histology and T2-weighted MRI measurements was 3.19 ± 4.87 mm and that the diﬀerence between histology and T1 postcontrast MRI measurements was 2.99 ± 4.41 mm. Kwon et al.  retrospectively
Fig. 3. Bland-Altman plots showing the agreement between MRI and intraoperative and postoperative pathological sections for measuring the infiltrating depth of a tongue tumor. Solid lines represent the bias between the two measurement methods, and dotted lines represent the 95% confidence intervals for the diﬀerences. MRI: magnetic resonance imaging.
Table 4 Factors associated with the diﬀerence between MRI and intraoperative and postoperative pathological sections in measuring the infiltrating depths of tongue tumors.