The characteristics of the tumors are
The characteristics of the tumors are shown in Table 2. The tumor morphology indicated 69 (74%) exophytic tumors and 24 (26%) endophytic tumors. Forty-five patients (48%) had anterior commissure invasion, and 30 (32%) had tumors extending to the contralateral true vocal fold. Twenty-two patients (24%) had vocal process invasion, and 7 (8%) had tumor invasion to the body of the arytenoid cartilage. Eighteen patients (19%) had vocal fold limitation. Twenty-eight patients (30%) had tumors extending to the ventricle, 9 (10%) had false vocal folds, and 18 (19%) had subglottic invasion. Seventeen patients (18%) had paraglottic invasion, and 8 (9%) had perichondrium of thyroid cartilage invasion. Fourteen patients (15%) had positive surgical margin (including carcinoma in situ and severe dysplasia).
The oncologic results after primary TLM are shown in Table 3. In a median follow-up of 35 months (range 8–100 months), 11 patients (12%) had local recurrence. Five patients (5%) had second primary malignancies, including four in the head and neck area and one in the lung. The 5-year local control, overall survival, and disease-specific survival rates were 87%, 95%, and 96%, respectively. Ten of the 11 patients with local recurrence underwent salvage treatment, and one patient received only palliative tracheotomy due to advanced age. The 10 patients were salvaged with TLM only (n = 4) or with postoperative RT (n = 6). One patient had total laryngectomy due to paratracheal lymph node recurrence. Ninety-one of the 93 patients still had a preserved W 54011 at last follow-up, for a final laryngeal preservation rate of 98%.
Table 4 shows the predictors of local recurrence. In univariate analysis, the following were significant factors for local recurrence: primary T stage, previous MLS, salvage treatment, type of cordectomy, tumor morphology, limitation of vocal folds, anterior commissure, arytenoid cartilage, false vocal fold, paraglottic space, perichondrium of thyroid cartilage invasion, positive surgical margin, and difficult exposure during MLS. In multivariate analysis (Table 5), the following were independent factors for local recurrence: arytenoid cartilage invasion, DLE, previous MLS, positive surgical margin, and endophytic tumors. Bilateral vocal fold lesions was not a significant factor of local recurrence (91% vs. 78%, p = 0.104). Patients with arytenoid cartilage extension had a 6.5-fold higher local recurrence rate than those without invasion. Patients with DLE had a 4.6-fold higher rate of local recurrence than those without DLE, and those with previous MLS had a 3.1-fold higher rate of local recurrence than those without previous MLS. Patients with a positive surgical margin had a 2.7-fold higher rate of local recurrence than those with a negative margin, and those with an endophytic tumor had a 2.6-fold higher rate of local recurrence than those with an exophytic tumor.
Discussion In the current study, TLM provided satisfactory oncologic results for patients with early and select cases of moderately advanced glottis cancer. The 5-year local control, overall survival, and disease-specific survival rates were 87%, 95%, and 96%, respectively. Although 11 patients (12%) developed local recurrence after TLM, 91 of the 93 patients (98%) had final laryngeal preservation after salvage treatment. Our results are comparable with a previous report from Schrijvers et al., who reported a higher laryngeal preservation rate after TLM vs. RT for patients with T1a glottis cancer. TLM can be used multiple times in cases of local recurrence, and postoperative RT can still be administered later in those patients with positive surgical margin. Our results showed that arytenoid cartilage extension, DLE, previous MLS, positive surgical margin, and an endophytic tumor were independent predictors of local recurrence in patients with glottic cancer who underwent TLM. The other factors, such as anterior comissure invasion and bilateral vocal fold lesions, were not independent predictors in patients who underwent TLM if exposure was good during operation. Patients with arytenoid cartilage extension had a 6.5-fold higher rate of local recurrence than those without invasion, and the 5-year local control rates were 38% and 91% in patients with or without arytenoid cartilage extension, respectively. Once tumors invade into this area, they grow close to the cricoarytenoid joint and may impair the mobility of the arytenoid cartilages, making it difficult to complete the resection. Peretti et al. also mentioned that posterior paraglottic involvement in close proximity to the cricoarytenoid joint was generally associated with local persistence of the tumor. Moreover, failure in this area can dramatically jeopardize the laryngeal preservation rate after partial laryngectomy. Special techniques and experience are needed to improve the surgical field for resection, and surgeons are recommended to employ wide resection of the primary tumor as well as the inner perichondrium of the thyroid lamina and the cricoid arch. The piecemeal technique may improve exposure of the posterior paraglottic space, and facilitate tumor depth and margin estimation. Postoperative RT is indicated for patients without adequate surgical margin.