Introduction: There are multiple studies looking at swallow function after transoral surgery (TOS) for oropharyngeal carcinoma. However, most do not include locally advanced (T3 and T4) tumors nor delineate the association between the extent of the defect and the impact of reconstruction on swallow function.
Methods: A retrospective review was performed of patients who underwent definitive transoral resection (TORS, TLM, or loupe magnification) of locally advanced oropharyngeal cancers at a single institution from 1997 to 2016. Patients who underwent definitive surgical therapy with at least 1 year of postoperative follow-up were included in the study. The primary outcome measure was swallow function, as measured by the Functional Outcomes Swallowing Scale (FOSS) at 1 year after surgery. Patients with a FOSS score 0-2 were classified with acceptable swallow function, and those with 3-5 were classified with poor function. Operative reports were carefully reviewed, and the extent of resection for each patient at each oropharyngeal subsite (ie: soft palate, lateral pharyngeal wall, base of tongue (BOT)) was documented and graded by % resected and depth of resection. Extirpation of additional structures, exposure of carotid artery, as well as reconstructive maneuvers (local, regional or free flap) were also recorded. The following variables were analyzed pertaining to their impact on swallow function: tumor stage, comorbidity, extent of surgical resection based on oropharyngeal subsite, reconstruction, neck dissection, and adjuvant therapy.
Results: Eighty-three patients met inclusion criteria. In this cohort, 26 patients (31.3%) underwent reconstruction: 18 local flaps, 2 regional flaps, and 4 free flaps. Fifty-six patients (67.5%) had acceptable swallowing at 1 year. Out of all subsites, only BOT resection >50% compared to <50% was significantly associated with poor swallowing (50% vs 22.6%; p = .011). Neither extent of defect, resection of other subsites, epiglottis resection, medial pterygoid resection, pharyngotomy, hypoglossal nerve transection, nor T-stage statistically impacted swallow outcomes. The majority of patients (n = 75; 92.6%) received adjuvant therapy, and among them, 27 (33.3%) had poor swallow outcomes. Of the 6 patients who did not receive adjuvant therapy, none had poor swallow outcomes (35.1% vs 0%, p = .077). Flap reconstruction was not statistically associated with poor swallow function (0.8% difference, CI -39.3 – 37.7). Of the 30 patients with >50% BOT resection, 15 (50%) had acceptable swallow outcomes. The 15 patients with >50% BOT resection and poor swallow function were more likely to have epiglottic resection (33.3% vs 6.7%), bilateral neck dissections (78.9% vs 50%), T4 primary tumor (73.3% vs 33.3%), and adjuvant chemoradiation (60% vs 26.7%).
Conclusion: This study is the largest to date to provide a swallow function analysis for advanced T3 and T4 oropharyngeal cancers removed transorally that is based on a detailed defect size analysis. Base of tongue resection >50% was associated with poor long-term swallow outcomes. Resection of other surgical subsites and reconstruction were not found to statistically impact swallow function.