Return to search

Surgical management of pharyngoesophageal tumours

Pharyngoesophageal (PE) tumours are tumours involving simultaneously the hypopharynx and the cervical oesophagus. The challenge in its surgical management lies in its deep-seated location behind the manubrium bone in the cervicothoracic region, in close proximity to great vessels in the lower neck and superior mediastinum.

Classically curative surgery is in the form of total pharyngo-laryngo-oesophagectomy (PLO) and gastric pull-up (GPU) via a three-phase one-stage operation. However PLO and GPU is a major undertaking associated with high operative morbidity and reported in-hospital mortality rates of up to 10%.

With a comprehensive preoperative work-up we demonstrated accurate tumour diagnosis and staging, with a 100% negative predictive rate. Together with vigilant postoperative surveillance and compliant follow-up, incidence of synchronous and metachronous tumours were low at 11.9% and 1.7% respectively.

Manubrial resection (MR) provided access to PE tumours in the cervicothoracic region enabling resection under direct vision with adequate resection margins - pharyngo-laryngo-cervico-oesophagectomy (PLCO). The trachea was resected and re-sited as a mediastinal tracheostoma in case of posterior tracheal wall invasion. Paratracheal and paraoesophageal lymph node dissection was performed in case of nodal metastasis. MR provided ample space for reconstruction of the resultant defect. Furthermore, it enabled access to vessels in the superior mediastinum to support microvascular tissue transfer. Intra-thoracic volume changes on maximal inspiration and expiration measured using computed tomography scan did not show significant difference pre- and post- MR. With attention to operative details, MR proved to be safe with minimal functional disturbance.

Free jejunal (FJ) flap was the preferred reconstructive modality as it offered the lowest pharyngocutaneous fistula and anastomotic stricture rates, and donor site morbidities. All patients resumed unrestricted oral diet postoperation. Videofluoroscopic swallowing studies (VFSS) and high resolution manometry (HRM) demonstrated significantly prolonged transit times for all bolus consistencies compared with normal subjects due to asynchronous contractions between the FJ and the oesophageal remnant, presence of retrograde propulsion and residue accumulation within the FJ. However, patients reported significant improvement in swallowing outcomes and associated quality of life (QOL) compared with preoperation (65.3% vs. 42.7%, p=0.02). Majority of patients were able to speak conveniently with a modality of their choice.

MR, PLCO and FJ flap showed significantly lower operative morbidities (58.3% vs. 85.7%, p=0.05), shorter hospital stay (42.5 vs. 50.7 days, p=0.37), and lower in-hospital mortality (8.3% vs. 9.5%, p=0.52) compared with PLO and GPU. None required intensive care unit postoperation. In resecting less, oncological outcomes and survival were not inferior to PLO and GPU. FJ patients were able to resume oral diet sooner than GPU with a higher functional oral intake scale (FOIS) at 6 months (100.0% vs. 92.8%). Shorter transit times for all bolus consistencies were demonstrated in VFSS and HRM of GPU patients due to the lack of contractions within the gastric tube. Swallowing, speech and associated QOL outcomes were comparable between the 2 groups.

In conclusion, MR, PLCO and FJ flap should be adopted in the surgical management of patients with isolated PE tumours. / published_or_final_version / Surgery / Master / Master of Surgery

Identiferoai:union.ndltd.org:HKU/oai:hub.hku.hk:10722/212561
Date January 2015
CreatorsChow, Ling-yu, Velda, 周令宇
PublisherThe University of Hong Kong (Pokfulam, Hong Kong)
Source SetsHong Kong University Theses
LanguageEnglish
Detected LanguageEnglish
TypePG_Thesis
RightsCreative Commons: Attribution 3.0 Hong Kong License, The author retains all proprietary rights, (such as patent rights) and the right to use in future works.
RelationHKU Theses Online (HKUTO)

Page generated in 0.0022 seconds