Management of malignant submandibular gland tumour: 20‐year experience in a single institution

Abstract

AbstractAimOur primary objective is to establish the epidemiology of submandibular gland malignant tumour in Hong Kong. Our second objective is to investigate the correlation of positive surgical margin and survival in patients with primary malignant submandibular gland tumour. Our third objective is to review the accuracy of fine needle aspiration in histological diagnosis.Patients and MethodsA retrospective study was conducted. Patients who underwent surgery and were discharged from Queen Mary Hospital and Tung Wah Hospital from December 31, 1997‐December 31, 2017 were included. Data was analysed by Microsoft Excel and IBM SPSS Statistics 25.ResultThere were 239 patients with submandibular gland removal between 1997‐2017 in the Queen Mary and Tung Wah Hospitals. Among this group, 32 patients were diagnosed with malignant submandibular gland. Data of these 32 patients were reviewed and analysed. The male‐to‐female ratio was 3:1. The median age at diagnosis was 55.5. Mean follow up time was 44.8 months. Patients presented with various histological subtypes of submandibular gland tumour, the most common being lymphoepithelioma‐like carcinoma and adenocystic carcinoma (28.1% each), followed by carcinoma ex pleomorphic adenoma (12.5%), mucoepidermoid carcinoma (9.4%), salivary duct carcinoma (9.4%), undifferentiated carcinoma (6.3%) and squamous cell carcinoma (6.3%). The tumour histological subtype was unknown in 40.6% of patients at the time of surgery. The positive surgical margin rate was 50%. The overall 5‐year locoregional control rate, disease‐free survival and overall survival were 81.83%, 75.14% and 76.57% respectively.ConclusionLack of preoperative histological diagnosis is identified as the only predictor of positive surgical margin, OR 7.22, P < 0.0163. It is concluded that TNM staging and lymphovascular invasion are the two significant prognostic factors in management of malignant submandibular gland tumour. However, we cannot draw a statistically significant conclusion on whether positive surgical margin would affect the overall survival or locoregional control or disease‐free survival. (P < 0.997, P < 0.74, P < 0.950 respectively).