• 2022-09
  • 2022-08
  • 2022-07
  • 2022-06
  • 2022-05
  • 2022-04
  • 2020-08
  • 2020-07
  • 2018-07
  • br Statistical analysis br The results were mainly presented


    Statistical analysis
    The results were mainly presented with descriptive statistical methods (number and %), because there was a small number of study patients. Survival analyses were conducted to identify potential risk factors for poor outcomes. The primary endpoints were progression-free survival and overall survival. Progression-free and overall survival durations were calculated as the time elapsed from the first day of treatments until the time of disease progression or any deaths, respec-tively. Cases without events by the last clinical follow-up were cen-sored.
    Clinico-pathological variables, degree of facial weakness and treat-ment modalities were included in the analysis. For progression-free survival, adenoid cystic carcinoma pathology was classified as high-grade tumors, because these tumors are usually locally infiltrative. Adenoid cystic carcinoma pathology was separated from other high-grade and non-high-grade tumors in the analysis of overall survival, because it grows slowly and most patients with these tumors have long survival, even with advanced tumor stage.
    From univariable analysis, the significant variables were chosen to construct a multivariable Cox proportional hazard model for survival. Multicollinearity among variables was tested by calculating variance inflation factor. The possible association between variables (tumor grade vs. metastasis, presence of distant Adriamycin vs. initial non-sur-gical treatment) was also statistically tested. Statistical analyses were executed using SPSS version 20.0 (IBM Corporation, Armonk, NY, USA). All tests were two-sided and P < 0.05 indicated statistical sig-nificance.
    Subject characteristics
    Among 45 enrolled patients, adenoid cystic carcinoma and salivary duct carcinoma (high-grade) were the most frequent pathologies, causing facial weakness (Table 1). Of note, high-grade tumors including adenoid cystic carcinoma represented 86.7% of all cases. Regarding tumor size, 77.8% of tumors were more than 4 cm in maximal diameter. We performed neck dissection in 22 patients: elective lymph node dissection in 11 and therapeutic lymph node dissection in 11. However, neck dissection was not indicated in 23 patients (patients with adenoid cystic carcinoma pathology or non-surgical management). As a result, cases diagnosed as N0 were 53.3% and N(+) were 46.6% (Table 1). Five patients with systemic metastasis had undergone palliative treat-ment alone. Overall, most parotid gland cancers causing facial weak-ness were advanced tumor stage and high-grade tumors.
    Complete facial paralysis (House-Brackmann grades 5–6) was ob-served in 40% of patients at presentation. Incomplete facial weakness (House-Brackmann grades 2–4) or paralysis on specific face subsites was observed in 60% of patients.
    All patients without systemic disease (M0) were subjected to cura-tive treatment, including primary surgery with adjuvant treatment (n = 33) and primary radiation (or chemoradiation) (n = 7).
    Oncological outcomes and survival analysis
    Overall and progression-free survival rates among all patients were 67.2% and 44.0% at 3 years, 40.1% and 38.5% at 5 years, respectively (Fig. 2). Because of the limited number of patients, oncological out-comes are only presented in the descriptive manner (Table 2). As ex-pected, the group with initial non-surgical treatment showed the worst survival outcomes. The two groups who underwent surgery and facial nerve sacrifice with or without nerve graft had similar oncological outcomes (Supplementary Fig. S1). The group with surgery and facial nerve preservation showed the best outcomes. This was probably due to low tumor burden (no lymph node metastasis) at diagnosis.  Oral Oncology 89 (2019) 144–149