OBJECTIVETo evaluate therapeutic effect in a consecutive series of patients with node-positive of N1 and N2 head and neck squamous cell carcinoma (HNSCC) who underwent selective neck dissection as part of their treatment in a single institution.

METHODSRetrospectively analyzed 36 patients with node positive head and neck squamous cell carcinoma, which had undergone 40 selective neck dissection (SND) and 16 radical neck dissections (RND) and the therapeutic effects of the SND in HNSCC patients with node positive. Kaplan-Meier method was used to estimate the survival analysis among the different levels, and log-rank method for comparison of the different distribution of the several influential factors of the survival; the fisher's exact test was used to test the difference of the neck recurrence between the groups that with or without lymph node extracapsular spread. Cox proportional hazard model was applied to screen the potentially significant prognostic factors.

RESULTSThe 3 and 5 years survival rate of this group were 76.8% and 54.3% retrospectively. The 3 and 5 years' survival rate were both 100% in the N1 + N2a group, while 59.4% and 32.0% in the N2b + N2c group correspondingly. With log-rank test between the two groups above, significant statistical difference was revealed (P = 0.003) in the 5 years' survival rates. To the groups that with or without extracapsular spread (ECS), the recurrence rate were 36.4% and 3.4% respectively and significant statistical difference existed (P = 0.015). In the groups that with or without ECS, the 3 years' survival rate were 45.5% and 81.8% , and 5 years' survival rate were 39.7% and 65.5% respectively, also the significant statistical difference (P = 0.0148 and P = 0.0423 respectively) had presented.

CONCLUSIONSSND is appropriate for treatment of N1 neck of the patient with HNSCC; However, SND should be taken very cautious to the N2 neck and node-positive neck with extracapsular spread, MRND or RND might be a better choice to prevent the recurrence in the neck.