Minimally invasive surgery using endoscopic vision is widely employed for the treatment of thyroid diseases. We have performed endoscopic thyroidectomy by the axillary approach (axillary approach) and video-assisted thyroidectomy via a 3 cm cervical incision (video-assisted approach). In this study, we evaluated the efficacy of these two procedures. Each procedure was performed in 20 consecutive consenting patients. The degree of invasiveness after surgery was compared using postoperative results. The amount of pain and satisfaction with surgery was evaluated by grade (1-5) using a patient questionnaire. All thyroidectomies were completed successfully. No recurrent laryngeal nerve palsies occurred. Operating time for the video-assisted approach was significantly shorter than that for the axillary approach (p < 0.01). The amount of pain for the axillary approach on 1, 3, and 5 days after operation, respectively, was graded 3.2+/-0.7, 2.1+/-0.6, and 1.6+/-0.7 compared to 2.7+/-1.1, 1.7+/-0.7, and 1.1+/-0.2 for the video-assisted approach. The postoperative course was significantly less painful in patients undergoing the video-assisted approach on postoperative days 3 and 5 (p < 0.01). The degrees of satisfaction for the axillary approach and the video-assisted procedure were 1.2+/-0.4 and 2.4+/-1.0, respectively (p < 0.01). The video-assisted approach is less "invasive" than the axillary approach, but the axillary approach may be indicated for patients who are anxious about the visible cosmetic results.