Objective To find out the possible causes of pillar pain after surgical treatment for carpal tunnel syndrome through anatomical and clinical study. Methods Through the anatomical study of the surgical area of carpal tunnel syndrome in 6 cadavers (12 sides), the distribution rule of the nerve branches in this area was found. According to this rule, the incision design and entry route were improved and applied to the clinic, and the occurrence of pillar pain was analyzed. A statistical comparison was made between the occurrence of pillar pain in routine surgical and modified surgical in order to find out the possible causes of the occurrence of pillar pain. Results The nerve innervation in the surgical area of carpal tunnel syndrome was derived from the palmar cutaneous branch of the median nerve. The palmar cutaneous branch of the median nerve emerged off the palmar fascia 8.1 mm at the distal side to the zero point. There were 3 types of main branch including 16.7% in single branch, 33.3% in double branches and 50% in three branches. The nerve trunk of the single and double branches walked in the radial side of the thenar crease with several irregular fine branches to the ulnar side. The ulnar branch of the three branches went from the main trunk to the ulnar side at the site to zero point averaging 11.2 mm and crossed the thenar crease at the site to zero point averaging 14.6 mm. According to this, the operative method was designed: the incision was located at 0.5 cm of the ulnar side of the thenar crease. The incision was about 3.0 cm long, and the proximal end of the incision was at least 2.0 cm to zero point. After incision, only the transverse carpal ligament was cut to protect the palmar aponeurosis at the proximal end of the incision to zero point and avoid the injury of the ulnar branch of palmar cutaneous branch. 102 patients were performed with the improved surgery while 110 cases were treated by routine surgery. The incidence of pillar pain in the improved group was 12.7%, and that in the control group was 34.5% one month after the operation. The incidence of pillar pain in the improved group was 14.7%, and that in the control group was 41.0% 3 months after the operation. The two groups had significant differences (P<0.05). There was no significant differences between the two groups 6 months after the operation, and the pillar pain disappeared in all the patients one year after the operation. Conclusion The injury of the ulnar branch of the palmar cutaneous branch of median nerve may be an important cause of pillar pain after surgical treatment for carpal tunnel syndrome. Key words: Carpal tunnel syndrome; Anatomy; Palmar cutaneous branch; Pillar pain