Transferring the long head of the triceps nerve to the axillary nerve is a widely used technique for restoring abduction and elevation of the shoulder in adults with partial brachial plexus lesions. This procedure can be performed using either an anterior (axillary) or a posterior approach. This anatomical study aimed to compare the distance between the microsurgical suture of the nerve transfer and the axillary nerve's entry into the middle and anterior deltoid, to determine which approach provides the shortest nerve regrowth path.

The anatomical study was carried out on 12 subjects (i.e. 24 shoulders). For each subject, the transfer of the long portion of the triceps nerve to the anterior division of the axillary nerve was simulated. The microsurgical suture area was marked on the axillary nerve using stitches or clips. The nerve was then dissected and released until it entered the middle and anterior bundles of the deltoid via a lateral trans-deltoid route. The nerve was then cut flush with the muscle and recovered via the axillary route on ten shoulders (five right, five left) and via the posterior route on ten shoulders (five right, five left). The distance between the marker and the end of the axillary nerve was measured. The time taken to locate the teres minor branch after locating the axillary nerve through the anterior and posterior approaches was also noted.

On average, the distance between the suture and the nerve's entry into the deltoid was 74 mm for the anterior approach (ranging from 65 to 80 mm), and 62.4 mm for the posterior approach (ranging from 38 to 69 mm). There was a significant difference in length between the two approaches. The teres minor branch was located more quickly with the anterior approach (average time: 4 minutes, range 2 to 6 minutes) than with the posterior approach (average time: 18.1 minutes, range 16 to 21 minutes). This difference was significant (p < 0.05).

In a nerve transfer, the distance between the suture and the recipient muscle affects the time taken for reinnervation and therefore the outcome, given the progressive degradation of the motor end plates from the initial lesion. This study shows that the distance is significantly shorter via the posterior route. Reinnervation of the deltoid muscle should therefore be faster and of better quality via this route. These results must be confirmed by a clinical study.