It arises from the lower border and medial surface of the zygomatic arch and inserts into the lateral surface of the mandible (FIG 1).

The muscle consists of three layers—superficial, middle, and deep.

The superficial muscle fibers pass backward and downward; the deeper fibers are more vertical. The motor nerve to the masseter is derived from the mandibular division of the trigeminal nerve. It leaves the infratemporal fossa by passing over the lateral pterygoid muscle and mandibular notch. It enters the masseter muscle from its deep surface and courses down between its deep and middle layers by having a diameter of 1 to 2 mm. It becomes more superficial distally toward the anterior inferior part of the muscle and runs alongside the (friable) vascular pedicle that arises from the inferior alveolar artery.

The nerve usually divides into upper (temporal) and lower (cervical) trunks at first, from which these branches arise. In practice, they constitute five variably interconnected groups of branches hereafter referred to as divisions.

Despite the various studies of facial nerve anatomy, it may be more useful to envision the facial nerve as consisting of physiological and functional divisions that do not necessarily correlate to the anatomic distribution of branches but instead relate to muscular groups and actions.

The course of the nerve follows a line drawn from 0.5 cm below the tragus to 1.5 cm above the eyebrow, but 3D modeling demonstrates the branches cover a triangular area bordered between a line through the palpable zygomaticofrontal and zygomaticotemporal sutures and a vertical line 1 cm in front of the auditory meatus.

At the lateral margin of the frontalis, these branches run deep to frontalis and orbicularis oculi (which it innervates), eventually reaching the glabella region to supply the depressors of the brow.

Distally, it runs on the undersurface of the zygomaticus muscle complex before ramifying more medially in the face. The buccal branches exit the anterior border of the parotid gland and run inferior to the parotid duct and on the surface of the buccal fat pad deep to the SMAS layer to supply the muscles of the mouth and cheek.

Both zygomatic and buccal divisions have a highly overlapping distribution of innervation that allows transection of one or more branches without weakness; this is the basis of cross-facial nerve grafting.

Cadaveric dissections demonstrate that the nerve always comprises at least three major branches that run on the undersurface of the platysma and superficial to the facial vessels.

Once the patient has been denervated for over 2 years and the orbicularis oculi has atrophied beyond salvage, it becomes very challenging to recreate a blink.

In contrast, a smile can be reliably restored years after irreversible denervation with a free gracilis or temporalis transfer.

This is also the reason the author generally reserves temporalis transfer for cases in which native facial musculature is unsalvageable, as the temporalis generally does not restore eye closure.

Lower lip motion and depression complete a symmetric outcome.

If there is muscle viability, nerve-to-masseter and minihypoglossal transfers are typically performed; if there is no viable muscle, functioning muscle transfer is required.

The ideal patient is one that has a relatively short segmental defect, no tumor in the proximal facial nerve trunk, and no prior radiation (FIG 2).

An anterolateral thigh free flap provides not only soft tissue replacement to fill the defect but also donor nerve graft in the form of a motor branch to the vastus lateralis muscle. The nerve can be left attached to the pedicle as a vascularized nerve graft or, more commonly, harvested separately to allow for greater freedom of inset.

While not ideal, nerve-to-masseter transfer is highly reliable, and some patients may not require an upper eyelid weight and frequent eye drops once reinnervation is complete.

The patient is evaluated preoperatively to confirm normally functioning masseteric and hypoglossal nerves. If there is a deficiency, these nerve donors are avoided, and other donors such as the spinal accessory nerve can be considered.

One significant drawback of this technique is the variability in strength of the outcome compared with nerve transfers that deliver reliable power but cannot compete with cross-face nerve grafting in terms of spontaneity.

It is also most commonly performed in two stages, which may not be feasible in cases of more advanced disease.

For patients in whom the ipsilateral nerve is unusable, however, this is the only procedure that provides spontaneous motion.