Facial anatomy in layers: what lies beneath the skin

The face is conventionally described as five concentric layers, consistent from scalp to neck in their order if not in their thickness. From the surface: skin, subcutaneous fat, the musculoaponeurotic layer (SMAS and the mimetic muscles invested within it), deep fat and areolar tissue, and the deep fascia and periosteum overlying bone. Every injectable decision is, in effect, a decision about which of these five to enter and where to stop.

Why the model is worth holding

The layered model does two things that a regional description cannot. It makes depth transferable between regions, so that what is learned about the plane containing the facial artery in one area carries over to the next. And it converts the question “where is the vessel” into the more answerable question “which layer is the vessel in here”, which is what the published anatomical work actually reports.

The layers change by region

Layer three is the layer that varies most. In the scalp it is the galea aponeurotica; in the temple it becomes the superficial temporal fascia; in the midface it is the SMAS proper; in the neck it is the platysma. It is continuous throughout, but its thickness, its adherence to the layers around it and the muscles it invests differ substantially.

Layer two also varies. The subcutaneous fat is thick and mobile over the cheek, thin and adherent over the temple and the nasal dorsum, and effectively absent at the eyelid, where the skin overlies orbicularis with almost no intervening cushion.

Layer four is the least uniform of all. In some regions it is a well-defined space with identifiable boundaries; in others it is condensed into ligamentous attachments that bind layer three directly to periosteum. The retaining ligaments are precisely the points at which this layer disappears.

Two regional consequences follow. The temple is thin, its layers are compressed into a few millimetres, and the middle temporal vein and superficial temporal artery lie in defined planes within that compression. The glabella and nasal dorsum have little subcutaneous cushion and carry vessels with documented anastomotic connection to the ophthalmic circulation, which is why they recur in the vision-loss literature disproportionately to the volume injected there.

What a “safe plane” means

This is where the terminology misleads. A safe plane is not a region without vessels. The face is vascularised throughout, and no injectable plane is empty.

What the safe-zone literature describes is a plane in which a fascial, muscular or periosteal barrier lies between the needle tip and the significant vessels of that region. Supraperiosteal placement in parts of the midface is described as comparatively safe not because the arteries are absent but because the principal branches run superficial to the deep fascia, with that fascia interposed. Superficial subdermal placement in some areas is described the same way for the mirror-image reason: the vessels of concern run deeper.

The corollary is that safety in this framework depends entirely on knowing which layer the needle tip is currently in — and that is a proprioceptive and anatomical judgement, not a measurement. Depth confidence degrades as tissue is entered obliquely, as it is displaced by the advancing instrument, and in patients whose layers are thinner than the anatomical average.

The published danger zones are the corollary again: regions where the barrier is absent, where the vessel is superficial and unprotected, or where the arterial supply has a documented retrograde route to the orbit.

SubSMAS spaces

Beneath the SMAS lie a series of gliding spaces — described in the anatomical literature with defined floors, roofs and ligamentous boundaries — which allow the superficial fascia to move over the deeper structures during expression. Anatomical study of these spaces identifies them as relatively avascular compartments, since the vessels tend to run in their boundaries rather than through their interiors.

Two implications follow for injectable practice. Material placed within such a space distributes along it, so the region influenced may be considerably larger than the deposit. And the ligamentous boundaries between adjacent spaces are, by definition, where the vessels are: crossing a boundary is not the same manoeuvre as remaining within a compartment, even though the depth reading at the needle tip may be unchanged.