Synkinesis and Facial Reanimation After Bell's Palsy: Why the Face That Came Back Moves Wrong, Why Botox Is the Main Treatment, and When Surgery Can Rebuild a Smile

Bell's palsy arrives overnight. A person brushes their teeth and water runs out of one corner of the mouth, or they look in the mirror and one side of the face has simply stopped. The eyebrow sits low, the eye will not close, the smile pulls only to one side. It is frightening, and the first thing every patient hears, correctly, is that most people get better.

What patients hear less often is what "better" can look like a year later. For a meaningful minority, movement comes back but comes back wrong. The eye narrows every time they smile. The corner of the mouth twitches when they blink. The neck cords tighten when they speak. The cheek on the affected side feels permanently clenched, and in photographs the smile looks strained rather than absent. This is synkinesis, and it is the most common long-term problem after facial paralysis that partly recovers. It is also a problem that sits exactly where cosmetic and reconstructive surgery meet, because patients often arrive at an aesthetic practice asking for Botox, a brow lift, or filler to "even things out," without anyone having named the underlying condition. This piece is about what synkinesis is, how it is treated, how facial reanimation works when the face never recovered at all, and how to find the right hands. It sits alongside the piece on facial nerve injury after a facelift, which covers the surgical cause of the same kind of weakness, and the piece on facial asymmetry, which covers the asymmetry people are born with.

What synkinesis is, and why the nerve rewires itself

The short answer: synkinesis is involuntary movement of one part of the face that happens whenever another part is moved on purpose (the eye closing during a smile, the chin dimpling during a blink), and it develops because the facial nerve, after it is damaged, regrows its fibers into the wrong muscles; it usually appears several months into recovery, is most likely after a severe initial paralysis, and often comes with tightness, facial fatigue, and sometimes tearing while eating.

The facial nerve is the seventh cranial nerve. It leaves the brainstem, travels through a narrow bony canal in the skull, exits just below the ear, and fans out through the parotid gland into branches that run to the forehead, the eyelids, the cheek, the mouth, and the neck. Each of those branches carries thousands of individual fibers, and in a healthy face each fiber has a single destination. The brain sends "close the eye" down one set of fibers and "smile" down another.

Bell's palsy is thought to be an inflammation of the nerve inside its bony canal, most likely triggered by a viral reactivation, with herpes simplex the leading suspect. The swollen nerve is squeezed in a space with no room, and depending on how hard it is squeezed, the fibers are either stunned and recover quickly or are damaged enough that they must regrow from the point of injury. Regrowth is where the trouble starts. Regenerating fibers follow whatever channel they encounter, and a fiber that once served the smile can end up in the muscle that closes the eye. From then on, every command to smile also reaches the eye. Surgeons call this aberrant regeneration, and the visible result is synkinesis.

The common patterns have names. Oral-ocular synkinesis, the eye narrowing or closing with a smile or with pursing the lips, is the one most patients notice first. Ocular-oral synkinesis runs the other way: a blink pulls the mouth corner. The platysma, the thin sheet of muscle in the neck, often joins in, producing tight bands when the patient talks or smiles, which is the same muscle discussed in the piece on the Nefertiti lift. The mentalis muscle can dimple and pucker the chin, the problem described in the piece on the orange-peel chin. A related miswiring, gustatory tearing (sometimes called crocodile tears), sends fibers meant for the salivary glands to the tear gland, so the eye waters during meals.

Synkinesis is not only about stray movement. The miswired muscles also tend to be overactive at rest, which is called hypertonicity. The affected side can look fuller in the cheek, with a deeper fold beside the nose and a narrower eye, and patients commonly describe tightness, a pulling sensation, or aching after a long day of talking. One consequence surprises people: the smile on the affected side is often weak not because the smiling muscles have failed, but because the muscles that pull the mouth down and back are firing at the same time and cancelling the effort. That observation is the basis of most modern treatment.

Who ends up with synkinesis is fairly predictable, and so is the window in which the odds can be improved. The American Academy of Otolaryngology-Head and Neck Surgery clinical practice guideline recommends oral corticosteroids within 72 hours of onset, and a major randomized trial in the New England Journal of Medicine found that early prednisolone meaningfully improved the rate of complete recovery.

The natural history of Bell's palsy is broadly favorable. Most patients begin to improve within a few weeks, and the large majority recover fully or nearly fully. The ones who do not are disproportionately those whose paralysis was complete rather than partial at its worst point, whose recovery did not begin within the first several weeks, and who are older. Those are exactly the patients whose nerve fibers have had to regrow, which is why they are also the patients who develop synkinesis. Published estimates of how often it occurs vary widely, because studies define it differently and follow patients for different lengths of time, but it is a common outcome after severe Bell's palsy rather than a rare one.

Early treatment matters because it changes how many fibers have to regrow in the first place. The 2007 Scottish trial published in the New England Journal of Medicine randomized patients within 72 hours of onset and found that prednisolone raised the proportion who had fully recovered at three months and at nine months compared with placebo, while the antiviral acyclovir on its own added nothing. The AAO-HNS guideline, updated in 2013, recommends steroids within 72 hours for patients 16 and older without contraindications, recommends against antiviral therapy alone, and allows antivirals to be offered in combination with steroids. It also recommends eye protection for anyone who cannot close the eye, and reassessment or referral for patients who have not recovered by three months.

Two warnings belong here. First, Bell's palsy is a diagnosis made after other causes are excluded. Shingles of the facial nerve (Ramsay Hunt syndrome, which usually comes with a painful rash in or around the ear), Lyme disease, middle ear infection, trauma, stroke, and tumors of the parotid gland or the nerve itself can all cause facial weakness. A paralysis that came on slowly over weeks, that affects other nerves, that recurs on the same side, or that shows no recovery at all after several months is not behaving like Bell's palsy and needs imaging. Second, the eye is the urgent part. An eye that does not close fully dries out overnight, and the cornea can ulcerate. Lubricating drops by day, ointment at night, and taping or a moisture chamber are not optional, and the piece on dry eye after eyelid surgery describes why exposure is so damaging to the cornea.

"Synkinesis is the face recovering its power without recovering its wiring. The treatment is rarely about adding movement. It is almost always about quieting the muscles that are fighting the ones the patient wants to use."

Botulinum toxin and retraining: the treatment that does most of the work

The short answer: for synkinesis, the mainstay is small, targeted doses of botulinum toxin into the overactive muscles on the affected side (most often around the eye, the neck, the chin, and the muscles that pull the mouth corner down), combined with neuromuscular retraining by a physical therapist who specializes in the face; the toxin use is off-label, the doses are far smaller and more selective than cosmetic dosing, and most patients need repeat treatment every three to four months indefinitely.

Botulinum toxin blocks the signal from nerve to muscle, and in a synkinetic face that is precisely what is wanted in the muscles that are misfiring. A few units into the orbicularis oculi, the ring muscle around the eye, reopens an eye that narrows during a smile. Injections into the platysma soften the neck banding. Treating the depressor anguli oris, the small muscle that pulls the mouth corner down, lets the smile muscles work without opposition, which is why a correctly dosed injection on the affected side can make a smile look stronger rather than weaker. The mentalis, the buccinator in the cheek, and the muscles of the upper lip are other common targets. Specialists sometimes also treat the healthy side lightly, so that a normal, strong smile does not make the weak side look worse by comparison. The mechanics are the same as in the piece on the marionette lines and downturned mouth corners, but the purpose and the stakes are different.

This is not cosmetic Botox with a different label. The doses are small, the injection points are chosen after watching the patient make specific movements, and an injector who overdoses the cheek or the upper lip can produce a drooping mouth, drooling, or trouble with speech, which on an already weak side is a real functional setback. The choice of brand matters less than the map, a point made in the piece comparing Botox and Dysport. The use for synkinesis is off-label in the United States, although botulinum toxin is approved for related conditions such as blepharospasm and hemifacial spasm, and it has been standard practice in facial nerve centers for decades.

Neuromuscular retraining is the other half. A therapist trained in facial rehabilitation teaches the patient to make slow, small, isolated movements while watching in a mirror, to smile without letting the eye close, and to release chronic tightness with massage and stretching. The goal is not strength. The large, forceful exercises that were once handed out for facial palsy, including chewing gum and electrical stimulation, tend to make synkinesis worse, because they reinforce mass movement. The evidence for retraining comes mostly from smaller studies, but it is consistent, and in facial nerve centers it is paired with toxin rather than offered as an alternative.

Grading is how progress is tracked. The House-Brackmann scale, which runs from grade I (normal) to grade VI (total paralysis), is the one most patients hear, but it was designed for overall nerve function and is crude for synkinesis. The Sunnybrook Facial Grading System scores resting symmetry, voluntary movement, and synkinesis separately, and the eFACE, a tool developed at Massachusetts Eye and Ear, uses a graphical scale for the same purpose. A practice that treats synkinesis should be scoring the face and taking standardized photographs and video before every round of injections. A practice that simply eyeballs it is not doing the work, for the same reason set out in the piece on how to read a before-and-after gallery.

Toxin wears off, typically over about three months, and synkinesis does not. Patients should expect repeat treatments for years, and should know that the response can seem to fade over time, which the piece on why Botox stops working explains.

When injections are not enough: neurolysis, myectomy, and the static procedures

The short answer: for patients whose synkinesis or tightness keeps limiting the smile despite good toxin and therapy, facial nerve surgeons can cut selected branches of the facial nerve that drive the counterproductive muscles (selective neurolysis), or remove small muscles such as the depressor anguli oris or a strip of the platysma (selective myectomy); separate procedures such as brow lifts, eyelid weights, and lower lid tightening address the eye and brow, and on a weak face every one of them must be judged first by what it does to eye closure.

Selective neurolysis is the most important of these operations and the least familiar to cosmetic patients. Through an incision similar to a facelift incision, the surgeon uses a nerve stimulator to map the individual buccal branches of the facial nerve that run to the midface. Branches that mainly drive the smile are preserved; branches that mainly drive the eye closure, cheek tightness, or lip pursing that fight the smile are divided. Published series from academic centers report meaningful improvements in smile and synkinesis scores in selected patients, and the procedure is usually paired with toxin afterward rather than replacing it. It is an operation that relies on detailed mapping of the nerve in real time, and it should be done by surgeons who do it routinely.

Selective myectomy is simpler. Removing the depressor anguli oris, usually through a small incision inside the lower lip, permanently removes one of the main muscles pulling the smile downward, and partial removal of the platysma can release neck banding. Surgeons often test the effect first with a toxin injection, which acts as a preview: if weakening the muscle helped, removing it will too.

The eye and brow get their own procedures. When the forehead branch does not recover, the brow on that side sinks and can hood the upper lid, a problem also described in the piece on hooded eyes versus brow ptosis. A brow lift on the paralyzed side, usually a direct or endoscopic technique, fixes the position, but a surgeon who lifts too high can leave the lid unable to close over the eye. When the eyelid does not close, a small gold or platinum weight placed in the upper lid lets gravity help close it, and a lower lid that has sagged away from the eye is tightened with a lateral canthal procedure similar to those in the piece on canthoplasty. A standard cosmetic upper blepharoplasty on a lid that already closes poorly can make exposure worse, which is why an ordinary eyelid operation on a paralyzed face is not ordinary at all.

Facial reanimation: when the face never came back

The short answer: when paralysis is complete and does not recover, surgeons can restore movement dynamically, by connecting a working nerve (most often the nerve to the masseter chewing muscle, sometimes a nerve graft from the healthy side of the face) to the facial nerve or to a transplanted muscle; if the native facial muscles have been without a nerve for too long, typically more than a year or two, a new muscle is needed, and the standard choice is a segment of the gracilis muscle from the inner thigh transferred with its blood supply by microsurgery.

This is a smaller group than the synkinesis group after Bell's palsy, because true permanent paralysis after Bell's palsy is uncommon. Reanimation patients more often come from tumor surgery (acoustic neuroma and parotid cancer), trauma, congenital conditions such as Moebius syndrome, and, rarely, cosmetic surgery. But the tools are the same, and the timeline dictates which tool applies.

Time matters because facial muscles that lose their nerve supply slowly waste away. In the first months after an injury, the muscles are still viable and can be reinnervated if a working nerve is brought to them. The masseteric nerve transfer does exactly this: the small branch that powers the masseter, the chewing muscle at the angle of the jaw, is redirected to the facial nerve's smile branches. It is powerful and fast to recover, but the patient initially has to bite down to smile, and although many learn to smile without thinking about it, not all do. A cross-face nerve graft instead routes a long nerve graft, usually taken from the lower leg, from a smile branch on the healthy side across the upper lip to the paralyzed side. It is slower and weaker, but it carries a signal that is genuinely emotional, which is why some surgeons combine the two. The hypoglossal nerve, which moves the tongue, is another donor, used more often to restore resting tone.

Once the native muscles have gone too long without a nerve (the window is commonly described as roughly 12 to 24 months, though it varies), reinnervation no longer helps, and a new muscle must be brought in. The gracilis transfer takes a slim segment of a thigh muscle that most people can spare, sews its artery and vein to vessels in the face under a microscope, anchors it to the corner of the mouth and the cheekbone, and connects its nerve to the masseteric nerve or to a cross-face graft. Movement appears over months, not weeks. The alternative, the temporalis tendon transfer (lengthening temporalis myoplasty), reroutes the tendon of a chewing muscle at the temple to the mouth corner, avoids microsurgery, and gives a smile powered by biting. Static procedures, such as a sling of tendon or fascia to hold the mouth corner up at rest, do not move but give symmetry at rest for patients who are not candidates for the larger operations.

These are major reconstructive operations with long recoveries, and they are performed at academic centers and specialized facial nerve practices by facial plastic and reconstructive surgeons or plastic surgeons with microsurgical training, often within a team that includes an ophthalmologist, an otolaryngologist, and a therapist. Many of the same microsurgical techniques underpin breast reconstruction after mastectomy. Because the aim is to restore function, insurance generally covers reanimation surgery and often covers toxin for synkinesis; the piece on medical necessity covers how that determination is made.

Who should treat it, and what the cosmetic menu can and cannot add

The short answer: the first stop for synkinesis or persistent facial paralysis should be a facial nerve center or a facial plastic surgeon, oculoplastic surgeon, or plastic surgeon who treats facial palsy regularly, because the treatments that help are highly specific and the ones that harm are often ordinary cosmetic procedures done without the diagnosis; once the function is managed, conventional aesthetic work such as filler, a facelift, or eyelid surgery can be part of the plan, but only with the nerve in mind.

The ASPS and the AAFPRS both count facial reanimation among the reconstructive work their members do, but only a fraction of their members do it routinely, and the difference between a surgeon who does one reanimation a year and one who does one a week is large. The questions in the piece on surgeon case volume apply directly: how many of these do you do, what grading scale do you use, and can I see your standardized video results rather than still photographs. Board certification, covered in the piece on board certifications, is necessary but does not tell a patient whether a surgeon has specific training in facial nerve work. A second consultation at a center that specializes in facial palsy is reasonable for anyone offered a major reconstructive operation.

The failure mode in aesthetic practice is predictable. A patient with mild synkinesis walks into a med spa and asks for their "lazy side" to be lifted, or for filler in the deeper fold on the tight side of the face. Filler in a face with hypertonic muscles can make the fullness worse. Cosmetic toxin doses in the brow or cheek can collapse a smile that was just holding. A brow lift or eyelid surgery on a weak face can leave the eye unable to close. None of this is exotic; it is ordinary cosmetic work done without the diagnosis, and it is a strong argument for the questions raised in the piece on med spa supervision.

With the function managed, aesthetic procedures can help. Fat grafting can restore volume on a side that has thinned, as described in the piece on fat transfer to the face. A facelift can improve the resting position of a paralyzed side, although the surgeon should know the nerve anatomy on both sides cold. Lip and mouth-corner work, discussed in the piece on uneven lips, can refine an asymmetry that remains after reanimation. And the psychological weight of living with a face that does not express what a person feels is real and underrated; the piece on emotional recovery after plastic surgery is a useful companion, and facial palsy patient organizations offer peer support that surgeons cannot.

The honest summary

Most people with Bell's palsy get their face back, and early steroids improve the odds. The people who do not recover completely often end up with synkinesis rather than paralysis: a face that moves, but moves together, with an eye that closes during a smile, a neck that tightens during speech, and a cheek that feels permanently clenched. It is common, it has a name and a grading scale, and it is treatable.

The treatment that does most of the work is unglamorous: small, carefully mapped doses of botulinum toxin into the muscles that misfire, every three or four months, paired with patient retraining by a facial therapist. Selective neurolysis and small muscle removals help people who plateau. For the smaller group whose face never recovered, nerve transfers and the gracilis muscle transfer can restore a real smile, but they are major operations with a timeline that starts ticking the day the paralysis begins, which is why early referral to a facial nerve center matters.

The mistake to avoid is treating a nerve problem as a cosmetic one. Filler, standard Botox, and a routine brow or eyelid lift can all make a synkinetic face worse, and on a face that does not close its eye properly, some of them are dangerous. Get the diagnosis, get a grade, protect the eye, and find someone who sees facial palsy every week. Then, and only then, is the cosmetic menu worth a look.