The paralytic eye and symmetrisation — What must be solved before movement returns
Lagophthalmos and the lid weight with the weight correction almost nobody applies, ectropion and lower lid suspension, and botulinum toxin for contralateral hyperkinesis and synkinesis, with doses by muscle.
Content intended for healthcare professionals. It does not replace clinical assessment and is not written as patient information.
Key points
- The eye has absolute priority over any aesthetic or movement consideration: an exposed cornea is lost in weeks and does not come back.
- The lid weight that works in the adhesive trial is not the one to implant: the angle changes on moving from skin to tarsus, and in almost 30 % of cases the chosen one falls short.
- Asymmetry does not come only from the paralysed side: the healthy side pulls. Weakening contralateral hyperkinesis symmetrises without touching the affected side.
- In acute palsy there is an additional argument for treating the healthy side: some work suggests that reducing its overactivity accelerates motor recovery of the paralysed side by lessening inhibitory feedback.
- There are no internationally validated clinical guidelines for toxin in facial palsy. The practical rule is to start with low doses and review at two weeks to decide whether more is needed.
Lagophthalmos and the lid weight
Loading the upper lid so that gravity does the work the orbicularis no longer does has been an established procedure since the 1970s, simple and effective. Its performance depends on two decisions. The first is where it goes: placed high on the tarsus so that it is hidden with the eye open, the aesthetic result is good; placed near the lid margin, closure improves but the weight shows and extrudes more often. The second is how much it weighs, and that is the point most often overlooked. The classic rule is to choose the minimum weight that brings the upper lid to within 2-4 mm of the lower, because large weights migrate. But the choice is made by taping a trial weight onto the skin, and that is not where it will end up. The consequence needs stating with numbers.
The weight correction that must be applied
When the weight sits on the skin of the upper lid, the angle between the vertical and its axis is 22° with the eye open and 0° with it closed. When it sits inside, on the tarsal plate, those angles become 40° and 18°. The useful force depends on the cosine of that angle, so the same mass does not produce the same effect in the two positions: the implanted weight works less well than the trial one. In practice this means that in almost 30 % of cases the weight chosen at trial proves insufficient once implanted, and that closure also becomes slower. The practical correction is direct: choose a weight about 0.2 g heavier than the one that worked in the trial. If 1.2 g closed the eye at trial, implant 1.4 g. Two further nuances: if skin resection is planned, add the weight of the resected skin; and if the brow is to be lifted, repeat the trial while lifting it with a finger, because some skin is repositioned out of the lid. It is also worth considering levator recession in the same operation so that a heavier weight does not leave a ptosis.
What the weight does not solve
- Gravity-assisted closure is far slower than the reflex blink. The patient must learn to close and hold the eye shut for one or two seconds to give the lid time to descend.
- Firm fixation to the tarsus and a high position on it are the two technical details separating a good result from an extrusion.
- An autologous alternative with the same logic is upper lid lipofilling, of the order of 2 ml equivalent to 1.6 g: it adds mass without a foreign body, though with less control over the exact figure.
- When active rather than merely passive closure is sought, the ground is muscle transposition — temporalis in the adult, frontalis in the young patient with severe ocular involvement — or free muscle transfer, the only option that can restore a synchronous blink.
The lower lid: ectropion and epiphora
Paralytic ectropion separates the inferior punctum from the globe, and that loss of contact produces epiphora, which may be worsened by aberrant reinnervation of the lacrimal gland. The classic options are tarsorrhaphy, lateral canthoplasty and canthopexy: in lateral canthoplasty the canthal ligament is detached and reinserted higher on the frontal periosteum for stability. Where structural support is needed, a tendon strip is used at the lower lid margin — palmaris longus, plantaris or fascia lata — avoiding alloplastic material where possible. One described variant uses two suture anchors, medial and lateral, allowing both lid anchoring points to be displaced cranially: elevating the medial portion is precisely what restores contact between punctum and sclera and prevents ectropion recurring. One calibration detail worth retaining: the lower lid margin should end up 1.5 mm above the pupil, under more tension than seems necessary, allowing for the descent that follows over the next weeks.
Why toxin symmetrises
Facial palsy produces static and dynamic asymmetry, hyperkinesis of the non-paralysed side and synkinesis, and all three benefit from the same mechanism: botulinum toxin denervates overactive muscles. In acute palsy its use is essentially symmetrising — the healthy side is treated so it does not drag the face — with a far from trivial added benefit: reducing overactivity on the healthy side appears to facilitate facial rebalancing and support more effective neuromuscular retraining. In chronic palsy the problem is different: faulty reinnervation makes some territories activate erroneously, distorting expression and producing anything from painful spasm to functional difficulty in the upper and lower face. There the toxin does not symmetrise but selectively switches off what moves wrongly. The advantage over surgery is obvious and worth stating: adverse effects are self-limiting and resolve within the toxin’s own pharmacological window.
Indicative doses by muscle
| Muscle | Range per session | Plane and depth |
|---|---|---|
| Orbicularis oculi | 2.5-10 U on the affected side; 4-10 U on the healthy one | Intracutaneous at 2-3 mm. The endopalpebral portion, 2-4 U at 1 mm and very superficial |
| Frontalis | 1.5-18 U affected, 2-10 U healthy. The affected-side range is so wide because it depends on flaccid hypotonia versus paradoxical hyperkinesis | Intramuscular or intracutaneous, 2-4 mm |
| Corrugator supercilii | 4-10 U affected, 2-4 U healthy | Intramuscular, 6-8 mm at the origin and 2 mm at the insertion |
| Levator labii superioris alaeque nasi | 2-7.5 U affected, 2.5-7.5 U healthy | Intramuscular at 4-8 mm. On the healthy side, to reduce upper tooth show |
| Depressor anguli oris | 0.5-10 U | Deep intramuscular at the bony origin, at 1-3 points along its line |
| Depressor labii inferioris | 1-3 U affected, 2.5-5 U healthy | Intramuscular at 2-3 mm. On the healthy side, to reduce lower tooth show on smiling |
| Platysma | 10-60 U affected, 20-34 U healthy. It has the widest range of all | Intracutaneous or superficial intramuscular at 1-2 mm, in lateral and anterior bands |
| Buccinator | 2-20 U | By the intraoral route, perpendicular to the mucosa. Useful to improve commissure excursion |
| Mentalis | 2.5-10 U affected, 3-5 U healthy | Deep intramuscular, 6-10 mm |
| Lacrimal gland | 1-20 U | Intraglandular by the transconjunctival route. This is the manoeuvre for aberrant-reinnervation tearing |
What can go wrong with toxin
Adverse effects are mostly self-limiting, but they should be warned about beforehand because for weeks they can be disabling. Injections into orbicularis oris, orbicularis oculi and platysma cause transient difficulty drinking, eating, articulating or puckering. When treating oro-ocular synkinesis, expect that it may worsen blepharoptosis, lagophthalmos and diplopia: that is, treating one problem may aggravate another. And overtreatment produces aesthetic and expressive alterations — smile dysfunction, brow ptosis — or even loss of function. Hence the sensible regimen is the one several groups have proposed: initial doses below standard and review at two weeks to decide whether to add. It is slower, but it is the only way not to overshoot in a face that is already unbalanced.
Details that improve the result
- Examination is muscle by muscle, not at a glance: global resting tone and then each muscle separately at maximum contraction with a standardised battery — brow elevation, gentle and forced lid closure, nose wrinkling, smiling, lip protrusion, whistling and lower lip depression.
- Standardised photography in a fixed sequence at every visit is what allows comparison: neutral face, small smile, big smile, raised brows, furrowed brows, puckered lips, lower lip down and upper lip up.
- The Sunnybrook scale is used in follow-up for its sensitivity to change: in one published case the score went from 47 before treatment to 58 six weeks after, coinciding with the pharmacological peak.
- EMG or ultrasound guidance improves precision in small or hard-to-palpate muscles, which is exactly where an error of plane is paid for with an unwanted effect.
- Formulation matters for duration: in a single-masked randomised trial the three preparations were equivalent at 1-2 weeks, but at 4 weeks improvement was 17 % with incobotulinumtoxinA versus 41 % and 42 % for the other two, suggesting the former needs higher doses or shorter intervals.
Toxin and functional microsurgery, in the same plan
They are not alternatives but parts of the same treatment, and it is worth framing it that way from the first consultation. Toxin prepares the ground before microsurgery — reducing the contralateral hyperkinesis that will compete with the reanimated muscle — and refines the result afterwards, modulating reanimated or transplanted muscles, with documented improvements in disease-specific quality of life. Coordination between the injector and the surgeon is what allows that staged use, and its absence is why many patients receive one or the other rather than both. With one caveat that should be said aloud: no internationally validated clinical guidelines exist, and the available evidence is limited by heterogeneous designs, small samples, inconsistent outcome measures and short follow-up.
What to do with the eye
Is there lagophthalmos with corneal risk?
Is there epiphora?
References
- 1.Renga M, D’Emilio R, Salti G, Mogavero S, Papagni M, Biglioli F, Lozza A. Facial paralysis treatment and facial symmetrization with botulinum neurotoxin: a narrative review with illustrative clinical cases. Toxins. 2026;18(6):253.
- 2.Hontanilla B. Weight measurement of upper eyelid gold implants for lagophthalmos in facial paralysis. Plast Reconstr Surg. 2001;108(6):1539-1543.
- 3.Jobe RP. A technique for lid loading in the management of the lagophthalmos of facial palsy. Plast Reconstr Surg. 1974;53(1):29-32.
- 4.Hontanilla B, Gómez-Ruiz R. Surgical correction of lower eyelid paralysis with suture screw anchors. J Plast Reconstr Aesthet Surg. 2009;62(12):1598-1601.
- 5.Hontanilla B, Marre D. Eyelid reanimation with gold weight implant and tendon sling suspension: evaluation of excursion and velocity using the Facial Clima system. J Plast Reconstr Aesthet Surg. 2013;66(4):518-524.
- 6.Ross BG, Fradet G, Nedzelski JM. Development of a sensitive clinical facial grading system. Otolaryngol Head Neck Surg. 1996;114(3):380-386.
- 7.Tzou CHJ, Rodríguez-Lorenzo A, eds. Facial Palsy: Techniques for Reanimation of the Paralyzed Face. Cham: Springer; 2021.
Related specialty: Facial Paralysis