boundaries
The operation is the decision about where to stop. Everything before it — the examination, the ultrasound, the CT where one was needed — exists to make that decision earlier and better informed. Everything about the result follows from it, including the things a patient will notice years later.
That is an unfamiliar way to describe surgery, because the intuitive picture of a removal is a container being emptied. The picture is wrong in a face injected several times over several years. There is no container. There is a material that has spread through planes it was never meant to occupy, fibrous tissue the body has built around it, and normal structures displaced, thinned or drawn into the same mass. Emptying is not available as an option. Choosing is.
01Why complete removal in one operation is often not the goal
Asked whether everything can be taken out, the accurate answer is that it usually cannot, and that attempting it is frequently the worse of the two available mistakes. The clinic states this directly: because normal tissue has to be preserved while the material is removed, complete removal is not achievable in most cases by surgical means, and it is not set as the objective.
The reason is that the material and the normal tissue are no longer separate things in the places where it matters most. Years of fibrosis bind them. A deposit lying against the infraorbital nerve, or beneath the muscle in the temple, or spread in a thin sheet through the deep cheek fat, cannot be lifted out as a unit. It can only be dissected out of the tissue it has bound itself to, and every millimetre of that dissection is taken from something.
What is taken is predictable. Normal fat, which does not grow back and whose loss shows as hollowing. Small nerve branches, whose loss shows as numbness or as a movement that no longer matches the other side. Skin blood supply, whose loss shows worst of all. An operation that removes ninety per cent of a material and leaves a depressed, numb, asymmetric cheek has not succeeded because of the ninety per cent.
What the operation is measured by
The measure is not volume removed. It is whether the material that was producing symptoms has been reduced, and whether the normal structures that were working before the operation are still working after it. Pain, recurrent swelling, firm nodules, visible distortion, material showing through thin skin: these are the targets. Material that is producing none of them, and that sits against a nerve, is frequently better left where it is.
The measure is not how much came out. It is whether what was causing the trouble is smaller and whether everything that was working still works.
There are cases where one operation does take out essentially all of it. A recent, discrete, encapsulated deposit in the subcutaneous plane with a clear border and little adhesion can be removed in a single procedure, and it should be. The staged, partial approach is a response to a particular situation, not a doctrine applied to every face. The assessment described on the previous page is what distinguishes the two.
02Deciding to leave something
The decision to leave part of a material is made twice: once before the operation on the imaging, and again during it on what is actually found. The second occasion overrules the first. Imaging is a prediction, and a prediction about adhesion is the least reliable thing imaging produces.
Four questions decide it. Is this part of the material producing a symptom? What is it lying against? How far would the dissection have to extend to take it, and through what? And what will the site look like once it is gone, given what else has already been removed around it?
The clinic’s stated working rule is that material which cannot be felt and which lies close to nerve or muscle is not pursued. That rule does most of the work. A deposit that the examining hand cannot find is by definition not producing the contour problem the patient came about, and if it sits in a plane where pursuing it means dissecting along a nerve, the arithmetic is straightforward.
The rule has an exception that matters. Where symptoms are clearly arising from a region — persistent pain, repeated episodes of swelling in the same place — material that cannot be palpated may still be the cause, and may still need to be found. The rule is about not chasing silent material, not about ignoring evidence.
| Consideration | Argues for removal | Argues for leaving it |
|---|---|---|
| Symptoms | Pain, recurrent swelling, a palpable nodule, visible distortion, material showing through the skin | No symptom attributable to that part |
| What it lies against | Normal fat, fascia, or a plane with nothing important in it | A named nerve, a vessel, the orbital septum, the lid margin |
| Adhesion | A definable border and a plane that opens | No border; the material and the tissue separate only by cutting tissue |
| Extent of dissection needed | Reachable from the access already made | Requires extending dissection into an untouched region |
| Consequence of the empty space | Enough surrounding tissue to fill or be re-draped | Removal would leave a depression that cannot be corrected at the same sitting |
| Type of material | Inflammatory, calcified, or known to provoke a persistent reaction | Quiet, stable and long unchanged |
Leaving material is described to patients before the operation rather than reported afterwards. It is a planned outcome and not a failure of the procedure, and where it is likely, it is named at the consultation along with what would prompt a further operation later. Material that stays is followed. If it starts to produce symptoms, it is reassessed, and further removal is one of the options at that point.
03Finding the boundary
Everything above assumes that the boundary between normal tissue and foreign material can be identified in theatre. Often it can. In the faces that have had several rounds of treatment elsewhere, it frequently cannot be identified with confidence, and the operation becomes a matter of reading several imperfect signals at once.
Before the incision
Palpation with the patient sitting up, and marking on the skin, are done before anaesthesia because tissue feels different once it is infiltrated and the face is flat. Ultrasound marking is added where the extent is greater than what can be felt. These marks are the only record of the boundary that is not affected by what the operation itself does to the tissue.
Colour and texture
Normal facial fat is yellow, soft, lobular and yields when a closed instrument is opened within it. Material that has been in place for years is usually paler, firmer and more uniform, and it does not yield in the same way. Grafted fat that failed to survive is often grey or white rather than yellow. An oil cyst has a firm wall and a liquid content that is unmistakable once it is entered. These differences are real, and they are least reliable in exactly the cases where they are most needed, because the older the material the less the difference shows.
How the plane behaves
The most dependable intraoperative sign is not appearance but behaviour. Normal tissue planes open. A blunt instrument introduced into normal subcutaneous fat separates it along its natural divisions with very little force. Where a foreign material sits, that separation stops. The tissue holds, and further progress requires sharp dissection rather than spreading. The point at which spreading stops working is, in practice, the working definition of the edge.
The capsule
Where the body has built a distinct capsule around a deposit, the capsule is the boundary and the operation becomes considerably more straightforward. Finding the outer surface of a capsule and staying on it allows the whole deposit to be lifted with its wall intact. Not every material provokes a capsule, and one that has been broken by an earlier attempt at removal elsewhere may no longer be a continuous structure.
When the boundary cannot be found
Where fibrosis has blurred the boundary past recognition, the operation is conducted from the other direction. Rather than following the material, dissection follows the structures that must be preserved: the nerve is identified and kept in view, the plane immediately above the periosteum is used as a floor, the undersurface of the dermis is respected as a ceiling. Removal then proceeds within the space defined by those, and what remains outside that space remains.
Two opposite errors are possible at the boundary, and both are described to patients. Removing too much causes hollowing, contour irregularity and nerve injury, none of which is straightforward to correct. Removing too little leaves the symptoms in place and may mean a further operation. The balance is set individually according to the site, the material and what the patient came in complaining of.04Where nerve risk rises
Risk is not distributed evenly across the face. It concentrates where a nerve becomes superficial, where it is fixed and cannot be displaced, and where it passes through a plane that injected material also occupies. Those three conditions coincide in a small number of well-described places, and the plan is more conservative in each of them.
The temple and the lateral brow
The temporal branch of the facial nerve crosses the zygomatic arch and runs within the superficial temporal fascia towards the frontalis muscle. Over the arch it is superficial and has almost no soft tissue protecting it. Injury produces a brow that does not lift and a forehead that does not wrinkle on that side, and recovery is not assured. Material placed in the temple, which is common, frequently sits in or near the same fascial layer the nerve travels in.
The forehead and glabella
The supraorbital and supratrochlear nerves emerge at the upper orbital rim and run upwards in the forehead, becoming superficial as they ascend. They are sensory, so injury produces numbness and sometimes persistent discomfort in the forehead and scalp rather than weakness. The glabella carries an additional and separate concern, because its vascular anatomy communicates with the orbit, which is the reason old injections in this region are approached with particular care.
The midface and the infraorbital nerve
The infraorbital nerve leaves its foramen a short distance below the orbital rim, roughly in line with the pupil, and fans out across the cheek, lower lid, side of the nose and upper lip. It is the single most relevant structure in midface removal, because the medial cheek is where filler is most often placed and because the nerve is fixed at the foramen and cannot be moved aside. Injury produces numbness across a region the patient uses constantly, and that numbness is noticed every day.
The jawline and the marginal mandibular branch
The marginal mandibular branch of the facial nerve runs near the lower border of the mandible, deep to the platysma, crossing the facial vessels. Its course relative to the bony border varies between individuals and with the position of the head. It supplies the muscles that depress the lower lip, so injury produces an asymmetric lower lip on smiling and speaking. It is a small branch with limited redundancy, which is why removal along the jawline and around the chin is approached deep to the platysma with the plane kept deliberately deep.
The cheek and the zygomatic and buccal branches
The zygomatic and buccal branches supply the muscles that elevate the lip and close the eye. They cross the cheek deep to the SMAS and, unlike the temporal and marginal mandibular branches, they connect with each other, so an injury to one small branch is less often permanent. That is a reason for relative rather than absolute reassurance. Material in this region is frequently deep enough to sit in the same plane.
The lower eyelid
The lower lid carries a risk that is structural rather than neural. Removing material that has become entangled with the orbicularis muscle, the septum and the lid support can leave the lid retracted or turned outwards, and lid position is far harder to restore than volume. This is the region in which the least is taken relative to what is present, and the threshold for leaving material is correspondingly lower.
| Region | Structure at risk | Consequence of injury |
|---|---|---|
| Temple and lateral brow | Temporal branch of the facial nerve | Brow does not lift; forehead does not wrinkle on that side |
| Forehead and glabella | Supraorbital and supratrochlear nerves | Numbness and sometimes persistent discomfort in forehead and scalp |
| Medial cheek and infraorbital region | Infraorbital nerve | Numbness of cheek, lower lid, side of nose and upper lip |
| Cheek, deep to the SMAS | Zygomatic and buccal branches | Weakness of lip elevation or eye closure; often temporary given branch overlap |
| Jawline and chin | Marginal mandibular branch | Asymmetric lower lip on smiling and speaking |
| Lower eyelid | Orbicularis, septum and lid support | Lid retraction or outward turning; difficult to correct |
05Operating while inflammation is active
The clinic’s stated principle is that an inflammatory reaction is settled with medical treatment before removal is undertaken. The reasons are practical rather than theoretical, and they are the same reasons at every site.
Inflamed tissue bleeds. Bleeding obscures the operative field, and a field that cannot be seen is a field in which the boundary cannot be read and a nerve cannot be identified. The entire method described above depends on seeing what is being separated, and inflammation removes that.
Inflamed tissue also loses its planes. Oedema and acute fibrin make everything adherent to everything, so the distinction between material and normal tissue becomes harder to draw at exactly the moment the temptation to take more is greatest. More normal tissue comes out than intended and less of the material than expected.
Wound healing is worse in inflamed tissue and infection risk is higher where a foreign material is present. An operation performed into active inflammation risks turning a localised problem into a wider one.
What settling means
Stabilisation usually means a period of medical treatment, and the time required depends on the degree of inflammation and on what kind of material is involved. It is measured in weeks rather than days, and the treatment is reviewed rather than simply completed. Where the response is partial, the interval is extended rather than the operation brought forward.
There is one point about medical treatment worth stating plainly, because it explains why some patients arrive having already been treated more than once. Steroid treatment can reduce inflammation and reduce the bulk of the reaction without removing the material that provoked it. The lump becomes smaller and the tenderness settles, and the cause remains where it was. When the effect wears off the same problem returns, sometimes with more fibrosis around it than before. A course of steroid that works is useful as a way of reaching a quiet enough field to operate in; it is not usually a substitute for the operation.
When waiting is the wrong answer
An abscess is drained rather than observed. Material threatening the skin, where a deposit is producing thinning, dusky discolouration or breakdown, is not left to settle either, because the skin will not recover on its own once that process is established. In both situations the intervention is focused on the immediate problem, and the considered removal of the rest of the material is planned separately for later.
- Why timing is not a formality
- Bleeding
- An inflamed field obscures the boundary and hides the nerve
- Loss of planes
- Oedema makes normal tissue and material inseparable at the moment precision is most needed
- Healing and infection
- Wounds do worse, and a foreign material raises the risk further
- Typical interval
- Weeks rather than days, extended where the response is partial
- Exceptions
- Abscess and impending skin breakdown are treated immediately, and the rest planned later
06When a dissolving agent is tried first
Where the material is identified as hyaluronic acid, or where the history and the imaging make that likely, hyaluronidase is considered before an operation is planned. It is the one situation in this field in which a non-surgical option genuinely resolves the problem, and it would be poor practice to operate without having established whether it applies.
It is attempted where the injection history is consistent with hyaluronic acid, where the material appears on ultrasound as a well-defined pocket rather than as dense echogenic tissue, where symptoms are mild enough to allow time for a trial, and where an operation would require dissection near a structure worth avoiding. In the last of those cases the balance is clear even when the probability is moderate.
It is not attempted, or is not expected to resolve matters, where the material is known or strongly suspected to be calcium-based, collagen-based or a synthetic polymer. Polycaprolactone-based products such as Ellansé are not broken down by hyaluronidase, and neither are the permanent materials. Grafted fat does not respond to it, and neither does the fibrous tissue that forms around any of these, which is frequently the part the patient is actually feeling.
Where a trial is undertaken, what it produced is recorded rather than remembered impressionistically, because a partial response says something useful about how many materials are present. Where the response is incomplete, the operation is planned against what is left rather than against what was there at the start.
Caution applies in one specific situation. In a region where an operation is already likely, repeated high-dose hyaluronidase changes the tissue and acts on the patient’s own hyaluronic acid as well as on the product. Where there is no realistic prospect of a response, a trial is not a harmless way of deferring a decision.
07Which regions justify an endoscope
An endoscope earns its place where the material is deep, where the nerves are unforgiving, and where the alternative is either a long incision in visible skin or a dissection carried out largely by feel. The clinic identifies the forehead, the glabella and the temple as the regions where it is used, and the reasoning follows from the anatomy described above.
Three things are gained. The first is magnification, which matters where a nerve branch is a millimetre wide and lies within the fascia being dissected. The second is illumination at depth, since a deep pocket reached through a small opening is otherwise dark at exactly the point where seeing matters most. The third is the position of the incision: access through the hair-bearing scalp allows a forehead or temple to be approached without an incision on the forehead itself.
There are regions where it adds little. Superficial subcutaneous material that can be felt and reached directly does not need it. The deep cheek can often be approached through the mouth, with direct vision and no external scar. Lower lid work is done through the conjunctiva or at the lash line under direct vision, in a space too small and too shallow for an endoscope to help.
The choice of instrument is downstream of the assessment, not upstream of it. What the ultrasound and CT established about plane, depth and extent decides which of the available approaches is appropriate, and more than one may be used in the same operation on the same face. The temporal endoscopic approach used at this clinic was developed by its founding director, Dr Jung Su Lee.
| Situation | Approach usually considered |
|---|---|
| Soft material, little adhesion, subcutaneous plane | Cannula through a small access point |
| Adherent material within reach of the access | Ultrasonic instrumentation to separate adhesion, with direct vision |
| Deep material in the forehead, glabella or temple | Endoscope through a scalp incision |
| Hard, encapsulated or calcified material | Open dissection through a planned incision |
| Deep cheek | Intraoral approach where the position allows it |
| Lower eyelid | Conjunctival or skin approach under direct vision |
08Where to stop
Everything on this page is one decision described from several directions. How much comes out is settled by what is producing symptoms, what lies against the material, whether a boundary can be found, whether the tissue is quiet enough to work in, and what the site will look like once the space is empty. Those considerations are weighed in advance and then weighed again with the tissue in view, and the second weighing is the one that governs.
The point at which an operation stops is not the point at which nothing is left. It is the point at which continuing would cost more than it returns. Recognising that point is the substance of the work, and it is why two faces with the same imaging findings can properly end up with two different operations.
Removal is also not the end of the sequence. A space that a material occupied for years leaves hollowing, slack skin and irregularity behind it, and the question of what to do about that belongs to a separate decision with its own timing. It is considered from the beginning, because the amount that comes out and what will be needed afterwards are the same question asked twice.