Nerve injury: the risk, and how it is reduced

Most numbness after dental work is the local anaesthetic and nothing else. An injection at the back of the lower jaw — an inferior alveolar block — numbs the lower teeth, the lower lip and chin on that side, and usually the tongue as well. Two to five hours is the ordinary range, depending on the agent used and on whether a block or an infiltration was given. A longer-acting agent, chosen so that the first evening is more comfortable, can hold the lip numb until bedtime and sometimes into the night. None of that is a nerve injury.

What is not ordinary is numbness that is unchanged the following morning, numbness that resolves and then returns, numbness that appears for the first time several days after surgery, or sensation that is altered rather than simply absent — tingling, prickling, burning, an electric quality, or a taste that has changed. Those are the findings that need to be reported, and reported on the day you notice them.

The reason the timing is strict is biological rather than administrative. Where a nerve has been divided or compressed, the reported results of surgical repair are poorer the longer the interval between the injury and the operation. Scar tissue forms at the site, the connection between the nerve and the tissue it supplies degrades, and the brain's map of that area changes. A referral made at four weeks and a referral made at four months are not the same referral, even when the words on them are identical.

• Numbness of the lower lip, chin, tongue or floor of the mouth that has not begun to change by the morning after surgery.
• Numbness that had resolved and then came back.
• Tingling, prickling, crawling, or a sensation that the area is tight or swollen when it is not.
• Burning, aching, shooting or electric-shock pain in the lip, chin or tongue.
• Taste that is blunted, absent, metallic or unpleasant.
• Biting the lip or the tongue without feeling it, or liquid escaping from the corner of the mouth.
• An area that is numb to touch but painful when touched — that combination matters more than either finding alone.

There is a second pattern that is urgent for an entirely different reason. Altered sensation that arrives alongside swelling which is growing rather than settling, or alongside fever, is more likely to reflect infection than injury, and infection in the lower jaw is managed on a same-day clock rather than a nerve clock.

Both nerves involved in lower wisdom tooth surgery are branches of the mandibular division of the trigeminal nerve, and both are purely sensory. Neither of them moves anything. A nerve injury of this kind does not make the face droop, does not weaken the smile and does not affect the eyelid — those functions belong to the facial nerve, which lies elsewhere and is not in the surgical field for a wisdom tooth. Sensation is affected on one side only, the side that was operated on.

The inferior alveolar nerve

This nerve enters the jaw through an opening on the inner surface of the ascending part of the mandible, then travels forward inside the bone in a canal, giving branches to the lower molars and premolars along the way. It leaves the bone at the mental foramen, near the second premolar, and its terminal branch supplies the skin of the chin, the skin and red border of the lower lip, and the gum in front of the lower teeth.

The roots of a lower wisdom tooth can sit above that canal with bone between them, rest directly on it, be grooved by it, or — less commonly — straddle it, with the canal passing between two roots. Injury produces numbness or altered feeling in the lower lip and chin on that side. Taste is unaffected. The tongue is unaffected.

The lingual nerve

The lingual nerve runs downwards and forwards in soft tissue, passing close to the inner surface of the jaw in the wisdom tooth region before turning towards the tongue. Its position there is variable. It commonly lies against the bone at or just below the crest of the lingual plate; in some people it sits at or above that crest, which places it directly beneath the gum behind the wisdom tooth. No radiograph or scan taken for this surgery shows it. It does not appear on a panoramic film and it does not appear on a CBCT scan, because it is soft tissue and those studies image bone and teeth.

It supplies general sensation to the front two-thirds of the tongue, the floor of the mouth and the gum on the tongue side. It also carries the taste fibres of the chorda tympani to that part of the tongue. Injury therefore produces numbness of one side of the tongue and altered or absent taste together — which is why a change in taste after lower wisdom tooth surgery is a specific finding and not a vague one.

Two smaller nerves

The long buccal nerve supplies a patch of cheek and the gum on the cheek side, and it is sometimes divided during the incision. The resulting numbness is usually small, often unnoticed, and generally recovers. The mylohyoid nerve occasionally accounts for a small area of altered sensation under the chin. Neither is in the same category as the two nerves above, but both are mentioned so that a numb patch of cheek is not mistaken for something larger.

Numbness is the word patients are given at consent, and it describes only one of several outcomes. The clinical vocabulary is more specific, and it is worth knowing because it is the vocabulary that will be used if you have to describe what you are experiencing over the telephone.

The practical consequences follow from the anatomy. A numb lower lip does not report a bite, so the inside of it can be chewed at a meal without any signal. Liquid escapes from the corner of the mouth because the lip cannot feel where the rim of the cup sits. Shaving, applying lipstick, and drinking anything hot all become tasks done by sight rather than by feel. Speech is usually intelligible with lip involvement alone; it changes more noticeably when the tongue is involved.

Where the lingual nerve is affected, the tongue side of the mouth stops reporting texture and temperature accurately as well as touch, and the tongue is bitten in the same way the lip is. Taste and flavour are not the same thing — most of what is experienced as flavour comes from smell — so food often still tastes of something, but of less, or of something metallic that was not there before.

Distress is not proportional to the size of the numb area. A large area that is simply numb and comfortable can often be adapted to over a few weeks. A small area that burns, or that hurts when touched, can be intrusive out of all proportion to its size, and it is the type of sensation rather than its extent that determines how it is managed. If your work depends on precise lip control — a wind instrument, for example — that is a fact to raise before the surgery is planned, not afterwards.

Temporary alteration of sensation after removal of a lower wisdom tooth is uncommon. Alteration that is still present a year later is rare. Both statements are true, and they describe different things, which is the source of most of the confusion patients encounter when they read about this.

Published figures vary widely between studies, because the teeth studied, the way sensation was tested, who did the testing and how long patients were followed for all differ. A single percentage applied to every tooth in every jaw would not describe your tooth. What can be said usefully is comparative: the risk attaching to a shallow, erupted lower wisdom tooth sitting well clear of the canal is not the risk attaching to a deeply impacted distoangular tooth whose roots are grooved by the canal on a scan. The number that matters is the one that belongs to your tooth, and it is arrived at from your imaging at the consultation.

Upper wisdom teeth are a separate matter entirely. Neither the inferior alveolar nerve nor the lingual nerve is in the surgical field for an upper third molar; the structure of concern there is the maxillary sinus. A numb lip is not a feature of upper wisdom tooth removal, which is one of the reasons upper and lower teeth are consented separately rather than as a set of four.

There is also a pattern to recovery, and it is the reason the follow-up is structured the way it is. Where sensation is going to return without intervention, some change is usually detectable within the first two months, and improvement then continues over the months that follow. Improvement first appearing beyond a year is unusual. This is why an unchanged examination at eight weeks is treated as a finding in its own right rather than as a reason to keep waiting.

Signs on a panoramic radiograph

The radiographic signs that indicate a close relationship between a lower wisdom tooth and the inferior alveolar canal were set out by Rood and Shehab in 1990, and they remain what a panoramic film is read for. Seven signs are described:

• Darkening of the root where it crosses the canal.
• Sudden deflection or bending of the root at the canal.
• Narrowing of the root where the canal crosses it.
• A dark and bifid root apex.
• Interruption of one or both white lines that outline the canal.
• Diversion of the canal from its expected course.
• Narrowing of the canal itself.

Interruption of the white lines, diversion of the canal and darkening of the root are the three usually treated as carrying the most weight. Any of them on a lower third molar film is the point at which a three-dimensional scan is discussed, because a two-dimensional image has reached the limit of what it can answer.

Features of the tooth and of the patient

• Depth. A deeply buried tooth sits closer to the canal and requires more bone removal to reach.
• Angle. Distoangular and horizontal impactions typically require more bone removal at the back of the tooth and more sectioning than a vertical one.
• Root form. Bulbous, hooked or widely divergent roots grip the bone and resist delivery, and resistance is met with force.
• Age. Roots are fully formed and bone is denser from the mid-twenties onward, and nerve recovery tends to be slower with increasing age.
• Existing infection around the tooth, which alters tissue planes and bleeding.
• Previous surgery in the same area, where scar tissue obscures the normal landmarks.

Features that specifically threaten the lingual nerve

The lingual nerve is not endangered by the same things. What matters for it is roots angled towards the tongue side, a lingual plate of bone that is thin or already perforated by the tooth, any need to lift and hold the tissue on the tongue side, bone removal or sectioning carried out on that aspect, and the individual anatomical variation that places the nerve at or above the crest of the lingual plate. That last one cannot be predicted from any image, and it is the limit of what pre-operative planning can do for this nerve.

A small number of injuries relate to the injection rather than to the surgery, and the lingual nerve is the one usually involved. If a sharp electric shock runs down the tongue or into the lip while local anaesthetic is being given, say so at the time. It is recorded, and it changes what is watched for afterwards.

Some of the factors above are modifiable by how the operation is planned and carried out. Some are not. A root that is genuinely perforated by the canal carries a risk that belongs to the anatomy, and no technique removes it.

A panoramic radiograph is a flattened image of a curved structure. It shows the canal and the roots in the same vertical column and tells you they overlap. What it cannot tell you is whether the canal lies on the cheek side of the roots, the tongue side, directly beneath them, or between them — and that is precisely the information that changes how a tooth is approached.

A cone-beam CT resolves that. It shows the position of the canal in three planes, whether the thin cortical wall around the canal is intact or lost, whether a root is merely touching the canal, grooved by it, or perforated by it, and how much bone lies between the tooth and each surface of the jaw. It also shows the thickness and shape of the lingual plate, though not the lingual nerve itself.

Four things follow from that information. The consent conversation becomes specific to your tooth rather than generic. The choice between complete removal and coronectomy becomes a decision made on evidence rather than an estimate. The surgical approach is settled in advance — where to section the tooth, which direction to move each fragment, and where an elevator must not go. And occasionally the scan supports the opposite decision entirely, that a quiet, deeply buried tooth with an intimate canal relationship is reviewed at intervals rather than removed.

It carries more radiation than a panoramic film and considerably less than a medical CT of the same region, which is why it is taken to answer a specific question — most often, one of the radiographic signs appearing on the panoramic film — rather than for every lower wisdom tooth. If a scan is recommended to you, ask what question it is expected to answer, and you will be given one. Expect to see the images yourself, on screen, with the canal and the roots pointed out.

Coronectomy is the deliberate removal of the crown of a lower wisdom tooth with the roots left in the jaw. The roots are reduced so they sit below the crest of the surrounding bone, the pulp is not instrumented, and the gum is closed over the site. The roots are not elevated, rocked or touched at their tips.

That last sentence is the whole rationale. The instrument that injures the inferior alveolar nerve is usually the one applied to a root that is sitting on the canal. Coronectomy is the decision not to apply it. It addresses the reasons the tooth was a problem in the first place — the crown, the pocket behind the second molar, decay, recurrent pericoronitis — while leaving alone the part of the tooth that sits against the nerve.

• Considered when: the roots are in genuine contact with the canal on a CBCT scan; the tooth is otherwise sound with a healthy pulp; there is no infection at the root tips and no cyst around the tooth; and follow-up over years is realistic for that patient.
• Usually not appropriate when: the tooth is mobile, since loose roots cannot be left behind; decay extends into the roots or the pulp is already dead or infected; there is periapical infection or a cyst involving the roots; the tooth lies horizontally in a position where removing the crown would inevitably loosen the roots; or the patient's medical circumstances make a retained root inadvisable, which includes significant immunosuppression and planned radiotherapy involving the jaws.

The things that can go wrong with a coronectomy are known and are part of the consent. The roots can become mobile while the crown is being sectioned off, in which case they have to be removed, and the nerve risk of a complete removal returns with them — this is why the consent covers both operations before the day, since the decision to convert is sometimes made with the tooth already open.

Retained roots commonly migrate away from the canal over the following months and years, usually moving upward through the bone. That is generally favourable, and where a root eventually needs removing it is by then further from the nerve than it was at the first operation. Infection of the retained roots is uncommon and is managed like any other post-surgical infection, occasionally by removing them. Rarely, a root migrates far enough to appear through the gum and requires a second, smaller procedure. Radiographic review at intervals is part of the plan, and it is agreed before the operation rather than raised afterwards.

Coronectomy is a considered operation chosen against a scan, and it is discussed and consented before the day of surgery.

What is used to keep you comfortable during the procedure is agreed at the consultation, and depends on the procedure itself, the number of teeth being removed and your medical history. Every lower wisdom tooth removal involves local anaesthetic, which is given as an inferior alveolar block and numbs the lower teeth, the lip and chin, and usually the tongue on that side.

Here, the approach to the tooth is from the cheek side. The incision and the flap are designed so that the tissue on the tongue side is not lifted where lifting it can be avoided. Where access to that side is unavoidable, the tissue is raised in a single layer directly against the bone and held in a way that protects it rather than stretching it, because stretch is a mechanism of injury in its own right.

Dividing the tooth rather than delivering it whole is the other structural decision. A sectioned tooth comes out through a smaller opening, needs less bone removed, and requires far less force. Force is the underlying problem: an elevator working against a root gripped by dense bone transmits that force to whatever lies on the other side of the root, and on a lower wisdom tooth that can be the canal.

Beyond that, the discipline is about what is not done. No instrument is placed beyond the root tips. The floor of the socket is not curetted blindly. Suction is not applied directly into the depths of the socket where the canal may be open. Where the plan is built from a CBCT, the direction each fragment is moved is decided before the first cut.

If the canal is opened during the removal — which does happen, and is not in itself an injury — the nerve is left undisturbed, the socket is not curetted, and the finding is recorded. You are told about it before you leave the building. That matters, because it changes what both you and the practice watch for over the following days, and because you should not be left to work out on your own whether a numb lip the next morning is worth mentioning.

None of this removes the possibility. Technique changes the odds. It does not set them to zero, and a tooth whose roots are perforated by the canal carries a risk that belongs to the anatomy rather than to the operation.

The first step is the telephone call, made on the day you notice. Everything below depends on it having been made.

At the assessment, the affected area is mapped and drawn on a diagram so that a change can later be measured rather than remembered. Light touch, the ability to distinguish sharp from blunt, two-point discrimination and — where the tongue is involved — taste are all tested and recorded with a date. That examination is repeated at intervals, because the direction of travel is what determines the decision, and a direction needs at least two points to establish.

The pattern of onset carries information too. Numbness present from the moment the local anaesthetic should have worn off suggests something that happened during the operation. Numbness or tingling that appears at one to three days, having been normal before, more often reflects pressure from swelling or a haematoma, and it often settles as the swelling does — but that is a finding to be watched and recorded, not assumed. Late onset accompanied by increasing pain and swelling points towards infection and is a different problem with a different treatment.

Referral for assessment by a unit that undertakes trigeminal nerve repair is made where sensation is completely absent, where the symptoms are unpleasant rather than merely absent, or where repeat testing shows no change. Lingual nerve injuries are generally referred earlier than inferior alveolar nerve injuries, because a divided lingual nerve has no bony canal holding its two ends in alignment and spontaneous recovery is correspondingly less predictable. Where the sensation is painful rather than absent, medication of the kind used for neuropathic pain is sometimes appropriate, and it is prescribed against your full medication list rather than in isolation.

While sensation is altered, a few practical habits prevent the secondary injuries that a numb area invites.

• Check the lip and the inside of the cheek for bite marks after eating, since the bite itself will not be felt.
• Test the temperature of drinks against a part of the mouth that has normal sensation, or against the back of the hand — a numb tongue or lip will not report a scald until afterwards.
• Chew on the other side until sensation is reliable again.
• Take extra care shaving over a numb chin or lip.
• Brush the affected side thoroughly and deliberately — a numb area does not report trapped food or plaque.
• Keep a short dated note of what you can and cannot feel, week by week. Written observations taken at the time are more useful at an assessment than recollection is.