Wisdom teeth removal: what actually happens

Most people reading this are deciding about an operation that has not been booked. Some are reading it in the middle of a problem, which is why the urgent material is first: the things that cannot wait are not the things that hurt the most. If you have already had surgery here, the written instructions you were given govern, and this page is general.

Pain is a poor guide to urgency. Severe pain often accompanies something entirely local — an exposed nerve, a dry socket — and a spreading infection sometimes hurts less as it spreads, because the pressure that caused the pain has escaped into the surrounding tissue. Swelling tells you where the infection has reached, and that is the question that changes management. Fever, feeling generally unwell, and any change in swallowing or in the voice are the other signals worth acting on.

Some patients are moved up that list by their medical history rather than by their symptoms. A dental infection in someone who is immunosuppressed, having chemotherapy, has poorly controlled diabetes, or takes a bone-modifying medication moves faster and is seen sooner, at a stage that would otherwise be an ordinary appointment. If you take an anticoagulant or antiplatelet medication and a socket starts bleeding, call after the first thirty minutes of pressure rather than working through the ladder below. Say which medication it is when you call.

Third molars arrive last, usually somewhere between seventeen and twenty-five, into whatever room the twenty-eight teeth ahead of them have left. The space behind the second molar is the constraint, and it is set by the shape of the jaw rather than by anything the patient did. Where there is room and the tooth comes up straight, it is a tooth like any other. Where there is not, it stops part of the way, tips into its neighbour, or stays in the bone.

Impaction is a statement about position, not about symptoms. A tooth can be fully impacted and silent for a decade, and a tooth can be fully erupted and cause trouble every few months. Position and disease are two separate findings, and they are assessed separately rather than treated as one answer.

Partial eruption is the state that produces most of the problems. When a cusp breaks through and the rest of the crown stays covered, a flap of gum sits over the tooth with a pocket underneath that no toothbrush reaches. Bacteria collect there, and pericoronitis — inflammation, often with infection, of that overlying gum — follows. It tends to recur, because the anatomy that caused it does not change between episodes.

The four things that actually go wrong

• Recurrent pericoronitis. The usual reason a third molar announces itself, and the usual reason one is removed.
• Decay on the back surface of the second molar, where a forward-tipped third molar contacts it below the gum line. That surface is difficult to reach with a drill and difficult to restore well, which is how a sound tooth ends up needing treatment because of its neighbour.
• Bone loss and a deep pocket behind the second molar, which behaves like gum disease confined to one site.
• A cyst forming from the follicle around an unerupted crown. Uncommon, usually producing no symptoms, and typically found on a radiograph taken for another reason. Resorption of the second molar root by a horizontal third molar sits in the same category: silent, uncommon, and picked up on a film.

None of that is the default outcome. Many third molars never cause any of it, and a tooth that is fully erupted, cleanable and biting against something is a tooth, not a diagnosis. The purpose of assessing them is to find the ones where something is happening or is likely to, and to record that the rest are being watched.

One reason that is sometimes given for removal does not hold up: crowding of the front teeth. Third molars are not an established cause of lower incisor crowding, and preventing crowding is not used as an indication for removal here. Where crowding is the only finding on the film, ask what else the film shows.

Two questions run the decision. Is there disease now, and if not, how likely is disease later. The first is answered from the examination and the films. The second is a judgement about a specific tooth in a specific mouth rather than a rule that can be applied at a distance.

Findings that usually settle it in favour of removal: pericoronitis that has happened more than once, decay in either tooth at the contact point, a periodontal pocket with bone loss behind the second molar, a lesion around the crown that is wider than a normal follicle or that has widened between films, resorption of the second molar root, or a tooth that is unrestorable in its own right.

Findings that support monitoring: a deeply buried tooth, no communication with the mouth, no decay on the neighbour, no pocket, a normal follicular outline, and no symptoms. That tooth can be reviewed with radiographs at intervals rather than removed, and monitoring is written down as the plan when it applies. Where removal of a symptomless tooth is recommended, the reasoning for that particular tooth is given, and you can ask to have it written down as well.

Age is a genuine argument and an easily overstated one. In the late teens and early twenties the roots are often not fully formed and the surrounding bone is more forgiving, and recovery is generally quicker than the same operation two decades later. That is an argument about probability rather than a deadline, and recovery varies between individuals more than any classification predicts.

Whether all four come out at one appointment is decided rather than assumed. Four impactions in one sitting means one recovery instead of two, and a harder few days. Two at a time, or one side at a time, keeps one side of the mouth working. There is no single correct answer. What is appropriate depends on the teeth, the anatomy, the medical history and what you want, and it is agreed with you before a date is set.

The consultation is a history, an examination, imaging, and a discussion of what the imaging showed. Surgery is normally booked for a separate visit. That is partly practical, because the arrangements for the day have to be made in advance, and partly deliberate: a consent conversation that ends in an operation an hour later is not a decision made with much room to think. Where there is infection, uncontrolled pain or a tooth that is actively damaging its neighbour, the sequence compresses.

Bring an accurate medication list. It is the single most useful thing you can arrive with: everything prescribed, plus supplements and anything bought over the counter. Also bring photo identification, your insurance information, and any radiograph taken elsewhere in the past few months, or ask your general dentist to send it ahead. A patient under eighteen needs a parent or legal guardian present for consent.

Several categories of medication change surgical planning, and none of them should be stopped without instruction. Anticoagulants and antiplatelet agents. Bisphosphonates and other bone-modifying agents, including ones given by injection years ago. Immunosuppressants. Diabetes medication, including the GLP-1 class. Anything for a heart valve or a recent stent. A previous complication under anaesthesia, in you or in a close relative, matters as well, and so do smoking, pregnancy, and any condition that affects healing or clotting.

The examination covers the teeth in question, the gum and bone around them, the bite, how wide the jaw opens, and a soft-tissue check of the tongue, floor of the mouth, cheeks and neck. That last part happens whatever you came in for. Anything unexpected there is explained at the time and referred on to the appropriate service.

Then the discussion. The findings are shown against the images, on screen, with the anatomy pointed out rather than described in the abstract. The options always include not operating. Expect to be told what happens if nothing is done, and expect the answer to be specific to your teeth rather than general. A second opinion is a reasonable thing to want, and asking for the films to take with you is a reasonable thing to ask.

The visits that follow surgery are short, and there are usually one or two of them. A post-operative problem is settled by examining the site rather than over the phone, so it is worth knowing in advance where you will be travelling to for that. This practice is in Aventura, in north Miami-Dade.

Fees are not published on this site. Benefits are verified with your plan and you are given a written estimate, itemised by procedure code, showing the expected plan payment and your share separately. An estimate is a calculation from a snapshot: the codes planned from the examination and the films, and the benefit information the plan returned that day. Benefits are determined by your insurer, and the final amount follows the procedure that was actually performed.

A panoramic radiograph shows all the teeth, both jaws and the sinuses in one image, and it is the standard starting point for third molars. It establishes which teeth are present, their angle, how deep they sit, the shape of the roots, and the apparent relationship between the lower roots and the inferior alveolar nerve canal.

That last one is the reason a second study is sometimes needed. A panoramic film is a flattened two-dimensional image of a curved three-dimensional structure, so a root that appears to cross the canal may be sitting to the cheek side of it, to the tongue side of it, or genuinely against it. Rood and Shehab set out the panoramic signs that suggest a true relationship in 1990, and they are still the ones looked for: darkening of the root where it crosses the canal, interruption of the white lines of the canal, diversion or narrowing of the canal, narrowing or deflection of the root, and a dark or bifid root apex.

Those signs mean look further. They do not mean injury, and they are not a prediction about you. Where one or more is present, a cone-beam scan answers the question the flat film cannot: whether there is intact bone between the root and the canal, and which side of the canal the root actually sits on.

A cone-beam scan can change surgical planning — flap design, how the tooth is divided, whether the roots are approached at all. What it has not been shown to do is reduce the rate of permanent nerve injury, and that distinction is worth stating plainly, because it means a scan is taken to inform a decision rather than to make the surgery safer by itself. A limited-field scan carries a higher radiation dose than a panoramic film and a lower one than a medical CT of the head, so it is taken where a question depends on it rather than as routine. If a scan is recommended, ask what question it is expected to answer. There is a specific answer for every scan taken here.

Consent is the conversation, and the document records it. The risks that matter are the ones attached to your teeth, so a lower tooth near the canal is consented differently from an upper tooth near the sinus, and four teeth are consented as four separate propositions rather than as a set.

Lower teeth, and the two nerves

The inferior alveolar nerve runs in a canal beneath the lower molar roots and supplies feeling to the lower lip and chin on that side. It is purely sensory: injury alters sensation and does not weaken the lip or change the movement of the face. The lingual nerve runs in the soft tissue on the tongue side and supplies sensation and taste to that side of the tongue. Its position varies between people, and it is occasionally at or above the level of the bone crest, which is why technique on that side is deliberate rather than routine.

Temporary altered sensation of the lip and chin follows a small minority of lower third molar removals, and it is more likely where a scan shows the roots in contact with the canal. Most changes that recover do so over the first weeks to months. Permanent alteration is uncommon, and it is described specifically as part of the consent conversation for the tooth it applies to. Altered sensation can be numbness, tingling, pins and needles, or a patch that feels ordinary until it is touched.

How long ordinary post-operative numbness lasts depends on which agent was used. Shorter-acting local anaesthetic wears off over a few hours; the longer-acting agents often chosen for lower blocks can keep the lip, chin and half the tongue numb for much of the day. Ask which was used before you leave, so you know what is expected and, more importantly, when it stops being expected.

Upper teeth, and the sinus

Upper third molars are often a less involved removal, and they carry their own specific risks. The floor of the maxillary sinus sits close above the roots, so a small opening into the sinus can occur at the time of removal. Where it is recognised at the time it is managed at that visit; many close with the measures taken then, and some need a further procedure. The bone behind the tooth can fracture, and a tooth sitting high and back can be displaced into the space behind the jaw. These are uncommon, and they are why an upper tooth is planned from a film rather than assumed to be simple.

Where the sinus has been involved, the instructions afterwards include not blowing the nose, sneezing with the mouth open, and avoiding anything that raises pressure in the nose for a period that is specified on the day. Separately from that period: air or fluid passing between the nose and the mouth, a nasal discharge on one side only, or a whistling sensation when you drink is reported the same day, at any point after surgery.

Risks common to both

• Dry socket — loss of the clot, usually day three to five, most often in a lower socket.
• Infection of the surgical site, which announces itself as swelling that grows again after day three, fever, or discharge.
• Bleeding that restarts, most often on the first night.
• Swelling, bruising along the jaw and neck, and restricted jaw opening, all of which are expected rather than complications.
• Damage to a filling, crown or the adjacent tooth, particularly where a restoration is already large.
• A root fragment that is deliberately left because retrieving it would do more harm than leaving it.
• Fracture of the jaw, which is rare and associated with deeply impacted teeth in dense bone.
• A tooth that proves more difficult than the film suggested, so the plan changes mid-procedure or the case is staged.

Where the roots are genuinely bound to the canal, coronectomy is sometimes the more proportionate operation: the crown is removed and the root structure is deliberately left in place. It is not suitable for every tooth — a mobile tooth, active infection involving the roots, and certain root positions rule it out — and it is decided against the scan and the medical history. The downside is that retained roots often migrate away from the canal over the following year or two, and a proportion eventually need retrieving in a second operation. That is set out before it is chosen rather than afterwards.

Local anaesthetic is the foundation of everything else. It blocks sensation at the surgical site and nothing more: awareness, memory and coordination are untouched. Onset takes a few minutes, and how long the numbness lasts depends on the agent used and on whether a nerve block or an infiltration was given.

What that means in the chair is worth knowing in advance. You will feel pressure, and you will feel vibration, and you will hear the handpiece and the suction. Pressure is not pain, and patients who have not been told to expect it can interpret it as pain starting. If something is sharp rather than firm, say so — more anaesthetic is given, and there is no prize for tolerating it.

A numb lip and cheek are easy to bite without noticing, which is the commonest self-inflicted injury after an extraction. Wait for sensation to return before eating anything that needs chewing, and keep an eye on children and teenagers, who are the ones most likely to test it.

What is used to keep you comfortable beyond local anaesthetic is agreed at the consultation, and it depends on the procedure, the number of teeth and your medical history. Whatever is agreed brings its own requirements with it — a fasting window, an escort home, restrictions on driving or work afterwards — and those are given to you in writing before the day rather than described on the morning. Where a fasting window is given to you, it is the one pre-operative instruction with no discretion attached to it, and an appointment is rescheduled rather than run if it has not been followed.

Local anaesthetic does nothing for anxiety. If the sounds, the duration or a previous experience are the difficulty rather than the pain, say so at the consultation. It changes what is planned and how the appointment is sequenced, and it is an easier conversation before the day than during it.

The appointment starts with confirming who you are, which teeth are being removed, and that the consent still reflects what you want. Any change since the consultation — a new medication, a cold, a period of illness, a pregnancy — is asked about again at that point, and any of them can change what happens next. The plan can still be altered or postponed on the day, by you or by the surgeon.

Local anaesthetic is given and time is allowed for it to take effect properly, which is longer than most people expect for a lower block. Numbness is tested before anything else happens.

For a straightforward erupted tooth, the tooth is loosened with an elevator and delivered with forceps. Upper teeth often come out this way, because the bone around them is thin enough to expand rather than requiring removal. For an impacted tooth the sequence is longer: an incision, the gum lifted as a flap, then a channel of bone taken from the cheek side and from over the crown with a surgical handpiece under continuous sterile irrigation — bone cut without cooling dies, so the water is not optional. Only as much bone is removed as the access requires, because what is left supports the second molar afterwards.

The tooth is then usually divided. The crown is separated from the roots, and the roots from each other, so each piece comes out through a smaller opening. Sectioning sounds more invasive and is generally the opposite: it removes far less bone than delivering a tipped tooth whole. Once the tooth is out, the socket is irrigated, the follicle removed, sharp bone edges smoothed, and the flap replaced with sutures that dissolve over one to two weeks depending on the material used.

How long any of this takes is not knowable from outside the mouth. Two teeth with the same classification on the same film can behave completely differently once the crown is exposed, so what is given at the consultation is a range rather than a figure.

At the end you are given gauze to bite on, written instructions, and a number to call. Read the instructions before you leave rather than at home, and if someone is taking you home, the instructions are gone through with them too — the first evening is when questions arrive, and the person answering them should not be doing it from memory.

Recovery runs a recognisable curve, and knowing its shape is what makes a deviation obvious. The single most useful thing to know is not what is normal but what is not: pain that improves and then worsens, or swelling that grows again after day three, are the two patterns that mean something has changed, and both are dealt with quickly. The table below is a general guide, not a diagnosis, and it does not replace the written instructions you were given or being examined.

Ice is used on the day of surgery rather than after it. Twenty minutes on and twenty off, against the cheek over each site, for the whole of the first day, or as your instruction sheet directs. It is soothing and may limit how much swelling forms, and cold applied once swelling has already built does very little. Swelling still varies a great deal between people, and a badly swollen day two does not mean you did anything wrong. From day two the ice stops helping and warmth starts to.

Follow the instruction sheet you were given for what to take for pain and when to take it. It is set against your medical history and everything else you take, and it governs over any timing in a general article. The usual pattern is to take the first dose before the local anaesthetic wears off rather than waiting for discomfort to arrive, because staying ahead of it takes less medication than catching up to it. If you are unsure what you have been given, or how it combines with something you already take, call before adding anything to it.

No rinsing at all for the first twenty-four hours. From day two, warm salt water — a teaspoon in a cup — allowed to fall out of the mouth rather than spat, after meals. No straws, because the negative pressure can pull a young clot out of a socket. Smoking is the strongest thing you can change: forty-eight hours is the minimum worth aiming for and seventy-two is preferable, and vaping applies the same suction.

Jaw stiffness peaks with the swelling, because the muscles that close the jaw sit directly over the surgical field and object to being retracted. Opening may be limited to a couple of finger-widths at its worst. Gentle opening and closing several times a day helps; forcing it wide does not, and neither does keeping it shut. Most patients return to desk work in two to three days, though that varies with the number of teeth and how deeply they were impacted, and there is nothing wrong with taking longer. Physical work and exercise wait longer again, because raising the blood pressure raises the chance of restarting the bleeding.

Sockets close over at the surface during the second and third weeks, while the bone underneath fills over months. A lower socket traps food for two to four weeks after surgery, which is expected rather than a complication, and an irrigating syringe issued at the post-operative visit clears it without disturbing the tissue. Nothing rigid goes into the socket. A socket that starts to smell or taste foul, becomes painful again, or develops swelling around it is examined rather than assumed to be food.

Consultations cover a lot of ground quickly, and the question people most often remember afterwards is the one they did not ask. These are the ones that change what happens.

1. Which of my third molars are you recommending removing, and what is the finding on each one individually?
2. What happens if I leave this tooth for a year and have it filmed again?
3. What does the film show about the relationship between the roots and the nerve canal, and does it change what you would do?
4. If a scan is recommended, what question will it answer that the panoramic film cannot?
5. Is coronectomy something you would consider for this tooth, and if not, why not?
6. Are you recommending all four at once or staging them, and what is the reasoning either way?
7. What is used to keep me comfortable, and what does that require of me before and after the appointment?
8. Which local anaesthetic will be used, and how long should the numbness last before I should be calling about it?
9. What are the specific risks for my lower teeth and for my upper teeth, and how do they differ?
10. Who do I call at nine in the evening on the second night, and what is the number?
11. What would make you stop mid-procedure and stage the rest for another day?

Ask for the answers to be written on the estimate or the instruction sheet where they are specific to you. A decision you understand is the point of the consultation, and there is no version of this in which asking too many questions is a problem.