What should happen if something goes wrong

Most of what goes wrong after third molar surgery can be dealt with unhurriedly. A small number of things cannot, and in those the waiting does damage that the original event did not. Sorting one group from the other is the difference between a telephone call you make today and a conversation you have at a two-week review, by which point some of the options have closed.

Below that threshold sits a second group: findings that are not emergencies but are reported on the day you notice them rather than saved for the next appointment. Each is time-critical for a different reason, and none of them improves for being handled politely.

• Numbness or altered feeling in the lower lip, chin or tongue that is still present the morning after surgery, that goes and returns, or that appears for the first time days later. Tingling, prickling, burning, an electric quality or a change in taste all count.
• After an upper tooth: air or liquid passing between the mouth and the nose, a nasal quality to the voice, drink coming back down the nose, or a whistle when you breathe with the lips closed.
• A tooth, root or fragment that was not accounted for at the end of the operation, or a piece you were told had moved somewhere and would be watched.
• A bite that no longer meets the way it did, or an inability to bring the back teeth together on one side, after a lower tooth.
• Pain increasing from the third day onward rather than easing, particularly if it radiates to the ear.
• Bleeding that restarts. If you take an anticoagulant or an antiplatelet medication, telephone after the first thirty minutes of firm pressure rather than working further through the usual ladder, and say which medication it is when you call.
• Fever with facial swelling that is expanding rather than settling after the third day.

This list exists because the commonest route by which a time-critical complication goes untreated is not concealment. It is a patient who noticed something on Saturday, decided it was probably normal, did not want to be a nuisance, and mentioned it at the review the following Thursday. Five days is nothing for most findings on this page and a great deal for one of them.

No. A complication is an adverse event that can occur when an operation is performed competently, on a properly assessed patient, with reasonable technique. An error is a departure from reasonable practice. Some complications follow errors. Most do not. The two get conflated in both directions, and each conflation costs a patient something different.

The first conflation is the patient's: something went wrong, therefore somebody was careless. Third molar surgery involves cutting bone within millimetres of a nerve that cannot be seen, in a part of the mouth that is difficult to reach, in a jaw whose anatomy varies more than almost any other site a dentist operates in. A proportion of operations performed exactly as they should be will produce a numb lip, an opening into the sinus, or a chipped restoration on the tooth in front. Naming those possibilities beforehand is what a consent conversation is for.

The second conflation belongs to the surgeon, and it is the more damaging of the two: treating "this is a recognised complication" as the end of the conversation instead of the start of it. A recognised complication explains why something happened. It says nothing whatever about what happens next, and what happens next is the part a patient can actually judge.

So the question worth asking is usually not whether an error occurred. That question is hard to answer, frequently unanswerable from the outside, and it is not the one that determines how you come out of this. The question that matters is whether the response was a proper one — prompt, plainly stated, written down, and pointed at the right destination. A careful surgeon, a complication, and a well-run response is a good outcome. The same complication met with vagueness is a different situation entirely, whatever caused it.

What happened, in ordinary words; why it happened; what it means for you; what is being done about it; and what to watch for and telephone about. Told on the day, before you leave, and repeated in writing. A patient should not have to extract a complication from a surgeon, and should not first learn of one from a record they requested three months later.

The "on the day" part carries a complication of its own. A patient who has had general anesthesia is not in a state to absorb a serious conversation at discharge and will frequently remember none of it. That is a known property of the drugs used, not a failure of attention. Where something occurs under general anesthesia, the reasonable pattern is: told to the adult escorting you at discharge, written into the discharge instructions you take home, and gone through again by telephone the following day, when you can hold onto it.

Plain words matter more than they sound. "There was a small communication with the antrum" and "the socket opened into your sinus, and here is what that means for blowing your nose" describe the same finding. Only one of them survives the journey home.

What a disclosure should contain

• The event itself, named, in words you can repeat to someone else.
• Why it happened — the anatomy, the position of the tooth, what was found once the operation was under way.
• What it means for you specifically, in terms of symptoms, timescale, and what you can and cannot do.
• What was already done about it during the operation.
• What happens next, including who else is being involved, and the date of the next review rather than an invitation to come back if it does not settle.
• What would make you telephone before that date.

None of that requires an apology, an admission, or a form of words drafted by anybody. It is a clinical handover to the person the clinical information belongs to, and it is the same handover a surgeon would want if the patient were their own relative.

An operative note is written at the time, by the person who operated, and it is the document everything afterwards is built on. The referral letter is drawn from it. A second opinion is formed from it. Your own memory of the conversation will, within a year, be less reliable than you expect it to be. If a complication is not in the note, it effectively did not happen as far as every clinician who sees you next is concerned.

A note covering a complication should name the event in the same words that were used to you, describe what was found and what was done about it, record what was said to you and when, and state the plan. Thin notes are a practical problem long before they are anything else: an ear, nose and throat surgeon reading "difficult extraction, uneventful" has been told nothing, and will start again from the beginning with imaging you have already had.

The elements worth looking for when you read your own note:

• The date, the teeth operated on identified by number, and who performed the operation.
• The anaesthetic used, by name and amount, and who monitored it.
• What was found — the position of the tooth, whether it was sectioned, whether bone was removed.
• The complication, named, with the point at which it was recognised.
• What was done about it at the time, including anything left in place deliberately and the reason for leaving it.
• What was said to you or to your escort, and when.
• The plan, with dates on it, and any referral made, with the name of the service.

Records are retained for a minimum period set by the Board of Dentistry, measured in years from your last visit rather than from the operation, so a patient rarely needs to hurry. The operative note itself, though, is written once and at the time. A summary composed later is a different document written for a different purpose, and it is the note you want.

A written plan naming five things: the finding, the date of the next examination, what would change the plan, who else needs to be involved and by when, and what to telephone about before that date. "We will keep an eye on it" is not a plan. It contains no date, and it quietly moves the judgement about when to worry onto the patient.

Dates are the part that most often goes missing. "Come back if it is not settling" sounds accommodating and is not: it asks somebody with no clinical training to decide what settling looks like for a finding they have never encountered before, at exactly the moment when the natural human response is to give it a little longer. A review that is booked happens. A review that is merely offered often does not, and the ones that do not happen are disproportionately the ones that mattered.

"Who else" is the second half. Several complications after third molar surgery are managed by somebody other than the surgeon who performed the operation, and that is ordinary rather than remarkable.

• A persistent opening into the maxillary sinus, or a root displaced into it, involves an ear, nose and throat service or an oral and maxillofacial unit that operates on the sinus.
• A nerve injury that has not begun to change involves a surgeon who performs microneurosurgical repair — a limited group of people, and not every oral and maxillofacial surgeon is among them.
• Damage to the second molar in front — a cracked cusp, a dislodged crown or filling, a tooth that has become mobile — involves your own general dentist, because the restorative decision belongs to them.
• A fracture of the mandible involves a service that fixes jaw fractures, and it involves them the same day rather than at a review.
• A retained root fragment being deliberately observed involves radiographic review at stated intervals, agreed in advance, with a date on the first one.

Writing the plan down is what makes the handover real. A referral described in the room and never dictated is a referral that exists only in the description. It is entirely reasonable to ask, before you leave, for the name of the service you are being sent to, the date the letter goes, and how you will know it arrived — and to write those three answers on the back of your appointment card yourself.

No, and it does not need a reason attached to it. Asking to be referred, or to have another surgeon look at the same imaging, is an ordinary step in the management of a complication, and in several situations it is the correct step no matter how well the operation was performed. A surgeon who receives the request as a personal challenge has told you something useful about how the rest of this episode is likely to go.

There is a structural reason the referral matters, quite separate from anyone's feelings about it. The surgeon who performed the operation is not the right person to judge whether their own work caused a nerve injury, and is not the right person to decide alone whether that injury should be explored. Not because of dishonesty, but because the judgement needs an independent examination and the person who made the incision cannot supply one. A surgeon who says this first, unprompted, is doing the correct thing.

The same holds for capability. Microneurosurgical repair of a lingual or inferior alveolar nerve is performed by a small number of surgeons, and sending you to one reflects what a given service does, not a verdict on the person referring. Nobody sends a jaw fracture to a colleague as a confession.

If the asking is the part that feels difficult, the sentence is short: "I would like this looked at by someone who does nerve repairs. Can you make that referral, and can I have a copy of the letter and my imaging?" Nothing in that characterises anybody. It requests an action and a document, both of which are ordinary things to request.

Where the answer is that a referral is not needed yet, that answer should arrive with reasoning specific to your finding — what has been observed so far, what timescale it is being judged against, and what would change the position. "It is too early" is a sound answer for some findings and the wrong answer for others, which is precisely why it needs the reasoning attached rather than offered on its own.

Ask in writing. A short letter or email to the practice requesting a complete copy of your record — clinical notes, the operative note, consent documents, referral letters, and the imaging in its original file format. Federal health privacy rules give a patient a right of access to their own record and put a limit measured in days on how long a response may take.

Under that federal right of access, a copy is ordinarily provided within thirty days of the request, with a single thirty-day extension available if you are told about it in writing and given the reason. The record should be supplied in the form and format you asked for where it can readily be produced that way, which is the clause that matters for imaging. A reasonable, cost-based charge for copies is permitted. Florida separately requires a dentist to furnish records on a patient's written request.

Put the word "complete" in the request and then enumerate, because a request for "my records" is frequently answered with a treatment summary, which is a shorter document written afterwards for a different reader. Ask specifically for the clinical notes from every visit, the operative note, the signed consent form, the anaesthesia record, any referral letters sent or received, and all radiographs and scans.

Imaging travels as data, not as a picture

A panoramic radiograph exports as an image file. A cone-beam scan is a three-dimensional data set and should travel as its original DICOM data, on a drive or through a secure link. A flattened picture exported from a cone-beam study is one view chosen by whoever exported it, and a photograph of a monitor loses contrast in exactly the region that matters — the faint white lines of the nerve canal against a root. Any surgeon giving you a second opinion needs the data, not a picture of it.

The manner of the response carries information of its own. A complete record supplied promptly, in the format asked for, with the operative note in it, tells you the episode was documented as it happened. A record that arrives slowly, in pieces, or as a summary tells you something different — though administrative delay is common, and on its own it is a reason to ask again in writing rather than to draw a conclusion.

A patient who cannot tell ordinary recovery from a complication ends up in one of two places: alarmed by something that was always going to happen, or silent about something that mattered. The second is the one worth designing against, and the way to avoid both is to know in advance what the uneventful version looks like.

• Swelling that increases for two to three days and then begins to subside. Swelling that appears or increases after the fourth day is the pattern that is not ordinary.
• Bruising that tracks down the jaw and into the neck over several days and passes through yellow-green before it clears. It follows gravity and looks worse than it is.
• Limited mouth opening, usually worst on days two to four and easing over one to two weeks. Opening that is getting tighter while the face is swelling is a different finding.
• A socket with a greyish-white base in the second week. That is granulation tissue, not infection and not a retained fragment.
• Small sharp pieces of bone working their way out through the gum over the following weeks. Uncomfortable, ordinary, and usually removed in a minute or two if they are a nuisance.
• A sore throat, and an ache in the jaw joint or the muscles from having the mouth held open. It settles over days.
• Numbness of the lip and tongue for two to five hours after the local anaesthetic, sometimes into the evening with a long-acting agent. Numbness still present the next morning is a different thing entirely.

Dry socket sits awkwardly between the two categories and is worth naming separately. It usually appears on the third to fifth day as pain that is worsening rather than easing, often radiating to the ear, with an empty-looking socket and an unpleasant odour. It is a genuine complication, it is common, it is not an error, and it responds to a dressing, frequently within an hour of being seen. Its relevance here is that it is both the commonest reason a patient telephones and the commonest thing to be told to wait out, when being seen is quicker.

The distinction being drawn on this page is not between serious and trivial. Dry socket is not dangerous and needs an appointment. A numb lip that does not hurt at all is the most time-critical finding described here. How much something hurts is a poor guide to how urgent it is, which is why the list of same-day calls at the top of this article is written in terms of findings rather than in terms of how bad they feel.

Occasionally the sequence above does not happen. The account stays vague, the plan has no dates in it, the referral is described but never made, or the calls are not returned. At that point the useful move is not to escalate the tone. It is to move everything into writing, because a written question has to be answered by somebody, whereas a spoken one can be absorbed by a busy afternoon.

1. Put the question in one dated sentence. "On 14 May my lower right wisdom tooth was removed. Please tell me in writing what happened during the operation, what it means, and what the plan is."
2. Ask, in the same letter, for a complete copy of your record and for your imaging in its original file format.
3. Ask who you are being referred to, and on what date the letter will be sent.
4. Where the finding is time-critical and no answer has come, arrange to be seen elsewhere rather than continuing to wait. An oral and maxillofacial service, an urgent care service or a hospital department can examine you, record what is present today, and start a referral. Waiting for a reply is not a neutral act when the window is measured in weeks.
5. Keep your own dated note of symptoms — what is numb, what has changed, what you cannot do — written on the day rather than reconstructed later. It is the only contemporaneous record you control, and it is genuinely useful to whoever sees you next.
6. Florida licence records are public and searchable through the Board of Dentistry, which is also where the standards a dentist practises to are published.

Almost none of this is adversarial and none of it needs to be. Most of these situations resolve the moment a written request lands, because most of what looks like avoidance from the outside is an office that is busy and a message that never reached the person who could answer it. Writing it down is what converts somebody's intention into somebody's task.

The underlying point is simple enough to carry into any consultation, including a first one where nothing has gone wrong at all. Ask what happens if something does. A surgeon who can answer that question specifically — who gets told, how quickly, what gets written, who you would be sent to and how fast — has thought about it before it happened. That answer is available to you before you consent to anything, and it is a more informative question than most of the ones patients prepare.