Trigeminal Neuralgia Treatment
Trigeminal Neuralgia Treatment
Treatment here has an unusual shape: the first step is a medicine borrowed from epilepsy, and the steps after it are procedures that deliberately trade something for the pain. What is worth understanding early is that none of these options is a choice between success and failure — they are choices about what a person is willing to live with, and they are best made while the pain is controlled rather than in the middle of a cluster. This page describes what is given first, what is given when that stops working, what the procedures actually involve, and what a written plan should contain.
What is given first
A medicine from the antiseizure family, and this surprises people who have been told their pain is not a seizure. The reason is not that the two conditions are related; it is that these medicines quiet an irritable nerve, and they are the best-studied medicines that do so for this pain.
Carbamazepine is the one with the longest record, and oxcarbazepine is used alongside it, often because it is tolerated better or interacts with fewer other medicines. Both are started at a low dose and raised slowly. That is not caution for its own sake: the dose that controls the pain is found by climbing towards it, and climbing quickly produces the side effects that make people abandon a medicine that would have worked.
Two things are worth insisting on at the first prescription. The first is monitoring: these medicines require blood tests before starting and periodically afterwards, and the sodium level is the one most often checked because a low sodium can develop without any symptom until it matters. The second is a genetic test before carbamazepine in people of Asian ancestry, specifically for the HLA-B*15:02 allele, which is associated with a severe skin reaction. This is a recommendation that applies before the first tablet rather than after a rash appears.
Where neither is tolerated, or where the pain does not respond, other medicines are used — lamotrigine and baclofen most often, and gabapentin or pregabalin where the picture is less clearly paroxysmal. What should be said about these honestly is that the evidence behind them is thinner, and that the benefit of adding a second medicine has to be weighed against the sedation of taking two.
What the first month should look like is worth stating: attacks becoming shorter and less frequent rather than vanishing, the dose being adjusted rather than fixed, and side effects that are commonest after each increase and often settle within days. A medicine abandoned in the first fortnight for drowsiness that would have settled is the commonest avoidable failure of the first step.
When the medicine stops working
This is the question that most people arrive with, and it is worth separating the three things it can mean, because they lead to three different answers.
The first is that the pain has got louder, not that the medicine has got weaker. Clusters intensify, and a dose that held in the first week of a cluster may not hold in the third. Raising the dose is reasonable here, up to the point where side effects outweigh benefit — and that point is reached by most people eventually rather than by nobody.
The second is that the condition has changed shape. Where a constant background ache has appeared alongside the paroxysms, more of the same medicine rarely helps: it treats one pain and not the other, and escalating it for the component it does not touch is the commonest way someone ends up on a high dose with sedation and no additional relief. Naming the second pain is usually more useful than raising the dose.
The third is that something has become a trigger that was not one before — dental work, a local anaesthetic injection, an infection, or simply a cluster arriving while a dose was missed for a few days. Here the answer is often to remove the provocation rather than to change the prescription.
Two rules hold whatever the reason. Never stop abruptly: this medicine is stopped by reducing it, and stopping it suddenly is a reliable way to bring a controlled pain back at full force. And do not escalate by telephone indefinitely — a pain that has stopped responding to a dose that once controlled it is a reason to be reviewed, because it is the point at which the conversation turns to a procedure rather than a further increase.
The procedures, and what each asks of you
Three families, and the difference between them is not which works best but what each one trades. It is worth reading this section while the pain is controlled.
| Option | What it is | What it asks of you |
|---|---|---|
| Microvascular decompression | An operation behind the ear under general anaesthesia. The surgeon looks for the blood vessel in contact with the nerve at the root entry zone and places a small pad between them, leaving the nerve intact | A general anaesthetic and an inpatient stay, and the risks that accompany any operation behind the ear, including hearing loss on that side and the general risks of anaesthesia. It is the one option that does not deliberately numb the face |
| Percutaneous lesioning | Radiofrequency rhizotomy, glycerol rhizolysis or balloon compression, performed through a needle passed to the nerve through the cheek under sedation or short anaesthesia | A short stay, and the intended trade of the pain for a degree of facial numbness on that side — which is the effect, not a complication, and which some people find harder to live with than they expected |
| Stereotactic radiosurgery | Gamma Knife or CyberKnife, delivered as an outpatient with no incision and no anaesthetic, aiming a focused dose along the nerve | A delay: relief is not immediate, and is typically described in weeks to months rather than days. It is the option chosen most often by people unfit for an operation or unwilling to have one |
What should be said about all three is that none of them is irreversible in the sense of being a dead end, and none of them can be promised to end the condition. Pain can return after any of them, at intervals that vary, and the question of what is done then is a live one rather than a failure.
What decides between them in practice is a combination of what the scan showed, how fit a person is for an operation, and what they are willing to trade. Where imaging has shown a vessel in contact with the nerve and the person is fit, decompression is discussed earlier rather than last, because it is the only option that addresses the contact without damaging the nerve. Where the scan shows nothing, or an operation carries too much risk, the lesioning and radiosurgery options are the ones discussed.
One thing worth asking before consenting to any of them: what the team's own experience of the procedure is, and what they would expect the recovery to look like week by week. That answer is more useful than any figure quoted in a leaflet, and a team that does the procedure regularly will have one.
Living with the medicine
The medicines that control this pain are not difficult to take, but they are not inert either, and the practical detail is what keeps people on them.
Timing. Drowsiness and unsteadiness are commonest after an increase and in the first hours after a dose. Where the schedule allows it, taking the larger part of the daily dose at night is often the single change that lets someone keep working. Modified-release preparations exist for this reason.
Monitoring. Blood tests before starting and periodically afterwards, with attention to the sodium level, the blood count and liver function. What to watch for between tests is worth writing down: unusual bruising or bleeding, a sore throat or fever that does not settle, yellowing of the eyes or skin, a rash, or a headache with vomiting and confusion. Several of these mean stop and be seen rather than wait for the next appointment.
Interactions. These medicines interact with others, including some antibiotics, some antidepressants, some blood thinners and some heart medicines, and the interaction can go in either direction. What to do is simple and worth doing every time: carry a written list of everything being taken, and have it checked before anything new is added — including anything bought over the counter or prescribed by a dentist.
Alcohol, driving and machinery. Alcohol adds to the drowsiness and, for some people, to the unsteadiness. Driving while newly drowsy or newly on a higher dose is a question to ask directly rather than to judge alone. And pregnancy planning is a conversation to have before rather than during: these medicines carry considerations in pregnancy, and the time to plan is before a remission tempts a change.
Missed doses and illness. A missed dose is taken when remembered unless it is close to the next one, and never doubled. During an illness with vomiting, or before an operation where nothing can be taken by mouth, the plan has to be agreed in advance rather than improvised — because an abrupt gap in this medicine is not a neutral event.
When the pain is not this condition
A significant part of what is treated as trigeminal neuralgia turns out, on careful assessment, not to be it — and treating the wrong pain with the medicine for this one is the second commonest way people end up on a high dose with no benefit.
Where the pain is constant rather than paroxysmal, the medicines used for persistent neuropathic pain are the relevant ones, and the evidence for the antiseizure medicine used in this condition is weaker for that picture. Where the pain is dental or from the jaw joint, the treatment is dental or dental-surgical, and no amount of carbamazepine substitutes for it. Where the pain is from shingles, the treatment is the neuropathic pain pathway and the prevention is vaccination.
Where the pain is secondary to something structural — a tumour, a cyst, a vascular malformation, or demyelination — the treatment is of that cause, and the facial pain is treated alongside it rather than instead of it. This is the group in which a normal scan reported without the trigeminal protocol causes real delay, and it is the group most often identified in a records review.
And where the pain is glossopharyngeal neuralgia, provoked by swallowing and felt in the throat, the ear or the angle of the jaw, it is a different nerve: the medicines are similar, the procedures are not, and the workup of the throat is not the workup of the face.
What connects all five is one sentence: the pattern has to be named before the treatment is chosen. Where a scan has not been aimed along the nerve, where nobody has asked how long an attack lasts, or where teeth have been removed for a pain that continued unchanged, the next step is naming rather than escalating.
What a treatment plan should leave in writing
Six things, and a person is entitled to ask for each of them at the point the first prescription is written rather than two years later.
First, the medicine, the dose and the target, stated as numbers rather than as a name. Second, what monitoring it requires and how often, including which blood test and which symptom should stop it. Third, what has been excluded before this was treated, naming dental and structural causes specifically, and stating whether the imaging included the trigeminal protocol. Fourth, what the dose ceiling is — the point at which the answer stops being a higher dose. Fifth, which procedure would be discussed, and on what trigger: not a promise to offer one, but a statement of what would open the conversation. Sixth, how a remission is handled — whether the medicine is reduced, how quickly, and what would restart it.
That sixth item is the one most often missing, and it is the one that decides what the next few years look like — because in this condition the pain-free intervals arrive, and what happens in them is planned rather than improvised. A written plan for a remission is worth more than a plan for an attack.