Intended for healthcare professionals

Clinical Review

Trigeminal neuralgia and its management

BMJ 2007; 334 doi: (Published 25 January 2007) Cite this as: BMJ 2007;334:201
  1. Luke Bennetto, honorary clinical research fellow1,
  2. Nikunj K Patel, consultant neurosurgeon2,
  3. Geraint Fuller, consultant neurologist3
  1. 1Institute of Clinical Neurosciences, University of Bristol, Frenchay Hospital, Bristol BS16 1LE
  2. 2Department of Neurosurgery, Frenchay Hospital
  3. 3Gloucester Royal Hospital, Gloucester GL1 3NN
  1. Correspondence to: G Fuller geraint.fuller{at}

    Trigeminal neuralgia is a severe unilateral paroxysmal facial pain, often described by patients as the “the world's worst pain.”

    Summary points

    • Trigeminal neuralgia is a rare but characteristic pain syndrome

    • Most cases are still referred to as idiopathic, although many are associated with vascular compression of the trigeminal nerve

    • A minority of cases are symptomatic of multiple sclerosis or nerve compression by tumour

    • The condition is variable and patients may have just one episode

    • Most patients respond well to drugs; carbamazepine is usually the first line treatment

    • If drug treatment fails or is not tolerated, surgical treatments are available

    • Ablative surgical treatments are associated with facial sensory loss, almost no risk of severe complications or death, and a high rate of pain recurrence; microvascular decompression has a risk of severe complications or death, albeit very low, and a lower relapse rate

    How common is trigeminal neuralgia?

    The diagnosis is made by general practitioners in 27 per 100 000 people each year1 in the United Kingdom. However, previous population based studies with a strict case definition estimated the rate to be 4-13 per 100 000 people each year.2 3 Almost twice as many women are affected as men.3 The incidence gradually increases with age and is rare below 40.

    If it is uncommon why do I need to read this?

    The condition causes severe pain, which responds poorly to analgesics, but when recognised it can be treated. The pathophysiology of trigeminal neuralgia is becoming clearer. A wide range of medical and surgical treatments has been developed and introduced, usually without randomised clinical trials. As a result, uncertainty remains about how best to use the available treatments.

    Data sources and selection criteria

    We searched Medline, National Institute for Health and Clinical Excellence (NICE), and Cochrane databases using the terms “trigeminal neuralgia”, “tic douloureux”, and “facial pain”. All related articles from the NICE and Cochrane databases were obtained. We searched Medline citation lists by title and abstract for relevant publications, including randomised control trials and review articles. The citation lists of these articles were hand searched for further relevant articles.

    How is the diagnosis made?

    Trigeminal neuralgia is a clinical diagnosis. The key feature is a sudden and severe lancinating pain, which usually lasts from a few seconds to two minutes, within the trigeminal nerve distribution, typically the maxillary or mandibular branches (fig 1). The pain is often evoked by trivial stimulation of appropriately named “trigger zones.” Occasionally the pain is so severe that it prevents eating or drinking. The nerves affected are usually stereotyped for a particular patient and lie within the sensory distribution of the trigeminal nerve. Box 1 lists the diagnostic criteria for the classic form of the disease.

    Box 1 Diagnostic criteria for classic trigeminal neuralgia4

    • Paroxysmal attacks of pain lasting from a fraction of a second to two minutes that affect one or more divisions of the trigeminal nerve

    • Pain has at least one of the following characteristics

      • intense, sharp, superficial, or stabbing

      • precipitated from trigger areas or by trigger factors

    • Attacks are similar in individual patients

    • No neurological deficit is clinically evident

    • Not attributed to another disorder


    Fig 1 Distribution of trigeminal neuralgia.3 In another 1% of patients it also affects all three divisions and rarely it can be bilateral (though paroxysms are not synchronous)

    In many cases the pain does not fit these criteria exactly because of a persistent ache between paroxysms or mild sensory loss. Such disease has been labelled as “atypical”5 or “mixed” trigeminal neuralgia.6 Patients with atypical disease are more likely to have symptomatic rather than idiopathic disease, and they are often more refractory to treatment6 than those with classic trigeminal neuralgia.7 Atypical trigeminal neuralgia should not be confused with atypical facial pain (table 1).

    Table 1

     Common conditions that are usually easy to distinguish from trigeminal neuralgia

    View this table:

    What causes trigeminal neuralgia?

    Increasing evidence (box 2) suggests that 80-90% of cases that are technically still classified as idiopathic are caused by compression of the trigeminal nerve (fig 2) close to its exit from the brainstem by an aberrant loop of artery or vein.5 10

    Box 2 Evidence that vascular compression commonly causes trigeminal neuralgia510

    • An aberrant loop of artery, or less commonly vein, is found to be compressing the root entry zone of the trigeminal nerve in 80-90% of patients at surgery

    • The trigeminal nerve is demyelinated next to the compressing vessel

    • Eliminating the compression by surgery provides long term relief in most patients

    • Intraoperative assessments report immediate improvement in trigeminal conduction on decompression

    • Sensory function recovers after decompression

    • Other causes, such as compression by tumours or the demyelinating plaques of multiple sclerosis, produce similar lesions of the root entry zone of the trigeminal nerve


    Fig 2 Compression of the trigeminal nerve root by an aberrant loop of the superior cerebellar artery. Photograph provided by Hugh B Coakham

    Importantly, compression is of the root entry zone, where axons are coated with central nervous system myelin, rather than peripheral nerve myelin. Similarly, vascular compression of the facial and glossopharyngeal nerves is thought to be responsible for most cases of hemifacial spasm and glosspharyngeal neuralgia, respectively.10 w1

    Less than 10% of patients will have symptomatic disease associated with an identifiable cause other than a vascular compressive lesion—usually a benign tumour or cyst—10 11 or multiple sclerosis. About 1-5% of patients with multiple sclerosis develop trigeminal neuralgia.5

    Why is trigeminal neuralgia paroxysmal?

    The paroxysmal pain of trigeminal neuralgia is distinct from many other neuropathic pains, which often have a constant, burning quality. Why does constant compression or damage of the trigeminal nerve produce paroxysmal rather than constant pain? Although not confirmed the “ignition hypothesis”w2 offers a plausible explanation.

    • Compression of the root entry zone leads to demyelination and axonal damage10 w3 w4

    • Damaged axons become electrically hyperexcitable, exude neurotransmitters and potassium into the interstitial space, and, crucially, can repetitively fire at high frequency (resonance)w2—a rare property in healthy sensory axons unless artificially stimulated by an electric shock

    • Demyelination leaves bare axons, often subserving different sensory modalities such as light touch and pain, in contact with one another; this allows ephaptic transmission directly between them

    • A “spark”—often from light touch or cold in a trigger zone—“ignites” a self fuelling “inferno” of electrical discharge dependent upon all the above mechanisms. This rapidly burns itself out as axons become hyperpolarised, leading to the sudden and merciful cessation in pain and characteristic refractory period of a minute or so during which further pain cannot be triggered.w2

    Most facial pain is not trigeminal neuralgia

    Other causes of facial pain are much more common than trigeminal neuralgia. This can often lead to delay in diagnosis as patients see dentists and doctors who consider more common alternatives first.

    Common causes of facial pain are usually straightforward to eliminate clinically or after dental examination (table 1); rarer alternatives may evade consideration (table 2). Many alternatives affect the forehead only, which is rare for trigeminal neuralgia, so trigeminal neuralgia affecting the forehead should be diagnosed with caution. Although tearing and other autonomic features can occur in trigeminal neuralgia, these features combined with forehead pain should prompt consideration of the trigeminal autonomic cephalalgias (for example, cluster headache).4

    Table 2

    Rare conditions that may be difficult to distinguish from trigeminal neuralgia (which has been included for comparison)

    View this table:

    What investigations are needed?

    Investigations are done to:

    • Clarify the differential diagnosis; for example, by taking dental x rays

    • Investigate whether there is an identifiable cause of the disease, particularly with a view to surgical cure. This is best done using magnetic resonance imaging.

    As 5-10%11 of cases are caused by tumours, multiple sclerosis, abnormalities of the skull base, or arteriovenous malformations the threshold for magnetic resonance imaging should be low. A brain scan should be obtained in younger patients; those with atypical clinical features, including sensory loss or a dull burning pain between paroxysms; and patients who do not respond to initial medical therapy.

    With recent improvements in magnetic resonance imaging techniques (fig 3), vascular compression is being demonstrated radiologically in increasing numbers of patients with trigeminal neuralgia. Studies where radiologists are blinded to the side of the pain have produced good results in terms of predicting subsequent surgical findings and response to surgery.12 w6 Postmortem studies have found vessels in contact with the trigeminal nerve in 3-12%5 of asymptomatic patients. Thus, currently these imaging techniques should be used to explore surgical treatment options in clinically diagnosed trigeminal neuralgia, rather than in making a diagnosis.


    Fig 3 Axial constructive interference in steady state magnetic resonance imaging through the brainstem of patient with bilateral compression of trigeminal nerve (TN) roots by aberrant loops of the superior cerebellar arteries

    Which magnetic resonance imaging techniques are best?

    The magnetic resonance imaging techniques used to detect neurovascular compression are variable and continually improving. Patel et al from our centre reported 90.5% sensitivity and 100% specificity when using 1.5T magnetic resonance imaging and contrast enhanced magnetic resonance angiography to detect vascular compression of the trigeminal root.12 The imaging protocol used to obtain these figures includes a fast spin echo T2 weighted axial sequence of the whole brain with parameters of 2515/120/2 (TR/TE/excitations), a slice thickness of 6 mm with an intersection gap of 0.6 mm, a matrix size of 512_512, and a field of view of 23 cm. This is followed by thin slice, high resolution (3 mm thick with 0.3 mm intersection gap; 512_512 matrix) fast spin echo T2 weighted scans performed in the axial and coronal planes through the posterior fossa. The first procedure uses parameters of 2873/120/4 and a field of view of 20 cm, and the second uses 1496/120/4 with a field of view of 19 cm. Conventional magnetic resonance imaging is generally poor at identifying neurovascular relations. We assess neurovascular compression in our neurosurgical centre by means of a gadolinium enhanced (using intravenous 0.1 mmol/kg Magnevist) three dimensional magnetic resonance angiography fast field echo sequence obtained in the axial plane. The parameters are 17/6/2, flip angle 24_, matrix size 256_512, field of view 11 cm, slice thickness 0.6 mm, and no gap.

    What is the natural course of trigeminal neuralgia?

    Because the severity of the pain demands intervention, no studies of the natural course of the disease are available. One study collected information from linked primary and secondary care records over a 40 year period.3 They found that 29% of patients had only one episode of pain, 19% had two, 24% had three, and 28% had four to 11. Each episode lasted from one day to four years (median 49 days). After the first episode 65% of patients had a second within five years, though in 23% the gap was more than 10 years. Similar ranges of delay were seen from the second to the third event. This highlights the wide spectrum from single to frequent episodes, with each episode being of variable duration.

    Most of the case series reporting surgical and other treatments for trigeminal neuralgia are from tertiary care centres and therefore represent the most severely affected patients.

    What are the best medical treatments?

    Drug treatments for trigeminal neuralgia have been the subject of several Cochrane systematic reviews.13 14 15 These reviews bring together the small number of trials available in a condition that poses difficulties for study design (box 3). Perhaps unsurprisingly the evidence is mostly weak.

    Box 3 Problems associated with studies of trigeminal neuralgia

    Related to the condition
    • Condition rare

    • Condition variable, with spontaneous remission and relapse

    • Condition has uncertain natural history

    • Condition is a clinically defined syndrome with no gold standard diagnosis

    • The pain is so severe that placebo control studies are considered unethical

    Related to the study
    • Diagnostic criteria variably interpreted and applied

    • Lack of standardised outcome measures

    • Quality of life rarely measured

    • Open unblinded non-randomised studies predominate

    • Based on case series, usually from tertiary referral centres

    Related to treatment
    • Carbamazepine, the standard treatment, has a long half life and induces liver enzymes, making crossover studies difficult

    • Most drugs used are out of patent, which limits funding opportunities

    Related to surgery16
    • Difficult to achieve blinding or use sham operations

    • Different types of risks are being compared with different procedures (see text)

    Available evidence shows that carbamazepine is the drug of choice (table 3).15 17 Many patients develop adverse effects, however, though most can continue taking the drug. If the patient responds well, a controlled release preparation can be substituted and the dose can gradually be reduced.

    Table 3

     Drug options for trigeminal neuralgia

    View this table:

    What is less clear is what to do if a patient is intolerant or allergic to carbamazepine, or if the drug is ineffective. In the absence of clear evidence of the effectiveness of other drugs, the choice between other agents can be made on the basis of adverse effects and ease of use (table 3).

    If carbamazepine has adverse effects (table 3)

    Oxcarbazepine is a prodrug of carbamazepine that is often better tolerated; it provides a logical,18 if largely unproved,17 alternative when carbamazepine has provided pain relief but has had unacceptable adverse effects. The risk of allergic crossreactivity between carbamazepine and oxcarbazepine is about 25%, so oxcarbazepine is best avoided in carbamazepine allergy.

    Gabapentin is effective and widely used for neuropathic pain, though it lacks evidence in trigeminal neuralgia.14 Use of gabapentin therefore relies on the similarities between trigeminal neuralgia and other neuropathic pain, rather than their obvious differences. Familiarity with use in other neuropathic pain has led many clinicians to choose this as second line for trigeminal neuralgia.

    Lamotrigine and baclofen have been suggested as alternative second line agents on the basis of small studies in trigeminal neuralgia (see table 2). In practice, lamotrigine needs to be titrated over many weeks and has limited value in severe pain. Other drugs to consider are phenytoin, clonazepam, valproate, mexiletine, and topiramate.17

    If carbamazepine is ineffective

    If pain relief is incomplete with carbamazepine options include adding a second agent or switching drugs. Similar considerations regarding choice of second line agent discussed above will apply.

    Failure of medical therapy should prompt a review of the diagnosis19 (see table 1). If pain control cannot be achieved or drugs cause unacceptable adverse effects, surgical options should be considered.

    What are the surgical treatment options?

    Two types of surgical procedure are available (table 4):

    • Microvascular decompression, where the posterior fossa is explored and the compressing vessel and trigeminal nerve root are separated

    • Ablative treatment that lesion the trigeminal nerve in different ways.

    Table 4

    Surgical procedures for trigeminal neuralgia

    View this table:

    As with most surgical procedures the literature comprises mainly case series. Such series are difficult to compare (see box 3), as they vary in the populations of patients studied; diagnostic criteria, outcome measures, and follow-up methods used; and the presentation of the data.7 20 Despite this, broadly comparable high quality studies show relatively consistent responses to each procedure (fig 4).


    Fig 4 Pain after surgical treatment of classic trigeminal neuralgia. Data come from series with broadly comparable diagnostic criteria, assessment criteria, and follow-up rates,6 7 w13 w15-w25 but combining the data has limitations (see box 3).

    Predicting outcome for individual patients with confidence is difficult. Case series suggest that patients with classic trigeminal neuralgia, evidence of vascular compression, shorter duration of disease, and no previous surgery respond better to all treatment options.20 21 In such patients, microvascular decompression can be considered the “gold standard” surgical procedure, and it offers the best long term cure rates. Outcome also varies with the case load of operating surgeons.22

    Deciding on which procedure to use involves choosing between two different types of risk. All procedures have a high initial response rate, except for stereotactic radiosurgery, which usually takes maximum effect at one to two months (fig 4).23 Microvascular decompression has the best chance of long term pain relief, with very low risk of facial sensory loss and other minor complications; however, it has a small risk of death (around 0.4%24). These risks vary according to other comorbidities that alter operative risk. In contrast, ablative procedures are less effective in the long term and more likely to produce facial numbness and other minor complications; indeed their effectiveness is often greatest when this is the case.7 They have a much lower risk of death or major complication, however, and are used to a greater extent in patients with high operative risk or as a partially diagnostic procedure in atypical disease. Gamma knife stereotactic radiosurgery has recently been approved23 by the UK's National Institute for Health and Clinical Excellence for treatment of trigeminal neuralgia, but access to this treatment is limited.

    Making a decision between these options depends on the patient's perception of the two very different types of risk. Patients need to be well informed of the options, the related risks, and the likely outcomes to make such a decision. After having microvascular decompression most patients said they wish they had undergone the procedure sooner.25 Choosing between different ablative procedures, which seem to have similar effects, may be influenced by factors such as the range of adverse effects and the way in which the procedure is undertaken (see table 4).

    The patient with trigeminal neuralgia who faces these difficult choices needs to be provided with the best available information and support from the doctors and surgeons involved in their care. They may find information from patient associations and other publications useful.

    Patient's perspective

    It was 1992 when I developed pain of a “shock nature” that seemed to run around the top of my teeth on the right side. I saw the dentist who said it was nothing to do with my teeth. My doctor put me on tegretol, which made me feel like a “zombie” and I became unsteady on my feet. By now the pain was travelling in towards the brain and down towards the jaw. It was excruciating, and the only way I could “hold” the pain was to grit my teeth together very hard until the pain died away. Sleep was not interrupted, but on waking the slightest movement of the mouth would set the spasms off again, so I didn't eat or talk much. I lost a lot of weight. I went to see a neurosurgeon who advised me to have an operation to “cuff” the responsible nerve, and despite the risks I agreed as the pain was intolerable. After the operation the pain had gone. I was alive again.

    Ten years later the pain returned without warning. I saw the neurosurgeon again and this time I had a “radiofrequency lesion” of my trigeminal nerve. I am at present pain free and enjoying life as much as an 86 year old man can.

    Research questions

    • What is the natural history of trigeminal neuralgia?

    • Why do some patients with vascular compression of the trigeminal nerve get the disease and others not?

    • Many drugs have some effect, but which is the best drug to use? A pragmatic comparative study is needed to clarify these choices.

    • Can the response to drugs and to different surgical interventions be predicted on clinical or radiological grounds?

    • Which ablative procedure is most effective for the fewest adverse effects?

    • How do ablative procedures compare with microvascular decompression?

    Additional educational resources for patients

    • Zakrzewska JM. Insights—facts and stories behind trigeminal neuralgia. Gainesville, FL: Trigeminal Neuralgia Association, 2006

    • Trigeminal Neuralgia Association UK, Bromley BR2 9XS (URL:; Tel: +44 020 84629122)—Provides excellent information and patient support groups

    • Trigeminal Neuralgia Association, Gainesville, FL 32605-6402, USA (URL:; Tel: +01 800 923 3608 or +01 352 331 7009)—Provides excellent information and patient support groups

    • Brain and Spine Foundation, London SW9 6EJ (URL:; Tel: +44 020 7793 5900)—Provides patient information and helpline


    • Contributors: LB and GF reviewed the literature and wrote the article. NKP reviewed and revised the article. GF is guarantor.

    • Competing interests: None declared.


    View Abstract

    Log in

    Log in through your institution


    * For online subscription