Making a start in the patient with chronic dizziness

The first step is to define the current symptoms; here it is critical to establish exactly what the patient means by ‘dizziness’. Clarify if there is a sensation of self-motion, and if so its nature:

• Spinning, like a merry-go-round?

• Rocking, like-a-boat?

• Linear as in falling or thrusting?

• Tilting?

• Floating (more non-specific)?

Alternatively, there may be a feeling of disequilibrium rather than dizziness—that is, there is no problem with an abnormal self-motion sensation but rather patients feel unsteady on their feet or even as if they are walking on a mattress.

While the description of the current symptom is important, it is also important to know the evolution of the symptoms—that is, how the symptoms started and how they have changed? This evolution can follow one of three patterns (figure 1):

• the patient who started with one or more attacks of rotational vertigo

• the patient with a persisting history of disequilibrium

• the patient with neither of these.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Three pathways that patients may follow and thus develop chronic dizziness. Episodic vertigo (eg, migraine, Ménière's disease, benign paroxysmal positional vertigo (BPPV)) gradually leading to chronic dizziness (group 1), a single attack of vertigo (typically vestibular neuritis) with some initial recovery but many residual symptoms (group 2) and chronic, slowly progressive conditions (eg, polyneuropathies, idiopathic bilateral vestibular failure) (group 3).

The chronic dizzy patient with a history of one or more attacks of vertigo

Patients in this category report one or more previous vertigo attacks:

• Vestibular neuritis: there will be a single but disabling attack of vertigo lasting for a few days.

• Benign paroxysmal positional vertigo (BPPV): multiple brief episodes (lasting seconds) of rotational vertigo on looking up or lying down and turning over in bed. Patients can have recurrent, undiagnosed and untreated BPPV for decades.

• Migrainous vertigo: recurrent attacks of vertigo lasting from minutes to a few days (usually hours). Look for typical accompanying migrainous features such as headache, photophobia, visual auras, etc.4 There are no interictal abnormalities.

• Ménière's disease (endolymphatic hydrops): typical attacks of auditory distortion, auditory fullness, tinnitus and vertigo. There is eventual progressive unilateral audiovestibular failure.5

Why do some patients not recover fully after one or more episodes of vertigo?

We suspect this is the consequence of incomplete vestibular compensation (box 1). Vestibular compensation can be most easily observed in the nystagmic response following vestibular neuritis (figure 2 shows horizontal eye movement recordings following a permanent unilateral peripheral vestibular lesion). The top trace shows massive nystagmus. The bottom trace shows that a month later there is virtually no nystagmus. The brainstem vestibular nuclei and cerebellum play key roles in the resolution of vestibular nystagmus after an acute peripheral vestibular lesion.6,–,8 Animal studies allude to the potential inimical influences on the brainstem processes of vestibular compensation—for example, disturbances of visual input (visual cortex ablation in primates impairs compensation),9 and excessive stress, physical or emotional, by its effect on corticosteroid activity.10 11 Our own empirical observation suggests that active migraine retards vestibular compensation and we have a low threshold for using migraine prophylaxis in patients requiring vestibular rehabilitation.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Top: Horizontal eye movement recordings a few days after complete unilateral vestibular loss (labyrinthectomy). Note the prominent nystagmus (alternating slow and fast eye movements) present almost exclusively in the dark, thus illustrating the initial role played by visual input in vestibular compensation. Bottom: A month later the nystagmus is hardly detectable in the dark, indicating central non-visually mediated compensation.

Recent evidence12 suggests that the vestibular mechanisms that control eye movements and patients' symptoms are mediated by two separate but linked systems: one brainstem and the other a perceptual (presumably cortical) system. Thus in acute vertigo, eye movements and perception are strongly coupled whereas in chronic vertigo eye movements correlate poorly with symptoms.13 14 Although multiple cortical loci that mediate vestibular perception have been identified,15,–,17 the basic mechanisms underlying vestibular cortical function are poorly understood. Thus we know much less about the factors which may influence the process of vestibular compensation at the cortical level compared with the brainstem level.

Whatever the cause of failure to compensate, anxiety and/or depression are associated with long term symptoms and disability.18 It is of course plausible that depression can directly interfere with vestibular compensation (or indirectly by the impact of chronically elevated cortisol on brainstem compensation at least).10 Furthermore, as with any other chronic disease, psychological adjustment to protracted or recurrent vestibular dysfunction may play a critical role.19 In one study, comorbid anxiety and depression were found in 57% of Ménière's and 65% of vestibular migraine patients but only in 22% of those with vestibular neuritis and 15% with BPPV.20 The entanglement of chronic vestibular migraine with anxiety and depression may sometimes be impossible to disentangle.21

Whatever the relationship between dizziness and anxiety, the treatment solution requires a multidisciplinary approach.

Assessment

Firstly, a thorough balance and vertigo assessment is required although what really matters are symptoms—(dys)function and (dis)abilities per se—rather than just making an aetiological diagnosis. Many patients with vestibular pathology develop secondary problems—for example, muscular pain from increased muscle tension, particularly in the neck, stress, fatigue and even chronic anxiety. These secondary problems can affect their ability to participate in a rehabilitation programme. Inactivity, whether from bed rest, fear, anxiety or other factors, also delays and impairs complete compensation.

Functional assessment is partly incorporated in the conventional neurological examination, such as the Romberg test, gait assessment, including tandem or heel to toe gait. In addition, observing the patient through sequential actions which can be rated or timed may be useful—for example, the ‘Get up and Go’ test which is a quick screening tool for detecting balance problems in the elderly22 or the Dynamic Gait Index23 that examines how the gait is able to adapt to various task demands. These examinations allow a therapist to determine performance levels of functionally relevant tasks, have predictive value (eg, for falls risk) and are also useful for evaluating the effectiveness of therapy.

Systems assessment partly implies detecting any additional sensory–motor or musculoskeletal impairments. Somatosensory, visual and motor function can be assessed with a conventional neurological examination. The type and effectiveness of postural reactions can be examined with gentle or vigorous pushes/pulls to the trunk to elicit ankle, hip or stepping motor postural strategies.24

Symptom assessment involves identifying the primary vestibular symptom and its associated autonomic and psychological correlates. The latter may require additional reassurance or treatment if patients develop anxiety or hyperventilation.

Dizziness triggers (eg, visual stimuli such as large field stimuli (a moving train) or self-motions (travel sickness)). This is critical in guiding therapy since the principle is ‘we’ll work on whatever turns your dizziness on' (ie, desensitisation treatment), and hence the exercises for the patient will largely be based on the findings here (see ‘Management of the chronic dizzy patient’ below).

Anxiety and/or depression are associated with long term symptoms and disability

The patient with progressive disequilibrium

It is not always easy to separate patients with chronic dizziness following episodes of vestibular vertigo from those with true gait unsteadiness on the basis of history alone. It is important at the outset to sort out if the problem lies in the head or the legs, or both. The patient with subjective chronic dizziness may acknowledge that the abnormal sensation is ‘in the head’. They may say they feel as if they were drunk but, in contrast with true drunkenness or gait unsteadiness, friends or colleagues will not notice anything wrong with their balance. In contrast, the patient who describes disequilibrium due to gait unsteadiness will usually volunteer that loss of balance is noticed by themselves and observers alike.

Falling is a key objective measure in disequilibrium and falls frequency (or the absolute number of falls) should be recorded. In addition, falls can help to differentiate patients with disequilibrium (who fall) from those with chronic dizziness (who rarely fall).

The diagnosis of the patient with disequilibrium is heavily guided by the collateral history, in particular any symptoms associated with the underlying disease responsible for the gait unsteadiness. Table 1 provides a few questions useful in guiding the examination and laboratory investigations required to reach a diagnosis. As in any other area of neurology, an MRI of the brain is not enough, and it is definitely no substitute for a neurological history and examination.

Bilateral vestibular failure

Most neurologists will usually consider the diagnosis of bilateral vestibular failure when patients report oscillopsia during walking and balance problems in the dark after a course of gentamycin (not deaf) or meningitis (usually deaf). However, they often miss the idiopathic cases (which are at least as common) where the only history is of slowly progressive balance difficulty (sometimes very mild) and oscillopsia during walking, running or riding a vehicle.25 26 In some cases there are episodes of vertigo or spontaneous paroxysmal oscillopsia gradually leading into this syndrome. The common causes of bilateral vestibular failure are shown in box 2.

Bilateral vestibular failure can be diagnosed in the clinic with simple clinical tests (figure 3), the most popular being the head thrust or head impulse test (figure 4). For further explanation, see our previous article in Practical Neurology.27

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Bedside assessment of the vestibulo-ocular reflex (VOR). All tests rely on the fact that the function of the VOR is to maintain steady gaze (and hence vision) during head movements. During the doll's head–eyes manoeuvre, the patient fixates the examiner's nose while the examiner rotates the head (instruct patient to relax the neck). In patients with bilateral absence of the VOR, even slow head oscillations (1 Hz) render them unable to track the examiner's nose smoothly and small catch up saccades are observed. For the head thrust (or head impulse) test, the head movements imposed by the examiner are discrete, very brisk (ie, high acceleration) and of small amplitude. If the patient has lost the VOR, he or she will need 1–2 catch up saccades at the end of the head movement to be able to re-fixate the examiner's nose (see figure 4). During dynamic funduscopy, the clinician identifies the optic disc and instructs the patient to maintain fixation on a target across the room. Then the head is rotated from side to side. If the VOR is abnormal, again small saccades can be seen as jerky jumps of the disc (make sure not to obstruct the patient's line of sight).49 During dynamic visual acuity testing, the patient's acuity is determined binocularly with the head stationary. Then the patient's head is actively oscillated from side to side (and/or up and down) at 1–2 Hz and visual acuity determined again during the head oscillation. Some normal people may loose 1 line of visual acuity (eg, 6/6 to 6/9) but patients with bilateral vestibular failure usually loose 3 or more lines. The head movement has to be actively imposed by the examiner because the patients develop an astute compensatory tactic of briefly stopping the head movement to take a snapshot of the visual chart.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

A normal (top) and abnormal (bottom) head thrust test.50 (For a video demonstration, see www.imperial.ac.uk/medicine/balance/research.) In the bottom picture, the right horizontal semicircular canal system is abnormal (hence the catch up saccades to the left when the head turns towards the abnormal labyrinth). In patients with bilateral vestibular dysfunction, the abnormal catch up saccades may be seen with head turns in any direction.

The patient with neither vertigo nor disequilibrium

Sometimes patients refer to a sense of vague and chronic dizziness but without any history of ‘true’ vertigo or any disequilibrium. Hopefully direct questions will guide one to specific organs or systems at fault but this line of enquiry may be negative. A full clinical examination is warranted, including checking for orthostatic hypotension, particularly in the elderly taking drugs for high blood pressure. We suggest having a low threshold for lying and standing blood pressure measurements, even when the history is not typical. Good predictors for a cardiovascular diagnosis in this age group are syncope, dizziness described as light headedness, the need to sit or lie down during symptoms, pallor with symptoms, symptom precipitation by prolonged standing and coexisting vascular disease.28 But in the elderly, the history may be unreliable due to age related cognitive decline which affects not just recall of any discrete event but sometimes loss of the relevant sensory percept—for example, some older patients do not perceive dizziness during caloric irrigation despite a vigorous nystagmus response. Hence common vestibular diagnoses can be found if specifically looked for in the elderly who complain of vague disequilibrium, in particular one should always check for BPPV.

In general, patients with vague imbalance pose a diagnostic challenge and often numerous investigations are requested—blood tests to rule out general medical conditions such as anaemia, hypothyroidism or other endocrine conditions, diabetes or hypoglycaemia, brain scans and vestibular function tests. With a non-specific history with no pointers to a known disease such as Ménière's and negative investigations, many patients are diagnosed with psychogenic dizziness—rightly or wrongly. Observe, during history taking, if the patient appears to hyperventilate and enquire about frank hyperventilation and anxiety episodes, as well as features which may indicate hyperventilation such as perioral or distal paraesthesia. In our opinion, the active hyperventilation test is not reliable because hyperventilation induces dizziness and unsteadiness in everybody.29 Only when the patient recognises his or her own typical symptoms during voluntary hyperventilation is the test useful. Blood gases and the opinion of a chest physician may be justified in some cases.30

Management of the chronic dizzy patient

There are four equally important components to the management of any patient with dizziness or vertigo, whether acute, recurrent or chronic:

• the treatment of the specific vestibular condition if there is one (eg, BPPV, migraine, etc)

• short term non-specific pharmacological treatment of vertigo and associated nausea

• physical rehabilitation

• the provision of information, counselling and reassurance.

However, not all patients require action in all four domains. For instance, patients with BPPV usually require repositioning treatment and nothing else. Patients with migrainous vertigo, if they do not have interictal or chronic symptoms, may just require antimigraineous drugs but no rehabilitation. In contrast, patients with chronic dizziness typically require counselling and rehabilitation, but no drugs.

Long term vestibular suppressants and tranquilisers are counterproductive

Disease specific treatment

If an active underlying vestibular disorder is identified, such as BPPV or Ménière's disease, this needs to be treated because, with each vertigo attack, the chronic symptoms tend to increase.27 31 It is particularly important to treat migraine since migraine can both cause vertigo and be triggered by it.32 33

Non-disease specific pharmacological treatment

Vertigo or nausea can be reduced with medications but only for the acute attack or the recurrent episode. Long term vestibular suppressants and tranquilisers are inimical to the process of vestibular compensation. These drugs should only be used for truly acute vertigo and stopped as soon as vertigo begins to recede.

The doctor should briefly describe the process of vestibular compensation to the patient and explain that, for compensation to occur, the brain must feel some vertigo. This sensation will function as a warning signal and so compensation processes will be set in motion. Essentially, ‘no vertigo, no warning signal and hence no compensation’. Despite this explanation it can be difficult to wean patients off medication they may have taken for a long time and one may have to compromise and let rehabilitation begin before complete medication withdrawal.

General support

Reassurance, information and counselling are all important. Patients with long term dizziness have often wandered from clinic to clinic, from specialist to specialist, sometimes for months or years. On the basis of normal brain scans some doctor is likely to have said ‘there’s nothing wrong with you', or ‘there’s nothing that medical science can do for you' or, worse, ‘it’s all in your mind'. This is not only the wrong approach but it is usually not true. Patients may have suffered a genuine vestibular insult in the past which may not show up in conventional vestibular testing but of course failures in the vestibular compensation process and/or added psychological problems complicate the situation (box 1). Because rehabilitation works even in patients with many years of chronic dizzy symptoms,24 you want your patient to cooperate in this process. So you need to explain the principles of vestibular compensation and rehabilitation. There are many useful web pages and leaflets which are helpful (eg, the British Brain and Spine Foundation http://www.bbsf.org.uk or the Vestibular Disorders Association http://www.vestibular.org). Motivated patients, actively engaged in the rehabilitation process, fare better than patients who develop an external locus of control, feeling they have no power over their clinical outcome.19

You may also need to mention that anxiety and depression are very common in patients with dizziness but that this does not mean the symptoms are imaginary or, worse, the result of malingering. Most patients have psychological complications. Separating ‘organic from psychogenic’ or ‘primary from secondary’ is deeply engrained in our medical training but nowhere is it more difficult to do this than in the patient with chronic ‘dizziness’. The effort to distinguish organic from psychogenic may not even always be worthwhile; indeed, many patients are willing to undertake cognitive behavioural or other psychotherapies if they see this as part of a global ‘body and soul’ effort. Some patients may need antidepressants, particularly on initiating the rehabilitation process. There is no evidence that antidepressants interfere with vestibular compensation or rehabilitation, and some studies suggest a positive effect.34

Rehabilitation

The complexity of the rehabilitation programme depends on how much, if any, rehabilitation and advice your patient has already had, and how much access you have to vestibular rehabilitation services and how good these services are. For the patient who has been ill advised to stay in bed and to take tablets when dizzy, simple factual explanation, encouragement to begin activity and progressive reduction in medication may be enough. The other end of the spectrum is the patient who has already completed a course of conventional vestibular rehabilitation but has developed visual vertigo (see below). This patient may need specialised vestibular rehabilitation, including optic flow techniques.24 Most chronic dizzy patients are likely to be in between these two extremes.

A key aspect is understanding that the control of balance emerges from an interaction between many sensorimotor systems. Thus the purpose of rehabilitation is to maximise the natural ability of the CNS to compensate for lesions in the vestibular system. The neural basis for vestibular compensation is distributed throughout the nervous system such that lesions in the cerebellum, cortex, spinal cord, brainstem or sensory systems can prevent or reduce the capacity for compensation. Vestibular compensation is a plastic process that allows the CNS to restore functional symmetry after a unilateral peripheral vestibular lesion (figure 2). Patients with both unilateral and bilateral vestibular loss also compensate by a process of sensory substitution by which they learn to use non-vestibular information for balance—namely, visual and proprioceptive inputs. Although this is obviously a useful compensatory process it can also create some new problems—for example, patients unduly dependent on visual input for balance and spatial orientation may become dizzy in the presence of visual motion stimulation (‘visual vertigo’ as discussed below).

We will now outline how to organise a rehabilitation programme because many doctors may not have access to an audiologist or physiotherapist specifically trained in vestibular rehabilitation. An important comorbidity to consider prior to embarking on vestibular rehabilitation is migraine which is often aggravated by vestibular exercises. We thus advise that any active migraine should be effectively treated before starting vestibular rehabilitation.32 33

Rehabilitation treatment of balance disorders

Vestibular (‘Cawthorne-Cooksey’) exercises consist of eye, head and postural exercises of increasing complexity (box 3).35 They are intended to include those eye, head or body positions and movements which provoke vertigo. A list of exercises can be given to the patient with the instruction to go down the list identifying which ones are symptom provoking; these the patients should concentrate on.

The purpose of rehabilitation is to maximise the natural ability of the CNS to compensate for lesions in the vestibular system

The exercises should be performed for 10–15 min twice a day. ‘Pacing’ is crucial because unless the exercises are performed slowly at first they will induce unacceptable vertigo and nausea. The patient should gradually increase the pace and difficulty of the exercises as the provoked dizziness progressively abates. It is important to foster positive but realistic expectations. Patients must be told that their symptoms will at first worsen and that improvement may be uneven. Although generic exercise programmes, either provided as leaflets or group physiotherapy sessions, achieve good results, customised therapy results are superior.36 In addition, the rehabilitation professional will work on all other limitations and impairments noted during the balance therapy assessment. This will include re-training of appropriate postures such as ankle, hip and stepping strategies. Similarly, he or she will attempt to redress any surface or visual dependence (figure 5 and bottom of figure 6). Decreasing a patient's hypersensitivity to visual motion cues can be done by balancing during exposure to optokinetic stimuli, virtual reality sets or simply watching complex computer games on a big screen.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5

Examples of the type of visual stimulation employed in the desensitisation of patients with chronic dizziness and visual vertigo. Left: A rotating disc. Centre: Gait exercises during planetarium projection. Right: Optokinetic stimuli delivered by the Eyetrek or virtual reality systems; the patient stands on foam to decrease the accuracy of the lower limb proprioceptive input. Modified from Pavlou et al.24

• Download figure
• Open in new tab
• Download powerpoint

Figure 6

Different head and eye movements to be included as part of a vestibular rehabilitation programme. During ‘gaze stability’ exercises, the patient is stimulating the vestibulo-ocular reflex (VOR)—that is, keeping the eyes steady on a fixated object as the head moves progressively faster. During ‘gaze transfer’, the patient practises the normal head and eye synergy for transferring gaze from one object to another. During ‘VOR suppression’, the patient tries to fixate or read from an object that she carries around (like reading a newspaper on a bus).

All exercises or situations that are symptom provoking have to be made progressively more complex as the patient is desensitised. For instance, if a patient is dizzy when turning the head from side to side, this exercise will be initially practised just while seated, then standing, walking, walking on a mattress or standing on one foot at a time, or with eyes closed. The principle is adding complexity to cover as many circumstances as possible so that when the patient has to move their head in an everyday life situation all possibilities have been practised beforehand.

Not all patients need so much work up. In many cooperative patients an explanation of the principles of vestibular compensation and how this is achieved through graded activity is enough. The next step up would be a list of the exercises in box 3, with an explanation to identify the symptom provoking ones, a twice a day activity schedule and an indication to progressively intensify the pace. However, better to encourage an audiologist or physiotherapist in your hospital to become interested in balance rehabilitation.

Last but not least, some specific syndromes of chronic dizziness to be aware of

Visual vertigo

These are the patients with chronic dizziness whose symptoms are worse in certain ‘visually busy’ surroundings. The syndrome has been given different names such as visual vertigo,37 38 visuo-vestibular mismatch39 and space and motion discomfort.40 Frequent triggers are walking between shelves in supermarket aisles, and viewing movement of large visual objects such as clouds, windswept trees, rivers flowing, disco lights, crowds, traffic, curtains or films with car chase scenes. Repetitive visual patterns like the stacks of cans on supermarket shelves, ironing striped shirts or walking past a repetitive patterned fence seem to be relevant. Some patients also mention that moving their eyes, reading and flickering or fluorescent light can make them feel dizzy.

Some patients with vestibular lesions become overly sensitive to visual stimuli. However, an increased sensitivity to visual signals is a normal response to vestibular disease (‘sensory substitution’) but this process may be exaggerated in patients with visual vertigo.39 41,–,43 Why some patients develop visual dependence and visual vertigo is not known but migraine appears to be a predisposing factor.44 We do know, however, that visual vertigo can be improved by rehabilitation which includes customised vestibular therapy with additional visual motion desensitisation24 (figure 3).

Motorist disorientation syndrome

Driving, particularly on motorways, can be uncomfortable for patients with vestibular disorders and chronic dizziness. Occasionally, they report a sensation that their car is tilting or veering to one side which is so compelling that they consult their mechanic or change their car before seeing their doctor. This syndrome45 also seems to be, at least partly, visually determined because patients describe problems in visually deprived areas (top of a hill) and visually challenging conditions (simultaneously overtaking and being overtaken by a car). Indeed the coexistence of visual vertigo and motorist disorientation syndrome in the same patient is not rare.

All exercises or situations that are symptom provoking have to be made progressively more complex as the patient is desensitised

Going round a bend, as when driving in a roundabout, can also disorient patients but here there may be a predominantly vestibular mediated component. It must be kept in mind that driving through a curve is different to just turning round while walking. The radius of curvature is large while driving so the conditions are equivalent to being centrifuged and that is why you feel pushed sideways against the car door. In this case the unusual stimulation (sideways linear acceleration) acting on a damaged otolith system may be responsible for the symptoms. Psychological components, usually in the form of panic and avoidance behaviour, can contribute. In fact, in patients with no vestibular history or findings, a psychological disorder may be the only mechanism responsible.46 In these patients, however, there is usually more ‘panic’ and less ‘veering and tilting’ of the car.

In those with motorist disorientation syndrome, a previous history of vestibular disease and no panic component, treatment is based on vestibular rehabilitation with the addition of visuo-vestibular conflict and optic flow stimuli. In patients with mostly panic symptoms but no vestibular disease, the treatment is predominantly psychiatric, often combining medication with cognitive behavioural therapy.47 Many patients fall between these two extremes and are accordingly managed with a combination of treatments.

Psychological presentations, phobic postural vertigo, panic attacks

The term ‘phobic postural vertigo’ describes patients with transient sensations of unsteadiness in whom postural balance is entirely normal on extensive clinical examination.48 Patients may deny psychological disturbance but excessive anxiety or an obsessive–compulsive personality may be apparent. Panic attacks, in which patients describe autonomic symptoms, catastrophic thoughts and avoidance behaviour can also turn up in dizzy patient clinics mimicking vestibular initiated chronic dizziness.

A complicating factor is that in approximately 30% of patients with these ‘psychogenic’ syndromes a vestibular onset such as BPPV, migraine or vestibular neuritis can be elicited.48 Another complication is that some symptoms of the visual vertigo and motorist disorientation syndromes overlap with phobic postural vertigo and panic. It may well be that some of these presentations are just normal modes of reaction that humans have to different disorders. A vestibular disorder or a psychological problem may lead to a similar final clinical result.

In terms of treatment, each patient needs to be considered on his or her own merits. Two examples at opposite ends of a spectrum would be the patient who has no pre-existing psychiatric history but just vestibular neuritis leading to chronic dizziness with visual vertigo and another patient with general anxiety and panic attacks in the supermarket in whom, despite direct questioning and specialised balance examination, no vestibular features are detected. There is little doubt that the former patient should receive vestibular treatment, usually rehabilitation including visual motion desensitisation, and that the latter should be in the hands of a psychiatrist or cognitive behavioural therapist. The many patients in between these two examples often require combined treatment. Ideally, vestibular therapists, whatever their background (audiologists or physiotherapists) should have counselling skills so they can deal with the very common anxiety, depressive or phobic components that so many patients have (table 2).