|Home | About | Journals | Submit | Contact Us | Français|
Vertigo and dizziness are among the most common complaints with a lifetime prevalence of about 30%. The various forms of vestibular disorders can be treated with pharmacological therapy, physical therapy, psychotherapeutic measures or, rarely, surgery. In this review, the current pharmacological treatment options for peripheral and central vestibular, cerebellar and ocular motor disorders will be described. They are as follows for peripheral vestibular disorders. In vestibular neuritis recovery of the peripheral vestibular function can be improved by treatment with oral corticosteroids. In Menière's disease a recent study showed long-term high-dose treatment with betahistine has a significant effect on the frequency of the attacks. The use of aminopyridines introduced a new therapeutic principle in the treatment of downbeat and upbeat nystagmus and episodic ataxia type 2 (EA 2). These potassium channel blockers presumably increase the activity and excitability of cerebellar Purkinje cells, thereby augmenting the inhibitory influence of these cells on vestibular and cerebellar nuclei. A few studies showed that baclofen improves periodic alternating nystagmus, and gabapentin and memantine, pendular nystagmus. However, many other eye movement disorders such as ocular flutter opsoclonus, central positioning, or see-saw nystagmus are still difficult to treat. Although progress has been made in the treatment of vestibular neuritis, downbeat and upbeat nystagmus, as well as EA 2, state-of-the-art trials must still be performed on many vestibular and ocular motor disorders, namely Menière's disease, bilateral vestibular failure, vestibular paroxysmia, vestibular migraine, and many forms of central eye movement disorders.
In the first part of this article, the treatment of common peripheral and central vestibular disorders are described [Strupp et al. 2007a; Brandt et al. 2005]. The second part focuses on the clinically most relevant forms of nystagmus, in particular downbeat and upbeat nystagmus, along with their pathophysiology and topographic diagnosis, and current therapy [Leigh and Zee, 2006; Strupp and Brandt, 2006].
The terms vertigo and dizziness cover a number of multisensory and sensorimotor syndromes of various aetiologies and pathogeneses, which can be elucidated only with an interdisciplinary approach. After headache, it is one of the most frequent presenting symptoms, not only in neurology. The lifetime prevalence is almost 30% [Neuhauser, 2007]. A survey of over 30,000 persons showed that the prevalence of vertigo as a function of age lies around 17% and rises up to 39% in those over 80 years of age [Davis and Moorjani, 2003].
The prerequisite of every treatment of vertigo and dizziness is a correct diagnosis, which can be simply made in most patients on the basis of the patient history and the clinical examination even without any laboratory examinations. Depending on the aetiology, the various forms of vestibular disorders can be treated with pharmacological therapy, physical therapy, psychotherapeutic measures or, rarely, surgery. Before beginning any treatment, the patient should be told that the prognosis is generally good for two reasons: (a) vertigo often takes a favourable natural course (e.g. the peripheral vestibular function improves or central vestibular compensation of the vestibular tone imbalance takes place) and (b) most forms can be successfully treated (mainly with drugs or physiotherapy). Several agents are now available for the specific treatment of certain forms of vestibular and ocular motor disorders.
Three typical forms of peripheral vestibular disorders can be differentiated by their characteristic signs and symptoms: (1) bilateral peripheral loss of vestibular function (bilateral vestibulopathy), characterised by oscillopsia during head movements and instability of gait and posture; (2) acute/subacute unilateral failure of vestibular function (most often caused by vestibular neuritis), characterised by a rotatory vertigo, oscillopsia, and a tendency to fall toward the affected ear; and (3) paroxysmal, inadequate stimulation or inhibition of the peripheral vestibular system, characterised by attacks of vertigo and oscillopsia. This occurs in benign paroxysmal positioning vertigo, but also in Menière's disease or vestibular paroxysmia.
Benign paroxysmal positioning vertigo (BPPV) is the most common cause of vertigo, not only in the elderly. It is characterised by brief attacks of rotatory vertigo and simultaneous positioning rotatory-vertical nystagmus toward the undermost ear elicited by extending the head or positioning the head or body toward the affected ear. It is called benign because it often resolves spontaneously within weeks to months; in some cases, however, it can last for years. The canalolithiasis hypothesis of freely floating ‘heavy otoconia’ is compatible with all features of BPPV: latency, duration, course of attacks, direction of nystagmus, reversal of nystagmus, fatigability and, most important, the efficacy of positioning ‘liberatory manoeuvres’ of the head [Brandt and Steddin, 1993]. Brandt and Daroff  were the first to devise an effective exercise programme which required the simple performance of a series of head positioning movements. Semont et al.  recommended that the patient's position should be changed from the inducing position by a tilt of 180 degrees to the opposite side. Epley  proposed another variation that involved turning the patient's trunk and head into a head-hanging position. It can also be explained by the mechanism of canalolithiasis. The liberatory maneuvers according to Semont et al.  or Epley  are successful in more than 95% of the patients if performed correctly [Strupp et al. 2007a; Brandt et al. 2005].
Vestibular neuritis is the third most common cause of peripheral vestibular vertigo (the first and the second are BPPV and Meniére's disease). It accounts for 7% of the patients who present at outpatient clinics specialising in the treatment of dizziness [Brandt et al. 2005] and has an incidence of 3.5 per 100,000 population [Sekitani et al. 1993]. The key signs and symptoms of vestibular neuritis are the acute onset of sustained rotatory vertigo, horizontal spontaneous nystagmus toward the unaffected ear with a rotational component, postural imbalance with Romberg's sign, that is, falls with the eyes closed toward the affected ear, and nausea. Caloric testing invariably shows ipsilateral hypo-responsiveness or nonresponsiveness. In the past, either inflammation of the vestibular nerve or labyrinthine ischaemia was proposed to cause vestibular neuritis. Currently a viral cause is favoured. The evidence, however, remains circumstantial. Herpes simplex virus type 1 (HSV-1) DNA has been detected on autopsy with the use of polymerase chain reaction in about two-thirds of human vestibular ganglia [Theil et al. 2002, 2000; Arbusow et al. 2000, 1999; Schulz et al. 1998]. This, as well as the expression of CD8-positive T-lymphocytes, cytokines and chemokines, indicates that the vestibular ganglia are latently infected with HSV-1 [Theil et al. 2003].
A prospective randomised, double-blind, two-by-two factorial trial was performed, in which patients with acute vestibular neuritis were randomly assigned to treatment with placebo, methylprednisolone (100 mg/day, doses tapered by 20mg every third day), valacyclovir (valociclovir, 1g t.i.d. for 7 days), or methylprednisolone plus valacyclovir. Vestibular function was determined by caloric irrigation, with the use of the vestibular paresis formula (to measure the extent of unilateral caloric paresis), within 3 days after the onset of symptoms and 12 months afterwards. A total of 141 patients underwent randomisation. The mean improvement in peripheral vestibular function at 12-month follow-up was 39.6 percentage points in the placebo group, 62.4 percentage points in the methylprednisolone group, 36.0 percentage points in the valacyclovir group, and 59.2 percentage points in the methylprednisolone plus valacyclovir group (Figure 1). Analysis of variance showed that methylprednisolone had a significant effect, but valacyclovir did not. Therefore, this study showed that methylprednisolone alone significantly improves the recovery of peripheral vestibular function in patients with vestibular neuritis, whereas valacyclovir is not required [Strupp et al. 2004b]. Symptom outcome at 12 months was not addressed for two reasons. First, animal experiments show that steroids improve central vestibular compensation. Thus, parameters other than vestibular paresis, such as postural imbalance or ‘vertigo and dizziness’, would not help differentiate between the effects of steroids on the recovery of peripheral vestibular function and on central vestibular compensation. Second, there are no validated scales for measuring vertigo and dizziness. All in all, this cheap and well-tolerated therapy can be recommended as the pharmaceutical treatment of choice for vestibular neuritis.
Menière's disease is clinically characterised by recurrent spontaneous attacks of vertigo, fluctuating hearing loss, tinnitus and aural fullness. Its incidence varies between 7.5 per 100,000 to 160 per 100,000 persons [Minor et al. 2004]. Endolymph hydrops is assumed to be the pathological basis of Menière's disease, either due to a too high production or a too low absorption of the endolymph. The increased endolymphatic pressure causes periodic rupturing or leakage (by the opening of nonselective, stretch-activated ion channels [Yeh et al. 1998] of the membrane separating the endolymph from the perilymph space. Therefore, pathophysiologically it makes sense to reduce the production and increase the absorption of endolymph. The clinical aims of treatment of Menière's disease are to stop vertigo, reduce or abolish tinnitus, and preserve and even reverse hearing loss. Most studies focus on the most distressing symptom of Menière's disease: recurrent attacks of vertigo.
There is a plethora of treatment strategies for Menière's disease. Destructive procedures involving the lateral semicircular canal and vestibule have been proposed since 1904. The first endo-lymphatic sac decompression was performed in 1926. This method is still used in some settings despite its evident ineffectiveness. Restricting salt and fluid intake and diuretics were first proposed in 1934. Salt restriction and diuretics are still recommended, although in one double-blind study diuretics did not have any effect [van-Deelen and Huizing, 1986]. Vestibulotoxic drugs have been in use since 1948; local intra-tympanic delivery has been performed since 1956 (for references see Smith et al. 2005). It is remarkable that despite the high incidence of Menière's disease and the large number of studies published on its treatment over the last few decades, there are still only very few state-of-the-art prospective, placebo-controlled, double-blind trials. Moreover, there are significant differences in the treatment regimen of Menière's disease between Europe and the US. In the US, low-salt diet, diuretics, and intratympanic injection of gentamicin and corticosteroids are preferred. In Europe betahistine is more often used, in the US rarely; it is remarkable that in a recent review on Menière's disease by two US authors the word betahistine does not even appear [Sajjadi and Paparella, 2008]. A national survey among UK otolaryngologists on the treatment of Menière's disease revealed that 94% used betahistine, 63% diuretics, 71% salt restriction, 52% sac decompression, and approximately 50% insertion of a grommet [Smith et al. 2005]. Local gentamicin instillation has become more and more popular since its introduction in the UK 10 years ago: approximately two-thirds of the otolaryngologists use this method.
Several studies have been published on intratym-panic gentamicin application for the treatment of Menière's disease. Initially multiple intratympa-nic injections of gentamicin were given until patients developed vestibular hypofunction. This led to a good control of attacks of vertigo, which, however, was accompanied by a high rate of sensorineural hearing loss (approximately 50%). Especially after the demonstration of a delayed onset of ototoxic effects [Magnusson et al. 1991], the regimen was changed in two ways: (1) single instillations at fixed interims of several days or weeks, or (2) single-shot injections and follow-up. Following the latter regimen, a prospective uncontrolled study with a follow-up time of 2–4 years on 57 patients showed that in 95% vertigo attacks could be controlled [Lange et al. 2004]. Fifty-three per cent of these patients needed only one injection of 12mg gentamicin, 32% two or three injections. A recent meta-ana-lysis on 15 trials with 627 patients on gentamicin injection showed that complete vertigo control was achieved in about 75% of patients and complete or substantial control in about 93%. The success rate was not affected by the gentamicin treatment regimen; that is, fixed versus titration [Cohen-Kerem et al. 2004]. Hearing level and word recognition were not adversely affected, regardless of treatment regimen. The authors, however, pointed out that the level of evidence reflected in the relevant articles is insufficient, especially because of relatively poor study designs – none of the trials was double-blind or had a blinded prospective control. Meanwhile there is good evidence that the beneficial effect of gentamicin is due to its damage to the hair cells. A complete ablation of function, however, does not seem necessary in order to control vertigo [Carey et al. 2002].
In a retrospective chart review, Barrs  evaluated the effects of intratympanic injections of dexamethasone in 34 patients. After a single course of weekly injections of 10mg/ml dexamethasone for 1 month, only 24% of the patients reported vertigo control. Another 24% responded to the repeat series of injections. All in all, approximately one-half of the patients with Menière's disease achieved control of vertigo with one or more courses of intratympanic injections of corticosteroids. The safety of intratympanic dexamethasone injections was evaluated by transient evoked otoacoustic emission. No change was found in 26 patients after five injections of 4mg dexamethasone [Yilmaz et al. 2005].
In Europe betahistine is more often used, mainly on the basis of a study by Meier in 1985 and more recent meta-analyses [James and Thorp, 2004; Claes and Van-de-Heyning, 1997]. Betahistine is an H1 agonist and H3 antagonist. It improves the microcirculation by acting on the precapillary sphincters of the stria vascularis [Dziadziola et al. 1999]. There is evidence that it reduces the production and increases the absorption of endolymph. In an open trial on 112 patients with Menière's disease it was shown that a higher dosage of betahistine-dihydrochloride (48 mg t.i.d.) and a long-term treatment (12 months) seems to be more effective than a low dosage (16–24mg t.i.d.) and short-term treatment (Figure 2) [Strupp et al. 2008]. These data are the basis for a recently begun prospective, randomised, double-blind dose-finding study comparing placebo with 16mg and 48mg t.i.d. betahistine-dihydrochloride. Finally, it must, however, be pointed out that up to now no state-of-the-art studies have been conducted in this field despite the large number of trials.
Vestibular paroxysmia is characterised by short attacks of rotatory or to-and-fro vertigo. These attacks last for seconds to minutes and may occur up to 30 times a day. Like in trigeminal neuralgia, hemifacial spasm or superior oblique myokymia, it is assumed that a neurovascular cross-compression of the eighth cranial nerve is the cause of vestibular paroxysmia [Brandt and Dieterich, 1994]. Therapy with low doses of carbamazepine (200–600mg per day) or oxcarbazepine has an early therapeutic onset, and thus provides a positive response useful in the diagnostics of the disease. This was demonstrated in an open trial. [Hufner et al. 2008]. In case of intolerance, gabapentin, valproic acid, or phenytoin is a possible alternative. Currently, a prospective, placebo-controlled, double-blind trial is underway.
In this review the pharmacological treatment of the most important central vestibular, ocular motor, and cerebellar disorders will be presented, namely vestibular migraine, episodic ataxia type 2, and downbeat, upbeat and other forms of nystagmus.
Vestibular migraine is the most common cause of central recurrent attacks of vertigo. Characteristic features include recurrent attacks of various combinations of vertigo, ataxia of stance and gait, visual disorders, and other brainstem symptoms accompanied or followed by occipitally located head pressure, pain, nausea or vomiting [Neuhauser and Lempert, 2004; Furman et al. 2003; Brandt and Dieterich, 1994]. There is, however, an ongoing debate as to whether it is a clinical entity. Treatment is the same as for migraine with aura; that is, for prophylactic therapy the use of betablockers (metoprolol or propranol), valproic acid or topiramate for at least 3–6 months. A few treatment studies on vestibular migraine have been performed. Tricyclic antidepressants in combination with diet showed a good response in a trial on 81 patients [Reploeg and Goebel 2002]. For zolmitriptan the response rate in acute attacks was 38% versus 22% in a study on 19 patients [Neuhauser et al. 2003]. Another open trial on 10 patients demonstrated that lamotrigine (100 mg per day as a single dose) had a significant effect on the occurrence of headache and a more marked effect on vertigo [Bisdorff, 2004]. Again, a placebo-controlled multicentre trial is warranted. So far, only the standard treatment of migraine with aura can be recommended for vestibular migraine.
Episodic ataxia type 2 (EA2) is clinically characterised by recurrent attacks of ataxia, provoked by stress or exercise, which last for several hours to days [Strupp et al. 2007b; Jen et al. 2004; Griggs and Nutt, 1995]. Associated findings during the nonattack interval include central ocular motor and vestibular dysfunction, mainly downbeat nystagmus. Patients with EA2 can often be successfully treated with acetazolamide [Griggs et al. 1978]. Genetically EA2 is an autosomal dominant hereditary disorder caused by mutations of the calcium channel gene CACNA1A [Ophoff et al. 1996], which encodes the CaV21 subunit of the PQ-calcium channel expressed mainly in the Purkinje cells. On the basis of the functional changes of the PQ-channel mutation, which leads to a reduced calcium current, it can be assumed that the inhibitory effect of Purkinje cells is reduced in EA2 [Kullmann, 2002]. This causes the disinhibition of the deep cerebellar nuclei and thus ataxia and downbeat nystagmus. Since aminopyridines (as potassium channel blockers) were shown to improve downbeat nystagmus (see below) most likely by increasing the inhibitory influence of the Purkinje cells (this hypothesis was supported by animal experiments [Etzion and Grossman, 2001]), we evaluated its effects on the occurrence of attacks with EA2 [Strupp et al. 2004a]. In three patients with EA2 (two with proven mutations of the CACNA1A gene) attacks could be prevented with the potassium channel blocker 4-aminopyridine (5 mg t.i.d.). Attacks recurred after treatment was stopped; subsequent treatment alleviated the symptoms (mean follow-up time greater than 12 months). These effects might be due to an improvement of the impaired functioning of Purkinje cells. It must be pointed out that these three patients did not respond to the standard treatment with acetylzolamide any more. Again on the basis of this open trial a placebo-controlled study is currently in progress. The clinical findings were supported by an animal study on the calcium channel mutant tottering mouse. Aminopyridines blocked the attacks characteristic of the tottering mouse via cerebellar potassium channels by increasing the threshold for attack initiation without mitigating the character of the attack [Weisz et al. 2005].
Nystagmus can be defined as periodic, most often involuntary eye movements that normally consist of a slow (causative or pathological) phase and a quick eye phase, which brings the eye back to the initial position. Nystagmus is quite common: its prevalence lies around 0.1% [Stayte et al. 1993]. The most common forms of acquired nystagmus are downbeat and upbeat nystagmus. Both can be treated nowadays with aminopyridines in their capacity as potassium channel blockers. More rare forms are congenital nystagmus, acquired fixation pendular nystagmus, and period alternating nystagmus. If they cause symptoms, mainly involuntary movement of the visual surrounding (oscillopsia), treatment with mematine or gabapentin should be tried. To improve treatment of the different forms of nystagmus, further randomised controlled trials will be necessary to test different agents on the basis of our current knowledge of the pathophysiology of nystagmus.
In the following the most common and clinically most relevant forms of nystagmus and their pathophysiology as well as current therapy will be described. For the frequency of the different forms, see Table 1. In Table 2 all the features are summarised. The treatment of nystagmus is based on four principles: medical treatment, optical devices, surgery to weaken certain eye muscles, and somatosensory or auditory stimuli. Medical treatment is the most relevant and successful means of treatment (for reviews see Leigh and Zee, 2006; Strupp and Brandt, 2006; Straube et al. 2004; Leigh and Tomsak, 2003).
Downbeat nystagmus (DBN) is the most common form of acquired persisting fixation nystagmus (Table 1) [Wagner et al. 2008]. It is characterised by slow upward drifts and fast downward phases. Slow-phase velocity increases on lateral and downward gaze and convergence, although there may be atypical presentations with enhancement of the DBN on upward gaze or suppression on convergence [Leigh and Zee, 2006; Pierrot-Deseilligny and Milea, 2005; Baloh and Spooner, 1981] From a clinical point of view, it is important to look for DBN in lateral gaze because it might otherwise be overlooked.
The most common presenting symptoms are unsteadiness of gait and to-and-fro vertigo [Wagner et al. 2008] On further inquiry, the patients frequently report blurred vision or oscillopsia that increases on lateral gaze. DBN is often associated with other ocular motor, cerebellar and vestibular disorders, predominantly smooth pursuit deficits and impairment of the optokinetic reflex and visual fixation suppression of the vestibuloocular reflex (VOR) [Leigh and Zee 2006; Glasauer et al. 2005a, 2004, 2003b; Straumann et al. 2000; Halmagyi et al. 1983].
The aetiology of DBN is diverse. In a recent study 117 patients were reviewed to establish whether analysis of a large collective and improved diagnostic means would reduce the number of cases with ‘idiopathic DBN’ and thus change the aetiological spectrum [Wagner et al. 2008]. In 62% (n¼ 72) of them, the aetiology was identified (‘secondary DBN’), the most frequent ones being cerebellar degeneration (n¼23) and cerebellar ischaemia (n¼10). In 38% (n¼45), no cause was found (‘idiopathic DBN’). A major finding was a high comorbidity of both idiopathic and secondary DBN with bilateral vestibulopathy (36%) and an association with polyneuropathy and cerebellar ataxia even without cerebellar pathology on MRI. From this study one can conclude that ‘idiopathic DBN’ remains common despite improved diagnostic techniques.
Animal studies in monkeys have shown that bilateral ablation of the cerebellar flocculus and paraflocculus result in DBN and an integrator deficit [Zee et al. 1981], lasting deficits in pursuit eye movements, impaired horizontal VOR adaptation [Rambold et al. 2002; Lisberger et al. 1984] and visual suppression of caloric nystagmus [Takemori and Cohen, 1974]. The upward drift of DBN consists of a gaze-evoked drift, which is hypothesised to be due to an impaired neural integrator function, and a spontaneous upward drift during gaze straight ahead [Glasauer et al. 2003a; Straumann et al. 2000]. Three different pathomechanisms are thought to cause the spontaneous upward drift: first, a tone imbalance of the central vestibular pathways of the vertical eye movements [Bohmer and Straumann, 1998; Dieterich and Brandt, 1995; Halmagyi et al. 1983; Baloh and Spooner, 1981], including otolith pathways as suggested by the finding that DBN is gravity-dependent [Sprenger et al. 2006; Marti et al. 2002]; second, an imbalance of the vertical smooth pursuit tone in which the imbalance of upward visual velocity commands results in spontaneous upward drift [Zee et al. 1974]; and third, a mismatch in the three-dimensional neural coordinate system for vertical saccade generation due to a defect of the neural velocity-to-position integrator for gaze holding [Glasauer et al. 2003a].
Marti et al.  have proposed a mechanism by which floccular deficiency causes DBN. They suggest that the distribution of the on-directions of vertical gaze-velocity Purkinje cells (PCs) is inherently asymmetrical. These cells are predominantly activated with ipsiversive and downward gaze velocity, but only ~10% of them show on-directions for upward-gaze velocity [Partsalis et al. 1995]. With functional magnetic resonance imaging (fMRI) and F-fluorodeoxyglucose-positron emission tomography, it was recently shown that patients with DBN have diminished activation/metabolism of both floccular lobes (Figure 3) [Bense et al. 2006; Kalla et al. 2006]. This supports the view that a functional deficiency of the flocculi causes not only a defect in downward pursuit but also DBN [Marti et al. 2005]. More recent studies using voxel-based morphometry demonstrated an atrophy in certain areas of the cerebellum, which are mainly related to ocular motor function [Hufner et al. 2007; Kalla et al. 2006].
Since the inhibitory influence of GABAergic Purkinje cells is assumed to be impaired in DBN, several agents that act on this receptor have been investigated. The GABAA agonist, clonazepam, improved DBN (dosage 0.5 mg t.i.d. to 1 mg b.i.d.), but these studies were not controlled [Young and Huang 2001; Currie and Matsuo 1986]. The GABAB agonist, baclofen, is assumed to reduce DBN [Dieterich et al. 1991], but as was shown in a double-blind crossover trial in a few patients, only one out of six responded to baclofen [Averbuch-Heller et al. 1997]. Further, the alpha-2-delta calcium channel antagonist gabapentin was assumed to have a positive effect on DBN, but again only one out of six patients responded positively [Averbuch-Heller et al. 1997].
On the basis of the assumed pathomechanism of DBN, the effects of aminopyridines were evaluated in a randomised, controlled, crossover trial involving 17 patients with DBN due to cerebellar atrophy, infarction, Arnold–Chiari malformation, or unknown aetiology [Strupp et al. 2003]. Mean peak slow-phase velocity of DBN was measured before and 30min after randomised ingestion of 20mg of 3,4-DAP or oral placebo. 3,4-DAP reduced peak slow-phase velocity of DBN from 7.2deg/s mean before treatment to 3.1 deg/s 30min after ingestion (p50.001) (Figure 4). The mean peak slow-phase velocity decreased in 10 of 17 patients by more than 50%. Except for transient perioral or digital paresthesia (three patients) and nausea and headache (one patient), no other side effects were observed. The authors demonstrated that the single dose of 3,4-DAP significantly improved DBN and visual acuity, and also reduced distressing oscillopsia. From a clinical point of view, it must be kept in mind that only 50% of all patients with DBN respond to this treatment, mainly those without structural lesions of the cerebellum or brainstem. The assumed underlying mechanism is that aminopyridines increase the activity and excitability of the Purkinje cells (as was found in animal experiments [Etzion and Grossman, 2001]), thereby augmenting the physiological inhibitory influence of the vestibular cerebellum on the vestibular nuclei. Meanwhile the effect of aminopyridines on the gravity dependence of DBN has also been evaluated [Helmchen et al. 2004] and an improvement of postural imbalance in DBN was demonstrated [Sprenger et al. 2005].
The underlying mechanism of action of 4-AP in DBN was also investigated in two studies using the magnetic search-coil technique [Kalla et al. 2007, 2004]. The major findings of these studies were as follows: first, 4-AP improved not only DBN, but also smooth pursuit and the gain of the vertical vestibulo-ocular reflex [Kalla et al. 2004] Second, 4-AP improved fixation by restoring gaze-holding ability and neural integrator function (Figure 5); further, as regards its aetiology-dependent efficacy in DBN, 4-AP may work best when DBN is associated with cerebellar atrophy [Kalla et al. 2007] (Figure 5). If DBN is caused by a structural lesion, 4-AP does not improve DBN in most cases. A PET study showed that 4-AP – in parallel to improving DBN – increases the metabolic activity of the flocculus [Bense et al. 2006] All these studies give additional support both to the above hypothesis about the pathophysiology of DBN and the way that aminopyridines act.
Upbeat nystagmus (UBN), that is, UBN with gaze straight ahead, is an ocular motor disorder that manifests with oscillopsia due to retinal slip of the visual scene and postural instability. It is the second most common cause of acquired nystagmus. UBN usually increases with upgaze. Analogously to DBN, it is associated with impaired upward pursuit. UBN can be caused by lesions in different brainstem and cerebellar regions such as the pontomesencephalic junction, medulla, or cerebellar vermis. Lesions in the pathways mediating upward eye movements, in particular, from the vestibular nuclei through the brachium conjunctivum to the ocular motor nuclei, might result in slow downward drift of the eyes, which is corrected by fast upward movements [Leigh and Zee, 2006]. Other hypotheses are that UBN is caused by an imbalance of vertical vestibulo-ocular reflex tone or a mismatch in the neural coordinate systems of saccade generation and neural velocity-to-position integration.
The symptoms persist as a rule for several weeks but are not permanent in most of the patients. Because the eye movements generally have larger amplitudes, oscillopsia in upbeat nystagmus is very distressing and impairs vision. Upbeat nystagmus due to damage to the pontomesencephalic brainstem is frequently combined with a unilateral or bilateral internuclear ophthalmoplegia, indicating that the MLF is affected. The main aetiologies are bilateral lesions in MS, brainstem ischaemia or tumour, Wernicke's encephalopathy, cerebellar degeneration and dysfunction of the cerebellum due to intoxication.
GABAergic substances like baclofen have been used to treat UBN and DBN, but they have had only moderate success. One study demonstrated a beneficial effect of baclofen (5–10 mg t.i.d.), but this trial was not controlled [Dieterich et al. 1991]. In a single patient with UBN it was shown that 4-aminopyridine (4-AP) reduced the peak slow-phase velocity in the light from 8.6 to 2.0 deg/s [Glasauer et al. 2005b]. 4-AP did not affect UBN in darkness, but it obviously activated pathways carrying visual information, which could then be used for UBN suppression in the light. Therefore, it was concluded that 4-AP reduces the downward drift in UBN by augmenting smooth pursuit commands. We propose that 4-AP helps to activate parallel pathways that can assume the function of the lesioned structures [Glasauer et al. 2005c]. 4-AP may strengthen these parallel pathways by increasing the excitability of cerebellar PCs [Etzion and Grossman 2001]. It may also evoke complex spikes in PCs similar to those elicited by climbing fibre stimulation [Cavelier et al. 2002].
Other forms of nystagmus which are associated with oscillopsia and in some patients with imbalance are acquired pendular nystagmus, periodic alternating nystagmus, convergence retraction nystagmus, central positioning or positional nystagmus (see above) and seesaw nystagmus. Congenital nystagmus often does not cause any symptoms. Square wave jerks, ocular flutter (mainly horizontal saccades), and opsoclonus (horizontal, vertical, and torsional saccades) belong to the saccadic intrusions or saccadic oscillations and are not classified as a nystagmus. In this review the features, pathophysiology, and treatment of congenital nystagmus, periodic alternating nystagmus and acquired fixation nystagmus will be summarised, because they are the clinically most relevant forms (see also Table 2).
This form of nystagmus most often beats horizontally and changes its direction every 60–180 seconds. Afflicted subjects complain of oscillopsia. Patients often turn their head in the direction of the quick phase and in this way bring their eyes in the direction of the slow phase of PAN to reduce oscillopsia – in accordance with Alexander's law. The diagnosis requires quite a long time of examination, otherwise one might overlook PAN. Like many other forms of nystagmus, PAN is most often caused by cerebellar dysfunction, in particular by lesions of the nodulus or the uvula. These lesions impair the velocity-storage mechanism as was shown in animal experiments and the oscillations are assumed to be caused by an ‘over-compensation’ or instability of the optokinetic-vestibular system [Leigh et al. 1981]. The treatment of choice is the GABAergic drug, baclofen, in a dosage of 5–10mg t.i.d., which abolishes PAN in most patients [Straube, 2005a, 2005b, Straube et al. 2004; Stahl et al. 2002]. There have been no randomised, controlled trials so far.
Acquired pendular nystagmus may have horizontal, vertical or torsional components. The amplitude varies, and in part the eye movements are not conjugate [Leigh et al. 2002; Stahl et al. 2000]. The clinical features and associated symptoms, in particular palatal tremor [Kim et al. 2007; Moon et al. 2003], often depend on the underlying disease. The three most common causes are multiple sclerosis, brainstem ischaemia, and Whipple's disease. In patients with multiple sclerosis APN has a frequency of 3–6Hz and is often associated with other central ocular motor disorders such as internuclear ophthalmoplegia or upbeat nystagmus. APN can also be associated with palatal tremor oculopalatal tremor. In such patients there is often a synchronisation of the nystagmus with the palatal tremor. MRI of patients in the chronic state often reveals a pseudohypertrophy of the inferior olivary nucleus [Deuschl and Wilms, 2002]. In a recent correlation between APN and MRI changes it was demonstrated that a dissociated APN predicts asymmetric (unilateral) inferior olivary pseudohypertrophy on MRI; however, symmetric pendular nystagmus was associated with either unilateral or bilateral signal changes in the inferior olivary nucleus [Kim et al. 2007; Moon et al. 2003]. It is assumed that oculopalatal tremor is caused by damage to the paramedian tract projections and denervation of the dorsal cap of the inferior olive, leading to an instability of eye velocity to position integration.
Several agents have been recommended for APN. One is trihexiphenidyl [Jabbari et al. 1987; Herishanu and Louzoun, 1986] but a doubleblind study demonstrated that only one of six patients responded to this treatment [Leigh et al. 1991]. Memantine, a glutamate antagonist, was also recommended [Starck et al. 1997], but its efficacy has not been proven. There are convincing data for gabapentin from a double-blind study by Averbuch-Heller et al. . They found a significant improvement in visual acuity and reduction of nystagmus with gabapentin in 10 of 15 patients but not with baclofen. The retrobulbar application of botulinum toxin was also recommended, but this was tested in only a small series of patients [Leigh et al. 1992] and was not always successful [Tomsak et al. 1995]. From a practical point of view, we now recommend using gabapentin (300–600mg t.i.d.) for acquired pendular nystagmus. Memantine and trihexiphenidyl are second and third choices, respectively.
Congenital nystagmus often develops during the first months of life. Some of the cases are familial and genetically heterogeneous. Autosomal dominant, autosomal recessive and X-linked patterns of inheritance have been reported. Linkage analysis suggested the existence of at least three distinct loci for both autosomal dominant and X-linked forms, although so far only one disease gene was identified on chromosome Xq26.2 [Self and Lotery 2007]. Congenital nystagmus is clinically characterised by the following criteria: fixation nystagmus (i.e. no decrease of the intensity during fixation); nystagmus most often beating horizontally; large variability of form and form frequency and velocity; intensity depending on gaze position; often a position (the so-called neutral zone) with a minimal intensity which the patient prefers and which leads to an appropriate head turn. Examination with the optokinetic drum often shows an inversion of the direction or during vertical optokinetic stimulation, a diagonal nystagmus. It is important to know that most patients do not have any complaints, namely they have no oscillopsia despite a high intensity of nystagmus. This is most likely due to an impairment of visual motion perception in these subjects. Since most patients do not have any medical complaints, treatment is generally not necessary. In patients with oscillopsia, one might try gabapentin or memantine. In a randomised, controlled, double-blind study it was demonstrated that memantine at a dosage of 10–40mg per day (as well as gabapentin at a dosage of 600–2400mg per day) caused a significant decrease of the intensity of the nystagmus and an increase of visual acuity [McLean et al. 2007]. This, however, was not associated with visual acuity during regular daily activities or improvement of the patients’ disease-related quality of life.
Considerable progress has been made over the last decades in the description of the clinical characteristics of different forms of nystagmus, its pathophysiology, and aetiology. However, because there are several forms of nystagmus and underlying central vestibular, ocular motor and in particular cerebellar disorders, effective drugs are still awaiting prospective, randomised, placebo-controlled and – due to the low prevalence of some of these disorders – multicentre trials. It is high time that these studies were performed. Several drugs could be potentially effective (cited in alphabetical order): acetazolamide, aminopyridines, anticholinergics (benztropine, scopolamine, trihexyphenidyl), baclofen, barbiturates, benzodiazepines, cannabinoids, carbamazepine, gabapentin, lamotrigine, memantine, phenytoin, selective serotonin reuptake inhibitors, tricyclic antidepressants, topiramate, triptans or valproic acid. In other words, there is still a lot to do. Knowledge of the possible effects of these agents – most of which act specifically on certain receptors or ion channels – will also further improve our insights into the patho-physiology of the underlying disorders.
Michael Strupp, Professor of Neurology and Clinical Neurophysiology, University of Munich, Klinikum Grosshadern, Munich, Germany ; Email: ed.nehcneum-inu.dem@ppurtS.leahciM.
Thomas Brandt, Institute of Clinical Neuroscience, University of Munich, Klinikum Grosshadern, Munich, Germany.