Introduction

Central positional nystagmus (CPN) is characterized by nystagmus triggered by changes in head position, and is caused by cerebellar and/or brainstem disorders involving the central vestibulo-cerebellar pathways, either in a permanent or transient fashion. These include structural lesions (e.g., tumors, stroke, multiple sclerosis), auto-immune (paraneoplastic and non-paraneoplastic), degenerative and genetic ataxias, vestibular migraine, and toxicity, among other [1•]. CPN can be paroxysmal or persistent (duration < 1 and > 1 min, respectively), and its direction is frequently vertical (e.g., downbeating or upbeating) in head hanging positions and horizontal (e.g., apogeotropic or geotropic) in lateral supine positions, often being multiplanar in the same individual. In contrast to its peripheral counterpart, i.e., benign paroxysmal positional vertigo (BPPV), CPN is frequently associated with additional central ocular motor and neurological signs, and does not respond to liberatory maneuvers [1•]. CPN is believed to reflect an abnormal integration of semicircular canal–related signals by the cerebellar nodulus, uvula, and/or tonsil, ultimately providing an erroneous estimation of head tilt and/or eye position coordinates [2–5]. CPN can be associated with intense vertigo, oscillopsia, and dysautonomia, particularly the paroxysmal form, and therefore, treatment might be needed [2, 3, 6, 7]. The latter is usually directed primarily towards the underlying disorder whenever possible, with the additional goal of alleviating clinical symptoms. Despite CPN being highly prevalent in several CNS disorders (e.g., 15% of patients with diffuse cerebellar disease [8], 90% of patients during an attack of vestibular migraine [9], 50% of patients with isolated cerebellar stroke [10], and up to 80% of patients with infratentorial tumors [11]), objective and standardized assessment of CPN before and after treatment has been scarcely reported in the literature. We provide a critical appraisal of current therapeutic strategies for CPN.

Methods

A PubMed search until 15 April 2022 using the key words “Central Positional Nystagmus”, “Central Positional Nystagmus Treatment”, “Positional Nystagmus” OR “Positioning Nystagmus” OR “Positional Vertigo” OR “Positioning Vertigo” AND “Central” AND “Treatment” was performed. Pertinent articles were assessed, and their references used to identify further publications of interest.

Treatment

We will discuss CPN treatment based on common etiologies and representative cases. See Table 1 for suggested recommendations on CPN treatment and Table 2 for a detailed list of CPN patients reported in the literature who underwent a specific treatment for CPN, and had information on CPN phenotype and outcome of CPN and/or related symptoms available.

Table 1 Treatment of central positional nystagmus — suggested recommendations
Full size table
Table 2 Treatment of central positional nystagmus — detailed cases
Full size table

Structural Lesions

Strategic lesions causing CPN commonly involve the cerebellar nodulus, uvula, tonsil, and/or areas containing fibers connecting the latter structures with the vestibular nuclei (e.g., cerebellar peduncles, lateral medulla, and dorsolateral region of the 4th ventricle) [1•].

Tumor

In patients with CPN associated with infratentorial (mostly cerebellar) tumors, both positional nystagmus, vertigo, and/or posturally evoked vomiting may be fully or partially improved after tumoral resection [12, 13] and post-operative radiation [14–16], albeit patients might be left with new ocular motor and vestibular deficits attributable to the surgical procedure [14, 16]. The above treatments might improve CPN by alleviating the compression of critical structures and/or pathways within the central vestibulo-cerebellar network, known to participate in CPN’s generation, when dysfunctional. Baloh and colleagues provided electronystagmographic assessment in two representative cases. The first patient, an 18-year-old female, had paroxysmal downbeating CPN, vertigo, and nausea in head hanging position due to a cerebellar astrocytoma. Six months after resection and radiation, there was no positional vertigo. The second patient, a 16-year-old female, had paroxysmal upbeating CPN, vertigo, and nausea in supine position due to a midline cerebellar astrocytoma. Five months after sub-total lesion resection and radiation, positional symptoms and CPN subsided [16]. Tumor resection, however, does not always grant symptoms and/or CPN resolution. Lea et al. described a 65-year-old male with paroxysmal downbeating CPN and vertigo on lateral head hanging associated with a hemangioblastoma involving the left cerebellar peduncle and extending into and deforming the midbrain and pons. Six months after resection of the lesion, CPN and vertigo were still present [17]. Still, in the above context, surgery and chemotherapy/radiotherapy constitute mainstay therapies which can be life-saving and might significantly increase life expectancy and, therefore, are performed regardless of CPN outcome.

Pharmacological treatment directed to the putative underlying mechanism of CPN, constitutes another line of therapy for tumor-related CPN, particularly in patients where CPN appeared or was not improved after surgery. Aminopyridines, i.e., 3,4-diaminopyridine (DAP) [18] and 4-aminopyridine (4AP) [19], have been used anecdotally, showing some benefit. DAP was used in 43-year-old male with intense paroxysmal downbeating CPN, vertigo, and vertical oscillopsia in straight head hanging, caused by surgical removal of a cerebellar nodulus and uvula multi-nodular dysembryoplastic neuroepithelial tumor. Ninety minutes after ingestion of DAP 20 mg, there was a modest (21%) reduction of CPN slow phase mean velocity (72 to 56°/s) [18]. Data on patient’s symptoms after DAP was not provided. Possible interference of nystagmus fatigability in the results, a phenomenon which might be seen in CPN, was not discussed either [20]. Strupp and colleagues used 4AP in a 45-year-old female with paroxysmal downbeating CPN and vertigo on backward head movements and supine position, associated with a cerebellar vermis hemangioblastoma, after showing no improvement with surgery. After 3 days of 4-AP 5 mg t.i.d., symptoms and CPN disappeared, having reappeared 2 days after 4-AP treatment was stopped [19]. Paroxysmal downbeating CPN in the above cases was ascribed to a deficient uvulo-nodular inhibition of the brainstem velocity storage, vertical angular vestibulo-ocular reflex and/or otolith-ocular reflex during rapid head tilts. Aminopyridines, possibly by blocking potassium channels and increasing of Purkinje cells excitability [21], may have partially restored the uvulo-nodular deficit, this way reducing the asymmetric charge of the velocity storage [18] or decreasing the overactive vertical reflexes [19]. Further increase in regional cerebral glucose metabolism in the nodulus in one of the patients during 4AP treatment supports the above mechanism [19]. Unfortunately, none of the reports provided data on the long-term efficacy of DAP or 4AP in CPN and related symptoms.

Stroke

Strategic infratentorial vascular lesions may cause CPN [10, 22, 23]. While CPN and positional vertigo might go unnoticed during the acute phase unless detailed video-oculographic assessment is performed, CPN may persist over time and be associated with disabling positional vertigo, particularly in patients with downbeating (but not upbeating, apparently) CPN [23]. Also here, pharmacological treatment may significantly relieve symptoms and CPN, although reports on the efficacy of treatment of stroke-related CPN are scarce. Recently, a 58-year-old male with paroxysmal downbeating CPN and vertigo in head hanging positions after intraventricular hemorrhage was described. Mild paroxysmal upbeating CPN when uprighting, and apogeotropic CPN during the supine head roll were also present. An initial trial with baclofen at 20 mg/day for 1 week failed due to intolerance to the drug. Subsequent treatment with clonazepam 1 mg/day, improved positional vertigo, and maximal slow phase velocity of downbeat CPN decreased from 32 to 6°/s. At the 2-month follow-up there was complete resolution of CPN and vertigo. Symptoms and CPN reappeared after clonazepam cessation, and were again abated once clonazepam was resumed [24]. Clonazepam enhances the activity of the inhibitory neurotransmitter gamma-aminobutyric acid (GABA). In this case, clonazepam might have abated CPN by enhancing the inhibitory GABAergic synaptic transmission from the cerebellar nodulus and uvula to the vestibular nuclei, this way stabilizing a presumed exaggerated postacceleratory secondary response due to disinhibited irregular vestibular afferents [3, 24]. In another patient with paroxysmal downbeating CPN in head hanging positions after a lateral medullary stroke, CPN and/or symptoms did not improve with the introduction, 20 months after the onset of symptoms, of baclofen 10 mg bid for 2 months, followed by 4AP 10 mg bid for 2 months (mean slow phase velocity, pre- [16.4°/s] and 2 months post-treatment [14.4°/s]) (patient 18 from reference [23]).

Multiple Sclerosis

CPN is relatively rare in multiple sclerosis (MS) (~ 6%) [25], and BPPV [26] and vestibular migraine probably account for the majority of MS patients with recurrent positional vertigo. Similarly to stroke-related CPN, CPN and related symptoms in MS can present only acutely during a relapse, or persist afterwards and become chronic [23, 27, 28]. Anagnostou et al. reported a 55-year-old female with persistent torsional CPN in right head hanging position, possibly associated with a gadolinium enhancing demyelinating plaque in the inner part of the left superior cerebellar peduncle. The patient was started on 500 mg/day intravenous methylprednisolone for 5 days, leading to the disappearance of CPN and symptoms 5 days later [27]. The same group had previously reported a 60-year-old female with persistent downbeating CPN, vertigo, and vomiting in lateral supine, convincingly associated with a gadolinium enhancing demyelinating plaque in the right superior cerebellar peduncle. As in the former case, administration of intravenous methylprednisolone (1000 mg/d for five days), in this case combined with 2 mg/day clonazepam, “brought about gradual improvement of the symptoms” [28]. In the above cases, a strategic lesion in the superior cerebellar peduncle might have disrupted central otolithic connections between deep cerebellar structures and the vestibular nuclei [27]. Such disruption is believed to create a mismatch between eye position estimated by the burst generator and the neural integrator, hypothetically due to altered otolith input to one of these systems [4]. Methylprednisolone might have improved CPN by exerting its anti-inflammatory effects at several levels, including the inhibition of the inflammatory edema and the restoration of the blood–brain barrier disruption (which has been described to occur shortly after administration) [29], and modulation of cytokine and chemokine production, adhesion-molecule expression, and inflammatory-cell accumulation [30]. All of the above mechanisms may have helped to restore the otolithic connections between the cerebellum and the vestibular nuclei. In MS, two additional mechanisms should be taken into account for explaining paroxysmal events such as CPN, namely transversely spreading ephaptic activation of axons [31] and ion channel dysfunction [32] in partially demyelinated lesions. This is relevant, because paroxysmal events in MS can dramatically respond to antiepileptics (e.g., carbamazepine, phenytoin, lamotrigine, lacosamide, levetiracetam) and acetazolamide [31–33]. In another patient with paroxysmal downbeating CPN in head hanging positions after right pontomedullary demyelination, CPN and vertigo did not improve with the introduction, 16 months after the onset of symptoms, of 4AP 10 mg (mean slow phase velocity, pre- [8.8°/s] and 90-min post-treatment [10°/s]). She nevertheless subjectively improved positional vertigo (without a relevant change in CPN) after the introduction, 2 years after the onset of symptoms, of phenytoin 50 mg/day (patient 21 from reference [23]).

Autoimmune Ataxia

Immune-mediated cerebellar ataxias (IMCAs) constitute a group of ataxias that might respond well to induction and maintenance immunotherapy (e.g., steroids, immunoglobulins, immunosuppressants such as mycophenolate, plasmapheresis and/or rituximab), in addition to a gluten-free diet or tumor-directed therapy, when applicable. These include paraneoplastic cerebellar degeneration (PCD) and antiglutamic acid decarboxylase antibody (GAD) ataxia, among other [34]. Rarely, PCD may present with predominant positional vertigo and CPN, apart from subacute cerebellar ataxia and other cerebellar signs. CPN response to treatment in these cases is heterogenous, may or may not follow tumor response, may fail over time, and prognosis is largely dependent on the overall survival dictated by the primary tumor type and stage [35–37]. Since PCD often constitutes an immunological response against neuronal antigens expressed both by an underlying tumor and the cerebellum, tumor-directed therapy (e.g., resection, chemotherapy, radiotherapy, other) and immunotherapies (e.g., steroids, immunoglobulin, plasmapheresis, cyclophosphamide, rituximab) might theoretically improve CPN [38]. In a 28-year-old male with a seminoma and paraneoplastic paroxysmal torsional nystagmus during right head hanging and downbeating after sitting up, spontaneous and positional vertigo, gaze evoked nystagmus, and gait and limb ataxia, bilateral orchiectomy, radiotherapy, intravenous methylprednisolone, and oral cyclophosphamide were associated with substantial improvement of vertigo and ataxia [37]. While remission of PCD/CPN is possible after excision of the primary tumor, relapses may follow as exemplified in a 63-year-old male with a chondrosarcoma and paraneoplastic paroxysmal torsional CPN during right supine, spontaneous horizontal nystagmus, ocular flutter, and ataxia. Four weeks after tumor resection, PCD/CPN disappeared. However, 6 weeks later, symptoms returned and progressed inexorably, showing no response to chemotherapy, plasmapheresis, or to baclofen (10 mg four times daily) [35]. Additionally, CPN and tumor-related responses may be disparate. In another patient, a 25-year-old male with Hodgkin lymphoma and paraneoplastic paroxysmal downbeating CPN and vertigo in supine, prone, and head-hanging positions, CPN did not respond to chemotherapy (2 cycles of adriamycin, bleomycin, vinblastine, and dacarbazine) or radiotherapy (20 Gy radiation), while the tumor itself fully remitted [36].

Non-paraneoplastic IMCAs can also manifest with symptomatic CPN, including anti-GAD Ab- and anti-GQ1b Ab-related ataxias. In one representative case, a 68-year-old diabetic female presented with intense paroxysmal upbeating CPN, vertigo, and nausea in supine and head hanging positions and mild asymptomatic geotropic CPN in lateral supine. High titters of anti-GAD65 Ab were identified (239.07 U/mL). Interestingly, an initial trial with baclofen 2.5 mg 3 times daily fully abated CPN. However, the effect waned after 2 weeks, and the dosage could not be increased because of gastric intolerance. Subsequently, the patient was given intravenous immunoglobulin, 30 g/day over 5 days, followed by 30 g monthly administration, showing complete and sustained resolution of CPN and vertigo. Six months later, anti-GAD65 Ab levels had fallen to 40.88 U/mL [39]. In anti-GAD ataxia, an anti-GAD Ab–mediated autoimmune attack on the cerebellum and/or brainstem, causing a selective deficiency in gamma-aminobutyric acid (GABA) neurotransmission, has been hypothesized. Baclofen is a GABA agonist known to affect the vestibular and oculomotor systems. GABAB receptors are widely distributed in the cerebellum, and GABA is the major transmitter of the Purkinje cell output that ends in the vestibular nuclei [40]. In the above case, baclofen might have helped to restore nodulus/uvula GABAergic inhibitory output over the vestibular nuclei, this way abating CPN [39]. On the other hand, immunoglobulin use might have indirectly restored the same inhibitory output by exerting an immunomodulatory and anti-inflammatory effect [30, 39].

Recently, high titers of anti-GQ1b Ab have been associated with acute CPN, including persistent apogeotropic and geotropic CPN during lateral supine, and downbeating and upbeating CPN in head hanging positions [41, 42•]. Remission of anti-GQ1b Ab–related CPN, in cases where the outcome was detailed, was spontaneous over 3 to 12 months, without requiring specific treatment [41]. Intravenous immunoglobulin, plasmapheresis, and intravenous steroids have been reported as successful therapeutics in acute vestibular syndrome related with anti-GQ1b Ab. However, their use in CPN anti-GQ1b Ab–related CPN has not been detailed [42•]. Of note, plasmapheresis has a broad application in IMCAs. It probably acts by removing pathologic antibodies, immune complexes, and cytokines and may also have an immunomodulatory effect [43].

Degenerative and Genetic Ataxia

Patients with degenerative and genetic ataxia associated with CPN seem to demonstrate more frequently positional vertigo and nausea, and therefore, specific treatment might be needed [8]. Mutations in the brain-specific P/Q type Ca2 + channel alpha1 subunit gene, CACNA1A, have been identified in three clinically distinct disorders, namely spinocerebellar ataxia type 6 (SCA6), familial hemiplegic migraine type 1 (FHM1), and episodic ataxia type 2 (EA2) [44]. SCA6 usually manifests as pure cerebellar ataxia with a late onset. Around 80% of SCA6 patients show CPN (mostly downbeating) and positional vertigo, often preceding the development of ataxia [45]. SCA6 patients may also have ictal episodes of ataxia lasting hours to days, often precipitated by physical exertion or emotional stress [46]. In 3 SCA6 patients with inter-ictal positional vertigo/downbeating CPN and truncal ataxia, and ictal episodes of ataxia (present in two of the patients), the use of acetazolamide 500 to 750 mg/day markedly decreased ictal episodes of ataxia, while showing no benefit on CPN and/or positional vertigo [46]. FHM1 on the other hand is mainly characterized by episodes of recurrent hemiplegia during the aura phase of a migraine headache, but progressive cerebellar ataxia is present in ~ 50% of patients, usually occurring independently of the migraine attacks [44, 47]. Thus, CPN may also be a feature in these patients. Suzuki et al. reported a 47-year-old patient with FMH1 and downbeating CPN, vertigo, and oscillopsia in supine position, in whom CPN slowly but completely resolved after 14 months of treatment with acetazolamide 500 mg, twice daily. Headaches were equally abolished [47]. In a similar FHM1 patient however, only headaches improved, while downbeating CPN remained unchanged after the same dosage of acetazolamide [48]. Finally, EA2 is characterized by paroxysmal attacks of ataxia triggered by stress and exercise. While downbeating CPN has been described in EA2, and both acetazolamide 750 mg/day and 4AP 20 mg/day have been shown to reduce EA2 attacks up to 60%, there are no reports evaluating CPN response to treatment [49, 50••]. Levetiracetam might reduce the attacks in EA2 patients in whom the former drugs are contraindicated, non-efficient or associated with intolerable adverse effects [51]. CPN in the above genetic ataxias probably represents an early manifestation of cerebellar atrophy (and neuronal loss), sometimes associated with cerebellar vermian atrophy in MRI [47]. The associated paroxysmal/ictal episodes (e.g., ataxia, headache), on the other hand, are believed to reflect a transient calcium channel dysfunction [52]. Acetazolamide, a carbonic anhydrase inhibitor, probably exerted its effect in the above cases by aiding in stabilizing transient abnormal ion channel function, through its modulation of pH (i.e., acidification) [53]. This might explain why acetazolamide was more effective in resolving paroxysmal/ictal episodes than permanent CPN and ataxia [46, 48]. In one FHM1 case, however, CPN was slowly abated after acetazolamide. Here, CPN might have represented an ictal phenomenon precipitated by abrupt head movements and not a permanent dysfunction of the vertical vestibulo-ocular reflex cancellation due to cerebellar neuronal loss [47]. Importantly, reports showing positive therapeutic responses of CPN in CACNA1A mutation–related diseases should be cautiously interpreted, since spontaneous resolution (i.e., without treatment) of vertigo has been observed in 4/13 SCA6 patients, before the onset of the ataxia [52].

In multiple system atrophy (MSA), a neurodegenerative disorder characterized by varying severity of parkinsonian, autonomic, and cerebellar features, the presence of CPN (mostly downbeating) constitutes an important diagnostic clue for MSA in a patient presenting with a parkinsonian disorder [54]. Unfortunately, also here, there are no reports concerning CPN treatment and/or progression over time.

Vestibular Migraine

Vestibular migraine (VM) is one of the most common causes of recurrent vertigo. During VM attacks, 64% of patients refer positional vertigo, and CPN, usually mild, persistent and with variable direction, is observed in ~ 90% of cases [9]. Since there is no strong evidence that any treatment is effective for improving VM in the long-term, patients are usually treated with drugs commonly used in the preventive treatment of migraine (e.g., betablockers, antidepressants, antiepileptics, calcium channel blockers, etc.). In recent systematic reviews and meta-analyses, it was shown that propranolol and venlafaxine seem to show slightly more benefit than other drugs in the short-term follow-up (< 12 weeks), and venlafaxine may be superior to other drugs in improving depressive symptoms. Still, several other drugs also demonstrated improvement in dizziness handicap inventory (DHI) score and vertigo attacks’ frequency and severity [55•, 56•]. Unfortunately, the only double-blind randomized placebo-controlled trial investigating the effectiveness of preventive treatment with metoprolol was inconclusive, due to poor patient accrual [57]. Studies specifically addressing CPN and positional vertigo response to preventive treatment in VM are rare. In a landmark paper by Marianne Dieterich and Thomas Brandt, among 90 VM patients, 3 had attacks consisting of severe positional vertical CPN (not habituating). As in all other patients in this series, in these three, migraine prophylaxis abolished the attacks [58]. Following this paper, several other small studies consistently showed that VM patients with predominant positional vertigo and CPN also responded to migraine preventive treatment, by improving their attacks’ frequency and/or severity. For instance, VM attacks consisting of geotropic or apogeotropic CPN during lateral supine (and head hanging positions in one of the patients) were completely abolished with the use of topiramate 25 mg twice daily in one patient and 200 mg twice daily in another [59, 60]. In another series, all VM patients (n = 13) with a positional vertigo attack associated with apogeotropic or geotropic persistent CPN in lateral supine positions, experienced symptom resolution after initiation of preventive treatment [61]. However, CPN data after treatment was not provided. Importantly, in VM, ictal CPN seems to spontaneously resolve, improve, or change its direction over time [62, 63]. Yet, it is not clear if preventive treatment can influence/fasten this process (i.e., CPN progression during an attack). El-Badry et al. for instance showed that vertical persistent CPN (both upbeating and downbeating) and positional vertigo completely subsided in 12 out of 13 VM patients, after initiating cinnarizine, 37.5 mg twice daily, or topiramate, 50 mg once daily at night [64]. However, in this study, it was not completely clear whether CPN was observed during (i.e., ictal CPN) or between VM attacks (i.e., inter-ictal CPN). In another VM series, 10 patients with acute persistent vertical and horizontal CPN and vertigo in lateral head hanging and supine positions, showed complete resolution of CPN and vertigo after topiramate with gradual increasing dose from 25 up to 100 mg [65]. According to the authors, after 6 months of complete improvement, any trial of gradual stopping of topiramate from the patient’s effective dose would make CPN and vertigo return [65]. It must be noted that in the above retrospective studies, in the absence of a placebo arm and a prospective design, it is difficult to appreciate the potential short-term effects of migraine preventive therapy on ictal CPN. Even conclusions about the long-term effects of these drugs should be carefully weighted, taking into account the natural course of VM, which can spontaneously vary over time [66, 67]. Additionally, in some VM treatment studies, one additional potential confounder in the results is the inclusion of patients taking additional vestibular suppressants in the acute phase, which can theoretically influence CPN intensity/phenotype [64]. Prophylactic drugs in VM may act on one or more of the following putative pathomechanisms: spreading depression, vasoconstriction, ion channel defect, sterile inflammation, release of neuropeptides, reduced threshold for trigeminovascular stimulation, and/or locus coeruleus and dorsal raphe nucleus hypersensitization, all ultimately affecting the cerebellar nodulus, uvula and tonsil, and/or the vestibular nuclei [63, 68, 69, 70].

On the other hand, abortive treatment in a VM attack could eventually dampen ictal CPN and positional vertigo more quickly. In one study addressing the effectiveness of zolmitriptan in VM attacks, results were inconclusive, due to limited study power [67]. In a recent work however, the use of non-invasive vagus nerve stimulation (NVNS) in a 58-year-old female patient with a VM attack, dramatically abolished persistent right beating CPN and vertigo during left head hanging and supine positions, 15 min after stimulation [71•]. As VM attacks can last as little as 5 min, further studies are needed to accurately assess the role of NVNS in VM-related CPN. NVNS seems to stimulate vagal fibers which travel to brainstem loci that host trigemino-vestibulovagal connections and further modulate the neuronal activity in the vestibular nuclei, vestibulocerebellum and their connections [71, 72]. Alternative mechanisms of NVNS include the inhibition of neurogenic inflammation, trigeminal nucleus caudalis activity, and the firing rate of trigeminocervical neurons [72].

Toxicity

Drugs with primary action in the central nervous system, putatively involving the cerebellar nodulus, uvula, and/or tonsil, may cause CPN [73–75]. The few reports available on CPN due to toxicity, which further provided CPN outcome and/or related symptoms after intervention, include the following: a 70-year-old female with paroxysmal downbeating CPN, vertigo, and dysautonomia during head hanging while on pregabalin 150 mg twice a day [73]; a 31-year-old female with downbeating CPN, vertigo, and dysautonomia during head hanging positions while on lamotrigine 600 mg/day [74]; a 70-year-old male with downbeating CPN, mild vertigo, and intense vomiting during off-axis head-trunk tilt and lateral supine, while on amiodarone 600 mg/day [75]. Discontinuation (pregabalin) or decreased dosing of the offending drug (lamotrigine), and/or the use of specific therapies including midazolam and clonazepam (amiodarone), all led to resolution or relief of positional vertigo and dysautonomia. CPN resolved 5 days, within 6 months, and possibly 5 weeks after intervention, respectively [73–75].

Vestibular suppressants (e.g., dimenhydrinate, flunarizine, cinnarizine, promethazine) and antiemetics (e.g., ondansetron, domperidone, metoclopramide) may temporarily improve CPN-related symptoms [12], but should be otherwise used judiciously only for short periods of time, taking into account their side effects and possible negative influence on vestibular adaptation [76].

Conclusions

Currently, CPN treatment options are based on anecdotal data. Moreover, in the few reports where objective assessment of CPN before and after treatment is available, rigorous assessment of the impact of related symptoms (vertigo, oscillopsia, dysautonomia, etc., by using, e.g., DHI) is usually lacking, as well as long-term evaluation of CPN and symptoms. Since CPN intensity and related symptoms may not always correlate or coincide in an individual patient, both objective and subjective assessments are needed to correctly interpret treatment responses. Moreover, since the natural progression of CPN and related symptoms is largely unknown in the majority of the conditions listed in the current paper, it becomes difficult to contextualize a single measurement in time showing CPN improvement after a specific therapeutic intervention. This issue becomes even more relevant when it is known that CPN intensity spontaneously fluctuates and is influenced by repeat positioning and velocity of the precipitant maneuver [49, 77]. Perhaps as an alternative to the use of single CPN measurements in the clinic, prone to several types of biases, continuous vestibular monitoring during daily activities could become a promising tool in CPN treatment trials, as it has the potential to provide a long-term, objective recording of eye movements under more natural conditions, data which could eventually better correlate with patients’ symptoms [78]. In sum, prospective randomized multicentric treatment trials including different CPN etiologies are highly needed to better support treatment strategies in CPN.