Characteristics and mechanism of apogeotropic central positional nystagmus

Characteristics and mechanism of apogeotropic central positional nystagmus Abstract Here we characterize persistent apogeotropic type of central positional nystagmus, and compare it with the apogeotropic nystagmus of benign paroxysmal positional vertigo involving the lateral canal. Nystagmus was recorded in 27 patients with apogeotropic type of central positional nystagmus (22 with unilateral and five with diffuse cerebellar lesions) and 20 patients with apogeotropic nystagmus of benign paroxysmal positional vertigo. They were tested while sitting, while supine with the head straight back, and in the right and left ear-down positions. The intensity of spontaneous nystagmus was similar while sitting and supine in apogeotropic type of central positional nystagmus, but greater when supine in apogeotropic nystagmus of benign paroxysmal positional vertigo. In central positional nystagmus, when due to a focal pathology, the lesions mostly overlapped in the vestibulocerebellum (nodulus, uvula, and tonsil). We suggest a mechanism for apogeotropic type of central positional nystagmus based on the location of lesions and a model that uses the velocity-storage mechanism. During both tilt and translation, the otolith organs can relay the same gravito-inertial acceleration signal. This inherent ambiguity can be resolved by a ‘tilt-estimator circuit’ in which information from the semicircular canals about head rotation is combined with otolith information about linear acceleration through the velocity-storage mechanism. An example of how this mechanism works in normal subjects is the sustained horizontal nystagmus that is produced when a normal subject is rotated at a constant speed around an axis that is tilted away from the true vertical (off-vertical axis rotation). We propose that when the tilt-estimator circuit malfunctions, for example, with lesions in the vestibulocerebellum, the estimate of the direction of gravity is erroneously biased away from true vertical. If the bias is toward the nose, when the head is turned to the side while supine, there will be sustained, unwanted, horizontal positional nystagmus (apogeotropic type of central positional nystagmus) because of an inappropriate feedback signal indicating that the head is rotating when it is not. vertigo, nystagmus, positional nystagmus, velocity-storage mechanism, nodulus Introduction A change in the orientation of the head with respect to the pull of gravity may induce paroxysmal (transient) or persistent positional nystagmus in central as well as peripheral vestibular disorders (Büttner et al., 1999; Bisdorff et al., 2009; Kim and Zee, 2014). In peripheral disorders, the most common is benign paroxysmal positional vertigo/nystagmus (BPPV), which occurs when gravity pulls otoconia that are free-flowing in the canals or adherent to the cupula (Kim and Zee, 2014). In contrast, positional nystagmus in central vestibular disorders may be caused by impaired central processing of canal and otolith cues (Büttner et al., 1999). Since the therapeutic strategy and prognosis differ between these disorders, one must distinguish between them in clinical practice. Among the central types of positional nystagmus, we previously demonstrated that central paroxysmal positional nystagmus (CPPN) has several characteristics that may differentiate it from BPPV (Choi et al., 2015). It was hypothesized that CPPN is initiated by enhanced post-rotatory canal signals and is directed to the vector of inhibited canals during the positioning (Choi et al., 2015; Choi and Kim, 2017). Among the types of persistent central positional nystagmus (CPN), two have been commonly reported in various disorders affecting the brainstem or cerebellum: downbeat nystagmus while supine or prone, and apogeotropic horizontal nystagmus during ear-down positions while supine (the direction of nystagmus beating ‘away from the earth’ with head turned to either side while supine) (Marti et al., 2002; von Brevern et al., 2005; Sprenger et al., 2006; Anderson et al., 2008; Yu-Wai-Man et al., 2009; Lee et al., 2014). Since downbeat nystagmus is well recognized as a central form of nystagmus, persistent positional downbeat nystagmus poses little diagnostic difficulty. In contrast, apogeotropic CPN may mimic the peripheral form of lateral canal BPPV related to cupulolithiasis (apogeotropic BPPV), one of the common causes of positional nystagmus (Baloh et al., 1995; Honrubia et al., 1996). Indeed, several studies have reported apogeotropic nystagmus as an isolated or a predominant sign of central vestibular disorders (Moon et al., 2009; Nam et al., 2009; Kim et al., 2012). However, to our knowledge, no study has systematically investigated the characteristics of apogeotropic CPN in comparison with those of BPPV. Furthermore, the neuroanatomical correlates of apogeotropic CPN have not been settled in humans even though clinical studies and animal experiments have implicated the vestibulocerebellum, especially the nodulus and uvula (Sheliga et al., 1999; Moon et al., 2009; Nam et al., 2009; Kim et al., 2012). Moreover, a fundamental question is how central lesions generate apogeotropic positional nystagmus. Abnormalities in central graviceptive pathways have been implicated based on the reproducibility of the nystagmus in specific head positions (Sheliga et al., 1999; Glasauer et al., 2001). Since an internal estimate of gravitational orientation is the only difference between the left and right ear-down positions (Fig. 1A), the oppositely-directed nystagmus [e.g. left beating on right-ear-down (RED) and right beating on left-ear-down positions (LED)] indicates that processing of information about gravity must be involved. However, this hypothesis has not been tested using human data and computational simulations of processing within central vestibular networks. Figure 1 View largeDownload slide Gravity orientation according to the head positions, nomenclature of apogeotropic nystagmus, and determination of nystagmus intensity. (A) Gravitational orientation. The gravity direction is the same between the head and earth coordinate frames in head upright position. During the positional test, the direction of gravity along the interaural axis was the only difference between the LED and RED positions while supine in head coordinate reference frame. (B) Ipsiversive and contraversive nystagmus are defined according to the direction of spontaneous nystagmus while supine, and ipsilesional and contralesional nystagmus are defined according to the lesion side (left-sided lesion is assumed in this figure). (C) Illustration showing how the intensities of ipsiversive (ipsilesional), contraversive (contralesional), and position-induced nystagmus are determined. g = gravity; LB/RB = left/right beating nystagmus; SPVs = slow-phase velocities. Positive value indicates rightward slow-phase velocities. Figure 1 View largeDownload slide Gravity orientation according to the head positions, nomenclature of apogeotropic nystagmus, and determination of nystagmus intensity. (A) Gravitational orientation. The gravity direction is the same between the head and earth coordinate frames in head upright position. During the positional test, the direction of gravity along the interaural axis was the only difference between the LED and RED positions while supine in head coordinate reference frame. (B) Ipsiversive and contraversive nystagmus are defined according to the direction of spontaneous nystagmus while supine, and ipsilesional and contralesional nystagmus are defined according to the lesion side (left-sided lesion is assumed in this figure). (C) Illustration showing how the intensities of ipsiversive (ipsilesional), contraversive (contralesional), and position-induced nystagmus are determined. g = gravity; LB/RB = left/right beating nystagmus; SPVs = slow-phase velocities. Positive value indicates rightward slow-phase velocities. In this study, we compared apogeotropic CPN with apogeotropic BPPV, analysed the location of lesions, suggest a mechanism for apogeotropic CPN, and simulated a mathematical model of the central interactions between the inputs from the semicircular canals and the otolith organs that can produce apogeotropic CPN. Materials and methods Patients This study was approved by the institutional review board of Seoul National University Bundang Hospital. We evaluated 27 patients (19 males and eight females, 58.5 ± 15.0 years) with positional vertigo and apogeotropic CPN. All patients underwent detailed neuro-otological evaluation by the senior author (J.S.K). The diagnosis of apogeotropic CPN was based on the following criteria: (i) presence of apogeotropic horizontal nystagmus; (ii) persistent nystagmus without fatigability; (iii) no resolution of apogeotropic nystagmus with repeated canalith repositioning manoeuvres; and (iv) documentation of the pathologies involving the brainstem or cerebellum with brain imaging or other neurological findings. Underlying disorders included brainstem or cerebellar infarction (n = 15), cerebellar tumour (n = 5), spinocerebellar ataxia (n = 4), multiple system atrophy (n = 1), and cerebellar haemorrhage (n = 2). The lesions were unilateral in the 22 patients who had a circumscribed lesion (infarction or tumour). For comparison, we studied 20 patients (seven males, 65.0 ± 15.7 years) with spontaneous nystagmus while sitting and supine from apogeotropic BPPV. The diagnosis was based on the following criteria: (i) presence of apogeotropic horizontal nystagmus; (ii) resolution of the vertigo and nystagmus with appropriate canalith repositioning manoeuvres; and (iii) absence of other neurological findings suggesting a central lesion. Evaluation Eye movements were recorded binocularly at a sampling rate of 60 Hz using 3D video-oculography (SensoMotoric Instruments). Digitized eye position data were analysed using MATLAB software (version R2011b, MathWorks, Natick, MA). Spontaneous nystagmus was recorded in the sitting position both with and without visual fixation. Gaze-evoked nystagmus, head-shaking nystagmus, and horizontal saccades and smooth pursuit were also evaluated. The detailed methods and normative data for saccades and smooth pursuit are described elsewhere (Yang et al., 2009). Positional nystagmus was induced without fixation in each position while patients were asked to keep their eyes in the straight-ahead position. After assessing for spontaneous nystagmus while sitting with the head upright, patients lay supine with the head 30° up from the earth horizontal plane. The patient’s head was then turned by about 90° to either side (Lee and Kim, 2010). Each position was maintained for at least 60 s. The intensity of horizontal nystagmus in each position was determined by averaging the slow-phase velocities of 10 consecutive beats when the nystagmus was relatively stable. Patients with CPN had bithermal caloric tests and an evaluation for the ocular tilt reaction. The ocular tilt reaction was considered present when the patients showed any head tilt, abnormal ocular torsion, skew deviation, and deviation of the subjective visual vertical. All additional tests were performed within 7 days of the eye movement recordings. The technical details and normative data of each test have been reported (Choi et al., 2007). Pattern analyses of apogeotropic nystagmus We evaluated the patterns of nystagmus in two ways. First, we analysed apogeotropic nystagmus during left/right-ear-down positions with respect to spontaneous horizontal nystagmus while supine. Thus, we adopted the terms ‘ipsiversive’ and ‘contraversive’. Ipsiversive nystagmus was defined when the beating direction of apogeotropic nystagmus was the same as that of spontaneous horizontal nystagmus while supine. In contrast, contraversive nystagmus indicated that the apogeotropic nystagmus beat in the opposite direction of the spontaneous horizontal nystagmus in the supine position. In patients with no spontaneous nystagmus while supine, the ipsiversive or contraversive nystagmus was determined by the direction of spontaneous nystagmus while sitting. For example, in patients with left beating spontaneous nystagmus while supine, the ipsiversive nystagmus would be observed in the RED position while the contraversive nystagmus was induced in the LED position (Fig. 1B). We determined the direction and intensity (absolute mean slow-phase velocities) of spontaneous horizontal nystagmus in the sitting, supine, and left/right-ear-down positions (Fig. 1C), and compared the intensity between ipsiversive and contraversive apogeotropic nystagmus. Second, we analysed apogeotropic nystagmus according to the lesion side. Thus, 22 patients with CPN and a circumscribed lesion and 20 patients with BPPV were included for analyses after excluding five patients with CPN from the multiple system atrophy and spinocerebellar ataxia cohorts. In this analysis, the term ‘ipsilesional’ was applied when the nystagmus beat toward the lesion side while ‘contralesional’ was adopted when the nystagmus beat toward the intact side. Therefore, in patients with a left-sided lesion, the left beating spontaneous nystagmus while sitting and supine was ‘ipsilesional’. In ear-down position, the left beating nystagmus during RED position was ‘ipsilesional’ apogeotropic nystagmus while the right beating nystagmus during LED position was ‘contralesional’ (Fig. 1B). Then, we determined the patterns of positional nystagmus in relation to the location of the lesion by comparing the intensity of ipsilesional and contralesional nystagmus (Fig. 1C). Positional effect on apogeotropic nystagmus To determine the positional effects on the apogeotropic nystagmus, we measured position-induced nystagmus (PIN) by subtracting the slow-phase velocities (SPVs) of spontaneous horizontal nystagmus while supine from that of apogeotropic nystagmus in each ear-down position using the following equations (Fig. 1C);   PINRED=SPVsofnystagmusonRED−SPVsofnystagmusonsupine (1)  PINLED=SPVsofnystagmusonLED−SPVsofnystagmusonsupine (2) For example, in a patient with left beating nystagmus while supine (slow-phase velocity, 3°/s), left beating nystagmus (10°/s) in the RED position has a 7°/s of PINRED and right beating nystagmus (−2°/s) in the LED position has a −5°/s of PINLED. Statistical analyses Normality of the data was determined using the Shapiro-Wilk test. In each subject, the intensities of ipsiversive (and ipsilesional) and contraversive (contralesional) apogeotropic nystagmus were compared using the Mann-Whitney test with statistical significance level of P < 0.05. Within each group, the Wilcoxon-signed rank test was used to compare the intensity of spontaneous nystagmus while sitting versus supine and ipsiversive versus contraversive apogeotropic nystagmus. The intensity of PIN was also compared between RED and LED positions in each group. Thus, the level of statistical significance for Wilcoxon-signed rank test was corrected using the Bonferroni method, and the corrected level of significance was set at 0.0167 (0.05/3) since the comparisons were performed three times in each group. In addition, by plotting PINLED as a function of PINRED, we determined the positional effects on apogeotropic nystagmus in both apogeotropic CPN and BPPV groups. Brain imaging and lesion analysis All patients with CPN underwent brain MRI according to the protocol described previously (Huh and Kim, 2011). Seventeen patients with an ischaemic or haemorrhagic stroke and four patients with a cerebellar tumour (three with hemangioblastoma and one with meningioma) underwent MRI within 14 days (4.8 ± 5.8 days) from the oculographic recording. One patient with cerebellar hemangioblastoma underwent MRIs 6 months before and 3 months after oculography, and there were no interval changes in the extent of the lesion. Five patients with degenerative disorders (spinocerebellar ataxia and multiple system atrophy) underwent MRI within 3 months before the oculography. We analysed MRI scans of 22 patients with circumscribed lesions. Diffusion-weighted and T2-weighted images were spatially normalized to a stereotaxic space using both linear and non-linear transformations with SPM8 (www.fil.ion.ucl.ac.uk/spm/software/spm8). Using MRIcron, the lesions were overlaid after flipping the right-sided lesions to the left (www.mccauslandcenter.sc.edu/mricro/mricron). Results Clinical features of apogeotropic CPN and BPPV Apogeotropic CPN All patients with apogeotropic CPN reported vertigo provoked or aggravated by positional changes. Twenty-two patients showed limb or truncal ataxia and eight also had sensory or bulbar dysfunction. With visual fixation while sitting, spontaneous horizontal nystagmus was present in only one patient (Patient 22). In contrast, spontaneous horizontal nystagmus was observed in 24 (88.9%) patients without fixation in darkness. Horizontal gaze-evoked nystagmus was present in 10 (37.0%) and downbeat nystagmus was induced after horizontal head shaking in 14 (51.9%) patients. Twenty-three (85.2%) patients showed abnormal saccades including increased latency in 20, hypometria in 16, and hypermetria in five. Horizontal smooth pursuit was impaired in 19 (70.4%) patients. Thus, in summary, all patients with apogeotropic CPN had additional ocular motor or other neurological findings indicative of central lesions. The findings in each patient are summarized in Table 1. Table 1 Findings in the patients with a persistent form of apogeotropic central positional nystagmus ID  Aetiology  Lesion side  Symptoms and signs  SN fix  SN non-fix  GEN  HSN  Pursuit gain  Saccades  OTR  CP  Sp. vertigo  Po. vertigo  Ataxia  Others  10°/s  20°/s  Latency  Gain  1  Cbll hemangioblastoma  L  −  +  −  −  −  L  −  D  −  ↓ (L)  Del (R)  ↓ (B)  R/Contralesional  −  2  Cbll hemangioblastoma  L  −  +  −  −  −  L  +  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  L/Ipsilesional    3  Cbll infarction (PICA)  L  +  +  −  −  −  L  −  D  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  4  Cbll infarction (PICA)  L  −  +  −  −  −  L  −  −  −  −  Del (B)  −  −    5  Cbll infarction (SCA)  L  +  +  Li/Tr  −  −  L  −  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  6  Cbll infarction (PICA)  L  −  +  Tr  −  −  L  +  R  ↓ (R)  ↓ (R)  Del (L)  ↓ (R)  R/Contralesional  −  7  Cbll infarction (PICA)  L  +  +  Li/Tr  Sen / Bul  −  L  −  D  −  −  Del (B)  ↓ (L)  R/Contralesional    8  Cbll infarction (PICA)  L  +  +  Tr  −  −  L  −  D  −  −  −  ↓ (B)  R/Contralesional  −  9  Cbll hemangioblastoma  L  +  +  Li/Tr  Bul  −  L/D  −  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  10  Cbll meningioma  L  −  +  Tr    −  L/D  −  R/D  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  R/Contralesional  −  11  Cbll haemorrhage  L  +  +  Tr  −  −  L/U  −  R/D  ↓ (R)  ↓ (R)  −  −  R/Contralesional  −  12  Cbll infarction (PICA)  L  +  +  Tr  −  −  −  −  D  ↓ (B)  ↓ (B)  −  ↓ (B)  R/Contralesional  −  13  Cbll infarction (PICA) + LMI  L  +  +  Li/Tr  −  −  R  +  D  ↓ (B)  ↓ (B)  −  ↓ (L)  R/Contralesional  −  14  PICA infarction  R  +  +  −  −  −  R  −  −  ↓ (B)  ↓ (B)  Del (R)  ↓ (R)  −  −  15  Cbll haemorrhage  R  +  +  Li/Tr  −  U  R  +  U  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  L/Contralesional  −  16  Cbll infarction (SCA)  R  +  +  Li/Tr  −  −  R  −  −  ↓ (R)  ↓ (R)  Del (B)  ↓ (L)  L/Contralesional  −  17  Cbll hemangioblastoma  R  +  +  Li/Tr  −  −  R  −  D          −    18  Cbll infarction (PICA)  R  +  +  Li/Tr  −  −  R  −  −          L/Contralesional  −  19  Cbll infarction (PICA)  R  +  +  Li/Tr  Bul  U  R/U  +  L/D  ↓ (R)  ↓ (R)  Del (R)  ↓ (L)  L/Contralesional  −  20  Cbll infarction (PICA)  R  +  +  Tr  −  −  R/U  +  −  −  −  Del (L)  ↓ (L)  −  −  21  Cbll infarction (PICA)  R  +  +  Li/Tr  −  −  R/D  −  R          L/Contralesional    22  Cbll infarction (PICA) + LMI  R  +  +  Li/Tr  Sen / Bul  L  L/U  −  R/D  ↓ (L)  ↓ (L)  Del (B)  −  R/Ipsilesional  −  23  Spinocerebellar ataxia type 6  Ud  +  +  Li/Tr  Bul  D  L/D  +  R/D  ↓ (B)  ↓ (B)  Del (B)  ↓ (R)  −  −  24  Spinocerebellar ataxia type 6  Ud  −  +  Li/Tr  Bul  −  −  +  L  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  −    25  Spinocerebellar ataxia type 6  Ud  −  +  Li/Tr  Bul  −  −  +  −  −  ↓ (B)  Del (B)  ↓ (L)  −    26  Multisystem atrophy  Ud  −  +  Tr  −  −  R  −  R  −  ↓ (B)  Del (B)  (L) ↑  −    27  Spinocerebellar ataxia type 6  Ud  +  +  Li/Tr  Bul  −  R  +  L/D  −  −  Del (B)  (L) ↑  R    ID  Aetiology  Lesion side  Symptoms and signs  SN fix  SN non-fix  GEN  HSN  Pursuit gain  Saccades  OTR  CP  Sp. vertigo  Po. vertigo  Ataxia  Others  10°/s  20°/s  Latency  Gain  1  Cbll hemangioblastoma  L  −  +  −  −  −  L  −  D  −  ↓ (L)  Del (R)  ↓ (B)  R/Contralesional  −  2  Cbll hemangioblastoma  L  −  +  −  −  −  L  +  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  L/Ipsilesional    3  Cbll infarction (PICA)  L  +  +  −  −  −  L  −  D  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  4  Cbll infarction (PICA)  L  −  +  −  −  −  L  −  −  −  −  Del (B)  −  −    5  Cbll infarction (SCA)  L  +  +  Li/Tr  −  −  L  −  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  6  Cbll infarction (PICA)  L  −  +  Tr  −  −  L  +  R  ↓ (R)  ↓ (R)  Del (L)  ↓ (R)  R/Contralesional  −  7  Cbll infarction (PICA)  L  +  +  Li/Tr  Sen / Bul  −  L  −  D  −  −  Del (B)  ↓ (L)  R/Contralesional    8  Cbll infarction (PICA)  L  +  +  Tr  −  −  L  −  D  −  −  −  ↓ (B)  R/Contralesional  −  9  Cbll hemangioblastoma  L  +  +  Li/Tr  Bul  −  L/D  −  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  10  Cbll meningioma  L  −  +  Tr    −  L/D  −  R/D  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  R/Contralesional  −  11  Cbll haemorrhage  L  +  +  Tr  −  −  L/U  −  R/D  ↓ (R)  ↓ (R)  −  −  R/Contralesional  −  12  Cbll infarction (PICA)  L  +  +  Tr  −  −  −  −  D  ↓ (B)  ↓ (B)  −  ↓ (B)  R/Contralesional  −  13  Cbll infarction (PICA) + LMI  L  +  +  Li/Tr  −  −  R  +  D  ↓ (B)  ↓ (B)  −  ↓ (L)  R/Contralesional  −  14  PICA infarction  R  +  +  −  −  −  R  −  −  ↓ (B)  ↓ (B)  Del (R)  ↓ (R)  −  −  15  Cbll haemorrhage  R  +  +  Li/Tr  −  U  R  +  U  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  L/Contralesional  −  16  Cbll infarction (SCA)  R  +  +  Li/Tr  −  −  R  −  −  ↓ (R)  ↓ (R)  Del (B)  ↓ (L)  L/Contralesional  −  17  Cbll hemangioblastoma  R  +  +  Li/Tr  −  −  R  −  D          −    18  Cbll infarction (PICA)  R  +  +  Li/Tr  −  −  R  −  −          L/Contralesional  −  19  Cbll infarction (PICA)  R  +  +  Li/Tr  Bul  U  R/U  +  L/D  ↓ (R)  ↓ (R)  Del (R)  ↓ (L)  L/Contralesional  −  20  Cbll infarction (PICA)  R  +  +  Tr  −  −  R/U  +  −  −  −  Del (L)  ↓ (L)  −  −  21  Cbll infarction (PICA)  R  +  +  Li/Tr  −  −  R/D  −  R          L/Contralesional    22  Cbll infarction (PICA) + LMI  R  +  +  Li/Tr  Sen / Bul  L  L/U  −  R/D  ↓ (L)  ↓ (L)  Del (B)  −  R/Ipsilesional  −  23  Spinocerebellar ataxia type 6  Ud  +  +  Li/Tr  Bul  D  L/D  +  R/D  ↓ (B)  ↓ (B)  Del (B)  ↓ (R)  −  −  24  Spinocerebellar ataxia type 6  Ud  −  +  Li/Tr  Bul  −  −  +  L  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  −    25  Spinocerebellar ataxia type 6  Ud  −  +  Li/Tr  Bul  −  −  +  −  −  ↓ (B)  Del (B)  ↓ (L)  −    26  Multisystem atrophy  Ud  −  +  Tr  −  −  R  −  R  −  ↓ (B)  Del (B)  (L) ↑  −    27  Spinocerebellar ataxia type 6  Ud  +  +  Li/Tr  Bul  −  R  +  L/D  −  −  Del (B)  (L) ↑  R    B = both-direction; Bul = bulbar sign; Cbll = cerebellar; CP = caloric paresis; D = downbeat; GEN = gaze-evoked nystagmus; HSN = head-shaking induced nystagmus; L = left side, left beating, and left-ward; Li = limb; LMI = lateral medullary infarction; OTR = ocular tilt reaction; PICA = posterior inferior cerebellar artery territory; Po = positional; R = right side, right beating, and right-ward; SCA = superior cerebellar artery territory; Sen = sensory sign; SN fix = spontaneous nystagmus with visual fixation; SN non-fix = SN without visual fixation; Sp = spontaneous; Tr = truncal; U = upbeat; Ud = undetermined lesion side; + = present; − = normal or absent; ↓ = hypometria and decreased gain; ↑ = hypermetria; Blank cell indicates that the examination was not performed. Apogeotropic BPPV All patients with apogeotropic BPPV showed horizontal nystagmus while sitting (‘pseudo-spontaneous nystagmus’), which beat to the affected side in 15 and to the intact side in five. In all patients, the apogeotropic HC-BPPV resolved; in 11 with the reversed-Gufoni manoeuvre and in nine after being converted into the geotropic type that was then successfully treated with the Gufoni or barbecue manoeuvre (Supplementary Table 1). Patterns of positional nystagmus Apogeotropic CPN in relation to the spontaneous nystagmus while supine In darkness, 24 of the 27 patients in the CPN group showed horizontal nystagmus in the sitting position while 22 patients showed horizontal nystagmus in the supine position. The modulation of spontaneous nystagmus between the positions was presented in Table 2. Table 2 The intensity of spontaneous nystagmus in the sitting and supine positions and the intensity of apogeotropic positional nystagmus according to the direction of spontaneous nystagmus while supine Presence of SN in the sitting and supine positions in the apogeotropic CPN group    SN (+)  SN (−)  SN (n)  Sitting  24  3  24    SN (+)  SN (−)  SN (+)  SN (−)  SN (n)  Supine  19  5  3  0  22    The intensity of spontaneous horizontal nystagmus in the sitting and supine positions    Sitting  Supine  P    Apogeotropic CPN group  1.6 (0.8–2.9)  0.8 (0.5–1.9)  0.16  Apogeotropic BPPV group  1.4 (0–2.3)  6.0 (3.5–11.1)  < 0.001    The intensity of apogeotropic nystagmus according to the direction of SN while supine    Ipsiversive  Contraversive  P    Apogeotropic CPN group  5.3 (3.8–8.3)  3.1 (2.2–5.5)  <0.001  Apogeotropic BPPV group  22.6 (13.4–46.5)  7.6 (5.0–13.3)  <0.001  Presence of SN in the sitting and supine positions in the apogeotropic CPN group    SN (+)  SN (−)  SN (n)  Sitting  24  3  24    SN (+)  SN (−)  SN (+)  SN (−)  SN (n)  Supine  19  5  3  0  22    The intensity of spontaneous horizontal nystagmus in the sitting and supine positions    Sitting  Supine  P    Apogeotropic CPN group  1.6 (0.8–2.9)  0.8 (0.5–1.9)  0.16  Apogeotropic BPPV group  1.4 (0–2.3)  6.0 (3.5–11.1)  < 0.001    The intensity of apogeotropic nystagmus according to the direction of SN while supine    Ipsiversive  Contraversive  P    Apogeotropic CPN group  5.3 (3.8–8.3)  3.1 (2.2–5.5)  <0.001  Apogeotropic BPPV group  22.6 (13.4–46.5)  7.6 (5.0–13.3)  <0.001  Statistical analysis was performed using the Wilcoxon-signed rank test and the intensity of nystagmus is shown as the median (IQR). SN = spontaneous nystagmus; SN(n): number of patients with spontaneous nystagmus; SN (+) = presence of SN; SN (−) = absence of SN. The intensity of horizontal nystagmus did not differ between the sitting and supine positions in the CPN group [median (interquartile range, IQR), 1.6 (0.8–2.9) versus 0.8 (0.5–1.9), Wilcoxon-signed rank test, P = 0.160] (Fig. 2A and Table 2). Figure 2 View largeDownload slide Modulation of nystagmus in CPN and BPPV. (A) Spontaneous horizontal nystagmus in sitting and supine. While the intensities of spontaneous nystagmus did not differ between the sitting and supine positions in apogeotropic CPN, the nystagmus markedly increased in the supine position in patients with apogeotropic BPPV. (B) Apogeotropic nystagmus according to the direction of spontaneous nystagmus while supine. In both CPN and BPPV groups, the ipsiversive (nystagmus in the direction of spontaneous nystagmus while supine) apogeotropic nystagmus was greater than the contraversive one. The thin lines indicate individual data while the thick lines denote the median value. (C and D) PIN in CPN and BPPV groups. The CPN group (C) shows symmetrical PIN in the ear-down position to either side (Spearman’s correlation r = −0.97, P < 0.001). In contrast, the BPPV group (D) exhibits a marked asymmetry of the PIN (Spearman’s correlation r = −0.48, P = 0.034). Note that there are 4-fold differences in the magnitude scales between panels C and D. Figure 2 View largeDownload slide Modulation of nystagmus in CPN and BPPV. (A) Spontaneous horizontal nystagmus in sitting and supine. While the intensities of spontaneous nystagmus did not differ between the sitting and supine positions in apogeotropic CPN, the nystagmus markedly increased in the supine position in patients with apogeotropic BPPV. (B) Apogeotropic nystagmus according to the direction of spontaneous nystagmus while supine. In both CPN and BPPV groups, the ipsiversive (nystagmus in the direction of spontaneous nystagmus while supine) apogeotropic nystagmus was greater than the contraversive one. The thin lines indicate individual data while the thick lines denote the median value. (C and D) PIN in CPN and BPPV groups. The CPN group (C) shows symmetrical PIN in the ear-down position to either side (Spearman’s correlation r = −0.97, P < 0.001). In contrast, the BPPV group (D) exhibits a marked asymmetry of the PIN (Spearman’s correlation r = −0.48, P = 0.034). Note that there are 4-fold differences in the magnitude scales between panels C and D. During the ear-down positions, the ipsiversive nystagmus was greater than the contraversive one [5.3 (3.8–8.3) versus 3.1 (2.2–5.5), Wilcoxon-signed rank test, P < 0.001] (Fig. 2B and Table 2). Furthermore, all 22 patients with spontaneous horizontal nystagmus while supine showed greater ipsiversive than contraversive nystagmus (Mann-Whitney test, P < 0.05). In contrast, the remaining five patients without horizontal nystagmus while supine (Patients 5, 10, 18, 21, and 27) showed nearly symmetrical apogeotropic nystagmus between the right and left ear-down positions (Mann-Whitney test, P > 0.05). The intensity and direction of positional nystagmus in patients with CPN are summarized (Supplementary Table 1). Apogeotropic CPN according to the lesion side Of the 22 patients with CPN and circumscribed MRI lesions, 13 had a left-sided and nine had a right-sided lesion. The lesions were limited to the cerebellum in all patients except two (Patients 13 and 22) who also had lesions in the lateral medulla. The spontaneous horizontal nystagmus usually beat to the lesion side while sitting (86.4%) and supine (68.2%). Fifteen (68.2%) patients showed ipsilesional spontaneous nystagmus while supine, and the ipsilesional (ipsiversive, during contralesional head turn) apogeotropic nystagmus was greater than the contralesional nystagmus (Fig. 3A). Likewise, in three patients with contralesionally beating nystagmus while supine, the contralesional (ipsiversive, during ipsilesional head turn) apogeotropic nystagmus was greater than the ipsilesional one (Fig. 3B). In contrast, four patients without spontaneous nystagmus while supine developed nearly symmetrical apogeotropic nystagmus with the head turned to either side. The patterns of apogeotropic positional nystagmus are summarized according to the lesion side (Table 3). Table 3 Patterns of apogeotropic nystagmus in the CPN and BPPV groups   CPN group with circumscribed MRI lesion (n = 22)  BPPV group (n = 20)  Patterns of SN in the supine positions according to the lesion side    Ipsilesional  Contralesional  None  Ipsilesional  Supine  15  3  4  20  Patterns of apogeotropic nystagmus relative to the lesion side  Ear-down  Ipsilesional N > Contralesional N  Ipsilesional N < Contralesional N  Ipsilesional N = Contralesional N  Ipsilesional N > Contralesional N  Patterns of apogeotropic nystagmus relative to the direction of spontaneous nystagmus while supinea  Ear-down  Ipsiversive N > Contraversive N  Ipsiversive N > Contraversive N  Ipsiversive N = Contraversive N  Ipsiversive N > Contraversive N    CPN group with circumscribed MRI lesion (n = 22)  BPPV group (n = 20)  Patterns of SN in the supine positions according to the lesion side    Ipsilesional  Contralesional  None  Ipsilesional  Supine  15  3  4  20  Patterns of apogeotropic nystagmus relative to the lesion side  Ear-down  Ipsilesional N > Contralesional N  Ipsilesional N < Contralesional N  Ipsilesional N = Contralesional N  Ipsilesional N > Contralesional N  Patterns of apogeotropic nystagmus relative to the direction of spontaneous nystagmus while supinea  Ear-down  Ipsiversive N > Contraversive N  Ipsiversive N > Contraversive N  Ipsiversive N = Contraversive N  Ipsiversive N > Contraversive N  aNote that apogeotropic nystagmus appears to be more intense in the direction of spontaneous nystagmus while supine in patients with CPN irrespective of the lesion side. When the spontaneous nystagmus while supine was absent, the ipsiversive or contraversive nystagmus was determined according to the direction of spontaneous nystagmus while sitting. Contralesional N = contralesional beating nystagmus during ipsilesional head turn; Ipsilesional N = ipsilesional beating nystagmus during contralesional head turn; SN = spontaneous nystagmus. Figure 3 View largeDownload slide Illustration of spontaneous and apogeotropic nystagmus in CPN group and overlay lesion plots in 22 patients with apogeotropic CPN from unilateral lesions. (A) A patient (Patient 6) with a left-sided nodulus lesion shows ipsilesional left beating nystagmus while supine and apogeotropic nystagmus in the ear-down position to either side with the ipsiversive left beating positional nystagmus greater than the contraversive right beating one. (B) Another patient (Patient 22) with infarctions involving right lateral medulla and cerebellum shows contralesional left beating nystagmus while supine and greater ipsiversive nystagmus in right ear-down position. In each recording, upward deflection indicated rightward eye motion. (C) A patient with apogeotropic BPPV involving the left horizontal semicircular canal shows ipsilesional left beating nystagmus while supine and apogeotropic nystagmus during head turn to either side while supine. The ipsiversive left beating nystagmus in the right ear-down position was stronger than the contraversive right beating nystagmus in the left ear-down position. (D) The lesions are mostly overlapped (red colour) in the nodulus (X), uvula (IX), and tonsil (H IX) on the spatially unbiased atlas template of the cerebellum and brainstem (SUIT, ver. 2.5.3). The numbers of overlapping lesions are illustrated by different colours from violet (n = 1) to red (n = 12). (E) Illustration of the areas corresponding to the nodulus, uvula, and tonsil in two representative templates of the SUIT. LH = horizontal position of the left eye. SPV = slow-phase velocity. Figure 3 View largeDownload slide Illustration of spontaneous and apogeotropic nystagmus in CPN group and overlay lesion plots in 22 patients with apogeotropic CPN from unilateral lesions. (A) A patient (Patient 6) with a left-sided nodulus lesion shows ipsilesional left beating nystagmus while supine and apogeotropic nystagmus in the ear-down position to either side with the ipsiversive left beating positional nystagmus greater than the contraversive right beating one. (B) Another patient (Patient 22) with infarctions involving right lateral medulla and cerebellum shows contralesional left beating nystagmus while supine and greater ipsiversive nystagmus in right ear-down position. In each recording, upward deflection indicated rightward eye motion. (C) A patient with apogeotropic BPPV involving the left horizontal semicircular canal shows ipsilesional left beating nystagmus while supine and apogeotropic nystagmus during head turn to either side while supine. The ipsiversive left beating nystagmus in the right ear-down position was stronger than the contraversive right beating nystagmus in the left ear-down position. (D) The lesions are mostly overlapped (red colour) in the nodulus (X), uvula (IX), and tonsil (H IX) on the spatially unbiased atlas template of the cerebellum and brainstem (SUIT, ver. 2.5.3). The numbers of overlapping lesions are illustrated by different colours from violet (n = 1) to red (n = 12). (E) Illustration of the areas corresponding to the nodulus, uvula, and tonsil in two representative templates of the SUIT. LH = horizontal position of the left eye. SPV = slow-phase velocity. Apogeotropic BPPV All patients in the BPPV group showed ipsilesional-beating horizontal nystagmus while supine (‘lying-down nystagmus’). In contrast to the nystagmus in the CPN group, the horizontal nystagmus was greater in the supine than in the sitting position [median (IQR) 6.0 (3.5–11.1) versus 1.4 (0–2.3), Wilcoxon-signed rank test, P < 0.001] (Fig. 2A and Table 2). During the ear-down positions, the BPPV group also showed the ipsiversive (and ipsilesional, during contralesional head turn) nystagmus greater than the contraversive (and contralesional, during ipsilesional head turn) nystagmus [22.6 (13.4–46.5) versus 7.6 (5.0–13.3), Wilcoxon-signed rank test, P < 0.001] (Figs 2B, 3C, and Table 2). The intensities and directions of positional nystagmus in patients with BPPV are also summarized (Supplementary Table 1). Positional effects on apogeotropic CPN and BPPV The intensity of PIN in patients with CPN was 4.4 (3.2–6.1) and 4.6 (3.0–6.4) in RED and LED positions, respectively. Meanwhile the BPPV group showed 13.1 (8.0–25.8) and 17.9 (11.0–33.6) PIN intensity in RED and LED positions. In both CPN and BPPV groups, there was no difference in the intensity of PIN between the RED and LED positions (Wilcoxon-signed rank test, P > 0.05). PINLED and PINRED in the CPN group, were significantly correlated (Spearman’s correlation r = −0.97, P < 0.001), indicating the PIN was nearly symmetrical in individual patients with CPN (Fig. 2C). However, this was not the case for the BPPV group (Spearman’s correlation r = −0.48, P = 0.034); there was a marked asymmetry of PIN in individual patients with apogeotropic BPPV (Fig. 2D). Lesion analysis When the lesions were overlapped on the SUIT (spatially unbiased atlas template of the cerebellum and brainstem) (Diedrichsen, 2006), the overlay density plots showed that the nodulus (lobule X), uvula (lobule IX), and tonsil (lobule H IX) were most frequently involved in the 22 patients with apogeotropic CPN from circumscribed lesions (Fig. 3D and E) (Schmahmann et al., 1999; Stoodley and Schmahmann, 2009). Modelling Estimated gravitation-induced nystagmus in patients with apogeotropic CPN In apogeotropic CPN, the PIN was symmetric with opposing directions during head turn to either side while supine. Given that the orientation of the head relative to gravity is the only difference between the RED and LED positions (Fig. 1A), we assume that the PIN was from the effects of gravity. Our assumption was in accord with the results of a previous animal study in which injection of muscimol into the nodulus and uvula of a monkey produced apogeotropic positional nystagmus with a linear relationship between the intensity of the nystagmus and the gravitational orientation of the head (Sheliga et al., 1999). This predicts that the intensities of gravity-induced nystagmus (GIN) should be similar between the right and left ear-down positions. Thus, we may estimate the gravity-induced nystagmus in the LED, supine, and RED positions based on the previous assumptions: (i) the contribution of the canal signals (C) to the nystagmus is independent of static head orientation (Zupan et al., 2000); and (ii) the intensities of gravity-induced nystagmus (GINs) in RED and LED are the same but in the opposing directions (GINRED = −GINLED). Thus, the slow-phase velocities (SPVs) of apogeotropic nystagmus during both ear-down positions can be represented as follows:   SPVsinRED=C+GINRED (3)  SPVsinLED=C+GINLED=C−GINRED (4) We can calculate the canal contribution (C) to the nystagmus and the gravity-induced nystagmus during ear-down positions by adding and subtracting the two equations:   C=[Equation3+Equation4]/2 (5)  GINRED=[Equation3−Equation4]/2 (6) For example, in a patient with left beating nystagmus (slow-phase velocity, 8°/s) in the RED position, and right beating nystagmus (−2°/s) in the LED position, GINRED is 5°/s and C is 3°/s. We can also calculate the gravity-induced nystagmus while supine by subtracting the slow-phase velocity of C from that of spontaneous nystagmus while supine. The intensity of C was 0.9 (0.5–1.7), and the intensity of gravity-induced nystagmus during each ear-down position was 4.6 (3.1–6.2) in the CPN group. The estimated gravity-induced nystagmus according to head position is shown in Fig. 4A. In the supine position, the gravity-induced nystagmus was 0.1 (0–0.2), and these findings could be simulated by our model. Figure 4 View largeDownload slide Gravity-induced nystagmus and apogeotropic CPN model. (A) Estimated gravity-induced nystagmus (GIN) in both ear-down and supine positions are plotted in patients with apogeotropic CPN. The thick horizontal line and the top and bottom boundaries of the box represent the mean of GIN with ± 2 standard deviations at each position. Positive value indicates rightward slow-phase velocities (SPVs). (B) A schematic model modified from Laurens and Angelaki (2011). The red lightning symbol is the assumed site of the lesion within the model. (C) In the normal condition, the estimated gravity direction in the ear-down position corresponds to the actual gravity direction. In patients with CPN, the estimated gravity direction is biased toward the nose along the nasooccipital axis of the head while in ear-down positions. The persistent directional mismatch between the actual and estimated gravities can cause the rotational cue (feedback) to bring the estimate of gravity toward the actual one. An empty circle is a summation point. g = estimated gravity; GIA = gravitoinertial acceleration; i = inertia; w = estimated angular velocity; x and ∫ are mathematical terms representing a vector cross product and integral. Figure 4 View largeDownload slide Gravity-induced nystagmus and apogeotropic CPN model. (A) Estimated gravity-induced nystagmus (GIN) in both ear-down and supine positions are plotted in patients with apogeotropic CPN. The thick horizontal line and the top and bottom boundaries of the box represent the mean of GIN with ± 2 standard deviations at each position. Positive value indicates rightward slow-phase velocities (SPVs). (B) A schematic model modified from Laurens and Angelaki (2011). The red lightning symbol is the assumed site of the lesion within the model. (C) In the normal condition, the estimated gravity direction in the ear-down position corresponds to the actual gravity direction. In patients with CPN, the estimated gravity direction is biased toward the nose along the nasooccipital axis of the head while in ear-down positions. The persistent directional mismatch between the actual and estimated gravities can cause the rotational cue (feedback) to bring the estimate of gravity toward the actual one. An empty circle is a summation point. g = estimated gravity; GIA = gravitoinertial acceleration; i = inertia; w = estimated angular velocity; x and ∫ are mathematical terms representing a vector cross product and integral. Model simulation The influence of gravity on eye position and velocity has been explained by two mechanisms. First, changes in the direction of the pull of gravity relative to the head, as with head tilts, changes primary eye position (i.e. an alteration of the orientation of Listing’s plane), though this mechanism would be expected to produce only a tiny positional horizontal nystagmus (Glasauer et al., 2001). Second, the direction of gravity affects the vestibulo-ocular reflex via the velocity-storage mechanism (VSM) (Raphan and Cohen, 1985; Yakusheva et al., 2007). The VSM is part of a central circuit that improves the low-frequency response of the rotational vestibulo-ocular reflex, and helps provide an internal estimate of the direction of gravity relative to the head. The latter signal is used to resolve the ambiguity between static head tilt relative to upright and translation of the head (Glasauer, 1992; Laurens and Angelaki, 2011; Laurens et al., 2013). For better understanding, we briefly describe here how the proposed model makes this distinction (Fig. 4B) (Laurens and Angelaki, 2011; Laurens et al., 2013). The otolith organs provide information on the gravito-inertial acceleration (GIA), a sum of gravitational acceleration (g) and translation-driven inertial acceleration (i) but they do not discriminate between the two. Thus, the otolith-driven vectors of gravito-inertial acceleration can be identical during tilt and translation so it is the VSM that enables a discrimination of tilt from translation. From the VSM, the brain obtains information about the position of the head related to gravity (tilt estimator) by integrating the cross product of canal-driven head velocity signals and gravity (∫ gravity × head velocity). Then, by subtracting the head tilt from the gravito-inertial acceleration (i = GIA − g), an appropriate translational cue is extracted to generate command for the translational vestibulo-ocular reflex. There are two feedback loops within the VSM model; one is the somatogravic loop that explains the faulty perception of tilt during low-frequency translation (somatogravic illusion) (Angelaki, 1998; Merfeld et al., 2005). The other is the rotational feedback loop, which enhances the accuracy of the tilt estimator. Because the VSM estimates head tilt by integrating head velocity, any inaccuracy of canal signals would cause a bias for head tilt. The directional mismatch between the actual gravity and the estimated one causes an erroneous cue for translation. To minimize this, the rotational feedback compensates for the directional mismatch by providing a corrective rotational cue. In normal subjects this model predicts that sustained nystagmus can be generated without direct canal stimulation. For example, a sustained nystagmus is elicited during constant-velocity, off-vertical axis rotations in normal subjects even though the signal from the canals has decayed away (Angelaki and Hess, 1996; Laurens et al., 2010). Therefore, to simulate sustained apogeotropic CPN, which also occurs after stimulation of the canals has ceased, we implemented a model of VSM in MATLAB/Simulink (The MathWorks, Natick, MA) and applied several different hypothetical lesion sites. The model simulation produced gravity-related, symmetric apogeotropic nystagmus in both ear-down positions assuming the lesion involves the vestibulocerebellar pathway that relays the estimated gravity to rotational feedback (Fig. 4B, red lightning symbol). We assumed that disruption of this pathway results in loss of information about the estimated gravity and also causes a small bias (0.02g) toward the nose along the naso-occipital axis of the head. Thus, in the right ear-down position, the actual gravity points toward the right ear while the estimated gravity is directed between the right ear and nose (Fig. 4C). Such a lesion generates a horizontal apogeotropic nystagmus of ∼6.4°/s) (Fig. 5). Note that neither the main VSM in the brainstem nor the direct otolith-ocular pathways have to be damaged to generate apogeotropic nystagmus. Further information on modelling is presented in the Supplementary material. Figure 5 View largeDownload slide Results of model simulation. A right-handed, head-fixed coordinate system is used (z-axis pointing upward, x-axis pointing forward, and y-axis pointing leftward). Gravity points downward in the upright head position. (A) The actual head angular velocity (w) and gravity orientation (g) during positioning from sitting to right-ear-down position. The inserted cartoons indicate actual head position. In the model, the head starts from upright and tilts back 90° at 5 s, and it finally reaches right-ear-down position at 20 s. The simulation is then continued for a complete duration of 70 s. (B) The simulation results show an estimated angular velocity (upper row) and a gravity orientation (lower row) in healthy subjects (left) and patients with CPN (right). (C) The estimated angular velocity along the rostro-caudal axis during right ear-down position while supine. After reaching the final position, the angular head velocity remains zero in healthy subjects (blue). In patients with CPN (red), the lesion-induced tilt toward the nose of the estimate of gravity induces a persistent estimate of angular head velocity, which in turn produces rightward slow-phase eye velocity of 6.2°/s and left-beating nystagmus (apogeotropic nystagmus). Figure 5 View largeDownload slide Results of model simulation. A right-handed, head-fixed coordinate system is used (z-axis pointing upward, x-axis pointing forward, and y-axis pointing leftward). Gravity points downward in the upright head position. (A) The actual head angular velocity (w) and gravity orientation (g) during positioning from sitting to right-ear-down position. The inserted cartoons indicate actual head position. In the model, the head starts from upright and tilts back 90° at 5 s, and it finally reaches right-ear-down position at 20 s. The simulation is then continued for a complete duration of 70 s. (B) The simulation results show an estimated angular velocity (upper row) and a gravity orientation (lower row) in healthy subjects (left) and patients with CPN (right). (C) The estimated angular velocity along the rostro-caudal axis during right ear-down position while supine. After reaching the final position, the angular head velocity remains zero in healthy subjects (blue). In patients with CPN (red), the lesion-induced tilt toward the nose of the estimate of gravity induces a persistent estimate of angular head velocity, which in turn produces rightward slow-phase eye velocity of 6.2°/s and left-beating nystagmus (apogeotropic nystagmus). Discussion This is the first study to systematically evaluate apogeotropic APN from two perspectives. First, we show a way to differentiate apogeotropic CPN from apogeotropic BPPV based on the patterns of nystagmus and accompanying neurological signs. Second, we provide a mechanism for apogeotropic CPN using a model of the VSM that suggests how the brain both maintains gravity orientation and generates an appropriate translational vestibulo-ocular reflex in response to an inertial linear acceleration. Clinical findings of apogeotropic CPN All patients with CPN suffered from vertigo induced or aggravated by positional changes. Most of our patients with apogeotropic CPN also showed neurological signs indicating a central pathology, which included gaze-evoked nystagmus (Cnyrim et al., 2008; Kattah et al., 2009), perverted (cross-coupled) head-shaking nystagmus (Huh and Kim, 2011; Choi et al., 2016), and hypermetric saccades (Waespe and Wichmann, 1990), in addition to limb or truncal ataxia. However, dizziness/vertigo was the only symptom in five patients (18.5%) making differentiation between apogeotropic CPN and BPPV difficult in clinical practice. How do the patterns of positional nystagmus give a clue to differentiate CPN from BPPV? Pattern of apogeotropic nystagmus in CPN and BPPV Generally, in CPN patients with circumscribed lesions, the spontaneous horizontal nystagmus is ipsilesional while supine and the ipsilesional apogeotropic nystagmus is more intense when the head is turned to the contralesional side (Fig. 3A). This is the pattern reported previously (Sheliga et al., 1999; Nam et al., 2009; Kim et al., 2012). Of interest, however, two patients (Patients 13 and 22) with contralesional spontaneous nystagmus from lesions involving both the inferior cerebellum and lateral medulla showed contralesional (but ipsiversive) apogeotropic nystagmus greater than the ipsilesional one (Fig. 3B). Taken together, apogeotropic nystagmus is more intense in the direction of spontaneous nystagmus while supine in patients with CPN irrespective of the lesion side. Since apogeotropic BPPV also shows asymmetric apogeotropic nystagmus and ipsiversive greater than contraversive nystagmus, an asymmetric pattern of apogeotropic nystagmus during the supine head roll test alone does not permit differentiation of the central from the peripheral form of apogeotropic nystagmus. On the other hand, the spontaneous horizontal nystagmus differed little between the sitting and supine positions in the CPN group, while the nystagmus usually became considerably greater in the supine than in the sitting position in the BPPV group. Thus, augmentation of spontaneous nystagmus while supine favors the diagnosis of apogeotropic HC-BPPV. In apogeotropic CPN, the similar intensity of the nystagmus between the sitting and supine positions suggests that changes in the gravity orientation from sitting to supine do not affect the horizontal nystagmus. In contrast, in apogeotropic BPPV, the increment of spontaneous horizontal nystagmus in the supine position may be explained by the change in the orientation relative to gravity of the cupula, laden with otolithic debris, within the lateral semicircular canal (Koo et al., 2006; Lee et al., 2007). Pathophysiological mechanism of apogeotropic CPN based on the lesions and modelling Our analysis suggested that apogeotropic CPN can result from summation of incorrectly interpreted canal-induced nystagmus (C) and gravity-induced nystagmus in both ear-down positions. We tested this idea and how it relates to vestibulo-cerebellar lesions using a model of the VSMs. Under normal circumstances, the brain appropriately estimates the direction of gravity and there is no directional mismatch between the actual and estimated gravity during the ear-down positions (Fig. 4C). Therefore, the horizontal eye velocity would be zero during ear-down position in healthy subjects. However, as shown in the constant off-vertical axis rotation paradigm (OVAR), a directional mismatch between the actual gravity provided by the otolith organs and the internally-estimated gravity could generate a compensatory rotational feedback signal that brings the estimated gravity direction into a veridical one (Laurens and Angelaki, 2011). In our VSM model, if a lesion disrupts the pathway providing the estimated gravity to the rotational feedback loop and also produces a positive bias toward the nose along with nasooccipital axis (x = 0.02g), the compensatory rotational feedback would generate constant horizontal apogeotropic gravity-induced nystagmus (Fig. 5). If there were a negative bias, away from the nose along the nasooccipital axis, the compensatory rotational feedback would generate a constant horizontal geotropic gravity-induced nystagmus. In this study, patients with apogeotropic CPN showed lesions that mostly overlapped in the vestibulocerebellum including the nodulus, uvula, and tonsil. The nodulus and uvula may function as the ‘tilt estimator’ in humans (Lee et al., 2017), as was suggested in monkeys (Angelaki and Hess, 1995). And the Purkinje cells in the nodulus and uvula correlate with translational (inertial) inputs (Yakusheva et al., 2007). Complete resection of the nodulus and uvula in monkeys causes a loss of post-rotational tilt suppression (Waespe et al., 1985; Wearne et al., 1998), which also has been attributed to the rotational feedback mechanism in the VSM model (Laurens and Angelaki, 2011). All these functions are compatible with the anatomical characteristics of the nodulus and uvula. It was shown that ∼70% of the primary vestibular afferents project to the nodulus and uvula. Especially those from the canals predominantly synapse with the nodulus while those from the otolithic organs mainly connect with the ventral uvula. The nodulus and uvula also have reciprocal connections with the neurons in the vestibular nucleus (Büttner-Ennever, 1999; Voogd and Barmack, 2006). The correlation of hypometabolism of the nodulus with the degree of subjective visual vertical tilt in patients with acute vestibular neuritis also supports these functional connections (Alessandrini et al., 2014). However, we assume that the tilt estimators need not be completely damaged for generation of CPN since in monkeys, partial resection of the nodulus and uvula preserve post-rotational tilt suppression (Wearne et al., 1998) and apogeotropic nystagmus occurs in monkeys with only unilateral chemical inactivation of the nodulus/uvula (Sheliga et al., 1999). Even in our patients with lesions involving the nodulus and uvula, the lesions were mostly unilateral (partial). In patients with lesions sparing the nodulus/uvula (such as Patient 22), we assume that the neural fibres connecting the vestibular nucleus and nodulus/uvula were affected. Thus, we conclude that apogeotropic CPN results when the lesions affect the VSM pathway that is involved in the generation and transfer of estimated gravity. Limitations Our study has some limitations. First, the patients with apogeotropic CPN in this study usually had infarctions in the territory of the posterior inferior cerebellar artery or tumours involving the cerebellar vermis. Therefore, the results of lesion overlap might have been biased. Moreover, MRIs may not detect a tiny lesion involving the brainstem or cerebellum. Second, even though an erroneous tilt of the estimated gravity is the key concept of our modelling of apogeotropic CPN, we could not measure this subjective phenomenon in our patients. Finally, we did not have the data to model the effects of gravity on torsional and vertical eye movements in these patients. The model can be critically tested in the future with a three-dimensional approach to positional nystagmus in patients with CPN. Conclusion In summary, at a clinical level, while the pattern of apogeotropic nystagmus with the supine head roll test is similar between CPN and BPPV, associated neurological findings and the modulation pattern of nystagmus between sitting and supine positions usually permit a differentiation between apogeotropic CPN from apogeotropic BPPV. From the point of pathophysiology, apogeotropic CPN may be attributed to dysfunction of the central graviceptive pathway and can be simulated by lesions affecting the cerebellar and brainstem circuits involved in the generation of centrally created estimates of gravity that are used to determine whether one is rotating, translating or tilted. These signals are then passed to the vestibular nuclei to produce the appropriate type of compensatory eye movement. Funding This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (NRF-2016R1D1A1B04935568), and by the German Federal Ministry of Research and Education (BMBF Grant No. 01BO 0901). Conflicts of interest J-S.K serves as an Associate Editor of Frontiers in Neuro-otology and on the editorial boards of the Journal of Clinical Neurology, Frontiers in Neuro-ophthalmology, Journal of Neuro-ophthalmology, Journal of Vestibular Research, Medicine, and Journal of Neurology. S.G. serves as an academic editor for PLOS ONE and a reviewer for the European Commission. He is also on the editorial board of the Journal of Neurophysiology and Journal of Experimental Psychology Human Perception and Performance. He receives research support from the German Research Foundation (DFG) and the German Federal Ministry of Education and Research (BMBF) and is a shareholder of EyeSeeTec GmbH. D.S.Z serves as an Associate Editor of Frontiers in Neurootology and a member of the Editorial Board of The Cerebellum. He received speaker’s honoraria from Abbott pharmaceuticals and Micromed and royalties from Oxford University Press. Supplementary material Supplementary material is available at Brain online. Abbreviations Abbreviations BPPV benign paroxysmal positional vertigo/nystagmus CPN central positional nystagmus L/RED left/right ear down PIN position-induced nystagmus VSM velocity-storage mechanism References Alessandrini M, Micarelli A, Chiaravalloti A, Candidi M, Bruno E, Di Pietro B, et al.   Cerebellar metabolic involvement and its correlations with clinical parameters in vestibular neuritis. J Neurol  2014; 261: 1976– 85. Google Scholar CrossRef Search ADS PubMed  Anderson T, Luxon L, Quinn N, Daniel S, Marsden CD, Bronstein A. Oculomotor function in multiple system atrophy: clinical and laboratory features in 30 patients. Mov Disord  2008; 23: 977– 84. Google Scholar CrossRef Search ADS PubMed  Angelaki DE. Three-dimensional organization of otolith-ocular reflexes in rhesus monkeys. III. Responses to translation. J Neurophysiol  1998; 80: 680– 95. Google Scholar CrossRef Search ADS PubMed  Angelaki DE, Hess BJ. Inertial representation of angular motion in the vestibular system of rhesus monkeys. II. Otolith-controlled transformation that depends on an intact cerebellar nodulus. J Neurophysiol  1995; 73: 1729– 51. Google Scholar CrossRef Search ADS PubMed  Angelaki DE, Hess BJ. Three-dimensional organization of otolith-ocular reflexes in rhesus monkeys. I. Linear acceleration responses during off-vertical axis rotation. J Neurophysiol  1996; 75: 2405– 24. Google Scholar CrossRef Search ADS PubMed  Baloh RW, Yue Q, Jacobson KM, Honrubia V. Persistent direction-changing positional nystagmus: another variant of benign positional nystagmus? Neurology  1995; 45: 1297– 301. Google Scholar CrossRef Search ADS PubMed  Bisdorff A, Von Brevern M, Lempert T, Newman-Toker DE. Classification of vestibular symptoms: towards an international classification of vestibular disorders. J Vestib Res  2009; 19: 1– 13. Google Scholar PubMed  Büttner U, Helmchen C, Brandt T. Diagnostic criteria for central versus peripheral positioning nystagmus and vertigo: a review. Acta Otolaryngol  1999; 119: 1– 5. Google Scholar CrossRef Search ADS PubMed  Büttner-Ennever JA. A review of otolith pathways to brainstem and cerebellum. Ann N Y Acad Sci  1999; 871: 51– 64. Google Scholar CrossRef Search ADS PubMed  Choi JY, Jung I, Jung JM, Kwon DY, Park MH, Kim HJ, et al.   Characteristics and mechanism of perverted head-shaking nystagmus in central lesions: video-oculography analysis. Clin Neurophysiol  2016; 127: 2973– 8. Google Scholar CrossRef Search ADS PubMed  Choi JY, Kim JH, Kim HJ, Glasauer S, Kim JS. Central paroxysmal positional nystagmus: characteristics and possible mechanisms. Neurology  2015; 84: 2238– 46. Google Scholar CrossRef Search ADS PubMed  Choi JY, Kim JS. Nystagmus and central vestibular disorders. Curr Opin Neurol  2017; 30: 98– 106. Google Scholar CrossRef Search ADS PubMed  Choi KD, Oh SY, Kim HJ, Koo JW, Cho BM, Kim JS. Recovery of vestibular imbalances after vestibular neuritis. Laryngoscope  2007; 117: 1307– 12. Google Scholar CrossRef Search ADS PubMed  Cnyrim CD, Newman-Toker D, Karch C, Brandt T, Strupp M. Bedside differentiation of vestibular neuritis from central “vestibular pseudoneuritis”. J Neurol Neurosurg Psychiatry  2008; 79: 458– 60. Google Scholar CrossRef Search ADS PubMed  Diedrichsen J. A spatially unbiased atlas template of the human cerebellum. Neuroimage  2006; 33: 127– 38. Google Scholar CrossRef Search ADS PubMed  Glasauer S. Interaction of semicircular canals and otoliths in the processing structure of the subjective zenith. Ann N Y Acad Sci  1992; 656: 847– 9. Google Scholar CrossRef Search ADS PubMed  Glasauer S, Dieterich M, Brandt T. Central positional nystagmus simulated by a mathematical ocular motor model of otolith-dependent modification of Listing's plane. J Neurophysiol  2001; 86: 1546– 54. Google Scholar CrossRef Search ADS PubMed  Honrubia V, Bell TS, Harris MR, Baloh RW, Fisher LM. Quantitative evaluation of dizziness characteristics and impact on quality of life. Am J Otol  1996; 17: 595– 602. Google Scholar PubMed  Huh YE, Kim JS. Patterns of spontaneous and head-shaking nystagmus in cerebellar infarction: imaging correlations. Brain  2011; 134 (Pt 12): 3662– 71. Google Scholar CrossRef Search ADS   Kattah JC, Talkad AV, Wang DZ, Hsieh YH, Newman-Toker DE. HINTS to diagnose stroke in the acute vestibular syndrome: three-step bedside oculomotor examination more sensitive than early MRI diffusion-weighted imaging. Stroke  2009; 40: 3504– 10. Google Scholar CrossRef Search ADS PubMed  Kim HA, Yi HA, Lee H. Apogeotropic central positional nystagmus as a sole sign of nodular infarction. Neurol Sci  2012; 33: 1189– 91. Google Scholar CrossRef Search ADS PubMed  Kim JS, Zee DS. Clinical practice. Benign paroxysmal positional vertigo. N Engl J Med  2014; 370: 1138– 47. Google Scholar CrossRef Search ADS PubMed  Koo JW, Moon IJ, Shim WS, Moon SY, Kim JS. Value of lying-down nystagmus in the lateralization of horizontal semicircular canal benign paroxysmal positional vertigo. Otol Neurotol  2006; 27: 367– 71. Google Scholar CrossRef Search ADS PubMed  Laurens J, Angelaki DE. The functional significance of velocity storage and its dependence on gravity. Exp Brain Res  2011; 210: 407– 22. Google Scholar CrossRef Search ADS PubMed  Laurens J, Meng H, Angelaki DE. Computation of linear acceleration through an internal model in the macaque cerebellum. Nat Neurosci  2013; 16: 1701– 8. Google Scholar CrossRef Search ADS PubMed  Laurens J, Straumann D, Hess BJ. Processing of angular motion and gravity information through an internal model. J Neurophysiol  2010; 104: 1370– 81. Google Scholar CrossRef Search ADS PubMed  Lee HJ, Kim ES, Kim M, Chu H, Ma HI, Lee JS, et al.   Isolated horizontal positional nystagmus from a posterior fossa lesion. Ann Neurol  2014; 76: 905– 10. Google Scholar CrossRef Search ADS PubMed  Lee SH, Choi KD, Jeong SH, Oh YM, Koo JW, Kim JS. Nystagmus during neck flexion in the pitch plane in benign paroxysmal positional vertigo involving the horizontal canal. J Neurol Sci  2007; 256: 75– 80. Google Scholar CrossRef Search ADS PubMed  Lee SH, Kim JS. Benign paroxysmal positional vertigo. J Clin Neurol  2010; 6: 51– 63. Google Scholar CrossRef Search ADS PubMed  Lee SU, Choi JY, Kim HJ, Park JJ, Zee DS, Kim JS. Impaired tilt suppression of post-rotatory nystagmus and cross-coupled head-shaking nystagmus in cerebellar lesions: Image Mapping Study. Cerebellum  2017; 16: 95– 102. Google Scholar CrossRef Search ADS PubMed  Marti S, Palla A, Straumann D. Gravity dependence of ocular drift in patients with cerebellar downbeat nystagmus. Ann Neurol  2002; 52: 712– 21. Google Scholar CrossRef Search ADS PubMed  Merfeld DM, Park S, Gianna-Poulin C, Black FO, Wood S. Vestibular perception and action employ qualitatively different mechanisms. I. Frequency response of VOR and perceptual responses during translation and tilt. J Neurophysiol  2005; 94: 186– 98. Google Scholar CrossRef Search ADS PubMed  Moon IS, Kim JS, Choi KD, Kim MJ, Oh SY, Lee H, et al.   Isolated nodular infarction. Stroke  2009; 40: 487– 91. Google Scholar CrossRef Search ADS PubMed  Nam J, Kim S, Huh Y, Kim JS. Ageotropic central positional nystagmus in nodular infarction. Neurology  2009; 73: 1163. Google Scholar CrossRef Search ADS PubMed  Raphan T, Cohen B. Velocity storage and the ocular response to multidimensional vestibular stimuli. Rev Oculomot Res  1985; 1: 123– 43. Google Scholar PubMed  Schmahmann JD, Doyon J, McDonald D, Holmes C, Lavoie K, Hurwitz AS, et al.   Three-dimensional MRI atlas of the human cerebellum in proportional stereotaxic space. Neuroimage  1999; 10 (3 Pt 1): 233– 60. Google Scholar CrossRef Search ADS   Sheliga BM, Yakushin SB, Silvers A, Raphan T, Cohen B. Control of spatial orientation of the angular vestibulo-ocular reflex by the nodulus and uvula of the vestibulocerebellum. Ann N Y Acad Sci  1999; 871: 94– 122. Google Scholar CrossRef Search ADS PubMed  Sprenger A, Rambold H, Sander T, Marti S, Weber K, Straumann D, et al.   Treatment of the gravity dependence of downbeat nystagmus with 3, 4-diaminopyridine. Neurology  2006; 67: 905– 7. Google Scholar CrossRef Search ADS PubMed  Stoodley CJ, Schmahmann JD. Functional topography in the human cerebellum: a meta-analysis of neuroimaging studies. Neuroimage  2009; 44: 489– 501. Google Scholar CrossRef Search ADS PubMed  von Brevern M, Zeise D, Neuhauser H, Clarke AH, Lempert T. Acute migrainous vertigo: clinical and oculographic findings. Brain  2005; 128 (Pt 2): 365– 74. Voogd J, Barmack NH. Oculomotor cerebellum. Prog Brain Res  2006; 151: 231– 68. Google Scholar CrossRef Search ADS PubMed  Waespe W, Cohen B, Raphan T. Dynamic modification of the vestibulo-ocular reflex by the nodulus and uvula. Science  1985; 228: 199– 202. Google Scholar CrossRef Search ADS PubMed  Waespe W, Wichmann W. Oculomotor disturbances during visual-vestibular interaction in wallenbergs lateral medullary syndrome. Brain  1990; 113 (Pt 3): 821– 46. Google Scholar CrossRef Search ADS PubMed  Wearne S, Raphan T, Cohen B. Control of spatial orientation of the angular vestibuloocular reflex by the nodulus and uvula. J Neurophysiol  1998; 79: 2690– 715. Google Scholar CrossRef Search ADS PubMed  Yakusheva TA, Shaikh AG, Green AM, Blazquez PM, Dickman JD, Angelaki DE. Purkinje cells in posterior cerebellar vermis encode motion in an inertial reference frame. Neuron  2007; 54: 973– 85. Google Scholar CrossRef Search ADS PubMed  Yang Y, Kim JS, Kim S, Kim YK, Kwak YT, Han IW. Cerebellar hypoperfusion during transient global amnesia: an MRI and Oculographic Study. J Clin Neurol  2009; 5: 74– 80. Google Scholar CrossRef Search ADS PubMed  Yu-Wai-Man P, Gorman G, Bateman DE, Leigh RJ, Chinnery PF. Vertigo and vestibular abnormalities in spinocerebellar ataxia type 6. J Neurol  2009; 256: 78– 82. Google Scholar CrossRef Search ADS PubMed  Zupan LH, Peterka RJ, Merfeld DM. Neural processing of gravito-inertial cues in humans. I. Influence of the semicircular canals following post-rotatory tilt. J Neurophysiol  2000; 84: 2001– 15. Google Scholar CrossRef Search ADS PubMed  © The Author(s) (2018). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Brain Oxford University Press

Characteristics and mechanism of apogeotropic central positional nystagmus

Loading next page...
 
/lp/ou_press/characteristics-and-mechanism-of-apogeotropic-central-positional-qyQhWjgWdu
Publisher
Oxford University Press
Copyright
© The Author(s) (2018). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
0006-8950
eISSN
1460-2156
D.O.I.
10.1093/brain/awx381
Publisher site
See Article on Publisher Site

Abstract

Abstract Here we characterize persistent apogeotropic type of central positional nystagmus, and compare it with the apogeotropic nystagmus of benign paroxysmal positional vertigo involving the lateral canal. Nystagmus was recorded in 27 patients with apogeotropic type of central positional nystagmus (22 with unilateral and five with diffuse cerebellar lesions) and 20 patients with apogeotropic nystagmus of benign paroxysmal positional vertigo. They were tested while sitting, while supine with the head straight back, and in the right and left ear-down positions. The intensity of spontaneous nystagmus was similar while sitting and supine in apogeotropic type of central positional nystagmus, but greater when supine in apogeotropic nystagmus of benign paroxysmal positional vertigo. In central positional nystagmus, when due to a focal pathology, the lesions mostly overlapped in the vestibulocerebellum (nodulus, uvula, and tonsil). We suggest a mechanism for apogeotropic type of central positional nystagmus based on the location of lesions and a model that uses the velocity-storage mechanism. During both tilt and translation, the otolith organs can relay the same gravito-inertial acceleration signal. This inherent ambiguity can be resolved by a ‘tilt-estimator circuit’ in which information from the semicircular canals about head rotation is combined with otolith information about linear acceleration through the velocity-storage mechanism. An example of how this mechanism works in normal subjects is the sustained horizontal nystagmus that is produced when a normal subject is rotated at a constant speed around an axis that is tilted away from the true vertical (off-vertical axis rotation). We propose that when the tilt-estimator circuit malfunctions, for example, with lesions in the vestibulocerebellum, the estimate of the direction of gravity is erroneously biased away from true vertical. If the bias is toward the nose, when the head is turned to the side while supine, there will be sustained, unwanted, horizontal positional nystagmus (apogeotropic type of central positional nystagmus) because of an inappropriate feedback signal indicating that the head is rotating when it is not. vertigo, nystagmus, positional nystagmus, velocity-storage mechanism, nodulus Introduction A change in the orientation of the head with respect to the pull of gravity may induce paroxysmal (transient) or persistent positional nystagmus in central as well as peripheral vestibular disorders (Büttner et al., 1999; Bisdorff et al., 2009; Kim and Zee, 2014). In peripheral disorders, the most common is benign paroxysmal positional vertigo/nystagmus (BPPV), which occurs when gravity pulls otoconia that are free-flowing in the canals or adherent to the cupula (Kim and Zee, 2014). In contrast, positional nystagmus in central vestibular disorders may be caused by impaired central processing of canal and otolith cues (Büttner et al., 1999). Since the therapeutic strategy and prognosis differ between these disorders, one must distinguish between them in clinical practice. Among the central types of positional nystagmus, we previously demonstrated that central paroxysmal positional nystagmus (CPPN) has several characteristics that may differentiate it from BPPV (Choi et al., 2015). It was hypothesized that CPPN is initiated by enhanced post-rotatory canal signals and is directed to the vector of inhibited canals during the positioning (Choi et al., 2015; Choi and Kim, 2017). Among the types of persistent central positional nystagmus (CPN), two have been commonly reported in various disorders affecting the brainstem or cerebellum: downbeat nystagmus while supine or prone, and apogeotropic horizontal nystagmus during ear-down positions while supine (the direction of nystagmus beating ‘away from the earth’ with head turned to either side while supine) (Marti et al., 2002; von Brevern et al., 2005; Sprenger et al., 2006; Anderson et al., 2008; Yu-Wai-Man et al., 2009; Lee et al., 2014). Since downbeat nystagmus is well recognized as a central form of nystagmus, persistent positional downbeat nystagmus poses little diagnostic difficulty. In contrast, apogeotropic CPN may mimic the peripheral form of lateral canal BPPV related to cupulolithiasis (apogeotropic BPPV), one of the common causes of positional nystagmus (Baloh et al., 1995; Honrubia et al., 1996). Indeed, several studies have reported apogeotropic nystagmus as an isolated or a predominant sign of central vestibular disorders (Moon et al., 2009; Nam et al., 2009; Kim et al., 2012). However, to our knowledge, no study has systematically investigated the characteristics of apogeotropic CPN in comparison with those of BPPV. Furthermore, the neuroanatomical correlates of apogeotropic CPN have not been settled in humans even though clinical studies and animal experiments have implicated the vestibulocerebellum, especially the nodulus and uvula (Sheliga et al., 1999; Moon et al., 2009; Nam et al., 2009; Kim et al., 2012). Moreover, a fundamental question is how central lesions generate apogeotropic positional nystagmus. Abnormalities in central graviceptive pathways have been implicated based on the reproducibility of the nystagmus in specific head positions (Sheliga et al., 1999; Glasauer et al., 2001). Since an internal estimate of gravitational orientation is the only difference between the left and right ear-down positions (Fig. 1A), the oppositely-directed nystagmus [e.g. left beating on right-ear-down (RED) and right beating on left-ear-down positions (LED)] indicates that processing of information about gravity must be involved. However, this hypothesis has not been tested using human data and computational simulations of processing within central vestibular networks. Figure 1 View largeDownload slide Gravity orientation according to the head positions, nomenclature of apogeotropic nystagmus, and determination of nystagmus intensity. (A) Gravitational orientation. The gravity direction is the same between the head and earth coordinate frames in head upright position. During the positional test, the direction of gravity along the interaural axis was the only difference between the LED and RED positions while supine in head coordinate reference frame. (B) Ipsiversive and contraversive nystagmus are defined according to the direction of spontaneous nystagmus while supine, and ipsilesional and contralesional nystagmus are defined according to the lesion side (left-sided lesion is assumed in this figure). (C) Illustration showing how the intensities of ipsiversive (ipsilesional), contraversive (contralesional), and position-induced nystagmus are determined. g = gravity; LB/RB = left/right beating nystagmus; SPVs = slow-phase velocities. Positive value indicates rightward slow-phase velocities. Figure 1 View largeDownload slide Gravity orientation according to the head positions, nomenclature of apogeotropic nystagmus, and determination of nystagmus intensity. (A) Gravitational orientation. The gravity direction is the same between the head and earth coordinate frames in head upright position. During the positional test, the direction of gravity along the interaural axis was the only difference between the LED and RED positions while supine in head coordinate reference frame. (B) Ipsiversive and contraversive nystagmus are defined according to the direction of spontaneous nystagmus while supine, and ipsilesional and contralesional nystagmus are defined according to the lesion side (left-sided lesion is assumed in this figure). (C) Illustration showing how the intensities of ipsiversive (ipsilesional), contraversive (contralesional), and position-induced nystagmus are determined. g = gravity; LB/RB = left/right beating nystagmus; SPVs = slow-phase velocities. Positive value indicates rightward slow-phase velocities. In this study, we compared apogeotropic CPN with apogeotropic BPPV, analysed the location of lesions, suggest a mechanism for apogeotropic CPN, and simulated a mathematical model of the central interactions between the inputs from the semicircular canals and the otolith organs that can produce apogeotropic CPN. Materials and methods Patients This study was approved by the institutional review board of Seoul National University Bundang Hospital. We evaluated 27 patients (19 males and eight females, 58.5 ± 15.0 years) with positional vertigo and apogeotropic CPN. All patients underwent detailed neuro-otological evaluation by the senior author (J.S.K). The diagnosis of apogeotropic CPN was based on the following criteria: (i) presence of apogeotropic horizontal nystagmus; (ii) persistent nystagmus without fatigability; (iii) no resolution of apogeotropic nystagmus with repeated canalith repositioning manoeuvres; and (iv) documentation of the pathologies involving the brainstem or cerebellum with brain imaging or other neurological findings. Underlying disorders included brainstem or cerebellar infarction (n = 15), cerebellar tumour (n = 5), spinocerebellar ataxia (n = 4), multiple system atrophy (n = 1), and cerebellar haemorrhage (n = 2). The lesions were unilateral in the 22 patients who had a circumscribed lesion (infarction or tumour). For comparison, we studied 20 patients (seven males, 65.0 ± 15.7 years) with spontaneous nystagmus while sitting and supine from apogeotropic BPPV. The diagnosis was based on the following criteria: (i) presence of apogeotropic horizontal nystagmus; (ii) resolution of the vertigo and nystagmus with appropriate canalith repositioning manoeuvres; and (iii) absence of other neurological findings suggesting a central lesion. Evaluation Eye movements were recorded binocularly at a sampling rate of 60 Hz using 3D video-oculography (SensoMotoric Instruments). Digitized eye position data were analysed using MATLAB software (version R2011b, MathWorks, Natick, MA). Spontaneous nystagmus was recorded in the sitting position both with and without visual fixation. Gaze-evoked nystagmus, head-shaking nystagmus, and horizontal saccades and smooth pursuit were also evaluated. The detailed methods and normative data for saccades and smooth pursuit are described elsewhere (Yang et al., 2009). Positional nystagmus was induced without fixation in each position while patients were asked to keep their eyes in the straight-ahead position. After assessing for spontaneous nystagmus while sitting with the head upright, patients lay supine with the head 30° up from the earth horizontal plane. The patient’s head was then turned by about 90° to either side (Lee and Kim, 2010). Each position was maintained for at least 60 s. The intensity of horizontal nystagmus in each position was determined by averaging the slow-phase velocities of 10 consecutive beats when the nystagmus was relatively stable. Patients with CPN had bithermal caloric tests and an evaluation for the ocular tilt reaction. The ocular tilt reaction was considered present when the patients showed any head tilt, abnormal ocular torsion, skew deviation, and deviation of the subjective visual vertical. All additional tests were performed within 7 days of the eye movement recordings. The technical details and normative data of each test have been reported (Choi et al., 2007). Pattern analyses of apogeotropic nystagmus We evaluated the patterns of nystagmus in two ways. First, we analysed apogeotropic nystagmus during left/right-ear-down positions with respect to spontaneous horizontal nystagmus while supine. Thus, we adopted the terms ‘ipsiversive’ and ‘contraversive’. Ipsiversive nystagmus was defined when the beating direction of apogeotropic nystagmus was the same as that of spontaneous horizontal nystagmus while supine. In contrast, contraversive nystagmus indicated that the apogeotropic nystagmus beat in the opposite direction of the spontaneous horizontal nystagmus in the supine position. In patients with no spontaneous nystagmus while supine, the ipsiversive or contraversive nystagmus was determined by the direction of spontaneous nystagmus while sitting. For example, in patients with left beating spontaneous nystagmus while supine, the ipsiversive nystagmus would be observed in the RED position while the contraversive nystagmus was induced in the LED position (Fig. 1B). We determined the direction and intensity (absolute mean slow-phase velocities) of spontaneous horizontal nystagmus in the sitting, supine, and left/right-ear-down positions (Fig. 1C), and compared the intensity between ipsiversive and contraversive apogeotropic nystagmus. Second, we analysed apogeotropic nystagmus according to the lesion side. Thus, 22 patients with CPN and a circumscribed lesion and 20 patients with BPPV were included for analyses after excluding five patients with CPN from the multiple system atrophy and spinocerebellar ataxia cohorts. In this analysis, the term ‘ipsilesional’ was applied when the nystagmus beat toward the lesion side while ‘contralesional’ was adopted when the nystagmus beat toward the intact side. Therefore, in patients with a left-sided lesion, the left beating spontaneous nystagmus while sitting and supine was ‘ipsilesional’. In ear-down position, the left beating nystagmus during RED position was ‘ipsilesional’ apogeotropic nystagmus while the right beating nystagmus during LED position was ‘contralesional’ (Fig. 1B). Then, we determined the patterns of positional nystagmus in relation to the location of the lesion by comparing the intensity of ipsilesional and contralesional nystagmus (Fig. 1C). Positional effect on apogeotropic nystagmus To determine the positional effects on the apogeotropic nystagmus, we measured position-induced nystagmus (PIN) by subtracting the slow-phase velocities (SPVs) of spontaneous horizontal nystagmus while supine from that of apogeotropic nystagmus in each ear-down position using the following equations (Fig. 1C);   PINRED=SPVsofnystagmusonRED−SPVsofnystagmusonsupine (1)  PINLED=SPVsofnystagmusonLED−SPVsofnystagmusonsupine (2) For example, in a patient with left beating nystagmus while supine (slow-phase velocity, 3°/s), left beating nystagmus (10°/s) in the RED position has a 7°/s of PINRED and right beating nystagmus (−2°/s) in the LED position has a −5°/s of PINLED. Statistical analyses Normality of the data was determined using the Shapiro-Wilk test. In each subject, the intensities of ipsiversive (and ipsilesional) and contraversive (contralesional) apogeotropic nystagmus were compared using the Mann-Whitney test with statistical significance level of P < 0.05. Within each group, the Wilcoxon-signed rank test was used to compare the intensity of spontaneous nystagmus while sitting versus supine and ipsiversive versus contraversive apogeotropic nystagmus. The intensity of PIN was also compared between RED and LED positions in each group. Thus, the level of statistical significance for Wilcoxon-signed rank test was corrected using the Bonferroni method, and the corrected level of significance was set at 0.0167 (0.05/3) since the comparisons were performed three times in each group. In addition, by plotting PINLED as a function of PINRED, we determined the positional effects on apogeotropic nystagmus in both apogeotropic CPN and BPPV groups. Brain imaging and lesion analysis All patients with CPN underwent brain MRI according to the protocol described previously (Huh and Kim, 2011). Seventeen patients with an ischaemic or haemorrhagic stroke and four patients with a cerebellar tumour (three with hemangioblastoma and one with meningioma) underwent MRI within 14 days (4.8 ± 5.8 days) from the oculographic recording. One patient with cerebellar hemangioblastoma underwent MRIs 6 months before and 3 months after oculography, and there were no interval changes in the extent of the lesion. Five patients with degenerative disorders (spinocerebellar ataxia and multiple system atrophy) underwent MRI within 3 months before the oculography. We analysed MRI scans of 22 patients with circumscribed lesions. Diffusion-weighted and T2-weighted images were spatially normalized to a stereotaxic space using both linear and non-linear transformations with SPM8 (www.fil.ion.ucl.ac.uk/spm/software/spm8). Using MRIcron, the lesions were overlaid after flipping the right-sided lesions to the left (www.mccauslandcenter.sc.edu/mricro/mricron). Results Clinical features of apogeotropic CPN and BPPV Apogeotropic CPN All patients with apogeotropic CPN reported vertigo provoked or aggravated by positional changes. Twenty-two patients showed limb or truncal ataxia and eight also had sensory or bulbar dysfunction. With visual fixation while sitting, spontaneous horizontal nystagmus was present in only one patient (Patient 22). In contrast, spontaneous horizontal nystagmus was observed in 24 (88.9%) patients without fixation in darkness. Horizontal gaze-evoked nystagmus was present in 10 (37.0%) and downbeat nystagmus was induced after horizontal head shaking in 14 (51.9%) patients. Twenty-three (85.2%) patients showed abnormal saccades including increased latency in 20, hypometria in 16, and hypermetria in five. Horizontal smooth pursuit was impaired in 19 (70.4%) patients. Thus, in summary, all patients with apogeotropic CPN had additional ocular motor or other neurological findings indicative of central lesions. The findings in each patient are summarized in Table 1. Table 1 Findings in the patients with a persistent form of apogeotropic central positional nystagmus ID  Aetiology  Lesion side  Symptoms and signs  SN fix  SN non-fix  GEN  HSN  Pursuit gain  Saccades  OTR  CP  Sp. vertigo  Po. vertigo  Ataxia  Others  10°/s  20°/s  Latency  Gain  1  Cbll hemangioblastoma  L  −  +  −  −  −  L  −  D  −  ↓ (L)  Del (R)  ↓ (B)  R/Contralesional  −  2  Cbll hemangioblastoma  L  −  +  −  −  −  L  +  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  L/Ipsilesional    3  Cbll infarction (PICA)  L  +  +  −  −  −  L  −  D  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  4  Cbll infarction (PICA)  L  −  +  −  −  −  L  −  −  −  −  Del (B)  −  −    5  Cbll infarction (SCA)  L  +  +  Li/Tr  −  −  L  −  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  6  Cbll infarction (PICA)  L  −  +  Tr  −  −  L  +  R  ↓ (R)  ↓ (R)  Del (L)  ↓ (R)  R/Contralesional  −  7  Cbll infarction (PICA)  L  +  +  Li/Tr  Sen / Bul  −  L  −  D  −  −  Del (B)  ↓ (L)  R/Contralesional    8  Cbll infarction (PICA)  L  +  +  Tr  −  −  L  −  D  −  −  −  ↓ (B)  R/Contralesional  −  9  Cbll hemangioblastoma  L  +  +  Li/Tr  Bul  −  L/D  −  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  10  Cbll meningioma  L  −  +  Tr    −  L/D  −  R/D  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  R/Contralesional  −  11  Cbll haemorrhage  L  +  +  Tr  −  −  L/U  −  R/D  ↓ (R)  ↓ (R)  −  −  R/Contralesional  −  12  Cbll infarction (PICA)  L  +  +  Tr  −  −  −  −  D  ↓ (B)  ↓ (B)  −  ↓ (B)  R/Contralesional  −  13  Cbll infarction (PICA) + LMI  L  +  +  Li/Tr  −  −  R  +  D  ↓ (B)  ↓ (B)  −  ↓ (L)  R/Contralesional  −  14  PICA infarction  R  +  +  −  −  −  R  −  −  ↓ (B)  ↓ (B)  Del (R)  ↓ (R)  −  −  15  Cbll haemorrhage  R  +  +  Li/Tr  −  U  R  +  U  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  L/Contralesional  −  16  Cbll infarction (SCA)  R  +  +  Li/Tr  −  −  R  −  −  ↓ (R)  ↓ (R)  Del (B)  ↓ (L)  L/Contralesional  −  17  Cbll hemangioblastoma  R  +  +  Li/Tr  −  −  R  −  D          −    18  Cbll infarction (PICA)  R  +  +  Li/Tr  −  −  R  −  −          L/Contralesional  −  19  Cbll infarction (PICA)  R  +  +  Li/Tr  Bul  U  R/U  +  L/D  ↓ (R)  ↓ (R)  Del (R)  ↓ (L)  L/Contralesional  −  20  Cbll infarction (PICA)  R  +  +  Tr  −  −  R/U  +  −  −  −  Del (L)  ↓ (L)  −  −  21  Cbll infarction (PICA)  R  +  +  Li/Tr  −  −  R/D  −  R          L/Contralesional    22  Cbll infarction (PICA) + LMI  R  +  +  Li/Tr  Sen / Bul  L  L/U  −  R/D  ↓ (L)  ↓ (L)  Del (B)  −  R/Ipsilesional  −  23  Spinocerebellar ataxia type 6  Ud  +  +  Li/Tr  Bul  D  L/D  +  R/D  ↓ (B)  ↓ (B)  Del (B)  ↓ (R)  −  −  24  Spinocerebellar ataxia type 6  Ud  −  +  Li/Tr  Bul  −  −  +  L  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  −    25  Spinocerebellar ataxia type 6  Ud  −  +  Li/Tr  Bul  −  −  +  −  −  ↓ (B)  Del (B)  ↓ (L)  −    26  Multisystem atrophy  Ud  −  +  Tr  −  −  R  −  R  −  ↓ (B)  Del (B)  (L) ↑  −    27  Spinocerebellar ataxia type 6  Ud  +  +  Li/Tr  Bul  −  R  +  L/D  −  −  Del (B)  (L) ↑  R    ID  Aetiology  Lesion side  Symptoms and signs  SN fix  SN non-fix  GEN  HSN  Pursuit gain  Saccades  OTR  CP  Sp. vertigo  Po. vertigo  Ataxia  Others  10°/s  20°/s  Latency  Gain  1  Cbll hemangioblastoma  L  −  +  −  −  −  L  −  D  −  ↓ (L)  Del (R)  ↓ (B)  R/Contralesional  −  2  Cbll hemangioblastoma  L  −  +  −  −  −  L  +  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  L/Ipsilesional    3  Cbll infarction (PICA)  L  +  +  −  −  −  L  −  D  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  4  Cbll infarction (PICA)  L  −  +  −  −  −  L  −  −  −  −  Del (B)  −  −    5  Cbll infarction (SCA)  L  +  +  Li/Tr  −  −  L  −  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  6  Cbll infarction (PICA)  L  −  +  Tr  −  −  L  +  R  ↓ (R)  ↓ (R)  Del (L)  ↓ (R)  R/Contralesional  −  7  Cbll infarction (PICA)  L  +  +  Li/Tr  Sen / Bul  −  L  −  D  −  −  Del (B)  ↓ (L)  R/Contralesional    8  Cbll infarction (PICA)  L  +  +  Tr  −  −  L  −  D  −  −  −  ↓ (B)  R/Contralesional  −  9  Cbll hemangioblastoma  L  +  +  Li/Tr  Bul  −  L/D  −  −  ↓ (B)  ↓ (B)  Del (B)  ↓ (B)  R/Contralesional  −  10  Cbll meningioma  L  −  +  Tr    −  L/D  −  R/D  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  R/Contralesional  −  11  Cbll haemorrhage  L  +  +  Tr  −  −  L/U  −  R/D  ↓ (R)  ↓ (R)  −  −  R/Contralesional  −  12  Cbll infarction (PICA)  L  +  +  Tr  −  −  −  −  D  ↓ (B)  ↓ (B)  −  ↓ (B)  R/Contralesional  −  13  Cbll infarction (PICA) + LMI  L  +  +  Li/Tr  −  −  R  +  D  ↓ (B)  ↓ (B)  −  ↓ (L)  R/Contralesional  −  14  PICA infarction  R  +  +  −  −  −  R  −  −  ↓ (B)  ↓ (B)  Del (R)  ↓ (R)  −  −  15  Cbll haemorrhage  R  +  +  Li/Tr  −  U  R  +  U  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  L/Contralesional  −  16  Cbll infarction (SCA)  R  +  +  Li/Tr  −  −  R  −  −  ↓ (R)  ↓ (R)  Del (B)  ↓ (L)  L/Contralesional  −  17  Cbll hemangioblastoma  R  +  +  Li/Tr  −  −  R  −  D          −    18  Cbll infarction (PICA)  R  +  +  Li/Tr  −  −  R  −  −          L/Contralesional  −  19  Cbll infarction (PICA)  R  +  +  Li/Tr  Bul  U  R/U  +  L/D  ↓ (R)  ↓ (R)  Del (R)  ↓ (L)  L/Contralesional  −  20  Cbll infarction (PICA)  R  +  +  Tr  −  −  R/U  +  −  −  −  Del (L)  ↓ (L)  −  −  21  Cbll infarction (PICA)  R  +  +  Li/Tr  −  −  R/D  −  R          L/Contralesional    22  Cbll infarction (PICA) + LMI  R  +  +  Li/Tr  Sen / Bul  L  L/U  −  R/D  ↓ (L)  ↓ (L)  Del (B)  −  R/Ipsilesional  −  23  Spinocerebellar ataxia type 6  Ud  +  +  Li/Tr  Bul  D  L/D  +  R/D  ↓ (B)  ↓ (B)  Del (B)  ↓ (R)  −  −  24  Spinocerebellar ataxia type 6  Ud  −  +  Li/Tr  Bul  −  −  +  L  ↓ (B)  ↓ (B)  Del (B)  (B) ↑  −    25  Spinocerebellar ataxia type 6  Ud  −  +  Li/Tr  Bul  −  −  +  −  −  ↓ (B)  Del (B)  ↓ (L)  −    26  Multisystem atrophy  Ud  −  +  Tr  −  −  R  −  R  −  ↓ (B)  Del (B)  (L) ↑  −    27  Spinocerebellar ataxia type 6  Ud  +  +  Li/Tr  Bul  −  R  +  L/D  −  −  Del (B)  (L) ↑  R    B = both-direction; Bul = bulbar sign; Cbll = cerebellar; CP = caloric paresis; D = downbeat; GEN = gaze-evoked nystagmus; HSN = head-shaking induced nystagmus; L = left side, left beating, and left-ward; Li = limb; LMI = lateral medullary infarction; OTR = ocular tilt reaction; PICA = posterior inferior cerebellar artery territory; Po = positional; R = right side, right beating, and right-ward; SCA = superior cerebellar artery territory; Sen = sensory sign; SN fix = spontaneous nystagmus with visual fixation; SN non-fix = SN without visual fixation; Sp = spontaneous; Tr = truncal; U = upbeat; Ud = undetermined lesion side; + = present; − = normal or absent; ↓ = hypometria and decreased gain; ↑ = hypermetria; Blank cell indicates that the examination was not performed. Apogeotropic BPPV All patients with apogeotropic BPPV showed horizontal nystagmus while sitting (‘pseudo-spontaneous nystagmus’), which beat to the affected side in 15 and to the intact side in five. In all patients, the apogeotropic HC-BPPV resolved; in 11 with the reversed-Gufoni manoeuvre and in nine after being converted into the geotropic type that was then successfully treated with the Gufoni or barbecue manoeuvre (Supplementary Table 1). Patterns of positional nystagmus Apogeotropic CPN in relation to the spontaneous nystagmus while supine In darkness, 24 of the 27 patients in the CPN group showed horizontal nystagmus in the sitting position while 22 patients showed horizontal nystagmus in the supine position. The modulation of spontaneous nystagmus between the positions was presented in Table 2. Table 2 The intensity of spontaneous nystagmus in the sitting and supine positions and the intensity of apogeotropic positional nystagmus according to the direction of spontaneous nystagmus while supine Presence of SN in the sitting and supine positions in the apogeotropic CPN group    SN (+)  SN (−)  SN (n)  Sitting  24  3  24    SN (+)  SN (−)  SN (+)  SN (−)  SN (n)  Supine  19  5  3  0  22    The intensity of spontaneous horizontal nystagmus in the sitting and supine positions    Sitting  Supine  P    Apogeotropic CPN group  1.6 (0.8–2.9)  0.8 (0.5–1.9)  0.16  Apogeotropic BPPV group  1.4 (0–2.3)  6.0 (3.5–11.1)  < 0.001    The intensity of apogeotropic nystagmus according to the direction of SN while supine    Ipsiversive  Contraversive  P    Apogeotropic CPN group  5.3 (3.8–8.3)  3.1 (2.2–5.5)  <0.001  Apogeotropic BPPV group  22.6 (13.4–46.5)  7.6 (5.0–13.3)  <0.001  Presence of SN in the sitting and supine positions in the apogeotropic CPN group    SN (+)  SN (−)  SN (n)  Sitting  24  3  24    SN (+)  SN (−)  SN (+)  SN (−)  SN (n)  Supine  19  5  3  0  22    The intensity of spontaneous horizontal nystagmus in the sitting and supine positions    Sitting  Supine  P    Apogeotropic CPN group  1.6 (0.8–2.9)  0.8 (0.5–1.9)  0.16  Apogeotropic BPPV group  1.4 (0–2.3)  6.0 (3.5–11.1)  < 0.001    The intensity of apogeotropic nystagmus according to the direction of SN while supine    Ipsiversive  Contraversive  P    Apogeotropic CPN group  5.3 (3.8–8.3)  3.1 (2.2–5.5)  <0.001  Apogeotropic BPPV group  22.6 (13.4–46.5)  7.6 (5.0–13.3)  <0.001  Statistical analysis was performed using the Wilcoxon-signed rank test and the intensity of nystagmus is shown as the median (IQR). SN = spontaneous nystagmus; SN(n): number of patients with spontaneous nystagmus; SN (+) = presence of SN; SN (−) = absence of SN. The intensity of horizontal nystagmus did not differ between the sitting and supine positions in the CPN group [median (interquartile range, IQR), 1.6 (0.8–2.9) versus 0.8 (0.5–1.9), Wilcoxon-signed rank test, P = 0.160] (Fig. 2A and Table 2). Figure 2 View largeDownload slide Modulation of nystagmus in CPN and BPPV. (A) Spontaneous horizontal nystagmus in sitting and supine. While the intensities of spontaneous nystagmus did not differ between the sitting and supine positions in apogeotropic CPN, the nystagmus markedly increased in the supine position in patients with apogeotropic BPPV. (B) Apogeotropic nystagmus according to the direction of spontaneous nystagmus while supine. In both CPN and BPPV groups, the ipsiversive (nystagmus in the direction of spontaneous nystagmus while supine) apogeotropic nystagmus was greater than the contraversive one. The thin lines indicate individual data while the thick lines denote the median value. (C and D) PIN in CPN and BPPV groups. The CPN group (C) shows symmetrical PIN in the ear-down position to either side (Spearman’s correlation r = −0.97, P < 0.001). In contrast, the BPPV group (D) exhibits a marked asymmetry of the PIN (Spearman’s correlation r = −0.48, P = 0.034). Note that there are 4-fold differences in the magnitude scales between panels C and D. Figure 2 View largeDownload slide Modulation of nystagmus in CPN and BPPV. (A) Spontaneous horizontal nystagmus in sitting and supine. While the intensities of spontaneous nystagmus did not differ between the sitting and supine positions in apogeotropic CPN, the nystagmus markedly increased in the supine position in patients with apogeotropic BPPV. (B) Apogeotropic nystagmus according to the direction of spontaneous nystagmus while supine. In both CPN and BPPV groups, the ipsiversive (nystagmus in the direction of spontaneous nystagmus while supine) apogeotropic nystagmus was greater than the contraversive one. The thin lines indicate individual data while the thick lines denote the median value. (C and D) PIN in CPN and BPPV groups. The CPN group (C) shows symmetrical PIN in the ear-down position to either side (Spearman’s correlation r = −0.97, P < 0.001). In contrast, the BPPV group (D) exhibits a marked asymmetry of the PIN (Spearman’s correlation r = −0.48, P = 0.034). Note that there are 4-fold differences in the magnitude scales between panels C and D. During the ear-down positions, the ipsiversive nystagmus was greater than the contraversive one [5.3 (3.8–8.3) versus 3.1 (2.2–5.5), Wilcoxon-signed rank test, P < 0.001] (Fig. 2B and Table 2). Furthermore, all 22 patients with spontaneous horizontal nystagmus while supine showed greater ipsiversive than contraversive nystagmus (Mann-Whitney test, P < 0.05). In contrast, the remaining five patients without horizontal nystagmus while supine (Patients 5, 10, 18, 21, and 27) showed nearly symmetrical apogeotropic nystagmus between the right and left ear-down positions (Mann-Whitney test, P > 0.05). The intensity and direction of positional nystagmus in patients with CPN are summarized (Supplementary Table 1). Apogeotropic CPN according to the lesion side Of the 22 patients with CPN and circumscribed MRI lesions, 13 had a left-sided and nine had a right-sided lesion. The lesions were limited to the cerebellum in all patients except two (Patients 13 and 22) who also had lesions in the lateral medulla. The spontaneous horizontal nystagmus usually beat to the lesion side while sitting (86.4%) and supine (68.2%). Fifteen (68.2%) patients showed ipsilesional spontaneous nystagmus while supine, and the ipsilesional (ipsiversive, during contralesional head turn) apogeotropic nystagmus was greater than the contralesional nystagmus (Fig. 3A). Likewise, in three patients with contralesionally beating nystagmus while supine, the contralesional (ipsiversive, during ipsilesional head turn) apogeotropic nystagmus was greater than the ipsilesional one (Fig. 3B). In contrast, four patients without spontaneous nystagmus while supine developed nearly symmetrical apogeotropic nystagmus with the head turned to either side. The patterns of apogeotropic positional nystagmus are summarized according to the lesion side (Table 3). Table 3 Patterns of apogeotropic nystagmus in the CPN and BPPV groups   CPN group with circumscribed MRI lesion (n = 22)  BPPV group (n = 20)  Patterns of SN in the supine positions according to the lesion side    Ipsilesional  Contralesional  None  Ipsilesional  Supine  15  3  4  20  Patterns of apogeotropic nystagmus relative to the lesion side  Ear-down  Ipsilesional N > Contralesional N  Ipsilesional N < Contralesional N  Ipsilesional N = Contralesional N  Ipsilesional N > Contralesional N  Patterns of apogeotropic nystagmus relative to the direction of spontaneous nystagmus while supinea  Ear-down  Ipsiversive N > Contraversive N  Ipsiversive N > Contraversive N  Ipsiversive N = Contraversive N  Ipsiversive N > Contraversive N    CPN group with circumscribed MRI lesion (n = 22)  BPPV group (n = 20)  Patterns of SN in the supine positions according to the lesion side    Ipsilesional  Contralesional  None  Ipsilesional  Supine  15  3  4  20  Patterns of apogeotropic nystagmus relative to the lesion side  Ear-down  Ipsilesional N > Contralesional N  Ipsilesional N < Contralesional N  Ipsilesional N = Contralesional N  Ipsilesional N > Contralesional N  Patterns of apogeotropic nystagmus relative to the direction of spontaneous nystagmus while supinea  Ear-down  Ipsiversive N > Contraversive N  Ipsiversive N > Contraversive N  Ipsiversive N = Contraversive N  Ipsiversive N > Contraversive N  aNote that apogeotropic nystagmus appears to be more intense in the direction of spontaneous nystagmus while supine in patients with CPN irrespective of the lesion side. When the spontaneous nystagmus while supine was absent, the ipsiversive or contraversive nystagmus was determined according to the direction of spontaneous nystagmus while sitting. Contralesional N = contralesional beating nystagmus during ipsilesional head turn; Ipsilesional N = ipsilesional beating nystagmus during contralesional head turn; SN = spontaneous nystagmus. Figure 3 View largeDownload slide Illustration of spontaneous and apogeotropic nystagmus in CPN group and overlay lesion plots in 22 patients with apogeotropic CPN from unilateral lesions. (A) A patient (Patient 6) with a left-sided nodulus lesion shows ipsilesional left beating nystagmus while supine and apogeotropic nystagmus in the ear-down position to either side with the ipsiversive left beating positional nystagmus greater than the contraversive right beating one. (B) Another patient (Patient 22) with infarctions involving right lateral medulla and cerebellum shows contralesional left beating nystagmus while supine and greater ipsiversive nystagmus in right ear-down position. In each recording, upward deflection indicated rightward eye motion. (C) A patient with apogeotropic BPPV involving the left horizontal semicircular canal shows ipsilesional left beating nystagmus while supine and apogeotropic nystagmus during head turn to either side while supine. The ipsiversive left beating nystagmus in the right ear-down position was stronger than the contraversive right beating nystagmus in the left ear-down position. (D) The lesions are mostly overlapped (red colour) in the nodulus (X), uvula (IX), and tonsil (H IX) on the spatially unbiased atlas template of the cerebellum and brainstem (SUIT, ver. 2.5.3). The numbers of overlapping lesions are illustrated by different colours from violet (n = 1) to red (n = 12). (E) Illustration of the areas corresponding to the nodulus, uvula, and tonsil in two representative templates of the SUIT. LH = horizontal position of the left eye. SPV = slow-phase velocity. Figure 3 View largeDownload slide Illustration of spontaneous and apogeotropic nystagmus in CPN group and overlay lesion plots in 22 patients with apogeotropic CPN from unilateral lesions. (A) A patient (Patient 6) with a left-sided nodulus lesion shows ipsilesional left beating nystagmus while supine and apogeotropic nystagmus in the ear-down position to either side with the ipsiversive left beating positional nystagmus greater than the contraversive right beating one. (B) Another patient (Patient 22) with infarctions involving right lateral medulla and cerebellum shows contralesional left beating nystagmus while supine and greater ipsiversive nystagmus in right ear-down position. In each recording, upward deflection indicated rightward eye motion. (C) A patient with apogeotropic BPPV involving the left horizontal semicircular canal shows ipsilesional left beating nystagmus while supine and apogeotropic nystagmus during head turn to either side while supine. The ipsiversive left beating nystagmus in the right ear-down position was stronger than the contraversive right beating nystagmus in the left ear-down position. (D) The lesions are mostly overlapped (red colour) in the nodulus (X), uvula (IX), and tonsil (H IX) on the spatially unbiased atlas template of the cerebellum and brainstem (SUIT, ver. 2.5.3). The numbers of overlapping lesions are illustrated by different colours from violet (n = 1) to red (n = 12). (E) Illustration of the areas corresponding to the nodulus, uvula, and tonsil in two representative templates of the SUIT. LH = horizontal position of the left eye. SPV = slow-phase velocity. Apogeotropic BPPV All patients in the BPPV group showed ipsilesional-beating horizontal nystagmus while supine (‘lying-down nystagmus’). In contrast to the nystagmus in the CPN group, the horizontal nystagmus was greater in the supine than in the sitting position [median (IQR) 6.0 (3.5–11.1) versus 1.4 (0–2.3), Wilcoxon-signed rank test, P < 0.001] (Fig. 2A and Table 2). During the ear-down positions, the BPPV group also showed the ipsiversive (and ipsilesional, during contralesional head turn) nystagmus greater than the contraversive (and contralesional, during ipsilesional head turn) nystagmus [22.6 (13.4–46.5) versus 7.6 (5.0–13.3), Wilcoxon-signed rank test, P < 0.001] (Figs 2B, 3C, and Table 2). The intensities and directions of positional nystagmus in patients with BPPV are also summarized (Supplementary Table 1). Positional effects on apogeotropic CPN and BPPV The intensity of PIN in patients with CPN was 4.4 (3.2–6.1) and 4.6 (3.0–6.4) in RED and LED positions, respectively. Meanwhile the BPPV group showed 13.1 (8.0–25.8) and 17.9 (11.0–33.6) PIN intensity in RED and LED positions. In both CPN and BPPV groups, there was no difference in the intensity of PIN between the RED and LED positions (Wilcoxon-signed rank test, P > 0.05). PINLED and PINRED in the CPN group, were significantly correlated (Spearman’s correlation r = −0.97, P < 0.001), indicating the PIN was nearly symmetrical in individual patients with CPN (Fig. 2C). However, this was not the case for the BPPV group (Spearman’s correlation r = −0.48, P = 0.034); there was a marked asymmetry of PIN in individual patients with apogeotropic BPPV (Fig. 2D). Lesion analysis When the lesions were overlapped on the SUIT (spatially unbiased atlas template of the cerebellum and brainstem) (Diedrichsen, 2006), the overlay density plots showed that the nodulus (lobule X), uvula (lobule IX), and tonsil (lobule H IX) were most frequently involved in the 22 patients with apogeotropic CPN from circumscribed lesions (Fig. 3D and E) (Schmahmann et al., 1999; Stoodley and Schmahmann, 2009). Modelling Estimated gravitation-induced nystagmus in patients with apogeotropic CPN In apogeotropic CPN, the PIN was symmetric with opposing directions during head turn to either side while supine. Given that the orientation of the head relative to gravity is the only difference between the RED and LED positions (Fig. 1A), we assume that the PIN was from the effects of gravity. Our assumption was in accord with the results of a previous animal study in which injection of muscimol into the nodulus and uvula of a monkey produced apogeotropic positional nystagmus with a linear relationship between the intensity of the nystagmus and the gravitational orientation of the head (Sheliga et al., 1999). This predicts that the intensities of gravity-induced nystagmus (GIN) should be similar between the right and left ear-down positions. Thus, we may estimate the gravity-induced nystagmus in the LED, supine, and RED positions based on the previous assumptions: (i) the contribution of the canal signals (C) to the nystagmus is independent of static head orientation (Zupan et al., 2000); and (ii) the intensities of gravity-induced nystagmus (GINs) in RED and LED are the same but in the opposing directions (GINRED = −GINLED). Thus, the slow-phase velocities (SPVs) of apogeotropic nystagmus during both ear-down positions can be represented as follows:   SPVsinRED=C+GINRED (3)  SPVsinLED=C+GINLED=C−GINRED (4) We can calculate the canal contribution (C) to the nystagmus and the gravity-induced nystagmus during ear-down positions by adding and subtracting the two equations:   C=[Equation3+Equation4]/2 (5)  GINRED=[Equation3−Equation4]/2 (6) For example, in a patient with left beating nystagmus (slow-phase velocity, 8°/s) in the RED position, and right beating nystagmus (−2°/s) in the LED position, GINRED is 5°/s and C is 3°/s. We can also calculate the gravity-induced nystagmus while supine by subtracting the slow-phase velocity of C from that of spontaneous nystagmus while supine. The intensity of C was 0.9 (0.5–1.7), and the intensity of gravity-induced nystagmus during each ear-down position was 4.6 (3.1–6.2) in the CPN group. The estimated gravity-induced nystagmus according to head position is shown in Fig. 4A. In the supine position, the gravity-induced nystagmus was 0.1 (0–0.2), and these findings could be simulated by our model. Figure 4 View largeDownload slide Gravity-induced nystagmus and apogeotropic CPN model. (A) Estimated gravity-induced nystagmus (GIN) in both ear-down and supine positions are plotted in patients with apogeotropic CPN. The thick horizontal line and the top and bottom boundaries of the box represent the mean of GIN with ± 2 standard deviations at each position. Positive value indicates rightward slow-phase velocities (SPVs). (B) A schematic model modified from Laurens and Angelaki (2011). The red lightning symbol is the assumed site of the lesion within the model. (C) In the normal condition, the estimated gravity direction in the ear-down position corresponds to the actual gravity direction. In patients with CPN, the estimated gravity direction is biased toward the nose along the nasooccipital axis of the head while in ear-down positions. The persistent directional mismatch between the actual and estimated gravities can cause the rotational cue (feedback) to bring the estimate of gravity toward the actual one. An empty circle is a summation point. g = estimated gravity; GIA = gravitoinertial acceleration; i = inertia; w = estimated angular velocity; x and ∫ are mathematical terms representing a vector cross product and integral. Figure 4 View largeDownload slide Gravity-induced nystagmus and apogeotropic CPN model. (A) Estimated gravity-induced nystagmus (GIN) in both ear-down and supine positions are plotted in patients with apogeotropic CPN. The thick horizontal line and the top and bottom boundaries of the box represent the mean of GIN with ± 2 standard deviations at each position. Positive value indicates rightward slow-phase velocities (SPVs). (B) A schematic model modified from Laurens and Angelaki (2011). The red lightning symbol is the assumed site of the lesion within the model. (C) In the normal condition, the estimated gravity direction in the ear-down position corresponds to the actual gravity direction. In patients with CPN, the estimated gravity direction is biased toward the nose along the nasooccipital axis of the head while in ear-down positions. The persistent directional mismatch between the actual and estimated gravities can cause the rotational cue (feedback) to bring the estimate of gravity toward the actual one. An empty circle is a summation point. g = estimated gravity; GIA = gravitoinertial acceleration; i = inertia; w = estimated angular velocity; x and ∫ are mathematical terms representing a vector cross product and integral. Model simulation The influence of gravity on eye position and velocity has been explained by two mechanisms. First, changes in the direction of the pull of gravity relative to the head, as with head tilts, changes primary eye position (i.e. an alteration of the orientation of Listing’s plane), though this mechanism would be expected to produce only a tiny positional horizontal nystagmus (Glasauer et al., 2001). Second, the direction of gravity affects the vestibulo-ocular reflex via the velocity-storage mechanism (VSM) (Raphan and Cohen, 1985; Yakusheva et al., 2007). The VSM is part of a central circuit that improves the low-frequency response of the rotational vestibulo-ocular reflex, and helps provide an internal estimate of the direction of gravity relative to the head. The latter signal is used to resolve the ambiguity between static head tilt relative to upright and translation of the head (Glasauer, 1992; Laurens and Angelaki, 2011; Laurens et al., 2013). For better understanding, we briefly describe here how the proposed model makes this distinction (Fig. 4B) (Laurens and Angelaki, 2011; Laurens et al., 2013). The otolith organs provide information on the gravito-inertial acceleration (GIA), a sum of gravitational acceleration (g) and translation-driven inertial acceleration (i) but they do not discriminate between the two. Thus, the otolith-driven vectors of gravito-inertial acceleration can be identical during tilt and translation so it is the VSM that enables a discrimination of tilt from translation. From the VSM, the brain obtains information about the position of the head related to gravity (tilt estimator) by integrating the cross product of canal-driven head velocity signals and gravity (∫ gravity × head velocity). Then, by subtracting the head tilt from the gravito-inertial acceleration (i = GIA − g), an appropriate translational cue is extracted to generate command for the translational vestibulo-ocular reflex. There are two feedback loops within the VSM model; one is the somatogravic loop that explains the faulty perception of tilt during low-frequency translation (somatogravic illusion) (Angelaki, 1998; Merfeld et al., 2005). The other is the rotational feedback loop, which enhances the accuracy of the tilt estimator. Because the VSM estimates head tilt by integrating head velocity, any inaccuracy of canal signals would cause a bias for head tilt. The directional mismatch between the actual gravity and the estimated one causes an erroneous cue for translation. To minimize this, the rotational feedback compensates for the directional mismatch by providing a corrective rotational cue. In normal subjects this model predicts that sustained nystagmus can be generated without direct canal stimulation. For example, a sustained nystagmus is elicited during constant-velocity, off-vertical axis rotations in normal subjects even though the signal from the canals has decayed away (Angelaki and Hess, 1996; Laurens et al., 2010). Therefore, to simulate sustained apogeotropic CPN, which also occurs after stimulation of the canals has ceased, we implemented a model of VSM in MATLAB/Simulink (The MathWorks, Natick, MA) and applied several different hypothetical lesion sites. The model simulation produced gravity-related, symmetric apogeotropic nystagmus in both ear-down positions assuming the lesion involves the vestibulocerebellar pathway that relays the estimated gravity to rotational feedback (Fig. 4B, red lightning symbol). We assumed that disruption of this pathway results in loss of information about the estimated gravity and also causes a small bias (0.02g) toward the nose along the naso-occipital axis of the head. Thus, in the right ear-down position, the actual gravity points toward the right ear while the estimated gravity is directed between the right ear and nose (Fig. 4C). Such a lesion generates a horizontal apogeotropic nystagmus of ∼6.4°/s) (Fig. 5). Note that neither the main VSM in the brainstem nor the direct otolith-ocular pathways have to be damaged to generate apogeotropic nystagmus. Further information on modelling is presented in the Supplementary material. Figure 5 View largeDownload slide Results of model simulation. A right-handed, head-fixed coordinate system is used (z-axis pointing upward, x-axis pointing forward, and y-axis pointing leftward). Gravity points downward in the upright head position. (A) The actual head angular velocity (w) and gravity orientation (g) during positioning from sitting to right-ear-down position. The inserted cartoons indicate actual head position. In the model, the head starts from upright and tilts back 90° at 5 s, and it finally reaches right-ear-down position at 20 s. The simulation is then continued for a complete duration of 70 s. (B) The simulation results show an estimated angular velocity (upper row) and a gravity orientation (lower row) in healthy subjects (left) and patients with CPN (right). (C) The estimated angular velocity along the rostro-caudal axis during right ear-down position while supine. After reaching the final position, the angular head velocity remains zero in healthy subjects (blue). In patients with CPN (red), the lesion-induced tilt toward the nose of the estimate of gravity induces a persistent estimate of angular head velocity, which in turn produces rightward slow-phase eye velocity of 6.2°/s and left-beating nystagmus (apogeotropic nystagmus). Figure 5 View largeDownload slide Results of model simulation. A right-handed, head-fixed coordinate system is used (z-axis pointing upward, x-axis pointing forward, and y-axis pointing leftward). Gravity points downward in the upright head position. (A) The actual head angular velocity (w) and gravity orientation (g) during positioning from sitting to right-ear-down position. The inserted cartoons indicate actual head position. In the model, the head starts from upright and tilts back 90° at 5 s, and it finally reaches right-ear-down position at 20 s. The simulation is then continued for a complete duration of 70 s. (B) The simulation results show an estimated angular velocity (upper row) and a gravity orientation (lower row) in healthy subjects (left) and patients with CPN (right). (C) The estimated angular velocity along the rostro-caudal axis during right ear-down position while supine. After reaching the final position, the angular head velocity remains zero in healthy subjects (blue). In patients with CPN (red), the lesion-induced tilt toward the nose of the estimate of gravity induces a persistent estimate of angular head velocity, which in turn produces rightward slow-phase eye velocity of 6.2°/s and left-beating nystagmus (apogeotropic nystagmus). Discussion This is the first study to systematically evaluate apogeotropic APN from two perspectives. First, we show a way to differentiate apogeotropic CPN from apogeotropic BPPV based on the patterns of nystagmus and accompanying neurological signs. Second, we provide a mechanism for apogeotropic CPN using a model of the VSM that suggests how the brain both maintains gravity orientation and generates an appropriate translational vestibulo-ocular reflex in response to an inertial linear acceleration. Clinical findings of apogeotropic CPN All patients with CPN suffered from vertigo induced or aggravated by positional changes. Most of our patients with apogeotropic CPN also showed neurological signs indicating a central pathology, which included gaze-evoked nystagmus (Cnyrim et al., 2008; Kattah et al., 2009), perverted (cross-coupled) head-shaking nystagmus (Huh and Kim, 2011; Choi et al., 2016), and hypermetric saccades (Waespe and Wichmann, 1990), in addition to limb or truncal ataxia. However, dizziness/vertigo was the only symptom in five patients (18.5%) making differentiation between apogeotropic CPN and BPPV difficult in clinical practice. How do the patterns of positional nystagmus give a clue to differentiate CPN from BPPV? Pattern of apogeotropic nystagmus in CPN and BPPV Generally, in CPN patients with circumscribed lesions, the spontaneous horizontal nystagmus is ipsilesional while supine and the ipsilesional apogeotropic nystagmus is more intense when the head is turned to the contralesional side (Fig. 3A). This is the pattern reported previously (Sheliga et al., 1999; Nam et al., 2009; Kim et al., 2012). Of interest, however, two patients (Patients 13 and 22) with contralesional spontaneous nystagmus from lesions involving both the inferior cerebellum and lateral medulla showed contralesional (but ipsiversive) apogeotropic nystagmus greater than the ipsilesional one (Fig. 3B). Taken together, apogeotropic nystagmus is more intense in the direction of spontaneous nystagmus while supine in patients with CPN irrespective of the lesion side. Since apogeotropic BPPV also shows asymmetric apogeotropic nystagmus and ipsiversive greater than contraversive nystagmus, an asymmetric pattern of apogeotropic nystagmus during the supine head roll test alone does not permit differentiation of the central from the peripheral form of apogeotropic nystagmus. On the other hand, the spontaneous horizontal nystagmus differed little between the sitting and supine positions in the CPN group, while the nystagmus usually became considerably greater in the supine than in the sitting position in the BPPV group. Thus, augmentation of spontaneous nystagmus while supine favors the diagnosis of apogeotropic HC-BPPV. In apogeotropic CPN, the similar intensity of the nystagmus between the sitting and supine positions suggests that changes in the gravity orientation from sitting to supine do not affect the horizontal nystagmus. In contrast, in apogeotropic BPPV, the increment of spontaneous horizontal nystagmus in the supine position may be explained by the change in the orientation relative to gravity of the cupula, laden with otolithic debris, within the lateral semicircular canal (Koo et al., 2006; Lee et al., 2007). Pathophysiological mechanism of apogeotropic CPN based on the lesions and modelling Our analysis suggested that apogeotropic CPN can result from summation of incorrectly interpreted canal-induced nystagmus (C) and gravity-induced nystagmus in both ear-down positions. We tested this idea and how it relates to vestibulo-cerebellar lesions using a model of the VSMs. Under normal circumstances, the brain appropriately estimates the direction of gravity and there is no directional mismatch between the actual and estimated gravity during the ear-down positions (Fig. 4C). Therefore, the horizontal eye velocity would be zero during ear-down position in healthy subjects. However, as shown in the constant off-vertical axis rotation paradigm (OVAR), a directional mismatch between the actual gravity provided by the otolith organs and the internally-estimated gravity could generate a compensatory rotational feedback signal that brings the estimated gravity direction into a veridical one (Laurens and Angelaki, 2011). In our VSM model, if a lesion disrupts the pathway providing the estimated gravity to the rotational feedback loop and also produces a positive bias toward the nose along with nasooccipital axis (x = 0.02g), the compensatory rotational feedback would generate constant horizontal apogeotropic gravity-induced nystagmus (Fig. 5). If there were a negative bias, away from the nose along the nasooccipital axis, the compensatory rotational feedback would generate a constant horizontal geotropic gravity-induced nystagmus. In this study, patients with apogeotropic CPN showed lesions that mostly overlapped in the vestibulocerebellum including the nodulus, uvula, and tonsil. The nodulus and uvula may function as the ‘tilt estimator’ in humans (Lee et al., 2017), as was suggested in monkeys (Angelaki and Hess, 1995). And the Purkinje cells in the nodulus and uvula correlate with translational (inertial) inputs (Yakusheva et al., 2007). Complete resection of the nodulus and uvula in monkeys causes a loss of post-rotational tilt suppression (Waespe et al., 1985; Wearne et al., 1998), which also has been attributed to the rotational feedback mechanism in the VSM model (Laurens and Angelaki, 2011). All these functions are compatible with the anatomical characteristics of the nodulus and uvula. It was shown that ∼70% of the primary vestibular afferents project to the nodulus and uvula. Especially those from the canals predominantly synapse with the nodulus while those from the otolithic organs mainly connect with the ventral uvula. The nodulus and uvula also have reciprocal connections with the neurons in the vestibular nucleus (Büttner-Ennever, 1999; Voogd and Barmack, 2006). The correlation of hypometabolism of the nodulus with the degree of subjective visual vertical tilt in patients with acute vestibular neuritis also supports these functional connections (Alessandrini et al., 2014). However, we assume that the tilt estimators need not be completely damaged for generation of CPN since in monkeys, partial resection of the nodulus and uvula preserve post-rotational tilt suppression (Wearne et al., 1998) and apogeotropic nystagmus occurs in monkeys with only unilateral chemical inactivation of the nodulus/uvula (Sheliga et al., 1999). Even in our patients with lesions involving the nodulus and uvula, the lesions were mostly unilateral (partial). In patients with lesions sparing the nodulus/uvula (such as Patient 22), we assume that the neural fibres connecting the vestibular nucleus and nodulus/uvula were affected. Thus, we conclude that apogeotropic CPN results when the lesions affect the VSM pathway that is involved in the generation and transfer of estimated gravity. Limitations Our study has some limitations. First, the patients with apogeotropic CPN in this study usually had infarctions in the territory of the posterior inferior cerebellar artery or tumours involving the cerebellar vermis. Therefore, the results of lesion overlap might have been biased. Moreover, MRIs may not detect a tiny lesion involving the brainstem or cerebellum. Second, even though an erroneous tilt of the estimated gravity is the key concept of our modelling of apogeotropic CPN, we could not measure this subjective phenomenon in our patients. Finally, we did not have the data to model the effects of gravity on torsional and vertical eye movements in these patients. The model can be critically tested in the future with a three-dimensional approach to positional nystagmus in patients with CPN. Conclusion In summary, at a clinical level, while the pattern of apogeotropic nystagmus with the supine head roll test is similar between CPN and BPPV, associated neurological findings and the modulation pattern of nystagmus between sitting and supine positions usually permit a differentiation between apogeotropic CPN from apogeotropic BPPV. From the point of pathophysiology, apogeotropic CPN may be attributed to dysfunction of the central graviceptive pathway and can be simulated by lesions affecting the cerebellar and brainstem circuits involved in the generation of centrally created estimates of gravity that are used to determine whether one is rotating, translating or tilted. These signals are then passed to the vestibular nuclei to produce the appropriate type of compensatory eye movement. Funding This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (NRF-2016R1D1A1B04935568), and by the German Federal Ministry of Research and Education (BMBF Grant No. 01BO 0901). Conflicts of interest J-S.K serves as an Associate Editor of Frontiers in Neuro-otology and on the editorial boards of the Journal of Clinical Neurology, Frontiers in Neuro-ophthalmology, Journal of Neuro-ophthalmology, Journal of Vestibular Research, Medicine, and Journal of Neurology. S.G. serves as an academic editor for PLOS ONE and a reviewer for the European Commission. He is also on the editorial board of the Journal of Neurophysiology and Journal of Experimental Psychology Human Perception and Performance. He receives research support from the German Research Foundation (DFG) and the German Federal Ministry of Education and Research (BMBF) and is a shareholder of EyeSeeTec GmbH. D.S.Z serves as an Associate Editor of Frontiers in Neurootology and a member of the Editorial Board of The Cerebellum. He received speaker’s honoraria from Abbott pharmaceuticals and Micromed and royalties from Oxford University Press. Supplementary material Supplementary material is available at Brain online. Abbreviations Abbreviations BPPV benign paroxysmal positional vertigo/nystagmus CPN central positional nystagmus L/RED left/right ear down PIN position-induced nystagmus VSM velocity-storage mechanism References Alessandrini M, Micarelli A, Chiaravalloti A, Candidi M, Bruno E, Di Pietro B, et al.   Cerebellar metabolic involvement and its correlations with clinical parameters in vestibular neuritis. J Neurol  2014; 261: 1976– 85. Google Scholar CrossRef Search ADS PubMed  Anderson T, Luxon L, Quinn N, Daniel S, Marsden CD, Bronstein A. Oculomotor function in multiple system atrophy: clinical and laboratory features in 30 patients. Mov Disord  2008; 23: 977– 84. Google Scholar CrossRef Search ADS PubMed  Angelaki DE. Three-dimensional organization of otolith-ocular reflexes in rhesus monkeys. III. Responses to translation. J Neurophysiol  1998; 80: 680– 95. Google Scholar CrossRef Search ADS PubMed  Angelaki DE, Hess BJ. Inertial representation of angular motion in the vestibular system of rhesus monkeys. II. Otolith-controlled transformation that depends on an intact cerebellar nodulus. J Neurophysiol  1995; 73: 1729– 51. Google Scholar CrossRef Search ADS PubMed  Angelaki DE, Hess BJ. Three-dimensional organization of otolith-ocular reflexes in rhesus monkeys. I. Linear acceleration responses during off-vertical axis rotation. J Neurophysiol  1996; 75: 2405– 24. Google Scholar CrossRef Search ADS PubMed  Baloh RW, Yue Q, Jacobson KM, Honrubia V. Persistent direction-changing positional nystagmus: another variant of benign positional nystagmus? Neurology  1995; 45: 1297– 301. Google Scholar CrossRef Search ADS PubMed  Bisdorff A, Von Brevern M, Lempert T, Newman-Toker DE. Classification of vestibular symptoms: towards an international classification of vestibular disorders. J Vestib Res  2009; 19: 1– 13. Google Scholar PubMed  Büttner U, Helmchen C, Brandt T. Diagnostic criteria for central versus peripheral positioning nystagmus and vertigo: a review. Acta Otolaryngol  1999; 119: 1– 5. Google Scholar CrossRef Search ADS PubMed  Büttner-Ennever JA. A review of otolith pathways to brainstem and cerebellum. Ann N Y Acad Sci  1999; 871: 51– 64. Google Scholar CrossRef Search ADS PubMed  Choi JY, Jung I, Jung JM, Kwon DY, Park MH, Kim HJ, et al.   Characteristics and mechanism of perverted head-shaking nystagmus in central lesions: video-oculography analysis. Clin Neurophysiol  2016; 127: 2973– 8. Google Scholar CrossRef Search ADS PubMed  Choi JY, Kim JH, Kim HJ, Glasauer S, Kim JS. Central paroxysmal positional nystagmus: characteristics and possible mechanisms. Neurology  2015; 84: 2238– 46. Google Scholar CrossRef Search ADS PubMed  Choi JY, Kim JS. Nystagmus and central vestibular disorders. Curr Opin Neurol  2017; 30: 98– 106. Google Scholar CrossRef Search ADS PubMed  Choi KD, Oh SY, Kim HJ, Koo JW, Cho BM, Kim JS. Recovery of vestibular imbalances after vestibular neuritis. Laryngoscope  2007; 117: 1307– 12. Google Scholar CrossRef Search ADS PubMed  Cnyrim CD, Newman-Toker D, Karch C, Brandt T, Strupp M. Bedside differentiation of vestibular neuritis from central “vestibular pseudoneuritis”. J Neurol Neurosurg Psychiatry  2008; 79: 458– 60. Google Scholar CrossRef Search ADS PubMed  Diedrichsen J. A spatially unbiased atlas template of the human cerebellum. Neuroimage  2006; 33: 127– 38. Google Scholar CrossRef Search ADS PubMed  Glasauer S. Interaction of semicircular canals and otoliths in the processing structure of the subjective zenith. Ann N Y Acad Sci  1992; 656: 847– 9. Google Scholar CrossRef Search ADS PubMed  Glasauer S, Dieterich M, Brandt T. Central positional nystagmus simulated by a mathematical ocular motor model of otolith-dependent modification of Listing's plane. J Neurophysiol  2001; 86: 1546– 54. Google Scholar CrossRef Search ADS PubMed  Honrubia V, Bell TS, Harris MR, Baloh RW, Fisher LM. Quantitative evaluation of dizziness characteristics and impact on quality of life. Am J Otol  1996; 17: 595– 602. Google Scholar PubMed  Huh YE, Kim JS. Patterns of spontaneous and head-shaking nystagmus in cerebellar infarction: imaging correlations. Brain  2011; 134 (Pt 12): 3662– 71. Google Scholar CrossRef Search ADS   Kattah JC, Talkad AV, Wang DZ, Hsieh YH, Newman-Toker DE. HINTS to diagnose stroke in the acute vestibular syndrome: three-step bedside oculomotor examination more sensitive than early MRI diffusion-weighted imaging. Stroke  2009; 40: 3504– 10. Google Scholar CrossRef Search ADS PubMed  Kim HA, Yi HA, Lee H. Apogeotropic central positional nystagmus as a sole sign of nodular infarction. Neurol Sci  2012; 33: 1189– 91. Google Scholar CrossRef Search ADS PubMed  Kim JS, Zee DS. Clinical practice. Benign paroxysmal positional vertigo. N Engl J Med  2014; 370: 1138– 47. Google Scholar CrossRef Search ADS PubMed  Koo JW, Moon IJ, Shim WS, Moon SY, Kim JS. Value of lying-down nystagmus in the lateralization of horizontal semicircular canal benign paroxysmal positional vertigo. Otol Neurotol  2006; 27: 367– 71. Google Scholar CrossRef Search ADS PubMed  Laurens J, Angelaki DE. The functional significance of velocity storage and its dependence on gravity. Exp Brain Res  2011; 210: 407– 22. Google Scholar CrossRef Search ADS PubMed  Laurens J, Meng H, Angelaki DE. Computation of linear acceleration through an internal model in the macaque cerebellum. Nat Neurosci  2013; 16: 1701– 8. Google Scholar CrossRef Search ADS PubMed  Laurens J, Straumann D, Hess BJ. Processing of angular motion and gravity information through an internal model. J Neurophysiol  2010; 104: 1370– 81. Google Scholar CrossRef Search ADS PubMed  Lee HJ, Kim ES, Kim M, Chu H, Ma HI, Lee JS, et al.   Isolated horizontal positional nystagmus from a posterior fossa lesion. Ann Neurol  2014; 76: 905– 10. Google Scholar CrossRef Search ADS PubMed  Lee SH, Choi KD, Jeong SH, Oh YM, Koo JW, Kim JS. Nystagmus during neck flexion in the pitch plane in benign paroxysmal positional vertigo involving the horizontal canal. J Neurol Sci  2007; 256: 75– 80. Google Scholar CrossRef Search ADS PubMed  Lee SH, Kim JS. Benign paroxysmal positional vertigo. J Clin Neurol  2010; 6: 51– 63. Google Scholar CrossRef Search ADS PubMed  Lee SU, Choi JY, Kim HJ, Park JJ, Zee DS, Kim JS. Impaired tilt suppression of post-rotatory nystagmus and cross-coupled head-shaking nystagmus in cerebellar lesions: Image Mapping Study. Cerebellum  2017; 16: 95– 102. Google Scholar CrossRef Search ADS PubMed  Marti S, Palla A, Straumann D. Gravity dependence of ocular drift in patients with cerebellar downbeat nystagmus. Ann Neurol  2002; 52: 712– 21. Google Scholar CrossRef Search ADS PubMed  Merfeld DM, Park S, Gianna-Poulin C, Black FO, Wood S. Vestibular perception and action employ qualitatively different mechanisms. I. Frequency response of VOR and perceptual responses during translation and tilt. J Neurophysiol  2005; 94: 186– 98. Google Scholar CrossRef Search ADS PubMed  Moon IS, Kim JS, Choi KD, Kim MJ, Oh SY, Lee H, et al.   Isolated nodular infarction. Stroke  2009; 40: 487– 91. Google Scholar CrossRef Search ADS PubMed  Nam J, Kim S, Huh Y, Kim JS. Ageotropic central positional nystagmus in nodular infarction. Neurology  2009; 73: 1163. Google Scholar CrossRef Search ADS PubMed  Raphan T, Cohen B. Velocity storage and the ocular response to multidimensional vestibular stimuli. Rev Oculomot Res  1985; 1: 123– 43. Google Scholar PubMed  Schmahmann JD, Doyon J, McDonald D, Holmes C, Lavoie K, Hurwitz AS, et al.   Three-dimensional MRI atlas of the human cerebellum in proportional stereotaxic space. Neuroimage  1999; 10 (3 Pt 1): 233– 60. Google Scholar CrossRef Search ADS   Sheliga BM, Yakushin SB, Silvers A, Raphan T, Cohen B. Control of spatial orientation of the angular vestibulo-ocular reflex by the nodulus and uvula of the vestibulocerebellum. Ann N Y Acad Sci  1999; 871: 94– 122. Google Scholar CrossRef Search ADS PubMed  Sprenger A, Rambold H, Sander T, Marti S, Weber K, Straumann D, et al.   Treatment of the gravity dependence of downbeat nystagmus with 3, 4-diaminopyridine. Neurology  2006; 67: 905– 7. Google Scholar CrossRef Search ADS PubMed  Stoodley CJ, Schmahmann JD. Functional topography in the human cerebellum: a meta-analysis of neuroimaging studies. Neuroimage  2009; 44: 489– 501. Google Scholar CrossRef Search ADS PubMed  von Brevern M, Zeise D, Neuhauser H, Clarke AH, Lempert T. Acute migrainous vertigo: clinical and oculographic findings. Brain  2005; 128 (Pt 2): 365– 74. Voogd J, Barmack NH. Oculomotor cerebellum. Prog Brain Res  2006; 151: 231– 68. Google Scholar CrossRef Search ADS PubMed  Waespe W, Cohen B, Raphan T. Dynamic modification of the vestibulo-ocular reflex by the nodulus and uvula. Science  1985; 228: 199– 202. Google Scholar CrossRef Search ADS PubMed  Waespe W, Wichmann W. Oculomotor disturbances during visual-vestibular interaction in wallenbergs lateral medullary syndrome. Brain  1990; 113 (Pt 3): 821– 46. Google Scholar CrossRef Search ADS PubMed  Wearne S, Raphan T, Cohen B. Control of spatial orientation of the angular vestibuloocular reflex by the nodulus and uvula. J Neurophysiol  1998; 79: 2690– 715. Google Scholar CrossRef Search ADS PubMed  Yakusheva TA, Shaikh AG, Green AM, Blazquez PM, Dickman JD, Angelaki DE. Purkinje cells in posterior cerebellar vermis encode motion in an inertial reference frame. Neuron  2007; 54: 973– 85. Google Scholar CrossRef Search ADS PubMed  Yang Y, Kim JS, Kim S, Kim YK, Kwak YT, Han IW. Cerebellar hypoperfusion during transient global amnesia: an MRI and Oculographic Study. J Clin Neurol  2009; 5: 74– 80. Google Scholar CrossRef Search ADS PubMed  Yu-Wai-Man P, Gorman G, Bateman DE, Leigh RJ, Chinnery PF. Vertigo and vestibular abnormalities in spinocerebellar ataxia type 6. J Neurol  2009; 256: 78– 82. Google Scholar CrossRef Search ADS PubMed  Zupan LH, Peterka RJ, Merfeld DM. Neural processing of gravito-inertial cues in humans. I. Influence of the semicircular canals following post-rotatory tilt. J Neurophysiol  2000; 84: 2001– 15. Google Scholar CrossRef Search ADS PubMed  © The Author(s) (2018). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

BrainOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial