Journal of the Neurological Sciences 368 (2016) 249–253
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Clinical short communication
Spontaneous, headshaking, and positional nystagmus in post-lateral medullary infarction dizziness Kazumitsu Amari a, Yosuke Kudo b, Kosuke Watanabe b, Masahiro Yamamoto b, Koji Takahashi c, Osamu Tanaka c, Ken Johkura b,⁎ a b c
Department of Neuroendovascular Therapy, Yokohama Brain and Spine Center, Yokohama, Japan Department of Neurology, Yokohama Brain and Spine Center, Yokohama, Japan Department of Clinical Laboratory, Yokohama Brain and Spine Center, Yokohama, Japan
a r t i c l e
i n f o
Article history: Received 3 May 2016 Received in revised form 25 June 2016 Accepted 11 July 2016 Available online 15 July 2016 Keywords: Lateral medullary infarction Dizziness Spontaneous nystagmus Head-shaking nystagmus Positional nystagmus
a b s t r a c t Background and purpose: Lateral medullary infarction (LMI) sometimes causes long-lasting dizziness. However, the characteristics of nystagmus in patients with post-LMI dizziness are unknown. We undertook a prospective, comparative study of nystagmus in patients with and without post-LMI dizziness to determine the characteristic pattern of nystagmus of chronic post-LMI dizziness. Methods: We evaluated and compared nystagmus under spontaneous, head-shaking, and positional testing conditions in 12 patients with post-LMI dizziness and in 6 patients without post-LMI dizziness. Results: In the dizziness group, contralateral spontaneous nystagmus, ipsilateral head-shaking nystagmus, and horizontal direction-changing geotropic positional nystagmus were observed in patients in whom the LMI had occurred b60 days previously (subacute period). In patients with dizziness in whom the LMI had occurred N 90 days previously (chronic period), the nystagmus was ipsilateral under all conditions. In the non-dizziness group, ipsilateral nystagmus was observed in 1 of the 2 subacute patients only after head-shaking and in 1 of the 4 chronic patients only during positional testing. Conclusions: Ipsilateral nystagmus observed under all spontaneous, head-shaking, and positional testing conditions characterizes chronic post-LMI dizziness. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Lateral medullary infarction (LMI), or Wallenberg syndrome, is a well-known vascular syndrome, usually caused by occlusion of the posterior inferior cerebellar artery or vertebral artery [1]. Vertigo and/or dizziness, usually associated with other neurologic symptoms such as gait instability, dysphagia, hoarseness, impaired facial and contralateral limb/body sensation, is common in the acute phase of LMI [2]. Dizziness can persist in the chronic phase of LMI [3]. However, the characteristics and mechanism of chronic post-LMI dizziness are not well described. In LMI, various patterns of nystagmus have been reported. Spontaneous contralateral beating nystagmus (beating away from the lesion side), presumably originating from ipsilateral central vestibular disruption [4], are frequently observed. Ipsilateral beating nystagmus (beating
⁎ Corresponding author at: Department of Neurology, Yokohama Brain and Spine Center, 1-2-1 Takigashira, Isogo-ku, Yokohama 235-0012, Japan. E-mail address:
[email protected] (K. Johkura).
http://dx.doi.org/10.1016/j.jns.2016.07.019 0022-510X/© 2016 Elsevier B.V. All rights reserved.
toward the lesion side) is also seen after head shaking [5]. Although head-shaking nystagmus is caused by various mechanisms, ipsilateral beating nystagmus following head-shaking in LMI is thought to result from ipsilateral damage of cerebellar inhibition on the velocity-storage mechanism of the vestibulo-ocular reflex (VOR) [5,6]. To determine the characteristic pattern of nystagmus in patients with chronic post-LMI dizziness, we conducted a prospective, comparative study to evaluate spontaneous, head-shaking, and positional nystagmus in patients with and without post-LMI dizziness. 2. Methods Because our institution, Yokohama Brain and Spine Center, is an acute stroke care and rehabilitation center, not only acute but also chronic stroke patients are hospitalized. Among these patients, between September 2014 and August 2015, we recruited 18 consecutive patients with pure LMI for participation in the study. All 18 were enrolled regardless of the time from LMI onset, but patients with a history of another neurologic or psychologic disease were excluded. Patients with a definite diagnosis of a peripheral vestibular disorder, such as benign paroxysmal positional vertigo, vestibular neuritis, or Ménière's disease,
250
K. Amari et al. / Journal of the Neurological Sciences 368 (2016) 249–253
Table 1 Study patients, including their clinical characteristics and study variables. Patient/Age/Sex
Time from LMI onset (days)
Dizziness (DHI score)
Lesion side
Spontaneous nystagmus (velocity, °/s; amplitude, °)a
Head-shaking nystagmus (velocity, °/s; amplitude, °)a
Positional nystagmus (velocity, °/s; amplitude, °)a
Sensory topography
1/43/M
4
L
2/53/M
4
3/41/M
7
4/49/M
12 25
6/50/M
57
7/80/M
90
8/67/M
102
9/54/M
461
10/45/M
711
11/61/M
1245
12/48/F
1607
13/66/M
34
Ipsilateral (9.18; 8.97) Ipsilateral (8.17; 4.31) Ipsilateral (4.87; 4.54) Ipsilateral (2.58; 3.24) Ipsilateral (9.01; 8.36) Ipsilateral (6.50; 6.58) Ipsilateral (8.58; 3.72) Ipsilateral (9.68; 5.00) Ipsilateral (2.90; 5.88) Ipsilateral (5.58; 5.37) Ipsilateral (5.84; 8.68) Ipsilateral (6.82; 5.67) None
14/49/F
35
15/46/M
368
Contralateral (4.53; 4.10) Contralateral (1.45; 2.99) Contralateral (3.44; 3.41) Contralateral (2.66; 6.52) Contralateral (3.35; 6.30) Contralateral (3.59; 5.62) Ipsilateral (1.81; 2.62) Ipsilateral (5.02; 4.32) Ipsilateral (1.68; 4.91) Ipsilateral (2.54; 3.31) Ipsilateral (3.34; 8.80) Ipsilateral (3.84; 2.49) Contralateral (6.84; 6.41) Contralateral (1.50; 2.19) None
Geotropic (10.67; 13.64) Ipsilateral (7.21; 6.16) Geotropic (12.49; 4.34) Geotropic (6.04; 5.84) Geotropic (2.99; 6.01) Geotropic (2.51; 5.16) Ipsilateral (4.89; 3.00) Ipsilateral (3.89; 5.72) Ipsilateral (2.93; 5.44) Ipsilateral (4.75; 5.81) Ipsilateral (10.95; 10.39) Ipsilateral (2.15; 2.76) Geotropic (4.04; 3.05) Contralateral (3.07; 3.29) None
Contralateral face Contralateral limbs Ipsilateral face Contralateral upper limb Ipsilateral face Contralateral limbs Contralateral upper limb
5/58/M
16/63/M
1008 1327
Contralateral (1.79; 2.38) None
None
17/55/M
Ipsilateral (1.55; 2.30) None
18/59/M
1927
+ (NE) + (NE) + (NE) + (NE) + (26) + (28) + (48) + (50) + (48) + (20) + (28) + (20) − (NE) − (0) − (0) − (0) − (0) − (0)
Ipsilateral face Contralateral limbs Ipsilateral face Contralateral limbs/body Contralateral limbs/body
R R L R L R R R R L L L L L L R L
Contralateral (2.42; 3.94)
Ipsilateral (2.65; 2.08) None
Contralateral (3.84; 8.67) Contralateral (6.24; 6.40)
Contralateral (4.86; 5.68)
Ipsilateral face Contralateral limbs Ipsilateral face Ipsilateral face Contralateral limbs Ipsilateral face Contralateral limbs Ipsilateral face Contralateral limbs/body Contralateral limbs Ipsilateral face Contralateral limbs/body Ipsilateral face Ipsilateral face Contralateral limbs Ipsilateral face
DHI, Dizziness Handicap Inventory; +dizziness present; −dizziness absent; NE, not examined; VOR, vestibulo-ocular reflex; C-VEMP, cervical vestibular evoked myogenic potential; O-VEMP, ocular vestibular evoked myogenic potential. Fr = {(VOR gain in darkness) – (VOR gain during fixation)}/(VOR gain in darkness) × 100(%). a Nystagmus are shown with the mean slow-phase velocities (degree/s, °/s) and amplitudes (degree, °). b Mean ± SD in age-matched normal controls (n = 16) are shown in parentheses as normative values.
were also excluded. Ischemic lesions in the lateral medulla were confirmed by diffusion-weighted MRI (DWI) performed within 1 day after LMI onset. The 18 patients were divided into 2 groups according to the presence or absence of dizziness: 12 with dizziness (dizziness group) and 6 patients without dizziness (non-dizziness group). For the purpose of the study, LMI was classified as subacute or chronic, depending on the time that had passed since LMI onset, with subacute LMI referring to the stroke phase 0–60 days after LMI onset and chronic stroke referring to the stroke phase beyond 60 days. The severity of patients' dizziness was scored according to the Dizziness Handicap Inventory (DHI), which consists of 25 questions representing the impact on daily life in 3 domains: the physical, functional, and emotional domains [7]. The DHI was applied only during the chronic phase of LMI because it is intended only for evaluation of chronic dizziness. Although dizziness in the subacute phase patients was not scored, all patients with dizziness in the subacute phase could not walk without assistance because of unsteady feeling. Spontaneous nystagmus, head-shaking nystagmus, and positional nystagmus were evaluated in both the dizziness group and the non-dizziness group. Nystagmus was recorded in darkness by means of video-oculography performed with the use of Frenzel goggles and a built-in charge-coupled device camera with infrared illumination. Head-shaking nystagmus was induced by application of a passive head-shaking maneuver; the patient's head was manually shaken horizontally in a sinusoidal fashion at a rate of about 2.8 Hz with an amplitude of about ± 10° for 15 s
[5]. In addition to nystagmus, the horizontal VOR, cervical vestibular evoked myogenic potential (C-VEMP), and ocular (O)-VEMP were evaluated. Horizontal VORs were recorded in darkness by means of a videooculography-based VOR recording and analysis system (IRN-2, Morita Mfg. Corp., Kyoto, Japan) [8], with the patient seated on a rotating armchair with a headrest (S-2, Nagashima Medical Instruments Corp., Tokyo, Japan). The chair was programed to rotate in sinusoidal fashion for 30 s (frequency, 0.6 Hz; amplitude, 60°). VOR gain, i.e., the eye velocity to head velocity ratio, was determined. After recording of the VOR, fixation suppression of the VOR was tested by asking the patient to fixate on a target (a red circle, 1 cm in diameter) located 50 cm from the eyes and rotating with the armchair. The fixation suppression rate (Fr) was calculated on the basis of the VOR gain in darkness (G) and the VOR gain during fixation (Gf) according to the following equation: Fr = (G − Gf)/G × 100(%) [9]. C-VEMP was recorded from the tonically contracting ipsilateral sternocleidomastoid muscle during monoaural stimulation with 105-dB, 500-Hz short tone bursts (MEB 2312 testing system, Nihon Kohden, Tokyo, Japan; bandwidth 20–2000 Hz, 200 averaged signals). Latencies of the first positive wave (p13) and second negative wave (n23) and peak-to-peak p13–n23 amplitude were measured [10]. O-VEMP was recorded from the contralateral lower eyelid (inferior oblique muscles) during stimulation with 105-dB, 500-Hz short tone bursts (MEB 2312, Nihon Kohden; bandwidth 20– 2000 Hz, 200 averaged signals). The patient was instructed to
K. Amari et al. / Journal of the Neurological Sciences 368 (2016) 249–253
251
Table 1 Study patients, including their clinical characteristics and study variables. Infarct area (mm2)
VOR gain (-0.45 ± 0.15)b
Fr (78.58 ± 11.81)b
C-VEMP amplitude (μV) (28.91 ± 18.12)b
O-VEMP amplitude (μV) (38.84 ± 6.02)b
Ipsi-lateral
Contra-lateral
Ipsi-lateral
Contra-lateral
Ipsi-lateral
Contra-lateral
Ipsi-lateral
Contra-lateral
+
−0.64
−0.36
37.50
97.22
172.03
433.35
2.94
9.20
34.96
+
−0.22
−0.10
0.00
20.00
453.16
204.59
4.84
2.41
36.09
+
−0.62
−0.13
33.89
76.92
183.87
503.60
7.16
14.71
35.50
+
−0.59
−0.51
93.22
101.96
90.42
61.78
14.35
22.70
24.52
−
−0.61
−0.32
73.77
84.38
202.00
673.55
3.04
16.22
30.75
−
−0.54
−0.39
64.81
97.44
119.42
173.21
2.96
5.47
29.79
−
−0.50
−0.15
40.00
-
209.59
532.99
2.87
5.55
25.25
−
−0.49
−0.35
-
37.14
40.78
87.58
0.00
2.53
38.23
−
−0.64
−0.52
26.56
38.46
229.03
474.30
8.50
26.28
38.91
−
−0.57
−0.39
57.89
66.66
162.72
304.72
7.71
8.19
32.21
−
−0.77
−0.59
35.06
47.45
404.17
378.23
12.70
6.97
20.44
−
−0.55
−0.22
47.27
40.91
82.15
358.70
5.78
11.60
30.78
+
−0.34
−0.26
76.47
92.31
289.28
237.67
22.48
11.48
34.86
−
−0.43
−0.40
74.42
85.00
456.57
508.15
2.67
5.36
20.46
−
−0.73
−0.55
60.27
61.82
495.94
540.83
20.80
19.52
26.11
−
−0.69
−0.32
20.29
59.38
494.54
153.84
30.62
4.64
55.75
−
−0.44
−0.39
38.64
61.54
32.99
268.70
10.18
27.76
50.00
−
−0.75
−0.83
82.67
74.70
155.79
292.02
24.55
22.11
26.53
Hoarseness/dysarthria
continue looking straight ahead and upward 30°. Latencies of the first negative wave (n1) and second positive wave (p1) and peakto-peak n1–p1 amplitude were measured [10]. For quantification of
ipsilateral-contralateral differences, ratios of ipsilateral/contralateral VOR gain, Fr, C-VEMP amplitude, O-VEMP amplitude were calculated.
Fig. 1. Representative diffusion-weighted image obtained soon after LMI onset (A) and video-oculographic recordings of nystagmus in the subacute phase (B) (Patient 4; 12 days after LMI onset). Contralateral spontaneous nystagmus, ipsilateral head-shaking nystagmus, and horizontal direction-changing geotropic positional nystagmus are evident.
252
K. Amari et al. / Journal of the Neurological Sciences 368 (2016) 249–253
Fig. 2. Representative diffusion-weighted image obtained soon after LMI onset (A) and video-oculographic recordings of nystagmus in the chronic phase (B) (Patient 10; 711 days after LMI onset). The nystagmus is uniformly ipsilateral under all conditions: at rest, after head shaking, and during positional testing.
In addition, patients' sensory symptom topographies and infarct areas were examined. Infarct areas were measured on diffusion-weighted images obtained within 1 day after LMI onset, by which ischemic lesions in the lateral medulla were confirmed, with the use of an image analysis system (SYNAPSE, Fujifilm Corp, Tokyo, Japan); when the lesion was traced manually on the MR image, the area was calculated automatically. Hoarseness/dysarthria, the other associated symptoms in our study patients, were also evaluated qualitatively in this study. Ratios of ipsilateral/contralateral VOR gain, Fr, C-VEMP amplitude, and O-VEMP amplitude were compared between the dizziness group and the non-dizziness group. Infarct areas were also compared between the 2 groups. Differences were analyzed by Mann-Whitney U test. PRISM software (version 6.03; GraphPad Software Inc.) was used for all statistical analyses, and p b 0.05 was considered statistically significant. The study was approved by our institutional ethics committee. Informed consent was obtained from all 18 patients for their participation in the study. 3. Results The 18 study patients, their clinical characteristics, and their study variables are shown in Table 1. Patients 1–12 comprised the dizziness group, and Patients 13–18 comprised the non-dizziness group. Patients in each group are listed according to the time from LMI onset. The high DHI scores (20–50) in the dizziness group indicate that chronic dizziness after LMI severely disturbed patients' activities of daily living. In the dizziness group, 6 patients for whom the time from LMI onset was b60 days (subacute phase; Patients 1–6) showed contralateral spontaneous nystagmus and ipsilateral head-shaking nystagmus;
horizontal direction-changing geotropic positional nystagmus was seen in 5 of these 6 patients (Table 1, Fig. 1). The remaining 6 patients for whom the time from LMI onset was 90 days or more (chronic phase; Patients 7–12) showed ipsilateral spontaneous nystagmus, ipsilateral head-shaking nystagmus, and ipsilateral nystagmus during positional testing (Table 1, Fig. 2). In the non-dizziness group, ipsilateral nystagmus was seen in 1 of the 2 subacute phase patients (Patient 14) only after head shaking and in 1 of the 4 chronic phase patients (Patients 16) only during positional testing; contralateral nystagmus were most frequently observed in the non-dizziness group (Table 1). Absolute VOR gains were greater in the ipsilateral direction (head rotated toward the lesion side) than in the contralateral direction in all patients in the dizziness group and in 5 of the 6 patients in the non-dizziness group, regardless of the direction of spontaneous nystagmus (Table 1); the mean ipsilateral/contralateral VOR gain ratio was statistically greater in the dizziness group (p = 0.03) (Table 2). The Fr tended to be lower in the ipsilateral direction than in the contralateral direction in both groups. C-VEMP amplitudes were smaller on the ipsilateral side than on the contralateral side in 13 (72.2%) of the 18 LMI patients, and O-VEMP amplitudes were smaller on the ipsilateral side in 12 (66.7%) of the 18 patients. However, there was no statistical between-group difference in the ipsilateral/contralateral Fr, O-VEMP amplitude, or C-VEMP amplitude ratios. There was no certain tendency in C-VEMP or O-VEMP latency. There was no difference in sensory topography between the dizziness group and the non-dizziness group (Table 1). Mild hoarseness/dysarthria was seen in patients soon after LMI onset in both the dizziness and nondizziness groups (Patients 1–4, and 13); chronic phase patients did not have hoarseness/dysarthria in this study (Table 1). In all patients,
Table 2 Results of vestibular examinations per study group. Vestibular examination
Dizziness group (n = 12)
Non-dizziness group (n = 6)
p Valuea
VOR gain ratio (ipsilateral/contralateral), mean (SD) Fr ratio (ipsilateral/contralateral), mean (SD) C-VEMP amplitude ratio (ipsilateral/contralateral), mean (SD) O-VEMP amplitude ratio (ipsilateral/contralateral), mean (SD)
2.04 (1.07) 0.87 (0.33) 0.72 (0.59) 0.69 (0.62)
1.32 (0.45) 0.79 (0.27) 1.15 (1.08) 1.94 (2.35)
0.03 0.54 0.30 0.13
VOR, vestibulo-ocular reflex; C-VEMP, cervical vestibular evoked myogenic potential; O-VEMP, ocular vestibular evoked myogenic potential; SD, standard deviation. VOR gain ratio = (ipsilateral VOR gain)/(contralateral VOR gain). Fr = {(VOR gain in darkness) − (VOR gain during fixation)}/(VOR gain in darkness) × 100(%). Fr ratio = (ipsilateral Fr)/(contralateral Fr). C-VEMP amplitude ratio = (ipsilateral C-VEMP amplitude)/(contralateral C-VEMP amplitude). O-VEMP amplitude ratio = (ipsilateral O-VEMP amplitude)/(contralateral O-VEMP amplitude). a By Mann-Whitney U test.
K. Amari et al. / Journal of the Neurological Sciences 368 (2016) 249–253
medullary infarctions were within 1 DWI slice (thickness 7.5 mm). The mean infarct areas in the dizziness group and the non-dizziness group were 31.45 mm2 and 35.62 mm2, respectively, and did not differ statistically (p = 0.89). 4. Discussion In the subacute phase of LMI, contralateral spontaneous nystagmus, ipsilateral head-shaking nystagmus, and horizontal direction-changing geotropic positional nystagmus were most commonly observed in the dizziness group patients (Patients 1–6); contralateral spontaneous nystagmus, ipsilateral head-shaking nystagmus, and horizontal directionchanging geotropic positional nystagmus were also observed in the non-dizziness group patients (Patients 13 and 14). There was no conspicuous difference in the nystagmus pattern between the dizziness and the non-dizziness groups patients in the subacute phase of LMI. In the chronic phase of LMI, however, the nystagmus pattern was different between the two groups; the nystagmus in the dizziness group patients (Patients 7–12) was ipsilateral under all conditions including at rest, after head shaking, and during positional testing, whereas, in the nondizziness group patients (Patients 15–18), ipsilateral nystagmus was seen in only 1 patient and only during positional testing; contralateral nystagmus were most frequently observed in the 4 chronic phase patients. Thus, ipsilateral nystagmus in all spontaneous, head-shaking, and positional testing conditions was the characteristic of chronic dizziness after LMI. Because ipsilateral head-shaking nystagmus in LMI was thought to be caused by disinhibition of the ipsilateral velocity storage mechanism of the VOR [5,6], our patients' ipsilateral head-shaking nystagmus seen in the dizziness group seemed to derive from the disinhibited (increased) ipsilateral velocity-storage mechanism. VOR gains, in this study, were calculated during 30-s 0.6 Hz sinusoidal stimulation, corresponding “slow” head-shaking. Thus, the greater ipsilateral absolute VOR gains regardless of the direction of spontaneous nystagmus in this study suggested that the velocity-storage mechanism was also increased in the lower acceleration VOR, although peripheral vestibular imbalance or asymmetric central adaptation, as a possible mechanism, could not completely be excluded. The VOR gain ratios were statistically greater in the dizziness group than in the non-dizziness group, suggesting more increased velocity-storage mechanism of the VOR in the dizziness group. Increased unilateral velocity-storage mechanism might influence on eye movements not only during sinusoidal stimulation (head-shaking) but also during other conditions. Thus, although the precise mechanism was unclear, ipsilateral nystagmus under spontaneous and positional testing conditions in the chronic phase of dizziness group patients (Patients 7–12) might also be explained by the previously reported disinhibition of the velocity-storage mechanism. Contralateral nystagmus occurring in the subacute phase of LMI, probably caused by the ipsilateral vestibular nuclear and/or nerve root damage [4,5,11], might be resolved by central compensation [5] at that time. However, like post-thalamic stroke pain [12], central disinhibition might progress over several months. To confirm this, it is necessary to observe the change of the nystagmus pattern from the subacute phase of LMI to the chronic phase in each patient. Head impulse test and caloric test, unfortunately not performed in this study, might be useful to know the ipsilateral vestibular nuclear and/or nerve root damage. In the early phase of LMI, ipsilateral spontaneous nystagmus was reported to occur when the rostral medulla was involved [5]. However, in our study, ipsilateral spontaneous nystagmus was not observed in the subacute phase of LMI; ipsilateral spontaneous nystagmus was seen only in the chronic post-LMI dizziness patients. Horizontal direction-changing geotropic positional nystagmus is usually seen in association with lateral-canal canalolithiasis-type benign paroxysmal positional vertigo [13] and is rarely observed in patients with central lesions [14]. However, in our study, horizontal direction-changing geotropic positional nystagmus was frequently observed in patients in
253
the subacute phase of LMI (Patients 1, 3–6, and 13). In more than half of these patients, amplitudes of C- and O-VEMPs, reflecting otolithmediated vestibular function, were smaller on the affected side. Otolithmediated vestibular dysfunction might cause direction-changing positional nystagmus [10]. However, it was unclear whether our patients' horizontal direction-changing geotropic positional nystagmus was actually related to an otolith-mediated vestibular dysfunction. Ocular torsion, which was also mediated by otolith, was not evaluated in this study. We found no between-group difference in the infarct areas, suggesting that infarct size is not related to the chronic dizziness seen after LMI. However, DWI studies, such as those that we performed, are not adequate for precise investigation; geometric distortion and hyperintense signals (artifacts) occur frequently in DWI of the lower brainstem [15]. Full extent of lesion might not be apparent within 1 day after LMI onset. Further detailed study of neuroanatomical correlations by another method should be necessary to conclude that infarct size is unrelated to the dizziness. We also found sensory symptom topography to be unrelated to the chronic dizziness. There was no statistical betweengroup difference in the ipsilateral/contralateral Fr, C-VEMP amplitude, or O-VEMP amplitude ratios. Thus, based on our study, ipsilateral nystagmus under spontaneous, head-shaking, and positional testing conditions is the solely informative characteristic of chronic dizziness after LMI. Conflict of interest The authors have no conflicts of interest to disclose. Acknowledgment We thank Dr. David S. Zee, of the Department of Neurology, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA, for reviewing the manuscript. References [1] R.L. Sacco, L. Freddo, J.A. Bello, J.G. Odel, S.T. Onesti, J.P. Mohr, Wallenberg's lateral medullary syndrome. Clinical-magnetic resonance imaging correlations, Arch. Neurol. 50 (1993) 609–614. [2] J.S. Kim, Pure lateral medullary infarction: clinical-radiological correlation of 130 acute, consecutive patients, Brain 126 (Pt. 8) (2003) 1864–1872. [3] G. Nelles, K.A. Contois, S.L. Valente, J.L. Higgins, D.H. Jacobs, J.D. Kaplan, M.S. Pessin, Recovery following lateral medullary infarction, Neurology 50 (1998) 1418–1422. [4] H. Rambold, C. Helmchen, Spontaneous nystagmus in dorsolateral medullary infarction indicates vestibular semicircular canal imbalance, J. Neurol. Neurosurg. Psychiatry 76 (2005) 88–94. [5] K.D. Choi, S.Y. Oh, S.H. Park, J.H. Kim, J.W. Koo, J.S. Kim, Head-shaking nystagmus in lateral medullary infarction: patterns and possible mechanisms, Neurology 68 (2007) 1337–1344. [6] K.D. Choi, J.S. Kim, Head-shaking nystagmus in central vestibulopathies, Ann. N. Y. Acad. Sci. 1164 (2009) 338–343. [7] G.P. Jacobson, C.W. Newman, The development of the Dizziness Handicap Inventory, Arch. Otolaryngol. Head Neck Surg. 116 (1990) 424–427. [8] K. Funabiki, Y. Naito, K. Matsuda, I. Honjo, A new vestibulo-ocular reflex recording system designed for routine vestibular clinical use, Acta Otolaryngol. 119 (1999) 413–419. [9] K. Johkura, T.N. Yoshida, Y. Kudo, Y. Nakae, T. Momoo, Y. Kuroiwa, Cilostazol versus aspirin therapy in patients with chronic dizziness after ischemic stroke, Clin. Neurol. Neurosurg. 114 (2012) 876–880. [10] K. Johkura, Y. Kudo, Y. Amano, K. Takahashi, Vestibular examinations in apogeotropic positional nystagmus caused by cerebellar tumor, Neurol. Sci. 36 (2015) 1051–1052. [11] T. Uemura, B. Cohen, Effects of vestibular nuclei lesions on vestibulo-ocular reflexes and posture in monkeys, Acta Otolaryngol. Suppl. 315 (1973) 1–71. [12] Z.S. Nasreddine, J.L. Saver, Pain after thalamic stroke: right diencephalic predominance and clinical features in 180 patients, Neurology 48 (1997) 1196–1199. [13] D. Nuti, P. Vannucchi, P. Pagnini, Benign paroxysmal positional vertigo of the horizontal canal: a form of canalolithiasis with variable clinical features, J. Vestib. Res. 6 (1996) 173–184. [14] P. Bertholon, S. Tringali, M.B. Faye, J.C. Antoine, C. Martin, Prospective study of positional nystagmus in 100 consecutive patients, Ann. Otol. Rhinol. Laryngol. 115 (2006) 587–594. [15] M. Iima, A. Yamamoto, V. Brion, T. Okada, M. Kanagaki, K. Togashi, D. Le Bihan, Reduced-distortion diffusion MRI of the craniovertebral junction, AJNR Am. J. Neuroradiol. 33 (2012) 1321–1325.