The association of head shaking nystagmus with head-bending and lying-down nystagmus in horizontal canal benign paroxysmal positional vertigo

Abstract

BACKGROUND:

In benign paroxysmal positional vertigo (BPPV), the otolithic debris may alter the dynamics of the endolymph or cupula during head-shaking. This dynamic may generate head-shaking nystagmus (HSN) but exact pathomechanism of HSN in BPPV has not been elucidated. The association of positional nystagmus induced by head-bending or lying-down with HSN may help to understand the dynamics of HSN.

OBJECTIVE:

To assess the presence, pattern, and relationship with head-bending nystagmus (HBN) and lying-down nystagmus (LDN) of HSN in horizontal canal (HC)-BPPV.

METHODS:

We recruited 173 patients with HC-BPPV (76 geotropic and 97 apogeotropic). We analyzed the pattern of HSN, and correlation with HBN and LDN.

RESULTS:

Half of patients (83/173, 48%) with HC-BPPV showed HSN. The directional preponderance of HSN was also not found in patients with geotropic or apogeotropic HC BPPV (p = 0.488). The presence of HSN was related with the occurrence of HBN in both geotropic (p = 0.005) and apogeotropic type (p = 0.001). The direction of HSN was same with HBN and was opposite to LDN in both geotropic and apogeotropic type.

CONCLUSIONS:

HSN was frequently found in patients with HC-BPPV and related with HBN and LDN. HSN in BPPV might be contributed by the otolith movements related with endolymph dynamics.

Keywords

Benign paroxysmal positional vertigo pathologic nystagmus semicircular canals

1 Introduction

Head-shaking nystagmus (HSN) is a useful bedside test demonstrating a left-right asymmetry in dynamic vestibular function[2 , 18]. In unilateral peripheral vestibulopathy, the HSN is explained by a central velocity-storage mechanism and an increasing peripheral vestibular input in the healthy side compared to the impaired side (i.e., vestibular asymmetry) according to Ewald second law [4]. In central vestibulopathy, HSN indicates a central vestibular tone imbalance leading to a direction-dependent asymmetric charge/discharge of the velocity storage mechanism due to non-linearities [2, 7].

In relation to velocity storage mechanism, central or peripheral vestibular asymmetry is essential for generating HSN. However, there have been some studies reported about HSN in benign paroxysmal positional vertigo (BPPV), which is not related with vestibular asymmetry [10 , 14]. BPPV involving the horizontal canal (HC-BPPV) is characterized by short-lasting episodes of positional vertigo and direction-changing horizontal nystagmus induced by supine head roll to either side while supine. HC-BPPV is not because of vestibular asymmetry from pathologic disease but because of mechanical changes from movement of free floating otolithic debris in the endolymph of the semicircular canal (i.e., canalolithiasis) or debris near or attached to the cupula (i.e., cupulolithiasis) [5, 19]. According to a previous study, the asymmetric BPPV stimulus is probably too weak to charge the velocity integrator and generate the HSN in HC-BPPV [14]. Furthermore, HSN did not have localizing value to indicate the affected side [10], especially in geotropic HC-BPPV [13] and this does not support that HSN in BPPV is from asymmetry of vestibular system. Because the velocity storage mechanism can’t support the mechanism of HSN in HC-BPPV and the otolithic debris in the endolymph or attached to the cupula may alter the dynamics of the endolymph or the cupula during horizontal head-shaking [17], we assumed that this dysnamics may affect to the generation of HSN in HC-BPPV. Positional nystagmus induced by head-bending and lying-down reflects dynamics of otoconia in BPPV. By investigating the association between HSN and head-bending nystagmus (HBN) and lying-down nystagmus (LDN), we could provide an indirect evidence to support the assumption that HSN in HC-BPPV is related with otoconial movements inside the canal. The purpose of our study was to describe the presence and pattern of horizontal HSN, and relationship with HBN and LDN in HC-BPPV, and to suggest a possible pathomechanism of HSN based on those findings.

2 Material and methods

We prospectively collected patients with a diagnosis of HC-BPPV, who visited Keimyung University Dongsan medical center from January 2012 to February 2014. We analyzed the presence and pattern, and association with HBN and LDN of HSN. We asked the patients when the positional vertigo started to get information about the duration of symptoms.

2.1 Inclusion and exclusion criteria

To determine HC-BPPV, patients were tested by the supine head roll test. The diagnosis of HC-BPPV was based on the following: 1) a history of brief episodes of positional vertigo, 2) direction changing horizontal nystagmus beating toward the ground (i.e., geotropic nystagmus) or to the ceiling (i.e., apogeotropic nystagmus) in head turned to either side in the supine position and 3) no other identifiable cause of the central nervous system. We excluded patients with poor cooperation, multiple canals involvement, secondary causes of BPPV such as head trauma, vestibular migraine, Meniere’s disease or other inner ear diseases, identifiable CNS disorders, or undetermined lesion side. To exclude patients with underlying peripheral or central vestibular disorders, all patients received detailed bedside neuro-otologic examinations including spontaneous and gaze-evoked nystagmus, horizontal head impulse test, horizontal and vertical smooth pursuit and saccades, limb ataxia, and balance function in addition to routine neurologic examinations. The patients with BPPV involving vertical canals or multi-canals were excluded with Dix-Hallpike test. The affected side in HC-BPPV was determined by the difference of the intensities of nystagmus on both sides during supine head roll. The asymmetry of nystagmus during supine head roll was indicated by the symmetry index (SI) that was calculated as (SPV_Rt–SPV_Lt)/(SPV_Rt + SPV_Lt)×100 where SPV is the maximal slow phase velocity of each nystagmus. We only include HC-BPPV patients with greater than 10% of SI [12].

2.2 Neuro-otologic evaluation

Nystagmus was first observed without fixation using a video-Frenzel goggle system (SLMED, Seoul, Republic of Korea). Eye movements were also recorded using 3-dimensional video-oculography (SMI, Teltow, Germany). Supine head roll, Dix-hallpike test, head-bending, lying-down, and head-shaking test were performed in order (Fig. 1). The affected ear was determined by comparing the intensity of the nystagmus, with an assumption that the induced nystagmus is more intense when the head is rotated to the affected side in the geotropic type and to the intact side in the apogeotropic type in HC-BPPV.

Fig. 1

Flow diagram of study population. HC-BPPV=horizontal canal benign paroxysmal positional vertigo.

Patients were asked to sit in the head upright position with the eyes looking forward during 30 s and then to bend the head at 60° forward the pitch axis for 50 s for the observation of nystagmus (i.e., HBN) [12]. Thereafter, the patients were asked to take an upright-sitting position with the eyes looking forward for 30 s and then to quickly lie down for 50 s; the nystagmus (i.e., LDN) was then examined [6].

Patients performed passive head-shaking maneuver twice. The head was tilted 30 degrees forward in the plane of the horizontal semicircular canal and was shaken in a sinusoidal pattern at the frequency of 2 to 3 Hz for 20 cycles. We measured the maximal SPV of the first continuous 3 beats of HSN with the patient’s head tilted at the first head-shaking maneuver. The presence of HSN was defined only when the SPV of the induced nystagmus exceeded 3 degrees per second and when the nystagmus lasted more than 5 seconds [13]. When pseudo-spontaneous nystagmus was shown, the presence of HSN was defined only when the SPV of the induced nystagmus exceeded 3 degrees per second after subtracting the SPV of pseudo-spontaneous nystagmus. In the second head-shaking maneuver, the nystagmus was observed with raised their chin after head-shaking with head tilt for excluding the effect of head tilt on HSN.

We used Chi-squared test for comparison of sex, affected side, the presence and pattern of HSN, HBN and LDN between geotropic and apogeotropic groups. We performed Mann-Whitney U test for comparison of age, duration of symptoms and the velocity of HSN, HBN and LDN between two groups. All analysis was performed using SPSS version 22 (SPSS, Chicago, IL). A statistically significant association between an exposure and the outcome was declared at a p-value <0.05. All of the experiments complied with the tenets of the Declaration of Helsinki, and the study protocol was also reviewed and approved by Institutional Review Board of Keimyung University Dongsan Medical Center.

3 Results

173 patients with HC-BPPV (76 geotropic and 97 apogeotropic types) were included in this study. Female was predominant (56.1%). Mean age was 61.7 years (SD = 14.0). The sex did not differ between geotropic and apogeotropic HC-BPPV, but apogeotropic HC-BPPV were older than geotropic group (mean age, 63.9±15.2 vs. 58.9±11.9 years, p = 0.016). The symptoms duration from attack to the test showed no difference between two groups. (p = 0.102) The affected side also did not differ between two groups (left ear involvement: 53.9% vs. 46.4%, p = 0.360) (Table 1).

Table 1
Demographics of patients with horizontal canal benign paroxysmal positional vertigo

Geotropic type N = 76 Apogeotropic type N = 97 P value

Age (years) 58.9±11.9 63.9±15.2 0.016

Sex (male), n (%) 29(38.2%) 47(48.5%) 0.217

Duration of symptom (days) 10.4±16.5 13.6±26.7 0.102

Affected side (Lt), n (%) 41(53.9%) 45(46.4%) 0.360

	Geotropic type N = 76	Apogeotropic type N = 97	P value
Age (years)	58.9±11.9	63.9±15.2	0.016
Sex (male), n (%)	29(38.2%)	47(48.5%)	0.217
Duration of symptom (days)	10.4±16.5	13.6±26.7	0.102
Affected side (Lt), n (%)	41(53.9%)	45(46.4%)	0.360

Totally, 48% (83/173) of the patients with HC-BPPV showed HSN. There was no difference of presence of HSN between patients with geotropic HC-BPPV (31/76, 40.8%) and apogeotropic HC-BPPV (52/97, 53.6%) (p = 0.125). (Supplementary table 1) The HSN was horizontal in all patients with geotropic and apogeotropic HC-BPPV. In geotropic HC-BPPV group, 54.8% (17/31) patients showed HSN beating toward the affected side and the others showed HSN beating away from the affected side. In apogeotropic HC-BPPV group, 51.9 % (27/52) patients showed HSN to the affected side and the others showed HSN to the healthy side. There was no difference of the direction of HSN between geotropic and apogeotropic type (p = 0.488). (Supplementary table 2) The maximal SPV of HSN was 9.72±4.57 degree/sec and did not different between geotropic and apogeotropic types (9.45±4.95 vs. 9.88±4.38, p = 0.327). (Supplementary table 3) The SPV of HSN was significantly larger than that of HBN (p = 0.00). When the patient raised their chin after head-shaking, the direction was same to HSN done by head tilting 30° forward in all patients (Supplementary video).

There were no differences of the number of patients showed HBN (p = 0.120) and LDN (p = 0.147) between geotropic and apogeotropic type. (Supplementary table 1) HBN and LDN showed no directional preponderance in both geotropic and apogeotropic type HC-BPPV. (Supplementary table 2). The maximal SPV of HSN were not different between geotropic and apogeotropic type. (Supplementary table 3)

The occurrence of HSN was significantly related with the presence of HBN in both geotropic (p = 0.005) and apogeotropic HC-BPPV (p = 0.001) (Table 2). The direction of HSN was also same to that of HBN in geotropic (p = 0.000) and apogeotropic HC-BPPV (p = 0.000) (Table 3), whereas was opposite to that of LDN in geotropic (p = 0.005) and apogeotropic HC-BPPV (p = 0.000). Illustrative case with HSN and HBN of same direction in apogeotropic HC-BPPV is shown in supplementary video.

Table 2

Comparison of the presence of head-bending and lying-down nystagmus in patients with or without head-shaking nystagmus in horizontal canal benign paroxysmal positional vertigo

		Presence of HBN	Presence of LDN
Geotropic type (N = 76)	Patients with HSN (n = 31)	22 (71.0%)	24 (77.4%)
	Patients without HSN (n = 45)	17 (37.8%)	20 (45.5%)
	P value	0.005	0.008
Apogeotropic type (N = 97)	Patients with HSN (n = 52)	41 (78.8%)	39 (75%)
	Patients without HSN (n = 45)	21 (46.7%)	29 (64.4%)
	P value	0.001	0.275

HSN (head-shaking nystagmus), HBN (head-bending nystagmus), LDN (lying-down nystagmus).

Table 3

Comparison of the direction of head-bending nystagmus in patients with head-shaking nystagmus in horizontal canal benign paroxysmal positional vertigo

		Patients with HBN to affected side	Patients with HBN to healthy side	Patients with no HBN
Geotropic type with HSN (n = 31)	Patients with HSN to affected side (n = 17)	9 (52.9%)	3 (17.6%)	5 (29.4%)
	Patients with HSN to healthy side (n = 14)	1 (7.1%)	9 (61.3%)	4 (28.6%)
	P value	0.000
Apogeotropic type with HSN (n = 52)	Patients with HSN to affected side (n = 27)	18(66.7%)	3 (11.1%)	6 (22.2%)
	Patients with HSN to healthy side (n = 25)	1(4%)	19 (76%)	5 (20%)
	P value	0.000

HSN (head-shaking nystagmus), HBN (head-bending nystagmus).

4 Discussion

About half of the patients with HC-BPPV showed HSN, and HSN was significantly related with the presence and direction of HBN in both geotropic and apogeotropic HC-BPPV. Previous studies have also assessed the presence and pattern of HSN in HC-BPPV, but this is the first study investigating the association between HSN, HBN and LDN in a large sample size of patients with HC-BPPV. We found the presence of relationship between HSN and HBN in HC-BPPV and suggested that pathophysiology of HSN in HC-BPPV may be related with movement of otoconia rather than velocity-storage mechanism. It has been thought that HSN is a sensitive test to detect central or peripheral vestibulopathy, however, our results showed that HC-BPPV could show HSN not due to vestibular imbalance but due to dynamics of otoconia.

HSN may be occurred after 20 cycles of passive head rotation at a rate of about 2 to 3 Hz with the head tilted 30 degrees forward in the plane of the horizontal semicircular canal. Generally, it is known that normal subjects have no HSN or only 1 to 2 beats [20], and in even healthy older adults, the incidence of HSN was only 2% [3]. Unlike previous studies (6–8) that 20–30% of patients with HC-BPPV showed HSN, overall 50% of patients showed HSN in HC-BPPV in our study. HSN was more common in apogeotropic than in geoptropic type in the previous study [13], but there was no significant difference of the number of patients showed HSN between geotropic and apogeotropic HC-BPPV, even the absolute number is still higher in apogeotropic type in our study. In terms of the pattern on HSN in patients with HC-BPPV, previous studies showed contradictory findings. One study reported that BPPV did not have specific patterns of HSN [10]. Another study reported that HSN in apogeotropic HC-BPPV beat predominantly away from the affected side, whereas patients with geotropic type did not show any directional preponderance of HSN [13]. The authors suggested that HSN may be a lateralizing sign in apogeotropic HC-BPPV. However, in our study, HSN did not have any localizing values in patients with HC-BPPV of both types. This finding supports that HSN in HC-BPPV may be not related with vestibular asymmetry and velocity-storage mechanism. The presence and pattern of HSN are strongly related with HBN in HC-BPPV.

Peripheral vestibular asymmetry induces an asymmetry of velocity storage after head-shaking because there is a greater peripheral input when the head rotates toward the intact side (Ewald’s second law) and the intact labyrinth will not be able to decrease its discharge below zero for head rotations toward the paretic side at high speeds. Similarly, the asymmetry of velocity storage ascribed to disruption of ipsilesional nodulo-uvular inhibition of the velocity storage mechanism may also be responsible of HSN in central vestibulopathy. Because HC-BPPPV is not related with asymmetric velocity storage, we could consider another possibility as the mechanism of HSN in HC-BPPV. It has been suggested that head-shaking test applies on a strong mechanical stimulus on the otolithic mass. With its alternate accelerating and decelerating power, head-shaking may cause detachment of the debris from the cupula or migration of the particles from the anterior into the posterior arm of the HC. The presence and direction of HSN in HC-BPPV were definitely correlated with HBN in our study. Therefore, we presume that the head-shaking can make the movement of the otolithic debris and head shaking’s effect is similar to one seen when head bending is occurred to semicircular canal. Because of a significant concordance rate in the direction of HSN and HBN in HC-BPPV, we assume that the mechanical effect from endolymph movement by the otolith debris is more effective than vestibular physiology (i.e., vestibular asymmetry according to Ewald second law) in producing HSN, which usually explains the mechanism of HSN in patients with vestibular imbalance. Density and viscosity of otolith and endolymph, and gravity during head oscillation may contribute the dynamics of otolith inside the canal and may make HSN. Because two maneuvers require similar head position (i.e., head tilting forward 30° and 60° for head-shaking test and head-bending maneuver, respectively), we could doubt whether the similarities between HSN and HBN may be originated from bending effect of the head. However, when patients raise their chin after head-shaking, the direction were same with HSN and the SPV of HSN with head bending were larger than that of HBN. Therefore, the possibility that head-bending movement itself during head-shaking may have a primary role in producing the nystagmus after head-shaking, would be low.

HBN and LDN have been suggested as a valuable sign for lateralizing the involved canal in HC-BPPV, especially when patients show symmetrical nystagmus during supine head roll test [6 , 12]. The direction of HBN is mostly toward the affected side and LDN is mostly beat to the healthy side in geotropic type, whereas the directions of HBN and LDN in apogeotropic type were usually opposite to those seen in geotropic type [1 , 15]. However, other studies showed a contradictive finding that LDN and HBN did not seem to have a lateralizing value in patients with HSC-BPPV [16, 21]. Our study also showed that the direction of HBN and LDN did not reflect the affected side in HC-BPPV.

Our study has some limitations. First, the main limitation of our study is that we cannot show a clear evidence for the physiology proposed of HSN in HC-BPPV because our study is not an experimental study but a clinical study. We need a further investigation to demonstrate a direct mechanism for HSN using an experimental BPPV model. Second, because BPPV can be affected with head position, preceding positional test could affect the results of head shaking. However, considering a large sample of the patients, our main finding on definite association between HSN and HBN is a clinically relevant one. Third, because we did not perform caloric test or video head impulse test and brain imaging, we could not completely exclude secondary causes of BPPV. However, we believed that most of patients were categorized as having no definite cause of BPPV through detailed history taking and neurological examinations.

Author contributions statement

Drs. HA Kim and H Lee conducted the design and conceptualization of the study, interpretation of the data, and drafting and revising the manuscript. Dr. HA Kim wrote the manuscript and analyzed the data.

Conflict of interest statement

None.

Footnotes

Acknowledgments

Dr. Lee serves on the editorial boards of the Research in Vestibular Science, Frontiers in Neuro-otology, and Current Medical Imaging Review.

Dr. Kim reports no disclosures.

Source of Funding: “This work was supported by the National Research Foundation of Korea (NRF) Grant funded by the Korea Government (MSIP) (No. 2014R1A5A2010008)”

The supplementary material is available in the electronic version of this article: .

References

Asprella Libonati

, Diagnostic and treatment strategy of lateral semicircular canal canalolithiasis, Acta otorhino-laryngologica italica 25 (2005), 277–283.

Choi

K.D.

, Oh

S.Y.

, Park

S.H.

, Kim

J.H.

, Koo

J.W.

and Kim

J.S.

, Head-shaking nystagmus in lateral medullary infarction: patterns and possible mechanisms, Neurology 68 (2007), 1337–1344.

Davalos-Bichara

and Agrawal

, Normative results of healthy older adults on standard clinical vestibular tests, Otology & Neurotology 35 (2014), 297–300.

Hain

T.C.

, Fetter

and Zee

D.S.

, Head-shaking nystagmus in patients with unilateral peripheral vestibular lesions, American Journal of Otolaryngology 8 (1987), 36–47.

Hall

S.F.

, Ruby

R.R.

and McClure

J.A.

, The mechanics of benign paroxysmal vertigo, Journal of Otolaryngology 8 (1979), 151–158.

Han

B.I.

, Oh

H.J.

and Kim

J.S.

, Nystagmus while recumbent in horizontal canal benign paroxysmal positional vertigo, Neurology 66 (2006), 706–710.

Huh

Y.E.

and Kim

J.S.

, Patterns of spontaneous and head-shaking nystagmus in cerebellar infarction: imaging correlations, Brain 134 (2011), 3662–3671.

Huh

Y.E.

, Koo

J.W.

, Lee

and Kim

J.S.

, Head-shaking aids in the diagnosis of acute audiovestibular loss due to anterior inferior cerebellar artery infarction, Audiology and Neuro-Otology 18 (2013), 114–124.

Kim

H.A.

, Park

S.W.

, Kim

, Kang

B.G.

, Lee

, Han

B.I.

, Seok

J.I.

, Chung

E.J.

, Kim

and Lee

, Efficacy of mastoid oscillation and the Gufoni maneuver for treating apogeotropic horizontal benign positional vertigo: a randomized controlled study, Journal of Neurology 264 (2017), 848–855.

10.

Kim

M.B.

, Huh

S.H.

and Ban

J.H.

, Diversity of head shaking nystagmus in peripheral vestibular disease, Otology & Neurotology 33 (2012), 634–639.

11.

Koo

J.W.

, Moon

I.J.

, Shim

W.S.

, Moon

S.Y.

and Kim

J.S.

, Value of lying-down nystagmus in the lateralization of horizontal semicircular canal benign paroxysmal positional vertigo, Otology & Neurotology 27 (2006), 367–371.

12.

Lee

S.H.

, Choi

K.D.

, Jeong

S.H.

, Oh

Y.M.

, Koo

J.W.

and Kim

J.S.

, Nystagmus during neck flexion in the pitch plane in benign paroxysmal positional vertigo involving the horizontal canal, Journal of the Neurological Sciences 256 (2007), 75–80.

13.

Lee

S.U.

, Kim

H.-J.

and Kim

J.S.

, Pseudo-spontaneous and head-shaking nystagmus in horizontal canal benign paroxysmal positional vertigo, Otology & Neurotology 35 (2014), 495–500.

14.

Lopez-Escamez

J.A.

, Zapata

, Molina

M.I.

and Palma

M.J.

, Dynamics of canal response to head-shaking test in benign paroxysmal positional vertigo, Acta Oto-Laryngologica 127 (2007), 1246–1254.

15.

Nuti

, Mandalàe;

and Salerni

, Lateral canal paroxysmal positional vertigo revisited, Annals of the New York Academy of Sciences 1164 (2009), 316–323.

16.

J.H.

, Song

S.K.

, Lee

J.S.

, Choi

J.C.

, Kang

S.Y.

and Kang

J.H.

, Lying-down nystagmus and head-bending nystagmus in horizontal semicircular canal benign paroxysmal positional vertigo: are they useful for lateralization? BMC Ophthalmology 14 (2014), 136–136.

17.

S.Y.

, Kim

J.S.

, Jeong

S.H.

, Oh

Y.M.

, Choi

K.D.

, Kim

B.K.

, Lee

S.H.

, Lee

H.S.

, Moon

I.S.

and Lee

J.J.

, Treatment of apogeotropic benign positional vertigo: comparison of therapeutic head-shaking and modified Semont maneuver, Journal of Neurology 256 (2009), 1330–1336.

18.

Pérez

, Llorente

J.L.

, Gómez

J.R.

, Del Campo

, López

and Suárez

, Functional significance of peripheral head-shaking nystagmus, Laryngoscope 114 (2004), 1078–1084.

19.

Schuknecht

H.F.

, Cupulolithiasis, Archives of Otolaryngology 90 (1969), 765–778.

20.

Welgampola

M.S.

, Bradshaw

A.P.

, Lechner

and Halmagyi

G.M.

, Bedside Assessment of Acute Dizziness and Vertigo, Neurologic Clinics 33 (2015), 551–vii.

21.

Yetiser

and Ince

, Diagnostic Role of Head-Bending and Lying-Down Tests in Lateral Canal Benign Paroxysmal Positional Vertigo, Otology & Neurotology 36 (2015), 1231–1237.