Biasing the semicircular canal cupula in excitatory direction decreases the gain of the vestibuloocular reflex for head impulses

1 Introduction

In a previous study we have shown that by cold thermal irrigation it is possible to reduce the gain of the angular vestibuloocular reflex (aVOR) for horizontal ipsilateral head impulses when the lateral semicircular canal is in the earth-vertical position [12]. In this experimental setup the direction of the slow phase of the ensuing thermal nystagmus was opposite to the direction of the eye movement necessary to compensate for the high acceleration, high velocity horizontal head perturbation (head impulse). In the study of Mantokoudis et al. [5], when head impulse testing (HIT) was carried out immediately after rotational stimulation, the slow phase velocity of the postrotational nystagmus and the velocity of the compensatory eye movements behaved in an additive way: the HIT gain could be decreased or increased by adding the positive or negative postrotational velocity [5]. During postrotatory nystagmus, the cupulae are deflected on both sides, on one side away from the vestibulum, on the other side vestibulopetally. The cupulae return to their position with a short latency (time constant 3-4 sec), however, the ensuing nystagmus is prolonged by central neural circuit, called the velocity integrator. By simply reviewing the data, it is not possible to identify the cause of gain changes measured by HIT during postrotatory nystagmus: does the cupular deviation play a role or is the change due to central processes? We suspected that this effect was probably brought about by the brainstem velocity integrator circuits in the experiments of Mantokoudis et al, because by the time the HIT-measurement goggles were fitted after having stopped the rotation, the cupula possibly would have returned to its neutral position. Also, in our similar experimental setup, when only one end-organ was stimulated, the effect was not elicited bilaterally (which it might have been in the case of a central mechanism). In our setup using postcaloric reaction, during the stimulation the video-goggles are in place and the impulses can be started immediately after cessation of the irrigation, when the temperature gradient is still present in the temporal bone [4]. Our hypothesis was that immediately after unilateral caloric reaction the brainstem circuits do not play a major role, only peripheral mechanisms and that biasing the cupula in either direction decreases its sensitivity.

In our previous experiments the head impulses were carried out during a sustained postcaloric deflection of the cupula in inhibitory direction. The aim of the present experiment was to answer the question if a sustained excitatory deflection increases the gain of the reflex in healthy volunteers. If the excitatory deflection increased the gain of the vestibuloocular reflex, then possibly brainstem mechanisms dominate caused by asymmetric afferent activity; if the gain decreased in spite of the excitatory deflection, then biasing the cupula in either direction causes a peripheral sensitivity decrease. In order to deflect the cupula cold caloric irrigation was applied in prone (forward head hanging) position. In this position cold thermal irrigation elicited an excitatory caloric nystagmus with an ipsilateral fast phase.

2 Methods

10 healthy individuals (2 woman, 8 men, average age 21.4 years; min.:18; max.:25) with no history of otologic or cochleovestibular disorders were recruited. Eardrum pathology was excluded by otoscopy before the experiments. The study was approved by the institutional review board of the Petz Aladar Hospital, Györ, Hungary (PAMOK Hospital Protocol number 76-1-18/2015). The protocol was in accordance with the ethical standards laid down in the 1964/2013 (7th revision) Declaration of Helsinki for research involving human subjects and the investigators have obtained written informed consent from each participant. Before the measurements, the left eardrum was examined. Video HIT (vHIT) was carried out with the ICS Impulse® video goggles (GN Otometrics, Taastrup, Denmark) in the plane of the lateral semicircular canal (SCC). We used cold caloric irrigation in order to excite horizontal semicircular canal in the left ear, because in prone position the right side cannot be irrigated without risking the electronics, which is built in the the ICS Impulse® video googles on the right side.

For head impulses, the position of the right eye was recorded with 250 samples/s. Baseline testing and caloric stimulation was done in prone (forward head hanging) position with the lateral SCCs in an earth-vertical position using the Reid's plane (with respect to this the canal is pitched upward by 25 degrees [2]) Subjects were tested in a well-lit room (to ensure a small pupil) with an eye-level target straight ahead (and below because of the prone position) at a distance of 120 cm in front of them. Care was taken to elicit reproducible head impulses (e.g. without touching the head trap etc.). Caloric irrigation lasted 40 seconds and was carried out in the left ear in all subjects using water at a temperature of 24°C. After calibration a baseline measurement was carried out using 4–6 high-acceleration head impulses to the right and left (peak head velocity >120°/s) in a randomly alternating fashion over a time interval of 30 sec. Then this protocol was repeated at the end of the thermal irrigation. This protocol created 4 to 6 gain values of individual impulses for every subject, for every side, per baseline measurement and again for the thermal irrigation. Raw data of head and eye velocity traces were exported into Matlab (The MathWorks, Natick, MA), and the aVOR gain (maximal eye velocity divided by inverted maximal head velocity) was determined off-line. Impulses, which were contaminated by a nystagmus fast phase, were discarded, and the first 4 impulses (four to the right and four to the left) were used. The time interval in which the 4 impulses were collected was from 10 seconds but not later as 30 seconds after irrigation. The slow phase velocity of the nystagmus could be judged by the vertical upward shift of the eye velocity curve (see also Fig. 4).

Statistics were done using IBM SPSS Statistics for Windows V.26 software. Significance of gain decrease was calculated using a mixed model ANCOVA (using the slow phase velocity values before each head impulse as a covariate). Thereby it was possible to examine if the caloric nystagmus had an influence on the gain decrease together with taking multiple measures per individual into account. Additionally a Pearson correlation was calculated between the difference of gain values and SPV. The baseline gain values between impulses to the left and to the right were compared using two way repeated measures ANOVA.

3 Results

Mean gain values (±SD) for the baseline measurements were 0.96±0.08 during impulses to the left and 0.99±0.07 to the right (n = 40, 10 individuals, 4 impulses each), impulses to the right yielding slightly but significantly higher gain values (p = 0.02). After cold irrigation on the left side the gain on the left decreased to 0.77±0.1 and it was 1.04±0.12 for the impulses to the right. After left cold caloric irrigation aVOR gain decreased highly significantly during impulses to the irrigated side (p < 0.001), while it did not change significantly during impulses to the right (Mean±SD = 1.04±0.12; p = 0.65) (Fig. 2.) As an illustration we also provide raw data (experiment 801) in Fig. 3. The inclusion of the slow phase velocity of the caloric nystagmus showed no significant interactions (to left left: p = 0.81; to the right: p = 0.94). A Pearson correlation showed no significant influence of the caloric nystagmus as calculated by the difference of gain values and SPV.

OPEN IN VIEWER

Fig.1

Experiment 1-2. Impulses to the left after left irrigation in supine position (nose-up) with cold water as published in [12]. Experiment 3-4. Impulses to the left after left irrigation in prone position (nose-down) with cold water. In the original experiment (1,2) the slow phase of the saccade necessary to compensate for the head impulse (arrows before the eyes) and the slow phase of the caloric nystagmus (dashed arrow) are in the opposite direction. In the present protocol (3,4) they are in the same direction.

OPEN IN VIEWER

Fig.2

Mean gain values (±SD) for head impulses before and shortly after cold thermal irrigation on the left (n = 40, 10 individuals, 4 impulses each). ^***p < 0.001; n.s.: not significant.

OPEN IN VIEWER

Fig.3

Raw baseline data and data after caloric reaction (experiment 801). Upper trace: baseline measurement in during head impulse testing in prone position. Lower trace: head impulses after cold thermal irrigation of the left ear. Leftward head impulses cause upward deflection, the curve tracking the eye movements is inverted (leftward fast phases cause downward deflection). The gain of the ipsilateral head impulse (to the left) was diminished. Downward spikes represent caloric nystagmus beats to the left.

OPEN IN VIEWER

Fig.4

Comparison of samples from the former experiment [12] (upper panel) and the present experiment (lower panel). Leftward head impulses cause upward deflection, the curve tracking the eye movements is inverted (leftward fast phases cause downward deflection). The slow phase velocity of the nystagmus is indicated by the vertical shift of the eye velocity curve. Upper panel: in that previous experiment the right ear was irrigated with cold water in supine positon, which resulted in a nystagmus with the fast phases to the left (the eye tracks are inverted, therefore the fast phases point downward in the panel). The ipsilateral head impulse (to the right) results in decreased aVOR gain. The resulting compensatory saccade cannot be differentiated from the fast phases of the nystagmus. Lower panel: in the present experiment the left ear was irrigated with cold water in prone position, which resulted again in a nystagmus with the fast phases to the left. The gain of the ipsilateral head impulse was diminished and it was followed by a compensatory saccade (fast phase to the right: upward), which is in opposite direction compared to the fast phase of the caloric nystagmus beats.

Baseline VOR gains for head impulses to the right were slightly but significantly higher than those to the right, as they were in our previous experiments [13]. Most likely, this is related to the fact that the rVOR was always recorded from the right eye, and velocity gains show higher values with head impulses in the direction of the eye that is recorded [15]. The main new result of this study was that the ipsilateral high velocity gain of the aVOR decreases if the cupula is biased in the excitatory direction. Together with the former experiment [12], in which the aVOR decreased during inhibitory bias, these results show that the aVOR gain decreases when the cupula is biased in either, inhibitory of excitatory direction (Fig. 4).

In our experimental setup, this behaviour of the VOR gain showed a profound difference to the results of Mantokoudis et al. [5], who could increase or decrease the VOR gain measured by HIT, depending on the direction of the slow phase of underlying postrotatory nystagmus. Another significant difference between the two outcomes was that in our setup changes occurred only during ipsilateral impulses, not bilaterally as in their experiment. We therefore may conjecture that in their experiment the effect was caused by the brainstem velocity storage mechanism [10] and in ours the effect was mainly peripheral and did not involve a strong postrotatory velocity integration.

As a possible peripheral cause, cupular and/or afferent bias should be considered as a possible mechanism. However, the accepted theories of SCC biomechanics do not offer any explanation for this phenomenon. According to van Buskirk’s model of SCC biomechanics [14] the cupula deflection due to the head impulse is incremental and has to be added to the initial cupula deflection at the start of the maneuver. This would predict a bias of the perceived rotatory movement in the direction of the post-caloric deflection.

Also more complex models of cupular mechanics [11] or endolymphatic flow [1] are unable to explain why opposite post-caloric deflection before the head-impulse would lead to the same (inhibitory) bias of the gain. Moreover, any potential non-linear effects due to excessive deformation of the cupula appear unlikely because the post-caloric cupular deflection is relatively small and lies within range of physiological cupula displacements. Unilateral negative bias of the cupula and/or afferent activity with deviation in one direction should be accompanied with a positive bias during deviation in the other.

Therefore, we conclude that SCC biomechanics or a saturation/cut-off of the afferent spike activity are unable to explain the observed results.

Another possibility would be a direct temperature effect acting on the fibers of the vestibular nerve. In this case, hypothetically, cooling the temporal bone might directly influence afferent function. According to an appealing theoretical framework, afferent vestibular information is transmitted by tonic (velocity sensitive) and phasic (acceleration sensitive) signal streams, possibly corresponding to regularly and irregularly discharging vestibular primary afferents and second-order vestibular neurons, respectively [7]. The phasic pathway has a role during transient, high acceleration stimuli such as applied during head impulse testing and behaves nonlinearly, because its gain rises with head velocity (proportional to the cube of head velocity) at higher frequencies, while the tonic pathway has a role over a broad range of rotational frequencies and velocities and has linear response properties. In normal circumstances the tonic pathway has a bigger role, the phasic pathway, being more modifiable, gains in importance during vestibular compensation after unilateral vestibular lesion [7]. Park et al. showed that the resting rate and sensitivity of irregular afferents are strongly affected by temperature in mice [8]. Hübner et al asked if it is possible to modify the overall VOR-gain in mice by temperature differences and, although they found that the VOR-gain could be modified mainly by low frequency sinusoidal rotations, during transient stimuli they found no effect of temperature on the VOR [3]. In our experiments, however, a direct cooling effect on the irregular/phasic afferentation seems not likely, because of the results of earlier experiments with an identical setup for measuring the effect of inhibitory cupular deviation. In this setup, head-impulses were applied in both horizontal directions after warm- and cold-water caloric stimulation [13] and as a control experiment, head-impulse tests were applied after warm- and cold-water caloric stimulation with the plane of lateral SCC-oriented earth horizontal (the subject’s head inclined 30° from supine). In this horizontal position no caloric nystagmus could be elicited and there were no changes of rVOR gains compared with the baseline [13].

A possible explanation of the decreased VOR gain comes into light when considering the mechanoelectrical transduction (MET) machinery of vestibular hair cells themselves and their behaviour when their stereocilia bundle gets deflected due to a biased cupula. The amount of depolarization of a hair cell for a stimulus of a given amplitude follows a sigmoid law whose shape describes the opening probability of the MET channels of the hair cell under consideration (more precisely, their Boltzmann statistics – review in Avan et al, 2013). The resting position of the hair bundle defines the operating point (OP) of the cell along the depolarization vs. stimulus amplitude curve. The slope of the curve around the OP, by determining the amount of depolarization for a given stimulus, governs the transduction gain: the steeper the slope, the larger the gain. A cupular bias, by moving the OP along the curve toward positions with a different slope, is expected to alter this gain. To predict in which direction, the OP position in the absence of bias is needed. The position of maximum efficiency is the one with a 50% opening probability of MET channels, the inflection point where the slope is maximum. That normal vestibular cells may well be poised there can be inferred by the observations of amplification in the semicircular canals of the oyster toadfish [9], and of spontaneous oscillations of the stereocilia of bullfrog saccular cells. These two properties are characteristic of so-called self-tuned critical oscillators near a Hopf bifurcation, which requires their being poised at the OP where efficiency is maximum. Being tuned on the verge of instability, their sensitivity to weak stimuli is maximum. Any displacement of the OP, in whichever direction, decreases the cell’s response to a stimulus regardless of its size, which might translate in a decreased gain of the VOR. The fact that the critical-oscillator behaviour has been demonstrated only in rather primitive species does not necessarily weaken the hypothesis, as it can be safely assumed that the evolutionary advantage afforded by this mechanism has led to its preservation in mammals with more advanced sensory organs.

Possible limitations of the study are the low number of observations and the possible confounding effects of the utricular input. However, we think that utricular activation/inhibition played a negligible role in our results because caloric reaction does not influence utricular afferent activity and other parameters were not changed during stimulation. Also, head impulses in prone position cannot be considered as a physiological situation, however this can also be said about the caloric stimulation, which is nevertheless widely accepted as a stimulation tool. As an additional consideration, we must keep in mind the possible addition effect of cerebellar inhibition and/or efferent vestibular function, until it will be clear, how these processes operate during sustained, non-physiological stimulations [6]. We noted a trend as if immediately after caloric stimulation the gain increased slightly during head impulses to the right (contralateral to the irrigated side) and this would mean that the slow velocity of the ensuing nystagmus influenced slightly the gain by a central mechanism, by adding to the velocity of the compensatory saccade. Then gain decrease on the left side would also be a little less than in the case of the inhibitory stimulation (cold caloric stimulation in supine position). However, due to the comparatively low number of points and their statistical deviation, this effect was not significant by the statistical method used. Considering this question further experiments will be necessary.

In conclusion, we suggest that the gain decrease after cold thermal irrigation during biasing in excitatory and inhibitory direction may be caused by the deflection of the hair cell stereocilia away from their optimal operating point. From clinical standpoint this new result may be interesting in pathological constellations, when the cupula is chronically biased in a specific direction, away from its optimal operating point, such as in changes of specific gravity (pathological light cupula or alcohol-induced nystagmus) or when the cupula is loaded by misplaced otoconia, such as in cupulolithiasis.

Conflict of interest statement

The authors report no conflict of interest.