- Email: [email protected]
Changes in utricular function during artificial endolymph injections in guinea pigs
Hearing Research 304 (2013) 70e76
Contents lists available at SciVerse ScienceDirect
Hearing Research journal homepage: www.elsevier.com/locate/heares
Research paper
Changes in utricular function during artificial endolymph injections in guinea pigs D.J. Brown a, *, Y. Chihara a, b, Y. Wang a a b
The Brain and Mind Research Institute, Sydney Medical School, The University of Sydney, 100 Mallett Street, Camperdown 2050, Australia National Institute of Sensory Organs, National Tokyo Medical Center, Tokyo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 March 2013 Received in revised form 7 May 2013 Accepted 27 May 2013 Available online 18 June 2013
Various theories suggest endolymphatic hydrops may cause a rupture of the membranous labyrinth or may force open the utriculo-saccular duct, resulting in a sudden change in inner ear function. Here, we have used slow injections of artificial endolymph into either scala media or the utricle of anaesthetised guinea pigs to investigate the effects of hydrops. Vestibular function was continuously monitored in addition to the measurements of cochlear function developed in our laboratory (Brown et al. Hear Res, 2013). Scala media injection induced consistent functional changes, which occurred in two stages. Initial changes involved were associated with an increased hydrostatic pressure in scala media that only affected cochlear function. After 3e4 ml of endolymph had been injected, cochlear function spontaneously recovered, and was often shortly followed by a transient increase or decrease in utricular sensitivity, with the effects varying between animals. Endolymph injection directly into the utricle produced variable effects across animals, although in 2 experiments it produced similar changes as those observed for scala media injections, suggesting that the fluid pathway between scala media and the utricle was continuous in these animals. The mechanism underlying the sudden, spontaneous functional changes is not yet clear, but we tentatively suggest that in some cases it may be caused by the utriculo-saccular duct suddenly opening to alleviate an elevated hydrostatic pressure in the pars inferior, resulting in a change in utricular function due to an increase in its volume. These changes are comparable to the sudden or fluctuating functional changes in Ménière’s sufferers, and support the hypothesis that endolymphatic hydrops can directly cause some symptoms of this syndrome. Crown Copyright Ó 2013 Published by Elsevier B.V. All rights reserved.
1. Introduction In support of Schuknecht’s 1976 rupture theory of vertigo attacks in Ménière’s Disease (MD), recent experimental studies have suggested that rapid injection of artificial endolymph into scala media in guinea pigs can cause a rupture of the membranous labyrinth (Valk et al., 2004, 2006), a mixing of perilymph and endolymph (Kakigi et al., 2010), and a severe loss of vestibular function (Kingma and Wit, 2010), supposedly due to a potassium intoxication of vestibular hair cells. In a recent study involving slow injections of artificial endolymph into scala media in guinea pigs (Brown et al., 2013) we observed repeated episodes of a slow decline in cochlear sensitivity followed by a spontaneous, sudden
Abbreviations: CAP, compound action potential; SP, summating potential; VsEP, vestibular evoked potential; EP, endocochlear potential; BCV, bone conducted vibration * Corresponding author. Tel.: þ61 2 93510748. E-mail addresses: [email protected], [email protected] (D.J. Brown), [email protected] (Y. Chihara).
recovery of function that could not be easily explained on the basis of a labyrinth rupture. A rupture of the membranous labyrinth allowing mixing of perilymph and endolymph would be expected to cause a severe reduction in cochlear and vestibular sensitivity (Zenner, 1986; Kobayashi et al., 1999; Marcon and Patuzzi, 2008; Kingma and Wit, 2010). Instead, as proposed in previous studies (Schuknecht and Belal, 1975; Salt and Rask-Andersen, 2004), we tentatively suggested that at certain stages of hydrops development the utriculo-saccular duct (also termed the valve of Bast) or the endolymphatic sinus may suddenly open (Gibson, 2010) allowing a rapid “de-compression” of endolymph pressure in the cochlea, causing a sudden recovery of cochlear function and possibly a sudden change in vestibular function. Whatever the mechanism underlying the abrupt recovery of cochlear function observed during artificial hydrops development, it could potentially explain recent clinical findings that are also difficult to explain on the basis of a rupture. That is, it has recently been demonstrated that in the early stages of Ménière’s Disease utricular and semicircular canal sensitivity is enhanced (Manzari et al., 2010, 2011), and that Ménière’s sufferers do not lose their
0378-5955/$ e see front matter Crown Copyright Ó 2013 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.heares.2013.05.011
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
hearing at the time of a vertigo attack (McNeill et al., 2009). An abrupt opening of physiological valves in the inner ear could potentially allow rapid changes in cochlear or vestibular function due to the relief of hydrops pressure in the pars inferior, without causing a mixing of perilymph and endolymph. Such a mechanism could also cause a sudden increase in the volume of the utricle and semicircular canals (pars superior), resulting in a mechanical displacement of the utricular macula and a stretching of the cupula, which can potentially change their sensitivity (Rabbitt et al., 2001; Chihara et al., 2013). To determine the effects of slow endolymph injections on utricular function, we have slowly injected artificial endolymph (at 40e80 nl/min) into the membranous labyrinth in guinea pigs whilst monitoring several measures of both cochlear (as in our previous study; Brown et al., 2013) and utricular function. As a sidenote, to the best of our knowledge this is the first time simultaneous measurements of both cochlear and vestibular function have been performed during in vivo animal experiments. Vestibular function can be estimated by monitoring the Vestibular Evoked short-latency Potential (VsEP), which is a gross compound action potential of neurons innervating the jerksensitive hair cells in the otoliths, in response to brief head acceleration pulses (Bohmer, 1995; Plotnik et al., 1999; Jones and Jones, 1999; Jones et al., 2011). Moreover, we recently provided strong evidence that the VsEP evoked by bone-conducted vibration (BCV) pulses (in the presence of acoustic masking noise to suppress cochlear responses) is almost entirely a response of neurons innervating the utricle (Chihara et al., 2013). We typically measure the VsEP from the recording electrode inserted into the facial nerve canal, which provides a close electrogenic proximity to the superior vestibular nerve from which the VsEP response originates. Kingma and Wit (2010) recently demonstrated that the VsEP amplitude decreased abruptly during rapid injections of artificial endolymph, supposedly due to a rupture of the membranous labyrinth. However, the rapid-rate of endolymph injection in this study prevented the assessment of subtle changes in the vestibular function, which, an earlier study by Kingma and Wit (2009) suggested may potentially include an enhanced VsEP amplitude due to an increase in endolymph pressure. The present study investigated the changes in vestibular function during relatively slow injections of artificial endolymph into scala media. We also utilized several measurements of cochlear function, such as Compound Action Potential (CAP) thresholds, Summating Potential (SP)/CAP ratios and cochlear microphonic analysis, as an indicator of the sudden changes in endolymph pressure based on our previous study (Brown et al., 2013). However, here we have omitted the SP/CAP and cochlear microphonic results in favour of focussing on the injection induced changes in vestibular sensitivity. The CAP thresholds alone are used here as an indicator of the injection induced changes in cochlear function. We have also attempted to inject endolymph directly into the utricle, to determine if physiological changes in the VsEP with endolymph injection are due to a hydrostatic pressure increase specifically in the utricle, and also to test the putative role of the physiological valves that separate the pars superior from the pars inferior, namely the utriculo-saccular duct (Schuknecht and Belal, 1975) and the endolymphatic sinus (Salt and Rask-Andersen, 2004). 2. Methods 2.1. Animal preparation & surgery Experiments were performed on 19 normal adult guinea pigs (Cavia porcellus) of either sex with body weights between 250 and 450 g. Partial results from 2 animals used in this study (GP120 and
71
GP139; Fig. 2A, D and N) were presented in our previous publication (Brown et al., 2013). All animal preparation, surgery and experimental protocol were approved by The University of Sydney’s Animal Ethics Committee. Animals were anaesthetised in a gas anaesthesia chamber with 3% isoflurane, and also given a onceoff 0.05 ml intraperitoneal injection of Temgesic (Buprenorpherine; Reckitt Benckiser, Auckland NZ) and 0.1 ml of Atrosine (0.6 mg/ml atropine sulphate; Apex Laboratories, NSW, Australia). Once sedate and the foot-withdrawal reflex was absent, animals were moved to the experimental setup in the sound-isolation booth, tracheotomized, and thereafter artificially ventilated with a mixture of oxygen and isoflurane (1e2%) by an artificial respirator pump. The animal’s core temperature was regulated to 38 C (2 C) by a rectal temperature probe and an electric heating pad, and the sound-isolated room was heated with an infrared heating lamp. After the main surgery, animals were given a 0.1 ml intramuscular injection of the neuromuscular blocker Pauvulon (2 mg/ml pancuronium bromide; Astra Pharmaceuticals, L.P.) to suppress middle ear reflexes and general muscle activity. To expose the cochlea, the bulla was opened via a dorso-lateral approach, providing a direct view of the round window and the basal cochlear turn. A thin tissue wick was placed in the bulla to minimize any acoustical, functional or electrical changes due to fluid build-up in the bulla. To expose the utricle and semicircular canals, the bulla was opened just superficial to the external auditory canal, providing a view of the where the superior and lateral ampullae meet with the utricle. In all experiments the animal was placed in a custom-designed ear bar, which housed a low-noise microphone (ER10Bþ, Etymotic Inc., IL, USA), and the ends of 3 silastic tubes (20 cm long) which were connected to 3 separate headphone speakers, forming a closed acoustic sound-field in the ear canal. To produce skull vibration stimuli a standard audiometric bone oscillator (Radioear, New Eagle, PA, USA), surrounded by custom electromagnetic shielding, was rigidly attached to the caudal surface of the guinea pigs skull via a metal bolt (Fig. 1). This involved removing a 2 mm 4 mm rectangular section of caudal skull bone contralateral to the saggital suture line (i.e. overlying the contralateral vestibular
Fig. 1. Schematic diagram of the experimental setup. BCV was delivered via a B71 bone vibrator surgically attached to the skull, with the skull vibration measured with an accelerometer also attached to the skull bone. Endolymph was injected into the labyrinth via a 5e10 mm tip glass micropipette inserted into either scala media via the round window, or into the utricle via a small hole drilled into the bone overlying the horizontal and superior semicircular canal commissure. The glass pipette was also used to measure the EP. Acoustic stimuli were delivered via speaker tubing sealed into the ear-bars, which also housed a low-noise microphone for sound calibration. Gross cochlear responses were measured from an AgCl wire sealed into scala tympani of the first cochlear turn, with the VsEP measured from a wire inserted 1 cm into the facial nerve canal.
72
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
system) using a handheld micro-drill, then inserting the flat “T” shaped end of a 4 cm long bolt into the opening (superficial to the dura), rotating the bolt by 90 , then tightening the bolt to the bone using a nut on the shank of the bolt (see Curthoys et al., 2006). To measure the skull vibration, a 3-axis accelerometer (Dimension Engineering, OH, USA) was also bolted to the temporal bone using the same technique used to attach the B71, but located over the ipsilateral vestibular system (i.e. ipsilateral to the saggital suture line on the caudal skull bone). While all 3 axis of the accelerometer output were monitored during recordings, for the purpose of simplicity, here we have only focussed on acceleration perpendicular to the plane of the frontal bone (i.e. transverse vibration of the skull). 2.2. Physiological measures To measure cochlear responses, a 0.2 mm diameter hole was made in the first turn of the cochlea, through which the exposed tip of a teflon-coated AgCl recording wire was placed, and then sealedin using superglue. To inject artificial endolymph and to record the endocochlear potential, a 5e10 mm diameter tip, 5e10 MU glass micropipette was inserted into scala media via the round window and basilar membrane. In this study, the composition of artificial endolymph consisted of (in mM): KCl, 126; NaCl, 1; KHCO3, 25; MgCl2, 0.025; CaCl2, 0.025; K2HPO4, 1.4; mannitol, 25; and had a pH of 7.4 and an osmolarity of 310e320 mOsmol, as previously described by Marcus et al. (1983). In 6 animals, the micropipette was inserted into the utricle through a small hole made in the horizontal semicircular canal ampulla/utricle commissure. Prior to making a hole, the bone was shaved slightly and a thin layer of hydrophilic glue (“Gorilla Glue”, Cincinnati, Ohio) was applied to the bone and allowed to dry for 5e 10 min. A hole was then made through the glue and bone, and the pipette tip was inserted into the utricle under micromanipulation control, before being sealed in place with superglue (Selleys Plastic Glue, DuluxGroupPty Ltd., Australia) to prevent fluid leakage from the hole. The difficulty with this procedure was that often the glass pipette appeared to produce either a large tear in the membranous labyrinth bounding the utricle (which was not easily visible due to the transparency of the membrane when covered with fluid), or it appeared to push the membranous labyrinth away from the bony wall. It was assumed that the pipette had been successfully inserted and sealed into the utricle on the basis of two findings. First, when inserting the pipette, the voltage at the tip of the glass pipette ’jumped’ abruptly by þ3 to þ5 mV (relative to the potential measured at the fluid overlying the membranous labyrinth), which previous studies have suggested is the resting potential of endolymph in the pars superior (Marcus and Marcus, 1987). Second, the functional changes observed later in the experiment appeared to support a correct insertion, as discussed later. For both scala media and utricular injections, the glass micropipette was connected to a 10 ml glass syringe, attached to an UltramicroPump III perfusion pump (WPI, Sarasota, FL). Prior to inserting the glass pipette, the tip of the pipette was visually inspected while injecting into air to confirm an immediate injection of fluid, and to confirm there was no “lag” in the injection onset. Physiological potentials were amplified using a customdesigned low-noise AC and high-impedance DC amplifier. Amplified signals were then digitized using a battery powered Wi-Fi data acquisition device (NI 9205, National Instruments, TX), and wirelessly transmitted to PC. The 3 speakers, the B71 bone conductor and the low-noise microphone signal were driven via an external USB soundcard (UltraLite-mk3 Hybrid, MOTU Cambridge, MA). All recordings and stimulus generation were performed with custom designed LabVIEW programs (National Instruments, TX).
The CAP responses were measured from the scala tympani wire in response to a train of high-frequency acoustic tone-bursts. This tone-burst train consisted of 15, 11, 6, 4 and 2 kHz tone-bursts, each 10 ms in duration and separated by 20 ms of silence, with a 1 ms cosine ramp on and off. CAP thresholds were determined programmatically using a custom software algorithm that adjusted the sound level of each tone-burst independently, until such time that the correlation between successive averaged responses (with 8 averages for each response) was close to 0.4 (0.05). This technique is similar to performing a visual detection of CAP thresholds, where the visual inspection of the response relative to the noise is used to subjectively determine a correlation threshold. To evoke VsEP responses, we aimed to produce a BCV stimulus that produced a brief (<1 ms) biphasic acceleration of the animal’s skull. However, delivering a 1 ms BCV pulse to the B71 resulted in a complex resonance of the animal’s skull (as determined from the skull-fixed accelerometer). This typically resulted in a complex VsEP waveform which differed between animals. As detailed in a recent study (Chihara et al., 2013), to partially correct for this resonance we used a standard engineering technique. This involved measuring the skull acceleration in response to a simple BCV stimulus consisting of a single 1 kHz sinusoid (with a Gaussian window), and then dividing the FFT of the stimulus by the FFT of the skull acceleration pattern. This resultant FFT was then multiplied by the FFT of the simple stimulus, and an inverse FFT was performed on this ’corrected’ spectrum, to provide a “corrected stimulus”. When delivered to the B71, this corrected stimulus generated a single-cycle 1 kHz, transverse acceleration of the animal’s skull. In short, we examined the vibration properties the head and corrected for the nonlinearities of the bone-conducted vibration, which is similar to calibrating a sound-field using a microphone, and then delivering a calibrated sound stimulus. This simple 1 ms acceleration of the animal’s skull resulted in a relatively simple VsEP waveform that was more consistent across animals. Broadband acoustic masking noise with a reasonably flat frequency response (10 dB, 10 Hz to 16 kHz) was simultaneously presented with the BCV stimulus, to both ear canals at 105 dB SPL. Acoustic noise was only presented during VsEP measurement, which generally lasted 3e4 s, allowing for 40e80 averages per averaged VsEP waveform, where successive BCV stimuli were presented at a rate of 16/second. The measurement of EP, CAP thresholds, and VsEP were performed systematically in series, with the LabVIEW programs (each measure was obtained using a separate LabVIEW program) continuously looping through each measurement subset. This allowed for a relatively high temporal resolution of measurements throughout endolymph injection experiments. 3. Results 3.1. Functional changes during slow injections of endolymph into scala media Following the insertion of the glass micropipette into scala media, all functional measurements were stable relative to the changes induced by the injection, at least over a 2 h period (data not shown, but see Brown et al., 2013). Fig. 2 shows examples of the functional changes during endolymph injection measured in 5 animals. Around 10 min after beginning the injection (which is indicated as a black line in Fig. 2KeO), there was a gradual increase in CAP thresholds, starting with the 2 kHz CAP threshold similar to that observed in our previous study (Fig. 2AeE), and a gradual decline in the EP (Fig. 2KeO). During this phase of the injection, the VsEP amplitude remained relatively stable in most animals, although in some animals it increased slightly, and in others it
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
73
Fig. 2. Changes in cochlear and vestibular function during endolymph injection into scala media in 5 representative animals. AeE) CAP thresholds to 10 ms duration tone-bursts of 15, 11, 6, 4 and 2 kHz frequencies. FeJ) The N1eP1 amplitude of the VsEP, KeO) The EP measured from the scala media pipette. The duration of the injection is indicated as a horizontal line in KeO, with the injection rate (in nl/minute) and injected volume (in ml, in parentheses) shown above each line. Dark arrows in AeE indicate the episodes of cochlear function recovery. Roman numerals indicate periods at which the representative responses used in Fig. 3 were made. Asterisks indicate periods where the VsEP amplitude was transiently increased.
gradually decreased (Fig. 2FeJ). In all experiments, after 2.5e3.5 ml of endolymph had been injected, the changes in cochlear function abruptly recovered (thick black arrow Fig. 2AeE). Following the recovery of cochlear function, the VsEP amplitude either remained relatively stable, rapidly increased, or rapidly decreased several minutes after the abrupt recovery. While changes in cochlear function were consistent and resembled that reported previously (Brown et al., 2013), there was no apparent consistency in the changes in the VsEP amplitude across 14 different injection experiments. Examples of the VsEP waveform before and after the rapid decrease in VsEP amplitude is shown in Fig. 3, with the timing of each waveform corresponding to the time points indicated with roman numerals in Fig. 2G. As a simple test of the effects of different injection rates, the initial injection was performed at a rate of 40 nl/min in 11 animals, 80 nl/ min in 2 animals, and 100 nl/min in 1 animal. The injection was typically ceased approximately 5e10 min after an abrupt recovery episode occurred, although on a two occasions the injection was continued for an additional 30e40 min following the recovery episode. On average, the volume of endolymph injected prior to an abrupt recovery episode was 3.3 ml 0.5 ml. The rate of the injection did not appear to have any obvious additional impact on the functional changes, other than reducing the time taken to induce the abrupt recovery episode. Often, multiple injections were performed in the same experiment, allowing 30e40 min for the recovery of the injection induced changes (see Fig. 2, gp139, gp152, gp154). In all experiments, if the VsEP amplitude did not decrease following the first injection and cochlear recovery episode, it did after the second injection (Fig. 2H and I). In one experiment, the injection was performed 3 times over a 5.5 h period, resulting in multiple reductions and recoveries of the VsEP in the same animal (Fig. 2J).
Fig. 3. Examples of VsEP responses at various stages during endolymph injection into scala media. Responses correspond to the time-points in Fig. 2, indicated using Roman Numerals I and II. The VsEP was evoked by a BCV pulse which produced a vibration of the animal’s skull shown at the bottom of this Figure.
While there were subtle differences between the effects of endolymph injection between different animal experiments, such as the volume required to induce a sudden recovery of cochlear function, changes were reliably and consistently reproduced in 8 experiments where endolymph was injected at a rate of 40 nl/min (in 3 other experiments utilizing 40 nl/min injection into scala media, the experiment was intentionally stopped just after the abrupt recovery episode, for the purpose of imaging the inner ear morphology at this stage). The averaged changes in each measurement (with standard deviations) throughout the first endolymph injection in these 8 experiments are shown in Fig. 4. Here,
Fig. 4. Average functional measurements during 40 nl/min endolymph injection into scala media in 8 animals, with standard deviations. A) CAP thresholds to a 2 kHz toneburst, B) the N1eP1 amplitude of the VsEP evoked by a 1e2 g BCV pulse, and C) the EP measured from the scala media pipette. In each animal, the timing of each measurement was first referenced to the initial major episode of recovery of cochlear function, and then down-sampled to a measurement every 2 min to enable averaging of the data cross-animals. The average onset and offset of the injection relative to the recovery episode is shown in C, along with standard deviation. The average volume of endolymph injected was 3.2 0.55 ml.
74
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
for the purposes of averaging the time-charts across different animals, the timing of each individual time-chart was referenced to the first major recovery of cochlear function, before being averaged with the results from other animals. The average start and end of the injection period relative to the abrupt recovery episode, with standard deviation are indicated in Fig. 4C. 3.2. Utricular injection In an attempt to determine if the reduction in VsEP amplitude was due to a rupture of the membranous labyrinth, or due to a hydrostatic pressure increase in the pars superior, endolymph was injected directly into the utricle in 6 animals. In 2 of these experiments, CAP thresholds increased throughout the initial stage of the injection, however, there was only a gradual decline in the VsEP amplitude (Fig. 5, left two columns). The EP changed relatively little throughout these experiments, although the initial EP measured in the utricle and was only 5 mV (see Section 2.2). Again, after about 3.6 ml of endolymph had been injected into the membranous labyrinth there was a sudden recovery of cochlear function (Fig. 5, black arrows), although the recovery of function was not as distinctive as that observed in scala media injections. In 4 of these utricular injection experiments, there was little change in cochlear function throughout or after the injection, but the VsEP amplitude rapidly decreased soon after starting the injection, when only 1e2 ml had been injected (Fig. 5, grey arrows). The VsEP amplitude did not recover following this rapid decrease. 4. Discussion 4.1. Functional changes throughout endolymph injections As suggested in our previous study (Brown et al., 2013), the initial changes in cochlear function observed during endolymph injection into scala media, namely the increase in CAP thresholds were most likely due to an increase in the hydrostatic pressure of endolymph causing a displacement of the basilar membrane towards scala tympani (discussed below). The present study introduced simultaneous measures of utricular function with acute endolymph injections, demonstrating that while cochlear sensitivity decreased due to the increase in endolymph pressure, there was little change in utricular function. This suggests that either the utricle is insensitive to hydrops, or that the injected endolymph did not cause a displacement of the utricular macula. As observed in our previous study, after 3e4 ml of endolymph had been injected
into scala media, cochlear function abruptly recovered. This recovery was typically followed by marked changes in utricular sensitivity, often a rapid reduction in sensitivity, but on occasions a transient increase in sensitivity was observed.
4.2. Functional changes are related to changes in hydrostatic pressure As demonstrated previously endolymph injection can cause a gradual increase, then sudden drop in endolymph pressure (Valk et al., 2006; Kingma and Wit, 2010). While we did not directly measure hydrostatic pressure here, we have assumed the changes in cochlear function were directly caused by changes in endolymph pressure, causing a displacement of the basilar membrane. Displacement of the basilar membrane can bias the average resting conductance of the outer hair cells, impairing the function of the cochlear amplifier and causing a loss of acoustic sensitivity (Zwicker, 1977; Patuzzi et al., 1984). While the functional changes may have been due to altered ionic composition of the fluids or changes in the receptor current, potentially affecting outer hair cell function, there are at least two reasons why the changes were more likely due to hydrostatic bias of the basilar membrane. First, that low-frequency thresholds were elevated before higher frequency thresholds suggests that the changes were induced by an increased hydrostatic pressure. That is, the compliance of the basilar membrane increases towards the apex of the cochlea, and more apical regions of the basilar membrane would be displaced first during increased endolymph pressure, and thus hydrops should cause the most severe loss of hearing sensitivity to low-frequency sounds (Tonndorf, 1983). Second, given that the hearing loss spontaneously recovered so abruptly (often within 10 s) after 3e3.5 ml of endolymph had been injected, it is more likely that the initial hearing loss and sudden recovery was dominated by mechanical changes, and moreover, due to a sudden relief from an abnormally large hydrostatic pressure in the endolymph.
4.3. The pressure relief event e rupture or shunt? There are at least two likely explanations for the sudden relief of pressure allowing recovery of cochlear function, while causing a sudden change in utricular sensitivity. The pressure release may be either due to a rupture of the membranous labyrinth (Schuknecht, 1976; Valk et al., 2006; Kingma and Wit, 2010), or a sudden shunt of endolymph from the pars inferior into the pars superior via some
Fig. 5. Changes in cochlear and vestibular function during endolymph injection into the utricle in 4 representative animals. AeD) CAP thresholds EeH) The VsEP amplitude, IeL) The EP measured from the utricle pipette. The duration of the injection is indicated as a horizontal line in IeL, with the injection rate and injected volume (in parentheses) shown near each line. Dark arrows in A and B indicate episodes of cochlear function recovery. Light arrows indicate where the VsEP amplitude decreased.
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
form of physiological valve mechanism (Schuknecht and Belal, 1975; Salt and Rask-Andersen, 2004). In support of the rupture theory, previous researchers have demonstrated that increasing the Kþ concentration of perilymph causes a significant reduction in the VsEP amplitude, and that injecting more than 3 ml of endolymph into scala media also results in a reduction of the VsEP (Kingma and Wit, 2010). While we did observe a rapid reduction in the VsEP amplitude following the recovery of cochlear function in several of our experiments, in others we observed either no change (Fig. 2H and I), or a transient increase (Fig. 2F and G). In these experiments, it is unlikely that a rupture of the labyrinth allowing endolymph and perilymph mixing was the cause of the enhanced VsEP amplitude because increasing the Kþ concentration in perilymph only produces a reduction in the VsEP amplitude. Previous imaging studies have demonstrated ruptures of the labyrinth following endolymph injection in guinea pigs (Valk et al., 2004); however, the imaging techniques used in these previous studies can lead to artificial tears in thin membranes, due to the need for decalcification and optical clearing. In a previous study which used imaging techniques free from these procedures, namely X-ray micro-tomography, no evidence of any ruptures following 4 ml of injected endolymph were found, nor was any mixing of the inner ear fluids evident when a contrast enhancement agent was added to the injected endolymph (see Fig. 6D2 from Brown et al., 2013). Further suggesting that the pressure relief during endolymph injections is not due to a rupture is the finding that repeated endolymph injections in the same cochlea often results in successive loss of cochlear sensitivity, followed by a successive abrupt recovery episodes and a repeated loss of utricular sensitivity (Fig. 2, gp154, gp139, gp152). That is, while experimentally induced ruptures of the labyrinth have been shown to heal over several days (Duvall III and Rhodes, 1967), it is difficult to envisage how such a tear could heal within 30e40 min to allow a second increase in endolymph pressure with injection, and a second abrupt pressure relief and cochlear recovery, with a subsequent loss of vestibular function. Since its initial description, the utriculo-saccular duct (also called the valve of Bast) has been suggested to act like a valve, separating the utricle and semicircular canals from the saccule and cochlea, except in cases of endolymphatic hydrops which might force the duct open and allow a shunt of endolymph into the pars superior (Schucknecht and Belal, 1975). Likewise, the endolymphatic sinus has been suggested to act like a valve, opening to allow saccular endolymph to flow towards the endolymphatic duct (Salt and Rask-Andersen, 2004). Given the sensitivity of the cochlea to static pressure gradients between endolymph and perilymph, it seems reasonable that we might expect to observe some sudden change in cochlear function related to the opening of the utriculosaccular duct or endolymphatic sinus, if indeed these structures act like a physiological valve (Fig. 6). What’s less clear, is what we might expect to happen to utricular sensitivity if utricular volume suddenly increased. It has been demonstrated that around 2/3rds of the utricular macula rests on a compliant membrane separating the perilymphfilled vestibule from the endolymph-filled utricle in guinea pigs (Uzun-Coruhu et al., 2007). A recent study (Chihara et al., 2013) demonstrated that slight displacement of the utricular macula altered its sensitivity. Furthermore, an increased hydrostatic pressure in the utricle was suggested as the cause of a slight increase in the amplitude of the VsEP during acute endolymph injections in guinea pigs (Kingma and Wit, 2009). It is therefore possible that an increase in utricle volume could cause a mechanical displacement of the utricular macula, resulting in a change in its sensitivity.
75
Fig. 6. Schematic diagram of the putative changes in membranous labyrinth during hydrops development. Here, hydrops initially develops in the pars inferior only because the utriculo-saccular duct is closed. At some stage during hydrops development, the utriculo-saccular duct opens, allowing an influx of endolymph into the utricle and semicircular canals (SSCs), and a reduction in the volume of the pars inferior. This might allow recovery from a pressure-induced hearing loss, but simultaneously cause a mechanical displacement of the utricular macula, altering its sensitivity to BCV.
4.4. Utricular injections In an attempt to test if changes in utricular volume alter its sensitivity, we injected artificial endolymph directly into the utricle. As previously stated, this was a far more technically challenging technique than scala media injections, because the pipette must be inserted through the compliant and translucent membranous labyrinth without tearing it. In two of these experiments, injection at a rate of 80 nl/min resulted in a slow increase in CAP thresholds and a subsequent sudden recovery of these changes, but only a slow decline in the VsEP amplitude (Fig. 5E and F). The fact that these changes closely resembled those consistently observed in scala media injections suggests that we had successfully sealed and injected fluid into the utricle via the glass pipette, resulting in increased endolymph pressure in not only the utricle, but also the cochlea, and therefore also suggests that the utriculo-saccular duct was patent in these experiments. In another four utricular injection experiments, there was no change in cochlear function during an injection of 3e4 ml of endolymph, but there was a severe and sudden reduction in the VsEP amplitude. Potentially, the reduction in the VsEP here was either because the glass pipette was not properly sealed into the utricle, allowing endolymph to mix with vestibular perilymph, causing a depolarization of the utricular hair cell, or alternatively, in these ears the utriculo-saccular duct may have been closed, resulting in a relatively rapid increase in utricular volume, causing a rapid decrease in utricular sensitivity. Ultimately, with less certainty regarding the actual injection of endolymph into the utricle, it is difficult to interpret functional changes observed in these utricle injection experiments. We are currently exploring new
76
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
techniques for more accurately determining where the injected endolymph is distributed in these injection experiments, which will ultimately allow us to determine the patency and function of the utriculo-saccular duct and endolymphatic sinus. 4.5. Artificial endolymph injection as a model for vertigo attacks in Ménière’s Disease Acute endolymph injection provides a simple experimental animal model of pathological endolymphatic hydrops (Salt and Plontke, 2010), and is a simple way to investigate possible causes of symptoms in Ménière’s Disease. While hydrops itself does not universally induce the symptoms of Ménière’s Disease (Merchant et al., 2005), this might be explained on the grounds that in some non-Ménière’s ears endolymph volume may increase sufficiently slowly that hyperplasia of the membranous labyrinth prevents an increase in endolymph pressure (due to an increased compliance of the labyrinth), and there is therefore no mechanical displacement of the cochlear and vestibular hair cells. However, while previous studies have found differences in cell densities between non-MD and MD patients with hydrops, the results are complex and vary with cell type and location (Cureoglu et al., 2004). It is also important to note that while Ménière’s sufferers appear to have hydrops consistently, they are not in a consistent stage of vertigo or hearing loss, and some discontinuous physiological event is required to explain the sudden onset of vertigo in Ménière’s. The theory that fluctuations in endolymph pressure causes fluctuating displacement of cochlear hair cells is not new (Tonndorf, 1983), and can explain why low-frequency sensitivity fluctuates most severely in Ménière’s sufferers (McNeill et al., 2009). The changes in utricular sensitivity we have observed in the present study, particularly the transiently enhanced sensitivity, are interesting given recent clinical findings. Manzari et al. (2011) recently demonstrated that the ocular Vestibular-Evoked Myogenic Potential in response to BCV (which provides an estimate of utricular sensitivity), is often enhanced, or at least fluctuates in amplitude in early Ménière’s Disease sufferers following a vertigo attack. An enhanced utricular sensitivity in early-stage Ménière’s Disease sufferers is difficult to attribute to a rupture of the labyrinth. These recent clinical findings and the changes in VsEP amplitude we have observed during endolymph injections offer a potential insight into the mechanism underlying vertigo attacks of early stage Ménière’s sufferers. Acknowledgements This study was supported in part by an NHMRC project grant APP1044219, and by funds held by The Sydney Medical School Foundation, and raised by the Ménière’s Research Fund Inc., a notfor-profit organization in NSW, Australia. References Bohmer, A., 1995. Short latency vestibular evoked responses to linear acceleration stimuli in small mammals: masking effects and experimental applications. Acta Otolaryngol. Suppl. 520 (Pt 1), 120e123. Brown, D.J., Chihara, Y., Curthoys, I.S., Wang, Y., Bos, M., 2013. Changes in cochlear function during acute endolymphatic hydrops development in guinea pigs. Hear. Res. 296, 96e106. Chihara, Y., Wang, Y., Brown, D.J., 2013. Evidence for the Utricular Origin of the Vestibular Short-latency Evoked Potential (VsEP) to Bone Conducted Vibration in Guinea Pig, [Epub ahead of print] PPMID: 23780310. Cureoglu, S., Schachern, P.A., Paul, S., Paparella, M.M., Singh, R.K., 2004. Cellular changes of Reissner’s membrane in Meniere’s disease: human temporal bone
study. Otolaryngol. Head Neck Surg. 130 (1), 113e119. Flock A, Flock B. Microlesions in Reissner’s membrane evoked by acute hydrops. Audiol Neurootol. 2003; 8(2):59e69. Curthoys, I.S., Kim, J., McPhedran, S.K., Camp, A.J., 2006. Bone conducted vibration selectively activates irregular primary otolithic vestibular neurons in the guinea pig. Exp. Brain Res. 175 (2), 256e267. Duvall 3rd, A.J., Rhodes, V.T., 1967. Ultrastructure of the organ of Corti following intermixing of cochlear fluids. Ann. Otol. Rhinol. Laryngol. 76 (3), 688e708. Gibson, W.P., 2010. Hypothetical mechanism for vertigo in Meniere’s disease. Otolaryngol. Clin. North Am. 43 (5), 1019e1027. Jones, T.A., Jones, S.M., 1999. Short latency compound action potentials from mammalian gravity receptor organs. Hear. Res. 136, 75e85. Jones, T.A., Jones, S.M., Vijayakumar, S., Brugeaud, A., Bothwell, M., Chabbert, C., 2011. The adequate stimulus for mammalian linear vestibular evoked potentials (VsEPs). Hear. Res. 280, 133e140. Kakigi, A., Salt, A.N., Takeda, T., 2010. Effect of artificial endolymph injection into the cochlear duct on perilymph potassium. ORL J. Otorhinolaryngol. Relat. Spec. 71 (Suppl 1), 16e18. Kingma, C.M., Wit, H.P., 2009. Acute endolymphatic hydrops has no direct effect on the vestibular evoked potential in the guinea pig. J. Vest. Res. 19, 27e32. Kingma, C.M., Wit, H.P., 2010. The effect of changes in perilymphatic Kþ on the vestibular evoked potential in the guinea pig. Eur. Arch. Otorhinolaryngol. 267, 1679e1684. Kobayashi, T., Gyo, K., Yanagihara, N., 1999. Combined rupture of Reissner’s membrane and round window: an experimental study in guinea pigs: experimental double-membrane rupture. Am. J. Otol. 20 (2), 179e182. Manzari, L., Tedesco, A.R., Burgess, A.M., Curthoys, I.S., 2010. Ocular and cervical vestibular-evoked myogenic potentials to bone conducted vibration in Meniere’s disease during quiescence vs during acute attacks. Clin. Neurophysiol. 121 (7), 1092e1101. Manzari, L., Burgess, A.M., MacDougall, H.G., Bradshaw, A.P., Curthoys, I.S., 2011. Rapid fluctuations in dynamic semicircular canal function in early Meniere’s disease. Eur. Arch. Otorhinolaryngol. 268 (4), 637e639. Marcon, S., Patuzzi, R., 2008. Changes in cochlear responses in guinea pig with changes in perilymphatic Kþ. Part I: summating potentials, compound action potentials and DPOAEs. Hear. Res. 237 (1e2). Marcus, N.Y., Marcus, D.C., 1987. Potassium secretion by nonsensory region of gerbil utricle in vitro. Am. J. Physiol. 253 (4 Pt 2), 613e621. Marcus, D.C., Rokugo, M., Ge, X.X., Thalmann, R., 1983. Response of cochlear potentials to presumed alterations of ionic conductance: endolymphatic perfusion of barium, valinomycin and nystatin. Hear. Res. 12 (1), 17e30. McNeill, C., Cohen, M.A., Gibson, W.P., 2009. Changes in audiometric thresholds before, during and after attacks of vertigo associated with Meniere’s syndrome. Acta Otolaryngol. 129 (12), 1404e1407. Merchant, S.N., Adams, J.C., Nadol Jr., J.B., 2005. Pathophysiology of Ménière’s syndrome: are symptoms caused by endolymphatic hydrops? Otol. Neurotol. 26 (1), 74e81. Patuzzi, R., Sellick, P.M., Johnstone, B.M., 1984. The modulation of the sensitivity of the mammalian cochlea by low frequency tones. III. Basilar membrane motion. Hear. Res. 13 (1), 19e27. Plotnik, M., Sichel, J.Y., Elidan, J., Honrubia, V., Sohmer, H., 1999. Origins of the short latency vestibular evoked potentials (VsEPs) to linear acceleration impulses. Am. J. Otol. 20, 238e243. Rabbitt, R.D., Yamauchi, A.M., Boyle, R., Highstein, S.M., 2001. How endolymph pressure modulates semicircular canal primary afferent discharge. Ann. N. Y Acad. Sci. 942, 313e321. Salt, A.N., Plontke, S.K., 2010. Endolymphatic hydrops: pathophysiology and experimental models. Otolaryngol. Clin. N. Am., 43971e43983. Salt, A.N., Rask-Andersen, H., 2004. Responses of the endolymphatic sac to perilymphatic injections and withdrawals: evidence for the presence of a one-way valve. Hear. Res. 191 (1e2), 90e100. Schuknecht, H., 1976. Pathophysiology of endolymphatic hydrops. Arch. Oto-RhinoLaryngol. 212, 253e262. Schuknecht, H.F., Belal, A.A., 1975. The utriculo-endolymphatic valve: its functional significance. J. Laryngol. Otol. 89 (10), 985e996. Tonndorf, J., 1983. Vestibular signs and symptoms in Meniere’s disorder: mechanical considerations. Acta Otolaryngol. 95 (5e6), 421e430. Uzun-Coruhlu, H., Curthoys, I.S., Jones, A.S., 2007. Attachment of the utricular and saccular maculae to the temporal bone. Hear. Res. 233 (1e2), 77e85. Valk, W.L., Wit, H.P., Albers, F.W., 2004. Evaluation of cochlear function in an acute endolymphatic hydrops model in the guinea pig by measuring low-level DPOAEs. Hear. Res. 192 (1e2), 47e56. Valk, W.L., Wit, H.P., Albers, F.W., 2006. Rupture of Reissner’s membrane during acute endolymphatic hydrops in the guinea pig: a model for Meniere’s disease? Acta Otolaryngol. 126 (10), 1030e1035. Zenner, H.P., 1986. Motile responses in outer hair cells. Hear. Res. 22, 83e90. Zwicker, E., 1977. Masking period patterns produced by very low frequency maskers and their possible relation to basilar membrane displacement. J. Acoust. Soc. Am. 61 (4), 1031e1040.
Contents lists available at SciVerse ScienceDirect
Hearing Research journal homepage: www.elsevier.com/locate/heares
Research paper
Changes in utricular function during artificial endolymph injections in guinea pigs D.J. Brown a, *, Y. Chihara a, b, Y. Wang a a b
The Brain and Mind Research Institute, Sydney Medical School, The University of Sydney, 100 Mallett Street, Camperdown 2050, Australia National Institute of Sensory Organs, National Tokyo Medical Center, Tokyo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 March 2013 Received in revised form 7 May 2013 Accepted 27 May 2013 Available online 18 June 2013
Various theories suggest endolymphatic hydrops may cause a rupture of the membranous labyrinth or may force open the utriculo-saccular duct, resulting in a sudden change in inner ear function. Here, we have used slow injections of artificial endolymph into either scala media or the utricle of anaesthetised guinea pigs to investigate the effects of hydrops. Vestibular function was continuously monitored in addition to the measurements of cochlear function developed in our laboratory (Brown et al. Hear Res, 2013). Scala media injection induced consistent functional changes, which occurred in two stages. Initial changes involved were associated with an increased hydrostatic pressure in scala media that only affected cochlear function. After 3e4 ml of endolymph had been injected, cochlear function spontaneously recovered, and was often shortly followed by a transient increase or decrease in utricular sensitivity, with the effects varying between animals. Endolymph injection directly into the utricle produced variable effects across animals, although in 2 experiments it produced similar changes as those observed for scala media injections, suggesting that the fluid pathway between scala media and the utricle was continuous in these animals. The mechanism underlying the sudden, spontaneous functional changes is not yet clear, but we tentatively suggest that in some cases it may be caused by the utriculo-saccular duct suddenly opening to alleviate an elevated hydrostatic pressure in the pars inferior, resulting in a change in utricular function due to an increase in its volume. These changes are comparable to the sudden or fluctuating functional changes in Ménière’s sufferers, and support the hypothesis that endolymphatic hydrops can directly cause some symptoms of this syndrome. Crown Copyright Ó 2013 Published by Elsevier B.V. All rights reserved.
1. Introduction In support of Schuknecht’s 1976 rupture theory of vertigo attacks in Ménière’s Disease (MD), recent experimental studies have suggested that rapid injection of artificial endolymph into scala media in guinea pigs can cause a rupture of the membranous labyrinth (Valk et al., 2004, 2006), a mixing of perilymph and endolymph (Kakigi et al., 2010), and a severe loss of vestibular function (Kingma and Wit, 2010), supposedly due to a potassium intoxication of vestibular hair cells. In a recent study involving slow injections of artificial endolymph into scala media in guinea pigs (Brown et al., 2013) we observed repeated episodes of a slow decline in cochlear sensitivity followed by a spontaneous, sudden
Abbreviations: CAP, compound action potential; SP, summating potential; VsEP, vestibular evoked potential; EP, endocochlear potential; BCV, bone conducted vibration * Corresponding author. Tel.: þ61 2 93510748. E-mail addresses: [email protected], [email protected] (D.J. Brown), [email protected] (Y. Chihara).
recovery of function that could not be easily explained on the basis of a labyrinth rupture. A rupture of the membranous labyrinth allowing mixing of perilymph and endolymph would be expected to cause a severe reduction in cochlear and vestibular sensitivity (Zenner, 1986; Kobayashi et al., 1999; Marcon and Patuzzi, 2008; Kingma and Wit, 2010). Instead, as proposed in previous studies (Schuknecht and Belal, 1975; Salt and Rask-Andersen, 2004), we tentatively suggested that at certain stages of hydrops development the utriculo-saccular duct (also termed the valve of Bast) or the endolymphatic sinus may suddenly open (Gibson, 2010) allowing a rapid “de-compression” of endolymph pressure in the cochlea, causing a sudden recovery of cochlear function and possibly a sudden change in vestibular function. Whatever the mechanism underlying the abrupt recovery of cochlear function observed during artificial hydrops development, it could potentially explain recent clinical findings that are also difficult to explain on the basis of a rupture. That is, it has recently been demonstrated that in the early stages of Ménière’s Disease utricular and semicircular canal sensitivity is enhanced (Manzari et al., 2010, 2011), and that Ménière’s sufferers do not lose their
0378-5955/$ e see front matter Crown Copyright Ó 2013 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.heares.2013.05.011
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
hearing at the time of a vertigo attack (McNeill et al., 2009). An abrupt opening of physiological valves in the inner ear could potentially allow rapid changes in cochlear or vestibular function due to the relief of hydrops pressure in the pars inferior, without causing a mixing of perilymph and endolymph. Such a mechanism could also cause a sudden increase in the volume of the utricle and semicircular canals (pars superior), resulting in a mechanical displacement of the utricular macula and a stretching of the cupula, which can potentially change their sensitivity (Rabbitt et al., 2001; Chihara et al., 2013). To determine the effects of slow endolymph injections on utricular function, we have slowly injected artificial endolymph (at 40e80 nl/min) into the membranous labyrinth in guinea pigs whilst monitoring several measures of both cochlear (as in our previous study; Brown et al., 2013) and utricular function. As a sidenote, to the best of our knowledge this is the first time simultaneous measurements of both cochlear and vestibular function have been performed during in vivo animal experiments. Vestibular function can be estimated by monitoring the Vestibular Evoked short-latency Potential (VsEP), which is a gross compound action potential of neurons innervating the jerksensitive hair cells in the otoliths, in response to brief head acceleration pulses (Bohmer, 1995; Plotnik et al., 1999; Jones and Jones, 1999; Jones et al., 2011). Moreover, we recently provided strong evidence that the VsEP evoked by bone-conducted vibration (BCV) pulses (in the presence of acoustic masking noise to suppress cochlear responses) is almost entirely a response of neurons innervating the utricle (Chihara et al., 2013). We typically measure the VsEP from the recording electrode inserted into the facial nerve canal, which provides a close electrogenic proximity to the superior vestibular nerve from which the VsEP response originates. Kingma and Wit (2010) recently demonstrated that the VsEP amplitude decreased abruptly during rapid injections of artificial endolymph, supposedly due to a rupture of the membranous labyrinth. However, the rapid-rate of endolymph injection in this study prevented the assessment of subtle changes in the vestibular function, which, an earlier study by Kingma and Wit (2009) suggested may potentially include an enhanced VsEP amplitude due to an increase in endolymph pressure. The present study investigated the changes in vestibular function during relatively slow injections of artificial endolymph into scala media. We also utilized several measurements of cochlear function, such as Compound Action Potential (CAP) thresholds, Summating Potential (SP)/CAP ratios and cochlear microphonic analysis, as an indicator of the sudden changes in endolymph pressure based on our previous study (Brown et al., 2013). However, here we have omitted the SP/CAP and cochlear microphonic results in favour of focussing on the injection induced changes in vestibular sensitivity. The CAP thresholds alone are used here as an indicator of the injection induced changes in cochlear function. We have also attempted to inject endolymph directly into the utricle, to determine if physiological changes in the VsEP with endolymph injection are due to a hydrostatic pressure increase specifically in the utricle, and also to test the putative role of the physiological valves that separate the pars superior from the pars inferior, namely the utriculo-saccular duct (Schuknecht and Belal, 1975) and the endolymphatic sinus (Salt and Rask-Andersen, 2004). 2. Methods 2.1. Animal preparation & surgery Experiments were performed on 19 normal adult guinea pigs (Cavia porcellus) of either sex with body weights between 250 and 450 g. Partial results from 2 animals used in this study (GP120 and
71
GP139; Fig. 2A, D and N) were presented in our previous publication (Brown et al., 2013). All animal preparation, surgery and experimental protocol were approved by The University of Sydney’s Animal Ethics Committee. Animals were anaesthetised in a gas anaesthesia chamber with 3% isoflurane, and also given a onceoff 0.05 ml intraperitoneal injection of Temgesic (Buprenorpherine; Reckitt Benckiser, Auckland NZ) and 0.1 ml of Atrosine (0.6 mg/ml atropine sulphate; Apex Laboratories, NSW, Australia). Once sedate and the foot-withdrawal reflex was absent, animals were moved to the experimental setup in the sound-isolation booth, tracheotomized, and thereafter artificially ventilated with a mixture of oxygen and isoflurane (1e2%) by an artificial respirator pump. The animal’s core temperature was regulated to 38 C (2 C) by a rectal temperature probe and an electric heating pad, and the sound-isolated room was heated with an infrared heating lamp. After the main surgery, animals were given a 0.1 ml intramuscular injection of the neuromuscular blocker Pauvulon (2 mg/ml pancuronium bromide; Astra Pharmaceuticals, L.P.) to suppress middle ear reflexes and general muscle activity. To expose the cochlea, the bulla was opened via a dorso-lateral approach, providing a direct view of the round window and the basal cochlear turn. A thin tissue wick was placed in the bulla to minimize any acoustical, functional or electrical changes due to fluid build-up in the bulla. To expose the utricle and semicircular canals, the bulla was opened just superficial to the external auditory canal, providing a view of the where the superior and lateral ampullae meet with the utricle. In all experiments the animal was placed in a custom-designed ear bar, which housed a low-noise microphone (ER10Bþ, Etymotic Inc., IL, USA), and the ends of 3 silastic tubes (20 cm long) which were connected to 3 separate headphone speakers, forming a closed acoustic sound-field in the ear canal. To produce skull vibration stimuli a standard audiometric bone oscillator (Radioear, New Eagle, PA, USA), surrounded by custom electromagnetic shielding, was rigidly attached to the caudal surface of the guinea pigs skull via a metal bolt (Fig. 1). This involved removing a 2 mm 4 mm rectangular section of caudal skull bone contralateral to the saggital suture line (i.e. overlying the contralateral vestibular
Fig. 1. Schematic diagram of the experimental setup. BCV was delivered via a B71 bone vibrator surgically attached to the skull, with the skull vibration measured with an accelerometer also attached to the skull bone. Endolymph was injected into the labyrinth via a 5e10 mm tip glass micropipette inserted into either scala media via the round window, or into the utricle via a small hole drilled into the bone overlying the horizontal and superior semicircular canal commissure. The glass pipette was also used to measure the EP. Acoustic stimuli were delivered via speaker tubing sealed into the ear-bars, which also housed a low-noise microphone for sound calibration. Gross cochlear responses were measured from an AgCl wire sealed into scala tympani of the first cochlear turn, with the VsEP measured from a wire inserted 1 cm into the facial nerve canal.
72
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
system) using a handheld micro-drill, then inserting the flat “T” shaped end of a 4 cm long bolt into the opening (superficial to the dura), rotating the bolt by 90 , then tightening the bolt to the bone using a nut on the shank of the bolt (see Curthoys et al., 2006). To measure the skull vibration, a 3-axis accelerometer (Dimension Engineering, OH, USA) was also bolted to the temporal bone using the same technique used to attach the B71, but located over the ipsilateral vestibular system (i.e. ipsilateral to the saggital suture line on the caudal skull bone). While all 3 axis of the accelerometer output were monitored during recordings, for the purpose of simplicity, here we have only focussed on acceleration perpendicular to the plane of the frontal bone (i.e. transverse vibration of the skull). 2.2. Physiological measures To measure cochlear responses, a 0.2 mm diameter hole was made in the first turn of the cochlea, through which the exposed tip of a teflon-coated AgCl recording wire was placed, and then sealedin using superglue. To inject artificial endolymph and to record the endocochlear potential, a 5e10 mm diameter tip, 5e10 MU glass micropipette was inserted into scala media via the round window and basilar membrane. In this study, the composition of artificial endolymph consisted of (in mM): KCl, 126; NaCl, 1; KHCO3, 25; MgCl2, 0.025; CaCl2, 0.025; K2HPO4, 1.4; mannitol, 25; and had a pH of 7.4 and an osmolarity of 310e320 mOsmol, as previously described by Marcus et al. (1983). In 6 animals, the micropipette was inserted into the utricle through a small hole made in the horizontal semicircular canal ampulla/utricle commissure. Prior to making a hole, the bone was shaved slightly and a thin layer of hydrophilic glue (“Gorilla Glue”, Cincinnati, Ohio) was applied to the bone and allowed to dry for 5e 10 min. A hole was then made through the glue and bone, and the pipette tip was inserted into the utricle under micromanipulation control, before being sealed in place with superglue (Selleys Plastic Glue, DuluxGroupPty Ltd., Australia) to prevent fluid leakage from the hole. The difficulty with this procedure was that often the glass pipette appeared to produce either a large tear in the membranous labyrinth bounding the utricle (which was not easily visible due to the transparency of the membrane when covered with fluid), or it appeared to push the membranous labyrinth away from the bony wall. It was assumed that the pipette had been successfully inserted and sealed into the utricle on the basis of two findings. First, when inserting the pipette, the voltage at the tip of the glass pipette ’jumped’ abruptly by þ3 to þ5 mV (relative to the potential measured at the fluid overlying the membranous labyrinth), which previous studies have suggested is the resting potential of endolymph in the pars superior (Marcus and Marcus, 1987). Second, the functional changes observed later in the experiment appeared to support a correct insertion, as discussed later. For both scala media and utricular injections, the glass micropipette was connected to a 10 ml glass syringe, attached to an UltramicroPump III perfusion pump (WPI, Sarasota, FL). Prior to inserting the glass pipette, the tip of the pipette was visually inspected while injecting into air to confirm an immediate injection of fluid, and to confirm there was no “lag” in the injection onset. Physiological potentials were amplified using a customdesigned low-noise AC and high-impedance DC amplifier. Amplified signals were then digitized using a battery powered Wi-Fi data acquisition device (NI 9205, National Instruments, TX), and wirelessly transmitted to PC. The 3 speakers, the B71 bone conductor and the low-noise microphone signal were driven via an external USB soundcard (UltraLite-mk3 Hybrid, MOTU Cambridge, MA). All recordings and stimulus generation were performed with custom designed LabVIEW programs (National Instruments, TX).
The CAP responses were measured from the scala tympani wire in response to a train of high-frequency acoustic tone-bursts. This tone-burst train consisted of 15, 11, 6, 4 and 2 kHz tone-bursts, each 10 ms in duration and separated by 20 ms of silence, with a 1 ms cosine ramp on and off. CAP thresholds were determined programmatically using a custom software algorithm that adjusted the sound level of each tone-burst independently, until such time that the correlation between successive averaged responses (with 8 averages for each response) was close to 0.4 (0.05). This technique is similar to performing a visual detection of CAP thresholds, where the visual inspection of the response relative to the noise is used to subjectively determine a correlation threshold. To evoke VsEP responses, we aimed to produce a BCV stimulus that produced a brief (<1 ms) biphasic acceleration of the animal’s skull. However, delivering a 1 ms BCV pulse to the B71 resulted in a complex resonance of the animal’s skull (as determined from the skull-fixed accelerometer). This typically resulted in a complex VsEP waveform which differed between animals. As detailed in a recent study (Chihara et al., 2013), to partially correct for this resonance we used a standard engineering technique. This involved measuring the skull acceleration in response to a simple BCV stimulus consisting of a single 1 kHz sinusoid (with a Gaussian window), and then dividing the FFT of the stimulus by the FFT of the skull acceleration pattern. This resultant FFT was then multiplied by the FFT of the simple stimulus, and an inverse FFT was performed on this ’corrected’ spectrum, to provide a “corrected stimulus”. When delivered to the B71, this corrected stimulus generated a single-cycle 1 kHz, transverse acceleration of the animal’s skull. In short, we examined the vibration properties the head and corrected for the nonlinearities of the bone-conducted vibration, which is similar to calibrating a sound-field using a microphone, and then delivering a calibrated sound stimulus. This simple 1 ms acceleration of the animal’s skull resulted in a relatively simple VsEP waveform that was more consistent across animals. Broadband acoustic masking noise with a reasonably flat frequency response (10 dB, 10 Hz to 16 kHz) was simultaneously presented with the BCV stimulus, to both ear canals at 105 dB SPL. Acoustic noise was only presented during VsEP measurement, which generally lasted 3e4 s, allowing for 40e80 averages per averaged VsEP waveform, where successive BCV stimuli were presented at a rate of 16/second. The measurement of EP, CAP thresholds, and VsEP were performed systematically in series, with the LabVIEW programs (each measure was obtained using a separate LabVIEW program) continuously looping through each measurement subset. This allowed for a relatively high temporal resolution of measurements throughout endolymph injection experiments. 3. Results 3.1. Functional changes during slow injections of endolymph into scala media Following the insertion of the glass micropipette into scala media, all functional measurements were stable relative to the changes induced by the injection, at least over a 2 h period (data not shown, but see Brown et al., 2013). Fig. 2 shows examples of the functional changes during endolymph injection measured in 5 animals. Around 10 min after beginning the injection (which is indicated as a black line in Fig. 2KeO), there was a gradual increase in CAP thresholds, starting with the 2 kHz CAP threshold similar to that observed in our previous study (Fig. 2AeE), and a gradual decline in the EP (Fig. 2KeO). During this phase of the injection, the VsEP amplitude remained relatively stable in most animals, although in some animals it increased slightly, and in others it
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
73
Fig. 2. Changes in cochlear and vestibular function during endolymph injection into scala media in 5 representative animals. AeE) CAP thresholds to 10 ms duration tone-bursts of 15, 11, 6, 4 and 2 kHz frequencies. FeJ) The N1eP1 amplitude of the VsEP, KeO) The EP measured from the scala media pipette. The duration of the injection is indicated as a horizontal line in KeO, with the injection rate (in nl/minute) and injected volume (in ml, in parentheses) shown above each line. Dark arrows in AeE indicate the episodes of cochlear function recovery. Roman numerals indicate periods at which the representative responses used in Fig. 3 were made. Asterisks indicate periods where the VsEP amplitude was transiently increased.
gradually decreased (Fig. 2FeJ). In all experiments, after 2.5e3.5 ml of endolymph had been injected, the changes in cochlear function abruptly recovered (thick black arrow Fig. 2AeE). Following the recovery of cochlear function, the VsEP amplitude either remained relatively stable, rapidly increased, or rapidly decreased several minutes after the abrupt recovery. While changes in cochlear function were consistent and resembled that reported previously (Brown et al., 2013), there was no apparent consistency in the changes in the VsEP amplitude across 14 different injection experiments. Examples of the VsEP waveform before and after the rapid decrease in VsEP amplitude is shown in Fig. 3, with the timing of each waveform corresponding to the time points indicated with roman numerals in Fig. 2G. As a simple test of the effects of different injection rates, the initial injection was performed at a rate of 40 nl/min in 11 animals, 80 nl/ min in 2 animals, and 100 nl/min in 1 animal. The injection was typically ceased approximately 5e10 min after an abrupt recovery episode occurred, although on a two occasions the injection was continued for an additional 30e40 min following the recovery episode. On average, the volume of endolymph injected prior to an abrupt recovery episode was 3.3 ml 0.5 ml. The rate of the injection did not appear to have any obvious additional impact on the functional changes, other than reducing the time taken to induce the abrupt recovery episode. Often, multiple injections were performed in the same experiment, allowing 30e40 min for the recovery of the injection induced changes (see Fig. 2, gp139, gp152, gp154). In all experiments, if the VsEP amplitude did not decrease following the first injection and cochlear recovery episode, it did after the second injection (Fig. 2H and I). In one experiment, the injection was performed 3 times over a 5.5 h period, resulting in multiple reductions and recoveries of the VsEP in the same animal (Fig. 2J).
Fig. 3. Examples of VsEP responses at various stages during endolymph injection into scala media. Responses correspond to the time-points in Fig. 2, indicated using Roman Numerals I and II. The VsEP was evoked by a BCV pulse which produced a vibration of the animal’s skull shown at the bottom of this Figure.
While there were subtle differences between the effects of endolymph injection between different animal experiments, such as the volume required to induce a sudden recovery of cochlear function, changes were reliably and consistently reproduced in 8 experiments where endolymph was injected at a rate of 40 nl/min (in 3 other experiments utilizing 40 nl/min injection into scala media, the experiment was intentionally stopped just after the abrupt recovery episode, for the purpose of imaging the inner ear morphology at this stage). The averaged changes in each measurement (with standard deviations) throughout the first endolymph injection in these 8 experiments are shown in Fig. 4. Here,
Fig. 4. Average functional measurements during 40 nl/min endolymph injection into scala media in 8 animals, with standard deviations. A) CAP thresholds to a 2 kHz toneburst, B) the N1eP1 amplitude of the VsEP evoked by a 1e2 g BCV pulse, and C) the EP measured from the scala media pipette. In each animal, the timing of each measurement was first referenced to the initial major episode of recovery of cochlear function, and then down-sampled to a measurement every 2 min to enable averaging of the data cross-animals. The average onset and offset of the injection relative to the recovery episode is shown in C, along with standard deviation. The average volume of endolymph injected was 3.2 0.55 ml.
74
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
for the purposes of averaging the time-charts across different animals, the timing of each individual time-chart was referenced to the first major recovery of cochlear function, before being averaged with the results from other animals. The average start and end of the injection period relative to the abrupt recovery episode, with standard deviation are indicated in Fig. 4C. 3.2. Utricular injection In an attempt to determine if the reduction in VsEP amplitude was due to a rupture of the membranous labyrinth, or due to a hydrostatic pressure increase in the pars superior, endolymph was injected directly into the utricle in 6 animals. In 2 of these experiments, CAP thresholds increased throughout the initial stage of the injection, however, there was only a gradual decline in the VsEP amplitude (Fig. 5, left two columns). The EP changed relatively little throughout these experiments, although the initial EP measured in the utricle and was only 5 mV (see Section 2.2). Again, after about 3.6 ml of endolymph had been injected into the membranous labyrinth there was a sudden recovery of cochlear function (Fig. 5, black arrows), although the recovery of function was not as distinctive as that observed in scala media injections. In 4 of these utricular injection experiments, there was little change in cochlear function throughout or after the injection, but the VsEP amplitude rapidly decreased soon after starting the injection, when only 1e2 ml had been injected (Fig. 5, grey arrows). The VsEP amplitude did not recover following this rapid decrease. 4. Discussion 4.1. Functional changes throughout endolymph injections As suggested in our previous study (Brown et al., 2013), the initial changes in cochlear function observed during endolymph injection into scala media, namely the increase in CAP thresholds were most likely due to an increase in the hydrostatic pressure of endolymph causing a displacement of the basilar membrane towards scala tympani (discussed below). The present study introduced simultaneous measures of utricular function with acute endolymph injections, demonstrating that while cochlear sensitivity decreased due to the increase in endolymph pressure, there was little change in utricular function. This suggests that either the utricle is insensitive to hydrops, or that the injected endolymph did not cause a displacement of the utricular macula. As observed in our previous study, after 3e4 ml of endolymph had been injected
into scala media, cochlear function abruptly recovered. This recovery was typically followed by marked changes in utricular sensitivity, often a rapid reduction in sensitivity, but on occasions a transient increase in sensitivity was observed.
4.2. Functional changes are related to changes in hydrostatic pressure As demonstrated previously endolymph injection can cause a gradual increase, then sudden drop in endolymph pressure (Valk et al., 2006; Kingma and Wit, 2010). While we did not directly measure hydrostatic pressure here, we have assumed the changes in cochlear function were directly caused by changes in endolymph pressure, causing a displacement of the basilar membrane. Displacement of the basilar membrane can bias the average resting conductance of the outer hair cells, impairing the function of the cochlear amplifier and causing a loss of acoustic sensitivity (Zwicker, 1977; Patuzzi et al., 1984). While the functional changes may have been due to altered ionic composition of the fluids or changes in the receptor current, potentially affecting outer hair cell function, there are at least two reasons why the changes were more likely due to hydrostatic bias of the basilar membrane. First, that low-frequency thresholds were elevated before higher frequency thresholds suggests that the changes were induced by an increased hydrostatic pressure. That is, the compliance of the basilar membrane increases towards the apex of the cochlea, and more apical regions of the basilar membrane would be displaced first during increased endolymph pressure, and thus hydrops should cause the most severe loss of hearing sensitivity to low-frequency sounds (Tonndorf, 1983). Second, given that the hearing loss spontaneously recovered so abruptly (often within 10 s) after 3e3.5 ml of endolymph had been injected, it is more likely that the initial hearing loss and sudden recovery was dominated by mechanical changes, and moreover, due to a sudden relief from an abnormally large hydrostatic pressure in the endolymph.
4.3. The pressure relief event e rupture or shunt? There are at least two likely explanations for the sudden relief of pressure allowing recovery of cochlear function, while causing a sudden change in utricular sensitivity. The pressure release may be either due to a rupture of the membranous labyrinth (Schuknecht, 1976; Valk et al., 2006; Kingma and Wit, 2010), or a sudden shunt of endolymph from the pars inferior into the pars superior via some
Fig. 5. Changes in cochlear and vestibular function during endolymph injection into the utricle in 4 representative animals. AeD) CAP thresholds EeH) The VsEP amplitude, IeL) The EP measured from the utricle pipette. The duration of the injection is indicated as a horizontal line in IeL, with the injection rate and injected volume (in parentheses) shown near each line. Dark arrows in A and B indicate episodes of cochlear function recovery. Light arrows indicate where the VsEP amplitude decreased.
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
form of physiological valve mechanism (Schuknecht and Belal, 1975; Salt and Rask-Andersen, 2004). In support of the rupture theory, previous researchers have demonstrated that increasing the Kþ concentration of perilymph causes a significant reduction in the VsEP amplitude, and that injecting more than 3 ml of endolymph into scala media also results in a reduction of the VsEP (Kingma and Wit, 2010). While we did observe a rapid reduction in the VsEP amplitude following the recovery of cochlear function in several of our experiments, in others we observed either no change (Fig. 2H and I), or a transient increase (Fig. 2F and G). In these experiments, it is unlikely that a rupture of the labyrinth allowing endolymph and perilymph mixing was the cause of the enhanced VsEP amplitude because increasing the Kþ concentration in perilymph only produces a reduction in the VsEP amplitude. Previous imaging studies have demonstrated ruptures of the labyrinth following endolymph injection in guinea pigs (Valk et al., 2004); however, the imaging techniques used in these previous studies can lead to artificial tears in thin membranes, due to the need for decalcification and optical clearing. In a previous study which used imaging techniques free from these procedures, namely X-ray micro-tomography, no evidence of any ruptures following 4 ml of injected endolymph were found, nor was any mixing of the inner ear fluids evident when a contrast enhancement agent was added to the injected endolymph (see Fig. 6D2 from Brown et al., 2013). Further suggesting that the pressure relief during endolymph injections is not due to a rupture is the finding that repeated endolymph injections in the same cochlea often results in successive loss of cochlear sensitivity, followed by a successive abrupt recovery episodes and a repeated loss of utricular sensitivity (Fig. 2, gp154, gp139, gp152). That is, while experimentally induced ruptures of the labyrinth have been shown to heal over several days (Duvall III and Rhodes, 1967), it is difficult to envisage how such a tear could heal within 30e40 min to allow a second increase in endolymph pressure with injection, and a second abrupt pressure relief and cochlear recovery, with a subsequent loss of vestibular function. Since its initial description, the utriculo-saccular duct (also called the valve of Bast) has been suggested to act like a valve, separating the utricle and semicircular canals from the saccule and cochlea, except in cases of endolymphatic hydrops which might force the duct open and allow a shunt of endolymph into the pars superior (Schucknecht and Belal, 1975). Likewise, the endolymphatic sinus has been suggested to act like a valve, opening to allow saccular endolymph to flow towards the endolymphatic duct (Salt and Rask-Andersen, 2004). Given the sensitivity of the cochlea to static pressure gradients between endolymph and perilymph, it seems reasonable that we might expect to observe some sudden change in cochlear function related to the opening of the utriculosaccular duct or endolymphatic sinus, if indeed these structures act like a physiological valve (Fig. 6). What’s less clear, is what we might expect to happen to utricular sensitivity if utricular volume suddenly increased. It has been demonstrated that around 2/3rds of the utricular macula rests on a compliant membrane separating the perilymphfilled vestibule from the endolymph-filled utricle in guinea pigs (Uzun-Coruhu et al., 2007). A recent study (Chihara et al., 2013) demonstrated that slight displacement of the utricular macula altered its sensitivity. Furthermore, an increased hydrostatic pressure in the utricle was suggested as the cause of a slight increase in the amplitude of the VsEP during acute endolymph injections in guinea pigs (Kingma and Wit, 2009). It is therefore possible that an increase in utricle volume could cause a mechanical displacement of the utricular macula, resulting in a change in its sensitivity.
75
Fig. 6. Schematic diagram of the putative changes in membranous labyrinth during hydrops development. Here, hydrops initially develops in the pars inferior only because the utriculo-saccular duct is closed. At some stage during hydrops development, the utriculo-saccular duct opens, allowing an influx of endolymph into the utricle and semicircular canals (SSCs), and a reduction in the volume of the pars inferior. This might allow recovery from a pressure-induced hearing loss, but simultaneously cause a mechanical displacement of the utricular macula, altering its sensitivity to BCV.
4.4. Utricular injections In an attempt to test if changes in utricular volume alter its sensitivity, we injected artificial endolymph directly into the utricle. As previously stated, this was a far more technically challenging technique than scala media injections, because the pipette must be inserted through the compliant and translucent membranous labyrinth without tearing it. In two of these experiments, injection at a rate of 80 nl/min resulted in a slow increase in CAP thresholds and a subsequent sudden recovery of these changes, but only a slow decline in the VsEP amplitude (Fig. 5E and F). The fact that these changes closely resembled those consistently observed in scala media injections suggests that we had successfully sealed and injected fluid into the utricle via the glass pipette, resulting in increased endolymph pressure in not only the utricle, but also the cochlea, and therefore also suggests that the utriculo-saccular duct was patent in these experiments. In another four utricular injection experiments, there was no change in cochlear function during an injection of 3e4 ml of endolymph, but there was a severe and sudden reduction in the VsEP amplitude. Potentially, the reduction in the VsEP here was either because the glass pipette was not properly sealed into the utricle, allowing endolymph to mix with vestibular perilymph, causing a depolarization of the utricular hair cell, or alternatively, in these ears the utriculo-saccular duct may have been closed, resulting in a relatively rapid increase in utricular volume, causing a rapid decrease in utricular sensitivity. Ultimately, with less certainty regarding the actual injection of endolymph into the utricle, it is difficult to interpret functional changes observed in these utricle injection experiments. We are currently exploring new
76
D.J. Brown et al. / Hearing Research 304 (2013) 70e76
techniques for more accurately determining where the injected endolymph is distributed in these injection experiments, which will ultimately allow us to determine the patency and function of the utriculo-saccular duct and endolymphatic sinus. 4.5. Artificial endolymph injection as a model for vertigo attacks in Ménière’s Disease Acute endolymph injection provides a simple experimental animal model of pathological endolymphatic hydrops (Salt and Plontke, 2010), and is a simple way to investigate possible causes of symptoms in Ménière’s Disease. While hydrops itself does not universally induce the symptoms of Ménière’s Disease (Merchant et al., 2005), this might be explained on the grounds that in some non-Ménière’s ears endolymph volume may increase sufficiently slowly that hyperplasia of the membranous labyrinth prevents an increase in endolymph pressure (due to an increased compliance of the labyrinth), and there is therefore no mechanical displacement of the cochlear and vestibular hair cells. However, while previous studies have found differences in cell densities between non-MD and MD patients with hydrops, the results are complex and vary with cell type and location (Cureoglu et al., 2004). It is also important to note that while Ménière’s sufferers appear to have hydrops consistently, they are not in a consistent stage of vertigo or hearing loss, and some discontinuous physiological event is required to explain the sudden onset of vertigo in Ménière’s. The theory that fluctuations in endolymph pressure causes fluctuating displacement of cochlear hair cells is not new (Tonndorf, 1983), and can explain why low-frequency sensitivity fluctuates most severely in Ménière’s sufferers (McNeill et al., 2009). The changes in utricular sensitivity we have observed in the present study, particularly the transiently enhanced sensitivity, are interesting given recent clinical findings. Manzari et al. (2011) recently demonstrated that the ocular Vestibular-Evoked Myogenic Potential in response to BCV (which provides an estimate of utricular sensitivity), is often enhanced, or at least fluctuates in amplitude in early Ménière’s Disease sufferers following a vertigo attack. An enhanced utricular sensitivity in early-stage Ménière’s Disease sufferers is difficult to attribute to a rupture of the labyrinth. These recent clinical findings and the changes in VsEP amplitude we have observed during endolymph injections offer a potential insight into the mechanism underlying vertigo attacks of early stage Ménière’s sufferers. Acknowledgements This study was supported in part by an NHMRC project grant APP1044219, and by funds held by The Sydney Medical School Foundation, and raised by the Ménière’s Research Fund Inc., a notfor-profit organization in NSW, Australia. References Bohmer, A., 1995. Short latency vestibular evoked responses to linear acceleration stimuli in small mammals: masking effects and experimental applications. Acta Otolaryngol. Suppl. 520 (Pt 1), 120e123. Brown, D.J., Chihara, Y., Curthoys, I.S., Wang, Y., Bos, M., 2013. Changes in cochlear function during acute endolymphatic hydrops development in guinea pigs. Hear. Res. 296, 96e106. Chihara, Y., Wang, Y., Brown, D.J., 2013. Evidence for the Utricular Origin of the Vestibular Short-latency Evoked Potential (VsEP) to Bone Conducted Vibration in Guinea Pig, [Epub ahead of print] PPMID: 23780310. Cureoglu, S., Schachern, P.A., Paul, S., Paparella, M.M., Singh, R.K., 2004. Cellular changes of Reissner’s membrane in Meniere’s disease: human temporal bone
study. Otolaryngol. Head Neck Surg. 130 (1), 113e119. Flock A, Flock B. Microlesions in Reissner’s membrane evoked by acute hydrops. Audiol Neurootol. 2003; 8(2):59e69. Curthoys, I.S., Kim, J., McPhedran, S.K., Camp, A.J., 2006. Bone conducted vibration selectively activates irregular primary otolithic vestibular neurons in the guinea pig. Exp. Brain Res. 175 (2), 256e267. Duvall 3rd, A.J., Rhodes, V.T., 1967. Ultrastructure of the organ of Corti following intermixing of cochlear fluids. Ann. Otol. Rhinol. Laryngol. 76 (3), 688e708. Gibson, W.P., 2010. Hypothetical mechanism for vertigo in Meniere’s disease. Otolaryngol. Clin. North Am. 43 (5), 1019e1027. Jones, T.A., Jones, S.M., 1999. Short latency compound action potentials from mammalian gravity receptor organs. Hear. Res. 136, 75e85. Jones, T.A., Jones, S.M., Vijayakumar, S., Brugeaud, A., Bothwell, M., Chabbert, C., 2011. The adequate stimulus for mammalian linear vestibular evoked potentials (VsEPs). Hear. Res. 280, 133e140. Kakigi, A., Salt, A.N., Takeda, T., 2010. Effect of artificial endolymph injection into the cochlear duct on perilymph potassium. ORL J. Otorhinolaryngol. Relat. Spec. 71 (Suppl 1), 16e18. Kingma, C.M., Wit, H.P., 2009. Acute endolymphatic hydrops has no direct effect on the vestibular evoked potential in the guinea pig. J. Vest. Res. 19, 27e32. Kingma, C.M., Wit, H.P., 2010. The effect of changes in perilymphatic Kþ on the vestibular evoked potential in the guinea pig. Eur. Arch. Otorhinolaryngol. 267, 1679e1684. Kobayashi, T., Gyo, K., Yanagihara, N., 1999. Combined rupture of Reissner’s membrane and round window: an experimental study in guinea pigs: experimental double-membrane rupture. Am. J. Otol. 20 (2), 179e182. Manzari, L., Tedesco, A.R., Burgess, A.M., Curthoys, I.S., 2010. Ocular and cervical vestibular-evoked myogenic potentials to bone conducted vibration in Meniere’s disease during quiescence vs during acute attacks. Clin. Neurophysiol. 121 (7), 1092e1101. Manzari, L., Burgess, A.M., MacDougall, H.G., Bradshaw, A.P., Curthoys, I.S., 2011. Rapid fluctuations in dynamic semicircular canal function in early Meniere’s disease. Eur. Arch. Otorhinolaryngol. 268 (4), 637e639. Marcon, S., Patuzzi, R., 2008. Changes in cochlear responses in guinea pig with changes in perilymphatic Kþ. Part I: summating potentials, compound action potentials and DPOAEs. Hear. Res. 237 (1e2). Marcus, N.Y., Marcus, D.C., 1987. Potassium secretion by nonsensory region of gerbil utricle in vitro. Am. J. Physiol. 253 (4 Pt 2), 613e621. Marcus, D.C., Rokugo, M., Ge, X.X., Thalmann, R., 1983. Response of cochlear potentials to presumed alterations of ionic conductance: endolymphatic perfusion of barium, valinomycin and nystatin. Hear. Res. 12 (1), 17e30. McNeill, C., Cohen, M.A., Gibson, W.P., 2009. Changes in audiometric thresholds before, during and after attacks of vertigo associated with Meniere’s syndrome. Acta Otolaryngol. 129 (12), 1404e1407. Merchant, S.N., Adams, J.C., Nadol Jr., J.B., 2005. Pathophysiology of Ménière’s syndrome: are symptoms caused by endolymphatic hydrops? Otol. Neurotol. 26 (1), 74e81. Patuzzi, R., Sellick, P.M., Johnstone, B.M., 1984. The modulation of the sensitivity of the mammalian cochlea by low frequency tones. III. Basilar membrane motion. Hear. Res. 13 (1), 19e27. Plotnik, M., Sichel, J.Y., Elidan, J., Honrubia, V., Sohmer, H., 1999. Origins of the short latency vestibular evoked potentials (VsEPs) to linear acceleration impulses. Am. J. Otol. 20, 238e243. Rabbitt, R.D., Yamauchi, A.M., Boyle, R., Highstein, S.M., 2001. How endolymph pressure modulates semicircular canal primary afferent discharge. Ann. N. Y Acad. Sci. 942, 313e321. Salt, A.N., Plontke, S.K., 2010. Endolymphatic hydrops: pathophysiology and experimental models. Otolaryngol. Clin. N. Am., 43971e43983. Salt, A.N., Rask-Andersen, H., 2004. Responses of the endolymphatic sac to perilymphatic injections and withdrawals: evidence for the presence of a one-way valve. Hear. Res. 191 (1e2), 90e100. Schuknecht, H., 1976. Pathophysiology of endolymphatic hydrops. Arch. Oto-RhinoLaryngol. 212, 253e262. Schuknecht, H.F., Belal, A.A., 1975. The utriculo-endolymphatic valve: its functional significance. J. Laryngol. Otol. 89 (10), 985e996. Tonndorf, J., 1983. Vestibular signs and symptoms in Meniere’s disorder: mechanical considerations. Acta Otolaryngol. 95 (5e6), 421e430. Uzun-Coruhlu, H., Curthoys, I.S., Jones, A.S., 2007. Attachment of the utricular and saccular maculae to the temporal bone. Hear. Res. 233 (1e2), 77e85. Valk, W.L., Wit, H.P., Albers, F.W., 2004. Evaluation of cochlear function in an acute endolymphatic hydrops model in the guinea pig by measuring low-level DPOAEs. Hear. Res. 192 (1e2), 47e56. Valk, W.L., Wit, H.P., Albers, F.W., 2006. Rupture of Reissner’s membrane during acute endolymphatic hydrops in the guinea pig: a model for Meniere’s disease? Acta Otolaryngol. 126 (10), 1030e1035. Zenner, H.P., 1986. Motile responses in outer hair cells. Hear. Res. 22, 83e90. Zwicker, E., 1977. Masking period patterns produced by very low frequency maskers and their possible relation to basilar membrane displacement. J. Acoust. Soc. Am. 61 (4), 1031e1040.