the vestibular sedative flunarizine upon systems · flunarizine 5 mgtwice daily for oneweekandthen...

Journal of Neurology, Neurosurgery, and Psychiatry 1983;46:716-724

The effects of the "Vestibular Sedative" drug,Flunarizine upon the vestibular and oculomotorsystemsJONATHAN ELL, MICHAEL GRESTY

From the MRC Neuro-otology Unit, The National Hospital for Nervous Diseases, Queen Square, London,UK.

SUMMARY The effects of the "vestibular sedative" drug Flunarizine upon the oculomotor func-tions of pursuit and voluntary saccades and upon the vestibular response (to rotational stimuli)were assessed in twenty volunteer subjects. The study was then extended to three patients withchronic imbalance of central origin who had reported a beneficial symptomatic response to thedrug. Three of the volunteer subjects were found to have a directional preponderance (presumedto arise from peripheral dysfunction). In the remaining seventeen normal subjects Flunarizinewas found to reduce the amplitudes of fast phases of vestibular nystagmus. The directionalpreponderance in the other three subjects was redressed through production of fast phases whichwere of lower and more uniform amplitude. In the patients, in addition to a reduction in fastphase amplitude, there was a reduction or abolition of after nystagmus. In no case was any

reduction in slow phase velocity observed. Pursuit and voluntary saccades were unaffected by thedrug. It was concluded, on the basis that the fast phases of nystagmus are centrally generated, thatFlunarizine has a central action rather than a depressant effect upon the vestibular end organ. Inview of known oculomotor physiology and pharmacology it is proposed that vestibular sedativesact by depression of Type II vestibular neurons, and modification of the functional relationshipsbetween the vestibular nuclei, the perihypoglossal nuclei and the flocculus of the cerebellum. Atrial of vestibular active drug is indicated particularly in patients in whom asymmetry of thevestibular response and/or abnormal after nystagmus is demonstrated.

The mainstay of treatment of states of disequilib-rium of various causes, both peripheral and central,remains the so-called "vestibular sedative" drugs.No rationale exists for the application of these drugsin individual cases so, not surprisingly, response totreatment is often disappointing. Several criticismsmay be levelled at previous investigations into ves-tibular sedatives. Acute dose studies have attemptedto define the mode of action of such drugs upon thevestibular system, both in animals and man, by emp-loying objective measurements of the vestibularresponse'-3. Chronic dose studies have relied uponpsychophysical rating scale assessments of subjec-

Address for reprint requests: Dr JJ EII, The MRC Neuro-otologyUnit, The National Hospital for Nervous Diseases, Queen Square,London WC1N 3BG, UK.

Received 11 January 1983Accepted 19 March 1983

tive experiences.4 All reports have ignored the con-tribution of drowsiness (drug induced or otherwise)to modification of the vestibular response5-7. Thereis a common (and unwarranted) assumption thatvestibular sedatives act upon the vestibular endorgan.3 Little consideration has been given to poss-ible actions upon central vestibular connections andthis has biased the interpretation of data. In additionthere has been a total disregard of the possibilitythat vestibular sedatives may influence oculomotorfunction, despite the fact that vision and eye move-ments contribute equally with the vestibular systemto the maintenance of equilibrium.The purpose of this present study was to examine

the effects of a vestibular sedative on a range ofobjective tests of vestibular and oculomotor func-tions in a chronic, double blind crossover,placebo controlled trial in normal volunteers. Thedrug selected for assessment was Flunarizine,

716

Protected by copyright.

on July 30, 2020 by guest.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.46.8.716 on 1 A

ugust 1983. Dow

nloaded from

http://jnnp.bmj.com/

"Vestibular Sedatives"

[(E)- 1 -[bis(4-fluorophenyl)methyl]-4-(3-phenyl-2-propenyl)piperazine dihydrochloride], on thegrounds that our clinical experience suggested that itwas helpful in disorders of equilibrium of centralorigin. The study was extended to selected patientswho had reported a positive beneficial response totreatment with the drug. Flunarizine is a difluorin-ated derivative of Cinnarizine, [(E)-1-(di-phenylmethyl)-4-(3-phenyl-2-propenyl) piperazine],which is more potent on a weight for weight basisand has a longer half life. It is principally proteinbound in the serum [99%]8 and is detectable 20minutes after oral ingestion. Maximal concentra-tions are achieved between two and four hours afterdosing. Sleepiness has been reported four to sixhours after ingestion.9 Plasma levels have shownconsiderable individual differences on chronic dosestudies. Side effects have been reported as doserelated rather than related to plasma levels.10Rationale for the use of Flunarizine is at mostspeculative and its administration is ad hoc, as withall vestibular sedatives.

Subjects and methods

Twenty control subjects (10 male, 10 female), age range24 to 40 years with no previous history of vertigo or imbal-ance were assessed. Informed consent after a full explana-tion of the trial was obtained from each subject.At the comencement of the trial each subject underwent

a baseline assessment of vestibular and oculomotor func-tions as detailed below. The subjects then took placebo orFlunarizine 5 mg twice daily for one week and then had asecond assessment. Following a seven day washout periodthey then took placebo or Flunarizine, issued on a doubleblind cross-over design, for a further week at the end ofwhich they were-tested for a third time. The subjects wereinstructed to refrain from smoking and ingestion ofalcoholic beverages during the 24 hours prior to each testsession.Each test session consisted of the following standardised

tasks. (1) Assessment of saccade velocity and latencyevoked by target jumps which were randomised with arectangular distribution in both time and amplitude. Thetarget jumps ranged from 5 to 400. The minimum inter-jump interval was 1-5 seconds and the maximum was 3-5seconds. (2) Smooth pursuit was assessed using a deter-ministic, sinusoidal target motion the frequency of whichramped linearly up and down in frequency from startingrightwards up to 0-8 Hz down to stopping leftwards. Pur-suit was tested both with and without background illumina-tion. (3) The subjects were oscillated in yaw in a rotatingchair in darkness using a sinusoidal motion ramped linearlyup and down in frequency from starting rightwards up to1-0 Hz down to stopping leftwards. A constant peak veloc-ity of 800/s was maintained and the duration of the stimuluswas two minutes.Eye movements were measured using DC coupled

electro-oculography and stored on magnetic tape for offline computer analysis. An electrostatic ink jet recorder

print out was used for qualitative assessments and manualmeasurements.Serum Flunarizine levels were measured with high per-

formance liquid chromatography, using 254 nm UV detec-tion, as described by Woestenborghs et al. II

Computer analysis was made of saccade latency andpeak velocity in relationship to amplitude of target jumpsand of accuracy of pursuit using proportion of the eyemovement occupied by saccadic intrusions plotted againstpeak target velocity as an index of performance. Details ofthe method of analysis have been published previously.'2 13The target used for the pursuit and saccade tests was the

projected beam of a helium-neon laser reflected from aservo-controlled rotating mirror on to a tangent screen.The target subtended 0-002 radians at a distance from thesubject of 3 metres.

Following analysis of the results from normal controls.the study was extended to an examination of the effects ofFlunarizine in three patients who had previously reported apositive symptomatic response to that drug. As a consequ-ence of the findings in normal subjects the test procedurewas modified. The patients were examined using EOG forthe presence of gaze-evoked nystagmus in the light anddark. Full field optokinetic stimulation for two minutes tothe right and left was given. Following each direction ofstimulation they were observed in darkness for the pres-ence of optokinetic after nystagmus (OKAN). They werealso given rapid acceleration to and deceleration from 40°/sconstant angular velocity rotation in a rotating chair inaddition to the ramped sinusoidal stimuli described previ-ously.Each patient was given a baseline assessment and then

took Flunarizine 10 mg nocte for four weeks after whichthey were reassessed.

Results

FINDINGS IN NORMAL SUBJECTSThe relevant negative findings were that as assessedwith placebo Flunarizine had no effect on saccadevelocity or latency in response to random targetjumps; pursuit was neither impaired nor improved;there was no difference between pursuit with andwithout background both within and between testsessions.

In baseline testing, three of the normal subjectswere found to have an asymmetry in the pattern ofnystagmus they produced to rightwards and left-wards rotational stimuli. In all 20 subjects the veloc-ity of the slow phase of vestibular nystagmus wasunaffected by Flunarizine. In the 17 subjects withsymmetrical responses, however, a significant reduc-tion in the amplitudes of fast phases of nystagmuswas found. Results of statistical analysis of saccadeamplitudes sampled at each peak of chair velocityare given in the table. A highly significant differencebetween saccade amplitudes during both the controland placebo test sessions and the on-drug sessionwas demonstrated. Frequency of saccades during the

717



j.com/

J Neurol N


ugust 1983. Dow

nloaded from


7Eli, Gresty

Table Statistical difference between saccade amplitudesmeasured at peak velocity under drug, placebo and controlregimes.

Patients1 C> D P < 0-01 Wilcoxon matched-pairs signed-ranks test2 C > D P S 0-01 analysed over right/left directions and3 C> D P <0-01 5 evenly spaced epochs of 3-5 cycles each.

17 Normal subjectsC = P p > 0-05 Friedman two-way analysis of variance andC > D p < 0-05 Wilcoxon matched-pairs signed ranks test,

for each subject saccade amplitudesP > D p < 0-01 were summed over 5 epochs.

D = saccade amplitude under drug. P = saccade amplitude underplacebo. C = saccade amplitude under control condition.Two-tailed tests.

on-drug sessions could not by accurately measuredas the saccades were often of such low amplitudethat they were at the limit of resolution of the EOG.The nature of the asymmetries in the three sub-

jects mentioned above were complex but followed a

similar pattern both for placebo and baseline testingsessions (fig 1). Although the velocities of the slowphase components of nystagmus for each directionof rotation were not significantly different, the fre-quencies and amplitudes of fast phases of nystagmuswere unequal for the two directions of motion. Afurther consequence of the asymmetries was thatthere could be a directional bias of deviation of theeyes from the primary position.Under the influence of the drug the directional

preponderance of vestibular nystagmus in these sub-jects was wholly or partly redressed (fig 1). In addi-tion, the amplitude of saccades generated duringvestibular stimulation was reduced.

Plasma levels of Flunarizine showed great varia-tion, with values ranging from 4-8 ng/ml to 38-2nglml (mean = 15-7 ng/ml). Three subjects experi-enced unacceptable drowsiness during the first threeto four days of ingestion of the drug. The drowsinesswas no longer noted by the time of the on-drugassessment. Plasma levels in these subjects were 9-8,17.6 and 19-8 ng/ml. However, three other subjectsreported a significant degree of psychomotor retar-dation whilst taking placebo! Plasma levels in thethree subjects in whom directional proponderancewas redressed were 15-2, 19*8 and 20-1 ng/ml. Theremaining subjects were unaware as to whether theywere taking the drug or not. No additional sideeffects were encountered.

FINDINGS IN PATIENTS WITH CENTRAL

VESTIBULAR DYSFUNCTIONThree patients successfully completed the testingregime. Each had a different disorder and the nys-tagmographic findings in each were different and so

are treated separately.Patient number I A 63-year-old male began to

R

L Control eyesSi

80°/s //X /pk stimulus

Drug.

Control 52

Drug

ControlS3

Drug l1sFig 1 Vestibular responses to sinusoidal oscillatorystimulation in three asymptomatc control subjects (SI, S2,S3) found to have directionalpreponderance. Upper traces:E;OG (baseline test session); middle traces: stimulus; lowertraces: EOG (on-drug session). Attention is drawn torightwards and leftwards directional asymmetries in theamplitudes and frequencies of the fast phases of thenystagmus (that is rapid transients which interrupt theoverall sinusoidal shape ofthe eye movement). Forexample, in the uppermost trace there are two or three timesas many fastphases in the leftwards direction as there are inthe rightwards direction (SI control). After drug there aremore or less an equal number offast phases in eachdirection.

experience giddy attacks following a fall five yearspreviously. The attacks could be spontaneous orprecipitated by head movement. When walking hetended to veer to the right. Examination two yearsafter the accident revealed benign positional nys-tagmus with the right ear dependent. This resolvedsubsequently. On examination, he had an essentialtremor. There was a mild slurring dysarthria. Fol-lowing eye movements were broken up in all planes.No spontaneous nystagmus was seen on directinspection of the eyes.EOG findings on and off the drug EOG showedfirst degree nystagmus to the left in the dark only.This was absent whilst on drug. Pursuit was impairedbut better in the presence of background illumina-tion. Pursuit was not altered on drug. Optokineticresponses were symmetrical. OKAN to right 35 s,(on drug, 8 s), OKAN to the left 15 s, (on drug, 19s). Responses to rotational vestibular stimuli were as

718



j.com/

J Neurol N


ugust 1983. Dow

nloaded from



follows: starting right 43 s, (on drug, 50 s); startingleft 44 s, (on drug, 10 s), with spontaneous reversalof nystagmus lasting 31 s, (abolished on drug).Stopping from right 64 s, (on drug, 18 s); stoppingfrom left 35 s, (on drug, 38 s).

Responses to oscillatory vestibular stimulationboth on and off drug are illustrated in fig 2. Off drugthe responses were asymmetrical against the left asdetermined by frequency of saccadic intrusions. Ondrug the record was virtually devoid of large amp-litude fast phases of nystagmus, consisting almostentirely of sinusoidal modulation of eye position. Onclose examination however the overall modulation isdetermined by runs of small amplitude saccades atvery high frequencies. These, unfortunately, are atthe limit of resolution of the EOG. Asymmetry isstill present in the record and arises from a hyperac-tive response to the right with respect to the left.Patient number 2 A 45-year-old male, in 1980,became aware that on executing a left hand tum inhis car it felt as though he were about to overturn.Symptoms of mild, intermittent imbalance persistedup to the time of assessment. On examination heveered to the left when walking with his eyes closed;Romberg's test was positive.EOG findings on and off the drug EOG showedfirst degree nystagmus to the left with prolongationof the slow phase durations in darkness which wasabolished on eye closure. Pursuit was impaired.Neither the spontaneous nystagmus nor thederanged pursuit was affected by the administrationof Flunarizine and there was no symptomatic reliefon this occasion. Optokinetic stimuli producedsymmetrical responses. Before the drug, optokineticafter nystagmus (OKAN) to the right lasted for 23 sand to the left 22 s. On Flunarizine OKAN to theright lasted for 29 s, OKAN left for 12 s. Responsesto rotational vestibular stimulation were as follows:duration of post rotatory nystagmus starting right,40 s (on drug, 19 s); starting left, 24 s (on drug, 32s), with spontaneous reversal lasting 22 s which wasabolished on drug; stopping from right, 50 s (ondrug, 16 s) and stopping from left 30 s (on drug, 17s).Responses to oscillatory vestibular stimulation

both on and off drug are illustrated in fig 2. Therecords show that the response off treatment com-prised nystagmus which was abnormal in that thepattern of saccadic intrusion throughout the entiretrace was highly irregular. For cycles of stimuluswith similar periods, at times saccade amplitudeswere very large and at others quite small. Within anygiven cycle the nystagmus to right and left could bedissimilar, as could the responses to adjacent cyclesto stimulus. In contrast, the responses whilst on drugwere symmetrical in direction at all frequencies of

10LL\ / eyesL Control

80'/5s An '~~ Plpk X stimulus

Drug

Control

v ki\

N

\> ~~ As ~N" P2

Drug

Control

// "'-'~~~~~~~P3

Drug ls

Fig 2 Vestibular response to sinusoidal oscillatorystimulation in three patients (P1, P2, P3) who reported apositive symptomatic response to Flunarizine. Upper traces:EOG (baseline test session); middle traces: (stimulus);lower traces: EOG (on-drug session).

stimulation; the response to stimuli at similar fre-quencies were of uniform appearance which, oninspection, was due to relative uniformity of theamplitudes of the fast phase of nystagmus.Patient number 3 A 48-year-old woman suddenly,in 1978, developed nausea, vomiting, loss of taste,paraesthesia of the arms and a tendency to veer tothe right when walking. These all settled rapidly, butshe was left with residual sensations of mild imbal-ance particularly on moving the head rapidly orwhen decelerating in the car. On examination, apartfrom mild impairment of pursuit, there were noabnormalities.EOG findings on and off the drug Off drug therewas a first degree nystagmus to the right in the darkonly. Pursuit movements were impaired. Neither thenystagmus nor the pursuit was altered by the drug.Optokinetic responses were symmetrical. OKAN tothe right lasted 47 s (on drug, 69 s); to the left 36 s(on drug, 40 s). Responses to rotational vestibularstimuli; Starting right, 55 s (on drug, 55 s); withspontaneous reversal lasting 150 s, (on drug, 25 s).Starting left, 43 s (on drug, 36 s); with spontaneousreversal lasting 70 s (on drug, 10 s.). Stopping from

719



j.com/

J Neurol N


ugust 1983. Dow

nloaded from


Ell, Gresty

60

- 50in40

-4'a

U 3U:

a20

10

Patient 2

1 2

Patient 3

4 5Epoch

Fig 3 Histogram ofsaccade amplitudes, right vs left, ofpatients Nos 2 and 3 for baseline and on-drug testsessions. Saccades are averaged over five equally spaced epochs of3 5 cycles each. Epochs I and 5 areover low frequencies, epoch 3 is over maximum frequencies.

right 54 s (on drug, 42 s); with spontaneous reversallasting 75 s (abolished on drug); stopping from left80 s (on drug, 45 s); with spontaneous reversal last-ing 42 s (abolished on drug).Responses to oscillatory stimulation are illus-

trated in fig 2. The off drug trace is probably withinnormal limits. On drug, the amplitude of saccades isreduced. At certain points in the record slow phasegeneration is impaired with. respect to gain andphase. In addition there is distortion of the slowphase which should, ideally, be sinusoidal. Max-imum saccade amplitude is reduced and saccadeamplitude is more uniform.A histogram displaying the reduction in saccade

amplitude over five equally spaced epochs of threeand a half cycles each in patients Nos. 2 and 3 isshown in fig 3. It can be seen that the degree ofasymmetry of saccade amplitude, right versus left,has also been lessened.

Discussion

The results of this chronic dose study demonstratethat, in normal subjects, Flunarizine exerts nomeasurable effect on the pursuit system or upon thegeneration of voluntary saccades. In normal subjectsa reduction in the amplitudes of fast phases of ves-

tibular nystagmus was found. In asymptomatic sub-jects in whom there was evidence of asymmetry ofvestibular function (of probable peripheral origin)the drug redressed or partially redressed the asym-metry by an overall reduction of saccade amplitude,so rendering saccades more uniform in amplitudeand frequency. In patients with imbalance of centralorigin who had reported a beneficial effect,Flunarizine lessened asymmetries of vestibular func-tion by modification of the pattern and amplitude offast phases of vestibular nystagmus and reduction ofthe duration of after nystagmus. In no instance wasany effect upon the generation of the slow phases ofvestibular nystagmus noted. We shall argue thatthese results indicate that Flunarizine has a centralsite of action.Other studies have concluded that Flunarizine

and related drugs have a peripheral action on thebasis that effects on slow phase velocity were noted.A review of the background physiology and phar-macology of the vestibular system will show that thisis not necessarily the case.

In a caloric induced vestibular nystagmus in anormal subject the maximum velocity of the slowphase of nystagmus induced by an ampullopetalstimulus is an appropriate measure of (peripheral)vestibular function.'4 Duration of nystagmus is very

720

2 3 4 5



j.com/

J Neurol N


ugust 1983. Dow

nloaded from



much influenced by the phenomenon of adaptation.This is mediated mainly through the central nervoussystem,'5 though some experimental evidence fromresponses to prolonged rotation suggest a peripheralcontribution by direction specific receptors.'6 1' Theelementary, three neuron vestibulo-ocular reflexarc'8 19 traverses the median longitudinal fasciculus(MLF). This is the pathway responsible for genera-tion of the slow phases of vestibular nystagmus. Thepolysynaptic second pathway (implicated in the gen-eration of fast phases of nystagmus) passes to theoculomotor nuclei via brainstem accessory nucleiand nuclei of the reticular formation.20 Anderson etal.2' have described the effect of amylobarbitone onthe vestibular response: increasing doses first causea fall in the number of nystagmic beats and laterabolition of saccadic intrusions. Rashbass and Rus-sell5 found no decrease in slow phase velocity withamylobarbitone and this finding was confirmed byHaciska.22 She postulated that barbiturates exerteda depressant action specifically upon polysynapticvestibular pathways, sparing the MLF and therebyhaving no effect upon slow phase velocity. In view ofthe fact that internuclear ophthalmoplegia can occa-sionally be observed in the context of barbiturateintoxication, it is more probable that amylobar-bitone in low concentrations inhibits conduction inpolysynaptic pathways, sparing the MLF only untilhigher drug levels are attained.

Studies of the effects of administration ofd-amphetamine upon vestibular nystagmus22 23 alsoshowed no effect on slow phase velocity. Haciskafound an improvement in the regularity of both fre-quency and amplitude of nystagmus during eye clos-ure. Importantly, Collins noted that level of arousalwas far more important a factor in determining vari-ation in slow phase velocity, though the alertingaction of amphetamine did smooth out individualvariations in the amount of mental activity requiredto avert onset of a state of reverie. He stressed thatthis fact had to be borne in mind in the evaluation ofdrug effects. Barbiturates have a direct depressantaction on the reticular activating system ;24amphetamines have an excitant action.25 By infer-ence, one can assert that the effects of the vestibularresponses described above reflect these actions.The studies in man by Philipszoon' on Cinnarizine

and those by Oosterveld on Flunarizine3 have beenacute dose studies. In the latter study, a significantreduction in both slow phase velocity and durationof nystagmus was found. This effect may well havebeen due to temporary decrease in level of arousal.The chronic dose study of Flunarizine by Ooster-veld4 suggests that the drug is symptomatically be-neficial, but objective assessments of vestibularfunction were not undertaken. Both these workers

imply that these drugs act by depression of eitherperipheral labyrinthine tonic discharge rate or sen-sitivity to cupular deflection.

If "vestibular sedatives" redress tonus imbalanceby any inhibitory effect, either upon tonic or phasicdischarges, then the only means by which they couldbe effective would be by reducing resting or phasicfiring rates bilaterally to near zero. This is not thecase. Two findings were common to all our subjects.In no case was slow phase velocity altered. Fast (sac-cadic) components were modified in amplitude anddegree of variability of amplitude and, in somecases, the degree of asymmetry of insertion of sac-cades was diminished. Because saccade amplitudewas often reduced to near (or very likely) below thelimit of resolution of the EOG we could not obtainaccurate saccade counts. However, naked eyeinspection of the records indicated that, overall, thefrequency of insertion of saccades was increased.Such an increase in saccade frequency along with thedecrease in saccade amplitude would be in keepingwith the finding of no alteration in slow phase veloc-ity. The three patients described in this presentstudy all had definitive evidence of a balance disor-der of central origin. In the three volunteer subjectswho fortuitously showed objective evidence of ves-tibular imbalance there were no additional abnor-malities that would indicate a central nervous dis-turbance and we assume that their asymmetricalresponses were of peripheral origin. There was adefinite effect of Flunarizine in all six cases, thoughnot exactly the same modifications were found ineach subject. We believe that the absence of anyeffect on slow phase velocity in the present trial isattributable to the fact that our subjects haddeveloped tolerance of the sedative effects of thedrug and that the previously reported effects uponslow phase velocity reflect drowsiness followingacute dosage. Our findings indicate that the drugacts centrally. The reasons underlying this conclu-sion shall now be outlined and discussed.A striking effect in the patients described above

was the abolition, or near abolition, of spontaneousreversal of nystagmus both from accelerative anddecelerative stimuli. In view of our understanding ofthe mechanisms underlying the production of afternystagmus, this observation lends strong support tothe hypothesis of a central site of action. Cells of thevestibular nuclei are termed Type I ("on" response)or Type II ("off' response),26. Secondary nystagmusarises from after discharge in antagonistic neurons.Since these antagonistic neurons are mainly of TypeII it follows that secondary nystagmus is in the oppo-site direction to the initial nystagmus. In general, thegreater the adaptation during stimulation in the"on" direction, the greater the rebound following an

721



j.com/

J Neurol N


ugust 1983. Dow

nloaded from


Ell, Gresty

"off' stimulation.27 A velocity-storage integrator/ssystem has been identified and characterised28 whichis involved in the generation of after nystagmus fol-lowing both vestibular and optokinetic stimulation.The anatomical location of this system had not beenabsolutely established, but it is probably situated inthe nucleus prepositus hypoglossi and is under theinfluence of the cerebellum.A second effect, seen both in subjects and

patients, was a general reduction in the amplitudesof the fast phases of vestibular nystagmus. Groups ofneurons related to the fast phases of vestibular nys-tagmus have been identified and extensively investi-gated in the cat. Premotor neurons participating inthe activation or suppression of abducens motoractivity during angular acceleration in yaw havebeen termed excitatory burst neurons (EBNs)29, andinhibitory burst neurons (IBNs),30. EBNs arelocated in the dorsomedial reticular formationimmediately rostral to and projecting to the ipsilat-eral abducens nucleus. IBNs are located just caudalto the abducens nucleus and project to the con-tralateral abducens nucleus. The so-called pausecells of the pontine paramedian reticular formationexert a direct effect upon the IBNs.31 Discharge ofthe EBNs causes inhibition of Type I vestibularneurons and excitation of both Type II vestibularneurons and IBNs. In addition, it is known thatstimulation of reticulospinal neurons located ros-troventral to the abducens nucleus can augment fre-quency and amplitude of nystagmus related dis-charges, probably through facilitation of nearbycentres rather that direct input.32 A schematic rep-resentation of these centres is shown in fig 4.To take a simplistic view of the system, the net

result of this neuronal circuitry is to effect efficientinactivation of the generation of the slow phases ofvestibular nystagmus (mediated by Type I neurons)and to reset eye position by generation of the fastphases of nystagmus. It can readily be appreciatedthat a synchronous generation of burst activity inEBNs, IBNs and Type II cells is necessary for wellcoordinated saccadic activity to occur. The thresholdof EBN discharge rate necessary for initiation of afast phase must be determined by an interreaction ofthese three cell types, with, of course, a contributionfrom equivalent contralateral structures. Whatdetermines saccade amplitude is as yet unknown.Amplitude may either be precoded at the time ofinitiation of the fast phase or be simply a function ofthe duration of EBN discharge (and Type I cellinhibition) versus pause cell activity. In the case ofnystagmus generated with the subject in darknessthe latter option seems more likely as no retinalerror information is available.

If one accepts that adaptation occurs primarily

to VI

Fig 4 Schematic representation ofbrainstem centresinvolved in the generation offast phases of vestibularnystagmus. TI = Type I vestibular neuron (excitatory), T II= Type II vestibular neuron (inhibitory). EBN =

Excitatory reticular burst neuron, IBN = Inhibitoryreticular burst neuron. VI = Abducens nucleus, VIII =Vestibular nerve. PPRF = Pontine paramedian reticularformation.

through central mechanisms, then the abolition ofspontaneous reversal of nystagmus induced by start-ing and stopping stimuli (trapezoidal wave form)seen in the three patients must indicate a central siteof action. It is not clear which structures or pathwaysmay be affected by such drugs but, in view of ourknowledge of the inhibitory influence of the cerebel-lum on the vestibulo-ocular reflex," we suspect thatthe connections between vestibular nuclei and thecerebellar flocculus are involved. In addition, we

suggest that the function characteristics of thevelocity-storage integrators situated in the nucleusprepositus hypoglossi (perihypoglossal nucleus inman) are also altered by the drug.The effect of administration of Flunarizine upon

saccade amplitude suggests an action on the reticu-lar burst neurons or their connections. As the drughas sedative effects in higher doses, an excitatoryaction seems unlikely. We posfulate that it actsneurochemically through suppression of Type II(inhibitory) vestibular neurons. This results in thelowering of the threshold for the insertion of sac-cades during vestibular nystagmus. Smaller amp-litude, more frequent saccades are produced and

722



j.com/

J Neurol N


ugust 1983. Dow

nloaded from



asymmetry is diminished through a relative decreasein the degree of direction-related amplitude asym-metry.

In conclusion, considering the effects ofFlunarizine observed, we propose that the term"vestibular sedative" is a misnomer. "Vestibularisostatic" would be more apt a description. In gen-eral neurological practice the "dizzy" patient pres-ents a spectrum of symptomatology ranging fromintense rotational vertigo through to mild unsteadi-ness. Sensations of imbalance may arise from disor-der of the visual, proprioceptive and vestibular sys-tems, the cerebellum or any combination of these.Unless asymmetry of vestibular function had beendemonstrated by objective tests (such as caloric androtational tests and ENG), and the abnormalitiesdetected are compatible with the patient' s symptomsit is unreasonable to expect a positive response totreatment with a vestibular active drug. A trial ofFlunarizine (or any similar drug), is worthwhile inpatients with objective evidence of asymmetry of thevestibular response (whether of peripheral or cen-tral origin), or with abnormal spontaneous reversalof rotational induced nystagmus.

We acknowledge the following: Janssen Phar-maceutical Company, for support and assistancewith the study. The Department of Medical Elec-tronics, St. Bartholomew's Hospital, London, fordevelopment of computer programmes and comput-ing facilities. Dr Ell was supported by an AlexanderPiggott Wernher Training Fellowship.

References

' Philipszoon AJ. Influence of cinnarizine on the labryinthand on vertigo. Clin Pharmocol Ther 1962;3:184-90.

2 Jongkees LBW. The influence of some drugs upon thefunction of the labryinth. Acta Otolaryngol.1961;53:281-5.

3 Oosterveld WJ. Vestibular pharmacology of flunarizinecompared to that of cinnarizine. ORL 1974;36:157-64.

4 Oosterveld WJ. Flunarizine in vertigo. ORL1982;44:72-80.

5Rashbass C, Russell GFM. Action of a barbiturate drug(amylobarbitone sodium) on the vestibulo-ocularreflex. Brain 1961;84:329-35.

6 Collins WE. Effects of mental set upon vestibular nys-tagmus. J Exp Psychol 1962;63:191-7.

' Blegvad B. Caloric vestibular reaction in unconsciouspatients. Arch Otolaryngol 1962;75:506-14.

Meuldermans W, Hurkmans R, Heykants J. A compara-

tive study of plasma protein binding and the distribu-tion of flunarizine and cinnarizine in human blood.Jenssen Research Products Information Service.1978;serial number:R 14 950/30.

Heykants J, Wynants J, Hendrickx J, Scheijgrond HW.

Absorption of flunarizine in volunteers after a singleoral dose of 50 mg. Janssen Research Products Infor-mation Service. 1976;serial number:R 14 950/18.

10 Heykants J, De Cree J, Hong Ch. Steady-state plasmnalevels of flunarizine in chronically treated patients.Arzneim.-Forsch. 1979;29(ii),: 1168-71.

Woestenborghs R. Hendrickx J, Lenoir H, Embrechts L,Heykants J. An improved HPLC-method for thedetermination of flunarizine (R 14 950) in plasma andanimal tissues. Janssen Research Products InformationService. 1980;serial number:R 14 950/14.

12 Smith AT, Bittencourt PRM, Lloyd DSL, Richens A.An efficient technique for determining the characteris-tics of saccadic eye movements using a minicomputer.J Biomed Eng 1981;3:39-43.

13 Bittencourt PRM, Smith AT, Lloyd DSL, Richens A.Determination of smooth pursuit eye movement vel-ocity in humans by computer. EEG Clin Neurophysiol1982;54:399-405.

4 Hinchcliffe R. Validity of measures of caloric testresponse Acta Otolaryngol 1967;63:69-73.

15 Hood JD. Persistence of response in the caloric test.Aerospace Med 1973;44:444-9.

16 Hailpike CS, Hood JD. Fatigue and adaptation of thecupular mechanism of the human horizontal semicir-cular canal: an experimental investigation. Proc R SocSer B 1953;141:542-61.

1 Young LR, Oman CM. Model of vestibular adaptationto horizontal rotations. Aerospace Med1969;40: 1076-80.

18 Lorente de No R. Vestibulo-ocular reflex arc. ArchNeurol Psychiatry 1933;30:245-91.

Szentagothai J. The elementary vestibulo-ocular reflexarc. J Neurophysiol 1950;13:395-407.

20 Lorente de No R. Analysis of the activity of the chains ofinternuncial neurons. J Neurophysiol 1938;1:207-44.

21 Anderson PJ, Diamond SP, Bergman PS, Nathamson M.Electro-oculographic investigation of the caloricresponse. Neurology (Minneap) 1958;8:741-9.

22 Haciska DT. .The influence of drugs on caloric-inducednystagmus. Acta Otolaryngol 1973;75:477-84.

23 Collins WE, Schroeder DJ, Elam GW. Effects ofd-amphetamine and of secobarbital on optokineticand rotation-induced nystagmus. Aviat Space EnvironMed 1975;46:357-64.

24 Moruzzi G, Magoun HW. Brainstem reticular formationand activation of the EEG. EEG Clin Neurophysiol1949;1:455-73.

25 Bradley PB, Elkes J. The effects of some drugs on theelectrical activity of the brain. Brain 1957;80:77-117.

26 Precht W. The physiology of the vestibular nuclei. In:Kornhuber HH, ed. Vestibular System Part I: BasicMechanisms. Berlin: Springer-Verlag, 1974.

27 Crampton GH. Certain cell responses within the vestibu-lar nuclei of cat have characteristics which may berelated to secondary nystagmus. In: Hood JD, ed. Ves-tibular mechanisms in Health and Disease. London:Academic Press, 1978;59-65.

28 Raphan T, Cohen B, Matsuo V. A velocity-storagemechanism responsible for optokinetic nystagmus(OKN), optokinetic after-nystagmus (OKAN) andvestibular nystagmus. In: Baker R and Cohen B, eds.

723



j.com/

J Neurol N


ugust 1983. Dow

nloaded from


724

Control of Gaze by Brain Stem Neurons. Amsterdam:Elsevier/North Holland Biomedical Press, 1977;37-47.

29 Igusa Y, Sasaki S, Shimazu H. Excitatory premotor burstneurons in the cat pontine reticular formation relatedto the quick phase of vestibular nystagmus. Brain Res1980;182:451-6.

30 Hikosaka 0, Kawakami T. Inhibitory reticular neurons

related to the quick phase of vestibular nystagmus-their location and projection. Exp Brain Res1977;27:377-96.

3' Markham CH, Nakao S, Curthoys IS. Direct projection

Ell, Gresty

of pauser neurons to nystagmus-related pontomedul-lary reticular burst neurons. Neurosci Abstr1979;5:496.

32 Sasaki S, Shimazu H. Reticulovestibular organizationparticipating in generation of horizontal fast eyemovement. In: Cohen B, ed. Vestibular andOculomotor Physiology: International Meeting of theBarany Society. Ann NYAcad Sci 1981;374:130-143.

33 Ito M. Cerebellar control of the vestibular neurons:Physiology and Pharmacology. In: Brodal A, Pom-peinao 0, eds. Progress in Brain Research. Amster-dam: Elsevier, 1972;37:387-390.



j.com/

J Neurol N


ugust 1983. Dow

nloaded from


the vestibular sedative flunarizine upon systems · flunarizine 5 mgtwice daily for oneweekandthen...

Documents