REDUCTION OF GAIN AND TIME CONSTANT OF VESTIBULO-OCULAR REFLEX IN MAN INDUCED BY DIAZEPAM AND THIOPENTAL

Transcription

1 Journal of Vestibular Researh, Vol. 1, pp , 1990 Printed in the US. ll rights reserved /90 $ Copyright 1990 Pergamon Press pi REDUCTION OF GIN ND TIME CONSTNT OF VESTIBULO-OCULR REFLEX IN MN INDUCED BY DIZEPM ND THIOPENTL Serge Padoan, * Kari Korttila,t Mans Magnusson, * IImari Pyykko,:j: and Luyna Shalen* *Department of Otorhinolaryngology, University Hospital of Lund, Lund, Sweden, Departments of tnesthesiology and totolaryngology, University Hospital of Helsinki, Helsinki, Finland Reprint address: Dr Serge Padoan, Department of Otorhinolaryngology, University Hospital of Lund, S Lund, Sweden o bstrat - The effet of intravenous administration of two sedatives, diazepam (0.3 mg/kg) and thiopental (6.0 mg/kg),. on the vestibulo-oular reflex (VOR) in man was investigated on 9 volunteers. The VOR was evoked with a veloity step rotation test and gain and time onstant of perrotatory and postrotatory nystagmus were measured. Both drugs redued VOR gain. Diazepam-indued redution lasted 8 h and thiopental-indued redution 1 h. redution of the VOR time onstant was found lasting about 1 h for both drugs, but with a tendeny for the thiopental effet to last longer. These findings, not previously desribed in man, differ from what has been found in maaques and rabbits injeted with diazepam. The redution of gain and time onstant were not orrelated with the blood onentration of either drug. The present results suggest that in man the VOR gain and time onstant are both redued by different types of sedatives although with different time ourses. On the basis of previously shown effet of alertness on the VOR, it is hypothesized that diazepam and thiopental, besides having a speifi effet on entral nervous system strutures important to the VOR, also indue redution of the VOR through a general sedation of the ens. o Keywords - VOR; diazepam; thiopenta!; time onstant. Introdution Benzodiazepines and barbiturates are known to affet voluntary and reflexive eye movements in humans (1-4), and benzodiazepines are known to affet reflexive eye movements in animals (5-7). The neurophysiologial orrelate for this effet is presumed to be a depressant ation on neurons important to eye movements (1,2,5-8). Nevertheless an unspeifi effet of the drugs is also plausible, the possible mehanism being general depression of the entral nervous system (CNS), ausing redution of alertness. n effet of alertness on vestibulo-oular reflex (VOR) gain has been desribed previously (9). Reently an effet of arousal on the veloity storage mehanism has been desribed, ausing a prolongation of the time onstant of the VOR in rabbit (10) and inreasing optokineti afternystagmus (OKN) in man (11). Thus redution of alertness due to CNS depression aomplished by benzodiazepines or barbiturates ould be expeted to redue the time onstant of the VOR. To our knowledge no studies on the effet of benzodiazepines or barbiturates on the human VOR refer to the time onstant. s animal studies (5,7) indiate an inreasing effet of diazepam on the time onstant and a reduing effet on gain with different time ourses, it seemed important to test the validity of these findings in humans. Furthermore most studies on man have been performed with oral administration of drugs, while in animal studies drugs have been given intravenously. Besides speies differenes, time RECEIVED 20 February 1990; REVISED 12 pril 1990; CCEPTED 13 pril

2 98 ourses may thus differ enough to make it diffiult to ompare results. In this experiment, doses were hosen orresponding approximately to those used in linial pratie for intravenous indution of anesthesia. We also onsidered it linially important to eluidate the effet of these drugs on reflexive eye movements and alertness, given the wide use of intravenous (Lv.) administration of benzodiazepines for short-term anesthesia of outpatients, who only a few hours after an operation may have to deal with ompliated traffi situations. The aim of the present study was to investigate the effet of i.v. diazepam and thiopentalon VOR gain and time onstant in man, and to probe the orrelation of the effets with blood onentrations of the drugs. Materials Nine paid volunteers, 5 females and 4 males, age 22 to 38 (mean age 30.3), took part in the study. They were all healthy and had no history of vertigo or of any neurologial or metaboli disorder. ll had normal hearing, vision, and weight and were nonsmokers. Methods Eah subjet was tested 3 times at 2-week intervals, whih enabled these drugs to be virtually eliminated between trials. On eah test oasion, the subjets were assigned to diazepam, thiopental, or no drug aording to a latin square design. No plaebo was used, as the effets of the drugs were so obvious to the subjets. Both drugs were given intravenously in quantities equivalent to 0.3 mglkg body weight for diazepam, and 6.0 mglkg body weight for thiopental, orresponding to doses used for in-dution of anesthesia with thiopental and half that amount with diazepam (12). The drugs were given at 8 M and the tests were performed at 9 M, at noon, and at 4 PM. Food intake was standardized for the partiipants and onsisted of two heese sand- S. Padoan et al. wihes and a up of tea after onluding the tests at noon. loholi beverages or any kind of drugs were forbidden during the 48 h prior to eah trial. Before eah reording, blood samples were taken with a disposable annula from a ubital vein in the arm ontralateral to the arm in whih the drugs were given. The samples were frozen and subsequently analyzed by gas hromatography for onentration of the respetive drug. Beause of inomplete blood olletion, thiopental blood onentrations were analyzed in only 7 subjets at 9 M and in only 8 subjets at noon and 4 PM. The subjets were seated in a rotation hair, provided with an oipital support fixating the head in the enter of the vertial rotation axis and with the ear-eye axis horizontal. Clokwise and ounterlokwise (CW and CCW) rotation-indued vestibular nystagmus was ahieved by 60 0 /s 2 aeleration and deeleration in darkness for 2 s eah giving a veloity step of Is. Rotation lasted 90 s. Eye movements were reorded with DC eletrooulography (13). nalysis Starting from the first beat after reahing the plateau of eah veloity step, the peak veloity of the slow phase of the nystagmus beat ourring every 5 s was alulated manually from the reordings. The mean of the slow phase veloity of the orresponding nystagmus beats from the 4 stimuli, ie, the veloity steps of aelerationl deeleration of CW and CCW rotation, was alulated. To these mean nystagmus veloities an exponential deay was fitted and the time onstant was alulated by a omputer (DEC Professional 380 with RS/I software, Bolt Berank & Newman, Cambridge, M). Gain was alulated as the initial peak slow phase veloity divided by the veloity step. Individual values of gain and time onstant for eah drug test oasion were nofmalized, ie, divided by their respetive ontrol value, to yield a relative value. This was done to avoid bias from possible variations in alert-

3 100 S. Padoan et al o O.B O.4j' 0.2 o 9 11m o 12 11m 4 pm -- ontrol t-hmelln +/- sam O.O+---I --I-4I--rI--1--II---4I---+I- o DIZEPM CONCENTRTION (ng/ml) pm 1 ep gam THIOPENfL CONCENfRTION (flg/m1) B Figure 1. The relationship between drug onentration (thiopental or diazepam) and VOR parameter (relative gain or relative time onstant). Individual values are indiated. Results\for 9 M, noon, and 4 PM are denoted by squares, irles, and triangles, respetively. The f (X) = 1.0 line represents the arithmeti ontrol value. Means and standard errors of the mean from the three test oasions are indiated by the orresponding filled symbols onneted by a line. (Figure ontinued on faing page.) provement in orrelation (Pearson's oeffiient for a log-log fit at the 3 trials, ie, 9 M, noon, and 4 PM, was 0.097,0.039, and between thiopental onentration and the time onstant; 0.580, 0.488, and between thiopental onentration and gain; 0.020, 40, and between diazepam onentration and the time onstant; 0.116, 0.482, and between diazepam onentration and gain).

4 Redution of Gain and Time Constant of VOR C gam o 9 8m o b. 4 pm -- ontrol liie8n +/- S8m P o 200 4, BOO BOO 2000 DIZEPM CONCENTRTION (ng/ml) l.b C) 4pm O.B C gam [J nnopentl CONCENTRTION (Jlg/ml) o Figure 1 ontinued. C Disussion In the present study two sedatives, thiopental and diazepam redued human VOR gain and time onstant. pparently diazepam had a longer effet on the gain than on the time onstant, while thiopental tended to show an opposite effet profile. In regression analysis, however, individual values of either VOR parameter were not orrelated with blood onentrations of either drug. redution of VOR gain by diazepam and thiopental is in aordane with most previous studies in man and in animals. The im.,. pairment of the VOR aused by both drugs is presumed to be a result of their effet on infratentorial neuronal populations of the entral nervous system, in the ase of diazepam,

5 102 S. Padoan et al. 1.2 R 1 E L T 0.8 I V E 0 6. V 0.4 L U E S 0.2 gain timeonstant Thiopental Diazepam DRUGS Figure 2. The differene between the effet of two drugs on the two VOR parameters at the two later measurements, thiopental affeting the--time onstant more than diazepam and diazepam affeting gain more than thiopental. Error bars indiate standard error of the mean. in the vestibular nulei (1-8,15,16). The VOR time onstant in rabbits is prolonged by inreased alertness (10) as is OKN in man (11). Diazepam and thiopental may thus be expeted to diminish gain and, by sedation, to shorten the VOR time onstant. Differenes between Humans and nimals There are few previous reports on the effet of sedating agents on the VOR time onstant. In the maaque, diazepam redues gain but inreases the VOR time onstant (5). In the rabbit the same drug does not seem to affet the time onstant although it redues gain (6,7). The differene between these findings and our results in man, with a lear redution in VOR gain and time onstant, may be due to the differenes between the speies studied, the amount of drug given, and to the intervals hosen for measuring the VOR. Very little sedation was reported in the maaque, though 0.7 mg/kg diazepam was used (5). This an be ompared to the present 0.3 mg/kg, whih learly auses sedation in humans. Furthermore, gaze nystagmus was notied immediately after injetion of diazepam on maaques. We observed no pronouned gaze nystagmus after diazepam and a transient gaze nystagmus after thiopental, whih is onsistent with previous reports (17). In the rabbit experiment (7), mg/kg diazepam was used and one group of rabbits showed exitatory response in VOR. Furthermore, the first test of the VOR funtion in the present human experiment ould not be done until 1 h after injetion beause the subjets were simply unable to ooperate earlier. In the above mentioned experiment on maaques, VOR tests were om-

6 Redution of Gain and Time Constant of VOR mened only 5 min after injetion and the effet on the VOR time onstant lasted less than 1 h (signifiantly only 16 min), while the effet on gain lasted several hours. In the rabbit experiment, tests also started shortly after injetion and gain was signifiantly redued for 1 h. Moreover, one should bear in mind that the pharmaologial profile of benzodiazepines varies between speies to the extent that in some ases alertness instead of ens depression is ahieved (18). In a study of benzodiazepine binding in human brains (19), a good orrelation has been shown between in vitro inhibition of (3H)flunitrazepam binding to human erebral ortex homogenate and in vivo tests of pharmaologial poteny of several benzodiazepines in humans, while orrelation with in vivo poteny tests on monkeys was low. ommon finding in our study and in the animal experiments mentioned above is that benzodiazepines affet VOR gain and time onstant with different time ourses. The interpretation of Blair and Gavin (5); that this ould be due to diazepam having multiple sites of ation, is plausible. In our experiment thiopental also tended to have different time ourses in its effet on the 2 parameters of the VOR, but in a fashion opposite to that of diazepam. Possible Contributing Effet of Sedation To some extent benzodiazepines and barbiturates share sites and mehanisms of ation in the ens (18). Benzodiazepine binding is low in the human brainstem and intermediate in the erebellum (19), whih has extensive projetions to and from the vestibular nulei (20). Nevertheless, the retiular formation and the vestibular nulei are both loated in the brainstem and seem to be major sites of the ens depressant ation of benzodiazepines (21,22). However, the VOR has also been shown to be influened by mental ativity (9,23). Thus diazepam or thiopental ould also affet the VOR either by reduing alertness and/or by ating on higher enters of the 103 ens disrupting the ability to hoose a frame of referene, the latter known to be of importane when willingly modulating the VOR (23). Lak of Correlation between Drug Conentration and Effet There seems to be some ontroversy regarding the orrelation between serum levels of benzodiazepines and peak veloity of voluntary saades (24,25). The apparent lak of orrelation between the effet of the diazepam and thiopental on the VOR and their onentration in blood, as observed in the present study, might be explained by individ-. ual variations in pharmaokinetis and pharmaodynamis of diazepam and thiopental (26-28). Furthermore, large interindividual variations in the VOR, as found in the present experiment, make it neessary to investigate larger groups of subjets to demonstrate any possible signifiant orrelation. Conlusions The effets of two sedatives, diazepam and thiopental, on the human vestibulo-oular reflex were investigated in the present study. VOR evoked by a veloity step was redued by both drugs. But while diazepam-indued redution of gain lasted up to 8 h, thiopental was only effetive for 1 h. The time onstant was only redued for 1 h by both drugs, although thiopental-indued redution tended to last longer. Different sedating agents are thus demonstrated to have different profiles in their redution of the human VOR. Both speifi and unspeifi ation in the brainstem due to sedation may be responsible for the impairing effet of diazepam and thiopental on the human VOR. knowledgments-the present work was supported by the Swedish Medial Researh Counil, grant no: 17X-05693, and the S6derbergs foundations.

7 104 S. Padoan et al. REFERENCES 1. Rothenberg SJ, Selkoe D. Speifi oulomotor defiit after diazepam. Psyhopharmaol ;74: Tedeshi G, Bittenourt PRM, Smith T, Rihens. Speifi oulomotor defiits after amylobarbitone. Psyhopharmaol ;79: Mai M, Dayal VS, Tomlinson RD, Farkashidy J. Study of pursuit and vestibulo-oular anellation. Otolaryngo! Head Nek Surg. 1986;95(5): Collins WE, Shroeder DJ, Elam GW. Effets of D amphetamine and of seobarbital on optokineti and rotation-indued nystagmus. viat Spae Environ Med. 1975;46(4): Blair SM, Gavin M. Modifiation of vestibulo-ou- 1979; 105 : o. Barmak NH, Pettorossi VE. Vestibulo-oular reflexes in rabbits: redution by intravenous injetion of diazepam. rh Neurol. 1980;37: Matsunaga T, Shiraishi T, Kubo T. Differential effets of diazepam upon vestibulo- and visual-oulomotor responses in the rabbit. ta Otolaryngol 1983 ;393(Suppl.): shoff JC. Veranderungen rasher Blikbewegungen (Saaden) beim Menshen unter Diazepam (Valium (R». rh Psyhiat Nervenkr. 1968;211: Collins WE. rousal and vestibular habituation. In: Kornhuber HH, ed. Handbook of sensory physiology, vol 6, the vestibular system, part 2. New York: Springer Verlag; 1974: Magnusson M. Effet of alertness on the vestibulooular reflex and on the slow rise in optokineti nystagmus in rabbits. m J Otolaryngo!. 1986;7: Magnusson M, Pyykk6 I, Jiintti V. Effet of alertness and visual attention on optokineti nystagmus in humans. m J Otolaryngol. 1985;6: Marshall BE, Wollman H. General anesthetis. In: Goodman LS, Gilman Z, eds. Pharmaologial basis of therapeutis, 7th ed. New York: Mamillan; 1985; Henriksson NG, Pfaltz CR, Torok N, Rubin W. synopsis of the vestibular system. (Sandoz Monographs). Basel, Switzerland: Sandoz, Ltd; Cox DR, Hinkley DV. In: Theoretial statistis. London: Chapman & Hall, Ltd; Gabersek V. tion du diazepam sur l'hyperrefletivite vestibulaire. ta Oto-Rhino-Laryngol Belgia, 1975 ;29(1) : Ryu JH, MCabe BF. Effets of diazepam and dimenhydrinate on the resting ativity of the vestibular neuron. erospae Med. 1974; Bender MB, O'Brien FH. The influene of barbiturate on various forms of nystagmus. mer J Ophtha!. 1946;29: Harvey SC. Hypnotis and sedatives. In: Goodman LS, Gilman Z, eds. Pharmaologial basis of therapeutis, 7th ed. New York: Mamillan; 1985; Speth RC, Wastek GJ, Johnson PC, Yarnamura HI. Benzodiazepine binding in human brain: haraterization using (3H) flunitrazepam. Life Si. 1978;22: Baloh RW, Honrubia V. Clinial neurophysiology of the vestibular system. Philadelphia: F Davis; Przybyla C, Wang SC. Lous of entral depressant ation of diazepam. J Pharmaol Exp Ther. 1968; 163(2): Kawasaki K, Matsushita, Satoh M, Takagi H. The lateral vestibular nuleus is a site for the depressant ation of benzodiazepines on the rossed extensor reflex. Brain Res. 1988;461 : Barr CC, Shultheis LW, Robinson, D. Voluntary, non-visual ontrol of the human vestibulo-oular reflex. ta Otolaryngol. 1976;81: Bittenourt PRM, Wade P, Smith T, Rihens. The relationship between peak veloity of saadi eye movements and serum benzodiazepine onentration. Br J Clin Pharmaol ;12: Salonen M, altonen L, antaa E, Kanto J. Saadi eye movements in determination of the residual effets of the benzodiazepines. Int J Clin Pharmaol Ther Toxio!. 1986;24(5): Taylor JL, Tinklenberg JR. Cognitive impairment and benzodiazepines. In: Meltzer HY, ed. Psyhopharmaology: the third generation of progress. New York: Raven Press; Kaplan S. Pharmaokinetis of the benzodiazepines. In: Priest RG, Vi anna Filho U, mrein R, Skreta M, eds. Benzodiazepines today and tomorrow. MTP Press Limited, International Medial Publishers; Lanaster, England: Falon House; Rihter J, Holtman JR. Barbiturates: their in vivo effets and potential biohemial mehanisms. Prog Neurobiol. 1982;18: