Acta Otolaryngol (Stockh) I993; II3: 339-344

The Site of Impulse Generation in Transcranial Magnetic Stimulationof the Facial Nerve

I. RIMPILAINEN,I I. PYYKKO,' G. BLOMSTEDT,3 T. KUURNEA ANd P. KARMA,

From the tRognor Gtunit l&titute, Tampere (Jnioersity of Technology, Tampere, Finlud, ud Deparlmen! of ClinicalNewophysiology, Sahlgrens Unioersity Hospital, Gorhenburg, Sweden,.'Depanmeilt oI Otofiinolerngolog),Helsinki tlnioersity Centra! Hospirol, Helsinki,3Deportment of Neurosurgery, Hekinki t)niversity Centrdl Hospital,

Helsinki and'Department of Neurovrgery, Tompere Unioersity Hospital, Tampere, Finland

Rimpiliinen I, Pyykkd Ii Btomstedt G, I(uurne T, Karma P. The site of impuke generation in transctonial mdgnetic

simulai;n- Acta Otolaryngol (stockh) 1993; 113:339-344.

The facial nerve en be stimulatcd itr its intracranial course through transcmnial magnetic stimulation (TMS), we studied

the site of impulse generation produed by TMS by comparing thc lateocics of the muwle evoked potentials (MEPS)

elicited with TMS and intracranial eletrical stimulation (tES) ofthe facial nerve during ncurosurgical posterior fossa

procedures. In a series of 25 patients, the mean latency of the TMS eliciled MEPS, recorded in the orbicularis oris murle,was 5.0 ms (SD 0.58). Also IES of the distal part of the facial trerve in the intcrnal acoustis mcatus showed a mean latency

of 5.0 ms (SD 0.68). Proximal IES itr the root entry zone of the facial nerve, and intemediate IES between root entry zone

and marus, produed MEPs with signifiently longer latencies comparcd to TMS and distal IES (p < 0.05). The findiogs

suggest that the TMS indued facial neNe activation, lcading to a MEP rcsponsc, takcs pla@ within the internal acoustic

mqtus. r(e, words: location of facial nerae actiuation.

INTRODUCTION

The intracranial part of the facial nerve can be stimu-lated through transcranial magnetic stimulation(TMS) (l). The method has been used for facialnerve (FN) flunction tests in Bell's palsy (2). How-ever, clinical applications of the method require a

more exact knowledge of the FN impulse generationsite. TMS may also provide information on othercranial nerves (3, 4).

The latencies of muscle evoked potentials (MEPs)elicited rvith TMS are 1.0-1.2 ms longer than thoseelicited with electrical stimulation extracranially at thestylomastoid foramen ( l, 5, 6). The impulse generationsite can be calculated from this difference in latency,given that the FN conduction velocity is approximatelyl5-50 m/s (7, 8). The TMS site, with this method, maycome close to the internal acoustic meatus.

In order to determine the TMS impulse generationsite more precisely, we compared the MEPs elicitedwith TMS to those elicited with intracranial electricalstimulation (IES) of the facial nerve in 25 patients.The two methods were compared during posteriorfossa surgery where low constant current electricalimpulses can be given under visual control along thecourse of the nerve.

MATERIAL AND METHODS

Patients. The clinical details of the 25 patienis oper-ated on at the Helsinki University Central Hospitalare presented in Table I.

In 8 patients, a single dose of 2 mg panctrronium, a

muscle relaxant with a relatively long-lasting effect,was given at the beginning of anaesthesia. In patients1 and 21, pancuronium was also given during theoperation. Between the two neurophysiological exam-inations, no relaxants were given. In the rest of thepatients, only short-lasting muscle relaxants were ad-ministered for endotracheal intubation (Table I).

With the patient placed in the lateral position we

used a retrosigmoid approach, and debulked anddissected the tumor free from the cerebellum, thebrain stem and the cranial nerves. We used IES toidentify the FN. We opened the meatus with a highspeed drill and completed the dissection alternatinglyfrom the proximal and the distal direction. In somepatients with FN tumor infiltration, some smallpieces of tumor were left along the preserved FN.

Equipment. We used a commercially available four-channel Nihon Kohden Neuropack ENMG device forthe electrical stimulation and for the recording ofelectrically as well as magnetically induced responses.The magnetic stimulation was performed with threedifferent commercial devices (Table I).

Intracranial electrical stimulation (IES). We used a

direct cathodal surfac€ contact, using a small-tip elec-trode and a 0.2 mA electrical impulse with a durationof 0.1-0.2ms. The anode was flxed to the scalp,making the stimulation practically monopolar. Repet-itive I to l0 Hz frequency impulses were given.

The itistal FN stimuli were given in the internalacoustic meatus and the proximal ones at the FN

hron6$-Nro-hos66v--600voo-d-N Nms---N o s

N WN6hw v6vho -6--

nhnhr6o-N* mNh--Nv_NNO_O-_dN

-609006@$vaY-6nN NNaroshhhshh-hv-vh$$hNrsahvdoo

-N-Vh9r66Q-NOS49r@60*N6<h--NNNNNN

< (dddddddddddddd dddddd6

Q OOOOOOOOQOOOOOO OOOOOOO

,N=r)a-aaa=)aaaa=daao F o o o o o o o o o o o o o o o,E d d d o d o ocScccic=t aEc=CCaCSEECCCEEo 9^o o o o o o o o o o o o o o o a o oq\qq

Q 5 O O oA O O O O O oQO o o o ii o o o oo o ooEoqoooooooouoooox66...aSd N d 6 d 6 d d 6,6 6 d 6 tr d 6 N d d d d

z+z-l

L A ' 277

Tf r , ,l , o oT o r oo-- oq d1"z; L ^ -23. Izz i"+?l:l FT l- tL++ +J r++1-+++o+++d+f r rua__l a>>>>>-*_>>aq>>>!>r' + + + J I i:lJ ;J.J.JjiJiJJJJ iJJ:

LOL';ll-; :--i ++l - l-+*r {-*r - L +:++ l-rTfrrffLLfqLLr-,f---

- - - da! ooooqqqqaoo_ coo

o o o o o o o o o o o o oL666'o666'o666-o6t-'a. & L a L o L L a q L r tr E= E'E's'E'E'EES_ _tEEEEEEEEtEtE'= O O O O O O O O U9? @{@@{@@ {@a@@{3^= = = = = = = = =r+{+rr{+rJ6-5D9CDA9

??.CE E E E E E E E E E E EA E E E E E E E E E

-tqq add6ae E a@@{{@@@@@@{@d ddsMmddss

6 d S d 6 6 d 6 6 6 6rt 6 6 6 d dua4zlz z z z z zz zzzv z 2 2z zzzz z

93 9 9 lt-g I .999 19.! d q d o d o o o o o o d d o d o d d d o.j fij o otr=strdEd6666dCe6CdEEddE.ddEE EE EE E E E E E EEE EE EEEE EEE E E

(,s

.AOE

&e

JO

oE

o

aJ

.2oc,so

.9

tr

q

ozF

..!9trUE

=! 5

;EI

do -z;@d.b

o.I 3!

o P=

9o!

659';n"

ir\ O

-=> dEQ}

iX!2ll =6s9u;raQ5> @c ..=o9L

5=u

- ilq-\=E;g

9 E4

aa 9ll @-e

L=

e;;!

: dNh

i t d>; PE-Q.:c;

G -o n

E .!2^a. € *:1s &;w- oJ^:t h qsi r a-

: \E B

c .E i€ _. d c: tr

>xroo p x g!# *"'E g

I. Rimpilijinen et al- Acta Orotaryngot (Slockh) ill

Acta Otolaryrgot (Stockh) tn Sile of inlpulse generatiotl in trunscranial mognetic. slitnulaliotl 341

root entry zone near the brain stem (9). Cisternal

stimuli proximal to the meatus but distal to the rootentry zone were called intermediate stimuli.

Transcranial magnetic slitnulatiott (i"MS). The coilwas placed tangentially on the head in the temporo-oc-cipitat region. The estimated distance between the

plastic cobt of the coil and the scalp varied between

0.5-l cm. There was some intra- and interindividualvariation in the location of the coil due to instrumen-tation on the operational area. Also the type and shape

ofthe coil ring varied. Dantec S I00 coil was used withtwo patients, Magstirit High Power coil ring with l3patients, and Magstim butterfly coil ring with l0patients. In one patient, Cadrvell Focal Point coil ringwas used in addition to butterfly coil (Table I).Impulses, 35o/o to 5Oo/oof the maximum devicecapacitywere sufficient to evoke responses comparable to thosewith the distal IES.

Recording of responses- Muscle responses wererecorded on the side ofthe stimuli. In patients l, 2, 3,

I 5, and 20, a 5 mm surface electrode was placed on thenasolabial fold. The referential electrode was located

ipsilaterally on the nose at the nasal bone level.

In'patients 4 to 25, including l5 and 20, a standard0. I mm coaxial concentric needle electrode was used forrecording the responses. The electrode was inserted inthe superior orbicularis oris muscle close to the nasola-bial fold. We found the needle electrode more reliablebecause of its constant impedance throughout Iongoperations, and it also proved more sensitive for nervemonitoring during surgery. In cases with both surlaceand needle recordings, the responsds recorded rvithneedle electrode were used for statistical comparison.

The amplifier bandpass was set at 3 Hz (high pass)

ro 3000 Hz (low pass).We measured the muscle response latency from the

beginning of the deflection. The amplitude was the

maximum peak-to-peak voltage difference.The facial nerve was sensitive to saline flushing and

manipulation. Anaesthetics and changes in body tem-perature (10) may also modily facial nerve and my-oneuraljunction conductivity. Therefore in all IES andTMS measurements, we accepted the response with theshortest latency for statistical analysis, because it was

probably Ieast modified by external factors. The re-

sponses in Figs. I and 3 do not neccessarily show theresponses with shortest latencies in each stimulationsite but are rather time and phase related to surgery.

Statistical analysis. Student's ,-test was used tocompare the significance of differences.

RESULTS

Table II shows the individual responses. With TMSthe mean minimum latency of responses was 5.0 ms

o'' *l

6ms

r.0 dl

I

6ms

0., mvl

1_

o , *l

, o *l

F€. /. Five series ol four consecutive facial motor evokedpotentials recorded with a needle electrode from the supe-rior orbicularis oris muscle (Patient ll). Recordings wereperformed before and after lacial nerve injury I cm distallrom tire pons. (A) Transcranial magnetic stimulation(TMS) before the injury, at 10.40 hours. (B) Intracranialelectrical stimulation (lES) at the internal acoustic meatus

before the lesion, at 16.55 hours. (C) IES at root entry zoneat 20.00 hours, some minutes before the lesion. (D) IES onthe distal side of nerve anastomosis after the injury. (E)TMS alter the injury, at 21.37 hours.

(SD 0.58). With distal IES the corresponding latency

was 5.0 ms (SD 0.68). With intermediate and proxi-mal iES the mean latencies were 6.0 ms (SD 0.74)

and 6.7 (SD 0.89), respectively, and they differedsignificantly from the responses elicited with TMSand distal IES (p < 0.05). The MEP responses were

usually polyphasic, as shown in Fig. l. The latency

variation with the different IES sites and the TMSwas evident also with surface recording (Fig' 2).

The. continuity of the facial nerve was brokenproximal to the internal acoustic meatus in one

5ns

342 I. Ritnpildinen et al. Acla Otolaryngol (Slockh) lll

=]

q\t

e,

s

:I

t

;

o-

t:tq

B

.!

!*lJ

5F

@I].)

;o

a,r.l

E

oA

qri

!

E

tr

qtr-l

E

E

qcl

;E

&

q9..1

.s!E

qE1

=a

a

oa

o

!

E

,"t 9ezEF

c9

J_5

E

oo

o

,6c

J

DFUE

.eqqla ?':'lvlq nc-anY\qqqqqr.:cqq qSqeo-ooO o-oo* o-*-JNjJoiocj-j_: o_jdcioiN

ve6+6n oo $ oOOOOOO d- O d

q!nq

? q n i o? n \q ! q q o? q \o o o d o N r $ o 6 _ o- .-.

- o o * * o - N - d o oi oi oi -i ci j ai j -i o _: 6i

..1 ?\q.: qVl v.] !.1q e g $ o6ooo@ooo 6r oH + * o j - 6 - - 6i d d ci d j 6i i.i o 6i -i oi d d 6i

nq\qt!? qggq-] cl @dooo6a6oooeecaes 6\irrr \de\drqisriuiUodEix;r;

eqi\nq@96696

6A@OhhOO96rohN6*O-O$-rro+ ri + ri + ri d + d++ ri -i + + e + e v + ri + ri ri

q - q q q n g c 9a 1\ r: t q v'l qqn n q q a't \ v')V h hn nh€ vhS VhhS $ h VaV9 nW riviy n-idw<i

N

-Not6Or€6O-NmShar66O-No+6 _-NNNNNN

\qPt"l

9qn"t

? f t ? !: q q q R cl n c? vt:9 q -:..j - ? -: n -:. q -.9,1ooooooooooooooooooooooood nooooItt tllttt -l

r--o?qq

qi-q

w o90*$ 660$O

€e

Acta Otolaryngol (Srockh) I ll Sile of inrpulse gcncration in transcraninl tnagnelic stitttuloliott 343

0.4 mV

Fig. 2. Three series of trvo consecutive facial motor evokedpotentials recorded rvith a surface electrode from the supc-rjor orbicularis oris muscle (Patient 20). (A)'lranscranialnagnetic stimulation (TMS), latency 4.5 ms. ( B) Intracra-nial electrical stimulation (IES) at the internal acousticrxeatus, latency 4.5 ms. (C) IES at the root entry zone,latency 6.0 ns.

patient (19) (Fig. 3). In two other patients (1, 22),

with the preserved anatomicai nerve continuity, theelectrophysiological nerve conductivity rvas broken atthe sarne level. lvlEPs could be elicited rvith TMS ineach of these patients, and tlie latencies equalled tothose registered belore the nerve lesions.

In one patient (ll), the nerve continuity rvas lostI crn distal from ihe pons. The irltermecliate IESlatency was thereafler 6.7 ms. The TMS latenciesrvere 4.8 rns belore and after lhe injury (Fig. l).

Furthermore, in patient 2 the intermediate IES)atency increased to 9.0 ms rvhile the nerve rvas beingpulled. In cornparison, the minimum latency lor eiec-

tric stimulatio:r at the root entry zone belore theconduction delay rvas 6.0 rrs. In patient 7, the proxi-mal IES iatency was prolonged fronr the minimumvalue of 9.0 rns to lE.0 ms during the dissection ol the

tumor. Correspondingly, the anrplitudes ol the re-

sponses were decreased from the maximuur ol LI mVto 0.02 mV. The reason for latency increases was

:lost iikely a partial nerve in.jury. The TMS responses

remained essentia)ly as before the lesion in both olthe patients.

DISCUSSION

During posterior fossa operations the intracranialpart ol the faciai nerve is accessible to direct elect.ricalstimulation. The precise point of the stirnulation is

possible using a small-tip electrode and a Iow current.The site of impulse generated with TMS can bedetermined by comparing the latencies ol TMS re-

iIt_

6

Fr!. J. Four series o[ fir,e consecutive facial motor evokedpotentials, recorded from the superior orbicularis oris mus-cle in patient 19. Therc u,as a lacial nerve injury withcomplete conduction block at the entrance of the internalacoustic meatus. (A) Transcranial nragnetic stimulation(TMS) before the injury, al 10.18 hours. (B) TMS alter thcinjury at 19.34 hours. (C) Intracranial electrical stinrulation(lES) at the iDternal acoustic neatus belore the injury at16.18 irours. (D) IES near the brain stem belore the injuryat 18.03 hours. IES proximal to the site of tlre lesjon afterthe injury did not elic;l any response.

sponses to. those elicited with IES in predeterrninedparts of the facial nerve.

Responses elicited with TMS had the latency differ-ence range ol -0.8 to 0.5 ms when compared rvith

distal IES at the internal acoustic meatus, the mean

latencies being practically the same, 5.0 ms, with bothstimulation methods. Responses from intermediateand proximal stimulation invariably showed longerlatencies. (p <0.05). The lacial nerve continuity (2patients) or conductivity (4 patients) was lost orimpaired near the internal acoustic meatus or in the

intermediate area in 6 cases. In each o[ these patients

the TMS responses were intact alter the injury, but

the'IES responses, proximal or at the site of the

lesion, were absent or longer than those lrom TMSand distal IES. These findings suggest that TMS-in-duced facial nerve activation leading to MEP-re-sponse takes place near the internal acoustic meatus(11, 12). Based on the variation in the range oflatency differences with TMS and distal IES, weconsider that the site of impulse generation may alsovary at some extent between the individuals.

MEPs lrom intraoperative TMS have on an aver-age 0.6 ms longer latencies compared to those inhealthy volunteers (l). Facial nerve compression andtr.rmor inliltration may be some reasons lor this differ-ence- The lorver body temperature during the opera-tion, and anaesthetics may also prolong theconduction times.

The shape of the coil may have an effect onthe magnetic fielil stimulated (13). In this study,however, our comparison of round and butterfly coilsshowed no diff'erence in the site ol impulse genera-tion. Probably because all tnagnetic stimulators gen-erate a current widely distributed to the brain, thetype of stirnulator does not affect the responses.

The magnetic field inducEd by a nragnetic coil isIarge (13) and covers the whole intracranial part ofthe facial uerve. Probably the magnetic field activatesa longer segment of the nerve rather than an exactsite in the nerve. Possibly a proximal activation,simuitaneously with a more distal ond, may be inhib-ited by the refractory period of the nerve. Thereforethe depolarization of the distal nerve segment only,i.e. the segment near the internal acoustic meatus,probably generates the impulse initiating the MEp.

REFERENCES

I. Rimpildinen I, Eskola H, Hzikkinen V, Karma p. Tran-scranial facial nerve stimulation in normal subjects.Electromyogr Clin Neurophysiol l99l; 3l: 259-63.

2. Rimpildinen I, Karma P, Laranne J, Eskola H, Hrikki-ncn V. Magnetic facial nerve stimulation in Bel)'s palsy.Acta Otolaryngol (Stockh) 1992; ll2:3ll-6.

J. Bcnecke R, Meyer B-U, Schonle P, Conrad P. Transcra-nial magnetic stimualtion of the human brain: responscsin muscles suppiied by cranjal nerves. Exp Brain Res1988;71:623-32.

4. Rimpildinen I, Karma P, Eskola H, IJdkkinen V, Mas_netic lacial nervc stintulation in normal subiects: thriegroups of rcsponses. Acta Ololar)ngol tsrolkh,; loo-.;492:99*102.

5. Maccabee PJ, Amassian VE, Cracco Re, Cracco JB,Anziska BJ. Intracranial stinrulation of facial nervc inhuntans with the magnetic coil. E)ectroelcephalogr ClinNeurophysiol 1988; 70: 350-.1.

6. Schrieler TN, Mjlls KR, Murray NMF, Hess C\\r.Evalua[ion of proxintal lacial nerve coDduction bvtranscraniil magneric stinrrrl.,rion. J Ncurol Neuro.urqPsychiatry i988; 5i: 60-6.

7. Es.len E. TIrc acutc [r.ial pal,ies. Berlrrr: SpnngerYerlag, 1977.

8, Richrrond IL, Mahla M. Use of antidromic recordinslo n)onitor fatial r.errc function ir.trooperatirely. Ncu.rosurgery 1985; 16: 458-62.

9. Msller AR, Jannetta PJ. Henilacial spasm: results ofelectrophl,siologic recorciing durirg microvascular deconpressation operations. Neurology I985; 35: 969-7.1.

10. Brorvning JL, Heizer ML, Baskin DS. Varjations incorticomotor ard somatosellsory evoked polentials:effects oltentperature, halothane anesthesia, and arter ialpartial pressure ol CO2. Anesth Analg 1992; 14: 613-E.

11. Rosler KM, Hess CW, Schmid UD. Investigation offacial motor palhrv:rys by electrical and ntaqnetic s[inulation: sites and mechanisnts of excitation. J NcurolNeurosurg Psychiatry 1989; 52: I 149-56.

i2. Schmid UD, Moller Al{, Schmid J. Transcranial mag-nctie stintularion ercitcs thc I"brrir,tl,inc scqmenr of rlr-lacial nerve: an intraoperativc electrophy,sio)ogical studl,in nran. Ncurosci Lett l99l; 121:213 6.

13. Jalinous R. Technical and oractical aspects olmagneticnerve stimulation. J Clin Neuropvsiol 1991; 8: i0 25.

Address lor correspondence:P. KarrnaDcpartment of Otorhinolarl,ngologyHelsinki Univcrsity Central HospitalIlaartnraninkatu.lSF-00290 HelsinkiFin la nri