J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.49.4.411 on 1 April 1986. Downloaded from Journal of Neurology, Neurosurgery, and Psychiatry 1986;49:411-418 Intracranial stimulation of the trigeminal nerve in man I. Direct motor responses GIORGIO CRUCCU From the Pain Relief Foundation and Department ofNeurosciences, Walton Hospital, Liverpool, UK SUMMARY Direct electrical stimulation of the intracranial portion of the trigeminal nerve was per- formed in 23 subjects undergoing retrogasserian thermocoagulation for the treatment of idiopathic trigeminal neuralgia. In 16 subjects, who were having the operation for the first time, neurological examination was normal, as was neurophysiological testing of trigeminal function. Seven subjects were being operated for the second time, owing to a recurrence of symptoms. In all the subjects being operated for the first time, direct motor responses were obtained from ipsilateral temporalis, masseter and anterior belly of the digastric. The motor conduction velocity was equal for the fibres directed to all three muscles. This was estimated to be 54 m/s in the masseteric nerve and 55-68 m/s in the intracranial portion of the trigeminal nerve. Patients who had undergone previous thermo- coagulation had a considerably slower conduction velocity. It is supposed that myelin sheaths had been damaged at the first operation. Protected by copyright. The trigeminal nerve is a mixed nerve subserving sen- stimulate motor and sensory fibres, thus allowing the sation of the face and the muscles of mastication. The recording of both direct and reflex responses. This motor and sensory fibres, however, run in separate paper describes the motor responses in the masti- nerves for most of their course, partially merging only catory muscles from stimulation of the retrogasserian in the retrogasserian rootlets and in a short portion of rootlets and the mandibular nerve, in subjects with- the mandibular nerve.' While the cutaneous nerves, out neurological defects who were having the oper- through the supraorbital, infraorbital and mental for- ation for the first time, and in subjects who had been amina, emerge on the surface of the face, the motor already submitted to a surgical retrogasserian lesion, nerves lie in the depth of the facial structures and their either by "controlled differential thermo- terminal branches enter the muscle bellies on their coagulation"6 7 or by "partial root section".89 Both deep surface. This makes the stimulation of trigem- techniques are supposed to spare the motor fibres. inal efferent and afferent alpha fibres difficult. When compared with the main limb nerves, electro- Methods physiology has provided little information about the http://jnnp.bmj.com/ human trigeminal nerve. The conduction velocity of The study has been carried out in 23 subjects, who gave the motor fibres and the afferent sensory fibre groups informed consent, undergoing retrogasserian rhizotomy by have not been directly radiofrequency thermocoagulation6 for the treatment of id- measured, the cortical evoked iopathic neuralgia in the third (V3) or second (V2) division. potentials are far from being certainly identified2 3 There were I1 men and 12 women. The right side of the face and even the more widely used trigeminal reflexes are was affected in 14 cases and the left in nine cases. obscure in many details.45 Sixteen cases were diagnosed as idiopathic trigeminal neu- The operation of percutaneous trigeminal rhi- ralgia, eight in V3 and eight in V2. Their ages ranged from zotomy provides a unique opportunity to reach the 43 to 67 years, the mean age being 56 years. These subjects on September 27, 2021 by guest. intracranial portion of the trigeminal nerve and to were selected among those without abnormal neurological or electrodiagnostic findings using known electro- physiological techniques: blink reflex and jaw jerk, masti- Present address and address for reprint requests: Dr G Cruccu, Di- catory EMG.'0-14 In particular, the supraorbital nerve was partimento Neuroscienze, Universita' di Roma, Viale Universita', 30, stimulated transcutaneously at the supraorbital groove and 00185 Roma, Italy. the infraorbital nerve via a pair of fine needle-electrodes in- serted into the homonymous foramen. In both cases the Received 12 April 1985 and in revised form 25 June 1985. right and left side were stimulated with square wave 0 1-03 Accepted 27 June 1985 ms pulses of adequate intensity and surface recordings were 411 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.49.4.411 on 1 April 1986. Downloaded from 412 Cruccu Table 1 Direct responses in the masticatory muscles (mean + standard deviation ofnegative components) Masseter Temporalis Digastric Onset Amplitude Duration Onset Amplitude Duration Onset Amplitude Duration latency (p V) (ms) latency (pu V) (ms) latency (UV) (ms) (ms) (ms) (ms) Stimulation at 2-2 3100 3 9 2-2 4300 5 3 2-94 4400 4-3 foramen ovale +0 19 +2300 +07 +0-36 +4000 +09 +0 33 +3700 +0 5 (I 6 subjects) Stimulation at +0-26 -900 01 0-25 -1200 +014 +0-24 -1100 +003 clivus +0-14 +1800 +012 +0 19 +3000 +044 +0 09 +2000 +0-23 (paired differences with foramen ovale in p < 001 NS p < 005 p < 001 NS NS p < 001 NS NS 8 V3 subjects) Stimulation at -0 55 + 1200 -0.5 zygomatic arch +0-21 + 1500 +0-6 (paired differences with foramen ovale in p < 0-01 NS p < 0 05 8 V3 subjects) taken from the inferior orbicularis oculi muscle bilaterally. The percutaneous method employed for the lesion of the The jawjerk was evoked by tapping the chin with a triggered sensory root has been described in detail by several au- hammer and recorded from the masseter and temporal mus- thors.67 15 The ganglion was approached by passing a can- cles bilaterally. nula through the foramen ovale, guided by means of x rays Seven cases had previously undergone an operation for and image intensification with a TV monitor. The intra- the treatment of facial pain which had been diagnosed as operative recordings were first performed when radiographs Protected by copyright. idiopathic trigeminal neuralgia (table 2). Five cases were in lateral projection showed that the tip of the cannula had having the same operation, that is radiofrequency thermo- reached the foramen ovale and then when the level of the coagulation, for the second time, and two cases had had clivus was attained. In the first position the needle was pre- surgical partial section of the root. In two cases there had sumed to be in the mandibular nerve just distal to the gan- been no pain relief and in the others the neuralgia had re- glion, and in the second within the trigeminal root.1 ' The curred. All presented with sensory deficits to a variable ex- stimulating electrode consisted of an insulated stylet with a tent or neurophysiological abnormalities. Most patients bare tip of 2 mm which protruded by 1 mm from the were receiving anticonvulsants (mainly carbamazepine) up teflon-coated surgical needle, whose tip was bared for 5 mm. to 24 hours before surgery. During the operation the sub- The stylet was connected to the cathode and the cannula to jects were fully conscious, while they were always fully ana- the anode of the stimulator. Square-wave negative pulses esthetised during the coagulation, by means ofa short-acting (O 5-50 mA, O l-0 5 ms) were delivered manually at a low barbiturate (methohexitone IV). rate or repetitively at 1 Hz. Table 2 Masseter responses in re-operated subjects Case, sex Previous M-wave M-wave Intracranial Jaw Clinical-EMG Comment & age (yr) operation latency amplitude velocity reflex function http://jnnp.bmj.com/ Case 4 Thermocoagulation 29 ms 2-1 mV 28-34 m/s Delayed Normal Demyelination F 61 for left V2 (abnormal) (abnormal) 2 years before Case 5 Thermocoagulation 2-2 ms 0 3 mV 3540 m/s Absent Abnormal Demyelination M 63 for right V2-V3 (abnormal) (abnormal) & fibre loss 3 months before Case 11 Thermocoagulation 2-1 ms 12 mV 60-73 m/s Normal Normal Normal M 68 for left V3 10 months before Case 13 Thermocoagulation 2 4 ms 2-2 mV 30-32 m/s Delayed Normal Demyelination F 46 for right V3 (abnormal) on September 27, 2021 by guest. I month before Case 14 Partial right Absent Absent Non evaluable Absent Abnormal Massive fibre M 78 root section loss 20 years before Case 19 Partial left 2-2 ms 18 mV 48-56 m/s Normal Normal Normal F 66 root section 12 years before Case 22 Thermocoagulation 2-6 ms 2-4 mV 40-49 m/s Normal Normal Doubtful M 70 for left V2 (borderline) (borderline) 5 years before J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.49.4.411 on 1 April 1986. Downloaded from Intracranial stimulation of the trigeminal nerve in man I. Direct motor responses 413 Signals were recorded by surface electrodes from the ip- (fig 1). Concentric needle recordings demonstrated silateral temporal, masseter and anterior belly of the di- that the potentials actually originated in these mus- gastric muscles, and by needle-electrodes from the cles. The current threshold was 0-5-4 mA, without orbicularis oculi. In the first cases, recording from the con- significant differences among the three muscles. Re- tralateral facial and masticatory muscles was also per- sponses of maximum voltage, starting with a negative formed, as well as recordings by concentric EMG needles inserted into the muscle bellies. Signals were amplified, component were recorded with the electrode disposi- filtered (8-3200 Hz) and displayed by means of standard tion shown in fig 1. The average features of the M- electromyography, and recorded on tape. Off-line processing waves to supramaximal stimulation are given in table included rectification and averaging when adequate, A/D 1.