J Neurosurg 75:244-250, 1991
Diagnostic relevance of trigeminal evoked potentials following infraorbital nerve stimulation MASSIMO LEANDRI, M.D., AND EMILIO FAVALE, M.D.
Department of Neurology, Universio,r Genoa. Genoa, Italy v- A new tool in neurophysiological exploration of the trigeminal nerve has recently been introduced. It has been demonstrated that stimulation of the infraorbital nerve trunk gives rise to very reliable scalp responses reflecting the activity of the afferent pathway between the maxillary nerve and the brain stem. The authors demonstrate that alterations of such trigeminal evoked responses fit with documented pathological processes at various locations along the trigeminal pathway (maxillary sinus, parasellar region, and within the brainstem parenchyma). They report the findings in 68 patients suffering from "idiopathic" trigeminal neuralgia. Alterations of the response were detected in 33 cases, suggesting that some damage of the nerve had taken place either at the root entry zone into the pons (23 cases) or slightly distal to it (10 cases). Such results support the hypothesis that trigeminal neuralgia may be due to a compression of the trigeminal root at the pons entry zone. KEY WORDS
~
evoked potentials
N man, the neurophysiological investigation of cranial nerves is hampered by their particular anatomical situation. They run most of their course far from the reach of the electrodes conventionally used to stimulate and record neural activity. Furthermore, because of the short distances involved, the round shape of the head, and its conspicuous number of muscles, the recording electrodes lie close to, or are electrically linked to, many sources of artifacts. Among the cranial nerves, the trigeminal nerve is of particular importance because it provides sensory innervation to the entire face and a large part of the head, and it is involved in diseases of intriguing etiology, like trigeminal neuralgia. This nerve offers unique opportunities for the study of particular issues in neurophysiology such as pain sensation. A new technique which allows a reliable estimate of the function of the trigeminal nerve by means of evoked potentials has recently been described.~3 This consists in electrically stimulating the infraorbital nerve by means of two fine needles inserted into the infraorbital foramen. The technique is slightly invasive (not more than methods routinely used in electromyography or in evaluating the sensory conduction velocity of peripheral nerves), but ensures artifact-free massive stimulation of the trigeminal afferents. Following this stimulation, a series of very early waves can be recorded from the scalp. The first three of these, W~, W2, and W3, occur
I
244
9
trigeminal nerve
9
neurophysiology
9
neuropathy
at the mean latencies (+ standard deviation) of 0.9 + 0.08 msec, 1.84 _+ 0.11 msec, and 2.54 _+ 0.18 msec, respectively.~6 Simultaneous intraoperative recordings ~-~ have demonstrated that these waves arise from the proximal part of the maxillary nerve (W~), the proximal part of the retrogasserian root (W2), and the intratruncal fibers directed to the trigeminal nuclei (W3). These first three waves, therefore, offer a reliable tool to assess the function of the part of the trigeminal pathway between the infraorbital nerve and the brain stem? -'~5~ In the present paper we report the clinical applications of this technique in patients with well-documented lesions at various sites along this portion of the trigeminal pathway and in patients with "essential" trigeminal neuralgia. Our aim is to demonstrate the correlation between alterations of the evoked response and the lesion site, and to offer some reliable evidence about the origin of tic douloureux. Clinical Material and Methods
Stimulating and Recording Technique The trigeminal nerve potentials have been evoked by electrical stimulation of the infraorbital nerve. The stimulating electrodes were two fine Teflon-insulated needles, with a length of 37 m m and a diameter of 0.3 ram. The conductive tip had an area of 0.56 sq mm, with approximate impedance of 5000 ohm. The needle J. Neurosurg. / Volume 75~August, 1991
Trigeminal evoked potentials electrodes were inserted into the infraorbital foramen according to the technique previously described.'" Electrical pulses (0.05 msec in width) were delivered at the frequency of 3 pulses/sec. Intensity was set at four times the sensory threshold, which ranged from 0.08 to 0.15 mA. The test was repeated twice to ensure reproducibility of results. The side affected by pain and the healthy side were both stimulated to permit comparison of results. The recording electrodes were stainless steel needles inserted subcutaneously with impedance of approximately 2000 ohm. The electrode connected to the noninverting input of the amplifier was placed at the vertex (Cz), and the other, connected to the inverting input, was positioned on the neck over the spinous process of the C-7 vertebra. The signal from the Cz-C7 derivation was amplified and averaged (1000 to 2000 responses, at 500 kHz); the result of averaging was drawn on an x-y digital plotter (the overall bandpass was 10,000 to 15,000 Hz, 3 dB/oct).
Analysis of Results Delay or absence of one or more of the waveform components was defined as an abnormal response. As suggested by Leandri, et al., '~ interpeak intervals (Wt-W2 and Wj-W3) and side-to-side asymmetry of these interpeak intervals were considered the relevant and sensitive parameters to detect a delay of the W2 or W3 peak. Reference to the upper limits of normality '6 was made in all cases. Case Material We have studied 68 patients suffering from "essential" trigeminal neuralgia and three with radiologically demonstrated lesions. All the patients studied gave their informed consent to the procedures. The three patients were selected to illustrate the effect of a lesion located at a different portion along the trigeminal pathway and to establish the possibility of a specific alteration of the evoked response. Their cases are reported briefly below. Case 1: Maxillary Sinusitis. This 45-year-old man first came to our attention in 1978, when he presented with intense right fronto-orbital pain and paresis of the fight sixth cranial nerve. These symptoms subsided after about 2 months. Arteriography of the right carotid artery was unremarkable; an erylhrocyte sedimentation rate of 75 mm/hr was the only abnormal blood test. Approximately every 15 days after that, the patient complained of excruciating orbital pain and paresthesiae in the second trigeminal division lasting 24 to 48 hours. In December, 1984, after an episode of pain, he developed an impairment of the right sixth nerve function which lasted for several months. He was hospitalized in January, 1985. Neurological examination demonstrated only paresis of the right sixth cranial nerve. The trigeminal evoked potentials were monitored for the first time and were definitely abnormal. Skull x-ray J. Neurosurg. / Volume 75/August, 1991
FIG. I. Computerized tomography scan in Case 1 showing filling of the right maxillarysinus.
films and computerized tomography (CT) scans demonstrated a hyperdense lesion of the fight maxillary sinus (Fig. 1). Surgical exploration of the maxillary sinus revealed a diffuse inflammatory process from which granulomatous tissue could be partially evacuated. The patient was treated with antibiotic and steroid therapy. His symptoms began to improve soon after, and repeat trigeminal evoked potential testing at this time showed results within normal limits. Case 2." Giant Aneurysm of the Carotid Siphon. This 63-year-old woman was referred to our institution in October, 1984, with a series of neurological signs and symptoms suggesting a partial cavernous sinus syndrome on the fight side. Four years earlier she had first suffered diplopia due to impairment of the fight sixth cranial nerve; subsequently, she developed defective function of the oculomotor nerve, signs of sympathetic paralysis, slight exophthalmos, pain, and hypesthesia of the first two trigeminal divisions. She experienced long periods of almost complete remission and was prone to underestimate her disease. In 1980, x-ray films, electroencephalography (EEG), and CT were reported within normal limits. Although no objective sensory loss could be detected on the right side of her face, the patient complained of altered sensation; this symptom encouraged us to perform a trigeminal evoked potential study, which gave abnormal results. A new CT scan was then obtained, which showed a large mass in a right parasellar location (Fig. 2). Arteriography demonstrated a "giant aneurysm" of the internal carotid artery within the cavernous sinus. Case 3. Demyelinating Plaques of the Pons. This 39-year-old woman began to suffer trigeminal pain about 5 years before admission. It was a typical "tic douloureux," localized on the right side of her face, mainly in the second trigeminal division. She was free 245
M. Leandri and E. Favale
Fl(;. 2. Computerizcd Iomographx scan in Case 2 showing a large spherical mass in a pamsellar position on the fight, which is likely to impinge on the lrigeminal nerve,
of pain for long periods every year. About 6 to 7 months before consulting us, she began to complain of a continuous, burning dysesthesia involving the right side of her face, where she still suffered bouts of typical tic douloureux. Neurological examination showed only a slight hypesthesia of the first and second trigeminal divisions on the right. Trigeminal evoked potentials were tested and found to be abnormal. Magnetic resonance imaging demonstrated various sites of altered signal around the ventricles and within the brain stem, possibly representing demyelinating plaques. Some of these were scattered around the right trigeminal root entry zone (Fig. 3). A diagnosis of multiple sclerosis was confirmed by alterations found in other neurophysiological tests, including brain-stem auditory evoked responses and somatosensory and visual evoked potentials. Further confirmation was provided by the finding of oligoclonal banding on examination of the cerebrospinal fluid.
Patients With "Essential" Trigeminal Neuralgia. The 68 patients suffering from "essential" trigeminal neuralgia were seen at the Headache and Pain Center of the University of Genoa between October, 1983, and November, 1989. The patients ranged in age between 48 and 83 years (mean 72 years). Pain involved the first division in three cases, the second division in 19 cases, the third division in eight cases, the first and second divisions in 11 cases, the second and third divisions in 26 cases, and all three divisions in one case. Thus, the second division was involved in 57 of the 68 patients. The right side of the face was affected in 39 cases, the left side in 28 cases, and in one case the neuralgia shifted from the right to the left side. The patients had suffered trigeminal pain for an approximate time varying between 1 and 15 years (mean 6 years). None of our patients had undergone any surgical 246
FiG. 3. Magnetic resonance image in Case 3 showing two or three small demyelinating plaques within the pons, around the trigeminal root entry zone on the right, treatment, whereas most had been taking carbamazepine or other drugs for some time. Neurological examination was normal in all cases. Standard radiographic investigation, EEG, and CT were within normal limits in all cases. Results
Patients bl'ith Lesions Case I." Maxillat:v Sinusilis. When trigeminal evoked potential testing was performed for the first time in this patient, a definite abnormality of the W~ component and of the subsequent waves was observed (Fig. 4, lower &['t trace). The latency of the main peak of W~ was slightly delayed: furthermore, the overall duration of this triphasic component was increased, which resulted in further delay of the subsequent components W, and W> These features are particularly evident when the trace is compared to that obtained after stimulation on the healthy side (Fig. 4 upper le[~ trace). These neurophysiological findings suggested defective conduction before or possibly at the point of origin of W,: that is, before or at the entrance of the maxilla~' nerve into the gasserian ganglion. The second session of trigeminal evoked potential testing, performed when the patient was clinically improved, showed a normal response, suggesting a good recovery, of trigeminal nerve function (Fig. 4 right traces). Case 2." Gianl Aneurvsm of the Carotid Siphon. In this patient stimulation of the right infraorbital nerve gave rise to an abnormal response, with the two components W_, and W~ missing, whereas W~ was normal (Fig. 5). This suggested a lesion of the trigeminal pathway located at some point between the sites of origin of W~ and We: that is, between the gasserian ganglion and the proximal part of the retrogasserian root. The J. Neurosurg. / f~blume 75/August, 1991
Trigeminal evoked potentials
FIG. 4. Trigeminal evoked responses from Case 1. Le& Trigeminal evoked potential recordings from the healthy (upper trace) and affected (lower trace) sides before treatment was started. Stimulation of the affected side gives rise to an abnormally delayed W~ peak (latency calculated at the main positive peak, marked by an asterisk). Arrows mark the two negative peaks delimiting the Wt component and so define its duration. The duration of W~ on the affected side is much longer than on the healthy side. As a result of the W~ abnormalities, the Wz and W3 components also show longer latencies, Right: Responses obtained 8 weeks after therapy began. The upper trace was obtained after stimulation of the healthy side and the lower trace after stimulation of the affected side. No significant differences are now evident. Calibrations: I msec/division for x axis and 0.75 uV/division for y axis. Negative upward. Although the analysis time was l0 msec, in this and the following figures only the first, relevant, part of the trace is shown.
FIG. 5. Trigeminal evoked potentials from Case 2. Calibrations: I msec/division for x axis and 0.75 uV/division for y axis. Negative upward. Upper Trace: Normal response from the healthy side. Lower Trace. Response after stimulation of the af('ected side. The W2 and W~ components are missing, while W~ is within normal limits.
trigeminal evoked potentials obtained on the left side were of course normal.
Case 3." Demyelinating Plaques of the Pons. In this patient stimulation of the right infraorbital nerve evoked an abnormal response, where the W~ component was missing (Fig. 6). The first two components, W~ and W2, were within normal limits. These findings suggested a lesion of the trigeminal pathway between the points of origin of W: and W~: that is, between the entrance of the trigeminal root into the ports and the trigeminal nuclei. Patients I4"7th "Essential" Trigeminal Neuralgia The trigeminal evoked potentials were found altered in 33 of the 68 patients suffering from "essential" trigeminal neuralgia. In particular, the W3 component was missing in seven cases, and the asymmetry of the W~-W3 interval exceeded the upper limits of normality on the affected side in 26 patients. Ten of the patients ,1. Neurosurg. / Vohtrne 75/Augtc~t, 1991
FIG. 6. Trigeminal evoked potentials from Case 3. Calibrations: I msec/division for x axis and 1 uV/division for y axis. Negative upward. Upper Trace: Response after stimulation of the healthy side. Lower Trace: Response after stimulation of the affected side. The W3 component is missing, while W~ and W2 do not differ significantly from normal.
247
M. Leandri and E. Favale
FIG. 7. Trigeminal evoked responses in some representativecases of "essential" trigeminal neuralgia. The three sets are obtained from Cases 617 (left), 603 (center), and 5 t7 (right). Calibrations: 1 msec/divisionfor x axis and 1uV/division for y axis. Negativeupward. Upper Traces: Responsesafter stimulation of the healthy side. Lower Traces: Responses after stimulation of the pain-affectedside showinga delay of W3 (left), delay of both W2 and W~ (center), and absence of W3 (right).
with missing or altered W.~ peaks also had abnormal asymmetry of the W~-W2 interval. One patient, who had a prolonged W~-W3 interval but no significant sideto-side asymmetry, was also considered abnormal. Examples of the alterations seen in this group are shown in Fig. 7. A complete summary of the altered neurophysiological findings in the 33 patients is presented in Table 1. Painful involvement of the first and second divisions was seen in seven of these cases, the second and third divisions in 15, the second division alone in eight, the first division alone in one, and the third division alone in two. Overall, the second division was involved in 30 cases. We could not detect any relationship between the degree of alteration and the patient's age, duration of illness, or pain intensity. The percentage of patients with painful involvement of the second division was higher in the group with altered responses (30 of 33 cases, or 90.9%) than in the group with normal responses (27 of 35 cases, or 77.14%). No other clinical differences were noted between the two groups. Discussion
Historical Perspective A brief review of the neurophysiological methods used in the past to investigate the trigeminal nerve seems appropriate in order to clarify the reasons that prompted us to use a new technique in our research. The classic method of studying this nerve was the blink reflex. Unfortunately, the blink reflex has a high degree of variability: the latency of its first component, R~, ranges from approximately 8 to 14 msecfl and the 248
related side-to-side asymmetry is considered abnormal only if it exceeds 1.2 to 1.5 msec. ~j~9 Slight damage to the trigeminal nerve would probably cause a delay on the reflex well within its normal range (even when compared with the healthy side), and may therefore be undetectable. Furthermore, like other reflexes, the blink has a good biological safety factor, so that the loss or damage of a few afferent fibers is compensated at the reflex center, and the latency of the elicited muscle contraction is unaffected. Similarly, other reflexes of the trigeminal nerve (the jaw jerk, the corneal reflex, and the trigeminocervical reflex) share the drawbacks of the blink, and also have not been found consistently altered in diseases with minor damage to the nerve, such as trigeminal neuralgia. Higher precision and sensitivity are offered, at least in principle, by evoked potential measurement, which is now routinely used for assessment of the somatosensory pathways from the upper and lower limbs. A number of authors have also used this method to study trigeminal afferents,23s-~223but the results obtained are very much open to criticism because of the heavy contamination of responses by muscular activity. ~2.t3.~ It is only recently that reliable evoked potentials could be obtained by electrically stimulating the peripheral branches of the trigeminal nerve within their foramina. L~.~a.~v For this study of trigeminal evoked potentials in a clinical setting, we electrically stimulated the infraorbital nerve because, of the three peripheral trigeminal branches, this has been the most extensively studied with published normative data. ~ In addition to the first three waves (W~, W2, and W3), stimulation of this nerve evokes further waves at slightly later latencies, called J. Neurosurg. / Volume 75/August, 1991
Trigeminal evoked potentials TABLE 1 Latencies (msecj o[altered responses in 33 0/68 CO.'(eS ()[' "essential" trigeminal neuralgia* Hospital File No.
Right Side W~ W2 W~
Wa
Left Side W, W3
15 0.97 1.93 2.53 0.95 1.93 2.78+ 24 1.00 1.87 2.35 0.94 1.89 2.71+ 28 0.88 1.82 2.58 0.80 1.95~ 2.74t 30 1.10 2.20 -0.98 2.10 2.90 32 0.81 [.84 2.65? 0.98 1.95 2.52 47 1.01 1.95 2.55 1.00 2.145- 2.76? 62 0,82 1.80 2.65 0.84 1.96"t 2.85?1 . 90 0.83 1.92 2.50 0.86 1.92 2.72? 123 0.85 1.907 2.65? 0.86 1.77 2.45 126 0.95 2.09 2.99 0.95 2.15 3.317 134 0.90 1.89 2.79 1.03 2.22? 3.78? 154 1.09 1.91 2.48 1.01 1.92 2.65? 164 0.98 1.88 2.89 0.95 2.00? 3.20? 182 0.95 1.85 2.56 0.91 1.91 3.04+ 184 1.07 2.15 3.53? 0.97 2.07 3.29 207 1.06 2.33? 3.125- 1.01 2.05 2.81 208 1.03 2.00 2.66 0.87 1.79 2.72? 219 0.81 1.87 2.67 0.84 2.03 -251 0.87 1.78 2.69? 0.96 1.80 2.60 316 0.90 1.93 3.09 0.90 1.92 3.30? 332 0.81 1.68 2.33 0.95 1.75 2.65? 432 0.95 1.88 -1.10 2.05 2.85 436 0.92 2,00 3.00? 0.87 1.85 2.58 453 1.03 1,93 2.87? 0.88 1,78 2.40 456 0.77 1,75 3.09? 0.80 1.77 2.59 457 1.09 2.19 3.40? 0.90 1.99 2.68 505 0.82 1,72 -0.91 1.69 2.40 517 0.96 2.09"I" -0.93 1.92 2.67 519 0.95 2,00 -0.90 1.95 2.69 566 1.08 2.02 3.88 1.04 2.04 -578 0.90 1.957 2.66? 1.10 1.90 2.44 603 0.81 1.787 3.187 0.82 1.62 2.50 617 0.88 1.90 2.76"I" 0.90 1.81 2,53 * All subjectshad both sidesinvestigated,as side-to-sideasymmetry, was considered essential to define the abnormality of a response. -= missingcomponent. t Delayedcomponents:delay was calculatedon the basis of interpeak intervalsand their side-to-sideasymmetry.
P4, N4, P6, and NT; '~ however, they are not as constant as the first three waves, nor has their origin been as precisely demonstrated (although there is little doubt that they arise from neural structures). It is worth adding that, at latencies longer than 8 to 10 msec, it is not possible to record any trustworthy activity from the scalp, at least in the waking subject; the occurrence of reflexes (blink and other reflexes) heavily contaminates the responses, whatever method is used to elicit activity in the trigeminal afferents. '~ Measurement of scalp responses after infraorbital nerve stimulation has been employed in patients with tumors at the base of the skull, allowing precise localization of the damage site along the trigeminal pathway. ~5 Alterations of the response have been found even in cases where no clinical involvement of the trigeminal nerve was evident, a fact suggesting that the method has high sensitivity. The diagnostic power of the trigeminal evoked responses and their site of origin J. Neurosurg. / Volume 75/August. 1991
are confirmed by the findings in some patients after surgical [esioning of the retrogasserian root. '~ C o m m e n t s on Results The first three cases reported here with lesions involving three specific sites of the afferent pathway (that is, the portion peripheral to the gasserian ganglion, the gasserian ganglion itself and/or the retrogasserian root, and the portion inside the brain stem) each gave rise to a definite alteration of the response. This confirmed the origin of the first three waves as demonstrated by intraoperative recordings. ~-" The degree of precision and sensitivity of trigeminal potentials evoked by infraorbital nerve stimulation are comparable to those of the sophisticated technique used to test the conduction velocity of peripheral nerves. Such properties make measurement of the trigeminal evoked potentials ideal for investigating the possible slight damage that may be present in patients suffering from "essential" trigeminal neuralgia. Previous preliminary results from our laboratory, showed that only nine (23.7%) of 38 cases suffering from trigeminal neuralgia had abnormalities of the trigeminal evoked potentials. '~ In the present updated series of 68 patients, we found that the percentage of abnormal findings had risen to 48.5%. This difference is due not only to an increase in the number of cases examined but also to more strict upper limits of normality as defined by a comprehensive study on normative data.'6 The majority (71%) of alterations seen in the present series involved the W3 component, suggesting a lesion at the root entry zone or just after it; only 29% of the alterations seemed to be due to more distal lesions of the trigeminal root. These results are in agreement with the hypothesis put forward by several authors (see the reviews by Adams ~ and Selby2~ that the pathology of the root entry zone is of crucial importance in the occurrence of tic douloureux. This hypothesis has been the object of much controversy, as the only evidence of damage to the root entry zone has been visual inspection during surgery. New MR techniques allow noninvasive examination of that region; ~~ however, few cases have been investigated and the images produced are not always absolutely convincing. Thus, measurement of the trigeminal evoked potentials seems to provide the only available objective evidence of damage of the fifth cranial nerve in trigeminal neuralgia. Clinical and Research Developments Measurement of the trigeminal evoked potentials is open to further improvements, which should enhance its clinical usefulness. In our laboratory,, we have been able to record scalp responses evoked after selective stimulation of the supraorbital and of the mental nerve trunks; 14'7 they have waveforms and latencies similar to those obtained after stimulation of the infraorbital nerve. These methods, when properly developed, will permit a complete examination of the trigeminal divi249
M. Leandri and E. Favale sions. One of the predictable consequences could be an increase of altered findings in trigeminal neuralgia, further confirming the hypothesis of damage to the root entry zone. An interesting development of the technique we have used to stimulate the infraorbital, supraorbital, and mental nerves is that the needles positioned to stimulate the nerve trunks are also in an ideal position to record volleys from nerves if a stimulus is applied to the peripheral branches. We have so far been able to record from the infraorbital nerve trunk the compound action potential evoked by stimulation of the upper lip and gum. ~xBy improving this technique, we hope to provide direct evidence relating to the type of peripheral fibers involved in physiological and pathological processes in man. The identification of fibers responsible for the various trigeminal reflexes or the assessment of damage caused by herpetic neuropathy are examples of possible applications of this technique. So far these issues have been subject to indirect speculation? ~eJ The usefulness of trigeminal evoked potential measurement is not limited to the diagnosis of nerve damage, but extends to the operating theater, as already pointed out. ~2.tS''sExamples of intraoperative monitoring both with orthodromi&' and antidromic 7 stimulation have been provided, so that reliable testing of the trigeminal function could be performed with the patient fully anesthetized. We believe that the reliable neurophysiological investigation of the trigeminal nerve sought by Adams ~in his recent review and by any other researcher interested in the problem, is now within reach of most laboratories.
References I. Adams CBT: Microvascular compression: an alternative view and hypothesis. J Neurosurg 70:1-12, 1989 2. Bennett MH, Jannetta PJ: Trigeminal evoked potentials in humans. Eleetroencephalogr Clin Neurophysiol 48: 517-526, 1980 3. Buettner VW, Petruch F, Scheglmann K, et al: Diagnostic significance of cortical somatosensory evoked potentials following trigeminal nerve stimulation, in Courjon J, Mauguiere F, Revol M (eds): Clinical Applications of Evoked Potentials in Neurology. New York: Raven Press, 1982, pp 339-345 4. Cruccu G, Agostino R, Inghilleri M, et al: The masseter inhibitory reflex is evoked by innocuous stimuli and mediated by A-beta afferent fibres. Exp Brain Res 77: 447-450, 1989 5. Cruccu G, Berardelli A, Manfredi M: Afferents for the human corneal reflex, a Neuroi 234:64, 1987 6. Cruccu G, Bowsher D: Intracranial stimulation of the trigeminal nerve in man. II. Reflex responses, a Neurol Neurosnrg Psychiatry 49:419-427, 1986 7. Cruccu G, Inghilleri M, Manfredi M, et al: Intracranial stimulation of the trigeminal nerve in man. lII. Sensory potentials. J Neurol Neurosurg Psychiatry 50: 1323-1330, 1987
250
8. Drechsler F: Short and long latency cortical potentials following trigemina[ nerve stimulation in man, in Barber C (ed): Evoked Potentials. Baltimore: University Press, 1980. pp 415-422 9. Goodgold J, Eberstein A: Electrodiagnosis of Neuromuscular Diseases. Baltimore: Williams & Wilkins, 1983, p 260 10. Hutchins LG, Harnsberger HR, Hardin CW, et al: The radiologic assessment of trigeminal neuropathy. AJR 153: 1275-1282, 1989 11. Kimura J, Powers JM, Van Allen MW: Reflex response of orbicu[aris oculi muscle to supraorbital nerve stimulation. Arch Neurol 21:193-199, 1969 12. Leandri M, Campbell JA: Origin of early waves evoked by infraorbital nerve stimulation in man. Eleetroencephalogr Clin Neurophysiul 65:13-19, 1986 13. Leandri M, Parodi CI, Fevale E: Early evoked potentials detected from the scalp of man following infraorbital nerve stimulation. Eleetroencephalogr Clin Neurophysiol 62:99-107. 1985 14. Leandri M, Parodi CI, Favale E: Early scalp responses evoked by stimulation of the supraorbital nerve in man. Electroencephalogr Clin Neurophysiol 74:367-377, 1989 [ 5. Leandri M, Parodi CI, Favale E: Early trigeminal evoked potentials in tumours of the base of the skull and trigeminal neuralgia. Electroencephalogr Clin Neurophysiol 71: 114-124, 1988 [6. Leandri M. Parodi CI, Favale E: Normative data on scalp responses evoked by infraorbital nerve stimulation. Electroencephalogr Clin Neurophysiol 71:415-421, 1988 17. Leandri M, Parodi CI, Rigardo S, et al: Early scalp responses evoked by stimulation of the mental nerve in humans. Neurology 40:315-320, 1990 18. Leandri M, Parodi CI, Zattoni J, et al: Subcovtical and cortical responses following infraorbital nerve stimulation in man. Electroencephalogr Clin Neuropathol 66: 253-262, 1987 [9. Ongerboer de Visser BW, Goor C: Electromyographic and reflex study in idiopathic and symptomatic trigeminal neuralgias: latency of the jaw and blink reflexes. J Neurol Neurosurg Psychiatry 37:1225-1230, 1974 20. Selby G: Diseases of the fifth cranial nerve, in Dick PJ, Thomas PK, Lambert EH, et al (eds): Peripheral Neuropathy, ed 2. Philadelphia: WB Saunders, 1984, Vol 2, 1224-1265 21. Shahani B: The human blink reflex. J Neurol Neurosurg Psychiatry 33:792-800, 1970 22. Singh N, Sachdev KK, Brisman R: Trigeminal nerve stimulation: short latency somatosensory evoked potentials. Neurology 32:97-10l, 1982 23. Stohr M, Petruch F: Somatosensory evoked potentials following stimulation of the trigeminaI nerve in man. J Neurol 220:95-98, 1979 24. Tash RR, Sze G, Leslie DR: Trigeminal neuralgia: MR imaging features. Radiology 172:767-770. 1989 Manuscript received June 28, 1990. Accepted in final form December 26, 1990. This paper is dedicated to Prof. Carlo Loeb of Genoa, Italy, on the occasion of his retirement. Address reprint requests to." Massimo Leandri, M.D., Department of Neurology, University of Genoa, Via De Toni 5, [6132 Genoa, Italy.
,L Neurosurg. / Volume 75/August, 199I
Diagnostic relevance of trigeminal evoked potentials following infraorbital nerve stimulation MASSIMO LEANDRI, M.D., AND EMILIO FAVALE, M.D.
Department of Neurology, Universio,r Genoa. Genoa, Italy v- A new tool in neurophysiological exploration of the trigeminal nerve has recently been introduced. It has been demonstrated that stimulation of the infraorbital nerve trunk gives rise to very reliable scalp responses reflecting the activity of the afferent pathway between the maxillary nerve and the brain stem. The authors demonstrate that alterations of such trigeminal evoked responses fit with documented pathological processes at various locations along the trigeminal pathway (maxillary sinus, parasellar region, and within the brainstem parenchyma). They report the findings in 68 patients suffering from "idiopathic" trigeminal neuralgia. Alterations of the response were detected in 33 cases, suggesting that some damage of the nerve had taken place either at the root entry zone into the pons (23 cases) or slightly distal to it (10 cases). Such results support the hypothesis that trigeminal neuralgia may be due to a compression of the trigeminal root at the pons entry zone. KEY WORDS
~
evoked potentials
N man, the neurophysiological investigation of cranial nerves is hampered by their particular anatomical situation. They run most of their course far from the reach of the electrodes conventionally used to stimulate and record neural activity. Furthermore, because of the short distances involved, the round shape of the head, and its conspicuous number of muscles, the recording electrodes lie close to, or are electrically linked to, many sources of artifacts. Among the cranial nerves, the trigeminal nerve is of particular importance because it provides sensory innervation to the entire face and a large part of the head, and it is involved in diseases of intriguing etiology, like trigeminal neuralgia. This nerve offers unique opportunities for the study of particular issues in neurophysiology such as pain sensation. A new technique which allows a reliable estimate of the function of the trigeminal nerve by means of evoked potentials has recently been described.~3 This consists in electrically stimulating the infraorbital nerve by means of two fine needles inserted into the infraorbital foramen. The technique is slightly invasive (not more than methods routinely used in electromyography or in evaluating the sensory conduction velocity of peripheral nerves), but ensures artifact-free massive stimulation of the trigeminal afferents. Following this stimulation, a series of very early waves can be recorded from the scalp. The first three of these, W~, W2, and W3, occur
I
244
9
trigeminal nerve
9
neurophysiology
9
neuropathy
at the mean latencies (+ standard deviation) of 0.9 + 0.08 msec, 1.84 _+ 0.11 msec, and 2.54 _+ 0.18 msec, respectively.~6 Simultaneous intraoperative recordings ~-~ have demonstrated that these waves arise from the proximal part of the maxillary nerve (W~), the proximal part of the retrogasserian root (W2), and the intratruncal fibers directed to the trigeminal nuclei (W3). These first three waves, therefore, offer a reliable tool to assess the function of the part of the trigeminal pathway between the infraorbital nerve and the brain stem? -'~5~ In the present paper we report the clinical applications of this technique in patients with well-documented lesions at various sites along this portion of the trigeminal pathway and in patients with "essential" trigeminal neuralgia. Our aim is to demonstrate the correlation between alterations of the evoked response and the lesion site, and to offer some reliable evidence about the origin of tic douloureux. Clinical Material and Methods
Stimulating and Recording Technique The trigeminal nerve potentials have been evoked by electrical stimulation of the infraorbital nerve. The stimulating electrodes were two fine Teflon-insulated needles, with a length of 37 m m and a diameter of 0.3 ram. The conductive tip had an area of 0.56 sq mm, with approximate impedance of 5000 ohm. The needle J. Neurosurg. / Volume 75~August, 1991
Trigeminal evoked potentials electrodes were inserted into the infraorbital foramen according to the technique previously described.'" Electrical pulses (0.05 msec in width) were delivered at the frequency of 3 pulses/sec. Intensity was set at four times the sensory threshold, which ranged from 0.08 to 0.15 mA. The test was repeated twice to ensure reproducibility of results. The side affected by pain and the healthy side were both stimulated to permit comparison of results. The recording electrodes were stainless steel needles inserted subcutaneously with impedance of approximately 2000 ohm. The electrode connected to the noninverting input of the amplifier was placed at the vertex (Cz), and the other, connected to the inverting input, was positioned on the neck over the spinous process of the C-7 vertebra. The signal from the Cz-C7 derivation was amplified and averaged (1000 to 2000 responses, at 500 kHz); the result of averaging was drawn on an x-y digital plotter (the overall bandpass was 10,000 to 15,000 Hz, 3 dB/oct).
Analysis of Results Delay or absence of one or more of the waveform components was defined as an abnormal response. As suggested by Leandri, et al., '~ interpeak intervals (Wt-W2 and Wj-W3) and side-to-side asymmetry of these interpeak intervals were considered the relevant and sensitive parameters to detect a delay of the W2 or W3 peak. Reference to the upper limits of normality '6 was made in all cases. Case Material We have studied 68 patients suffering from "essential" trigeminal neuralgia and three with radiologically demonstrated lesions. All the patients studied gave their informed consent to the procedures. The three patients were selected to illustrate the effect of a lesion located at a different portion along the trigeminal pathway and to establish the possibility of a specific alteration of the evoked response. Their cases are reported briefly below. Case 1: Maxillary Sinusitis. This 45-year-old man first came to our attention in 1978, when he presented with intense right fronto-orbital pain and paresis of the fight sixth cranial nerve. These symptoms subsided after about 2 months. Arteriography of the right carotid artery was unremarkable; an erylhrocyte sedimentation rate of 75 mm/hr was the only abnormal blood test. Approximately every 15 days after that, the patient complained of excruciating orbital pain and paresthesiae in the second trigeminal division lasting 24 to 48 hours. In December, 1984, after an episode of pain, he developed an impairment of the right sixth nerve function which lasted for several months. He was hospitalized in January, 1985. Neurological examination demonstrated only paresis of the right sixth cranial nerve. The trigeminal evoked potentials were monitored for the first time and were definitely abnormal. Skull x-ray J. Neurosurg. / Volume 75/August, 1991
FIG. I. Computerized tomography scan in Case 1 showing filling of the right maxillarysinus.
films and computerized tomography (CT) scans demonstrated a hyperdense lesion of the fight maxillary sinus (Fig. 1). Surgical exploration of the maxillary sinus revealed a diffuse inflammatory process from which granulomatous tissue could be partially evacuated. The patient was treated with antibiotic and steroid therapy. His symptoms began to improve soon after, and repeat trigeminal evoked potential testing at this time showed results within normal limits. Case 2." Giant Aneurysm of the Carotid Siphon. This 63-year-old woman was referred to our institution in October, 1984, with a series of neurological signs and symptoms suggesting a partial cavernous sinus syndrome on the fight side. Four years earlier she had first suffered diplopia due to impairment of the fight sixth cranial nerve; subsequently, she developed defective function of the oculomotor nerve, signs of sympathetic paralysis, slight exophthalmos, pain, and hypesthesia of the first two trigeminal divisions. She experienced long periods of almost complete remission and was prone to underestimate her disease. In 1980, x-ray films, electroencephalography (EEG), and CT were reported within normal limits. Although no objective sensory loss could be detected on the right side of her face, the patient complained of altered sensation; this symptom encouraged us to perform a trigeminal evoked potential study, which gave abnormal results. A new CT scan was then obtained, which showed a large mass in a right parasellar location (Fig. 2). Arteriography demonstrated a "giant aneurysm" of the internal carotid artery within the cavernous sinus. Case 3. Demyelinating Plaques of the Pons. This 39-year-old woman began to suffer trigeminal pain about 5 years before admission. It was a typical "tic douloureux," localized on the right side of her face, mainly in the second trigeminal division. She was free 245
M. Leandri and E. Favale
Fl(;. 2. Computerizcd Iomographx scan in Case 2 showing a large spherical mass in a pamsellar position on the fight, which is likely to impinge on the lrigeminal nerve,
of pain for long periods every year. About 6 to 7 months before consulting us, she began to complain of a continuous, burning dysesthesia involving the right side of her face, where she still suffered bouts of typical tic douloureux. Neurological examination showed only a slight hypesthesia of the first and second trigeminal divisions on the right. Trigeminal evoked potentials were tested and found to be abnormal. Magnetic resonance imaging demonstrated various sites of altered signal around the ventricles and within the brain stem, possibly representing demyelinating plaques. Some of these were scattered around the right trigeminal root entry zone (Fig. 3). A diagnosis of multiple sclerosis was confirmed by alterations found in other neurophysiological tests, including brain-stem auditory evoked responses and somatosensory and visual evoked potentials. Further confirmation was provided by the finding of oligoclonal banding on examination of the cerebrospinal fluid.
Patients With "Essential" Trigeminal Neuralgia. The 68 patients suffering from "essential" trigeminal neuralgia were seen at the Headache and Pain Center of the University of Genoa between October, 1983, and November, 1989. The patients ranged in age between 48 and 83 years (mean 72 years). Pain involved the first division in three cases, the second division in 19 cases, the third division in eight cases, the first and second divisions in 11 cases, the second and third divisions in 26 cases, and all three divisions in one case. Thus, the second division was involved in 57 of the 68 patients. The right side of the face was affected in 39 cases, the left side in 28 cases, and in one case the neuralgia shifted from the right to the left side. The patients had suffered trigeminal pain for an approximate time varying between 1 and 15 years (mean 6 years). None of our patients had undergone any surgical 246
FiG. 3. Magnetic resonance image in Case 3 showing two or three small demyelinating plaques within the pons, around the trigeminal root entry zone on the right, treatment, whereas most had been taking carbamazepine or other drugs for some time. Neurological examination was normal in all cases. Standard radiographic investigation, EEG, and CT were within normal limits in all cases. Results
Patients bl'ith Lesions Case I." Maxillat:v Sinusilis. When trigeminal evoked potential testing was performed for the first time in this patient, a definite abnormality of the W~ component and of the subsequent waves was observed (Fig. 4, lower &['t trace). The latency of the main peak of W~ was slightly delayed: furthermore, the overall duration of this triphasic component was increased, which resulted in further delay of the subsequent components W, and W> These features are particularly evident when the trace is compared to that obtained after stimulation on the healthy side (Fig. 4 upper le[~ trace). These neurophysiological findings suggested defective conduction before or possibly at the point of origin of W,: that is, before or at the entrance of the maxilla~' nerve into the gasserian ganglion. The second session of trigeminal evoked potential testing, performed when the patient was clinically improved, showed a normal response, suggesting a good recovery, of trigeminal nerve function (Fig. 4 right traces). Case 2." Gianl Aneurvsm of the Carotid Siphon. In this patient stimulation of the right infraorbital nerve gave rise to an abnormal response, with the two components W_, and W~ missing, whereas W~ was normal (Fig. 5). This suggested a lesion of the trigeminal pathway located at some point between the sites of origin of W~ and We: that is, between the gasserian ganglion and the proximal part of the retrogasserian root. The J. Neurosurg. / f~blume 75/August, 1991
Trigeminal evoked potentials
FIG. 4. Trigeminal evoked responses from Case 1. Le& Trigeminal evoked potential recordings from the healthy (upper trace) and affected (lower trace) sides before treatment was started. Stimulation of the affected side gives rise to an abnormally delayed W~ peak (latency calculated at the main positive peak, marked by an asterisk). Arrows mark the two negative peaks delimiting the Wt component and so define its duration. The duration of W~ on the affected side is much longer than on the healthy side. As a result of the W~ abnormalities, the Wz and W3 components also show longer latencies, Right: Responses obtained 8 weeks after therapy began. The upper trace was obtained after stimulation of the healthy side and the lower trace after stimulation of the affected side. No significant differences are now evident. Calibrations: I msec/division for x axis and 0.75 uV/division for y axis. Negative upward. Although the analysis time was l0 msec, in this and the following figures only the first, relevant, part of the trace is shown.
FIG. 5. Trigeminal evoked potentials from Case 2. Calibrations: I msec/division for x axis and 0.75 uV/division for y axis. Negative upward. Upper Trace: Normal response from the healthy side. Lower Trace. Response after stimulation of the af('ected side. The W2 and W~ components are missing, while W~ is within normal limits.
trigeminal evoked potentials obtained on the left side were of course normal.
Case 3." Demyelinating Plaques of the Pons. In this patient stimulation of the right infraorbital nerve evoked an abnormal response, where the W~ component was missing (Fig. 6). The first two components, W~ and W2, were within normal limits. These findings suggested a lesion of the trigeminal pathway between the points of origin of W: and W~: that is, between the entrance of the trigeminal root into the ports and the trigeminal nuclei. Patients I4"7th "Essential" Trigeminal Neuralgia The trigeminal evoked potentials were found altered in 33 of the 68 patients suffering from "essential" trigeminal neuralgia. In particular, the W3 component was missing in seven cases, and the asymmetry of the W~-W3 interval exceeded the upper limits of normality on the affected side in 26 patients. Ten of the patients ,1. Neurosurg. / Vohtrne 75/Augtc~t, 1991
FIG. 6. Trigeminal evoked potentials from Case 3. Calibrations: I msec/division for x axis and 1 uV/division for y axis. Negative upward. Upper Trace: Response after stimulation of the healthy side. Lower Trace: Response after stimulation of the affected side. The W3 component is missing, while W~ and W2 do not differ significantly from normal.
247
M. Leandri and E. Favale
FIG. 7. Trigeminal evoked responses in some representativecases of "essential" trigeminal neuralgia. The three sets are obtained from Cases 617 (left), 603 (center), and 5 t7 (right). Calibrations: 1 msec/divisionfor x axis and 1uV/division for y axis. Negativeupward. Upper Traces: Responsesafter stimulation of the healthy side. Lower Traces: Responses after stimulation of the pain-affectedside showinga delay of W3 (left), delay of both W2 and W~ (center), and absence of W3 (right).
with missing or altered W.~ peaks also had abnormal asymmetry of the W~-W2 interval. One patient, who had a prolonged W~-W3 interval but no significant sideto-side asymmetry, was also considered abnormal. Examples of the alterations seen in this group are shown in Fig. 7. A complete summary of the altered neurophysiological findings in the 33 patients is presented in Table 1. Painful involvement of the first and second divisions was seen in seven of these cases, the second and third divisions in 15, the second division alone in eight, the first division alone in one, and the third division alone in two. Overall, the second division was involved in 30 cases. We could not detect any relationship between the degree of alteration and the patient's age, duration of illness, or pain intensity. The percentage of patients with painful involvement of the second division was higher in the group with altered responses (30 of 33 cases, or 90.9%) than in the group with normal responses (27 of 35 cases, or 77.14%). No other clinical differences were noted between the two groups. Discussion
Historical Perspective A brief review of the neurophysiological methods used in the past to investigate the trigeminal nerve seems appropriate in order to clarify the reasons that prompted us to use a new technique in our research. The classic method of studying this nerve was the blink reflex. Unfortunately, the blink reflex has a high degree of variability: the latency of its first component, R~, ranges from approximately 8 to 14 msecfl and the 248
related side-to-side asymmetry is considered abnormal only if it exceeds 1.2 to 1.5 msec. ~j~9 Slight damage to the trigeminal nerve would probably cause a delay on the reflex well within its normal range (even when compared with the healthy side), and may therefore be undetectable. Furthermore, like other reflexes, the blink has a good biological safety factor, so that the loss or damage of a few afferent fibers is compensated at the reflex center, and the latency of the elicited muscle contraction is unaffected. Similarly, other reflexes of the trigeminal nerve (the jaw jerk, the corneal reflex, and the trigeminocervical reflex) share the drawbacks of the blink, and also have not been found consistently altered in diseases with minor damage to the nerve, such as trigeminal neuralgia. Higher precision and sensitivity are offered, at least in principle, by evoked potential measurement, which is now routinely used for assessment of the somatosensory pathways from the upper and lower limbs. A number of authors have also used this method to study trigeminal afferents,23s-~223but the results obtained are very much open to criticism because of the heavy contamination of responses by muscular activity. ~2.t3.~ It is only recently that reliable evoked potentials could be obtained by electrically stimulating the peripheral branches of the trigeminal nerve within their foramina. L~.~a.~v For this study of trigeminal evoked potentials in a clinical setting, we electrically stimulated the infraorbital nerve because, of the three peripheral trigeminal branches, this has been the most extensively studied with published normative data. ~ In addition to the first three waves (W~, W2, and W3), stimulation of this nerve evokes further waves at slightly later latencies, called J. Neurosurg. / Volume 75/August, 1991
Trigeminal evoked potentials TABLE 1 Latencies (msecj o[altered responses in 33 0/68 CO.'(eS ()[' "essential" trigeminal neuralgia* Hospital File No.
Right Side W~ W2 W~
Wa
Left Side W, W3
15 0.97 1.93 2.53 0.95 1.93 2.78+ 24 1.00 1.87 2.35 0.94 1.89 2.71+ 28 0.88 1.82 2.58 0.80 1.95~ 2.74t 30 1.10 2.20 -0.98 2.10 2.90 32 0.81 [.84 2.65? 0.98 1.95 2.52 47 1.01 1.95 2.55 1.00 2.145- 2.76? 62 0,82 1.80 2.65 0.84 1.96"t 2.85?1 . 90 0.83 1.92 2.50 0.86 1.92 2.72? 123 0.85 1.907 2.65? 0.86 1.77 2.45 126 0.95 2.09 2.99 0.95 2.15 3.317 134 0.90 1.89 2.79 1.03 2.22? 3.78? 154 1.09 1.91 2.48 1.01 1.92 2.65? 164 0.98 1.88 2.89 0.95 2.00? 3.20? 182 0.95 1.85 2.56 0.91 1.91 3.04+ 184 1.07 2.15 3.53? 0.97 2.07 3.29 207 1.06 2.33? 3.125- 1.01 2.05 2.81 208 1.03 2.00 2.66 0.87 1.79 2.72? 219 0.81 1.87 2.67 0.84 2.03 -251 0.87 1.78 2.69? 0.96 1.80 2.60 316 0.90 1.93 3.09 0.90 1.92 3.30? 332 0.81 1.68 2.33 0.95 1.75 2.65? 432 0.95 1.88 -1.10 2.05 2.85 436 0.92 2,00 3.00? 0.87 1.85 2.58 453 1.03 1,93 2.87? 0.88 1,78 2.40 456 0.77 1,75 3.09? 0.80 1.77 2.59 457 1.09 2.19 3.40? 0.90 1.99 2.68 505 0.82 1,72 -0.91 1.69 2.40 517 0.96 2.09"I" -0.93 1.92 2.67 519 0.95 2,00 -0.90 1.95 2.69 566 1.08 2.02 3.88 1.04 2.04 -578 0.90 1.957 2.66? 1.10 1.90 2.44 603 0.81 1.787 3.187 0.82 1.62 2.50 617 0.88 1.90 2.76"I" 0.90 1.81 2,53 * All subjectshad both sidesinvestigated,as side-to-sideasymmetry, was considered essential to define the abnormality of a response. -= missingcomponent. t Delayedcomponents:delay was calculatedon the basis of interpeak intervalsand their side-to-sideasymmetry.
P4, N4, P6, and NT; '~ however, they are not as constant as the first three waves, nor has their origin been as precisely demonstrated (although there is little doubt that they arise from neural structures). It is worth adding that, at latencies longer than 8 to 10 msec, it is not possible to record any trustworthy activity from the scalp, at least in the waking subject; the occurrence of reflexes (blink and other reflexes) heavily contaminates the responses, whatever method is used to elicit activity in the trigeminal afferents. '~ Measurement of scalp responses after infraorbital nerve stimulation has been employed in patients with tumors at the base of the skull, allowing precise localization of the damage site along the trigeminal pathway. ~5 Alterations of the response have been found even in cases where no clinical involvement of the trigeminal nerve was evident, a fact suggesting that the method has high sensitivity. The diagnostic power of the trigeminal evoked responses and their site of origin J. Neurosurg. / Volume 75/August. 1991
are confirmed by the findings in some patients after surgical [esioning of the retrogasserian root. '~ C o m m e n t s on Results The first three cases reported here with lesions involving three specific sites of the afferent pathway (that is, the portion peripheral to the gasserian ganglion, the gasserian ganglion itself and/or the retrogasserian root, and the portion inside the brain stem) each gave rise to a definite alteration of the response. This confirmed the origin of the first three waves as demonstrated by intraoperative recordings. ~-" The degree of precision and sensitivity of trigeminal potentials evoked by infraorbital nerve stimulation are comparable to those of the sophisticated technique used to test the conduction velocity of peripheral nerves. Such properties make measurement of the trigeminal evoked potentials ideal for investigating the possible slight damage that may be present in patients suffering from "essential" trigeminal neuralgia. Previous preliminary results from our laboratory, showed that only nine (23.7%) of 38 cases suffering from trigeminal neuralgia had abnormalities of the trigeminal evoked potentials. '~ In the present updated series of 68 patients, we found that the percentage of abnormal findings had risen to 48.5%. This difference is due not only to an increase in the number of cases examined but also to more strict upper limits of normality as defined by a comprehensive study on normative data.'6 The majority (71%) of alterations seen in the present series involved the W3 component, suggesting a lesion at the root entry zone or just after it; only 29% of the alterations seemed to be due to more distal lesions of the trigeminal root. These results are in agreement with the hypothesis put forward by several authors (see the reviews by Adams ~ and Selby2~ that the pathology of the root entry zone is of crucial importance in the occurrence of tic douloureux. This hypothesis has been the object of much controversy, as the only evidence of damage to the root entry zone has been visual inspection during surgery. New MR techniques allow noninvasive examination of that region; ~~ however, few cases have been investigated and the images produced are not always absolutely convincing. Thus, measurement of the trigeminal evoked potentials seems to provide the only available objective evidence of damage of the fifth cranial nerve in trigeminal neuralgia. Clinical and Research Developments Measurement of the trigeminal evoked potentials is open to further improvements, which should enhance its clinical usefulness. In our laboratory,, we have been able to record scalp responses evoked after selective stimulation of the supraorbital and of the mental nerve trunks; 14'7 they have waveforms and latencies similar to those obtained after stimulation of the infraorbital nerve. These methods, when properly developed, will permit a complete examination of the trigeminal divi249
M. Leandri and E. Favale sions. One of the predictable consequences could be an increase of altered findings in trigeminal neuralgia, further confirming the hypothesis of damage to the root entry zone. An interesting development of the technique we have used to stimulate the infraorbital, supraorbital, and mental nerves is that the needles positioned to stimulate the nerve trunks are also in an ideal position to record volleys from nerves if a stimulus is applied to the peripheral branches. We have so far been able to record from the infraorbital nerve trunk the compound action potential evoked by stimulation of the upper lip and gum. ~xBy improving this technique, we hope to provide direct evidence relating to the type of peripheral fibers involved in physiological and pathological processes in man. The identification of fibers responsible for the various trigeminal reflexes or the assessment of damage caused by herpetic neuropathy are examples of possible applications of this technique. So far these issues have been subject to indirect speculation? ~eJ The usefulness of trigeminal evoked potential measurement is not limited to the diagnosis of nerve damage, but extends to the operating theater, as already pointed out. ~2.tS''sExamples of intraoperative monitoring both with orthodromi&' and antidromic 7 stimulation have been provided, so that reliable testing of the trigeminal function could be performed with the patient fully anesthetized. We believe that the reliable neurophysiological investigation of the trigeminal nerve sought by Adams ~in his recent review and by any other researcher interested in the problem, is now within reach of most laboratories.
References I. Adams CBT: Microvascular compression: an alternative view and hypothesis. J Neurosurg 70:1-12, 1989 2. Bennett MH, Jannetta PJ: Trigeminal evoked potentials in humans. Eleetroencephalogr Clin Neurophysiol 48: 517-526, 1980 3. Buettner VW, Petruch F, Scheglmann K, et al: Diagnostic significance of cortical somatosensory evoked potentials following trigeminal nerve stimulation, in Courjon J, Mauguiere F, Revol M (eds): Clinical Applications of Evoked Potentials in Neurology. New York: Raven Press, 1982, pp 339-345 4. Cruccu G, Agostino R, Inghilleri M, et al: The masseter inhibitory reflex is evoked by innocuous stimuli and mediated by A-beta afferent fibres. Exp Brain Res 77: 447-450, 1989 5. Cruccu G, Berardelli A, Manfredi M: Afferents for the human corneal reflex, a Neuroi 234:64, 1987 6. Cruccu G, Bowsher D: Intracranial stimulation of the trigeminal nerve in man. II. Reflex responses, a Neurol Neurosnrg Psychiatry 49:419-427, 1986 7. Cruccu G, Inghilleri M, Manfredi M, et al: Intracranial stimulation of the trigeminal nerve in man. lII. Sensory potentials. J Neurol Neurosurg Psychiatry 50: 1323-1330, 1987
250
8. Drechsler F: Short and long latency cortical potentials following trigemina[ nerve stimulation in man, in Barber C (ed): Evoked Potentials. Baltimore: University Press, 1980. pp 415-422 9. Goodgold J, Eberstein A: Electrodiagnosis of Neuromuscular Diseases. Baltimore: Williams & Wilkins, 1983, p 260 10. Hutchins LG, Harnsberger HR, Hardin CW, et al: The radiologic assessment of trigeminal neuropathy. AJR 153: 1275-1282, 1989 11. Kimura J, Powers JM, Van Allen MW: Reflex response of orbicu[aris oculi muscle to supraorbital nerve stimulation. Arch Neurol 21:193-199, 1969 12. Leandri M, Campbell JA: Origin of early waves evoked by infraorbital nerve stimulation in man. Eleetroencephalogr Clin Neurophysiul 65:13-19, 1986 13. Leandri M, Parodi CI, Fevale E: Early evoked potentials detected from the scalp of man following infraorbital nerve stimulation. Eleetroencephalogr Clin Neurophysiol 62:99-107. 1985 14. Leandri M, Parodi CI, Favale E: Early scalp responses evoked by stimulation of the supraorbital nerve in man. Electroencephalogr Clin Neurophysiol 74:367-377, 1989 [ 5. Leandri M, Parodi CI, Favale E: Early trigeminal evoked potentials in tumours of the base of the skull and trigeminal neuralgia. Electroencephalogr Clin Neurophysiol 71: 114-124, 1988 [6. Leandri M. Parodi CI, Favale E: Normative data on scalp responses evoked by infraorbital nerve stimulation. Electroencephalogr Clin Neurophysiol 71:415-421, 1988 17. Leandri M, Parodi CI, Rigardo S, et al: Early scalp responses evoked by stimulation of the mental nerve in humans. Neurology 40:315-320, 1990 18. Leandri M, Parodi CI, Zattoni J, et al: Subcovtical and cortical responses following infraorbital nerve stimulation in man. Electroencephalogr Clin Neuropathol 66: 253-262, 1987 [9. Ongerboer de Visser BW, Goor C: Electromyographic and reflex study in idiopathic and symptomatic trigeminal neuralgias: latency of the jaw and blink reflexes. J Neurol Neurosurg Psychiatry 37:1225-1230, 1974 20. Selby G: Diseases of the fifth cranial nerve, in Dick PJ, Thomas PK, Lambert EH, et al (eds): Peripheral Neuropathy, ed 2. Philadelphia: WB Saunders, 1984, Vol 2, 1224-1265 21. Shahani B: The human blink reflex. J Neurol Neurosurg Psychiatry 33:792-800, 1970 22. Singh N, Sachdev KK, Brisman R: Trigeminal nerve stimulation: short latency somatosensory evoked potentials. Neurology 32:97-10l, 1982 23. Stohr M, Petruch F: Somatosensory evoked potentials following stimulation of the trigeminaI nerve in man. J Neurol 220:95-98, 1979 24. Tash RR, Sze G, Leslie DR: Trigeminal neuralgia: MR imaging features. Radiology 172:767-770. 1989 Manuscript received June 28, 1990. Accepted in final form December 26, 1990. This paper is dedicated to Prof. Carlo Loeb of Genoa, Italy, on the occasion of his retirement. Address reprint requests to." Massimo Leandri, M.D., Department of Neurology, University of Genoa, Via De Toni 5, [6132 Genoa, Italy.
,L Neurosurg. / Volume 75/August, 199I
