- Email: [email protected]
Somatosensory evoked potentials and magnetic resonance imaging in intraspinal neoplasms
Electroencephalography and clinical Neurophysiology, 1990, 77: 101-111
101
Elsevier Scientific Publishers Ireland, Ltd.
EVOPOT 89096
Somatosensory evoked potentials and magnetic resonance imaging in intraspinal neoplasms 1 Bahman Jabbari, Carl Geyer *, Margaret Schlatter, Barbara Scherokman, Michael Mitchell, John W. McBurney, Celia Elbrecht and Carl H. Gunderson Neurology Service and * Radiology Department, Walter Reed Army Medical Center, Washington, DC (U.S.A.), and Department of Neurology, Uniformed Services University of the Health Sciences, Bethesda, MD ( l~ S.A.) (Accepted for publication: 17 July 1989)
Summary
Median and posterior tibial somatosensory evoked potentials (SEPs) were studied on 25 patients with pathologically proven intraspinal neoplasms, and the results were compared and correlated with the details of clinical examination and the information derived from magnetic resonance imaging (MR). M R was abnormal in all cases and in 23 of 25 (92%) demonstrated an intraspinal expansile lesion. SEP was abnormal in 19 of 25 patients (76%). Abnormal SEPs were found in 18 of 19 patients (94%) with cervical or thoracic neoplasms but only in 1 of 6 patients (16%) with the tumor in the thoracolumbar or lumbar region. SEP-MR correlation was significant ( P < 0.05) for thoracic intraspinal neoplasms where all 9 had an abnormal SEP showing a similar pattern of normal median and abnormal posterior tibial study. Clinically, all 7 patients with posterior column sensory deficits had abnormal SEP (100%). Abnormal SEPs were seen in 7 of 11 (63%) patients with spinothalamic deficits and in 4 of 8 (50%) of those with normal sensory examinations. Four of 9 patients (44%) with a normal neurological examination or an examination disclosing ambiguous results indistinguishable from a peripheral pathology had an abnormal SEP strongly suggesting a central sensory disorder. Comparison of preoperative and postoperative SEPs did not disclose useful prognostic information pertaining to the functional recovery. Key words:: Somatosensory evoked potentials; Magnetic resonance imaging; Intraspinal neoplasm
Spinal cord tumors account for 15% of central nervous system neoplasms (Sloof et al. 1984). An incidence of 1.3/100,000 of population has been estimated for primary spinal cord tumors compared to 6-7/100,000 for brain tumors (Percy et al. 1972). Approximately 90% of spinal cord tumors occur in adults. The symptoms of spinal cord tumors are variable but the initial complaints
1 The opinions expressed in this paper are the private views of the authors and should not be construed as the views of the Department of Defense of the U.S. Army, or the Uniformed Services University of the Health Sciences.
Correspondence to: Dr. B. Jabbari, Neurology Service, Walter Reed Army Medical Center, P.O. Box 310, Washington, DC 20307-5001 (U.S.A.).
may be vague, sometimes limited to non-localizing pain (Schliack and Stille 1975). In such cases with no clear indication for invasive procedures, noninvasive tests may provide invaluable information as to the altered spinal cord anatomy or physiology. Somatosensory evoked potentials (SEPs) have been shown to be sensitive to a variety of intraspinal lesions (Giblin, 1960; Halliday and Wakefield 1964; Norl and Desmedt 1980; Chiappa 1983; Lueders et al. 1983; Yu and Jones 1985; Eisen 1986; Seyal and Gabor 1987). SEP abnormalities were previously described in small groups of patients with intraspinal tumors without correlation with magnetic resonance imaging (MR) (Noel and Desmedt 1980; Riffel et al. 1984; Magui~re et al. 1985). We now report SEP abnormalities in 25
0168-5597,/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland, Ltd.
102
B. JABBARI ET AL.
pathologically proven cases of intraspinal tumors and correlate the findings with the details of clinical examination and that of magnetic resonance imaging.
mentation rate, FANA, ACE, RF and serum B12 measurement were obtained in several patients. CSF was tested for cell count, protein, VDRL, glucose and culture for all patients either initially or at the time of myelography. Several patients had further testing for gamma-globulin and basic myelin protein. Tissue diagnosis was obtained in all patients. The post-surgical follow-up periods varied from 6 to 48 months (average 21 months).
Subjects and methods The study group consisted of 25 patients with intraspinal tumors. Patients were selected from admissions to the neurology and neurosurgery wards and neurology consult service of Walter Reed Army Medical Center between July 1983 and January 1989. Clinically the patient either had a radiculopathy, or a myelopathy or both. Three patients had a normal neurological examination. There were 19 males and 6 females. Their ages ranged from 9 to 69 years (mean 31 years). All patients had routine blood work including CBC, VDRL, electrolytes and assessment of renal and hepatic function. Additional tests such as sedi-
N2o
MR and radiological methods All patients underwent M R study of the spine. MR examinations were performed on 2 scanners 0.5 and 1.5 Tesla with no significant difference in image resolution. The first 21 patients had a nonenhanced MR and the last 4 had a gadolinium study. Sagittal T1 and T2 weighed spin-echo sequences with T R / T E intervals of 600/30 and 2 0 0 0 / 4 0 / 8 0 were obtained. The slice thickness was 5 /tm with a 1.5 # m gap. Axial T1 weighed spin-echo sequences with T R / T E interval of
2.0
7.0
Fz.Cc
I C4-C3 2.0 7.0
Al-Cc
I Fz'Cz' P/N13
~
t
7.0
N22
PF
2.0
2.0
7.0
Fz'Erb
I IC-PF
0 8 16 24 32 40 48 56 64 72 80
G10 10 20 30 40 50 60 70 80 90 100
Fig. 1. Normal (left) median and posterior tibial study in a control subject.
SEPs AND MR IN INTRASPINAL NEOPLASMS
600/30 and 5/~m slice thickness were performed at the level of the lesion. Surface coils were utilized in such a fashion to cover the entire spine in 22 patients
Evoked potential methods Median and posterior tibial SEPs were obtained in all patients. Two patients had additional ulnar SEPs. m square wave constant current electrical pulse of 200/zsec duration was used for SEP testing. The recording was performed in a quiet and semi-dark room and the patients were encouraged to relax. N o sedation was used. The median nerve was stimulated at the wrist at a rate of 5.1/sec and the posterior tibial nerve was stimulated at the ankle at a rate of 3.1/sec with an intensity to produce a moderate twitch in the corresponding muscles. Responses from the median nerve stimulation were recorded at ipsilateral Erb's point, C6 cervical level, and contralateral somatosensory region of the scalp (Cz) with Fz and ear reference (Fig. 1A). Cz was designated as a point 2 cm posterior to the C3 and C4 locations of the international 10-20 system. The responses obtained from posterior tibial stimulation were recorded at L 1 - T 1 2 level referred to iliac crest and at scalp Cz' (2 cm behind Cz) referred to Fz and between C3-C4 scalp electrodes (Fig. 1B). The last 14 patients of this study had the peripheral response over the popliteal fossa (PF) recorded additionally. A filter range of TABLE I Controls. Mean age: 36 years; median: 80 subjects; post-tibial: 30 subjects. Measure Erb-N20 (RT) Erb-N20 (LT) Erb-N20 (side difference) P / N 13-N 20 (RT) P/N13-N20 (LT) P/NI3-N20 (side difference) N20/P25 amp (RT) N20/P25 amp (LT) N20/P25 amp (side difference) N22-P40 (RT) N22-P40 (LT) N22-P40 (side difference)
Mean
S.D.
9.42 9.43 0.42 6.21 6.06 0.54 1.77 1.82 0.46
0.62 0.71 0.36 0.66 0.72 0.51 0.93 1.11 0.38
16.5 16.6 0.04
1.3 1.4 1.0
103
10-3000 was used for both median and posterior tibial study. In each patient 2 trials of 1000 averaged responses were obtained. The data were compared with that of 80 normal asymptomatic volunteers for median and 30 individuals for posterior tibial study currently used as control data in our laboratory (Table I). The values obtained from the posterior tibial stimulation were corrected for the patient's height. The following criteria were used as SEP abnormality: (1) absence of the major central components of median or posterior tibial study ( P / N I 3 , N 2 0 / P 2 5 , P40). (2) Increased central conduction times beyond 3 S.D. either unilateral, bilateral or side-to-side differences (Table I, controis).
Results Table II demonstrates the results of SEP and M R studies and the type of the lesions. M R was abnormal in all patients and demonstrated evidence of a focal intraspinal neoplasm in 23 of 25 patients (92%). In 1 patient (patient 16, Table II), M R showed a syrinx at C 2 - C 7 level but no evidence of a tumor. At surgery, a small tumor nodule was found at C6 level which was pathologically verified as hemangioblastoma. In the other patient M R revealed evidence of syrinx and cord atrophy at cervical area. A small locus of low grade astrocytoma was found at surgery in the wall of the syrinx (patient 15, Table II). Both of these patients had a non-enhanced M R study. SEP was abnormal in 19 of 25 patients (76%). However, 5 of the 6 patients with a normal SEP (83%) had a lesion at the level of conus medullaris. SEP abnormalities could be divided either on the basis of localization or pattern. The localizationrelated abnormalities consisted of 3 types: (1) Median nerve study abnormal, posterior tibial normal: this group consisted of 3 patients (patients 3, 11 and 17, Table II). All 3 patients had a high cervical cord neoplasm (Fig. 2). (2) Median study normal, posterior tibial abnormal: there were 12 patients in this group. Nine patients had a thoracic tumor. Two patients had a cervico-thoracic neoplasm which did not
104
B. JABBARI ET AL
T A B L E II SEP, M R and pathology. M, median; U, ulnar; PT, post-tibial; CCT, central conduction time. Age/Sex
SEP finding
M R localization
Pathology
l
20/M
30/M
Increased T2 signal, intradural at C7 to T 3 - T 4 disk spaces Increased T2 signal, rt side at C6 level
Lipoma
2
3
67/F
4
25/M
5
26/M
6
48/M
7
30/M
8
46/M
9
26/F
10
l l/M
11
9/M
12
14/M
13
12/M
14
69/F
15
15/M
16
30/M
M: normal Pt: absent P40s bilaterally M: normal Pt: increased C C T rt stimulation M: absent N / P 1 3 , rt side study PT: normal M: normal PT: absence of N22 and P40s bilaterally M: normal PT: C C T It > rt > 3 S.D. rt P40 low amplitude M: normal PT: C C T It > rt > 3 S.D. rt P40 low amplitude M and U: normal PT: C C T It > rt > 3 S.D. rt P40 low amplitude Median: normal PT: C C T rt > It > 3 S.D. P40 ampl low, It side Median: normal PT: absent P40s bilaterally M and U: normal PT: absent P40s bilaterally M: no P / N 1 3 and N20 potentials on both sides. PT: normal M and PT: increased C C T bilaterally N o P / N 1 3 s both sides M: rt, no P / N 1 3 and N20 PT: RT, increased C C T It, no P40 M: normal PT: absence of LP and P40s bilateral M: normal PT: C C T It > rt > 3 S.D. M and PT: normal
17
36/M
Enlarged upper cord and medulla with high signal intensity (T2) and syrinx
18
21/M
M: P / N 1 3 and N20 potentials absent bilaterally PT: normal M and PT: normal
Low signal, intradural, extramedullary mass at C 1 - C 2 level Low intensity signal, intradural-extramedullary at T 1 0 - T l l level Low intensity signal, intradural-extramedullary at T10 level High intensity T2 signal intradural-extramedullary at T10 level High intensity T2 signal intradural-extramedullary at T10 level T4 lesion with cord compression Lesion at T12-L1 with flow void characteristics High T2 signal intensity mass T7, with cord compression Enlarged cervical cord and and medulla with large cysts Enlarged cervical cord and lower medulla, cystic
Neurofibroma
Meningioma
Meningioma
Teratoma
Schwannoma
Schwannoma
Metastatic adenocarcarcinoma Angioma
Aneurysmal bone cyst Astrocytoma
Ependymoma
Enlarged upper cord to the pons with cavity formation Mass with low T2 signal at T 7 - T 9
Astrocytoma cystic
Cord atrophy and small syrinx, C 3 - C 6 Small syrinx C 2 - C 7
Astrocytoma
Increased T1, decreased T2 signal mass at T12 level
Meningioma
Syrinx with hemangioblastoma No surgery possible ependymoma Lipoma
105
SEPs AND MR IN INTRASPINALNEOPLASMS TABLE II (continued) Age/Sex
SEP finding
MR localization
Pathology
19
28/M
M and PT: normal
20 21
39/F 22/F
M and P: normal M and P: normal
Myxo-papillary ependymoma Angioma Schwannoma
22
50/M
M: absent P/N13 and N20 potentials bilaterally PT: absent P40s bilaterally M and PT: normal M: normal PT: CCT It > rt > 3 S.D. M: Lt P/N13 a b s e n t PT: normal
Conus mass with increased signal at T12 level Mass LI-L2 Mass with increased T2 signal, Tll T12 level Large cervical syrinx
23 24
30/M 40/M
25
37/M
extend above C6 level (Figs. 3 and 4). One patient had syrinx and tumor at C 3 - C 6 level. (3) Both median and posterior tibial studies were abnormal: there were 4 patients in this group. Three of them had extensive neoplasms involving the cervical cord and the lower brain-stem (patients 12, 13 and 22, Table II). One patient had a cervical cord tumor and syrinx (patient 25). Both median and posterior tibial studies were normal in 6 patients. Five of these patients had a mass at the high lumbar region involving the most distal part of the spinal cord (patients 18-21 and 23, Table II). One patient had a small hemangioblastoma embedded in the wall of a syrinx at the lower cervical region which escaped MR detection as well (patient 16, Table II). Two patterns of central SEP abnormality were noted: (1) Absence of central potentials (N20, P / N 1 3 , P40): this group consisted of 11 patients. Six of these patients had extensive cervico-medullary lesions and 4 patients had thoracic lesions with evidence of cord compression (Figs. 2 and 3) and 1 patient had a thoracolumbar angioma. (2) Increased central conduction time only: there were 8 patients in this group, 7 with unilateral and 1 with bilateral abnormality. The latter patient also demonstrated marked reduction of amplitude and poor reproducibillity of P / N 1 3
Gabolinium MR: tumor module at C2-C3 Mass conus to L3 Widened cord T7-T12 Cervicomedullary syrinx Gadolinium MR: enhancement C6 level
Recurrent Ependymoma Ependymoma Astrocytoma Ependymoma C2-C6
potentials bilaterally (patient 12, Table II). In all 7 patients with unilateral abnormalities the finding consisted of increased N 2 2 - P 4 0 central conduction time obtained in the posterior tibial study. Five of these patients had normal absolute values but side to side difference exceeded 3 S.D. In 4 of these 5 patients the amplitude of P40 potential was also diminished on the side of prolonged CCT (Fig. 4). SEP was abnormal in all 7 patients who clinically demonstrated impairment of the position sense (100%), in 7 of 11 (63%) patients with spinothalamic sensory abnormalities only and in 4 of 8 patients (50%) with no sensory deficits. Follow-up post-surgical repeat SEPs were obtained in 11 patients. Two patients showed an improved pattern which corresponded with clinical improvement. Six patients showed no change and 3 patients demonstrated a deterioration of SEP pattern. N o correlation with the clinical course was noted for the patients with unchanged or worsened SEPs.
Discussion
Early studies of median SEP in humans have shown a close correlation between SEP abnormalities and pathology of posterior columns or lemnis-
106
B. JABBARI ET AL.
A Cc-AI ................... ~ . , ~,,..+..+...+.
,~
. f ~
.... ~
,..~,+,.~,
~
.......~
-
,-+
4.0
4.0
4.0
?
.............
,,
V,.N?/~w~-
........ - ~ ........ ,,"2 ~
4,0
........~.< . ~ > , ±
4.0
C2"Fz
•
'
'
'kL/
'
'"v
-
~r
v~>CJ"-"
'
4.0
'
l
I!/
.0
G1-
L
I .........
0
I .........
8
I .........
16
1 .........
24
I .........
32
I.........
40
I
.........
48
I .........
56
l .........
64
L .........
72
I
80
G1I .........
0
I .........
8
I .........
16
I .........
24
I .........
32
L .........
40
I .........
48
L .........
56
I .........
64
~ .........
72
1
80
Fig. 2. Patient 11. A: median SEP shows absence of P/N13 and N20 components bilaterally - posterior tibial studies are normal. B: on sagittal T1 weighted pulse sequence there is a large expensile mass extending from medulla, to upper thoracic region. There are several cavities of decreased signal, primarily limited to the medulla.
cal system, m a n i f e s t e d clinically as proprioceptive deficit ( G i b l i n 1960; H a l l i d a y a n d Wakefield 1964). In addition, recent studies d e m o n s t r a t e d that SEP can provide evidence for subclinical dis-
r u p t i o n of the central sensory pathways. SEP a b n o r m a l i t i e s have b e e n reported in n o n - s e n s o r y focal b r a i n disorders ( M o i l et al. 1984), in C N S lesions w i t h o u t p r o p r i o c e p t i o n deficits (Pedersen
SEPs A N D M R IN I N T R A S P I N A L N E O P L A S M S
107
2.C
C4-C3
N22 IC-L1
IC -PF
J
± G1. t
1~
0
......
i .........
10
I .........
20
I .........
30
I .........
40
I .........
50
I .........
60
I .........
70
I .........
80
I .........
90
I
100
Fig. 3. Patient 1 (Table II). A: left posterior tibial SEP study shows well formed popliteal fossa and caudal cord potential (N22) but P40 is absent. This patient's median SEP is normal. B: there is an expensile mass from the level of C7 to T3 which is intradural with increased signal on sagittal T1 weighted image. The pathology proved to be lipoma.
and Trojaborg, 1984), and in bilateral brain disorders known not to produce sensory symptoms (Bollen et al. 1985). Although SEPs have been described before in patients with intraspinal neoplasms (NoEl and Desmedt 1980; Riffel et al. 1984; Mauguirre et al. 1985), there has been no systematic study of this subject correlating SEP data with details of clinical and MR findings. Cracco (1973) described absence of potentials rostral to a neoplastic cord lesion in some patients with clinical evidence of advanced disease. NoEl and Desmedt (1980) reported 7 patients with a cervical myelopathy due to tumors and cervical spondylosis. These patients had a normal median study, but lower limb stimulation (sural nerve) disclosed abnormal results. The authors stated that such an SEP pattern in the presence of a cervical myelopathy should suggest its compressive nature. In another study of 6 patients with spinal cord tumors (Riffel et al.
1984), increased central conduction times in posterior tibial SEPs were observed in 3 patients. The authors, however, considered decreased amplitude ratio of P 4 0 / N 2 2 ( < 0.47) a more powerful predictor of abnormality as it was noted in the majority of their patients. Maugui~re et al. (1985) carefully studied median and posterior tibial SEPs in 20 patients with intraspinal neoplasms. Not all patients had both studies and some of them had the test postoperatively. M R was not performed (or reported) on these patients. They have found abnormal SEPs in several patients with normal sensory examination and believed this 'electroclinical dissociation' to be a pattern of extramedullary neoplasms. The majority of their patients with cervical neoplasms had normal median SEPs. SEP-clinical correlations
Several relevant clinical questions might be asked from the data accumulated in our study:
"etUOUUeA~qas~ sen~ s!q.L "alSS) JO ueql .l;~le~)J~ so!lsuolaeaeq3 IgU~!S seq ss-etu onss!l ljos aql 'a3uznbas aslnd p a l t t ~ ! ~ E l Ieu°ao3 uo :~) "slooa aalau ~u!punoJJns ao pao3 lV.u!ds aql jo sags!aalaeaeq~ ieu~!s seq uo.tsaI s!ql aouznbas aslnd paltl~!anx [ I |euoao3 u O "slooa anaau aql Jo ~sano3 ~ql ~UOle ~u!puzlx~ 'ssetu anss!l 1jos paq!aastunaa!3 ila~ e s! uo!~za 3!3eaoql aaddn pue lea!Aaa3 aZaOI ltl~!a ~ql U! :fl "IetUJOU S! d3S ug!pzlAI "017d JO ~pnl!ldtue u! zseaa3ap pa~Iaetu pug (01rd-EEN) atu!l uo!lanpuo~ leJlUa3 pase~azu! SA~OqSd~JS le!q!l ao!Jalsod lq~!a :V "(II aIqeZ) Z lua!led "P "~.tq
OOI.
P .....
06 I ........
05 T.........
0/. I .........
09 T .........
OG T.........
0IF I .........
Og I .........
OZ I .........
01.
0
OOL
OI
oe
OL
N
U
Ot
O[
e~
O~
e
[ .........
-I.O
-L9 L'I'S
O'l
0"|
Ll-rl
0"|
o'1 tNI
|'it
e'l
0"l
O~
"3V i ~ I'SV~flVf "fl
V 80[
SEPs A N D MR IN INTRASPINAL NEOPLASMS
(1) Are the SEPs a good screening test for patients who have a normal neurological examination but who may, nonetheless, have spinal tumors? One of 3 such patients (nos. 3, 21 and 23) in our study had an abnormal SEP, strongly suggesting a central pathology. (2) Do SEPs help to establish the central origin of the symptoms when the neurological examination could be consistent with a purely peripheral pathology? In our study 3 of 6 patients (nos. 7, 16, 18, 19, 22, 25) with such ambiguous examination had cenlrally abnormal SEPs. (3) Do SEPs help establ;.sh the bilaterality of involvement and thus suggest (but not prove) a spinal cord localization with symptoms and signs confined to one-half of the body and sparing the face? Two of four (50%, nos. 13, 15, 18, 20) fit this description. (4) Can particular SEP findings or pattern of findings be used either to suggest or to exclude a spinal cord tumor in a patient who clinically presents with a myelopathy? No - the observed patterns of SEP abnormality were non-specific for tumors. (5) When neurological examination is abnormal and strongly suggestive of a central pathology, can SEP predict the location of the tumor more accurately than the neurological examination and thus allow a more restricted radiologic evaluation? Overall, clinical examination of such patients was more accurate than SEP localization. However, SEP localization was very closely correlated with clinical examination in patients with findings confined to the legs due to thoracic neoplasms. Moreover, as described above 4 of 9 patients with either a normal or ambiguous examination had a centrally abnormal SEP study. S E P - M R correlations and lesion localization Magnetic resonance imaging is now the non-invasive procedure of choice for intraspinal neoplasms (Williams et al. 1987). We found good correlation between SEP and M R for cervical and thoracic neoplasms as SEP was abnormal in all 17 patients of this category in whom M R disclosed a mass lesion. Median SEP was abnormal (with or without posterior tibial abnormality) in 7 of 8 patients with M R evidence of cervical tumor above
109
the C6 level whereas in the group with neoplasms below C6 level, all 9 patients disclosed an abnormal posterior tibial SEP and a normal median study. However, only 1 of 6 patients with cauda equinaconus medullaris syndrome demonstrated an abnormal SEP. Our low yield in this area may be due to our method of recording which did not include lumbosacral recording sites. Using such a montage Lehmkuhl et al. (1988) found abnormal SEPs in all 8 patients with trauma to cauda equina or conus medullaris region. Two of our patients with a small nodule of neoplasm in the wall of a cervical syrinx had normal median SEPs but M R also failed to detect the neoplastic lesion (nos. 15 and 16). It should be noted, however, that both patients had nonenhanced MRs and a vascular lesion like hemangioblastoma would ordinarily produce enhancement. Patient 15 provided exception to NoEl and Desmedt's view that in cervical myelopathy the 'normal median-abnormal posterior tibial' pattern indicates an extradural compressive pathology. Both median and posterior tibial SEPs were markedly abnormal in a patient with a post-surgical cervical syrinx in whom M R with gadolinium injection disclosed recurrence of tumor. In our series, the high yield of median SEP in cervical neoplasms contrasts with the low yield reported by other investigators (Maugui~re et al. 1985; Lehmkuhl et al. 1988). We believe this is due to the fact that the majority of our patients with cervical neoplasms had extensive intramedullary lesions which are more likely to produce abnormalities of median SEPs. On the other hand, the yield of medium SEP in intraspinal cervical neoplasms may be further improved by modification and extension of the current SEP recording techniques. Using a circular array of electrodes around the neck, Emerson and Pedley (1986) reported a dissociation between N 13 (posterior neck) and P13 (anterior neck) potentials in several patients with intramedullary lesions in whom conventional median nerve studies showed no abnormality. Other authors have found ulnar SEPs more informative than median or posterior tibial SEPs in patients with compressive cervical myelopathies (Veilleux and Daube 1987). Only 2 of our patients had the ulnar study. However,
110 u l n a r a n d m e d i a n S E P s w e r e n o r m a l in 1 p a t i e n t w i t h a C 6 - T 1 s c h w a n n o m a ( p a t i e n t 7, T a b l e II). P o s t o p e r a t i v e i m p r o v e m e n t of S E P c o r r e l a t e d w i t h c l i n i c a l i m p r o v e m e n t in 2 p a t i e n t s . T h e i m p o r t a n c e o f p o s t o p e r a t i v e S E P s in p r e d i c t i n g f u n c t i o n a l r e c o v e r y is c o n f o u n d e d b y several v a r i a b l e s , the m o s t i m p o r t a n t p o s s i b l y b e i n g t u m o r type. The assessment of the predictive value of postope r a t i v e S E P in i n d i c a t i n g f u n c t i o n a l r e c o v e r y a w a i t s a l a r g e r s t u d y c o n t r o l l e d for this m a j o r variable. I n c o n c l u s i o n , o u r d a t a i n d i c a t e t h a t S E P is a s e n s i t i v e m e a s u r e o f s p i n a l c o r d i n v o l v e m e n t by i n t r a s p i n a l t u m o r s at c e r v i c a l o r t h o r a c i c levels. M R is the p r o c e d u r e of c h o i c e for d e t e c t i n g s p i n a l c o r d n e o p l a s m s . H o w e v e r , M R is s e l d o m i n d i c a t e d in p a t i e n t s w i t h v a g u e c o m p l a i n t s , n o n l o c a l i z i n g signs o r n o r m a l e x a m i n a t i o n s . I n o u r s t u d y , 44% o f s u c h p a t i e n t s d e m o n s t r a t e d a c e n trally a b n o r m a l S E P s t u d y w h i c h c l e a r l y d i s c l o s e d an i n d i c a t i o n for M R of t h e spine. S E P m a y p r o v e to b e e v e n m o r e u s e f u l in c h i l d r e n w i t h i n t r a s p i n a l t u m o r s in w h o m t h e s y m p t o m s are o f t e n v a g u e a n d S E P a b n o r m a l i t i e s a r e u n l i k e l y to r e p r e s e n t m u l t i p l e sclerosis o r d e g e n e r a t i v e s p i n e disease. This study was performed by the W.R.A.M.C. Department of Clinical Investigation, Project No. 7132. The authors wish to thank Mrs. Michelle Welton, Miss Diane Bailey and Mrs. Ferne Robinson for assistance in the preparation of this manuscript and S.F.C. Howard Mays for performing SEPs.
References Bollen, E.L., Arts, R.J., Roose, R.A.C., Van der Velde, E.A. and Buruma, O.J.S. Somatosensory evoked potentials in Huntington's chorea. Electroenceph. clin. Neurophysiol., 1985, 62: 235-240. Chiappa, K.H. Evoked Potentials in Clinical Medicine. Raven Press, New York, 1983: 203-312. Cracco, R.Q. Spinal evoked response: peripheral nerve stimulation in man. Electroenceph. clin. Neurophysiol., 1973, 35: 379-386. Desmedt, J.E. and Cheron, G. Prevertebral (oesophageal) recording of subcortical somatosensory evoked potentials in man. Electroenceph. clin. Neurophysiol., 1981, 52: 257-275. Eisen, A.A. Non-invasive measurement of spinal cord conduc-
B. JABBARI ET AL. tion: review of the presently available methods. Muscle Nerve, 1986, 9: 95-103. Emerson, R.G. and Pedley, T.A. Effect of cervical spinal cord lesions on early components of the median nerve somatosensory evoked potentials. Neurology, 1986, 36: 20-26. Giblin, D.R. The effects of the lesions of the nervous system on cerebral responses after peripheral nerve stimulation. Electroenceph. clin. Neurophysiol., 1960, 12: 262. Halliday, A.M. and Wakefield, G.S. Cerebral evoked potentials in patients with dissociated sensory loss. J. Neurol. Neurosurg. Psychiat., 1964, 26: 21-28. Lehmkuhl, D.L., Dimitrijevid, M.R. and Zidar, J. Lumbosacral evoked potentials and cortical somatosensory evoked potentials in patients with lesions of conus medullaris and cauda equina. Electroenceph. clin. Neurophysiol., 1988, 71: 161-169. Lueders, H., Lesser, R., Hahn, J., Little, J. and Kelm, G. Subcortical somatosensory evoked potentials to median nerve stimulation. Brain, 1983, 106: 341-372. Maugui~re, F., lba~ez, V. et Fisher, C. Les potentiels 6voqu6s somesth6siques dans les tumeurs intra-rachidiennes. Rev. EEG Neurophysiol., 1985, 15: 95-106. Mori, E., Yamadori, A., Kmdo, Y. and Tabuchi, M. Ataxic hemiparesis from small capsular hemorrhage. Computed tomography and somatosensory evoked potentials, Arch. Neurol,, 1984, 41: 1050-1053. Mulder, D.W. and Dale, A.J.D. Spinal cord tumors and disks. In: A.B. Baker and R:J. Joynt (Eds.), Clinical Neurology. Harper and Row, Philadelphia, PA, 1975: Ch. 45. No~l, P. and Desmedt, J.E. Cerebral and far-field somatosensory evoked potentials in neurological disorders involving the cervical spinal cord, brain stem thalamus and cortex. In: Progress in Clinical Neurophysiology, Vol. 7. Karger, Basel, 1980: 216-218. Pedersen, L. and Trojaborg, W. Visual, auditory and somatosensory pathway involvement in hereditary cerebellar ataxia, Freidreich's ataxia, and familial spastic paraplegia. Electroenceph, clin. Neurophysiol., 1984, 52: 283-297. Percy, A.K., Elveback, R.K., Okazaki, H. et al. Neoplasms of the central nervous system. Epidemiologic considerations. Neurology, 1972, 22: 40. Riffel, B., St~3hr, M. and K~Srner, S. Spinal and cortical evoked potentials following stimulation of the posterior tibial nerve in the diagnosis and localization of spinal cord diseases. Electroenceph. clin. Neurophysiol., 1984, 58: 400-407. Schliack, H. and Stille, D. Clinical symptomatology of intraspinal tumors. In: P.J. Vinken and G.W. Bruyn (Eds.), Handbook of Clinical Neurology, Vol, 19. North-Holland Publ., Amsterdam, 1975: 23-49. Seyal, M. and Gabor, A.J. Generators of human spinal somatosensory evoked potentials. J. Clin, Neurophysiol., 1987, 4: 177-187. Sloof, J.L., Kernohan, J.W. and McCarthy, C.S. Primary Intra-Medullary Neoplasms of the Spinal Cord and Filum Terminale. Saunders, Philadelphia, PA, 1984.
SEPs AND MR IN INTRASPINAL NEOPLASMS Veilleux, M. and Daube, J.R. The value of ulnar somatosensory evoked potentials in cervical myelopathy. Electroenceph, clin. Neurophysiol., 1987, 68: 415-423. Williams, A.L., Haughton, V.M., Pojunas, K.W., Danials, D.L. and Kilgore, D.P. Differentiation of intra-medullary neo-
111 plasms and cyst by MR. Am. J. Neuroradiol., 1987, 8: 527-532. Yu, Y.L. and Jones, S.J. Somatosensory evoked potentials in cervical spondylosis. Brain, 1985, 108: 273-300.
101
Elsevier Scientific Publishers Ireland, Ltd.
EVOPOT 89096
Somatosensory evoked potentials and magnetic resonance imaging in intraspinal neoplasms 1 Bahman Jabbari, Carl Geyer *, Margaret Schlatter, Barbara Scherokman, Michael Mitchell, John W. McBurney, Celia Elbrecht and Carl H. Gunderson Neurology Service and * Radiology Department, Walter Reed Army Medical Center, Washington, DC (U.S.A.), and Department of Neurology, Uniformed Services University of the Health Sciences, Bethesda, MD ( l~ S.A.) (Accepted for publication: 17 July 1989)
Summary
Median and posterior tibial somatosensory evoked potentials (SEPs) were studied on 25 patients with pathologically proven intraspinal neoplasms, and the results were compared and correlated with the details of clinical examination and the information derived from magnetic resonance imaging (MR). M R was abnormal in all cases and in 23 of 25 (92%) demonstrated an intraspinal expansile lesion. SEP was abnormal in 19 of 25 patients (76%). Abnormal SEPs were found in 18 of 19 patients (94%) with cervical or thoracic neoplasms but only in 1 of 6 patients (16%) with the tumor in the thoracolumbar or lumbar region. SEP-MR correlation was significant ( P < 0.05) for thoracic intraspinal neoplasms where all 9 had an abnormal SEP showing a similar pattern of normal median and abnormal posterior tibial study. Clinically, all 7 patients with posterior column sensory deficits had abnormal SEP (100%). Abnormal SEPs were seen in 7 of 11 (63%) patients with spinothalamic deficits and in 4 of 8 (50%) of those with normal sensory examinations. Four of 9 patients (44%) with a normal neurological examination or an examination disclosing ambiguous results indistinguishable from a peripheral pathology had an abnormal SEP strongly suggesting a central sensory disorder. Comparison of preoperative and postoperative SEPs did not disclose useful prognostic information pertaining to the functional recovery. Key words:: Somatosensory evoked potentials; Magnetic resonance imaging; Intraspinal neoplasm
Spinal cord tumors account for 15% of central nervous system neoplasms (Sloof et al. 1984). An incidence of 1.3/100,000 of population has been estimated for primary spinal cord tumors compared to 6-7/100,000 for brain tumors (Percy et al. 1972). Approximately 90% of spinal cord tumors occur in adults. The symptoms of spinal cord tumors are variable but the initial complaints
1 The opinions expressed in this paper are the private views of the authors and should not be construed as the views of the Department of Defense of the U.S. Army, or the Uniformed Services University of the Health Sciences.
Correspondence to: Dr. B. Jabbari, Neurology Service, Walter Reed Army Medical Center, P.O. Box 310, Washington, DC 20307-5001 (U.S.A.).
may be vague, sometimes limited to non-localizing pain (Schliack and Stille 1975). In such cases with no clear indication for invasive procedures, noninvasive tests may provide invaluable information as to the altered spinal cord anatomy or physiology. Somatosensory evoked potentials (SEPs) have been shown to be sensitive to a variety of intraspinal lesions (Giblin, 1960; Halliday and Wakefield 1964; Norl and Desmedt 1980; Chiappa 1983; Lueders et al. 1983; Yu and Jones 1985; Eisen 1986; Seyal and Gabor 1987). SEP abnormalities were previously described in small groups of patients with intraspinal tumors without correlation with magnetic resonance imaging (MR) (Noel and Desmedt 1980; Riffel et al. 1984; Magui~re et al. 1985). We now report SEP abnormalities in 25
0168-5597,/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland, Ltd.
102
B. JABBARI ET AL.
pathologically proven cases of intraspinal tumors and correlate the findings with the details of clinical examination and that of magnetic resonance imaging.
mentation rate, FANA, ACE, RF and serum B12 measurement were obtained in several patients. CSF was tested for cell count, protein, VDRL, glucose and culture for all patients either initially or at the time of myelography. Several patients had further testing for gamma-globulin and basic myelin protein. Tissue diagnosis was obtained in all patients. The post-surgical follow-up periods varied from 6 to 48 months (average 21 months).
Subjects and methods The study group consisted of 25 patients with intraspinal tumors. Patients were selected from admissions to the neurology and neurosurgery wards and neurology consult service of Walter Reed Army Medical Center between July 1983 and January 1989. Clinically the patient either had a radiculopathy, or a myelopathy or both. Three patients had a normal neurological examination. There were 19 males and 6 females. Their ages ranged from 9 to 69 years (mean 31 years). All patients had routine blood work including CBC, VDRL, electrolytes and assessment of renal and hepatic function. Additional tests such as sedi-
N2o
MR and radiological methods All patients underwent M R study of the spine. MR examinations were performed on 2 scanners 0.5 and 1.5 Tesla with no significant difference in image resolution. The first 21 patients had a nonenhanced MR and the last 4 had a gadolinium study. Sagittal T1 and T2 weighed spin-echo sequences with T R / T E intervals of 600/30 and 2 0 0 0 / 4 0 / 8 0 were obtained. The slice thickness was 5 /tm with a 1.5 # m gap. Axial T1 weighed spin-echo sequences with T R / T E interval of
2.0
7.0
Fz.Cc
I C4-C3 2.0 7.0
Al-Cc
I Fz'Cz' P/N13
~
t
7.0
N22
PF
2.0
2.0
7.0
Fz'Erb
I IC-PF
0 8 16 24 32 40 48 56 64 72 80
G10 10 20 30 40 50 60 70 80 90 100
Fig. 1. Normal (left) median and posterior tibial study in a control subject.
SEPs AND MR IN INTRASPINAL NEOPLASMS
600/30 and 5/~m slice thickness were performed at the level of the lesion. Surface coils were utilized in such a fashion to cover the entire spine in 22 patients
Evoked potential methods Median and posterior tibial SEPs were obtained in all patients. Two patients had additional ulnar SEPs. m square wave constant current electrical pulse of 200/zsec duration was used for SEP testing. The recording was performed in a quiet and semi-dark room and the patients were encouraged to relax. N o sedation was used. The median nerve was stimulated at the wrist at a rate of 5.1/sec and the posterior tibial nerve was stimulated at the ankle at a rate of 3.1/sec with an intensity to produce a moderate twitch in the corresponding muscles. Responses from the median nerve stimulation were recorded at ipsilateral Erb's point, C6 cervical level, and contralateral somatosensory region of the scalp (Cz) with Fz and ear reference (Fig. 1A). Cz was designated as a point 2 cm posterior to the C3 and C4 locations of the international 10-20 system. The responses obtained from posterior tibial stimulation were recorded at L 1 - T 1 2 level referred to iliac crest and at scalp Cz' (2 cm behind Cz) referred to Fz and between C3-C4 scalp electrodes (Fig. 1B). The last 14 patients of this study had the peripheral response over the popliteal fossa (PF) recorded additionally. A filter range of TABLE I Controls. Mean age: 36 years; median: 80 subjects; post-tibial: 30 subjects. Measure Erb-N20 (RT) Erb-N20 (LT) Erb-N20 (side difference) P / N 13-N 20 (RT) P/N13-N20 (LT) P/NI3-N20 (side difference) N20/P25 amp (RT) N20/P25 amp (LT) N20/P25 amp (side difference) N22-P40 (RT) N22-P40 (LT) N22-P40 (side difference)
Mean
S.D.
9.42 9.43 0.42 6.21 6.06 0.54 1.77 1.82 0.46
0.62 0.71 0.36 0.66 0.72 0.51 0.93 1.11 0.38
16.5 16.6 0.04
1.3 1.4 1.0
103
10-3000 was used for both median and posterior tibial study. In each patient 2 trials of 1000 averaged responses were obtained. The data were compared with that of 80 normal asymptomatic volunteers for median and 30 individuals for posterior tibial study currently used as control data in our laboratory (Table I). The values obtained from the posterior tibial stimulation were corrected for the patient's height. The following criteria were used as SEP abnormality: (1) absence of the major central components of median or posterior tibial study ( P / N I 3 , N 2 0 / P 2 5 , P40). (2) Increased central conduction times beyond 3 S.D. either unilateral, bilateral or side-to-side differences (Table I, controis).
Results Table II demonstrates the results of SEP and M R studies and the type of the lesions. M R was abnormal in all patients and demonstrated evidence of a focal intraspinal neoplasm in 23 of 25 patients (92%). In 1 patient (patient 16, Table II), M R showed a syrinx at C 2 - C 7 level but no evidence of a tumor. At surgery, a small tumor nodule was found at C6 level which was pathologically verified as hemangioblastoma. In the other patient M R revealed evidence of syrinx and cord atrophy at cervical area. A small locus of low grade astrocytoma was found at surgery in the wall of the syrinx (patient 15, Table II). Both of these patients had a non-enhanced M R study. SEP was abnormal in 19 of 25 patients (76%). However, 5 of the 6 patients with a normal SEP (83%) had a lesion at the level of conus medullaris. SEP abnormalities could be divided either on the basis of localization or pattern. The localizationrelated abnormalities consisted of 3 types: (1) Median nerve study abnormal, posterior tibial normal: this group consisted of 3 patients (patients 3, 11 and 17, Table II). All 3 patients had a high cervical cord neoplasm (Fig. 2). (2) Median study normal, posterior tibial abnormal: there were 12 patients in this group. Nine patients had a thoracic tumor. Two patients had a cervico-thoracic neoplasm which did not
104
B. JABBARI ET AL
T A B L E II SEP, M R and pathology. M, median; U, ulnar; PT, post-tibial; CCT, central conduction time. Age/Sex
SEP finding
M R localization
Pathology
l
20/M
30/M
Increased T2 signal, intradural at C7 to T 3 - T 4 disk spaces Increased T2 signal, rt side at C6 level
Lipoma
2
3
67/F
4
25/M
5
26/M
6
48/M
7
30/M
8
46/M
9
26/F
10
l l/M
11
9/M
12
14/M
13
12/M
14
69/F
15
15/M
16
30/M
M: normal Pt: absent P40s bilaterally M: normal Pt: increased C C T rt stimulation M: absent N / P 1 3 , rt side study PT: normal M: normal PT: absence of N22 and P40s bilaterally M: normal PT: C C T It > rt > 3 S.D. rt P40 low amplitude M: normal PT: C C T It > rt > 3 S.D. rt P40 low amplitude M and U: normal PT: C C T It > rt > 3 S.D. rt P40 low amplitude Median: normal PT: C C T rt > It > 3 S.D. P40 ampl low, It side Median: normal PT: absent P40s bilaterally M and U: normal PT: absent P40s bilaterally M: no P / N 1 3 and N20 potentials on both sides. PT: normal M and PT: increased C C T bilaterally N o P / N 1 3 s both sides M: rt, no P / N 1 3 and N20 PT: RT, increased C C T It, no P40 M: normal PT: absence of LP and P40s bilateral M: normal PT: C C T It > rt > 3 S.D. M and PT: normal
17
36/M
Enlarged upper cord and medulla with high signal intensity (T2) and syrinx
18
21/M
M: P / N 1 3 and N20 potentials absent bilaterally PT: normal M and PT: normal
Low signal, intradural, extramedullary mass at C 1 - C 2 level Low intensity signal, intradural-extramedullary at T 1 0 - T l l level Low intensity signal, intradural-extramedullary at T10 level High intensity T2 signal intradural-extramedullary at T10 level High intensity T2 signal intradural-extramedullary at T10 level T4 lesion with cord compression Lesion at T12-L1 with flow void characteristics High T2 signal intensity mass T7, with cord compression Enlarged cervical cord and and medulla with large cysts Enlarged cervical cord and lower medulla, cystic
Neurofibroma
Meningioma
Meningioma
Teratoma
Schwannoma
Schwannoma
Metastatic adenocarcarcinoma Angioma
Aneurysmal bone cyst Astrocytoma
Ependymoma
Enlarged upper cord to the pons with cavity formation Mass with low T2 signal at T 7 - T 9
Astrocytoma cystic
Cord atrophy and small syrinx, C 3 - C 6 Small syrinx C 2 - C 7
Astrocytoma
Increased T1, decreased T2 signal mass at T12 level
Meningioma
Syrinx with hemangioblastoma No surgery possible ependymoma Lipoma
105
SEPs AND MR IN INTRASPINALNEOPLASMS TABLE II (continued) Age/Sex
SEP finding
MR localization
Pathology
19
28/M
M and PT: normal
20 21
39/F 22/F
M and P: normal M and P: normal
Myxo-papillary ependymoma Angioma Schwannoma
22
50/M
M: absent P/N13 and N20 potentials bilaterally PT: absent P40s bilaterally M and PT: normal M: normal PT: CCT It > rt > 3 S.D. M: Lt P/N13 a b s e n t PT: normal
Conus mass with increased signal at T12 level Mass LI-L2 Mass with increased T2 signal, Tll T12 level Large cervical syrinx
23 24
30/M 40/M
25
37/M
extend above C6 level (Figs. 3 and 4). One patient had syrinx and tumor at C 3 - C 6 level. (3) Both median and posterior tibial studies were abnormal: there were 4 patients in this group. Three of them had extensive neoplasms involving the cervical cord and the lower brain-stem (patients 12, 13 and 22, Table II). One patient had a cervical cord tumor and syrinx (patient 25). Both median and posterior tibial studies were normal in 6 patients. Five of these patients had a mass at the high lumbar region involving the most distal part of the spinal cord (patients 18-21 and 23, Table II). One patient had a small hemangioblastoma embedded in the wall of a syrinx at the lower cervical region which escaped MR detection as well (patient 16, Table II). Two patterns of central SEP abnormality were noted: (1) Absence of central potentials (N20, P / N 1 3 , P40): this group consisted of 11 patients. Six of these patients had extensive cervico-medullary lesions and 4 patients had thoracic lesions with evidence of cord compression (Figs. 2 and 3) and 1 patient had a thoracolumbar angioma. (2) Increased central conduction time only: there were 8 patients in this group, 7 with unilateral and 1 with bilateral abnormality. The latter patient also demonstrated marked reduction of amplitude and poor reproducibillity of P / N 1 3
Gabolinium MR: tumor module at C2-C3 Mass conus to L3 Widened cord T7-T12 Cervicomedullary syrinx Gadolinium MR: enhancement C6 level
Recurrent Ependymoma Ependymoma Astrocytoma Ependymoma C2-C6
potentials bilaterally (patient 12, Table II). In all 7 patients with unilateral abnormalities the finding consisted of increased N 2 2 - P 4 0 central conduction time obtained in the posterior tibial study. Five of these patients had normal absolute values but side to side difference exceeded 3 S.D. In 4 of these 5 patients the amplitude of P40 potential was also diminished on the side of prolonged CCT (Fig. 4). SEP was abnormal in all 7 patients who clinically demonstrated impairment of the position sense (100%), in 7 of 11 (63%) patients with spinothalamic sensory abnormalities only and in 4 of 8 patients (50%) with no sensory deficits. Follow-up post-surgical repeat SEPs were obtained in 11 patients. Two patients showed an improved pattern which corresponded with clinical improvement. Six patients showed no change and 3 patients demonstrated a deterioration of SEP pattern. N o correlation with the clinical course was noted for the patients with unchanged or worsened SEPs.
Discussion
Early studies of median SEP in humans have shown a close correlation between SEP abnormalities and pathology of posterior columns or lemnis-
106
B. JABBARI ET AL.
A Cc-AI ................... ~ . , ~,,..+..+...+.
,~
. f ~
.... ~
,..~,+,.~,
~
.......~
-
,-+
4.0
4.0
4.0
?
.............
,,
V,.N?/~w~-
........ - ~ ........ ,,"2 ~
4,0
........~.< . ~ > , ±
4.0
C2"Fz
•
'
'
'kL/
'
'"v
-
~r
v~>CJ"-"
'
4.0
'
l
I!/
.0
G1-
L
I .........
0
I .........
8
I .........
16
1 .........
24
I .........
32
I.........
40
I
.........
48
I .........
56
l .........
64
L .........
72
I
80
G1I .........
0
I .........
8
I .........
16
I .........
24
I .........
32
L .........
40
I .........
48
L .........
56
I .........
64
~ .........
72
1
80
Fig. 2. Patient 11. A: median SEP shows absence of P/N13 and N20 components bilaterally - posterior tibial studies are normal. B: on sagittal T1 weighted pulse sequence there is a large expensile mass extending from medulla, to upper thoracic region. There are several cavities of decreased signal, primarily limited to the medulla.
cal system, m a n i f e s t e d clinically as proprioceptive deficit ( G i b l i n 1960; H a l l i d a y a n d Wakefield 1964). In addition, recent studies d e m o n s t r a t e d that SEP can provide evidence for subclinical dis-
r u p t i o n of the central sensory pathways. SEP a b n o r m a l i t i e s have b e e n reported in n o n - s e n s o r y focal b r a i n disorders ( M o i l et al. 1984), in C N S lesions w i t h o u t p r o p r i o c e p t i o n deficits (Pedersen
SEPs A N D M R IN I N T R A S P I N A L N E O P L A S M S
107
2.C
C4-C3
N22 IC-L1
IC -PF
J
± G1. t
1~
0
......
i .........
10
I .........
20
I .........
30
I .........
40
I .........
50
I .........
60
I .........
70
I .........
80
I .........
90
I
100
Fig. 3. Patient 1 (Table II). A: left posterior tibial SEP study shows well formed popliteal fossa and caudal cord potential (N22) but P40 is absent. This patient's median SEP is normal. B: there is an expensile mass from the level of C7 to T3 which is intradural with increased signal on sagittal T1 weighted image. The pathology proved to be lipoma.
and Trojaborg, 1984), and in bilateral brain disorders known not to produce sensory symptoms (Bollen et al. 1985). Although SEPs have been described before in patients with intraspinal neoplasms (NoEl and Desmedt 1980; Riffel et al. 1984; Mauguirre et al. 1985), there has been no systematic study of this subject correlating SEP data with details of clinical and MR findings. Cracco (1973) described absence of potentials rostral to a neoplastic cord lesion in some patients with clinical evidence of advanced disease. NoEl and Desmedt (1980) reported 7 patients with a cervical myelopathy due to tumors and cervical spondylosis. These patients had a normal median study, but lower limb stimulation (sural nerve) disclosed abnormal results. The authors stated that such an SEP pattern in the presence of a cervical myelopathy should suggest its compressive nature. In another study of 6 patients with spinal cord tumors (Riffel et al.
1984), increased central conduction times in posterior tibial SEPs were observed in 3 patients. The authors, however, considered decreased amplitude ratio of P 4 0 / N 2 2 ( < 0.47) a more powerful predictor of abnormality as it was noted in the majority of their patients. Maugui~re et al. (1985) carefully studied median and posterior tibial SEPs in 20 patients with intraspinal neoplasms. Not all patients had both studies and some of them had the test postoperatively. M R was not performed (or reported) on these patients. They have found abnormal SEPs in several patients with normal sensory examination and believed this 'electroclinical dissociation' to be a pattern of extramedullary neoplasms. The majority of their patients with cervical neoplasms had normal median SEPs. SEP-clinical correlations
Several relevant clinical questions might be asked from the data accumulated in our study:
"etUOUUeA~qas~ sen~ s!q.L "alSS) JO ueql .l;~le~)J~ so!lsuolaeaeq3 IgU~!S seq ss-etu onss!l ljos aql 'a3uznbas aslnd p a l t t ~ ! ~ E l Ieu°ao3 uo :~) "slooa aalau ~u!punoJJns ao pao3 lV.u!ds aql jo sags!aalaeaeq~ ieu~!s seq uo.tsaI s!ql aouznbas aslnd paltl~!anx [ I |euoao3 u O "slooa anaau aql Jo ~sano3 ~ql ~UOle ~u!puzlx~ 'ssetu anss!l 1jos paq!aastunaa!3 ila~ e s! uo!~za 3!3eaoql aaddn pue lea!Aaa3 aZaOI ltl~!a ~ql U! :fl "IetUJOU S! d3S ug!pzlAI "017d JO ~pnl!ldtue u! zseaa3ap pa~Iaetu pug (01rd-EEN) atu!l uo!lanpuo~ leJlUa3 pase~azu! SA~OqSd~JS le!q!l ao!Jalsod lq~!a :V "(II aIqeZ) Z lua!led "P "~.tq
OOI.
P .....
06 I ........
05 T.........
0/. I .........
09 T .........
OG T.........
0IF I .........
Og I .........
OZ I .........
01.
0
OOL
OI
oe
OL
N
U
Ot
O[
e~
O~
e
[ .........
-I.O
-L9 L'I'S
O'l
0"|
Ll-rl
0"|
o'1 tNI
|'it
e'l
0"l
O~
"3V i ~ I'SV~flVf "fl
V 80[
SEPs A N D MR IN INTRASPINAL NEOPLASMS
(1) Are the SEPs a good screening test for patients who have a normal neurological examination but who may, nonetheless, have spinal tumors? One of 3 such patients (nos. 3, 21 and 23) in our study had an abnormal SEP, strongly suggesting a central pathology. (2) Do SEPs help to establish the central origin of the symptoms when the neurological examination could be consistent with a purely peripheral pathology? In our study 3 of 6 patients (nos. 7, 16, 18, 19, 22, 25) with such ambiguous examination had cenlrally abnormal SEPs. (3) Do SEPs help establ;.sh the bilaterality of involvement and thus suggest (but not prove) a spinal cord localization with symptoms and signs confined to one-half of the body and sparing the face? Two of four (50%, nos. 13, 15, 18, 20) fit this description. (4) Can particular SEP findings or pattern of findings be used either to suggest or to exclude a spinal cord tumor in a patient who clinically presents with a myelopathy? No - the observed patterns of SEP abnormality were non-specific for tumors. (5) When neurological examination is abnormal and strongly suggestive of a central pathology, can SEP predict the location of the tumor more accurately than the neurological examination and thus allow a more restricted radiologic evaluation? Overall, clinical examination of such patients was more accurate than SEP localization. However, SEP localization was very closely correlated with clinical examination in patients with findings confined to the legs due to thoracic neoplasms. Moreover, as described above 4 of 9 patients with either a normal or ambiguous examination had a centrally abnormal SEP study. S E P - M R correlations and lesion localization Magnetic resonance imaging is now the non-invasive procedure of choice for intraspinal neoplasms (Williams et al. 1987). We found good correlation between SEP and M R for cervical and thoracic neoplasms as SEP was abnormal in all 17 patients of this category in whom M R disclosed a mass lesion. Median SEP was abnormal (with or without posterior tibial abnormality) in 7 of 8 patients with M R evidence of cervical tumor above
109
the C6 level whereas in the group with neoplasms below C6 level, all 9 patients disclosed an abnormal posterior tibial SEP and a normal median study. However, only 1 of 6 patients with cauda equinaconus medullaris syndrome demonstrated an abnormal SEP. Our low yield in this area may be due to our method of recording which did not include lumbosacral recording sites. Using such a montage Lehmkuhl et al. (1988) found abnormal SEPs in all 8 patients with trauma to cauda equina or conus medullaris region. Two of our patients with a small nodule of neoplasm in the wall of a cervical syrinx had normal median SEPs but M R also failed to detect the neoplastic lesion (nos. 15 and 16). It should be noted, however, that both patients had nonenhanced MRs and a vascular lesion like hemangioblastoma would ordinarily produce enhancement. Patient 15 provided exception to NoEl and Desmedt's view that in cervical myelopathy the 'normal median-abnormal posterior tibial' pattern indicates an extradural compressive pathology. Both median and posterior tibial SEPs were markedly abnormal in a patient with a post-surgical cervical syrinx in whom M R with gadolinium injection disclosed recurrence of tumor. In our series, the high yield of median SEP in cervical neoplasms contrasts with the low yield reported by other investigators (Maugui~re et al. 1985; Lehmkuhl et al. 1988). We believe this is due to the fact that the majority of our patients with cervical neoplasms had extensive intramedullary lesions which are more likely to produce abnormalities of median SEPs. On the other hand, the yield of medium SEP in intraspinal cervical neoplasms may be further improved by modification and extension of the current SEP recording techniques. Using a circular array of electrodes around the neck, Emerson and Pedley (1986) reported a dissociation between N 13 (posterior neck) and P13 (anterior neck) potentials in several patients with intramedullary lesions in whom conventional median nerve studies showed no abnormality. Other authors have found ulnar SEPs more informative than median or posterior tibial SEPs in patients with compressive cervical myelopathies (Veilleux and Daube 1987). Only 2 of our patients had the ulnar study. However,
110 u l n a r a n d m e d i a n S E P s w e r e n o r m a l in 1 p a t i e n t w i t h a C 6 - T 1 s c h w a n n o m a ( p a t i e n t 7, T a b l e II). P o s t o p e r a t i v e i m p r o v e m e n t of S E P c o r r e l a t e d w i t h c l i n i c a l i m p r o v e m e n t in 2 p a t i e n t s . T h e i m p o r t a n c e o f p o s t o p e r a t i v e S E P s in p r e d i c t i n g f u n c t i o n a l r e c o v e r y is c o n f o u n d e d b y several v a r i a b l e s , the m o s t i m p o r t a n t p o s s i b l y b e i n g t u m o r type. The assessment of the predictive value of postope r a t i v e S E P in i n d i c a t i n g f u n c t i o n a l r e c o v e r y a w a i t s a l a r g e r s t u d y c o n t r o l l e d for this m a j o r variable. I n c o n c l u s i o n , o u r d a t a i n d i c a t e t h a t S E P is a s e n s i t i v e m e a s u r e o f s p i n a l c o r d i n v o l v e m e n t by i n t r a s p i n a l t u m o r s at c e r v i c a l o r t h o r a c i c levels. M R is the p r o c e d u r e of c h o i c e for d e t e c t i n g s p i n a l c o r d n e o p l a s m s . H o w e v e r , M R is s e l d o m i n d i c a t e d in p a t i e n t s w i t h v a g u e c o m p l a i n t s , n o n l o c a l i z i n g signs o r n o r m a l e x a m i n a t i o n s . I n o u r s t u d y , 44% o f s u c h p a t i e n t s d e m o n s t r a t e d a c e n trally a b n o r m a l S E P s t u d y w h i c h c l e a r l y d i s c l o s e d an i n d i c a t i o n for M R of t h e spine. S E P m a y p r o v e to b e e v e n m o r e u s e f u l in c h i l d r e n w i t h i n t r a s p i n a l t u m o r s in w h o m t h e s y m p t o m s are o f t e n v a g u e a n d S E P a b n o r m a l i t i e s a r e u n l i k e l y to r e p r e s e n t m u l t i p l e sclerosis o r d e g e n e r a t i v e s p i n e disease. This study was performed by the W.R.A.M.C. Department of Clinical Investigation, Project No. 7132. The authors wish to thank Mrs. Michelle Welton, Miss Diane Bailey and Mrs. Ferne Robinson for assistance in the preparation of this manuscript and S.F.C. Howard Mays for performing SEPs.
References Bollen, E.L., Arts, R.J., Roose, R.A.C., Van der Velde, E.A. and Buruma, O.J.S. Somatosensory evoked potentials in Huntington's chorea. Electroenceph. clin. Neurophysiol., 1985, 62: 235-240. Chiappa, K.H. Evoked Potentials in Clinical Medicine. Raven Press, New York, 1983: 203-312. Cracco, R.Q. Spinal evoked response: peripheral nerve stimulation in man. Electroenceph. clin. Neurophysiol., 1973, 35: 379-386. Desmedt, J.E. and Cheron, G. Prevertebral (oesophageal) recording of subcortical somatosensory evoked potentials in man. Electroenceph. clin. Neurophysiol., 1981, 52: 257-275. Eisen, A.A. Non-invasive measurement of spinal cord conduc-
B. JABBARI ET AL. tion: review of the presently available methods. Muscle Nerve, 1986, 9: 95-103. Emerson, R.G. and Pedley, T.A. Effect of cervical spinal cord lesions on early components of the median nerve somatosensory evoked potentials. Neurology, 1986, 36: 20-26. Giblin, D.R. The effects of the lesions of the nervous system on cerebral responses after peripheral nerve stimulation. Electroenceph. clin. Neurophysiol., 1960, 12: 262. Halliday, A.M. and Wakefield, G.S. Cerebral evoked potentials in patients with dissociated sensory loss. J. Neurol. Neurosurg. Psychiat., 1964, 26: 21-28. Lehmkuhl, D.L., Dimitrijevid, M.R. and Zidar, J. Lumbosacral evoked potentials and cortical somatosensory evoked potentials in patients with lesions of conus medullaris and cauda equina. Electroenceph. clin. Neurophysiol., 1988, 71: 161-169. Lueders, H., Lesser, R., Hahn, J., Little, J. and Kelm, G. Subcortical somatosensory evoked potentials to median nerve stimulation. Brain, 1983, 106: 341-372. Maugui~re, F., lba~ez, V. et Fisher, C. Les potentiels 6voqu6s somesth6siques dans les tumeurs intra-rachidiennes. Rev. EEG Neurophysiol., 1985, 15: 95-106. Mori, E., Yamadori, A., Kmdo, Y. and Tabuchi, M. Ataxic hemiparesis from small capsular hemorrhage. Computed tomography and somatosensory evoked potentials, Arch. Neurol,, 1984, 41: 1050-1053. Mulder, D.W. and Dale, A.J.D. Spinal cord tumors and disks. In: A.B. Baker and R:J. Joynt (Eds.), Clinical Neurology. Harper and Row, Philadelphia, PA, 1975: Ch. 45. No~l, P. and Desmedt, J.E. Cerebral and far-field somatosensory evoked potentials in neurological disorders involving the cervical spinal cord, brain stem thalamus and cortex. In: Progress in Clinical Neurophysiology, Vol. 7. Karger, Basel, 1980: 216-218. Pedersen, L. and Trojaborg, W. Visual, auditory and somatosensory pathway involvement in hereditary cerebellar ataxia, Freidreich's ataxia, and familial spastic paraplegia. Electroenceph, clin. Neurophysiol., 1984, 52: 283-297. Percy, A.K., Elveback, R.K., Okazaki, H. et al. Neoplasms of the central nervous system. Epidemiologic considerations. Neurology, 1972, 22: 40. Riffel, B., St~3hr, M. and K~Srner, S. Spinal and cortical evoked potentials following stimulation of the posterior tibial nerve in the diagnosis and localization of spinal cord diseases. Electroenceph. clin. Neurophysiol., 1984, 58: 400-407. Schliack, H. and Stille, D. Clinical symptomatology of intraspinal tumors. In: P.J. Vinken and G.W. Bruyn (Eds.), Handbook of Clinical Neurology, Vol, 19. North-Holland Publ., Amsterdam, 1975: 23-49. Seyal, M. and Gabor, A.J. Generators of human spinal somatosensory evoked potentials. J. Clin, Neurophysiol., 1987, 4: 177-187. Sloof, J.L., Kernohan, J.W. and McCarthy, C.S. Primary Intra-Medullary Neoplasms of the Spinal Cord and Filum Terminale. Saunders, Philadelphia, PA, 1984.
SEPs AND MR IN INTRASPINAL NEOPLASMS Veilleux, M. and Daube, J.R. The value of ulnar somatosensory evoked potentials in cervical myelopathy. Electroenceph, clin. Neurophysiol., 1987, 68: 415-423. Williams, A.L., Haughton, V.M., Pojunas, K.W., Danials, D.L. and Kilgore, D.P. Differentiation of intra-medullary neo-
111 plasms and cyst by MR. Am. J. Neuroradiol., 1987, 8: 527-532. Yu, Y.L. and Jones, S.J. Somatosensory evoked potentials in cervical spondylosis. Brain, 1985, 108: 273-300.