635

Future Ambulation Prognosis as Predicted by Somatosensory Evoked Potentials in Motor Complete and Incomplete Quadriplegia
Stanley R. Jacobs, MD, Natalie K. Yeaney, BS, Gerald J. Herbison, AID, John F. Ditunno, Jr., hiD
ABSTRACT. Jacobs SR, Yeaney NK, Herbison G J, Ditunno Jr. JF. Future ambulation prognosis as predicted by somatosensory evoked potentials in motor complete and incomplete quadriplegia. Arch Phys Meal Rehabil 1995;76:635-41. Objectil~e: The purpose of this prospective study was to determine the efficacy of tibial somatosensory evoked potentials (SEPs) in predicting ambulation in tetraplegic individuals. Design: This was a prospective study of a cohort of cervical spinal cord-injured patients who had SEPs recorded within 72 hours to 2 weeks post-SCI and whose ambulation outcome was followed up to 2 years post-SCI. Setting: Regional Spinal Cord Injury (SCI) Center. Patients: All male and female subjects admitted to the center from 1988 to 1991 between the ages of 15 and 60 years who demonstrated C4 through T~ complete and incomplete acute SCIs were asked to participate in this study. Measurements: The tibial nerve cortical SEPs were graded as either present or absent. The waveforms were also graded as less than 0.5/tV or --0.5/tV. Quadriceps strength plus touch and pin sensation were tested within 72 hours to 2 weeks post-SCI. Ambulation was rated as absent, exercise, household, or community. The ambulatory and clinical status were assessed monthly for 3 months, and then at 6, 12, 18, and 24 months postSCI. Statistical analysis using the two-tailed Fisher's exact test was performed relating the initial clinical and SEP data to ambulation outcome up to 24 months post-SCI. Results: All 13 subjects with a right and/or left quadriceps manual muscle test (MMT) greater than 0/5 became ambulatory. Of the 9 subjects with an initial bilateral quadriceps MMT = 05, only 1 recovered enough lower limb function to ambulate (p = .0001). One of the 7 subjects with absent touch sensation in the lower limbs became ambulatory, whereas 14 of the 15 subjects with touch sensation present became ambulatory (p = .002). All 7 subjects with absent pin sensation in the lower limbs were nonambulatory, and 14 of 15 subjects with pin sensation present became ambulatory (p < .0001). Of the 9 subjects with bilaterally absent cortical SEP waveforms, 2 became ambulatory. Twelve of the 13 subjects with a cortical SEP wave present became ambulatory (p = .0015). Of the 10 subjects with a cortical SEP wave amplitude less than 0.5/xV, only two became ambulatory, whereas all 12 subjects with an amplitude - 0.5/xV became almbulatory (p = .00014). In no subject did the SEP predict future ambulation where the clinical examination did not also predict recovery of ambulation. Conclusion: Both the early postinjury clinical evaluation and the SEP predicted ambulation outcome to a significant degree, but the SEP offered no additional prognostic accuracy over that provided by the clinical examination. 1995 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitt~ion

Providing a prognosis for recovery of ambulation in the acutely spinal cord-injured (SCI) patient is helpful in assisting the rehabilitation team to efficiently plan functional and vocational training, as well as timely home renovations. ~ It is well documented that quadriplegic SCI patients who remain Frankel A at 2 weeks postinjury very rarely ambulate. 2~*However, the ambulation prognosis varies for patients whose initial Frankel score is B, C, or D. s-8 Although initial bedside neurological evaluation is helpful in predicting amFrom the Department of Rehabilitation Medicine, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA. Supported in part by awards from the National Institute on Disability and Rehabilitation Research to the Regional Spinal Cord Injury Center of Delaware Valley (G008535135) and the National Rehabilitation Research and Training Center in Spinal Cord Injury (H133B80017). Submitted for publication July 22, 1994. Accepted in revised form January 31, 1995. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Stanley R. Jacobs, MD, Department of Rehabilitation Medicine, Thomas Jefferson University Hospital, Suite 9410, Gibbon Building, 111 S. l I Street, Philadelphia, PA 1910'7. 1995 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/95/7607-316953.00/0

bulation, the evaluation has not indicated with complete accuracy the patients who would or would not eventually ambulate. Studies have shown initial bedside examination of pin appreciation and strength predicts ambulation outcome with an 85% to 90% accuracy. 2"3 Crozier, 2 Li, 6 Young, 7 and their associates have examined patients' initial motor power, and Crozier, 2 Li, 6 and their colleagues have evaluated initial sensory function to predict the patients' final lower-extremity outcome. Crozier and associates reported the relationship between initial sensory function and ambulation in Frankel B patients and determined that those who had pin sensation in the lower limbs in addition to light touch sensation had a much better probability of ambulating than those who did not (p < .0002). 2 In a subsequent report, Crozier and colleagues 3 determined that patients who achieved a greaterthan 3/5 quadriceps manual muscle test (MMT) grade by 2 months post-SCI would become functional ambulators (p < .003). Katz, 5 Li, 6 Ziganow, 8 and their associates have recently studied whether a neurophysiologic test such as the somatosensory evoked potential (SEP) might further improve the ability to predict final lower limb function. However, those

Arch Phys Med Rehabil Vol 76, July 1995

636

AMBULATION PREDICTED BY SEP, Jacobs week of the injury owing to dementia or psychosis. The study also excluded patients who were medically unstable owing to the development of sepsis, hypoxia, serious immune compromise, severe metabolic disturbance, hypotension, or deep vein thrombosis. Patients with skin diseases or burns that would prevent using recording electrodes were excluded from the project as were patients who had severe lower limb injuries that precluded peripheral nerve conduction studies. Subjects taking diazepam were not included in the study because it affects the middle latency SEP measurement. n Eight patients were treated with high-dose methylprednisolone. A cervical spine operation was performed on 13 patients. The clinical assessment and treatment program were the same for the patients included in this study as for all patients admitted to the spinal cord injury center.

who studied the early post-SCI SEP only compared the initial SEP with lower limb muscle scores or improvement scores, but they did not evaluate whether the SEP could predict ambulation outcome. 5-8 Young performed SEPs on 300 acutely injured SCI patients and found a low correlation (r = 0.33) between the 6-week SEP and the 6-week motor score. 7 He did not report the correlation of the initial SEP to the 6-week motor score or to ambulation. Li and colleagues 6 compared the prognostic value of the SEP with the neurological examination in predicting the Barthel Index at 6 months post-SCI. The initial combined mean tibial and ulnar SEP score correlated with the final Barthel Index almost as well as did a combination of motor, pin, and position sense scores (R2 0.75 v R 2 0.80; p < .0001). However, the Barthel Index does not indicate final ambulation status. 9 Katz and colleagues correlated the initial dermatomal evoked potential (DEP) and SEP score with subsequent motor improvement, s He found both the initial bedside examination and the evoked potentials to be reasonable predictors of further motor improvement, but evoked potentials added little prognostic information to the clinical examination concerning motor score improvement. Although Katz compared the initial motor scores to final ambulation outcome, they did not relate the initial SEP or DEP specifically to final ambulation status. 5 Although Li and colleagues 6 and Ziganow 8 have believed the SEP could be an accurate lower limb motor outcome predictor and/or a supplement to the initial clinical examination, no one has determined whether the tibial SEP could predict which quadriplegic patients will ambulate. Because ambulation is the most important lower limb function, the authors undertook this study to determine whether the early post-SCI SEP could predict ambulation outcome more accurately than the clinical examination.

Clinical Assessment
Neurological testing was conducted according to the standards of the American Spinal Injury Association (ASIA). 1 The initial clinical and electrophysiological tests were performed between 72 hours and 2 weeks post-SCI because Herbison and colleagues demonstrated that a 72-hour to 2week examination was superior to the initial 24-hour examination in forecasting motor and functional recoveryJ 2 Because of scheduling conflicts, the clinical and SEP examinations could not be performed on the same day. The clinical assessment included testing motor power, 1 touch m and pinprick m sensation, deep tendon reflexes (DTR), 13and ambulation s t a t u s . 2"3'14 The ambulation and subsequent neurological status were then assessed monthly for 3 months, and at 6, 12, 18, and 24 months post-SCI on all patients who met the selection criteria for the study. If a subject became a community ambulator within 1 year post-SCI, the ambulation status was no longer evaluated for the purpose of this study. Ambulation status was rated as nonambulatory, exercise only, household, or communityJ 5 Patients could use devices to be considered an exercise, household, or community ambulator. Other than ask the subject if he or she was an exercise, household, or community ambulator, no further attempt was made to quantify ambulationJ 5 The DTR was rated according to a 0-to-5 scale as per a modified method of Hermant3: 0, absent reflexes; 1, hyporeflexive; 2, normal; 3, increased deep tendon reflexes and/or 1 to 2 beats of clonus; 4, 3 to 4 beats of clonus; and 5, more than 5 beats of clonus.

METHODS Materials and Methods

Twenty-two cervical spinal cord-injured patients were evaluated by examining initial touch and pin sensation, motor strength, and the tibial SEP and relating them to the attainment of functional ambulation.

Patient Selection
This prospective study focused on 22 sequentially admitted cervical spinal cord-injured patients who had early SEPs recorded and whose ambulation outcome was observed up to 2 years postinjury. Informed consent was obtained from each patient. Male and female subjects between the ages of 15 and 60 who demonstrated C4 through T1 complete and incomplete acute SCIs were eligible for the study. The Frankel grades of these patients ranged from class A to class D.m Individuals were asked to participate in the study if they had been admitted to a regional spinal cord injury center within 72 hours of the onset of injury. SEPs were not performed on patients unless they met the criteria for this study. Subjects with probable brain or brainstem injury were excluded from the project as were subjects who could not cooperate with a sensory or motor examination within 1

Evoked Potential Testing

All patients had fight and left tibial SEPs performed between 72 hours and 2 weeks postinjury. SEPs were acquired using a four-channel Nicolet Compact System a by a trained technician. Recordings were obtained from the popliteal space, spinous processes L2-L4 and T10-T12, as well as cephalic leads over Fpz' and Cz'. Subjects were prepared according to the method of DeLisa and associates 16 and the American Encephalographic Society t7 based on the 10/20 system of electroencephalogram (EEG) and SEP recording. Impedances were kept at or below 3K Ohm with no more than a 2K Ohm difference between electrodes. Separate fight

Arch Phys Med Rehabil Vol 76, July 1995

AMBULATION PREDICTED BY SEP, Jacobs

637

C7 Frankel D 19 y. o. male SEP

Scale for tracings 4,8 2.44~v 1
10 milli~z

.i
M4 o M3 M5 M1

N2
-%_...~ ~...~, .,~,_~,-~ ~ ,~ ~.~., J s

,*

C 3, . Fpz, e ",----._.._. ._.__~-,.~_f-N 17

.k ~ _,,---'-"~'~-.~--~----~

K,:,r,r~.#/ - I -~--_... -.....

-,.. ~ ' ~

i;II
Fig 1 - - T h i s is 1the tibial nerve cortical potential from a subject with C7 Frankel D quadriparesis. Note the symmetrical perfectly duplicated W-like wave form.

T10 "T12 "-*

. . . . . . P - ~ ' - ' " - ....... --

L1 L2 Tiblal n. popllteal space

MS
Stim. Rate

Duration Recording sensitivity 500 sweeps averaged

2.1 Hz 200 ix,sec 10p.V/Div

and left tibial SF_,Ps were obtained by stimulation posterior to the medial rnalleolus. The evoked responses from 500 stimulations were recorded and averaged. The tibial nerve stimulation parameters were as follows: 0.2 milliseconds duration, frequency 2.3Hz, and enough intensity to cause a brisk hallux flexJion twitch Two runs were obtained for each nerve studied to ensure reliability If the runs were not adequately matched, 3 or 4 runs were obtained whenever possible Typical trac,ings of normal, moderately abnormal, and absent waveforms are shown in figures 1, 2, and 3. It was possible to identify the following peak responses in several subjects: P1, N1, P2, N2. However, in some subjects it was impossible to determine whether the single peak response recorded was a delayed P1 or N1 versus an absent P1 or N1 with preserved P2 or N2 peaks. Therefore, as per the method of Li, 6 the largest cortical SEP peak to peak amplitude within lOOms was measured.
Data Analysis

was defined as achieving household or community ambulation. If the subjects were nonambulatory or performed only exercise ambulation, they were considered nonambulatory. All patients whose final status was nonambulatory were observed for at least 1 year post-SCI. Statistical analysis. The two-tailed Fisher's exact test was performed relating the initial SEP data sensation, and motor power, to final ambulation status. The Fisher's exact test was the only statistical analysis used in this study. There were too few patients to statistically analyze the effect of methylprednisolone or surgery on final ambulation status. RESULTS Quadriceps strength, pin sensation, and touch sensation below the level of the lesion all significantly predicted ambulation outcome. Sensory limb scores were no more predictive of ambulation outcome than the presence or absence of pin and touch sensation in the lower limbs. Unfortunately the combined DTR score was so variable that ambulation outcome could not be assessed in relation to the spasticity score. Of the 9 subjects with an initial 0/5 bilateral quadriceps MMT, only 1 recovered enough lower limb function to ambulate. All 13 subjects with initial right or left quadriceps strength greater than 0/5 ambulated, whereas with a quadricep M M T score equal to 0, no subject ambulated. Seven subjects had an absence of touch sensation at the time of their initial examination. However, only 1 became ambulatory and pin was present in this subject. Fifteen subjects had initial touch sensation: 14 became ambulatory and 1 remained nonambulatory but had no pin appreciation below the lesion (No. 16, fig 4). Given the presence of touch sensation, the probability of a subject ambulating was 87%.

SEPs. Cortical SEP signals were recorded as present or absent. 6 In addition, these subjects were categorized as having peak-to-peak cortical SEP response of either less than 0.5#V or ->0.5#V. Sensation. The presence or absence of sensation was defined as having any sensation below the level of the lesion. The absence of sensation was defined as having no sensation below the level of the lesion. Separate analyses were performed for pin and touch) M o t o r power. The presence of lower-extremity motor power was defined as having a quadriceps M M T of greater than 0/5 in both lower extremities. F i n a l ambulation status. The final ambulation status

Arch Phys Med Rehabil Vol 76, July 1995

638

AMBULATION PREDICTED BY SEP, Jacobs

C 7 Frankel C 55 y. o. female SEP

Scale for tracings 4,8
1.221N I C 3, M8 M4" M7 --11 M3" M6M2 M5 M1.

- Fpz,

10-T12 L1 L2 PN
I I I I I I I I I I

Tlblal n.
popllteal space

MS

Stim. Rate Duration Recording sensitivity 500 sweeps averaged

2.1 Hz 200 psec 101~V/Div

Fig 2mThis is the tibial nerve evoked cortical SEP of a patient with C7 Frankel C quadriparesis. Although the waveforms are present, the early components are markedly diminished.

Given the absence of touch sensation, the probability of a subject ambulating was 14%. The significance of the statistical findings was p = .002. Of the 7 subjects with absent initial pin sensation in the lower extremities, none became ambulatory. Fourteen of the 15 subjects with initial pin sensation were ambulatory by 1

year postinjury, and only 1 subject remained non-ambulatory. The probability of a subject ambulating given the presence of initial pin sensation was 93%. Given the absence of initial pin sensation, no subject ambulated. When no matching set of potentials could be obtained, the SEP was scored as absent rather than deleting the data

C 5 Frankel A 23 y. o. male SEP

Scale for tracings 4,8
M 8 iir-~,

"lB~i

0.611AV ' 1o milllsec

P* A

M3_

..~T~"~_"

710"712

M2" M1I i . I I . . f . . I . . I . . . I I I

L tlblal popllteal space

Fig 3mThese are the tibial nerve cortical SEP signals from a subject with C5 Frankel A quadriplegia. Note the absence of any discernible signal.

MS
Stim. Rate Duration Recording sensitivity 500 sweeps averaged 2.1 Hz 200 ixsec 101xV/Div

Arch Phys Med Rehabil Vol 76, July 1995

AMBULATION PREDICTED BY SEP, Jaeobs

639

Frankel R / L I n i t i a l R / L F i n a l T o u c h
Grade CSD Quad 5/5 MMT Quad MMT 5/5 P P P P P P P P P P P P A P P P A A A A A A

Pin SEP SEP A or P A or P A or P Amp. P P P P P P P P P P P P P P P A A A A A A A P P P P A P P P P P P P P A A P A A A A A A 2.56 1.36 1.20 1.80 0.00 2.16 2.10 6.83 1.95 4.90 2.05 2.78 2.09 0.00 0.00 0,46 0.001 0.00] 0.00 0.00 0.00 0.00

Final Ambulation C C C C C C C H H C C C C N H N N N N N N N

C5D C6D C5D

C5D

5/5 5/5 5/5

4+/5

5/5 5/5 5/5

5/5

C6C
C5C

5/4+
4+/1+

5/4+
4+/4

C4C
C4C

3/4
3+/3+
3/2+ 2+/2+ 1+/2+

5/ 5
4+/4
4+/5 4+/4+ 3+/4 5/5

10 11 12
13 Fig 4 - - T h e sphtal injury level, the clinical examina- 14 tion results, the SEP, and the ambulation outcome of 15 the subjects in ttds study. %Quad, Quadriceps; *A 16 or P: absent or present; XAmp, amplitude in evoked 17 potential; **C, community ambulation; H, household. 18 A, absent; Amp, ,'vnaplitude; 19 C, community ambulator; H, household ambulator; 20 ladt, initial; MMT, manual muscle test; N, nonambu- 21 latory; Quad, quadriceps; SEP, somatosensory evoked 22 potential; Tot, total.

C7C

C5C
C6C

C4C C6C C5B C5B

1/1
0/0 0/0
0/0

0/1+ 0/0
0/1+

C5A
C5A CSA C4A

0/0
0/0 0/0 0/0

o/o
0/0 0/0 0/0

C5A
C5A

0/o 0/ 0

0/ 0 0/ 0

from consideration. This occurred in 1 patient who eventually ambulated. Nine subjects had no discernable cortical waveforms during the initial SEP testing. Of the 9 subjects with absent cortical waveforms, 2 became ambulatory. Twelve of the 13 subjects with a conical SEP wave present became ambulatory, whereas only 1 subject remained nonambulatory. Given the presence of a tibial cortical SEP wave, the probability of a subject ambulating was 92%. If the tibial SEP wave was absent, the probability of a subject ambulating was 22% (p = .0015). Of the 10 subjects with a cortical SEP wave amplitude less than 0.5#V, 2 became functionally ambulatory and 8 were nonambulatory. All 12 subjects with a cortical SEP wave amplitude -> 0.5#V became ambulatory. Given an amplitude -- 0.51~V, the probability of a subject ambulating was 100%. Given an amplitude less than 0.5#V, the probability of a subject ambulating was only 20%.

DISCUSSION The early SCI clinical examination and the tibial SEP both forecast SCI ambulation outcome with a significant degree of accuracy. Although the early post-SCI tibial SEP forecasts ambulation outcome, it is not superior to the clinical examination. This prediction of SCI motor recovery using SEPs and the clinical examination can be explained by known anatomic and physiological relationships as well as animal SCI histopathologic and functional outcome studies. 18-22 Anatomic and physiological studies demonstrate that lower limb touch and tibial SEPs are carried in the medial portion of the posterior columns. Touch is also carried in the anterior spinothalamic tract (fig 5). 23 Pin prick is carried in the lateral spinothalamic tract. The lateral spinothalamic tract is much closer to the lateral corticospinal tract than is the medial portion of the posterior

Arch Phys Med Rehabil Vo! 76, July 1995

640

AMBULATION PREDICTED BY SEP, Jacobs

FASCICULUS

CORTICOSPINAL TRACT

LATEFIAL
SPINOTHALAMIC TRACT ANTERIOR SPINOTHALAMIC TRACT

Fig 5 - - A simplified cross-section diagram of the spinal cord stressing the spatial relationships of the corticospinal tract to the posterior columns and the lateral spinothalamic tract. column and the anterior spinothalamic t r a c t s . 24 Therefore, sparing of the axons in the lateral spinothalamic tract may be associated with a similar preservation of axons in the corticospinal tracts making pin a highly significant predictor of ambulation outcome. The correlation between preservation of pin sensation and ambulation in the present study was first reported by Crozier and associates. 2 They provided a similar anatomic explanation as to how the preservation of pin sensation predicts ambulation outcome. Lower limb touch is carried in axons in the medial portion of the posterior columns and the anterior spinothalamic tract. These two spinal cord touch tracts are approximately the same distance from the lateral corticospinal tract (fig 5). Both tracts are located relatively far from the lateral corticospinal tract compared to the pin-carrying tract. The ambulatory outcome predictive power of touch preservation, therefore, probably is not related to proximity of the two tracts to the corticospinal tract. Histopathologic reconstruction of the spindle-shaped spinal cord injury from animal studies and human reports demonstrate that the injury of axons is diffuse and similarly distributed in all the major long tracts. 18-22The similar damage in all tracts explains the predictive value of touch sensation for ambulation. However, the damage to the tracts is not always identical as demonstrated by Jellinger 25 and Quencer and colleagues) 6 A recent postmortem study of central cord injury by Quencer suggests that corticospinal tracts can be selectively damaged. 26 This may account for the occasional prognostic failure of touch sensation and the SEPs for ambulation (fig 4). The initial impression was that a cortical SEP amplitude threshold in microvolts would more accurately predict ambulation than qualitatively rating the SEP as present or absent. However, the data support the idea that nonlinear ratings such as a simple qualitative rating of the SEP as present or absent appear to forecast ambulation as well as a SEP amplitude of greater than or equal to 0.5#V.

The results of this study support the nonlinear relationship of the amount of spinal cord damage to the recovery of ambulation as shown by Blight and Decresito) 8'~9 Blight demonstrated that animals will ambulate if only 10% to 25% of the spinal cord in preserved. ]8A9'21 There is controversy as to whether a quantitative measure of the SEP is superior to a qualitative rating. Ziganow was one of the first to question the value of a qualitative interpretation of post-SCI SEPs. She found a very high correlation between a combined post-SCI SEP amplitude and latency score and a 10-week combined muscle strength sensory score of incomplete quadriplegic patients) However, 'no attempt was made to separately correlate future ambulation status with the early postinjury SEP. Young summarized data on 300 SCI patients and found that admission SEPs predicted motor recovery in only 40% to 50% of patients at 6 weeks p o s t i n j u r y . 7 HOWever, he did not comment on how soon after the injury the admission SEP was obtained. If the initial post-SCI SEP used to predict ambulation outcome was obtained during a physiologically unstable post-SCI period, this may partially explain why the SEP motor recovery correlation in Young's study w a s l o w . 7 In addition, Young did not define motor recovery, and ambulation was not correlated with the SEP. 7 Li used a patient's combination of ulnar and tibial quantitative SEP values to predict a combined upper- and lowerextremity functional score using the Barthel Index. 6 Although the Barthel Index measures various upper- and lower-extremity functions, it does not separately measure ambulation. An initial quantitative SEP score to Barthel Index correlation at 6 months was high. 6 However, no data was given to allow the clinician to predict ambulation outcome from the SEP data. 6 Katz recently stated that the SEP added little or no useful functional prognostic information beyond that provided by the initial examination, s He followed his patients for 1 year, but he did not directly relate either the early post-SCI clinical examination or the initial SEP to ambulation outcome. Although the SEP in this study significantly predicted ambulation outcome, this would only be clinically relevant if the time-consuming SEP was clearly superior to the bedside clinical examination. The results of this study demonstrated that the bedside clinical examination also strongly predicted the recovery of ambulation. All 13 subjects with an early right and/or left quadriceps muscle grade greater than 0 ambulated, and only 1 of 9 subjects with a right and/or left quadriceps muscle grade of 0 ambulated. Fourteen of 15 subjects with normal or partially intact initial pin sensation in the lower limb walked, and all 7 subjects with absent initial pin sensation below the lesion did not ambulate (p < .0001). Therefore, this suggests that the early post-SCI tibial nerve SEP provides no prognostic advantage for predicting ambulation outcome over the standardized clinical examination. Despite the fact that the SEP provides no prognostic advantage over the clinical examination, some investigators 28'29 suggested that the tibial SEP could be helpful when the bedside examination is unreliable. 27'2s Clinical prognosis used to counsel patients and their families should predict that a specific patient can regain a specific lost function such as eating, transfers, and/or ambulation as

Arch Phys Med Rehabil Vol 76, July 1995

AMBULATION PREDICTED BY SEP, Jacobs opposed to just providing total functional score or group data. In the present study, the early post-spinal cord injury tibial nerve SEP was not a better predictor of ambulation outcome than a early 72-hour to 2-week post-SCI standardized motor and sensory examination.
Acknowledgment: The authors would like to express their appreciation to Barbara Wolff, Jocelyn Collazo, Mariea Stewart, Kira Ozols, Gregory Purnsley, Patt Williams, and Patricia Herbison for their assistance in conducting this study. References
i. Ditunno Jr JF, Stover SL, Freed MM, Ahn JH. Motor recovery of the upper extremities in traumatic quadriplegia: a multicenter study. Arch Phys Med Rehabil 1992;73:431-6. 2. Crozier KS, Graziiani V, Ditunno Jr JF, Herbison GJ. Spinal cord injury: prognosis for ambulation based on sensory examination in patients who are initially moto:r complete. Arch Phys Med Rehabil 1991 ; 72:119-21. 3. Crozier KS, Cheng LL, Graziani V, Zorn G, Herbison GJ, Ditunno Jr JF. Spinal cord injury: prognosis for ambulation based on quadriceps recovery. Paraplegia 1992;30:762-7. 4. Waters RL, Yaku:ra JS, Adtdns RH, Sie I. Recovery following complete paraplegia. Arch Phys Med Rehabil 1992;73:784-9. 5. Katz RT, Tolkeikis RJ, Knuth AE. Somatosensory-evoked and dermatomal-evoked potentials are not clinically useful in the prognostication of acute spinal cord injury. Spine 1991; 16:730-5. 6. Li C, Houlden DA, Rowed DW. Somatosensory evoked potentials and neurological grades as predictors of outcome in acute spinal cord injury. J Neurosurgery 1990;72:600-9. 7. Young W. Somatosensory evoked potentials (SEPs) in spinal cord injury. In: Schranml J, Jones SJ, editors. Spinal cord monitoring. Berlin: Springer-Verlag, 1985:127-42. 8. Ziganow S. Neurometric evaluation of the cortical somatosensory evoked potential in acute incomplete spinal cord injuries. Electroenceph Clin Neurophysie,1 1986;65:86-93. 9. Mahoney FI, Baithel DW. Functional evaluation: the Barthel Index. MD Med Journal 1965; 14:61-5. 10. American Spinal Injury Association. Standards for neurological classification of spinal injury patients. Chicago: ASIA, 1990. 11. Prevec TS. Effect of valium on the somatosensory evoked potentials. In: Destedt JE, editor. Clinical uses of cerebral, brainstem, and somatosensory evoked potentials. Basel: Karger, 1980:311-8. 12. Herbison GJ, Zerby SA, Cohen ME, Marino RI, Ditunno Jr JF. Motor

641

13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

23. 24. 25. 26. 27. 28.

power differences within the first two weeks post-SCI in cervical spinal cord injured subjects. J Neurotrauma 1992;9:373-80. Herman R. The myotatic reflex: clinico-physiological aspects of spasticity and contractures. Brain 1970;93:273-3t2. Chehrazi B, Wagner FC, Collins WF, Freeman DH. A scale for evaluation of spinal cord injury. J Neurosurg 1981;54:310-5. Hussey RW, Stauffer ES. Spinal cord injury: requirements for ambulation. Arch Phys Med Rehabil 1973;54:544-7. DeLisa JA, Mackenzie K, Baran EM. Manual of nerve conduction velocity and somatosensory evoked potentials. New York: Raven Press, 1987:151-200. American Encephalographic Society. Guidelines for clinical evoked potential studies. J Clin Neurophysiol 1984; 1:3-53. Blight AR. Cellular morphology of chronic spinal cord injury in the cat: analysis of myelinated by line-sampling. Neurosci 1983; 10:52142. Blight AR, Decresito V. Morphometric analysis of experimental spinal cord injury in the cat. The relationship of injury and intensity to survival of myelinated axons. Neuroscience 1986; 19:321-41. Blight AR. Spinal cord injury models: neurophysiology. J Neurotrauma 1992; 2:147-50. Bresnahan JC, Beattie MS, Stokes BT, Conway KM. Three dimensional computer assisted analysis of graded contusion lesions in the spinal cord of the rat. J Neurotrauma 1990;8:91-101. Fehlings MG, Tator CH, Linden RD. The relationships among the severity of spinal cord injury, motor and somatosensory evoked potentials and spinal cord blood flow. Electroenceph Clin Neurophysiol 1989; 74:241-59. Chiappa K, editor. Evoked potentials in clinical medicine. New York: Raven Press, 1990:372-3. Carpenter MB. Human neuroanatomy. Baltimore: Williams & Wilkins, 1976:259, 271. Jellinger K. Neuropathology of cord injuries. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology. Injuries of the spine and spinal cord, part 1. New York: American Elsevier, 1976:43-121. Quencer RM, Bunge RP, Egnor M, Green BA, Puckett W, Naidich TP, et al. Acute traumatic central cord syndrome: MRI-pathological correlations. Diag Neuroradiol 1992;34:85-94. Houlden DA, Schwartz ML, Klettke KA. Neurophysiologic diagnosis in uncooperative trauma patients: confounding factors. J Trauma 1992;33:244-51. Kovindhaa, Mahachai R. Short latency somatosensory evoked potentials of the tibial nerve in spinal cord injuries. Paraplegia 1992; 30:5026.

Suppliers
a. Nicolet Compact System; 5225 Verona Road, Madison, WI 53711-4495.

Arch Phys Med Rehabil Vol 76, July 1995