By using motor evoked potential (MEP) created by transcranial electric stimulation over the motor cortex and F-wave measurement from the peripheral nerve stimulation, it is possible to estimate the spinal cord motor conduction velocity (SCMCV) in the diseased state. Twenty-four patients with spinal cord injury (SCI) between T I and T11 neurological levels participated in this study. MEP in leg muscle was absent in all neurologically complete paraplegics. In 16 patients with neurologically incomplete SCI, MEP was obtained in 13 patients. The SCMCV estimated from C7 to T I 2 spinal levels was 32.1 (SD = 9.4) m/s. This was significantly slower than 63.3 (SD = 8.6) m/s in 40 normal controls. This noninvasive, indirect method is measurable, and can provide valuable electrophysiologicaldata in the assessment of motor function in patients with SCI. Key words: Spinal cord motor conduction motor evoked potential F wave spinal cord injury MUSCLE & NERVE 14:990-996 1991
ESTIMATE OF MOTOR CONDUCTION IN HUMAN SPINAL CORD: SLOWED CONDUCTION IN SPINAL CORD INJURY CHEIN-WE1 CHANG, MD, and I-NAN LIEN, MD
Many electrophysiological assessments have been used for the evaluation of spinal cord dysfunction. Most of the studies consisted of somatosensory evoked potentials (SSEPs) in assessing spinal cord integrit and follow-up clinical evaluation. 1,16?17,30,43 The SSEP is primarily an estimate of dysfunction of afferent sensory pathways in the dorsal column.’2 It has limited use in the detection of efferent motor pathways of the spinal cord. In 1980, Merton and Morton2’ developed a noninvasive technique for obtaining the motor evoked potential (MEP) by transcranial electric stimulation of the motor cortex. This technique has since been used for the clinical evaluation of electrophysiolo ical function of the central motor pathwaj9,p1,36,42 and injured spinal Cord.5,1?,8 24 32 1,44
From the Department of Physical Medicine and Rehabilitation, National Taiwan University, College of Medicine, Taipei, Taiwan, Republic of China. Acknowledgments: This study was supported by the National Science Council, ROC (grant NSC 77-0412-8002-154). Address reprint requests to Chein-Wei Chang, MD, Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, No. 1, Chang-Te Street, Taipei 10016, Taiwan, ROC. Accepted for publication October 9, 1990. CCC 0148-639X/91/0100990-08 $04.00 0 1991 John Wiley & Sons, Inc.
990
Motor Spinal Cord Conduction in SCI
Invasive methods developed by Levy and Y 0 1 - k ~and ~ Boyd and colleagues’ were available for measurement of spinal cord motor conduction velocity (SCMCV). Also, several noninvasive methods have been introduced. Marsden26 and Merton,27 as well as Rothwell and colleague^,^' used MEP studies utilizing scalp and spinal stimulations. They computed a “central latency” by subtracting peripheral nerve conduction time from the total time from the motor cortex to the limb muscle. Robinson and colleagues34 developed a measurement of central motor conduction time by subtracting peripheral nerve conduction time from the latency study of F waves. By application of such a technique, the conduction velocities within the pyramidal tract from motor cortex to spinal cord could be estimated. However, this estimation of SCMCV is somewhat inaccurate, because it is not known whether the measurement of the central conduction distance represents both a corticomedullary pathway and the spinal cord conduction route. Recently, Swash and S n o o k ~and ~ ~Ingram2’ , ~ ~ developed a method for the measurement of SCMCV by using 2 stimulation sites on the spinal cord, and recording from the lower leg or from the pelvic floor muscles. This method of stimulation is simple, but it needs a high stimulus intensity (800 to 1500 V) to activate the anatomically deep-seated spinal
MUSCLE K NERVE
October 1991
cord. Also, the lower stimulation site of the spinal cord could easily be confused with nerve root stimulation. Berger and Shahani7 introduced a method for determining SCMCV. They used a monopolar needle to stimulate C5 spinal cord, and recorded the muscle action potential (MAP) from the ipsilateral anterior tibia1 (AT) muscle. They defined the mean conduction index as the distance from C5 to L4 divided by the central conduction time obtained from the cord stimulation and motor conduction time of the peroneal nerve determined from F-wave latency. Their measurement of the cord distance from C5 to L4 was controversial. Also high-intensity stimuli (usually 250 V, with duration of 200 ps to 1 ms) were required for near-cord stimulation. Needle stimulation using such high intensities may cause nerve or spinal cord damage and should be used with caution.33 Although MEP studies for evaluation of spinal cord dysfunction have been available for many years, no reliable noninvasive technique for the measurement of the SCMCV has been developed. In this study, we describe our method for determination of SCMCV using transcranial electric stimulation and F-wave latencies from the peripheral nerves. In the following section, we assess its value in evaluating motor dysfunction in patients with SCI.
SUBJECTS AND METHODS
Twenty-four patients with tramatic spinal cord injury (SCI) between T 1 and T11 participated in this study. There were 16 men and 8 women, ranging from 17 to 45 years (mean 33 years) in age. Their body heights were between 154 and 176 cm (mean 166 cm). According to Frankel's classification for neurological function in SCI, l9 8 patients were categorized as scale A with complete loss of both motor and sensory functions below the segmental level involved. The other 16 patients had incomplete SCI. One patient was categorized as scale B with some sensation present below the level of the lesion, but the motor paralysis was complete at that level. Eight patients were scale C with some muscle power present below the lesion, but it was of no practical use to the patient. Four patients were scale D with useful muscle power below the level of the lesion. They can move the lower extremities and can walk with or without aids. Three patients were scale E with complete recovery of neurological functions. There was no muscle weakness, sensory loss, or
Motor Spinal Cord Conduction in SCI
sphincter disturbance, but they presented abnormal reflexes. Otherwise, 40 normal subjects, matched for age, served as controls. The motor cortex was stimulated using a Digitimer D-180 cortical stimulator connected to a Medelec MS 92a EMG machine. The stimulator can deliver a single electric shock up to 750 V with a constant decay time of 50 ps and peak current up to 1200 mA. For activation of the contralateral abductor pollicis brevis (APB) muscle, the cathode was placed at the vertex, and the anode placed 6 to 7 cm lateral to a line drawn from the vertex to the external auditory meatus. For activation of the AT muscle, the anode was placed at the vertex and the cathode placed 6 cm anteriorly. For recording the compound muscle action potential, paired electrodes were placed on the bellytendons of APB and AT muscles. Sweep speed of the EMG oscilloscope was at 10 ms/division, and amplifier bandwidth was between 10 and 10 kHz. T o obtain a maximal amplitude of MAP, stimuli were applied by using slowly increased intensity together with a small voluntary muscle contraction (less than 20% of maximum force measured by a dynamometer). The maximal stimulus intensity was usually
ESTIMATE OF MOTOR CONDUCTION IN HUMAN SPINAL CORD: SLOWED CONDUCTION IN SPINAL CORD INJURY CHEIN-WE1 CHANG, MD, and I-NAN LIEN, MD
Many electrophysiological assessments have been used for the evaluation of spinal cord dysfunction. Most of the studies consisted of somatosensory evoked potentials (SSEPs) in assessing spinal cord integrit and follow-up clinical evaluation. 1,16?17,30,43 The SSEP is primarily an estimate of dysfunction of afferent sensory pathways in the dorsal column.’2 It has limited use in the detection of efferent motor pathways of the spinal cord. In 1980, Merton and Morton2’ developed a noninvasive technique for obtaining the motor evoked potential (MEP) by transcranial electric stimulation of the motor cortex. This technique has since been used for the clinical evaluation of electrophysiolo ical function of the central motor pathwaj9,p1,36,42 and injured spinal Cord.5,1?,8 24 32 1,44
From the Department of Physical Medicine and Rehabilitation, National Taiwan University, College of Medicine, Taipei, Taiwan, Republic of China. Acknowledgments: This study was supported by the National Science Council, ROC (grant NSC 77-0412-8002-154). Address reprint requests to Chein-Wei Chang, MD, Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, No. 1, Chang-Te Street, Taipei 10016, Taiwan, ROC. Accepted for publication October 9, 1990. CCC 0148-639X/91/0100990-08 $04.00 0 1991 John Wiley & Sons, Inc.
990
Motor Spinal Cord Conduction in SCI
Invasive methods developed by Levy and Y 0 1 - k ~and ~ Boyd and colleagues’ were available for measurement of spinal cord motor conduction velocity (SCMCV). Also, several noninvasive methods have been introduced. Marsden26 and Merton,27 as well as Rothwell and colleague^,^' used MEP studies utilizing scalp and spinal stimulations. They computed a “central latency” by subtracting peripheral nerve conduction time from the total time from the motor cortex to the limb muscle. Robinson and colleagues34 developed a measurement of central motor conduction time by subtracting peripheral nerve conduction time from the latency study of F waves. By application of such a technique, the conduction velocities within the pyramidal tract from motor cortex to spinal cord could be estimated. However, this estimation of SCMCV is somewhat inaccurate, because it is not known whether the measurement of the central conduction distance represents both a corticomedullary pathway and the spinal cord conduction route. Recently, Swash and S n o o k ~and ~ ~Ingram2’ , ~ ~ developed a method for the measurement of SCMCV by using 2 stimulation sites on the spinal cord, and recording from the lower leg or from the pelvic floor muscles. This method of stimulation is simple, but it needs a high stimulus intensity (800 to 1500 V) to activate the anatomically deep-seated spinal
MUSCLE K NERVE
October 1991
cord. Also, the lower stimulation site of the spinal cord could easily be confused with nerve root stimulation. Berger and Shahani7 introduced a method for determining SCMCV. They used a monopolar needle to stimulate C5 spinal cord, and recorded the muscle action potential (MAP) from the ipsilateral anterior tibia1 (AT) muscle. They defined the mean conduction index as the distance from C5 to L4 divided by the central conduction time obtained from the cord stimulation and motor conduction time of the peroneal nerve determined from F-wave latency. Their measurement of the cord distance from C5 to L4 was controversial. Also high-intensity stimuli (usually 250 V, with duration of 200 ps to 1 ms) were required for near-cord stimulation. Needle stimulation using such high intensities may cause nerve or spinal cord damage and should be used with caution.33 Although MEP studies for evaluation of spinal cord dysfunction have been available for many years, no reliable noninvasive technique for the measurement of the SCMCV has been developed. In this study, we describe our method for determination of SCMCV using transcranial electric stimulation and F-wave latencies from the peripheral nerves. In the following section, we assess its value in evaluating motor dysfunction in patients with SCI.
SUBJECTS AND METHODS
Twenty-four patients with tramatic spinal cord injury (SCI) between T 1 and T11 participated in this study. There were 16 men and 8 women, ranging from 17 to 45 years (mean 33 years) in age. Their body heights were between 154 and 176 cm (mean 166 cm). According to Frankel's classification for neurological function in SCI, l9 8 patients were categorized as scale A with complete loss of both motor and sensory functions below the segmental level involved. The other 16 patients had incomplete SCI. One patient was categorized as scale B with some sensation present below the level of the lesion, but the motor paralysis was complete at that level. Eight patients were scale C with some muscle power present below the lesion, but it was of no practical use to the patient. Four patients were scale D with useful muscle power below the level of the lesion. They can move the lower extremities and can walk with or without aids. Three patients were scale E with complete recovery of neurological functions. There was no muscle weakness, sensory loss, or
Motor Spinal Cord Conduction in SCI
sphincter disturbance, but they presented abnormal reflexes. Otherwise, 40 normal subjects, matched for age, served as controls. The motor cortex was stimulated using a Digitimer D-180 cortical stimulator connected to a Medelec MS 92a EMG machine. The stimulator can deliver a single electric shock up to 750 V with a constant decay time of 50 ps and peak current up to 1200 mA. For activation of the contralateral abductor pollicis brevis (APB) muscle, the cathode was placed at the vertex, and the anode placed 6 to 7 cm lateral to a line drawn from the vertex to the external auditory meatus. For activation of the AT muscle, the anode was placed at the vertex and the cathode placed 6 cm anteriorly. For recording the compound muscle action potential, paired electrodes were placed on the bellytendons of APB and AT muscles. Sweep speed of the EMG oscilloscope was at 10 ms/division, and amplifier bandwidth was between 10 and 10 kHz. T o obtain a maximal amplitude of MAP, stimuli were applied by using slowly increased intensity together with a small voluntary muscle contraction (less than 20% of maximum force measured by a dynamometer). The maximal stimulus intensity was usually
