0148-396X/85/1702-0253$02.00/0 NEUROSURGERY Copyright © 1985 by the Congress of Neurological Surgeons Clinical and laboratory reports Vol. 17, No. 2, 1985 Printed in U.S.A. Evaluation of Spinal Cord Stimulation for Postapoplectic Spastic Hemiplegia Saburo Nakamura, M.D., and Takashi Tsubokawa, M.D. Department of Neurosurgery, Nihon University School of Medicine, Tokyo, Japan We used spinal cord stimulation (SCS) in patients with postapoplectic spastic hemiplegia in an attempt to reduce the spasticity. Three patients with spastic hemiplegia due to apoplexia were selected for the treatment. Reduction of spasticity was observed 3 to 9 days after the stimulation. Electrophysiological evaluation of the spasticity from the H reflex revealed a remarkable improvement in all three patients. The mechanism of reduction of spasticity has not yet been clarified, although a direct or indirect effect on the reticulospinal tract is thought to play a role. (Neurosurgery 17:253- 259, 1985) Key words: Apoplexy, Hemiplegia, H reflex, Spasticity, Spinal cord stimulation Spasticity involves increased muscle tone, which is related to an increased rate of sensitivity of the stretch reflex and is one of the characteristic symptoms of pyramidal tract disor- ders. According to Bobath (2), all hemiplegia in adult cases of cerebral vascular disorder, brain tumor, or head injury is accompanied by spasticity of various degrees. When the spas- ticity is significantly strong, spontaneous or passive movement is restricted and the activity of the patient is disturbed. Various physical, pharmacological, and surgical procedures have been used as treatments for spasticity (12, 16, 21). In recent years, physiological treatments including cerebellar or spinal cord stimulation have also been developed (4, 6, 8). Such methods, involving modification of motor function by the stimulation of sensory pathways, have been applied in a large number of patients with multiple sclerosis. Cook and Weinstein (7) and Dooley and Sharkey (14) reported that there was improvement of spasticity in approximately 60% of their patients. In the present paper, the effect of spinal cord stimulation (SCS) in patients with spastic hemiplegia due to apoplexy is evaluated and discussed. MATERIALS Three cases were selected from among patients with spastic hemiplegia caused by hypertensive intracerebral hemorrhage or cerebral infarction. The spasticity had been present for more than 2 months without improvement and was thought to obstruct the physical treatment. METHODS Radiofrequency stimulation In all patients, an alternating pulse stimulating system (Medtronic, Inc., Minneapolis, Minnesota) was implanted. This system used an external stimulation transmitter with an antenna and an implanted radio receiver connected to two 253 electrodes situated epidurally over the midline of the posterior column at about C-5 and C-7 (Fig. 1). The parameters of stimulation were as follows: a range of voltages from 1.0 to 1.5 V, a pulse width of 0.3 to 0.5 ms, and a frequency of 350 to 100 Hz. The stimulation was applied continuously for 12 to 14 hours during daytime every day. Assessment of spasticity As an objective evaluation of spasticity, H reflex studies were performed. In all patients, the H reflex was obtained by percutaneous stimulation of the posterior tibial nerve at the popliteal fossa. The reflex was recorded with 2 silver disc surface electrodes, with the active electrode on the soleus muscle and the reference electrode on the Achilles tendon. The electrical stimulation was given as single impulses at a voltage of 100 to 150 V and a 0.5-ms duration (Nihonkoden Stimulation Unit SEM-4101; Nihonkoden, Tokyo, Japan), and the electrical responses (magnified and monitored on a Nihonkoden Oscilloscope VC-9) were recorded. The reflex studies were performed in the basal condition (before SCS stimulation) as well as during SCS. The following items were analyzed: (a) the threshold eliciting the H reflex, (b) the ratio between the responses H maximum and M maximum (H/M ratio), and (c) the recovery cycle of the H reflex. Concerning the last item, a paired shock (H1, H2) technique was used with the conditioning stimulation (H1) at the same intensity as that used for the test stimulation (H2). The interval between the paired shocks varied between 10 and 1000 ms. At each interval, 10 responses were averaged. Clinical evaluations Functional evaluations and neurological examinations were based on the following: (a) strength and muscle tone, (b) vigor of deep tendon reflexes, (c) duration of clonus in muscles, (d) range of active and passive movements in the extremities, (@) speed of gait (walking time taken over a 10-m distance) and patient’s walking posture, and (/) evaluation of motor func- tion of the extremities by Brunnstrom’s test (3). 254 NAKAMURA AND TSUBOKAWA Fic. 1. Electrodes of Medtronic’s SCS system implanted at the lower cervical spinal cord extradurally. Lefi, anteroposterior view: right, lateral view. RESULTS The clinical course and therapeutic effects are described for each case individually. Case 1 This 74-year-old man was hospitalized elsewhere on March 7, 1981, for left-sided hemiplegia. The computed tomographic (CT) diagnosis was cerebral infarction in the right putamen and the posterior limb of the internal capsule (Fig. 2). The patient began physical exercise on April 21, 1981. He regained the ability to move his leg slightly and was able to sit by himself at the end of April. After August 20, 1981, spasticity appeared in the left extremities and motor function of the leg was suppressed. His upper extremity had lost its motor func- tion completely immediately after the stroke. The patient was admitted to our hospital to undergo sys- temic exercise on September 1981. His spasticity showed no improvement after antispastic medication and functional training for 6 months. SCS was begun on March 15, 1982. On the 4th day, the spasticity of the lower extremity was reduced and stepping forward improved remarkably. On the 7th day, the spasticity of the upper extremity was also reduced and extension at the elbow joint widened from 90° to 130°. Neurological exami- nations revealed that knee clonus disappeared on the 122nd day and ankle clonus lessened after 7 months. His speed of gait, which had recovered to as fast as SO seconds/10 m before stimulation, increased to 28 seconds/10 Neurosurgery, Vol. 17, No. 2 Fic. 2. Case 1. CT scan shows an infarction in the right internal capsule. August 1985 m after 2 months. 23.5 seconds/10 m after 7 months. and 15 seconds/10 m after 2 years. However. the Brunnstrom test revealed no remarkable improvement (Table 1). On electrophysiological sssment (Fig. 3), the H/M ratio, which was calculated as 0.85 before stimulation, fell to 0.77 on the 68th day and was 0.80 on the | 73rd day. The recovery cycle of the H reflex showed its delay as 50 ms. recovery time as 100 ms. and recovery rate as 120% before treatment. On the 68th day, the recovery time was prolonged to 170 ms and the recovery rate had decreased to 110%, whereas the delay was prolonged to 75 ms on the 1 73rd day. Case 2 This 47-year-old man suddenly complained of headache, right-sided hemiplegia, and dysarthria on April 20, 1981. He was hospitalized elsewhere with right-sided hemiplegia. The CT diagnosis was pontine hemorrhage (Fig. 4). At the begin- ning of May 1981, spontaneous movement of his right leg was recognized. The patient became able to stand by himself SCS FOR SPASTIC HEMIPLEGIA 255 on May 15, 1981, and could lift his right hand slightly at the beginning of June 1981. He was transferred to our hospital for systemic exercise on June 16, 1981. After 6 months, his gait was improved slightly. but severe spasticity appeared in his right extremities and the hand lost its function completely. SCS was begun on May 4, 1982. From the 9th day, the spasticity in the right upper extremity was reduced, movement of the extremities when he walked became smooth, and his walking posture improved. His speed of gait was relatively fast, being measured as 10 seconds/10 m, and increased to 8 seconds/10 m at the 19th month after SCS. The Brunnstrom test showed slight improvement from IV-2 to V-1 from the 6th day, but in the upper extremity it indicated a temporary fall to IV-2 in the 5th month, followed by a return to VI-I in the 6th month (Table 2). The H/M ratio was 1.05 before stimulation and fell to 0.87 on the 17th day and 0.75 on the 122nd day (Fig. 5). The recovery cycle of the H reflex showed its delay as 25 ms, TABLE | Data for Case I* 7 1/23/82 3/4 4/6 5/15 10/7 4/14/83 5/18/84 Knee clonus + + = = - - - Ankle clonus + + + + co = Gait speed (sec/10 m) 50 - = 28 23.5 15 15 Brunnstrom stage U/E II-2 IH-2 Ill-2 III-2 II-2 IIl-2 TII-2 L/E IL-3 HI-3 III-3 1-3 III-3 IL-3 TI-3 U/E, upper extremities: L/E. lower extremities. 150 “Patient | is a 74-year-old man. Onset. March 4. 1981. left-sided hemiplegia. Diagnosis, cerebral infarction. SCS started on March 15, 1982. pre-SCS post-SCS(68D) 100 7-----F--- post-SCS(173D) H/M Ratio 50 Pre—SCS Post—SCS 0.85 0.77 0.80 (68D) (173D) 100 1000 500 msec Fic. 3. Case 1. Recovery curves of H waves and H/M ratio. H/M ratio: quotient M maximum/H maximum: pre-SCS, before spinal cord stimulation: post-SCS. after spinal cord stimulation: D. day. 256 NAKAMURA AND TSUBOKAWA recovery time as 40 ms, and recovery rate as 155% before stimulation. which indicated remarkably active spasticity. On the 17th day after stimulation. these data showed improve- ment. The delay was measured at as long as 75 ms, the recovery time was 125 ms. and the recovery rate was reduced Fic. 4. Case 2. CT scan shows localized hemorrhage in the pons. The white line is an artifact. Neurosurgery, Vol. 17, No. 2 to 110%. On the 122nd day, there was further improvement, with delay prolonged to 90 ms and the recovery time pro- longed to 200 ms. Case 3 This 47-year-old man had a 3-year history of hypertension controlled with medication. On October 11, 1981, he suffered an episode of disturbed consciousness with right-sided hemi- plegia. He was taken to a neighboring hospital in a drowsy state with internuclear ophtalmoplegia. The CT diagnosis was pontine hemorrhage. Ten days after this episode, the patient’s consciousness gradually improved. He regained the ability to move his right hand slightly and was able to move his right leg at the end of November 1981. The patient could raise himself and sit down. On January 1, 1982, he could walk with a cane for 5 to 6 steps. He was transferred to our hospital to undergo systemic exercise on March 15, 1982. In August, his right extremities TABLE 2 Data for Case 2¢ Z 2/26/82. 4/55/10 10/7 12/2/83 4/13/84 Knee clonus - - - - = = Ankle clonus - - - = = = Gait speed (sec/10m) 10 95 10 10 8 8 Brunnstrom stage U/E V-L 0 V-1 V-1TV-2) V-1 V-1 L/E IV-2. 1V-2 IV-2 V-1 V-1 V-1 “ Patient 2 is a 47-year-old man. Onset, April 20, 1981. right-sided hemiplegia. Diagnosis, pontine hemorrhage. SCS started on March 4, 1982. U/E, upper extremities; L/E, lower extremities. % 150 1oo}-4----- | --- AL post-SCS(17D) post-SCS(122p) H/M Ratio 50 Pre—SCS Post—SCS 1.05 0.87 0.75 | (17D) (122D) —$__ a 100 500 Fic. S. Case 2. Recovery curves of H waves and H/M ratio. August 1985 ceased to improve in motor function and their function was estimated as IV-2 by the Brunnstrom test. SCS was begun on December 3, 1982. A reduction in the spasticity of the right extremities, especially the upper extrem- ity, was noted 3 days after the stimulation. On the 13th day. however, his speed of gait dropped to 180 seconds/10 m and his motor function was reduced to Grade IV-1 on the Brunns- trom test. Nevertheless, the gait speed returned to the prestim- ulation level on the 48th day and increased to 73 seconds/10 m the 88th day, with the Brunnstrom test evaluation as Grade IV-2 (Table 3). The effect of SCS appeared about 20 minutes after the start of SCS in the morning as a reduction of “stiffness of the extremities.” It continued for 30 minutes after turning-off of the SCS at night and was then gradually reduced. By electro- physiological assessment (Fig. 6), the H/M ratio was 1.23 before stimulation, 1.05 on the Sth day, and 0.82 on the 154th day after stimulation. The recovery cycle of the H reflex showed its delay as 50 ms, recovery time as 150 ms, and recovery rate as 120% before stimulation, which indicated TABLE 3 Data for Case 3% 10/19/82 12/16 1/21/83. 3/29 5/28 Knee clonus = = ca - = Ankle clonus = + + + + Gait speed (sec/10 m) 90 180 90 73 «7S Brunnstrom stage U/E IV-2 IV-1 IV-1 IV-2 IV-2 L/E IV-2 IV-1 IV-1 IV-2. IV-2 “Patient 3 is a 47-year-old man. Onset. October 11. 1981. right- sided hemiplegia. Diagnosis, pontine hemorrhage. SCS started on December 3. 1982. U/E. upper extremities: L/E. lower extremities. SCS FOR SPASTIC HEMIPLEGIA 257 active spasticity. On the 5th day after stimulation, the recovery rate was reduced to 100%, whereas on the 154th day the delay was estimated as 75 ms, the recovery time as 200 ms, and the recovery rate as 100%. Because this patient moved to a distant location, his status has not been evaluated since May 28, 1983. DISCUSSION It has been reported that the spasticity that appears after hypertensive intracerebral hemorrhage or cerebral infarction varies in its occurrence according to the location of the lesion. The time interval between the onset of hemiplegia and ap- pearance of spasticity varies from several days to weeks. Although spasticity is one of the characteristic pyramidal symptoms, its pathogenesis in relation to the location of the lesion, supposed participation of the extraphyramidal tract (1, 11, 24), time lag for the appearance of spasticity, and supposed participation of the sprouting phenomenon of the G-la fibers in the spinal reflex arc (19, 20) remain to be clarified. In spite of the efficacy of spasticity in maintaining the posture of the extremities, it may mask the retained muscle power, interfere with transfer activities, yield contractures, and contribute to decubitus formation. In any case, strong spasticity has disadvantages in relation to physical motion. Three different approaches in reducing spasticity have been used: pharmacological, physical, and surgical. Physical ther- apy is useful in determining the effect of therapeutic efforts. However, it has a low therapeutic efficiency. Pharmacological treatment uses drugs that are intended to reduce the spasticity. However, such medications are somewhat limited in their usefulness because of varying degrees of impairment of differ- ent muscle groups and fairly diffuse and equal effects of the treatment, which bring about a reduction in muscle tone in undesired muscle and result in a lowering of motor action. Surgical treatments are quite selective, but invasive, and be- % 150|- pre-SCS post-SCS(5pD) VQ | one gy men nnn post-SCS(154D) H/M Ratio 50 — — — Pre—SCS Post—SCS 4.23 105 0.84 0.82 _ (5D) (8D) (154D) 00 500 , * (000 msec Fic. 6. Case 3. Recovery curves of H waves and H/M ratio. tv mn 8 NAKAMURA AND TSUBOKAWA —43—| Reticulo-spinal tract Spino-reticular fibers ES Fasciculus gracilis, cuneatus, Fic. 7, Diagram of the reduction mechanism of spasticity. A direct or indirect effect on the reticulospinal tract by SCS is postulated. EF. extensor: FS, electrical stimulation: GC, nucleus reticularis gigan- tocellularis. cause of their specificity may necessitate extensive subsequent surgical intervention. Although surgical intervention may be long-lasting. its effects on the spasticity are not permanent because regeneration may occur. On the other hand, physiological treatments such as periph- eral nerve. spinal cord. and cerebellar stimulation offer non- invasive procedures that utilize the sensory input to modify the impaired muscle tone through the nervous system. SCS was initially used for the treatment of patients with multiple sclerosis by Cook and Weinstein in 1973 (7). After their report, abundant experience in the treatment of patients with multi- ple sclerosis (5. 8. 13. 17. 18, 23. 25), spinal cord injury. familial spastic paraplegia. cerebellar degeneration, amy- otrophic lateral sclerosis. pseudobulbar palsy. and cerebral palsy has been reported (22). SCS was applied here to three patients with postapoplectic spastic hemiplegia. A reduction of spasticity was recognized in all cases both clinically and electrophysiologically. In Case 1. lessening of spasticity was recognized initially in the leg both subjectively and objectively. Extension of the elbow joint was then increased. and ankle clonus became minimized and finally disappeared. In Case 2. the spasticity in the upper extremity. which was stronger than that in the lower extremity before stimulation. was remarkably diminished by the treat- ment. A similar effect was also recognized in Case 3 after stimulation. Electrophysiological assessments revealed a de- crease in the H/M ratio. normalization of the delay in recov- ery time in the recovery cycle of the H reflex. and reduction of the recovery rate in all three patients. The detailed reduction mechanism of SCS for spasticity has yet to be elucidated. It is clear from the pathophysiological findings in the present three patients that the stimulation does Neurosurgery, Vol. 17, No. 2 not exert its effect from the dorsal column to the sensory cortex and further to the pyramidal or extrapyramidal tract via the ventral posterolateral nucleus or ventral posteromedial nucleus because these tracts are already interrupted by the apoplectic lesions. As one hypothesis, therefore, we consider that the stimulation may influence the spastic muscle by an inhibitory effect via the spinoreticular tract, which terminates in the reticular formation in the medulla oblongata, or directly on the reticulospinal tract (15) (Fig. 7). Concerning the effect of SCS, Campos et al. reported that it developed relatively quickly after the beginning of stimu- lation (4). They observed an effect within 3 days after implan- tation, with continuous improvement of function for up to 3 months later. In our experience, reduction of spasticity after SCS began was recognized from the 4th day in Case 1, 9th day in Case 2, and 3rd day in Case 3, respectively. Improve- ment of spasticity was still recognized by electrophysiological assessment at 173 days in Case 1, 122 days in Case 2, and 154 days in Case 3. This suggests that continuous improve- ment of spasticity may be obtained for at least 4 months or more after the beginning of stimulation. According to the report of Campos et al. (4), stimulation was continued for 12 months. We propose, however, that stimulation should be allowed to continue until the spasticity is resolved or no further effect of the stimulation can be recognized. As regards the neurophysiological mechanisms underlying the effectiveness of SCS, Davis and his coworkers support the theory that dorsal column stimulation (DCS) increases the amount of background afferent activity in a partially deaffer- ented or weakened neuronal system (10). That is, DCS acti- vates impulses orthodromically and antidromically in the dorsal column. However, it has not been possible with SCS or DCS to reestablish lost voluntary movement of the extrem- ities. One essential purpose of the treatment is to decrease the spasticity that interferes with the residual motor function. SCS is thus not indicated in patients with complete para- plegia, hemiplegia, or quadriplegia. However, it can benefit the patient with incomplete motor palsy and spasticity by decreasing the spasticity and helping to augment the residual motor function. Although several complications such as in- fection of the system, pocket hematoma, or radio receiver failure have been reported (9), we have not experienced such complications in our patients to date. The electrode tip position for obtaining the greatest benefit from spinal cord stimulation still remains controversial, and electrode displacement has proved to be a problem using the prototype needle electrode. We hope that new developments in electrode design will overcome the problems of electrode displacement. SCS seems to be safe and yields relatively selective effects in spastic motor disorders. As a result, the technique is effec- tive for treating postapoplectic spastic palsy and can give some benefit of functional recovery in patient rehabilitation. Nevertheless, the use of SCS for such patients still requires clarification of a suitable starting time, parameter settings. and the period of stimulation. Received for publication. November 14. 1984: accepted. March 3. 1985. Reprint requests: Saburo Nakamura. Department of Neurosurgery. Nihon University School of Medicine. 30-1 Oyaguchi Kamimachi. Itabashi-ku. Tokyo 173, Japan. REFERENCES J}. Ashby P. Andrews C. Knowles L, Lance JW: Pyramidal and extrapyramidal control of tonic mechanisms in the cat. Brain 95:21-30, 1972. 2. Bobath B: Adult Hemiplegia: Evaluation and Treatment. Lon- in . Campos RJ. Dimitrij August 1985 don. Heinemann Medical Books. 1978. pp 1-5. . Brunnstrom $: Movement therapy in hemiplegia: A neurophys- iological approach. New York. Harper and Row. 1970. pp 24- Pen . Campos RJ. Dimitrijevic MM. Faganel J. Sharkey PC: Clinical evaluation of the effect of spinal cord stimulation on motor performance in patients with upper motor neuron lesions. Appl Neurophysiol 44:141-151. 1981. ijevic MM. Sharkey PC: Clinical evaluation of the effects of spinal stimulation on motor performance in patients with upper motor neurone lesion. in Proceedings of the Sixth International Symposium on the External Control of Hu man Extremities. Belgrade. Yugoslav Committee for Electronics and Automation. 1978. pp 569-574. . Cook AW. Taylor JK. Nidzgorski F: Results of spinal cord stimulation in multiple sclerosis. Appl Neurophysiol 44:55-61. 1981, . Cook AW, Weinstein SP: Chronic dorsal column stimulation in multiple sclerosis: Preliminary report. NY State J Med 73:2868— 2872. 1973. Davis R. Flitter MA. Bolton D. Komrad M: Clinical efficacy and safety of chronic spinal stimulation used in multiple sclerosis and other demyelinating diseases, in Proceedings of the Sixth Inter- national Symposium on the External Control of Human Extrem- ities. Belgrade. Yugoslav Committee for Electronics and Auto- mation. 1978, pp 557-566. . Davis R. Gray E: Technical factors important to dorsal column stimulation. App! Neurophysiol 44: 160-170, 1981. . Davis R. Gray E. Kudzma J: Beneficial augmentation following dorsal column stimulation in some neurological diseases. Appl Neurophysiol 44:37-49. 1981. . Denny-Brown D: The extrapyramidal system and postural mech- anisms. Clin Pharmacol Ther 5:812-827. 1964. . Dimitrijevic MR. Sherwood AM: Spasticity: Medical and surgical treatment. Neurology (NY) 30:19-27. 1980. . Dooley DM. Nisonson I: Treatment of patients with degenerative of the central nervous system by electrical stimulation siol 44:71-76, 1981. disease: of the spinal cord. App! Neurop| . Dooley DM. Sharkey J: Electrical stimulation of the spinal cord in patients with demyelinating and degenerative diseases of the central nervous system. Appl Neurophysiol 44:218-224, 1981. . Grillner S, Lund S: The origin of a descending pathway with monosynaptic action on flexor motoneurones. Acta Physiol Scand 74:278-284. 1968. . Gros C: Spasticity—clinical classification and surgical treatment. in Krayenbiihl H, et al (eds); Advances and Technical Standards in Neurosurgery. Wien, Springer-Verlag. 1979, vol 6, pp 55-97. . Hawkes CH. Fawcett D. Cooke ED, Emson PC. Paul EA. Bow- cock SA: Dorsal column stimulation in multiple sclerosis: Effects on bladder. leg blood flow. and peptides. Appl Neurophysiol 44:62-70. 1981. . Illis LS. Sedgwick EM, Tallis RC: Spinal cord stimulation in multiple sclerosis: Clinical results. J Neurol Neurosurg Psychiatry 43:1-114. 1980. SCS FOR SPASTIC HEMIPLEGIA 259 19. Liu CN. Chamber WW: Intraspinal sprouting of dorsal root anxons. Arch Neurol Psychiatry 79:46-61, 1958. 20. McCouch GP. Austin GM. Liu CN, Liu CY: Sprouting as a cause of spasticity. J Neurophysiol 21:205-216, 1958. 21. Siegfried J. Kranick J-U, Haas H, Adorjani C, Meyer M, Thoden U: Electrical spinal cord stimulation for spastic movement dis- orders. Appl Neurophysiol 41:134-141, 1978. 22. Siegfried J. Lazorthes Y. Broggi G: Electrical spinal cord stimu- lation for spastic movement disorders. App! Neurophysiol 44:77- 92. 1981. 23. Sharkey PC: Medical and surgical management of spasticity. Clin Neurosurg 28:589-596, 1981. 24. Tower SS: Extrapyramidal action from the cat's cerebral cortex: Motor and inhibitory. Brain 59:408-444, 1936. 25. Walz JM, Pani KC: Spinal cord stimulation in disorders of the motor system. in Proceedings of the Sixth International Svympo- sium on the External Control of Human Extremities. Belgrade. Yugoslav Committee for Electronics and Automation, 1978. pp 545-556. COMMENT The treatment of abnormalities of motor function and tone by the use of electrical stimulation has now been investigated for nearly 10 years without any definitive answers concerning the utility of the technique in neurosurgical practice. The authors produce a convincing, although anecdotal report of three patients with spastic hemiplegia who were improved. One of the major difficulties with the assessment of such patients is the lack of quantifiable measures on which every- one will agree. It is not possible to do a controlled trial because of the great variability of these diseases. The patient must serve as his or her own control. Assessment by independent personnel seems to be the best we can do now to try to verify the benefit of these stimulation procedures. Even this cannot be blinded because the patients feel something and therefore report to all those examining them concerning the success or failure of their therapy. Patient observations in such a circum- stance are notoriously inaccurate and the unconscious biases of all those involved in therapy may influence patient report- ing. There is a real need for a system of quantification to help settle the questions of the value of stimulation for disorders of motor tone. Nevertheless, this is an interesting, convincing report, and the concept of improving the spastic hemiplegic through this relatively innocuous technique certainly deserves further exploration by those with special interests in electrical stimulation of the nervous system for the modification of functional disorders. Don M. Long, M.D. Baltimore, Maryland