Journal of the Neurological Sciences, 100 (1990) 85-93
Elsevier
85
JNS 03428
The combined effects of clonidine and cyproheptadine with interactive training on the modulation of locomotion in spinal cord injured subjects J. F u n g , J . E . S t e w a r t a n d H . B a r b e a u School of Physical & Occupational Therapy McGill University,Montreal Quebec (Canada)
(Received 25 April, 1990) (Revised, received 17 July, 1990) (Accepted 20 July, 1990) Key words: Spinal cord injury; Clonidine; Cyproheptadine; Locomotion; Rehabilitation; Spasticity
Summary The combined effects of a noradrenergic agonist, clonidine, and a serotonergic antagonist, cyproheptadine, together with an interactive locomotor training program incorporating progressive body weight support and treadmill walking exercise, were investigated in two chronic spinal cord injured subjects. Both subjects had no independent locomotor ability due to severe spasticity. Kinematic, temporal distance and electromyographic (EMG) data were collected during treadmill walking. The EMG activity of the lower limb muscles, initially characterized by tonic discharge and abnormal timing, became more phasic with less clonus following medication, which was related to a change in the kinematic pattern. Further kinematic and functional improvement were gained by training. Previously wheelchair-bound, both patients became functionally ambulatory overground with the aid of Canadian crutches. Thus, a potentially effective strategy for facilitating the expression of the locomotor pattern following spinal cord injury is proposed. This preliminary study showed that such a treatment strategy could possibly lead to a recovery of locomotor function in some chronic, wheelchair-bound spinal cord injured patients who had previously been stabilized on conventional therapies.
Introduction The locomotor pattern of spinal cord injured (SCI) subjects with incomplete lesions is often impaired by spasticity as manifested by clonus, spasms, and hyperactive spinal reflexes. Disturbance in the spinal cord circuitry can give rise to disordered muscular activation patterns and abnormal reflex activities, as well as affecting the ability to cope with weight beating, speed, and balance during walking. Interventions which aim at alleviating these problems would seem beneficial to optimize the gait pattern. In addressing the problem of motor disturbance in SCI subjects, recent research studies have shown that the serotonergic (5-HT) antagonist, cyproheptadine, and noradrenergic (NA) agonist, clonidine, are both potentially effective in reducing clinical signs of spasticity (Nance et al., 1985, 1989; Tuckman et al., 1982) and improving the locomotor function of spastic paretic patients (Stewart et al., 1987; Wainberg et al., 1986, 1990). The rationale for using these new drugs stems from spinal animal studies (Btdard Correspondence to: Hugues Barbeau, PhD, Associate Professor, School of Physical & Occupational Therapy, McGill University, 3654 Drummond Street, Montreal, Quebec H3G 1Y5, Canada.
et al., 1987; Rossignol et al., 1986; Barbeau et al., 1987a; Barbeau and Rossignol, 1990). Following chronic, complete spinal cord transection in the adult cat, in which all the descending systems have degenerated, monoaminergic agents (5-HT and NA) that act on the receptors below the transection site were shown to modify spinal reflex activity and modulate the locomotor pattern. The problems of external influences have also been investigated in spastic SCI subjects during treadmill locomotion (Barbeau et al., 1988; Visintin and Barbeau, 1988). It was revealed that an increase in speed, as well as a decrease in hand-rail support for balance, can accentuate the spastic features and deteriorate the locomotor pattern. A preliminary study (Visintin and Barbeau, 1989), done on seven spastic SCI subjects, has shown that supporting a percentage of the body weight during treadmill walking can facilitate the expression of a more normal gait pattern. As the demand of loading and balance decreased, there was also an increase in speed and step length. Hence, a novel training strategy consisting of providing body weight support (BWS) during treadmill locomotion was proposed, such that neurological gait can be retrained by progressively reducing the external BWS, until the patient can walk full weight bearing on the treadmill. Such an 'interactive loco-
0022-510X/90/$03.50 © 1990Elsevier Science Publishers B.V. (BiomedicalDivision)
86 motor training' approach was found to be an important factor for the recovery of locomotion in the adult spinal cat (Barbeau and Rossignol, 1987). Thus, based on the findings in animal studies and preliminary human studies, we propose to combine the approaches of pharmacological intervention and locomotor training to achieve an optimal locomotor pattern in spastic paretic subjects. It is the intent of the present study, to determine whether it is possible, and to what extent, that locomotor function can be recovered in SCI patients who previously had no locomotor capability despite administration of conventional antispastic medication and rehabilitation. As a first step, two chronic SCI patients, classified as Frankel's category C, having incomplete sensation and no useful motor function below the level of lesion, were included in this preliminary study. The potential effects of combining the action of cyproheptadine and clonidine, with an interactive gait training program that incorporates both progressive weight bearing and treadmill walking exercise, were examined. Once the use of these novel strategies is justified, the long-term goal is to extend the study to a large population of SCI patients.
Methods
This preliminary study included two young and motivated wheelchair-bound male SCI subjects presenting with symptoms of spasms and clonus at rest. Both subjects (S 1 : 26 years old, T4_7, 11 months post-lesion; $2:23 years old, C7_8, 8 months post-lesion) had sustained incomplete spinal cord injury due to motor vehicle accident. Except for a grade 2 muscle power in the right hip and knee flexors, S 1 had no apparent voluntary movement in the lower extremities. It is possible that existing voluntary movement could be masked by severe spasticity. $2 also had no voluntary movement or motor control in either lower limb, as tested in various resting positions (lying and side lying). In both subjects, especially $2, voluntary effort exerted in the resting position resulted in a marked extensor spasm which could last up to 10s. Limited by such severe spasticity, the subjects could stand with support but were unable to take steps independently overground. Despite diminished tactile sensation and decreased pressure sensitivity in either lower limb, exaggerated flexion withdrawal of the lower limbs frequently occurred in both subjects with the pin-prick test. In the present study, only the more impaired lower extremity, which had no apparent voluntary control and increased tonic stretch reflexes as well as augmented spasms and clonus, was chosen for evaluation (i.e., left lower limb for S 1 and right lower limb for $2). Both patients had been stabilized on another antispastic medication (baclofen) and were discharged from rehabili-
tation after reaching a plateau stage in their locomotor function before entering the study. Prior to the experimental evaluation, each patient attended an orientation session in the gait laboratory to become familiarized with the assessment procedures. Adequate habituation to the treadmill and body weight support system was given to ensure that any changes observed in gait outcomes were due to experimental intervention rather than the acclimatization process. The experimental design consisted of an initial double blind randomized crossover drug trial which compared the effects of each medication with placebo, followed by the combination of both cyproheptadine (S 1 & $2:24 rag/day) and clonidine (SI: 0.175 mg/day; $2:0.20 mg/day). The individual effects of cyproheptadine (Wainberg et al., 1986, 1990) and clonidine (Stewart etal., 1987) have been reported earlier. This study emphasizes the combined effects of both medications. The control evaluation was conducted while the patients were on placebo medication. The medication was then administered orally, starting with 6 mg/day of cyproheptadine, increasing systematically to 24 mg/day over a 2-week period. The patients were stabilized on this optimal dose for a further two weeks. The side effects of cyproheptadine included an increase in appetite and drowsiness at the time of dosage increase. In the subsequent stage, clonidine was added, starting with an initial daily dosage of 0.05 mg/day and increasing systematically over another two-week period to an optimal dosage of 0.175-0.20 mg/day which produced minimal side effects. The side effects included dryness of eyes and mouth, lethargy, and dizziness, but the range of dosage administered (0.05-0.20 mg/day) was well below that prescribed for anti-hypertensive purposes. Both patients continued on baclofen therapy, and were stabilized on the optimal dosages of the combined medications for at least 2weeks before participating in the post-medication evaluation. While on the combined medications, the patients participated in an interactive locomotor rehabilitation program, in which gait retraining was performed on a treadmill, with the subject mechanically supported in a comfortable overhead suspension harness. External body weight support (BWS) could be provided, with the proportion (0-100~o) calibrated and determined by a strain gauge transducer. This BWS system, which permits gait evaluation of the severely disabled subjects who could not walk independently overground, has been described in detail previously (Barbeau et al., 1987b). Manual assistance could be provided as necessary to aid stepping movements of the lower limb. S1 was trained 3 sessions per week for 2 months, and $2, 5 sessions per week for 3 weeks. Each session lasted 1-2 h, and was divided into walking trials of variable duration, depending on the subject's tolerance and endurance. Optimal BWS was provided and progressively
87 reduced until the subject could walk full weight bearing (0 BWS), coping comfortably with the adjustable treadmill speed. The post-training evaluation was carried out at the end of the training period. Care was taken to ensure that the patient had an empty bladder during each of the evaluations. Electromyographic (EMG), temporal distance, and kinematic data were collected during treadmill locomotion before and after each medication period, as well as pre- and post-training. E M G were recorded from the lower extremity muscles: medial hamstrings (MH), vastus lateralis (VL), tibialis anterior (TA), and medial gastrocnemius (GA). Footswitches placed beneath the heel, ball, and big toe region of each foot provided data on temporal aspects of the gait cycle. The movement of the trunk and lower extremity in the sagittal plane was videotaped simultaneously. Data analysis was performed off-line to normalize the EMG data to the gait cycle, from the initial foot floor contact (0~o) to the subsequent one (100~o), producing an ensemble of 6-10 consecutive cycles (Fig. 1A, B, C). An ensemble average (Fig. 3B and D) was generated with EMG values averaged across the ensemble for every 0.4~o
A. Placebo
B. Postmedication
C. Posttraining
100 # V
0 100 °70 GAIT CYCLE Fig. 1. An example of an EMG ensemble processed in (A) placebo, (B) post-medication, and (C)post-training evaluations. The example shownis the gastrocnemiusmuscleof $2 normalizedto the gaitcycle.The solid line across the cycles indicates stance-swingtransition.
of the gait cycle. The raw EMG amplitude was further normalized to the ensemble average peak (100~o) for between-subject or between-session comparison. Temporal gait descriptors, such as stance, swing, and double support durations, were also normalized to 100~o of the gait cycle for comparison. The trunk, hip, knee, and ankle angular displacements were measured directly from the monitor screen at each 5 ~ of one representative gait cycle. The cycle chosen for kinematic analysis was selected within the cycles digitized for EMG analysis (Fig. 1).
Results and discussion
The video pictures and joint angular displacement profiles of S 1 and $2, during each evaluation, are shown in Figs. 2 and 3 (A & C) respectively. During the placebo evaluation (Fig. 2A and D; Fig. 3A and C, open diamonds), both subjects required at least 50~ BWS and manual assistance to take steps on the treadmill even at a minimal speed of 0.26 m . s - 1 . Their gait pattern was characterized by a flexed posture, especially in the trunk for S I (Figs. 2A and 3A), and in the hip and knee for $2 (Figs. 2D and 3C), indicating the inability to efficiently bear weight on the lower extremity and to maintain an upright posture while advancing on the treadmill. Furthermore, both subjects frequently encountered flexor spasms as soon as the foot was lifted off the treadmill, giving rise to an excessive flexion in the hip and knee during swing (S 1 : 85 ° and 105 °, $2:45 ° and 85 °, respectively). The initial footfloor contact was made with the toe, as indicated by the flexed hip and knee, as well as the plantarflexed (S1) or neutral ($2) ankle joint. For S1, there was a tendency towards hip and knee extension in early stance (10~o of the gait cycle), but yielded to flexion immediately on weight acceptance (see arrow with asterisk). Although the hip progressively underwent extension until lift off, the knee remained flexed and ankle plantarflexed throughout stance. For $2, the hip and knee remained flexed in early stance, with the knee and ankle progressively yielding more into flexion and plantarflexion respectively throughout the rest of stance, showing a total incapacity to cope with weight acceptance. As a result of the flexor spasm, the mean stance-swing transition occurred as early as 54.1~o (+8.4~o, n = 10) of the gait cycle for S1, and 44.3~o (+ 12.7~, n = 9) for $2. Following the administration of clonidine in combination with cyproheptadine, both subjects gained the ability to walk unassisted, independently on the treadmill at the same speed of 0.26 m. s- 1. S1 still required 20~o BWS, whereas $2 could walk full weight bearing at 0~o BWS. With the improved ability to cope with weight, both subjects assumed an upright posture with a near neutral trunk angu-
88
A
S1
B
C
D
$2
E
F
Fig. 2. Representative video pictures at critical gait events of S I and $2 taken respectively during (A and D)placebo evaluation; (B and E)postmedication evaluation; and (C and F) post-training evaluation.
89
Z
t,/)
--
...7
........................
• _
: : : : _
•
-J
(/)
i
i
o .~,
i
i
~'-
~
:Á.->
~....~f:'~ : :
"t.-,
~
~....-.~'~--L
× "
"~ :
O
~
'-
. . ~
,
(
~
-~< ~
"~
,
.
0
;~
'
o
0
~
O
~ ,-, = ,=,..~
~
~-
-
~
o
~
0
c:
e~
~
~m ' o - ~~
i
°'""
o>o
'
o
o
°
°
~
~
o
"-6 ~
o
>
~
~.~'=
0
o:-
°° I
"~ '~'~" ~ ~
~
o:I7ONV ~ N r l l d l
o:I7~DNV dlH
o : I 7 O N V :J:INN
o:i7ONV :17~tNV
•
. ~. ~ ~ ,.
.~ ~ ~-.o ~
....
''"
"'..'''.''"
•
Z
'7
"
::::~
.,.I
'"
......
.
.
.
•
.
.
.
.
.
.
.
.
0
5
.:.
~
~,
~
"~
5 0 to 20~o; $2: from > 50 to 0~o). This type of disordered activation pattern in the quadriceps has been described by Knutsson (1983) as 'crutch spasticity', which has functional implications for the knee to stiffen into extension and the limb to function as a crutch, in response to weight acceptance during stance. Such an activation pattern of VL and T A was found to be modified when BWS was provided (Visintin and Barbeau, 1989). With interactive gait training, the E M G profiles (Figs. 3C and 1D, thin solid lines) remained similar to that of the post-medication evaluation. The same was observed for the index values. However, the joint angular displacement patterns, as described earlier, were further improved with training. Thus it seems that the final expression of the locomotor pattern can still be optimized even though the muscle activation patterns remain similar. It would be interesting, as a continuation of the present study, to investigate whether the effects of medications and training can still be retained when medications are gradually withdrawn. The functional profiles of both patients are summarized in Table 2. Functionally, S 1 gained the ability to walk independently overground post medication with Canadian crutches and a Klenzac brace, which was later changed to a small soft ankle brace following training. $2 could walk independently overground with a walker following medication, and with Canadian crutches following training. This is remarkable since both patients were wheelchair-bound to begin with. Moreover, both patients showed a marked improvement in their endurance for treadmill walking. From being able to take only 6-10 steps with maximal BWS and manual assistance during the placebo evaluation, both S 1 and $2 could walk for long periods (5-8 min non-stop) without BWS or manual assistance following medication and training. The two patients also demonstrated a marked decrease in spasms and clonus. Ankle clonus, as measured in the quiet sitting position, was reduced from being sus-
tained in both patients to 1-2 beats in S 1, and 5-10 beats in $2. Since both patients were on a stabilized dosage of baclofen before entering the study, the reduction in spasms and clonus, following administration of clonidine and cyproheptadine, further suggests that the two monoaminergic agents could be used as effective antispastic medications. In conclusion, we have proposed a potentially effective approach for the rehabilitation of spastic paretic gait by combining pharmacological intervention with an interactive gait retraining program. This preliminary study shows that locomotion in the lower extremities can be recovered in chronic SCI patients who have previously undergone conventional therapy and remained wheelchair-bound. The combined medications, clonidine and cyproheptadine, acting on receptors in the spinal cord, possibly modify the level of neuronal hyperexcitability (Brdard et al., 1987) and modulate the expression of the locomotor pattern, as reflected by the changes in the muscle activation pattern and timing. However, the neurophysiologic mechanisms behind the alteration in reflex activity, in relation to the locomotor pattern, have yet to be investigated. The interactive gait training, incorporating progressive weight bearing with treadmill walking exercises, can further optimize the functional gain, as reflected by the kinematic changes, through improving the patient's ability to cope with external demands such as loading and speed. We are in the process of validating this new intervention approach in SCI subjects who will be stratified according to the severity of lesion and the impairment of locomotion. Acknowledgements This project is supported by the Medical Research Council of Canada. J. Fung received a studentship from the Rick Hansen Man in Motion Legacy Fund for spinal cord research. H. Barbeau is a Chercheur-Boursier supported by the Fonds de la Recherche en Sant6 du Qurbec. The authors appreciate the contributions of Lois Finch, Martha Visintin, and Lora Salvo in the training of the patients.
References Barbeau, H. and S. Rossignol (1987) Recovery of locomotion after chronic spinalization in the adult cat. Brain Res., 412: 84-95. Barbeau, H. and S. Rossignol (1990) The effect of serotonergic drugs on the locomotor pattern and on cutaneous reflexes of the adult chronic spinal cat. Brain Res. 514: 55-67. Barbeau, H., C. Julien and S. Rossignol (1987a) The effects of clonidine and yohimbine on locomotion and cutaneous reflexes in the adult chronic spinal eat. Brain Res., 437: 83-96. Barbeau, H., M. Wainberg and L. Finch (1987b) Description and application of a system for locomotor rehabilitation. Med. Biol. Eng. Comput., 25: 341-344. Barbeau, H., J. Fung, J. Stewart and M. Visintin (1988) Impairment of spastic paraparetic gait: Implications for new rehabilitation strategies. Proceedings of the fifth Biennial Conference of Canadian Society for Biomechanics pp. 12-16.
93 B6dard, P.J., L.E. Tremblay, H. Barbeau, M. Filion, R. Maheux, C.L. Richards and T. DiPaolo (1987) Action of 5-hydroxytryptamine, substance P, thyrotropin-releasing hormone and clonidine on motorneurone excitability. Can. J. Neurol. Sci., 14: 506-509. Conrad, B., R. Benecke and H-M. Meinck (1985) Gait disturbances in paraspastic patients. In: P.J. Delwaide and R.R. Young (Eds.), Clinical Neurophysiology in Spastieity, Elsevier Sciences Publishers, Amsterdam, pp. 155-174. Fung, J. and H. Barbeau (1989) A dynamic EMG profile index to quantify muscular activation disorder in spastic paretic gait. Electroenceph. Clin. Neurophysiol., 73: 233-244. Knutsson, E. and C. Richards (1979) Different types of disturbed motor control in gait of hemiparetic patients. Brain, 102: 405-430. Knutsson, E. (1983) Analysis of gait and isokinetic movements for evaluation ofantispastic drugs or physical therapies. In: J.E. Desmedt (Ed.), Motor Control Mechanisms in Health and Diseases, Raven Press, New York, pp. 1013-1034. Nance, P.W., A.H. Shears and D.M. Nance (1985) Clonidine in spinal cord injury. J. Can. Med. Assoc., 133: 41-42. Nance, P.W., A.H. Shears and D.M. Nance (1989) Reflex changes induced by clonidine in spinal cord injured patients. Paraplegia, 27: 296-301. Pierrot-Deseilligny, E. and L. Mazieres (1985) Spinal mechanisms underlying spasticity. In: P.J. Delwaide and R.R. Young (Eds.), Clinical Neurophysiology in Spasticity, Elsevier Sciences Publishers, Amsterdam, pp. 63-76.
Rossignol, S., H. Barbeau and C. Julien (1986) Locomotion of the adult chronic spinal and its modification by monoaminergic agonists and antagonists. In: M.E. Goldberger, A. Gorio and M. Murray (Eds.), Development and Plasticity of the Mammalian Spinal Cord, Liviana Press, Padova, pp. 323-345. Tuckman, J., D.S. Chu, C.R. Petrillo and N.E. Naftchi (1982) Clinical trial of an alpha adrenergic receptor stimulating drug (clonidine) for the treatment of spasticity in spinal cord injured patients. In: N.E. Naftchi (Ed.), Spinal Cord Injuries, Spectrum Publications, pp. 133-137. Stewart, J.E., H. Barbeau and S. Gauthier (1987) The effects of clonidine on clinical spasticity and in modulation of the locomotor pattern in chronic spastic spinal cord patients. Soc. Neurosei. Abstr., 13: 353. Visintin, M. and H. Barbeau (1988) The effects of treadmill speed on spastic paraparetic gait. Soc. Neurosci. Abstr. 14: 262. Visintin, M. and H. Barbeau (1989) The effects of body weight support on the locomotor pattern of spastic paretic patients. Can. J. Neurol. Sci., 16: 315-325. Wainberg, M., H. Barbeau and S. Gauthier (1986) Quantitative assessment of the effect of cyproheptadine on spastic paretic gait: a preliminary study. J. Neurol., 233:311-314. Wainberg, M., H. Barbeau and S. Gauthier (1990) The effects of cyproheptadine on locomotion and on spasticity in spinal cord injured patients. J. Neurol. Neurosurg. Psycbiat., 53: 754-763.
Elsevier
85
JNS 03428
The combined effects of clonidine and cyproheptadine with interactive training on the modulation of locomotion in spinal cord injured subjects J. F u n g , J . E . S t e w a r t a n d H . B a r b e a u School of Physical & Occupational Therapy McGill University,Montreal Quebec (Canada)
(Received 25 April, 1990) (Revised, received 17 July, 1990) (Accepted 20 July, 1990) Key words: Spinal cord injury; Clonidine; Cyproheptadine; Locomotion; Rehabilitation; Spasticity
Summary The combined effects of a noradrenergic agonist, clonidine, and a serotonergic antagonist, cyproheptadine, together with an interactive locomotor training program incorporating progressive body weight support and treadmill walking exercise, were investigated in two chronic spinal cord injured subjects. Both subjects had no independent locomotor ability due to severe spasticity. Kinematic, temporal distance and electromyographic (EMG) data were collected during treadmill walking. The EMG activity of the lower limb muscles, initially characterized by tonic discharge and abnormal timing, became more phasic with less clonus following medication, which was related to a change in the kinematic pattern. Further kinematic and functional improvement were gained by training. Previously wheelchair-bound, both patients became functionally ambulatory overground with the aid of Canadian crutches. Thus, a potentially effective strategy for facilitating the expression of the locomotor pattern following spinal cord injury is proposed. This preliminary study showed that such a treatment strategy could possibly lead to a recovery of locomotor function in some chronic, wheelchair-bound spinal cord injured patients who had previously been stabilized on conventional therapies.
Introduction The locomotor pattern of spinal cord injured (SCI) subjects with incomplete lesions is often impaired by spasticity as manifested by clonus, spasms, and hyperactive spinal reflexes. Disturbance in the spinal cord circuitry can give rise to disordered muscular activation patterns and abnormal reflex activities, as well as affecting the ability to cope with weight beating, speed, and balance during walking. Interventions which aim at alleviating these problems would seem beneficial to optimize the gait pattern. In addressing the problem of motor disturbance in SCI subjects, recent research studies have shown that the serotonergic (5-HT) antagonist, cyproheptadine, and noradrenergic (NA) agonist, clonidine, are both potentially effective in reducing clinical signs of spasticity (Nance et al., 1985, 1989; Tuckman et al., 1982) and improving the locomotor function of spastic paretic patients (Stewart et al., 1987; Wainberg et al., 1986, 1990). The rationale for using these new drugs stems from spinal animal studies (Btdard Correspondence to: Hugues Barbeau, PhD, Associate Professor, School of Physical & Occupational Therapy, McGill University, 3654 Drummond Street, Montreal, Quebec H3G 1Y5, Canada.
et al., 1987; Rossignol et al., 1986; Barbeau et al., 1987a; Barbeau and Rossignol, 1990). Following chronic, complete spinal cord transection in the adult cat, in which all the descending systems have degenerated, monoaminergic agents (5-HT and NA) that act on the receptors below the transection site were shown to modify spinal reflex activity and modulate the locomotor pattern. The problems of external influences have also been investigated in spastic SCI subjects during treadmill locomotion (Barbeau et al., 1988; Visintin and Barbeau, 1988). It was revealed that an increase in speed, as well as a decrease in hand-rail support for balance, can accentuate the spastic features and deteriorate the locomotor pattern. A preliminary study (Visintin and Barbeau, 1989), done on seven spastic SCI subjects, has shown that supporting a percentage of the body weight during treadmill walking can facilitate the expression of a more normal gait pattern. As the demand of loading and balance decreased, there was also an increase in speed and step length. Hence, a novel training strategy consisting of providing body weight support (BWS) during treadmill locomotion was proposed, such that neurological gait can be retrained by progressively reducing the external BWS, until the patient can walk full weight bearing on the treadmill. Such an 'interactive loco-
0022-510X/90/$03.50 © 1990Elsevier Science Publishers B.V. (BiomedicalDivision)
86 motor training' approach was found to be an important factor for the recovery of locomotion in the adult spinal cat (Barbeau and Rossignol, 1987). Thus, based on the findings in animal studies and preliminary human studies, we propose to combine the approaches of pharmacological intervention and locomotor training to achieve an optimal locomotor pattern in spastic paretic subjects. It is the intent of the present study, to determine whether it is possible, and to what extent, that locomotor function can be recovered in SCI patients who previously had no locomotor capability despite administration of conventional antispastic medication and rehabilitation. As a first step, two chronic SCI patients, classified as Frankel's category C, having incomplete sensation and no useful motor function below the level of lesion, were included in this preliminary study. The potential effects of combining the action of cyproheptadine and clonidine, with an interactive gait training program that incorporates both progressive weight bearing and treadmill walking exercise, were examined. Once the use of these novel strategies is justified, the long-term goal is to extend the study to a large population of SCI patients.
Methods
This preliminary study included two young and motivated wheelchair-bound male SCI subjects presenting with symptoms of spasms and clonus at rest. Both subjects (S 1 : 26 years old, T4_7, 11 months post-lesion; $2:23 years old, C7_8, 8 months post-lesion) had sustained incomplete spinal cord injury due to motor vehicle accident. Except for a grade 2 muscle power in the right hip and knee flexors, S 1 had no apparent voluntary movement in the lower extremities. It is possible that existing voluntary movement could be masked by severe spasticity. $2 also had no voluntary movement or motor control in either lower limb, as tested in various resting positions (lying and side lying). In both subjects, especially $2, voluntary effort exerted in the resting position resulted in a marked extensor spasm which could last up to 10s. Limited by such severe spasticity, the subjects could stand with support but were unable to take steps independently overground. Despite diminished tactile sensation and decreased pressure sensitivity in either lower limb, exaggerated flexion withdrawal of the lower limbs frequently occurred in both subjects with the pin-prick test. In the present study, only the more impaired lower extremity, which had no apparent voluntary control and increased tonic stretch reflexes as well as augmented spasms and clonus, was chosen for evaluation (i.e., left lower limb for S 1 and right lower limb for $2). Both patients had been stabilized on another antispastic medication (baclofen) and were discharged from rehabili-
tation after reaching a plateau stage in their locomotor function before entering the study. Prior to the experimental evaluation, each patient attended an orientation session in the gait laboratory to become familiarized with the assessment procedures. Adequate habituation to the treadmill and body weight support system was given to ensure that any changes observed in gait outcomes were due to experimental intervention rather than the acclimatization process. The experimental design consisted of an initial double blind randomized crossover drug trial which compared the effects of each medication with placebo, followed by the combination of both cyproheptadine (S 1 & $2:24 rag/day) and clonidine (SI: 0.175 mg/day; $2:0.20 mg/day). The individual effects of cyproheptadine (Wainberg et al., 1986, 1990) and clonidine (Stewart etal., 1987) have been reported earlier. This study emphasizes the combined effects of both medications. The control evaluation was conducted while the patients were on placebo medication. The medication was then administered orally, starting with 6 mg/day of cyproheptadine, increasing systematically to 24 mg/day over a 2-week period. The patients were stabilized on this optimal dose for a further two weeks. The side effects of cyproheptadine included an increase in appetite and drowsiness at the time of dosage increase. In the subsequent stage, clonidine was added, starting with an initial daily dosage of 0.05 mg/day and increasing systematically over another two-week period to an optimal dosage of 0.175-0.20 mg/day which produced minimal side effects. The side effects included dryness of eyes and mouth, lethargy, and dizziness, but the range of dosage administered (0.05-0.20 mg/day) was well below that prescribed for anti-hypertensive purposes. Both patients continued on baclofen therapy, and were stabilized on the optimal dosages of the combined medications for at least 2weeks before participating in the post-medication evaluation. While on the combined medications, the patients participated in an interactive locomotor rehabilitation program, in which gait retraining was performed on a treadmill, with the subject mechanically supported in a comfortable overhead suspension harness. External body weight support (BWS) could be provided, with the proportion (0-100~o) calibrated and determined by a strain gauge transducer. This BWS system, which permits gait evaluation of the severely disabled subjects who could not walk independently overground, has been described in detail previously (Barbeau et al., 1987b). Manual assistance could be provided as necessary to aid stepping movements of the lower limb. S1 was trained 3 sessions per week for 2 months, and $2, 5 sessions per week for 3 weeks. Each session lasted 1-2 h, and was divided into walking trials of variable duration, depending on the subject's tolerance and endurance. Optimal BWS was provided and progressively
87 reduced until the subject could walk full weight bearing (0 BWS), coping comfortably with the adjustable treadmill speed. The post-training evaluation was carried out at the end of the training period. Care was taken to ensure that the patient had an empty bladder during each of the evaluations. Electromyographic (EMG), temporal distance, and kinematic data were collected during treadmill locomotion before and after each medication period, as well as pre- and post-training. E M G were recorded from the lower extremity muscles: medial hamstrings (MH), vastus lateralis (VL), tibialis anterior (TA), and medial gastrocnemius (GA). Footswitches placed beneath the heel, ball, and big toe region of each foot provided data on temporal aspects of the gait cycle. The movement of the trunk and lower extremity in the sagittal plane was videotaped simultaneously. Data analysis was performed off-line to normalize the EMG data to the gait cycle, from the initial foot floor contact (0~o) to the subsequent one (100~o), producing an ensemble of 6-10 consecutive cycles (Fig. 1A, B, C). An ensemble average (Fig. 3B and D) was generated with EMG values averaged across the ensemble for every 0.4~o
A. Placebo
B. Postmedication
C. Posttraining
100 # V
0 100 °70 GAIT CYCLE Fig. 1. An example of an EMG ensemble processed in (A) placebo, (B) post-medication, and (C)post-training evaluations. The example shownis the gastrocnemiusmuscleof $2 normalizedto the gaitcycle.The solid line across the cycles indicates stance-swingtransition.
of the gait cycle. The raw EMG amplitude was further normalized to the ensemble average peak (100~o) for between-subject or between-session comparison. Temporal gait descriptors, such as stance, swing, and double support durations, were also normalized to 100~o of the gait cycle for comparison. The trunk, hip, knee, and ankle angular displacements were measured directly from the monitor screen at each 5 ~ of one representative gait cycle. The cycle chosen for kinematic analysis was selected within the cycles digitized for EMG analysis (Fig. 1).
Results and discussion
The video pictures and joint angular displacement profiles of S 1 and $2, during each evaluation, are shown in Figs. 2 and 3 (A & C) respectively. During the placebo evaluation (Fig. 2A and D; Fig. 3A and C, open diamonds), both subjects required at least 50~ BWS and manual assistance to take steps on the treadmill even at a minimal speed of 0.26 m . s - 1 . Their gait pattern was characterized by a flexed posture, especially in the trunk for S I (Figs. 2A and 3A), and in the hip and knee for $2 (Figs. 2D and 3C), indicating the inability to efficiently bear weight on the lower extremity and to maintain an upright posture while advancing on the treadmill. Furthermore, both subjects frequently encountered flexor spasms as soon as the foot was lifted off the treadmill, giving rise to an excessive flexion in the hip and knee during swing (S 1 : 85 ° and 105 °, $2:45 ° and 85 °, respectively). The initial footfloor contact was made with the toe, as indicated by the flexed hip and knee, as well as the plantarflexed (S1) or neutral ($2) ankle joint. For S1, there was a tendency towards hip and knee extension in early stance (10~o of the gait cycle), but yielded to flexion immediately on weight acceptance (see arrow with asterisk). Although the hip progressively underwent extension until lift off, the knee remained flexed and ankle plantarflexed throughout stance. For $2, the hip and knee remained flexed in early stance, with the knee and ankle progressively yielding more into flexion and plantarflexion respectively throughout the rest of stance, showing a total incapacity to cope with weight acceptance. As a result of the flexor spasm, the mean stance-swing transition occurred as early as 54.1~o (+8.4~o, n = 10) of the gait cycle for S1, and 44.3~o (+ 12.7~, n = 9) for $2. Following the administration of clonidine in combination with cyproheptadine, both subjects gained the ability to walk unassisted, independently on the treadmill at the same speed of 0.26 m. s- 1. S1 still required 20~o BWS, whereas $2 could walk full weight bearing at 0~o BWS. With the improved ability to cope with weight, both subjects assumed an upright posture with a near neutral trunk angu-
88
A
S1
B
C
D
$2
E
F
Fig. 2. Representative video pictures at critical gait events of S I and $2 taken respectively during (A and D)placebo evaluation; (B and E)postmedication evaluation; and (C and F) post-training evaluation.
89
Z
t,/)
--
...7
........................
• _
: : : : _
•
-J
(/)
i
i
o .~,
i
i
~'-
~
:Á.->
~....~f:'~ : :
"t.-,
~
~....-.~'~--L
× "
"~ :
O
~
'-
. . ~
,
(
~
-~< ~
"~
,
.
0
;~
'
o
0
~
O
~ ,-, = ,=,..~
~
~-
-
~
o
~
0
c:
e~
~
~m ' o - ~~
i
°'""
o>o
'
o
o
°
°
~
~
o
"-6 ~
o
>
~
~.~'=
0
o:-
°° I
"~ '~'~" ~ ~
~
o:I7ONV ~ N r l l d l
o:I7~DNV dlH
o : I 7 O N V :J:INN
o:i7ONV :17~tNV
•
. ~. ~ ~ ,.
.~ ~ ~-.o ~
....
''"
"'..'''.''"
•
Z
'7
"
::::~
.,.I
'"
......
.
.
.
•
.
.
.
.
.
.
.
.
0
5
.:.
~
~,
~
"~
5 0 to 20~o; $2: from > 50 to 0~o). This type of disordered activation pattern in the quadriceps has been described by Knutsson (1983) as 'crutch spasticity', which has functional implications for the knee to stiffen into extension and the limb to function as a crutch, in response to weight acceptance during stance. Such an activation pattern of VL and T A was found to be modified when BWS was provided (Visintin and Barbeau, 1989). With interactive gait training, the E M G profiles (Figs. 3C and 1D, thin solid lines) remained similar to that of the post-medication evaluation. The same was observed for the index values. However, the joint angular displacement patterns, as described earlier, were further improved with training. Thus it seems that the final expression of the locomotor pattern can still be optimized even though the muscle activation patterns remain similar. It would be interesting, as a continuation of the present study, to investigate whether the effects of medications and training can still be retained when medications are gradually withdrawn. The functional profiles of both patients are summarized in Table 2. Functionally, S 1 gained the ability to walk independently overground post medication with Canadian crutches and a Klenzac brace, which was later changed to a small soft ankle brace following training. $2 could walk independently overground with a walker following medication, and with Canadian crutches following training. This is remarkable since both patients were wheelchair-bound to begin with. Moreover, both patients showed a marked improvement in their endurance for treadmill walking. From being able to take only 6-10 steps with maximal BWS and manual assistance during the placebo evaluation, both S 1 and $2 could walk for long periods (5-8 min non-stop) without BWS or manual assistance following medication and training. The two patients also demonstrated a marked decrease in spasms and clonus. Ankle clonus, as measured in the quiet sitting position, was reduced from being sus-
tained in both patients to 1-2 beats in S 1, and 5-10 beats in $2. Since both patients were on a stabilized dosage of baclofen before entering the study, the reduction in spasms and clonus, following administration of clonidine and cyproheptadine, further suggests that the two monoaminergic agents could be used as effective antispastic medications. In conclusion, we have proposed a potentially effective approach for the rehabilitation of spastic paretic gait by combining pharmacological intervention with an interactive gait retraining program. This preliminary study shows that locomotion in the lower extremities can be recovered in chronic SCI patients who have previously undergone conventional therapy and remained wheelchair-bound. The combined medications, clonidine and cyproheptadine, acting on receptors in the spinal cord, possibly modify the level of neuronal hyperexcitability (Brdard et al., 1987) and modulate the expression of the locomotor pattern, as reflected by the changes in the muscle activation pattern and timing. However, the neurophysiologic mechanisms behind the alteration in reflex activity, in relation to the locomotor pattern, have yet to be investigated. The interactive gait training, incorporating progressive weight bearing with treadmill walking exercises, can further optimize the functional gain, as reflected by the kinematic changes, through improving the patient's ability to cope with external demands such as loading and speed. We are in the process of validating this new intervention approach in SCI subjects who will be stratified according to the severity of lesion and the impairment of locomotion. Acknowledgements This project is supported by the Medical Research Council of Canada. J. Fung received a studentship from the Rick Hansen Man in Motion Legacy Fund for spinal cord research. H. Barbeau is a Chercheur-Boursier supported by the Fonds de la Recherche en Sant6 du Qurbec. The authors appreciate the contributions of Lois Finch, Martha Visintin, and Lora Salvo in the training of the patients.
References Barbeau, H. and S. Rossignol (1987) Recovery of locomotion after chronic spinalization in the adult cat. Brain Res., 412: 84-95. Barbeau, H. and S. Rossignol (1990) The effect of serotonergic drugs on the locomotor pattern and on cutaneous reflexes of the adult chronic spinal cat. Brain Res. 514: 55-67. Barbeau, H., C. Julien and S. Rossignol (1987a) The effects of clonidine and yohimbine on locomotion and cutaneous reflexes in the adult chronic spinal eat. Brain Res., 437: 83-96. Barbeau, H., M. Wainberg and L. Finch (1987b) Description and application of a system for locomotor rehabilitation. Med. Biol. Eng. Comput., 25: 341-344. Barbeau, H., J. Fung, J. Stewart and M. Visintin (1988) Impairment of spastic paraparetic gait: Implications for new rehabilitation strategies. Proceedings of the fifth Biennial Conference of Canadian Society for Biomechanics pp. 12-16.
93 B6dard, P.J., L.E. Tremblay, H. Barbeau, M. Filion, R. Maheux, C.L. Richards and T. DiPaolo (1987) Action of 5-hydroxytryptamine, substance P, thyrotropin-releasing hormone and clonidine on motorneurone excitability. Can. J. Neurol. Sci., 14: 506-509. Conrad, B., R. Benecke and H-M. Meinck (1985) Gait disturbances in paraspastic patients. In: P.J. Delwaide and R.R. Young (Eds.), Clinical Neurophysiology in Spastieity, Elsevier Sciences Publishers, Amsterdam, pp. 155-174. Fung, J. and H. Barbeau (1989) A dynamic EMG profile index to quantify muscular activation disorder in spastic paretic gait. Electroenceph. Clin. Neurophysiol., 73: 233-244. Knutsson, E. and C. Richards (1979) Different types of disturbed motor control in gait of hemiparetic patients. Brain, 102: 405-430. Knutsson, E. (1983) Analysis of gait and isokinetic movements for evaluation ofantispastic drugs or physical therapies. In: J.E. Desmedt (Ed.), Motor Control Mechanisms in Health and Diseases, Raven Press, New York, pp. 1013-1034. Nance, P.W., A.H. Shears and D.M. Nance (1985) Clonidine in spinal cord injury. J. Can. Med. Assoc., 133: 41-42. Nance, P.W., A.H. Shears and D.M. Nance (1989) Reflex changes induced by clonidine in spinal cord injured patients. Paraplegia, 27: 296-301. Pierrot-Deseilligny, E. and L. Mazieres (1985) Spinal mechanisms underlying spasticity. In: P.J. Delwaide and R.R. Young (Eds.), Clinical Neurophysiology in Spasticity, Elsevier Sciences Publishers, Amsterdam, pp. 63-76.
Rossignol, S., H. Barbeau and C. Julien (1986) Locomotion of the adult chronic spinal and its modification by monoaminergic agonists and antagonists. In: M.E. Goldberger, A. Gorio and M. Murray (Eds.), Development and Plasticity of the Mammalian Spinal Cord, Liviana Press, Padova, pp. 323-345. Tuckman, J., D.S. Chu, C.R. Petrillo and N.E. Naftchi (1982) Clinical trial of an alpha adrenergic receptor stimulating drug (clonidine) for the treatment of spasticity in spinal cord injured patients. In: N.E. Naftchi (Ed.), Spinal Cord Injuries, Spectrum Publications, pp. 133-137. Stewart, J.E., H. Barbeau and S. Gauthier (1987) The effects of clonidine on clinical spasticity and in modulation of the locomotor pattern in chronic spastic spinal cord patients. Soc. Neurosei. Abstr., 13: 353. Visintin, M. and H. Barbeau (1988) The effects of treadmill speed on spastic paraparetic gait. Soc. Neurosci. Abstr. 14: 262. Visintin, M. and H. Barbeau (1989) The effects of body weight support on the locomotor pattern of spastic paretic patients. Can. J. Neurol. Sci., 16: 315-325. Wainberg, M., H. Barbeau and S. Gauthier (1986) Quantitative assessment of the effect of cyproheptadine on spastic paretic gait: a preliminary study. J. Neurol., 233:311-314. Wainberg, M., H. Barbeau and S. Gauthier (1990) The effects of cyproheptadine on locomotion and on spasticity in spinal cord injured patients. J. Neurol. Neurosurg. Psycbiat., 53: 754-763.
