Brain Reseurch Bulletin, Vol. 22, pp. 53-56. 0 Pergamon Press plc, 1989. Printed in the U.S.A. 0361.9230/89 $3.00 + .OO Late Flexion Reflex in Paraplegic Patients. Evidence for a Spinal Stepping Generator’ zyxwvutsrqpon B. BUSSEL, A. ROBY-BRAMI, A. YAKOVLEFF AND N. BENNIS INSERM U 215, Handicap Moteur Neurologique et croissance Hopital Raymond Poincark, 92380 Garches, France FLEXION withdrawal reflexes are well known in patients with spinal lesions. They were recognized in patients with spinal cord section since the pioneer studies performed after the first world war. Flexor reflexes are usually the prominent characteristic of spinal automatic activity, and it is generally assumed that this is linked to the release of the flexion reflex pathways from su- praspinal control. The electrophysiological study of the flexion reflexes in man began only in 1954 with Pedersen (17). The flexion reflex was elicited by electrical stimulation of a peripheral distal nerve and was recorded by Electromyography (EMG) in ipsilateral flexor muscles. In normal subjects, as well as in patients with lesions of the central nervous system, it was demonstrated (6, 10, 11, 17) that the minimum latency of the flexion reflex (100 msec or less), considering the length of the conduction pathway in man, was consistent with the classical description of flexion reflexes (14,20). These previous studies focused on the early response since in normal human or in patients who could voluntarily move their legs, the later responses could be due to a voluntary reaction independently from spinal reflexes. Recently, we systematically studied the flexor reflexes in a group of patients with a clinically complete spinal cord section (18). As described before (S), we observed that these reflexes could be separated into early and late EMG responses (Fig. 1). The early response did not differ from the flexion reflexes described in normal man (11,22) and was similar to the flexion reflex described in spinal animals. The late response was BUSSEL, B., A. ROBY-BRAMI, A. YAKOVLEFF AND N. BENNIS. Lateflexion reflex in paraplegic patients. Evidence for a spinal stepping generator. BRAIN RES BULL 22(l) 53-56, 1989. -We demonstrated previously that electrical stimulation of the Flexor Reflex Afferents (FRA) induces a late flexion reflex with a central conduction time longer than 100 msec. Its la- tency is prolonged by increasing the intensity or the duration of the stimulation. This late reflex is therefore similar to the late flexion reflex observed in acute spinal cat with DOPA. Some findings suggest that in man the late flexion reflex could be inhibited at a premotoneuronal level by contralateral FRA stimulation. In relation to the late flexion reflex, a late contralateral facilitation of soleus monosynaptic reflex (MSR) was observed. Rhythmical activity was observed in only one patient who had an exceptional form of spinal myoclonus. This myoclonus could be modulated by FRA stimulation. These facts show that the reflex organization in paraplegic patients is similar to the one described in acute spinal cat with DOPA and therefore suggest that a spinal stepping generator could exist in humans. Man Paraplegia Late flexion reflex Spinal stepping generator characterized by its latency which could last longer than 130 msec, and vary with the intensity of the stimulation. Strangely, its latency was regularly longer when stronger stimulations were applied (Fig. 1). Its threshold was lower than the threshold of the early response. We investigated the nature of this late response. It obviously originated from spinal reflexes since the patients had a com- plete spinal cord section. However, it could be due to any other previous motor response: the movement activating afferents which could, in turn, act back at spinal level to elicit a later response. We demonstrated (18) that both the late response and its latency increase did not depend on any previous motor response (whatever it was, either a direct motor response or an early flexion reflex) and that they were directly related to the electrical stimulation. We estimated that its minimum central delay was of more than 100 msec. In addition we observed that the latency of the late reflex was increased if the duration of the electrical trains was increased (Fig. 2). Thus, the late response could be considered as a proper spi- nal reflex whose main property was that its latency could increase if the intensity of the stimulation was increased or if the duration of the train of electrical pulses was increased. These properties showed that the late reflex was similar to the late reflex described by Lundberg and co-workers in the acute spinal cat with DOPA [see review in (2,15)]: in the acute spi- nal cat, DOPA depresses the early flexion reflex and induces the appearance of a late flexion reflex, by releasing it from an ‘Service de reeducation neurologique du Professeur Held. 53