Electromyographic activity associated with spontaneous functional recovery after spinal cord injury in rats Sibille Kaegi, 1 Martin E. Schwab, 1 Volker Dietz 2 and Karim Fouad 1,2, * 1 Brain Research Institute, University of Zurich and Department of Biology ETH Zurich, 8057 Zurich, Switzerland 2 ParaCare, University Hospital Balgrist, University of Zurich, 8008 Zurich, Switzerland Keywords: animal model, locomotion, rehabilitation, spinal cord injury Abstract This investigation was designed to study the spontaneous functional recovery of adult rats with incomplete spinal cord injury (SCI) at thoracic level during a time course of 2 weeks. Daily testing sessions included open ®eld locomotor examination and electromyographic (EMG) recordings from a knee extensor (vastus lateralis, VL) and an ankle ¯exor muscle (tibialis anterior, TA) in the hindlimbs of treadmill walking rats. The BBB score (a locomotor score named after Basso et al., 1995, J. Neurotrauma, 12, 1±21) and various measures from EMG recordings were analysed (i.e. step cycle duration, rhythmicity of limb movements, ¯exor and extensor burst duration, EMG amplitude, root-mean-square, activity overlap between ¯exor and extensor muscles and hindlimb coupling). Directly after SCI, a marked drop in locomotor ability occurred in all rats with subsequent partial recovery over 14 days. The recovery was most pronounced during the ®rst week. Signi®cant changes were noted in the recovery of almost all analysed EMG measures. Within the 14 days of recovery, many of these measures approached control levels. Persistent abnormalities included a prolonged ¯exor burst and increased activity overlap between ¯exor and extensor muscles. Activity overlap between ¯exor and extensor muscles might be directly caused by altered descending input or by maladaptation of central pattern generating networks and/or sensory feedback. Introduction Following injuries of the spinal cord in adult mammals, anatomical repair and in particular axonal regeneration is very limited (reviewed in Schwab & Bartholdi, 1996). Nevertheless, signi®cant spontaneous locomotor improvements have been observed in humans and animals with incomplete spinal cord injury (SCI) (Eidelberg et al., 1981; Wernig & Muller, 1992; Basso et al., 1994; Jiang & Drew, 1996; Harkema et al., 1997; Dietz et al., 1998; Rossignol et al., 1999; Merkler et al. 2001). Functional recovery is likely based on multiple factors, involving recovery from spinal shock (Holaday & Faden, 1983; Basso et al., 1994; Hiersemenzel et al. 2000), remyelination (Gensert & Goldman, 1997), and plastic changes in the locomotor system below and above the lesion. Not much is known about the mechanisms driving plastic changes within the adult spinal cord. The only evidence gained from animal (Edgerton et al., 1997), as well as human studies (Wernig & Muller, 1992; Dietz et al., 1994; Dietz et al., 1995) is that locomotor training has a strong impact on functional recovery. Possible mechanisms underlying the adaptive changes after treadmill training are the modulation in glycinergic inhibition of spinal locomotor networks, the upregulation in BDNF and neurotrophin-3 (NT-3) expression (de Leon et al., 1999; Gomez- Pinilla et al. 2001), adaptive changes within spinal re¯ex pathways involved in the control of stepping (Pearson, 2001), and/or plastic changes in the anatomy or synaptic ef®cacy of spared descending ®bers. Studies on the locomotor recovery of incomplete SCI in cats have indicated that the muscle activation patterns, organized by spinal pattern generating networks, are undergoing extensive alter- ations, which might be involved in compensatory movement strategies (Helgren & Goldberger, 1993; Gorska et al., 1996; Jiang & Drew, 1996; Brustein & Rossignol, 1998). The present investigation was designed to study spontaneous locomotor recovery after incomplete SCI in adult rats, a frequently used model to study treatment strategies. Detailed knowledge of the parameters of functional recovery is essential to analyse the underlying mechanisms of recovery that have been reported to occur after various treatment approaches following SCI (McDonald et al., 1999; Ramon-Cueto et al. 2000; Coumans et al. 2001; Merkler et al. 2001). We hypothesize that adaptive changes in spinal networks controlling locomotion are involved in the spontaneous recovery of locomotion, and that maximizing these processes through speci®c treatments is a promising way to enhance functional recovery. Thus, detailed knowledge of the spontaneous adaptive capacity of spinal locomotor centres will help in the interpretation of behavioural data of SCI animals after speci®c treatments and will allow the design of combined pharmaceutical interventions with rehabilitative strategies. Using EMG recordings during the recovery phase 14 days after a SCI in adult rats, we gained detailed insight into various changes occurring in the stepping pattern, with many of them undetectable for behavioural tests or kinematic analysis. Materials and methods Experiments were carried out on 14 female and two male adult Lewis rats (150±250 g). In 13 female animals, a dorsal spinal cord lesion Correspondence: Dr K. Fouad, at *present address below below. E-mail: Karim.Fouad@ualberta.ca *Present address: Faculty of Rehabilitation Medicine, University of Alberta, Edmonton, T6G 2G4, Canada. Received 13 February 2002, revised 8 May 2002, accepted 17 May 2002 doi:10.1046/j.1460-9568.2002.02076.x European Journal of Neuroscience, Vol. 16, pp. 249±258, 2001 ã Federation of European Neuroscience Societies