Animal experiments were carried out in accordance with established practices as described in the National Institutes of Health Guide for Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee of the Asan Institute for Life Sciences, Seoul, Korea. Female Beagle (Orient bio, Seongnam, Korea) dogs weighing 8 kg to 12 kg at the time of injury were used. Before injury, all dogs were housed one per cage under simulated daylight conditions with alternating 12-hour light-dark cycles and had free access to food and water.

The dogs were anesthetized using an intramuscular injection of Zoletil (25 mg/kg) and Rompun (5 mg/kg) together with atropine sulfate (0.25 mg/dog). Anesthesia was maintained by inhalation of 70% N₂O and 1% isoflurane (Hana Pharm Co. Ltd., Hwaseong, Korea) through an endotracheal tube. The isoflurane concentration was increased to 2% during induction of the SCI. Dogs were laid in ventral recumbence on the operating bed with their limbs immobilized and pulled caudally. The posterior midline approach was selected. Usual muscle dissection was performed using electrocautery, and laminectomy was performed using an electrical drill. We identified the spinal cord between the L1 and L2 spinous processes of the vertebrae; a silicone balloon catheter (size 6 Fr; Sewoon Medical Co. Ltd., Cheonan, Korea) was then inserted into the vertebral canal. The balloon catheter was inserted in the cranial direction, to a distance corresponding to that previously measured between T10 and L1, thereby positioning it for inflation at the center of T10. The balloon was then inflated to the maximum volume (1.5 mL, via saline injection) in the spinal extradural space for 1, 2, or 3 days. After surgery, the skin was closed, and the dogs were administered a prophylactic intramuscular injection of cefazolin (20 mg/kg). For pain management, the animals intramuscularly injected with ketoprofen (10 mg/kg) once daily for 1 week. After injury, all dogs were housed individually in the same pre-injury environment. Their bladders were emptied manually twice daily until spontaneous control of urination returned.

Hindlimb motor function was assessed after 1, 2, or 3 days of compression SCI. The open field test was performed for all dogs as previously described.1,2,4) Before this test, each animal was allowed to adapt to an open field for 5 min. Once a dog demonstrated acclimatization on the open field, locomotor performance was evaluated for 5 min. We evaluated hindlimb motor function every 2 weeks from 1 day to 3 months after surgery.

Under inhalation anesthesia, spinal cord motor evoked potential (MEP) in response to electrical or magnetic stimulation (spinal-MEP) was recorded from the forelimbs and hindlimbs of dogs subjected to compression for 3 days using a Medelec Synergy system (Oxford Instrument Medical Inc., Surrey, UK). First, an electrical stimulus was delivered through two needle-type electrodes placed on the vertex, with an intensity of 100 mA, pulse duration of 0.5 ms, and pulse rate of 4.0 Hz. The recorded potentials were amplified and averaged over 100 impulses/recording. Second, a magnetic stimulation system (Magstim M200®; Magstim, Whiteland, UK) was used to analyze motor function. This apparatus provided a brief magnetic pulse that stimulated the spinal cord. Clinical and experimental applications of this apparatus have been reported elsewhere.3,5,15) An operational setting of 80% of the maximum output power was used, and the duration of the stimulus was 0.01 ms. Stimuli were delivered using an 8-figure coil (double 70-mm coil) applied to both the vertex and cervical spine, with identical tests performed on both injured and control dogs. Following stimulation of the spinal cord, Medelec Synergy® (Oxford Instrument Medical Inc., Surrey, UK) was used to record compound muscle action potentials (CMAP) with a needle electrode placed on the left triceps surae and tibia. The average of 10 recordings was calculated, and CMAP curves were obtained. A reduction in amplitude to less than 50% of the amplitude for controls was considered to represent critical SCI.

The Basso, Beattie, and Bresnahan (BBB) locomotor rating scale scores for dogs subjected to compression for 3 days remained at 0 or 1, without any obvious improvement, throughout the observation period. The tonus of their hindlimbs remained flaccid. These dogs moved using only their forelimbs, with no joint contracture evident in the hindlimbs. The hindlimbs were dragged with the dorsal surface of the paws against the floor. No voluntary hindlimb movement was observed. In contrast, motor function in dogs that had experienced compression for only 1 or 2 days was variable; all these dogs showed spastic tonus in their hindlimbs. Further, these dogs showed improvement in their BBB scores over the observation period.

For forelimbs of dogs subjected to compression for 3 days, the amplitudes of MEPs induced by electrical stimulation showed minimal changes. However, there was a significant reduction in the MEP amplitude for the hindlimbs of these dogs (Figure 1). Figures 2 and 3 show the CMAPs of a dog subjected to compression for 3 days. Forelimb movements were evoked by magnetic stimulation at the cervical spine and cranial areas (Figure 2) of these dogs. However, hindlimb movements were not evoked by magnetic stimulation at these areas, as shown in Figure 3.