Forty-one 8-week-old Sprague-Dawley rats (200-230 g) were used. After anesthesia from ether inhalation and intraperitoneal administration of ketamine (100 mg/kg) and xylazine (10 mg/kg), T10 of the spine of each rat was laminectomized to expose the dura mater, which was pressed with a tension gage (Fig. 1). To create single crush and sustained crush models, stepwise compressions were performed at intensities of 2-140 g for 1-150 seconds. The bladders of the rats were manually voided twice a day for a week after the injury. To prevent infection, 1.0 mL of Bactramin (Roche, Basel, Switzerland) was mixed in 500 mL of bottled water provided for hydration for 2 weeks following spinal cord injury. Food was provided on the cage floor, and the rats had no difficulty reaching their water bottles. All animals were treated and cared for in accordance with the Nagoya University School of Medicine Guidelines pertaining to the treatment of experimental animals.

The cranial bones were exposed and bores were drilled 3 mm lateral and 2 mm posterior to the bregma where bipolar stimulating needles were inserted. A Nihon Koden Neuropack 8 (Nihon Koden Corp., Tokyo, Japan) was used as the stimulator. Except for the stimulus intensity, the stimulus conditions were approximately the same as actual spinal surgery (a train of four pulses at 2-ms intervals, stimulus intensities of 10-60 mA, and 20 additions with a phase switch after 10 additions). A Nihon Koden Neuropack (MEB-2200, ver. 04.02) used in spinal surgery was used to derive the control waveform by inserting bipolar needles in both perifemoral muscles (Fig. 1). The ground electrode was placed subcutaneously between the coil and the recording electrodes.

Derivation was started immediately after the infliction of injury at every 30 seconds for up to 15 minutes under the same conditions as the foregoing waveform measurement. We reviewed amplitude decrease and morphology change. Morphology change was having occurred in the event of any of the following: change in the number of waves in the waveform, prolonged duration, and shift in the location of the peak latency. For convenience, integrated intensity was defined as compression time multiplied by compression intensity.

The locomotor performance of 16 animals was analyzed using the Basso, Beattie and Bresnahan (BBB, 0-21 pts) open-field score for 4 weeks (1 day, 3 days, 5 days, 1 week, 2 weeks, 3 weeks, and 4 weeks) [4]. The evaluations were made by two blind observers for all analyzed rats.

Statistical analyses were performed with an unpaired two-tailed Student's t-test for single comparisons and one-way analysis of variance (ANOVA) for multiple comparisons. For the locomotor performance scores, repeated measures ANOVA and the Mann-Whitney U-test were used. In all statistical analyses, values of p<0.05 were considered to indicate significance. To obtain the data for statistical analyses, the investigators were blinded to the genotypes in all procedures.

Four different types of waveforms were obtained: no change, amplitude decrease only, morphology change only, and amplitude and morphology change. In the 50 limbs of 25 animals, type 1 accounted for 14% (7/50 limbs), type 2 for 36% (18/50 limbs), type 3 for 18% (9/50 limbs), and type 4 for 32% (16/50 limbs) (Fig. 2). No change of latency was seen in any rat.

While the integrated intensity was naturally low in the group with no change, the amplitude group and the morphology group produced their respective waveforms at approximately the same level of intensity. At high intensities, numerous waveforms were derived that exhibited amplitude decrease and morphology change (Fig. 3). In a box plot, the amplitude group and the morphology group were significantly correlated with high intensities (p<0.05) (Fig. 4).