A new technique for minimal invasive complete spinal cord injury in minipigs

This study was carried out under the protocols approved by the Ethics and Deontology Committee for Research on Animals, Victor Babes University of Medicine, Timisoara, Romania. We certify that all applicable institutional and governmental regulations concerning the ethical use of animals were followed during the course of this research. The study was performed to minimize group size and animal suffering. Ten female Göttingen minipigs (Ellegaard Minipigs, Dalmose, Denmark) aged 6 months (weight 21–26 kg) were included in this study. The animals were group housed in the “Horia Cernescu” Animal Facility at the University of Agronomy and Veterinary Medicine in Temeshaw, Romania. All animals were housed in stable groups of three animals per pen and had 2 months of acclimatization. Prior to the SCI procedure, the health state of all animals was documented.

Minipigs were premedicated with an intramuscular cocktail of ketamine (15 mg/kg) and xylazine (2 mg/kg). Deep sedo-analgesia was established and maintained by intravenous propofol (10 mg/kg). Fentanyl (0.05 mg/kg) was given as analgesia prior to the surgery and a second time prior to the spinal cord compression. We preferred sedo-analgesia to general anesthesia with automated ventilation similar to the process of needle and balloon catheter placement. Prior to inflation of the balloon catheter, fentanyl was again given intravenously, to reduce possible pain perception. The observed muscular reaction during balloon catheter inflation was also observed when severing the SC during SC transection in deeply generally anesthetized pigs. After induction of anesthesia, the animals were placed in extended prone position. An initial CT image (Siemens, Somatom Sensation, Bucarest, Romania) of the spinal cord was performed, and the dimensions of the spinal cord were measured. The depth and angle for needle guidance to the intervertebral thoracic space T13/L1 were calculated. Needle positioning and individual virtual piercing routes were marked on the skin. After a minimal skin incision, the spinal canal (SC) was punctured by a 2-part 17.5-G puncture needle (Renodrain Nephrostomy Puncture Needle, Urotech, Achenmühle, Germany) under CT guidance. The needle was introduced into the epidural fatty tissue without piercing the underlying dura mater. Cerebrospinal fluid leakage was checked by aspiration testing. By means of the Seldinger technique, a hydrophilic guide wire (.035 × 150 cm, NiCore guidewire, Bard Medical, Covington, USA) was dorsally inserted via the puncture needle to the epidural space until thoracic spinal cord level Th8. The needle was removed and a 10-French dilator, followed by a 12-French dilator, was inserted via the guidewire into the epidural space to widen the channel. Thereafter, a kyphoplasty balloon catheter (Guardian Inflatable Bone Expander System, Taipeh, Taiwan, Balloon catheter 2ea) with radio-opaque markers was inserted via guide wire to the thoracic spinal cord level T12. All steps were carefully followed by CT scans. Once the position was confirmed by CT, the balloon was inflated to 2 atm. Balloon filling and spinal cord compression was confirmed in the CT and maintained for 30 min. The pressure was constantly monitored by a manometer to ensure continuous and constant pressure. After completion, the balloon was deflated and the balloon catheter including the guide wire was removed. The skin incision did not require a suture.

Animals were provided with antibiotics (ceftazidimim 40 mg/kg) and analgesics (butomidor 0.1 mg/kg) for 5 consecutive days post injury after standard protocols. A daily health examination of the injury site and skin analysis was performed. After 4 and 16 weeks, neurological assessments were conducted. Measurement of motor function, sensory function and muscle hypertonia was performed after standard protocols [7].

Motor function: Recovery of motor function was assessed using the 14-point scoring system (porcine neurological motor score—PNM score). Using this system, (i) movement in all three joints in the lower extremities and tail, with no weight support and (ii) the degree of recovery of ambulatory function, with weight support could be assessed [7]. Testing of motor function was performed after 4 and 16 weeks of follow-up time in the open field and on a treadmill by two experienced investigators including a video assessment to define the ultimate PMC scoring.

Sensory function: Sensory function was assessed by the presence of the withdrawal response to a mechanical stimulus. For this purpose, the minipigs were placed in a hammock. The toes of the front and hind limbs were progressively compressed with Halsted forceps. If sensation was present, a vigorous withdrawal response and/or vocalization was observed and the stimulus was immediately stopped. The response was scored as present or absent after 4 and 16 weeks follow-up time.

Muscle hypertonia: The presence of muscle hypertonia was defined as a spontaneous (stimulus-independent) increase in muscle tone. A positive response was assigned when the partially or fully paralyzed extremity was extended or flexed in the knee or hip joint, and returned to its original position.

To determine the injury dimension, two MRIs were performed at a Siemens Magnetom Avanto 1.5 T (Siemens Romania, Bucharest, Romania). The first complete spine MRI (T1-weighted and T2-weighted scans) was done 6 h after SCI according to standard acquisition procedures, and the second complete spine MRI (T1-weighted and T2-weighted scans) was performed 4 weeks after SCI. The corresponding Digital Imaging and Communications in Medicine (DICOM) files were correlated to each other.

At the end of the follow-up period, the minipigs were sacrificed by an overdose intracardial barbiturate injection in deep general anesthesia. The spinal cord tissue was taken at level Th9 to L1, single segments of the rostral (Th9-Th10), lesion site (Th11-Th12), and caudal (Th13-L1) segments prepared and immersion-fixed in 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS) at 4 °C for 12 h. Thereafter, the specimens were extensively washed in 0.1 M PBS and stored in 0.1 M PBS with 0.05% sodium azide (NaN₃) at 4 °C until cryo-sectioning. For cryo-sectioning, the segments were incubated in 15% sucrose in 0.1 M PBS for 24 h and in 30% sucrose in 0.1 PBS for 48 h. Thereafter, the segments were cryofrozen via Tissue Tek OCT™ compound (Hartenstein, Germany). Transverse serial sections (10 μm) were obtained in a Leica cryostat, collected on Superfrost Plus slides, and stored at − 20 °C until further processing. Slides were thawed, washed with 0.1 M PBS with 0.1% Tween 20 (PBST), and blocked with 0.1% bovine serum albumin, 0.1% Triton in PBST (blocking buffer) for 2 h at room temperature. Primary antibodies (anti-200-kD neurofilament heavy (ab8135, Abcam, Cambridge, UK), anti-NeuN (clone A60, Merck Millipore, Darmstadt, Germany), anti-Iba1 (ab5076, abcam), anti-GFAP (ab4648, abcam), and anti-Olig-2 (ab109186, abcam) diluted in blocking buffer were incubated for 24 h at 4 °C. After several washes with PBST, sections were incubated with species-specific Alexa™ fluorochrome-labeled secondary antibodies for 4 h at room temperature. After several washes with PBST and PBS, sections were mounted with an anti-fade mounting medium (Thermo Fischer Scientific, MA, USA) and representative images of each segment were obtained using an Olympus VS 120 (Olympus Corporation, Tokyo, Japan) and post-processed with the OlyVIA Software (Olympus Corporation, Tokyo, Japan) and ImageJ (open source program, developed by the National Institutes of Health, USA).

In minipigs, an extradural spinal cord compression was performed by inserting a balloon catheter via the Seldinger technique at the thoracic spinal cord level T12. Inflating the balloon to 2 atm for 30 min led to a complete SCI. All procedures were CT-controlled and images taken for verification of technique. The initial CT scan showed no skeletal abnormalities or deformations in any of our minipigs which could have hampered a needle puncture after Seldinger technique and insertion of the kyphoplasty catheter in any of the ten minipigs. Moreover, it depicted similar thoracic spinal cord dimensions of the ten minipigs with the length of the spinal cord from T11 to T13 being 4.51 ± 0.08 cm and the diameter of the spinal cord being 0.97 ± 0.03 cm at T12 and 0.94 ± 0.02 cm at T13 (Fig. 1). A needle angle of 25–30°, depending on the individual anatomical features, was chosen to correctly and safely puncture the intervertebral space (Fig. 2a). The test to aspirate cerebrospinal fluid (CSF) was negative in all minipigs and the tip of the needle was not placed deeper than the epidural fatty tissue. The hydrophilic guidewire was smoothly rooted up to Th8. Only twice, the guidewire had to be corrected after initial caudal placement by retraction and rerouting to cranial Th8. Dilatation of the intervertebral space with both 10- and 12-French dilator was feasible and possible from an anatomical point of view. Careful dilatation was necessary to avoid spinal cord damage and damage of the dilator which would hamper a safe removal of the dilator. The dimensions of the kyphoplasty catheter fitted to the bony structures of the spinal channel (Fig. 2b). However, any balloon catheter had to be measured in advance to match the dimension of the spinal cord. The balloon with air filling at 2 atm was possible in all minipigs, and no losses were registered during the 30-min compression time. The insertion procedure and placement was easily possible in all ten minipigs. Altogether, the SCI procedure was easily standardized. The procedure was performed exactly the same in all ten minipigs. The procedure took approximately 20 to 30 min for correct epidural balloon catheter placement plus 30 min for the complete compression of the spinal cord.

All ten minipigs survived the spinal cord compression procedure without any complications. No swelling or inflammation was observed above the injury site during follow-up by the regular health checkups. After 5 days post injury, no further antibiotic or analgesic coverage was needed. Body balance, to get in an upright position and to move around the pen by using front limbs solely, was quickly regained after SCI.

All minipigs showed stable severe neurological deficits (grade 0 in the PNM score) in the hind limbs which remained unchanged at follow-up time points of 4 and 16 weeks post SCI. All animals showed a baseline increase in muscle tone (i.e., spastic hypertonia) in the hind limbs during the 4 months follow-up time. Moreover, spastic tail movements were observed starting after 4 weeks of SCI in all SCI minipigs.

Analysis of withdrawal response to a mechanical stimulus for sensational analysis showed a normal response for the front limbs, while the response of the hind limbs was absent during the 4 months follow-up time in all minipigs.

To evaluate the extent of the initial damage and to visualize later alterations in the lesion size/dimension, MRI image analysis was performed. The complete spine MRI image assessment at the sub-acute time point of 6-h post injury and the early chronic time point at 4 weeks post injury suggested that (i) the injury was rather complete and (ii) that the injury site remained clearly discernible with comparable lesion size at the two time points (Fig. 3a–d).

When dissecting the spinal cord, the epidural sack was not lesioned due to the balloon catheter compression in any of the ten SCI minipigs. Only after opening the epidural sack, the macroscopic alterations were seen at the spinal cord level of balloon catheter compression.

The immunohistochemical analysis clearly depicted a normal appearing structural organization of the spinal cord tissue rostal and caudal to the lesion site (Fig. 4a, c, d, f). The central canal was clearly discernible rostral to the lesion site and continued to be seen after the end of the lesion zone at level T13/L1 towards caudal. Neuronal soma were present in the gray matter and their axons appeared dense and well organized in the white matter. Oligodendrocytes, astrocytes, and microglia appeared normally distributed and shaped.

Within the lesion site, a massive disorganization was observed with the presence of cystic cavities in the central gray matter lined by reactive astrocytes (Fig. 4b). The central canal was not to be defined within the lesion site. Furthermore, neuronal cell death was observed by a complete loss of neuronal soma signal and only unorganized fragments of axons present at all (Fig. 4e). The diameter of the spinal cord at the lesion region was approximately 50% smaller compared to the same thoracic region of the healthy reference minipig spinal cords.