Fusion capacity after the implantation of interspinous fusion device - a biomechanical analysis and preliminary histological evaluation in pigs


 BackgroudThe interspinous fusion device (ISFD) was widely used in the degenerative lumbar diseases. Previous clinical studies have demonstrated favorable efficacy. To further explore interspinous fusion capacity after ISFD implantation, we performed biomechanical analysis and preliminary histological evaluation.MethodsTwelves experimental pigs were selected and accepted a simulated operation of ISDF implantation. The animals were grouped six groups based on different procedures: sham surgery; unilateral laminotomy (UL); bilateral laminotomy (BL); ISFD + intact; ISFD + UL; ISFD + BL. The graft-bed site was filled with purified bone graft material without any autograft bone. After six months feeding, all experimental animals were sacrificed and the corresponding lumbar vertebrae was obtained. The samples were fixed on the spinal test system and the motion range of flexion-extension, lateral bending and rotation were tested through a multiaxial robotic system. The samples were prepared to cut using a hard tissue-slicing machine. Then, tissue sections stained with hematoxylin-eosin (HE) and toluidine blue were observed to evaluate the new-bone formation.ResultsLaminotomy increased lumbar mobility in comparison with intact model, especially for BL. The implantation of ISDF can reduce the range of motion in all motion directions. Well-defined bone tissue can be observed in the histological images after 6 months. In the histological part of toluidine blue staining, the area of newly formed bone accounted for 8.721%. The hyperchromatic osteoblasts cells and density bone tissue can be observed in HE staining slides.ConclusionsThe interspinous osteogenesis in histology confirmed the possibility of interspinous fusion. Moreover, the implantation of ISDF provide the necessary stabilization for promoting fusion.


Introduction
Lumbar degenerative disease is one of the most common diseases in elderly patients, which could cause severe lower back pain and in uence the quality of life (1). Conservative treatment is preferred during the early symptom stage. If conservative management fails, surgical intervention should be considered. Posterior lumbar interbody fusion (PLIF) is considered as gold standard (2). In recent years, with the development of minimally invasive technology, all variety of interspinous fusion devices begin to be explored in the treatment of lumbar degenerative diseases. Several studies reported the usage of interspinous xation device (ISFD) to stabilize the intervertebral fusion segment (3,4). However, some literatures began to explore the application of ISFD as a stand-alone device in lumbar degenerative disease, and reported good clinical e cacy (5)(6)(7).
ISFD can be xed into the spinous process through locking mechanism, which can provide certain stability after decompression. Moreover, the designed cavity of ISFD is lled with the mixture of the autogenous bone and bone graft material to promote interspinous fusion. It is vital to achieve interspinous fusion for long-term clinical e cacy. If bone fusion cannot be achieved, the instrument will be fail in the long term(8). This is a general rule that stability provided by internal xation is temporary and long term stability depends on bony fusion. Otherwise, it is easy to occur spinous process fracture and instrument failure, which further lead to symptom recurrence. At present, only a clinical study reported 84% interspinous fusion in patients with Grade I degenerative spondylolisthesis through the CT evaluation at 34.4 months after surgery (9). However, it is still di cult to judge whether true fusion can be achieved through radiological evaluation because of metal artifacts interference from ISDF. The interspinous fusion after the implantation of ISDF was still understudied area. Cunningham performed a biomechanical and histologic evaluation experiment to explore the osteogenesis capacity of a new bone graft substitute in sheep PLF model (10). Therefore, we conducted an larger-bodied animal trial to explore interspinous fusion capacity after the implantation of ISDF through biomechanical analysis and preliminary histological evaluation.

Materials And Methods
The interspinous fusion device applied in our study was BacFuse (Pioneer Surgical Technology Inc Marquette, MI, US) which including a rigid barrel and staggered spike on the lateral plate to optimize xation ( Fig. 1, A). The device applied for animal was custom-made and was available in 8mm and 10 mm height. The bone graft material (Hydroxyapatite/collagen) used was originated from Beijing Allgens Medical Science & Technology Co Ltd.

1.Preparation and grouping of specimens
Twelves experimental pigs were randomly grouped into six groups and then accepted different surgical operations. Group A: sham surgery; Group B: unilateral laminotomy (UL); Group C: bilateral laminotomy (BL); Group D: ISFD; Group E: ISFD + UL; Group F: ISFD + BL.

2.Surgery
All pigs were operated on with general anesthesia by professional animal anesthesiologist. The pigs were placed in the prone position on the operative table. After con rmation of the target segment, a skin incision was made along the midline. Keep the supraspinous ligament in the intact condition. The fascia and paraspinal muscle were stripped gradually, exposing L3/4 spinous processes and lamina. Bone decompression was performed based on requirement of different groups. Remove the lower part of L3 lamina and upper of L4 lamina to imitate laminotomy. The lateral half part of facet joint should be retained in the procedure of laminotomy as much as possible. The different size bone les were inserted into interspinous space to expand and polish the adjacent spinous processes. The cavity of ISDF was lled with puri ed bone graft material without any autograft bone (Fig. 1, A). Implant suitable size instrument according to the tension of supraspinous ligament. The instrument was placed in the base part of interspinous processes. The compressive clamp was used to clamp tightly spinous processes, and the entire system was locked by screwdriver ( Fig. 1, B). Then, the incision was then closed. After surgery, all pigs were prescribed with oral anti-bacterial drug. The next 2 weeks all pigs were observed closely and keep the incision in dry condition.
After six months feeding, all miniature pigs were sacri ced and corresponding spinal segment (L3/4) was obtained. All samples were performed X-ray uoroscopy (Fig. 2). All muscle tissue were removed and a intact spinal motion element was retained, such as vertebral bodies, discs, facet joints and all ligamentous structure.
The cranial and caudal of samples were potted in polymethacrylate in a metal xture for subsequent spinal mechanical tests.

2.Biomechanical test
The specimens embedded with polymethacrylate were installed in the spinal test system (MH5S, Yaskawa electric corporation, Kitakyushu, Japan). The L4 vertebral body was xed on the base and the cranial of L3 vertebral body was connected to a robotic arm. Make sure the spinal segment was in neutral position and the L3/4 disc was horizontal. Markers were placed on each vertebral segments, the displacement of these markers was captured by the photoelectric camera system (Optotrak; Northern Digital Inc., Waterloo, Ontario, Canada). The motion range of exion-extension, lateral bending and rotation were calculated by the computer system ( Fig. 3A, B). The force-moment sensor was applied to measure the applied load and provided feedback to the robot. The sensor can measure axial forces and torque and provide feedback to ensure that pure torque was applied along the main axis of motion of the spine. First, the biomechanical machine run in the speed of 1.0°/s with 3.0 NM moments in the intact model to get a series of standard angle. Subsequently, each of the test constructs was subjected to six load-unload cycles in each of the physiologic planes, generating exion, extension, left /right lateral bending, and left /right axial rotation load-displacement curves. The construct was sprayed frequently with normal saline to keep sample in a moist condition. The coordinate system for each vertebra and a set of adjacent vertebrae was based on the international community biomechanics 2002 revised standard (11). In order to reduce the in uence of spinal viscoelasticity, the mean value of the results of the last three times was taken for data analysis.

Histological experiment
The bone tissues including ISDF were obtained and submerged into ethylene diamine tetraacetic acid (EDTA) to decalcify. The tissues were cut to satisfy requirement and further dehydrate in the tissue dehydration liquid. Then, the tissues were deal with para n immersion and embedding. Tissue slices were made by using a hard tissue-slicing machine (German EXAKT-300CP) to cut 4 microns thick slices (Fig. 3C). After slicing specimens, the slices were stained with toluidine blue and hematoxylin-eosin. Then the ready-made para n sections were observed and analyzed under electronic microscope.

Data Analysis
All the data were measured and evaluated by an external third party. Statisticians do not participate in entire decision-making process. The normal distribution and continuity of test data are based on Kolmogorov-Smirnov test. The range of motions of each group were compared using a repeated measures ANOVA. SPSS 20.0 and GraphPad Prism 6.0 were used for statistical analyses, P < 0.05 was statistically signi cant.

Biomechanical evaluation
The range of motion (ROM) angles for each group in each direction was presented in Table 1. Compared with Group A, the ROM of Group B and Group C increased signi cantly in the movement of in exion, extension, left/right lateral bending and left/right rotation (P < 0.05, vs group B, P < 0.05 vs. Group C). The ROM of Group C in any movements signi cantly increased comparing with that of Group B (P < 0.05).
The ROM of Group D, E, F with the implantation of ISDF was signi cant less than that of Group A, B, C in any motion models, respectively (P < 0.05).

Extension and Flexion
As can be seen from Table 1, the ROM of exion in group B increased by 60.95% and that in group C increased by 97.16%. After applying the ISFD, the ROM of group D decreased signi cantly compared with group A (P < 0.05), a decrease of about 21.43%. Group E and F showed a decrease in ROM compared with group B and C (P < 0.05), about 39.98% and 34.53%, respectively.
The extension ROM in group B increased by 42.61% compared with that in group A, and that in group C increased by 76.52%. After the implantation of ISFD, the ROM of group D signi cantly decreased compared with group A (P < 0.05), a decrease of about 26.70%. Group E and F showed a decrease in ROM compared with group B and C (P < 0.05), by 22.07% and 21.76%, respectively.

Lateral bending
The ROM of left lateral bending in group B increased by 34.20% and that in group C increased by 78.67%. After applying the ISFD, the ROM of group D decreased signi cantly compared with group A (P < 0.05), a decrease of about 19.32%. Group E and F had signi cantly decreased in motion compared with group B and C (P < 0.05), about 28.05% and 27.30%, respectively.
The ROM of right lateral bending in group B increased by 26.56% and that in group C increased by 66.30%. After applying the ISFD, the ROM of group D decreased signi cantly compared with group A (P < 0.05), a decrease of about 21.45%. Group E and F had signi cantly decreased in motion compared with group B and C (P < 0.05), about 21.49% and 25.44%, respectively.

Axial rotation
The ROM of left rotation in group B increased by 88.20% and that in group C increased by 175.85%. After the implantation of ISDF, the ROM of group D decreased signi cantly compared with group A (P < 0.05), a decrease of about 49.25%. Group E and F had signi cantly decreased in motion compared with group B and C (P < 0.05), about 24.97% and 18.67%, respectively.
The ROM of right lateral bending in group B increased by 108.60% and that in group C increased by 212.21%. After the implantation of ISFD, the ROM of group D decreased signi cantly compared with group A (P < 0.05), a decrease of about 33.77%. Group E and F had signi cantly decreased in motion compared with group B and C (P < 0.05), about 17.21% and 21.59%, respectively.

HE staining
As described in Fig. 5, the cavity of ISFD was occupied with osteoblast cells and bone tissues. The surround black part was metal frame. The red staining part indicated new bone formation which was constituted with osteoblast cells and bone tissues. The brown part was the precursor condition of new bone formation. The blank part was caused of tissue exfoliation.

Discussion
Several studies reported the application of ISFD alone in lumbar degenerative diseases with good clinic e cacy(6, 7, 12). According to the initial design philosophy, nal interspinous process fusion was an expected goal (13). However, whether this fusion device can provide enough stability after decompression and meet the requirement of fusion needed to be studied.
At present, most of related biomechanical literature focused to analyze the application of ISFD as an adjunct to stabilize intervertebral fusion (3, 4, 14). Techy demonstrated that ISFD provided similar stability in the exion-extension motion and inferior stability in the lateral bending and rotation comparing with bilateral pedicle screws and rods system(3). Karahalios and Gonzalez et al. reported that ISFD as an adjunctive construct with anterior lumbar interbody fusion provided equivalent stability with bilateral pedicle screws and rods (4,14). In this present study, we investigated the stability of lumbar motion unit with ISFD stand-alone in the fresh animal samples. The results indicated that ROM of lumbar motion unit in the intact model signi cantly decreased after ISFD implantation in all directions. Similarly to our results, Karahalios reported that ISFD alone limited exionextension 75% ROM and reduced lateral bending and rotation 29% ROM when it was used in an intact cadaveric sample (4). Differ from traditional interspinous distraction devices, ISFD can be gripped and xed to the spinous process through medial plates spikes, which limit spinal motion and provide stability.
Laminectomy/laminotomy seems to be effectively ease the neurological symptom in clinic. However, bone decompression may in uence lumbar stability which can prompt the degeneration progress. In this study, the ROM in exion-extension, lateral bending and rotation signi cantly increased in the laminotomy groups, especially for BL. This may be one reason some patients experienced decompression alone occurred to symptom recurrence after several years (15,16). The implantation of ISFD after laminotomy not only stabilize spine but also prevent degeneration when true interspinous fusion has be achieved. Gonzalez et al concluded that the ISFD may be a suitable device to provide immediate exion-extension balance after a UL, but the bilateral screw and rods system provides greater stability in lateral bending and axial rotation motions. Anyhow, the implantation of ISFD can provide certain stability which may be required for interspinous fusion.
When it comes to interspinous fusion, several studies has reported related contents (4,13,17,18). It is vital important for implanting ISFD to achieve interspinous bone fusion. Spadea S resected the L3 spinous process and implanted it as a bone implant substitute in L4/5 interspinous space in lumbar disc herniation patients. After several years, interspinous bone fusion can be observed directly in several revision patients (17). Tian reported a case of lumbar spinal stenosis with Co ex implantation after decompression. His symptoms were completely relieved postoperatively. After 4.5 years follow-up, X-ray examination showed de nite bone fusion between the spinous processes in the Co ex implantation segment. The author thought that it was a phenomenon of heterotopic ossi cation, but it did con rm the possibility of interspinous fusion(18). Karahali performed the biomechanical study of anterior lumbar interbody fusion combined with posterior interspinous xation. From postoperative imaging, it was found that interspinous fusion was achieved between upper and lower spinous processes through bone graft groove(4). In our study, we implanted the bone graft material without any autograft bone in the barrel. However, well-de ned bone tissue can be observed in the histological images after 6 months. Previous literature has reported this kind of bone graft material used in the tibial bone defect model of rabbit. Bone defects can be repaired at 2 months and bone graft material can be fully resorbed at 3 months (19). In our study, we rst report the usage of bone graft material in the ISFD without any autograft bone and observed effective bone formation, which provide strong evidence to support interspinous fusion theory.
However, this study also has several shortcomings. First, the experimental model used in the study were nonupright animals, which spinal anatomy and motion pattern may be different with human beings. Moreover, we just presented histological pictures at 6 months after surgery. In the further study, the histological pictures at different time points will be presented to demonstrate entirely osteogenetic process more truer.

Conclusion
The histological interspinous osteogenesis con rmed the possibility of interspinous fusion. Moreover, the implantation of ISFD provide the necessary stabilization for promoting fusion.

Declarations
Ethics approval and consent to participate This study was approved by life ethics committee of Beijing Friendship Hospital.

Not applicable
Availability of data and materials Related data and materials can be obtained.

Competing interests
All authors declare that they have no con ict of interest.

No
Authors' contributions TH and CMM designed the research. BL, JX and SQC performed related experiments. BL and CMM analyzed the extractable data and wrote the rst draft of the paper. BL searched the literature, extracted and analyzed the data. TH was responsible for the statistical analysis and reviewed the manuscript and had the primary responsibility for the paper's nal content.   The results of hematoxylin-eosin staining. The blue part represented the new formation bone and the black was the precursor condition of new bone formation.