Abstract
Background
Current techniques for instrumented posterolateral lumbar fusion include the use of titanium rods and pedicle screws to provide mechanical fixation, while autograft and/or allograft bone is used to promote osseous union across the fusion site. Although bone graft substitutes may be used for this purpose, they can resorb before a solid fusion is established, leading to nonunion. Tetracalcium phosphate-phosphoserine (TTCP-PS) is a novel, bioresorbable, osteoconductive, wet-field mineral-organic bone adhesive and bone graft substitute that has been previously shown to promote bone healing in a number of applications, including cranial bone flap fixation.
Objective
The goal of this study was to investigate the use of TTCP-PS for fusion in a posterolateral lumbar ovine model and to compare it to a commercially available bone graft substitute.
Study Design
An ovine lumbar fusion model was used to evaluate the safety and efficacy of TTCP-PS mineral-organic bone adhesive compared with a currently marketed bone graft substitute as a control (Actifuse ABX) over 1 year. These substances were implanted bilaterally in 20 sheep between the L3-L4 and L4-L5 transverse processes, stabilized with a pedicle screw fixation system.
Methods
Postoperative x-ray images, computed tomography scans, and histological findings were assessed at 12, 26, and 52 weeks after implantation to evaluate lumbar fusion across groups and time points.
Results
After euthanasia at 12 weeks, no statistically significant differences in the bone deposition region of interest at any of the substance implantation sites were observed. At 26 weeks, histologic analysis demonstrated new bone formation percentage to be significantly greater (P = 0.001) in the TTCP-PS sites (57%) than in the control group sites (31%). By 52 weeks, no statistically significant difference was found between groups, although the mean percentage bone growth area remained higher for the bone adhesive (52%) compared with the control (45%).
Conclusions
TTCP-PS was safe and effective in an ovine posterolateral lumbar fusion model. TTCP-PS promoted earlier and greater (26 weeks) bone bridging in this study than a currently marketed bone graft substitute (Actifuse ABX) while producing greater levels of bone formation at 1 year after surgery. This accelerated healing may aid implant stabilization and reduce the risk of nonunion in posterolateral lumbar fusions in humans.
Clinical Relevance
TTCP-PS was demonstrated to be safe and effective as a bone graft substitute in this large animal posterolateral spinal fusion model. Its use resulted in superior new bone volume, greater mineral density of the fusion mass, and more rapid bone ingrowth than the control. These properties of TTCP-PS bone adhesive may improve the reliability and success of spinal fusion procedures.
Level of Evidence
2.
Keywords: bone adhesive, biomaterial, lumbar fusion, bioresorbable, osteoconduction, osteointegration, bone bridging, regenerative, ovine lumbar spine
Introduction
The current technique for instrumented posterolateral lumbar fusion (PLF) utilizes titanium rods and pedicle screws to mechanically fixate the vertebrae and autograft or allograft bone to promote osseous growth across the fusion site.1–3 Although autologous bone is the “gold standard” for these surgeries, synthetic bone graft substitutes have been developed to eliminate the morbidity associated with autologous bone graft harvesting.4 Unfortunately, these substitute materials may resorb before a solid fusion is established, leading to nonunion and possible revision surgery.5
Tetracalcium phosphate-phosphoserine (TTCP-PS) is a wet-field, self-setting, load-bearing biomaterial that has been developed as an osteoconductive, bioresorbable bone adhesive and can serve as a substitute for traditional bone grafting.6–10 It has previously been shown to promote bone healing in a number of applications, including cranial bone flap fixation.11 This compound is composed of TTCP-PS. It can be augmented with porous beta-tricalcium granules and poly(lactic-co-glycolic) acid fibers. The adhesive cures minutes after application and produces a durable, load-bearing bond to both wet bone tissue and metal, augmenting the mechanical support of traditionally used titanium hardware. This Good Laboratory Practice (GLP) study aimed to test the bone regenerative efficacy and local tissue effects of the TTCP-PS material in an ovine posterolateral spine fusion model.
Methods
Regenerative Bone Adhesive
TTCP-PS (Tetranite Bone Adhesive, RevBio, Inc.) was provided by the manufacturer, packaged and sealed in powder form in bowls, and sterilized by gamma irradiation (20–30 kGy). The powdered material was composed of tetracalcium phosphate, phosphoserine, porous beta-tricalcium phosphate granules, and poly(lactic-co-glycolic) acid fibers (10:90) mixed with 1.68 mL water for injection at a powder-to-liquid ratio of 5.5 wt/vol.
Actifuse ABX (ApaTech)
The control substance (C) used was Actifuse ABX, a comparable marketed product intended for use in posterolateral fusions. Actifuse APX is composed of silicate-substituted hydroxyapatite granules11 and was selected due to its network of interconnected pores and osteostimulatory properties. It was provided ready to use and was molded to obtain the desired shape for implantation.
Xia Spinal System (Stryker)
The Stryker Xia Spinal System was used to stabilize the L3 to L4 or L4 to L5 arthrodesis in the sheep. Constructs consisted of 4 × 20 mm polyaxial screws, Xia 4.5 mm rod blockers, and 4.5 × 200 mm titanium rods.
Test System
A sheep model was selected for implantation. The size and geometry of sheep vertebrae are appropriate for evaluation of posterolateral fusion.12,13 Twenty healthy female sheep (53–73 kg, age >2 years) were used. Sheep were included if they were free of pre-existing spinal pathology, infection, or systemic disease, as verified by physical examination and imaging. Each sheep was implanted bilaterally at 1 spinal segment (either L3–L4 or L4–L5, depending upon whether the sheep had 6 or 7 lumbar vertebrae). Each sheep acted as its own control, with TTCP-PS implanted between the transverse processes on 1 side and the control substance on the other side.
Husbandry conditions conformed to the European requirements (EEC Directive 86/609). The sheep were fed a standard laboratory diet of hay, commercial pelleted food, and water. Standardized housing conditions were provided and ensured by Biomatech, a NAMSA company (Chasse-sur-Rhone, France). The protocol of this study was approved by the Biomatech Ethical Committee prior to the start of the experiment. Apart from the computed tomography (CT) scans, this study was conducted in accordance with both European (2004/10/EC Directive) and United States Food and Drug Administration (21 CFR 58) GLP regulations.
Primary outcome measures included radiographic assessment of spinal fusion using x-ray imaging and CT, as well as histological analysis of new bone formation and bony integration of the implanted graft substitute.
Study Design
In this prospective randomized study conducted at an independent GLP laboratory, TTCP-PS and control substance were implanted in sheep between the L3 to L4 or L4 to L5 transverse processes stabilized by a spinal pedicle fixation system (Figure 1). Twenty sheep were implanted bilaterally for 12, 26, or 52 weeks (6 sheep per time point plus 2 spares following the recommendations of the ISO 10993 standard and direct regulatory feedback), at which time they were euthanized (Table 1). Each implanted substance was randomly assigned by draw of the hat to either the right or left side, allowing for 6 implantation sites per substance at each time point.
Figure 1.

Implantation site computed tomography.
Table 1.
Study design.
| Time-Period | No. of Sheep | Sites | No. of Sites | |
| Test Article 1 (TTCP-PS) | Control Article (C) | |||
| 12 wk | 6 | Right and left spaces between L3–L4 and L4–L5 transverse processes | 6 | 6 |
| 26 wk | 6 | 6 | 6 | |
| 52 wk | 6 (+2 reserves) |
6 (+2 reserves) |
6 (+2 reserves) |
|
Abbreviation: TTCP-PS, Tetracalcium phosphate-phosphoserine.
PLF Surgical Procedure
Preoperative Measures
Premedication and anesthesia were performed by intravenous injection of a thiopental-pentobarbital mixture and atropine, followed by inhalation of an O2-isoflurane mixture. Each sheep received analgesic treatment and prophylactic perioperative antibiotics. Dorsoventral preoperative x-ray imaging was performed to assess the exact number of lumbar vertebrae and any skeletal abnormalities present.
All surgeries were performed by an experienced preclinical veterinary surgeon (Biomatech) using standard aseptic techniques. Each sheep was placed in sternal recumbency. A longitudinal cutaneous incision was performed, overlying the palpable spinous processes from L2 to L5. Two paramedian fascial incisions were made through the lumbodorsal fascia, and the underlying muscle was dissected with scissors along the spinous processes. The articular processes, lateral pars interarticularis, and medial dorsal part of the transverse processes were then exposed.
Hardware Fixation Procedure
The medial dorsal aspect of each transverse process and lateral pars interarticularis was decorticated with a high-speed burr. The Xia Spinal System was used to perform an L3 to L4 or L4 to L5 arthrodesis.
The caudal aspect of each superior articular process was decorticated using bone rongeurs to expose the underlying cancellous bone to prepare the entry point of the transpedicular screws. A modular tap was manually introduced to prepare the pedicle screw pathway, and the polyaxial screws were inserted using a polyaxial screwdriver. The rod was cut and bent to fit the required spinal length and contour before positioning into the 2 screws. The rod was then fixed into the 2 screws with a rod blocker using the persuader and blocker inserter. This technique was repeated for the contralateral side.
Biomaterial Implantation
After hardware fixation, the TTCP-PS was prepared by dry mixing with a spatula to ensure powder homogeneity. An aliquot of 1.68 mL of water for injection (Proamp, Aguettant) was added to the mixing bowl, and the components were mixed for 45 seconds. Mixing resulted in a cohesive putty with a working time of 2 minutes after the start of mixing. Approximately 4 cc of the putty was manually applied to the bony surfaces of the lateral gutters within 1 to 2 minutes from the start of mixing and allowed to cure to a hardened porous (35% porosity) scaffold within 10 minutes of mixing.
The control substance was ready to use and was modeled until the appropriate shape for implantation was obtained. A volume of 4.5 cc of the control substance was implanted in each sheep. No water, saline, autologous blood, or bone marrow was added or mixed into the control.
TTCP-PS and the control substance were placed bilaterally (TTCP-PS on 1 side and control on the other, randomly assigned to right or left by draw of the hat) between the L3 to L4 and L4 to L5 transverse processes and along the lateral pars and facet complex (Figure 2). The fascia was closed using a continuous suture; the subcutaneous and cutaneous tissues were closed using standard procedures. The wound was disinfected using an iodine solution, after which a dressing was applied to the surgical site.
Figure 2.

Intraoperative macroscopic photo—implantation design. TTCP-PS, tetracalcium phosphate-phosphoserine.
Treatment and Clinical Observations
An injection of buprenorphine was delivered subcutaneously once daily for 2 days after surgery. A nonsteroidal anti-inflammatory drug was administered daily by intramuscular injection for 7 days after surgery. An antibiotic was given every other day for 8 days after surgery. Surgical skin clips were removed after complete healing, approximately 2 weeks after surgery. The sheep were observed daily for any clinical abnormalities during the follow-up period.
CT Scans
CT scans were performed on 6 sheep per time period 2 weeks after surgery and at 2-month intervals (±5 days) thereafter. CT scans were performed by Voxcan, an independent subcontractor selected and approved by Biomatech. All CT scans were performed at 120 kV and 200 mA with slices of 1.5 mm and a pixel size of 0.351 mm in the plane. The density of the intertransverse bridge, made of implanted substance and newly formed bone, was divided into 5 categories: 400 to 600, 600 to 800, 800 to 1000, 1000 to 1250, or 1250+ mg/cm3.
The total volume percentage of each density range was calculated for each intertransverse bridge, and the distribution of the volume in each density range was compared between the implanted substances and over time. Statistical analysis was performed on calculated fusion mass volumes for each implanted site using R software (R Development Core Team, 2007). Statistical analysis examined whether a statistically significant difference existed across all volumes at all time points in the density range.
Postoperative X-Ray Imaging
For each time period, x-ray imaging was performed at Biomatech on the 6 sheep that were not used for CT 2 weeks after surgery and at 2-month intervals (±3 days) thereafter. X-ray images were scored based on the presence of radiolucent areas in the intertransverse bridge, indicating the presence of gaps in the bone bridge.
Termination
At each time point (12, 26, and 52 weeks following implantation), designated sheep were weighed and euthanized via a lethal injection of pentobarbital solution. A macroscopic examination of the implanted sites was conducted, and gross changes in tissues surrounding the site were recorded according to size, shape, color, consistency, and distribution. Macrophotographs were taken, and the spine was harvested from L1 to L7 using an oscillating saw.
Histologic Preparation
Before fixation, each implanted site was separated into 2 hemicolumns of the left and right sides. All sites, and 1 piece of each of the nonimplanted substances, were dehydrated in alcohol of increasing concentration, cleared in xylene, and embedded in polymethyl methacrylate for slide preparation. One longitudinal section per site was conducted based on x-ray images located at the center of the fusion mass parallel to the spinal axis (Figure 3). To prepare ground sections, a microcutting and grinding technique adapted from Donath et al was utilized.14 All histological sections were stained using modified Paragon stain.
Figure 3.

Histological segment location.
Histopathologic Analysis
Analysis of the local tissue effects of the implanted substances was adapted from ISO 10993-6 standard.15 Analysis examined 3 secondary regions of interest (ROIs)—a cranial, medial, and caudal section of each implanted site. A qualitative and semiqualitative histopathologic evaluation of each site was conducted by an independent histopathologist to grade the following parameters: fibrin, inflammatory reaction, necrosis, tissue degeneration, neovascularization, fibrosis, fibrous encapsulation, osteolysis, and other relevant parameters of the tissues surrounding the implantation site.
The bone healing response was qualitatively and semiqualitatively assessed by analyzing osteoblastic cells, new bone formation, osteointegration, osteoconduction, bone density, bone remodeling, implanted substance degradation, implanted substance migration, bone growth toward fusion, implanted substance interruption, and bone-implanted substance continuum, a measure of the bridge made of bone-implanted substance coupling without evidence of interruption.
Histomorphometry Analysis
Histology sections were imaged at 10× magnification for histomorphometry analysis (Nikon E800, Nikon Inc., Melville, NY, USA). Images were segmented by pixel color to distinguish between bone, soft tissue, and empty area within the primary posterolateral fusion ROI using ImagePro software (ImagePro, Media Cybernetics, Silver Spring, MD, USA). The posterolateral fusion ROI was selected by contouring any new tissue caudal of the cranial transverse process and cranial of the caudal transverse process (Figure 4). Statistical comparisons were made between treatment groups and time points via a 2-way analysis of variance with an alpha value (α) of 0.05 and Tukey post hoc test (MATLAB R2020A, Natick, MA, USA).16
Figure 4.

Histology method—segmentation. The posterolateral fusion region of interest was contoured for histomorphometry analysis by encapsulating all neo tissues within the outer (cranial/caudal) edges and posterior of the transverse processes.
Results
Perioperative Observations
TTCP-PS performed as intended in the operating room. The adhesive was mixed by hand, resulting in a runny, paste-like material that gradually increased in viscosity over ~45 seconds, yielding a malleable, tacky, nonexothermic adhesive. The tacky adhesive became putty-like and was manually placed in the exposed and decorticated posterolateral intertransverse fusion site and conformed to ensure contact with trabecular bone before curing to a hardened state.
Postoperative X-Ray Imaging
X-ray images were scored on a scale of 0 to 2 based on the quality of the bone-implanted substance bridge across the implantation site. A score of 0 indicated the presence of a bridge with 1 or several large radiolucent areas, a score of 1 indicated the presence of a bridge with 1 or several small radiolucent areas, and a score of 2 indicated the presence of a bridge without radiolucent areas.
Large radiolucent areas in the intertransverse bridge were observed in both groups at the 2-week time point, with a mean postoperative x-ray score of 0.0. Large radiolucent areas were still observed in most cases at the 10-week time point in the control group, with a mean score of 0.2. However, only small radiolucent areas were observed in most cases at the 10-week time point in the TTCP-PS group, with a mean score of 0.7.
Only small radiolucent areas were observed in both groups at the 18-week time point, with mean scores ranging from 0.7 to 1.0 (Table 2). No radiolucent areas were observed in most cases from 26 weeks, with mean scores ranging from 1.7 to 2.0. However, half of the control substance implantation sites still showed small radiolucent areas in the bridge up to 26 or 50 weeks, with a mean score of 1.3.
Table 2.
Postoperative x-ray scores.
| Score at Given Time Period | |||||||
| Article | 2 wk | 10 wk | 18 wk | 26 wk | 34 wk | 42 wk | 50 wk |
| Control (C) | |||||||
| Mean | 0.0 | 0.2 | 0.7 | 1.3 | 1.3 | 1.3 | 1.3 |
| SD | 0.0 | 0.4 | 0.5 | 0.5 | 0.5 | 0.5 | 0.5 |
| Test (TTCP-PS) | |||||||
| Mean | 0.0 | 0.7 | 1.0 | 2.0 | 2.0 | 2.0 | 2.0 |
| SD | 0.0 | 0.5 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
Abbreviation: TTCP-PS, tetracalcium phosphate-phosphoserine.
Note: Scoring system: 0 = presence of a bridge with 1 or several large radiolucent areas; 1 = presence of a bridge with 1 or several small radiolucent areas; 2 = presence of a bridge without radiolucent areas.
CT Scans
CT scans were analyzed for the mean total volume percentages of the intertransverse bridge by density ranges in Hounsfield units and by time point (Figure 5). TTCP-PS demonstrated a distributed density across each density range starting at week 2 and maintained this distribution throughout the study. However, the control substance, which started at a lower density distribution, demonstrated an increase in the higher density distribution ranges.
Figure 5.

Area density distribution. TTCP-PS, tetracalcium phosphate-phosphoserine.
CT-quantified bone-material composite volume was determined using manual segmentation of each of the fusion masses. Additionally, the composite volume measurements, which included both bone and biomaterial, showed statistically significant (P < 0.01) higher volume in TTCP-PS (6.05 ± 0.52 mm3) than in the control group (3.68 ± 0.53 mm3) across all time points (Figure 6).
Figure 6.

Total material volume of TTCP-PS and control. TTCP-PS, tetracalcium phosphate-phosphoserine.
Macroscopic Observations
Macroscopic observations of the implantation sites were conducted at termination.
Control Group
In most cases (5 of 6 sites) at the 12-week time point, the control substance was not visible and/or not detectable by palpation between the transverse processes. Slight muscular tissue discoloration was observed in 2 of 6 sites at the 26-week time point. The control substance was visible and/or detectable by palpation between the transverse processes in 3 of 6 sites. When visible, the substance was whitish and hard with a granular surface and had a bridge shape in 2 of the 3 detectable sites. At the 52-week time point, the control substance was visible and detectable via palpation between the transverse processes in 5 of the 8 sites. When visible, the substance was hard, with either a smooth or granular surface, and a bridge shape.
TTCP-PS Group
Very slight muscular tissue discoloration was observed at 12 weeks and 26 weeks (4 of 6 sites at each time point). No discoloration was observed at 52 weeks. In most cases, TTCP-PS was visible and/or detectable by palpation between the transverse processes at 12 weeks (5 of the 6 sites), at 26 weeks (in all 6 sites), and at 52 weeks (all 6 sites). The TTCP-PS was whitish, gray, or brown at both 12 and 52 weeks, and whitish or gray at 26 weeks. When visible, it was hard, granular, or smooth and had a bridge shape (4 of 5 detectable sites at 12 weeks and all sites at 26 weeks). At 52 weeks, TTCP-PS had a bridge shape in all detectable sites, with a hard and smooth consistency. These observations were confirmed by x-ray imaging at each time point, as visibility and detectability by touch correlated with minimal or no radiolucent areas in the intertransverse bridge on postoperative x-rays images.
Histological Analysis
Semiquantitative Histopathologic Analysis
Initial, independent analysis across the implantation sites was conducted, and several parameters were evaluated based on the ISO 10993-6 standard when considering the cell type and tissue response. Results (Table 3) demonstrate that TTCP-PS was “nonreactive” compared with the control sample. Additionally, TTCP-PS showed increased scores compared with control when considering bone parameters such as new bone formation, osteointegration, osteoconduction, and bone bridging.
Table 3.
Bioreactivity scores of TTCP-PS vs control per ISO 10993-6:2016.
| Variable | TTCP-PS | Control | ||||
| 12 wk | 26 wk | 52 wk | 12 wk | 26 wk | 52 wk | |
| Fibrin | 0 | 0 | 0 | 0 | 0.2 | 0 |
| Polymorphonuclear cells | 0 | 0 | 0 | 0 | 0.2 | 0 |
| Lymphocytes | 0 | 0 | 0.4 | 0 | 0.7 | 0 |
| Plasma cells | 0 | 0 | 0 | 0 | 0.5 | 0 |
| Macrophages | 2 | 1.2 | 1.2 | 2.2 | 2.2 | 1.3 |
| Giant/osteoclastic cells | 2 | 1 | 1 | 2.2 | 2.2 | 1.2 |
| Necrosis | 0.2 | 0 | 0 | 0 | 0 | 0 |
| Subtotal (×2) | 42 | 22 | 26 | 52.8 | 72 | 30 |
| Tissue generation, muscle | 0.6 | 1 | 0 | 1 | 1 | 0 |
| Neovascularization | 2 | 2 | 2.4 | 2 | 2 | 2.2 |
| Fibrosis/fibrous encapsulation | 2 | 0.4 | 1.2 | 2 | 2.3 | 2 |
| Osteolysis | 0 | 0 | 0.2 | 0 | 0 | 0 |
| Subtotal | 23 | 17 | 19 | 30 | 31.8 | 25.2 |
| Group total | 65 | 39 | 45 | 82.8 | 103.8 | 55.2 |
| Average | 13 | 7.8 | 9 | 13.8 | 17.3 | 9.2 |
| Osteoblastic cells | 3 | 3 | 2 | 2.2 | 2.7 | 2.2 |
| Bone neoformation | 2.8 | 3.4 | 2.8 | 1.8 | 1.8 | 2.3 |
| Osteointegration | 3 | 4 | 3.2 | 2.3 | 2.3 | 2.7 |
| Osteoconduction | 2.8 | 3.8 | 3.2 | 1.8 | 1.8 | 2.5 |
| Bone density | 2.8 | 2.8 | 3.2 | 1.8 | 1.8 | 2.3 |
| Bone remodeling | 2 | 2.4 | 2.8 | 1 | 2 | 2.7 |
| Material degradation | 1 | 2 | 3 | 1.2 | 2 | 2.7 |
| Implant migration | 1 | 1 | 0.6 | 1 | 1 | 1 |
| Bone growth toward fusion | 3 | 2.8 | 3 | 1.3 | 1.8 | 2.5 |
| Implant interruption | 1.8 | 0.4 | 2 | 4 | 4 | 4 |
| Bone bridging | 2/5 | 4/5 | 2/5 | 0/6 | 0/6 | 1/6 |
| 12 wk | 26 wk | 52 wk | ||||
| Reactivity scores (TTCP-PS - Control)* | 0 | 0 | 0 | |||
Abbreviation: TTCP-PS, tetracalcium phosphate-phosphoserine.
Used to determine reactivity ranking. A negative difference is recorded as zero.
At the 12-week time point (Figure 7-top), the TTCP-PS sites showed higher levels of osteointegration, bone growth toward fusion, bone-implant continuum, new bone formation, and lower levels of implant interruption than the control substance sites.
Figure 7.

Histological results at 12, 26, and 52 weeks after surgery using modified paragon stain demonstrate native bone (blue), osteoid/new bone formation (pink/purple), and implanted substance (TTCP-PS vs control, black) at the various time points. TTCP-PS, tetracalcium phosphate-phosphoserine.
These levels were consistent among the TTCP-PS sites at the 26-week time point as well (Figure 7, middle), with the addition of higher levels of osteoconduction than in the control samples.
At the 52-week time point, the control samples displayed results closer to those of the TTCP-PS samples, although the TTCP-PS samples maintained higher levels in every analysis parameter (Figure 7, bottom).
Histomorphometry Analysis
Histomorphometric analysis of the primary ROI at the substance implantation sites was conducted, and several parameters were evaluated (percentage of new bone, fibrous tissue, implanted material, and void space; Figure 8).
Figure 8.

Total area analysis by histomorphometry. TTCP-PS, tetracalcium phosphate-phosphoserine.
At the 12-week time point, there was no statistical difference in the percentage area of new bone formation between the 2 groups; however, TTCP-PS showed a higher average bone percentage than control within the ROI (36% in TTCP-PS, 24% in control, P = 0.10). The TTCP-PS sites contained a statistically higher mean percentage area of remaining implant material (40.4% in TTCP-PS, 24.8% in the control, P = 0.007). The 12-week mean percentage area of fibrous tissue ingrowth in the control sites neared half of the ROI (49.3%), while the mean percentage in TTCP-PS sites was less than a quarter of the indicated area (22.5%). Both substances displayed statistically similar (P = 0.41) percentage void areas (1.3% in TTCP-PS, 2.2% in the control) at this time point.
By the 26-week time point, the mean percentage of bone area in TTCP-PS sites rose to over half of the ROI (57%), while the percentage bone area in the control sites remained below one-third of the observed area (31%), a statistically significant difference (P = 0.001). The remaining implant material was reduced by approximately 50% for both groups from 12 to 26 weeks, from 40.4% to 21.1% for TTCP-PS, and 24.8% to 14.1% for control. The mean percentage area of fibrous tissue ingrowth decreased in TTCP-PS sites (from 22.5% to 15.6%) but remained similar to the 12-week level in the control sites (49%, a difference of <1% across the time points), also a statistically significant difference (P < 0.001). The mean percentage void area remained similar for both substances at 26 weeks (6.1% for TTCP-PS, 6.2% for control, P = 0.97).
After 52 weeks, no statistical difference existed between the percentage bone areas in TTCP-PS sites and control sites, although the mean percentage bone area remained higher for TTCP-PS (52% in TTCP-PS, 45% in the control, P = 0.40). A similar trend was observed for mean percentage implant area (17.7% in TTCP-PS, 10.5% in the control, P = 0.073) as well as mean percentage fibrous tissue (22.4% in TTCP-PS, 29.5% in the control, P = 0.47) remaining in both groups. The percentage void area in the control sites at this time point was almost twice as large as that of the TTCP-PS sites (7.7% in TTCP-PS, 14.8% in the control, P = 0.08), despite a lack of statistical difference.
Systemic Effects
Detailed macroscopic and histopathologic observations of the lymph nodes at termination revealed that the implantation of both substances adjacent to the vertebral transverse processes of sheep for all time points did not produce any pathologic effects on the iliac lymph nodes. At termination, a few sheep exhibited abnormal masses on the liver or the abdominal conjunctive tissues in the abdominal cavity. These masses correlated histologically with parasitic cysts, necrotic granulomas, fat necrosis, or granulomatous inflammation and were not related to the TTCP-PS or control substances.
Discussion
This study was designed to investigate whether a novel wet-field bone adhesive (TTCP-PS) could be used as a bone substitute for lumbar spine fusion in an ovine model compared with a marketed control. The results from this study not only demonstrate proof-of-concept but also feasibility in a large animal model and the future potential of this biomaterial for clinical use.
Both TTCP-PS and the control substance appeared to be systemically safe in this sheep model. No pathologic regional lymph node changes were observed, consistent with non-toxicity of both implanted substances. Several isolated liver and abdominal masses were associated with parasitic cysts unrelated to the fusion materials.
Macroscopic observations, as well as analysis of x-ray images, CT density and volume, and histomorphology, demonstrate that TTCP-PS appeared to promote more rapid and complete bone healing than the control.
Gross observation and palpation of fusion areas showed subjectively greater retained and/or new intertransverse fusion mass in the TTCP-PS group than in controls at early test points, suggesting more rapid bone healing and implant volume maintenance with TTCP-PS.
Plain radiographs demonstrated more rapid healing of the TTCP-PS subjects with fewer large radiolucent areas in the graft at 10 and 18 weeks as compared with the controls and persistent small radiolucent areas in the controls at 26 weeks.
CT densitometry showed that TTCP-PS produced a higher initial density distribution in the fusion masses. The controls initially had a lower density distribution, but this gradually increased over time. Even so, the control group had a lower overall density as compared with the TTCP-PS group. Although the density of the TTCP-PS fusion masses did not significantly change over time, histological analysis demonstrated substantial bone substitution. The TTCP-PS group had significantly higher fusion volumes than controls, but it was difficult to distinguish TTCP-PS from the surrounding bone due to the similarity of CT density of bone and TTCP-PS.
CT data showed higher volume for the TTCP-PS fusion masses at all time points (P < 0.01). CT volume was high for both groups, though. Manual segmentation and the difficulty of differentiating between the biomaterials and the adjacent bone (eg, transverse processes) may have contributed to this.
Histopathology and Histomorphology
Semiquantitative histopathology of the intertransverse tissues showed a greater degree of osteointegration, new bone growth, and continuity between transverse processes at all time periods for TTCP-PS compared with the control. Quantifying the histology in the regions of interest, TTCP-PS consistently showed statistically significantly greater material retention, bone ingrowth, and reduced fibrous tissue ingrowth than the control. At 52 weeks, however, both implanted substances were statistically similar, although a greater qualitative level of bone healing was observed with TTCP-PS.
It appears from this analysis that TTCP-PS may promote more rapid and complete bone healing in a posterolateral transverse process fusion model than a commercially available, comparable bone graft substitute. Clinically, TTCP-PS demonstrates properties that satisfy many of the desired characteristics of an ideal bone adhesive: biocompatibility, non-toxicity, adequate adhesion, rapid wet-field setting, practical clinical handling, bioresorption, and bone healing promotion.17
Improved bone healing may be related to both the bone regenerative and adhesive properties of TTCP-PS. Tetracalcium phosphate cements have well-characterized osteoconductive properties principally derived from their conversion during hydrolytic setting to hydroxyapatite. Adhesion to the fusion area in the lateral gutters provides close apposition of graft material to the host bony surfaces, facilitating fusion. Migration of other graft material has been reported with lower fusion rates in a sheep PLF model.18 Graft adhesion, such as occurs with TTCP-PS, may limit migration of graft material and promote fusion.
Clinical Significance
For posterolateral onlay spinal fusion, the use of TTCP-PS may improve upon the current standard of care by allowing for more predictable graft volume maintenance and fusion through the graft mass. Possibly, with its load-bearing and adhesive capabilities, TTCP-PS may allow for a reduction of the amount of hardware needed to achieve structural stability to allow for fusion of the intended site (although this remains to be studied).
Limitations and Future Research
This project was intended to evaluate the use of TTCP-PS in posterolateral spinal fusion procedures using a large animal model. Human clinical use will require further study, specifically the correlation between graft volume maintenance and the turnover of the TTCP-PS into regenerated bone of the fusion mass. Its load-bearing and adhesive properties were not the subject of this investigation. How these mechanical characteristics vary longitudinally with resorption and replacement of TTCP-PS by native bone will need to be fully defined. Furthermore, the role of TTCP-PS as an adjunct to or even a potential replacement for internal fixation will need to be established clinically.
Conclusions
TTCP-PS was demonstrated to be safe and effective as a bone graft substitute in a large animal transverse process spinal fusion model. In this study, the bioresorbable bone adhesive appeared to have some advantages compared with a commercially available bone graft substitute. TTCP-PS promoted more rapid bone ingrowth, a greater density of the fusion mass, and a greater volume of fusion bone as seen through clinical observations, radiographic data, and histologic assessments. These properties of TTCP-PS may improve the reliability and success of posterolateral spinal fusion procedures.
References
- 1. Bono CM, Lee CK. Critical analysis of trends in fusion for degenerative disc disease over the past 20 years: influence of technique on fusion rate and clinical outcome. Spine (Phila Pa 1976). 2004;29(4):455–463. 10.1097/01.brs.0000090825.94611.28 [DOI] [PubMed] [Google Scholar]
- 2. Deyo RA, Gray DT, Kreuter W, Mirza S, Martin BI. United States trends in lumbar fusion surgery for degenerative conditions. Spine (Phila Pa 1976). 2005;30(12):1441–1445. 10.1097/01.brs.0000166503.37969.8a [DOI] [PubMed] [Google Scholar]
- 3. Rajaee SS, Bae HW, Kanim LEA, Delamarter RB. Spinal fusion in the United States: analysis of trends from 1998 to 2008. Spine (Phila Pa 1976). 2012;37(1):67–76. 10.1097/BRS.0b013e31820cccfb [DOI] [PubMed] [Google Scholar]
- 4. Surgery BA. Actifuse. Baxter Advanced Surgery. https://advancedsurgery.baxter.com/actifuse. 6 June 2023.
- 5. Guppy KH, Royse KE, Norheim EP, et al. Operative nonunion rates in posterolateral lumbar fusions: analysis of a cohort of 2591 patients from a national spine registry. World Neurosurg. 2021;145:e131–e140. 10.1016/j.wneu.2020.09.142 [DOI] [PubMed] [Google Scholar]
- 6. Sheikh Z, Abdallah M-N, Hanafi AA, Misbahuddin S, Rashid H, Glogauer M. Mechanisms of in vivo degradation and resorption of calcium phosphate based biomaterials. Materials (Basel). 2015;8(11):7913–7925. 10.3390/ma8115430 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Kirillova A, Kelly C, von Windheim N, Gall K. Bioinspired mineral-organic bioresorbable bone adhesive. Adv Healthc Mater. 2018;7(17):e1800467. 10.1002/adhm.201800467 [DOI] [PubMed] [Google Scholar]
- 8. Kirillova A, Nillissen O, Liu S, Kelly C, Gall K. Reinforcement and fatigue of a bioinspired mineral-organic bioresorbable bone adhesive. Adv Healthc Mater. 2021;10(2):e2001058. 10.1002/adhm.202001058 [DOI] [PubMed] [Google Scholar]
- 9. Kirillova A, Kelly C, Liu S, Francovich J, Gall K. High‐strength composites based on 3D printed porous scaffolds infused with a bioresorbable mineral–organic bone adhesive. Adv Eng Mater. 2022;24(7):2101367. 10.1002/adem.202101367 [DOI] [Google Scholar]
- 10. Kesseli FP, Lauer CS, Baker I, Mirica KA, Van Citters DW. Identification of a calcium phosphoserine coordination network in an adhesive organo-apatitic bone cement system. Acta Biomater. 2020;105:280–289. 10.1016/j.actbio.2020.01.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Actifuse ABX: Instructions for Use. 2023. https://advancedsurgery.baxter.com/sites/g/files/ebysai3391/files/2022-01/ACTIFUSE_ABX_IFU_25005_10.2009.pdf.
- 12. Wilke H-J, Kettler A, Claes LE. Are sheep spines a valid biomechanical model for human spines? Spine (Phila Pa 1976). 1997;22(20):2365–2374. 10.1097/00007632-199710150-00009 [DOI] [PubMed] [Google Scholar]
- 13. Kettler A, Liakos L, Haegele B, Wilke H-J. Are the spines of calf, pig and sheep suitable models for pre-clinical implant tests? Eur Spine J. 2007;16(12):2186–2192. 10.1007/s00586-007-0485-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Donath K, Breuner G. A method for the study of undecalcified bones and teeth with attached soft tissues. J Oral Pathology Medicine. 1982;11(4):318–326. 10.1111/j.1600-0714.1982.tb00172.x [DOI] [PubMed] [Google Scholar]
- 15. [ISO] IOfS. Biological Evaluation of Medical Devices – Part 6: Tests for Local Effects after Implantation, ISO 10993-6:2016. www.iso.org/obp/ui/#iso:std:iso:10993:-6. 11 November 2023.
- 16. Gadomski BC, Labus KM, Puttlitz CM, et al. Evaluation of lumbar spinal fusion utilizing recombinant human platelet derived growth factor-b chain homodimer (rhpdgf-BB) combined with a bovine collagen/β-tricalcium phosphate (β-TCP) matrix in an ovine model. JOR SPINE. 2021;4(3):e1166. 10.1002/jsp2.1166 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Bingol HB, Bender JCME, Opsteen JA, Leeuwenburgh SCG. Bone adhesive materials: from bench to bedside. Mater Today Bio. 2023;19(100599):100599. 10.1016/j.mtbio.2023.100599 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Axelsen MG, Overgaard S, Jespersen SM, Ding M. Comparison of synthetic bone graft ABM/p-15 and allograft on uninstrumented posterior lumbar spine fusion in sheep. J Orthop Surg Res. 2019;14(1):2. 10.1186/s13018-018-1042-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
