Assessment of Bioplex Interbody Fusion Device in a Sheep Lumbar Fusion Model

Abstract

Objective

The purpose of this study was to evaluate the bioPlex bioresorbable interbody device in a sheep lumbar fusion model and compare it to the concorde®, a standard carbon fiber interbody cage.

Background

Lumbar interbody fusion devices are made from a variety of materials, including titanium alloys, carbon-fiber, and PEEK. The BioPlex Continuous Phase Composite (CPC) is a unique bioresorbable material comprised of Pro Osteon 500R and 70:30 Poly (L/D, L-lactic acid). The BioPlex device is radiolucent, resorbable and due to its bulk nanoporosity of 8%, has a more consistent degradation profile as compared to a polymer alone.

Methods

A total of twenty five male Suffolk sheep were used in this study; nineteen of which were implanted with a bioPlex or concorde device at the L3-L4 and L5-L6 levels using a modified transforaminal/lateral approach. A discectomy was performed and each implant (filled with autologous bone) was placed within the disc space. The sheep were sacrificed at 6, 12, 24 months postimplantation. Fusion was assessed via motion, radiographic and histological data.

Results

The BioPlex and Concorde implanted levels had significantly less motion (p<0.05) than the normal controls in flexion/extension and lateral bending at 6, 12, and 24 months. No significant difference in motion was detected between the bioPlex and concorde implants. CT fusion scores correlated with the motion analysis in all the three cases.

Conclusion

In comparison to the concorde device, the bioPlex implant appears to have equivalent radiographic and biomechanical fusion success.

Introduction

Spinal fusion is one of the most widely used modalities for treating degenerative spinal disorders, especially in the lumbar region. The ultimate goal of fusion is to obtain a solid union between two or more vertebrae. Although alternative methods to treat degenerative disc disease are increasing in popularity, the efficacy of methods such as total disc replacement are not yet fully understood, especially in the lumbar spine¹.

In 2003, more than 325,000 spinal fusions were performed, of which approximately 162,000 involved the lumbar spine². There are a number of metrics by which success may be judged with respect to fusion such as: a healed radiographic appearance, pain relief, and prevention of neurologic injury. Though there are a number of fusion approaches/techniques, an interbody fusion is the most biomechanically stable of all available fusion option; fusion rates are higher and stabilization is more effectively maintained with interbody fusion^3-4.

Based on the initial experiences of Bagby with a stainless steel basket⁵, interbody fusion cage technologies have evolved rapidly over the past two decades. In addition to cage design, the choice of material has expanded (e.g., stainless steel, titanium, carbon fiber, PEEK, and ceramics)^5-8. Although carbon fiber and PEEK cages provide an advantage over metal cages because they are radiolucent and less stiff, long-term problems such as the release of wear debris/particles and breakage of the cage have been reported^9-12. The inherent limitations of current cage devices and the temporary role of fusion cages were incentives for the development of bioresorbable polylactic acid-based cages. BioPlex Continuous Phase Composite (CPC) is one such unique bioresorbable material comprised of Pro Osteon 500R and 70:30 Poly L/D, L-lactic acid (PLDLLA).

The properties of the composite are significantly improved by filling the macroporosity of Pro Osteon 500R with PLDLLA polymer. As a base material, Pro Osteon 500R is characterized by a 100% inter-connected, porous structure that allows for polymer penetration through the entire macroporosity. The ceramic consists of a thin layer of hydroxyapatite over a calcium carbonate skeleton.

The BioPlex composite has the ability to support bone growth within the implant walls which results in more bone present at the implant site. The presence of microporosity within the BioPlex ceramic phase allows for degradation products to readily and easily be removed from within the implant preventing lactic acid build up and associated implant complications. The presence of microporosity also gives BioPlex an optimal degradation profile that allows for the gradual loading of the newly forming bone. The presence of residual Pro Osteon within the polymer during the late stages of polymer degradation allows for the remaining calcium carbonate to react with lactic acid thereby neutralizing the acid. Pro Osteon also allows the composite to be visualized on radiographs.

In this study, we compared BioPlex interbody fusion device with Concorde; a popular FDA approved carbon fiber interbody device. Motion, radiographic and histological analyses were the parameters used to compare the two devices.

Materials and Methods

Specimen Preparation

A total of twenty-five skeletally mature male Suffolk sheep were used for this study. Six sheep served as controls, while interbody fusion devices were introduced at levels L3-L4 and L5-L6 for the remaining nineteen sheep (Figure 1). Each animal and spinal level was randomly assigned an implant (i.e., either a BioPlex or a Concorde interbody device).

Figure 1. Diagram showing distribution of sheep.

For both implants, a modified transforaminal approach was used. Corticocancellous autogenous bone graft was harvested locally. A posterior midline approach was used to expose the posterior aspect of the spine. A dorsal midline skin incision was made from T10-S1. This extended incision was necessary to ensure sufficient muscle retraction. The fascia and supraspinous ligament were incised around each spinous process and on the midline between each treated motion segment. The incision was deepened to the laminae to complete the midline muscle separation. The multifidus was incised and freed from the mamillary processes. The incision was made directly on the bone to limit hemorrhage and damage to the underlying dorsal nerve root and vessels.

Lateral retraction of the muscles exposed the mamillary, cranial, caudal, and transverse processes and nerve roots and vessels. A hemifacetectomy of the cranial and caudal processes was then preformed. The pars interarticularis was removed, allowing access to the lateral aspect of the disc. A near-complete discectomy was performed from this approach. The disc space was then implanted with a BioPlex or Concorde cage filled with autograft. Pedicle screws and rods were used for posterior stabilization of the spinal units in four animals to check if posterior instrumentation accelerated fusion. No instrumentation was used for the remaining animals.

The sheep were euthanized at 6, 12, or 24 months after surgery with Euthasol followed by removal of the spinal column. The spine was scanned using CT before storing them in double plastic freezer bags at -20°C. Prior to testing, the specimens were thawed overnight. The thawed specimens were carefully cleaned by taking off all the musculature while avoiding disruption of any ligaments, joints and discs. The levels of interest (L3-L4 and L5-L6) were then dissected from each spine, wrapped in saline drenched gauze and stored again in freezer bags. On the day of testing, the samples were removed from the freezer and thawed.

Experimental Setup

The caudal vertebra of the functional spinal unit (FSU) was secured on a custom made fixture using bolts to achieve a rigid fixation. A loading frame was attached to the cephalad vertebra. A sensor consisting of three IRed markers was rigidly attached to each vertebral body (Figure 2).

Figure 2. Specimen mounted on the fixture with loading frame and markers attached.

A custom designed testing rig was used to apply pure moments using dead weights and a pulley system. Maximum moments of 6.0 Nm were applied in four incremental steps of 1.5 Nm in flexion, extension, lateral bending and axial rotation. The resulting motions were recorded using an optoelectronic camera system - Optotrak 3020 (Northern Digital, Waterloo, Ontario, Canada). The angular rotations of the cephalad vertebra with respect to the caudal vertebra were calculated for each loading condition using the software Motion Monitor system (Innsport, Chicago, IL). Relative angles were calculated as Euler angles with the sequence Z (lateral bending), Y (axial rotation), X (flexion - extension).

The six intact / control specimens underwent mechanical testing first followed by the implanted specimens. Fusion was assessed by comparing the motion of the instrumented specimens with that of the control (intact) specimens.

Radiographic Analysis

The CT scans of each specimen were analyzed by an Orthopaedic Surgeon and the specimens were given scores on a scale of 1–5 to quantify fusion (Table 1).

Table 1.

Fusion Assessment scale

Fusion Assessment	Radiographic criteria Present
5. Definitely Fused	Presence of [1] and [2]. Absence of [3]
4. Probably Fused	Presence of [1], partial [2]. Absence of [3]
3. Indeterminate	Presence of [1] or [2] and [3] on one image
2. Probably Not Fused	Absence of [1], [2], or presence of [3] on one image
1. Definitely Not Fused	Presence of [3]

Radiographic Criteria:

[1] Bridging trabecular bone on at least two sequential images

[2] Cortication of the peripheral edges of the fusion mass

[3] Presence of an identifiable cleft on sequential images

Histology

Following the experimental investigation, histological examination was performed on each implanted specimen using parasagittal sections of the fusion mass. The slides were stained using H&E stain. The slides were then analyzed for the presence of new bone formation (NBF), quality of NBF, and inflammation and presence of residual graft.

Data Analysis

Student t-test with a confidence interval of 95% (p = 0.05) was used to analyze all the data.

Results

Control (Intact) Specimens

Table 2 summarizes the range of motion for the 6 Nm moment (i.e., maximum moment) in flexion, extension, right & left lateral bending and right & left axial rotation. No significant differences (p>0.05) in motion were observed between the two functional levels (L3-L4 and L5-L6) tested.

Table 2.

Average range of motion observed in intact specimens at maximum moment (6Nm)

	Range of Motion (degrees)
	L3/L4	L5/L6
Flexion	4.63 ± 1.53	5.14 ± 2.11
Extension	-4.78 ± 1.07	-5.23 ± 2.46
Right Lateral Bending	5.85 ± 1.18	5.34 ± 0.69
Left Lateral bending	-5.09 ± 0.96	-5.01 ± 0.42
Right Axial Rotation	-2.03 ± 0.57	-1.53 ± 0.81
Left Axial Rotation	1.59 ± 0.44	0.92 ± 0.34

Implanted Specimens

Figure 3 summarizes the motions for each of the 6 Nm loading modes at the various time points (6, 12, or 24 months) for both the BioPlex and Concord implanted specimens. The motion of implanted levels (both BioPlex and Concorde) decreased gradually over time. The Bio-Plex and Concord implanted specimens experienced a significant decrease in motion (p<0.05) at each interval of time, as compared to the control specimens. However, no significant difference (p>0.05) was observed between the two implants.

Figure 3. comparison of motion over time.

Four specimens in the six month group had posterior instrumentation. Though there was a trend for posterior instrumentation to reduce the motion in comparison to the non-instrumented specimens, the reduction in motion was not significant.

Radiographic Results

A summary of mean fusion scores and mean fusion quality scores for all implanted specimens is shown in Figure 4. In accordance with the motion analysis results, the radiographic analysis also showed an improvement in fusion over time. When BioPlex and Concorde scores were compared, there was no significant difference (p>0.05) in the scores at all time intervals. Comparing instrumented and un-instrumented specimens, BioPlex showed no significant difference which was in agreement with the motion analysis results.

Figure 4. comparison of Radiographic scores over time.

Histology Results

Figure 5 shows sample histology slides for the specimens implanted with the BioPlex and Concorde cages sacrificed at 6 , 12, and 24 months. At 6 months, there were clear signs of new bone formation in both cases.

Figure 5. Histology slides at the various time intervals.

At 12 months, new bone formation was extensive for both implants. Bone formation in and around the Bio- Plex cage was clearly visible, accompanied by implant resorption. The Concorde device, in contrast, is clearly visible with bone formation around the implant.

At 24 months, the BioPlex implant had resorbed completely and the space was filled with bone, thereby fusing the two vertebrae. The Concord implant, being composed of carbon fiber remained clearly visible with bone growth surrounding the implant.

Discussion

The aim of the present study was to evaluate the BioPlex interbody fusion device in a sheep model and compare it with the current standard, the Concorde carbon fiber cage. We were also interested whether posterior instrumentation accelerated fusion and hence four specimens had additional posterior instrumentation accompanying the interbody cages. The specimens were tested for motion in the sagittal plane (flexion-extension), coronal plane (lateral bending), and the transverse plane (axial rotation).

The six month specimens were divided into two groups depending on the use of posterior instrumentation. The instrumented group had four specimens (eight FSUs; four BioPlex and four Concorde). Although the specimens with posterior instrumentation showed a reduced motion in all directions as compared to the un-instrumented specimens, the reduction was not significant. This was observed in both the Concorde and BioPlex implanted specimens.

The total number of un-instrumented specimens for the six month group was five (nine FSUs), but four FSUs had to be excluded from mechanical testing giving us five FSUs; three of BioPlex and two of Concorde. Both the BioPlex and Concorde implanted un-instrumented specimens showed a significant reduction in flexion, extension and lateral bending when compared with control. The reduction in axial rotation was not significant. This might be due to the fact that the motion observed in axial rotation in the lumbar spine itself is low. The instrumented specimens, however, demonstrated a significant reduction in motion, not only in flexion–extension and lateral bending, but also in axial rotation. The differences in the motion observed between the BioPlex and Concorde implanted specimens were not significant for both the instrumented and un-instrumented specimens. In these specimens, most reduction was observed for lateral bending where the average reduction was more than 80% for both BioPlex and Concorde. Meanwhile the average reduction in flexion–extension and axial rotation was about 50%.

In the 12 month group, there were a total of four specimens (i.e. eight FSUs; four BioPlex and four Concorde). There was a significant reduction in motion in all directions for both the BioPlex and Concorde implanted specimen. In this group, most reduction was observed in flexion – extension (75% in BioPlex and 90% in Concorde) followed by lateral bending and axial rotation (60% in BioPlex and 75 % in Concorde). In this group too, there was no statistically significant difference between BioPlex and Concorde.

There were six specimens in the 24 month group (12 FSUs; six BioPlex and six Concorde). In one BioPlex FSU, the level did not fuse due to improper placement of the implant during surgery. This sample was excluded from the analysis yielding five FSUs of BioPlex and six of Concorde. The motion observed in all of these specimens was least in comparison with the other groups giving us about 95% reduction in lateral bending, more than 90% reduction in flexion – extension, and more than 85% reduction in axial rotation. The reduction in motion was significant in all direction at a confidence interval of 99%. When compared with each other, the two implants did not differ statistically.

Considering all the un-instrumented groups together (i.e. 6 month, 12 month and 24 month), we see a gradual decrease in motion over time for flexion – extension and axial rotation. In lateral bending, the reduction was gradual for Concorde but not for BioPlex.

Other fusion related studies mostly compare stiffness as opposed to range of motion. Kanayama et al.¹³, a group from Japan also conducted a study on fusion in sheep. The sheep underwent posterolateral spinal fusion surgery at L3-L4 and L4-L5. Since the study was focused on the effect of instrumentation, one of the two levels had posterior instrumentation. The time period of the study was 8 and 16 weeks which was much less as compared to our study. When the six month results of our study were compared with 16 week results from the paper, the six month specimens (both instrumented and un-instrumented) were much stiffer.

There have been other similar fusion studies performed on various models. In a literature review by Oxland and Lund¹⁴, anterior and posterior approaches for cage insertion in human cadaveric models are compared with some early animal results. Comparing the results in this paper which includes studies by Wilder et al¹⁵, Brodke et al.¹⁶ and Butts et al.¹⁷ with our study, the 12 and 24 month results of our study are much better.

Radiographic results correlate to the mechanical testing results in most cases. As seen in testing results, radiographic scores also showed an increase in fusion and fusion quality over time. Both fusion and fusion quality scores showed no statistically significant difference between specimens implanted with BioPlex and Concorde interbody fusion devices for all time intervals. Comparing instrumented specimens with un-instrumented ones, BioPlex showed no difference but there was a statistically significant difference in both scores for specimens implanted with Concorde cages.

Histological analysis showed that new bone formation was similar between groups with the BioPlex having slightly more in the six month group with a slight transition to Concorde by 24 months. Quality of NBF was similar at six months with generally better quality at 12 and 24 months in the Concorde group. Interestingly at 24 months, about 50% of the Bioplex group had similar quality as Concorde; however another 50% had poor quality. This also inversely corresponded to inflammation scores in which Bioplex had similar to slightly lower scores. In general the residual graft was more readily detected in the Bioplex group at all time points.

The current study serves as the first study to biomechanically, radiographically and histologically demonstrate BioPlex as an interbody device for spinal fusion. The biomechanical results demonstrated that BioPlex was equally effective in fusing the spine as the standard Concorde cage. Radiographic assessment results corroborated the biomechanical data. According to the histologic analysis, there were clear signs of bone penetration, active degradation and long term biocompatibility.

The lack of any implant complications was the result of a variety of properties unique to BioPlex. Bone in growth, residual calcium carbonate, and nanoporosity within the ceramic phase all contribute to the overall biocompatibility of BioPlex throughout the entire degradation process. The presence of bone within the implant walls makes it easier for the tissue to resorb any released lactic acid. Overall, the properties of the BioPlex composite result in a unique material that eliminates implant related complications often associated with 100% polymer devices.

Summary Points

• 1.Specimens implanted with a BioPlex or Concorde implant showed a significant reduction in motion.
• 2.No significant difference in motion was observed between specimens implanted with the Bioplex and Concorde devices.
• 3.The BioPlex implants resorbed completely within 24 months.