Nianbin Hu

Adjacent Level Intradiscal Pressure and Segmental Kinematics Following A Cervical Total Disc Arthroplasty : An In Vitro Human Cadaveric Model

Study Design. In vitro investigation of cervical adjacent level intradiscal pressures (IDPs) following a total disc replacement arthroplasty. Objectives. The current in vitro study was undertaken to compare adjacent level IDPs and operative level kinematics following a cervical arthroplasty versus an arthrodesis procedure. Summary of Background Data. Clinical data indicate the incidence of symptomatic transition syndrome to be as high as 3% annually following a cervical interbody arthrodesis. Recent developments in the motion preservation technology should, in theory, minimize transition syndrome at the adjacent levels. Methods. A total of 10 human cadaveric cervical spines were used in this investigation. Following intact analysis, all specimens were sequentially reconstructed at C5–C6 with 1) total disc replacement (TDR), 2) allograft dowel, and 3) allograft dowel + anterior cervical plate. Testing was performed in displacement control under axial rotation, flexion/extension, and lateral bending loading modes. IDPs were recorded at C4–C5 and C6–C7 whereas peak range of motion (ROM) and NZ were monitored at C5–C6 level. Results. Similar IDPs were recorded between the intact condition and a TDR reconstruction at both adjacent levels under all loading modes (P > 0.05). However, the C4–C5 IDP values produced under flexion/extension testing for both arthrodesis treatments were significantly higher than the means obtained for the intact and disc replacement groups (P < 0.05). Similar intergroup differences were observed at the C6–C7 level; however, statistical significance was achieved during all three loading methods (P < 0.05). C5–C6 ROM analysis indicated a significantly lower ROM for both arthrodesis constructs compared with intact and TDR groups during flexion/extension testing (P < 0.05). No differences were recorded between the intact and the total disc replacement group under any loading conditions (P > 0.05). Conclusion. This is a first study to document that a cervical disc replacement arthroplasty procedure maintains adjacent level IDPs and reconstruction level kinematics near the preoperative values. Consequently, total disc replacement may provide an alternative to conventional surgical management of cervical discogenic pathology decreasing the incidence of symptomatic transition syndrome.

Biomechanical evaluation of total disc replacement arthroplasty: an in vitro human cadaveric model.

Study Design. This in vitro biomechanical study was undertaken to quantify the multidirectional intervertebral kinematics following total disc replacement arthroplasty compared to conventional stabilization techniques. Objective. Using an in vitro human cadaveric model, the primary objective was to compare the multidirectional flexibility properties and map the center of intervertebral rotation of total disc arthroplasty versus conventional threaded fusion cages and cages augmented with transpedicular fixation for single-level spinal instrumentation. Summary of Background Data. The utilization of motion-preserving implants versus instrumentation systems, which stabilize the operative segments, necessitates improved understanding of their comparative biomechanical properties. Methods. A total of eight human cadaveric lumbosacral spines (L2 to sacrum) were utilized in this investigation and biomechanically evaluated under the following L4–L5 reconstruction conditions: 1) intact spine; 2) SB Charitè disc prosthesis; 3) BAK cages; and 4) BAK cages + ISOLA pedicle screw/rod fixation (anteroposterior). The superior (L3–L4) and inferior (L5–S1) intervertebral levels remained uninstrumented to quantify adjacent level properties. Multidirectional flexibility included pure, unconstrained moments (±8 Nm) in axial rotation, flexion–extension, and lateral bending, with quantification of the operative and adjacent level range of motion and neutral zone, which were normalized to the intact spine condition. Results. The SB Charitè prosthesis indicated an average percentage increase in axial rotation range of motion by 44% compared to the intact condition (P < 0.05), whereas the BAK and anteroposterior reconstructions decreased range of motion by 29% and 80%, respectively (P < 0.05). The SB Charitè was significantly different from BAK and combined anteroposterior reconstructions (P < 0.05). Flexion–extension indicated a minor increase in range of motion for the SB Charitè (3%) versus the intact disc (P > 0.05), whereas the BAK and anteroposterior stabilization groups resulted in significant decreases in range of motion (BAK = 57%, anteroposterior = 93%) (P < 0.05) when compared to the intact and SB Charitè conditions. Based on flexion–extension radiographs, the intervertebral centers of rotation were in the posterior one-third of the operative intervertebral disc only for the SB Charitè reconstruction and intact spine condition, with definitive evidence of physiologic intervertebral translation (intact 2.06 ± 77 mm; SB III = 1.9 ± 0.98 mm). Conclusions. Total disc arthroplasty serves as the next frontier in the surgical management of discogenic spinal pathology. The SB Charitè restored motion to the level of the intact segment in flexion–extension and lateral bending and increased motion in axial rotation. The anterior annular resection necessary for device implantation and unconstrained design of the prosthesis account for this change in rotation. The normal lumbar flexion–extension axis of rotation is an ellipse rather than a single point. Only disc replacement rather than pedicle instrumentation or BAK interbody instrumentation preserves the kinematic properties and normal mapping of segmental motion at the operative and adjacent intervertebral disc levels.

Biomechanical analysis of rotational motions after disc arthroplasty : Implications for patients with adult deformities

Study Design. An anatomic and biomechanical bench-top basic scientific comparative analysis to determine the appropriateness of total disc replacement (TDR) in a lumbar spine with scoliotic tendencies. Objectives. Only limited data are currently available studying the application of disc replacement adjacent to scoliosis fusions. Theoretically, motion preservation should help delay the continuum of lumbar degeneration adjacent to scoliosis fusions and rotationally unstable lumbar segments. Summary of Background Data. As a tertiary referral center for failed TDR, we noticed an alarming number of lumbar spinal rotational iatrogenic instability patterns but none occurring in the cervical spine. It is appropriate to analyze the bench-top rotational stability of disc replacement to predict whether this new technology is feasible for a larger prospective clinical study in the treatment of degenerative scoliosis. Methods. Measurements were taken from 60 human specimens from the Hamann-Todd Osteological Collection: 1) to determine the rotational arc of influence (AOI) = the angle formed from the center of axial rotation to the outermost extent of the facet joints; and 2) to determine the relative anatomic size discrepancy between the left and right facets proportionately with the cross-sectional area of the intervertebral disc = facet/endplate ratio (FER). Biomechanical testing was performed using fresh frozen human cadaveric spines with the following conditions to determine the rotational stability: 1) intact; 2) resection of ALL, anulus, disc, and PLL simulating the preparation for a TDR; 3) a more radical anular resection; 4) entire 360° anular resection; and 4) insertion of the respective unconstrained-type disc replacement. Using a 6 degrees of freedom spine simulator, unconstrained pure moments of ±8.0 Nm (lumbar) and ±3.0 Nm (cervical) were used for axial rotation with quantification of the operative level range of motion and neutral zone, with data normalized to the intact spine condition. Results. There were anatomic limitations in the lumbar spine that make it less desirable to apply uncon-strained disc replacements; indeed, the spine was at risk for iatrogenic lumbar scoliosis. The anulus fibrosis, anterior longitudinal ligament, and the posterior longitudinal ligament are critical structures in preventing iatrogenic scoliosis. The lumbar facet joints are more posteriorly located and are smaller relative to the intervertebral disc, compared with this association in the cervical spine. Because the facet capsular ligaments are mechanically less effective with lower tensile strength in the lumbar spine, multiple-level arthroplasty tends to accentuate scoliotic tendencies; this is independent of prosthetic design and surgical technique. Discussion. Implantation of the lumbar TDR never restored the motion segment back to the rotational stability of the intact segment achieving a range of 120% to 140% rotational range of motion compared with the intact condition. This rotational instability proved to be additive as a two-level lumbar TDR resulted in between 240% and 260% increase in rotational instability compared with the intact condition. Conclusion. The neutral zone of the intact cervical spine was restored even using an unconstrained cervical TDR. The greater inherent rotational constraints of the cervical spine make it more amenable to stable multilevel arthroplasty compared with the lumbar spine.

General principles of total disc replacement arthroplasty: seventeen cases in a nonhuman primate model.

Study Design. To investigate the biomechanical, histochemical, and biologic ingrowth characteristics of two different lumbar disc prostheses—AcroFlex (DePuy-AcroMed) and the SB Charitè (DePuy-AcroMed)—for total disc replacement arthroplasty. Methods. A total of 17 mature baboons (n = 17, Papio cynocephalus) underwent L5–L6 total disc replacement procedures. The AcroFlex device (n = 10 levels) consisted of sintered titanium beaded ingrowth surfaces, bound together by a hexene-based polyolefin rubber core. The SB Charitè (n = 7 levels) device prosthetic vertebral end plates were cobalt chrome, covered by two layers of thin titanium with an electrochemically bonded hydroxyapatite coating and an ultra-high molecular weight polyethylene core. Results. Following 6-month survival periods, the range of motion of the SB Charitè and intact nonoperative controls under axial compression, flexion–extension, and lateral bending showed no statistical difference (P > 0.05). However, both exhibited greater range of motion compared to the AcroFlex treatments (P < 0.05). Plain film radiographic analysis showed no lucencies or loosening of any metallic prosthetic vertebral endplate. Gross histopathologic analysis of the AcroFlex and SB Charité prosthesis demonstrated excellent ingrowth at the level of the implant–bone interface, without evidence of fibrous tissue or synovium. Histochemical assays showed no local or systemic accumulation of particulate wear debris (titanium, ultra-high molecular weight polyethylene, or cobalt chrome) nor cytokines (TNF-&agr;, PGE2, IL-1, IL-2, or IL-6). Porous ingrowth calculations showed the mean ingrowth (linear apposition) ranging from 47.9% ± 9.12 for the SB Charitè device and 54.59% ± 13.24 for the AcroFlex device. Conclusions. The porous ingrowth, percentage pore ingrowth coverage at the bone–metal interface was more favorable for total disc replacement compared to that reported for cementless total joint components in the appendicular skeleton (range 10–30%). The reason for the improved degree of porous ingrowth in total disc replacement prostheses is probably due to ligamentotaxis causing sustained compression across the metal–bone interface. This project serves as the first comprehensive in vivo investigation comparing two different types of unconstrained disc prostheses with alternate in-growth surfaces and establishes an excellent research model in the evaluation of lumbar total disc replacement arthroplasty.

A comparison of retraction pressure during anterior cervical plate surgery and cervical disc replacement: a cadaveric study.

Background Context Dysphagia is a well-recognized complication after anterior cervical discectomy and fusion, observed in as high as 50% of cases by videofluoroscopic evaluation postoperatively. Esophageal injury due to surgical retraction is a complication due to which swallowing difficulties may ensue. There are limited published data evaluating the effect of soft tissue retraction on intraesophageal pressures during anterior cervical instrumentation procedures. Purpose The purpose of this study was to (a) measure the intraesophageal pressure secondary to retraction during anterior instrumentation, (b) determine whether any pressure differences exist between plating and cervical disc replacement, and (c) determine whether the surgical level or length of the plate influences the magnitude of intraesophageal pressure during retraction. Study Design An analysis of soft tissue retraction pressure was performed for anterior single-level and 3-level cervical plating and cervical disc replacement procedures. Methods Using a 4-cm transverse incision, a Smith-Robinson anterior approach to the cervical spine was performed on 7 fresh, frozen cadavers. The correct placement of an esophageal pressure-transducing catheter was confirmed by laryngoscopy, manual palpation of the esophagus, and fluoroscopic imaging. Three surgical instrumentation groups were used for comparisons: (a) single-level plate (b) single-level Porous Coated Motion cervical disc replacement, and (c) 3-level plate. Hand-held appendiceal retractors were used to retract the soft tissues during screw insertion into the plate and during application of the disc prosthesis into the interspace. Care was taken to exert just enough force on the retractors to allow the surgeon to move the desired implant into the correct position. In addition the individual performing the retraction was blinded to the procedure being performed—1-level plating, 3-level plating, or disk replacement. Fluoroscopy confirmed that the pressure sensors were directly behind the retractors during data acquisition. Results Significantly greater intraesophageal pressures were demonstrated for single-level cervical plating at C5–6 compared to that at C3–4 (P=0.036). Similarly, significantly greater pressures were recorded at C5–6 versus C3–4 for the 3-level plating group (P<0.001). In contrast, there was no statistically significant difference in pressures observed during disk replacement at C5–6 compared to that at C3–4 (P=0.084). Significantly greater pressures were recorded during single-level plating compared to disc replacement at both C3–4 (P=0.016) and C5–6 (P=0.016). Three-level plating demonstrated significantly greater pressures at C5–6 compared to disk replacement (P<0.001) but no statistically significant difference compared to disk replacement at C3–4 (P=0.333). The highest mean pressure, 154.5±49.5 mm Hg, was recorded at C5–6 level during insertion of the 3-level plates. Conclusions On the basis of the data presented here, anterior cervical plating results in significantly greater intraesophageal pressures when performed at C5–6 compared to C3–4. This holds regardless of whether the plate spans the distance from C3 to C6 (3-level plate) or the single C5–6 level. In addition, the insertion of the cervical disc replacement seems to require less esophageal retraction and hence reduced intraesophageal pressures when compared to anterior cervical plating.

Biomechanical comparison of single- and two-level cervical arthroplasty versus arthrodesis: effect on adjacent-level spinal kinematics.

BACKGROUND CONTEXT The use of motion-preserving spinal implants versus conventional arthrodesis instrumentation systems, which stabilize operative segments, necessitates improved understanding of their effect on spinal kinematics and the biomechanically optimal method for surgical reconstruction. PURPOSE The primary objective of this study was to measure operative- and adjacent-level kinematics after single- and two-level cervical arthroplasty and compare them with those after anterior cervical arthrodesis. A secondary objective was to locate the centers of intervertebral rotation at the operative and adjacent levels after arthroplasty and compare them to those after arthrodesis. STUDY DESIGN This biomechanical study used an in vitro human cadaveric model to compare the multidirectional flexibility kinematics of single- versus two-level cervical disc arthroplasty reconstructions. METHODS Eight cadaveric cervical spines (C2-T2) were biomechanically evaluated between Levels C4 and T1 in the intact condition and under the following reconstructions: single-level arthroplasty (C6-C7) using porous coated motion (PCM) device; single-level arthrodesis (C6-C7) using interbody cage with anterior plate; two-level arthroplasty (C5-C7) using PCM devices; two-level hybrid treatment of arthroplasty (C5-C6) using PCM device and arthrodesis (C6-C7) using cage/plate; and two-level arthrodesis (C5-C7) using cage/plate. Multidirectional flexibility testing used the Panjabi hybrid testing protocol, including pure moments for the intact condition with overall spinal motion replicated under displacement control for subsequent reconstructions. Unconstrained intact moments of +/-3.0 Nm were used for axial rotation, flexion-extension, and lateral bending testing with quantification of the operative- and adjacent-level range of motion (ROM) and neutral zone. The calculated centers of intervertebral rotation were compared for all intervertebral levels under flexion-extension conditions. RESULTS Axial rotation loading demonstrated a significant decrease in the C6-C7 ROM for the single-level arthrodesis group compared with the intact spine and the single-level arthroplasty group (p<.05). No differences were observed between the intact and single-level arthroplasty groups (p>.05). For the two-level hybrid treatment group, the C5-C6 ROM significantly increased compared with the intact, single-level arthroplasty, and two-level arthrodesis groups (p<.05). Moreover, a significant increase was observed in the adjacent-level (C7-T1) ROM for the two-level arthrodesis group compared with all other treatment groups (p<.05). Under flexion-extension, no differences were observed in C6-C7 ROM between the intact spine and single-level arthroplasty groups (p>.05). However, as expected, the single-level arthrodesis and two-level hybrid treatment groups demonstrated a decreased ROM at C6-C7 versus the intact spine and arthroplasty treatments (p<.05). In terms of adjacent-level effects, two-level arthrodesis (C5-C7) led to increased ROM in the inferior level (C7-T1) in axial rotation and flexion-extension compared with the intact spine and all other treatment groups (p<0.05). Lateral bending loading conditions demonstrated no significant difference among the treatment groups (p>.05). In flexion-extension, the centers of intervertebral rotation for the intact spine and single-level arthroplasty groups were localized in the central to posterior one-third of the inferior vertebral body for each motion segment: C5-C6, C6-C7, and C7-T1. The single-level arthrodesis group produced more diffuse centers of rotation, particularly at the operative (C6-C7) and inferior adjacent levels (C7-T1). CONCLUSIONS This study highlights the biomechanical effects of single- and two-level cervical arthroplasty versus single- and two-level arthrodesis on four functional spinal levels (C4-T1). Operative-level ROM was preserved with single- and two-level arthroplasty under all loading modes. The distal adjacent level (C7-T1) demonstrated the greatest increase among the four levels in ROM compared with the intact condition after two-level arthrodesis. These kinematic findings were corroborated by changes in the adjacent-level centers of rotation after arthrodesis and may suggest a biomechanical cause of adjacent-level disease secondary to cervical arthrodesis.

Porous coated motion cervical disc replacement: a biomechanical, histomorphometric, and biologic wear analysis in a caprine model.

Study Design. The biomechanical, histopathologic, and histomorphometric characteristics of cervical disc replacement were assessed in a caprine animal model. Objective. To investigate the biomechanical, porous ingrowth, and histopathologic characteristics of the Porous Coated Motion (PCM) Cervical Disc replacement (Cervitech, Inc., Rockaway, NJ). Summary of Background Data. As an alternative to anterior cervical interbody arthrodesis, an artificial cervical disc serves to replace the symptomatic degenerated disc, restore the functional biomechanical properties of the motion segment, and preserve neurologic function. Methods. There were 12 mature Nubian goats divided into 2 groups based on postoperative survival periods of 6 (n = 6) and 12 months (n = 6). Using an anterior surgical approach, a complete discectomy was performed at the C3–C4, followed by implantation of the PCM device. Functional outcomes of the disc prosthesis were based on computerized tomography (CT), multidirectional flexibility testing, undecalcified histology, histomorphometric, and immunocytochemical analyses. Results. There was no evidence of prosthesis loosening, or neurologic or vascular complications. CT showed the ability to image and assess the cervical spinal canal for the presence of compressive pathology in the area of the CoCrMo prosthesis. Multidirectional flexibility testing under axial rotation and lateral bending indicated no differences in the full range of intervertebral motion between the disc prosthesis and nonoperative controls (P > 0.05). Based on immunohistochemical and histologic analysis, there was no evidence of particulate debris, cytokines, or cellular apoptosis within the local or systemic tissues. Moreover, review of the spinal cord at the operative levels indicated no evidence of cord lesions, inflammatory reaction, wear particles, or significant pathologic changes in any treatment. Histomorphometric analysis at the metal-bone interface indicated the mean trabecular ingrowth of 40.5% ± 24.4% and 58.65% ± 28.04% for the 6 and 12-month treatments, respectively. Conclusion. To our knowledge, this serves as the first in vivo time-course study investigating the use of the PCM device for cervical arthroplasty. All 12 animals undergoing cervical disc replacement had no evidence of implant loosening, subluxation, or inflammatory reactions. PCM cervical arthroplasty permits unobstructed visualization of the spinal canal based on CT imaging. Segmental intervertebral motion was preserved under axial rotation and lateral bending loading conditions, while at the same time permitting porous osseointegration at the prosthesis-bone interface. Based on histopathologic review of all local and systemic tissues, there was no evidence of particulate wear debris, cytokines, cellular apoptosis, or significant pathologic changes in any treatment.

Ceramic granules enhanced with B2A peptide for lumbar interbody spine fusion: an experimental study using an instrumented model in sheep.

OBJECT New generations of devices for spinal interbody fusion are expected to arise from the combined use of bioactive peptides and porous implants. The purpose of this dose-ranging study was to evaluate the fusion characteristics of porous ceramic granules (CGs) coated with the bioactive peptide B2A2-K-NS (B2A) by using a model of instrumented lumbar interbody spinal fusion in sheep. METHODS Instrumented spinal arthrodesis was performed in 40 operative sites in 20 adult sheep. In each animal, posterior instrumentation (pedicle screw and rod) and a polyetheretherketone cage were placed in 2 single-level procedures (L2-3 and L4-5). All cages were packed with graft material prior to implantation. The graft materials were prepared by mixing (1:1 vol/vol) CGs with or without a B2A coating and morselized autograft. Ceramic granules were coated with B2A at 50, 100, 300, and 600 microg/ml granules (50-B2A/CG, 100-B2A/CG, 300-B2A/CG, and 600-B2A/CG, respectively), resulting in 4 B2A-coated groups plus a control group (uncoated CGs). Graft material from each of these groups was implanted in 8 operative sites. Four months after arthrodesis, interbody fusion status was assessed with CT, and the interbody site was further evaluated with quantitative histomorphometry. RESULTS All B2A/CG groups had higher CT-confirmed interbody fusion rates compared with those in controls (CGs only). Seven of 8 sites were fused in the 50-B2A/CG, 100-B2A/CG, and 300-B2A/CG groups, whereas 5 of 8 sites were fused in the group that had received uncoated CGs. New woven and lamellar bone spanned the fusion sites with excellent osseointegration. There was no heterotopic ossification or other untoward events attributed to the use of B2A/CG in any group. Each B2A/CG treatment produced more new bone than that in the CG group. CONCLUSIONS Bioactive treatment with B2A effectively enhanced the fusion capacity of porous CGs. These findings suggest that B2A/CG may well represent a new generation of biomaterials for lumbar interbody fusion and indicate that additional studies are warranted.

Preclinical evaluation of the Dynesys posterior spinal stabilization system: a nonhuman primate model

BACKGROUND CONTEXT Posterior dynamic spinal stabilization systems are intended to restore near-normal biomechanical function of the spine without inducing unnatural stresses to the spinal elements or eliciting a histopathological response. These devices must resist loosening within the challenging biomechanical environment of the lumbar spine. PURPOSE To determine the biomechanical effects of the Dynesys dynamic stabilization system (Zimmer, Inc., Warsaw, IN, USA) in the acute postoperative period and after 6 and 12 months in vivo; to examine the facet joints at the same postoperative intervals for signs of degeneration; and to measure the incidence of screw loosening after in vivo loading. STUDY DESIGN/SETTING This was an in vitro and in vivo animal survival study. METHODS Fourteen baboons were used. Eight animals underwent survival surgery to implant a posterior dynamic stabilization system spanning two lumbar levels. Six animals were sacrificed acutely, and their spines were biomechanically tested in the intact condition and with instrumentation implanted as described above. Six animals in the survival group were sacrificed at 6 months postoperatively and two animals at 12 months postoperatively. Their spines were biomechanically tested with instrumentation in situ and explanted. The facets were then processed using undecalcified technique. Microradiographs of the facets were examined for signs of arthrosis, inflammation, and degenerative changes. RESULTS The range of flexion-extension motion for the acute group of instrumented spines was 27% of the intact condition. After 6 months with instrumentation in situ, flexion-extension was 56% of the intact condition. After 12 months with instrumentation in situ, flexion-extension was 70% of the intact condition. With instrumentation explanted, flexion-extension at 6 and 12 months was not different from the intact condition (p>.05). Similar results were observed for lateral bending. There were no significant differences in axial rotation between any groups at any time point (p>.05). The facet joints at the operative and adjacent levels exhibited normal articular cartilage at both the 6- and 12-month postoperative time points. There was no evidence of facet arthrosis in any animal. At 6 months postoperatively, 0 of 36 screws exhibited radiolucency at the bone-metal interface. At 12 months postoperatively, 3 of 12 screws exhibited radiolucency. CONCLUSIONS After 12 months in vivo, spinal motions were stabilized by the dynamic instrumentation system. No facet arthrosis was observed at 6 and 12 months postoperatively. Explantation of the instrumentation restored motion to intact levels. A 25% rate of screw loosening (3 of 12 screws) was observed at the 12-month postoperative time point.

Comparative fixation methods of cervical disc arthroplasty versus conventional methods of anterior cervical arthrodesis: serration, teeth, keels, or screws?

OBJECT Using a synthetic vertebral model, the authors quantified the comparative fixation strengths and failure mechanisms of 6 cervical disc arthroplasty devices versus 2 conventional methods of cervical arthrodesis, highlighting biomechanical advantages of prosthetic endplate fixation properties. METHODS Eight cervical implant configurations were evaluated in the current investigation: 1) PCM Low Profile; 2) PCM V-Teeth; 3) PCM Modular Flange; 4) PCM Fixed Flange; 5) Prestige LP; 6) Kineflex/C disc; 7) anterior cervical plate + interbody cage; and 8) tricortical iliac crest. All PCM treatments contained a serrated implant surface (0.4 mm). The PCM V-Teeth and Prestige contained 2 additional rows of teeth, which were 1 mm and 2 mm high, respectively. The PCM Modular and Fixed Flanged devices and anterior cervical plate were augmented with 4 vertebral screws. Eight pullout tests were performed for each of the 8 conditions by using a synthetic fixation model consisting of solid rigid polyurethane foam blocks. Biomechanical testing was conducted using an 858 Bionix test system configured with an unconstrained testing platform. Implants were positioned between testing blocks, using a compressive preload of -267 N. Tensile load-to-failure testing was performed at 2.5 mm/second, with quantification of peak load at failure (in Newtons), implant surface area (in square millimeters), and failure mechanisms. RESULTS The mean loads at failure for the 8 implants were as follows: 257.4 +/- 28.54 for the PCM Low Profile; 308.8 +/- 15.31 for PCM V-Teeth; 496.36 +/- 40.01 for PCM Modular Flange; 528.03+/- 127.8 for PCM Fixed Flange; 306.4 +/- 31.3 for Prestige LP; 286.9 +/- 18.4 for Kineflex/C disc; 635.53 +/- 112.62 for anterior cervical plate + interbody cage; and 161.61 +/- 16.58 for tricortical iliac crest. The anterior plate exhibited the highest load at failure compared with all other treatments (p < 0.05). The PCM Modular and Fixed Flange PCM constructs in which screw fixation was used exhibited higher pullout loads than all other treatments except the anterior plate (p < 0.05). The PCM VTeeth and Prestige and Kineflex/C implants exhibited higher pullout loads than the PCM Low Profile and tricortical iliac crest (p < 0.05). Tricortical iliac crest exhibited the lowest pullout strength, which was different from all other treatments (p < 0.05). The surface area of endplate contact, measuring 300 mm(2) (PCM treatments), 275 mm(2) (Prestige LP), 250 mm(2) (Kineflex/C disc), 180 mm(2) (plate + cage), and 235 mm(2) (tricortical iliac crest), did not correlate with pullout strength (p > 0.05). The PCM, Prestige, and Kineflex constructs, which did not use screw fixation, all failed by direct pullout. Screw fixation devices, including anterior plates, led to test block fracture, and tricortical iliac crest failed by direct pullout. CONCLUSIONS These results demonstrate a continuum of fixation strength based on prosthetic endplate design. Disc arthroplasty constructs implanted using vertebral body screw fixation exhibited the highest pullout strength. Prosthetic endplates containing toothed ridges (>or= 1 mm) or keels placed second in fixation strength, whereas endplates containing serrated edges exhibited the lowest fixation strength. All treatments exhibited greater fixation strength than conventional tricortical iliac crest. The current study offers insights into the benefits of various prosthetic endplate designs, which may potentially improve acute fixation following cervical disc arthroplasty.