The Spine Journal 14 (2014) 128–136

Basic Science

Biomechanics of an integrated interbody device versus ACDF anteriorlocking plate in a single-level cervical spine fusion construct

Matthew I. Stein, MDa, Aniruddh N. Nayak, MSb, Roger B. Gaskins, III, MDb,Andres F. Cabezas, BSb, Brandon G. Santoni, PhDb,*, Antonio E. Castellvi, MDc

aDepartment of Orthopaedics and Sports Medicine, University of South Florida, 13220 USF Laurel Dr, Tampa, FL 33612, USAbPhillip Spiegel Orthopaedic Research Laboratory, Foundation for Orthopaedic Research and Education, 13020 N. Telecom Parkway, Tampa,

FL 33637, USAcCenter for Spinal Disorders, Florida Orthopaedic Institute, 13020 N. Telecom Parkway, Tampa, FL 33637, USA

Received 12 March 2012; revised 11 April 2013; accepted 24 June 2013

Abstract BACKGROUND CONTEXT: No profile, int

FDA device/drug

Author disclosures

RBG: Nothing to disc

(D, Paid directly to in

rectly to institution/em

ployer), NIH-NIBIIB

Grant: Centinel Spine

Central Spine SAB (A

(A); Scientific Advisor

matics SAB (A), Cro

Paid directly to institu

1529-9430/$ - see fro

http://dx.doi.org/10.10

egrated interbody cages are designed to act as im-plants for cervical spine fusion, which obviates the need for additional internal fixation, combiningthe functionality of an interbody device and the stabilizing benefits of an anterior cervical plate.Biomechanical data are needed to determine if integrated interbody constructs afford similar stabil-ity to anterior plating in single-level cervical spine fusion constructs.PURPOSE: The purpose of this study was to biomechanically quantify the acute stabilizing effectconferred by a single low-profile device design with three integrated screws (‘‘anchored cage’’), andcompare the range of motion reductions to those conferred by a standard four-hole rigid anteriorplate following instrumentation at the C5–C6 level. We hypothesized that the anchored cage wouldconfer comparable postoperative segmental rigidity to the cage and anterior plate construct.STUDY DESIGN: Biomechanical laboratory study of human cadaveric spines.METHODS: Seven human cadaveric cervical spines (C3–C7) were biomechanically evaluated us-ing a nondestructive, nonconstraining, pure-moment loading protocol with loads applied in flexion,extension, lateral bending (rightþleft), and axial rotation (leftþright) for the intact and instru-mented conditions. Range of motion (ROM) at the instrumented level was the primary biomechan-ical outcome. Spines were loaded quasi-statically up to 1.5 N-m in 0.5 N-m increments and ROM atthe C5–C6 index level was recorded. Each specimen was tested in the following conditions:1. Intact2. Discectomyþanchored cage (STA)3. Anchored cage (screws removed)þanterior locking plate (ALP)4. Anchored cage only, without screws or plates (CO)

RESULTS: ROM at the C5–C6 level was not statistically different in any motion plane betweenthe STA and ALP treatment conditions (pO.407). STA demonstrated significant reductions in flex-ion/extension, lateral bending, and axial rotation ROM when compared with the CO condition(p!.022).CONCLUSIONS: In this in vitro biomechanical study, the anchored cage with three integratedscrews afforded biomechanical stability comparable to that of the standard interbody cageþanteriorplate cervical spine fusion approach. Due to its low profile design, this anchored cage device may

status: Approved (STALIFC and CSLP).

:MIS:Nothing to disclose.ANN:Nothing to disclose.

lose. AFC: Nothing to disclose. BGS: Centinel Spine

stitution/employer); Grants: Medtronic (D, Paid di-

ployer), Globus (C, Paid directly to institution/em-

(D, Paid directly to institution/employer). AEC:

(C, Paid directly to institution/employer); Consulting:

), Alphatec Spine (A), Trans1 (A), Orthokinematics

y Board/Other Office: Alphatec Spine (A), Orthokine-

cker SAB (A); Fellowship Support: OREF grant (D,

tion/employer).

The disclosure key can be found on the Table of Contents and at www.

TheSpineJournalOnline.com.

Support: This study was supported in part by a research grant from Cen-

tinel Spine Inc. (West Chester, PA).

* Corresponding author. Biomechanics Operations, Phillip Spiegel Or-

thopaedic Research Laboratory, Foundation for Orthopaedic Research and

Education, Tampa, FL 33637, USA. Tel.: (813) 910-3662; fax: (813) 558-

6948.

E-mail address: bsantoni@foreonline.org (B.G. Santoni)

nt matter � 2014 Elsevier Inc. All rights reserved.

16/j.spinee.2013.06.088

129M.I. Stein et al. / The Spine Journal 14 (2014) 128–136

avoid morbidities associated with standard anterior plating, such as dysphagia. � 2014 ElsevierInc. All rights reserved.

Keywords: No-profile interbody device; Anterior locking plate; ACDF; Biomechanical evaluation; Cervical spine

Introduction

Anterior cervical discectomy and fusion (ACDF) hasbeen the standard surgical treatment for degenerative condi-tions of the cervical spine, including degenerative discdisease, cervical spondylotic myelopathy, and cervical discprolapse [1]. Since the description of the anterior approachfor cervical discectomy and fusion by Robinson and Smithin 1955, anterior cervical procedures have become quitecommon with generally good clinical results [2]. Althoughmotion-preserving disc arthroplasty procedures have beenrecently described, ACDF remains the standard surgicaltreatment, particularly for more elderly patients or thosewith a contraindication to disc prosthesis [3].

Studies have shown that the addition of an anterior platewith screws to an interbody cage/spacer provides enhancedstability and increased fusion rates [4,5]. Although success-ful in achieving fusion, anterior plating is not withoutcomplications. Dysphagia is the most common postopera-tive complication, and although its mechanism is poorlyunderstood, it has been linked to the anterior prominenceassociated with plate and screw constructs, the adhesionsthat form in response to the plate [2,6,7], and retractionof the pharynx/esophagus during the anterior approachand instrumentation of the cervical spine [8,9]. Althoughthe profile of current anterior plates is smaller than priordesigns, 2% to 60% of patients complain of dysphagiain the early postoperative period [7,10]. Although in manypatients these symptoms eventually disappear, not allbecome asymptomatic subsequent to ACDF, as the inci-dence of chronic dysphagia has been reported to range from3% to 21% [6,11–13]. In addition to prolonged surgicaltime, other potential risks associated with anterior platinginclude screw pullout, subsequent cage migration, and a po-tentially higher rate of adjacent-level degeneration [14–18],manifested as adjacent-level ossification due to placementof the anterior plate near the adjacent level disc.

The STALIFC (Centinel Spine, West Chester, PA, USA),from this point on referred to as the ‘‘anchored cage,’’ isa radiolucent cervical integrated interbody fusion deviceconstructed of polyether-ether-ketone (PEEK) with threeintegrated cancellous screws designed to provide lag com-pression between the adjacent vertebral bodies and conferanterior column fixation, bridging the index levels. Thedesign avoids the need for any additional internal fixationdevices and theoretically circumvents the aforementionedmorbidities associated with anterior plating while providingthe segmental rigidity necessary for cervical spinal fusion.

The purpose of this study was to biomechanically eval-uate, in a single-level fusion construct (C5–C6), the stabi-lizing properties of the anchored cage and compare this

with the standard cageþanterior plate ACDF construct,which has been shown to provide excellent rigidity andfusion outcomes [3,4]. We hypothesized that the anchoredcage would provide similar stability to that of the cage-þanterior plate and screws. This biomechanical evaluationcould provide clinicians with an anterior cervical fusiondevice that affords adequate segmental rigidity, withoutthe anterior prominence implicated as a cause for multiplesurgical morbidities.

Materials and methods

Specimen preparation

Seven cervical spines (C3–C7) were dissected fromfresh-frozen, human cadaveric specimens (three male, fourfemale) (mean age: 56.6 years; range: 50–64 years).The medical history of each donor was reviewed to excludetrauma, malignancy, or significant metabolic disease.Anterior-posterior and lateral radiographs were takento confirm that the procured specimens were free of signif-icant deformity or prior instrumentation. Bone mineral den-sity (BMD) values were assessed at each C6 vertebral levelby using dual energy X-ray absorptiometry (DEXA) (LunarProdigy Advance; GE Healthcare, Madison, WI, USA) us-ing an approach previously described for assessing bonequality in cadaveric tissue specimens denuded of extrane-ous soft tissues published by Wahnert et al. [19] Specimenswith BMD values indicating obvious osteoporosis were ex-cluded and replaced. The average BMD value at the C6level of the seven cadaveric specimens was 0.980 g/cm2

(range 0.786–1.159 g/cm2).Specimens were cleaned of musculature and adipose

tissue and all ligamentous structures were retained. Thespecimens were then rigidly potted at the cephalad and cau-dal (C3 and C7) ends using interference screws and high-strength resin. All cadaveric specimens were kept hydratedthroughout dissection, instrumentation, and biomechanicalevaluation by wrapping with saline-soaked gauze or spray-ing with 0.9% saline at regular intervals. Before bio-mechanical testing, all specimens were thawed overnight(8–10 hours) at room temperature (approximately 25�C).

Implants

To control for interspecimen variability, each specimenwas tested in each of the following conditions: (1) intact(INT); (2) following discectomy, decompression, and inser-tion of the anchored cage (STA); (3) anchored cage withoutscrewsþanterior locking plate (ALP); and (4) anchored cageonly (CO) with screws and anterior plate removed. The test

130 M.I. Stein et al. / The Spine Journal 14 (2014) 128–136

device (Fig. 1, Left) was a no-profile PEEK spacer (STA-LIFC; Centinel Spine), which combines an interbody cagewith integrated anterior screw fixation. It functions as an in-terbody spacer when the integrated screws are removed andis manufactured from PEEK-Optima polymer (CentinelSpine), which displays a Young’s modulus similar to corticalbone [20]. Two titanium markers embedded along the poste-rior aspect of the interbody space facilitate placement in thedisc space under fluoroscopy. In the sagittal plane, the devicedisplays a tapered profile with a larger height anteriorly thanposteriorly. The 6� inclusive angle facilitates restoration oflordosis on cage placement. The anteroposterior length ofthe device is 14 mm and its medio-lateral dimension is16.5 mm. In the current study, implanted cage heights rangedfrom 5.5 to 7.5 mm and the screw lengths ranged from 14 to16 mm. The experimental design allowed for an appropri-ately sized single PEEK cage to be placed in the C5–C6intervertebral disc space.

The locking plate (Fig. 1, Right) used was a rigid ante-rior plating system (CSLP; Synthes, West Chester, PA,USA), which is 21 mm wide and 2 mm thick. Plate sizes

Fig. 1. Devices studied. (Left) PEEK Optima anchored cage with three points o

within anchoring screws to create a rigid screw-plate interface.

used to instrument the C5–C6 level ranged from 14 to 18mm, measured from the cephalad to caudal hole pair, andthe screw lengths used ranged from 14 to 16 mm.

Surgical procedures

Following initial intact flexibility testing (INT), speci-mens were prepared for single-level cage instrumentation.A fellowship-trained and board-certified spine surgeon per-formed all procedures. A box-like incision was madethrough the anterior longitudinal ligament and annulus fol-lowed by discectomy at the C5–C6 level. Levels were con-firmed through the use of fluoroscopy. The disc materialwas removed using a small curette and rongeur. A completediscectomy was performed, including excision of the poste-rior longitudinal ligament (PLL) and part of the Luschka(unco-vertebral) joints. At least 50% of these joints werepreserved.

Following destabilization surgery, a correctly sized an-chored cage was implanted at the C5–C6 level for eachspecimen. All anchored cages were sized appropriate to

f lag screw fixation; (Right) locking titanium plate, which uses set screws

131M.I. Stein et al. / The Spine Journal 14 (2014) 128–136

individual specimen anatomy and surgeon’s experience.The cancellous screws were then sized and inserted. Appro-priate placement was confirmed through the use of fluoros-copy (Fig. 2, Top). Following insertion of the anchoredcage, each specimen underwent flexibility testing. Subse-quently, the anchoring screws were removed and a four-hole, rigid, anterior locking plate was applied overlyingthe interbody cage at the C5–C6 level (Fig. 2, Middle).The specimen was then tested in the ALP condition. Fol-lowing the completion of testing, the anterior plate wasremoved and each specimen underwent testing in the COcondition (Fig. 2, Bottom).

Sample size estimates were derived from range ofmotion (ROM) data published previously by Scholz et al.[21] in a cervical spine biomechanical study of similarscope to ours. Using the published mean and variance esti-mates and expecting a similar difference in our study, wefound that an n55 was sufficient to detect a differenceof 3.3� in flexion ROM with a power of 80% between theanchored cage device and the negative control. An n of7 powered our experiment at the 0.988 level and this sam-ple size is more in line with literature recommendations forcomparative spine biomechanical studies [22].

Biomechanical testing

Biomechanical testing was accomplished using a non-constraining, nondestructive, pure-moment flexibility pro-tocol using a system of cables and pulleys in similarfashion to previously reported methods [23,24]. We chosenot to use a compressive follower load to mimic the weightof the head because such loading can confound ROM data,as it may introduce secondary moments depending on thespecimen’s size and its positioning [25]. The applicationof a follower load is also difficult to replicate in the differ-ent loading planes. The caudal base (C7) of the specimenwas mounted on a 6-component load cell (AMTI, ModelMC3A-1000; AMTI Transducers, Watertown, MA, USA)and the specimen was allowed to move freely at the ceph-alad (C3) end. Pure moments up to 1.5 N-m in incrementsof 0.5 N-m were applied in flexion, extension, lateral bend-ing (rightþleft), and axial rotation (rightþleft) for the intactand instrumented conditions. To overcome the spine’s vis-coelastic effects, for each loading scenario, three precondi-tioning cycles up to 1.5 N-m were applied to the specimen,and incrementally applied moments were maintained forapproximately 30 seconds before recording ROM. Speci-mens were kept hydrated by regularly spraying with 0.9%saline. Intervertebral ROM was obtained using an optoelec-tronic motion analysis system (Optotrak Certus; NorthernDigital Inc., Waterloo, Ontario, Canada) with infraredlight-emitting diode marker arrays rigidly coupled to eachvertebral level.

With the use of a digitizing probe, a local coordinatesystem for each cervical vertebra was defined using threeanatomical landmarks per body using the convention

recommended by White and Panjabi [26] with the (þ)xdirection to the left, (þ)y directed cephalad, and the (þ)zdirected anteriorly. The intervertebral ROM (degrees) wascalculated as the range of the Euler angle corresponding toflexion-extension (X), axial rotation (Y), and lateral bending(Z). The Euler angle decomposition sequence was definedby the Optotrak system as Rz(4)�Ry

0(q)�Rx00(j). Although

it has been documented that coupled motion exists duringmoment loading of the cervical spine, particularly axial ro-tation motion subsequent to lateral bending loading, thiscoupling characteristic is relatively weak in the lower cervi-cal region [26]. ROM values are reported as the dominantEuler angle during a given loading mode. As secondary out-comes, plots of applied moment versus intervertebral rota-tion were generated, and general observations regardingthe neutral zone, defined as the region of little to no resis-tance tomotion on either side of the neutral position for amo-tion segment, at the C5–C6 index level as a function ofinstrumentation, were made. Finally, the effects of simulatedfusion over the C5–C6 level on the distribution of angularmotion over the remaining adjacent cervical spine levelswere assessed by comparing intact spine ROM versus thesimulated fusion conditions. The ROM of each specimen(C3–C7) was normalized to 100%. For each cervical motionsegment, the ROMwas subsequently expressed as a percent-age of the total intact cervical spine motion. Comparisons inchanges of the distribution of angular motion at the adjacentlevels as a function of instrumentation condition were madein flexion-extension.

Statistical analysis

Range of motion in each loading mode was recorded anddata analyzed using one-way repeated measures analysis ofvariance (RM-ANOVA) with four levels of treatment (INT,STA, ALP, CO). Post hoc tests were performed where indi-cated by RM-ANOVA results using Bonferroni correctionfor multiple comparisons. Effects of simulated fusion atC5–C6 on adjacent level motion were compared using thesame approach. All statistical analyses were performedwith SPSS statistical software (IBM, Chicago, IL, USA).

Results

All specimens were grossly inspected and radiographedfollowing instrumentation to ensure correct positioning ofthe hardware and to rule out any fractures. Following kine-matic evaluation, the specimens were re-inspected and nosigns of screw loosening or fracture were identified inany of the constructs.

Relative to the intact spine, ROM at the C5–C6 levelwas significantly reduced (p!.010) by instrumentation withthe anchored cage (STA) and the ALP. The anchored cagereduced flexion, extension, lateral bending, and axial rota-tion ROM at the index level by 62%, 48%, 66%, and51%, respectively. ROM reductions for the ALP in the

Fig. 2. Anterior-posterior and lateral radiographs showing single-level instrumentation of (Top) anchored cage device with screws engaged; (Middle) an-

chored cage without screwsþanterior locking plate; and (Bottom) anchored cage without any screws or anterior locking plate.

132 M.I. Stein et al. / The Spine Journal 14 (2014) 128–136

Fig. 3. Representative load versus displacement plots for the intact spine

as well as anchored cage (STA) and anterior locking plate (ALP) fixation at

the C5–C6 level. The neutral zone is virtually eliminated due to rigid fix-

ation at the index level.

Fig. 4. Range-of-motion (ROM) histograms in all loading modes (Flex-

ion-Extension/Lateral Bending/Axial rotation) for intact, anchored cage

(STA), anterior locking plate (ALP), and anchored cage only (CO) con-

structs. There were no significant differences in ROM between the STA

or ALP fixation groups for any loading mode (pO.05, dashed line). Signif-

icant decrease in ROM relative to CO was noted in all loading modes for

STA fixation (**p!.05).

Table

Results of the distribution of measured ROM at all cervical spine levels

Test condition C3–C7 C3–C4 C4–C5 C5–C6 C6–C7

Intact, deg 46.365.1 12.261.6 12.761.9 12.262.4 9.163.2

Normalized, % 100 26.361.8 27.562.2 26.766.0 19.566.5

Anchored cage

fusion, deg

41.166.6 12.561.8 12.661.9 6.062.3 9.663.0

Normalized, % 100 30.964.5* 30.864.2a 15.064.6y 23.165.5*

Anterior plate

fusion, deg

43.668.1 13.462.3 13.962.7 4.662.9 10.463.9

Normalized, % 100 31.465.4* 32.666.4b 12.768.4y 23.365.5*

Cage only, deg 52.368.8 14.962.7 14.362.2 12.164.4 11.064.6

Normalized, % 100 28.965.8 27.864.2 22.766.3 20.666.4

Note: Denotes a statistically significant increase (*) or decrease (y) rel-ative to the corresponding normalized value for the intact spine.

ap5.057 and bp5.052 for comparison to the corresponding normalized

value for the intact spine.

133M.I. Stein et al. / The Spine Journal 14 (2014) 128–136

same motion planes were 64%, 60%, 67%, and 59%, re-spectively. Qualitatively, the force-displacement relation-ships were altered as a function of STA and ALPinstrumentation (Fig. 3). Following simulated fusion atthe C5–C6 level, the neutral zone decreased relative tothe intact motion segment.

ROM comparison between anchored cage and COconstruct

The ROM at the instrumented level was compared in allthree planes of motion. The anchored cage significantly re-duced ROM by 57%, 40%, 47%, and 32% in flexion, exten-sion, bending, and rotation, respectively, relative to the COconstruct. These decreases in ROM were statistically sig-nificant in all motion planes (flexion: p5.013; extension:p5.022; lateral bending: p5.004; and axial rotation:p5.019) (Fig. 4).

ROM comparison between anchored cage and ALPconstruct

The ROM at the instrumented level was compared in allthree planes of motion. There was no significant differencein ROM (p$.90) found following instrumentation betweenthe anchored cage (STA) and the ALP in flexion, extension,lateral bending, or axial rotation (Fig. 4).

Motion at the nonoperative levels

Simulated fusion with the anchored interbody cage andALP at the C5–C6 level resulted in significantly increasednormalized motion at C3–C4 and C6–C7 (Table). Therewas a trend of increased normalized motion at the immedi-ately adjacent superior C4–C5 level for the anchored cage

and anterior plate groups (p5.057 and p5.052, respec-tively) (Table). Reconstruction with the cage only at thesingle cervical level did not result in increases in normal-ized adjacent level motion.

Discussion

The goal of the current study was to characterize the bio-mechanical stability afforded by an anchored interbodyPEEK spacer with cancellous screw fixation in a singlelevel cadaveric cervical spine fusion construct and comparethese biomechanical characteristics to those afforded byrigid anterior plating at the same level. The radiolucent cer-vical spine implant, while affording differing design param-eters, is similar to an anterior stabilization implant

134 M.I. Stein et al. / The Spine Journal 14 (2014) 128–136

previously described clinically [27] and biomechanically[28] for the lumbar spine. Current biomechanical study re-sults presented here indicate that the no-profile device canconfer comparable rigidity at the instrumented cervicalspine level to traditional rigid anterior plating.

Anterior cervical discectomy and fusion has become thegold standard for the treatment of cervical disc disease[29,30]. The advent of locked anterior plates that providestable fixation without the need for bicortical screw purchasesignificantly reduced the potentially catastrophic morbid-ities (dural penetration) associated with first-generation an-terior plates [31]. Current-generation plating techniquesare still associated with a number of intraoperative and post-operative complications. Intraoperatively, preparation of theanterior surface of the vertebral body for plating may endan-ger the trachea, carotid arteries, and esophagus. Althoughthe anterior profile of the plates has decreased over time,the incidence of early postoperative dysphagia has been re-ported to range from 2% to 67% [7,10], with current litera-ture reports of chronic dysphagia ranging from 3% to 21%[6,11–13]. Additional plate-associated, postoperative mor-bidities include screw and plate loosening, reported to occurat a prevalence of up to 15.4%, as well as screw fracture,plate fracture, and implant malposition, which, respectively,have a reported incidence of up to 6.7%, 21.4%, and 12.5%[32]. These hardware-related complications, particularly thedebilitating condition of dysphagia, have prompted the de-velopment of no-profile arthrodesis devices that have re-cently been reported to confer similar acute biomechanicalstability to traditional ACDF techniques [21,33,34]. Advan-tages of these anchored cage devices include the need fora smaller surgical incision and less soft tissue retraction rel-ative to anterior plating. The anatomical profile and fewerimplantation steps may be advantageous by decreasing oper-ating room time and reducing the incidence of dysphagia. Insupport of the latter, a recent clinical report by Scholz et al.[35] indicated that a cervical anterior implant with integratedscrews (Zero-P; Synthes) promotes fusion with an infrequentincidence (1 patient, 2.9%) of chronic dysphagia in 34 pa-tients having undergone single- or multi-level ADCF witha minimum of 6 months of follow-up. Future clinical studieswill better help to more clearly define the role of the an-chored interbody cage devices at mitigating the complica-tions, such as dysphagia [7,10], associated with anteriorplating.

Biomechanical results obtained in the current investiga-tion indicated that STA and ALP instrumentation signifi-cantly reduced motion in all loading modes comparedwith the intact condition. Further, no significant differencesin ROM were identified between either instrumentation.Compared with the spacer-only condition, ALP instrumen-tation reduced flexion-extension, lateral bending, and axialrotation by 57%, 47%, and 44%, respectively, suggestingthat the spacer-only construct is the least stable construct.This finding agrees with prior studies of similar scope[36,37] using identical or similar anterior locking plates.

For the same loading modes, Freeman et al. [37] reported69%, 45%, and 27% ROM reductions beyond the spacer-only condition with the addition on a one-level cervicaltitanium locking plate. Crawford and coworkers [36] re-ported ROM reductions of 65% in flexion, 62% in exten-sion, 59% in bending, and 57% in axial rotation with theaddition of a rigid four-hole anterior locking plate to the in-terbody cage.

Recent investigations have analyzed the biomechanicalperformance of various anchored interbody cage designs,and a comparison of our current study results to these re-lated studies is worth noting [21,33,34]. Scholz et al. [21]reported that an anchored cage with four points of lockedscrew fixation provided biomechanical stability equivalentto rigid and dynamic anterior plating in C3–C7 cadavericcervical spine constructs. Relative to the spacer-alone con-dition, Sholz et al. [21] reported significant ROM reduc-tions of 49% in flexion, 35% in extension, 56% inbending, and 57% in rotation, with the anchored interbodyspacer relative to the spacer alone in eight cervical spinesinstrumented at the C5–C6 level. In a comparative humancadaveric study of an anchored cage with two points oflag screw fixation versus the same anchored cage designevaluated by Scholz and coworkers [21], Majid et al. [33]demonstrated that both fusion constructs significantly re-duced ROM in all loading modes relative to the intactC5–C6 level. Further, they found no significant differencesin motion between the two fusion constructs. Our findingsevaluating an anchored interbody spacer with three pointsof lag screw fixation indicate significant ROM reductionsof 57%, 40%, 47%, and 32% in flexion, extension, bending,and rotation, respectively, relative to the spacer-only con-struct and are in good agreement with these prior studies[21,33]. Comparative biomechanical findings suggest thatthree points of fixed-angle anterior fixation confer compa-rable acute stability to other commercially available an-chored spacer designs.

The current investigation, as well as those previously de-scribed [21,33], suggests that low-profile anchored spacersmay be a suitable alternative to traditional rigid or dynamicanterior plating for the treatment of degenerative conditionsof the cervical spine at a single level. Additional indicationsfor ACDF include traumatic injuries to the subaxial spine,such as unstable flexion-distraction injuries [38]. A recentstudy by Wojewnik et al. [34] aided in defining the applica-bility of anchored spacer fixation in the setting of traumaticflexion-distraction injuries of increasing severity. In theirstudy, cervical spines (C3–C7) were instrumented at theC5–C6 level with either a locked anchored spacer (Zero-P; Synthes) or a variable-angle anchored spacer (Zero-PVA; Synthes). The instrumented spines were then tested us-ing a pure-moment loading protocol (61.5 N-m) in flexion-extension, lateral bending, and axial rotation. Thereafter,progressive posterior destabilization was performed in threestages at C5–C6 to induce progressive stages of flexion dis-traction injury and load versus displacement data were

135M.I. Stein et al. / The Spine Journal 14 (2014) 128–136

collected after each stage of injury. Similar to the work byScholz et al. [21], significant ROM reductions in all motionplanes greater than 72% and 45% were documented for thelocked and variable-angle anchored cage designs, respec-tively. However, whereas the locked device design main-tained stability at the C5–C6 level after progressiveposterior destabilization, the variable angle spacer did notsufficiently stabilize the flexion-distraction injuries. Thus,the current body of literature-based, biomechanical evi-dence supports the use of various anchored cage designswith 2 to 4 points of locked or variable angle screw fixationto confer adequate rigidity to a degenerated cervical level.However, more work is necessary to ascribe the stabilizingeffects of these anchored cage designs in the setting of cer-vical spine trauma.

Potential limitations of the present study largely relate touse of isolated cadaveric spines. Although fresh-frozen spec-imens afford a good representation of the clinical setting, thecervical spines biomechanically evaluated are devoid ofmuscular tissue that would normally be present in vivo andshould, in actuality, confer added stability. Thus, the testingconditions imposed here could be considered ‘‘worst case,’’as the effects of stabilizing muscle forces cannot be in-cluded. A second potential limitation relates to the repeatedmeasures nature of our experimental design. In the currentstudy, each specimen was tested with three different methodsof fixation, which may introduce the possibility of implantloosening as a result of repeat instrumentation, resulting inan artificial decrease in the reported rigidity of the construct.This may have particular relevance for the rigid anteriorplate treatment arm following removal of the cancellouslag screws of the integrated interbody device. However, weconfirmed radiographically and with direct visualization thateach tested construct showed no signs of damage, screwloosening, or breakage throughout the course of repeat in-strumentation and biomechanical evaluation. Following an-chored cage testing, the three integrated screws were torquedslightly to confirm that all retained purchase throughout test-ing. Subsequently, the integrated screws were removed andthe anterior plate with four-point screw fixation was attachedanteriorly and osseous purchase of all screws was confirmedmanually before kinematic evaluation. These screws wereagain checked following testing to ensure that none had lostbony purchase and then removed, along with the anteriorplate, to evaluate the stabilizing effect of the interbodyspacer. At no point during testing was the interbody cage re-moved from the disc space. We observed no visible instanceof hardware loosening throughout the course of testing inany of the seven specimens. The ability to test all threemethods in the same cadaveric specimen is similar to priorstudies evaluating various anchored cage designs [21,33],and provided us with important internal controls that servedto eliminate interspecimen effects on ROM outcomes in-cluding the variable of bone quality. Finally, cadaveric tissuetesting permits only the reporting of the acute effectson biomechanical stability conferred by various implants.

Although biomechanical data should not be extrapolated totime frames beyond the immediate postoperative, we believethe findings reported here support continued clinical use ofthe integrated interbody implant in single-level cervicalspine fusions.

Conclusion

Study results indicate that there are no statistical differ-ences in biomechanical stability conferred to the instru-mented level between the anchored cage device withthree points of anterior fixation and rigid anterior plate fix-ation. Although this study seems to validate the biomechan-ical design of the anchored cage for use in one-level fusionconstructs, clinical studies are needed to discern its long-term success and performance in vivo. These studies willneed to address fusion rates, applicability, and success inmanaging adjacent-level disease, as well as intraoperativeand postoperative complications associated with the useof this device. Although emerging data suggest that postop-erative dysphagia can be mitigated with the no-profiledevices, additional clinical data are necessary to defini-tively confirm these preliminary findings and make thisclinical conclusion.

Acknowledgments

The authors thank Bing Wang, MD, for his assistancewith tissue preparation and biomechanical testing.

References

[1] Bohlman HH, Emery SE, Goodfellow DB, Jones PK. Robinson ante-

rior cervical discectomy and arthrodesis for cervical radiculopathy.

Long-term follow-up of one hundred and twenty-two patients. J Bone

Joint Surg Am 1993;75:1298–307.

[2] Daniels AH, Riew KD, Yoo JU, et al. Adverse events associated with

anterior cervical spine surgery. J Am Acad Orthop Surg 2008;16:

729–38.

[3] Korinth MC. Treatment of cervical degenerative disc disease—cur-

rent status and trends. Zentralbl Neurochir 2008;69:113–24.

[4] Kaiser MG, Haid RW Jr, Subach BR, et al. Anterior cervical plating

enhances arthrodesis after discectomy and fusion with cortical allo-

graft. Neurosurgery 2002;50:229–36; discussion 236–8.

[5] Xie JC, Hurlbert RJ. Discectomy versus discectomy with fusion ver-

sus discectomy with fusion and instrumentation: a prospective ran-

domized study. Neurosurgery 2007;61:107–16; discussion 116–7.

[6] Fountas KN, Kapsalaki EZ, Nikolakakos LG, et al. Anterior cervical

discectomy and fusion associated complications. Spine 2007;32:

2310–7.

[7] Lee MJ, Bazaz R, Furey CG, Yoo J. Influence of anterior cervical

plate design on dysphagia: a 2-year prospective longitudinal

follow-up study. J Spinal Disord Tech 2005;18:406–9.

[8] Heese O, Fritzsche E, Heiland M, et al. Intraoperative measurement

of pharynx/esophagus retraction during anterior cervical surgery. Part

II: perfusion. Eur Spine J 2006;15:1839–43.

[9] Heese O, Schroder F, Westphal M, Papavero L. Intraoperative mea-

surement of pharynx/esophagus retraction during anterior cervical

surgery. Part I: pressure. Eur Spine J 2006;15:1833–7.

136 M.I. Stein et al. / The Spine Journal 14 (2014) 128–136

[10] Bazaz R, Lee MJ, Yoo JU. Incidence of dysphagia after anterior cer-

vical spine surgery: a prospective study. Spine 2002;27:2453–8.

[11] Kasimatis GB, Panagiotopoulos E, Gliatis J, et al. Complications of

anterior surgery in cervical spine trauma: an overview. Clin Neurol

Neurosurg 2009;111:18–27.

[12] Riley LH 3rd, Skolasky RL, Albert TJ, et al. Dysphagia after anterior

cervical decompression and fusion: prevalence and risk factors from

a longitudinal cohort study. Spine 2005;30:2564–9.

[13] Yue WM, Brodner W, Highland TR. Persistent swallowing and voice

problems after anterior cervical discectomy and fusion with allograft

and plating: a 5- to 11-year follow-up study. Eur Spine J 2005;14:

677–82.

[14] Cagli S, Isik HS, Zileli M. Cervical screw missing secondary to

delayed esophageal fistula: case report. Turk Neurosurg 2009;19:

437–40.

[15] Park JB, Cho YS, Riew KD. Development of adjacent-level ossifica-

tion in patients with an anterior cervical plate. J Bone Joint Surg Am

2005;87:558–63.

[16] Sahjpaul RL. Esophageal perforation from anterior cervical screw

migration. Surg Neurol 2007;68:205–9; discussion 209–10.

[17] Yang JY, Song HS, Lee M, et al. Adjacent level ossification develop-

ment after anterior cervical fusion without plate fixation. Spine

2009;34:30–3.

[18] Shih P, Simon PE, Pelzer HJ, Liu JC. Osteophyte formation after

multilevel anterior cervical discectomy and fusion causing a delayed

presentation of functional dysphagia. Spine J 2010;10:e1–5.

[19] Wahnert D, Hoffneier KL, Lehmann G, et al. Temperature influence

on DXA measurements: bone mineral density acquisition in frozen

and thawed human femora. BMC Musculoskelet Disord 2009;10:1–7.

[20] Williams DF, McNamara A, Turner RM. Potential of polyetherether-

ketone (PEEK) and carbon-fibre-reinforced PEEK in medical appli-

cations. J Mater Sci Lett 1987;6:188–90.

[21] Scholz M, Reyes PM, Schleicher P, et al. A new stand-alone cervical

anterior interbody fusion device. Spine 2009;34:156–60.

[22] Wilke HJ, Wenger K, Claes L. Testing criteria for spinal implants:

recommendations for the standardization of in vitro stability testing

of spinal implants. Eur Spine J 1998;7:148–54.

[23] Puttlitz CM, Melcher RP, Kleinstueck FS, et al. Stability analysis of

craniovertebral junction fixation techniques. J Bone Joint Surg Am

2004;86-A:561–8.

[24] Nayak A, Gutierrez S, Billys JB, et al. Biomechanics of lateral plate

and pedicle screw constructs in spines instrumented at two levels with

laterally placed interbody cages. Spine J 2013 May 16. [Epub ahead

of print].

[25] Cripton PA, Bruehlmann SB, Orr TE, et al. In vitro axial preload ap-

plication during spine flexibility testing: towards reduced apparatus-

related artefacts. J Biomech 2000;33:1559–68.

[26] White AA, Panjabi MM. Clinical biomechanics of the spine. Phila-

delphia, PA: J.B. Lippincott Company, 1978.

[27] Schleicher P, Gerlach R, Schar B, et al. Biomechanical comparison of

two different concepts for stand alone anterior lumbar interbody fu-

sion. Eur Spine J 2008;17:1757–65.

[28] Cain CM, Schleicher P, Gerlach R, et al. A new stand-alone anterior

lumbar interbody fusion device: biomechanical comparison with es-

tablished fixation techniques. Spine 2005;30:2631–6.

[29] Anderson DG, Albert TJ. Bone grafting, implants, and plating op-

tions for anterior cervical fusions. Orthop Clin North Am 2002;33:

317–28.

[30] Fraser JF, Hartl R. Anterior approaches to fusion of the cervical

spine: a metaanalysis of fusion rates. J Neurosurg Spine 2007;6:

298–303.

[31] Lowery GL, McDonough RF. The significance of hardware failure in

anterior cervical plate fixation. Patients with 2- to 7-year follow-up.

Spine 1998;23:181–6; discussion 186–7.

[32] Coe JD, Vaccaro AR. Complications of anterior cervical plating. In:

Clark CR, Benzel EC, Currier BL, eds. The cervical spine. Philadel-

phia, PA: Lippincott Williams & Wilkins, 2005:1147–69.

[33] Majid K, Chinthakunta S, Muzumdar A, Khalil S. A comparative bio-

mechanical study of a novel integrated plate spacer for stabilization

of cervical spine: an in vitro human cadaveric model. Clin Biomech

(Bristol, Avon) 2012;27:532–6.

[34] Wojewnik B, Ghanayem AJ, Tsitsopoulos PP, et al. Biomechanical

evaluation of a low profile, anchored cervical interbody spacer device

in the setting of progressive flexion-distraction injury of the cervical

spine. Eur Spine J 2013;22:135–41.

[35] Scholz M, Schnake KJ, Pingel A, et al. A new zero-profile implant

for stand-alone anterior cervical interbody fusion. Clin Orthop Relat

Res 2011;469:666–73.

[36] Crawford NR, Scholz M, Reyes PM, et al. Biomechanics of one-level

anterior cervical discectomy and plating using two screws versus four

screws. Spine J 2011;11:234–40.

[37] Freeman AL, Derincek A, Beaubien BP, et al. In vitro comparison

of bioresorbable and titanium anterior cervical plates in the imme-

diate postoperative condition. J Spinal Disord Tech 2006;19:

577–83.

[38] Smith GW, Robinson RA. The treatment of certain cervical-spine dis-

orders by anterior removal of the intervertebral disc and interbody fu-

sion. J Bone Joint Surg Am 1958;40-A:607–24.