Available online at www.sciencedirect.com

Surgical Neurology 70 (2008) 390–397www.surgicalneurology-online.com

Spine

Threaded interbody fusion cage for adjacent segment degenerative diseaseafter previous anterior cervical fusion

Paul Arnold, MD⁎, Scott Boswell, BS, Joan McMahon, MSADepartment of Neurosurgery, University of Kansas Medical Center, Kansas City, KS 66106, USA

Received 8 May 2007; accepted 9 July 2007

Abstract Background: Anterior discectomy and fusion have been used for over 50 years in the treatment of

Abbreviations: ACadjacent segment disCopenhagen Neck DiNeck Disability Index

⁎ Corresponding a66160, USA. Tel.: +1

E-mail address: p

0090-3019/$ – see frodoi:10.1016/j.surneu.2

degenerative disease of the cervical spine. However, as these procedures become more common, thelong-term consequences are becoming more evident. One such consequence is degeneration of anadjacent segment, which can occur in up to 17% of patients undergoing cervical fusion. A threadedinterbody fusion cage has often been used in a primary degenerative disorder of the cervical spine.However, there have been no studies in which these cages have been used in adjacent segments afterprevious cervical fusion. This is a retrospective review of 7 patients to determine the fusion rate,operative utility, and clinical outcomes using a threaded fusion cage construct in the treatment ofcervical adjacent segment degeneration.Methods: A standard low-profile interbody fusion cage was implanted after standard discectomyand local vertebral body bone graft in 7 patients with documented radiographic adjacent segmentdegeneration and clinical disease after anterior cervical fusion. Each patient underwent clinical andradiographic evaluation, and all 7 patients demonstrated signs of radiculopathy and/or myelopathy aswell as radiographic signs of degeneration referable to a motion segment adjacent to previouscervical arthrodesis. These evaluations were repeated postoperatively. Patients were also asked to filla preoperative and postoperative VAS, NDI, Prolo Economic-Functional Rating System, and CNDSfor evaluation of outcome.Results: Each procedure was performed without complication. The mean VAS pain scale decreased58% as a result of the surgery. The CNDS improved in all patients by an average of 42%. The NDIimproved in all patients, with an average increase of 42%. The Prolo Economic-Function Statusshowed that 4 patients had an excellent outcome and 3 patients had a good outcome. There was noincidence of pseudoarthrosis in any procedure at the 24-month follow-up.Conclusion: These preliminary results support the use of threaded interbody cages in adjacentsegment degeneration of the cervical spine after previous anterior cervical fusion. Pain andfunctional scores improved in all cases. This technique should be among the possibilities forsurgical treatment of degeneration of adjacent segments in patients with previous cervicalspinal fusion.© 2008 Elsevier Inc. All rights reserved.

Keywords: BAK/C; Adjacent segment degeneration; Adjacent segment disease; Cervical fusion cage

DF, anterior cervical discectomy and fusion; ASD,ease; BAK/C, Bagby and Kulisch Cage; CNDS,sability Scale; CT, computerized tomography; NDI,; VAS, Visual Analog Scale.uthor. Department of Neurosurgery, Kansas City, KS913 588 7587; fax: +1 913 588 7570.arnold@kumc.edu (P. Arnold).

nt matter © 2008 Elsevier Inc. All rights reserved.007.07.034

1. Introduction

During the past several decades, ACDF has been shownto be an effective treatment for upper-extremity radicularpain, axial neck pain, and myelopathic symptoms related todegenerative disease of the cervical spine. Success rates ofthe procedure have been reported as high as 90% to 95%[5,14,18].

391P. Arnold et al. / Surgical Neurology 70 (2008) 390–397

There are many methods that surgeons can use toachieve fusion. One such method is the placement ofautogenous bone harvested from the iliac crest betweenpathologic vertebral segments. However, this method,which is associated with the highest reported fusion rates[1,7,8,21,27,28,33], is associated with the many risksinherent to a second surgical wound, including donor sitepain, infection, nerve injury, and hematoma [7,15,32]. Asecond method for achieving fusion is the use of donor boneallograft, which circumvents donor-site morbidity of auto-graft. However, this method is associated with an increasedextrusion rate of substrate and a decreased fusion rate[1,7,8,21]. These problems may be overcome using anteriorcervical fixation. However, hardware failure, a potentialcomplication that may require additional surgery for cor-rection, has been reported in the use of cervical fixation[8,13,23,25,34].

One other method of fusion that has been developed toachieve bony fusion between pathologic vertebral segmentsis the use of threaded fusion cages. One such system is theBAK/C (BAK/C Cervical Interbody Fusion System, ZimmerSpine, Minneapolis, Minn). Cervical cages involve a processknown as “local autograft” to provide for bony fusion.During the implantation process, bone is gathered from drillreamings and placed within the cage. This bone thusprovides the foundation for bony fusion. Rates for fusion,complication rates, and duration of hospital stay all approachthose for other fusion techniques [15,16] while obviating theneed for bone graft from the iliac crest and the need forsubsequent surgical procedures for the alteration or removalof plates. Other potential advantages to this system include ashorter surgery time, a low-profile system, immediatefixation, and lower incidence of subsequent proceduresresulting from hardware failure.

Anterior cervical discectomy and fusion are becomingan increasingly more common procedure [12,18]. Becausethese procedures have become so prevalent, long-termsecondary sequelae have become more of a concern. Onesuch complication is adjacent segment degeneration anddisease (ASD). These degenerative processes can both beseen radiographically and have clinical symptoms. Adja-cent segment disease is defined as the development of newradicular or myelopathic signs and symptoms clinically,

Table 1Patient pre-operative demographic data

Caseno.

Age Sex Previousfusion

Currentfusion

NDI

Preoperative Postoperative3 mo

Postopera12 mo

1 45 Female C4-C7 C3-C4 40 38 342 51 Male C5-C6 C6-C7 30 22 203 52 Female C4-C6 C3-C4 0 0 04 40 Male C4-C6 C6-C7 41 38 385 42 Male C6-C7 C6-C7 12 5 26 26 Female C4-C6 C6-C7 40 24 217 53 Female C4-C4 C3-C4 40 22 0

which are referable to a motion segment adjacent to priorcervical arthrodesis [3,18]. In previous reports, theprevalence of symptomatic ASD has ranged from 7% to17% [18,19,35].

The current study was designed to evaluate the rate ofarthrodesis and successful clinical outcome when anteriorcervical interbody fusion was performed using a threadedinterbody fusion cage in degenerative disease of levelsadjacent to a previous cervical fusion. The success ofanterior interbody cervical fusion using a threadedinterbody system in these patients has not beenpreviously reported.

2. Materials and methods

2.1. Bagby and Kulisch Cage

The BAK/C interbody fusion system is a hollow,threaded, cylindrical cage that is made of titanium alloy.Each implant has multiple holes spaced evenly throughoutouter surface, which allow bony growth through the holesand throughout the center of the cage. These holes areangled, and the cage has V-shaped threads that shave localbone into the graft chamber, creating the local bone graft.This local bone can also be supplemented with boneremoved during decompression.

2.2. Patient data

From 2003 to 2005, 7 patients presented with ASD afteranterior cervical fusion, defined as new radicular ormyelopathic signs and symptoms referable to a motionsegment adjacent to a prior cervical arthrodesis. Of these 7cases, the previous procedure consisted of 3 single-leveldiscectomies, 3 two-level discectomies, and 1 three-leveldiscectomy. Mean duration of time between previous fusionand cage placement for ASD was 42 months. Patientpreoperative demographic data are included in Table 1.Approval from the institutional review board was obtainedfor this study.

2.3. Preoperative evaluation/Operative indication

Preoperative evaluation consisted of a complete historyand physical examination. Neurologic evaluation included

Copenhagen Prolopostoperative

Hospitaldays

tive Preoperative Postoperative3 mo

Postoperative12 mo

30 27 26 8 217 6 12 7 20 0 0 10 129 28 28 7 111 7 2 9 128 15 9 9 129 15 2 10 1

392 P. Arnold et al. / Surgical Neurology 70 (2008) 390–397

examination of strength, sensation, reflexes, and other signsof upper or lower motor neuron compression.

The management of these patients with ASD followedthe same principles as that of patients with primarycervical spondylosis. For those patients who developedacute or subacute neurological changes (eg, motordeficits, myelopathic signs and symptoms, or clinicalinstability), early surgical intervention was recommended.The rest of the patients received an aggressive course ofconservative management, including nonsteroidal drugs,analgesics, physical therapy, and epidural injections. Onlyupon failure of conservative treatment was surgicaltreatment recommended.

Preoperative evaluation also included plain radiographs,CT myelography, and/or magnetic resonance imaging. Thesestudies were used in conjunction with clinical presentation todetermine the appropriate levels for decompression andarthrodesis. An example of a CT myelogram showingdegeneration in segments adjacent to a previous fusion isprovided (Fig. 1).

Preoperative evaluation also included a completeotolaryngology assessment, including assessment of vocalfold function. This was necessary because surgerywas performed on the side contralateral to the originalsurgery. There were no cases of abnormalities seen onotolaryngology assessment.

Lastly, preoperative evaluation consisted of subjectiveassessment of pain and function. Pain was assessed using theVAS. This tool evaluates pain on a scale between 0 and 10,where 10 represents severe pain. The NDI and CNDS wereused to evaluate how pain affects the daily functions ofpatients. The NDI is a scale between 0 and 50, where50 represents complete dysfunction. The Copenhagen scaleis between 0 and 30, where 30 represents completedysfunction. The Prolo system was used to determinefunctional outcome of surgery. This system also uses a

Fig. 1. A 40-year-old man with a history of C4-C6 fusion presented with neck painand C: Axial CT myelogram.

scale from 0 to 10, where 9-10 represent excellent outcomes;7-8, good; 5-6, fair; and 0-4, poor.

2.4. Operative technique

General endotracheal anesthesia was used in allpatients. The spine was approached through a curvilineartransverse skin incision in the anterior cervical trianglecontralateral to the previous surgical incision. Thecontralateral side was chosen because scar tissue presentfrom the previous procedure would have made tissuedissection and manipulation more difficult. Anteriorcervical microdiscectomy and/or osteophytectomy wereperformed in every patient. The disc space was preparedwith a reaming drill specifically sized for the requiredimplant. Bone fragments from the reaming were saved forplacement into the interbody cage. Subsequent tappingand placement of either a single central cage or 2bilateral cages, at the discretion of the surgeon, wereaccomplished using fluoroscopic control and a drill-tubeguide system. After surgery, all patients were instructedto use either hard or soft cervical collars for the first6 weeks postoperatively.

2.5. Postoperative evaluation

After the surgery, patients were evaluated at 3-, 6-, 12-,and 24-month postoperatively. Evaluation included perti-nent history of new complaints, use of analgesic medica-tions, and a complete neurologic examination. Neurologicevaluation included strength, sensation, reflexes, and othersigns of upper or lower motor neuron compression.Flexion and extension lateral radiographs of the cervicalspine were taken at each visit and were used to confirm thepresence of absence of fusion. This was done byassessment of movement, presence or absence of halo orlucency around the device, and measurement of changes in

and radicular pain referable to a C6-C7 lesion. A: Sagittal CT myelogram. B

Fig. 2. Postoperative x-rays of a 40-year-old man (same patient as in Fig. 1)with degeneration in C6-C7 segments adjacent to a previous C4-6 fusion.Note the hollow, threaded interbody construct in C6-C7 disk space. A:Lateral. B: Anteroposterior, AP.

393P. Arnold et al. / Surgical Neurology 70 (2008) 390–397

interspinous distance. A postoperative radiograph showingthe construct is provided (Fig. 2).

Patients also were asked to subjectively reevaluate theirpain and functional outcome, again using the NDI, ProloEconomic-Functional Rating System, and CNDS. Thepatient's subjective perception of overall satisfaction with theoutcome of the procedure was graded as excellent, good, fair,

Fig. 3. Mean VAS of neck and arms showing subjective as

or poor. These subjective evaluations of pain and functionaloutcome occurred at both 3 and 12 months postoperatively.

3. Results

3.1. Clinical findings

There were 7 patients enrolled in this study, and all werepresent at 3-, 6-, 12-, and 24-month follow-up. Of thepatients in this study, 7 (100%) of 7 received single-levelarthrodesis at the location of the adjacent segment degenera-tion and disease. Of these patients, 4 received adjacentsegment arthrodesis at the C6-C7 level, and 3 patientsreceived arthrodesis at C3-C4.

Of the 7 patients, 6 presented because of extreme pain; 2presented with concurrent neurologic defects; 3 presentedwith neck pain, but no objective neurologic defects onexamination. This pain was characterized by a mean VASscore of 7.3. This mean decreased to 4.9 and 3.1 at 3 and12months, postoperatively. This correlates to a 33% and 58%decrease in neck pain at 3 and 12 months postoperatively,respectively (Fig. 3). Of the 7 patients, 6 presented withradicular arm pain characterized by a mean VAS score of 4.9,which decreased to a mean of 1.2 and 0.6 at 3 and 12 monthspostoperatively, respectively. Therefore, the surgery resultedin a 73% and 88% decrease in radicular arm pain at 3 and12 months postoperatively, respectively (Fig. 3). Individualresults of the VAS score are presented in Table 2.

The seventh patient presented with no pain but with 4+/4reflexes and new weakness of the left triceps and deltoid

sessment of pain preoperatively and postoperatively.

Table 2Individual results of the VAS Score

Case no. VAS neck VAS arms

Preoperative Postoperative 3 mo Postoperative 12 mo Preoperative Postoperative 3 mo Postoperative 12 mo

1 9.6 8.1 7 9.6 0 02 7.7 4.1 1.7 5.2 4.6 1.33 2.2 1.2 1.2 0.1 0.1 04 7.2 7.5 7 4.1 3 3.15 6.3 1.8 1 5.2 0.1 06 8.6 6.1 2.9 0.5 0 07 9.4 5.5 1.2 9 0.4 0

394 P. Arnold et al. / Surgical Neurology 70 (2008) 390–397

muscles on examination. All 7 patients presented withradiographic changes consistent with adjacent segmentdegeneration at segments correlating with their myelopathicor radicular findings.

Each incidence of clinically apparent neurologic findingswas resolved postoperatively at the 3-month examination.These findings remained consistent at the 6-, 12-, and24-month examination. In other words, the neurologic find-ings of loss of strength, hyperreflexia, changes in sensation,and other signs of upper or lower neuron compression thatwere preoperatively present in 3 patients were resolved attheir postoperative examinations.

One patient presented preoperatively with dysphagiasecondary to screw backout occurring in the plate constructused in the original fusion procedure. This issue wasaddressed during the second procedure as the screws were

ig. 4. Subjective functional measurement quantifying effect of pain on patient activities. Higher scores translate into more disability and inhibition of activitiesf daily living by pain.

Fo

removed and replaced with rescue screws. The patientreported no dysphagia upon follow-up at 3 and 12 months.There were no other cases of dysphagia reported.

Patient functional outcome also improved with the use ofthis construct. Preoperative NDI was a mean score of 29. Themean NDI scores at 3 and 12 months postoperatively were21.2 and 16.7, respectively. According to the CNDS, patientspresented with a mean of 20.1. The mean CNDS dropped to14.0 and 11.7 at 3 and 12 months postoperatively,respectively. These numbers correspond to a 30% and 42%decrease in NDI at 3 and 12 months postoperatively,respectively (Fig. 4). According to the CNDS, there was a28% and 42% decrease at 3 and 12 months postoperatively,respectively (Fig. 4).

According to the Prolo Economic-Functional RatingSystem, there were 4 excellent and 3 good outcomes. The

395P. Arnold et al. / Surgical Neurology 70 (2008) 390–397

mean Prolo score was 8.6 subjectively, each patient was“completely satisfied” with the result of the procedure andwould recommend the procedure to others.

3.2. Radiographic findings

Solid arthrodesis, as confirmed by flexion-extensionlateral radiographs, was achieved in 7 (100%) of 7 patients.There was no evidence of lucency or halo around the device,and there was no change in interspinous distance betweenflexion and extension views.

3.3. Hospital stay

The average duration of hospital stay postoperatively was1.3 days. Only 2 patients stayed longer than 1 day (Table 1).

3.4. Surgery-related complications

Postoperatively, there was no evidence of pseudoarthrod-esis or hardware failure. There were no cases of vertebralfracture or stenosis secondary to excessive bone formationaround the implant. There were no wound infections or casesof postoperative dysphagia.

4. Discussion

Anterior cervical discectomy and fusion have long beenused in the treatment of many cervical neck pathologies,especially in degenerative disorders. Ever since Robinsonand Smith [30] first described the procedure, surgeons havesteadily improved the procedure and their results, nowachieving fusion rates higher than 90% and improved pain orfunctional outcome in greater than 80% of patients[5,10,11,14-16].

Because of the high success rate of these proceduresin relieving patients' symptoms, the long-term conse-quences of fusion procedures were thought to be ofsecondary importance. However, as more of theseprocedures are performed and as patients live longerafter the procedure, there are concerns regarding the long-term consequences of these procedures. One suchconsequence is degeneration and disease of the segmentsadjacent to the primary fusion.

Adjacent segment degeneration is defined as radiographicchanges seen at levels adjacent to a previous spinal fusion[1]. This can manifest as osteophytes, disc degeneration,facet hypertrophy, spinal canal stenosis, and segmentalhypertrophy. Adjacent segment degeneration may occur in asmany as 25% of patients undergoing cervical fusion [2,14].However, radiographic evidence of degenerative changes inthe spine may not necessarily mean that clinical symptomsare present [3,4]. These changes can occur in many patientswho are asymptomatic. Therefore, ASD is defined asradicular or myelopathic signs and symptoms referable to amotion segment adjacent to a prior cervical arthrodesis[3,17]. All 7 patients who participated in our study hadclinical signs or symptoms that correlated to radiographic

changes indicating degeneration adjacent to a previouscervical arthrodesis.

Primary fusion of cervical segments has been theorizedto cause a more rapid degeneration of levels above orbelow the level or levels of the arthrodesis [3,9,18,20].This degenerative acceleration is thought to occur secon-dary to the mechanical and shearing forces exerted onadjacent segments from the lever movements of the fusedsegments. It should be noted that there is still debatein the literature whether there is truly acceleration or ifthe adjacent segment degeneration represents progressionof the natural history of the underlying degenerativedisease [18].

Primary spondylosis of the cervical spine has beentreated with discectomy and fusion using a threadedinterbody fusion cage for a long time, and with goodresults. Hacker et al [15] showed that the fusion rate ofinterbody fusion systems not only approached that ofstandard, noninstrumented ACDF but also surpassed it(98.5% and 80.5%, respectively). It has also equaled orsurpassed the fusion rate of ACDF with rigid plate fixationusing donor fibula [24] and tricortical iliac crest allograft[31]. It is thought to accomplish this by supplying structuralsupport for the organic, fusion-producing bone material thataccumulates in and around the device as it is inserted. Thus,cancellous bone serves as the fusion substrate rather thancortical bone that other procedures may offer. This isadvantageous because cancellous bone has been shown toincorporate into the fusion more quickly than corticalallograft [27,28,33].

Another benefit of local autograft use is that it avoidsthe morbidity of iliac crest autograft. These complicationscan include bleeding, fracture, meralgia paresthetica, andpain at the iliac donor site and can occur in up to 25% ofpatients [15,32].

Other potential benefits of a cage construct in interbodyfusion are a shorter duration of surgery and shorter hospitalstays. In a study by Cauthen et al [8], mean duration ofsurgery for the cage construct was 2.2 hours, whereas that fora dowel and plate construct was 3.5 hours. In the study byHacker et al [15], there was no statistical difference betweenthe mean duration of surgery between the interbody cagegroups and the control (ACDF with no plating techniques)group. A smaller incision and less tissue dissection requiredwhen using threaded cage constructs are among the reasonsfor the shorter duration of surgery. It is of note that a smallerincision and less tissue dissection may be of greater value inreoperation for adjacent segment degeneration, as thepresence of scar tissue may make tissue dissection moredifficult than primary cervical fusion. In addition, if acervical plate was used in the primary cervical fusion, it maybe necessary to remove that plate and add a longer plate forthe treatment of the adjacent segment degeneration. Thiswould significantly increase the duration of surgery. It oftenis not necessary to remove the plate if a threaded cageconstruct was used.

396 P. Arnold et al. / Surgical Neurology 70 (2008) 390–397

Regarding hospital stay, cage constructs have beenstatistically shown to have a shorter duration of stay(3.56 days) compared with allograft (5.04 days) andautograft (6.47 days) [22]. Our mean duration of hospitalstay of 1.3 days compares favorably to the average hospitalstay of 1.6 to 2.2 days in primary cervical interbody fusionusing plating techniques [8] and is similar to the 1.5 daysreported with the use of threaded cage constructs in primaryfusion [15]. There is growing support that cage constructsmay be cost-effective [6,8,22] because of decreased durationof hospital stay, lesser morbidity, and earlier return to work.

The rate of hardware failure, complications, and hard-ware migration also seems to be low in the interbody cageconstructs. In the study by Hacker et al [15], no graftcollapse or implant failures were reported in the cagepopulation, whereas there were 6 cases of failure in the bonegraft control group. In this same study of 346 patients, therewere 8 cases of implant-related complications, including6 cases of pseudoarthrosis, 1 case of vertebral fracture, and1 case of excessive osteogenesis around the implant leadingto cervical stenosis. In a study by Matge [26], the incidenceof symptomatic hardware migration in threaded interbodycages resulting in reoperation is 1 (1%) of 149. Thesenumbers have been supported in subsequent studies [8].These numbers are in comparison to hardware failure inplating techniques up to 35% in plating techniques [23]. At2 years of follow-up, there were no cases of hardwarefailure, hardware migration, or need for reoperation inour study.

The final advantage of local autograft systems is that theyhave lower rates of dysphagia, a problem more common inplating techniques. This is presumed to occur because thecervical cage is recessed below the vertebral surface and isessentially a “no-profile” device, whereas the platingsystems maintain an interface contact with the esophagus.In addition, a decrease in tissue dissection and totalprocedure time would result in a decrease in esophagealretraction, another potential cause of postoperative dyspha-gia. In the study by Hacker et al [15], dysphagia was presentin 0.6% of patients receiving the threaded interbody cage andin 2.1% of patients receiving a plating system. Dysphagiahas been reported to occur in up to 30% of patients usingplating systems [29].

In summary, the local autograft systems like those usedin our study have shown to be beneficial to standardACDF with iliac bone autograft or allograft because theyhave higher fusion success rates, lower complication rates,and concomitant improvement in clinical outcomes whileobviating the need for iliac crest autograft and itsassociated morbidity.

Despite the high incidence of ASD and the apparent valueof local autograft systems in primary disease, there have beenno previously published studies evaluating the application ofthese systems in ASD. The current study showed that acervical fixation cage achieves a high fusion rate andpredominantly good or excellent outcome ratings.

It is of note that no author involved in this study hasany financial interest in the devices or methodology usedin this study.

5. Conclusion

Although cervical spine surgery is an evolvingtechnique, the preliminary results presented here suggestthat interbody fusion cages are as good as, and in someaspects better than, standard allograft fusions in thetreatment of ASD. Their application should be consideredby surgeons as a possible treatment in this increasinglycommon disease.

References

[1] An HS, Simpson JM, Glover JM, et al. Comparison betweenallograft plus demineralized bone matrix versus autograft in anteriorcervical fusion. A prospective multicenter study. Spine 1995;20:2211-6.

[2] Baba H, Furusawa N, Imura S, et al. Late radiographic findings afteranterior cervical fusion for spondylotic myeloradiculopathy. Spine1993;18:2167-73.

[3] Bartolomei JC, Theodore N, Sonntag VK. Adjacent level degenerationafter anterior cervical fusion: a clinical review. Neurosurg Clin N Am2005;16:575-87.

[4] Boden SD, McCowin PR, Davis DO, et al. Abnormal magnetic-resonance scans of the cervical spine in asymptomatic subjects.A prospective investigation. J Bone Joint Surg Am 1990;72:1178-84.

[5] Bohlman HH, Emery SE, Goodfellow DB, et al. Robinson anteriorcervical discectomy and arthrodesis for cervical radiculopathy. Long-term follow-up of one hundred and twenty-two patients. J Bone JointSurg 1993;75:1298-307.

[6] CastroJr FP, Holt RT, Majd M, et al. A cost analysis of two anteriorcervical fusion procedures. J Spinal Disord 2000;13:511-4.

[7] Cauthen JC, Kinard RE, Vogler JB, et al. Outcome analysis ofnoninstrumented anterior cervical discectomy and interbody fusion in348 patients. Spine 1998;23:188-92.

[8] Cauthen JC, Theis RP, Allen AT. Anterior cervical fusion: acomparison of cage, dowel and dowel-plate constructs. Spine J2003;3:106-17.

[9] Cherubino P, Benazzo F, Borromeo U, et al. Degenerative arthritis ofthe adjacent spinal joints following anterior cervical spinal fusion:clinicoradiologic and statistical correlations. Ital J Orthop Traumatol1990;16:533-43.

[10] Chiles III BW, Leonard MA, Choudhri HF, et al. Cervical spondyloticmyelopathy: patterns of neurological deficit and recovery after anteriorcervical decompression. Neurosurgery 1999;44:762-9.

[11] Clements DH, O'Leary PF. Anterior cervical discectomy and fusion.Spine 1990;15:1023-5.

[12] Davis H. Increasing rates of cervical and lumbar spine surgery in theUnited States, 1979-1990. Spine 1994;19:1117-23.

[13] Geisler FH, Caspar W, Pitzen T, et al. Reoperation in patients afteranterior cervical plate stabilization in degenerative disease. Spine1998;23:911-20.

[14] Gore DR, Sepic SB. Anterior cervical fusion for degenerated orprotruded discs. A review of one hundred forty-six patients. Spine1984;9:667-71.

[15] Hacker RJ, Cauthen JC, Gilbert TJ, et al. A prospective randomizedmulticenter clinical evaluation of an anterior cervical fusion cage.Spine 2000;25:2646-54.

397P. Arnold et al. / Surgical Neurology 70 (2008) 390–397

[16] Hacker RJ. A randomized prospective study of an anterior cervicalinterbody fusion device with a minimum of 2 years of follow-upresults. J Neurosurg 2000;93(2 Suppl):222-6.

[17] Hilibrand AS, Carlson GD, Palumbo MA, et al. Radiculopathy andmyelopathy at segments adjacent to the site of a previous anteriorcervical arthrodesis. J Bone Joint Surg Am 1999;81:519-28.

[18] Hilibrand AS, Robbins M. Adjacent segment degeneration andadjacent segment disease: the consequences of spinal fusion? Spine J2004;4(6 Suppl):190S-4S.

[19] Ishihara H, Kanamori M, Kawaguchi Y, et al. Adjacent segmentdisease after anterior cervical interbody fusion. Spine J 2004;4:624-8.

[20] Kulkarni V, Rajshekhar V, Raghuram L. Accelerated spondyloticchanges adjacent to the fused segment following central cervicalcorpectomy: magnetic resonance imaging study evidence. J NeurosurgSpine 2004;100(1 Suppl Spine):2-6.

[21] Lofgren H, Johannsson V, Olsson T, et al. Rigid fusion after clowardoperation for cervical disc disease using autograft, allograft, orxenograft: a randomized study with radiostereometric and clinicalfollow-up assessment. Spine 2000;25:1908-16.

[22] Lopez-Oliva Munoz F, Garcia de las Heras B, Concejero Lopez V.Comparison of three techniques of anterior fusion in single-levelcervical disc herniation. Eur Spine J 1998;7:512-6.

[23] Lowery GL, McDonough RF. The significance of hardware failure inanterior cervical plate fixation. Patients with 2- to 7-year follow-up.Spine 1998;23:181-6.

[24] Martin Jr GJ, Haid Jr RW, MacMillan M. Anterior cervical discectomywith freeze-dried fibula allograft. Overview of 317 cases and literaturereview. Spine 1999;24:852-8.

[25] McLaughlin MR, Purighalla V, Pizzi FJ. Cost advantages of two-levelanterior cervical fusion with rigid internal fixation for radiculopathyand degenerative disease. Surg Neurol 1997;48:560-5.

[26] Matge G. Cervical cage fusion with 5 different implants: 250 cases.Acta Neurochir 2002;144:539-49.

[27] Muschler GF, Lane JM, Dawson EG. The biology of spinal fusion. In:Cotler MJ, Cotler HB, editors. Spinal fusion: science and technique.New York: Springer-Verlag; 1990.

[28] Muschler GF, Lane JM. Orthopedic surgery. In: Habal MB, Reddi AH,editors. Bone grafts and bone substitutes. Philadelphia: WB Saunders;1992. p. 375-407.

[29] Riley III LH, Skolasky RL, Albert TJ, et al. Dysphagia after anteriorcervical decompression and fusion: prevalence and risk factors from alongitudinal cohort study. Spine 2005;30:2564-9.

[30] Robinson RA, Smith GW. Anterolateral cervical disc removal andinterbody fusion for cervical disc syndrome. Bulletin Johns HopkinsHospital 1955;96:223-4.

[31] Samartzis D, Shen FH, Matthews DK, et al. Comparison of allograft toautograft in multilevel anterior cervical discectomy and fusion withrigid plate fixation. Spine J 2003;3:451-9.

[32] Sawin PD, Traynelis VC, Menezes AH. A comparative analysis offusion rates and donor-site morbidity for autogeneic rib and iliaccrest bone grafts in posterior cervical fusions. J Neurosurg1998;88:255-65.

[33] Suchomel P, Barsa P, Buchvald P, et al. Autologous versus allogenicbone grafts in instrumented anterior cervical discectomy and fusion: aprospective study with respect to bone union pattern. Eur Spine J2004;13:510-5.

[34] Tribus CB, Corteen DP, Zdeblick TA. The efficacy of anterior cervicalplating in the management of symptomatic pseudoarthrosis of thecervical spine. Spine 1999;24:860-4.

[35] Williams JL, Allen Jr MB, Harkess JW. Late results of cervicaldiscectomy and interbody fusion: some factors influencing the results.J Bone Joint Surg Am 1968;50:277-86.