161SProceedings of the NASS 22nd Annual Meeting / The Spine Journal 7 (2007) 1S–163S

OUTCOME MEASURES: Reoperation rate was defined as any patient

requiring reoperation, revision, or supplemental fixation at the index level.

Heterotopic ossification (HO) was classified at the adjacent levels at 24

months using the criteria described by McAfee.

METHODS: Patients were assessed pre-operatively and post-operatively,

6 weeks, 3, 6, 12, 18 and 24 months.

RESULTS: Adjacent level SCDD was reported in one (1) patient in short-

term (!42–210 days) and in two (2) patients in long-term (!210 days)

follow-up in the ACDF group out to 24 months for an incidence rate of

2.8%. No patients in the ProDisc-C group were reported to have adjacent level

SCDD out to 24 months. Class I ossification was identified at the adjacent

level in ten (10) ACDF patients representing 10.9% (10/92). Class II ossifica-

tion was identified at the adjacent level in two (2) ACDF patients representing

2.2% (2/92) of the ACDF patients. Class III ossification was identified at the

adjacent level in one (1) ACDF patients representing 1.1% (1/92) of the ACDF

patients. Class IV ossification was identified at the adjacent level in one (1)

ACDF patients representing 1.1% (1/92) of the ACDF patients. Therefore,

ACDF patients had an overall adjacent level HO rate of 15.2%. Class II ossi-

fication was identified in one (1) ProDisc-C patient at 24 months representing

0.97% (1/103). No Class I, III, or IV level ossifications were found in the Pro-

Disc-C group at adjacent levels therefore the overall HO rate was 0.97% for

the ProDisc-C group. The reoperation rate at 24 months were statistically dif-

ferent between treatments (p50.04). In the ACDF group, the reoperation rate

was 9.4% (10/106) compared to the ProDisc-C group was 1.9%(2/103).

CONCLUSIONS: This study supports previously published rates of

SCDD after ACDF of 2.8% but cervical TDR does not exhibit the same

degeneration pathology. In addition, the ACDF group had significantly dif-

ferent adjacent level HO rates at 2 year follow-up in comparison to the Pro-

Disc-C group. Reoperation rates at the index level were statistically

different between treatments. ACDF has been the historical standard be-

cause it is viewed as very successful, however the high reoperation rate,

adjacent level symptomatic degeneration, and high prevalence of HO

clearly have a negative clinical and economic effect on the system. All

three issues are not present following TDR at 2 years post-operatively.

FDA DEVICE/DRUG STATUS: ProDisc-C: Investigational/ Not approved.

doi: 10.1016/j.spinee.2007.07.380

P171. Computer Assisted vs Manual Coding Of Surgical Procedures

in a Private Practice Setting

Paul Schwaegler, MD1; 1Orthopedics International, Seattle, WA, USA

BACKGROUND CONTEXT: Today coding for spinal surgical interven-

tion is dauntingly complex. That coding and the resulting reimbursement

are intimately tied together. As providers we perform work, determine

the most accurate coding to represent that work, and then send that coding

scenario to the third party carrier. Stereotypically, the insurers receive the

bill, and then run that coding scenario through a number of filters to deter-

mine the reimbursement they think is fair to pay. Those ‘‘filters’’ include

not only the contracted rate of reimbursement, but also a payer set of rules

which determines how and if work will be reimbursed.

PURPOSE: Advanced computer aided coding can be extremely helpful by

bringing these insurer ‘‘filters’’ to the surgeon as he is coding the claim. By

addressing the rule sets prospectively while coding, the likelihood that the

surgery will be appropriately reimbursed is greatly enhanced. Medicare is

the largest single insurer, and its rule set is over 200,000 long. We set out to

determine whether computer aided coding could be of value when coding

spine surgery claims with this large carrier.

STUDY DESIGN/SETTING: Two groups of Medicare surgical proce-

dure claims were retrospective analyzed. Group A consisted of consecutive

claims manually coded in the first half of 2005. Group B consisted of con-

secutive claims filed during a similar time frame in 2006 which were coded

with the assistance of a coding software program. The two groups were

evaluated for coding accuracy and resulting reimbursement.

PATIENT SAMPLE: Group A consisted of 41 claims manually coded in

the first half of 2005. Group B consisted of 36 claims filed in 2006 and

coded with the assistance of a coding software program.

OUTCOME MEASURES: Outcome measures consisted at comparing

inappropriately denied reimbursement for the two groups.

METHODS: Methods involved reviewing all coding and reimbursement

from the two groups. We took each claim individually and examined each code

and each modifier for accuracy. Using the Medicare rule set (including the

Correct Coding Edits), we were able to determine how the claims should have

been coded, and how the misscoding affected the potential reimbursement.

RESULTS: Group A: There were 41 claims submitted in the first six

months of 2005, representing almost $110,000.00 ($109,649.69) of ex-

pected reimbursement. In the review, it was determined that eight of these

41 claims or 20% were under-reimbursed. The lost revenue was just under

$15,000 ($14,966.51), or 14%. Group B: 36 claims submitted, representing

expected reimbursement of just over $114,000 ($114,123.29). Using

computer aided coding, only three of the 36 or 8% of the claims were

under-reimbursed. More impressively, the lost revenue for those three

under-reimbursed claims totaled just over $1,500 ($1,534.20) or about 1%.

CONCLUSIONS: Computer aided coding can provide the surgeon with

information at the time of coding that can assist him in coding correctly.

Capturing the expected revenue on the ‘‘first pass’’ is incredibly valuable,

as the dispute process with many carriers is a maze of multiple appeals, all

of which must be timely, and each with its own set of rules that must be

followed. The cost of these appeals often times erodes the value of any

re-captured reimbursement we might ultimately gain. This makes it all

the more crucial to code correctly the first time.

FDA DEVICE/DRUG STATUS: This abstract does not discuss or include

any applicable devices or drugs.

doi: 10.1016/j.spinee.2007.07.381

P172. Hydroxyapatite-Coated Pedicle Screws for Dynamic Fixation

Significantly Improve Osseointegration at the Bone-Screw Interface

in an Ovine Model

Kevin Foley, MD1, Henry Bonin, MS2, Eric Lange, MS2, A. Simon Turner,

BVSC, MS3, Michael Beck, PhD4, Amy Lyons, MS3, Howard Seim, III,

DVM, DIPL ACVS3; 1University of Tennessee, Memphis, TN, USA;2Medtronic, Inc., Memphis, TN, USA; 3Colorado State University, Fort

Collins, CO, USA; 4Medtronic, Inc., Minneapolis, MN, USA

BACKGROUND CONTEXT: Pedicle-based dynamic fixation systems

have been introduced in the recent past. The risk of screw loosening following

these procedures remains significant, especially in patients with poor bone

quality. Surface augmentation of orthopedic implants with bioactive coatings

like hydroxyapatite (HA) has been shown to promote osseointegration and

may reduce the risk of screw loosening in dynamic fixation procedures.

PURPOSE: The purpose of this study was to determine what, if any, differ-

ences in osseointegration occurred between HA-coated and non-HA-coated

titanium pedicle screws when used in conjunction with a dynamic fixator in

an ovine model.

STUDY DESIGN/SETTING: An in vivo ovine model was used to eval-

uate various parameters of osseointegration of lumbar pedicle screws used

in conjunction with a dynamic fixator.

PATIENT SAMPLE: Twelve skeletally mature ewes were studied.

OUTCOME MEASURES: Extraction torque, percent connectivity (per-

cent surface area contact), bone implant contact.

METHODS: Single-level, bilateral pedicle screws were inserted in the lum-

bar spines of 12 sheep (48 screws). Each ipsilateral screw pair was connected

with a dynamic fixation device. Standard anodized titanium screws (4.5mm

diameter, 30mm length) were used in 6 sheep (24 standard screws); similar

titanium screws with a plasma-sprayed HA coating on the threaded portion

were used in the 6 others (24 HA–coated screws). Plain AP and lateral lumbar

radiographs along with axial CT scans were obtained at 3, 6, 9, and 12 months

postoperatively. Half of the sheep were euthanized at 6 months and half at

one year. Forty-two of the screws were removed from the harvested spines

using a torque wrench with digital readout and minimum torque measure-

ment of 0.5Nm. Peak extraction torque was recorded using a constant effort.

The other six screws were carefully harvested to capture approximately 5mm

of bone surrounding the screws. These 6 samples were analyzed using

162S Proceedings of the NASS 22nd Annual Meeting / The Spine Journal 7 (2007) 1S–163S

micro-CT and un-decalcified histology. Radiographs from 12-month sheep

following screw removal were analyzed for micro-fractures. Contact area

at the bone-screw interface was calculated using virtual volume subtraction

of the 3D reconstructed micro-CT scans. Stained, un-decalcified histology

was used to complement the micro-CT data.

RESULTS: Mean extraction torque for non-HA coated screws at 6 and

12 months postoperative was 1.64þ/�0.15 Nm and 1.49þ/�2.68 Nm,

respectively. Mean extraction torque for HA-coated screws at 6 and 12

months postoperative was 7.28þ/�1.41 Nm and 7.20þ/�0.97 Nm. Radio-

lucent zones (defined as greater than 1 mm in width) were not found on any

of the plain films or axial CTs. Micro-fractures were not found in radio-

graphs of vertebral bodies where screws were removed. Both 6 and

12-month HA-coated screws demonstrated 96.6% virtual connectivity on

micro-CT, whereas the 12-month non-HA-coated screws showed only

57.5% virtual connectivity. Measurements of bone-to-implant apposition

obtained with standard un-decalcified histology matched the micro-CT data.

CONCLUSIONS: In this study, HA coating significantly improved os-

seointegration at the bone-screw interface, as measured by peak extraction

torque, micro-CT analysis, and histology, in an ovine model of pedicle–

based lumbar dynamic fixation.

FDA DEVICE/DRUG STATUS: CD HORIZON Spinal System: Ap-

proved for this indication.

doi: 10.1016/j.spinee.2007.07.382

P173. Lumbar Degenerative Spondylolisthesis: MRI Accuracy

Christian Fras, MD1, Michael Fras, MD2; 1Department of Orthopaedic

Surgery, University of Pennsylvania, Philadelphia, PA, USA; 2Allegheny

General Hospital, Pittsburgh, PA, USA

BACKGROUND CONTEXT: Degenerative spondylolisthesis and steno-

sis of the lumbar spine is a common condition affecting the elderly. Surgi-

cally treatment frequently includes lumbar decompression and fusion.

Patients often present pre-operatively with only MRI studies of their lum-

bar spine, and no x-rays. It is unclear how often the MRI films accurately

diagnose the presence of the degenerative spondylolisthesis.

PURPOSE: The purpose of this study was to evaluate how frequently

MRI studies correctly diagnosed the presence of degenerative spondylolis-

thesis as established by x-rays.

STUDY DESIGN/SETTING: Retrospective review of all x-rays and

MRI’s of patients undergoing surgery for spinal stenosis and degenerative

spondylolisthesis.

PATIENT SAMPLE: All patients undergoing surgery for lumbar spinal

stenosis and degenerative spondylolisthesis.

OUTCOME MEASURES: Accuracy of MRI films in diagnosing the

presence of the degenerative spondylolisthesis.

METHODS: 25 consecutive patients who underwent lumbar decompres-

sion and fusion for L4-5 lumbar stenosis and spondylolisthesis were iden-

tified; all had Grade I degenerative spondylolisthesis on pre-operative

x-rays. The pre-operative MRI studies were then evaluated independently

of the x-rays for the presence of spondylolisthesis. The frequency with

which the MRI studies correctly identified the deformity noted on x-ray

was then determined.

RESULTS: Fourteen women and 11 men were identified; mean age was

66 (range: 47–78). All patients had degenerative Grade I spondylolisthesis

on x-ray. Review of MRI films revealed 6 of 25 patients to have no evi-

dence of spondylolisthesis apparent. The MRI report of 2 patients specif-

ically indicated that no spondylolisthesis was present, although direct

examination of these films revealed a clear deformity.

CONCLUSIONS: Lumbar stenosis with spondylolisthesis has been

shown by some studies to be better treated surgically by lumbar laminec-

tomy and fusion than by laminectomy alone. In this study, 24% of patients

treated surgically for lumbar stenosis and spondylolisthesis would have

had their spondylolisthesis missed if radiographic evaluation were limited

to MRI alone, and could consequently have had suboptimal surgical care

rendered. An additional 8% of patients who did have spondylolisthesis

visible on MRI had an accompanying radiology report failing to recognize

the deformity. The need for x-ray as well as MRI evaluation of the patient

with spinal stenosis is therefore emphasized, as is the need for the treating

surgeon to personally review all studies.

FDA DEVICE/DRUG STATUS: This abstract does not discuss or include

any applicable devices or drugs.

doi: 10.1016/j.spinee.2007.07.383

P174. Anterior Cervical Fusion with Bone-Morphogenic Protein-7

(OP-1) A Prospective Study of its Effects on Postoperative Laryngeal

Swelling

Jack Chen, MD1; 1Orthopaedic Specialty Institute, Orange, CA, USA

BACKGROUND CONTEXT: Recombinant human bone-mophogenic

protein-7 (trademark OP-1, Stryker) has shown clinical efficacy in induc-

ing posterolateral fusions in the lumbar spine. Its use in anterior cervical

spine surgery, however, has not been described.

PURPOSE: The purpose of this study is to evaluate the safety of its use in

terms of postoperative prevertebral soft-tissue swelling and complications

related to airway and swallowing difficulty.

STUDY DESIGN/SETTING: Prospective study. Tertiary referral center.

PATIENT SAMPLE: 44 consecutive patients were treated with anterior

cervical decompression and fusion. All patients were treated with allograft

rings packed with OP-1 and stabilized with anterior instrumentation.

OUTCOME MEASURES: Radiographic measurements of soft tissue

swelling. Clinical evaluation of laryngeal swelling.

METHODS: Lateral radiographs of the cervical spine were taken preoper-

atively, on postoperative day 2, and on postoperative day 7. We recorded the

day of when patients were able to tolerate regular diet. Adverse events related

to soft-tissue swelling, including prolonged hospitalization (greater than

48 hours), hospital readmission, or prolonged dysphagia were also recorded.

RESULTS: Nine patients had a 1-level fusion, 20 patients had 2-level fu-

sions, 11 patients had 3-level fusions, and 3 patients had 4-level fusions.

The average preoperative prevertebral soft tissue swelling measured 5.1

mm at C3 and 15.5 mm at C6. On postoperative day 2, the average mea-

surements were 11.6 mm at C3 and 18.3 mm at C6. This was compared to

previously published control measurements of 12.7 mm and 17.3 mm. On

postoperative day 7, the average swelling measured 10.7 mm and 20.4 mm,

respectively, at C3 and C6. Reported historical controls are 9 mm and 18

mm. Forty-three patients (97%) tolerated soft diet on postoperative day 2.

There were no hospital readmissions. One patient required reintubation and

evacuation of a postoperative hematoma.

CONCLUSIONS: Our series show that the incidence of soft-tissue swell-

ing complications with the use of rhBMP-7 is low. Furthermore, our mea-

surements of prevertebral swelling after surgery show trends comparable to

those reported after anterior fusion with tricortical iliac crest autogenous

graft and plate fixation.

FDA DEVICE/DRUG STATUS: OP-1 (Stryker): Not approved for this

indication.

doi: 10.1016/j.spinee.2007.07.384

P175. A Comparison of Two Retroperitoneal Surgical Approaches

for Total Disc Arthroplasty of the Lumbar Spine

John Bendo, MD1, Martin Quirno, MD1, Thomas Errico, MD1,

Jeffrey Spivak, MD1, Jeffrey Goldstein, MD1; 1NYU Hospital for Joint

Diseases, New York, NY, USA

BACKGROUND CONTEXT: The theoretical advantages of disc arthro-

plasty are motion preservation and improved clinical outcome, with a corre-

sponding decrease in the incidence of adjacent level disease. A number of

studies have demonstrated comparable clinical outcomes between TDA and

lumbar fusion recipients. Surgical technique has evolved to allow for proper

placement of various interbody implants and to avoid immediate and long-

term complications. Little is known about the impact of varying surgical

approaches on lumbar artificial disc implant position and clinical outcome.