Comparison of revision surgeries for one- to two-level cervical TDR and ACDF from 2002 to 2011

Accepted Manuscript Comparison of Revision Surgeries for 1-2 Level Cervical TDR and ACDF from 2002-2011 Sreeharsha V. Nandyala, BA, Research Coordinator Alejandro Marquez-Lara, MD, Research Coordinator Steven J. Fineberg, MD, Research Coordinator Kern Singh, MD, Associate Professor PII:

S1529-9430(14)00345-3

DOI:

10.1016/j.spinee.2014.03.037

Reference:

SPINEE 55832

To appear in:

The Spine Journal

Received Date: 1 November 2013 Revised Date:

5 March 2014

Accepted Date: 28 March 2014

Please cite this article as: Nandyala SV, Marquez-Lara A, Fineberg SJ, Singh K, Comparison of Revision Surgeries for 1-2 Level Cervical TDR and ACDF from 2002-2011, The Spine Journal (2014), doi: 10.1016/j.spinee.2014.03.037. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Sreeharsha V. Nandyala BA Research Coordinator Department of Orthopaedic Surgery Rush University Medical Center Chicago, IL

SC

Alejandro Marquez-Lara MD Research Coordinator Department of Orthopaedic Surgery Rush University Medical Center Chicago, IL

RI PT

Comparison of Revision Surgeries for 1-2 Level Cervical TDR and ACDF from 2002-2011

M AN U

Steven J. Fineberg MD Research Coordinator Department of Orthopaedic Surgery Rush University Medical Center Chicago, IL

TE D

Kern Singh MD Associate Professor Department of Orthopaedic Surgery Rush University Medical Center Chicago, IL

AC C

EP

Correspondence: Kern Singh, MD Associate Professor Department of Orthopaedic Surgery Rush University Medical Center 1611 W. Harrison St, Suite #300 Chicago, IL 60612 Phone: 312-432-2373 Fax: 708- 492-5373 E-mail: [email protected]

Disclosure: No funds were received in support of this work. No benefits in any form have been or will be received from any commercial party related directly or indirectly to the subject of this manuscript.

ACCEPTED MANUSCRIPT

Abstract

2

Background Context: Cervical total disc replacement (TDR) and anterior cervical

3

discectomy and fusion (ACDF) provide comparable outcomes for degenerative cervical

4

pathology. However, revisions of these procedures are not well characterized.

5

Purpose: To examine the rates, epidemiology, perioperative complications, and costs

6

between revision procedures as well as to compare these outcomes with those of primary

7

cases.

8

Study Design: Retrospective database analysis.

9

Patient Sample: 3,792 revision and 183,430 primary cases from the Nationwide

M AN U

SC

RI PT

1

Inpatient Sample (NIS) database from 2002-2011.

11

Outcome Measures: Incidence of revision cases, patient demographics, length of stay

12

(LOS), in-hospital costs, mortality, and perioperative complications.

13

Methods: Patients who underwent revision for either 1-2 level cervical TDR or ACDF

14

were identified. SPSS v.20 was utilized for statistical analysis with χ2 test for categorical

15

data and Independent-Samples T-test for continuous data. The relative risk for

16

perioperative complications with revisions was calculated in comparison to primary cases

17

utilizing a 95% confidence interval. An alpha level of <0.05 denoted statistical

18

significance.

19

Results: There were 3,536 revision 1-2 level ACDFs and 256 revision cervical TDRs

20

recorded in the NIS database from 2002-2011. The revision cervical TDR cohort

21

demonstrated a significantly greater LOS (3.18 vs 2.25; p<0.001), cost ($16,998 vs

22

$15,222; p=0.03), and incidence of perioperative wound infections (13.6 vs 5.3 per 1,000;

23

p<0.001) when compared with the ACDF revision cohort (p<0.001). There were no

AC C

EP

TE D

10

1

ACCEPTED MANUSCRIPT

differences in mortality between the revision surgical cohorts. When compared with

2

primary cases, both revision cohorts demonstrated a significantly greater LOS and cost.

3

Furthermore, patients who underwent revision demonstrated a greater incidence and risk

4

for perioperative wound infections, hematomas, dysphagia, and neurological

5

complications relative to the primary procedures.

6

Conclusions: This study demonstrated a significantly greater incidence of perioperative

7

wound infection, length of stay, and costs associated with a TDR revision when

8

compared with a revision ACDF. We propose that these differences are by virtue of the

9

inherently more invasive nature of revising total disc replacements. In addition, when

M AN U

SC

RI PT

1

10

compared with primary cases, revision procedures are associated with greater costs, LOS,

11

and complications including wound infections, dysphagia, hematomas, and neurological

12

events. These additional risks must be considered prior to opting for a revision procedure.

AC C

EP

TE D

13

2

ACCEPTED MANUSCRIPT

1

Introduction: Radiographic evidence demonstrates at least one degenerative finding in the

2

cervical spine on x-rays by age 65 in nearly 95% of men and 70% of women[1-3].

4

Cervical total disc replacement (TDR) and anterior cervical discectomy and fusion

5

(ACDF) are both indicated surgical interventions for this degenerative condition.

RI PT

3

Currently, five artificial disc devices are approved by the United States Food and

6

Drug Administration (FDA) for TDR involving single level anterior cervical

8

procedures[4]. These devices include the Bryan Disc (Medtronic Sofamor Danek,

9

Memphis, TN, USA), the Prestige Disc (Medtronic Sofamor Danek, Memphis, TN,

M AN U

SC

7

USA), the ProDisc-C (Synthes Spine, West Chester, PA, USA), the Secure C disc

11

(Globus Medical Inc, Audubon, PA, USA), and the PCM disc (NuVasive Inc, San Diego,

12

CA, USA)[4]. Several studies have consistently established the non-inferiority of TDR

13

with these approved products with regards to rates of revision when compared to ACDF

14

[5-7].

TE D

10

Although several studies and meta-analyses both within and outside of the FDA

16

Investigational Device Exemption (IDE) trials have assessed the revision rates of these

17

procedures, no studies to our knowledge have characterized the patient demographics,

18

length of stay (LOS), hospital costs, mortality, and perioperative complications

19

associated with revising these devices. The purpose of this study is to examine the rates,

20

epidemiology, perioperative complications, and costs between revision procedures as

21

well as to compare these outcomes with those of primary cases.

AC C

EP

15

22 23

Materials and Methods:

3

ACCEPTED MANUSCRIPT

1

The Nationwide Inpatient Sample (NIS) database is part of the Healthcare Cost and Utilization Project (HCUP) and is governed by the Agency for Healthcare Research

3

and Quality (AHRQ).[8] The NIS represents the largest all-payer healthcare database in

4

the United States approximating a 20% stratified sample of all hospital discharges. Each

5

entry contains data from a single hospital admission. The NIS issues admission diagnosis,

6

in-hospital complications, and procedural data utilizing the International Classification of

7

Disease-9th Edition-Clinical Modification (ICD-9-CM) codes.

8

Data Collection

SC

RI PT

2

Data from the Nationwide Inpatient Sample (NIS) was queried from 2002-2011.

10

Patients who underwent a 1-2 level (86.62) revision cervical TDR (846.6 or 846.9) with

11

or without concurrent anterior fixation (81.02) or a 1-2 level revision ACDF (81.30,

12

81.31, 81.32, or 81.39) were identified. The procedural codes for revision ACDF are also

13

utilized interchangeably for pseudarthrosis repair. Revision posterior cervical fusion

14

codes were not included because we could not ascertain which revision PCF cases were

15

subsequent to a primary ACDF. In addition, the procedural codes for revision TDR are

16

further qualified as the removal of a disc prosthesis with the synchronous insertion of a

17

new spinal disc prosthesis. We included revision TDR codes with or without concurrent

18

anterior fixation because one must assume that, even with documented revision cervical

19

TDR codes (846.6 or 846.9), the failed disc may not have been replaced with another

20

prosthesis, but rather, was converted into an anterior fixation. This ensured that we

21

captured all index level TDR revisions with the documented TDR revision codes

22

regardless of whether or not the prosthesis was actually replaced. Furthermore, only 1-2

AC C

EP

TE D

M AN U

9

4

ACCEPTED MANUSCRIPT

1

level revision procedures were included in order to ensure the inclusion of only index

2

level revision procedures in both cohorts. Patients were further stratified utilizing ICD-9-CM diagnosis codes for

4

degenerative cervical pathology including cervical spondylosis with or without

5

myelopathy, degenerative disc disease, and radiculitis (721.0–1, 722.0, 722.4, 722.71,

6

722.81, 722.91, 723.x). Only elective admissions were included to eliminate confounding

7

factors. Patients younger than 18 years were excluded from the study.

SC

RI PT

3

The revision surgical cohorts were analyzed for demographic distribution

9

evaluating age, gender, race, and the comorbid risk factors. In addition, the perioperative

10

outcomes for revision cases were compared with those of primary procedures. To assess

11

patient comorbidities, the Charlson Comorbidity Index was calculated utilizing data from

12

the NIS Disease Severity Measure Files [9]. The Charlson Comorbidity Index (CCI)

13

utilizes 22 different comorbidities and the patients’ age to predict the 10-year mortality

14

[10]. Several minor modifications were made to the CCI to account for the fact that not

15

all 22 comorbidities were recorded by the NIS database. These modifications include: 1)

16

a history of myocardial infarction was omitted, and 2) liver disease was given an adjusted

17

weight of 2 points rather than 1 point for mild disease and 3 points for moderate to severe

18

liver disease. Previous studies have established that slight modifications to the CCI have

19

minimal impact on the overall score and also demonstrate excellent agreement between

20

various modifications for calculating the CCI [11, 12]. Lastly, the in-hospital outcomes

21

including length of stay (LOS), hospital costs, and mortality were also assessed.

22

Data Analysis

AC C

EP

TE D

M AN U

8

5

ACCEPTED MANUSCRIPT

1

SPSS v.20 was utilized for statistical analysis with χ2 test for categorical data and independent sample T-test for continuous data. The relative risk for perioperative

3

complications associated with revision procedures was calculated in relation to primary

4

cases with a 95% confidence interval. An alpha level of <0.05 denoted statistical

5

significance.

RI PT

2

6

8

Results:

SC

7

There were 3,536 revision 1-2 level ACDFs and 256 revision cervical TDRs recorded from 2002-2011. Of the 256 revision cervical TDRs, 106 cases had concurrent

10

ACDF codes. The number of revision cases in both cohorts increased during this period

11

(Figure 1). There were no significant differences between the mean age (50.9 vs 50.3

12

years), comorbidity burden (CCI 2.06 vs 2.03), race, or teaching status (50.2% vs 46.8%)

13

between the revision ACDF and cervical TDR cohorts, respectively (Table 1).

TE D

14

M AN U

9

The hospital course varied considerably between the revision groups. The revision cervical TDR cohort incurred an additional 0.9 days in LOS and $1776 in total hospital

16

costs when compared with the revision ACDF treated patients (p<0.001) (Table 1). In

17

addition, the revision cervical TDR cohort demonstrated a significantly greater incidence

18

of perioperative infection (p= 0.03) (Table 1). However, mortality did not vary between

19

the revision cohorts.

AC C

20

EP

15

There were significant differences in the outcomes between the primary and

21

revision cases. ACDF revision cases demonstrated a greater LOS (2.25 vs 1.8 days) and

22

total hospital costs ($15,222 vs $12,781) when compared with primary procedures

23

(p<0.001) (Table 2). In addition, the ACDF revision cohort demonstrated a greater

6

ACCEPTED MANUSCRIPT

prevalence of perioperative wound infections, dysphagia, hematomas, and neurological

2

events when compared primary procedures (p<0.001) (Table 2). Similarly, the cervical

3

TDR revision cohort also demonstrated analogous variations in outcomes when compared

4

with primary cases (Table 2).

5

RI PT

1

Compared with primary procedures, both the revision ACDF and cervical TDR cohorts demonstrated significantly greater relative risks for perioperative wound

7

infections, hematomas, dysphagia, and neurological complications (Table 3).

SC

6

8

10

Discussion:

M AN U

9

Proponents of cervical disc replacement have clearly demonstrated the noninferiority of cervical arthroplasty when compared with ACDF [13-18]. The prospective,

12

randomized, controlled FDA IDE trials have reported no significant differences in the

13

LOS, costs, or the intraoperative course [5-7, 19]. No studies, to our knowledge, have

14

assessed these outcome parameters with regards to revision cases. In addition, this is the

15

first study to compare perioperative outcomes and costs between single-level revision and

16

primary anterior cervical arthrodesis and arthroplasty procedures on a national scale.

EP

17

TE D

11

The revision and reoperation rates for single-level ACDF and cervical TDR have come under considerable scrutiny[20]. In the present study, 256 revision TDRs and 3,536

19

revision ACDFs were documented, providing an NIS database generated “revision rate”

20

of 7.7% and 2.0%, respectively. It must be noted that this is not a true longitudinal

21

analysis of revision rates because the utilization of the NIS database precludes long-term

22

follow up. However, the documented proportion of revisions is noteworthy and warrants

23

additional analysis because the figures stem from an unbiased dataset. Most FDA IDE

AC C

18

7

ACCEPTED MANUSCRIPT

1

trials and other retrospective studies have suggested that the 2-year revision and

2

reoperation rates favor cervical arthroplasty.

3

Furthermore, multiple meta-analyses have demonstrated that cervical TDR likely carries superiority in this regard over anterior arthrodesis [21-23]. However, there is

5

“very low” to “low” quality evidence to support these conclusions [21-23]. A recent

6

Cochrane Database Review concluded that the superiority claims of cervical TDR with

7

regards to revision and reoperation are marred by significant bias and are not robust to

8

sensitivity analysis which factor in bias [24]. In addition, in the present study, the ACDF

9

revision rate of 2.0% is far lower than that of the control arms in the FDA IDE trials

M AN U

SC

RI PT

4

(6.2%-12.2%). Singh et retrospectively compared their own revision rates with that of the

11

single-level ACDF control cohorts of the IDE trials and demonstrated a 2-year repeat

12

surgery rate of 2.1% in the single-level ACDF cohort, similar to the present study. The

13

authors suggested that this discrepancy is likely related to the variations in the criteria and

14

threshold for repeat surgery[20]. Other explanations may involve the patient’s perception

15

of pain and vested interest for success depending upon whether or not the patient received

16

a TDR[20]. Despite the lack of a true longitudinal analysis, the surprisingly large

17

proportion of revisions documented during this period, particularly revision TDRs,

18

warrants further unbiased investigation.

EP

AC C

19

TE D

10

The revision cervical TDR cohort demonstrated a significantly greater LOS,

20

costs, and incidence of perioperative infection when compared with the revision ACDF

21

cohort. These differences between the revision cases are contrasted by the similarity in

22

outcomes between primary ACDF and cervical TDR, as published by previous reports.

23

The worsened outcomes for revision TDR are likely related to the more invasive nature

8

ACCEPTED MANUSCRIPT

of revising cervical arthroplasty, as this may require extensive surgical wound exposure

2

and dissection to remove the prosthesis. The longer hospitalization likely contributed to

3

the added expense associated with revising a cervical arthroplasty. Investigation into the

4

nuances involved with revising these procedures is warranted to gain further

5

understanding as to why revision cervical TDR is associated with worsened outcomes

6

over revision ACDF while primary cervical arthroplasty has been demonstrated to be

7

non-inferior by previous trials.

SC

9

There were dramatic differences between the perioperative outcomes of primary and revision cases. The revision cohorts incurred significantly greater costs, LOS, and

M AN U

8

RI PT

1

perioperative complications than the primary procedures. The older and more comorbid

11

patient population that underwent revision may explain these findings. In addition, repeat

12

surgical exposures are a known risk factor for greater perioperative complications. More

13

importantly, this study demonstrated the relative risk for complications associated with

14

revision compared with primary cases. Both revision cohorts demonstrated a drastic

15

increase in the risk for perioperative infection, dysphagia, hematomas, and neurological

16

events. In light of the additional hospital costs and hospital resource utilization associated

17

with revision, the risk increase for complications must be carefully considered prior to

18

opting for revising an ACDF or cervical TDR. This study clearly highlights that these

19

additional risks and costs must be weighed against the potential benefits of revision

20

surgery.

EP

AC C

21

TE D

10

Despite the novelty of this study, there are several pitfalls in utilizing a large-scale

22

database. The NIS database carries risks for inaccuracy partially due to the

23

ineffectiveness of ICD-9-CM coding in capturing complications[25]. Only complications

9

ACCEPTED MANUSCRIPT

that occur during the initial inpatient admission are recorded and those that occur after

2

discharge are not included. Campbell et al compared the inaccuracy of ICD-9-CM codes

3

with prospectively collected data and demonstrated that most complication rates are

4

under reported by ICD-9-CM Codes[25]. In addition, large administrative databases are

5

subject to inaccuracies in capturing correct diagnoses and procedures[25]. Variations in

6

coding practices or errors in data input may introduce additional inaccuracies in patient

7

selection from the NIS database. Our methodology for identifying complications may not

8

be perfect, as some complications, say dysphagia, may have been overlooked or

9

misdocumented. Furthermore, the NIS database does not facilitate for long-term follow

M AN U

SC

RI PT

1

up so we could not analyze the incidence of pseudarthrosis, persistent symptoms, or a

11

true longitudinal analysis to assess revision rates. The incidence of adjacent segment

12

degeneration following surgery is not a coded diagnosis, and therefore, will not be

13

reported in the NIS. Lastly, the NIS database does not provide an assessment of the

14

severity of the patient’s symptoms.

15

TE D

10

To address these limitations, we first utilized procedural codes and then diagnostic codes to identify patients to improve the specificity of our search. As

17

explained in the methodology, we chose to utilize a very stringent inclusion and

18

exclusion criteria. The revision codes utilized in this analysis have very specific

19

definitions for revision. The revision codes for ACDF are re-fusion codes that are applied

20

interchangeably for pseudarthrosis repair while the codes for revision cervical TDR

21

specify the removal of disc prosthesis with the synchronous insertion of new spinal disc

22

prosthesis. We included revision TDR codes with or without concurrent anterior fixation

23

because one must assume that, even with documented revision cervical TDR codes

AC C

EP

16

10

ACCEPTED MANUSCRIPT

(846.6 or 846.9), the failed disc may not have been replaced with another prosthesis, but

2

rather, was converted into an anterior fixation. This ensured that we captured all index

3

level TDR revisions with the documented TDR revision codes regardless of whether or

4

not the prosthesis was actually replaced. Lastly, only elective cases were included while

5

trauma and tumor diagnoses were excluded to reduce confounding factors.

6

RI PT

1

In order to address data entry inaccuracies, the HCUP utilizes benchmarks to

ensure quality data collection. The HCUP databases has been evaluated and benchmarked

8

against other sources of data including the National Hospital Ambulatory Medical Care

9

Survey and the AHA Annual Survey Databases. In addition, the HCUP performs a

M AN U

SC

7

number of edit checks on the validity of coding. This benchmark process enables the

11

HCUP databases to be linked with other databases including the American Hospital

12

Association Survey and the Area Resource File with accuracy.

13

Conclusion

TE D

10

In this analysis of the NIS database, the revision cervical TDR cohort

15

demonstrated a greater hospitalization, costs, and incidence of perioperative infection

16

than the revision ACDF cohort. We propose that these differences are by virtue of the

17

inherently more invasive nature of revising total disc replacements. In addition, when

18

compared with primary cases, revision procedures are associated with greater costs, LOS,

19

and incidence and risk for complications including wound infections, dysphagia,

20

hematomas, and neurological events. Interestingly, of note, the increased incidence of

21

revision cervical TDR needs to be further evaluated, as these rates appear to be greater

22

than reported in the current surgical literature.

AC C

EP

14

23

11

ACCEPTED MANUSCRIPT

References

EP

TE D

M AN U

SC

RI PT

1. Lehto IJ, Tertti MO, Komu ME, Paajanen HE, Tuominen J, Kormano MJ. Agerelated MRI changes at 0.1 T in cervical discs in asymptomatic subjects. Neuroradiology. 1994;36(1):49-53. 2. Boden SD, McCowin PR, Davis DO, Dina TS, Mark AS, Wiesel S. Abnormal magnetic-resonance scans of the cervical spine in asymptomatic subjects. A prospective investigation. The Journal of bone and joint surgery American volume. 1990;72(8):117884. 3. Gore DR, Sepic SB, Gardner GM, Murray MP. Neck pain: a long-term follow-up of 205 patients. Spine. 1987;12(1):1-5. 4. Moatz B, Tortolani PJ. Cervical disc arthroplasty: Pros and cons. Surgical neurology international. 2012;3(Suppl 3):S216-24. 5. Heller JG, Sasso RC, Papadopoulos SM, et al. Comparison of BRYAN cervical disc arthroplasty with anterior cervical decompression and fusion: clinical and radiographic results of a randomized, controlled, clinical trial. Spine. 2009;34(2):101-7. 6. Mummaneni PV, Burkus JK, Haid RW, Traynelis VC, Zdeblick TA. Clinical and radiographic analysis of cervical disc arthroplasty compared with allograft fusion: a randomized controlled clinical trial. Journal of neurosurgery Spine. 2007;6(3):198-209. 7. Murrey D, Janssen M, Delamarter R, et al. Results of the prospective, randomized, controlled multicenter Food and Drug Administration investigational device exemption study of the ProDisc-C total disc replacement versus anterior discectomy and fusion for the treatment of 1-level symptomatic cervical disc disease. The spine journal : official journal of the North American Spine Society. 2009;9(4):275-86. 8. Healthcare Cost and Utilization Project (HCUP): Overview of the Nationwide Inpatient Sample (NIS). Rockville, MD: Agency for Healthcare Research and Quality; 2002-2009; Available from: http://www.hcup-us.ahrq.gov/nisoverview.jsp. 9. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Journal of chronic diseases. 1987;40(5):373-83. 10. Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. Journal of clinical epidemiology. 1994;47(11):1245-51. 11. Ghali WA, Hall RE, Rosen AK, Ash AS, Moskowitz MA. Searching for an improved clinical comorbidity index for use with ICD-9-CM administrative data. Journal of clinical epidemiology. 1996;49(3):273-8. 12. Cleves MA, Sanchez N, Draheim M. Evaluation of two competing methods for calculating Charlson's comorbidity index when analyzing short-term mortality using administrative data. Journal of clinical epidemiology. 1997;50(8):903-8. 13. Coric D, Nunley PD, Guyer RD, et al. Prospective, randomized, multicenter study of cervical arthroplasty: 269 patients from the Kineflex|C artificial disc investigational device exemption study with a minimum 2-year follow-up: clinical article. Journal of neurosurgery Spine. 2011;15(4):348-58. 14. Jawahar A, Cavanaugh DA, Kerr EJ, 3rd, Birdsong EM, Nunley PD. Total disc arthroplasty does not affect the incidence of adjacent segment degeneration in cervical

AC C

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

12

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

spine: results of 93 patients in three prospective randomized clinical trials. The spine journal : official journal of the North American Spine Society. 2010;10(12):1043-8. 15. Mummaneni PV, Robinson JC, Haid RW, Jr. Cervical arthroplasty with the PRESTIGE LP cervical disc. Neurosurgery. 2007;60(4 Suppl 2):310-4; discussion 4-5. 16. Phillips FM, Lee JY, Geisler FH, et al. A prospective, randomized, controlled clinical investigation comparing PCM cervical disc arthroplasty with anterior cervical discectomy and fusion. 2-year results from the US FDA IDE clinical trial. Spine. 2013;38(15):E907-18. 17. Sasso RC, Smucker JD, Hacker RJ, Heller JG. Clinical outcomes of BRYAN cervical disc arthroplasty: a prospective, randomized, controlled, multicenter trial with 24-month follow-up. Journal of spinal disorders & techniques. 2007;20(7):481-91. 18. Zigler JE, Delamarter R, Murrey D, Spivak J, Janssen M. ProDisc-C and anterior cervical discectomy and fusion as surgical treatment for single-level cervical symptomatic degenerative disc disease: five-year results of a Food and Drug Administration study. Spine. 2013;38(3):203-9. 19. Burkus JK, Haid RW, Traynelis VC, Mummaneni PV. Long-term clinical and radiographic outcomes of cervical disc replacement with the Prestige disc: results from a prospective randomized controlled clinical trial. Journal of neurosurgery Spine. 2010;13(3):308-18. 20. Singh K, Phillips FM, Park DK, Pelton MA, An HS, Goldberg EJ. Factors affecting reoperations after anterior cervical discectomy and fusion within and outside of a Federal Drug Administration investigational device exemption cervical disc replacement trial. The spine journal : official journal of the North American Spine Society. 2012;12(5):372-8. 21. Gao Y, Liu M, Li T, Huang F, Tang T, Xiang Z. A meta-analysis comparing the results of cervical disc arthroplasty with anterior cervical discectomy and fusion (ACDF) for the treatment of symptomatic cervical disc disease. The Journal of bone and joint surgery American volume. 2013;95(6):555-61. 22. McAfee PC, Reah C, Gilder K, Eisermann L, Cunningham B. A meta-analysis of comparative outcomes following cervical arthroplasty or anterior cervical fusion: results from 4 prospective multicenter randomized clinical trials and up to 1226 patients. Spine. 2012;37(11):943-52. 23. Yin S, Yu X, Zhou S, Yin Z, Qiu Y. Is cervical disc arthroplasty superior to fusion for treatment of symptomatic cervical disc disease? A meta-analysis. Clinical orthopaedics and related research. 2013;471(6):1904-19. 24. Boselie TF, Willems PC, van Mameren H, de Bie RA, Benzel EC, van Santbrink H. Arthroplasty versus fusion in single-level cervical degenerative disc disease: a Cochrane review. Spine. 2013;38(17):E1096-107. 25. Campbell PG, Malone J, Yadla S, et al. Comparison of ICD-9-based, retrospective, and prospective assessments of perioperative complications: assessment of accuracy in reporting. Journal of neurosurgery Spine. 2011;14(1):16-22.

AC C

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

13

ACCEPTED MANUSCRIPT

1

Figure Legend

2

Figure 1: Trends of 1-2 Level Revision Procedures for ACDF and Cervical TDR.

AC C

EP

TE D

M AN U

SC

RI PT

3

14

ACCEPTED MANUSCRIPT

ACDF 3536 50.9 (10.3) 56.1%

RI PT

Table 1: Comparison of Revision Outcomes between 1-2 Level ACDF and Cervical TDR from 2002-2011 Cervical TDR 256 50.3 (10.8) 53.6%

TE D

M AN U

SC

Revision Procedures (%) Average Age in Years (SD) % Female Charlson Co-Morbidity Index 2.06 (1.4) 2.03 (1.5) (SD) 71.0% 81.9% % Caucasian 50.2% 46.8% Teaching Hospital Status (%) 2.25 (2.8) 3.18 (5.1) Length of Stay in Days (SD) 15,222 (10,536) 16,998 (13,852) Hospital Costs in $ (SD) In-Hospital Complications* 77.2 95.2 Overall Complication Rate 0.3 0.0 Pulmonary Embolism 1.3 0.0 Deep Venous Thrombosis 5.3 13.6 Infection 4.1 4.3 Cardiac 14.7 21.5 Hematoma 7.4 12.9 Neurologic Complications

P-VALUE 0.329 0.494 0.095

0.058 0.342 <0.001 0.034 0.37 0.787 0.589 <0.001 0.958 0.415 0.431

AC C

EP

SD= Standard Deviation, TDR= Total disc replacement, ACDF= Anterior cervical discectomy and fusion *Per 1,000

ACCEPTED MANUSCRIPT

Table 2: Comparison between Revision and Primary Procedural Outcomes between 1-2 Level ACDF and Cervical TDR from 2002-2011 Revision ACDF

180,100

3536 (2.0%)

49.2

50.9

<0.001

52.9%

56.1%

<0.001

Charlson Co-Morbidity Index

1.8

2.1

<0.001

Teaching Hospital Status (%)

49.8%

50.2%

Length of Stay in Days

1.8

2.25

Hospital Costs (US $)

$12,781

$15,222

% Female

46.8

50.3

<0.001

51.7%

53.6%

0.62

1.5

2.03

<0.001

0.58

48.4%

46.8%

0.79

<0.001

1.8

3.18

<0.001

<0.001

$14,580

$16,998

0.001

0.3

0.5

0.60

0.0

0.69

1.3

0.2

0.90

0.0

0.64

5.3

<0.001

1.80

13.6

0.001

4.1

0.7

3.30

4.3

0.8

4.56

14.7

<0.001

4.80

21.5

0.001

1.38

7.4

<0.001

0.90

12.9

<0.001

Deep Venous Thrombosis

0.81

Infection

0.86

Cardiac

3.60

Neurologic Complications

EP

0.51

AC C

Pulmonary Embolism

TE D

In-Hospital Complications*

Hematoma

P-VALUE

256 (7.7%)

SC

Average Age in Years

Revision Cervical TDR

3,330

M AN U

Cases

P-VALUE

Primary Cervical TDR

RI PT

Primary ACDF

ACCEPTED MANUSCRIPT

24.51

43.8

<0.001

16.22

42.9

<0.001

Mortality

0.01

0.3

0.4

0.30

0.0

0.79

RI PT

Dysphagia

SD= Standard Deviation, TDR= Total disc replacement, ACDF= Anterior cervical discectomy and fusion

AC C

EP

TE D

M AN U

SC

*Per 1,000

RI PT

ACCEPTED MANUSCRIPT

ACDF

M AN U

SC

Table 3: Relative risk for Perioperative Complications with 1-2 Level Revisions Compared with 1-2 Level Primary Procedures Cervical TDR

RR*

95% CI

P-Value

RR*

95% CI

Pulmonary Embolism

0.56

0.6

0.08-3.90

0.49

2.4

0.2-30.2

Deep Venous Thrombosis

0.21

1.7

0.73-4.1

0.66

1.8

0.1-24.1

Wound Infection

<0.001

6.5

4.4-9.6

<0.001

4.9

1.9-12.3

Cardiac

0.72

1.1

0.65-1.85

0.8

1.2

0.2-7.9

Hematoma

<0.001

3.2

2.4-4.2

0.004

3.5

1.6-7.6

Neurologic Complications

<0.001

4.6

3.1-6.7

0.001

8.4

4.4-16.1

Dysphagia

<0.001

1.8

1.6-2.1

<0.001

2.2

1.2-3.8

0.2

0.3

0.04-1.9

0.3

3.6

0.3-39.8

*Risk relative to primary procedures

EP

AC C

Mortality

TE D

P-Value

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT