SPINE Volume 40, Number 12, pp 917-925 ©2015, Wolters Kluwer Health, Inc. All rights reserved.

CLINICAL CASE SERIES

Posterolateral Lumbar Arthrodesis With and Without Interbody Arthrodesis for L4–L5 Degenerative Spondylolisthesis A Comparative Value Analysis Michael B. Gottschalk, MD,* Ajay Premkumar, BS,† Kyle Sweeney, MD,† Scott D. Boden, MD,‡ John Heller, MD‡, S. Tim Yoon, MD, PhD,† John M. Rhee, MD,† Steven K. Leckie, MD,§ Brett Braly, MD,¶ Andrew K. Simpson, MD, MHS, and Eric Lenehan, MD**

Study Design. Independent retrospective review of prospectively collected data, comparative cohort study. Objective. The objective of this study was to compare the clinical, radiographical, and cost/value of the addition of an interbody arthrodesis (IBA) to a posterolateral arthrodesis (PLA) in the surgical treatment of L4–L5 degenerative spondylolisthesis (DS). The authors hypothesized that the addition of IBA to PLA would produce added value while incurring minimal additional costs. Summary of Background Data. Many lumbar surgical advances have been made during the past several decades, yet there is a paucity of strong evidence-based validation, let alone comparative value analyses. The addition of an IBA to a PLA has become increasingly popular during the past 2 decades, yet the potential added value for the patient has not been carefully defined. Methods. Patients undergoing single-level arthrodesis for L4–L5 DS performed at our institution from 2004 to 2012 were identified. Exclusion criteria included multilevel arthrodesis, spinal stenosis requiring decompression at or above L2–L3, previous L4–L5 spinal From the *Department of Orthopaedic Surgery, Emory University School of Medicine, Atlanta, GA; †Department of Orthopaedic Surgery, Emory University School of Medicine, Atlanta, Ga; ‡The Emory Spine Center, Emory University School of Medicine, Atlanta, GA; §Plymouth Bay Orthopedic Associates, Beth Israel Deaconess Hospital Plymouth, Duxbury, MA; ¶Oklahoma Sports Science and Orthopaedics, Edmond, OK; Texas Back Institute, Dallas, TX; and **Department of Orthopedic Surgery, Greenville Health System, Greenville, SC. Acknowledgment date: October 13, 2014. First revision date: January 8, 2015. Acceptance date: February 23, 2015. The device(s)/drug(s) that is/are the subject of this article is/are not FDAapproved for this indication and is/are not commercially available in the United States. No funds were received in support of this work. Relevant financial activities outside the submitted work: board membership, consultancy, payment for lectures, grants, royalties, stocks. Address correspondence and reprint requests to Michael B. Gottschalk, MD, Department of Orthopaedic Surgery, Emory University School of Medicine, 59 Executive Park South Atlanta, GA 30329; E-mail: [email protected] or [email protected] DOI: 10.1097/BRS.0000000000000856 Spine

fusion, spondylolisthesis of greater than 33% of the vertebral body, and use of minimally invasive surgery. Radiographical fusion status, epidemiological, surgical, and functional outcomes, and cost/value data were recorded or calculated. Results. A total of 179 patients with follow-up meeting inclusion criteria were identified: 68 with PLA alone and 111 with PLA + IBA. No statistical differences were noted in Oswestry Disability Index, 36-item Short-Form Health Survey scores, fusion rates, or cost/value at 6 months and at more than 3 years despite the PLA cohort being significantly older with more medical comorbidities. When length of stay was normalized across cohorts, the addition of an IBA increased hospital costs ranging from $577 to $5276, but this did not reach statistical significance. Conclusion. This single-center review of open surgical treatment of L4–L5 DS demonstrated that the addition of IBA to PLA added cost while producing equivalent results in fusion rates, Oswestry Disability Index, and 36-item Short-Form Health Survey scores when compared with PLA alone. Key words: degenerative spondylolisthesis, spinal fusion, spinal stenosis, value analysis, 360° fusion, interbody fusion, TLIF, posterolateral fusion, cost-effectiveness, instrumentation. Level of Evidence: 3 Spine 2015;40:917–925

M

any technical advances in lumbar fusion have been made during the past several decades, yet there is a paucity of strong evidence-based validation, let alone comparative value analyses.1 The institution of the Affordable Care Act has accelerated a transition from volume- to valuebased care, and, as a result, literature supporting or refuting current surgical techniques is imperative. Lumbar fusions performed in 2003 accounted for 47% of Medicare dollars spent on spine surgery.2 A recent study investigating regional variation in the rates of lumbar fusions for Medicare enrollees demonstrated a 20-fold variation; this represented the largest variation in any surgical procedure.2 As degenerative www.spinejournal.com

917

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. SPINEIP1407_LR 917

14/05/15 7:56 AM

CLINICAL CASE SERIES

Addition of Interbody to Posterolateral Fusion • Gottschalk et al

spondylolisthesis (DS) accounts for a significant proportion of patients older than 60 years undergoing lumbar fusion surgery and is one of the better-documented indications for spinal arthrodesis, this disease subset was considered the focus of the present study.3 Despite the quality of available data, current and previous trends in lumbar spine surgery continue to demonstrate a rise in the number of combined posterolateral and interbody arthrodesis (PLA + IBA) for routine degenerative diagnoses.4–6 This can potentially be attributed to some of the purported advantages of the addition of an IBA such as increased fusion success, better foraminal distraction, and improved spinal alignment. Scant data exist comparing PLA alone with PLA + IBA for degenerative lumbar disorders.7,8 The objective of this study was to compare the clinical, radiographical, and cost/value of the addition of an IBA to a PLA in the surgical treatment of L4–L5 DS. The authors hypothesized that the addition of IBA to PLA will produce added value while incurring minimal additional costs.

MATERIALS AND METHODS Institutional review board approval was obtained prior to the onset of the study. A prospective database of patients undergoing spinal surgery by 4 senior spine surgeons at our institution was reviewed for all patients undergoing L4–L5 spinal arthrodesis for a diagnosis of DS from 2004 to 2012. Exclusion criteria included multilevel arthrodesis, spinal stenosis requiring decompression at or above L2–L3, previous L4–L5 spinal fusion, spondylolisthesis of greater than 33% of the vertebral body, and use of minimally invasive surgery.

Patients who had previously undergone lumbar decompression surgery (without arthrodesis) with known DS were still included. Patients were categorized into 2 groups: PLA alone and PLA + IBA. The PLA-alone group consisted of patients who underwent arthrodesis with pedicle screw instrumentation in all cases. The PLA + IBA group consisted of those patients who received pedicle screw instrumentation as well as an interbody device. The decision to add an IBA was at the discretion of the treating surgeon and typically unrelated to patient variables. Two of the surgeons almost exclusively perform PLA alone, whereas the other 2 surgeons almost always prefer PLA + IBA. A total of 238 patients were identified meeting the aforementioned criteria. Upon contact for long-term follow-up (e.g., follow-up at least 2 yr out from surgery), 13 patients were deceased and an additional 46 were unavailable for contact. A total of 179 patients meeting inclusion criteria were identified, with an average follow-up of more than 38.7 months (range, 25.16–117.32 mo): 68 having undergone PLA and 111 having received PLA + IBA. Study size was determined by convenience. Figure 1 demonstrates patient cohort designations. Prospectively recorded outcome data included visual analogue pain scale (VAS), Oswestry Disability Index (ODI) version 2, and 36-item Short-Form Health Survey (SF-36) version 2.9,10 Epidemiological variables were reviewed, which included age, height, weight, body mass index, medical comorbidities, Charlson Comorbidity Index, smoking status, and chief complaint at the time of preoperative visit. Surgical data reviewed included primary versus revision operation, number of levels decompressed, type of bone graft used, reoperation

Assessed for eligibility

eligibility Excluded ♦ Minimally Invasive Surgery ♦ Spinal stenosis requiring decompression at or above L2-L3 ♦ Previous L4-L5 Fusion ♦ Multilevel Fusion ♦ Spondylolisthesis >33% of vertebral body ♦ Deceased (n = 13, 6 Posterolateral Arthrodesis, 7 Posterolateral and Interbody Arthrodesis)

Available (n = 225) Excluded: 46 2-yr f/u (80%)

allocation Allocated to Posterolateral Arthrodesis (n = 68)

Allocated to Posterolateral and Interbody Arthrodesis (n = 111)

Avg clinical outcomes follow-up: 40.28 mo Avg clinical outcomes follow-up: 37.7 mo Avg radiographic follow-up: 28.49 mo Avg radiographic follow-up: 24.78 mo

918

Figure 1. Flow diagram demonstrating the exclusion and allocation of patients to the 2 cohorts, PLA and PLA + IBA. PLA indicates posterolateral arthrodesis; IBA, interbody arthrodesis; f/u, follow-up.

www.spinejournal.com

June 2015

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. SPINEIP1407_LR 918

14/05/15 7:56 AM

CLINICAL CASE SERIES rate, estimated blood loss, length of case in minutes, American Society of Anesthesiologists rating, complications, drain output, pre- and postoperative hemoglobin/hematocrit, and transfusion status. Financial data reviewed included hospital charges, hospital costs, length of stay (LOS), complication costs, and reoperation costs. Radiographical data reviewed included both plain radiographs and computed tomographic (CT) scans. Independent fellowship trained spine surgeons graded fusion status. Fusions were graded using previously established criteria and then converted to a binary scale as either fused or not fused.11,12 For long-term follow-up (e.g., follow-up at least 2 yr out from surgery), patients were consented via phone and asked to complete a survey by secure e-mail. The e-mail survey included VAS, ODI, and SF-36. Patients also received questions pertaining to additional spine surgery. For those patients who agreed to participate, anterior/posterior and lateral lumbar radiographs were obtained at no charge. Nonparametric tests including the Mann-Whitney U test and the Wilcoxon-Kruskal-Wallis (rank sums) test were used for continuous variables without a normal distribution. For nominal data, χ2 analysis (n >10) or the Fisher exact test (n