Bone graft substitutes and extenders are routinely used during spinal arthrodesis surgery in an effort to decrease the morbidity associated with iliac crest autograft harvest, the rising popularity of minimally invasive approaches, and technological advances that have been successful at achieving bony fusion.1–3 The presence of variability in its use is a reflection of uncertainty regarding effectiveness and outcomes, and unawareness regarding costs. An evidence-based approach to cost-effective bone graft decision making would consider both cost and outcome, and be guided by the goal of optimizing value. Optimizing value involves maximizing the patient-centered benefit or outcome while minimizing the cost and risk. Outcomes, costs, and risks are difficult to measure, and decision making in many areas of medicine, including bone graft alternatives, is made under conditions of uncertainty and limited information. Recently, significant controversy has developed regarding the use of recombinant human bone morphogenetic protein (rhBMP-2), which acts a potent molecule of osteoinduction because it has been associated with significant complications.4,5 Biologics and bone graft substitutes are being increasingly used and their costs may represent a significant percentage of hospital charges (17%–47%6–8 [calculated]). The purpose of this article is to provide a systematic review of the literature on bone graft substitutes and extenders with the intention of addressing the cost-effectiveness of bone graft extenders and substitutes in spine surgery.
This systematic review was designed to evaluate the following key questions:
- Is there evidence to suggest that the use of rhBMP-2 for spinal arthrodesis is cost-effective compared with the use of iliac crest bone graft (ICBG)?
- Is there evidence to suggest that the use of allograft or bone graft extenders (demineralized bone matrices, synthetic carriers, and allogeneic stem cell products) is cost-effective in spinal arthrodesis compared with the use of ICBG?
- Are there differences in cost-effectiveness with the use of allograft bone versus polyetheretherketone cages with bone graft substitutes in anterior cervical discectomy and fusion (ACDF)?
MATERIALS AND METHODS
Electronic Literature Search
A systematic search of PubMed/MEDLINE, the Cochrane Collaboration Library, EMBASE, the CRD database, and Tuft's CEA registry for literature published through December 2013 was conducted. The search strategy was based on the key questions and inclusion/exclusion criteria established a priori (Table 1). The search strategy is documented in the Supplemental Digital Content, available at http://links.lww.com/BRS/A896. Full formal economic analyses (including cost-minimization, cost-benefit, cost-effectiveness, and cost-utility studies) comparing the use of biological grafts with ICBG autograft in spinal fusion for thoracolumbar or cervical degenerative spinal pathologies were specifically sought, and articles were reviewed for inclusion/exclusion based on the criteria described in Table 1. Incomplete economic evaluations such as costing studies or studies that did not compare costs directly with outcomes for both the intervention and comparator of interest were excluded. A full list of studies excluded after full-text review is available in the Supplemental Digital Content, available at http://links.lww.com/BRS/A896. We also reviewed the references of all selected studies to identify any additional studies that were not found by our systematic literature search.
Data related to study characteristics, patient demographics, treatments and clinical data, and information on economic modeling parameters and primary study findings were abstracted (Supplemental Digital Content, available at http://links.lww.com/BRS/A896). Descriptive data and economic findings were reported as presented in the articles.
We performed all analyses on an individual study level because data pooling was not possible as a result of heterogeneity and inconsistency in reporting between studies.
The Quality of Health Economic Studies (QHES) instrument developed by Ofman et al9 was used to provide an initial basis for critical appraisal of included economic studies and is described in the Supplemental Digital Content, available at http://links.lww.com/BRS/A896. Factors important in critical appraisal of studies from an epidemiological perspective were also considered. Two reviewers independently applied the QHES to included studies. Discrepancies in ratings were discussed so that consensus could be reached and a final score obtained.
The search strategy is outlined in Figure 1. Out of an initial 88 citations that were identified, 10 studies were excluded after full-text review (Supplemental Digital Content, available at http://links.lww.com/BRS/A896). A total of 6 economic analyses10–15 were included, of which 5 addressed key question 1,10–14 1 addressed key question 2,15 and none addressed key question 3. A summary of included studies is provided in Table 2 and results in Table 3. QHES ratings were in the range from 68 to 88, with a possible maximum score of 100. Two studies (AHRQ10; Angevine et al15) received “high” scores and the remaining 4 studies11–14 received “fair” QHES scores (i.e., scores fall between 75–100 and 50–74, respectively16). QHES scores and study limitations are included in Table 3 (Supplemental Digital Content, available at http://links.lww.com/BRS/A896).
Key Question 1: rhBMP-2 Compared With ICBG
Five studies were included that compared fusion with BMP-2 to fusion with autogenous ICBG in patients with degenerative disease of the lumbar spine.10–14 In all cases, a 2-year time horizon was used, and no discounting was performed.
Four studies10,11,13,14 assessed the cost-effectiveness of rhBMP-2 compared with ICBG for anterior lumbar interbody fusion in 279 patients with single-level degenerative disc disease and disabling symptoms of at least 6 months' duration. All relied on a single noninferiority randomized trial17 for clinical data that served as the pivotal trial for FDA approval of Medtronic Sofamor Danek Inc., (Memphis, TN) Infuse (rhBMP-2). Two studies (AHRQ10; Garrison et al13) relied solely on this RCT, one (Alt et al11) also used data from 2 other nonrandomized trials of the same grafts inserted laparoscopically, and one (Polly et al14) also used expert opinion.
Two studies (AHRQ10; Garrison et al13/NHS) undertook cost-utility analyses (CUA) from a payer perspective. Both derived utility estimates from unpublished preoperative and 6-month Short Form-36 data from the trial.
The AHRQ study10 employed stationary Markov models to accommodate changing health states over time in terms of fusion status, need for secondary intervention, and donor site pain (Table 2). The analysis used Centers for Medicare & Medicaid Services cost data and included costs associated with the initial hospitalization and secondary interventions reported in 2007 US dollars. In the base case analysis, which assumed that BMP-2 use incurred no additional cost, BMP-2 was the dominant treatment strategy at 2 years (Table 3). The higher QALY for the rhBMP-2 group was attributed to donor site pain in the control group. RhBMP-2 remained the dominant strategy after 1-way sensitivity analyses were performed in which the risks of fusion and secondary interventions were varied. If rhBMP-2 was assumed to cost an additional $3000 (which accounts for approximately 10% of the mean diagnosis-related group cost for the index procedure), it was no longer the dominant treatment.
The National Health Service study used Monte Carlo simulations and clinical data included surgery parameters, reoperation rates, return-to-work, and fusion (Table 2). Costs were reported in pounds sterling; the year was not reported. Cost of treatment, BMP, reoperation, operating time, hospital stay were included and obtained from a variety of sources. The analysis suggested that the yearly cost of BMP use was £1,325,461 (95% CI, £583,547–£2,192,916) more than that of ICBG. However, use of BMP resulted in shorter operating time and length of hospital stay, as well as reduced reoperation rates. Its use resulted in approximately 11 (95% CI, −30 to 56) additional QALYs. Thus, anterior lumbar interbody fusion with BMP-2 versus ICBG was associated with £120,390 per QALY gained, and thus is not likely to be cost-effective, depending on the willingness to pay threshold. No sensitivity analysis was performed.
Two studies (Alt et al,11 Polly et al14) undertook cost-effectiveness analyses. In addition to the single RCT described in the earlier text, Alt et al11 obtained clinical data from 2 prospective nonrandomized cohort studies in which rhBMP-2 or ICBG were implanted via laparoscopic surgery, whereas Polly et al14 also relied on medical literature and expert opinion.
Alt et al11 considered the financial impact of rhBMP-2 use in Germany, France, and the United Kingdom from a societal perspective. Outcome measures used in the analysis included need for secondary surgery and return-to-work. Costs included in the analysis were those of the initial treatment, secondary surgery, and cost of lost productivity and were reported in 2008 Euros. Compared with ICBG, rhBMP-2 use resulted in savings in each country as a result of decreased rates of secondary surgery; savings were in the range from €337 to €756 per patient. Because patients who received rhBMP-2 returned to work on average 43 days sooner than those who received ICBG, rhBMP-2 use was associated with €7052 to €7783 savings from prevented lost productivity. Overall, the savings from avoiding lost productivity accounted for 82% to 92% of the overall savings associated with rhBMP-2 use. The authors concluded that, from a societal perspective, use of rhBMP-2 resulted in savings over time that offset the higher upfront cost of rhBMP-2 use compared with ICBG.
The study by Polly et al14 was sponsored by Medtronic Sofamor Danek Inc.; the analytic perspective was not stated. Clinical outcomes included fusion, hospital parameters, need for blood transfusion, and donor site pain and complications. Costs were reported in 2001 US dollars, determined from a variety of sources, and included hospitalization costs and all costs accrued through 2 years after the operation. Cost saved per prevention of donor site complications was the only outcome reported that was not costing only. In terms of cost-effectiveness outcomes, the authors found that there was approximately $549 savings per patient in avoidance of donor site pain and complications, which accounts for the cost saved in pain, infection, hematoma, and iliac crest fracture, among others. The authors concluded that the upfront price of BMP was likely to be offset by reductions in medical resource use and follow-up care.
One study (Carreon et al12) performed a CUA of rhBMP-2 versus ICBG for single- or multilevel lumbar posterolateral lumbar fusion with rhBMP-2 or ICBG in patients older than 60 years. Both clinical and actual cost data were taken from a single RCT conducted by the same authors. Utility estimates for each of the health states were derived from Short Form-36 data from the trial. Total inpatient and outpatient costs were included. A decision tree was used, and risks of complications, additional treatments, and revision surgery were included. A utility value and cost were assigned to each pathway throughout the decision tree. The analysis suggested that use of rhBMP-2 was associated with an overall savings of $2319 and improved utility score of 0.01 compared with ICBG. Thus, in this scenario, rhBMP-2 was more cost-effective than ICBG. The authors noted that there were no graft site complications, but a greater number of overall complications and persistent symptoms in the ICBG group. Furthermore, the costs associated with additional treatments, revision surgery, and complications were higher in the ICBG group, though the reasons for this were not explained. As a result, the higher upfront cost associated with rhBMP-2 use was offset by the higher cost of treating complications and symptoms that occurred in the ICBG group.
Key Question 2: Allograft or Bone Graft Extenders Compared With ICBG
One study15 was identified that compared fusion with allograft to fusion with autogenous ICBG in patients with degenerative disease of the cervical spine. No studies were identified that compared other bone graft extenders with ICBG for spinal fusion surgery.
Angevine et al15 performed a CUA of use of allograft compared with autograft (suggested to be ICBG) for single-level ACDF for patients with cervical spondylosis. The study was performed from a societal perspective using a 5-year time horizon.
Clinical data were obtained from literature and were based on 2 cohort studies and a variety of case series. In some cases, outcome rates were determined directly from the literature, and in other cases, assumptions were made on the basis of 1 or more studies. Clinical outcomes included clinical improvement after the index procedure and reoperation, time to return to work, symptomatic worsening, complications, reoperation, and adjacent segment disease. The study used decision tree analysis for outcomes in the first year and Markov models for outcomes between 1 and 5 years. Health state utilities were derived from Short Form-36 scores reported in a case series of 28 patients that underwent ACDF, and values were assigned to 3 states: preoperative (0.81), improved (1.0), and not improved (0.81). Sensitivity analysis was performed by varying utility values and discount rate. Costs were obtained from a retrospective analysis of 78 patients with cervical disc disease who received single-level ACDF with allograft or allograft with plating as well as from previously published costing studies. Costs accrued after hospital discharge were not included (Table 2). Costs were discounted at 3% per year. Compared with ACDF with ICBG, ACDF with allograft cost $60 more and resulted in an increase in QALY of 0.121. The corresponding incremental cost-effectiveness ratio was $496, suggesting that the 2 treatments are similarly cost-effective. Sensitivity analysis of utility values, postoperative recovery periods, and complication rates yielded similar incremental cost-effectiveness ratios. The authors concluded that ACDF with autograft has a similar cost-effectiveness ratio as ACDF with allograft.
No studies met the inclusion criteria to answer key question 3: allograft bone versus polyetheretherketone in ACDF.
Biological substitutes and extenders have become an integral part of spinal arthrodesis as both patient and surgeon preferences have led to a decrease in prevalence of iliac crest bone graft (ICBG) procedures. The rising rates of spine fusion surgery, the rising popularity of minimally invasive approaches dependent upon effective graft substitutes, and the technological advances that have led to improvement in fusion rates has brought the question of cost-effectiveness of graft choices to a priority status.1–3 Because there are many different regulatory pathways that govern the approval of use of these products, a wide variety of quantity and quality of data exists. Although the highest level of data is available for review in recombinant proteins, significant controversy has developed regarding the use of rhBMP-2, which acts a potent molecule of osteoinduction because it has been associated with significant complications.4,5 The use of BMP has become a contentious topic within the spine surgery community because of its cost, widespread off-label use, surgical risks, potential author bias from conflict-of-interest issues, and disagreement regarding clinical indications.18–21 Although rhBMP-2 has been widely used in spinal surgical procedures since 2003, a critical review of the cost-based literature with Infuse had not been performed. This dearth of data regarding efficacy and cost-effectiveness makes decision making for practitioners, administrators, and payers difficult. A systematic review of the literature evaluating the value of the use of biologics in spine fusion is useful in guiding an informed choice based upon cost-effectiveness.
Among the 5 studies that met the inclusion criteria to compare cost-effectiveness of rhBMP-2 versus ICBG, discordant conclusions were reached depending on the type of data used, cost-measurement methods, and study design. For example, the National Health Service study13 used cost of treatment and hospitalization data from the United Kingdom and concluded that rhBMP-2 was not cost-effective. Conversely, Alt et al11 reported data including return-to-work parameters from 3 different European countries and concluded that the increased loss of productivity seen from the ICBG group resulted in a savings with use of BMP per patient. Finally, the AHRQ study10 used Centers for Medicare & Medicaid Services cost data for initial and secondary interventions and concluded that rhBMP-2 would be cost-effective compared with ICBG if there was no additional cost for the product; however, at a price of $3000, there was no significant difference.
Conflicts of interest can introduce potential bias into the presentation of data of rhBMP-2 and conclusions derived thereof.18 In at least 2 of the studies12,14 included in our systematic review the authors had a financial conflict of interest with the parent company (Medtronic Sofamor Danek Inc.). Although the mere existence of such a conflict of interest does not discount the conclusions from the study, these relationships must be considered in the interpretation of the data. Although Rodgers et al22 concluded that the published industry funded study outcome data regarding rhBMP-2 did not differ significantly from confidential information, other studies have reported a significantly higher rate of favorable treatment outcomes when industry is involved.23,24
The systematic review of key question 1 demonstrated that rhBMP-2 likely does not present a cost-effective option over ICBG for routine spinal arthrodesis from a payer perspective, especially if the upfront costs for the product are high. However, if societal costs are considered including those from lost wages/productivity, then rhBMP-2 may be a more cost-effective option. In any case, instead of routine use, many authors have proposed that rhBMP-2 may offer the greatest benefit-risk ratio for patients who are at risk for a nonunion, undergo minimally invasive surgery, or who have had previous ICBG harvest.20 This particular patient population may be that which demonstrates the greatest cost-benefit ratio, but unfortunately was not analyzed in the studies that met the inclusion criteria for our systematic review. Because the market for rhBMP-2 varies depending on region and country, the price point for this product will be the determining factor in its cost-effectiveness.
One study met the inclusion criteria for key question 2. Angevine et al15 published a high-quality CUA that demonstrated similar cost-effectiveness ratios with the use of ICBG versus allograft in anterior cervical fusions. Although these data support the use of allograft in this setting, our question regarding the cost-effectiveness of other bone graft extenders such as demineralized bone matrix, synthetic carriers, and stem cell products could not be answered with the limited data available. Furthermore, because of the increasing clinical use of these products with local bone graft in lumbar fusion, future studies in this realm would be valuable.
This systematic review highlights the need for additional high-quality studies to address the cost-effectiveness of biological substitutes and extenders compared with ICBG in spinal arthrodesis. Ideally, such studies should be conducted prospectively and from a societal perspective, provide adequate time horizons and reliable measurements of costs, and use well-established economic models and sensitivity analyses. Although the spine community has gradually shifted away from the use of ICBG, only a handful of studies have conducted full formal economic analyses evaluating the cost-effectiveness of graft alternatives over ICBG. These studies will be critical for determining the future role that biological substitutes and graft extenders play in spinal arthrodesis because health care systems must make difficult decisions regarding inventory, resource utilization, and maintenance of quality patient care to its constituents.
- From a third-party payer perspective, the cost-effectiveness ratio of rhBMP-2 over ICBG in routine spinal arthrodesis is likely dependent upon its upfront costs that vary according to country.
- From a societal perspective, the use of rhBMP-2 may be more cost-effective than ICBG in spinal arthrodesis if lost productivity/wages are incorporated in the data analysis.
- Because of limited economic data regarding the use of biological substitutes/extenders in spinal arthrodesis, high-quality, prospective, cost analyses are necessary before definitive conclusions can be made regarding treatment options.
- A systematic review was conducted to evaluate the cost-effectiveness of lumbar or cervical spinal arthrodesis using rhBMP-2, allograft, or bone graft extenders compared with iliac crest autograft for the treatment of degenerative spinal condition.
- Regarding the cost-effectiveness of rhBMP-2 compared with iliac crest autograft in the lumbar spine, 5 studies provided discordant conclusions that varied with type of data used, cost-measurement methods, and study design. Whether rhBMP-2 is more cost-effective than autograft depends on the upfront cost of the bone graft and whether a payer or societal perspective is used.
- Regarding the cost-effectiveness of allograft to autograft in the cervical spine, one study suggested that, from a societal perspective, ACDF with allograft is similarly cost-effective as ACDF with autograft.
- No studies were identified that evaluated the cost-effectiveness of other bone graft extenders compared with iliac crest autograft.
- No studies were identified that evaluated use of polyetheretherketone cages with bone graft substitutes compared with allograft in ACDF.
The authors thank Erika Brodt for her assistance with data abstraction, Andrea Skelly for her advisory assistance, and Nancy Holmes and Chi Lam for their administrative assistance.
Supplemental digital content is available for this article. Direct URL citations appearing in the printed text are provided in the HTML and PDF version of this article on the journal's web site (www.spinejournal.com).
1. Banwart JC, Asher MA, Hassanein RS. Iliac crest bone graft
harvest donor site morbidity. A statistical evaluation. Spine (Phila Pa 1976) 1995;20:1055–60.
2. Goulet JA, Senunas LE, DeSilva GL, et al. Autogenous iliac crest bone graft
. Complications and functional assessment. Clin Orthop Relat Res 1997;339:76–81.
3. Hsu WK, Polavarapu M, Riaz R, et al. Nanocomposite therapy as a more efficacious and less inflammatory alternative to bone morphogenetic protein
-2 in a rodent arthrodesis model. J Orthop Res 2011;29:1812–9.
4. Vaidya R, Sethi A, Bartol S, et al. Complications in the use of rhBMP-2
cages for interbody spinal fusions. J Spinal Disord Tech 2008;21:557–62.
5. Haid RW Jr, Branch CL Jr, Alexander JT, et al. Posterior lumbar interbody fusion using recombinant human bone morphogenetic protein
type 2 with cylindrical interbody cages. Spine J 2004;4:527–38; discussion 38–9.
6. Cahill KS, Chi JH, Day A, et al. Prevalence, complications, and hospital charges associated with use of bone-morphogenetic proteins in spinal fusion procedures. JAMA 2009;302:58–66.
7. Deyo RA, Ching A, Matsen L, et al. Use of bone morphogenetic proteins in spinal fusion surgery for older adults with lumbar stenosis: trends, complications, repeat surgery, and charges. Spine 2012;37:222–30.
8. Glassman SD, Carreon LY, Campbell MJ, et al. The perioperative cost of Infuse bone graft in posterolateral lumbar spine fusion. Spine J 2008;8:443–8.
9. Ofman JJ, Sullivan SD, Neumann PJ, et al. Examining the value and quality of health economic analyses: implications of utilizing the QHES. J Manag Care Pharm 2003;9:53–61.
10. Agency for Healthcare Research and Quality. Bone Morphogenetic Protein
: the State of the Evidence of On-Label and Off-Label Use. Technology Assessment Report. Rockville, MD: Agency for Healthcare Research and Quality; 2010.
11. Alt V, Chhabra A, Franke J, et al. An economic analysis of using rhBMP-2
for lumbar fusion in Germany, France, and UK from a societal perspective. Eur Spine J 2009;18:800–6.
12. Carreon LY, Glassman SD, Djurasovic M, et al. RhBMP-2 versus iliac crest bone graft
for lumbar spine fusion in patients over 60 years of age: a cost-utility
study. Spine (Phila Pa 1976) 2009;34:238–43.
13. Garrison KR, Donell S, Ryder J, et al. Clinical effectiveness and cost-effectiveness
of bone morphogenetic proteins in the non-healing of fractures and spinal fusion: a systematic review. Health Technol Assess 2007;11:1–150; iii–iv.
14. Polly DW Jr, Ackerman SJ, Shaffrey CI, et al. A cost analysis of bone morphogenetic protein versus
autogenous iliac crest bone graft
in single-level anterior lumbar fusion. Orthopedics 2003;26:1027–37.
15. Angevine PD, Zivin JG, McCormick PC. Cost-effectiveness
of single-level anterior cervical discectomy and fusion
for cervical spondylosis. Spine (Phila Pa 1976) 2005;30:1989–97.
16. Spiegel BM, Targownik LE, Kanwal F, et al. The quality of published health economic analyses in digestive diseases: a systematic review and quantitative appraisal. Gastroenterology 2004;127:403–11.
17. Burkus JK, Gornet MF, Dickman CA, et al. Anterior lumbar interbody fusion using rhBMP-2
with tapered interbody cages. J Spinal Disord Tech 2002;15:337–49.
18. Carragee EJ, Hurwitz EL, Weiner BK. A critical review of recombinant human bone morphogenetic protein
-2 trials in spinal surgery: emerging safety concerns and lessons learned. Spine J 2011;11:471–91.
19. Carragee EJ, Mitsunaga KA, Hurwitz EL, et al. Retrograde ejaculation after anterior lumbar interbody fusion using rhBMP-2
: a cohort controlled study. Spine J 2011;11:511–6.
20. Resnick D, Bozic KJ. Meta-analysis of trials of recombinant human bone morphogenetic protein
-2: what should spine surgeons and their patients do with this information? Ann Intern Med 2013;158:912–3.
21. Fu R, Selph S, McDonagh M, et al. Effectiveness and harms of recombinant human bone morphogenetic protein
-2 in spine fusion: a systematic review and meta-analysis. Ann Intern Med 2013;158:890–902.
22. Rodgers MA, Brown JV, Heirs MK, et al. Reporting of industry funded study outcome data: comparison of confidential and published data on the safety and effectiveness of rhBMP-2
for spinal fusion. BMJ 2013;346:f3981.
23. Bailey CS, Fehlings MG, Rampersaud YR, et al. Industry and evidence-based medicine: believable or conflicted? A systematic review of the surgical literature. Can J Surg 2011;54:321–6.
24. Khan SN, Mermer MJ, Myers E, et al. The roles of funding source, clinical trial outcome, and quality of reporting in orthopedic surgery literature. Am J Orthop (Belle Mead NJ) 2008;37:E205–12; discussion E12.