Cervical degenerative disc disease (DDD) can manifest as neck pain, cervical radiculopathy, and/or myelopathy. Although most can be managed nonoperatively, surgical intervention may be required for persistent or progressive symptoms. Currently, anterior cervical discectomy and fusion (ACDF) remains the most common surgical intervention for 1- or 2-level radiculopathy or myelopathy.1,2 Despite the proven long-term clinical and radiographical success of ACDF, enthusiasm has grown for nonfusion alternatives including posterior cervical foraminotomy (PCF) for radiculopathy, or cervical disc arthroplasty (CDR), laminectomy with fusion, and laminoplasty for radiculopathy or myelopathy.1,2 The major, but contentious, argument for motion preservation is that the latter surgical procedures will prevent, or lessen, the incidence of adjacent segment degeneration and disease compared with ACDF. Both CDR and PCF have been shown to produce equivalent, and in some cases superior, clinical outcomes in patients with radiculopathy with the potential of avoiding the complications specific to ACDF (e.g., recurrent laryngeal nerve palsy, dysphagia, loss of range of motion, and adjacent segment disease).1–3 The same has been shown for laminoplasty versus ACDF for patients with myelopathy.4 Although comparisons on outcomes and complication rates among these procedures have been plentiful, comparisons on cost-effectiveness have been sparse.
The primary goal of this review was to perform an evidence-based synthesis of the literature, assessing the cost-effectiveness of surgical intervention for syndromes associated with cervical DDD. To accomplish this goal, we sought to answer the following key questions:
- Is there evidence to suggest that surgical intervention is cost-effective compared with nonoperative treatment of cervical myelopathy or radiculopathy?
- Is there evidence to suggest that ACDF is cost-effective compared with CDR for cervical myelopathy or radiculopathy?
- Is there evidence to suggest that anterior surgical procedures are cost-effective compared with posterior surgical procedures for cervical myelopathy?
- Is there evidence to suggest that anterior surgical procedures are cost-effective compared with posterior surgical procedures for cervical radiculopathy?
MATERIALS AND METHODS
General methods used for this systematic review are described in greater detail in the Supplemental Digital Content available at http://links.lww.com/BRS/A893 specific to this focus issue. Additional topic-specific methods are described below.
Electronic Literature Search
We conducted a systematic search of PubMed/MEDLINE, EMBASE, the Cochrane Collaboration Library, the Tufts Cost-Effectiveness Analysis registry, and the National Health Service Economic Evaluation Database for full economic evaluations published through January 16, 2014 based on the key questions and Population, Intervention, Comparator, Outcomes inclusion/exclusion criteria established a priori (Table 1). The search strategy is further documented in the Supplemental Digital Content available at http://links.lww.com/BRS/A893. Only economic evaluations that evaluated and synthesized the costs and consequences of surgical procedures or surgical intervention with nonsurgical management in patients with cervical degenerative disease were considered for inclusion. Discrepancies were discussed so that consensus could be reached regarding final inclusion of studies.
Data Extraction and Analysis
Data related to patient population characteristics, treatments, information on economic modeling parameters, and primary study findings, including those from sensitivity analyses, were abstracted (Supplemental Digital Content available at http://links.lww.com/BRS/A893). Descriptive data and economic findings were reported as presented in the articles.
The Quality of Health Economic Studies (QHES) instrument developed by Ofman et al5 was used to provide an initial basis for critical appraisal of the included economic studies and is described in the Materials and Methods section of this focus issue. Factors important in critical appraisal of studies from an epidemiological perspective were also considered. Two reviewers independently applied the QHES to the included studies. Discrepancies in ratings were discussed so that consensus could be reached and a final score obtained.
The literature search yielded 250 unique, potentially relevant citations (Figure 1). Because the key questions were closely related, 1 literature search addressed all key questions. Two of the study authors (D.J.F., D.C.N.) independently considered studies for inclusion and discrepancies in selection were resolved by discussion. After title/abstract review, 228 articles were excluded, the majority of which did not assess the population or comparators of interest or were not economic studies. Among 22 full-text articles reviewed, 17 were excluded, the primary reason being that studies did not report the outcomes of interest. The details of the excluded studies are listed in the Supplemental Digital Content available at http://links.lww.com/BRS/A893.
A total of 5 articles were retained that together addressed 3 of our 4 key questions (Tables 2, 3). Three cost-utility studies2,6,7 and 2 cost-effectiveness studies were included1,3. Additional detail about the included studies, including study limitations, is reported in Table 3 and the Supplemental Digital Content available at http://links.lww.com/BRS/A893.
KQ1: Is there evidence to suggest that surgical intervention is cost-effective compared with nonoperative treatment of cervical myelopathy or radiculopathy?
No full economic evaluations existed relative to this question.
KQ2: Is there evidence to suggest that ACDF is cost-effective compared with CDR for cervical myelopathy or radiculopathy?
Two economic evaluations compared the cost-effectiveness of ACDF with CDR for cervical myelopathy or radiculopathy (Tables 2, 3).1,2
Menzin et al1 performed a cost-benefit assessment (type of cost-effectiveness study), the focus of which was to evaluate the difference between incremental medical costs and gains in work productivity when comparing CDR with ACDF in patients with single-level disc disease and radiculopathy or myelopathy. In a randomized controlled trial, 541 patients were assigned to treatment with either CDR (n = 276) or ACDF (n = 265) and followed for 2 years. Electronic billing data available for a portion of the clinical trial population (n = 68; 12.6%) were used to determine inpatient facility costs.
Compared with ACDF, the mean total direct medical cost per patient during 2 years was $431 (standard error: $247) lower for the CDR group ($11,472 vs. ACDF $11,903). Furthermore, patients with CDR worked an average of 38 more days during follow-up. Patients made the decision to return to work, but it was not mentioned how that decision was reached or if collars were mandatory. The mean value of work productivity (estimated as average daily wage rate) was $6547 higher for the CDR cohort (CDR, $56,998; ACDF, $50,452). Taking into account the increased medical cost associated with CDR, CDR had a net benefit of $6978 per patient during 2 years postsurgery (CDR, $43,974; ACDF, $37,175).
Qureshi et al2 performed a cost-effectiveness study comparing CDR with ACDF using decision analysis. They used 3% discounting for a target population of patients aged 45 years who had single-level cervical disc disease with radiculopathy. Primary complications of ACDF were determined from the literature and applied to the model at 5% per year for pseudarthrosis and hardware failure, and 3% per year for adjacent segment degeneration. For CDR, hardware failure was the primary complication reported in the literature and was applied at a rate of 1.5% per year. The base disease state of cervical disc disease was assigned a utility value of 0.7 based upon literature for other diseases (osteoarthritis, neck pain). Utility values were ACDF (0.8), CDR (0.9), revision ACDF (0.75), and revision CDR (0.85).
CDR and ACDF generated a total lifetime (20-yr) cost of $11,987 and $16,823, respectively. These costs were estimates from the Nationwide Inpatient Sample database. Consequently, CDR had a net incremental lifetime savings of $4836. CDR resulted in a generation of 3.94 QALYs, whereas ACDF resulted in 1.92. The cost-effectiveness ratio associated with CDR was $3042/QALY versus $8760/QALY for ACDF, indicating that CDR dominates ACDF (CDR is less costly and more effective than ACDF). The model assumed that CDR prosthesis would survive 20 years. One-way sensitivity analyses were performed on variables in which the data used were based on imperfect information (i.e., assumptions, estimates from the literature) or lack of evidence-based literature.
KQ3: Is there evidence to suggest that anterior surgical procedures are cost-effective compared with posterior surgical procedures for cervical myelopathy?
Two economic evaluations used data from a prospective cohort study to compare the cost-effectiveness of anterior with posterior surgical procedures for treatment of cervical spondylotic myelopathy (CSM; Tables 2, 3).6,7 In both studies, the study design and patient population were similar; however, the economic data and analyses were performed differently. Patients aged 40 to 85 years with CSM and cervical spinal cord compression at 2 or more levels from degenerative spondylosis were eligible for study participation.
Ghogawala et al6 reported that all patients underwent preoperative radiographs to determine whether an anterior or posterior approach should be used. Fifty patients underwent anterior multilevel discectomy with fusion and plating (n = 28) or posterior midline cervical laminectomy with lateral mass screws and rods for rigid fixation (n = 22) at the discretion of the surgeon. An economic analysis was conducted on a subset of 41 patients (82%) for whom both baseline and 1-year EuroQol-5D (EQ-5D), as well as hospital charge data, were available. The economic analysis did not include costs associated with postoperative care, productivity loss, or subsequent hospitalizations. The mean unadjusted hospital cost for an initial posterior fusion was significantly higher than that for an initial anterior surgery with fusion (posterior, $29,465; anterior, $19,245). At 1 year, the EQ-5D was similar in those undergoing anterior surgery compared with posterior surgery (anterior, 0.16; posterior, 0.13; P>0.05). An incremental cost-effectiveness ratio (ICER) was not calculated.
Whitmore et al7 performed a cost-effectiveness analysis using data from 72 of 85 patients with CSM who underwent anterior (n = 41) or posterior (n = 31) surgery. Results were based on 1-year SF-36 Physical Component Summary and EQ-5D data. Economic data on each patient were captured prospectively. Data on work status, productivity loss, and costs of postoperative care, complications, or subsequent hospitalizations were not obtained. Direct costs were calculated with the cost-to-charge ratio (CCR) and Medicare reimbursement methods.
With the CCR method, the anterior approach was less costly than the posterior approach (anterior, $21,563 ± 8721; posterior, $27,942 ± 14,220; P = 0.02). Using the Medicare reimbursement method, it was found that there was no significant difference in costs (anterior, $17,538 ± 3115; posterior, $16,579 ± 4188; P = 0.29). This is an important point when interpreting cost-effectiveness studies: what method was used to calculate costs. Overall (both CCR and Medicare), patients who underwent anterior surgery had a greater, but nonsignificant, improvement in the change in SF-36 Physical Component Summary outcomes at 1 year compared with patients who underwent posterior surgery (anterior, 8.45 ± 9.57; posterior, 4.72 ± 10.33; P = 0.12). The change in EQ-5D at 1 year was similar between the 2 groups (anterior, 0.16 ± 0.19; posterior, 0.13 ± 0.20; P = 0.56). With the CCR method, the anterior approach dominated the posterior approach. Using the Medicare reimbursement method, the ICER for anterior surgery is $34,533 per QALY gained and is cost-effective compared with posterior surgery.
KQ4: Is there evidence to suggest that anterior surgical procedures are cost-effective compared with posterior surgical procedures for cervical radiculopathy?
One economic evaluation used data from a retrospective cohort study of 38 military patients with unilateral cervical radiculopathy to assess the cost-benefit of ACDF compared with PCF (Tables 2, 3).3 The 19 patients who underwent PCF were matched by age, treatment level, and surgeon with 19 patients who underwent ACDF from the corresponding author's institution.
The average time to return to unrestricted full duty was significantly less in the PCF group compared with ACDF group (PCF, 4.8 wk; ACDF, 19.6 wk; P < 0.0001). Eighteen of 19 patients who underwent ACDF returned to unrestricted full duty, whereas one returned to an administrative role (change from prior duties). In the PCF group, 18 of 19 patients returned to unrestricted full duty, and one was unable to return to full duty; the time to return to duty for this patient was excluded.
Total direct costs (i.e., hospital and instrumentation costs) in a 1-year period were $6508 less for PCF compared with ACDF (PCF, $3570; ACDF, $10,078). Indirect costs for the ranks of junior enlisted service member (E4), senior enlisted member (E7), junior officer (O1), and senior officer (O4) were as follows: $13,586; $17,797; $17,475; and $24,045, respectively. The total cost (direct and indirect) differences between procedures for an E4, E7, O1, and O4 service member were $20,094, $24,305, $23,983, and $30,553, respectively. PCF had a greater net economic benefit than ACDF because PCF was less costly and resulted in a shorter time to return to unrestricted full duty.
As the United States moves toward a value-based health care system, substantiation of treatments for various spinal conditions will become necessary. Value-based health care emphasizes identification of the optimal treatment option for a given pathology, where optimal is defined as the greatest gain in quality of life at the lowest cost. There have been few cost-effectiveness studies in the spine surgery literature, and even within the field, analyses of degenerative cervical conditions have been particularly neglected. In a literature review of all cost-utility analyses in spine care through 2010, Kepler et al8 identified 28 studies, of which only 4 (14%) involved the cervical spine. Similarly, the present review identified only 5 studies meeting the inclusion criteria. However, our goal in conducting this review was to identify what has been done and propose recommendations for future cost-effectiveness studies in this area.
There is a high degree of heterogeneity among the published studies on this topic. The studies identified in the present review differ with regard to quality, how costs are calculated, what perspective was used (i.e., societal/hospital/payer), as well as the study design. Most studies did not report the explicit methodology for modeling, and it was unclear what assumptions were made and what was the justification for the economic model chosen. The follow-up periods were limited in duration, and may not adequately represent the longer term costs and clinical outcomes associated with these procedures. For example, only 1 study2 conducted a sensitivity analysis and had a follow-up of greater than 2 years and performed proper discounting of costs per the recommendations of the 1996 Panel on cost-effectiveness in Health and Medicine.9 Scores on QHES in the range from 34 to 70, suggesting variation in the extent to which studies met quality standards. Other limitations are discussed in Tables 2 and 3 and the Supplemental Digital Content available at http://links.lww.com/BRS/A893.
Although we understand that there are certain clinical scenarios where a particular procedure is clearly the best choice, we are able to summarize what the included studies have shown. It must be recognized, however, that we think that each study had flaws in design that necessitate further research to validate the stated findings. First, CDR is more cost-effective than ACDF for patients with single-level cervical DDD and radiculopathy1,2 (Menzin et al1, 2 yr postsurgery; Qureshi et al, >10–11 yr postsurgery) or myelopathy (Menzin et al1, 2 yr postsurgery). Second, anterior surgery is more cost-effective than posterior surgery for patients with CSM at 1 year postsurgery.6,7 Finally, PCF is less costly than ACDF for patients with single-level cervical radiculopathy at 1-year postsurgery.3 Clearly, conclusions based on 1 or 2 studies need to be further validated by high-quality prospective investigations. For example, for anterior versus posterior surgery for CSM, studies commonly have different subsets of patients treated with anterior or posterior techniques. Specifically, anterior techniques are more commonly used in younger patients with more focal pathology, less neurological impairment, and less medical comorbidity. Conclusions on cost-effectiveness may be heavily influenced by patient demographics.10 In the study by Qureshi et al, the literature value assumptions for CDR and ACDF may not accurately represent true values. At the time of this study, health utility indices for patients with degenerative cervical spine conditions were not available, thus necessitating the need to make assumptions that affect the study findings. Since that time, Qureshi et al11 have published further literature defining health utility indices in the study population as well as long-term complications and their associated costs. Although the health utility indices and updated complication profile has been published, a repeat cost-effectiveness comparison is still lacking. The assumptions made by the authors regarding preoperative and postoperative health utility state and the lack of comprehensive cost analysis given the design of the study may have affected the value equation significantly. Further studies are necessary with the new information now available.
There was insufficient evidence to address our first key question because there were no full economic evaluations comparing surgical with conservative management for myelopathy or radiculopathy. Indeed, studies have been conducted that show that surgery is cost-effective for both CSM (e.g., Fehlings et al12) and radiculopathy (e.g., Carreon et al13). However, these studies were excluded from the present review because they did not involve comparisons between either 2 surgical procedures or surgery versus nonoperative management. Fehlings et al12 conducted a retrospective review (n = 70) on patients undergoing surgery for CSM. During 10 years of follow-up, patients gained a total of 0.64 QALYs with an average cost of Can$21,066, producing a favorable cost-utility ratio. Relative to the comparison between operative and nonoperative management for CSM, relatively few studies have been conducted on the topic. In 2013, Rhee et al14 performed a systematic review of nonoperative management for CSM. The authors identified only 5 studies and concluded that nonoperative management may yield equivalent outcomes to surgery for patients with mild myelopathy, but beyond mild symptoms, nonoperative management is inferior to surgery. Although future research may be helpful, it is unlikely to be undertaken because there is no longer therapeutic equipoise between operative and nonoperative approaches for CSM. Specially, Fehlings et al15 found that in the case of mild myelopathy, there is little evidence that nonoperative treatment may have a role; for moderate and severe myelopathy, nonoperative treatment results in outcomes inferior to those of surgery.
The primary goal of this review was to identify whether evidence exists regarding the cost-effectiveness of surgical interventions in patients with cervical DDD. Three of the 4 key questions were addressed, albeit by only 1 or 2 studies. This is not ideal given the high incidence of surgery for cervical DDD and continuing evolution of health care in this country. The few studies included suggest that CDR may be more cost-effective than ACDF for patients with cervical radiculopathy or myelopathy at 1-year follow-up, and that PCF may be more cost-effective than ACDF for radiculopathy at 2-year follow-up. However, the relatively low quality of evidence (low QHES scores) and the numerous study limitations necessitates further, more uniform, and longer follow-up analyses prior to making definitive conclusions and recommendations. These preliminary conclusions, however, do allow us to make several recommendations going forward. First, investigators wishing to conduct cost-effectiveness analyses on this topic should adopt a standardized cost-utility methodology for proper comparability and interpretation by policy makers and the public. This includes both comprehensive long-term follow-up costs and highly valid quality of life outcome questionnaire data. Second, these analyses should directly compare either 2 surgical interventions or surgical versus nonsurgical interventions using an ICER value and be specific about what indication patients are receiving treatment of (e.g., radiculopathy vs. myelopathy vs. neck pain alone). Third, other surgical interventions should be explored, including laminoplasty, anterior corpectomy, and hybrid approaches (i.e., combination ACDF/corpectomy), and different models of cervical disc prosthesis used in CDR. Finally, longer term follow-up is necessary for cost-effectiveness studies to define important aspects such as adjacent level surgery, failure rates, and clinical outcomes.
Cost-effectiveness analyses on surgical interventions for cervical DDD are lacking in the literature. Future research is necessary to validate the findings of the few analyses that do exist and to examine surgery versus nonoperative management in particular. Investigators seeking to perform such analyses should be transparent in their methodology and cost calculations with an ICER as the endpoint to properly compare interventions.
- Common surgical interventions for cervical spondylosis include ACDF, posterior foraminotomy, and CDR.
- Although many studies have been done comparing outcomes and complication rates between interventions, few have focused on costs or cost-effectiveness.
- In those studies that do focus on cost-effectiveness, CDR, ventral fusion, and foraminotomy are more cost-effective than ACDF, dorsal fusion, and ACDF, respectively.
- Future research is necessary to validate the findings of the few studies that do exist.
The authors thank Nancy Holmes and Ms 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. Menzin J, Zhang B, Zeumann PJ, et al. A health-economic assessment of cervical disc arthroplasty
compared with allograft fusion. Tech Orthop 2010;25:133–7.
2. Qureshi SA, McAnany S, Goz V, et al. Cost-effectiveness
analysis: comparing single-level cervical disc replacement and single-level anterior cervical discectomy and fusion: clinical article. J Neurosurg Spine 2013;19:546–54.
3. Tumialan LM, Ponton RP, Gluf WM. Management of unilateral cervical radiculopathy in the military: the cost-effectiveness
of posterior cervical foraminotomy compared with anterior cervical discectomy and fusion. Neurosurg Focus 2010;28:E17.
4. Liu T, Yang HL, Xu YZ, et al. ACDF
with the PCB cage-plate system versus
laminoplasty for multilevel cervical spondylotic myelopathy
. J Spinal Disord Tech 2011;24:213–20.
5. Ofman JJ, Sullivan SD, Neumann PJ, et al. Examining the value and quality of health economic analyses: implications of utilizing the QHES. JMCP 2003;9:53–61.
6. Ghogawala Z, Martin B, Benzel EC, et al. Comparative effectiveness of ventral vs
. dorsal surgery for cervical spondylotic myelopathy
. Neurosurgery 2011;68:622–30; discussion 30–1.
7. Whitmore RG, Schwartz JS, Simmons S, et al. Performing a cost analysis in spine outcomes research: comparing ventral and dorsal approaches for cervical spondylotic myelopathy
. Neurosurgery 2012;70:860–7; discussion 7.
8. Kepler CK, Wilkinson SM, Radcliff KE, et al. Cost-utility analysis in spine care: a systematic review. Spine J 2012;12:676–90.
9. Weinstein MC, Siegel JE, Gold MR, et al. Recommendations of the Panel on Cost-effectiveness
in Health and Medicine. JAMA 1996;276:1253–8.
10. Fehlings MG, Barry S, Kopjar B, et al. Anterior versus
posterior surgical approaches to treat cervical spondylotic myelopathy
: outcomes of the prospective multicenter AOSpine North America CSM study in 264 patients. Spine (Phila Pa 1976) 2013;38:2247–52.
11. Qureshi S, Goz V, McAnany S, et al. Health state utility of patients with single-level cervical degenerative disc disease: comparison of anterior cervical discectomy and fusion with cervical disc arthroplasty
. J Neurosurg Spine 2014;20:475–9.
12. Fehlings MG, Jha NK, Hewson SM, et al. Is surgery for cervical spondylotic myelopathy
cost-effective? A cost-utility analysis based on data from the AOSpine North America prospective CSM study. J Neurosurg Spine 2012;17:89–93.
13. Carreon LY, Anderson PA, Traynelis VC, et al. Cost-effectiveness
of single-level anterior cervical discectomy and fusion five years after surgery. Spine (Phila Pa 1976) 2013;38:471–5.
14. Rhee JM, Shamji MF, Erwin WM, et al. Nonoperative management of cervical myelopathy: a systematic review. Spine (Phila Pa 1976) 2013;38:S55–67.
15. Fehlings MG, Wilson JR, Yoon ST, et al. Symptomatic progression of cervical myelopathy and the role of nonsurgical management: a consensus statement. Spine (Phila Pa 1976) 2013;38:S19–20.