Cost-effectiveness of anterior surgical decompression surgery for cervical degenerative disk disease: a systematic review of economic evaluations

No clear consensus exists on which anterior surgical technique is most cost-effective for treating cervical degenerative disk disease (CDDD). One of the most common treatment options is anterior cervical discectomy with fusion (ACDF). Anterior cervical discectomy with arthroplasty (ACDA) was developed in an effort to reduce the incidence of clinical adjacent segment pathology and associated additional surgeries by preserving motion. This systematic review aims to evaluate the evidence regarding the cost-effectiveness of anterior surgical decompression techniques used to treat radiculopathy and/or myelopathy caused by CDDD. The search was conducted in PubMed, EMBASE, Web of Science, CINAHL, EconLit, NHS-EED and the Cochrane Library. Studies were included if healthcare costs and utility or effectivity measurements were mentioned. A total of 23 studies were included out of the 1327 identified studies. In 9 of the 13 studies directly comparing ACDA and ACDF, ACDA was the most cost-effective technique, with an incremental cost effectiveness ratio ranging from $2.900/QALY to $98.475/QALY. There was great heterogeneity between the costs of due to different in- and exclusion criteria of costs and charges, cost perspective, baseline characteristics, and calculation methods. The methodological quality of the included studies was moderate. The majority of studies report ACDA to be a more cost-effective technique in comparison with ACDF. The lack of uniform literature impedes any solid conclusions to be drawn. There is a need for high-quality cost-effectiveness research and uniformity in the conduct, design and reporting of economic evaluations concerning the treatment of CDDD. PROSPERO Registration: CRD42020207553 (04.10.2020).


Introduction
No consensus exists on which anterior surgical technique is more cost-effective to treat cervical degenerative disk disease (CDDD) resulting in cervical degenerative radiculopathy and/or myelopathy.
The incidence of degenerative pathologies is significantly increasing as the population ages [1,2]. Generalized spinal disk degeneration occurs in more than 90% of adults past the 5th decade of life [3]. This age group now represents 32.8% of the population in Europe and is projected to reach 40.6% by 2050 [2]. Complaints of radiculopathy and/or myelopathy differ in severity, but are often disabling. This leads to restrictions in daily life and loss of professional capability with absenteeism as a result, leading to an increase in societal costs. Healthcare costs are driven up further when patients require surgical treatment, in combination with associated hospitalization and rehabilitation.
One of the most common procedures for treating patients with radiculopathy and/or myelopathy as a consequence of single-or multilevel CDDD is anterior cervical discectomy with fusion (ACDF) or without (ACD) [4]. A recent study by Neifert et al. predicts a significant increase in ACDFprocedures in those aged [45][46][47][48][49][50][51][52][53][54] in the next 20 years which involves the working population [5][6][7]. Both techniques show good short-term clinical results [8]. However, patientreported satisfaction gradually drops to 68-96% after 7-20 years [9,10]. This is thought to be the consequence of the development of new complaints of radiculopathy and/or myelopathy at a segment adjacent to the site of previous surgery, also known as clinical adjacent segment pathology (CASP) [11]. The underlying mechanism is thought to be compensation for the loss of motion in the fused segment, resulting in overstraining of the adjacent segments in addition to progression of natural degeneration [12][13][14]. This occurs at an estimated cumulative rate of 1.6 to 4.2% per year after fusion surgery. Patients with CASP often require additional surgery, which drives up health-care costs even further [11,15,16].
Anterior cervical discectomy with arthroplasty (ACDA) was developed in an effort to reduce the incidence of CASP by preserving motion in the operated segment(s). Previously conducted research has shown that no significant differences in clinical or radiological outcomes among these different techniques [12,17,18]. Multiple trials with long-term follow-up report significantly lower additional surgery rates for ACDA in comparison with ACDF, both for single-and multilevel surgeries [19][20][21][22][23]. Proper interpretation remains difficult however, as reported additional surgery rates vary strongly, from 0 to 11.6% for ACDA, and 0 to 18.2% for ACDF in 2-7 years follow-up [24][25][26].
Presently, ACDA is often discouraged as the implant costs are higher than those for ACDF while the clinical outcomes are similar. However, preventing additional surgeries due to CASP and new complaints might be a reason for ACDA to be a more cost-effective technique in the long run.
There are high quality systematic reviews available concerning the clinical effectiveness of these different anterior decompression surgeries for CDDD [27,28]. To date, only one systematic review, dating from 2014, and one narrative review have been conducted concerning the cost-effectiveness of these techniques. Conclusive evidence is therefore still lacking [29,30]. This systematic review aims to evaluate the current evidence on cost-effectiveness of anterior cervical decompressive surgeries in patients with CDDD resulting in radiculopathy and/or myelopathy. The methodological quality of the included studies will be determined to be able to draw conclusions concerning the level of evidence.

Review protocol
This systematic review was executed in accordance with the PRISMA statement [31,32] and the five-step approach on preparing a systematic review of economic evaluations by Van Mastrigt et al. [33][34][35]. The study protocol has been published in the PROSPERO-database before commencing the study (registration number CRD42020207553).

Search strategy and eligibility criteria
Searches for clinical and economic studies were systematically carried out in PubMed, EMBASE (OVID), Web of Science (Clarivate Analytics), CINAHL (EBSCO), Econlit (EBSCO), NHS EED (CRD) and the Cochrane Library (CENTRAL) and Cochrane database of Systematic 1 3 Reviews. The search was conducted without using filters. The final search was conducted on September 16th, 2020 (Appendix File 1).
Studies were included if they met all of the following eligibility criteria: (1) CDDD resulting in radiculopathy and/or myelopathy, (2) anterior decompressive surgery (3) cost sources, health care and societal perspective costs, total costs (health care perspective costs plus societal perspective costs), costs per quality-adjusted life years (QALY) (4) utility measurement tool, gained qualityadjusted life years.

Study selection and data collection
Duplicates were removed, potential studies were screened on title and abstract, and full texts were assessed by two independent researchers (VS, XOZ) using all eligibility criteria. This assessment was done in "Rayyan" [36]. If necessary, consensus was reached between both authors through discussion.
Randomized controlled trials (RCTs), prospective comparative studies concerning cost-effectiveness and both full and partial economic evaluations were included. Cost-utility analyses (CUA) and cost-effectiveness analyses (CEA) were also included if they had a retrospective study design.
The following data were extracted from the included articles; study design, year of publication, country, baseline characteristics of patients, diagnosis, type of interventions, primary and secondary outcome measures as described above. The complete data collection sheet can be found in Appendix File 2. Cost-effectiveness of two interventions can be compared using the incremental cost-effectiveness ratio (ICER), where the difference in costs between two interventions will be divided by the difference in their effect [37].

Quality assessment
The methodological quality of the economic evaluations was assessed by two independent reviewers (VS, OZ) with the checklist of Consensus on Health Economic Criteria (CHEC) by Evers et al. [38]. The CHEC-list consists of 19 categories, including performed discounting and sensitivity analysis, where a single point can be assigned to each category with a maximum score of 19 points. A CHEC-list score between 15 and 17 points is rated as average methodological quality [38]. Risk of bias was assessed by two independent reviewers (VS, SM). RCTs were assessed with the Cochrane Risk of Bias tool in Review Manager 5.3 [39]. Non-randomized controlled trials were be assessed with the Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) tool [40]. Full risk of bias assessment sheets and CHEC list scores are outlined in Appendix File 3 and Appendix File 4.

Analysis
All costs were converted to American Dollars to the reference year 2020 with the use of a web-based tool, developed by the Campbell and Cochrane Economics Methods Group (CCEMG) and the Evidence for Policy and Practice Information and Coordinating Centre (EPPI-Centre) (v.1.6) [41]. If the index year was not mentioned in the study, subsequently the last year of patient inclusion or the year of publication was used as index year. Heterogeneity was assessed with the Chi-squared test in Review Manager 5.3 [42]. Descriptive statistics were used and reported in a narrative summary according to the Cochrane handbook [39]. Sensitivity analyses of the different study designs were checked for consistency of results.

Study selection
The systematic search in the databases yielded 1948 articles, 1327 of which remained after removal of duplicates. Of the 28 studies eligible for inclusion, 6 studies were based on data of overlapping patient cohorts. Of these, the most recent studies with the largest patient cohort and the longest follow-up time were included. One study was included through cross-referencing [43], resulting in a total of 23 studies . A flowchart of the selection process is presented in Fig. 1.

Study characteristics
Characteristics and results of the included studies are displayed in Table 1. Chi-squared was larger than 99%, indicating a very high heterogeneity between studies, and a metaanalysis was therefore not performed. Publication years ranged from 2005 [52] to 2019 [45]. Follow-up time ranged from 1 [48,49,58] to 10 years post-operative [55]. Probabilistic models simulated 5 years [45,56] to life-long follow-up [51]. Twenty studies were performed in the United States of America, one in Canada [50], one in Korea [45] and one in Spain [53]. Three prospective economic evaluations [48,58,66] were included; all others were retrospective CEAs or CUAs. In a total of 5 studies, a Markov Analysis was performed [44,45,47,51,56].

Quality of identified studies
The methodological quality of each economic evaluation was critically assessed according to the CHEC-criteria, which can be found in Appendix File 3. CHEC scores ranged from 8 to 18, with a mean score of 14.2. This indicates an overall moderate methodological quality.
The risk of bias assessment can be found in Appendix File 4. Based on the criteria for randomized studies overall risk of bias was high. According to the ROBINS-I tool, the overall quality of non-randomized studies was moderate.

Discussion
The objective of this systematic review was to evaluate the current evidence on cost-effectiveness of anterior cervical decompressive surgeries in patients with CDDD resulting in radiculopathy and/or myelopathy. The most important conclusion to be drawn from this review is the lack of literature on the subject and, more importantly, the large heterogeneity between studies in determining both costs and effectiveness. Solid conclusions can thus not be drawn regarding the most cost-effective technique to treat CDDD. Figure 2 reflects the large variety in results of the included study. These methodological challenges draw attention to the need for high quality cost-effectiveness research, and uniformity in methodology and reporting. Radcliff et al. described similar heterogeneity in a recent publication about economics of cervical disk replacement [72]. There was large heterogeneity in health care and societal perspective costs due to differences in calculation methods of costs and/or charges, included costs, different in-and exclusion criteria and baseline characteristics. It is noteworthy to mention that the lowest costs are reported in studies conducted outside of the United States of America (United Kingdom, Spain and Korea). This is not surprising considering the prominent difference in healthcare systems.
The Panel on Cost-Effectiveness in Health and Medicine in the United States recommends performing cost-effectiveness studies from the societal perspective [73]. Nevertheless, only 6 out of 23 studies included in this review have reported on societal perspective costs. The variety in the calculation of costs and charges was reliant on the inclusion of costs such as hospitalization, physician fees, procedural costs, pharmaceuticals, rehabilitation, caregivers and work absenteeism. Furthermore, it should be noted that some studies evaluated patients with radiculopathy or myelopathy alone, and other studies included all CDDD patients. There was also an important difference in the inclusion of singlelevel or multi-level surgeries. Another important factor is the large variation in study follow-up time. This may have a significant impact on cost-effectiveness as both the costs and clinical effectivity might be influenced by the occurrence of CASP and possible additional surgery rates [11,16].
The clinical effectivity of the surgical interventions was determined based on different utility measurement tools. Most studies used derivatives of the SF-questionnaires and converted these to SF-6D, however these converted questionnaires cannot equally be compared to their originals [74]. The reported effectiveness of the evaluated interventions was comparable in all studies and the difference in QALY-gain can be interpreted as not clinically significant. This is in line with previous literature that shows no clinical difference between ACDA and ACDF [17,18].
The theory that ACDA may be the superior surgical intervention for symptomatic CDDD has been the topic of discussion for multiple years. ACDA was developed in an effort to maintain motion and thereby prevent overstraining of the adjacent segments and consequent CASP. In spite of higher implant costs, ACDA might be a more cost-effective technique on the long term as CASP and additional surgeries with associated costs could be avoided. Due to the variances in reported data on costs and clinical effectivity, stated cost-effectiveness cannot equally be compared. However, there seems to be a trend of higher cost-effectiveness in ACDA in comparison with ACDF. The WTP is dependent on the burden of disease and varies strongly per country. In this study, WTP was only mentioned in US-orientated studies, of which 12 upheld a WTP of $50.000/QALY and 2 upheld a WTP $100.000/QALY. Three out of the 9 studies in favor of ACDA reported an ICER that passed the threshold of a WTP of $50.000/QALY, 2 out of the 3 studies in favor of ACDF passed this threshold. It can be stated that both techniques are expensive, which gives rise to the question of which technique is most cost-effective.
Besides determining the most cost-effective technique, it is important to improve cost-effectiveness, either by reducing costs or by improving clinical effectivity. Patient profiling may play a significant role. More research is needed to investigate which baseline characteristics predict better patient-reported outcome measures (PROMs) post-operatively, which could influence surgical decision-making. Moreover, it is strongly recommended that future cost-effectiveness analyses should be prospective and conducted uniformly. It would be profitable to have guidelines for the conduct of economic evaluations in spine surgery to create uniformity in methodology and reporting, as has recently been published for osteoporosis research [75].
In contrast to the limited number of narrative reviews concerning economics of cervical spine surgery, we provide a detailed and systematic overview of relevant literature. The strength of this systematic review is its methodological quality, as it was executed in accordance with the proposed PRISMA statement [31,32] and the five-step approach on preparing a systematic review of economic evaluations by Van Mastrigt et al. [33][34][35]. The main limitation that may have biased the result of this study was the inclusion of only full text and published studies but not conference proceedings, PhD dissertations or gray literature. Three studies were manufacturer-sponsored or had a possible financial benefit [53,64,69] and two studies did not provide information on funding [46,55]. Another limitation is that we did not differentiate between the costs of different manufacturers for disk protheses. We expect limited influence on our results, in contrast to an increase in heterogeneity of our data.
Results from this systematic review were mostly extracted from data from North-America and thus may not be relevant for other continents. Data from other continents, or even countries, is essential as costs between countries vary.

Conclusion
This review evaluated cost-effectiveness of anterior surgical decompression techniques for radiculopathy and/or myelopathy due to CDDD. The most important result is the lack of uniform literature concerning the subject, which impedes any solid conclusions to be drawn. The results of this review show that both techniques are expensive. It seems that ACDA might be more cost-effective than ACDF as the majority of studies individually report ACDA to be the more cost-effective technique. In conclusion, there is a need for high-quality cost-effectiveness research and uniformity in reporting data to allow solid conclusions to be drawn.
Funding None.

Conflict of interest
The authors declare that they have no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.