Abstract
Background:
Operative and nonoperative treatments for spinal metastases are expensive interventions with a high rate of complications. We sought to determine the cost-effectiveness of a surgical procedure compared with nonoperative management as treatment for spinal metastases.
Methods:
We constructed a Markov state-transition model with health states defined by ambulatory status and estimated the quality-adjusted life-years (QALYs) and costs for operative and nonoperative management of spine metastases. We considered 2 populations: 1 in which patients presented with independent ambulatory status and 1 in which patients presented with nonambulatory status due to acute (e.g., <48 hours) metastatic epidural compression. We defined the efficacy of each treatment as a likelihood of maintaining, or returning to, independent ambulation. Transition probabilities for the model, including the risks of mortality and becoming dependent or nonambulatory, were obtained from secondary data analysis and published literature. Costs were determined from Medicare reimbursement schedules. We conducted analyses over patients’ remaining life expectancy from a health system perspective and discounted outcomes at 3% per year. We conducted sensitivity analyses to account for uncertainty in data inputs.
Results:
Among patients presenting as independently ambulatory, QALYs were 0.823 for operative treatment and 0.800 for nonoperative treatment. The incremental cost-effectiveness ratio (ICER) for a surgical procedure was $899,700 per QALY. Among patients presenting with nonambulatory status, those undergoing surgical intervention accumulated 0.813 lifetime QALY, and those treated nonoperatively accumulated 0.089 lifetime QALY. The incremental cost-effectiveness ratio for a surgical procedure was $48,600 per QALY. The cost-effectiveness of a surgical procedure was most sensitive to the variability of its efficacy.
Conclusions:
Our data suggest that the value to society of a surgical procedure for spinal metastases varies according to the features of the patient population. In patients presenting as nonambulatory due to acute neurologic compromise, surgical intervention provides good value (ICER, $48,600 per QALY). There is a low value for a surgical procedure performed for patients who are ambulatory at presentation (ICER, $899,700 per QALY).
Level of Evidence:
Economic and Decision Analysis Level III. See Instructions for Authors for a complete description of levels of evidence.
Over the last 2 decades, surgery has gained wider acceptance as a treatment for spinal metastases, and nonoperative care remains a mainstay, particularly among those without neurologic compromise1,2. Recent research has suggested that surgical management provides greater preservation of ambulatory function1,3–7, and this has led some to suggest that outcomes may be superior if patients undergo surgical intervention at the initial presentation, compared with delaying the surgical procedure until nonoperative treatment fails1,3,8–10. Health-care costs associated with a surgical procedure are higher, but may be offset by improved pain and functional independence in patients who undergo surgical treatment1,4,7–9.
There is no consensus with regard to treatment for patients with spinal metastases, especially in cases in which the individual presents with intact neurologic function1,2. Given that surgical intervention does not cure a patient’s underlying oncologic condition, the expense of a surgical procedure and anticipated benefits must be weighed against remaining life expectancy4,8,10–13. Some recent studies have maintained that surgical intervention is a cost-effective treatment for patients with spinal metastases5,6. These investigations were conducted using fairly small samples and did not distinguish different levels of ambulation at presentation3,7,11–13. The outcomes for patients presenting with neurologic deficits prior to a surgical procedure may be very different from those with intact ambulatory capacity at baseline. Moreover, it is necessary to factor in the extent of clinical variation encountered in patients with spinal metastases in order to make more accurate estimates of cost-effectiveness. We conducted a cost-effectiveness analysis to establish the value of operative treatment compared with nonoperative management for patients with spinal metastases, accounting for ambulatory function at presentation.
Materials and Methods
Analytic Overview
We developed a Markov state-transition model simulating patients with spinal metastatic disease that could be treated using either operative or nonoperative management. We constructed the Markov model using TreeAge Pro (TreeAge Software) to depict the post-treatment course of patients using a series of transitions between health states over their remaining life expectancy (Fig. 1). We derived the transition probabilities from secondary data analysis and published literature. We assigned a utility to each health state (see Appendix). Utility values are used to convert a year of life into a quality-adjusted life year (QALY)14. Costs consisted of expenditures associated with the resources utilized within each health state. We expressed costs in 2019 U.S. dollars with 3% annual adjustment applied for quality of life and costs14. We applied annual cancer-specific mortality, adjusted to 30-day periods as the model runs on a monthly cycle.
Fig. 1:
The structure of the Markov model for operative and nonoperative treatment for spinal metastases. The circles represent the health states. A simulated patient starts in the model at the time of initial presentation and undergoes a surgical procedure or radiation (orange circle). Following treatment, the patient may proceed to an ambulatory independent, ambulatory dependent, or nonambulatory state (white circles) or die (blue circle). At 30-day intervals, patients may proceed to a different health state (straight arrow) or remain in their current health state (curved arrow). Due to the nature of epidural compression in those treated nonoperatively, transition from independent ambulatory to nonambulatory health states was modeled (longer dashed curved arrow). Transition from dependent to independent ambulatory health states was allowed in the first 6 months (dashed arrow). Transition from nonambulatory function to dependent ambulatory function was only modeled in the operative arm (dotted arrow).
Outcomes from the model included QALYs and lifetime direct medical costs. We used the difference in costs divided by the difference in QALYs between the simulated operative and nonoperative strategies to calculate an incremental cost-effectiveness ratio (ICER), which quantifies the value of resources spent against willingness-to-pay thresholds, or the maximal cost that society is willing to spend for each additional QALY gained14,15. In this study, we used willingness-to-pay thresholds of $100,000 and $150,000, which are commonly used thresholds in the United States5,14,15. Treatments that have ICERs below the willingness-to-pay threshold are considered cost-effective14–16.
Model Structure
The model’s health states are characterized by ambulatory function and defined as independent, dependent (requiring use of an assistive device such as a cane or walker), nonambulatory (bed or wheelchair-bound), or death1,4,12,13. We defined the efficacy of operative and nonoperative treatment as the likelihood of maintaining, or returning to, independent ambulation. Patients could transition to dependent and nonambulatory states due to pain or neurologic deterioration. Patients could transition to death from any health state. Transitions occur at 30-day intervals, based on current ambulatory state, treatments received, and any post-treatment events (e.g., complications, revision surgical procedure). The time frame of the model was 5 years, or until all simulated patients died, whichever occurred first.
Populations Under Consideration and Strategies
We considered 2 populations: 1 in which patients presented with independent ambulatory status, and 1 in which patients presented with nonambulatory status due to acute (e.g., <48 hours) metastatic epidural compression. In both populations, we modeled a patient with metastatic epidural canal compromise at T12 and L1 from a radiosensitive tumor (e.g., lung, breast, prostate)12.
We examined 2 clinical strategies: operative treatment or nonoperative treatment. Operative management consisted of a 2-level posterolateral decompression and partial corpectomy with interbody cage, autograft, and 6-level posterior instrumented fusion. Nonoperative treatment included conventional radiation therapy of 30 Gy, administered in 10 fractions2,12.
Data Inputs
We estimated the probability of entering a dependent or a nonambulatory state, as well as mortality following treatment based on cancer progression, treatment failure, and post-treatment complications. We derived these probabilities from retrospective encounter data of adult patients who were independent ambulators and underwent operative or nonoperative treatment for spinal metastases at 3 tertiary referral cancer centers in Boston, Massachusetts, between 2005 and 2017 (see Appendix). We used published data3,4,10–13,17,18 to derive remaining transition probabilities for events that occurred infrequently in our retrospective data. The 30-day mortality rate was estimated to be 3.0% in patients who underwent operative treatment and 2.5% for patients who underwent nonoperative treatment. The complication rate (see Appendix) was estimated at 51.0% for patients who underwent operative treatment13 and 6.9% for patients who underwent nonoperative treatment4 (Table I).
TABLE I.
Input Parameters for Transition Probabilities, Complications, and Mortality for Patients Undergoing Operative and Nonoperative Treatment in Our Cost-Effectiveness Analysis
| Parameter | Ambulatory Independent at Presentation* | Nonambulatory at Presentation* | ||
|---|---|---|---|---|
| Operative | Nonoperative | Operative | Nonoperative | |
| First month | ||||
| Independent | 28.7%† | 41.8%† | 10.1%17 | 0%‡ |
| Dependent | 42.5%† | 30.2%† | 87.3%17 | 0%‡ |
| Nonambulatory | 28.7%† | 28.0%† | 2.5%17 | 100%‡ |
| Complication | 51%13 | 6.9%4 | 51%13 | 6.9%4 |
| Mortality | 3%13 | 2.5%§ | 3%13 | 2.5%§ |
| Transition probabilities between states at each additional month | ||||
| Independent to dependent | 2.9%† | 1.8%† | 2.9%† | 1.8%† |
| Independent to nonambulatory | — | 0.6%† | — | 0.6%† |
| Dependent to independent | 4.8%†# | 1.2%†# | 4.8%†# | 1.2%†# |
| Dependent to nonambulatory | 1.1%† | 1.9%† | 1.1%† | 1.9%† |
| Nonambulatory to independent | — | — | — | — |
| Nonambulatory to dependent | 8.6%†# | — | 8.6%†# | — |
| Independent to mortality | 2.8%† | 2.8%† | 2.8%† | 2.8%† |
| Dependent to mortality | 6.5%† | 6.5%† | 6.5%† | 6.5%† |
| Nonambulatory to mortality | 21.9%† | 21.9%† | 21.9%† | 21.9%† |
The values are given as the rate.
The designated data element is derived from retrospective encounter data of 713 adult patients who were independent ambulators and underwent operative treatment (n = 370) or nonoperative treatment (n = 343) for spinal metastases at Massachusetts General Hospital and Brigham and Women’s Hospital/Dana Farber Cancer Center.
This designates an assumption made that radiation cannot cause someone to improve from nonambulatory status.
This designates an assumption made on the basis of operative overall mortality.
This designates a value that is only applied for the first 6 months.
Each ambulatory state was associated with a quality-of-life utility and costs, including those related to complications, readmissions, the need for a surgical procedure in individuals initially managed nonoperatively, and revision procedures for patients treated operatively. We derived quality-of-life values from prospective EuroQol 5-Dimension (EQ-5D) surveys. These surveys were completed by participants in the Prospective Observational study of Spinal metastasis Treatment (POST), 2017 to 20191. We transformed raw data to utilities using normative U.S. values for the EQ-5D19. Our utilities were adjusted for age, sex, operative or nonoperative treatment, length of follow-up, and repeated measures in the same individual. The utility for independent ambulatory status was 0.756, and the utilities were 0.599 for dependent ambulatory status and 0.175 for dependent nonambulatory status.
We used the 2019 Medicare fee schedule20 to calculate unit costs of operative and nonoperative treatment including hospital admission, outpatient physician visits, emergency room encounters, imaging, anesthesia and surgeon fees, post-treatment care and evaluations, prescription medications, cost for treatment of complications including readmission and revision surgical procedures, post-treatment rehabilitation, durable medical equipment, home nursing care, and hospice care (see Appendix). The frequency of resource use for each category was determined from expert consultation. The direct episode cost was $33,529 for primary surgical intervention and $12,932 for nonoperative management. Costs for patients treated with radiation and then undergoing a surgical procedure were $46,600.
Sensitivity Analyses
In sensitivity analyses, we evaluated how the cost-effectiveness of operative treatment changes among different surgical procedures as well as the efficacy of operative management and the rate of complications following a surgical procedure. To account for variation in the types of surgical procedures performed for spinal metastases, we considered 2 additional procedures, posterior decompression and posterior decompression with fusion, and different levels of hospital reimbursement based on varying the diagnosis-related group (DRG). To account for the variation in clinical efficacy of surgical interventions, we incrementally increased the efficacy of operative management, starting at the probability (28.7%) of remaining independently ambulatory immediately following operative treatment that was used in the base model. We also incrementally reduced the complication rate following a surgical procedure, starting at the probability (51.0%) that was used in the base model.
Because a number of point estimates used in the base case analysis were derived from relatively small samples, we performed probabilistic sensitivity analysis to assess how robust our determinations were with regard to the cost-effectiveness of a surgical procedure by simultaneously varying estimates of the inputs used for mortality, complications, utilities, and transition to dependent and nonambulatory states in 10,000 simulations (see Appendix). We used cost-effectiveness acceptability curves to display the percentage of simulations where an operative or nonoperative strategy was considered cost-effective.
Source of Funding
This research was funded by National Institutes of Health (NIH-NIAMS [National Institute of Arthritis and Musculoskeletal and Skin Diseases]) grants K23-AR071464 to Dr. Schoenfeld, K24-AR057827 to Dr. Losina, and P30-AR072577 to Dr. Katz and Dr. Losina.
Results
Independent Ambulatory Status at Presentation
In the model, patients undergoing a surgical procedure for spinal metastases accumulated 0.823 QALY, and those treated with a nonoperative strategy accumulated 0.800 QALY (Table II). For those who underwent a surgical procedure, costs amounted to $73,777. For those who underwent nonoperative treatment, costs amounted to $53,299. The ICER for surgical intervention compared with nonoperative management was estimated at $899,700 per QALY.
TABLE II.
Cost-Effectiveness of Nonoperative and Operative Treatment in Patients with Independent Ambulatory and Nonambulatory Status at Presentation
| Lifetime Medical Cost | QALY | ICER* | |
|---|---|---|---|
| Independent ambulatory status at presentation† | |||
| Nonoperative treatment | $53,299 | 0.800 | — |
| Operative treatment | $73,777 | 0.823 | $899,700 |
| Nonambulatory status at presentation‡ | |||
| Nonoperative treatment | $38,330 | 0.089 | — |
| Operative treatment | $73,481 | 0.813 | $48,600 |
The ICER per QALY was rounded to the nearest hundred dollars.
In the base case analysis, patients were considered to present with independent ambulatory status.
In this scenario, patients were considered to present with nonambulatory status due to acute neurologic compromise.
Nonambulatory Status at Presentation
When we conducted the analysis for patients presenting with nonambulatory status due to acute metastatic epidural compression, those undergoing surgical intervention accumulated 0.813 lifetime QALY, and those treated nonoperatively accumulated 0.089 lifetime QALY. The total direct medical costs for patients who underwent a surgical intervention amounted to $73,481, and the costs of nonoperative treatment amounted to $38,330. The ICER for surgical intervention compared with nonoperative management was $48,600 per QALY.
Sensitivity Analysis
In sensitivity analysis, we examined surgical costs across a range of procedures and hospital compensation rates by DRG, including a surgical procedure and radiation, as well as lower-cost procedures such as decompression and posterolateral fusion. Costs ranged from $32,118 to $71,688 per surgical episode. The ICER for a surgical procedure and radiation was $1,465,300 per QALY, and the ICER for decompression and posterolateral fusion was $837,900 per QALY.
We then incrementally increased the efficacy of operative treatment for patients presenting as independent ambulators at 1-percentage-point intervals from the rate used in the base model (28.7%). A surgical procedure was cost-effective at the $150,000 willingness-to-pay threshold once 47% (i.e., an 18-percentage-point increase from the base case) of those treated operatively remained independently ambulatory following treatment. A surgical procedure was cost-effective at the $100,000 willingness-to-pay threshold when 62% (i.e., a 33-percentage-point increase from the base case) of those treated operatively remained independently ambulatory following treatment. In the population presenting as nonambulatory due to acute neurologic compromise, 84% of patients treated nonoperatively would have to recover some degree of ambulatory function before a surgical procedure was no longer cost-effective at the $100,000 willingness-to-pay threshold. Varying the rate of complications following a surgical procedure did not have a substantial impact on our estimates. At a postoperative complication rate of 0%, the ICER for a surgical procedure in patients presenting as independently ambulatory was $884,100 per QALY. Among those presenting as nonambulatory, a complication rate of 0% resulted in an ICER of $48,300 per QALY.
In probabilistic sensitivity analysis for the population presenting with independent ambulatory status, a surgical procedure was cost-effective in 11% of the simulations at the willingness-to-pay threshold of $100,000 per QALY (Fig. 2). At the willingness-to-pay threshold of $150,000 per QALY, a surgical procedure was cost-effective in 20% of the simulations. In probabilistic sensitivity analysis for patients presenting as nonambulatory (Fig. 3), a surgical procedure was cost-effective in 71% of the simulations at the willingness-to-pay threshold of $100,000 per QALY. At the willingness-to-pay threshold of $150,000 per QALY, a surgical procedure was cost-effective in 82% of the simulations.
Fig. 2:
Cost-effectiveness acceptability curve indicating the probability of operative or nonoperative treatment being the cost-effective strategy at a given willingness-to-pay (WTP) threshold, utilizing a Monte Carlo simulation of 10,000 patients, in the scenario in which patients present as independent ambulators. The x axis represents the different WTP thresholds, with the y axis denoting the percent of the Monte Carlo iterations that were cost-effective for nonoperative treatment (in orange) or operative treatment (in blue). The Monte Carlo simulation simultaneously varied the parameters drawn from the distributions for mortality and treatment complication rates, utilities, and transition probabilities to dependent and nonambulatory states.
Fig. 3:
Cost-effectiveness acceptability curve indicating the probability of operative or nonoperative treatment being the cost-effective strategy at a given willingness-to-pay (WTP) threshold, utilizing a Monte Carlo simulation of 10,000 patients, in the scenario where patients present with nonambulatory status. The x axis represents the different WTP thresholds, with the y axis denoting the percent of the Monte Carlo iterations that were cost-effective for nonoperative treatment (in orange) or operative treatment (in blue). The Monte Carlo simulation simultaneously varied the parameters drawn from the distributions for mortality and complication rates, utilities, and transition probabilities to dependent and nonambulatory states.
Discussion
We investigated the cost-effectiveness of initial surgical intervention compared with nonoperative management for the treatment of patients with spinal metastases. From the health-care sector perspective, surgical intervention was not cost-effective at willingness-to-pay thresholds of $100,000 or $150,000 for patients presenting as independent ambulators. In the analysis focused on patients presenting as nonambulatory, a surgical procedure was cost-effective with an ICER of $48,600 per QALY.
Spinal metastases are challenging to treat, due to limited life expectancy among patients as well as the impact that the disease can have on quality of life and functional independence2,4,9,11,18. A surgical procedure is thought to provide a number of benefits relative to nonoperative care, but does have a nontrivial recovery period with reduced quality of life in the immediate postoperative period1–3. Both operative and nonoperative approaches are relatively expensive1,4,7,21,22. The initial direct costs for a surgical procedure are higher than nonoperative management, and in our population presenting with independent ambulation, these procedures resulted in modest preservation of QALYs overall and slightly improved longevity. The limited life expectancy of patients with spinal metastases means that many succumb to other aspects of the cancer condition even after the spinal disorders have been successfully treated1,2,7, a fact that impacts the cost-effectiveness of operative care.
Other investigations have previously reported that a surgical procedure for spinal metastases is cost-effective compared with nonoperative treatment5,6,23. Miyazaki et al. found that the ICER for surgical intervention approximated $42,000 per QALY, but the baseline EQ-5D among the 47 patients in that investigation was 0.046. Similar results were published by Depreitere et al. in a cohort of only 46 patients23. In that investigation, 50% of patients who underwent operative treatment had some degree of neurologic compromise. In their analysis of 92 patients treated for spinal metastases, Turner et al. concluded that the cost-effectiveness of a surgical procedure resulted from improved preservation of ambulatory function and lower postoperative health-care requirements5. Although our episode costs for operative and nonoperative care approximate those of Turner et al., our determinations with regard to postoperative health-care requirements are different. However, Turner et al. made no allowance for dynamic changes in ambulatory function over time. Therefore, the determinations of these direct clinical studies may be confounded by indication, restricted clinical variation, and a classification bias given that outcomes were not stratified by ambulatory status.
Our approach addresses uncertainties in key input parameters with sensitivity analyses that investigate the effects on our findings across wide plausible ranges of these input parameters. The results of our analyses suggest that the cost-effectiveness of a surgical procedure varies on the basis of the degree of ambulatory status at presentation, as well as the effectiveness of treatment in maintaining functional independence. In the population in which patients presented with nonambulatory status due to acute neurologic compromise from epidural compression, our findings support surgical intervention as cost-effective, with a favorable incremental cost-effectiveness ratio of $48,600 per QALY.
In contrast, in the scenario in which patients presented as independently ambulatory, nonoperative treatment maintained the higher probability of being cost-effective throughout the policy-relevant willingness-to-pay range of $50,000 to $150,000 per QALY. However, improvement in the effectiveness of operative management could result in a better value for a surgical procedure. For example, an 18-percentage-point shift in the proportion of patients who are independently ambulatory after a surgical procedure (e.g., 29% independently ambulatory to 47% independently ambulatory) would change the conclusions with regard to the cost-effectiveness of a surgical procedure at the willingness-to-pay threshold of $150,000 per QALY. As a cost-effectiveness analysis, our results are not intended to be used as a treatment algorithm directing patient care. However, the results would indicate that we cannot endorse surgical intervention as a standardized treatment strategy for patients who present with independent ambulatory function. In such scenarios, radiation and observation would seem to be the more appropriate strategy. At the same time, in conditions such as spinal metastatic disease, where expensive treatments have the ability to make dramatic changes in quality of life but only over a limited lifespan, there is unlikely to be a one-size-fits-all strategy that can balance health-care policy goals and individual patient priorities.
We recognize several limitations. Foremost, the data were largely derived from a retrospective review of clinical encounters within health-care facilities integrated under a single health-care corporation and may be subject to selection or surveillance bias. We attempted to address this to the fullest extent possible by varying input parameters in our sensitivity tests. Furthermore, the mortality figures that occurred in the model for both operative and nonoperative cohorts are consistent with prior published data12,24,25. We believe that these findings provide external validity for the model. Similarly, the limited data on quality of life, pain, and ambulatory function in large series of patients treated operatively and nonoperatively for spinal metastases necessitated assumptions in certain aspects of the model, including the types of procedures performed. We recognize that patients selected for a surgical procedure might have favorable clinical and cancer-related characteristics supporting such a decision. As a result, our findings would not be considered applicable to those undergoing a palliative surgical procedure. We also limited consideration to the most common modalities used to treat spinal metastases, focusing on surgical intervention and radiation, and only considered 2 clinical populations based on ambulatory status at presentation. We did not consider other interventions such as chemotherapy or immunotherapy, alone or in combination. Incorporating these often expensive treatment modalities would certainly increase health-care expenditures, with an uncertain impact on effectiveness. Likewise, given the lack of detailed information on the impact of treatments on work and out-of-pocket expenses for patients and their caregivers, we were unable to conduct analyses from a societal perspective.
In conclusion, our data suggest that the value to society of a surgical procedure for spinal metastases varies according to the features of the patient population. In patients presenting as nonambulatory due to acute neurologic compromise, surgical intervention provides good value (ICER, $48,600 per QALY). There is low value for a surgical procedure performed for patients who are ambulatory at presentation (ICER, $899,700 per QALY).
Supplementary Material
Acknowledgments
The authors wish to acknowledge the contributions of Dr. Kristin Redmond and Dr. Claire Safran-Norton for their assistance in determining resource utilization for the cohorts modeled in this study. Neither of these individuals was compensated for their efforts.
Footnotes
Disclosure: The Disclosure of Potential Conflicts of Interest forms are provided with the online version of the article (http://links.lww.com/JBJS/G617).
Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the NIH or the Federal government.
Appendix
Supporting material provided by the authors is posted with the online version of this article as a data supplement at jbjs.org (http://links.lww.com/JBJS/G618).
References
- 1.Schoenfeld AJ, Blucher JA, Barton LB, Schwab JH, Balboni TA, Chi JH, Shin JH, Kang JD, Harris MB, Ferrone ML. Design of the Prospective Observational study of Spinal metastasis Treatment (POST). Spine J. 2020. Apr;20(4):572–9. Epub 2019 Nov 8. [DOI] [PubMed] [Google Scholar]
- 2.Rose PS, Buchowski JM. Metastatic disease in the thoracic and lumbar spine: evaluation and management. J Am Acad Orthop Surg. 2011. Jan;19(1):37–48. [DOI] [PubMed] [Google Scholar]
- 3.Lo WY, Yang SH. Metastatic spinal cord compression (MSCC) treated with palliative decompression: surgical timing and survival rate. PLoS One. 2017. Dec 29;12(12):e0190342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Schoenfeld AJ, Losina E, Ferrone ML, Schwab JH, Chi JH, Blucher JA, Silva GS, Chen AT, Harris MB, Kang JD, Katz JN. Ambulatory status after surgical and nonsurgical treatment for spinal metastasis. Cancer. 2019. Aug 1;125(15):2631–7. Epub 2019 Apr 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Turner I, Kennedy J, Morris S, Crockard A, Choi D. Surgery and radiotherapy for symptomatic spinal metastases is more cost effective than radiotherapy alone: a cost utility analysis in a U.K. spinal center. World Neurosurg. 2018. Jan;109:e389–97. Epub 2017 Oct 5. [DOI] [PubMed] [Google Scholar]
- 6.Miyazaki S, Kakutani K, Sakai Y, Ejima Y, Maeno K, Takada T, Yurube T, Terashima Y, Ito M, Kakiuchi Y, Takeoka Y, Hara H, Kawamoto T, Sakashita A, Okada T, Kiyota N, Kizawa Y, Sasaki R, Akisue T, Minami H, Kuroda R, Nishida K. Quality of life and cost-utility of surgical treatment for patients with spinal metastases: prospective cohort study. Int Orthop. 2017. Jun;41(6):1265–71. Epub 2017 Apr 10. [DOI] [PubMed] [Google Scholar]
- 7.Schoenfeld AJ, Ferrone ML. The next generation in surgical research for patients with spinal metastases. Spine J. 2018. Oct;18(10):1956–8. Epub 2018 Jul 26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Choi D, Fox Z, Albert T, Arts M, Balabaud L, Bunger C, Buchowski JM, Coppes MH, Depreitere B, Fehlings MG, Harrop J, Kawahara N, Martin-Benlloch JA, Massicotte EM, Mazel C, Oner FC, Peul W, Quraishi N, Tokuhashi Y, Tomita K, Verlaan JJ, Wang M, Wang M, Crockard HA. Rapid improvements in pain and quality of life are sustained after surgery for spinal metastases in a large prospective cohort. Br J Neurosurg. 2016. Jun;30(3):337–44. Epub 2016 Feb 22. [DOI] [PubMed] [Google Scholar]
- 9.Tang Y, Qu J, Wu J, Liu H, Chu T, Xiao J, Zhou Y. Effect of surgery on quality of life of patients with spinal metastasis from non-small-cell lung cancer. J Bone Joint Surg Am. 2016. Mar 2;98(5):396–402. [DOI] [PubMed] [Google Scholar]
- 10.Paulino Pereira NR, Ogink PT, Groot OQ, Ferrone ML, Hornicek FJ, van Dijk CN, Bramer JAM, Schwab JH. Complications and reoperations after surgery for 647 patients with spine metastatic disease. Spine J. 2019. Jan;19(1):144–56. Epub 2018 Jun 1. [DOI] [PubMed] [Google Scholar]
- 11.Paulino Pereira NR, Janssen SJ, van Dijk E, Harris MB, Hornicek FJ, Ferrone ML, Schwab JH. Development of a prognostic survival algorithm for patients with metastatic spine disease. J Bone Joint Surg Am. 2016. Nov 2;98(21):1767–76. [DOI] [PubMed] [Google Scholar]
- 12.Shi DD, Chen YH, Lam TC, Leonard D, Balboni TA, Schoenfeld A, Skamene S, Cagney DN, Chi JH, Cho CH, Harris M, Ferrone ML, Hertan LM. Assessing the utility of a prognostication model to predict 1-year mortality in patients undergoing radiation therapy for spinal metastases. Spine J. 2018. Jun;18(6):935–40. Epub 2017 Oct 12. [DOI] [PubMed] [Google Scholar]
- 13.Schoenfeld AJ, Le HV, Marjoua Y, Leonard DA, Belmont PJ Jr, Bono CM, Harris MB. Assessing the utility of a clinical prediction score regarding 30-day morbidity and mortality following metastatic spinal surgery: the New England Spinal Metastasis Score (NESMS). Spine J. 2016. Apr;16(4):482–90. Epub 2015 Sep 25. [DOI] [PubMed] [Google Scholar]
- 14.Rajan PV, Qudsi RA, Wolf LL, Losina E. Cost-effectiveness analyses in orthopaedic surgery: Raising the bar. J Bone Joint Surg Am. 2017. Jul 5;99(13):e71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Basu A Estimating costs and valuations of non-health benefits in cost-effectiveness analysis. In: Neumann PJ, Sanders GD, Russell LB, Siegel JE, Ganiats TG, editors. Cost-effectiveness in health and medicine. 2nd ed. New York: Oxford University Press; 2017. [Google Scholar]
- 16.Siegel JE, Weinstein MC, Russell LB, Gold MR; Panel on Cost-Effectiveness in Health and Medicine. Recommendations for reporting cost-effectiveness analyses. JAMA. 1996. Oct 23–30;276(16):1339–41. [DOI] [PubMed] [Google Scholar]
- 17.Kato S, Hozumi T, Takeshita K, Kondo T, Goto T, Yamakawa K. Neurological recovery after posterior decompression surgery for anterior dural compression in paralytic spinal metastasis. Arch Orthop Trauma Surg. 2012. Jun;132(6):765–71. Epub 2012 Feb 14. [DOI] [PubMed] [Google Scholar]
- 18.Karhade AV, Thio QCBS, Ogink PT, Bono CM, Ferrone ML, Oh KS, Saylor PJ, Schoenfeld AJ, Shin JH, Harris MB, Schwab JH. Predicting 90-day and 1-year mortality in spinal metastatic disease: development and internal validation. Neurosurgery. 2019. Oct 1;85(4):E671–81. [DOI] [PubMed] [Google Scholar]
- 19.Ramos-Goñi JM, Rivero-Arias O . EQ5D: A command to calculate index values for the EQ-5D quality of life instrument. Stata J. 2011;11(1):120–5. [Google Scholar]
- 20.U.S. Centers for Medicare and Medicaid Services. Medicare Fee Schedules 2019. Accessed 2020 Mar 1. https://www.cms.gov/Medicare/Medicare.html
- 21.Hess G, Barlev A, Chung K, Hill JW, Fonseca E. Cost of palliative radiation to the bone for patients with bone metastases secondary to breast or prostate cancer. Radiat Oncol. 2012. Oct 12;7:168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Tipsmark LS, Bünger CE, Wang M, Morgen SS, Dahl B, Søgaard R. Healthcare costs attributable to the treatment of patients with spinal metastases: a cohort study with up to 8 years follow-up. BMC Cancer. 2015. May 5;15:354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Depreitere B, Turner I, Vandoren C, Choi D. Cost-utility analysis of surgery and radiotherapy for symptomatic spinal metastases in a Belgian specialist center. World Neurosurg. 2019. May;125:e537–43. Epub 2019 Feb 1. [DOI] [PubMed] [Google Scholar]
- 24.Ghori AK, Leonard DA, Schoenfeld AJ, Saadat E, Scott N, Ferrone ML, Pearson AM, Harris MB. Modeling 1-year survival after surgery on the metastatic spine. Spine J. 2015. Nov 1;15(11):2345–50. Epub 2015 Jul 6. [DOI] [PubMed] [Google Scholar]
- 25.Schoenfeld AJ, Ferrone ML, Schwab JH, Blucher JA, Barton LB, Tobert DG, Chi JH, Shin JH, Kang JD, Harris MB. Prospective validation of a clinical prediction score for survival in patients with spinal metastases: the New England Spinal Metastasis Score. Spine J. 2021. Jan;21(1):28–36. Epub 2020 Feb 19. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.