Sheeraz A. Qureshi, Steven McAnany, Vadim Goz, Steven M. Koehler and Andrew C. Hecht
In recent years, there has been increased interest in the use of cervical disc replacement (CDR) as an alternative to anterior cervical discectomy and fusion (ACDF). While ACDF is a proven intervention for patients with myelopathy or radiculopathy, it does have inherent limitations. Cervical disc replacement was designed to preserve motion, avoid the limitations of fusion, and theoretically allow for a quicker return to activity. A number of recently published systematic reviews and randomized controlled trials have demonstrated positive clinical results for CDR, but no studies have revealed which of the 2 treatment strategies is more cost-effective. The purpose of this study was to evaluate the cost-effectiveness of CDR and ACDF by using the power of decision analysis. Additionally, the authors aimed to identify the most critical factors affecting procedural cost and effectiveness and to define thresholds for durability and function to focus and guide future research.
The authors created a surgical decision model for the treatment of single-level cervical disc disease with associated radiculopathy. The literature was reviewed to identify possible outcomes and their likelihood following CDR and ACDF. Health state utility factors were determined from the literature and assigned to each possible outcome, and procedural effectiveness was expressed in units of quality-adjusted life years (QALYs). Using ICD-9 procedure codes and data from the Nationwide Inpatient Sample, the authors calculated the median cost of hospitalization by multiplying hospital charges by the hospital-specific cost-to-charge ratio. Gross physician costs were determined from the mean Medicare reimbursement for each current procedural terminology (CPT) code. Uncertainty as regards both cost and effectiveness numbers was assessed using sensitivity analysis.
In the reference case, the model assumed a 20-year duration for the CDR prosthesis. Cervical disc replacement led to higher average QALYs gained at a lower cost to society if both strategies survived for 20 years ($3042/QALY for CDR vs $8760/QALY for ACDF). Sensitivity analysis revealed that CDR needed to survive at least 9.75 years to be considered a more cost-effective strategy than ACDF. Cervical disc replacement becomes an acceptable societal strategy as the prosthesis survival time approaches 11 years and the $50,000/QALY gained willingness-to-pay threshold is crossed. Sensitivity analysis also indicated that CDR must provide a utility state of at least 0.796 to be cost-effective.
Both CDR and ACDF were shown to be cost-effective procedures in the reference case. Results of the sensitivity analysis indicated that CDR must remain functional for at least 14 years to establish greater cost-effectiveness than ACDF. Since the current literature has yet to demonstrate with certainty the actual durability and long-term functionality of CDR, future long-term studies are required to validate the present analysis.
Robert J. Owen, Adam Z. Khan, Steven J. McAnany, Colleen Peters and Lukas P. Zebala
The aim of this study was to compare the patient-reported outcome measures Neck Disability Index (NDI) and visual analog scale (VAS) with the Patient Reported Outcomes Measurement Information System (PROMIS) physical function (PF) and pain interference (PI) measures, respectively, and to determine their correlations in a surgical population longitudinally.
Legacy outcome measures such as NDI and VAS are essential for analyzing treatments in spine surgery for cervical disc herniations with radiculopathy. Despite their usefulness, administrative burdens impose limits on completion of these measures. PROMIS was developed as a patient outcome measure in order to improve reporting of patient symptoms and function and to reduce administrative burden. Despite early positive results of PROMIS in orthopedics, NDI and VAS scores have not been compared with PROMIS scores in patients with cervical disc herniations with radiculopathy.
Eighty patients undergoing surgery for cervical disc herniations with radiculopathy were included. All patients were treated at the same tertiary spine center. Patients were seen and PROMIS PF and PI, NDI, and VAS arm and neck pain scores were collected preoperatively and at 1 year postoperatively. Correlations between NDI, VAS, and PROMIS PF and PI were quantified using Pearson correlation coefficients. Two-tailed Student t-tests were used to demonstrate correlation significance, with alpha = 0.05.
All 80 (100%) patients completed all preoperative questionnaires. Fifty-seven (72%) and 75 (94%) patients completed all questionnaires at baseline and at the 6-month and 1-year follow-ups, respectively. PROMIS PF and NDI scores demonstrated a strong negative correlation, with Pearson r values of −0.81, −0.77, and −0.75 at baseline, 6 months, and 1 year. PROMIS PI and VAS neck pain scores demonstrated a moderately positive correlation, with Pearson r values of 0.51, 0.61, and 0.6. PROMIS PI and VAS arm pain scores demonstrated a moderately positive correlation, with Pearson r values of 0.46, 0.47, and 0.45.
PROMIS PF scores have a strong negative correlation with NDI scores at baseline and in the postoperative course in patients undergoing surgery for cervical disc herniations with radiculopathy. PROMIS PI scores have a moderately positive correlation with VAS neck and arm pain scores at baseline and in the postoperative course. Surgeons may factor these correlation results into the interpretation of patient-reported outcome measures in patients with cervical radiculopathy. Use of PROMIS PF and PI for this patient population may reduce administrative burden while providing reliable outcomes data.
Sheeraz Qureshi, Vadim Goz, Steven McAnany, Samuel K. Cho, Andrew C. Hecht, Rick B. Delamarter and Michael G. Fehlings
Cost-effectiveness analysis (CEA) of medical interventions has become increasingly relevant to the discussion of optimization of care. The use of utility scales in CEA permits a quantitative assessment of effectiveness of a given intervention. There are no published utility values for degenerative disc disease (DDD) of the cervical spine, anterior cervical discectomy and fusion (ACDF), or cervical disc replacement (CDR). The purpose of this study was to define health utility values for those health states.
The 36-Item Short Form Health Survey data from the ProDisc-C investigational device exemption study were obtained for single-level DDD at baseline and 24 months postoperatively after ACDF or CDR procedures. Patients in the original study were randomized to either ACDF or CDR. Utilizing a commercially available Short Form–6 dimensions program, utility scores were calculated for each health state using a set of parametric preference weights obtained from a sample of the general population using the recognized valuation technique of standard gamble.
The baseline health state utility (HSU) value for a patient with single-level DDD was 0.54 in both the ACDF and CDR groups. Postoperative changes in HSU values were seen in both intervention groups at 24 months. Cervical disc replacement had a HSU value of 0.72. Anterior cervical discectomy and fusion was found to have a postoperative utility state of 0.71. No statistically significant difference was found in the HSU for ACDF and CDR at 24 months of follow-up.
This study represents the first calculated HSU value for a patient with single-level cervical DDD. Additionally, 2 common treatment interventions for this disease state were assessed. Both treatments were found to have significant impact on the HSU values. These values are integral to future CEA of ACDF and CDR.
Sohrab Virk, Avani S. Vaishnav, Jung Kee Mok, Steven McAnany, Sravisht Iyer, Todd J. Albert, Catherine Himo Gang and Sheeraz A. Qureshi
Preoperative pain assessment is often used to gauge the amount of disability in patients with lumbar disc herniation. How high preoperative pain scores impact the clinical course and outcomes of patients after lumbar microdiscectomy is not always clear. Here, the authors aimed to determine whether patients reporting higher preoperative pain scores have worse outcomes after lumbar microdiscectomy than those reporting lower preoperative scores.
The authors performed a retrospective review of patients with symptomatic lumbar disc herniations that had failed to improve with nonsurgical methods and who had undergone tubular lumbar microdiscectomy. Health-related quality of life (HRQOL) scores had been collected in the preoperative and postoperative period. The anatomical severity of disease was graded based on lumbar disc health (Pfirrmann classification), facet degeneration, thecal sac cross-sectional area, and disc herniation grade. Data on each patient’s narcotic consumption and length of stay were collected. A Student t-test and chi-square test were used to compare patients with high preoperative pain scores (HP cohort) and those with lower preoperative scores (non-HP cohort).
One hundred thirty-eight patients were included in this analysis. The 47 patients in the HP cohort had taken more preoperative opioids (12.0 ± 21.2 vs 3.6 ± 11.1 morphine equivalent doses, p = 0.01). However, there was no statistically significant difference in Pfirrmann classification (p > 0.15), facet grade (p > 0.11), thecal sac cross-sectional area (p = 0.45), or disc herniation grade (p = 0.39) between the HP and non-HP cohorts. The latter cohort had statistically significant higher preoperative PROMIS scores (36.5 ± 7.0 vs 29.9 ± 6.4, p < 0.001), SF-12 mental component summary scores (48.7 ± 11.5 vs 38.9 ± 16.1, p < 0.001), and SF-12 physical component summary scores (PCS; 32.4 ± 8.6 vs 27.5 ± 10.0, p = 0.005), but a lower Oswestry Disability Index (56.4 ± 22.1 vs 35.4 ± 15.5, p < 0.001). There were only two time points after microdiscectomy when the HP cohort had worse HRQOL scores: at the 2-week follow-up for SF-12 PCS scores (32.4 ± 8.6 vs 29.3 ± 7.1, p = 0.03) and the 12-week follow-up for PROMIS scores (45.2 ± 9.5 vs 39.5 ± 7.1, p = 0.01). All other postoperative HRQOL measurements were similar between the two cohorts (p > 0.05).
A patient’s perceived severity of disease often does not correlate with the actual clinical pathology on imaging. Although patients who report high pain and have a symptomatic lumbar disc herniation may describe their pain as more extreme, they should be counseled that the outcomes of microdiscectomy are positive.
Francis C. Lovecchio, Avani S. Vaishnav, Michael E. Steinhaus, Yahya A. Othman, Catherine Himo Gang, Sravisht Iyer, Steven J. McAnany, Todd J. Albert and Sheeraz A. Qureshi
In an effort to prevent loss of segmental lordosis (SL) with minimally invasive interbody fusions, manufacturers have increased the amount of lordosis that is built into interbody cages. However, the relationship between cage lordotic angle and actual SL achieved intraoperatively remains unclear. The purpose of this study was to determine if the lordotic angle manufactured into an interbody cage impacts the change in SL during minimally invasive surgery (MIS) for lumbar interbody fusion (LIF) done for degenerative pathology.
The authors performed a retrospective review of a single-surgeon database of adult patients who underwent primary LIF between April 2017 and December 2018. Procedures were performed for 1–2-level lumbar degenerative disease using contemporary MIS techniques, including transforaminal LIF (TLIF), lateral LIF (LLIF), and anterior LIF (ALIF). Surgical levels were classified on lateral radiographs based on the cage lordotic angle (6°–8°, 10°–12°, and 15°–20°) and the position of the cage in the disc space (anterior vs posterior). Change in SL was the primary outcome of interest. Subgroup analyses of the cage lordotic angle within each surgical approach were also conducted.
A total of 116 surgical levels in 98 patients were included. Surgical approaches included TLIF (56.1%), LLIF (32.7%), and ALIF (11.2%). There were no differences in SL gained by cage lordotic angle (2.7° SL gain with 6°–8° cages, 1.6° with 10°–12° cages, and 3.4° with 15°–20° cages, p = 0.581). Subgroup analysis of LLIF showed increased SL with 15° cages only (p = 0.002). The change in SL was highest after ALIF (average increase 9.8° in SL vs 1.8° in TLIF vs 1.8° in LLIF, p < 0.001). Anterior position of the cage in the disc space was also associated with a significantly greater gain in SL (4.2° vs −0.3°, p = 0.001), and was the only factor independently correlated with SL gain (p = 0.016).
Compared with cage lordotic angle, cage position and approach play larger roles in the generation of SL in 1–2-level MIS for lumbar degenerative disease.