Hospital Cost Analysis of Neuromuscular Scoliosis Surgery

         Neuromuscular scoliosis (NMS) is commonly defined as a scoliotic deformity arising secondarily to muscle imbalance caused by an underlying neuropathic or myopathic disease. Generally, spinal deformities secondary to neuromuscular pathology present early in life and progress rapidly.1 In children and adolescents with neuromuscular disease, the incidence of spinal deformity depends on the underlying diagnosis. For example, the incidence of scoliosis secondary to cerebral palsy has been reported between 6% and 64%, while nearly 100% of patients with a thoracic spinal cord injury occurring before puberty develop scoliosis.2 Neuromuscular scoliosis patients represent a heterogenous and medically fragile subgroup of scoliotic patients. Despite the complexity and heterogeneity of co-morbidities, as well as significant risks and associated costs, recent studies have demonstrated that surgical treatment improves both medical outcomes and the quality of life in patients with progressive NMS.1,3-6 

          Therefore, multilevel spinal fusion surgery remains the primary surgical treatment option for progressive NMS.7 Surgical techniques for the treatment of scoliotic deformity have continued to evolve. Segmental spinal instrumenHospital Cost Analysis of Neuromuscular Scoliosis Surgery Christopher Diefenbach, M.D., Marc N. Ialenti, M.D., Baron S. Lonner, M.D., Jonathan R, Kamerlink, M.D., Kushagra Verma, M.D., and Thomas J. Errico, M.D. Proofs to: BLonner@nyc.rr.com Author: Refs #18 and #19 not cited and have be deleted, all remaining references have been renumbered Author: Figure 1 pie chart deleted was hard to read and all info is in table 3 Christopher Diefenbach, M.D., Marc N. Ialenti, M.D., Baron S. Lonner, M.D., Jonathan R. Kamerlink, M.D., Kushagra Verma, M.D., and Thomas J. Errico, M.D., are from the Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, New York, New York. Correspondence: Baron S. Lonner, M.D., Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, 820 Second Avenue, Suite 7A, New York, New York 10017; BLonner@nyc.rr.com. Bulletin of the Hospital for Joint Diseases 2013;71(4):00-00 273 tation, blood salvage techniques, and spinal cord monitoring, although difficult to establish in all NMS patients, are now standard components of surgical treatment.8-12

          The inevitable consequence of the utilization of these technological advances, however, is increased costs incurred at the time of surgery. In the current setting of rising health care expenditures with no immediate signs of abatement, with the goal of providing universal care, and the need to demonstrate value, it is increasingly important to provide care for NMS, and all patients, in the most cost-effective manner possible, without compromising results or patient safety.13,14 Characterization of the various costs incurred at the time of surgery and hospitalization will facilitate the identification of potential opportunities for cost control. The purpose of this study is to evaluate the distribution of hospital and operating room costs incurred during surgical correction of NMS. An accurate analysis of the surgical and hospital costs for NMS is of paramount importance to ensure future equitable allocation of financial resources in this patient population and to provide opportunities for cost containment.

          We hypothesize that there will be demographic and surgical parameters that correlate with higher cost. Methods Before commencement, this study was granted approval by our Institutional Review Board. A retrospective chart review of 74 consecutive patients who had undergone a primary spinal fusion or spinal fusion revision for neuromuscular scoliosis by three surgeons at two hospitals at our institution between January 2006 and December 2007 was completed. Demographic data collected included gender, age, preoperative height (measured in standing position or supine with a tape measure), preoperative weight, and preoperative BMI. Full length upright (when possible) scoliosis radiographs were obtained preoperatively and postoperatively. Radiographic measurements were performed by the three surgeons. Major coronal curvatures, as well as T5-T12 kyphosis angles, were assessed. Additionally, the construct type (ie. pedicle screws, hooks, or hybrid), as well as the number of screws, hooks, and cable anchors implanted, were recorded from the postoperative radiographs. In this study, pedicle fixation was defined as a pedicle screw construct if more than 80% of the anchors used were screws. Similarly, fixation was defined as a hook construct if more than 80% of the anchors used were hooks. All other combinations were defined as hybrid.

          Cost Determination We reviewed 13,470 individual, as well as aggregate, costs associated with the surgical interventions for the 74 patients. These costs for surgical correction, hospital stay, and postoperative care were collected and separated into 23 categories. Using the Eclipsys Sunrise Decision Support System™ (Eclipsys Corporation, Atlanta, GA) and the PeopleSoft Payroll system (Oracle®, Redwood Shores, CA), the Finance Department of our hospital was able to ascertain and compile cost data. Costs were obtained using the definitions previously reported by Finkler15 and Calderone and coworkers.16 Within the cost accounting module, costs from the general ledger, and payroll systems are assigned to departments. Each chargeable service within a department is assigned a relative value unit (RVU), which represents the labor and supply elements required to provide that service. Based upon these RVUs and the expenses from the general ledger, a labor and supply cost is developed within the system for each service. Some RVU are by the hour (OR, recovery room) while some are per day (ICU, floor). Physical therapy is determined on the level of care.

          Material costs are based on raw material costs, supply costs, and labor needed to acquire and maintain the material. Indirect departments (administration, accounting and billing, admitting, insurance, heat, light and power, plant maintenance, medical records, laundry and linen, purchasing, housekeeping, central supply, and so forth) are departments that do not provide direct chargeable patient care but support the overall medical center. These costs are allocated via a step down methodology to the direct departments based upon a variety of defined metrics Facility costs, including building depreciation, maintenance, heat, light and power are allocated by the square footage occupied by each department. Other costs are determined using time studies, hours of service, time spent, pounds of laundry, and cost requisitions. These indirect costs are allocated using the Eclipsys Sunrise Decision Support System™. This methodology is similar to that previously reported by Kamerlink and coworkers.17 Statistical Analysis Descriptive statistics including means and standard deviations were calculated for demographic, operative, and financial data.

          Linear correlation analysis was employed to assess the following variables for independent correlation with total cost: age, height, weight, BMI, major structural curve, minor structural curve, number of posterior levels fused, number of anterior levels fused, total number of (A/P) levels fused, estimated blood loss, total operative time, number of screws or hooks or wires, and length of hospital stay. The threshold for inclusion of individual variables in the linear correlation model was p < 0.05. All statistical analyses were performed in SPSS, version 15.0 (SPSS Inc., Chicago, IL). Source of Funding This study was department initiated with no outside funding. Results The mean age of the patients was 15.8 ± 7.3 years, and a majority of the patients (57%) were male. The most common patient diagnoses were found to be cerebral palsy (28%) and familial dysautonomia (14%), while all other diagnoses comprised 58% of the patients (see Table 1 for all comorbid 274 Bulletin of the Hospital for Joint Diseases 2013;71(4):00-00 diagnoses).

          The mean preoperative major curve magnitude was 60° (range: 11° to 89°). Demographic data and comorbid factors for the study patients can be found in Table 1. The most prevalent surgical approach was posterior (76%), and pedicle screws (75% of cases) were predominantly utilized for fixation. Neuromonitoring was successfully obtained in 50 patients (68%). The average length of hospitalization was 8.0 days (range: 3 to 47). There were six major complications (8%), which included unilateral lower extremity weakness, which fully resolved (one patient); persistent wound drainage (one patient); postoperative hematuria (one patient); hypertensive crisis, which was corrected (one patient); postoperative wound infection (one patient); and postoperative subarachanoid hemorrhage (one patient). The distribution of surgical approaches, additional procedures, operative data, distribution of anchor type, and intraoperative monitoring can be found in Table 2. The total cost for a patients’ primary procedure was found to be $50,096 ± $23,998. The highest individual cost was for implants, with a mean cost of $13,916.

         The cost of implants accounted for 24% of the total costs. The second highest individual cost was for inpatient room and ICU costs, with a mean of $12,483 (22% of total cost). Bone grafts were the third highest individual cost with a mean of $6,398 (11% of total cost). The breakdown of the individual costs and their respective percentage of total cost can be found in Table 3. Additionally, multivariate linear regressions were completed to demonstrate predictive factors for increased costs (Table 4). Increased major and minor structural curve, increased total (A/P) levels fused, and increased length of hospital stay predicted an increase in total costs. Discussion Socioeconomic pressures due to increasing healthcare costs, diminishing resources, and the imperative to provide healthcare to all elicits the need to focus on cost effective and cost efficient delivery of care.

         This requires careful analysis of the factors involved in costs for healthcare services provided. It should be noted that the term costs is erroneously used throughout the literature.15,16,18 Costs are defined as the resources that are consumed in order to produce a product.15 There are many ways that costs can be ascertained depending on the cost centers that are used (i.e., weighted procedure cost, per hour cost, surcharge cost, or per diem cost). In our study, we chose to report the cost of care for the hospitalization. For this study, we performed a cost analysis for consecutive NMS surgeries at a single hospital between 2006 and 2007. We were able to characterize the relative contributions of various items and services to total costs incurred during hospitalization.

          Not surprisingly, we found that implant costs comprised the largest single contributing factor to overall Table 1 Demographic Data Demographics Number (%) Median (Range) Age (yrs.) 15.8 ± 7.3 Males (% of group) 42 (57) Females (% of group) 32 (43) Height (cm) 148 ± 19.7 Weight (kg) 44.5 ± 19 BMI (kg/m2) 20.1 ± 7.14 Comorbid Conditions Number (%) Cerebral Palsy 21 (28) Familial Dysautonomia 10 (14) Other† 43 (58) *Plus-minus values are means ± SD. Numbers in parentheses are (%) of total. †Includes Asperger’s syndrome, Autistic-like behavior with scoliosis, Congenital encephalopathy, Congenitial insensitivity to pain with anhidrosis (HSAN type IV), Duchene muscular dystrophy, Facioscapulohumeral Muscular Dystrophy, Muscular Dystrophy, Paraplegia secondary to spinal cord tumor resection, Polio, Prader-Willi, Rett’s Syndrome, Seizure Disorder, Spina bifida, Spinal Muscular Atrophy Type 2, Syndromic scoliosis, Unspecified neuromuscular disorder, and VATER Table 2 Surgical Data Category Number (%) Median (Range) Operative approach Posterior 57(77) Combined 17(23) Procedural variables Posterior levels fused 12.9 ± 4.0 Anterior levels fused 5.1 ± 0.9 Operative time (hrs) 6.4 ± 2.5 Estimated blood loss (mL) 985.7 ± 800 Bone graft 67(91) Autograft 37(50) Allograft 62(84) Fluoroscopy 36(49) Neuromonitoring 50(68) Somatosensory evoked potential 49(66) Motor evoked potential 23(31) Instrumentation Hooks† 4 (5) Mean per patient 9.5 ± 2.1 Screws‡ 55(75) Mean per patient 21.6 ± 7.1 Hybrid 15 (20) Hospitalization Length of stay (days) 8.03 ± 6.5 *Plus-minus values are means ± SD. Values in parentheses are (%) of patients or procedures. †Patients with >80% hook fixation only. ‡Patients with >80% screw fixation only. Bulletin of the Hospital for Joint Diseases 2013;71(4):00-00 275 cost, accounting for 24%, followed by inpatient and intensive care unit costs (21.6%), bone graft materials (11.1%), and operating room costs (9.7%).

          While market forces will likely continue to drive the cost of implants, collaborative efforts between industry, individual hospitals, and physicians are needed to help curtail rising costs while maintaining the continued viability and productivity of each of these stakeholders. Competitive group pricing allows for reduced cost per implant and is a strategy commonly employed at our institution. Such an arrangement, while potentially cost saving in the short-term, may limit the surgeon’s ability to choose the implant system that is most appropriate for the patient and may not allow the most efficient performance of an operation by the individual surgeon if the highest quality implant systems are not utilized. The data provided herein provides a framework and allows for comparison across hospital systems, with the understanding that geographic and regional market differences will influence the cost, charge, and reimbursement differential between different hospitals.19 The type of insurer also comprises a significant element; i.e., government versus private and DRG-related bundled payments vs. itemized reimbursements.

         Multivariate linear regression identified four independent variables that predicted higher cost: degree of major structural curve, the degree of minor structural curve, the total number of anterior/posterior levels fused, and the length of hospital stay. Surgery for larger curves is inherently more extensive than for smaller single curves and requires longer ICU and overall hospital stays. Additionally, as curve magnitude increases, the total number of implants required to achieve and maintain correction of the curve increases. Seventy five percent of our patients had all-pedicle screw constructs, while 20% received hybrid constructs and 5% obtained hook constructs. Several studies have compared implant cost differences among hook, hybrid constructs, and pedicle screw constructs.

          Storer and colleagues demonstrated that hook and sublaminar wire constructs are less costly than pedicle screw constructs in AIS patients, while providing similar correction in the sagittal and coronal planes.20 Cheng and associates demonstrated that sublaminar wires provided similar correction to all-pedicle screw constructs but had more blood loss and longer fusion lengths.21 Kim and coworkers recently showed that compared with hybrid constructs, all-pedicle screw constructs produced greater radiographic curve correction and improved pulmonary function, but carried with it higher implant costs averaging $14,200.22 Two other recent studies have shown similar radiographic and clinical outTable 3 Costs by Category Following Surgery for Neuromuscular Scoliosis* Category Cost Percent of Total Cost† Implants $13,916± 10,087 24.0% Room/ICU $12,483± 10,177 21.6% Graft $6,398± 5,971 11.1% OR $5,618± 2,754 9.7% Housestaff $3,440± 4,065 5.9% OR misc $2,489± 1,845 4.3% Drugs $1,771± 1,289 3.1% Cell saver $1,518± 861 2.6% Housekeeping $1,437± 1,195 2.5% Vent $1,313± 1,545 2.3% Laboratory $1,029± 769 1.8% OR instruments $919± 845 1.6% Misc $896± 707 1.5% X-ray $799± 630 1.4% Neuro monitoring $790± 681 1.4% Anesthesia Equipment $730± 545 1.3% PT/OT/SPL $624± 637 1.1% Recovery $613± 282 1.1% Blood bank $523± 759 0.9% Pathology $270± 377 0.5% Pain management/PCA $163± 73 0.3% Fluoro $159± 83 0.3% *Plus-minus values are means ± SD. All units are $US unless otherwise specified. †Percent of sum total mean categorical costs, not percent of mean overall total. Table 4 Predictors of Increasing Costs (N = 62)* Category Predictor Coefficient (β, $US/unit) P value† Pearson r Cost* Major Structural curve (degrees) $2,262 0.0274 0.36 Minor Structural curve (degrees) $2,603 0.042 0.34 Total A/P levels fused (no.) $14,760 0.0023 0.38 Length of Stay (days) $4,313 < 0.0001 0.85 *Variables not found to correlate significantly with costs include age, body mass index, gender, Pre-op kyphosis, number of posterior levels fused, number of anterior levels fused, estimated blood loss, total operative time, number of screws or hooks or wires. †P values represent significance of linear correlation coefficients.

          276 Bulletin of the Hospital for Joint Diseases 2013;71(4):00-00 comes when comparing hybrid versus all-screw constructs and called into question whether or not these more expensive constructs should be considered the new gold standard.23,24 The relative contribution of implants to costs will vary by institution based on contracting agreements between the hospital and implant vendors and current trends have hooks priced greater than screws. Innovative pricing strategies and volume discounts have been helpful in curtailing the overall implant component costs at our institution. A study published in 2010 from our institution analyzed the hospital costs of adolescent idiopathic scoliosis (AIS) correction surgery.16 There are several similarities and interesting key differences between the current study and findings in the AIS population. The AIS study found the highest percentages of total cost derive from implants (29%), ICU and inpatient bed costs (22%), operating room costs (9.9%), and bone grafts (6%). The same variables provided the highest cost in the NMS population; however, bone grafts comprised a larger percentage of total cost even though only 91% of NMS patients had supplemental bone graft placed as opposed to 100% of AIS patients.

         The quantity of bone graft utilized per case may account for this difference. The AIS study found approach (anterior/combined), number of pedicle screws, and number of vertebral levels fused to be significant predictors of increased cost, whereas the current study found degree of major and minor structural curves, total number of anterior/posterior levels fused, and length of hospital stay to predict increased cost. Length of stay (LOS) predicted cost in NMS patients and not AIS patients likely because the average LOS was 8 ± 6.5 versus 4.6 ± 1.2, respectively; longer mean LOS and larger standard deviation for the NMS group may account for this. Major structural curve magnitude predicted cost in the NMS population and not AIS, likely because of the larger average major curve magnitude, 60° versus 50°, respectively as well as a larger standard deviation in the former group. The current study was limited in that we did not report costs associated with the professional component, i.e., the anesthesiologist and surgeons. Reimbursement rates likely vary from surgeon to surgeon, by geographic location, by insurer, and by institution. Future studies should assess the costs associated with the various scoliosis construct types in both the short term (i.e., perioperative hospital costs) and in the long term (revision rates, costs associated with re-hospitalization and re-operation). Such a comparison, however, was beyond the focus of the current study.

          A multicenter cost analysis from different geographic regions may highlight the inevitable disparities and variations in practice from region to region and from country to country. Future analysis may provide strategies to optimize resource allocation and to reduce costs. It is our hope that the current study may help identify potential areas where such improvements can be made. NMS patients are a medically fragile and challenging patient population to treat, thus demanding increased healthcare needs and associated costs. This report is the first to quantify the relative contributions from the myriad of factors that comprise surgical and hospital costs in the surgical treatment of NMS. We further identified four statistically significant independent predictors of higher cost: the degree of major structural curve, the degree of minor structural curve, the total number of anterior/posterior levels fused, and the length of hospital stay. Disclosure Statement None of the authors have a financial or proprietary interest in the subject matter or materials discussed, including, but not limited to, employment, consultancies, stock ownership, honoraria, and paid expert testimony.

References

1. Mercado E, Alman B, Wright JG. Does spinal fusion influence quality of life in neuromuscular scoliosis? Spine (Phila Pa 1976). 2007 Sep 1;32(19 Suppl):S120-5.

2. Driscoll SW, Skinner J. Musculoskeletal complications of neuromuscular disease in children. Phys Med Rehabil Clin N Am. 2008 Feb;19(1):163-94, viii.

3. Benson ER, Thomson JD, Smith BG, Banta JV. Results and morbidity in a consecutive series of patients undergoing spinal fusion for neuromuscular scoliosis. Spine (Phila Pa 1976). 1998 Nov 1;23(21):2308-17; discussion 2318.

4. Edwards BT, Zura R, Bertrand S, et al. Treatment of neuromuscular scoliosis with posterior spinal fusion using the Galveston technique: a retrospective review and results of 62 patients. J Long Term Eff Med Implants. 2003;13(6):437-44.

5. Gill I, Eagle M, Mehta JS, et al. Correction of neuromuscular scoliosis in patients with preexisting respiratory failure. Spine (Phila Pa 1976). 2006 Oct 1;31(21):2478-83.

6. Sarwark J, Sarwahi V. New strategies and decision making in the management of neuromuscular scoliosis. Orthop Clin North Am. 2007 Oct;38(4):485-96, v.

7. Gregg FO, Zhou H, Bertrand SL. Treatment of neuromuscular scoliosis with posterior spinal fusion using the galveston procedure: retrospective of eight years of experience with unit rod instrumentation. J Long Term Eff Med Implants. 2012;22(1):11-5.

8. Copley LA, Richards BS, Safavi FZ, Newton PO. Hemodilution as a method to reduce transfusion requirements in adolescent spine fusion surgery. Spine (Phila Pa 1976). 1999 Feb 1;24(3):219-22; discussion 223-4.

9. Ho C, Sucato DJ, Richards BS. Risk factors for the development of delayed infections following posterior spinal fusion and instrumentation in adolescent idiopathic scoliosis patients. Spine (Phila Pa 1976). 2007 Sep 15;32(20):2272-7.

10. Neilipovitz DT, Murto K, Hall L, et al. A randomized trial of tranexamic acid to reduce blood transfusion for scoliosis surgery. Anesth Analg. 2001Jul;93(1):82-7.

11. Ridgeway S, Tai C, Alton P, et al., Pre-donated autologous blood transfusion in scoliosis surgery. J Bone Joint Surg Br. 2003 Sep;85(7):1032-6.

12. Schwartz DM, Auerbach JD, Dormans JP, et al. Neurophysiological detection of impending spinal cord injury during scoliosis surgery. J Bone Joint Surg Am. 2007 Nov;89(11):2440- 9. Bulletin of the Hospital for Joint Diseases 2013;71(4):00-00 277

13. Bible JE, Lee RS, Friedlaender GE. The need for increased access to the U.S. health-care system. J Bone Joint Surg Am. 2009 Feb;91(2):476-84.

14. Weatherly, L.A., The rising cost of health care: strategic and societal considerations for employers. . HR Magazine. September, 2004.

15. Finkler SA. The distinction between cost and charges. Ann Intern Med. 1982 Jan;96(1):102-9.

16. Calderone RR, Garland DE, Capen DA, Oster H. Cost of medical care for postoperative spinal infections. Orthop Clin North Am. 1996 Jan;27(1):171-82.

17. Kamerlink JR, Quirno M, Auerbach JD, et al. Hospital cost analysis of adolescent idiopathic scoliosis correction surgery in 125 consecutive cases. J Bone Joint Surg Am. 2010 May;92(5):1097-104.

18. Whitecloud TS 3rd, Roesch WW, Ricciardi JE. Transforaminal interbody fusion versus anterior-posterior interbody fusion of the lumbar spine: a financial analysis. J Spinal Disord. 2001 Apr;14(2):100-3.

19. Gray DT, Hollingworth W, Onwudiwe N, Jarvik JG. Costs and state-specific rates of thoracic and lumbar vertebroplasty, 2001-2005. Spine (Phila Pa 1976). 2008 Aug 1;33(17):1905- 12.

20. Storer, SK, Vital MG, Hyman JE, et al. Correction of adolescent idiopathic scoliosis using thoracic pedicle screw fixation versus hook constructs. J Pediatr Orthop. 2005 JulAug;25(4):415-9.

21. Cheng I, Kim Y, Gupta MC, et al. Apical sublaminar wires versus pedicle screws--which provides better results for surgical correction of adolescent idiopathic scoliosis? Spine (Phila Pa 1976). 2005 Sep 15;30(18):2104-12.

22. Kim YJ, Bridwell KH, Lenke LG, et al. Pseudarthrosis in long adult spinal deformity instrumentation and fusion to the sacrum: prevalence and risk factor analysis of 144 cases. Spine (Phila Pa 1976). 2006 Sep 15;31(20):2329-36.

23. Lowenstein JE, Matsumoto H, Vitale MG, et al. Coronal and sagittal plane correction in adolescent idiopathic scoliosis: a comparison between all pedicle screw versus hybrid thoracic hook lumbar screw constructs. Spine (Phila Pa 1976). 2007 Feb 15;32(4):448-52.

24. Vora V, Crawford A, Babekhir N, et al. A pedicle screw construct gives an enhanced posterior correction of adolescent idiopathic scoliosis when compared with other constructs: myth or reality. Spine (Phila Pa 1976). 2007 Aug 1;32(17):1869-74