Otology & Neurotology 38:147–151 ß 2016, Otology & Neurotology, Inc.
Cost Analysis of Cerebrospinal Fluid Leaks and Cerebrospinal Fluid Leak Prevention in Patients Undergoing Cerebellopontine Angle Surgery Alexander Chern, Jacob B. Hunter, and Marc L. Bennett The Otology Group of Vanderbilt University, Department of Otolaryngology—Head and Neck Surgery, Vanderbilt University Medical Center, Nashville, Tennessee
Objective: To determine if cranioplasty techniques following translabyrinthine approaches to the cerebellopontine angle are cost-effective. Study Design: Retrospective case series. Patients: One hundred eighty patients with available financial data who underwent translabyrinthine approaches at a single academic referral center between 2005 and 2015. Intervention: Cranioplasty with a dural substitute, layered fat graft, and a resorbable mesh plate secured with screws Main Outcome Measures: billing data was obtained for each patient’s hospital course for translabyrinthine approaches and postoperative cerebrospinal fluid (CSF) leaks. Results: One hundred nineteen patients underwent translabyrinthine approaches with an abdominal fat graft closure, with a median cost of $25759.89 (range, $15885.65 – $136433.07). Sixty-one patients underwent translabyrinthine approaches with a dural substitute, abdominal fat graft, and a resorbable mesh for closure, with a median cost of $29314.97 (range, $17674.28–$111404.55). The median cost of a CSF leak was $50401.25 (range, $0–$384761.71). The
additional cost of a CSF leak when shared by all patients who underwent translabyrinthine approaches is $6048.15. The addition of a dural substitute and a resorbable mesh plate after translabyrinthine approaches reduced the CSF leak from 12 to 1.9%, an 84.2% reduction, and a median savings per patient of $2932.23. Applying our cohort’s billing data to previously published cranioplasty techniques, costs, and leak rate improvements after translabyrinthine approaches, all techniques were found to be cost-effective. Conclusion: Resorbable mesh cranioplasty is cost-effective at reducing CSF leaks after translabyrinthine approaches. Per our billing data and achieving the same CSF leak rate, cranioplasty costs exceeding $5090.53 are not cost-effective. Key Words: Acoustic neuroma—Cerebellopontine angle surgery—Cerebrospinal fluid leak—Costeffectiveness—Resorbable mesh cranioplasty—Vestibular schwannoma.
As the American healthcare system changes to a bundled payment system, attaining high value for patients will become paramount for healthcare delivery. Value in healthcare can be defined as outcomes achieved per dollar spent, and can be measured across multiple dimensions, such as survival and complication rates. By improving value, patients, providers, payers, and suppliers can mutually benefit while simultaneously maintaining an economically sustainable healthcare system. Thus,
thoroughly measuring and improving value can help drive progress in the American healthcare system, which currently spends significantly more per capita on healthcare than other developed countries, with no appreciable improvement in health outcomes (1,2). Given these issues, we sought to review the costs associated with our experience managing cerebrospinal fluid (CSF) leaks after vestibular schwannoma resections. In general, CSF leaks can result from either anterior or lateral skull base defects, manifesting as rhinorrhea, otorrhea, or meningitis. Many procedures, including cerebellopontine angle (CPA) tumor excision, can be complicated by CSF leaks, requiring significant costs and effort to manage. When postoperative CSF leaks do occur, patients have a prolonged hospital stay, increased overall cost of treatment, and an increased risk of meningitis (3). Until recently, placing an abdominal fat graft (AFG) in the surgical defect after CPA tumor surgery had been considered the standard of care for minimizing
Otol Neurotol 38:147–151, 2017.
Address correspondence and reprint requests to Marc L. Bennett, M.D., Department of Otolaryngology—Head and Neck Surgery, The Bill Wilkerson Center for Otolaryngology and Communication Sciences, 7209 Medical Center East, South Tower 1215 21st Avenue South, Nashville, TN 37232-8605; E-mail: [email protected]
Internal departmental funding was used without commercial sponsorship or support. Institutional review board approval: Data Integrated Study Console of Vanderbilt’s Research Enterprise (DISCOVR-E) IRB-141149. The authors disclose no conflicts of interest. DOI: 10.1097/MAO.0000000000001252
Copyright © 2016 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
A. CHERN ET AL.
postoperative CSF leaks after CPA surgery. Depending on the closure technique, there is still a CSF leak rate in 6 to 17% of cases after CPA surgery (4–6). To improve upon these outcomes, novel cranioplasty techniques have been developed to further reduce the CSF leak rate. These include using hydroxyapatite cement alone or in combination with a molded titanium mesh over the AFG (7,8). Our institution places a resorbable mesh plate over the AFG, which reduced our CSF leak rate after translabyrinthine craniectomies from 12% with an AFG to 1.9% with the additional resorbable mesh cranioplasty (3,9). By employing such novel cranioplasty techniques that reduce postoperative CSF leak rate, we can lower the average cost of translabyrinthine CPA tumor excision. However, these recently described cranioplasties also add expense compared with the traditional AFG, since they require additional materials (i.e., resorbable mesh). We hypothesize that applying our cranioplasty technique, using a resorbable mesh and dural substitute, to every translabyrinthine closure costs less than the sum of costs to manage all the CSF leaks that the cranioplasty technique prevents. Thus, recognizing that the changing landscape of the American healthcare system is leading toward a greater discussion toward value in healthcare, as well as the paucity of literature regarding costs associated with postoperative CSF leaks, we sought to determine if the resorbable mesh cranioplasty is cost-effective compared with the traditional AFG in preventing postoperative CSF leaks. METHODS Following approval from the Vanderbilt University Institutional Review Board (141149), a retrospective chart review was conducted. All patients 18 years and older who underwent translabyrinthine approaches to resect CPA tumors at Vanderbilt University Medical Center between June 2005 and April 2015 were identified. Patient variables that were analyzed included age at the time of surgery, and sex as determined from the largest single dimension on preoperative MR imaging, body mass index (BMI), and medical comorbidities, including diabetes and tobacco use. Patients were defined as diabetics if they took diabetic medication. Analyzed postoperative variables included diagnosis of CSF leak, length of subsequent hospital stay, and medications and or procedures used to manage the CSF leak (i.e., overclosure, lumbar drain, reexploration, ventriculoperitoneal shunt).
Surgical Technique Both our translabyrinthine surgical approach and closure technique have been previously described (10). To summarize the closure technique using the resorbable mesh plate, after tumor resection, the middle ear is tightly packed with autologous muscle, occasionally interspersed with Surgicel (Johnson & Johnson, New Brunswick, NJ, U.S.A.). An areolar tissue graft is then placed over the attic and posterior tympanotomy, if present. A 1 2-inch piece of Duraform (DePuys Synthes, West Chester, PA, U.S.A.) is placed, reconstituting the posterior fossa and internal auditory canal dura. Fat, harvested from the lower left abdominal wall and cut into 5 cm 1 cm strips, are placed lengthwise over the dural defect, lateral to the dural substitute, until the craniotomy site is filled. Afterward, either a 5 5-cm Rapid Resorbable Fixation System (DePuys Synthes,
West Chester, PA, U.S.A.) or ResorbX (KLS Martin, Jacksonville, FL, U.S.A.) is molded with a warm sponge to place a small dent in the middle of the mesh to provide increased counter pressure to the fat graft. This is then placed to close the bony defect. The mesh is secured at a minimum of three locations using titanium screws with an underlying goal to stop transmitted pulsations from the underlying AFG. The subcutaneous tissue and skin are closed in three layers, with the skin closed using either a running-locking nylon stitch or staples. The head is wrapped with a pressure dressing until discharge from the hospital or postoperative Day 4, whichever occurs first. All patients had their suture or staples routinely removed no sooner than 14 days after their surgery.
Cost Analysis Itemized cost data for each patient’s hospital course who underwent translabyrinthine CPA tumor surgery was obtained from Vanderbilt University Medical Center (VUMC); these costs included anesthesia, surgeon’s fee, materials (i.e., resorbable mesh plate and dural substitute), and cost of hospitalization, including medications during their initial hospital stay after the tumor excision. The cost of each translabyrinthine CPA tumor surgery was defined as the sum of these itemized costs. Itemized cost data was also obtained for the subset of patients who developed postoperative CSF leaks; these costs included the emergency room visit, antibiotics, outpatient visits, hospital stay, and any intervention (i.e., lumbar drain, reexploration, ventriculoperitoneal shunt insertion). The cost of a CSF leak repair was defined as the sum of these itemized costs. To determine if the resorbable mesh and dural substitute cranioplasty was cost-effective, the median savings per patient was calculated with the addition of the resorbable mesh and dural substitute cranioplasty after translabyrinthine tumor surgery using the following equation: S¼PL – C S is defined as the median savings (dollars) per patient. P is defined as the percentage relative leak rate reduction with the addition of a resorbable mesh plate and dural substitute. To calculate P, we used previously described CSF leak rates after translabyrinthine CPA tumor resections with AFG closures with and without resorbable mesh cranioplasties from our institution, 1.9% and 12%, respectively (9). Thus, (12–1.9%)/ 12% ¼ 0.842, or an 84.2% relative leak rate reduction. L is defined as the additional cost of a CSF leak when shared by all patients who underwent translabyrinthine CPA tumor surgery with AFG closure. This was determined by adding the total cost of all CSF leak repairs for all patients, as reported from the itemized cost data, divided by all patients who had a translabyrinthine CPA tumor resection. C is defined as the extra cost of a resorbable mesh cranioplasty in one patient. To calculate C, the itemized cost data was used to determine the cost of a resorbable mesh plate and dural substitute. Since two different types of resorbable mesh plates were used, a weighted average was calculated to represent the cost of a resorbable mesh plate and dural substitute, to approximate the extra cost of a resorbable mesh cranioplasty. Solving for the above equation when the median savings per patient (S) is zero determines the price limit at which a CSF leak repair, C, would not be cost-effective.
Statistical Analysis Patients were separated into two cohorts based whether an AFG was used alone or in combination with the additional dural
Otology & Neurotology, Vol. 38, No. 1, 2017
Copyright © 2016 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
COST-EFFECTIVENESS OF RESORBABLE MESH CRANIOPLASTY
substitute and resorbable mesh to close the surgical defect after a translabyrinthine craniectomy. Continuous features were summarized with medians and ranges. Categorical features were summarized with percentages. The x2 test was used to compare categorical features between the two cohorts, while t tests were used to compare continuous features. P values