- Email: [email protected]
Cost Analysis of Endovascular Coiling and Surgical Clipping for the Treatment of Ruptured Intracranial Aneurysms
Original Article
Cost Analysis of Endovascular Coiling and Surgical Clipping for the Treatment of Ruptured Intracranial Aneurysms Daniel Monsivais1, Miriam Morales1, Arthur Day1, Dong Kim1, Brian Hoh2, Spiros Blackburn1
BACKGROUND: Cost-effectiveness analyses for the treatment of aneurysmal subarachnoid hemorrhage are necessary to determine health policy, treatment guidelines, and screening protocols for cerebral aneurysms. To perform these modeling studies, detailed cost data are necessary and are currently lacking.
-
CONCLUSIONS: The detailed cost information reported in this article can be used to help establish appropriate, standardized costs nationally by improving transparency. It can also help provide critical information necessary to develop cost-effective treatment algorithms and screening protocols.
-
OBJECTIVE: The goal of this study was to determine detailed inpatient cost of aneurysmal subarachnoid hemorrhage.
-
-
INTRODUCTION
RESULTS: There were 269 patients treated, 209 were coiled and 60 were clipped. Mean age in the clipping group was 49 years and 55 years in the coil group (P [ 0.006). Other patient demographics and clinical characteristics were found to be statistically similar for both groups. Total cost per patient for treatment and hospital stay was $74,192 for clipping and $85,553 for coiling (P [ 0.06). Cost amplified with increasing Hunt and Hess grade in both clipping and coiling groups.
ubarachnoid hemorrhage (SAH) owing to ruptured cerebral aneurysms is associated with significant morbidity, mortality, and economic burden. There are approximately 30,000 cases occurring annually in the United States.1 Sixty percent of those cases are either fatal or the patient is left severely disabled, and the associated yearly economic burden is $1.75 billion.2 A cost-effective analysis (CEA) describes and contrasts costs and outcomes for the course of events that are expected to occur with a specific intervention, and the expected course of events without the intervention.3 Such studies are important as they help determine guidelines for treatment and preventative care. To better understand the cost impact of SAH on society and perform these analyses, detailed cost data are necessary. To date, there is 1 large scale study examining treatment for SAH patients in the United States, and this was published prior to widespread use of endovascular management of cerebral aneurysms.4 Although there are recent U.S. data on hospital charges and relative costs associated with treating patients with SAH, no
METHODS: A retrospective review of our ruptured subarachnoid hemorrhage database was performed to identify consecutive patients between January 2013 and December 2015. Patients were searched by International Classification of Disease 9 diagnosis and procedure codes. Patient demographics and clinical characteristics were acquired. The cost breakdown was compiled into a comprehensive itemized list encompassing all aspects of hospitalization. A mean cost based on resource used per patient was obtained.
-
Key words Cost-effective analysis - Endovascular coil - Intracranial aneurysm - Neurosurgical clipping - Quality adjusted life years - Subarachnoid hemorrhage -
Abbreviations and Acronyms CEA: Cost-effective analysis GOSE: Extended Glasgow Outcome Score HH: Hunt and Hess ISAT: International Subarachnoid Aneurysm Trial LOS: Length of stay
WORLD NEUROSURGERY -: e1-e6, - 2019
S
OR: Operating room SAH: Subarachnoid hemorrhage From the 1Department of Neurosurgery, University of Texas Health Science Center at Houston, Houston, Texas; and 2Lillian S. Wells Department of Neurosurgery, University of Florida, Gainesville, Florida, USA To whom correspondence should be addressed: Daniel Monsivais, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2018.12.028 Journal homepage: www.journals.elsevier.com/world-neurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.journals.elsevier.com/world-neurosurgery
e1
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
publication in the last 20 years has examined cost as determined from a societal perspective.5 There is recent literature that has examined cost using hospital charges as a surrogate marker of economic cost.6-9 However, cost values derived from hospital charges may not be reliable for CEA.10 Other countries have examined cost with the intention of facilitating CEA.11 In the United Kingdom, 1 study (a subgroup of International Subarachnoid Aneurysm Trial [ISAT] patients) examined hospital costs and found that there was no significant difference in costs between coiling and clipping in patients with SAH.12 To our knowledge, we provide the first report comparing resource usage and cost associated with each treatment modality using a detailed itemized lists of cost associated with resource consumption, rather than hospital charges or billing, that is a more accurate way to track true economic cost.10 With the evolution and proliferation of endovascular SAH treatment in the last 2 decades, updated cost information is necessary to direct health policy and standardization of care. These data can be used for outcomes related research and cost studies. METHODS After institutional review board approval, we performed a retrospective chart review from our Comprehensive Stroke Center database. The database was queried for patients with ruptured cerebral aneurysms treated between January 1, 2013 and December 31, 2015 at Memorial Hermann Hospital in the Texas Medical Center in Houston, Texas. This facility is a volume, tertiary, academic center. Patients were searched using SAH International Classification of Disease 9 diagnosis and procedure codes. Patient demographics (age, sex), aneurysm characteristics (size, anterior or posterior circulation), Hunt and Hess clinical grade (HH), treatment (surgical clip vs. endovascular coil), discharge disposition (home, rehabilitation, skilled nursing facility, long-term acute care hospital), and 6-month extended Glasgow Outcome Scores (GOSE) were acquired from the clinical database. The decision regarding the treatment modality to be selected for each patient was done based on the aneurysm geometry and the individual patient characteristics. For example, older patients with multiple medical co-morbidities and patients with higher HH grades were often treated with coiling. Patients with an aneurysm in which the shape and size were not technically ideal for endovascular treatment would often undergo surgical clipping, particularly if patients were younger, healthier and were a lower HH grade. Patients who were treated with clipping and coiling in combination were excluded. Patient privacy was protected in compliance with the Health Insurance Portability and Accountability Act. No identifying patient information was used in the study and therefore, individual patient consent was not obtained. The cost per patient was determined from our hospital resource usage database and accounting system. Cost was determined by detailed hospital itemization of consumables, not charges. Direct costs were defined as those directly attributable to a patient’s stay in the hospital and defined on an itemized list provided by the
e2
www.SCIENCEDIRECT.com
hospitals financial database. This list included the following: hospital room costs for the intensive care unit and wards (includes physician and nursing charges), operating room (OR) costs, respiratory care, medical supplies, medications, laboratory tests, and imaging studies. The neurosurgical procedure (clipping and coiling) costs included cost of supplies (sterile supplies, drapes, surgifoam, suture), cost of OR time and angiography suite time, and staff costs (surgeon, anesthesia, and nursing required for the procedure). Angiography consumables included vascular radiology, catheter, coil, glide wire, monitors, introducer set, and sheath. The ancillary service cost included physical therapy, occupational therapy, speech therapy, cardiology services, respiratory therapy, and emergency services. Additionally, nonneurosurgical procedures (tracheostomy, percutaneous gastrostomy tubes, inferior vena cava filters) were included in the total cost for each patient. Overhead costs were provided by our hospital financial institution using EPSI software (Allscripts Healthcare, Chicago, Illinois, USA). EPSI is used for financial data analysis by healthcare companies. Overhead costs were defined as those necessary to keep the hospital building functional—for example electrical power, running water, janitorial and maintenance services. The sum of the direct cost and overhead cost was determined to be the total cost. The total cost per patient differences were explored between clipping and coiling groups across patient HH grade, demographic, and clinical information. Age, length of stay (LOS), days on the floor, intermediate care unit, and intensive care unit were explored by clipping and coiling groups. Two group non-parametric statistical differences used the Wilcoxon rank sum test to examine clipping and coiling direct cost, overhead cost, total cost, and supplies cost differences. The KruskalWallis tests were used to examine cost differences by HH grade. Statistical differences used parametric or non-parametric tests, where appropriate, and 2-tailed P values <0.05 were considered significant. All statistical analyses were conducted in Stata SE software version 12 (StataCorp, College Station, Texas, USA). RESULTS Between January 1, 2013 and December 31, 2015, there was a total of 269 consecutive patients with SAH treated by clipping or coiling. Two hundred nine patients were treated by coiling, 60 patients were treated by clipping, and 7 patients were treated by both (this cohort was excluded). Patient demographics, HH grade, aneurysm size, LOS, and disposition were directly compared for both groups and found to be statistically similar. There was a statistically younger mean age in the clipping group (49 vs. 55 years, P ¼ 0.006; Tables 1 and 2). Total cost per patient for treatment and hospital stay was $74,192 in the clipping group and $85,553 in the coiling group (P ¼ 0.059; Table 3). Direct cost was $54,182 for patients in the clipping group and $65,601 (P ¼ 0.007) for patients in the coiling group. Overhead cost was approximately equal from both treatment groups, $20,010 in the clipping group versus $19,993 in the coiling group (P ¼ 0.47).
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.12.028
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
Table 1. Demographics Clipping Number [ 60
Coiling Number [ 209
Total Number [ 269
37 (62)
144 (69)
181
23
65
88
49 (14)
55 (14)
54 (14)
Black
11 (18)
42 (20)
53
White
23 (38)
91 (44)
114
4 (7)
7 (3)
11
All other
21 (35)
64 (31)
85
Unknown
1 (2)
5 (2)
6
Grade I
4 (6.7)
15 (7.2)
19
Grade II
29 (48.3)
77 (36.8)
106
Grade III
20 (33.3)
69 (33)
89
Grade IV
5 (8.3)
34 (16.3)
39
Grade V
2 (3.3)
13 (6.2)
15
<5 mm
29
76
105
5e10 mm
26
112
142
10e15 mm
4
18
22
15e20 mm
1
2
3
20e25 mm
0
1
1
P value
Characteristics Sex Female Male Age
0.293
0.0061
Race
Asian Pacific Islander
0.594
Hunt and Hess grade 0.401
Aneurysm size 0.261
Data demonstrates patient demographics including race, age, and sex. Clinical characteristics including Hunt and Hess grade, aneurysm size, and the treatment modality used to treat (surgical clipping vs. endovascular coiling). The P value for aneurysm size (0.261) only includes the 3 categories (<5 mm, 5e10 mm and 10e15 mm) because the other categories had an N that was too low to consider for P value comparisons with the other categories.
Table 2. Length of Stay by Mean, Standard Deviation, 25th Percentile, Median, and 75th Percentile Characteristic Average length of stay
Routine days
Intermediate care unit days
Intensive care unit days
Clipping
Coiling
Total
P value (Clip vs. Coil Only) 0.3528
15 (10)
16 (10)
16 (10)
9e13e17.5
10e14e20
10e14e19
2.7 (4.7)
2.7 (4.3)
2.7 (4.3)
0e1.5e3
0e1e3
0e1e3
1.6 (2.7)
1.5 (2.7)
1.5 (2.7)
0e.5e2
0e0e2
0e0e2
10.8 (4.98)
12 (6.1)
11.8 (5.9)
8.5e10e12.5
8e11e15
8e11e14
WORLD NEUROSURGERY -: e1-e6, - 2019
0.7052
0.7006
0.2652
www.journals.elsevier.com/world-neurosurgery
e3
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
Table 3. Mean Cost per Case
Table 5. Cost for Supplies per Case
Procedure
Clipping
Coiling
60
209
Patient volume
P value*
Procedure Patient volume
Clipping
Coiling
60
209
P value*
Direct cost per case
$54,182
$65,601
0.0073
Room (room cost only)
$17,272
$19,386
0.23
Overhead cost per case
$20,010
$19,932
0.456
Hospital services
$1,199
$1,094
0.38
Total cost (direct þ overhead) per case
$74,192
$85,553
0.0592
Hospital supplies
$3,154
$3,500
0.96
Medications
$4,174
$4,225
0.40
Labs
$2,050
$2,351
0.21
*Wilcoxon rank sum test P value.
Cost increased with worse HH grade, except for the grade 5 patient in the clipping group (N ¼ 2; average LOS ¼ 6.0 days; average cost ¼ $36,330; Table 4). This was owing to early deaths related to poor prognosis and withdrawal of care. In the coiling group, cost continued to increase for HH grade 5 patients (N¼14; average LOS ¼ 22.4 days; average cost ¼ $91,933). Excluding the HH grade 5 cohort in both groups for low numbers, the average cost changed slightly for clipping and coiling to $75,105 and $83,545, respectively (P ¼ 0.027), and the average LOS changed to 15.5 and 15.9 days (P ¼ 0.7), respectively. Room cost, hospital services, medications, laboratory tests, intravenous fluids, blood products, and imaging costs were statistically similar for both clipping and coiling groups. There was a statistical difference between angiography consumables ($4,224 vs. $25,872; P < 0.0001), OR consumables ($16,112 vs. $3,786; P < 0.0001), and anesthesia ($2,864 vs. $1,637; P < 0.0001; Table 5). This resulted in a total surgical and procedural cost of $23,200 for the clipping group and $31,295 for the coiling group. LOS and disposition were directly compared for both groups and found to be statistically similar. Clinical outcomes at 6 months were available for 266 patients in the form of GOSE. We found that the mean GOSE for patients who underwent clipping was 6.16 (N ¼ 60) and those who underwent coiling was 6.0 (N ¼ 206). DISCUSSION Optimized use of resources is becoming increasingly important in modern medicine as the cost of health care increases. One method
Table 4. Mean Cost by Hunt and Hess Clinical Grade Procedure
Clipping
Coiling
60
209
Grade 1
$58,208
$42,555
Grade 2
$44,601
$57,685
Grade 3
$66,555
$66,650
Grade 4
$64,177
$83,379
Grade 5
$36,330
$91,933
P value*
0.0308
0.0001
Patient volume
*Kruskal-Wallis test P value.
e4
www.SCIENCEDIRECT.com
IV fluids
$117
$142
0.03
Angiography consumables
$4,224
$25,872
<0.0001
OR consumables
$16,112
$3,786
<0.0001
Anesthesia
$2,864
$1,637
<0.0001
Blood products
$411
$542
0.27
Misc. imaging
$1,167
$1,212
0.88
Misc. radiology
$304
$448
0.63
IV, intravenous; OR, operating room; Misc., miscellaneous. *Wilcoxon rank sum test P value.
to understand the benefit of an intervention in health care is to perform standardized cost analyses correlated with subsequent effects on quality adjusted life years. For this to be done, accurate treatment costs associated with a specific disease entity are required. Although these data have been pursued more deliberately in spine literature, little cost data exists to accurately perform analyses in the cranial realm.13 Our overriding goal for this study was to determine and report accurate cost data. These data are unique from hospital charges or reimbursement—data points that can lead to misleading results if used in cost effectiveness analyses.10,13 This analysis calculated the cost of SAH and coiling was slightly higher than SAH and clipping, although this was not significant, and was $74,192 for clipping and $85,553 for coiling. Although the cost difference is notable, it is our opinion that this in no way implies preference for 1 treatment over another considering that equipoise was not established, and the aneurysms were treated on an individual basis to provide the best clinical outcome. Furthermore, patients undergoing coiling were older than those undergoing clipping, and this practice is consistent with literature, specifically the International Study of Unruptured Intracranial Aneurysms and ISAT, that suggests coiling may be a better treatment modality in older patients.14,15 Despite the older average age of the coiled patients, resource used for hospital admission and LOS was comparable in both coiling and clipping groups, as was clinical outcome. This is a testament to the usefulness of coiling technology in the older population in appropriately selected aneurysms. To get some perspective on the results from this article, we can compare our cost data to a previous publication from 1999, prior to the widespread use of endovascular coiling.5 As reported in 1997 U.S. dollars, the cost of treating an aneurysmal SAH was
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.12.028
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
$47,000.5 Inflating this to 2015 U.S. dollars makes this roughly $70,600, a figure that is comparable to our results for SAH and clipped aneurysms. Contrarily, an analysis of the ISAT costs in the United Kingdom cohort showed that clipping was roughly £20,000 in 2004 (w$46,000 in 2015) for all costs inclusive of 12 months post-SAH. This seems to be dramatically lower than current U.S. costs and may be because of differences in health care systems. These differences highlight the importance of local cost determination for accurate economic analyses. The mean LOS for patients in our study who underwent endovascular coiling was 16 days, which seems to reflect some national trends for SAH in the United States.8,16 There are other U.S. studies that had a slightly longer LOS for patients who underwent endovascular coiling (19.8e20.2 days).6,16 When compared with international data taken from the ISAT trial, our LOS is slightly shorter (16.0 days vs. 19.4 days).15 The overall mean LOS for patients who underwent surgical clipping in our study was 15.0 days, which varied from other studies in which the LOS was between 19.7 to 23 days.4,6,8,16 International data taken from ISAT showed an average LOS of 23.3 days in patients who underwent surgical clipping. In contrast to other studies comparing outcomes of clipping and coiling, in which treatment selection is through randomization, the patients in our article were selected for each treatment based on aneurysm geometry and patient demographics. The similar LOS and outcomes (GOSE at 6 months) among the 2 groups may demonstrate the benefit of optimized treatment selection for each patient. Comparing postcoiling and clipping interventions, our data showed that the cost of posttreatment angiograms during admission did not disproportionately influence the overall cost of treatment in 1 group over the other. We also did not find that there was a difference in the number of posttreatment angiograms during admission between patients who underwent clipping and those that underwent coiling, resulting in no difference in cost between the 2 groups.
2. de Rooij NK, Linn FH, van der Plas JA, Algra A, Rinkel GJ. Incident of subarachnoid hemorrhage: a systematic review with emphasis on region, age, and time trends. J Neurol Neurosurg Psychiatry. 2007; 78:1365-1372. 3. Siegel J, Milton S, Weinsten C, Russel L, Gold M. Recommendations for reporting cost-effective analysis. J Am Med. 1996;276:16. 4. Hoh BL, Chi YY, Lawson MF, Mocco J, Barker FG 2nd. Length of stay and total hospital charges of clipping versus coiling for ruptured and unruptured adult cerebral aneurysms in the
CONCLUSIONS The detailed cost information reported in this article can be used to help establish appropriate, standardized costs nationally by improving transparency. It can also help provide critical information necessary to develop cost-effective treatment algorithms and screening protocols.
Nationwide Inpatient Sample database 2002 to 2006. Stroke. 2010;41:337-342.
cost after cerebral aneurysm clipping. J Neurointerv Surg. 2016;8:316-322.
5. Johnston SC, Dudley RA, Gress DR, Ono L. Surgical and endovascular treatment of unruptured cerebral aneurysms at university hospitals. Neurology. 1999;52:1799-1805.
9. Bekelis K, Gottlieb DJ, Su Y, et al. Medicare expenditures for elderly patients undergoingsurgical clipping or endovascular intervention for subarachnoid hemorrhage. J Neurosurg. 2017;126: 805-810.
REFERENCES 1. Shea A, Reed S, Curtis L, et al. Characteristics of nontraumatic subarachnoid hemorrhage in the United States in 2003. Neurosurgery. 2007;61: 1131-1138.
To investigate why the cost of HH grade 1 patients who underwent clipping was higher than those who underwent coiling, each patient chart was examined individually. There was a total of 4 HH grade 1 patients who underwent surgical clipping. It was determined that 1 of these patients suffered an internal capsule infarct during surgical clip placement. Complications from the stroke contributed to a prolonged hospital stay resulting in higher cost of admission. Another patient suffered an infarct secondary to vasospasm that resulted in complications contributing to a prolonged hospital stay and increased cost of admission. The increased overall cost for admission in these 2 patients resulted in an increased average admission cost for HH grade 1 patients who underwent clipping. There are several limitations of our study. The sample size for clipping is relatively small compared to the coiling group (60 vs. 209), which reflects the practice pattern that more aneurysms are coiled than clipped at our institution. Additionally, we only examined costs for a single admission and did not examine costs for subsequent re-admissions, rehabilitation, nursing home, or long-term acute care hospital, and we did not examine the longterm costs associated with disability. Furthermore, overhead costs may have larger variability as wages, power, water, maintenance, and janitorial costs will differ across facilities and impact the total cost to treat patients. Also not included is the cost of long-term imaging follow-up required for coiled aneurysms, which is likely to be higher than clipped aneurysms. For complete clarification of health care costs in this population of patients, these data would need to be described via a longitudinal follow-up study.
6. Chang H, Shin S, Suh S, Kim B, Rho M. Costeffective analysis of endovascular coiling versus neurosurgical clipping for intracranial aneurysms in Republic of Korea. Neurointervention. 2016;11: 86-91. 7. Brinjikji W, Kallmes D, Lanzino G, Cloft H. Hospitalization costs for endovascular and surgical treatment of ruptured aneurysms in the United States are substantially higher than Medicare payments. Am J Neuroradiol. 2012;33: 1037-1040. 8. Bekelis K, Missios S, MacKenzi T, Labropoulos N, Roberts D. A predictive model of hospitalization
WORLD NEUROSURGERY -: e1-e6, - 2019
10. Finkler S. The distinction between cost and charges. Ann Intern Med. 1982;96:102-109. 11. Bairstow P, Dodgson A, Linto J, Khangure M. Comparison of cost and outcome of endovascular and neurosurgical procedures in the treatment of ruptured intracranial aneurysms. Australian Radiol. 2002;46:249-251. 12. Wolstenhome J, Rivero-Arias O, Gray A, et al. Treatment pathways, resource use, and cost of endovascular coiling versus surgical clipping after aSAH. Stroke. 2008;39:111-119.
www.journals.elsevier.com/world-neurosurgery
e5
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
13. Zygourakis C, Kahn J. Cost-effectiveness research in neurosurgery. Neurosurg Clin N Am. 2015;26:189-196. 14. Wiebers D, Whisnant J, Hutson J, Meissner I, Brown R. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet. 2003; 362:103-110. 15. Ryttlefors M, Per E, Kerr R, Molyneux A. International subarachnoid aneurysm trial of neurosurgical clipping versus endovascular coiling subgroup analysis of 278 elderly patients. Stroke. 2008;39:2720-2726.
e6
www.SCIENCEDIRECT.com
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
16. Hoh B, Chi Y, Dermott M, Lipori P, Lewis S. The effect of coiling versus clipping of ruptured and unruptured cerebral aneurysms on length of stay, hospital cost, hospital reimbursement, and surgeon reimbursement at the University of Florida. Neurosurgery. 2009;64: 614-621.
The content of this manuscript was presented as a poster at the Congress of Neurologic Surgeons 2017 Annual Meeting in Boston, Massachusetts. Received 8 August 2018; accepted 4 December 2018 Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2018.12.028 Journal homepage: www.journals.elsevier.com/worldneurosurgery
Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.12.028
Cost Analysis of Endovascular Coiling and Surgical Clipping for the Treatment of Ruptured Intracranial Aneurysms Daniel Monsivais1, Miriam Morales1, Arthur Day1, Dong Kim1, Brian Hoh2, Spiros Blackburn1
BACKGROUND: Cost-effectiveness analyses for the treatment of aneurysmal subarachnoid hemorrhage are necessary to determine health policy, treatment guidelines, and screening protocols for cerebral aneurysms. To perform these modeling studies, detailed cost data are necessary and are currently lacking.
-
CONCLUSIONS: The detailed cost information reported in this article can be used to help establish appropriate, standardized costs nationally by improving transparency. It can also help provide critical information necessary to develop cost-effective treatment algorithms and screening protocols.
-
OBJECTIVE: The goal of this study was to determine detailed inpatient cost of aneurysmal subarachnoid hemorrhage.
-
-
INTRODUCTION
RESULTS: There were 269 patients treated, 209 were coiled and 60 were clipped. Mean age in the clipping group was 49 years and 55 years in the coil group (P [ 0.006). Other patient demographics and clinical characteristics were found to be statistically similar for both groups. Total cost per patient for treatment and hospital stay was $74,192 for clipping and $85,553 for coiling (P [ 0.06). Cost amplified with increasing Hunt and Hess grade in both clipping and coiling groups.
ubarachnoid hemorrhage (SAH) owing to ruptured cerebral aneurysms is associated with significant morbidity, mortality, and economic burden. There are approximately 30,000 cases occurring annually in the United States.1 Sixty percent of those cases are either fatal or the patient is left severely disabled, and the associated yearly economic burden is $1.75 billion.2 A cost-effective analysis (CEA) describes and contrasts costs and outcomes for the course of events that are expected to occur with a specific intervention, and the expected course of events without the intervention.3 Such studies are important as they help determine guidelines for treatment and preventative care. To better understand the cost impact of SAH on society and perform these analyses, detailed cost data are necessary. To date, there is 1 large scale study examining treatment for SAH patients in the United States, and this was published prior to widespread use of endovascular management of cerebral aneurysms.4 Although there are recent U.S. data on hospital charges and relative costs associated with treating patients with SAH, no
METHODS: A retrospective review of our ruptured subarachnoid hemorrhage database was performed to identify consecutive patients between January 2013 and December 2015. Patients were searched by International Classification of Disease 9 diagnosis and procedure codes. Patient demographics and clinical characteristics were acquired. The cost breakdown was compiled into a comprehensive itemized list encompassing all aspects of hospitalization. A mean cost based on resource used per patient was obtained.
-
Key words Cost-effective analysis - Endovascular coil - Intracranial aneurysm - Neurosurgical clipping - Quality adjusted life years - Subarachnoid hemorrhage -
Abbreviations and Acronyms CEA: Cost-effective analysis GOSE: Extended Glasgow Outcome Score HH: Hunt and Hess ISAT: International Subarachnoid Aneurysm Trial LOS: Length of stay
WORLD NEUROSURGERY -: e1-e6, - 2019
S
OR: Operating room SAH: Subarachnoid hemorrhage From the 1Department of Neurosurgery, University of Texas Health Science Center at Houston, Houston, Texas; and 2Lillian S. Wells Department of Neurosurgery, University of Florida, Gainesville, Florida, USA To whom correspondence should be addressed: Daniel Monsivais, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2018.12.028 Journal homepage: www.journals.elsevier.com/world-neurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.journals.elsevier.com/world-neurosurgery
e1
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
publication in the last 20 years has examined cost as determined from a societal perspective.5 There is recent literature that has examined cost using hospital charges as a surrogate marker of economic cost.6-9 However, cost values derived from hospital charges may not be reliable for CEA.10 Other countries have examined cost with the intention of facilitating CEA.11 In the United Kingdom, 1 study (a subgroup of International Subarachnoid Aneurysm Trial [ISAT] patients) examined hospital costs and found that there was no significant difference in costs between coiling and clipping in patients with SAH.12 To our knowledge, we provide the first report comparing resource usage and cost associated with each treatment modality using a detailed itemized lists of cost associated with resource consumption, rather than hospital charges or billing, that is a more accurate way to track true economic cost.10 With the evolution and proliferation of endovascular SAH treatment in the last 2 decades, updated cost information is necessary to direct health policy and standardization of care. These data can be used for outcomes related research and cost studies. METHODS After institutional review board approval, we performed a retrospective chart review from our Comprehensive Stroke Center database. The database was queried for patients with ruptured cerebral aneurysms treated between January 1, 2013 and December 31, 2015 at Memorial Hermann Hospital in the Texas Medical Center in Houston, Texas. This facility is a volume, tertiary, academic center. Patients were searched using SAH International Classification of Disease 9 diagnosis and procedure codes. Patient demographics (age, sex), aneurysm characteristics (size, anterior or posterior circulation), Hunt and Hess clinical grade (HH), treatment (surgical clip vs. endovascular coil), discharge disposition (home, rehabilitation, skilled nursing facility, long-term acute care hospital), and 6-month extended Glasgow Outcome Scores (GOSE) were acquired from the clinical database. The decision regarding the treatment modality to be selected for each patient was done based on the aneurysm geometry and the individual patient characteristics. For example, older patients with multiple medical co-morbidities and patients with higher HH grades were often treated with coiling. Patients with an aneurysm in which the shape and size were not technically ideal for endovascular treatment would often undergo surgical clipping, particularly if patients were younger, healthier and were a lower HH grade. Patients who were treated with clipping and coiling in combination were excluded. Patient privacy was protected in compliance with the Health Insurance Portability and Accountability Act. No identifying patient information was used in the study and therefore, individual patient consent was not obtained. The cost per patient was determined from our hospital resource usage database and accounting system. Cost was determined by detailed hospital itemization of consumables, not charges. Direct costs were defined as those directly attributable to a patient’s stay in the hospital and defined on an itemized list provided by the
e2
www.SCIENCEDIRECT.com
hospitals financial database. This list included the following: hospital room costs for the intensive care unit and wards (includes physician and nursing charges), operating room (OR) costs, respiratory care, medical supplies, medications, laboratory tests, and imaging studies. The neurosurgical procedure (clipping and coiling) costs included cost of supplies (sterile supplies, drapes, surgifoam, suture), cost of OR time and angiography suite time, and staff costs (surgeon, anesthesia, and nursing required for the procedure). Angiography consumables included vascular radiology, catheter, coil, glide wire, monitors, introducer set, and sheath. The ancillary service cost included physical therapy, occupational therapy, speech therapy, cardiology services, respiratory therapy, and emergency services. Additionally, nonneurosurgical procedures (tracheostomy, percutaneous gastrostomy tubes, inferior vena cava filters) were included in the total cost for each patient. Overhead costs were provided by our hospital financial institution using EPSI software (Allscripts Healthcare, Chicago, Illinois, USA). EPSI is used for financial data analysis by healthcare companies. Overhead costs were defined as those necessary to keep the hospital building functional—for example electrical power, running water, janitorial and maintenance services. The sum of the direct cost and overhead cost was determined to be the total cost. The total cost per patient differences were explored between clipping and coiling groups across patient HH grade, demographic, and clinical information. Age, length of stay (LOS), days on the floor, intermediate care unit, and intensive care unit were explored by clipping and coiling groups. Two group non-parametric statistical differences used the Wilcoxon rank sum test to examine clipping and coiling direct cost, overhead cost, total cost, and supplies cost differences. The KruskalWallis tests were used to examine cost differences by HH grade. Statistical differences used parametric or non-parametric tests, where appropriate, and 2-tailed P values <0.05 were considered significant. All statistical analyses were conducted in Stata SE software version 12 (StataCorp, College Station, Texas, USA). RESULTS Between January 1, 2013 and December 31, 2015, there was a total of 269 consecutive patients with SAH treated by clipping or coiling. Two hundred nine patients were treated by coiling, 60 patients were treated by clipping, and 7 patients were treated by both (this cohort was excluded). Patient demographics, HH grade, aneurysm size, LOS, and disposition were directly compared for both groups and found to be statistically similar. There was a statistically younger mean age in the clipping group (49 vs. 55 years, P ¼ 0.006; Tables 1 and 2). Total cost per patient for treatment and hospital stay was $74,192 in the clipping group and $85,553 in the coiling group (P ¼ 0.059; Table 3). Direct cost was $54,182 for patients in the clipping group and $65,601 (P ¼ 0.007) for patients in the coiling group. Overhead cost was approximately equal from both treatment groups, $20,010 in the clipping group versus $19,993 in the coiling group (P ¼ 0.47).
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.12.028
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
Table 1. Demographics Clipping Number [ 60
Coiling Number [ 209
Total Number [ 269
37 (62)
144 (69)
181
23
65
88
49 (14)
55 (14)
54 (14)
Black
11 (18)
42 (20)
53
White
23 (38)
91 (44)
114
4 (7)
7 (3)
11
All other
21 (35)
64 (31)
85
Unknown
1 (2)
5 (2)
6
Grade I
4 (6.7)
15 (7.2)
19
Grade II
29 (48.3)
77 (36.8)
106
Grade III
20 (33.3)
69 (33)
89
Grade IV
5 (8.3)
34 (16.3)
39
Grade V
2 (3.3)
13 (6.2)
15
<5 mm
29
76
105
5e10 mm
26
112
142
10e15 mm
4
18
22
15e20 mm
1
2
3
20e25 mm
0
1
1
P value
Characteristics Sex Female Male Age
0.293
0.0061
Race
Asian Pacific Islander
0.594
Hunt and Hess grade 0.401
Aneurysm size 0.261
Data demonstrates patient demographics including race, age, and sex. Clinical characteristics including Hunt and Hess grade, aneurysm size, and the treatment modality used to treat (surgical clipping vs. endovascular coiling). The P value for aneurysm size (0.261) only includes the 3 categories (<5 mm, 5e10 mm and 10e15 mm) because the other categories had an N that was too low to consider for P value comparisons with the other categories.
Table 2. Length of Stay by Mean, Standard Deviation, 25th Percentile, Median, and 75th Percentile Characteristic Average length of stay
Routine days
Intermediate care unit days
Intensive care unit days
Clipping
Coiling
Total
P value (Clip vs. Coil Only) 0.3528
15 (10)
16 (10)
16 (10)
9e13e17.5
10e14e20
10e14e19
2.7 (4.7)
2.7 (4.3)
2.7 (4.3)
0e1.5e3
0e1e3
0e1e3
1.6 (2.7)
1.5 (2.7)
1.5 (2.7)
0e.5e2
0e0e2
0e0e2
10.8 (4.98)
12 (6.1)
11.8 (5.9)
8.5e10e12.5
8e11e15
8e11e14
WORLD NEUROSURGERY -: e1-e6, - 2019
0.7052
0.7006
0.2652
www.journals.elsevier.com/world-neurosurgery
e3
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
Table 3. Mean Cost per Case
Table 5. Cost for Supplies per Case
Procedure
Clipping
Coiling
60
209
Patient volume
P value*
Procedure Patient volume
Clipping
Coiling
60
209
P value*
Direct cost per case
$54,182
$65,601
0.0073
Room (room cost only)
$17,272
$19,386
0.23
Overhead cost per case
$20,010
$19,932
0.456
Hospital services
$1,199
$1,094
0.38
Total cost (direct þ overhead) per case
$74,192
$85,553
0.0592
Hospital supplies
$3,154
$3,500
0.96
Medications
$4,174
$4,225
0.40
Labs
$2,050
$2,351
0.21
*Wilcoxon rank sum test P value.
Cost increased with worse HH grade, except for the grade 5 patient in the clipping group (N ¼ 2; average LOS ¼ 6.0 days; average cost ¼ $36,330; Table 4). This was owing to early deaths related to poor prognosis and withdrawal of care. In the coiling group, cost continued to increase for HH grade 5 patients (N¼14; average LOS ¼ 22.4 days; average cost ¼ $91,933). Excluding the HH grade 5 cohort in both groups for low numbers, the average cost changed slightly for clipping and coiling to $75,105 and $83,545, respectively (P ¼ 0.027), and the average LOS changed to 15.5 and 15.9 days (P ¼ 0.7), respectively. Room cost, hospital services, medications, laboratory tests, intravenous fluids, blood products, and imaging costs were statistically similar for both clipping and coiling groups. There was a statistical difference between angiography consumables ($4,224 vs. $25,872; P < 0.0001), OR consumables ($16,112 vs. $3,786; P < 0.0001), and anesthesia ($2,864 vs. $1,637; P < 0.0001; Table 5). This resulted in a total surgical and procedural cost of $23,200 for the clipping group and $31,295 for the coiling group. LOS and disposition were directly compared for both groups and found to be statistically similar. Clinical outcomes at 6 months were available for 266 patients in the form of GOSE. We found that the mean GOSE for patients who underwent clipping was 6.16 (N ¼ 60) and those who underwent coiling was 6.0 (N ¼ 206). DISCUSSION Optimized use of resources is becoming increasingly important in modern medicine as the cost of health care increases. One method
Table 4. Mean Cost by Hunt and Hess Clinical Grade Procedure
Clipping
Coiling
60
209
Grade 1
$58,208
$42,555
Grade 2
$44,601
$57,685
Grade 3
$66,555
$66,650
Grade 4
$64,177
$83,379
Grade 5
$36,330
$91,933
P value*
0.0308
0.0001
Patient volume
*Kruskal-Wallis test P value.
e4
www.SCIENCEDIRECT.com
IV fluids
$117
$142
0.03
Angiography consumables
$4,224
$25,872
<0.0001
OR consumables
$16,112
$3,786
<0.0001
Anesthesia
$2,864
$1,637
<0.0001
Blood products
$411
$542
0.27
Misc. imaging
$1,167
$1,212
0.88
Misc. radiology
$304
$448
0.63
IV, intravenous; OR, operating room; Misc., miscellaneous. *Wilcoxon rank sum test P value.
to understand the benefit of an intervention in health care is to perform standardized cost analyses correlated with subsequent effects on quality adjusted life years. For this to be done, accurate treatment costs associated with a specific disease entity are required. Although these data have been pursued more deliberately in spine literature, little cost data exists to accurately perform analyses in the cranial realm.13 Our overriding goal for this study was to determine and report accurate cost data. These data are unique from hospital charges or reimbursement—data points that can lead to misleading results if used in cost effectiveness analyses.10,13 This analysis calculated the cost of SAH and coiling was slightly higher than SAH and clipping, although this was not significant, and was $74,192 for clipping and $85,553 for coiling. Although the cost difference is notable, it is our opinion that this in no way implies preference for 1 treatment over another considering that equipoise was not established, and the aneurysms were treated on an individual basis to provide the best clinical outcome. Furthermore, patients undergoing coiling were older than those undergoing clipping, and this practice is consistent with literature, specifically the International Study of Unruptured Intracranial Aneurysms and ISAT, that suggests coiling may be a better treatment modality in older patients.14,15 Despite the older average age of the coiled patients, resource used for hospital admission and LOS was comparable in both coiling and clipping groups, as was clinical outcome. This is a testament to the usefulness of coiling technology in the older population in appropriately selected aneurysms. To get some perspective on the results from this article, we can compare our cost data to a previous publication from 1999, prior to the widespread use of endovascular coiling.5 As reported in 1997 U.S. dollars, the cost of treating an aneurysmal SAH was
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.12.028
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
$47,000.5 Inflating this to 2015 U.S. dollars makes this roughly $70,600, a figure that is comparable to our results for SAH and clipped aneurysms. Contrarily, an analysis of the ISAT costs in the United Kingdom cohort showed that clipping was roughly £20,000 in 2004 (w$46,000 in 2015) for all costs inclusive of 12 months post-SAH. This seems to be dramatically lower than current U.S. costs and may be because of differences in health care systems. These differences highlight the importance of local cost determination for accurate economic analyses. The mean LOS for patients in our study who underwent endovascular coiling was 16 days, which seems to reflect some national trends for SAH in the United States.8,16 There are other U.S. studies that had a slightly longer LOS for patients who underwent endovascular coiling (19.8e20.2 days).6,16 When compared with international data taken from the ISAT trial, our LOS is slightly shorter (16.0 days vs. 19.4 days).15 The overall mean LOS for patients who underwent surgical clipping in our study was 15.0 days, which varied from other studies in which the LOS was between 19.7 to 23 days.4,6,8,16 International data taken from ISAT showed an average LOS of 23.3 days in patients who underwent surgical clipping. In contrast to other studies comparing outcomes of clipping and coiling, in which treatment selection is through randomization, the patients in our article were selected for each treatment based on aneurysm geometry and patient demographics. The similar LOS and outcomes (GOSE at 6 months) among the 2 groups may demonstrate the benefit of optimized treatment selection for each patient. Comparing postcoiling and clipping interventions, our data showed that the cost of posttreatment angiograms during admission did not disproportionately influence the overall cost of treatment in 1 group over the other. We also did not find that there was a difference in the number of posttreatment angiograms during admission between patients who underwent clipping and those that underwent coiling, resulting in no difference in cost between the 2 groups.
2. de Rooij NK, Linn FH, van der Plas JA, Algra A, Rinkel GJ. Incident of subarachnoid hemorrhage: a systematic review with emphasis on region, age, and time trends. J Neurol Neurosurg Psychiatry. 2007; 78:1365-1372. 3. Siegel J, Milton S, Weinsten C, Russel L, Gold M. Recommendations for reporting cost-effective analysis. J Am Med. 1996;276:16. 4. Hoh BL, Chi YY, Lawson MF, Mocco J, Barker FG 2nd. Length of stay and total hospital charges of clipping versus coiling for ruptured and unruptured adult cerebral aneurysms in the
CONCLUSIONS The detailed cost information reported in this article can be used to help establish appropriate, standardized costs nationally by improving transparency. It can also help provide critical information necessary to develop cost-effective treatment algorithms and screening protocols.
Nationwide Inpatient Sample database 2002 to 2006. Stroke. 2010;41:337-342.
cost after cerebral aneurysm clipping. J Neurointerv Surg. 2016;8:316-322.
5. Johnston SC, Dudley RA, Gress DR, Ono L. Surgical and endovascular treatment of unruptured cerebral aneurysms at university hospitals. Neurology. 1999;52:1799-1805.
9. Bekelis K, Gottlieb DJ, Su Y, et al. Medicare expenditures for elderly patients undergoingsurgical clipping or endovascular intervention for subarachnoid hemorrhage. J Neurosurg. 2017;126: 805-810.
REFERENCES 1. Shea A, Reed S, Curtis L, et al. Characteristics of nontraumatic subarachnoid hemorrhage in the United States in 2003. Neurosurgery. 2007;61: 1131-1138.
To investigate why the cost of HH grade 1 patients who underwent clipping was higher than those who underwent coiling, each patient chart was examined individually. There was a total of 4 HH grade 1 patients who underwent surgical clipping. It was determined that 1 of these patients suffered an internal capsule infarct during surgical clip placement. Complications from the stroke contributed to a prolonged hospital stay resulting in higher cost of admission. Another patient suffered an infarct secondary to vasospasm that resulted in complications contributing to a prolonged hospital stay and increased cost of admission. The increased overall cost for admission in these 2 patients resulted in an increased average admission cost for HH grade 1 patients who underwent clipping. There are several limitations of our study. The sample size for clipping is relatively small compared to the coiling group (60 vs. 209), which reflects the practice pattern that more aneurysms are coiled than clipped at our institution. Additionally, we only examined costs for a single admission and did not examine costs for subsequent re-admissions, rehabilitation, nursing home, or long-term acute care hospital, and we did not examine the longterm costs associated with disability. Furthermore, overhead costs may have larger variability as wages, power, water, maintenance, and janitorial costs will differ across facilities and impact the total cost to treat patients. Also not included is the cost of long-term imaging follow-up required for coiled aneurysms, which is likely to be higher than clipped aneurysms. For complete clarification of health care costs in this population of patients, these data would need to be described via a longitudinal follow-up study.
6. Chang H, Shin S, Suh S, Kim B, Rho M. Costeffective analysis of endovascular coiling versus neurosurgical clipping for intracranial aneurysms in Republic of Korea. Neurointervention. 2016;11: 86-91. 7. Brinjikji W, Kallmes D, Lanzino G, Cloft H. Hospitalization costs for endovascular and surgical treatment of ruptured aneurysms in the United States are substantially higher than Medicare payments. Am J Neuroradiol. 2012;33: 1037-1040. 8. Bekelis K, Missios S, MacKenzi T, Labropoulos N, Roberts D. A predictive model of hospitalization
WORLD NEUROSURGERY -: e1-e6, - 2019
10. Finkler S. The distinction between cost and charges. Ann Intern Med. 1982;96:102-109. 11. Bairstow P, Dodgson A, Linto J, Khangure M. Comparison of cost and outcome of endovascular and neurosurgical procedures in the treatment of ruptured intracranial aneurysms. Australian Radiol. 2002;46:249-251. 12. Wolstenhome J, Rivero-Arias O, Gray A, et al. Treatment pathways, resource use, and cost of endovascular coiling versus surgical clipping after aSAH. Stroke. 2008;39:111-119.
www.journals.elsevier.com/world-neurosurgery
e5
ORIGINAL ARTICLE DANIEL MONSIVAIS ET AL.
13. Zygourakis C, Kahn J. Cost-effectiveness research in neurosurgery. Neurosurg Clin N Am. 2015;26:189-196. 14. Wiebers D, Whisnant J, Hutson J, Meissner I, Brown R. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet. 2003; 362:103-110. 15. Ryttlefors M, Per E, Kerr R, Molyneux A. International subarachnoid aneurysm trial of neurosurgical clipping versus endovascular coiling subgroup analysis of 278 elderly patients. Stroke. 2008;39:2720-2726.
e6
www.SCIENCEDIRECT.com
COST ANALYSIS FOR RUPTURED INTRACRANIAL ANEURYSMS
16. Hoh B, Chi Y, Dermott M, Lipori P, Lewis S. The effect of coiling versus clipping of ruptured and unruptured cerebral aneurysms on length of stay, hospital cost, hospital reimbursement, and surgeon reimbursement at the University of Florida. Neurosurgery. 2009;64: 614-621.
The content of this manuscript was presented as a poster at the Congress of Neurologic Surgeons 2017 Annual Meeting in Boston, Massachusetts. Received 8 August 2018; accepted 4 December 2018 Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2018.12.028 Journal homepage: www.journals.elsevier.com/worldneurosurgery
Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.12.028