JOURNAL OF NEUROSCIENCE AND NEUROSURGERY (ISSN:2517-7400)

Background: Endovascular (EV) treatment of a SAH has increased in the last 20 years. Results of surgical (SC) and endovascular treatments are comparable, but EV has higher rates of retreatments. The cost of endovascular devices is high and makes this treatment expensive.

Object: To evaluate the results and costs of surgical treatment against endovascular in SAH.

Methods: Retrospective review of SAH patients was conducted, treated endovascularly (EV) and surgically (SC). A descriptive study of demographic, clinical and radiological aspects have been carried out. Clinical and angiographic results and economic costs have been evaluated in both globally and in each of the groups.

Results: 231 patients were reviewed (82.5% EV; 17.5% SC). A 71.2% of the patients presented favorable GOS (4-5) at 6 months (80% SC; 69, 3% EV). Mortality was 7,9% (all in SV group). The angiographic results were better in the SC group (88.6% of complete occlusion versus 50%), as well as the stability of the treatment (19.5% of retreatments versus 0%). The Average Length of Stay (ALOS) was higher in the SC group, however, overall costs after follow-up and retreatment were 12.3% higher in the EV group. 

Conclusion: Clinical results are comparable. The rate of complete occlusion of the SC group was higher and the need for retreatment lower. Costs of the EV group were significantly higher than the SC group due to the follow up and the rate of retreatment. Adequate selection of candidates for endovascular coiling could improve angiographic outcomes and save costs.

Aneurysmal Subarachnoid Haemorrhage (SAH) is an important cause of death and disability, with an incidence of 6-12/100,000 in Western world. Endovascular treatment (EV) has been gaining ground against Surgical Treatment (SC) in an exponential way, mainly as a result of the publication of the ISAT study in 2002 [1,2]. The conclusions of this study have been questioned in subsequent publications [1,3-5] and nowadays the results of both treatments could be considered similar. Retreatment rates in EV are around 20% (less than 1% in surgical ones) and re-bleeding (even the very late ones) are more frequent in this group. EV requires prolonged follow-up with imaging techniques that ensure the stability of the treatment. The current trend indicates a progressive increasing in the indications for endovascular treatment, in parallel with the development of new devices and materials of embolization. Indeed, an argument against EV could be the high cost of embolization material compared to the surgical one and in this sense, several publications try to elucidate the cost of one and another option for treatment [6-15].

The present study aims to evaluate the clinical and angiographic results as well as the costs of surgical and endovascular treatment of a series of cases with SAH treated in a unique hospital. The hypothesis is that the EV is more expensive than the SC without differences in the clinical results.

The SAH diagnosed patients treated between the years 2010 and 2015 were revised. Demographic data (age and sex), clinical (Fisher grade and Hunt-Hess scale at admission), anatomical characteristics of the aneurysm (location and size) and results (GOS at 6 months) of both series of treatment were analyzed. The average length of stay (ALOS) (ICU and ward days) and costs of treatment and retreatment (number of coils, stents, guide catheters, balloons, craniotomy, clips, etc.) were recorded, as well as expenses related to the follow-up of these patients (routine arteriographies, MRI angiography). The costs of stays, procedures and materials were obtained from hospital sources (Table 1). A descriptive study of both series was carried out and compared in terms of age, sex and clinical and radiological characteristics. For the calculation of costs, patients who died and those who were not adequately followed up were excluded. The follow-up of the intervened patients differed from embolized ones; generally, the SC patients under- went with a routine arteriography during the first admission, without requiring other standardized test. An arteriography at 6 months and another one at 12 months were performed in embolized patients to continue with an annual MRI angiography if the treatment was considered stable, for a period of not less than 5 years. If a recanalization was observed, patients were programmed for a new treatment, restarting the cycle of arteriographies and MRI angiography. The costs of stays, procedures and materials were obtained from hospital sources (Table 1). A descriptive study of both series was carried out and compared in terms of age, sex and clinical and radiological characteristics. For the calculation of costs, patients who died and those who were not adequately followed up were excluded. The follow-up of the intervened patients differed from embolized ones; generally, the SC patients under- went with a routine arteriography during the first admission, without requiring other standardized test. An arteriography at 6 months and another one at 12 months were performed in embolized patients to continue with an annual MRI angiography if the treatment was considered stable, for a period of not less than 5 years. If a recanalization was observed, patients were programmed for a new treatment, restarting the cycle of arteriographies and MRI angiography.

Due to the retrospective nature of this study, ethics committee approval and patient consent were not required.

In the descriptive analysis, the numerical variables were expressed as mean and standard deviation; median and percentiles in cases that did not meet the normality hypothesis. Qualitative variables were described by absolute and relative frequencies. The normality of the continuous variables was contrasted by the Kolmogorov-Smirnov test. A bivariate analysis was performed to compare possible differences between groups. For the quantitative variables, in the non-parametric cases, the Student’s t test or Mann Whitney were used; chi-square test of Pearson or Fisher were used for the qualitative ones. The significance level considered for all assays was 0.05. The software used was IBM SPSS Statistics 19. 

Between January 2010 and December 2015, 231 SAH patients with aneurysmal were treated. Two patients were excluded due to lack of basic data so that, 229 patients were considered finally, 189 of the EV group (82.5%) and 40 of the SC group (17.5%). Eighteen patients died before 6 months, which means a 7.9% overall mortality in the series. All deaths occurred in the endovascular treatment group (9.6% mortality). The mean follow- up of the series was 43.8 months (6-77), slightly higher in the SC group (49.5 versus 43 months).

Age range from 18 to 87 years (mean 55.6 years). About gender, bleeding).

Routinely arteriography was available in 89% of cases (67% SC and 96% EV). The reason for not performing this control was a poor clinical situation in the cases of the EV or the use of intraoperative ICG (technical indocyanine green) in the SC. Complete occlusion (CO) was achieved in 56.4% of the patients, leaving a rest of neck in the others. This occlusion was superior in the SC group (88.6% versus 50%). Of the cases treated endovascularly with CO at the time of treatment, recanalization was observed in 24.9% of patients. Of these, 68% had to be retreated while the rest keeps in follow-up (Table 3). The overall rate of retreatment in the EV group was 16.9%. Most patients were retreated only once (84%), while other patients were retreated 2 to 5 times. Retreatment rates of certain aneurysms are above average (ICA, MCA, BA, PCom and PICA). The average size of the recanalized and retreated aneurysms is also higher than the mean (7.8 mm and 8.5 mm respectively versus the 6.5 mm of the mean). The mean number of coils used per patient in the first treatment was 4.6 and in the retreatments 7.15. Similarly, the use of stents and flow diverters was more frequent in retreatments.

The ALOS in patients with SAH was 30 days, of which 7.7 days corresponded to the stay in the ICU (Table 3). There are certain factors that influence in it (Figure 2):

Follow-up and cost calculations exclude deceased patients [16] and those who were not adequately followed up [2], so that the calculations were performed with 209 patients (169 EV and 40 SC).

The total expenditure generated by these patients was 7,718,550€, of which 83% corresponds to the EV group. This represents an expense of 1,286,425€ per year. Of this expenditure, 68.1% corresponds to the hospitalization of the patient, 24.6% to the different interventions performed to treat ruptured aneurysm and 7.3% to follow-up. These percentages differ from the EV to the SC group. In this way, hospitalization costs are higher in the SC group and treatment and follow-up are higher in the EV group (Figure 3).

Considering only the first admission, surgical treatment is 6.7% more expensive than the endovascular one, due to the longer ALOS. When the follow-ups are added, the overall expenditure in the EV group exceeds the SC by 1.2%. When the retreatments are included, the mean expenditure per patient in the EV group is significantly higher than in the SC group (12.3% more expensive).

The average expenditure of embolization in the EV group was 6,585€, increasing to 10,500€ when retreatments were included. The total expenditure on embolization material reached 1,195,900€ (an average higher than 7,000€ per patient). In the SC group, the average cost per intervention was 3,810€ (this group did not require retreatments). Considering only the intervention to ensure the aneurysm, the endovascular treatment is 176% more expensive than the surgical one (Figure 4).

The controversy regarding the treatment of brain aneurysms has been and still is, a major point in Neurosurgery. The EV treatment has gained acceptance and nowadays most of the aneurysms are treated by endovascular procedures. This treatment is less aggressive and easier to perform than the surgical, but it is also less stable and the material s more expensive. We hypothesize that the expenses of EV treatment are bigger than the SC and there is no difference in clinical results.

The ISAT study meant a milestone in the management of intracranial aneurysmal pathology [17]. According to those results the morbimortality of the SC versus the EV treatment advised the last one as the first treatment option. After its initial impact, subsequent studies [1,3-5,18] showed different bias in this study (selection of surgeons and cases, stratification of results) and the same group published later another study with results of this same series after 5 years in which the supposed advantage of endovascular treatment was diluted, mainly due to cases of rebleeding and retreatment of the endovascular group [5]. Rebleeding occurs sometimes in embolized aneurysms and has high morbidity and mortality with a rate around 2% [4,16,19], closely related to the degree of occlusion of the aneurysm. Rebleeding uses to appear more often in the first year after treatment, but late rebleeding is possible [20].

Nowadays, it is possible to say that the clinical results of both techniques are similar in expert hands, with the endovascular techniques in their favor the less aggressiveness of the procedure and the surgical ones their stability along time.

Incomplete occlusion is frequent in endovascular treatments, with percentages that in the best and most populated series are around 25% [19, 21-23]. Total occlusion in surgical cases is more common, around 95% [24]. Incomplete occlusion does not imply retreatment, but evolution of the neck remains unpredictable, requiring a prolonged follow-up [25, 26]. Recanalization is estimated around 15-20% of treated cases [4,16,17,22] and are described even very late in the follow up, requiring most of the times retreatment.

Lower stability of the EV makes that the follow-up of treated patients by this technique have to be more tight and prolonged, with routinely arteriographies and serial image tests over long periods of time. For costs calculation of this therapeutic modality it is necessary to include the expenses derived from this follow-up, which involve readmissions and invasive tests, in addition to those costs generated by the retreatments. To date there is no consensus about how long these patients should be followed.

In the literature there are several studies that try to elucidate which of the two treatments is the most economical in broken aneurysms. Characteristics and results of this publications are summarized in Table 4. Most of them do not take into account followup or retreatments when calculating costs. Some works are indirect studies based on previously published series [6,8,9,13,24] while others are based on patient series [10,11,14,18]. Most of the papers do not include the expenses of follow up and retreatment. From the literature review, it can be deduced that costs of EV are generally higher than SC and in those publications which differ, costs of followup and retreatment are not included, an important part of the overall expenditure per patient.

It is interesting to mention the work of Zubair et al. [26] that studied costs of both treatment including 6 months of follow-up in 55 patients in a Pakistani hospital, concluding that EV is 62.5% more expensive than SC, without morbimortality results had favored to one or another group. It was mainly due to the fact that cost of hospitalization was relatively low, so the increase in hospital stay generated by the surgery (2 more days on average) did not impact on the final expenditure as the embolization materials. They pro- posed a decision algorithm based on the evaluation of a series of factors (patient age, Hunt-Hess, location and size of aneurysm), in order to the final recommendation taking into account risks associated with clipping and inherent costs to embolization (Figure 5).

Current series does not present clinically significant differences between the two groups in terms of age, sex, clinical status at admission or Fisher grade. Clinical results were similar in both groups, with a better result in the surgical series although the number of patients did not allow obtaining significant conclusions. Results were related to age, initial Hunt-Hess score and Fisher grade (Figure 1). Relationship of results with age has been established in other studies, such as Koffiberj et al. [14].

Complete occlusion rate of aneurysm is higher in SC group (88.6% versus 50%) as well as the stability of treatment. In EV group, a recanalization rate of around 25% and a retreatment rate of around 20% were recorded. It is interesting to highlight that both re- canalization and retreatment were more common in large aneurysms, as well as that the retreatment rate is above average in aneurysms whose surgical risk is relatively low, (MCA, PCom or CoA).

Regarding costs generated, results of these series indicate that EV, as a procedure, is considerably more expensive than SC (73% more expensive), a difference greater when retreatments are included (175% more expensive). This fact compensates the expenses derived from hospitalization between both modalities, which are on average 19% more expensive in SC, so that the cost of the first admission is discreetly favorable to EV (6% cheaper). However, this benefit is diluted over time because of follow-up and especially retreatments, so that the expenditure per patient including only follow-up is 1.2% higher in the EV group, a difference that is increased after including retreatment, where expenditure in the EV group is 12.4% higher (Figure 4). This difference should be higher if a correct follow up had been made.

In this series, as in others published in literature, the procedure itself constitutes a substantial part of the expenditure in endovascular group, specifically 27.8%, whereas in surgical group hospital stay generates more than 88% of expenditure (Figure 3). Price of embolization materials is mainly responsible for this high cost, which is significantly higher when, in addition to coils, stents or flow diverters are used.

It looks like evident that EV is advantageous in certain aneurysms, such as posterior circulation and especially in the vertebrobasilar segment, whereas the clinical results of surgical treatment are excellent in anterior-circulation aneurysms6, with the advantage of its stability, therefore, generalizing and proposing surgical treatment in all aneurysms is not reasonable. Size of the aneurysm is directly related to the cost of the procedure, so Zubair et al. [26] recommended surgical treatment for those cases of large aneurysms and favorable surgery. Following the model of Zubair (Figure 5), it could be estimated the savings that would occur in current series by selecting cases for surgical or endovascular treatment based on age, clinical status and complexity and size of the aneurysm. With these parameters, surgical treatment would be indicated in 48% of the patients in this series, compared to 18.3% in those that were performed. This means that, after adjusting the average costs per patient in this series of both techniques, these new percentages gives a total expense of 7,469,516€, about 250,000€ less than the expenditure finally realized (42,000€ per year).

The presented series has limitations. First is that it is a retrospective study and expenditure estimations have been made using various sources and sometimes indirect methods. The second is that it does not evaluate the differences that in terms of labor reincorporation rate could exist between both treatments, the socalled loss of profit, which could influence in the savings generated by one or other technique. 

In the shown series, the morbimortality results obtained by both treatments are similar, while the angiographic results and stability of SC are superior to EV. Although the endovascular treatment has a shorter ALOS, the total cost is significantly higher, mainly due to the price of the embolization materials, the rate of retreatment and the follow up required. An adequate selection of patients could improve the angiographic results, the rate of retreatment and thus save costs. 

1. Mitchell P, Kerr R, Mendelow AD, Molyneux A. Could late rebleedingoverturn the superiority of cranial aneurysm coil embolizationover clip ligation seen in the International SubarachnoidAneurysm Trial? J Neurosurg. 2008 Mar;108(3):437-442. 
   
2. Andaluz N, Zuccarello M. Recent trends in the treatment ofcerebral aneurysms: analysis of a nationwide inpatient database. J Neurosurg. 2008 Jun;108(6):1163-1169.
   
3. Bakker NA, Metzemaekers JD, Groen RJ, Mooij JJ, Van Dijk JM.International subarachnoid aneurysm trial 2009: endovascularcoiling of ruptured intracranial aneurysms has no significantadvantage over neurosurgical clipping. Neurosurgery. 2010May;66(5):961-962.
   
4. Molyneux AJ, Kerr RS, Birks J, Ramzi N, Yarnold J, et al. Risk ofrecurrent subarachnoid haemorrhage, death, or dependenceand standardised mortality ratios after clipping or coiling ofan intracranial aneurysm in the International SubarachnoidAneurysm Trial (ISAT): long-term follow-up. Lancet Neurol. 2009May;8(5):427-433.
   
5. Spetzler RF, Mc Dougall CG, Zabramski JM, Nakaji P, WallaceRC. The Barrow Ruptured Aneurysm Trial: 6-year results. J Neurosurg. 2015 Sep;123(3):609-617. 
   
6. Brinjikji W, Kallmes DF, Lanzino G, Cloft HJ. Hospitalization costsfor endovascular and surgical treatment of ruptured aneurysmsin the United States are substantially higher than Medicarepayments. AJNR Am J Neuroradiol. 2012 Jan;33(1):49-51.
   
7. Halkes PH, Wermer MJ, Rinkel GJ, Buskens E. Direct costsof surgical clipping and endovascular coiling of unrupturedintracranial aneurysms. Cerebrovasc Dis. 2006 Sep;11(2):86-91.
   
8. Chang HW, Shin SH, Suh SH, Kim BS, Rho MH. Cost-EffectivenessAnalysis of Endovascular Coiling versus Neurosurgical Clipping forIntracranial Aneurysms in Republic of Korea. Neurointervention.2016 Sep;11(2):86-91.
   
9. Hoh BL, Chi YY, Dermott MA, Lipori PJ, Lewis SB. The effectof coiling versus clipping of ruptured and unrupturedcerebral aneurysms on length of stay, hospital cost, hospitalreimbursement and surgeon reimbursement at the university ofFlorida. Neurosurgery. 2009 Apr;64(4):614-619.
   
10. Hoh BL, Chi YY, Lawson MA, MoccoF, Barker FJ. Length of stay andtotal hospital charges of clipping versus coiling for ruptured andunruptured adult cerebral aneurysms in the Nationwide InpatientSample database 2002 to 2006. Stroke.2010 Feb;41(2):337-342.
    
11. Javadpour M, Jain H, Wallace MC, Willinsky RA, Brugge T, et al.Analysis of cost related to clinical and angiographic outcomes ofaneurysm patients enrolled in the international subarachnoidaneurysm trial in a North American setting. Neurosurgery. 2015May;56: 886-894.
    
12. Koffijberg H, Buskens E, Rinkel GJ. Aneurysm occlusion in elderlypatients with aneurysmal subarachnoid haemorrhage: a costutility analysis. J Neurol Neurosurg Psychiatry. 2011 Jul;82:718-727.
    
13. Maud A, Lakshminarayan K, Suri MF, Vazquez G, Lanzino G, et al.Cost-effectiveness analysis of endovascular versus neurosurgicaltreatment for ruptured intracranial aneurysms in the UnitedStates. J Neurol Neurosurg Psychiatry. 2011 Jul;82(7):727.
    
14. Zubair Tahir M, Enam SA, Pervez Ali R, Bhatti A, Ul Haq T . Costeffectiveness of clip- ping vs coiling of intracranial aneurysmsafter subarachnoid hemorrhage in a developing country--aprospective study. Surg Neurol. 2009 Oct;72(4):355-360. 
    
15. Wolstenholme J, Rivero-Arias O, Gray A, Molyneux AJ, Kerr RS, etal. Treatment pathways, resource use and costs of endovascularcoiling versus surgical clipping after aSAH. Stroke. 2008Jan;39:1171-1179.
    
16. CARAT Investigators. Rates of delayed rebleeding fromintracranial aneurysms are low after surgical and endovasculartreatment. Stroke. 2006 Jun;37:1437-1442.
    
17. Molyneux A1, Kerr R, Stratton I, Sandercock P, Clarke M, etal. International Subarachnoid Aneurysm Trial (ISAT) ofneurosurgical clipping versus endovascular coiling in 2143patients with ruptured intracranial aneurysms: a randomisedtrial. Lancet. 2002 Oct; 26;360(9342):1267-1274.
    
18. Bairstow P, Dodgson A, Linto J, Khangure M. Comparison of cost and outcome of endovascular and neurosurgical procedures inthe treatment of ruptured intracranial aneurysms. AustralasRadiol. 2002 Sep; 46:249-251.
    
19. Johnston SC, Dowd CF, Higashida RT, Lawton MT, Duckwiler GR, et al. Predictors of rehemorrhage after treatment of rupturedintracranial aneurysms: the Cerebral Aneurysm Rerupture AfterTreatment (CARAT) study. Stroke. 2008 Jan;39:120-125.
    
20. Sluzewski M, Van Rooj WJ, Beute GN, Nijssen PC. Latere bleedingof ruptured intra- cranial aneurysms treated with detachablecoils. AJNR Am J Neuroradiol. 2005 Nov-Dec;26:2542-2549.
    
21. Murayama Y, Nien YL, Duckwiler G, Gobin YP, Jahan R, et al. Guglielmi detachable coil embolization of cerebral aneurysms:11 years’ experience. J Neurosurg. 2003 May;98:959-966.
    
22. Gallas S, Pasco A, Cottier JP, Gabrillargues J, Drouineau J, et al. A multicenter study of 705 ruptured intracranial aneurysmstreated with Guglielmi detachable coils. AJNR Am J Neuroradiol.2005 Aug;26(7):1723-1731.
    
23. Plowman RS, Clarke A, Clarke M, Byrne JV. Sixteen-year singlesurgeon experience with coil embolization for ruptured intracranialaneurysms: recurrence rates and incidence of late rebleeding.Clinical article. J Neurosurg. 2011 Mar;114(3):863-874.
    
24. Thornton J, Debrun GM, AletichVA, Bashir Q, Charbel FT, etal. Follow-up angiography of intracranial aneurysms treatedwith endovascular placement of Guglielmi detachable coils.Neurosurgery. 2002 Feb;50(2):249-250.
    
25. Byrne JV, Sohn MJ, Molyneux AJ, Chir B. Five-year experience in using coil embolization for ruptured intracranial aneurysms:outcomes and incidence of late rebleeding. J Neurosurg.1999;90:656-663.
    
26. Campi A, Ramzi N, Molyneux AJ, Summers PE, Kerr RS, etal. Retreatment of ruptured cerebral aneurysms in patientsrandomized by coiling or clipping in the InternationalSubarachnoid Aneurysm Trial (ISAT). Stroke. 2007;38:1538-1544.