Abstract
Background and Purpose
Intracranial aneurysm treatment using flow-diverters and flow-disruptors requires a higher level of expertise when compared to more traditional methods. Our hypothesis was that the procedural success and the rate of complications are dependent on the annual case load of a center.
Materials and Methods
Conducting a retrospective analysis on the Database of the German Society for Interventional Radiology for the years 2020 to 2021, we examined flow-diverter and flow-disruptor procedures. We categorized centers into four groups according to their annual case load and proceeded to analyze success rates, complication rates, and fluoroscopy times across these centers.
Results
No statistically significant differences were observed among the groups in both flow-diverter and flow-disruptor cases concerning fluoroscopy time and the incidence of technical complications. However, within the subgroup of flow-disruptor cases, centers with lower case load exhibited significantly higher rates of hemorrhagic and clinically relevant complications. Additionally, it was noted that the rate of therapeutic success in the flow-diverter group significantly increased in centers with higher case volumes.
Conclusion
Our findings support the intention towards centralization of medical care especially for complex neuroendovascular procedures. Furthermore, our findings are an argument to further develop a standardized educational and procedural algorithm based on defined case numbers and training modules for complex neurovascular procedures as already implemented by the Database of the German Society for Interventional Radiology.
Keywords: Flow-diverter, case load, flow-disruptor, Database of the German Society for Interventional Radiology, intracranial aneurysm
Introduction
Endovascular therapy of ruptured and unruptured intracranial aneurysms represents a safe and effective treatment option with high rates of functional independence and low rates of poor outcome.1,2 Standard coiling, stent-assisted coiling or the remodeling technique represent traditional techniques with proven efficacy for saccular aneurysm morphologies.3,4 However, complex morphologies as broad-based, large or fusiform aneurysms require modern endovascular techniques like intra- or extra-saccular flow modulating devices.5,6 This resulted in the introduction of extra-aneurysmal flow-diverters and intrasaccular so-called flow-disruptors that meanwhile proved a high clinical and anatomical success rate in complex otherwise difficult-to-treat morphologies.
In brief, an extraaneurysmal flow-diverter is a flexible, low-porosity stent that allows to occlude an aneurysm by redirecting the blood flow along the parent artery, away from the aneurysm without intrasaccular manipulation during the procedure. An intrasaccular flow-disruptor consists of a woven spherical implant that, when placed inside an aneurysm, allows to reduce the aneurysm inflow, resulting in an intra-aneurysmal thrombosis. The neck area of these devices acts as scaffold for neointimal growth.7,8
Procedures with those devices require a high level of neuroendovascular expertise as hemorrhagic or ischemic complications can occur.
Several studies addressing outcomes after aneurysm treatments whether treated by clipping or coiling reported lower morbidity and mortality rates and higher success rates in high-volume centers compared to centers with a lower number of cases.9,10
Our hypothesis was that these findings are transferable to the novel and more complex neuroendovascular procedures with flow-modulating devices. We thought to prove that the success rate and the frequency of complications in procedures using flow-diverters or flow-disruptors are dependent on the level of experience of the institution, assuming that in centers with higher case volumes, the success rate is higher while morbidity and mortality rates are lower.
We, therefore, analysed the Database of the German Society for Interventional Radiology and Minimally Invasive Therapy (DeGIR) regarding aneurysm treatment using either flow-diverters or flow-disruptors with regard to success rate and the ratio and severity of complications related to the case load of the institution.
Materials and methods
The DeGIR database is a nationwide pseudonymized register of (neuro)-interventional procedures carried out across German centers, intended for comprehensive quality evaluation and professional policy interests of the associated societies of Radiology and Neuroradiology. Participation is still voluntary for institutions; however, professional certifications of institutions require to enter the requested data into this registry. This explains the high participation rate among German (neuro)—interventional centers.
For the execution of this work, a data registry was made accessible by DeGIR following a positive ethics committee approval, containing nationwide pseudonymized data from the years 2020 and 2021. The inclusion criteria for our retrospective analysis were: Endovascular intracranial aneurysm treatment procedures using flow-diverters or flow-disruptors between. Procedures using standard coiling, stent-assisted coiling or balloon-remodeling were excluded.
The following data was recorded: Patient age, aneurysm morphology, aneurysm size, rupture status, technical success rate, aneurysmal occlusion initially after treatment, complications within the first 72 h following the procedure and fluoroscopy time of each procedure.
While patient age, aneurysm morphology, aneurysm size, and rupture status are case or patient-specific variables that are not influenceable by the experience of a center and the interventionalist these were excluded from further analysis following our hypothesis.
We defined fluoroscopy time, incidence of complications, success rate of the procedure, and aneurysms occlusion as variables that are at least in part influenced by the experience of the center or the interventionalist. This might be not only due to the different technical experience but also the potentially wider experience in case selection among higher volume centers.
The fluoroscopy time was analyzed as an indirect indicator for the duration of each procedure.
Aneurysm occlusion was defined as complete occlusion versus incomplete occlusion following both flow-diverter and flow-disruptor cases. More detailed data regarding specific occlusion scales tailored for hemodynamically active implants were not recorded within the DeGIR datasets.
Complications are subdivided into three subgroups within the database. These are hemorrhagic complications, clinically relevant complications and technical complications. While for hemorrhagic and technical complications only yes or no is requested, clinically relevant complications are recorded as mild clinical deficit, severe clinical deficit and death.
Success of a procedure is recorded as “successful” (aneurysm occlusion) versus “unsuccessful” (incomplete aneurysm occlusion). Among the cases with a result of incomplete aneurysm occlusion the further subgroups of “incomplete aneurysm occlusion unplanned” or “failure/abortion” of the intended procedure were defined as unsuccessful cases.
After identification of the absolute number of cases with flow-diverters or flow-disruptors at each hospital within the sample-period a stratification was made based on the center-specific case load.
Therefore, four groups were defined for both flow-diverter and flow-disruptor cases. For flow-diverter cases groups were subdivided as follows: Hospitals with a case load below 5 during the 2-year period were allocated into group 1, those with 6–11 cases and those with 12–24 cases formed groups 2 and 3, respectively. Group 4 comprises hospitals with more than 25 cases within 2 years.
For flow-disruptor cases groups were as follows: 1–3 cases in group 1, 4–7 cases in group 2 and 8–15 cases in group 3; centers with more than 16 cases within the 2 years formed group 4.
Statistical analysis
All statistical analyses were performed with IBM SPSS Statistics, Version 28.0.0.0 (Armonk, NY: IBM Corp.).
Quantitative variables were summarized in tables displaying the numbers of individuals, medians, maximum, and minimum values. Qualitative variables were described in terms of frequencies and percentage of the number of individuals examined.
A hierarchy of hypotheses was tested consecutively at an error level of alpha = 0.05 with two-sided parametric test. Chi-square tests were used to compare categoric variables. A p-value <0.05 was considered statistically significant.
Ethical statement
The study has been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. A separate informed consent from each patient before inclusion in this study was not required due to the retrospective design of the study.
Results
Among the 117 centers included in our study, a total of 3297 cases met the inclusion criteria, involving 3128 patients over a 2-year period. Within this cohort, there were 2080 flow-diverter cases (1976 patients) and 1217 flow-disruptor cases (1152 patients), distributed among 110 and 104 centers, respectively.
In the subgroup of flow-diverter cases 1899/2080 (91.3%) were saccular and 181/2080 (8.7%) were fusiform or mycotic aneurysms. 79.6% (1152/1899) of the saccular aneurysms were small (1–7 mm fundus diameter). 40.5% (842/2080) of the aneurysms in this group were acutely ruptured.
In the subgroup analysis of flow-disruptor cases, 90.5% (1101/1217) of the aneurysms were saccular and 9.5% (116/1217) were fusiform or mycotic aneurysms. 78.9% (869/1101) of the saccular aneurysms were small (1–7 mm fundus diameter). In total, 45.1% (549/1217) of the aneurysms treated with a flow-disruptor were acutely ruptured.
Center-specific case loads varied widely, ranging from 1 to 156 for the entire cohort. The range narrowed slightly in the subgroups of flow-diverter cases (1–129) and flow-disruptor cases (1–85).
The stratification of the centers into the four groups subdivided into flow-diverter and flow-disruptor is summarized in Table 1(a) and (b).
Table 1(a).
Flow-diverter cases categorized into four groups based on case volume.
| Flow-diverter-groups | Case load | Number of centers N |
Total cases N (%) |
Median |
|---|---|---|---|---|
| I | 1–5 | 30 | 88 (4.2%) | 3 |
| II | 6–11 | 28 | 261 (12.5%) | 10 |
| III | 12–24 | 25 | 419 (20.8%) | 16 |
| IV | 25–129 | 27 | 1257 (62.5%) | 38 |
| Total | 1–129 | 110 | 2080 (100%) | 11 |
1(b).
Flow-disruptor cases categorized into four groups based on case volume.
| Flow disruptor-groups | Case load | Number of centers N |
Total cases N (%) |
Median |
|---|---|---|---|---|
| I | 1–3 | 28 | 50 (4.1%) | 1.5 |
| II | 4–7 | 27 | 138 (11.3%) | 5 |
| III | 8–15 | 23 | 234 (19.2%) | 9 |
| IV | 16–85 | 26 | 795 (65.3%) | 25.5 |
| Total | 1–85 | 104 | 1217 (100%) | 7 |
Further analysis included the above-defined variables technical success, clinically relevant complications, hemorrhagic complications, and fluoroscopy time.
There was no statistically significant difference among the groups for both flow-diverter and flow-disruptor cases regarding fluoroscopy time (p = 0.796 for group of flow-diverter and p = 0.752 for group of flow-disruptor) and the occurrence of technical complications (p = 0.665 for group of flow-diverter and p = 0.777 for group of flow-disruptor). Therefore, these parameters are not shown in the tables below.
In the subgroup of flow-diverter cases hemorrhagic and clinically relevant complications were not significantly higher in centers with a lower case load. However, in flow-disruptor cases, centers with lower case volumes exhibited significantly higher rates of hemorrhagic and clinically relevant complications. For instance, the occurrence of hemorrhagic complications was 10% in group 1 compared to 1.4% in group 4 (p = 0.001), and clinically relevant complications occurred in 14% of cases in group 1 compared to 4.3% in group 4 (p = 0.02).
A summary of the variable “complications” is given in Table 2(a) and (b) for flow-diverter and flow-disruptor cases, respectively.
Table 2(a).
Complications in the subgroup of flow-diverter cases.
| Case load (flow-diverter) | Cases | Hemorrhagic complications | Clinically relevant complications | ||||
|---|---|---|---|---|---|---|---|
| Range | N | No | Yes | p | No | Yes | p |
| 1–5 | 88 | 87 (98.9%) | 1 (1.1%) | 0.188 | 82 (93.2%) | 6 (6.8%) | 0.671 |
| 6–11 | 261 | 257 (98.5%) | 4 (1.5%) | 250 (95.8%) | 11 (4.2%) | ||
| 12–24 | 432 | 419 (97.0%) | 13 (3.0%) | 409 (94.7%) | 23 (5.3%) | ||
| 25–129 | 1299 | 1280 (98.5%) | 19 (98.5%) | 1241 (95.5%) | 58 (4.5%) | ||
| Total | 2080 | 2043 (98.2%) | 37 (1.8%) | 1982 (95.3%) | 98 (4.7%) | ||
2(b).
Complications in the subgroup of flow-disruptor cases.
| Case load flow-disruptor | Cases | Hemorrhagic complications | Clinically relevant complications | ||||
|---|---|---|---|---|---|---|---|
| Range | N | No | Yes | p | No | Yes | p |
| 1–3 | 50 | 45 (90.0%) | 5 (10%) | 0.001 | 43 (86.0%) | 7 (14.0%) | 0.02 |
| 4–7 | 138 | 132 (95.7%) | 6 (4.3%) | 129 (93.5%) | 9 (6.5%) | ||
| 8–15 | 234 | 230 (98.3%) | 4 (1.7%) | 222 (94.9%) | 12 (5.1%) | ||
| 16–85 | 795 | 781 (98.2%) | 14 (1.4%) | 761 (95.7%) | 34 (4.3%) | ||
| Total | 1217 | 1188 (97.6%) | 29 (2.4%) | 1155 (94.9%) | 62 (5.1%) | ||
Furthermore, the analysis revealed a statistically significant difference in the rate of unsuccessful procedures among centers with lower case volumes within the subgroup of flow-diverters (6.8% in group 1 vs. 3.2% in group 4, p = 0.012). No such differences were found in the subgroups of flow-disruptors.”
Tables 3(a) and (b) illustrate the variable “success” of the intended procedure for flow-diverter and flow-disruptor cases.
Table 3(a).
Success rate in the subgroup of flow-diverter cases.
| Case load—Flow-diverter | Success | Total N | p | |
|---|---|---|---|---|
| No | Yes | |||
| 1–5 | 6 (6.8%) | 82 (93.2%) | 88 | 0 . 012 |
| 6–11 | 18 (6.9%) | 243 (93.1%) | 261 | |
| 12–24 | 13 (3.0%) | 419 (97%) | 432 | |
| 25–129 | 42 (3.2%) | 1257 (96.8%) | 1299 | |
| Total | 79 (3.8%) | 2001 (96.2%) | 2080 | |
Table 3(b).
Success rate in the subgroup of flow-disruptor cases.
| Case load—Flow-disruptor | Success | Total N | p | |
|---|---|---|---|---|
| No | Yes | |||
| 1–3 | 3 (6%) | 47 (94%) | 50 | 0.760 |
| 4–7 | 5 (3.6%) | 133 (96.4%) | 138 | |
| 8–15 | 7 (3.6%) | 227 (97%) | 234 | |
| 16–85 | 32 (4%) | 763 (96%) | 795 | |
| Total | 47 (3.9%) | 1170 (96.1%) | 1217 | |
Discussion
There is a general trend toward centralization of medical care following evidence about the fact that procedural outcomes and technical success of invasive procedures is linked to the case volume of a center and the individual experience of the operator.9–12
Our retrospective study investigated whether the center-specific case volume among German neurointerventional centers is associated with higher success and a lower complication rate regarding specialized neuroendovascular procedures with flow-diverters or flow-disruptors. Our data analysis is based on the nationwide DeGIR registry. The results prove our hypothesis in part. We found a statistically significant correlation between case volume and hemorrhagic and clinically relevant complications regarding flow-disruptor procedures. Furthermore, a correlation between flow-diverter cases and procedural success was proven. Other predefined variables as fluoroscopy time and technical complications did not show statistically relevant differences dependent on the center-specific case load. The procedural success rate did not exhibit statistically significant variations among the centers for flow-disruptor cases, nor did the rate of hemorrhagic and clinically relevant complications in the case of flow-diverters.
Initially, these varying results seem surprising, but with a more detailed view the findings might be comprehensible. First, the woven endobridge (WEB) device, which was used in the vast majority of all flow-disruptor cases, is a technically comparatively easy-to-apply device known for a high technical success rate. Mouchtouris et al. described in their series of 115 WEB-treated aneurysms a technical success rate of 92%. Based on the findings of our analysis this might allow for the conclusion that the success rate of aneurysm treatment with flow-disruptors does not correlate with the case load of a center because placement of a flow-disruptor is not extremely challenging for an interventionalist even with a lower experience. These findings are in line with the findings of Daube et al. 13 who did not found an influence of operator experience on technical and clinical results of WEB procedures of intracranial aneurysms. They concluded a short learning curve for this device.
However, these findings might be influenced by an unidentified selection bias. Cases selected for treatment with a flow-disruptor at a center with a lower case volume might potentially be ideal straightforward aneurysm morphologies (eg. Bifurcation-type aneurysms with a low angle between the parent artery and the aneurysm dome). Probably cases performed at low-volume centers are more often accompanied by an experienced interventionalist acting as a proctor. This might explain a success rate comparable to higher volume centers, where more complex cases are treated. Centers with a higher case volume are furthermore those with an educational commitment (university hospitals) where younger less experienced interventionalists are trained, which might influence the success rate and rate of complications negatively.
However, the rate of hemorrhagic and clinically relevant complications was significantly higher in centers with lower case volumes regarding flow-disruptor procedures. Even though the success rate of procedures with flow-disruptors seems to be independent from case volume this finding supports our hypothesis that complications are higher in centers with a lower case volume, which might indicate the lower experience of those centers. The discrepancy results from the definition of variables given by the entry mask of the DeGIR. A procedure can be successful even though a complication occurred and vice versa. Furthermore, the incidence of complications and their clinical consequence is not only related to the complication itself but also to the medical management and the accompanying processes around. The risk of disability and death following an intraprocedural hemorrhage is dependent on immediate and proper action (immediate reversal of anticoagulants, hypotension, endovascular bail out actions as balloons or emergency coiling, immediate external ventricular drainage). These actions require experienced teamworking based on standard operating processes, which will potentially be more effective in the setting of large volume centers compared to a center with a lower case load. 14
Another finding is the significantly higher rate of unsuccessful procedures in centers with lower case volumes regarding procedures where flow-diverters are used. These findings are matching also our personal experience with flow-diverters: proper placement of a flow-diverter with accurate opening and wall apposition of the device requires experience not only in the process of placing the device but also in selecting the correct size, which is crucial for a successful procedure. 15 Delgado Almandos et al. 16 analyzed the influence of the operator experience on intra-procedural difficulties in the deployment of a flow-diverter in 150 aneurysms treated with the pipeline device. They could prove that the risk of intra-procedural difficulties and neurological morbidity decreases significantly with increased operator experience.
Limitations
The main limitation of our analysis is associated with the underlying source data. The nationwide DeGIR database is a voluntary registry intended for quality and professional policy interests. We are aware that coding inaccuracies and errors occur and that quality parameters are assessed retrospectively by the operator, which will with no doubt create a selection bias. The parameters defined in order to prove our hypothesis were restricted to the predefined data frame of the registry. Another main limitation is the lack of information regarding the individual experience of the interventionalist, probably a more appropriate parameter given our hypothesis.
However, this database is the largest nationwide registry regarding neuroendovascular procedures and our results are mainly in line with comparable studies within the literature.
Conclusions
The results of our study underscore the importance of centralization of medical care for complex neuroendovascular procedures. Specifically, within the subgroup of flow-disruptor cases, our findings indicate a significant association between lower case volumes and higher rates of hemorrhagic and clinically relevant complications. This suggests that concentrating these procedures in centers with higher case volumes may lead to improved patient outcomes. Furthermore, our observation of increased therapeutic success in the flow-diverter group among centers with higher case volumes highlights the potential benefits of specialized expertise and experience.
Based on these results, we advocate for the development and implementation of standardized educational and procedural protocols, that are tailored to the complexities of neurovascular interventions. Such protocols should consider defined case numbers and structured training modules to ensure the delivery of high-quality care across all centers performing these procedures.
This analysis might also serve as an argument to refine the DeGIR questionnaire to create a more detailed dataset for upcoming scientific questions.
Footnotes
Author`s contribution: FY contributed to the data collection, manuscript preparation, and writing. VM contributed to the manuscript preparation and writing. WW contributed to the critical review of the manuscript. CA contributed to the statistical analysis and critical review of the manuscript. AB contributed to the critical review of the manuscript. SF contributed to the project development, data collection, manuscript preparation and writing.
Availability of data and material: On request to the corresponding author.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethical approval statement: This retrospective study, utilizing pseudonymized data, received ethical approval from the Ethics Committee of the Faculty of Medicine at Ruhr University Bochum with registration number 22-7490-BR.
Funding: The authors received no financial support for the research, authorship, and/or publication of this article.
ORCID iD: Furkan Yapici https://orcid.org/0009-0004-3165-5293
References
- 1.Haverkamp C, Kaier K, Shah M, et al. Cerebral aneurysms: Germany-wide real-world outcome data of endovascular or neurosurgical treatment from 2007 to 2019. J Neurointervent Surg 2024; 16: 365–371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Zaitoun MM, Malky IE, Winklhofer S, et al. Unassisted and multiple microcatheter coiling of distal basilar aneurysms: outcomes and literature review. Interv Neuroradiol 2022; 28: 169–176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gawlitza M, Soize S, Barbe C, et al. Aneurysm characteristics, study population, and endovascular techniques for the treatment of intracranial aneurysms in a large, prospective, multicenter cohort: results of the analysis of recanalization after endovascular treatment of intracranial aneurysm study. AJNR Am J Neuroradiol 2019; 40: 517–523. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Velasco Gonzalez A, Stracke O, Nordmeyer H, et al. Low rates of recanalization for wide-necked aneurysms treated with stenting after balloon-assisted coiling: combination of techniques delivers stable and improved results during follow-up. Neuroradiology 2018; 60: 1223–1230. [DOI] [PubMed] [Google Scholar]
- 5.Dmytriw AA, Salem MM, Yang VXD, et al. Endovascular flow disruption: a new frontier in endovascular aneurysm management. Neurosurgery 2020; 86: 170–181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.White TG, Santhumayor BA, Turpin J, et al. Flow diverter surface modifications for aneurysm treatment: a review of the mechanisms and data behind existing technologies. Interv Neuroradiol 2023. Online ahead of print. doi: 10.1177/15910199231207550 [DOI] [PubMed] [Google Scholar]
- 7.Dandapat S, Mendez-Ruiz A, Martinez-Galdamez M, et al. Review of current intracranial aneurysm flow diversion technology and clinical use. J Neurointervent Surg 2021; 13: 54–62. [DOI] [PubMed] [Google Scholar]
- 8.Hecker C, Broussalis E, Griessenauer CJet al. et al. A mini-review of intrasaccular flow diverters. J Neurointervent Surg 2022; 15: 70–74. [DOI] [PubMed] [Google Scholar]
- 9.Brinjikji W, Rabinstein AA, Lanzino G, et al. Patient outcomes are better for unruptured cerebral aneurysms treated at centers that preferently treat with endovascular coiling: a study oft he national inpatient sample 2001-2007. AJNR Am J Neuroradiol 2011; 32: 1065–1070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zacharia BE, Bruce SS, Carpenter AM, et al. Variability in outcome after elective cerebral aneurysm repair in high volume academic medical centers. Stroke 2014; 45: 1447–1452. [DOI] [PubMed] [Google Scholar]
- 11.Kim JY, Kang JK, Kim BJ, et al. Annual case volume and one-year mortality for endovascular treatment in acute ischemic stroke. J Korean Med Sci 2022; 37: e270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kawczynski MJ, van Kuijk AMJ, Olsthoon JR, et al. Type A aortic dissection: Optimal annual case volume for suergery. Eur Heart J 2023; 44: 4357–4372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Daube P, Cagnazzo F, Morganti R, et al. Influence of operator experience on the technical and clinical results of woven EndoBridge endovascular treatment for intracranial aneurysms. Clin Neurol Neurosurg 2021. Sep; 208: 106900. [DOI] [PubMed] [Google Scholar]
- 14.Ihn YK, Shin SH, Baik SKet al. et al. Complications of endovascular treatment for intracranial aneurysms: management and prevention. Interv Neuroradiol 2018. Jun; 24: 237–245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gravino G, Masri S, Chandran Aet al. et al. Management of WEB device migration and mal-position in endovascular treatment of cerebral aneurysms. Interv Neuroradiol 2022. Online ahead of print. doi: 10.1177/15910199221122857 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Delgado Almondes J, Kayan Y, Tenreiro A, et al. Clinical and angiographic outcome in patients with intracranial aneurysms treated with the pipeline embolization device: intra-procedural technical difficulties, major morbidity, and neurological mortality decrease significantly with increased operator experience in device deployment and patient management. Neuroradiology 2017. Dec; 59: 1291–1299. [DOI] [PubMed] [Google Scholar]