Article Text

Brain aneurysm and parent vessel remodeling after flow diversion treatment: a proposed modification for Cekirge-Saatci classification (mCSC)
  1. Ricardo A Hanel1,
  2. Gustavo M Cortez1,
  3. Demetrius Klee Lopes2,
  4. Isil Saatci3,
  5. H Saruhan Cekirge3,4
  1. 1 Lyerly Neurosurgery, Baptist Neurological Institute, Jacksonville, Florida, USA
  2. 2 Brain and Spine Institute, Advocate Aurora Health, Park Ridge, Illinois, USA
  3. 3 Radiology Department, Koru Health Group, Ankara, Turkey
  4. 4 Private Office, Saruhan Cekirge, Ankara, Turkey
  1. Correspondence to Dr Isil Saatci, Koru Health Group, Ankara, Turkey; isaatci{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Endovascular therapy is the preferred treatment approach for managing ruptured intracranial aneurysms after studies demonstrated an overall safer profile than surgical clipping.1 Evolving and refining technology also settled neurointervention as the main approach for treating unruptured aneurysms worldwide, with few exceptions.2 3 A key element to attest to the impact and establish the efficacy of endovascular surgery is the development of universal scales to quantify aneurysm occlusion after treatment. Numerous classifications have been proposed to document aneurysm occlusion and guide patient surveillance imaging.4–6 The Raymond-Roy classification was originally developed to measure the degree of aneurysm occlusion after coiling and has been the preferred tool for assessing intervention outcomes.4

The introduction of flow diverters (FDs) led to a significant paradigm shift.7 FDs primarily work by redirecting blood flow away from the aneurysm and providing scaffolding for endothelialization across the neck, which results in exclusion from circulation.8 The mechanism of parent vessel remodeling commonly evolves over weeks to months.8 It is not uncommon, however, that aneurysms with a side branch (coming from the dome or neck) or aneurysms involving a bifurcation take longer to occlude after flow diversion.9 This delay in occlusion results from the demand for blood flow generated by a pressure gradient between the parent vessel and branch.10

Several classifications have been proposed to address the limitations of the Raymond-Roy classification in aneurysms treated with FDs.6 11 12 For the specific assessment of aneurysms treated with such stents, proposed grading scales have commonly taken into account aneurysm filling, concomitantly with either the degree of contrast stasis inside the aneurysm or parent artery status.6 11 12 Nonetheless, none of these factors has been demonstrated to be a reliable marker of aneurysm occlusion. Particular limitations of evaluating contrast stagnation include variable injection power, hemodynamic parameters, acquisition parameters, and reading subjectivity.13

Cekirge and Saatci recently suggested a new aneurysm occlusion classification with five categories, which, in particular, intended to address incorporated branches arising from treated aneurysms and introduce the concept of aneurysm remodeling while keeping the basic categories of Raymond-Roy classification (class 1, total occlusion; class 2, neck remnant; and class 3, aneurysm remnant).13 The scale has been applied for evaluating radiological outcomes of aneurysms treated with FD, including complex bifurcation aneurysms.14 The natural history of aneurysms with incomplete occlusion post-coiling (residual aneurysm or neck remnant) had been associated with an increased risk of (re)bleeding, and retreatment may be advisable to achieve complete occlusion and secure the aneurysm in these cases.15 16 Nonetheless, residual aneurysms filling after FD treatment appears to be more favorable, thus calling for different nomenclature that emphasizes these differences in the mechanisms of occlusion and on the natural history of aneurysms treated with FDs.10 17 We propose a modification for the Cekirge-Saatci classification (mCSC) to create a universal language, improving the documentation of aneurysms treated with FDs and assisting in the differentiation of cases considered treatment failures requiring retreatment, expected delayed occlusion, or permanent occlusion with remodeling. After reviewing our personal experiences applying the CSC to aneurysms treated with FDs over the last 5 years, we have identified important stages that aneurysms undergo after FD placement, and thus, we would like to update the original scale to include these.

In our current proposal, as previously published on PREMIER 3-year data analysis,18 the original CSC occlusion classes 1, 2, and 4 remain unchanged. In summary, complete aneurysm occlusion is adjudicated as class 1, but in the presence of a branch arising from or involving the aneurysm sac, complete aneurysm occlusion is further subdivided according to the existence of a full patent branch (class 1A), reduced caliber branch (class 1B), or occluded branch (class 1C). Angiographic definition of residual neck filling (class 2) remains unaltered. Class 4 is reserved exclusively for immediate postoperative results to indicate the presence (class 4A) or absence (class 4B) of contrast stagnation based on the end-of-treatment digital subtraction angiogram after FD/modifier placement. Our proposed modification aims to elaborate on the concept of aneurysm remodeling, previously adjudicated as CSC class 5. This concept was initially portrayed in patients treated with FDs as either shrinkage of the aneurysm with an associated ‘infundibulum-like’ enlargement of the branch coming off the sac or a tortuous proximal course of the branch replacing the previous proximal aneurysm sac. It also refers to filling in the neck region which stays unchanged or reduced in at least two consecutive angiograms which are at least 6 months apart. Our revision is aiming to elucidate the long-term impact of size changes and stability of aneurysms with residual filling further depicting the dynamic remodeling following FD placement. The potential protection conferred to remodeled aneurysms after FD is reinforced by long-term data of prospective studies demonstrating no cases of aneurysm rupture after 180 days, despite aneurysm filling in some cases.19 20

To ensure that the observed changes in aneurysms are considered stable, two angiographic controls (at least 6 months apart, extending over 1 year of the index procedure) are required. Because aneurysms with an incorporated vessel branch differ from sidewall aneurysms without an incorporated vessel, the remodeling process is classified into different categories to cement the distinctive patterns as follows.

Decrease in size of an aneurysm with residual filling restricted to the site of an incorporated branch (applies to sidewall aneurysms with an incorporated branch or bifurcation aneurysms):

  • Class 5 —Stable remodeling: Aneurysm initially shows residual filling at the integrated branch site but with reduced dimensions compared with baseline (class 3), remaining stable (class 5) in size at subsequent control angiogram (figure 1A).

  • Class 5A —Progressive remodeling: Aneurysm initially shows residual filling at the integrated branch site but with reduced dimensions compared with baseline (class 3), followed by a further size decrease (class 5A) in subsequent follow-up (figure 1A,B).

Figure 1

Flow diagram examples of aneurysms treated with a flow diverter and adjudicated according to the modified Cekirge-Saatci classification (mCSC). (A) Sidewall aneurysm with an incorporated branch. (B) Bifurcation aneurysm. (C) Aneurysm with no incorporated branch. FU, follow-up.

Decrease in size of aneurysm with residual filling with no incorporated branches:

  • Class 3A Stable or progressive: Aneurysm with no integrated branch initially shows residual filling after treatment (class 3), either remaining stable or further decreasing in size (class 3A) in subsequent angiographic controls (figure 1C).

Increase in size of a filling aneurysm in the second or subsequent control angiogram, regardless of the presence of an incorporated branch:

  • Class 3C Growth: Aneurysm shows sac filling after treatment, with unchanged or reduced dimensions (class 3) on initial follow-up, followed by growth (class 3C) on subsequent angiogram (figure 1A,C).

If angiographic follow-up demonstrates that aneurysm sac filling and dimensions remain unaltered compared with baseline, the aneurysm is consistently adjudicated as class 3. Moreover, because two consecutive angiographic controls are needed to confirm the stability of the findings, we still classify as mCSC class 3 when a filling aneurysm with reduced dimensions is first encountered, and only in the subsequent follow-up the newly presented classes should be regarded (figure 1A–C). Online supplemental table summarizes the proposed classification after modification, and case examples are available in the supplementary material (online supplemental video 1).

Supplemental material

Supplementary video

The modified classification intends to emphasize the concept of aneurysm remodeling and the potential stability of lesions treated with FDs. The mechanisms of healing post-primary coiling and the significance of residual aneurysm filling after coiling cases with regards to the risk of aneurysm (re)rupture are likely to be different from FD-treated cases. This new classification is proposed to better express the timeline and progression of occlusion in cases of aneurysm healing after FD treatment for different types of residual aneurysm filling. Moreover, the classification may also be applied for the results of intrasaccular flow modifier treatments (online supplemental video 2).

Supplementary video

When evaluating the natural history of aneurysms treated with FDs, most occlude between 3 and 12 months after the index procedure, given the fact that the aneurysm segment has been covered completely with good device wall apposition.21 22 The presence of side branches incorporated by the aneurysm is a key factor commonly associated with delayed aneurysm occlusion.9 Although flow-diverters had appeared as an attractive option, continuous aneurysm filling in some cases raised questions about their efficacy in such situations.10 22 This is based on the remark that a pressure gradient in aneurysms incorporating end vessel branches could maintain sac filling. However, this process is often accompanied by progressive or stable aneurysm remodeling (reducing aneurysm dimensions) that may ultimately progress to aneurysm securing in the long term.

During a 5-year follow-up of the Pipeline for Uncoilable or Failed Aneurysms (PUFS) trial, the authors could attest that complete angiographic occlusion, regardless of additional treatment, increases over time, reinforcing the concept that the aneurysm healing process following FD placement occurs progressively.19 Chiu and colleagues depicted this gradual progression to occlusion over time in a retrospective review of patients with more than 2-year follow-up,21 an observation that was further reproduced by Bender et al in a larger set of patients.22 Previously, aneurysm occlusion was treated as a binary variable (occluded vs non-occluded) but, currently, the available evidence supports the progressive character of aneurysm remodeling over time in the setting of FD usage.13 Besides progressive occlusion, large prospective series, including the 5-year results of the PUFS and SCENT trials as well as 3-year data of the PREMIER trial, showed no cases of aneurysm ruptured after 180 days, even in cases of incomplete aneurysm occlusion, implying a likely different mechanism of aneurysm rupture protection provided by FD treatment.18–20

This new nomenclature provides a universal language to get across the different reorganizational patterns in a filling aneurysm treated with FD, whether having an incorporated branch or not. The modification does not intend to redefine the understanding of the natural history of these lesions but to clarify it. More precise naming may facilitate investigating appropriate indications for patient management, including imaging surveillance or impending indication of retreatment. Patients with complete aneurysm occlusion (class 1) have a very low risk for aneurysm recurrence and can have a more relaxed imaging follow-up. Classes 2, 3, and 5 represent residual neck or aneurysm filling. While the term ‘remodeling’ does not necessarily indicate the future stability of the filling portion in all aneurysms with an incorporated branch, more permissive surveillance may be adopted for patients with stable (class 5) or progressive remodeling (class 5A). In cases of class 3C, where documented growth is seen, discussion for retreatment should be emphasized.

There are limitations in the presented classification: 1) This classification is more complex than the previous ones and it may be deemed difficult to be adopted in daily practice. However, the flow modification techniques induce a multifactorial and more complicated aneurysm occlusion process compared with the straightforward ‘packing a sac’. Therefore, basic classifications would not be refined enough to describe the varied results. 2) The intra- and inter-observer variability are to be investigated to test its steadiness. 3) The variability of imaging modalities and protocols during angiographic follow-up may also limit objective evaluation, a drawback that applies to all occlusion classifications. 4) The real impact of remodeling may differ depending on aneurysm baseline characteristics (including size, location, and morphology) and the relative proportion of shrinkage; thus, it may not be scalable.

In conclusion, the proposed mCSC elaborates the concept of aneurysm healing and remodeling post-FD treatment by addressing morphologic changes over time, whether in the presence of an incorporated branch or not. It allows for a more specific denomination of residual aneurysms and may guide the decision for aneurysm surveillance or retreatment.

Supplemental material

Data availability statement

Data sharing not applicable as no datasets generated and/or analyzed for this study.

Ethics statements

Patient consent for publication

Ethics approval

Not applicable.


Supplementary materials


  • Twitter @cure4stroke

  • Contributors All authors have substantially contributed to the conception and design of the study. RAH: contributed to data interpretation, planning, conception and design, manuscript preparation. GMC: contributed to data interpretation, conception and design, manuscript preparation. DKL: contributed to data interpretation, planning, conception and design, manuscript preparation. IS: contributed to data interpretation, planning, conception and design, manuscript preparation. HSC: contributed to data interpretation, planning, conception and design, manuscript preparation.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests RAH is a consultant for Medtronic, Stryker, Cerenovous, Microvention, Balt, Phenox, Rapid Medical, and Q'Apel. He is on the advisory board for MiVI, eLum, Three Rivers, Shape Medical and Corindus. Unrestricted research grant from NIH, Interline Endowment, Microvention, Stryker, CNX. Investor/stockholder for InNeuroCo, Cerebrotech, eLum, Endostream, Three Rivers Medical Inc, Scientia, RisT, BlinkTBI, and Corindus. Unrestricted research grant from NIH, Interline Endowment, Microvention, Stryker, CNX. Investor/stockholder for InNeuroCo, Cerebrotech, eLum, Endostream, Three Rivers Medical Inc, Scientia, RisT, BlinkTBI, and Corindus. GMC has no disclosures to report. DKL is consultant for Medtronic, Stryker, Asahi, Rapid.AI and QApel. Advisory Board: Q’Apel, Siemens, investor/stockholder for Three Rivers, Q’Apel, Vastrax, Synchron, Viz.AI, Methinks, Bendit, eLum, NVI. IS is a consultant and has proctorship agreements with Medtronic and MicroVention Inc. HSC is a consultant and has proctorship agreements with Medtronic and MicroVention Inc. Shareholder of Neuravention, Sychron Inc, Bend It Tech, Sim & Cure and Vesalio LLC.

  • Provenance and peer review Not commissioned; internally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.