Article Text

Download PDFPDF
Original research
WEB as part of a multimodality treatment in complex, large, and partially thrombosed intracranial aneurysms: a single-center observational study of technical success, safety, and recurrence
  1. C Kabbasch1,
  2. A Mpotsaris1,
  3. M Reiner2,
  4. T Liebig3
  1. 1Department of Radiology and Neuroradiology, University Hospital of Cologne, Cologne, Germany
  2. 2Department of Neurosurgery, University Hospital of Cologne, Cologne, Germany
  3. 3Department of Neuroradiology, Charité University Hospital of Berlin, Berlin, Germany
  1. Correspondence to Dr C Kabbasch, Department of Radiology and Neuroradiology, University Hospital of Cologne, Kerpener Str 62, Cologne 50937, Germany; christoph.kabbasch{at}


Introduction Complex, large, wide-neck aneurysms, but particularly partially thrombosed intracranial aneurysms (PTIAs), have a greater tendency to recanalize after coil occlusion. The Woven Endovascular Bridge (WEB) combines shape memory wires braided to a relatively uniform, dense surface, which may limit its compaction and its incorporation into an existing aneurysm clot and may thus reduce PTIA recurrence.

Objective To carry out a retrospective analysis of our experience with the WEB as part of a complex treatment in conjunction with other implants to assess the efficacy and safety of this technique.

Methods Among 43 aneurysms treated with the WEB in our center, eight complex, large, wide-neck aneurysms were treated in conjunction with other implants (additional WEBs, coils, stents, flow diverters). Six of these eight aneurysms were PTIAs. All patients were followed up by DSA between 3½ and 38 months.

Results All eight (100%) of the complex, large, wide-neck aneurysms were treated successfully and without periprocedural adverse events (0%). At follow-up, the two non-thrombosed aneurysms were completely occluded, but all six PTIAs recurred (75%) and were re-treated. There was no morbidity or mortality in these eight patients.

Conclusions Treatment of large, complex aneurysms with the WEB in conjunction with other implants was technically successful and safe but did not prevent recurrence of partially thrombosed aneurysms in our center. PTIA re-treatment was possible and not limited by the previously placed WEB.

  • Aneurysm
  • Angiography
  • Catheter
  • Coil
  • Subarachnoid

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.


Complex, large, wide-neck aneurysms are challenging to treat by endovascular or surgical means. The treatment risk increases with aneurysm size, especially for surgical clipping.1–3 The likelihood of aneurysm recurrence after endovascular coil occlusion is greater for large aneurysms, especially when coils only are used compared with coils with additional remodeling strategies.4 In particular, partially thrombosed intracranial aneurysms (PTIAs) are vulnerable to recanalization after coiling compared with non-thrombosed cases.5–7 Recently, endosaccular flow disruptors have shown promising results in a large series and good clinical practice studies on wide-necked aneurysms.8 ,9 In these series, however, PTIAs were not included. The Woven Endovascular Bridge (WEB) is an intra-aneurysmal flow-disrupting device, available in three different shapes: the WEB dual-layer, single-layer, and single-layer sphere.10 Based on the WEB`s shape memory and its more uniform and densely woven surface, we expected that this technique would be less prone to compaction and also to integration into an intra-aneurysmal clot, and thus might be efficient in reducing recurrence of large and complex aneurysms, especially in the presence of intra-aneurysmal thrombus. A retrospective analysis of our experience with the WEB as part of a complex treatment in conjunction with other implants was performed to assess the efficacy and safety of this technique.

Materials and methods

Patient population

Between May 2011 and August 2015 a total of 43 aneurysms were treated with a WEB (Sequent Medical Inc, Aliso Viejo, California, USA) in our center. Among these 43 aneurysms, eight complex large or giant aneurysms could be identified, in which the WEB had to be used in conjunction with other implants owing to an aneurysm size that exceeded the largest available WEB. Six of these eight aneurysms were partially thrombosed, as proved by either CT or MRI, displaying a contrast filling cavity inside a significantly larger aneurysm, some of these with calcifications of the outer rim (figure 1A, see also table 1). The age of the patients was between the fifth and eighth decade. The mean values for height, diameter of the dome, and neck width were, respectively, 17.1/13.6/5.9 mm, with a maximum measure of the individual complex shape ranging from 12 to 28 mm. One complex aneurysm was ruptured but not thrombosed. All patients were followed up by angiography between 3½ and 38 months. Aneurysm occlusion class was evaluated according to the Raymond–Roy Occlusion Classification.11

Table 1

Patient overview

Figure 1

Patient 3: a giant partially thrombosed intracranial aneurysm of the internal carotid artery at the posterior cerebral artery fetal origin, previously clipped 10 years earlier, compressing the brainstem and partially calcified (A). Before (B) and after (C) endovascular treatment with coils, a Woven Endovascular Bridge (WEB) device, WEB, coils, and a flow diverter, which resulted immediately in complete occlusion. At 3 months, a recanalization of the previously thrombosed regions alongside the WEB can be seen (D). Two additional flow diverters were implanted; the 1-year control shows complete occlusion (E). The patient has largely recovered from his neurological symptoms.



All patients were treated under general anesthesia on a Philips biplane system (Philips, Best, the Netherlands). Planning, including sizing, was performed both on subtracted projection radiography and 3D reconstructions using the maximum intensity projection of volume-rendered datasets. Six patients were treated with dual-layer WEBs and two with a single-layer WEB. All WEBs could be placed as planned, none had to be retrieved or repositioned. There were no procedural adverse events. The individual rationale for using a WEB instead of other intrasaccular or remodeling devices was different from case to case. The most common reason was to avoid the need for long-term antiaggregation in patients who were expected to undergo surgery during their further management. In these cases (patients 1, 2, 4, 5), the WEB was implanted at the level of the neck like a cork, while the dome was filled in with coils. In two cases (patients 3 and 7, figures 1 and 2), the WEB was specifically chosen to avoid an unnecessary large mass of platinum coils that would obscure any radiographic display of the neck region on later follow-up angiograms and presumably invade the present intra-aneurysmal thrombus and thus lead to aneurysm recurrence. In one giant PTIA, two large WEBs were combined with coiling of the distal portion of the dome as well as coiling of the neck and sealing of the neck with a single Pipeline embolization device flow diverter initially (ev3, Irvine, USA) (patient 3, 28 mm PTIA, figure 1). Half of the patients had been receiving at least single antiaggregation at the time of their treatment. Figure 3 gives an overview of all cases, showing the status before treatment, immediately after treatment and at follow-up.

Figure 2

Patient 7: before (A) and 4 months after (B) placement of two Woven Endovascular Bridge WEB) devices. Both WEBs (lines in B) have become completely integrated into the pre-existing thrombus. The outline of the aneurysm is represented by the solid line. (C) The short-term angiographic result at 2 weeks after the initial treatment with complete occlusion and very good reconstruction of the basilar tip.

Figure 3

Overview of all cases, showing the status before treatment (A), immediately after (B), and at follow-up (C).


All patients were followed up by intra-arterial angiography between 3½ and 38 months. The two large non-thrombosed aneurysms (patients 1 and 8) had stable complete occlusions at the time of follow-up, while all PTIAs have recurred. The mechanism of recurrence was not uniform. In two PTIAs, the WEBs were extremely deformed, misplaced from their original position, and integrated into intra-aneurysmal thrombus in a region that had been filled with thrombus at the time of the treatment (patients 7 and 2, figures 2 and 4). In all other PTIAs (patients 3, 4, 5 and 6), the recurrence occurred alongside the WEB with a compartment or channel formation within a pre-existing intra-aneurysmal thrombus (figures 5 and 6). In three patients (patients 4, 5, and 6), this occurred together with a slight rotation or movement of the otherwise non-deformed or reperfused WEB itself. All aneurysms with a recurrence following this pattern were treated with additional coils or stent and coils. At further follow-up, this had resulted in a stable occlusion of the aneurysm at 7–16 months. However, in one patient who had multiple medical conditions, the only follow-up available was obtained during the second hospital stay. This patient was still alive and neurologically unchanged at 1 year after the treatment but refused further investigation.

Figure 4

Patient 2: a partially thrombosed intracranial aneurysm of the anterior communicating artery (A and B). The neck was sealed with a Woven Endovascular Bridge (WEB) single-layer sphere and the dome coiled through a trapped microcatheter, resulting in near complete occlusion with just a small recess at the neck (C); at 6 months the non-coiled portion of the dome has completely recanalized with the WEB compressed and eccentric displacement into a compartment that was filled with thrombus at the time of the initial treatment (D). This could easily be treated with stent-assisted coiling (not shown, further follow-up to be obtained).

Figure 5

Patient 4: a partially thrombosed intracranial aneurysm of an anterior communicating artery aneurysm showing ring enhancement on contrast-enhanced CT (A and B). After sealing the neck with a Woven Endovascular Bridge (WEB) device and coiling the dome (not shown) and initial complete occlusion, the WEB remained stable without compression or reperfusion of the implant itself, but recanalization alongside the WEB (C). The resulting ‘neo-neck’ could successfully be treated with stent-assisted coil occlusion (D).

Figure 6

Patient 5, partially thrombosed intracranial aneurysm of the terminal internal carotid artery (A). A Woven Endovascular Bridge (WEB) device was placed at the level of the neck to avoid stenting, while the dome was filled with coils (B). Despite immediate complete occlusion, a channel formed alongside the WEB (C) that could be coiled with the assistance of a stent, resulting in a stable occlusion at 2 years’ follow-up (D).


Large and giant aneurysms are challenging to treat by any means and there is a lack of dedicated implants to deal with this entity by endovascular treatment, which results in a high rate of recurrence, especially with standard coiling. Intrasaccular flow disruption with the WEB is an appealing alternative because of its shape memory and more uniform filling of the intrasaccular space. This is potentially valuable even if the size of the aneurysm exceeds that of the largest available WEB. PTIAs are a special subset of these complex lesions, often symptomatic for reasons other than subarachnoid hemorrhage.12 In our series, only one non-thrombosed, but complex, aneurysm was ruptured. It has to be pointed out that our series comprised complex aneurysms that were all treated, applying the WEB in an extraordinary fashion and in conjunction with other implants, if needed, in order to deal with an otherwise unmet need for specific dedicated implants. The choice of the implants in each case was based on the specific size and shape and the vascular anatomy (for explanation see right-hand column of table 1).

PTIAs have been regarded as a distinct entity because of their radiographic and histopathologic features. Ring enhancement of the outer margin (figure 5A) may suggest that a surrounding vascular network or proliferation of perianeurysmal vasa vasorum is involved in the pathogenesis.13 Histopathologic examinations have shown the existence of intrathrombotic vascular channels with endothelial lining as the basis for a hypothetical blood supply independent of the aneurysmal lumen and potentially responsible for recurrence and continuous growth even after complete occlusion at the level of the neck.14 PTIAs may also pose difficulties for surgical clipping, because frequently the aneurysmal domes tend to become embedded in deep brain parenchyma and the aneurysm neck tends to be wide.15 ,16 Likewise, it is generally accepted that endovascular coil occlusion in PTIAs is associated with an increased recurrence rate.5–7 Quite often, the entire coil mass is displaced and compacted or integrated into the thrombus at follow-up imaging. This phenomenon might be a result of the continuous water-hammer effect17 of the pulse pressure wave and because the individual coil loop with a quite small contact surface can invade the clot relatively easily. For the same reason, the WEB with its larger and more uniform surface might be expected to circumvent this shortcoming of standard coiling.

We thus evaluated our series of eight large, complex aneurysms treated with the WEB, including six partially thrombosed aneurysms. Eight aneurysms of our series of 43 exceeded the nominal maximum size that is recommended for treatment with an off-the-shelf WEB. Six of these aneurysms were partially thrombosed, as proved by CT or MRI. All treatments were performed as planned and without adverse events. On follow-up imaging only the non-thrombosed aneurysms (patients 1 and 8) were still sufficiently occluded, while all PTIAs had recurred despite initial complete occlusion. Two mechanisms of recurrence were identified: (1) deformation and migration/integration of the WEB into the pre-existing thrombus; (2) recanalization of regions that were thrombosed at the time of the initial treatment alongside the WEB—in one case even through the mesh of an additionally placed flow diverter (figure 1D).

In our small series, we found no specific pattern that would allow prediction of the type of recurrence. However, in those cases where the WEB was integrated into pre-existing clot, the reperfused lumen was similar to the untreated aneurysm (figures 2 and 4), whereas recanalization alongside typically resulted in new cavities that were (1) smaller than would be expected in cases of recurrence after regular coiling and (2) easy to treat with additional coils, partly because of the radiolucency of the WEB that allowed good visualization of the ‘neo-neck’ (figures 5 and 6). In light of these findings, the presence of an intramural rather than intraluminal thrombus in PTIAs, as suggested in previous publications, has to be questioned as we saw complete integration of the WEB into the clot. An intramural thrombus covered with endothelium would presumably prevent complete integration in such a short period, as seen in our series.


Being aware of all limitations of a small series like this, we conclude that the ability of the WEB to overcome some of the drawbacks of coil occlusion in large and complex aneurysms is limited. Although two large aneurysms were successfully occluded at follow-up, all PTIAs in our series did recur. We found two patterns of recurrence of PTIAs but are unable to predict a pattern for an individual case. Although large, complex aneurysms appear to be suitable targets for the WEB, even if the size exceeds that of the largest available version and a multi-modality technique needs to be applied, PTIAs remain challenging lesions, with a high probability of recurrence. Re-treatment is possible, however, and not limited by the previously placed WEB.



  • Contributors CK: manuscript writing, data collection, and patient treatment. AM: manuscript editing and data collection. MR: data collection and patient physical and neurological examination. TL: project development, manuscript writing, and patient treatment.

  • Competing interests AM: proctor and consultant for Penumbra Inc and Sequent Medical. TL: proctor and consultant for Sequent Medical.

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

  • Data sharing statement All data can be made available in an anonymized manner on request.